From 1813a68457bb45b378d5bbec615b167deff3bcfc Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Tue, 20 Jul 2010 15:22:54 -0700
Subject: x86: Move alloc_desk_mask variables inside ifdef

They are only useful with CONFIG_CPUMASK_OFFSTACK
Avoids hundreds of warnings with a gcc 4.6 -Wall build.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/irq.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index c03243ad84b4..fff1d77c3753 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -439,12 +439,12 @@ extern int set_irq_msi(unsigned int irq, struct msi_desc *entry);
 static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
 							bool boot)
 {
+#ifdef CONFIG_CPUMASK_OFFSTACK
 	gfp_t gfp = GFP_ATOMIC;
 
 	if (boot)
 		gfp = GFP_NOWAIT;
 
-#ifdef CONFIG_CPUMASK_OFFSTACK
 	if (!alloc_cpumask_var_node(&desc->affinity, gfp, node))
 		return false;
 
-- 
cgit v1.2.3


From e3239ff92a17976ac5d26fa0fe40ef3a9daf2523 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 4 Aug 2010 14:06:41 +1000
Subject: memblock: Rename memblock_region to memblock_type and
 memblock_property to memblock_region

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/arm/mm/init.c                       |   2 +-
 arch/arm/plat-omap/fb.c                  |   2 +-
 arch/microblaze/mm/init.c                |   4 +-
 arch/powerpc/mm/hash_utils_64.c          |   2 +-
 arch/powerpc/mm/mem.c                    |  26 ++---
 arch/powerpc/platforms/embedded6xx/wii.c |   2 +-
 arch/sparc/mm/init_64.c                  |   6 +-
 drivers/video/omap2/vram.c               |   2 +-
 include/linux/memblock.h                 |  24 ++---
 mm/memblock.c                            | 168 +++++++++++++++----------------
 10 files changed, 118 insertions(+), 120 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 7185b00650fe..d1496e65dc2d 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -237,7 +237,7 @@ static void __init arm_bootmem_free(struct meminfo *mi, unsigned long min,
 #ifndef CONFIG_SPARSEMEM
 int pfn_valid(unsigned long pfn)
 {
-	struct memblock_region *mem = &memblock.memory;
+	struct memblock_type *mem = &memblock.memory;
 	unsigned int left = 0, right = mem->cnt;
 
 	do {
diff --git a/arch/arm/plat-omap/fb.c b/arch/arm/plat-omap/fb.c
index 0054b9501a53..05bf22827404 100644
--- a/arch/arm/plat-omap/fb.c
+++ b/arch/arm/plat-omap/fb.c
@@ -173,7 +173,7 @@ static int check_fbmem_region(int region_idx, struct omapfb_mem_region *rg,
 
 static int valid_sdram(unsigned long addr, unsigned long size)
 {
-	struct memblock_property res;
+	struct memblock_region res;
 
 	res.base = addr;
 	res.size = size;
diff --git a/arch/microblaze/mm/init.c b/arch/microblaze/mm/init.c
index db5934989926..afd6494fbbc6 100644
--- a/arch/microblaze/mm/init.c
+++ b/arch/microblaze/mm/init.c
@@ -77,8 +77,8 @@ void __init setup_memory(void)
 
 	/* Find main memory where is the kernel */
 	for (i = 0; i < memblock.memory.cnt; i++) {
-		memory_start = (u32) memblock.memory.region[i].base;
-		memory_end = (u32) memblock.memory.region[i].base
+		memory_start = (u32) memblock.memory.regions[i].base;
+		memory_end = (u32) memblock.memory.regions[i].base
 				+ (u32) memblock.memory.region[i].size;
 		if ((memory_start <= (u32)_text) &&
 					((u32)_text <= memory_end)) {
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index 09dffe6efa46..b1a3784744db 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -660,7 +660,7 @@ static void __init htab_initialize(void)
 
 	/* create bolted the linear mapping in the hash table */
 	for (i=0; i < memblock.memory.cnt; i++) {
-		base = (unsigned long)__va(memblock.memory.region[i].base);
+		base = (unsigned long)__va(memblock.memory.regions[i].base);
 		size = memblock.memory.region[i].size;
 
 		DBG("creating mapping for region: %lx..%lx (prot: %lx)\n",
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 1a84a8d00005..a33f5c186fb7 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -86,10 +86,10 @@ int page_is_ram(unsigned long pfn)
 	for (i=0; i < memblock.memory.cnt; i++) {
 		unsigned long base;
 
-		base = memblock.memory.region[i].base;
+		base = memblock.memory.regions[i].base;
 
 		if ((paddr >= base) &&
-			(paddr < (base + memblock.memory.region[i].size))) {
+			(paddr < (base + memblock.memory.regions[i].size))) {
 			return 1;
 		}
 	}
@@ -149,7 +149,7 @@ int
 walk_system_ram_range(unsigned long start_pfn, unsigned long nr_pages,
 		void *arg, int (*func)(unsigned long, unsigned long, void *))
 {
-	struct memblock_property res;
+	struct memblock_region res;
 	unsigned long pfn, len;
 	u64 end;
 	int ret = -1;
@@ -206,7 +206,7 @@ void __init do_init_bootmem(void)
 	/* Add active regions with valid PFNs */
 	for (i = 0; i < memblock.memory.cnt; i++) {
 		unsigned long start_pfn, end_pfn;
-		start_pfn = memblock.memory.region[i].base >> PAGE_SHIFT;
+		start_pfn = memblock.memory.regions[i].base >> PAGE_SHIFT;
 		end_pfn = start_pfn + memblock_size_pages(&memblock.memory, i);
 		add_active_range(0, start_pfn, end_pfn);
 	}
@@ -219,16 +219,16 @@ void __init do_init_bootmem(void)
 
 	/* reserve the sections we're already using */
 	for (i = 0; i < memblock.reserved.cnt; i++) {
-		unsigned long addr = memblock.reserved.region[i].base +
+		unsigned long addr = memblock.reserved.regions[i].base +
 				     memblock_size_bytes(&memblock.reserved, i) - 1;
 		if (addr < lowmem_end_addr)
-			reserve_bootmem(memblock.reserved.region[i].base,
+			reserve_bootmem(memblock.reserved.regions[i].base,
 					memblock_size_bytes(&memblock.reserved, i),
 					BOOTMEM_DEFAULT);
-		else if (memblock.reserved.region[i].base < lowmem_end_addr) {
+		else if (memblock.reserved.regions[i].base < lowmem_end_addr) {
 			unsigned long adjusted_size = lowmem_end_addr -
-				      memblock.reserved.region[i].base;
-			reserve_bootmem(memblock.reserved.region[i].base,
+				      memblock.reserved.regions[i].base;
+			reserve_bootmem(memblock.reserved.regions[i].base,
 					adjusted_size, BOOTMEM_DEFAULT);
 		}
 	}
@@ -237,7 +237,7 @@ void __init do_init_bootmem(void)
 
 	/* reserve the sections we're already using */
 	for (i = 0; i < memblock.reserved.cnt; i++)
-		reserve_bootmem(memblock.reserved.region[i].base,
+		reserve_bootmem(memblock.reserved.regions[i].base,
 				memblock_size_bytes(&memblock.reserved, i),
 				BOOTMEM_DEFAULT);
 
@@ -257,10 +257,10 @@ static int __init mark_nonram_nosave(void)
 
 	for (i = 0; i < memblock.memory.cnt - 1; i++) {
 		memblock_region_max_pfn =
-			(memblock.memory.region[i].base >> PAGE_SHIFT) +
-			(memblock.memory.region[i].size >> PAGE_SHIFT);
+			(memblock.memory.regions[i].base >> PAGE_SHIFT) +
+			(memblock.memory.regions[i].size >> PAGE_SHIFT);
 		memblock_next_region_start_pfn =
-			memblock.memory.region[i+1].base >> PAGE_SHIFT;
+			memblock.memory.regions[i+1].base >> PAGE_SHIFT;
 
 		if (memblock_region_max_pfn < memblock_next_region_start_pfn)
 			register_nosave_region(memblock_region_max_pfn,
diff --git a/arch/powerpc/platforms/embedded6xx/wii.c b/arch/powerpc/platforms/embedded6xx/wii.c
index 5cdcc7c8d973..8450c29e9b2f 100644
--- a/arch/powerpc/platforms/embedded6xx/wii.c
+++ b/arch/powerpc/platforms/embedded6xx/wii.c
@@ -65,7 +65,7 @@ static int __init page_aligned(unsigned long x)
 
 void __init wii_memory_fixups(void)
 {
-	struct memblock_property *p = memblock.memory.region;
+	struct memblock_region *p = memblock.memory.region;
 
 	/*
 	 * This is part of a workaround to allow the use of two
diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index f0434513df15..16d8bee889ba 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -978,7 +978,7 @@ static void __init add_node_ranges(void)
 		unsigned long size = memblock_size_bytes(&memblock.memory, i);
 		unsigned long start, end;
 
-		start = memblock.memory.region[i].base;
+		start = memblock.memory.regions[i].base;
 		end = start + size;
 		while (start < end) {
 			unsigned long this_end;
@@ -1299,7 +1299,7 @@ static void __init bootmem_init_nonnuma(void)
 		if (!size)
 			continue;
 
-		start_pfn = memblock.memory.region[i].base >> PAGE_SHIFT;
+		start_pfn = memblock.memory.regions[i].base >> PAGE_SHIFT;
 		end_pfn = start_pfn + memblock_size_pages(&memblock.memory, i);
 		add_active_range(0, start_pfn, end_pfn);
 	}
@@ -1339,7 +1339,7 @@ static void __init trim_reserved_in_node(int nid)
 	numadbg("  trim_reserved_in_node(%d)\n", nid);
 
 	for (i = 0; i < memblock.reserved.cnt; i++) {
-		unsigned long start = memblock.reserved.region[i].base;
+		unsigned long start = memblock.reserved.regions[i].base;
 		unsigned long size = memblock_size_bytes(&memblock.reserved, i);
 		unsigned long end = start + size;
 
diff --git a/drivers/video/omap2/vram.c b/drivers/video/omap2/vram.c
index f6fdc2085f3e..0f2532bf0f04 100644
--- a/drivers/video/omap2/vram.c
+++ b/drivers/video/omap2/vram.c
@@ -554,7 +554,7 @@ void __init omap_vram_reserve_sdram_memblock(void)
 	size = PAGE_ALIGN(size);
 
 	if (paddr) {
-		struct memblock_property res;
+		struct memblock_region res;
 
 		res.base = paddr;
 		res.size = size;
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index a59faf2b5edd..86e7daf742f2 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -18,22 +18,22 @@
 
 #define MAX_MEMBLOCK_REGIONS 128
 
-struct memblock_property {
+struct memblock_region {
 	u64 base;
 	u64 size;
 };
 
-struct memblock_region {
+struct memblock_type {
 	unsigned long cnt;
 	u64 size;
-	struct memblock_property region[MAX_MEMBLOCK_REGIONS+1];
+	struct memblock_region regions[MAX_MEMBLOCK_REGIONS+1];
 };
 
 struct memblock {
 	unsigned long debug;
 	u64 rmo_size;
-	struct memblock_region memory;
-	struct memblock_region reserved;
+	struct memblock_type memory;
+	struct memblock_type reserved;
 };
 
 extern struct memblock memblock;
@@ -56,27 +56,27 @@ extern u64 memblock_end_of_DRAM(void);
 extern void __init memblock_enforce_memory_limit(u64 memory_limit);
 extern int __init memblock_is_reserved(u64 addr);
 extern int memblock_is_region_reserved(u64 base, u64 size);
-extern int memblock_find(struct memblock_property *res);
+extern int memblock_find(struct memblock_region *res);
 
 extern void memblock_dump_all(void);
 
 static inline u64
-memblock_size_bytes(struct memblock_region *type, unsigned long region_nr)
+memblock_size_bytes(struct memblock_type *type, unsigned long region_nr)
 {
-	return type->region[region_nr].size;
+	return type->regions[region_nr].size;
 }
 static inline u64
-memblock_size_pages(struct memblock_region *type, unsigned long region_nr)
+memblock_size_pages(struct memblock_type *type, unsigned long region_nr)
 {
 	return memblock_size_bytes(type, region_nr) >> PAGE_SHIFT;
 }
 static inline u64
-memblock_start_pfn(struct memblock_region *type, unsigned long region_nr)
+memblock_start_pfn(struct memblock_type *type, unsigned long region_nr)
 {
-	return type->region[region_nr].base >> PAGE_SHIFT;
+	return type->regions[region_nr].base >> PAGE_SHIFT;
 }
 static inline u64
-memblock_end_pfn(struct memblock_region *type, unsigned long region_nr)
+memblock_end_pfn(struct memblock_type *type, unsigned long region_nr)
 {
 	return memblock_start_pfn(type, region_nr) +
 	       memblock_size_pages(type, region_nr);
diff --git a/mm/memblock.c b/mm/memblock.c
index 43840b305ecb..6f407ccf604e 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -29,7 +29,7 @@ static int __init early_memblock(char *p)
 }
 early_param("memblock", early_memblock);
 
