From feeeaa87e8e6702f57ed3be7904ffd87cc044b82 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 13 Apr 2006 02:42:42 +0200
Subject: [PATCH] softmac: fix event sending

Softmac is sending custom events to userspace already, but it
should _really_ be sending the right WEXT events instead. This
patch fixes that.

Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211softmac.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h
index b971d8c82bdd..6b3693f05ca0 100644
--- a/include/net/ieee80211softmac.h
+++ b/include/net/ieee80211softmac.h
@@ -267,8 +267,9 @@ extern void ieee80211softmac_stop(struct net_device *dev);
 #define IEEE80211SOFTMAC_EVENT_AUTH_FAILED		5
 #define IEEE80211SOFTMAC_EVENT_AUTH_TIMEOUT		6
 #define IEEE80211SOFTMAC_EVENT_ASSOCIATE_NET_NOT_FOUND	7
+#define IEEE80211SOFTMAC_EVENT_DISASSOCIATED		8
 /* keep this updated! */
-#define IEEE80211SOFTMAC_EVENT_LAST			7
+#define IEEE80211SOFTMAC_EVENT_LAST			8
 /*
  * If you want to be notified of certain events, you can call
  * ieee80211softmac_notify[_atomic] with
-- 
cgit v1.2.3


From dc6de33674608f978ec29f5c2f7e3af458c06f78 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Thu, 20 Apr 2006 00:10:50 -0700
Subject: [NET]: Add skb->truesize assertion checking.

Add some sanity checking.  truesize should be at least sizeof(struct
sk_buff) plus the current packet length.  If not, then truesize is
seriously mangled and deserves a kernel log message.

Currently we'll do the check for release of stream socket buffers.

But we can add checks to more spots over time.

Incorporating ideas from Herbert Xu.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 7 +++++++
 include/net/sock.h     | 1 +
 net/core/skbuff.c      | 8 ++++++++
 net/core/stream.c      | 1 +
 4 files changed, 17 insertions(+)

(limited to 'include')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index c4619a428d9b..f8f234708b98 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -344,6 +344,13 @@ extern void	      skb_over_panic(struct sk_buff *skb, int len,
 				     void *here);
 extern void	      skb_under_panic(struct sk_buff *skb, int len,
 				      void *here);
+extern void	      skb_truesize_bug(struct sk_buff *skb);
+
+static inline void skb_truesize_check(struct sk_buff *skb)
+{
+	if (unlikely((int)skb->truesize < sizeof(struct sk_buff) + skb->len))
+		skb_truesize_bug(skb);
+}
 
 extern int skb_append_datato_frags(struct sock *sk, struct sk_buff *skb,
 			int getfrag(void *from, char *to, int offset,
diff --git a/include/net/sock.h b/include/net/sock.h
index af2b0544586e..ff8b0dad7b0f 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -454,6 +454,7 @@ static inline void sk_stream_set_owner_r(struct sk_buff *skb, struct sock *sk)
 
 static inline void sk_stream_free_skb(struct sock *sk, struct sk_buff *skb)
 {
+	skb_truesize_check(skb);
 	sock_set_flag(sk, SOCK_QUEUE_SHRUNK);
 	sk->sk_wmem_queued   -= skb->truesize;
 	sk->sk_forward_alloc += skb->truesize;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 09464fa8d72f..fb3770f9c094 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -112,6 +112,14 @@ void skb_under_panic(struct sk_buff *skb, int sz, void *here)
 	BUG();
 }
 
+void skb_truesize_bug(struct sk_buff *skb)
+{
+	printk(KERN_ERR "SKB BUG: Invalid truesize (%u) "
+	       "len=%u, sizeof(sk_buff)=%Zd\n",
+	       skb->truesize, skb->len, sizeof(struct sk_buff));
+}
+EXPORT_SYMBOL(skb_truesize_bug);
+
 /* 	Allocate a new skbuff. We do this ourselves so we can fill in a few
  *	'private' fields and also do memory statistics to find all the
  *	[BEEP] leaks.
diff --git a/net/core/stream.c b/net/core/stream.c
index 35e25259fd95..e9489696f694 100644
--- a/net/core/stream.c
+++ b/net/core/stream.c
@@ -176,6 +176,7 @@ void sk_stream_rfree(struct sk_buff *skb)
 {
 	struct sock *sk = skb->sk;
 
+	skb_truesize_check(skb);
 	atomic_sub(skb->truesize, &sk->sk_rmem_alloc);
 	sk->sk_forward_alloc += skb->truesize;
 }
-- 
cgit v1.2.3


From 7daa411b810d7eadfaabe3765ec5f827893dbb30 Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Wed, 12 Apr 2006 21:05:59 -0500
Subject: [PATCH] powerpc: IOMMU support for honoring dma_mask

Some devices don't support full 32-bit DMA address space, which we currently
assume. Add the required mask-passing to the IOMMU allocators.

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/iommu.c     | 36 ++++++++++++++++++++++++++----------
 arch/powerpc/kernel/pci_iommu.c | 40 ++++++++++++++++++++++++++++++++++++----
 arch/powerpc/kernel/vio.c       |  6 +++---
 include/asm-powerpc/iommu.h     |  7 ++++---
 4 files changed, 69 insertions(+), 20 deletions(-)

(limited to 'include')

diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
index d9a7fdef59b9..4eba60a32890 100644
--- a/arch/powerpc/kernel/iommu.c
+++ b/arch/powerpc/kernel/iommu.c
@@ -61,6 +61,7 @@ __setup("iommu=", setup_iommu);
 static unsigned long iommu_range_alloc(struct iommu_table *tbl,
                                        unsigned long npages,
                                        unsigned long *handle,
+                                       unsigned long mask,
                                        unsigned int align_order)
 { 
 	unsigned long n, end, i, start;
@@ -97,9 +98,21 @@ static unsigned long iommu_range_alloc(struct iommu_table *tbl,
 	 */
 	if (start >= limit)
 		start = largealloc ? tbl->it_largehint : tbl->it_hint;
-	
+
  again:
 
+	if (limit + tbl->it_offset > mask) {
+		limit = mask - tbl->it_offset + 1;
+		/* If we're constrained on address range, first try
+		 * at the masked hint to avoid O(n) search complexity,
+		 * but on second pass, start at 0.
+		 */
+		if ((start & mask) >= limit || pass > 0)
+			start = 0;
+		else
+			start &= mask;
+	}
+
 	n = find_next_zero_bit(tbl->it_map, limit, start);
 
 	/* Align allocation */
@@ -150,14 +163,14 @@ static unsigned long iommu_range_alloc(struct iommu_table *tbl,
 
 static dma_addr_t iommu_alloc(struct iommu_table *tbl, void *page,
 		       unsigned int npages, enum dma_data_direction direction,
-		       unsigned int align_order)
+		       unsigned long mask, unsigned int align_order)
 {
 	unsigned long entry, flags;
 	dma_addr_t ret = DMA_ERROR_CODE;
-	
+
 	spin_lock_irqsave(&(tbl->it_lock), flags);
 
-	entry = iommu_range_alloc(tbl, npages, NULL, align_order);
+	entry = iommu_range_alloc(tbl, npages, NULL, mask, align_order);
 
 	if (unlikely(entry == DMA_ERROR_CODE)) {
 		spin_unlock_irqrestore(&(tbl->it_lock), flags);
@@ -236,7 +249,7 @@ static void iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr,
 
 int iommu_map_sg(struct device *dev, struct iommu_table *tbl,
 		struct scatterlist *sglist, int nelems,
-		enum dma_data_direction direction)
+		unsigned long mask, enum dma_data_direction direction)
 {
 	dma_addr_t dma_next = 0, dma_addr;
 	unsigned long flags;
@@ -274,7 +287,7 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl,
 		vaddr = (unsigned long)page_address(s->page) + s->offset;
 		npages = PAGE_ALIGN(vaddr + slen) - (vaddr & PAGE_MASK);
 		npages >>= PAGE_SHIFT;
-		entry = iommu_range_alloc(tbl, npages, &handle, 0);
+		entry = iommu_range_alloc(tbl, npages, &handle, mask >> PAGE_SHIFT, 0);
 
 		DBG("  - vaddr: %lx, size: %lx\n", vaddr, slen);
 
@@ -479,7 +492,8 @@ void iommu_free_table(struct device_node *dn)
  * byte within the page as vaddr.
  */
 dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr,
-		size_t size, enum dma_data_direction direction)
+		size_t size, unsigned long mask,
+		enum dma_data_direction direction)
 {
 	dma_addr_t dma_handle = DMA_ERROR_CODE;
 	unsigned long uaddr;
@@ -492,7 +506,8 @@ dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr,
 	npages >>= PAGE_SHIFT;
 
 	if (tbl) {
-		dma_handle = iommu_alloc(tbl, vaddr, npages, direction, 0);
+		dma_handle = iommu_alloc(tbl, vaddr, npages, direction,
+					 mask >> PAGE_SHIFT, 0);
 		if (dma_handle == DMA_ERROR_CODE) {
 			if (printk_ratelimit())  {
 				printk(KERN_INFO "iommu_alloc failed, "
@@ -521,7 +536,7 @@ void iommu_unmap_single(struct iommu_table *tbl, dma_addr_t dma_handle,
  * to the dma address (mapping) of the first page.
  */
 void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size,
-		dma_addr_t *dma_handle, gfp_t flag)
+		dma_addr_t *dma_handle, unsigned long mask, gfp_t flag)
 {
 	void *ret = NULL;
 	dma_addr_t mapping;
@@ -551,7 +566,8 @@ void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size,
 	memset(ret, 0, size);
 
 	/* Set up tces to cover the allocated range */
-	mapping = iommu_alloc(tbl, ret, npages, DMA_BIDIRECTIONAL, order);
+	mapping = iommu_alloc(tbl, ret, npages, DMA_BIDIRECTIONAL,
+			      mask >> PAGE_SHIFT, order);
 	if (mapping == DMA_ERROR_CODE) {
 		free_pages((unsigned long)ret, order);
 		ret = NULL;
diff --git a/arch/powerpc/kernel/pci_iommu.c b/arch/powerpc/kernel/pci_iommu.c
index c336f3e31cff..c1d95e14bbed 100644
--- a/arch/powerpc/kernel/pci_iommu.c
+++ b/arch/powerpc/kernel/pci_iommu.c
@@ -59,6 +59,25 @@ static inline struct iommu_table *devnode_table(struct device *dev)
 }
 
 
+static inline unsigned long device_to_mask(struct device *hwdev)
+{
+	struct pci_dev *pdev;
+
+	if (!hwdev) {
+		pdev = ppc64_isabridge_dev;
+		if (!pdev) /* This is the best guess we can do */
+			return 0xfffffffful;
+	} else
+		pdev = to_pci_dev(hwdev);
+
+	if (pdev->dma_mask)
+		return pdev->dma_mask;
+
+	/* Assume devices without mask can take 32 bit addresses */
+	return 0xfffffffful;
+}
+
+
 /* Allocates a contiguous real buffer and creates mappings over it.
  * Returns the virtual address of the buffer and sets dma_handle
  * to the dma address (mapping) of the first page.
@@ -67,7 +86,7 @@ static void *pci_iommu_alloc_coherent(struct device *hwdev, size_t size,
 			   dma_addr_t *dma_handle, gfp_t flag)
 {
 	return iommu_alloc_coherent(devnode_table(hwdev), size, dma_handle,
-			flag);
+			device_to_mask(hwdev), flag);
 }
 
 static void pci_iommu_free_coherent(struct device *hwdev, size_t size,
@@ -85,7 +104,8 @@ static void pci_iommu_free_coherent(struct device *hwdev, size_t size,
 static dma_addr_t pci_iommu_map_single(struct device *hwdev, void *vaddr,
 		size_t size, enum dma_data_direction direction)
 {
-	return iommu_map_single(devnode_table(hwdev), vaddr, size, direction);
+	return iommu_map_single(devnode_table(hwdev), vaddr, size,
+			        device_to_mask(hwdev), direction);
 }
 
