21 files changed, 704 insertions, 211 deletions

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 6d110a4f7f65..120bd31a46ea 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -10,6 +10,7 @@ config SUPERH
 	select EMBEDDED
 	select HAVE_CLK
 	select HAVE_IDE
+	select HAVE_LMB
 	select HAVE_OPROFILE
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_IOREMAP_PROT if MMU
diff --git a/arch/sh/boards/mach-se/7724/setup.c b/arch/sh/boards/mach-se/7724/setup.c
index c050a8d76dfd..8fed45a2fb85 100644
--- a/arch/sh/boards/mach-se/7724/setup.c
+++ b/arch/sh/boards/mach-se/7724/setup.c
@@ -19,6 +19,7 @@
 #include <linux/smc91x.h>
 #include <linux/gpio.h>
 #include <linux/input.h>
+#include <linux/usb/r8a66597.h>
 #include <video/sh_mobile_lcdc.h>
 #include <media/sh_mobile_ceu.h>
 #include <asm/io.h>
@@ -302,6 +303,34 @@ static struct platform_device sh_eth_device = {
 	.resource = sh_eth_resources,
 };
 
+static struct r8a66597_platdata sh7724_usb0_host_data = {
+};
+
+static struct resource sh7724_usb0_host_resources[] = {
+	[0] = {
+		.start	= 0xa4d80000,
+		.end	= 0xa4d800ff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= 65,
+		.end	= 65,
+		.flags	= IORESOURCE_IRQ | IRQF_TRIGGER_LOW,
+	},
+};
+
+static struct platform_device sh7724_usb0_host_device = {
+	.name		= "r8a66597_hcd",
+	.id		= 0,
+	.dev = {
+		.dma_mask		= NULL,         /*  not use dma */
+		.coherent_dma_mask	= 0xffffffff,
+		.platform_data		= &sh7724_usb0_host_data,
+	},
+	.num_resources	= ARRAY_SIZE(sh7724_usb0_host_resources),
+	.resource	= sh7724_usb0_host_resources,
+};
+
 static struct platform_device *ms7724se_devices[] __initdata = {
 	&heartbeat_device,
 	&smc91x_eth_device,
@@ -311,6 +340,7 @@ static struct platform_device *ms7724se_devices[] __initdata = {
 	&ceu1_device,
 	&keysc_device,
 	&sh_eth_device,
+	&sh7724_usb0_host_device,
 };
 
 #define EEPROM_OP   0xBA206000
@@ -364,6 +394,7 @@ static void __init sh_eth_init(void)
 #define SW4140    0xBA201000
 #define FPGA_OUT  0xBA200400
 #define PORT_HIZA 0xA4050158
+#define PORT_MSELCRB 0xA4050182
 
 #define SW41_A    0x0100
 #define SW41_B    0x0200
@@ -373,6 +404,7 @@ static void __init sh_eth_init(void)
 #define SW41_F    0x2000
 #define SW41_G    0x4000
 #define SW41_H    0x8000
+
 static int __init devices_setup(void)
 {
 	u16 sw = ctrl_inw(SW4140); /* select camera, monitor */
@@ -385,6 +417,12 @@ static int __init devices_setup(void)
 		    (1 << 14)), /* RMII */
 		  FPGA_OUT);
 
+	/* turn on USB clocks, use external clock */
+	ctrl_outw((ctrl_inw(PORT_MSELCRB) & ~0xc000) | 0x8000, PORT_MSELCRB);
+
+	/* enable USB0 port */
+	ctrl_outw(0x0600, 0xa40501d4);
+
 	/* enable IRQ 0,1,2 */
 	gpio_request(GPIO_FN_INTC_IRQ0, NULL);
 	gpio_request(GPIO_FN_INTC_IRQ1, NULL);
diff --git a/arch/sh/include/asm/hwblk.h b/arch/sh/include/asm/hwblk.h
new file mode 100644
index 000000000000..51a46f496639
--- /dev/null
+++ b/arch/sh/include/asm/hwblk.h
@@ -0,0 +1,61 @@
+#ifndef __ASM_SH_HWBLK_H
+#define __ASM_SH_HWBLK_H
+
+#include <asm/clock.h>
+#include <asm/io.h>
+
+#define HWBLK_AREA_FLAG_PARENT (1 << 0) /* valid parent */
+
+#define HWBLK_AREA(_flags, _parent)		\
+{						\
+	.flags = _flags,			\
+	.parent = _parent,			\
+}
+
+struct hwblk_area {
+	unsigned long cnt;
+	unsigned char parent;
+	unsigned char flags;
+};
+
+#define HWBLK(_mstp, _bit, _area)		\
+{						\
+	.mstp = (void __iomem *)_mstp,		\
+	.bit = _bit,				\
+	.area = _area,				\
+}
+
+struct hwblk {
+	void __iomem *mstp;
+	unsigned char bit;
+	unsigned char area;
+	unsigned long cnt;
+};
+
+struct hwblk_info {
+	struct hwblk_area *areas;
+	int nr_areas;
+	struct hwblk *hwblks;
+	int nr_hwblks;
+};
+
+/* Should be defined by processor-specific code */
+int arch_hwblk_init(void);
+int arch_hwblk_sleep_mode(void);
+
+int hwblk_register(struct hwblk_info *info);
+int hwblk_init(void);
+
+/* allow clocks to enable and disable hardware blocks */
+#define SH_HWBLK_CLK(_name, _id, _parent, _hwblk, _flags)	\
+{							\
+	.name		= _name,			\
+	.id		= _id,				\
+	.parent		= _parent,			\
+	.arch_flags	= _hwblk,			\
+	.flags		= _flags,			\
+}
+
+int sh_hwblk_clk_register(struct clk *clks, int nr);
+
+#endif /* __ASM_SH_HWBLK_H */
diff --git a/arch/sh/include/asm/lmb.h b/arch/sh/include/asm/lmb.h
new file mode 100644
index 000000000000..9b437f657ffa
--- /dev/null
+++ b/arch/sh/include/asm/lmb.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_SH_LMB_H
+#define __ASM_SH_LMB_H
+
+#define LMB_REAL_LIMIT	0
+
+#endif /* __ASM_SH_LMB_H */
diff --git a/arch/sh/include/asm/perf_counter.h b/arch/sh/include/asm/perf_counter.h
index 61c2b40c802c..d8e6bb9c0ccc 100644
--- a/arch/sh/include/asm/perf_counter.h
+++ b/arch/sh/include/asm/perf_counter.h
@@ -4,4 +4,6 @@
 /* SH only supports software counters through this interface. */
 static inline void set_perf_counter_pending(void) {}
 