-static void memblock_dump(struct memblock_region *region, char *name)
+static void memblock_dump(struct memblock_type *region, char *name)
 {
 	unsigned long long base, size;
 	int i;
@@ -37,8 +37,8 @@ static void memblock_dump(struct memblock_region *region, char *name)
 	pr_info(" %s.cnt  = 0x%lx\n", name, region->cnt);
 
 	for (i = 0; i < region->cnt; i++) {
-		base = region->region[i].base;
-		size = region->region[i].size;
+		base = region->regions[i].base;
+		size = region->regions[i].size;
 
 		pr_info(" %s[0x%x]\t0x%016llx - 0x%016llx, 0x%llx bytes\n",
 		    name, i, base, base + size - 1, size);
@@ -74,34 +74,34 @@ static long memblock_addrs_adjacent(u64 base1, u64 size1, u64 base2, u64 size2)
 	return 0;
 }
 
-static long memblock_regions_adjacent(struct memblock_region *rgn,
+static long memblock_regions_adjacent(struct memblock_type *type,
 		unsigned long r1, unsigned long r2)
 {
-	u64 base1 = rgn->region[r1].base;
-	u64 size1 = rgn->region[r1].size;
-	u64 base2 = rgn->region[r2].base;
-	u64 size2 = rgn->region[r2].size;
+	u64 base1 = type->regions[r1].base;
+	u64 size1 = type->regions[r1].size;
+	u64 base2 = type->regions[r2].base;
+	u64 size2 = type->regions[r2].size;
 
 	return memblock_addrs_adjacent(base1, size1, base2, size2);
 }
 
-static void memblock_remove_region(struct memblock_region *rgn, unsigned long r)
+static void memblock_remove_region(struct memblock_type *type, unsigned long r)
 {
 	unsigned long i;
 
-	for (i = r; i < rgn->cnt - 1; i++) {
-		rgn->region[i].base = rgn->region[i + 1].base;
-		rgn->region[i].size = rgn->region[i + 1].size;
+	for (i = r; i < type->cnt - 1; i++) {
+		type->regions[i].base = type->regions[i + 1].base;
+		type->regions[i].size = type->regions[i + 1].size;
 	}
-	rgn->cnt--;
+	type->cnt--;
 }
 
 /* Assumption: base addr of region 1 < base addr of region 2 */
-static void memblock_coalesce_regions(struct memblock_region *rgn,
+static void memblock_coalesce_regions(struct memblock_type *type,
 		unsigned long r1, unsigned long r2)
 {
-	rgn->region[r1].size += rgn->region[r2].size;
-	memblock_remove_region(rgn, r2);
+	type->regions[r1].size += type->regions[r2].size;
+	memblock_remove_region(type, r2);
 }
 
 void __init memblock_init(void)
@@ -109,13 +109,13 @@ void __init memblock_init(void)
 	/* Create a dummy zero size MEMBLOCK which will get coalesced away later.
 	 * This simplifies the memblock_add() code below...
 	 */
-	memblock.memory.region[0].base = 0;
-	memblock.memory.region[0].size = 0;
+	memblock.memory.regions[0].base = 0;
+	memblock.memory.regions[0].size = 0;
 	memblock.memory.cnt = 1;
 
 	/* Ditto. */
-	memblock.reserved.region[0].base = 0;
-	memblock.reserved.region[0].size = 0;
+	memblock.reserved.regions[0].base = 0;
+	memblock.reserved.regions[0].size = 0;
 	memblock.reserved.cnt = 1;
 }
 
@@ -126,24 +126,24 @@ void __init memblock_analyze(void)
 	memblock.memory.size = 0;
 
 	for (i = 0; i < memblock.memory.cnt; i++)
-		memblock.memory.size += memblock.memory.region[i].size;
+		memblock.memory.size += memblock.memory.regions[i].size;
 }
 
-static long memblock_add_region(struct memblock_region *rgn, u64 base, u64 size)
+static long memblock_add_region(struct memblock_type *type, u64 base, u64 size)
 {
 	unsigned long coalesced = 0;
 	long adjacent, i;
 
-	if ((rgn->cnt == 1) && (rgn->region[0].size == 0)) {
-		rgn->region[0].base = base;
-		rgn->region[0].size = size;
+	if ((type->cnt == 1) && (type->regions[0].size == 0)) {
+		type->regions[0].base = base;
+		type->regions[0].size = size;
 		return 0;
 	}
 
 	/* First try and coalesce this MEMBLOCK with another. */
-	for (i = 0; i < rgn->cnt; i++) {
-		u64 rgnbase = rgn->region[i].base;
-		u64 rgnsize = rgn->region[i].size;
+	for (i = 0; i < type->cnt; i++) {
+		u64 rgnbase = type->regions[i].base;
+		u64 rgnsize = type->regions[i].size;
 
 		if ((rgnbase == base) && (rgnsize == size))
 			/* Already have this region, so we're done */
@@ -151,61 +151,59 @@ static long memblock_add_region(struct memblock_region *rgn, u64 base, u64 size)
 
 		adjacent = memblock_addrs_adjacent(base, size, rgnbase, rgnsize);
 		if (adjacent > 0) {
-			rgn->region[i].base -= size;
-			rgn->region[i].size += size;
+			type->regions[i].base -= size;
+			type->regions[i].size += size;
 			coalesced++;
 			break;
 		} else if (adjacent < 0) {
-			rgn->region[i].size += size;
+			type->regions[i].size += size;
 			coalesced++;
 			break;
 		}
 	}
 
-	if ((i < rgn->cnt - 1) && memblock_regions_adjacent(rgn, i, i+1)) {
-		memblock_coalesce_regions(rgn, i, i+1);
+	if ((i < type->cnt - 1) && memblock_regions_adjacent(type, i, i+1)) {
+		memblock_coalesce_regions(type, i, i+1);
 		coalesced++;
 	}
 
 	if (coalesced)
 		return coalesced;
-	if (rgn->cnt >= MAX_MEMBLOCK_REGIONS)
+	if (type->cnt >= MAX_MEMBLOCK_REGIONS)
 		return -1;
 
 	/* Couldn't coalesce the MEMBLOCK, so add it to the sorted table. */
-	for (i = rgn->cnt - 1; i >= 0; i--) {
-		if (base < rgn->region[i].base) {
-			rgn->region[i+1].base = rgn->region[i].base;
-			rgn->region[i+1].size = rgn->region[i].size;
+	for (i = type->cnt - 1; i >= 0; i--) {
+		if (base < type->regions[i].base) {
+			type->regions[i+1].base = type->regions[i].base;
+			type->regions[i+1].size = type->regions[i].size;
 		} else {
-			rgn->region[i+1].base = base;
-			rgn->region[i+1].size = size;
+			type->regions[i+1].base = base;
+			type->regions[i+1].size = size;
 			break;
 		}
 	}
 
-	if (base < rgn->region[0].base) {
-		rgn->region[0].base = base;
-		rgn->region[0].size = size;
+	if (base < type->regions[0].base) {
+		type->regions[0].base = base;
+		type->regions[0].size = size;
 	}
-	rgn->cnt++;
+	type->cnt++;
 
 	return 0;
 }
 
 long memblock_add(u64 base, u64 size)
 {
-	struct memblock_region *_rgn = &memblock.memory;
-
 	/* On pSeries LPAR systems, the first MEMBLOCK is our RMO region. */
 	if (base == 0)
 		memblock.rmo_size = size;
 
-	return memblock_add_region(_rgn, base, size);
+	return memblock_add_region(&memblock.memory, base, size);
 
 }
 
-static long __memblock_remove(struct memblock_region *rgn, u64 base, u64 size)
+static long __memblock_remove(struct memblock_type *type, u64 base, u64 size)
 {
 	u64 rgnbegin, rgnend;
 	u64 end = base + size;
@@ -214,34 +212,34 @@ static long __memblock_remove(struct memblock_region *rgn, u64 base, u64 size)
 	rgnbegin = rgnend = 0; /* supress gcc warnings */
 
 	/* Find the region where (base, size) belongs to */
-	for (i=0; i < rgn->cnt; i++) {
-		rgnbegin = rgn->region[i].base;
-		rgnend = rgnbegin + rgn->region[i].size;
+	for (i=0; i < type->cnt; i++) {
+		rgnbegin = type->regions[i].base;
+		rgnend = rgnbegin + type->regions[i].size;
 
 		if ((rgnbegin <= base) && (end <= rgnend))
 			break;
 	}
 
 	/* Didn't find the region */
-	if (i == rgn->cnt)
+	if (i == type->cnt)
 		return -1;
 
 	/* Check to see if we are removing entire region */
 	if ((rgnbegin == base) && (rgnend == end)) {
-		memblock_remove_region(rgn, i);
+		memblock_remove_region(type, i);
 		return 0;
 	}
 
 	/* Check to see if region is matching at the front */
 	if (rgnbegin == base) {
-		rgn->region[i].base = end;
-		rgn->region[i].size -= size;
+		type->regions[i].base = end;
+		type->regions[i].size -= size;
 		return 0;
 	}
 
 	/* Check to see if the region is matching at the end */
 	if (rgnend == end) {
-		rgn->region[i].size -= size;
+		type->regions[i].size -= size;
 		return 0;
 	}
 
@@ -249,8 +247,8 @@ static long __memblock_remove(struct memblock_region *rgn, u64 base, u64 size)
 	 * We need to split the entry -  adjust the current one to the
 	 * beginging of the hole and add the region after hole.
 	 */
-	rgn->region[i].size = base - rgn->region[i].base;
-	return memblock_add_region(rgn, end, rgnend - end);
+	type->regions[i].size = base - type->regions[i].base;
+	return memblock_add_region(type, end, rgnend - end);
 }
 
 long memblock_remove(u64 base, u64 size)
@@ -265,25 +263,25 @@ long __init memblock_free(u64 base, u64 size)
 
 long __init memblock_reserve(u64 base, u64 size)
 {
-	struct memblock_region *_rgn = &memblock.reserved;
+	struct memblock_type *_rgn = &memblock.reserved;
 
 	BUG_ON(0 == size);
 
 	return memblock_add_region(_rgn, base, size);
 }
 
-long memblock_overlaps_region(struct memblock_region *rgn, u64 base, u64 size)
+long memblock_overlaps_region(struct memblock_type *type, u64 base, u64 size)
 {
 	unsigned long i;
 
-	for (i = 0; i < rgn->cnt; i++) {
-		u64 rgnbase = rgn->region[i].base;
-		u64 rgnsize = rgn->region[i].size;
+	for (i = 0; i < type->cnt; i++) {
+		u64 rgnbase = type->regions[i].base;
+		u64 rgnsize = type->regions[i].size;
 		if (memblock_addrs_overlap(base, size, rgnbase, rgnsize))
 			break;
 	}
 
-	return (i < rgn->cnt) ? i : -1;
+	return (i < type->cnt) ? i : -1;
 }
 
 static u64 memblock_align_down(u64 addr, u64 size)
@@ -311,7 +309,7 @@ static u64 __init memblock_alloc_nid_unreserved(u64 start, u64 end,
 				base = ~(u64)0;
 			return base;
 		}
-		res_base = memblock.reserved.region[j].base;
+		res_base = memblock.reserved.regions[j].base;
 		if (res_base < size)
 			break;
 		base = memblock_align_down(res_base - size, align);
@@ -320,7 +318,7 @@ static u64 __init memblock_alloc_nid_unreserved(u64 start, u64 end,
 	return ~(u64)0;
 }
 
-static u64 __init memblock_alloc_nid_region(struct memblock_property *mp,
+static u64 __init memblock_alloc_nid_region(struct memblock_region *mp,
 				       u64 (*nid_range)(u64, u64, int *),
 				       u64 size, u64 align, int nid)
 {
@@ -350,7 +348,7 @@ static u64 __init memblock_alloc_nid_region(struct memblock_property *mp,
 u64 __init memblock_alloc_nid(u64 size, u64 align, int nid,
 			 u64 (*nid_range)(u64 start, u64 end, int *nid))
 {
-	struct memblock_region *mem = &memblock.memory;
+	struct memblock_type *mem = &memblock.memory;
 	int i;
 
 	BUG_ON(0 == size);
@@ -358,7 +356,7 @@ u64 __init memblock_alloc_nid(u64 size, u64 align, int nid,
 	size = memblock_align_up(size, align);
 
 	for (i = 0; i < mem->cnt; i++) {
-		u64 ret = memblock_alloc_nid_region(&mem->region[i],
+		u64 ret = memblock_alloc_nid_region(&mem->regions[i],
 					       nid_range,
 					       size, align, nid);
 		if (ret != ~(u64)0)
@@ -402,8 +400,8 @@ u64 __init __memblock_alloc_base(u64 size, u64 align, u64 max_addr)
 		max_addr = MEMBLOCK_REAL_LIMIT;
 
 	for (i = memblock.memory.cnt - 1; i >= 0; i--) {
-		u64 memblockbase = memblock.memory.region[i].base;
-		u64 memblocksize = memblock.memory.region[i].size;
+		u64 memblockbase = memblock.memory.regions[i].base;
+		u64 memblocksize = memblock.memory.regions[i].size;
 
 		if (memblocksize < size)
 			continue;
@@ -423,7 +421,7 @@ u64 __init __memblock_alloc_base(u64 size, u64 align, u64 max_addr)
 					return 0;
 				return base;
 			}
-			res_base = memblock.reserved.region[j].base;
+			res_base = memblock.reserved.regions[j].base;
 			if (res_base < size)
 				break;
 			base = memblock_align_down(res_base - size, align);
@@ -442,7 +440,7 @@ u64 memblock_end_of_DRAM(void)
 {
 	int idx = memblock.memory.cnt - 1;
 