 
@@ -100,7 +120,7 @@ static int pci_iommu_map_sg(struct device *pdev, struct scatterlist *sglist,
 		int nelems, enum dma_data_direction direction)
 {
 	return iommu_map_sg(pdev, devnode_table(pdev), sglist,
-			nelems, direction);
+			nelems, device_to_mask(pdev), direction);
 }
 
 static void pci_iommu_unmap_sg(struct device *pdev, struct scatterlist *sglist,
@@ -112,7 +132,19 @@ static void pci_iommu_unmap_sg(struct device *pdev, struct scatterlist *sglist,
 /* We support DMA to/from any memory page via the iommu */
 static int pci_iommu_dma_supported(struct device *dev, u64 mask)
 {
-	return 1;
+	struct iommu_table *tbl = devnode_table(dev);
+
+	if (!tbl || tbl->it_offset > mask) {
+		printk(KERN_INFO "Warning: IOMMU table offset too big for device mask\n");
+		if (tbl)
+			printk(KERN_INFO "mask: 0x%08lx, table offset: 0x%08lx\n",
+				mask, tbl->it_offset);
+		else
+			printk(KERN_INFO "mask: 0x%08lx, table unavailable\n",
+				mask);
+		return 0;
+	} else
+		return 1;
 }
 
 void pci_iommu_init(void)
diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c
index 13c655ba2841..971020cf3f7d 100644
--- a/arch/powerpc/kernel/vio.c
+++ b/arch/powerpc/kernel/vio.c
@@ -202,7 +202,7 @@ static dma_addr_t vio_map_single(struct device *dev, void *vaddr,
 			  size_t size, enum dma_data_direction direction)
 {
 	return iommu_map_single(to_vio_dev(dev)->iommu_table, vaddr, size,
-			direction);
+			~0ul, direction);
 }
 
 static void vio_unmap_single(struct device *dev, dma_addr_t dma_handle,
@@ -216,7 +216,7 @@ static int vio_map_sg(struct device *dev, struct scatterlist *sglist,
 		int nelems, enum dma_data_direction direction)
 {
 	return iommu_map_sg(dev, to_vio_dev(dev)->iommu_table, sglist,
-			nelems, direction);
+			nelems, ~0ul, direction);
 }
 
 static void vio_unmap_sg(struct device *dev, struct scatterlist *sglist,
@@ -229,7 +229,7 @@ static void *vio_alloc_coherent(struct device *dev, size_t size,
 			   dma_addr_t *dma_handle, gfp_t flag)
 {
 	return iommu_alloc_coherent(to_vio_dev(dev)->iommu_table, size,
-			dma_handle, flag);
+			dma_handle, ~0ul, flag);
 }
 
 static void vio_free_coherent(struct device *dev, size_t size,
diff --git a/include/asm-powerpc/iommu.h b/include/asm-powerpc/iommu.h
index d5677cbec200..18ca29e9105a 100644
--- a/include/asm-powerpc/iommu.h
+++ b/include/asm-powerpc/iommu.h
@@ -70,17 +70,18 @@ extern void iommu_free_table(struct device_node *dn);
 extern struct iommu_table *iommu_init_table(struct iommu_table * tbl);
 
 extern int iommu_map_sg(struct device *dev, struct iommu_table *tbl,
-		struct scatterlist *sglist, int nelems,
+		struct scatterlist *sglist, int nelems, unsigned long mask,
 		enum dma_data_direction direction);
 extern void iommu_unmap_sg(struct iommu_table *tbl, struct scatterlist *sglist,
 		int nelems, enum dma_data_direction direction);
 
 extern void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size,
-		dma_addr_t *dma_handle, gfp_t flag);
+		dma_addr_t *dma_handle, unsigned long mask, gfp_t flag);
 extern void iommu_free_coherent(struct iommu_table *tbl, size_t size,
 		void *vaddr, dma_addr_t dma_handle);
 extern dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr,
-		size_t size, enum dma_data_direction direction);
+		size_t size, unsigned long mask,
+		enum dma_data_direction direction);
 extern void iommu_unmap_single(struct iommu_table *tbl, dma_addr_t dma_handle,
 		size_t size, enum dma_data_direction direction);
 
-- 
cgit v1.2.3


From 2fd83038160531245099c3c5b3511fa4b80765eb Mon Sep 17 00:00:00 2001
From: Helge Deller <deller@parisc-linux.org>
Date: Thu, 20 Apr 2006 20:40:23 +0000
Subject: [PARISC] Further work for multiple page sizes

More work towards supporing multiple page sizes on 64-bit. Convert
some assumptions that 64bit uses 3 level page tables into testing
PT_NLEVELS. Also some BUG() to BUG_ON() conversions and some cleanups
to assembler.

Signed-off-by: Helge Deller <deller@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
---
 arch/parisc/Kconfig              | 31 ++++++++++++++++++++
 arch/parisc/kernel/asm-offsets.c |  3 ++
 arch/parisc/kernel/entry.S       | 36 ++++++++++++++---------
 arch/parisc/kernel/head.S        | 15 +++++-----
 arch/parisc/kernel/init_task.c   | 10 +++----
 arch/parisc/kernel/pacache.S     | 25 ++++++++--------
 arch/parisc/kernel/syscall.S     | 10 +++----
 arch/parisc/kernel/vmlinux.lds.S | 54 ++++++++++++++++++++--------------
 arch/parisc/mm/init.c            | 28 +++++++++---------
 include/asm-parisc/page.h        | 25 +++++++++++++---
 include/asm-parisc/pgtable.h     | 63 ++++++++++++++++++++++++++++------------
 11 files changed, 198 insertions(+), 102 deletions(-)

(limited to 'include')

diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 19f911c5dd58..910fb3afc0b5 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -138,6 +138,37 @@ config 64BIT
 	  enable this option otherwise. The 64bit kernel is significantly bigger
 	  and slower than the 32bit one.
 
+choice
+	prompt "Kernel page size"
+	default PARISC_PAGE_SIZE_4KB  if !64BIT
+	default PARISC_PAGE_SIZE_4KB  if 64BIT
+#	default PARISC_PAGE_SIZE_16KB if 64BIT
+
+config PARISC_PAGE_SIZE_4KB
+	bool "4KB"
+	help
+	  This lets you select the page size of the kernel.  For best
+	  performance, a page size of 16KB is recommended.  For best
+	  compatibility with 32bit applications, a page size of 4KB should be
+	  selected (the vast majority of 32bit binaries work perfectly fine
+	  with a larger page size).
+
+	  4KB                For best 32bit compatibility
+	  16KB               For best performance
+	  64KB               For best performance, might give more overhead.
+
+	  If you don't know what to do, choose 4KB.
+
+config PARISC_PAGE_SIZE_16KB
+	bool "16KB (EXPERIMENTAL)"
+	depends on PA8X00 && EXPERIMENTAL
+
+config PARISC_PAGE_SIZE_64KB
+	bool "64KB (EXPERIMENTAL)"
+	depends on PA8X00 && EXPERIMENTAL
+
+endchoice
+
 config SMP
 	bool "Symmetric multi-processing support"
 	---help---
diff --git a/arch/parisc/kernel/asm-offsets.c b/arch/parisc/kernel/asm-offsets.c
index e23c4e1e3a25..c11a5bc7c067 100644
--- a/arch/parisc/kernel/asm-offsets.c
+++ b/arch/parisc/kernel/asm-offsets.c
@@ -288,8 +288,11 @@ int main(void)
 	DEFINE(ASM_PGD_ENTRY_SIZE, PGD_ENTRY_SIZE);
 	DEFINE(ASM_PMD_ENTRY_SIZE, PMD_ENTRY_SIZE);
 	DEFINE(ASM_PTE_ENTRY_SIZE, PTE_ENTRY_SIZE);
+	DEFINE(ASM_PFN_PTE_SHIFT, PFN_PTE_SHIFT);
 	DEFINE(ASM_PT_INITIAL, PT_INITIAL);
 	DEFINE(ASM_PAGE_SIZE, PAGE_SIZE);
+	DEFINE(ASM_PAGE_SIZE_DIV64, PAGE_SIZE/64);
+	DEFINE(ASM_PAGE_SIZE_DIV128, PAGE_SIZE/128);
 	BLANK();
 	DEFINE(EXCDATA_IP, offsetof(struct exception_data, fault_ip));
 	DEFINE(EXCDATA_SPACE, offsetof(struct exception_data, fault_space));
diff --git a/arch/parisc/kernel/entry.S b/arch/parisc/kernel/entry.S
index 7c95d7663c29..d9e53cf0372b 100644
--- a/arch/parisc/kernel/entry.S
+++ b/arch/parisc/kernel/entry.S
@@ -502,18 +502,20 @@
 	 * all ILP32 processes and all the kernel for machines with
 	 * under 4GB of memory) */
 	.macro		L3_ptep pgd,pte,index,va,fault
+#if PT_NLEVELS == 3 /* we might have a 2-Level scheme, e.g. with 16kb page size */
 	extrd,u		\va,63-ASM_PGDIR_SHIFT,ASM_BITS_PER_PGD,\index
 	copy		%r0,\pte
-	extrd,u,*=	\va,31,32,%r0
+	extrd,u,*=	\va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0
 	ldw,s		\index(\pgd),\pgd
-	extrd,u,*=	\va,31,32,%r0
+	extrd,u,*=	\va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0
 	bb,>=,n		\pgd,_PxD_PRESENT_BIT,\fault
-	extrd,u,*=	\va,31,32,%r0
+	extrd,u,*=	\va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0
 	shld		\pgd,PxD_VALUE_SHIFT,\index
-	extrd,u,*=	\va,31,32,%r0
+	extrd,u,*=	\va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0
 	copy		\index,\pgd
-	extrd,u,*<>	\va,31,32,%r0
+	extrd,u,*<>	\va,63-ASM_PGDIR_SHIFT,64-ASM_PGDIR_SHIFT,%r0
 	ldo		ASM_PGD_PMD_OFFSET(\pgd),\pgd
+#endif
 	L2_ptep		\pgd,\pte,\index,\va,\fault
 	.endm
 
@@ -563,10 +565,18 @@
 	extrd,u,*= 	\pte,_PAGE_GATEWAY_BIT+32,1,%r0
 	depd		%r0,11,2,\prot	/* If Gateway, Set PL2 to 0 */
 
-	/* Get rid of prot bits and convert to page addr for iitlbt and idtlbt */
+	/* Enforce uncacheable pages.
+	 * This should ONLY be use for MMIO on PA 2.0 machines.
+	 * Memory/DMA is cache coherent on all PA2.0 machines we support
+	 * (that means T-class is NOT supported) and the memory controllers
+	 * on most of those machines only handles cache transactions.
+	 */
+	extrd,u,*=	\pte,_PAGE_NO_CACHE_BIT+32,1,%r0
+	depi		1,12,1,\prot
 
-	depd		%r0,63,PAGE_SHIFT,\pte
-	extrd,s		\pte,(63-PAGE_SHIFT)+(63-58),64-PAGE_SHIFT,\pte
+	/* Drop prot bits and convert to page addr for iitlbt and idtlbt */
+	extrd,u		\pte,(63-ASM_PFN_PTE_SHIFT)+(63-58),64-PAGE_SHIFT,\pte
+	depdi		_PAGE_SIZE_ENCODING_DEFAULT,63,63-58,\pte
 	.endm
 
 	/* Identical macro to make_insert_tlb above, except it
@@ -584,9 +594,8 @@
 
 	/* Get rid of prot bits and convert to page addr for iitlba */
 
-	depi		0,31,PAGE_SHIFT,\pte
+	depi		_PAGE_SIZE_ENCODING_DEFAULT,31,ASM_PFN_PTE_SHIFT,\pte
 	extru		\pte,24,25,\pte
-
 	.endm
 
 	/* This is for ILP32 PA2.0 only.  The TLB insertion needs
@@ -1201,10 +1210,9 @@ intr_save:
 	 */
 
 	/* adjust isr/ior. */
-
-	extrd,u         %r16,63,7,%r1    /* get high bits from isr for ior */
-	depd            %r1,31,7,%r17    /* deposit them into ior */
-	depdi           0,63,7,%r16      /* clear them from isr */
+	extrd,u         %r16,63,SPACEID_SHIFT,%r1	/* get high bits from isr for ior */
+	depd            %r1,31,SPACEID_SHIFT,%r17	/* deposit them into ior */
+	depdi           0,63,SPACEID_SHIFT,%r16		/* clear them from isr */
 #endif
 	STREG           %r16, PT_ISR(%r29)
 	STREG           %r17, PT_IOR(%r29)
diff --git a/arch/parisc/kernel/head.S b/arch/parisc/kernel/head.S
index 0b47afc20690..3e79e62f7b0b 100644
--- a/arch/parisc/kernel/head.S
+++ b/arch/parisc/kernel/head.S
@@ -76,16 +76,16 @@ $bss_loop:
 	mtctl		%r4,%cr24	/* Initialize kernel root pointer */
 	mtctl		%r4,%cr25	/* Initialize user root pointer */
 