+#define PERF_COUNTER_INDEX_OFFSET	0
+
 #endif /* __ASM_SH_PERF_COUNTER_H */
diff --git a/arch/sh/include/asm/suspend.h b/arch/sh/include/asm/suspend.h
index b1b995370e79..5c8ea28ff7a4 100644
--- a/arch/sh/include/asm/suspend.h
+++ b/arch/sh/include/asm/suspend.h
@@ -10,6 +10,15 @@ struct swsusp_arch_regs {
 	struct pt_regs user_regs;
 	unsigned long bank1_regs[8];
 };
+
+void sh_mobile_call_standby(unsigned long mode);
+
+#ifdef CONFIG_CPU_IDLE
+void sh_mobile_setup_cpuidle(void);
+#else
+static inline void sh_mobile_setup_cpuidle(void) {}
+#endif
+
 #endif
 
 /* flags passed to assembly suspend code */
diff --git a/arch/sh/include/asm/thread_info.h b/arch/sh/include/asm/thread_info.h
index 499e315f4be2..5123bcaa8509 100644
--- a/arch/sh/include/asm/thread_info.h
+++ b/arch/sh/include/asm/thread_info.h
@@ -51,7 +51,7 @@ struct thread_info {
 	.exec_domain	= &default_exec_domain,	\
 	.flags		= 0,			\
 	.cpu		= 0,			\
-	.preempt_count	= 1,			\
+	.preempt_count	= INIT_PREEMPT_COUNT,	\
 	.addr_limit	= KERNEL_DS,		\
 	.restart_block	= {			\
 		.fn = do_no_restart_syscall,	\
diff --git a/arch/sh/include/cpu-sh4/cpu/sh7722.h b/arch/sh/include/cpu-sh4/cpu/sh7722.h
index 738ea43c5038..48560407cbe1 100644
--- a/arch/sh/include/cpu-sh4/cpu/sh7722.h
+++ b/arch/sh/include/cpu-sh4/cpu/sh7722.h
@@ -221,4 +221,18 @@ enum {
 	GPIO_FN_KEYOUT3, GPIO_FN_KEYOUT4_IN6, GPIO_FN_KEYOUT5_IN5,
 };
 
+enum {
+	HWBLK_UNKNOWN = 0,
+	HWBLK_TLB, HWBLK_IC, HWBLK_OC, HWBLK_URAM, HWBLK_XYMEM,
+	HWBLK_INTC, HWBLK_DMAC, HWBLK_SHYWAY, HWBLK_HUDI,
+	HWBLK_UBC, HWBLK_TMU, HWBLK_CMT, HWBLK_RWDT, HWBLK_FLCTL,
+	HWBLK_SCIF0, HWBLK_SCIF1, HWBLK_SCIF2, HWBLK_SIO,
+	HWBLK_SIOF0, HWBLK_SIOF1, HWBLK_IIC, HWBLK_RTC,
+	HWBLK_TPU, HWBLK_IRDA, HWBLK_SDHI, HWBLK_SIM, HWBLK_KEYSC,
+	HWBLK_TSIF, HWBLK_USBF, HWBLK_2DG, HWBLK_SIU, HWBLK_VOU,
+	HWBLK_JPU, HWBLK_BEU, HWBLK_CEU, HWBLK_VEU, HWBLK_VPU,
+	HWBLK_LCDC,
+	HWBLK_NR,
+};
+
 #endif /* __ASM_SH7722_H__ */
diff --git a/arch/sh/kernel/cpu/Makefile b/arch/sh/kernel/cpu/Makefile
index eecad7cbd61e..3d6b9312dc47 100644
--- a/arch/sh/kernel/cpu/Makefile
+++ b/arch/sh/kernel/cpu/Makefile
@@ -19,4 +19,4 @@ obj-$(CONFIG_UBC_WAKEUP)	+= ubc.o
 obj-$(CONFIG_SH_ADC)		+= adc.o
 obj-$(CONFIG_SH_CLK_CPG)	+= clock-cpg.o
 
-obj-y	+= irq/ init.o clock.o
+obj-y	+= irq/ init.o clock.o hwblk.o
diff --git a/arch/sh/kernel/cpu/hwblk.c b/arch/sh/kernel/cpu/hwblk.c
new file mode 100644
index 000000000000..7c3a73deff24
--- /dev/null
+++ b/arch/sh/kernel/cpu/hwblk.c
@@ -0,0 +1,130 @@
+#include <linux/clk.h>
+#include <linux/compiler.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+#include <linux/spinlock.h>
+#include <asm/suspend.h>
+#include <asm/hwblk.h>
+#include <asm/clock.h>
+
+static DEFINE_SPINLOCK(hwblk_lock);
+
+static void hwblk_area_inc(struct hwblk_info *info, int area)
+{
+	struct hwblk_area *hap = info->areas + area;
+
+	hap->cnt++;
+	if (hap->cnt == 1)
+		if (hap->flags & HWBLK_AREA_FLAG_PARENT)
+			hwblk_area_inc(info, hap->parent);
+}
+
+static void hwblk_area_dec(struct hwblk_info *info, int area)
+{
+	struct hwblk_area *hap = info->areas + area;
+
+	if (hap->cnt == 1)
+		if (hap->flags & HWBLK_AREA_FLAG_PARENT)
+			hwblk_area_dec(info, hap->parent);
+	hap->cnt--;
+}
+
+static void hwblk_enable(struct hwblk_info *info, int hwblk)
+{
+	struct hwblk *hp = info->hwblks + hwblk;
+	unsigned long tmp;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hwblk_lock, flags);
+
+	hp->cnt++;
+	if (hp->cnt == 1) {
+		hwblk_area_inc(info, hp->area);
+
+		tmp = __raw_readl(hp->mstp);
+		tmp &= ~(1 << hp->bit);
+		__raw_writel(tmp, hp->mstp);
+	}
+
+	spin_unlock_irqrestore(&hwblk_lock, flags);
+}
+
+static void hwblk_disable(struct hwblk_info *info, int hwblk)
+{
+	struct hwblk *hp = info->hwblks + hwblk;
+	unsigned long tmp;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hwblk_lock, flags);
+
+	if (hp->cnt == 1) {
+		hwblk_area_dec(info, hp->area);
+
+		tmp = __raw_readl(hp->mstp);
+		tmp |= 1 << hp->bit;
+		__raw_writel(tmp, hp->mstp);
+	}
+	hp->cnt--;
+
+	spin_unlock_irqrestore(&hwblk_lock, flags);
+}
+
+static struct hwblk_info *hwblk_info;
+
+int __init hwblk_register(struct hwblk_info *info)
+{
+	hwblk_info = info;
+	return 0;
+}
+
+int __init __weak arch_hwblk_init(void)
+{
+	return 0;
+}
+
+int __weak arch_hwblk_sleep_mode(void)
+{
+	return SUSP_SH_SLEEP;
+}
+
+int __init hwblk_init(void)
+{
+	return arch_hwblk_init();
+}
+
+/* allow clocks to enable and disable hardware blocks */
+static int sh_hwblk_clk_enable(struct clk *clk)
+{
+	if (!hwblk_info)
+		return -ENOENT;
+
+	hwblk_enable(hwblk_info, clk->arch_flags);
+	return 0;
+}
+
+static void sh_hwblk_clk_disable(struct clk *clk)
+{
+	if (hwblk_info)
+		hwblk_disable(hwblk_info, clk->arch_flags);
+}
+
+static struct clk_ops sh_hwblk_clk_ops = {
+	.enable		= sh_hwblk_clk_enable,
+	.disable	= sh_hwblk_clk_disable,
+	.recalc		= followparent_recalc,
+};
+
+int __init sh_hwblk_clk_register(struct clk *clks, int nr)
+{
+	struct clk *clkp;
+	int ret = 0;
+	int k;
+
+	for (k = 0; !ret && (k < nr); k++) {
+		clkp = clks + k;
+		clkp->ops = &sh_hwblk_clk_ops;
+		ret |= clk_register(clkp);
+	}
+
+	return ret;
+}
diff --git a/arch/sh/kernel/cpu/sh4a/Makefile b/arch/sh/kernel/cpu/sh4a/Makefile
index ebdd391d5f42..3cafda696378 100644
--- a/arch/sh/kernel/cpu/sh4a/Makefile
+++ b/arch/sh/kernel/cpu/sh4a/Makefile
@@ -25,7 +25,7 @@ clock-$(CONFIG_CPU_SUBTYPE_SH7780)	:= clock-sh7780.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7785)	:= clock-sh7785.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7786)	:= clock-sh7786.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7343)	:= clock-sh7343.o
-clock-$(CONFIG_CPU_SUBTYPE_SH7722)	:= clock-sh7722.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7722)	:= clock-sh7722.o hwblk-sh7722.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7723)	:= clock-sh7723.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7724)	:= clock-sh7724.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7366)	:= clock-sh7366.o
diff --git a/arch/sh/kernel/cpu/sh4a/clock-sh7722.c b/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
index 40f859354f79..1fa9e1dd1cc8 100644
--- a/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
+++ b/arch/sh/kernel/cpu/sh4a/clock-sh7722.c
@@ -22,6 +22,8 @@
 #include <linux/kernel.h>
 #include <linux/io.h>
 #include <asm/clock.h>
+#include <asm/hwblk.h>
+#include <cpu/sh7722.h>
 