-	return (memblock.memory.region[idx].base + memblock.memory.region[idx].size);
+	return (memblock.memory.regions[idx].base + memblock.memory.regions[idx].size);
 }
 
 /* You must call memblock_analyze() after this. */
@@ -450,7 +448,7 @@ void __init memblock_enforce_memory_limit(u64 memory_limit)
 {
 	unsigned long i;
 	u64 limit;
-	struct memblock_property *p;
+	struct memblock_region *p;
 
 	if (!memory_limit)
 		return;
@@ -458,24 +456,24 @@ void __init memblock_enforce_memory_limit(u64 memory_limit)
 	/* Truncate the memblock regions to satisfy the memory limit. */
 	limit = memory_limit;
 	for (i = 0; i < memblock.memory.cnt; i++) {
-		if (limit > memblock.memory.region[i].size) {
-			limit -= memblock.memory.region[i].size;
+		if (limit > memblock.memory.regions[i].size) {
+			limit -= memblock.memory.regions[i].size;
 			continue;
 		}
 
-		memblock.memory.region[i].size = limit;
+		memblock.memory.regions[i].size = limit;
 		memblock.memory.cnt = i + 1;
 		break;
 	}
 
-	if (memblock.memory.region[0].size < memblock.rmo_size)
-		memblock.rmo_size = memblock.memory.region[0].size;
+	if (memblock.memory.regions[0].size < memblock.rmo_size)
+		memblock.rmo_size = memblock.memory.regions[0].size;
 
 	memory_limit = memblock_end_of_DRAM();
 
 	/* And truncate any reserves above the limit also. */
 	for (i = 0; i < memblock.reserved.cnt; i++) {
-		p = &memblock.reserved.region[i];
+		p = &memblock.reserved.regions[i];
 
 		if (p->base > memory_limit)
 			p->size = 0;
@@ -494,9 +492,9 @@ int __init memblock_is_reserved(u64 addr)
 	int i;
 
 	for (i = 0; i < memblock.reserved.cnt; i++) {
-		u64 upper = memblock.reserved.region[i].base +
-			memblock.reserved.region[i].size - 1;
-		if ((addr >= memblock.reserved.region[i].base) && (addr <= upper))
+		u64 upper = memblock.reserved.regions[i].base +
+			memblock.reserved.regions[i].size - 1;
+		if ((addr >= memblock.reserved.regions[i].base) && (addr <= upper))
 			return 1;
 	}
 	return 0;
@@ -511,7 +509,7 @@ int memblock_is_region_reserved(u64 base, u64 size)
  * Given a <base, len>, find which memory regions belong to this range.
  * Adjust the request and return a contiguous chunk.
  */
-int memblock_find(struct memblock_property *res)
+int memblock_find(struct memblock_region *res)
 {
 	int i;
 	u64 rstart, rend;
@@ -520,8 +518,8 @@ int memblock_find(struct memblock_property *res)
 	rend = rstart + res->size - 1;
 
 	for (i = 0; i < memblock.memory.cnt; i++) {
-		u64 start = memblock.memory.region[i].base;
-		u64 end = start + memblock.memory.region[i].size - 1;
+		u64 start = memblock.memory.regions[i].base;
+		u64 end = start + memblock.memory.regions[i].size - 1;
 
 		if (start > rend)
 			return -1;
-- 
cgit v1.2.3


From 411a25a80da328f5ae6b6c037872ffe867fcc130 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:38:56 -0700
Subject: memblock: No reason to include asm/memblock.h late

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 86e7daf742f2..4b6931327b22 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -16,6 +16,8 @@
 #include <linux/init.h>
 #include <linux/mm.h>
 
+#include <asm/memblock.h>
+
 #define MAX_MEMBLOCK_REGIONS 128
 
 struct memblock_region {
@@ -82,8 +84,6 @@ memblock_end_pfn(struct memblock_type *type, unsigned long region_nr)
 	       memblock_size_pages(type, region_nr);
 }
 
-#include <asm/memblock.h>
-
 #endif /* __KERNEL__ */
 
 #endif /* _LINUX_MEMBLOCK_H */
-- 
cgit v1.2.3


From 72d4b0b4e0e7fa858767e03972771a9f7c02b689 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 4 Aug 2010 14:38:47 +1000
Subject: memblock: Implement memblock_is_memory and memblock_is_region_memory

To make it fast, we steal ARM's binary search for memblock_is_memory()
and we use that to also the replace existing implementation of
memblock_is_reserved().

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h |  2 ++
 mm/memblock.c            | 42 ++++++++++++++++++++++++++++++++++--------
 2 files changed, 36 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 4b6931327b22..47bceb187058 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -56,6 +56,8 @@ extern u64 __init __memblock_alloc_base(u64 size,
 extern u64 __init memblock_phys_mem_size(void);
 extern u64 memblock_end_of_DRAM(void);
 extern void __init memblock_enforce_memory_limit(u64 memory_limit);
+extern int memblock_is_memory(u64 addr);
+extern int memblock_is_region_memory(u64 base, u64 size);
 extern int __init memblock_is_reserved(u64 addr);
 extern int memblock_is_region_reserved(u64 base, u64 size);
 extern int memblock_find(struct memblock_region *res);
diff --git a/mm/memblock.c b/mm/memblock.c
index 6f407ccf604e..aa88c62bce7f 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -487,17 +487,43 @@ void __init memblock_enforce_memory_limit(u64 memory_limit)
 	}
 }
 
+static int memblock_search(struct memblock_type *type, u64 addr)
+{
+	unsigned int left = 0, right = type->cnt;
+
+	do {
+		unsigned int mid = (right + left) / 2;
+
+		if (addr < type->regions[mid].base)
+			right = mid;
+		else if (addr >= (type->regions[mid].base +
+				  type->regions[mid].size))
+			left = mid + 1;
+		else
+			return mid;
+	} while (left < right);
+	return -1;
+}
+
 int __init memblock_is_reserved(u64 addr)
 {
-	int i;
+	return memblock_search(&memblock.reserved, addr) != -1;
+}
 
-	for (i = 0; i < memblock.reserved.cnt; i++) {
-		u64 upper = memblock.reserved.regions[i].base +
-			memblock.reserved.regions[i].size - 1;
-		if ((addr >= memblock.reserved.regions[i].base) && (addr <= upper))
-			return 1;
-	}
-	return 0;
+int memblock_is_memory(u64 addr)
+{
+	return memblock_search(&memblock.memory, addr) != -1;
+}
+
+int memblock_is_region_memory(u64 base, u64 size)
+{
+	int idx = memblock_search(&memblock.reserved, base);
+
+	if (idx == -1)
+		return 0;
+	return memblock.reserved.regions[idx].base <= base &&
+		(memblock.reserved.regions[idx].base +
+		 memblock.reserved.regions[idx].size) >= (base + size);
 }
 
 int memblock_is_region_reserved(u64 base, u64 size)
-- 
cgit v1.2.3


From 5b385f259fa4d356452e3b4729cbaf5213f4f55b Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 4 Aug 2010 13:40:38 +1000
Subject: memblock: Introduce for_each_memblock() and new accessors

Walk memblock's using for_each_memblock() and use memblock_region_base/end_pfn() for
getting to PFNs.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 52 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 52 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 47bceb187058..c914112cd24f 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -64,6 +64,7 @@ extern int memblock_find(struct memblock_region *res);
 
 extern void memblock_dump_all(void);
 
+/* Obsolete accessors */
 static inline u64
 memblock_size_bytes(struct memblock_type *type, unsigned long region_nr)
 {
@@ -86,6 +87,57 @@ memblock_end_pfn(struct memblock_type *type, unsigned long region_nr)
 	       memblock_size_pages(type, region_nr);
 }
 
+/*
+ * pfn conversion functions
+ *
+ * While the memory MEMBLOCKs should always be page aligned, the reserved
+ * MEMBLOCKs may not be. This accessor attempt to provide a very clear
+ * idea of what they return for such non aligned MEMBLOCKs.
+ */
+
+/**
+ * memblock_region_base_pfn - Return the lowest pfn intersecting with the region
+ * @reg: memblock_region structure
+ */
+static inline unsigned long memblock_region_base_pfn(const struct memblock_region *reg)
+{
+	return reg->base >> PAGE_SHIFT;
+}
+
+/**
+ * memblock_region_last_pfn - Return the highest pfn intersecting with the region
+ * @reg: memblock_region structure
+ */
+static inline unsigned long memblock_region_last_pfn(const struct memblock_region *reg)
+{
+	return (reg->base + reg->size - 1) >> PAGE_SHIFT;
+}
+
+/**
+ * memblock_region_end_pfn - Return the pfn of the first page following the region
+ *                      but not intersecting it
+ * @reg: memblock_region structure
+ */
+static inline unsigned long memblock_region_end_pfn(const struct memblock_region *reg)
+{
+	return memblock_region_last_pfn(reg) + 1;
+}
+
+/**
+ * memblock_region_pages - Return the number of pages covering a region
+ * @reg: memblock_region structure
+ */
+static inline unsigned long memblock_region_pages(const struct memblock_region *reg)
+{
+	return memblock_region_end_pfn(reg) - memblock_region_end_pfn(reg);
+}
+
+#define for_each_memblock(memblock_type, region)					\
+	for (region = memblock.memblock_type.regions;				\
+	     region < (memblock.memblock_type.regions + memblock.memblock_type.cnt);	\
+	     region++)
+
+
 #endif /* __KERNEL__ */
 
 #endif /* _LINUX_MEMBLOCK_H */
-- 
cgit v1.2.3


From 1e2b904026e9debf95f500b8980a00c43ac0f31c Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 4 Aug 2010 13:52:25 +1000
Subject: memblock: Remove obsolete accessors

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 23 -----------------------
 1 file changed, 23 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index c914112cd24f..7d70fdd43db4 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -64,29 +64,6 @@ extern int memblock_find(struct memblock_region *res);
 
 extern void memblock_dump_all(void);
 
-/* Obsolete accessors */
-static inline u64
-memblock_size_bytes(struct memblock_type *type, unsigned long region_nr)
-{
-	return type->regions[region_nr].size;
-}
-static inline u64
-memblock_size_pages(struct memblock_type *type, unsigned long region_nr)
-{
-	return memblock_size_bytes(type, region_nr) >> PAGE_SHIFT;
-}
-static inline u64
-memblock_start_pfn(struct memblock_type *type, unsigned long region_nr)
-{
-	return type->regions[region_nr].base >> PAGE_SHIFT;
-}
-static inline u64
-memblock_end_pfn(struct memblock_type *type, unsigned long region_nr)
-{
-	return memblock_start_pfn(type, region_nr) +
-	       memblock_size_pages(type, region_nr);
-}
-
 /*
  * pfn conversion functions
  *
-- 
cgit v1.2.3


From b693fffb189fbfe7e1e8317ce5838808be8666a0 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 4 Aug 2010 13:52:55 +1000
Subject: memblock: Remove memblock_find()

Nobody uses it anymore. It's semantics were ... weird

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h |  1 -
 mm/memblock.c            | 32 --------------------------------
 2 files changed, 33 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 7d70fdd43db4..776c7d945dcc 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -60,7 +60,6 @@ extern int memblock_is_memory(u64 addr);
 extern int memblock_is_region_memory(u64 base, u64 size);
 extern int __init memblock_is_reserved(u64 addr);
 extern int memblock_is_region_reserved(u64 base, u64 size);
-extern int memblock_find(struct memblock_region *res);
 
 extern void memblock_dump_all(void);
 
diff --git a/mm/memblock.c b/mm/memblock.c
index aa88c62bce7f..8a118b71cbec 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -531,35 +531,3 @@ int memblock_is_region_reserved(u64 base, u64 size)
 	return memblock_overlaps_region(&memblock.reserved, base, size) >= 0;
 }
 
-/*
- * Given a <base, len>, find which memory regions belong to this range.
- * Adjust the request and return a contiguous chunk.
- */
-int memblock_find(struct memblock_region *res)
-{
-	int i;
-	u64 rstart, rend;
-
-	rstart = res->base;
-	rend = rstart + res->size - 1;
-
-	for (i = 0; i < memblock.memory.cnt; i++) {
-		u64 start = memblock.memory.regions[i].base;
-		u64 end = start + memblock.memory.regions[i].size - 1;
-
-		if (start > rend)
-			return -1;
-
-		if ((end >= rstart) && (start < rend)) {
-			/* adjust the request */
-			if (rstart < start)
-				rstart = start;
-			if (rend > end)
-				rend = end;
-			res->base = rstart;
-			res->size = rend - rstart + 1;
-			return 0;
-		}
-	}
-	return -1;
-}
-- 
cgit v1.2.3


From 35a1f0bd07015dde66501b47cfb6ddc72ebe7346 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:38:58 -0700
Subject: memblock: Remove nid_range argument, arch provides
 memblock_nid_range() instead

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/sparc/mm/init_64.c  | 16 ++++++----------
 include/linux/memblock.h |  7 +++++--
 mm/memblock.c            | 13 ++++++++-----
 3 files changed, 19 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index dd68025ecdbe..0883113624b9 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -785,8 +785,7 @@ static int find_node(unsigned long addr)
 	return -1;
 }
 