-#ifdef CONFIG_64BIT
+#if PT_NLEVELS == 3
 	/* Set pmd in pgd */
 	load32		PA(pmd0),%r5
 	shrd            %r5,PxD_VALUE_SHIFT,%r3	
-        ldo             (PxD_FLAG_PRESENT+PxD_FLAG_VALID)(%r3),%r3	
+	ldo		(PxD_FLAG_PRESENT+PxD_FLAG_VALID)(%r3),%r3
 	stw		%r3,ASM_PGD_ENTRY*ASM_PGD_ENTRY_SIZE(%r4)
 	ldo		ASM_PMD_ENTRY*ASM_PMD_ENTRY_SIZE(%r5),%r4
 #else
 	/* 2-level page table, so pmd == pgd */
-        ldo             ASM_PGD_ENTRY*ASM_PGD_ENTRY_SIZE(%r4),%r4
+	ldo		ASM_PGD_ENTRY*ASM_PGD_ENTRY_SIZE(%r4),%r4
 #endif
 
 	/* Fill in pmd with enough pte directories */
@@ -99,7 +99,7 @@ $bss_loop:
 	stw		%r3,0(%r4)
 	ldo		(ASM_PAGE_SIZE >> PxD_VALUE_SHIFT)(%r3),%r3
 	addib,>		-1,%r1,1b
-#ifdef CONFIG_64BIT
+#if PT_NLEVELS == 3
 	ldo             ASM_PMD_ENTRY_SIZE(%r4),%r4
 #else
 	ldo             ASM_PGD_ENTRY_SIZE(%r4),%r4
@@ -107,13 +107,14 @@ $bss_loop:
 
 
 	/* Now initialize the PTEs themselves */
-	ldo		_PAGE_KERNEL(%r0),%r3 /* Hardwired 0 phys addr start */
+	ldo		0+_PAGE_KERNEL(%r0),%r3 /* Hardwired 0 phys addr start */
+	ldi		(1<<(KERNEL_INITIAL_ORDER-PAGE_SHIFT)),%r11 /* PFN count */
 	load32		PA(pg0),%r1
 
 $pgt_fill_loop:
 	STREGM          %r3,ASM_PTE_ENTRY_SIZE(%r1)
-	ldo		ASM_PAGE_SIZE(%r3),%r3
-	bb,>=		%r3,31-KERNEL_INITIAL_ORDER,$pgt_fill_loop
+	ldo		(1<<PFN_PTE_SHIFT)(%r3),%r3 /* add one PFN */
+	addib,>		-1,%r11,$pgt_fill_loop
 	nop
 
 	/* Load the return address...er...crash 'n burn */
diff --git a/arch/parisc/kernel/init_task.c b/arch/parisc/kernel/init_task.c
index 7e898fd64415..8384bf9cecd2 100644
--- a/arch/parisc/kernel/init_task.c
+++ b/arch/parisc/kernel/init_task.c
@@ -53,17 +53,17 @@ union thread_union init_thread_union
 	__attribute__((aligned(128))) __attribute__((__section__(".data.init_task"))) =
 		{ INIT_THREAD_INFO(init_task) };
 
-#ifdef __LP64__
+#if PT_NLEVELS == 3
 /* NOTE: This layout exactly conforms to the hybrid L2/L3 page table layout
  * with the first pmd adjacent to the pgd and below it. gcc doesn't actually
  * guarantee that global objects will be laid out in memory in the same order 
  * as the order of declaration, so put these in different sections and use
  * the linker script to order them. */
-pmd_t pmd0[PTRS_PER_PMD] __attribute__ ((aligned(PAGE_SIZE))) __attribute__ ((__section__ (".data.vm0.pmd"))) = { {0}, };
-
+pmd_t pmd0[PTRS_PER_PMD] __attribute__ ((__section__ (".data.vm0.pmd"), aligned(PAGE_SIZE)));
 #endif
-pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__ ((aligned(PAGE_SIZE))) __attribute__ ((__section__ (".data.vm0.pgd"))) = { {0}, };
-pte_t pg0[PT_INITIAL * PTRS_PER_PTE] __attribute__ ((aligned(PAGE_SIZE))) __attribute__ ((__section__ (".data.vm0.pte")))  = { {0}, };
+
+pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__ ((__section__ (".data.vm0.pgd"), aligned(PAGE_SIZE)));
+pte_t pg0[PT_INITIAL * PTRS_PER_PTE] __attribute__ ((__section__ (".data.vm0.pte"), aligned(PAGE_SIZE)));
 
 /*
  * Initial task structure.
diff --git a/arch/parisc/kernel/pacache.S b/arch/parisc/kernel/pacache.S
index 7a4f07e8d3c3..f600556414d1 100644
--- a/arch/parisc/kernel/pacache.S
+++ b/arch/parisc/kernel/pacache.S
@@ -65,7 +65,7 @@ flush_tlb_all_local:
 	 */
 
 	/* pcxt_ssm_bug	- relied upon translation! PA 2.0 Arch. F-4 and F-5 */
-	rsm	PSW_SM_I, %r19		/* save I-bit state */
+	rsm		PSW_SM_I, %r19		/* save I-bit state */
 	load32		PA(1f), %r1
 	nop
 	nop
@@ -84,8 +84,7 @@ flush_tlb_all_local:
 	rfi
 	nop
 
-1:      ldil		L%PA(cache_info), %r1
-	ldo		R%PA(cache_info)(%r1), %r1
+1:      load32		PA(cache_info), %r1
 
 	/* Flush Instruction Tlb */
 
@@ -212,8 +211,7 @@ flush_instruction_cache_local:
 	.entry
 
 	mtsp		%r0, %sr1
-	ldil		L%cache_info, %r1
-	ldo		R%cache_info(%r1), %r1
+	load32		cache_info, %r1
 
 	/* Flush Instruction Cache */
 
@@ -254,8 +252,7 @@ flush_data_cache_local:
 	.entry
 
 	mtsp		%r0, %sr1
-	ldil		L%cache_info, %r1
-	ldo		R%cache_info(%r1), %r1
+	load32 		cache_info, %r1
 
 	/* Flush Data Cache */
 
@@ -303,7 +300,8 @@ copy_user_page_asm:
 	 */
 
 	ldd		0(%r25), %r19
-	ldi		32, %r1                 /* PAGE_SIZE/128 == 32 */
+	ldi		ASM_PAGE_SIZE_DIV128, %r1
+
 	ldw		64(%r25), %r0		/* prefetch 1 cacheline ahead */
 	ldw		128(%r25), %r0		/* prefetch 2 */
 
@@ -368,7 +366,7 @@ copy_user_page_asm:
 	 * use ldd/std on a 32 bit kernel.
 	 */
 	ldw		0(%r25), %r19
-	ldi		64, %r1		/* PAGE_SIZE/64 == 64 */
+	ldi		ASM_PAGE_SIZE_DIV64, %r1
 
 1:
 	ldw		4(%r25), %r20
@@ -461,6 +459,7 @@ copy_user_page_asm:
 	sub		%r25, %r1, %r23		/* move physical addr into non shadowed reg */
 
 	ldil		L%(TMPALIAS_MAP_START), %r28
+	/* FIXME for different page sizes != 4k */
 #ifdef CONFIG_64BIT
 	extrd,u		%r26,56,32, %r26		/* convert phys addr to tlb insert format */
 	extrd,u		%r23,56,32, %r23		/* convert phys addr to tlb insert format */
@@ -551,6 +550,7 @@ __clear_user_page_asm:
 #ifdef CONFIG_64BIT
 #if (TMPALIAS_MAP_START >= 0x80000000)
 	depdi		0, 31,32, %r28		/* clear any sign extension */
+	/* FIXME: page size dependend */
 #endif
 	extrd,u		%r26, 56,32, %r26	/* convert phys addr to tlb insert format */
 	depd		%r25, 63,22, %r28	/* Form aliased virtual address 'to' */
@@ -566,10 +566,10 @@ __clear_user_page_asm:
 	pdtlb		0(%r28)
 
 #ifdef CONFIG_64BIT
-	ldi		32, %r1			/* PAGE_SIZE/128 == 32 */
+	ldi		ASM_PAGE_SIZE_DIV128, %r1
 
 	/* PREFETCH (Write) has not (yet) been proven to help here */
-/* #define	PREFETCHW_OP	ldd		256(%0), %r0 */
+	/* #define	PREFETCHW_OP	ldd		256(%0), %r0 */
 
 1:	std		%r0, 0(%r28)
 	std		%r0, 8(%r28)
@@ -591,8 +591,7 @@ __clear_user_page_asm:
 	ldo		128(%r28), %r28
 
 #else	/* ! CONFIG_64BIT */
-
-	ldi		64, %r1			/* PAGE_SIZE/64 == 64 */
+	ldi		ASM_PAGE_SIZE_DIV64, %r1
 
 1:
 	stw		%r0, 0(%r28)
diff --git a/arch/parisc/kernel/syscall.S b/arch/parisc/kernel/syscall.S
index af88afef41bd..479d9a017cd1 100644
--- a/arch/parisc/kernel/syscall.S
+++ b/arch/parisc/kernel/syscall.S
@@ -55,7 +55,7 @@
 	 * pointers.
 	 */
 
-	.align 4096
+	.align ASM_PAGE_SIZE
 linux_gateway_page:
 
         /* ADDRESS 0x00 to 0xb0 = 176 bytes / 4 bytes per insn = 44 insns */
@@ -632,7 +632,7 @@ cas_action:
 end_compare_and_swap:
 
 	/* Make sure nothing else is placed on this page */
-	.align 4096
+	.align ASM_PAGE_SIZE
 	.export end_linux_gateway_page
 end_linux_gateway_page:
 
@@ -652,7 +652,7 @@ end_linux_gateway_page:
 
 	.section .rodata,"a"
 
-	.align 4096
+	.align ASM_PAGE_SIZE
 	/* Light-weight-syscall table */
 	/* Start of lws table. */
 	.export lws_table
@@ -662,14 +662,14 @@ lws_table:
 	LWS_ENTRY(compare_and_swap64)	/* 1 - ELF64 Atomic compare and swap */
 	/* End of lws table */
 
-	.align 4096
+	.align ASM_PAGE_SIZE
 	.export sys_call_table
 .Lsys_call_table:
 sys_call_table:
 #include "syscall_table.S"
 
 #ifdef CONFIG_64BIT
-	.align 4096
+	.align ASM_PAGE_SIZE
 	.export sys_call_table64
 .Lsys_call_table64:
 sys_call_table64:
diff --git a/arch/parisc/kernel/vmlinux.lds.S b/arch/parisc/kernel/vmlinux.lds.S
index 6d6436a6b624..94dcc03a28ed 100644
--- a/arch/parisc/kernel/vmlinux.lds.S
+++ b/arch/parisc/kernel/vmlinux.lds.S
@@ -6,6 +6,7 @@
  *    Copyright (C) 2000 Michael Ang <mang with subcarrier.org>
  *    Copyright (C) 2002 Randolph Chung <tausq with parisc-linux.org>
  *    Copyright (C) 2003 James Bottomley <jejb with parisc-linux.org>
+ *    Copyright (C) 2006 Helge Deller <deller@gmx.de>
  *
  *
  *    This program is free software; you can redistribute it and/or modify
@@ -27,6 +28,7 @@
 /* needed for the processor specific cache alignment size */	
 #include <asm/cache.h>
 #include <asm/page.h>
+#include <asm/asm-offsets.h>
 	
 /* ld script to make hppa Linux kernel */
 #ifndef CONFIG_64BIT
@@ -68,7 +70,7 @@ SECTIONS
   RODATA
 
   /* writeable */
-  . = ALIGN(4096);		/* Make sure this is page aligned so
+  . = ALIGN(ASM_PAGE_SIZE);	/* Make sure this is page aligned so
   				   that we can properly leave these
 				   as writable */
   data_start = .;
@@ -81,23 +83,17 @@ SECTIONS
   __start___unwind = .;         /* unwind info */
   .PARISC.unwind : { *(.PARISC.unwind) }
   __stop___unwind = .;
- 
+
+  /* rarely changed data like cpu maps */
+  . = ALIGN(16);
+  .data.read_mostly : { *(.data.read_mostly) }
+
+  . = ALIGN(L1_CACHE_BYTES);
   .data : {			/* Data */
 	*(.data)
-	*(.data.vm0.pmd)
-	*(.data.vm0.pgd)
-	*(.data.vm0.pte)
 	CONSTRUCTORS
 	}
 