 /* SH7722 registers */
 #define FRQCR		0xa4150000
@@ -140,35 +142,37 @@ struct clk div6_clks[] = {
 	SH_CLK_DIV6("video_clk", &pll_clk, VCLKCR, 0),
 };
 
-#define MSTP(_str, _parent, _reg, _bit, _flags) \
-  SH_CLK_MSTP32(_str, -1, _parent, _reg, _bit, _flags)
+#define R_CLK &r_clk
+#define P_CLK &div4_clks[DIV4_P]
+#define B_CLK &div4_clks[DIV4_B]
+#define U_CLK &div4_clks[DIV4_U]
 
 static struct clk mstp_clks[] = {
-	MSTP("uram0", &div4_clks[DIV4_U], MSTPCR0, 28, CLK_ENABLE_ON_INIT),
-	MSTP("xymem0", &div4_clks[DIV4_B], MSTPCR0, 26, CLK_ENABLE_ON_INIT),
-	MSTP("tmu0", &div4_clks[DIV4_P], MSTPCR0, 15, 0),
-	MSTP("cmt0", &r_clk, MSTPCR0, 14, 0),
-	MSTP("rwdt0", &r_clk, MSTPCR0, 13, 0),
-	MSTP("flctl0", &div4_clks[DIV4_P], MSTPCR0, 10, 0),
-	MSTP("scif0", &div4_clks[DIV4_P], MSTPCR0, 7, 0),
-	MSTP("scif1", &div4_clks[DIV4_P], MSTPCR0, 6, 0),
-	MSTP("scif2", &div4_clks[DIV4_P], MSTPCR0, 5, 0),
-
-	MSTP("i2c0", &div4_clks[DIV4_P], MSTPCR1, 9, 0),
-	MSTP("rtc0", &r_clk, MSTPCR1, 8, 0),
-
-	MSTP("sdhi0", &div4_clks[DIV4_P], MSTPCR2, 18, 0),
-	MSTP("keysc0", &r_clk, MSTPCR2, 14, 0),
-	MSTP("usbf0", &div4_clks[DIV4_P], MSTPCR2, 11, 0),
-	MSTP("2dg0", &div4_clks[DIV4_B], MSTPCR2, 9, 0),
-	MSTP("siu0", &div4_clks[DIV4_B], MSTPCR2, 8, 0),
-	MSTP("vou0", &div4_clks[DIV4_B], MSTPCR2, 5, 0),
-	MSTP("jpu0", &div4_clks[DIV4_B], MSTPCR2, 6, CLK_ENABLE_ON_INIT),
-	MSTP("beu0", &div4_clks[DIV4_B], MSTPCR2, 4, 0),
-	MSTP("ceu0", &div4_clks[DIV4_B], MSTPCR2, 3, 0),
-	MSTP("veu0", &div4_clks[DIV4_B], MSTPCR2, 2, CLK_ENABLE_ON_INIT),
-	MSTP("vpu0", &div4_clks[DIV4_B], MSTPCR2, 1, CLK_ENABLE_ON_INIT),
-	MSTP("lcdc0", &div4_clks[DIV4_B], MSTPCR2, 0, 0),
+	SH_HWBLK_CLK("uram0", -1, U_CLK, HWBLK_URAM, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("xymem0", -1, B_CLK, HWBLK_XYMEM, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("tmu0", -1, P_CLK, HWBLK_TMU, 0),
+	SH_HWBLK_CLK("cmt0", -1, R_CLK, HWBLK_CMT, 0),
+	SH_HWBLK_CLK("rwdt0", -1, R_CLK, HWBLK_RWDT, 0),
+	SH_HWBLK_CLK("flctl0", -1, P_CLK, HWBLK_FLCTL, 0),
+	SH_HWBLK_CLK("scif0", -1, P_CLK, HWBLK_SCIF0, 0),
+	SH_HWBLK_CLK("scif1", -1, P_CLK, HWBLK_SCIF1, 0),
+	SH_HWBLK_CLK("scif2", -1, P_CLK, HWBLK_SCIF2, 0),
+
+	SH_HWBLK_CLK("i2c0", -1, P_CLK, HWBLK_IIC, 0),
+	SH_HWBLK_CLK("rtc0", -1, R_CLK, HWBLK_RTC, 0),
+
+	SH_HWBLK_CLK("sdhi0", -1, P_CLK, HWBLK_SDHI, 0),
+	SH_HWBLK_CLK("keysc0", -1, R_CLK, HWBLK_KEYSC, 0),
+	SH_HWBLK_CLK("usbf0", -1, P_CLK, HWBLK_USBF, 0),
+	SH_HWBLK_CLK("2dg0", -1, B_CLK, HWBLK_2DG, 0),
+	SH_HWBLK_CLK("siu0", -1, B_CLK, HWBLK_SIU, 0),
+	SH_HWBLK_CLK("vou0", -1, B_CLK, HWBLK_VOU, 0),
+	SH_HWBLK_CLK("jpu0", -1, B_CLK, HWBLK_JPU, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("beu0", -1, B_CLK, HWBLK_BEU, 0),
+	SH_HWBLK_CLK("ceu0", -1, B_CLK, HWBLK_CEU, 0),
+	SH_HWBLK_CLK("veu0", -1, B_CLK, HWBLK_VEU, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("vpu0", -1, B_CLK, HWBLK_VPU, CLK_ENABLE_ON_INIT),
+	SH_HWBLK_CLK("lcdc0", -1, P_CLK, HWBLK_LCDC, 0),
 };
 