-static unsigned long long nid_range(unsigned long long start,
-				    unsigned long long end, int *nid)
+u64 memblock_nid_range(u64 start, u64 end, int *nid)
 {
 	*nid = find_node(start);
 	start += PAGE_SIZE;
@@ -804,8 +803,7 @@ static unsigned long long nid_range(unsigned long long start,
 	return start;
 }
 #else
-static unsigned long long nid_range(unsigned long long start,
-				    unsigned long long end, int *nid)
+u64 memblock_nid_range(u64 start, u64 end, int *nid)
 {
 	*nid = 0;
 	return end;
@@ -822,8 +820,7 @@ static void __init allocate_node_data(int nid)
 	struct pglist_data *p;
 
 #ifdef CONFIG_NEED_MULTIPLE_NODES
-	paddr = memblock_alloc_nid(sizeof(struct pglist_data),
-			      SMP_CACHE_BYTES, nid, nid_range);
+	paddr = memblock_alloc_nid(sizeof(struct pglist_data), SMP_CACHE_BYTES, nid);
 	if (!paddr) {
 		prom_printf("Cannot allocate pglist_data for nid[%d]\n", nid);
 		prom_halt();
@@ -843,8 +840,7 @@ static void __init allocate_node_data(int nid)
 	if (p->node_spanned_pages) {
 		num_pages = bootmem_bootmap_pages(p->node_spanned_pages);
 
-		paddr = memblock_alloc_nid(num_pages << PAGE_SHIFT, PAGE_SIZE, nid,
-				      nid_range);
+		paddr = memblock_alloc_nid(num_pages << PAGE_SHIFT, PAGE_SIZE, nid);
 		if (!paddr) {
 			prom_printf("Cannot allocate bootmap for nid[%d]\n",
 				  nid);
@@ -984,7 +980,7 @@ static void __init add_node_ranges(void)
 			unsigned long this_end;
 			int nid;
 
-			this_end = nid_range(start, end, &nid);
+			this_end = memblock_nid_range(start, end, &nid);
 
 			numadbg("Adding active range nid[%d] "
 				"start[%lx] end[%lx]\n",
@@ -1317,7 +1313,7 @@ static void __init reserve_range_in_node(int nid, unsigned long start,
 		unsigned long this_end;
 		int n;
 
-		this_end = nid_range(start, end, &n);
+		this_end = memblock_nid_range(start, end, &n);
 		if (n == nid) {
 			numadbg("      MATCH reserving range [%lx:%lx]\n",
 				start, this_end);
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 776c7d945dcc..367dea6e95a0 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -46,8 +46,7 @@ extern long memblock_add(u64 base, u64 size);
 extern long memblock_remove(u64 base, u64 size);
 extern long __init memblock_free(u64 base, u64 size);
 extern long __init memblock_reserve(u64 base, u64 size);
-extern u64 __init memblock_alloc_nid(u64 size, u64 align, int nid,
-				u64 (*nid_range)(u64, u64, int *));
+extern u64 __init memblock_alloc_nid(u64 size, u64 align, int nid);
 extern u64 __init memblock_alloc(u64 size, u64 align);
 extern u64 __init memblock_alloc_base(u64 size,
 		u64, u64 max_addr);
@@ -63,6 +62,10 @@ extern int memblock_is_region_reserved(u64 base, u64 size);
 
 extern void memblock_dump_all(void);
 
+/* Provided by the architecture */
+extern u64 memblock_nid_range(u64 start, u64 end, int *nid);
+
+
 /*
  * pfn conversion functions
  *
diff --git a/mm/memblock.c b/mm/memblock.c
index 8a118b71cbec..13807f280ada 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -319,7 +319,6 @@ static u64 __init memblock_alloc_nid_unreserved(u64 start, u64 end,
 }
 
 static u64 __init memblock_alloc_nid_region(struct memblock_region *mp,
-				       u64 (*nid_range)(u64, u64, int *),
 				       u64 size, u64 align, int nid)
 {
 	u64 start, end;
@@ -332,7 +331,7 @@ static u64 __init memblock_alloc_nid_region(struct memblock_region *mp,
 		u64 this_end;
 		int this_nid;
 
-		this_end = nid_range(start, end, &this_nid);
+		this_end = memblock_nid_range(start, end, &this_nid);
 		if (this_nid == nid) {
 			u64 ret = memblock_alloc_nid_unreserved(start, this_end,
 							   size, align);
@@ -345,8 +344,7 @@ static u64 __init memblock_alloc_nid_region(struct memblock_region *mp,
 	return ~(u64)0;
 }
 
-u64 __init memblock_alloc_nid(u64 size, u64 align, int nid,
-			 u64 (*nid_range)(u64 start, u64 end, int *nid))
+u64 __init memblock_alloc_nid(u64 size, u64 align, int nid)
 {
 	struct memblock_type *mem = &memblock.memory;
 	int i;
@@ -357,7 +355,6 @@ u64 __init memblock_alloc_nid(u64 size, u64 align, int nid,
 
 	for (i = 0; i < mem->cnt; i++) {
 		u64 ret = memblock_alloc_nid_region(&mem->regions[i],
-					       nid_range,
 					       size, align, nid);
 		if (ret != ~(u64)0)
 			return ret;
@@ -531,3 +528,9 @@ int memblock_is_region_reserved(u64 base, u64 size)
 	return memblock_overlaps_region(&memblock.reserved, base, size) >= 0;
 }
 
+u64 __weak memblock_nid_range(u64 start, u64 end, int *nid)
+{
+	*nid = 0;
+
+	return end;
+}
-- 
cgit v1.2.3


From 27f574c223d2c09610058b3ec7a29582d63a3e06 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:39:00 -0700
Subject: memblock: Expose MEMBLOCK_ALLOC_ANYWHERE

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/mm/hash_utils_64.c | 2 +-
 include/linux/memblock.h        | 1 +
 mm/memblock.c                   | 2 --
 3 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index 4072b871497d..a542ff5ec8a9 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -625,7 +625,7 @@ static void __init htab_initialize(void)
 		if (machine_is(cell))
 			limit = 0x80000000;
 		else
-			limit = 0;
+			limit = MEMBLOCK_ALLOC_ANYWHERE;
 
 		table = memblock_alloc_base(htab_size_bytes, htab_size_bytes, limit);
 
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 367dea6e95a0..3cf3304e901d 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -50,6 +50,7 @@ extern u64 __init memblock_alloc_nid(u64 size, u64 align, int nid);
 extern u64 __init memblock_alloc(u64 size, u64 align);
 extern u64 __init memblock_alloc_base(u64 size,
 		u64, u64 max_addr);
+#define MEMBLOCK_ALLOC_ANYWHERE	0
 extern u64 __init __memblock_alloc_base(u64 size,
 		u64 align, u64 max_addr);
 extern u64 __init memblock_phys_mem_size(void);
diff --git a/mm/memblock.c b/mm/memblock.c
index e264e8c70892..0131684c42f8 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -15,8 +15,6 @@
 #include <linux/bitops.h>
 #include <linux/memblock.h>
 
-#define MEMBLOCK_ALLOC_ANYWHERE	0
-
 struct memblock memblock;
 
 static int memblock_debug;
-- 
cgit v1.2.3


From e63075a3c9377536d085bc013cd3fe6323162449 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:39:01 -0700
Subject: memblock: Introduce default allocation limit and use it to replace
 explicit ones

This introduce memblock.current_limit which is used to limit allocations
from memblock_alloc() or memblock_alloc_base(..., MEMBLOCK_ALLOC_ACCESSIBLE).

The old MEMBLOCK_ALLOC_ANYWHERE changes value from 0 to ~(u64)0 and can still
be used with memblock_alloc_base() to allocate really anywhere.

It is -no-longer- cropped to MEMBLOCK_REAL_LIMIT which disappears.

Note to archs: I'm leaving the default limit to MEMBLOCK_ALLOC_ANYWHERE. I
strongly recommend that you ensure that you set an appropriate limit
during boot in order to guarantee that an memblock_alloc() at any time
results in something that is accessible with a simple __va().

The reason is that a subsequent patch will introduce the ability for
the array to resize itself by reallocating itself. The MEMBLOCK core will
honor the current limit when performing those allocations.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/microblaze/include/asm/memblock.h |  3 ---
 arch/powerpc/include/asm/memblock.h    |  7 -------
 arch/powerpc/kernel/prom.c             | 20 +++++++++++++++++++-
 arch/powerpc/kernel/setup_32.c         |  2 +-
 arch/powerpc/mm/40x_mmu.c              |  5 +++--
 arch/powerpc/mm/fsl_booke_mmu.c        |  3 ++-
 arch/powerpc/mm/hash_utils_64.c        |  3 ++-
 arch/powerpc/mm/init_32.c              | 29 +++++++----------------------
 arch/powerpc/mm/ppc_mmu_32.c           |  3 +--
 arch/powerpc/mm/tlb_nohash.c           |  2 ++
 arch/sh/include/asm/memblock.h         |  2 --
 arch/sparc/include/asm/memblock.h      |  2 --
 include/linux/memblock.h               | 16 +++++++++++++++-
 mm/memblock.c                          | 19 +++++++++++--------
 14 files changed, 63 insertions(+), 53 deletions(-)

(limited to 'include/linux')

diff --git a/arch/microblaze/include/asm/memblock.h b/arch/microblaze/include/asm/memblock.h
index f9c2fa331d2a..20a8e257c77f 100644
--- a/arch/microblaze/include/asm/memblock.h
+++ b/arch/microblaze/include/asm/memblock.h
@@ -9,9 +9,6 @@
 #ifndef _ASM_MICROBLAZE_MEMBLOCK_H
 #define _ASM_MICROBLAZE_MEMBLOCK_H
 
-/* MEMBLOCK limit is OFF */
-#define MEMBLOCK_REAL_LIMIT	0xFFFFFFFF
-
 #endif /* _ASM_MICROBLAZE_MEMBLOCK_H */
 
 
diff --git a/arch/powerpc/include/asm/memblock.h b/arch/powerpc/include/asm/memblock.h
index 3c29728b56b1..43efc345065e 100644
--- a/arch/powerpc/include/asm/memblock.h
+++ b/arch/powerpc/include/asm/memblock.h
@@ -5,11 +5,4 @@
 
 #define MEMBLOCK_DBG(fmt...) udbg_printf(fmt)
 
-#ifdef CONFIG_PPC32
-extern phys_addr_t lowmem_end_addr;
-#define MEMBLOCK_REAL_LIMIT	lowmem_end_addr
-#else
-#define MEMBLOCK_REAL_LIMIT	0
-#endif
-
 #endif /* _ASM_POWERPC_MEMBLOCK_H */
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index fed9bf6187d1..3aec0b980f6a 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -98,7 +98,7 @@ static void __init move_device_tree(void)
 
 	if ((memory_limit && (start + size) > memory_limit) ||
 			overlaps_crashkernel(start, size)) {
-		p = __va(memblock_alloc_base(size, PAGE_SIZE, memblock.rmo_size));
+		p = __va(memblock_alloc(size, PAGE_SIZE));
 		memcpy(p, initial_boot_params, size);
 		initial_boot_params = (struct boot_param_header *)p;
 		DBG("Moved device tree to 0x%p\n", p);
@@ -655,6 +655,21 @@ static void __init phyp_dump_reserve_mem(void)
 static inline void __init phyp_dump_reserve_mem(void) {}
 #endif /* CONFIG_PHYP_DUMP  && CONFIG_PPC_RTAS */
 
+static void set_boot_memory_limit(void)
+{
+#ifdef CONFIG_PPC32
+	/* 601 can only access 16MB at the moment */
+	if (PVR_VER(mfspr(SPRN_PVR)) == 1)
+		memblock_set_current_limit(0x01000000);
+	/* 8xx can only access 8MB at the moment */
+	else if (PVR_VER(mfspr(SPRN_PVR)) == 0x50)
+		memblock_set_current_limit(0x00800000);
+	else
+		memblock_set_current_limit(0x10000000);
+#else
+	memblock_set_current_limit(memblock.rmo_size);
+#endif
+}
 
 void __init early_init_devtree(void *params)
 {
@@ -683,6 +698,7 @@ void __init early_init_devtree(void *params)
 
 	/* Scan memory nodes and rebuild MEMBLOCKs */
 	memblock_init();
+
 	of_scan_flat_dt(early_init_dt_scan_root, NULL);
 	of_scan_flat_dt(early_init_dt_scan_memory_ppc, NULL);
 
@@ -718,6 +734,8 @@ void __init early_init_devtree(void *params)
 
 	DBG("Phys. mem: %llx\n", memblock_phys_mem_size());
 
+	set_boot_memory_limit();
+
 	/* We may need to relocate the flat tree, do it now.
 	 * FIXME .. and the initrd too? */
 	move_device_tree();
diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index a10ffc85ada7..b7eb1ded3b5f 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -246,7 +246,7 @@ static void __init irqstack_early_init(void)
 	unsigned int i;
 