-  . = ALIGN(4096);
-  /* nosave data is really only used for software suspend...it's here
-   * just in case we ever implement it */
-  __nosave_begin = .;
-  .data_nosave : { *(.data.nosave) }
-  . = ALIGN(4096);
-  __nosave_end = .;
-
   . = ALIGN(L1_CACHE_BYTES);
   .data.cacheline_aligned : { *(.data.cacheline_aligned) }
 
@@ -105,12 +101,29 @@ SECTIONS
   . = ALIGN(16);
   .data.lock_aligned : { *(.data.lock_aligned) }
 
-  /* rarely changed data like cpu maps */
-  . = ALIGN(16);
-  .data.read_mostly : { *(.data.read_mostly) }
+  . = ALIGN(ASM_PAGE_SIZE);
+  /* nosave data is really only used for software suspend...it's here
+   * just in case we ever implement it */
+  __nosave_begin = .;
+  .data_nosave : { *(.data.nosave) }
+  . = ALIGN(ASM_PAGE_SIZE);
+  __nosave_end = .;
 
   _edata = .;			/* End of data section */
 
+  __bss_start = .;		/* BSS */
+  /* page table entries need to be PAGE_SIZE aligned */
+  . = ALIGN(ASM_PAGE_SIZE);
+  .data.vmpages : {
+	*(.data.vm0.pmd)
+	*(.data.vm0.pgd)
+	*(.data.vm0.pte)
+	}
+  .bss : { *(.bss) *(COMMON) }
+  __bss_stop = .;
+
+
+  /* assembler code expects init_task to be 16k aligned */
   . = ALIGN(16384); 		/* init_task */
   .data.init_task : { *(.data.init_task) }
 
@@ -126,6 +139,7 @@ SECTIONS
   .dlt : { *(.dlt) }
 #endif
 
+  /* reserve space for interrupt stack by aligning __init* to 16k */
   . = ALIGN(16384);
   __init_begin = .;
   .init.text : { 
@@ -166,7 +180,7 @@ SECTIONS
      from .altinstructions and .eh_frame */
   .exit.text : { *(.exit.text) }
   .exit.data : { *(.exit.data) }
-  . = ALIGN(4096);
+  . = ALIGN(ASM_PAGE_SIZE);
   __initramfs_start = .;
   .init.ramfs : { *(.init.ramfs) }
   __initramfs_end = .;
@@ -174,14 +188,10 @@ SECTIONS
   __per_cpu_start = .;
   .data.percpu  : { *(.data.percpu) }
   __per_cpu_end = .;
-  . = ALIGN(4096);
+  . = ALIGN(ASM_PAGE_SIZE);
   __init_end = .;
   /* freed after init ends here */
 	
-  __bss_start = .;		/* BSS */
-  .bss : { *(.bss) *(COMMON) }
-  __bss_stop = .; 
-
   _end = . ;
 
   /* Sections to be discarded */
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index 3796be67cd53..631712562656 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -6,6 +6,7 @@
  *    changed by Philipp Rumpf
  *  Copyright 1999 Philipp Rumpf (prumpf@tux.org)
  *  Copyright 2004 Randolph Chung (tausq@debian.org)
+ *  Copyright 2006 Helge Deller (deller@gmx.de)
  *
  */
 
@@ -371,8 +372,8 @@ static void __init setup_bootmem(void)
 
 void free_initmem(void)
 {
-	unsigned long addr;
-	
+	unsigned long addr, init_begin, init_end;
+
 	printk(KERN_INFO "Freeing unused kernel memory: ");
 
 #ifdef CONFIG_DEBUG_KERNEL
@@ -395,8 +396,11 @@ void free_initmem(void)
 	local_irq_enable();
 #endif
 	
-	addr = (unsigned long)(&__init_begin);
-	for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) {
+	/* align __init_begin and __init_end to page size,
+	   ignoring linker script where we might have tried to save RAM */
+	init_begin = PAGE_ALIGN((unsigned long)(&__init_begin));
+	init_end   = PAGE_ALIGN((unsigned long)(&__init_end));
+	for (addr = init_begin; addr < init_end; addr += PAGE_SIZE) {
 		ClearPageReserved(virt_to_page(addr));
 		init_page_count(virt_to_page(addr));
 		free_page(addr);
@@ -407,7 +411,7 @@ void free_initmem(void)
 	/* set up a new led state on systems shipped LED State panel */
 	pdc_chassis_send_status(PDC_CHASSIS_DIRECT_BCOMPLETE);
 	
-	printk("%luk freed\n", (unsigned long)(&__init_end - &__init_begin) >> 10);
+	printk("%luk freed\n", (init_end - init_begin) >> 10);
 }
 
 
@@ -639,11 +643,13 @@ static void __init map_pages(unsigned long start_vaddr, unsigned long start_padd
 				 * Map the fault vector writable so we can
 				 * write the HPMC checksum.
 				 */
+#if defined(CONFIG_PARISC_PAGE_SIZE_4KB)
 				if (address >= ro_start && address < ro_end
 							&& address != fv_addr
 							&& address != gw_addr)
 				    pte = __mk_pte(address, PAGE_KERNEL_RO);
 				else
+#endif
 				    pte = __mk_pte(address, pgprot);
 
 				if (address >= end_paddr)
@@ -874,8 +880,7 @@ unsigned long alloc_sid(void)
 			flush_tlb_all(); /* flush_tlb_all() calls recycle_sids() */
 			spin_lock(&sid_lock);
 		}
-		if (free_space_ids == 0)
-			BUG();
+		BUG_ON(free_space_ids == 0);
 	}
 
 	free_space_ids--;
@@ -899,8 +904,7 @@ void free_sid(unsigned long spaceid)
 
 	spin_lock(&sid_lock);
 
-	if (*dirty_space_offset & (1L << index))
-	    BUG(); /* attempt to free space id twice */
+	BUG_ON(*dirty_space_offset & (1L << index)); /* attempt to free space id twice */
 
 	*dirty_space_offset |= (1L << index);
 	dirty_space_ids++;
@@ -975,7 +979,7 @@ static void recycle_sids(void)
 
 static unsigned long recycle_ndirty;
 static unsigned long recycle_dirty_array[SID_ARRAY_SIZE];
-static unsigned int recycle_inuse = 0;
+static unsigned int recycle_inuse;
 
 void flush_tlb_all(void)
 {
@@ -984,9 +988,7 @@ void flush_tlb_all(void)
 	do_recycle = 0;
 	spin_lock(&sid_lock);
 	if (dirty_space_ids > RECYCLE_THRESHOLD) {
-	    if (recycle_inuse) {
-		BUG();  /* FIXME: Use a semaphore/wait queue here */
-	    }
+	    BUG_ON(recycle_inuse);  /* FIXME: Use a semaphore/wait queue here */
 	    get_dirty_sids(&recycle_ndirty,recycle_dirty_array);
 	    recycle_inuse++;
 	    do_recycle++;
diff --git a/include/asm-parisc/page.h b/include/asm-parisc/page.h
index 45e02aa5bf4b..c0dd461fb8f1 100644
--- a/include/asm-parisc/page.h
+++ b/include/asm-parisc/page.h
@@ -1,13 +1,30 @@
 #ifndef _PARISC_PAGE_H
 #define _PARISC_PAGE_H
 
-/* PAGE_SHIFT determines the page size */
-#define PAGE_SHIFT	12
-#define PAGE_SIZE	(1UL << PAGE_SHIFT)
-#define PAGE_MASK	(~(PAGE_SIZE-1))
+#if !defined(__KERNEL__)
+/* this is for userspace applications (4k page size) */
+# define PAGE_SHIFT	12	/* 4k */
+# define PAGE_SIZE	(1UL << PAGE_SHIFT)
+# define PAGE_MASK	(~(PAGE_SIZE-1))
+#endif
+
 
 #ifdef __KERNEL__
 #include <linux/config.h>
+
+#if defined(CONFIG_PARISC_PAGE_SIZE_4KB)
+# define PAGE_SHIFT	12	/* 4k */
+#elif defined(CONFIG_PARISC_PAGE_SIZE_16KB)
+# define PAGE_SHIFT	14	/* 16k */
+#elif defined(CONFIG_PARISC_PAGE_SIZE_64KB)
+# define PAGE_SHIFT	16	/* 64k */
+#else
+# error "unknown default kernel page size"
+#endif
+#define PAGE_SIZE	(1UL << PAGE_SHIFT)
+#define PAGE_MASK	(~(PAGE_SIZE-1))
+
+
 #ifndef __ASSEMBLY__
 
 #include <asm/types.h>
diff --git a/include/asm-parisc/pgtable.h b/include/asm-parisc/pgtable.h
index 4e34c6b44059..aec089eb8b85 100644
--- a/include/asm-parisc/pgtable.h
+++ b/include/asm-parisc/pgtable.h
@@ -59,16 +59,15 @@
 #define  ISTACK_SIZE  32768 /* Interrupt Stack Size */
 #define  ISTACK_ORDER 3
 
-/* This is the size of the initially mapped kernel memory (i.e. currently
- * 0 to 1<<23 == 8MB */
+/* This is the size of the initially mapped kernel memory */
 #ifdef CONFIG_64BIT
-#define KERNEL_INITIAL_ORDER	24
+#define KERNEL_INITIAL_ORDER	24	/* 0 to 1<<24 = 16MB */
 #else
-#define KERNEL_INITIAL_ORDER	23
+#define KERNEL_INITIAL_ORDER	23	/* 0 to 1<<23 = 8MB */
 #endif
 #define KERNEL_INITIAL_SIZE	(1 << KERNEL_INITIAL_ORDER)
 
-#ifdef CONFIG_64BIT
+#if defined(CONFIG_64BIT) && defined(CONFIG_PARISC_PAGE_SIZE_4KB)
 #define PT_NLEVELS	3
 #define PGD_ORDER	1 /* Number of pages per pgd */
 #define PMD_ORDER	1 /* Number of pages per pmd */
@@ -111,11 +110,15 @@
 #define MAX_ADDRBITS	(PGDIR_SHIFT + BITS_PER_PGD)
 #define MAX_ADDRESS	(1UL << MAX_ADDRBITS)
 
-#define SPACEID_SHIFT (MAX_ADDRBITS - 32)
+#define SPACEID_SHIFT	(MAX_ADDRBITS - 32)
 
 /* This calculates the number of initial pages we need for the initial
  * page tables */
-#define PT_INITIAL	(1 << (KERNEL_INITIAL_ORDER - PMD_SHIFT))
+#if (KERNEL_INITIAL_ORDER) >= (PMD_SHIFT)
+# define PT_INITIAL	(1 << (KERNEL_INITIAL_ORDER - PMD_SHIFT))
+#else
+# define PT_INITIAL	(1)  /* all initial PTEs fit into one page */
+#endif
 
 /*
  * pgd entries used up by user/kernel:
@@ -160,6 +163,10 @@ extern  void *vmalloc_start;
  * to zero */
 #define PTE_SHIFT	   	xlate_pabit(_PAGE_USER_BIT)
 
+/* PFN_PTE_SHIFT defines the shift of a PTE value to access the PFN field */
+#define PFN_PTE_SHIFT		12
+
+
 /* this is how many bits may be used by the file functions */
 #define PTE_FILE_MAX_BITS	(BITS_PER_LONG - PTE_SHIFT)
 
@@ -188,7 +195,8 @@ extern  void *vmalloc_start;
 /* The pgd/pmd contains a ptr (in phys addr space); since all pgds/pmds
  * are page-aligned, we don't care about the PAGE_OFFSET bits, except
  * for a few meta-information bits, so we shift the address to be
- * able to effectively address 40-bits of physical address space. */
+ * able to effectively address 40/42/44-bits of physical address space
+ * depending on 4k/16k/64k PAGE_SIZE */
 #define _PxD_PRESENT_BIT   31
 #define _PxD_ATTACHED_BIT  30
 #define _PxD_VALID_BIT     29
@@ -198,7 +206,7 @@ extern  void *vmalloc_start;
 #define PxD_FLAG_VALID    (1 << xlate_pabit(_PxD_VALID_BIT))
 #define PxD_FLAG_MASK     (0xf)
 #define PxD_FLAG_SHIFT    (4)
-#define PxD_VALUE_SHIFT   (8)
+#define PxD_VALUE_SHIFT   (8) /* (PAGE_SHIFT-PxD_FLAG_SHIFT) */
 
 #ifndef __ASSEMBLY__
 
@@ -246,6 +254,7 @@ extern  void *vmalloc_start;
 #define __S110  PAGE_RWX
 #define __S111  PAGE_RWX
 