 int __init arch_clk_init(void)
@@ -191,7 +195,7 @@ int __init arch_clk_init(void)
 		ret = sh_clk_div6_register(div6_clks, ARRAY_SIZE(div6_clks));
 
 	if (!ret)
-		ret = sh_clk_mstp32_register(mstp_clks, ARRAY_SIZE(mstp_clks));
+		ret = sh_hwblk_clk_register(mstp_clks, ARRAY_SIZE(mstp_clks));
 
 	return ret;
 }
diff --git a/arch/sh/kernel/cpu/sh4a/hwblk-sh7722.c b/arch/sh/kernel/cpu/sh4a/hwblk-sh7722.c
new file mode 100644
index 000000000000..00a1c02d82b1
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4a/hwblk-sh7722.c
@@ -0,0 +1,106 @@
+/*
+ * arch/sh/kernel/cpu/sh4a/hwblk-sh7722.c
+ *
+ * SH7722 hardware block support
+ *
+ * Copyright (C) 2009 Magnus Damm
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <asm/suspend.h>
+#include <asm/hwblk.h>
+#include <cpu/sh7722.h>
+
+/* SH7722 registers */
+#define MSTPCR0		0xa4150030
+#define MSTPCR1		0xa4150034
+#define MSTPCR2		0xa4150038
+
+/* SH7722 Power Domains */
+enum { CORE_AREA, SUB_AREA, CORE_AREA_BM };
+static struct hwblk_area sh7722_hwblk_area[] = {
+	[CORE_AREA] = HWBLK_AREA(0, 0),
+	[CORE_AREA_BM] = HWBLK_AREA(HWBLK_AREA_FLAG_PARENT, CORE_AREA),
+	[SUB_AREA] = HWBLK_AREA(0, 0),
+};
+
+/* Table mapping HWBLK to Module Stop Bit and Power Domain */
+static struct hwblk sh7722_hwblk[HWBLK_NR] = {
+	[HWBLK_TLB] = HWBLK(MSTPCR0, 31, CORE_AREA),
+	[HWBLK_IC] = HWBLK(MSTPCR0, 30, CORE_AREA),
+	[HWBLK_OC] = HWBLK(MSTPCR0, 29, CORE_AREA),
+	[HWBLK_URAM] = HWBLK(MSTPCR0, 28, CORE_AREA),
+	[HWBLK_XYMEM] = HWBLK(MSTPCR0, 26, CORE_AREA),
+	[HWBLK_INTC] = HWBLK(MSTPCR0, 22, CORE_AREA),
+	[HWBLK_DMAC] = HWBLK(MSTPCR0, 21, CORE_AREA_BM),
+	[HWBLK_SHYWAY] = HWBLK(MSTPCR0, 20, CORE_AREA),
+	[HWBLK_HUDI] = HWBLK(MSTPCR0, 19, CORE_AREA),
+	[HWBLK_UBC] = HWBLK(MSTPCR0, 17, CORE_AREA),
+	[HWBLK_TMU] = HWBLK(MSTPCR0, 15, CORE_AREA),
+	[HWBLK_CMT] = HWBLK(MSTPCR0, 14, SUB_AREA),
+	[HWBLK_RWDT] = HWBLK(MSTPCR0, 13, SUB_AREA),
+	[HWBLK_FLCTL] = HWBLK(MSTPCR0, 10, CORE_AREA),
+	[HWBLK_SCIF0] = HWBLK(MSTPCR0, 7, CORE_AREA),
+	[HWBLK_SCIF1] = HWBLK(MSTPCR0, 6, CORE_AREA),
+	[HWBLK_SCIF2] = HWBLK(MSTPCR0, 5, CORE_AREA),
+	[HWBLK_SIO] = HWBLK(MSTPCR0, 3, CORE_AREA),
+	[HWBLK_SIOF0] = HWBLK(MSTPCR0, 2, CORE_AREA),
+	[HWBLK_SIOF1] = HWBLK(MSTPCR0, 1, CORE_AREA),
+
+	[HWBLK_IIC] = HWBLK(MSTPCR1, 9, CORE_AREA),
+	[HWBLK_RTC] = HWBLK(MSTPCR1, 8, SUB_AREA),
+
+	[HWBLK_TPU] = HWBLK(MSTPCR2, 25, CORE_AREA),
+	[HWBLK_IRDA] = HWBLK(MSTPCR2, 24, CORE_AREA),
+	[HWBLK_SDHI] = HWBLK(MSTPCR2, 18, CORE_AREA),
+	[HWBLK_SIM] = HWBLK(MSTPCR2, 16, CORE_AREA),
+	[HWBLK_KEYSC] = HWBLK(MSTPCR2, 14, SUB_AREA),
+	[HWBLK_TSIF] = HWBLK(MSTPCR2, 13, SUB_AREA),
+	[HWBLK_USBF] = HWBLK(MSTPCR2, 11, CORE_AREA),
+	[HWBLK_2DG] = HWBLK(MSTPCR2, 9, CORE_AREA_BM),
+	[HWBLK_SIU] = HWBLK(MSTPCR2, 8, CORE_AREA),
+	[HWBLK_JPU] = HWBLK(MSTPCR2, 6, CORE_AREA_BM),
+	[HWBLK_VOU] = HWBLK(MSTPCR2, 5, CORE_AREA_BM),
+	[HWBLK_BEU] = HWBLK(MSTPCR2, 4, CORE_AREA_BM),
+	[HWBLK_CEU] = HWBLK(MSTPCR2, 3, CORE_AREA_BM),
+	[HWBLK_VEU] = HWBLK(MSTPCR2, 2, CORE_AREA_BM),
+	[HWBLK_VPU] = HWBLK(MSTPCR2, 1, CORE_AREA_BM),
+	[HWBLK_LCDC] = HWBLK(MSTPCR2, 0, CORE_AREA_BM),
+};
+
+static struct hwblk_info sh7722_hwblk_info = {
+	.areas = sh7722_hwblk_area,
+	.nr_areas = ARRAY_SIZE(sh7722_hwblk_area),
+	.hwblks = sh7722_hwblk,
+	.nr_hwblks = ARRAY_SIZE(sh7722_hwblk),
+};
+
+int arch_hwblk_sleep_mode(void)
+{
+	if (!sh7722_hwblk_area[CORE_AREA].cnt)
+		return SUSP_SH_STANDBY | SUSP_SH_SF;
+
+	if (!sh7722_hwblk_area[CORE_AREA_BM].cnt)
+		return SUSP_SH_SLEEP | SUSP_SH_SF;
+
+	return SUSP_SH_SLEEP;
+}
+
+int __init arch_hwblk_init(void)
+{
+	return hwblk_register(&sh7722_hwblk_info);
+}
diff --git a/arch/sh/kernel/cpu/shmobile/Makefile b/arch/sh/kernel/cpu/shmobile/Makefile
index 08bfa7c7db29..e8a5111e848a 100644
--- a/arch/sh/kernel/cpu/shmobile/Makefile
+++ b/arch/sh/kernel/cpu/shmobile/Makefile
@@ -4,3 +4,4 @@
 