 	/* interrupt stacks must be in lowmem, we get that for free on ppc32
-	 * as the memblock is limited to lowmem by MEMBLOCK_REAL_LIMIT */
+	 * as the memblock is limited to lowmem by default */
 	for_each_possible_cpu(i) {
 		softirq_ctx[i] = (struct thread_info *)
 			__va(memblock_alloc(THREAD_SIZE, THREAD_SIZE));
diff --git a/arch/powerpc/mm/40x_mmu.c b/arch/powerpc/mm/40x_mmu.c
index 1dc2fa5ce1bd..58969b51f454 100644
--- a/arch/powerpc/mm/40x_mmu.c
+++ b/arch/powerpc/mm/40x_mmu.c
@@ -35,6 +35,7 @@
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/highmem.h>
+#include <linux/memblock.h>
 
 #include <asm/pgalloc.h>
 #include <asm/prom.h>
@@ -47,6 +48,7 @@
 #include <asm/bootx.h>
 #include <asm/machdep.h>
 #include <asm/setup.h>
+
 #include "mmu_decl.h"
 
 extern int __map_without_ltlbs;
@@ -139,8 +141,7 @@ unsigned long __init mmu_mapin_ram(unsigned long top)
 	 * coverage with normal-sized pages (or other reasons) do not
 	 * attempt to allocate outside the allowed range.
 	 */
-
-	__initial_memory_limit_addr = memstart_addr + mapped;
+	memblock_set_current_limit(memstart_addr + mapped);
 
 	return mapped;
 }
diff --git a/arch/powerpc/mm/fsl_booke_mmu.c b/arch/powerpc/mm/fsl_booke_mmu.c
index cdc7526e9c93..e525f862d759 100644
--- a/arch/powerpc/mm/fsl_booke_mmu.c
+++ b/arch/powerpc/mm/fsl_booke_mmu.c
@@ -40,6 +40,7 @@
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/highmem.h>
+#include <linux/memblock.h>
 
 #include <asm/pgalloc.h>
 #include <asm/prom.h>
@@ -212,5 +213,5 @@ void __init adjust_total_lowmem(void)
 	pr_cont("%lu Mb, residual: %dMb\n", tlbcam_sz(tlbcam_index - 1) >> 20,
 	        (unsigned int)((total_lowmem - __max_low_memory) >> 20));
 
-	__initial_memory_limit_addr = memstart_addr + __max_low_memory;
+	memblock_set_current_limit(memstart_addr + __max_low_memory);
 }
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index a542ff5ec8a9..b05890e23813 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -696,7 +696,8 @@ static void __init htab_initialize(void)
 #endif /* CONFIG_U3_DART */
 		BUG_ON(htab_bolt_mapping(base, base + size, __pa(base),
 				prot, mmu_linear_psize, mmu_kernel_ssize));
-       }
+	}
+	memblock_set_current_limit(MEMBLOCK_ALLOC_ANYWHERE);
 
 	/*
 	 * If we have a memory_limit and we've allocated TCEs then we need to
diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c
index 6a6975dc2654..59b208b7ec6f 100644
--- a/arch/powerpc/mm/init_32.c
+++ b/arch/powerpc/mm/init_32.c
@@ -91,12 +91,6 @@ int __allow_ioremap_reserved;
 /* max amount of low RAM to map in */
 unsigned long __max_low_memory = MAX_LOW_MEM;
 
-/*
- * address of the limit of what is accessible with initial MMU setup -
- * 256MB usually, but only 16MB on 601.
- */
-phys_addr_t __initial_memory_limit_addr = (phys_addr_t)0x10000000;
-
 /*
  * Check for command-line options that affect what MMU_init will do.
  */
@@ -126,13 +120,6 @@ void __init MMU_init(void)
 	if (ppc_md.progress)
 		ppc_md.progress("MMU:enter", 0x111);
 
-	/* 601 can only access 16MB at the moment */
-	if (PVR_VER(mfspr(SPRN_PVR)) == 1)
-		__initial_memory_limit_addr = 0x01000000;
-	/* 8xx can only access 8MB at the moment */
-	if (PVR_VER(mfspr(SPRN_PVR)) == 0x50)
-		__initial_memory_limit_addr = 0x00800000;
-
 	/* parse args from command line */
 	MMU_setup();
 
@@ -190,20 +177,18 @@ void __init MMU_init(void)
 #ifdef CONFIG_BOOTX_TEXT
 	btext_unmap();
 #endif
+
+	/* Shortly after that, the entire linear mapping will be available */
+	memblock_set_current_limit(lowmem_end_addr);
 }
 
 /* This is only called until mem_init is done. */
 void __init *early_get_page(void)
 {
-	void *p;
-
-	if (init_bootmem_done) {
-		p = alloc_bootmem_pages(PAGE_SIZE);
-	} else {
-		p = __va(memblock_alloc_base(PAGE_SIZE, PAGE_SIZE,
-					__initial_memory_limit_addr));
-	}
-	return p;
+	if (init_bootmem_done)
+		return alloc_bootmem_pages(PAGE_SIZE);
+	else
+		return __va(memblock_alloc(PAGE_SIZE, PAGE_SIZE));
 }
 
 /* Free up now-unused memory */
diff --git a/arch/powerpc/mm/ppc_mmu_32.c b/arch/powerpc/mm/ppc_mmu_32.c
index f8a01829d64f..7d34e170e80f 100644
--- a/arch/powerpc/mm/ppc_mmu_32.c
+++ b/arch/powerpc/mm/ppc_mmu_32.c
@@ -223,8 +223,7 @@ void __init MMU_init_hw(void)
 	 * Find some memory for the hash table.
 	 */
 	if ( ppc_md.progress ) ppc_md.progress("hash:find piece", 0x322);
-	Hash = __va(memblock_alloc_base(Hash_size, Hash_size,
-				   __initial_memory_limit_addr));
+	Hash = __va(memblock_alloc(Hash_size, Hash_size));
 	cacheable_memzero(Hash, Hash_size);
 	_SDR1 = __pa(Hash) | SDR1_LOW_BITS;
 
diff --git a/arch/powerpc/mm/tlb_nohash.c b/arch/powerpc/mm/tlb_nohash.c
index d8695b02a968..7ba32e762990 100644
--- a/arch/powerpc/mm/tlb_nohash.c
+++ b/arch/powerpc/mm/tlb_nohash.c
@@ -432,6 +432,8 @@ static void __early_init_mmu(int boot_cpu)
 	 * the MMU configuration
 	 */
 	mb();
+
+	memblock_set_current_limit(linear_map_top);
 }
 
 void __init early_init_mmu(void)
diff --git a/arch/sh/include/asm/memblock.h b/arch/sh/include/asm/memblock.h
index dfe683b88075..e87063fad2ea 100644
--- a/arch/sh/include/asm/memblock.h
+++ b/arch/sh/include/asm/memblock.h
@@ -1,6 +1,4 @@
 #ifndef __ASM_SH_MEMBLOCK_H
 #define __ASM_SH_MEMBLOCK_H
 
-#define MEMBLOCK_REAL_LIMIT	0
-
 #endif /* __ASM_SH_MEMBLOCK_H */
diff --git a/arch/sparc/include/asm/memblock.h b/arch/sparc/include/asm/memblock.h
index f12af880649b..c67b047ef85e 100644
--- a/arch/sparc/include/asm/memblock.h
+++ b/arch/sparc/include/asm/memblock.h
@@ -5,6 +5,4 @@
 
 #define MEMBLOCK_DBG(fmt...) prom_printf(fmt)
 
-#define MEMBLOCK_REAL_LIMIT	0
-
 #endif /* !(_SPARC64_MEMBLOCK_H) */
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 3cf3304e901d..c4f6e53264ed 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -34,6 +34,7 @@ struct memblock_type {
 struct memblock {
 	unsigned long debug;
 	u64 rmo_size;
+	u64 current_limit;
 	struct memblock_type memory;
 	struct memblock_type reserved;
 };
@@ -46,11 +47,16 @@ extern long memblock_add(u64 base, u64 size);
 extern long memblock_remove(u64 base, u64 size);
 extern long __init memblock_free(u64 base, u64 size);
 extern long __init memblock_reserve(u64 base, u64 size);
+
 extern u64 __init memblock_alloc_nid(u64 size, u64 align, int nid);
 extern u64 __init memblock_alloc(u64 size, u64 align);
+
+/* Flags for memblock_alloc_base() amd __memblock_alloc_base() */
+#define MEMBLOCK_ALLOC_ANYWHERE	(~(u64)0)
+#define MEMBLOCK_ALLOC_ACCESSIBLE	0
+
 extern u64 __init memblock_alloc_base(u64 size,
 		u64, u64 max_addr);
-#define MEMBLOCK_ALLOC_ANYWHERE	0
 extern u64 __init __memblock_alloc_base(u64 size,
 		u64 align, u64 max_addr);
 extern u64 __init memblock_phys_mem_size(void);
@@ -66,6 +72,14 @@ extern void memblock_dump_all(void);
 /* Provided by the architecture */
 extern u64 memblock_nid_range(u64 start, u64 end, int *nid);
 
+/**
+ * memblock_set_current_limit - Set the current allocation limit to allow
+ *                         limiting allocations to what is currently
+ *                         accessible during boot
+ * @limit: New limit value (physical address)
+ */
+extern void memblock_set_current_limit(u64 limit);
+
 
 /*
  * pfn conversion functions
diff --git a/mm/memblock.c b/mm/memblock.c
index 0131684c42f8..770c5bfac2cd 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -115,6 +115,8 @@ void __init memblock_init(void)
 	memblock.reserved.regions[0].base = 0;
 	memblock.reserved.regions[0].size = 0;
 	memblock.reserved.cnt = 1;
+
+	memblock.current_limit = MEMBLOCK_ALLOC_ANYWHERE;
 }
 
 void __init memblock_analyze(void)
@@ -373,7 +375,7 @@ u64 __init memblock_alloc_nid(u64 size, u64 align, int nid)
 
 u64 __init memblock_alloc(u64 size, u64 align)
 {
-	return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ANYWHERE);
+	return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ACCESSIBLE);
 }
 
 u64 __init memblock_alloc_base(u64 size, u64 align, u64 max_addr)
@@ -399,14 +401,9 @@ u64 __init __memblock_alloc_base(u64 size, u64 align, u64 max_addr)
 
 	size = memblock_align_up(size, align);
 
-	/* On some platforms, make sure we allocate lowmem */
-	/* Note that MEMBLOCK_REAL_LIMIT may be MEMBLOCK_ALLOC_ANYWHERE */
-	if (max_addr == MEMBLOCK_ALLOC_ANYWHERE)
-		max_addr = MEMBLOCK_REAL_LIMIT;
-
 	/* Pump up max_addr */
-	if (max_addr == MEMBLOCK_ALLOC_ANYWHERE)
-		max_addr = ~(u64)0;
+	if (max_addr == MEMBLOCK_ALLOC_ACCESSIBLE)
+		max_addr = memblock.current_limit;
 
 	/* We do a top-down search, this tends to limit memory
 	 * fragmentation by keeping early boot allocs near the
@@ -527,3 +524,9 @@ int memblock_is_region_reserved(u64 base, u64 size)
 	return memblock_overlaps_region(&memblock.reserved, base, size) >= 0;
 }
 
+
+void __init memblock_set_current_limit(u64 limit)
+{
+	memblock.current_limit = limit;
+}
+
-- 
cgit v1.2.3


From cd3db0c4ca3d237e7ad20f7107216e575705d2b0 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:39:02 -0700
Subject: memblock: Remove rmo_size, burry it in arch/powerpc where it belongs

The RMA (RMO is a misnomer) is a concept specific to ppc64 (in fact
server ppc64 though I hijack it on embedded ppc64 for similar purposes)
and represents the area of memory that can be accessed in real mode
(aka with MMU off), or on embedded, from the exception vectors (which
is bolted in the TLB) which pretty much boils down to the same thing.