+
 extern pgd_t swapper_pg_dir[]; /* declared in init_task.c */
 
 /* initial page tables for 0-8MB for kernel */
@@ -272,7 +281,7 @@ extern unsigned long *empty_zero_page;
 #define pgd_flag(x)	(pgd_val(x) & PxD_FLAG_MASK)
 #define pgd_address(x)	((unsigned long)(pgd_val(x) &~ PxD_FLAG_MASK) << PxD_VALUE_SHIFT)
 
-#ifdef CONFIG_64BIT
+#if PT_NLEVELS == 3
 /* The first entry of the permanent pmd is not there if it contains
  * the gateway marker */
 #define pmd_none(x)	(!pmd_val(x) || pmd_flag(x) == PxD_FLAG_ATTACHED)
@@ -282,7 +291,7 @@ extern unsigned long *empty_zero_page;
 #define pmd_bad(x)	(!(pmd_flag(x) & PxD_FLAG_VALID))
 #define pmd_present(x)	(pmd_flag(x) & PxD_FLAG_PRESENT)
 static inline void pmd_clear(pmd_t *pmd) {
-#ifdef CONFIG_64BIT
+#if PT_NLEVELS == 3
 	if (pmd_flag(*pmd) & PxD_FLAG_ATTACHED)
 		/* This is the entry pointing to the permanent pmd
 		 * attached to the pgd; cannot clear it */
@@ -303,7 +312,7 @@ static inline void pmd_clear(pmd_t *pmd) {
 #define pgd_bad(x)      (!(pgd_flag(x) & PxD_FLAG_VALID))
 #define pgd_present(x)  (pgd_flag(x) & PxD_FLAG_PRESENT)
 static inline void pgd_clear(pgd_t *pgd) {
-#ifdef CONFIG_64BIT
+#if PT_NLEVELS == 3
 	if(pgd_flag(*pgd) & PxD_FLAG_ATTACHED)
 		/* This is the permanent pmd attached to the pgd; cannot
 		 * free it */
@@ -351,7 +360,7 @@ extern inline pte_t pte_mkwrite(pte_t pte)	{ pte_val(pte) |= _PAGE_WRITE; return
 ({									\
 	pte_t __pte;							\
 									\
-	pte_val(__pte) = ((addr)+pgprot_val(pgprot));			\
+	pte_val(__pte) = ((((addr)>>PAGE_SHIFT)<<PFN_PTE_SHIFT) + pgprot_val(pgprot));	\
 									\
 	__pte;								\
 })
@@ -361,20 +370,16 @@ extern inline pte_t pte_mkwrite(pte_t pte)	{ pte_val(pte) |= _PAGE_WRITE; return
 static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
 {
 	pte_t pte;
-	pte_val(pte) = (pfn << PAGE_SHIFT) | pgprot_val(pgprot);
+	pte_val(pte) = (pfn << PFN_PTE_SHIFT) | pgprot_val(pgprot);
 	return pte;
 }
 
-/* This takes a physical page address that is used by the remapping functions */
-#define mk_pte_phys(physpage, pgprot) \
-({ pte_t __pte; pte_val(__pte) = physpage + pgprot_val(pgprot); __pte; })
-
 extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 { pte_val(pte) = (pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot); return pte; }
 
 /* Permanent address of a page.  On parisc we don't have highmem. */
 
-#define pte_pfn(x) (pte_val(x) >> PAGE_SHIFT)
+#define pte_pfn(x)		(pte_val(x) >> PFN_PTE_SHIFT)
 
 #define pte_page(pte)		(pfn_to_page(pte_pfn(pte)))
 
@@ -499,6 +504,26 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr,
 
 #endif /* !__ASSEMBLY__ */
 
+
+/* TLB page size encoding - see table 3-1 in parisc20.pdf */
+#define _PAGE_SIZE_ENCODING_4K		0
+#define _PAGE_SIZE_ENCODING_16K	1
+#define _PAGE_SIZE_ENCODING_64K	2
+#define _PAGE_SIZE_ENCODING_256K	3
+#define _PAGE_SIZE_ENCODING_1M		4
+#define _PAGE_SIZE_ENCODING_4M		5
+#define _PAGE_SIZE_ENCODING_16M	6
+#define _PAGE_SIZE_ENCODING_64M	7
+
+#if defined(CONFIG_PARISC_PAGE_SIZE_4KB)
+# define _PAGE_SIZE_ENCODING_DEFAULT _PAGE_SIZE_ENCODING_4K
+#elif defined(CONFIG_PARISC_PAGE_SIZE_16KB)
+# define _PAGE_SIZE_ENCODING_DEFAULT _PAGE_SIZE_ENCODING_16K
+#elif defined(CONFIG_PARISC_PAGE_SIZE_64KB)
+# define _PAGE_SIZE_ENCODING_DEFAULT _PAGE_SIZE_ENCODING_64K
+#endif
+
+
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
 
-- 
cgit v1.2.3


From 6ca773cf8b9dc19989c9b44635292b1ba80f9112 Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@parisc-linux.org>
Date: Thu, 20 Apr 2006 04:44:07 +0000
Subject: [PARISC] Add new entries to the syscall table

Most are easy, but sync_file_range needed special handling when entering
through the 32-bit syscall table.

Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
---
 arch/parisc/kernel/sys_parisc.c    | 8 ++++++++
 arch/parisc/kernel/syscall_table.S | 8 +++++++-
 include/asm-parisc/unistd.h        | 8 +++++++-
 3 files changed, 22 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/arch/parisc/kernel/sys_parisc.c b/arch/parisc/kernel/sys_parisc.c
index d15a1d53e101..8b5df98e2b31 100644
--- a/arch/parisc/kernel/sys_parisc.c
+++ b/arch/parisc/kernel/sys_parisc.c
@@ -231,6 +231,14 @@ asmlinkage long parisc_fadvise64_64(int fd,
 			(loff_t)high_len << 32 | low_len, advice);
 }
 
+asmlinkage long parisc_sync_file_range(int fd,
+			u32 hi_off, u32 lo_off, u32 hi_nbytes, u32 lo_nbytes,
+			unsigned int flags)
+{
+	return sys_sync_file_range(fd, (loff_t)hi_off << 32 | lo_off,
+			(loff_t)hi_nbytes << 32 | lo_nbytes, flags);
+}
+
 asmlinkage unsigned long sys_alloc_hugepages(int key, unsigned long addr, unsigned long len, int prot, int flag)
 {
 	return -ENOMEM;
diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S
index bbeeb614cfab..e27b432f90a8 100644
--- a/arch/parisc/kernel/syscall_table.S
+++ b/arch/parisc/kernel/syscall_table.S
@@ -13,7 +13,7 @@
  *    Copyright (C) 2001 Helge Deller <deller at parisc-linux.org>
  *    Copyright (C) 2000-2001 Thomas Bogendoerfer <tsbogend at parisc-linux.org>
  *    Copyright (C) 2002 Randolph Chung <tausq with parisc-linux.org>
- *
+ *    Copyright (C) 2005-2006 Kyle McMartin <kyle at parisc-linux.org>
  *
  *    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
@@ -393,5 +393,11 @@
 	ENTRY_SAME(readlinkat)		/* 285 */
 	ENTRY_SAME(fchmodat)
 	ENTRY_SAME(faccessat)
+	ENTRY_SAME(unshare)
+	ENTRY_COMP(set_robust_list)
+	ENTRY_COMP(get_robust_list)	/* 290 */
+	ENTRY_SAME(splice)
+	ENTRY_OURS(sync_file_range)
+	ENTRY_SAME(tee)
 	/* Nothing yet */
 
diff --git a/include/asm-parisc/unistd.h b/include/asm-parisc/unistd.h
index c56fccbf34ad..0e1a30be2e30 100644
--- a/include/asm-parisc/unistd.h
+++ b/include/asm-parisc/unistd.h
@@ -780,8 +780,14 @@
 #define __NR_readlinkat		(__NR_Linux + 285)
 #define __NR_fchmodat		(__NR_Linux + 286)
 #define __NR_faccessat		(__NR_Linux + 287)
+#define __NR_unshare		(__NR_Linux + 288)
+#define __NR_set_robust_list	(__NR_Linux + 289)
+#define __NR_get_robust_list	(__NR_Linux + 290)
+#define __NR_splice		(__NR_Linux + 291)
+#define __NR_sync_file_range	(__NR_Linux + 292)
+#define __NR_tee		(__NR_Linux + 293)
 
-#define __NR_Linux_syscalls     288
+#define __NR_Linux_syscalls     294
 
 #define HPUX_GATEWAY_ADDR       0xC0000004
 #define LINUX_GATEWAY_ADDR      0x100
-- 
cgit v1.2.3


From 1b52d7c2210b9a64c5cba6aded478c8217a8853c Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@parisc-linux.org>
Date: Thu, 20 Apr 2006 21:16:32 +0000
Subject: [PARISC] Make ioremap default to _nocache

Since it is way more work to change most drivers to comply with parisc, take
the easy way out and make ioremap _NO_CACHE by default. This is in line with
what powerpc does.

Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
---
 include/asm-parisc/io.h | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/include/asm-parisc/io.h b/include/asm-parisc/io.h
index 29da31194b91..244f6b8883f4 100644
--- a/include/asm-parisc/io.h
+++ b/include/asm-parisc/io.h
@@ -126,24 +126,17 @@ static inline void gsc_writeq(unsigned long long val, unsigned long addr)
 
 extern void __iomem * __ioremap(unsigned long offset, unsigned long size, unsigned long flags);
 
-extern inline void __iomem * ioremap(unsigned long offset, unsigned long size)
-{
-	return __ioremap(offset, size, 0);
-}
-
-/*
- * This one maps high address device memory and turns off caching for that area.
- * it's useful if some control registers are in such an area and write combining
- * or read caching is not desirable:
+/* Most machines react poorly to I/O-space being cacheable... Instead let's
+ * define ioremap() in terms of ioremap_nocache().
  */
-extern inline void * ioremap_nocache(unsigned long offset, unsigned long size)
+extern inline void __iomem * ioremap(unsigned long offset, unsigned long size)
 {
-        return __ioremap(offset, size, _PAGE_NO_CACHE /* _PAGE_PCD */);
+	return __ioremap(offset, size, _PAGE_NO_CACHE);
 }
+#define ioremap_nocache(off, sz)	ioremap((off), (sz))
 
 extern void iounmap(void __iomem *addr);
 
-
 static inline unsigned char __raw_readb(const volatile void __iomem *addr)
 {
 	return (*(volatile unsigned char __force *) (addr));
-- 
cgit v1.2.3


From d0e15bed84db7a9b0ea85d2ad9707b5e6d2e38da Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@samba.org>
Date: Sun, 23 Apr 2006 10:42:04 +1000
Subject: powerpc: Fix define_machine so machine_is() works from modules

machine_is() was always returning 0 when used in a module, because
we weren't exporting the machine definitions.  This was why sound
wasn't working on powermacs when CONFIG_SND_POWERMAC=m.  Original
fix from Ben Herrenschmidt, further fixed by me.

Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/machdep.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include')

diff --git a/include/asm-powerpc/machdep.h b/include/asm-powerpc/machdep.h
index 5ed847680754..0f9254c18914 100644
--- a/include/asm-powerpc/machdep.h
+++ b/include/asm-powerpc/machdep.h
@@ -253,7 +253,11 @@ extern struct machdep_calls *machine_id;
 
 #define __machine_desc __attribute__ ((__section__ (".machine.desc")))
 
-#define define_machine(name) struct machdep_calls mach_##name __machine_desc =
+#define define_machine(name)					\
+	extern struct machdep_calls mach_##name;		\
+	EXPORT_SYMBOL(mach_##name);				\
+	struct machdep_calls mach_##name __machine_desc =
+
 #define machine_is(name) \
 	({ \
 		extern struct machdep_calls mach_##name \
-- 
cgit v1.2.3


From 818667f7c40dd0bd14029b5ac1d0f5282e12310e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 20 Apr 2006 20:02:03 +0200
Subject: [PATCH] softmac: fix SIOCSIWAP

There are some bugs in the current implementation of the SIOCSIWAP wext,
for example that when you do it twice and it fails, it may still try
another access point for some reason. This patch fixes this by introducing
a new flag that tells the association code that the bssid that is in use
was fixed by the user and shouldn't be deviated from.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/net/ieee80211softmac.h                  |  5 ++++-
 net/ieee80211/softmac/ieee80211softmac_assoc.c  | 20 ++++++++++++++----
 net/ieee80211/softmac/ieee80211softmac_module.c |  2 ++
 net/ieee80211/softmac/ieee80211softmac_wx.c     | 27 ++++++++++++++++---------
 4 files changed, 39 insertions(+), 15 deletions(-)

(limited to 'include')

diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h
index 6b3693f05ca0..b1ebfbae397f 100644
--- a/include/net/ieee80211softmac.h
+++ b/include/net/ieee80211softmac.h
@@ -96,10 +96,13 @@ struct ieee80211softmac_assoc_info {
 	 *
 	 * bssvalid is true if we found a matching network
 	 * and saved it's BSSID into the bssid above.
+	 *
+	 * bssfixed is used for SIOCSIWAP.
 	 */
 	u8 static_essid:1,
 	   associating:1,
-	   bssvalid:1;
+	   bssvalid:1,
+	   bssfixed:1;
 
 	/* Scan retries remaining */
 	int scan_retry;
diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c
index 4498023841dc..fb79ce7d6439 100644
--- a/net/ieee80211/softmac/ieee80211softmac_assoc.c
+++ b/net/ieee80211/softmac/ieee80211softmac_assoc.c
@@ -144,6 +144,12 @@ network_matches_request(struct ieee80211softmac_device *mac, struct ieee80211_ne
 	if (!we_support_all_basic_rates(mac, net->rates_ex, net->rates_ex_len))
 		return 0;
 
+	/* assume that users know what they're doing ...
+	 * (note we don't let them select a net we're incompatible with) */
+	if (mac->associnfo.bssfixed) {
+		return !memcmp(mac->associnfo.bssid, net->bssid, ETH_ALEN);
+	}
+
 	/* if 'ANY' network requested, take any that doesn't have privacy enabled */
 	if (mac->associnfo.req_essid.len == 0 
 	    && !(net->capability & WLAN_CAPABILITY_PRIVACY))
@@ -176,7 +182,7 @@ ieee80211softmac_assoc_work(void *d)
 		ieee80211softmac_disassoc(mac, WLAN_REASON_DISASSOC_STA_HAS_LEFT);
 
 	/* try to find the requested network in our list, if we found one already */
-	if (mac->associnfo.bssvalid)
+	if (mac->associnfo.bssvalid || mac->associnfo.bssfixed)
 		found = ieee80211softmac_get_network_by_bssid(mac, mac->associnfo.bssid);	
 	
 	/* Search the ieee80211 networks for this network if we didn't find it by bssid,
@@ -241,19 +247,25 @@ ieee80211softmac_assoc_work(void *d)
 			if (ieee80211softmac_start_scan(mac))
 				dprintk(KERN_INFO PFX "Associate: failed to initiate scan. Is device up?\n");
 			return;
-		}
-		else {
+		} else {
 			spin_lock_irqsave(&mac->lock, flags);
 			mac->associnfo.associating = 0;
 			mac->associated = 0;
 			spin_unlock_irqrestore(&mac->lock, flags);
 
 			dprintk(KERN_INFO PFX "Unable to find matching network after scan!\n");
+			/* reset the retry counter for the next user request since we
+			 * break out and don't reschedule ourselves after this point. */
+			mac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT;
 			ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_NET_NOT_FOUND, NULL);
 			return;
 		}
 	}
-	
+
+	/* reset the retry counter for the next user request since we
+	 * now found a net and will try to associate to it, but not
+	 * schedule this function again. */
+	mac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT;
 	mac->associnfo.bssvalid = 1;
 	memcpy(mac->associnfo.bssid, found->bssid, ETH_ALEN);
 	/* copy the ESSID for displaying it */
diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c
index 60f06a31f0d1..be83bdc1644a 100644
--- a/net/ieee80211/softmac/ieee80211softmac_module.c
+++ b/net/ieee80211/softmac/ieee80211softmac_module.c
@@ -45,6 +45,8 @@ struct net_device *alloc_ieee80211softmac(int sizeof_priv)
 	softmac->ieee->handle_disassoc = ieee80211softmac_handle_disassoc;
 	softmac->scaninfo = NULL;
 
+	softmac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT;
+
 	/* TODO: initialise all the other callbacks in the ieee struct
 	 *	 (once they're written)
 	 */
diff --git a/net/ieee80211/softmac/ieee80211softmac_wx.c b/net/ieee80211/softmac/ieee80211softmac_wx.c
index 00f0d4f71897..27edb2b5581a 100644
--- a/net/ieee80211/softmac/ieee80211softmac_wx.c
+++ b/net/ieee80211/softmac/ieee80211softmac_wx.c
@@ -27,7 +27,8 @@
 #include "ieee80211softmac_priv.h"
 
 #include <net/iw_handler.h>
-
+/* for is_broadcast_ether_addr and is_zero_ether_addr */
+#include <linux/etherdevice.h>
 
 int
 ieee80211softmac_wx_trigger_scan(struct net_device *net_dev,
@@ -83,7 +84,6 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev,
 			sm->associnfo.static_essid = 1;
 		}
 	}
-	sm->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT;
 
 	/* set our requested ESSID length.
 	 * If applicable, we have already copied the data in */
@@ -310,8 +310,6 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev,
 			    char *extra)
 {
 	struct ieee80211softmac_device *mac = ieee80211_priv(net_dev);
-	static const unsigned char any[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
-	static const unsigned char off[] = {0x00, 0x00, 0x00, 0x00, 0x00, 0x00};
 	unsigned long flags;
 
 	/* sanity check */
@@ -320,10 +318,17 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev,
 	}
 
 	spin_lock_irqsave(&mac->lock, flags);
-	if (!memcmp(any, data->ap_addr.sa_data, ETH_ALEN) ||
-	    !memcmp(off, data->ap_addr.sa_data, ETH_ALEN)) {
-		schedule_work(&mac->associnfo.work);
-		goto out;
+	if (is_broadcast_ether_addr(data->ap_addr.sa_data)) {
+		/* the bssid we have is not to be fixed any longer,
+		 * and we should reassociate to the best AP. */
+		mac->associnfo.bssfixed = 0;
+		/* force reassociation */
+		mac->associnfo.bssvalid = 0;
+		if (mac->associated)
+			schedule_work(&mac->associnfo.work);
+	} else if (is_zero_ether_addr(data->ap_addr.sa_data)) {
+		/* the bssid we have is no longer fixed */
+		mac->associnfo.bssfixed = 0;
         } else {
 		if (!memcmp(mac->associnfo.bssid, data->ap_addr.sa_data, ETH_ALEN)) {
 			if (mac->associnfo.associating || mac->associated) {
@@ -333,12 +338,14 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev,
 		} else {
 			/* copy new value in data->ap_addr.sa_data to bssid */
 			memcpy(mac->associnfo.bssid, data->ap_addr.sa_data, ETH_ALEN);
-		}	
+		}
+		/* tell the other code that this bssid should be used no matter what */
+		mac->associnfo.bssfixed = 1;
 		/* queue associate if new bssid or (old one again and not associated) */
 		schedule_work(&mac->associnfo.work);
         }
 
-out:
+ out:
 	spin_unlock_irqrestore(&mac->lock, flags);
 	return 0;
 }
-- 
cgit v1.2.3


From d8fe3f19203b1f5070358aaa292d33295258b448 Mon Sep 17 00:00:00 2001
From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Date: Mon, 24 Apr 2006 13:48:51 -0700
Subject: [SPARC]: __NR_sys removal

__NR_sys_sync_file_range part was lost somewhere...
[glibc is already checking __NR_sync_file_range]

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/asm-sparc/unistd.h   | 2 +-
 include/asm-sparc64/unistd.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include')

diff --git a/include/asm-sparc/unistd.h b/include/asm-sparc/unistd.h
index 45feff893b8e..32a48f623e2b 100644
--- a/include/asm-sparc/unistd.h
+++ b/include/asm-sparc/unistd.h
@@ -271,7 +271,7 @@
 #define __NR_getsid             252
 #define __NR_fdatasync          253
 #define __NR_nfsservctl         254
-#define __NR_sys_sync_file_range 255
+#define __NR_sync_file_range	255
 #define __NR_clock_settime	256
 #define __NR_clock_gettime	257
 #define __NR_clock_getres	258
diff --git a/include/asm-sparc64/unistd.h b/include/asm-sparc64/unistd.h
index 597f6923a46e..ca80e8aca128 100644
--- a/include/asm-sparc64/unistd.h
+++ b/include/asm-sparc64/unistd.h
@@ -273,7 +273,7 @@
 #define __NR_getsid             252
 #define __NR_fdatasync          253
 #define __NR_nfsservctl         254
-#define __NR_sys_sync_file_range 255
+#define __NR_sync_file_range	255
 #define __NR_clock_settime	256
 #define __NR_clock_gettime	257
 #define __NR_clock_getres	258
-- 
cgit v1.2.3


From 55fe5866366ae42f259f27ae5962eb267d9ce172 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 24 Apr 2006 17:16:28 -0700
Subject: [NETFILTER]: Fix compat_xt_counters alignment for non-x86

Some (?) non-x86 architectures require 8byte alignment for u_int64_t
even when compiled for 32bit, using u_int32_t in compat_xt_counters
breaks on these architectures, use u_int64_t for everything but x86.

Reported by Andreas Schwab <schwab@suse.de>.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index f6bdef82a322..38701454e197 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -361,7 +361,11 @@ struct compat_xt_entry_target
 
 struct compat_xt_counters
 {
+#if defined(CONFIG_X86_64) || defined(CONFIG_IA64)
 	u_int32_t cnt[4];
+#else
+	u_int64_t cnt[2];
+#endif
 };
 
 struct compat_xt_counters_info
-- 
cgit v1.2.3


From 912d35f86781e64d73be1ef358f703c08905ac37 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Wed, 26 Apr 2006 10:59:21 +0200
Subject: [PATCH] Add support for the sys_vmsplice syscall

sys_splice() moves data to/from pipes with a file input/output. sys_vmsplice()
moves data to a pipe, with the input being a user address range instead.

This uses an approach suggested by Linus, where we can hold partial ranges
inside the pages[] map. Hopefully this will be useful for network
receive support as well.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 arch/ia64/kernel/entry.S                    |   1 +
 arch/powerpc/kernel/systbl.S                |   1 +
 arch/powerpc/platforms/cell/spu_callbacks.c |   1 +
 fs/splice.c                                 | 292 ++++++++++++++++++++++++----
 include/asm-i386/unistd.h                   |   3 +-
 include/asm-ia64/unistd.h                   |   3 +-
 include/asm-powerpc/unistd.h                |   3 +-
 include/asm-x86_64/unistd.h                 |   4 +-
 include/linux/syscalls.h                    |   3 +
 9 files changed, 268 insertions(+), 43 deletions(-)

(limited to 'include')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index e30798811216..bcb80ca5cf40 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1610,5 +1610,6 @@ sys_call_table:
 	data8 sys_get_robust_list
 	data8 sys_sync_file_range		// 1300
 	data8 sys_tee
+	data8 sys_vmsplice
 
 	.org sys_call_table + 8*NR_syscalls	// guard against failures to increase NR_syscalls
diff --git a/arch/powerpc/kernel/systbl.S b/arch/powerpc/kernel/systbl.S
index 8d1522690501..0b98eea73c5e 100644
--- a/arch/powerpc/kernel/systbl.S
+++ b/arch/powerpc/kernel/systbl.S
@@ -324,6 +324,7 @@ COMPAT_SYS(ppoll)
 SYSCALL(unshare)
 SYSCALL(splice)
 SYSCALL(tee)
+SYSCALL(vmsplice)
 
 /*
  * please add new calls to arch/powerpc/platforms/cell/spu_callbacks.c
diff --git a/arch/powerpc/platforms/cell/spu_callbacks.c b/arch/powerpc/platforms/cell/spu_callbacks.c
index deb3afb94484..b283380a2a18 100644
--- a/arch/powerpc/platforms/cell/spu_callbacks.c
+++ b/arch/powerpc/platforms/cell/spu_callbacks.c
@@ -318,6 +318,7 @@ void *spu_syscall_table[] = {
 	[__NR_unshare]			sys_unshare,
 	[__NR_splice]			sys_splice,
 	[__NR_tee]			sys_tee,
+	[__NR_vmsplice]			sys_vmsplice,
 };
 
 long spu_sys_callback(struct spu_syscall_block *s)
diff --git a/fs/splice.c b/fs/splice.c
index 8c6030c762e2..0b2c1f060cae 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -27,6 +27,7 @@
 #include <linux/buffer_head.h>
 #include <linux/module.h>
 #include <linux/syscalls.h>
+#include <linux/uio.h>
 
 /*
  * Passed to the actors
@@ -38,6 +39,22 @@ struct splice_desc {
 	loff_t pos;			/* file position */
 };
 