 # Power Management & Sleep mode
 obj-$(CONFIG_PM)	+= pm.o sleep.o
+obj-$(CONFIG_CPU_IDLE)	+= cpuidle.o
diff --git a/arch/sh/kernel/cpu/shmobile/cpuidle.c b/arch/sh/kernel/cpu/shmobile/cpuidle.c
new file mode 100644
index 000000000000..4afdd975cc66
--- /dev/null
+++ b/arch/sh/kernel/cpu/shmobile/cpuidle.c
@@ -0,0 +1,102 @@
+/*
+ * arch/sh/kernel/cpu/shmobile/cpuidle.c
+ *
+ * Cpuidle support code for SuperH Mobile
+ *
+ *  Copyright (C) 2009 Magnus Damm
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <linux/suspend.h>
+#include <linux/cpuidle.h>
+#include <asm/suspend.h>
+#include <asm/uaccess.h>
+#include <asm/hwblk.h>
+
+static unsigned long cpuidle_mode[] = {
+	SUSP_SH_SLEEP, /* regular sleep mode */
+	SUSP_SH_SLEEP | SUSP_SH_SF, /* sleep mode + self refresh */
+};
+
+static int cpuidle_sleep_enter(struct cpuidle_device *dev,
+			       struct cpuidle_state *state)
+{
+	unsigned long allowed_mode = arch_hwblk_sleep_mode();
+	ktime_t before, after;
+	int requested_state = state - &dev->states[0];
+	int allowed_state;
+	int k;
+
+	/* convert allowed mode to allowed state */
+	for (k = ARRAY_SIZE(cpuidle_mode) - 1; k > 0; k--)
+		if (cpuidle_mode[k] == allowed_mode)
+			break;
+
+	allowed_state = k;
+
+	/* take the following into account for sleep mode selection:
+	 * - allowed_state: best mode allowed by hardware (clock deps)
+	 * - requested_state: best mode allowed by software (latencies)
+	 */
+	k = min_t(int, allowed_state, requested_state);
+
+	dev->last_state = &dev->states[k];
+	before = ktime_get();
+	sh_mobile_call_standby(cpuidle_mode[k]);
+	after = ktime_get();
+	return ktime_to_ns(ktime_sub(after, before)) >> 10;
+}
+
+static struct cpuidle_device cpuidle_dev;
+static struct cpuidle_driver cpuidle_driver = {
+	.name =		"sh_idle",
+	.owner =	THIS_MODULE,
+};
+
+void sh_mobile_setup_cpuidle(void)
+{
+	struct cpuidle_device *dev = &cpuidle_dev;
+	struct cpuidle_state *state;
+	int i;
+
+	cpuidle_register_driver(&cpuidle_driver);
+
+	for (i = 0; i < CPUIDLE_STATE_MAX; i++) {
+		dev->states[i].name[0] = '\0';
+		dev->states[i].desc[0] = '\0';
+	}
+
+	i = CPUIDLE_DRIVER_STATE_START;
+
+	state = &dev->states[i++];
+	snprintf(state->name, CPUIDLE_NAME_LEN, "C0");
+	strncpy(state->desc, "SuperH Sleep Mode", CPUIDLE_DESC_LEN);
+	state->exit_latency = 1;
+	state->target_residency = 1 * 2;
+	state->power_usage = 3;
+	state->flags = 0;
+	state->flags |= CPUIDLE_FLAG_SHALLOW;
+	state->flags |= CPUIDLE_FLAG_TIME_VALID;
+	state->enter = cpuidle_sleep_enter;
+
+	dev->safe_state = state;
+
+	state = &dev->states[i++];
+	snprintf(state->name, CPUIDLE_NAME_LEN, "C1");
+	strncpy(state->desc, "SuperH Sleep Mode [SF]", CPUIDLE_DESC_LEN);
+	state->exit_latency = 100;
+	state->target_residency = 1 * 2;
+	state->power_usage = 1;
+	state->flags = 0;
+	state->flags |= CPUIDLE_FLAG_TIME_VALID;
+	state->enter = cpuidle_sleep_enter;
+
+	dev->state_count = i;
+
+	cpuidle_register_device(dev);
+}
diff --git a/arch/sh/kernel/cpu/shmobile/pm.c b/arch/sh/kernel/cpu/shmobile/pm.c
index 8c067adf6830..de078d24ce56 100644
--- a/arch/sh/kernel/cpu/shmobile/pm.c
+++ b/arch/sh/kernel/cpu/shmobile/pm.c
@@ -1,5 +1,5 @@
 /*
- * arch/sh/kernel/cpu/sh4a/pm-sh_mobile.c
+ * arch/sh/kernel/cpu/shmobile/pm.c
  *
  * Power management support code for SuperH Mobile
  *
@@ -32,20 +32,17 @@
  *
  * R-standby mode is unsupported, but will be added in the future
  * U-standby mode is low priority since it needs bootloader hacks
- *
- * All modes should be tied in with cpuidle. But before that can
- * happen we need to keep track of enabled hardware blocks so we
- * can avoid entering sleep modes that stop clocks to hardware
- * blocks that are in use even though the cpu core is idle.
  */
 
+#define ILRAM_BASE 0xe5200000
+
 extern const unsigned char sh_mobile_standby[];
 extern const unsigned int sh_mobile_standby_size;
 
-static void sh_mobile_call_standby(unsigned long mode)
+void sh_mobile_call_standby(unsigned long mode)
 {
 	extern void *vbr_base;
-	void *onchip_mem = (void *)0xe5200000; /* ILRAM */
+	void *onchip_mem = (void *)ILRAM_BASE;
 	void (*standby_onchip_mem)(unsigned long) = onchip_mem;
 
 	/* Note: Wake up from sleep may generate exceptions!
@@ -55,11 +52,6 @@ static void sh_mobile_call_standby(unsigned long mode)
 	if (mode & SUSP_SH_SF)
 		asm volatile("ldc %0, vbr" : : "r" (onchip_mem) : "memory");
 
-	/* Copy the assembly snippet to the otherwise ununsed ILRAM */
-	memcpy(onchip_mem, sh_mobile_standby, sh_mobile_standby_size);
-	wmb();
-	ctrl_barrier();
-
 	/* Let assembly snippet in on-chip memory handle the rest */
 	standby_onchip_mem(mode);
 