We take that out of the generic MEMBLOCK data structure and move it into
arch/powerpc where it belongs, renaming it to "RMA" while at it.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/mmu.h  | 12 ++++++++++++
 arch/powerpc/kernel/head_40x.S  |  6 +-----
 arch/powerpc/kernel/paca.c      |  2 +-
 arch/powerpc/kernel/prom.c      | 29 ++++++++---------------------
 arch/powerpc/kernel/rtas.c      |  2 +-
 arch/powerpc/kernel/setup_64.c  |  2 +-
 arch/powerpc/mm/40x_mmu.c       | 14 +++++++++++++-
 arch/powerpc/mm/44x_mmu.c       | 14 ++++++++++++++
 arch/powerpc/mm/fsl_booke_mmu.c |  9 +++++++++
 arch/powerpc/mm/hash_utils_64.c | 22 +++++++++++++++++++++-
 arch/powerpc/mm/init_32.c       | 14 ++++++++++++++
 arch/powerpc/mm/init_64.c       |  1 +
 arch/powerpc/mm/ppc_mmu_32.c    | 15 +++++++++++++++
 arch/powerpc/mm/tlb_nohash.c    | 14 ++++++++++++++
 include/linux/memblock.h        |  1 -
 mm/memblock.c                   |  8 --------
 16 files changed, 125 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/arch/powerpc/include/asm/mmu.h b/arch/powerpc/include/asm/mmu.h
index 7ebf42ed84a2..bb40a06d3b77 100644
--- a/arch/powerpc/include/asm/mmu.h
+++ b/arch/powerpc/include/asm/mmu.h
@@ -2,6 +2,8 @@
 #define _ASM_POWERPC_MMU_H_
 #ifdef __KERNEL__
 
+#include <linux/types.h>
+
 #include <asm/asm-compat.h>
 #include <asm/feature-fixups.h>
 
@@ -82,6 +84,16 @@ extern unsigned int __start___mmu_ftr_fixup, __stop___mmu_ftr_fixup;
 extern void early_init_mmu(void);
 extern void early_init_mmu_secondary(void);
 
+extern void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				       phys_addr_t first_memblock_size);
+
+#ifdef CONFIG_PPC64
+/* This is our real memory area size on ppc64 server, on embedded, we
+ * make it match the size our of bolted TLB area
+ */
+extern u64 ppc64_rma_size;
+#endif /* CONFIG_PPC64 */
+
 #endif /* !__ASSEMBLY__ */
 
 /* The kernel use the constants below to index in the page sizes array.
diff --git a/arch/powerpc/kernel/head_40x.S b/arch/powerpc/kernel/head_40x.S
index a90625f9b485..8278e8bad5a0 100644
--- a/arch/powerpc/kernel/head_40x.S
+++ b/arch/powerpc/kernel/head_40x.S
@@ -923,11 +923,7 @@ initial_mmu:
 	mtspr	SPRN_PID,r0
 	sync
 
-	/* Configure and load two entries into TLB slots 62 and 63.
-	 * In case we are pinning TLBs, these are reserved in by the
-	 * other TLB functions.  If not reserving, then it doesn't
-	 * matter where they are loaded.
-	 */
+	/* Configure and load one entry into TLB slots 63 */
 	clrrwi	r4,r4,10		/* Mask off the real page number */
 	ori	r4,r4,(TLB_WR | TLB_EX)	/* Set the write and execute bits */
 
diff --git a/arch/powerpc/kernel/paca.c b/arch/powerpc/kernel/paca.c
index 139a773853f4..b9ffd7deeed7 100644
--- a/arch/powerpc/kernel/paca.c
+++ b/arch/powerpc/kernel/paca.c
@@ -117,7 +117,7 @@ void __init allocate_pacas(void)
 	 * the first segment. On iSeries they must be within the area mapped
 	 * by the HV, which is HvPagesToMap * HVPAGESIZE bytes.
 	 */
-	limit = min(0x10000000ULL, memblock.rmo_size);
+	limit = min(0x10000000ULL, ppc64_rma_size);
 	if (firmware_has_feature(FW_FEATURE_ISERIES))
 		limit = min(limit, HvPagesToMap * HVPAGESIZE);
 
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 3aec0b980f6a..c3c6a8857544 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -66,6 +66,7 @@
 int __initdata iommu_is_off;
 int __initdata iommu_force_on;
 unsigned long tce_alloc_start, tce_alloc_end;
+u64 ppc64_rma_size;
 #endif
 
 static int __init early_parse_mem(char *p)
@@ -492,7 +493,7 @@ static int __init early_init_dt_scan_memory_ppc(unsigned long node,
 
 void __init early_init_dt_add_memory_arch(u64 base, u64 size)
 {
-#if defined(CONFIG_PPC64)
+#ifdef CONFIG_PPC64
 	if (iommu_is_off) {
 		if (base >= 0x80000000ul)
 			return;
@@ -501,9 +502,13 @@ void __init early_init_dt_add_memory_arch(u64 base, u64 size)
 	}
 #endif
 
-	memblock_add(base, size);
-
+	/* First MEMBLOCK added, do some special initializations */
+	if (memstart_addr == ~(phys_addr_t)0)
+		setup_initial_memory_limit(base, size);
 	memstart_addr = min((u64)memstart_addr, base);
+
+	/* Add the chunk to the MEMBLOCK list */
+	memblock_add(base, size);
 }
 
 u64 __init early_init_dt_alloc_memory_arch(u64 size, u64 align)
@@ -655,22 +660,6 @@ static void __init phyp_dump_reserve_mem(void)
 static inline void __init phyp_dump_reserve_mem(void) {}
 #endif /* CONFIG_PHYP_DUMP  && CONFIG_PPC_RTAS */
 
-static void set_boot_memory_limit(void)
-{
-#ifdef CONFIG_PPC32
-	/* 601 can only access 16MB at the moment */
-	if (PVR_VER(mfspr(SPRN_PVR)) == 1)
-		memblock_set_current_limit(0x01000000);
-	/* 8xx can only access 8MB at the moment */
-	else if (PVR_VER(mfspr(SPRN_PVR)) == 0x50)
-		memblock_set_current_limit(0x00800000);
-	else
-		memblock_set_current_limit(0x10000000);
-#else
-	memblock_set_current_limit(memblock.rmo_size);
-#endif
-}
-
 void __init early_init_devtree(void *params)
 {
 	phys_addr_t limit;
@@ -734,8 +723,6 @@ void __init early_init_devtree(void *params)
 
 	DBG("Phys. mem: %llx\n", memblock_phys_mem_size());
 
-	set_boot_memory_limit();
-
 	/* We may need to relocate the flat tree, do it now.
 	 * FIXME .. and the initrd too? */
 	move_device_tree();
diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c
index d0516dbee762..1662777be5dd 100644
--- a/arch/powerpc/kernel/rtas.c
+++ b/arch/powerpc/kernel/rtas.c
@@ -934,7 +934,7 @@ void __init rtas_initialize(void)
 	 */
 #ifdef CONFIG_PPC64
 	if (machine_is(pseries) && firmware_has_feature(FW_FEATURE_LPAR)) {
-		rtas_region = min(memblock.rmo_size, RTAS_INSTANTIATE_MAX);
+		rtas_region = min(ppc64_rma_size, RTAS_INSTANTIATE_MAX);
 		ibm_suspend_me_token = rtas_token("ibm,suspend-me");
 	}
 #endif
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index d135f93cb0f6..4360944b60f0 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -487,7 +487,7 @@ static void __init emergency_stack_init(void)
 	 * bringup, we need to get at them in real mode. This means they
 	 * must also be within the RMO region.
 	 */
-	limit = min(slb0_limit(), memblock.rmo_size);
+	limit = min(slb0_limit(), ppc64_rma_size);
 
 	for_each_possible_cpu(i) {
 		unsigned long sp;
diff --git a/arch/powerpc/mm/40x_mmu.c b/arch/powerpc/mm/40x_mmu.c
index 58969b51f454..5810967511d4 100644
--- a/arch/powerpc/mm/40x_mmu.c
+++ b/arch/powerpc/mm/40x_mmu.c
@@ -141,7 +141,19 @@ unsigned long __init mmu_mapin_ram(unsigned long top)
 	 * coverage with normal-sized pages (or other reasons) do not
 	 * attempt to allocate outside the allowed range.
 	 */
-	memblock_set_current_limit(memstart_addr + mapped);
+	memblock_set_current_limit(mapped);
 
 	return mapped;
 }
+
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	/* We don't currently support the first MEMBLOCK not mapping 0
+	 * physical on those processors
+	 */
+	BUG_ON(first_memblock_base != 0);
+
+	/* 40x can only access 16MB at the moment (see head_40x.S) */
+	memblock_set_current_limit(min_t(u64, first_memblock_size, 0x00800000));
+}
diff --git a/arch/powerpc/mm/44x_mmu.c b/arch/powerpc/mm/44x_mmu.c
index d8c6efb32bc6..024acab588fd 100644
--- a/arch/powerpc/mm/44x_mmu.c
+++ b/arch/powerpc/mm/44x_mmu.c
@@ -24,6 +24,8 @@
  */
 
 #include <linux/init.h>
+#include <linux/memblock.h>
+
 #include <asm/mmu.h>
 #include <asm/system.h>
 #include <asm/page.h>
@@ -213,6 +215,18 @@ unsigned long __init mmu_mapin_ram(unsigned long top)
 	return total_lowmem;
 }
 
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	/* We don't currently support the first MEMBLOCK not mapping 0
+	 * physical on those processors
+	 */
+	BUG_ON(first_memblock_base != 0);
+
+	/* 44x has a 256M TLB entry pinned at boot */
+	memblock_set_current_limit(min_t(u64, first_memblock_size, PPC_PIN_SIZE));
+}
+
 #ifdef CONFIG_SMP
 void __cpuinit mmu_init_secondary(int cpu)
 {
diff --git a/arch/powerpc/mm/fsl_booke_mmu.c b/arch/powerpc/mm/fsl_booke_mmu.c
index e525f862d759..0be8fe24c54e 100644
--- a/arch/powerpc/mm/fsl_booke_mmu.c
+++ b/arch/powerpc/mm/fsl_booke_mmu.c
@@ -215,3 +215,12 @@ void __init adjust_total_lowmem(void)
 
 	memblock_set_current_limit(memstart_addr + __max_low_memory);
 }
+
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	phys_addr_t limit = first_memblock_base + first_memblock_size;
+
+	/* 64M mapped initially according to head_fsl_booke.S */
+	memblock_set_current_limit(min_t(u64, limit, 0x04000000));
+}
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index b05890e23813..83f534d862db 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -649,7 +649,7 @@ static void __init htab_initialize(void)
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	linear_map_hash_count = memblock_end_of_DRAM() >> PAGE_SHIFT;
 	linear_map_hash_slots = __va(memblock_alloc_base(linear_map_hash_count,
-						    1, memblock.rmo_size));
+						    1, ppc64_rma_size));
 	memset(linear_map_hash_slots, 0, linear_map_hash_count);
 #endif /* CONFIG_DEBUG_PAGEALLOC */
 
@@ -1248,3 +1248,23 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
 	local_irq_restore(flags);
 }
 #endif /* CONFIG_DEBUG_PAGEALLOC */
+
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	/* We don't currently support the first MEMBLOCK not mapping 0
+	 * physical on those processors
+	 */
+	BUG_ON(first_memblock_base != 0);
+
+	/* On LPAR systems, the first entry is our RMA region,
+	 * non-LPAR 64-bit hash MMU systems don't have a limitation
+	 * on real mode access, but using the first entry works well
+	 * enough. We also clamp it to 1G to avoid some funky things
+	 * such as RTAS bugs etc...
+	 */
+	ppc64_rma_size = min_t(u64, first_memblock_size, 0x40000000);
+
+	/* Finally limit subsequent allocations */
+	memblock_set_current_limit(ppc64_rma_size);
+}
diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c
index 59b208b7ec6f..742da43b4ab6 100644
--- a/arch/powerpc/mm/init_32.c
+++ b/arch/powerpc/mm/init_32.c
@@ -237,3 +237,17 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 }
 #endif
 
+
+#ifdef CONFIG_8xx /* No 8xx specific .c file to put that in ... */
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	/* We don't currently support the first MEMBLOCK not mapping 0
+	 * physical on those processors
+	 */
+	BUG_ON(first_memblock_base != 0);
+
+	/* 8xx can only access 8MB at the moment */
+	memblock_set_current_limit(min_t(u64, first_memblock_size, 0x00800000));
+}
+#endif /* CONFIG_8xx */
diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c
index 71f1415e2472..9e081ffbf0f2 100644
--- a/arch/powerpc/mm/init_64.c
+++ b/arch/powerpc/mm/init_64.c
@@ -328,3 +328,4 @@ int __meminit vmemmap_populate(struct page *start_page,
 	return 0;
 }
 #endif /* CONFIG_SPARSEMEM_VMEMMAP */
+
diff --git a/arch/powerpc/mm/ppc_mmu_32.c b/arch/powerpc/mm/ppc_mmu_32.c
index 7d34e170e80f..11571e118831 100644
--- a/arch/powerpc/mm/ppc_mmu_32.c
+++ b/arch/powerpc/mm/ppc_mmu_32.c
@@ -271,3 +271,18 @@ void __init MMU_init_hw(void)
 
 	if ( ppc_md.progress ) ppc_md.progress("hash:done", 0x205);
 }
+
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	/* We don't currently support the first MEMBLOCK not mapping 0
+	 * physical on those processors
+	 */
+	BUG_ON(first_memblock_base != 0);
+
+	/* 601 can only access 16MB at the moment */
+	if (PVR_VER(mfspr(SPRN_PVR)) == 1)
+		memblock_set_current_limit(min_t(u64, first_memblock_size, 0x01000000));
+	else /* Anything else has 256M mapped */
+		memblock_set_current_limit(min_t(u64, first_memblock_size, 0x10000000));
+}
diff --git a/arch/powerpc/mm/tlb_nohash.c b/arch/powerpc/mm/tlb_nohash.c
index 7ba32e762990..a086ed562606 100644
--- a/arch/powerpc/mm/tlb_nohash.c
+++ b/arch/powerpc/mm/tlb_nohash.c
@@ -446,4 +446,18 @@ void __cpuinit early_init_mmu_secondary(void)
 	__early_init_mmu(0);
 }
 