+struct partial_page {
+	unsigned int offset;
+	unsigned int len;
+};
+
+/*
+ * Passed to move_to_pipe
+ */
+struct splice_pipe_desc {
+	struct page **pages;		/* page map */
+	struct partial_page *partial;	/* pages[] may not be contig */
+	int nr_pages;			/* number of pages in map */
+	unsigned int flags;		/* splice flags */
+	struct pipe_buf_operations *ops;/* ops associated with output pipe */
+};
+
 /*
  * Attempt to steal a page from a pipe buffer. This should perhaps go into
  * a vm helper function, it's already simplified quite a bit by the
@@ -128,6 +145,19 @@ static void page_cache_pipe_buf_unmap(struct pipe_inode_info *info,
 	kunmap(buf->page);
 }
 
+static void *user_page_pipe_buf_map(struct file *file,
+				    struct pipe_inode_info *pipe,
+				    struct pipe_buffer *buf)
+{
+	return kmap(buf->page);
+}
+
+static void user_page_pipe_buf_unmap(struct pipe_inode_info *pipe,
+				     struct pipe_buffer *buf)
+{
+	kunmap(buf->page);
+}
+
 static void page_cache_pipe_buf_get(struct pipe_inode_info *info,
 				    struct pipe_buffer *buf)
 {
@@ -143,19 +173,33 @@ static struct pipe_buf_operations page_cache_pipe_buf_ops = {
 	.get = page_cache_pipe_buf_get,
 };
 
+static int user_page_pipe_buf_steal(struct pipe_inode_info *pipe,
+				    struct pipe_buffer *buf)
+{
+	return 1;
+}
+
+static struct pipe_buf_operations user_page_pipe_buf_ops = {
+	.can_merge = 0,
+	.map = user_page_pipe_buf_map,
+	.unmap = user_page_pipe_buf_unmap,
+	.release = page_cache_pipe_buf_release,
+	.steal = user_page_pipe_buf_steal,
+	.get = page_cache_pipe_buf_get,
+};
+
 /*
  * Pipe output worker. This sets up our pipe format with the page cache
  * pipe buffer operations. Otherwise very similar to the regular pipe_writev().
  */
-static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages,
-			    int nr_pages, unsigned long len,
-			    unsigned int offset, unsigned int flags)
+static ssize_t move_to_pipe(struct pipe_inode_info *pipe,
+			    struct splice_pipe_desc *spd)
 {
-	int ret, do_wakeup, i;
+	int ret, do_wakeup, page_nr;
 
 	ret = 0;
 	do_wakeup = 0;
-	i = 0;
+	page_nr = 0;
 
 	if (pipe->inode)
 		mutex_lock(&pipe->inode->i_mutex);
@@ -171,27 +215,19 @@ static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages,
 		if (pipe->nrbufs < PIPE_BUFFERS) {
 			int newbuf = (pipe->curbuf + pipe->nrbufs) & (PIPE_BUFFERS - 1);
 			struct pipe_buffer *buf = pipe->bufs + newbuf;
-			struct page *page = pages[i++];
-			unsigned long this_len;
 
-			this_len = PAGE_CACHE_SIZE - offset;
-			if (this_len > len)
-				this_len = len;
-
-			buf->page = page;
-			buf->offset = offset;
-			buf->len = this_len;
-			buf->ops = &page_cache_pipe_buf_ops;
+			buf->page = spd->pages[page_nr];
+			buf->offset = spd->partial[page_nr].offset;
+			buf->len = spd->partial[page_nr].len;
+			buf->ops = spd->ops;
 			pipe->nrbufs++;
+			page_nr++;
+			ret += buf->len;
+
 			if (pipe->inode)
 				do_wakeup = 1;
 
-			ret += this_len;
-			len -= this_len;
-			offset = 0;
-			if (!--nr_pages)
-				break;
-			if (!len)
+			if (!--spd->nr_pages)
 				break;
 			if (pipe->nrbufs < PIPE_BUFFERS)
 				continue;
@@ -199,7 +235,7 @@ static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages,
 			break;
 		}
 
-		if (flags & SPLICE_F_NONBLOCK) {
+		if (spd->flags & SPLICE_F_NONBLOCK) {
 			if (!ret)
 				ret = -EAGAIN;
 			break;
@@ -234,8 +270,8 @@ static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages,
 		kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
 	}
 
-	while (i < nr_pages)
-		page_cache_release(pages[i++]);
+	while (page_nr < spd->nr_pages)
+		page_cache_release(spd->pages[page_nr++]);
 
 	return ret;
 }
@@ -246,17 +282,24 @@ __generic_file_splice_read(struct file *in, loff_t *ppos,
 			   unsigned int flags)
 {
 	struct address_space *mapping = in->f_mapping;
-	unsigned int loff, offset, nr_pages;
+	unsigned int loff, nr_pages;
 	struct page *pages[PIPE_BUFFERS];
+	struct partial_page partial[PIPE_BUFFERS];
 	struct page *page;
 	pgoff_t index, end_index;
 	loff_t isize;
-	size_t bytes;
-	int i, error;
+	size_t total_len;
+	int error;
+	struct splice_pipe_desc spd = {
+		.pages = pages,
+		.partial = partial,
+		.flags = flags,
+		.ops = &page_cache_pipe_buf_ops,
+	};
 
 	index = *ppos >> PAGE_CACHE_SHIFT;
-	loff = offset = *ppos & ~PAGE_CACHE_MASK;
-	nr_pages = (len + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+	loff = *ppos & ~PAGE_CACHE_MASK;
+	nr_pages = (len + loff + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
 
 	if (nr_pages > PIPE_BUFFERS)
 		nr_pages = PIPE_BUFFERS;
@@ -266,15 +309,15 @@ __generic_file_splice_read(struct file *in, loff_t *ppos,
 	 * read-ahead if this is a non-zero offset (we are likely doing small
 	 * chunk splice and the page is already there) for a single page.
 	 */
-	if (!offset || nr_pages > 1)
-		do_page_cache_readahead(mapping, in, index, nr_pages);
+	if (!loff || spd.nr_pages > 1)
+		do_page_cache_readahead(mapping, in, index, spd.nr_pages);
 
 	/*
 	 * Now fill in the holes:
 	 */
 	error = 0;
-	bytes = 0;
-	for (i = 0; i < nr_pages; i++, index++) {
+	total_len = 0;
+	for (spd.nr_pages = 0; spd.nr_pages < nr_pages; spd.nr_pages++, index++) {
 		unsigned int this_len;
 
 		if (!len)
@@ -367,26 +410,29 @@ readpage:
 			 */
 			if (end_index == index) {
 				loff = PAGE_CACHE_SIZE - (isize & ~PAGE_CACHE_MASK);
-				if (bytes + loff > isize) {
+				if (total_len + loff > isize) {
 					page_cache_release(page);
 					break;
 				}
 				/*
 				 * force quit after adding this page
 				 */
-				nr_pages = i;
+				nr_pages = spd.nr_pages;
 				this_len = min(this_len, loff);
+				loff = 0;
 			}
 		}
 fill_it:
-		pages[i] = page;
-		bytes += this_len;
+		pages[spd.nr_pages] = page;
+		partial[spd.nr_pages].offset = loff;
+		partial[spd.nr_pages].len = this_len;
 		len -= this_len;
+		total_len += this_len;
 		loff = 0;
 	}
 
-	if (i)
-		return move_to_pipe(pipe, pages, i, bytes, offset, flags);
+	if (spd.nr_pages)
+		return move_to_pipe(pipe, &spd);
 
 	return error;
 }
@@ -1018,6 +1064,174 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 	return -EINVAL;
 }
 
+/*
+ * Map an iov into an array of pages and offset/length tupples. With the
+ * partial_page structure, we can map several non-contiguous ranges into
+ * our ones pages[] map instead of splitting that operation into pieces.
+ * Could easily be exported as a generic helper for other users, in which
+ * case one would probably want to add a 'max_nr_pages' parameter as well.
+ */
+static int get_iovec_page_array(const struct iovec __user *iov,
+				unsigned int nr_vecs, struct page **pages,
+				struct partial_page *partial)
+{
+	int buffers = 0, error = 0;
+
+	/*
+	 * It's ok to take the mmap_sem for reading, even
+	 * across a "get_user()".
+	 */
+	down_read(&current->mm->mmap_sem);
+
+	while (nr_vecs) {
+		unsigned long off, npages;
+		void __user *base;
+		size_t len;
+		int i;
+
+		/*
+		 * Get user address base and length for this iovec.
+		 */
+		error = get_user(base, &iov->iov_base);
+		if (unlikely(error))
+			break;
+		error = get_user(len, &iov->iov_len);
+		if (unlikely(error))
+			break;
+
+		/*
+		 * Sanity check this iovec. 0 read succeeds.
+		 */
+		if (unlikely(!len))
+			break;
+		error = -EFAULT;
+		if (unlikely(!base))
+			break;
+
+		/*
+		 * Get this base offset and number of pages, then map
+		 * in the user pages.
+		 */
+		off = (unsigned long) base & ~PAGE_MASK;
+		npages = (off + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+		if (npages > PIPE_BUFFERS - buffers)
+			npages = PIPE_BUFFERS - buffers;
+
+		error = get_user_pages(current, current->mm,
+				       (unsigned long) base, npages, 0, 0,
+				       &pages[buffers], NULL);
+
+		if (unlikely(error <= 0))
+			break;
+
+		/*
+		 * Fill this contiguous range into the partial page map.
+		 */
+		for (i = 0; i < error; i++) {
+			const int plen = min_t(size_t, len, PAGE_SIZE) - off;
+
+			partial[buffers].offset = off;
+			partial[buffers].len = plen;
+
+			off = 0;
+			len -= plen;
+			buffers++;
+		}
+
+		/*
+		 * We didn't complete this iov, stop here since it probably
+		 * means we have to move some of this into a pipe to
+		 * be able to continue.
+		 */
+		if (len)
+			break;
+
+		/*
+		 * Don't continue if we mapped fewer pages than we asked for,
+		 * or if we mapped the max number of pages that we have
+		 * room for.
+		 */
+		if (error < npages || buffers == PIPE_BUFFERS)
+			break;
+
+		nr_vecs--;
+		iov++;
+	}
+
+	up_read(&current->mm->mmap_sem);
+
+	if (buffers)
+		return buffers;
+
+	return error;
+}
+
+/*
+ * vmsplice splices a user address range into a pipe. It can be thought of
+ * as splice-from-memory, where the regular splice is splice-from-file (or
+ * to file). In both cases the output is a pipe, naturally.
+ *
+ * Note that vmsplice only supports splicing _from_ user memory to a pipe,
+ * not the other way around. Splicing from user memory is a simple operation
+ * that can be supported without any funky alignment restrictions or nasty
+ * vm tricks. We simply map in the user memory and fill them into a pipe.
+ * The reverse isn't quite as easy, though. There are two possible solutions
+ * for that:
+ *
+ *	- memcpy() the data internally, at which point we might as well just
+ *	  do a regular read() on the buffer anyway.
+ *	- Lots of nasty vm tricks, that are neither fast nor flexible (it
+ *	  has restriction limitations on both ends of the pipe).
+ *
+ * Alas, it isn't here.
+ *
+ */
+static long do_vmsplice(struct file *file, const struct iovec __user *iov,
+			unsigned long nr_segs, unsigned int flags)
+{
+	struct pipe_inode_info *pipe = file->f_dentry->d_inode->i_pipe;
+	struct page *pages[PIPE_BUFFERS];
+	struct partial_page partial[PIPE_BUFFERS];
+	struct splice_pipe_desc spd = {
+		.pages = pages,
+		.partial = partial,
+		.flags = flags,
+		.ops = &user_page_pipe_buf_ops,
+	};
+
+	if (unlikely(!pipe))
+		return -EBADF;
+	if (unlikely(nr_segs > UIO_MAXIOV))
+		return -EINVAL;
+	else if (unlikely(!nr_segs))
+		return 0;
+
+	spd.nr_pages = get_iovec_page_array(iov, nr_segs, pages, partial);
+	if (spd.nr_pages <= 0)
+		return spd.nr_pages;
+
+	return move_to_pipe(pipe, &spd);
+}
+
+asmlinkage long sys_vmsplice(int fd, const struct iovec __user *iov,
+			     unsigned long nr_segs, unsigned int flags)
+{
+	struct file *file;
+	long error;
+	int fput;
+
+	error = -EBADF;
+	file = fget_light(fd, &fput);
+	if (file) {
+		if (file->f_mode & FMODE_WRITE)
+			error = do_vmsplice(file, iov, nr_segs, flags);
+
+		fput_light(file, fput);
+	}
+
+	return error;
+}
+
 asmlinkage long sys_splice(int fd_in, loff_t __user *off_in,
 			   int fd_out, loff_t __user *off_out,
 			   size_t len, unsigned int flags)
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index d81d6cfc1bb4..eb4b152c82fc 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -321,8 +321,9 @@
 #define __NR_splice		313
 #define __NR_sync_file_range	314
 #define __NR_tee		315
+#define __NR_vmsplice		316
 