@@ -85,7 +77,15 @@ static struct platform_suspend_ops sh_pm_ops = {
 
 static int __init sh_pm_init(void)
 {
+	void *onchip_mem = (void *)ILRAM_BASE;
+
+	/* Copy the assembly snippet to the otherwise ununsed ILRAM */
+	memcpy(onchip_mem, sh_mobile_standby, sh_mobile_standby_size);
+	wmb();
+	ctrl_barrier();
+
 	suspend_set_ops(&sh_pm_ops);
+	sh_mobile_setup_cpuidle();
 	return 0;
 }
 
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index dd38338553ef..ceb409bf7741 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -30,6 +30,7 @@
 #include <linux/clk.h>
 #include <linux/delay.h>
 #include <linux/platform_device.h>
+#include <linux/lmb.h>
 #include <asm/uaccess.h>
 #include <asm/io.h>
 #include <asm/page.h>
@@ -233,39 +234,45 @@ void __init __add_active_range(unsigned int nid, unsigned long start_pfn,
 void __init setup_bootmem_allocator(unsigned long free_pfn)
 {
 	unsigned long bootmap_size;
+	unsigned long bootmap_pages, bootmem_paddr;
+	u64 total_pages = (lmb_end_of_DRAM() - __MEMORY_START) >> PAGE_SHIFT;
+	int i;
+
+	bootmap_pages = bootmem_bootmap_pages(total_pages);
+
+	bootmem_paddr = lmb_alloc(bootmap_pages << PAGE_SHIFT, PAGE_SIZE);
 
 	/*
 	 * Find a proper area for the bootmem bitmap. After this
 	 * bootstrap step all allocations (until the page allocator
 	 * is intact) must be done via bootmem_alloc().
 	 */
-	bootmap_size = init_bootmem_node(NODE_DATA(0), free_pfn,
+	bootmap_size = init_bootmem_node(NODE_DATA(0),
+					 bootmem_paddr >> PAGE_SHIFT,
 					 min_low_pfn, max_low_pfn);
 
-	__add_active_range(0, min_low_pfn, max_low_pfn);
-	register_bootmem_low_pages();
-
-	node_set_online(0);
+	/* Add active regions with valid PFNs. */
+	for (i = 0; i < lmb.memory.cnt; i++) {
+		unsigned long start_pfn, end_pfn;
+		start_pfn = lmb.memory.region[i].base >> PAGE_SHIFT;
+		end_pfn = start_pfn + lmb_size_pages(&lmb.memory, i);
+		__add_active_range(0, start_pfn, end_pfn);
+	}
 
 	/*
-	 * Reserve the kernel text and
-	 * Reserve the bootmem bitmap. We do this in two steps (first step
-	 * was init_bootmem()), because this catches the (definitely buggy)
-	 * case of us accidentally initializing the bootmem allocator with
-	 * an invalid RAM area.
+	 * Add all physical memory to the bootmem map and mark each
+	 * area as present.
 	 */
-	reserve_bootmem(__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET,
-			(PFN_PHYS(free_pfn) + bootmap_size + PAGE_SIZE - 1) -
-			(__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET),
-			BOOTMEM_DEFAULT);
+	register_bootmem_low_pages();
 
-	/*
-	 * Reserve physical pages below CONFIG_ZERO_PAGE_OFFSET.
-	 */
-	if (CONFIG_ZERO_PAGE_OFFSET != 0)
-		reserve_bootmem(__MEMORY_START, CONFIG_ZERO_PAGE_OFFSET,
+	/* Reserve the sections we're already using. */
+	for (i = 0; i < lmb.reserved.cnt; i++)
+		reserve_bootmem(lmb.reserved.region[i].base,
+				lmb_size_bytes(&lmb.reserved, i),
 				BOOTMEM_DEFAULT);
 
+	node_set_online(0);
+
 	sparse_memory_present_with_active_regions(0);
 
 #ifdef CONFIG_BLK_DEV_INITRD
@@ -296,12 +303,37 @@ void __init setup_bootmem_allocator(unsigned long free_pfn)
 static void __init setup_memory(void)
 {
 	unsigned long start_pfn;
+	u64 base = min_low_pfn << PAGE_SHIFT;
+	u64 size = (max_low_pfn << PAGE_SHIFT) - base;
 
 	/*
 	 * Partially used pages are not usable - thus
 	 * we are rounding upwards:
 	 */
 	start_pfn = PFN_UP(__pa(_end));
+
+	lmb_add(base, size);
+
+	/*
+	 * Reserve the kernel text and
+	 * Reserve the bootmem bitmap. We do this in two steps (first step
+	 * was init_bootmem()), because this catches the (definitely buggy)
+	 * case of us accidentally initializing the bootmem allocator with
+	 * an invalid RAM area.
+	 */
+	lmb_reserve(__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET,
+		    (PFN_PHYS(start_pfn) + PAGE_SIZE - 1) -
+		    (__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET));
+
+	/*
+	 * Reserve physical pages below CONFIG_ZERO_PAGE_OFFSET.
+	 */
+	if (CONFIG_ZERO_PAGE_OFFSET != 0)
+		lmb_reserve(__MEMORY_START, CONFIG_ZERO_PAGE_OFFSET);
+
+	lmb_analyze();
+	lmb_dump_all();
+
 	setup_bootmem_allocator(start_pfn);
 }
 #else
@@ -402,6 +434,7 @@ void __init setup_arch(char **cmdline_p)
 	nodes_clear(node_online_map);
 
 	/* Setup bootmem with available RAM */
+	lmb_init();
 	setup_memory();
 	sparse_init();
 
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index 9b352a1e3fb4..d2424b068b7b 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -21,6 +21,7 @@
 #include <linux/smp.h>
 #include <linux/rtc.h>
 #include <asm/clock.h>
+#include <asm/hwblk.h>
 #include <asm/rtc.h>
 
 /* Dummy RTC ops */
@@ -96,6 +97,7 @@ void __init time_init(void)
 	if (board_time_init)
 		board_time_init();
 
+	hwblk_init();
 	clk_init();
 
 	rtc_sh_get_time(&xtime);
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index f53c76acaede..674ed8feb8ae 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -50,12 +50,7 @@ SECTIONS
 		_etext = .;		/* End of text section */
 	} = 0x0009
 
-	. = ALIGN(16);		/* Exception table */
-	__ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) {
-		__start___ex_table = .;
-		*(__ex_table)
-		__stop___ex_table = .;
-	}
+	EXCEPTION_TABLE(16)
 
 	NOTES
 	RO_DATA(PAGE_SIZE)
@@ -71,69 +66,14 @@ SECTIONS
 		__uncached_end = .;
 	}
 
-	. = ALIGN(THREAD_SIZE);
-	.data : AT(ADDR(.data) - LOAD_OFFSET) {		/* Data */
-		*(.data.init_task)
-
-		. = ALIGN(L1_CACHE_BYTES);
-		*(.data.cacheline_aligned)
-
-		. = ALIGN(L1_CACHE_BYTES);
-		*(.data.read_mostly)
-
-		. = ALIGN(PAGE_SIZE);
-		*(.data.page_aligned)
-
-		__nosave_begin = .;
-		*(.data.nosave)
-		. = ALIGN(PAGE_SIZE);
-		__nosave_end = .;
-
-		DATA_DATA
-		CONSTRUCTORS
-	}
+	RW_DATA_SECTION(L1_CACHE_BYTES, PAGE_SIZE, THREAD_SIZE)
 