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	/* On Embedded 64-bit, we adjust the RMA size to match
+	 * the bolted TLB entry. We know for now that only 1G
+	 * entries are supported though that may eventually
+	 * change. We crop it to the size of the first MEMBLOCK to
+	 * avoid going over total available memory just in case...
+	 */
+	ppc64_rma_size = min_t(u64, first_memblock_size, 0x40000000);
+
+	/* Finally limit subsequent allocations */
+	memblock_set_current_limit(ppc64_memblock_base + ppc64_rma_size);
+}
 #endif /* CONFIG_PPC64 */
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index c4f6e53264ed..71b8edc6ede8 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -33,7 +33,6 @@ struct memblock_type {
 
 struct memblock {
 	unsigned long debug;
-	u64 rmo_size;
 	u64 current_limit;
 	struct memblock_type memory;
 	struct memblock_type reserved;
diff --git a/mm/memblock.c b/mm/memblock.c
index 770c5bfac2cd..73d903ebf3d4 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -49,7 +49,6 @@ void memblock_dump_all(void)
 		return;
 
 	pr_info("MEMBLOCK configuration:\n");
-	pr_info(" rmo_size    = 0x%llx\n", (unsigned long long)memblock.rmo_size);
 	pr_info(" memory.size = 0x%llx\n", (unsigned long long)memblock.memory.size);
 
 	memblock_dump(&memblock.memory, "memory");
@@ -195,10 +194,6 @@ static long memblock_add_region(struct memblock_type *type, u64 base, u64 size)
 
 long memblock_add(u64 base, u64 size)
 {
-	/* On pSeries LPAR systems, the first MEMBLOCK is our RMO region. */
-	if (base == 0)
-		memblock.rmo_size = size;
-
 	return memblock_add_region(&memblock.memory, base, size);
 
 }
@@ -459,9 +454,6 @@ void __init memblock_enforce_memory_limit(u64 memory_limit)
 		break;
 	}
 
-	if (memblock.memory.regions[0].size < memblock.rmo_size)
-		memblock.rmo_size = memblock.memory.regions[0].size;
-
 	memory_limit = memblock_end_of_DRAM();
 
 	/* And truncate any reserves above the limit also. */
-- 
cgit v1.2.3


From 2898cc4cdf208f15246b7a1c6951d2b126a70fd6 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 4 Aug 2010 13:34:42 +1000
Subject: memblock: Change u64 to phys_addr_t

Let's not waste space and cycles on archs that don't support >32-bit
physical address space.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h |  48 +++++++++----------
 mm/memblock.c            | 118 ++++++++++++++++++++++++-----------------------
 2 files changed, 84 insertions(+), 82 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 71b8edc6ede8..b65045a4ed08 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -21,19 +21,19 @@
 #define MAX_MEMBLOCK_REGIONS 128
 
 struct memblock_region {
-	u64 base;
-	u64 size;
+	phys_addr_t base;
+	phys_addr_t size;
 };
 
 struct memblock_type {
 	unsigned long cnt;
-	u64 size;
+	phys_addr_t size;
 	struct memblock_region regions[MAX_MEMBLOCK_REGIONS+1];
 };
 
 struct memblock {
 	unsigned long debug;
-	u64 current_limit;
+	phys_addr_t current_limit;
 	struct memblock_type memory;
 	struct memblock_type reserved;
 };
@@ -42,34 +42,34 @@ extern struct memblock memblock;
 
 extern void __init memblock_init(void);
 extern void __init memblock_analyze(void);
-extern long memblock_add(u64 base, u64 size);
-extern long memblock_remove(u64 base, u64 size);
-extern long __init memblock_free(u64 base, u64 size);
-extern long __init memblock_reserve(u64 base, u64 size);
+extern long memblock_add(phys_addr_t base, phys_addr_t size);
+extern long memblock_remove(phys_addr_t base, phys_addr_t size);
+extern long __init memblock_free(phys_addr_t base, phys_addr_t size);
+extern long __init memblock_reserve(phys_addr_t base, phys_addr_t size);
 
-extern u64 __init memblock_alloc_nid(u64 size, u64 align, int nid);
-extern u64 __init memblock_alloc(u64 size, u64 align);
+extern phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid);
+extern phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align);
 
 /* Flags for memblock_alloc_base() amd __memblock_alloc_base() */
-#define MEMBLOCK_ALLOC_ANYWHERE	(~(u64)0)
+#define MEMBLOCK_ALLOC_ANYWHERE	(~(phys_addr_t)0)
 #define MEMBLOCK_ALLOC_ACCESSIBLE	0
 
-extern u64 __init memblock_alloc_base(u64 size,
-		u64, u64 max_addr);
-extern u64 __init __memblock_alloc_base(u64 size,
-		u64 align, u64 max_addr);
-extern u64 __init memblock_phys_mem_size(void);
-extern u64 memblock_end_of_DRAM(void);
-extern void __init memblock_enforce_memory_limit(u64 memory_limit);
-extern int memblock_is_memory(u64 addr);
-extern int memblock_is_region_memory(u64 base, u64 size);
-extern int __init memblock_is_reserved(u64 addr);
-extern int memblock_is_region_reserved(u64 base, u64 size);
+extern phys_addr_t __init memblock_alloc_base(phys_addr_t size,
+		phys_addr_t, phys_addr_t max_addr);
+extern phys_addr_t __init __memblock_alloc_base(phys_addr_t size,
+		phys_addr_t align, phys_addr_t max_addr);
+extern phys_addr_t __init memblock_phys_mem_size(void);
+extern phys_addr_t memblock_end_of_DRAM(void);
+extern void __init memblock_enforce_memory_limit(phys_addr_t memory_limit);
+extern int memblock_is_memory(phys_addr_t addr);
+extern int memblock_is_region_memory(phys_addr_t base, phys_addr_t size);
+extern int __init memblock_is_reserved(phys_addr_t addr);
+extern int memblock_is_region_reserved(phys_addr_t base, phys_addr_t size);
 
 extern void memblock_dump_all(void);
 
 /* Provided by the architecture */
-extern u64 memblock_nid_range(u64 start, u64 end, int *nid);
+extern phys_addr_t memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid);
 
 /**
  * memblock_set_current_limit - Set the current allocation limit to allow
@@ -77,7 +77,7 @@ extern u64 memblock_nid_range(u64 start, u64 end, int *nid);
  *                         accessible during boot
  * @limit: New limit value (physical address)
  */
-extern void memblock_set_current_limit(u64 limit);
+extern void memblock_set_current_limit(phys_addr_t limit);
 
 
 /*
diff --git a/mm/memblock.c b/mm/memblock.c
index 73d903ebf3d4..81da63592a68 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -55,13 +55,14 @@ void memblock_dump_all(void)
 	memblock_dump(&memblock.reserved, "reserved");
 }
 
-static unsigned long memblock_addrs_overlap(u64 base1, u64 size1, u64 base2,
-					u64 size2)
+static unsigned long memblock_addrs_overlap(phys_addr_t base1, phys_addr_t size1,
+				       phys_addr_t base2, phys_addr_t size2)
 {
 	return ((base1 < (base2 + size2)) && (base2 < (base1 + size1)));
 }
 
-static long memblock_addrs_adjacent(u64 base1, u64 size1, u64 base2, u64 size2)
+static long memblock_addrs_adjacent(phys_addr_t base1, phys_addr_t size1,
+			       phys_addr_t base2, phys_addr_t size2)
 {
 	if (base2 == base1 + size1)
 		return 1;
@@ -72,12 +73,12 @@ static long memblock_addrs_adjacent(u64 base1, u64 size1, u64 base2, u64 size2)
 }
 
 static long memblock_regions_adjacent(struct memblock_type *type,
-		unsigned long r1, unsigned long r2)
+				 unsigned long r1, unsigned long r2)
 {
-	u64 base1 = type->regions[r1].base;
-	u64 size1 = type->regions[r1].size;
-	u64 base2 = type->regions[r2].base;
-	u64 size2 = type->regions[r2].size;
+	phys_addr_t base1 = type->regions[r1].base;
+	phys_addr_t size1 = type->regions[r1].size;
+	phys_addr_t base2 = type->regions[r2].base;
+	phys_addr_t size2 = type->regions[r2].size;
 
 	return memblock_addrs_adjacent(base1, size1, base2, size2);
 }
@@ -128,7 +129,7 @@ void __init memblock_analyze(void)
 		memblock.memory.size += memblock.memory.regions[i].size;
 }
 
-static long memblock_add_region(struct memblock_type *type, u64 base, u64 size)
+static long memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
 {
 	unsigned long coalesced = 0;
 	long adjacent, i;
@@ -141,8 +142,8 @@ static long memblock_add_region(struct memblock_type *type, u64 base, u64 size)
 
 	/* First try and coalesce this MEMBLOCK with another. */
 	for (i = 0; i < type->cnt; i++) {
-		u64 rgnbase = type->regions[i].base;
-		u64 rgnsize = type->regions[i].size;
+		phys_addr_t rgnbase = type->regions[i].base;
+		phys_addr_t rgnsize = type->regions[i].size;
 
 		if ((rgnbase == base) && (rgnsize == size))
 			/* Already have this region, so we're done */
@@ -192,16 +193,16 @@ static long memblock_add_region(struct memblock_type *type, u64 base, u64 size)
 	return 0;
 }
 
-long memblock_add(u64 base, u64 size)
+long memblock_add(phys_addr_t base, phys_addr_t size)
 {
 	return memblock_add_region(&memblock.memory, base, size);
 
 }
 
-static long __memblock_remove(struct memblock_type *type, u64 base, u64 size)
+static long __memblock_remove(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
 {
-	u64 rgnbegin, rgnend;
-	u64 end = base + size;
+	phys_addr_t rgnbegin, rgnend;
+	phys_addr_t end = base + size;
 	int i;
 
 	rgnbegin = rgnend = 0; /* supress gcc warnings */
@@ -246,17 +247,17 @@ static long __memblock_remove(struct memblock_type *type, u64 base, u64 size)
 	return memblock_add_region(type, end, rgnend - end);
 }
 
-long memblock_remove(u64 base, u64 size)
+long memblock_remove(phys_addr_t base, phys_addr_t size)
 {
 	return __memblock_remove(&memblock.memory, base, size);
 }
 
-long __init memblock_free(u64 base, u64 size)
+long __init memblock_free(phys_addr_t base, phys_addr_t size)
 {
 	return __memblock_remove(&memblock.reserved, base, size);
 }
 
-long __init memblock_reserve(u64 base, u64 size)
+long __init memblock_reserve(phys_addr_t base, phys_addr_t size)
 {
 	struct memblock_type *_rgn = &memblock.reserved;
 
@@ -265,13 +266,13 @@ long __init memblock_reserve(u64 base, u64 size)
 	return memblock_add_region(_rgn, base, size);
 }
 
-long memblock_overlaps_region(struct memblock_type *type, u64 base, u64 size)
+long memblock_overlaps_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
 {
 	unsigned long i;
 
 	for (i = 0; i < type->cnt; i++) {
-		u64 rgnbase = type->regions[i].base;
-		u64 rgnsize = type->regions[i].size;
+		phys_addr_t rgnbase = type->regions[i].base;
+		phys_addr_t rgnsize = type->regions[i].size;
 		if (memblock_addrs_overlap(base, size, rgnbase, rgnsize))
 			break;
 	}
@@ -279,20 +280,20 @@ long memblock_overlaps_region(struct memblock_type *type, u64 base, u64 size)
 	return (i < type->cnt) ? i : -1;
 }
 
-static u64 memblock_align_down(u64 addr, u64 size)
+static phys_addr_t memblock_align_down(phys_addr_t addr, phys_addr_t size)
 {
 	return addr & ~(size - 1);
 }
 
-static u64 memblock_align_up(u64 addr, u64 size)
+static phys_addr_t memblock_align_up(phys_addr_t addr, phys_addr_t size)
 {
 	return (addr + (size - 1)) & ~(size - 1);
 }
 
-static u64 __init memblock_alloc_region(u64 start, u64 end,
-				   u64 size, u64 align)
+static phys_addr_t __init memblock_alloc_region(phys_addr_t start, phys_addr_t end,
+					   phys_addr_t size, phys_addr_t align)
 {
-	u64 base, res_base;
+	phys_addr_t base, res_base;
 	long j;
 
 	base = memblock_align_down((end - size), align);
@@ -301,7 +302,7 @@ static u64 __init memblock_alloc_region(u64 start, u64 end,
 		if (j < 0) {
 			/* this area isn't reserved, take it */
 			if (memblock_add_region(&memblock.reserved, base, size) < 0)
-				base = ~(u64)0;
+				base = ~(phys_addr_t)0;
 			return base;
 		}
 		res_base = memblock.reserved.regions[j].base;
@@ -310,42 +311,43 @@ static u64 __init memblock_alloc_region(u64 start, u64 end,
 		base = memblock_align_down(res_base - size, align);
 	}
 
-	return ~(u64)0;
+	return ~(phys_addr_t)0;
 }
 
-u64 __weak __init memblock_nid_range(u64 start, u64 end, int *nid)
+phys_addr_t __weak __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid)
 {
 	*nid = 0;
 
 	return end;
 }
 
-static u64 __init memblock_alloc_nid_region(struct memblock_region *mp,
-				       u64 size, u64 align, int nid)
+static phys_addr_t __init memblock_alloc_nid_region(struct memblock_region *mp,
+					       phys_addr_t size,
+					       phys_addr_t align, int nid)
 {
-	u64 start, end;
+	phys_addr_t start, end;
 
 	start = mp->base;
 	end = start + mp->size;
 
 	start = memblock_align_up(start, align);
 	while (start < end) {
-		u64 this_end;
+		phys_addr_t this_end;
 		int this_nid;
 
 		this_end = memblock_nid_range(start, end, &this_nid);
 		if (this_nid == nid) {
-			u64 ret = memblock_alloc_region(start, this_end, size, align);
-			if (ret != ~(u64)0)
+			phys_addr_t ret = memblock_alloc_region(start, this_end, size, align);
+			if (ret != ~(phys_addr_t)0)
 				return ret;
 		}
 		start = this_end;
 	}
 