-#define NR_syscalls 316
+#define NR_syscalls 317
 
 /*
  * user-visible error numbers are in the range -1 - -128: see
diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h
index a40ebec6aeeb..7107763168bf 100644
--- a/include/asm-ia64/unistd.h
+++ b/include/asm-ia64/unistd.h
@@ -290,12 +290,13 @@
 #define __NR_get_robust_list		1299
 #define __NR_sync_file_range		1300
 #define __NR_tee			1301
+#define __NR_vmsplice			1302
 
 #ifdef __KERNEL__
 
 #include <linux/config.h>
 
-#define NR_syscalls			278 /* length of syscall table */
+#define NR_syscalls			279 /* length of syscall table */
 
 #define __ARCH_WANT_SYS_RT_SIGACTION
 
diff --git a/include/asm-powerpc/unistd.h b/include/asm-powerpc/unistd.h
index c612f1a62772..34325e292596 100644
--- a/include/asm-powerpc/unistd.h
+++ b/include/asm-powerpc/unistd.h
@@ -303,8 +303,9 @@
 #define __NR_unshare		282
 #define __NR_splice		283
 #define __NR_tee		284
+#define __NR_vmsplice		285
 
-#define __NR_syscalls		285
+#define __NR_syscalls		286
 
 #ifdef __KERNEL__
 #define __NR__exit __NR_exit
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index 98c36eae567c..feb77cb8c044 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -615,8 +615,10 @@ __SYSCALL(__NR_splice, sys_splice)
 __SYSCALL(__NR_tee, sys_tee)
 #define __NR_sync_file_range	277
 __SYSCALL(__NR_sync_file_range, sys_sync_file_range)
+#define __NR_vmsplice		278
+__SYSCALL(__NR_vmsplice, sys_vmsplice)
 
-#define __NR_syscall_max __NR_sync_file_range
+#define __NR_syscall_max __NR_vmsplice
 
 #ifndef __NO_STUBS
 
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index d3ebc0e68b2b..3996960fc565 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -574,6 +574,9 @@ asmlinkage long sys_splice(int fd_in, loff_t __user *off_in,
 			   int fd_out, loff_t __user *off_out,
 			   size_t len, unsigned int flags);
 
+asmlinkage long sys_vmsplice(int fd, const struct iovec __user *iov,
+			     unsigned long nr_segs, unsigned int flags);
+
 asmlinkage long sys_tee(int fdin, int fdout, size_t len, unsigned int flags);
 
 asmlinkage long sys_sync_file_range(int fd, loff_t offset, loff_t nbytes,
-- 
cgit v1.2.3


From 734cbc363b159caee158d5a83408c72d98bcacf0 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Tue, 25 Apr 2006 10:58:50 -0700
Subject: [PATCH] sky2: reschedule if irq still pending

This is a workaround for the case edge-triggered irq's. Several users
seem to have broken configurations sharing edge-triggered irq's. To avoid
losing IRQ's, reshedule if more work arrives.

The changes to netdevice.h are to extract the part that puts device
back in list into separate inline.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/sky2.c        | 20 ++++++++++++++++----
 include/linux/netdevice.h | 18 ++++++++++--------
 2 files changed, 26 insertions(+), 12 deletions(-)

(limited to 'include')

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 67b0eab16589..618fde8622ca 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -2093,6 +2093,7 @@ static int sky2_poll(struct net_device *dev0, int *budget)
 	int work_done = 0;
 	u32 status = sky2_read32(hw, B0_Y2_SP_EISR);
 
+ restart_poll:
 	if (unlikely(status & ~Y2_IS_STAT_BMU)) {
 		if (status & Y2_IS_HW_ERR)
 			sky2_hw_intr(hw);
@@ -2123,7 +2124,7 @@ static int sky2_poll(struct net_device *dev0, int *budget)
 	}
 
 	if (status & Y2_IS_STAT_BMU) {
-		work_done = sky2_status_intr(hw, work_limit);
+		work_done += sky2_status_intr(hw, work_limit - work_done);
 		*budget -= work_done;
 		dev0->quota -= work_done;
 
@@ -2133,9 +2134,22 @@ static int sky2_poll(struct net_device *dev0, int *budget)
 		sky2_write32(hw, STAT_CTRL, SC_STAT_CLR_IRQ);
 	}
 
-	netif_rx_complete(dev0);
+	local_irq_disable();
+	__netif_rx_complete(dev0);
 
 	status = sky2_read32(hw, B0_Y2_SP_LISR);
+
+	if (unlikely(status)) {
+		/* More work pending, try and keep going */
+		if (__netif_rx_schedule_prep(dev0)) {
+			__netif_rx_reschedule(dev0, work_done);
+			status = sky2_read32(hw, B0_Y2_SP_EISR);
+			local_irq_enable();
+			goto restart_poll;
+		}
+	}
+
+	local_irq_enable();
 	return 0;
 }
 
@@ -2153,8 +2167,6 @@ static irqreturn_t sky2_intr(int irq, void *dev_id, struct pt_regs *regs)
 	prefetch(&hw->st_le[hw->st_idx]);
 	if (likely(__netif_rx_schedule_prep(dev0)))
 		__netif_rx_schedule(dev0);
-	else
-		printk(KERN_DEBUG PFX "irq race detected\n");
 
 	return IRQ_HANDLED;
 }
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 40ccf8cc4239..01db7b88a2b1 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -829,19 +829,21 @@ static inline void netif_rx_schedule(struct net_device *dev)
 		__netif_rx_schedule(dev);
 }
 
-/* Try to reschedule poll. Called by dev->poll() after netif_rx_complete().
- * Do not inline this?
- */
+
+static inline void  __netif_rx_reschedule(struct net_device *dev, int undo)
+{
+	dev->quota += undo;
+	list_add_tail(&dev->poll_list, &__get_cpu_var(softnet_data).poll_list);
+	__raise_softirq_irqoff(NET_RX_SOFTIRQ);
+}
+
+/* Try to reschedule poll. Called by dev->poll() after netif_rx_complete(). */
 static inline int netif_rx_reschedule(struct net_device *dev, int undo)
 {
 	if (netif_rx_schedule_prep(dev)) {
 		unsigned long flags;
-
-		dev->quota += undo;
-
 		local_irq_save(flags);
-		list_add_tail(&dev->poll_list, &__get_cpu_var(softnet_data).poll_list);
-		__raise_softirq_irqoff(NET_RX_SOFTIRQ);
+		__netif_rx_reschedule(dev, undo);
 		local_irq_restore(flags);
 		return 1;
 	}
-- 
cgit v1.2.3


From 00522fb41a2a9bf0f98a007c0e2b516a3873148c Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Wed, 26 Apr 2006 14:39:29 +0200
Subject: [PATCH] splice: rearrange moving to/from pipe helpers

We need these for people writing their own ->splice_read/write hooks.

Signed-off-by: Jens Axboe <axboe@suse.de>
---
 fs/splice.c               | 35 +++++++++++------------------------
 include/linux/pipe_fs_i.h | 17 +++++++++++++++++
 2 files changed, 28 insertions(+), 24 deletions(-)

(limited to 'include')

diff --git a/fs/splice.c b/fs/splice.c
index 0b2c1f060cae..447ebc0a37f3 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -29,23 +29,13 @@
 #include <linux/syscalls.h>
 #include <linux/uio.h>
 
-/*
- * Passed to the actors
- */
-struct splice_desc {
-	unsigned int len, total_len;	/* current and remaining length */
-	unsigned int flags;		/* splice flags */
-	struct file *file;		/* file to read/write */
-	loff_t pos;			/* file position */
-};
-
 struct partial_page {
 	unsigned int offset;
 	unsigned int len;
 };
 
 /*
- * Passed to move_to_pipe
+ * Passed to splice_to_pipe
  */
 struct splice_pipe_desc {
 	struct page **pages;		/* page map */
@@ -192,8 +182,8 @@ static struct pipe_buf_operations user_page_pipe_buf_ops = {
  * Pipe output worker. This sets up our pipe format with the page cache
  * pipe buffer operations. Otherwise very similar to the regular pipe_writev().
  */
-static ssize_t move_to_pipe(struct pipe_inode_info *pipe,
-			    struct splice_pipe_desc *spd)
+static ssize_t splice_to_pipe(struct pipe_inode_info *pipe,
+			      struct splice_pipe_desc *spd)
 {
 	int ret, do_wakeup, page_nr;
 
@@ -432,7 +422,7 @@ fill_it:
 	}
 
 	if (spd.nr_pages)
-		return move_to_pipe(pipe, &spd);
+		return splice_to_pipe(pipe, &spd);
 
 	return error;
 }
@@ -666,17 +656,14 @@ out_nomem:
 	return ret;
 }
 
-typedef int (splice_actor)(struct pipe_inode_info *, struct pipe_buffer *,
-			   struct splice_desc *);
-
 /*
  * Pipe input worker. Most of this logic works like a regular pipe, the
  * key here is the 'actor' worker passed in that actually moves the data
  * to the wanted destination. See pipe_to_file/pipe_to_sendpage above.
  */
-static ssize_t move_from_pipe(struct pipe_inode_info *pipe, struct file *out,
-			      loff_t *ppos, size_t len, unsigned int flags,
-			      splice_actor *actor)
+ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out,
+			 loff_t *ppos, size_t len, unsigned int flags,
+			 splice_actor *actor)
 {
 	int ret, do_wakeup, err;
 	struct splice_desc sd;
@@ -795,7 +782,7 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
 	struct address_space *mapping = out->f_mapping;
 	ssize_t ret;
 
-	ret = move_from_pipe(pipe, out, ppos, len, flags, pipe_to_file);
+	ret = splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file);
 	if (ret > 0) {
 		struct inode *inode = mapping->host;
 
@@ -837,7 +824,7 @@ EXPORT_SYMBOL(generic_file_splice_write);
 ssize_t generic_splice_sendpage(struct pipe_inode_info *pipe, struct file *out,
 				loff_t *ppos, size_t len, unsigned int flags)
 {
-	return move_from_pipe(pipe, out, ppos, len, flags, pipe_to_sendpage);
+	return splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_sendpage);
 }
 
 EXPORT_SYMBOL(generic_splice_sendpage);
@@ -924,7 +911,7 @@ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out,
 
 		/*
 		 * We don't have an immediate reader, but we'll read the stuff
-		 * out of the pipe right after the move_to_pipe(). So set
+		 * out of the pipe right after the splice_to_pipe(). So set
 		 * PIPE_READERS appropriately.
 		 */
 		pipe->readers = 1;
@@ -1210,7 +1197,7 @@ static long do_vmsplice(struct file *file, const struct iovec __user *iov,
 	if (spd.nr_pages <= 0)
 		return spd.nr_pages;
 
-	return move_to_pipe(pipe, &spd);
+	return splice_to_pipe(pipe, &spd);
 }
 
 asmlinkage long sys_vmsplice(int fd, const struct iovec __user *iov,
diff --git a/include/linux/pipe_fs_i.h b/include/linux/pipe_fs_i.h
index ef7f33c0be19..0008d4bd4059 100644
--- a/include/linux/pipe_fs_i.h
+++ b/include/linux/pipe_fs_i.h
@@ -61,4 +61,21 @@ void __free_pipe_info(struct pipe_inode_info *);
 				 /* from/to, of course */
 #define SPLICE_F_MORE	(0x04)	/* expect more data */
 
+/*
+ * Passed to the actors
+ */
+struct splice_desc {
+	unsigned int len, total_len;	/* current and remaining length */
+	unsigned int flags;		/* splice flags */
+	struct file *file;		/* file to read/write */
+	loff_t pos;			/* file position */
+};
+
+typedef int (splice_actor)(struct pipe_inode_info *, struct pipe_buffer *,
+			   struct splice_desc *);
+
+extern ssize_t splice_from_pipe(struct pipe_inode_info *, struct file *,
+				loff_t *, size_t, unsigned int,
+				splice_actor *);
+
 #endif
-- 
cgit v1.2.3