 	_edata = .;			/* End of data section */
 
 	. = ALIGN(PAGE_SIZE);		/* Init code and data */
-	.init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
-		__init_begin = .;
-		_sinittext = .;
-		INIT_TEXT
-		_einittext = .;
-	}
-
-	.init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { INIT_DATA }
-
-	. = ALIGN(16);
-	.init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) {
-		__setup_start = .;
-		*(.init.setup)
-		__setup_end = .;
-	}
-
-	.initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) {
-		__initcall_start = .;
-		INITCALLS
-		__initcall_end = .;
-	}
-
-	.con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) {
-		__con_initcall_start = .;
-		*(.con_initcall.init)
-		__con_initcall_end = .;
-	}
-
-	SECURITY_INIT
-
-#ifdef CONFIG_BLK_DEV_INITRD
-	. = ALIGN(PAGE_SIZE);
-	.init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) {
-		__initramfs_start = .;
-		*(.init.ramfs)
-		__initramfs_end = .;
-	}
-#endif
+	__init_begin = .;
+	INIT_TEXT_SECTION(PAGE_SIZE)
+	INIT_DATA_SECTION(16)
 
 	. = ALIGN(4);
 	.machvec.init : AT(ADDR(.machvec.init) - LOAD_OFFSET) {
@@ -152,16 +92,11 @@ SECTIONS
 	.exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { EXIT_DATA }
 
 	. = ALIGN(PAGE_SIZE);
-	.bss : AT(ADDR(.bss) - LOAD_OFFSET) {
-		__init_end = .;
-		__bss_start = .;		/* BSS */
-		*(.bss.page_aligned)
-		*(.bss)
-		*(COMMON)
-		. = ALIGN(4);
-		_ebss = .;			/* uClinux MTD sucks */
-		_end = . ;
-	}
+	__init_end = .;
+	BSS(PAGE_SIZE)
+	. = ALIGN(4);
+	_ebss = .;			/* uClinux MTD sucks */
+	_end = . ;
 
 	/*
 	 * When something in the kernel is NOT compiled as a module, the
@@ -170,7 +105,7 @@ SECTIONS
 	 * it's a module.
 	 */
 	/DISCARD/ : {
-		*(.exitcall.exit)
+		EXIT_CALL
 	}
 
 	STABS_DEBUG
diff --git a/arch/sh/mm/fault_32.c b/arch/sh/mm/fault_32.c
index 71925946f1e1..08d0117d90fa 100644
--- a/arch/sh/mm/fault_32.c
+++ b/arch/sh/mm/fault_32.c
@@ -2,7 +2,7 @@
  * Page fault handler for SH with an MMU.
  *
  *  Copyright (C) 1999  Niibe Yutaka
- *  Copyright (C) 2003 - 2008  Paul Mundt
+ *  Copyright (C) 2003 - 2009  Paul Mundt
  *
  *  Based on linux/arch/i386/mm/fault.c:
  *   Copyright (C) 1995  Linus Torvalds
@@ -25,18 +25,84 @@ static inline int notify_page_fault(struct pt_regs *regs, int trap)
 {
 	int ret = 0;
 
-#ifdef CONFIG_KPROBES
-	if (!user_mode(regs)) {
+	if (kprobes_built_in() && !user_mode(regs)) {
 		preempt_disable();
 		if (kprobe_running() && kprobe_fault_handler(regs, trap))
 			ret = 1;
 		preempt_enable();
 	}
-#endif
 
 	return ret;
 }
 
+static inline pmd_t *vmalloc_sync_one(pgd_t *pgd, unsigned long address)
+{
+	unsigned index = pgd_index(address);
+	pgd_t *pgd_k;
+	pud_t *pud, *pud_k;
+	pmd_t *pmd, *pmd_k;
+
+	pgd += index;
+	pgd_k = init_mm.pgd + index;
+
+	if (!pgd_present(*pgd_k))
+		return NULL;
+
+	pud = pud_offset(pgd, address);
+	pud_k = pud_offset(pgd_k, address);
+	if (!pud_present(*pud_k))
+		return NULL;
+
+	pmd = pmd_offset(pud, address);
+	pmd_k = pmd_offset(pud_k, address);
+	if (!pmd_present(*pmd_k))
+		return NULL;
+
+	if (!pmd_present(*pmd))
+		set_pmd(pmd, *pmd_k);
+	else
+		BUG_ON(pmd_page(*pmd) != pmd_page(*pmd_k));
+
+	return pmd_k;
+}
+
+/*
+ * Handle a fault on the vmalloc or module mapping area
+ */
+static noinline int vmalloc_fault(unsigned long address)
+{
+	pgd_t *pgd_k;
+	pmd_t *pmd_k;
+	pte_t *pte_k;
+
+	/* Make sure we are in vmalloc area: */
+	if (!(address >= VMALLOC_START && address < VMALLOC_END))
+		return -1;
+
+	/*
+	 * Synchronize this task's top level page-table
+	 * with the 'reference' page table.
+	 *
+	 * Do _not_ use "current" here. We might be inside
+	 * an interrupt in the middle of a task switch..
+	 */
+	pgd_k = get_TTB();
+	pmd_k = vmalloc_sync_one(__va((unsigned long)pgd_k), address);
+	if (!pmd_k)
+		return -1;
+
+	pte_k = pte_offset_kernel(pmd_k, address);
+	if (!pte_present(*pte_k))
+		return -1;
+
+	return 0;
+}
+
+static int fault_in_kernel_space(unsigned long address)
+{
+	return address >= TASK_SIZE;
+}
+
 /*
  * This routine handles page faults.  It determines the address,
  * and the problem, and then passes it off to one of the appropriate
@@ -46,6 +112,7 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 					unsigned long writeaccess,
 					unsigned long address)
 {
+	unsigned long vec;
 	struct task_struct *tsk;
 	struct mm_struct *mm;
 	struct vm_area_struct * vma;
@@ -53,59 +120,30 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 	int fault;
 	siginfo_t info;
 
-	/*
-	 * We don't bother with any notifier callbacks here, as they are
-	 * all handled through the __do_page_fault() fast-path.
-	 */
-
 	tsk = current;
+	mm = tsk->mm;
 	si_code = SEGV_MAPERR;
+	vec = lookup_exception_vector();
 
-	if (unlikely(address >= TASK_SIZE)) {
-		/*
-		 * Synchronize this task's top level page-table
-		 * with the 'reference' page table.
-		 *
-		 * Do _not_ use "tsk" here. We might be inside
-		 * an interrupt in the middle of a task switch..
-		 */
-		int offset = pgd_index(address);
-		pgd_t *pgd, *pgd_k;
-		pud_t *pud, *pud_k;
-		pmd_t *pmd, *pmd_k;
-
-		pgd = get_TTB() + offset;
-		pgd_k = swapper_pg_dir + offset;
-
-		if (!pgd_present(*pgd)) {
-			if (!pgd_present(*pgd_k))
-				goto bad_area_nosemaphore;
-			set_pgd(pgd, *pgd_k);
+	/*
+	 * We fault-in kernel-space virtual memory on-demand. The
+	 * 'reference' page table is init_mm.pgd.
+	 *
+	 * NOTE! We MUST NOT take any locks for this case. We may
+	 * be in an interrupt or a critical region, and should
+	 * only copy the information from the master page table,
+	 * nothing more.
+	 */
+	if (unlikely(fault_in_kernel_space(address))) {
+		if (vmalloc_fault(address) >= 0)
 			return;
-		}
-
-		pud = pud_offset(pgd, address);
-		pud_k = pud_offset(pgd_k, address);
-
-		if (!pud_present(*pud)) {
-			if (!pud_present(*pud_k))
-				goto bad_area_nosemaphore;
-			set_pud(pud, *pud_k);
+		if (notify_page_fault(regs, vec))
 			return;
-		}
-
-		pmd = pmd_offset(pud, address);
-		pmd_k = pmd_offset(pud_k, address);
-		if (pmd_present(*pmd) || !pmd_present(*pmd_k))
-			goto bad_area_nosemaphore;
-		set_pmd(pmd, *pmd_k);
 
-		return;
+		goto bad_area_nosemaphore;
 	}
 
-	mm = tsk->mm;
-
-	if (unlikely(notify_page_fault(regs, lookup_exception_vector())))
+	if (unlikely(notify_page_fault(regs, vec)))
 		return;
 
 	/* Only enable interrupts if they were on before the fault */
@@ -115,8 +153,8 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 	perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, address);
 
 	/*
-	 * If we're in an interrupt or have no user
-	 * context, we must not take the fault..
+	 * If we're in an interrupt, have no user context or are running
+	 * in an atomic region then we must not take the fault:
 	 */
 	if (in_atomic() || !mm)
 		goto no_context;
@@ -132,10 +170,11 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 		goto bad_area;
 	if (expand_stack(vma, address))
 		goto bad_area;
-/*
- * Ok, we have a good vm_area for this memory access, so
- * we can handle it..
- */
+
+	/*
+	 * Ok, we have a good vm_area for this memory access, so
+	 * we can handle it..
+	 */
 good_area:
 	si_code = SEGV_ACCERR;
 	if (writeaccess) {
@@ -173,10 +212,10 @@ survive:
 	up_read(&mm->mmap_sem);
 	return;
 
-/*
- * Something tried to access memory that isn't in our memory map..
- * Fix it, but check if it's kernel or user first..
- */
+	/*
+	 * Something tried to access memory that isn't in our memory map..
+	 * Fix it, but check if it's kernel or user first..
+	 */
 bad_area:
 	up_read(&mm->mmap_sem);
 
diff --git a/arch/sh/mm/numa.c b/arch/sh/mm/numa.c
index 095d93bec7cd..9b784fdb947c 100644
--- a/arch/sh/mm/numa.c
+++ b/arch/sh/mm/numa.c
@@ -9,6 +9,7 @@
  */
 #include <linux/module.h>
 #include <linux/bootmem.h>
+#include <linux/lmb.h>
 #include <linux/mm.h>
 #include <linux/numa.h>
 #include <linux/pfn.h>
@@ -26,6 +27,15 @@ EXPORT_SYMBOL_GPL(node_data);
 void __init setup_memory(void)
 {
 	unsigned long free_pfn = PFN_UP(__pa(_end));
+	u64 base = min_low_pfn << PAGE_SHIFT;
+	u64 size = (max_low_pfn << PAGE_SHIFT) - min_low_pfn;
+
+	lmb_add(base, size);
+
+	/* Reserve the LMB regions used by the kernel, initrd, etc.. */
+	lmb_reserve(__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET,
+		    (PFN_PHYS(free_pfn) + PAGE_SIZE - 1) -
+		    (__MEMORY_START + CONFIG_ZERO_PAGE_OFFSET));
 
 	/*
 	 * Node 0 sets up its pgdat at the first available pfn,
@@ -45,24 +55,23 @@ void __init setup_memory(void)
 
 void __init setup_bootmem_node(int nid, unsigned long start, unsigned long end)
 {
-	unsigned long bootmap_pages, bootmap_start, bootmap_size;
-	unsigned long start_pfn, free_pfn, end_pfn;
+	unsigned long bootmap_pages;
+	unsigned long start_pfn, end_pfn;
+	unsigned long bootmem_paddr;
 
 	/* Don't allow bogus node assignment */
 	BUG_ON(nid > MAX_NUMNODES || nid == 0);
 
-	/*
-	 * The free pfn starts at the beginning of the range, and is
-	 * advanced as necessary for pgdat and node map allocations.
-	 */
-	free_pfn = start_pfn = start >> PAGE_SHIFT;
+	start_pfn = start >> PAGE_SHIFT;
 	end_pfn = end >> PAGE_SHIFT;
 
+	lmb_add(start, end - start);
+
 	__add_active_range(nid, start_pfn, end_pfn);
 
 	/* Node-local pgdat */
-	NODE_DATA(nid) = pfn_to_kaddr(free_pfn);
-	free_pfn += PFN_UP(sizeof(struct pglist_data));
+	NODE_DATA(nid) = __va(lmb_alloc_base(sizeof(struct pglist_data),
+					     SMP_CACHE_BYTES, end_pfn));
 	memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
 
 	NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
@@ -71,16 +80,17 @@ void __init setup_bootmem_node(int nid, unsigned long start, unsigned long end)
 
 	/* Node-local bootmap */
 	bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn);
-	bootmap_start = (unsigned long)pfn_to_kaddr(free_pfn);
-	bootmap_size = init_bootmem_node(NODE_DATA(nid), free_pfn, start_pfn,
-				    end_pfn);
+	bootmem_paddr = lmb_alloc_base(bootmap_pages << PAGE_SHIFT,
+				       PAGE_SIZE, end_pfn);
+	init_bootmem_node(NODE_DATA(nid), bootmem_paddr >> PAGE_SHIFT,
+			  start_pfn, end_pfn);
 
 	free_bootmem_with_active_regions(nid, end_pfn);
 
 	/* Reserve the pgdat and bootmap space with the bootmem allocator */
 	reserve_bootmem_node(NODE_DATA(nid), start_pfn << PAGE_SHIFT,
 			     sizeof(struct pglist_data), BOOTMEM_DEFAULT);
-	reserve_bootmem_node(NODE_DATA(nid), free_pfn << PAGE_SHIFT,
+	reserve_bootmem_node(NODE_DATA(nid), bootmem_paddr,
 			     bootmap_pages << PAGE_SHIFT, BOOTMEM_DEFAULT);
 
 	/* It's up */