-	return ~(u64)0;
+	return ~(phys_addr_t)0;
 }
 
-u64 __init memblock_alloc_nid(u64 size, u64 align, int nid)
+phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid)
 {
 	struct memblock_type *mem = &memblock.memory;
 	int i;
@@ -359,23 +361,23 @@ u64 __init memblock_alloc_nid(u64 size, u64 align, int nid)
 	size = memblock_align_up(size, align);
 
 	for (i = 0; i < mem->cnt; i++) {
-		u64 ret = memblock_alloc_nid_region(&mem->regions[i],
+		phys_addr_t ret = memblock_alloc_nid_region(&mem->regions[i],
 					       size, align, nid);
-		if (ret != ~(u64)0)
+		if (ret != ~(phys_addr_t)0)
 			return ret;
 	}
 
 	return memblock_alloc(size, align);
 }
 
-u64 __init memblock_alloc(u64 size, u64 align)
+phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align)
 {
 	return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ACCESSIBLE);
 }
 
-u64 __init memblock_alloc_base(u64 size, u64 align, u64 max_addr)
+phys_addr_t __init memblock_alloc_base(phys_addr_t size, phys_addr_t align, phys_addr_t max_addr)
 {
-	u64 alloc;
+	phys_addr_t alloc;
 
 	alloc = __memblock_alloc_base(size, align, max_addr);
 
@@ -386,11 +388,11 @@ u64 __init memblock_alloc_base(u64 size, u64 align, u64 max_addr)
 	return alloc;
 }
 
-u64 __init __memblock_alloc_base(u64 size, u64 align, u64 max_addr)
+phys_addr_t __init __memblock_alloc_base(phys_addr_t size, phys_addr_t align, phys_addr_t max_addr)
 {
 	long i;
-	u64 base = 0;
-	u64 res_base;
+	phys_addr_t base = 0;
+	phys_addr_t res_base;
 
 	BUG_ON(0 == size);
 
@@ -405,26 +407,26 @@ u64 __init __memblock_alloc_base(u64 size, u64 align, u64 max_addr)
 	 * top of memory
 	 */
 	for (i = memblock.memory.cnt - 1; i >= 0; i--) {
-		u64 memblockbase = memblock.memory.regions[i].base;
-		u64 memblocksize = memblock.memory.regions[i].size;
+		phys_addr_t memblockbase = memblock.memory.regions[i].base;
+		phys_addr_t memblocksize = memblock.memory.regions[i].size;
 
 		if (memblocksize < size)
 			continue;
 		base = min(memblockbase + memblocksize, max_addr);
 		res_base = memblock_alloc_region(memblockbase, base, size, align);
-		if (res_base != ~(u64)0)
+		if (res_base != ~(phys_addr_t)0)
 			return res_base;
 	}
 	return 0;
 }
 
 /* You must call memblock_analyze() before this. */
-u64 __init memblock_phys_mem_size(void)
+phys_addr_t __init memblock_phys_mem_size(void)
 {
 	return memblock.memory.size;
 }
 
-u64 memblock_end_of_DRAM(void)
+phys_addr_t memblock_end_of_DRAM(void)
 {
 	int idx = memblock.memory.cnt - 1;
 
@@ -432,10 +434,10 @@ u64 memblock_end_of_DRAM(void)
 }
 
 /* You must call memblock_analyze() after this. */
-void __init memblock_enforce_memory_limit(u64 memory_limit)
+void __init memblock_enforce_memory_limit(phys_addr_t memory_limit)
 {
 	unsigned long i;
-	u64 limit;
+	phys_addr_t limit;
 	struct memblock_region *p;
 
 	if (!memory_limit)
@@ -472,7 +474,7 @@ void __init memblock_enforce_memory_limit(u64 memory_limit)
 	}
 }
 
-static int memblock_search(struct memblock_type *type, u64 addr)
+static int memblock_search(struct memblock_type *type, phys_addr_t addr)
 {
 	unsigned int left = 0, right = type->cnt;
 
@@ -490,17 +492,17 @@ static int memblock_search(struct memblock_type *type, u64 addr)
 	return -1;
 }
 
-int __init memblock_is_reserved(u64 addr)
+int __init memblock_is_reserved(phys_addr_t addr)
 {
 	return memblock_search(&memblock.reserved, addr) != -1;
 }
 
-int memblock_is_memory(u64 addr)
+int memblock_is_memory(phys_addr_t addr)
 {
 	return memblock_search(&memblock.memory, addr) != -1;
 }
 
-int memblock_is_region_memory(u64 base, u64 size)
+int memblock_is_region_memory(phys_addr_t base, phys_addr_t size)
 {
 	int idx = memblock_search(&memblock.reserved, base);
 
@@ -511,13 +513,13 @@ int memblock_is_region_memory(u64 base, u64 size)
 		 memblock.reserved.regions[idx].size) >= (base + size);
 }
 
-int memblock_is_region_reserved(u64 base, u64 size)
+int memblock_is_region_reserved(phys_addr_t base, phys_addr_t size)
 {
 	return memblock_overlaps_region(&memblock.reserved, base, size) >= 0;
 }
 
 
-void __init memblock_set_current_limit(u64 limit)
+void __init memblock_set_current_limit(phys_addr_t limit)
 {
 	memblock.current_limit = limit;
 }
-- 
cgit v1.2.3


From 9d3c30f5a17ec35894eadb7171f724643dce19c3 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:39:04 -0700
Subject: memblock: Remove unused memblock.debug struct member

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index b65045a4ed08..0fe6dd56a4b4 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -32,7 +32,6 @@ struct memblock_type {
 };
 
 struct memblock {
-	unsigned long debug;
 	phys_addr_t current_limit;
 	struct memblock_type memory;
 	struct memblock_type reserved;
@@ -55,9 +54,11 @@ extern phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align);
 #define MEMBLOCK_ALLOC_ACCESSIBLE	0
 
 extern phys_addr_t __init memblock_alloc_base(phys_addr_t size,
-		phys_addr_t, phys_addr_t max_addr);
+					 phys_addr_t align,
+					 phys_addr_t max_addr);
 extern phys_addr_t __init __memblock_alloc_base(phys_addr_t size,
-		phys_addr_t align, phys_addr_t max_addr);
+					   phys_addr_t align,
+					   phys_addr_t max_addr);
 extern phys_addr_t __init memblock_phys_mem_size(void);
 extern phys_addr_t memblock_end_of_DRAM(void);
 extern void __init memblock_enforce_memory_limit(phys_addr_t memory_limit);
-- 
cgit v1.2.3


From 4734b594c6ca1be796d30c82d93fdf5160f45124 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 28 Jul 2010 14:31:29 +1000
Subject: memblock: Remove memblock_type.size and add memblock.memory_size
 instead

Right now, both the "memory" and "reserved" memblock_type structures have
a "size" member. It represents the calculated memory size in the former
case and is unused in the latter.

This moves it out to the main memblock structure instead

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/mm/mem.c    | 2 +-
 include/linux/memblock.h | 2 +-
 mm/memblock.c            | 8 ++++----
 3 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 52df5428ece4..f661f6c527da 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -301,7 +301,7 @@ void __init mem_init(void)
 		swiotlb_init(1);
 #endif
 
-	num_physpages = memblock.memory.size >> PAGE_SHIFT;
+	num_physpages = memblock_phys_mem_size() >> PAGE_SHIFT;
 	high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);
 
 #ifdef CONFIG_NEED_MULTIPLE_NODES
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 0fe6dd56a4b4..c9c7b0f344a5 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -27,12 +27,12 @@ struct memblock_region {
 
 struct memblock_type {
 	unsigned long cnt;
-	phys_addr_t size;
 	struct memblock_region regions[MAX_MEMBLOCK_REGIONS+1];
 };
 
 struct memblock {
 	phys_addr_t current_limit;
+	phys_addr_t memory_size;	/* Updated by memblock_analyze() */
 	struct memblock_type memory;
 	struct memblock_type reserved;
 };
diff --git a/mm/memblock.c b/mm/memblock.c
index 81da63592a68..5ae413e9afd8 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -49,7 +49,7 @@ void memblock_dump_all(void)
 		return;
 
 	pr_info("MEMBLOCK configuration:\n");
-	pr_info(" memory.size = 0x%llx\n", (unsigned long long)memblock.memory.size);
+	pr_info(" memory size = 0x%llx\n", (unsigned long long)memblock.memory_size);
 
 	memblock_dump(&memblock.memory, "memory");
 	memblock_dump(&memblock.reserved, "reserved");
@@ -123,10 +123,10 @@ void __init memblock_analyze(void)
 {
 	int i;
 
-	memblock.memory.size = 0;
+	memblock.memory_size = 0;
 
 	for (i = 0; i < memblock.memory.cnt; i++)
-		memblock.memory.size += memblock.memory.regions[i].size;
+		memblock.memory_size += memblock.memory.regions[i].size;
 }
 
 static long memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
@@ -423,7 +423,7 @@ phys_addr_t __init __memblock_alloc_base(phys_addr_t size, phys_addr_t align, ph
 /* You must call memblock_analyze() before this. */
 phys_addr_t __init memblock_phys_mem_size(void)
 {
-	return memblock.memory.size;
+	return memblock.memory_size;
 }
 
 phys_addr_t memblock_end_of_DRAM(void)
-- 
cgit v1.2.3


From bf23c51f1f49d3960f3cd8e3d2e7f943d9c41042 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:39:06 -0700
Subject: memblock: Move memblock arrays to static storage in memblock.c and
 make their size a variable

This is in preparation for having resizable arrays.

Note that we still allocate one more than needed, this is unchanged from
the previous implementation.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h |  7 ++++---
 mm/memblock.c            | 10 +++++++++-
 2 files changed, 13 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index c9c7b0f344a5..150be938b910 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -18,7 +18,7 @@
 
 #include <asm/memblock.h>
 
-#define MAX_MEMBLOCK_REGIONS 128
+#define INIT_MEMBLOCK_REGIONS 128
 
 struct memblock_region {
 	phys_addr_t base;
@@ -26,8 +26,9 @@ struct memblock_region {
 };
 
 struct memblock_type {
-	unsigned long cnt;
-	struct memblock_region regions[MAX_MEMBLOCK_REGIONS+1];
+	unsigned long cnt;	/* number of regions */
+	unsigned long max;	/* size of the allocated array */
+	struct memblock_region *regions;
 };
 
 struct memblock {
diff --git a/mm/memblock.c b/mm/memblock.c
index 5ae413e9afd8..3c474502d92b 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -18,6 +18,8 @@
 struct memblock memblock;
 
 static int memblock_debug;
+static struct memblock_region memblock_memory_init_regions[INIT_MEMBLOCK_REGIONS + 1];
+static struct memblock_region memblock_reserved_init_regions[INIT_MEMBLOCK_REGIONS + 1];
 
 static int __init early_memblock(char *p)
 {
@@ -104,6 +106,12 @@ static void memblock_coalesce_regions(struct memblock_type *type,
 
 void __init memblock_init(void)
 {
+	/* Hookup the initial arrays */
+	memblock.memory.regions	= memblock_memory_init_regions;
+	memblock.memory.max		= INIT_MEMBLOCK_REGIONS;
+	memblock.reserved.regions	= memblock_reserved_init_regions;
+	memblock.reserved.max	= INIT_MEMBLOCK_REGIONS;
+
 	/* Create a dummy zero size MEMBLOCK which will get coalesced away later.
 	 * This simplifies the memblock_add() code below...
 	 */
@@ -169,7 +177,7 @@ static long memblock_add_region(struct memblock_type *type, phys_addr_t base, ph
 
 	if (coalesced)
 		return coalesced;
-	if (type->cnt >= MAX_MEMBLOCK_REGIONS)
+	if (type->cnt >= type->max)
 		return -1;
 
 	/* Couldn't coalesce the MEMBLOCK, so add it to the sorted table. */
-- 
cgit v1.2.3


From d2cd563ba82c424083b78e0ce97d68bfb04d1242 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:39:14 -0700
Subject: memblock: Add arch function to control coalescing of memblock memory
 regions

Some archs such as ARM want to avoid coalescing accross things such
as the lowmem/highmem boundary or similar. This provides the option
to control it via an arch callback for which a weak default is provided
which always allows coalescing.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h |  2 ++
 mm/memblock.c            | 19 ++++++++++++++++++-
 2 files changed, 20 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 150be938b910..e5e8f9db3a84 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -72,6 +72,8 @@ extern void memblock_dump_all(void);
 
 /* Provided by the architecture */
 extern phys_addr_t memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid);
+extern int memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1,
+				   phys_addr_t addr2, phys_addr_t size2);
 
 /**
  * memblock_set_current_limit - Set the current allocation limit to allow
diff --git a/mm/memblock.c b/mm/memblock.c
index 0787790b1ce0..8715f09434df 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -241,6 +241,12 @@ static int memblock_double_array(struct memblock_type *type)
 	return 0;
 }
 
+extern int __weak memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1,
+					  phys_addr_t addr2, phys_addr_t size2)
+{
+	return 1;
+}
+
 static long memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
 {
 	unsigned long coalesced = 0;
@@ -262,6 +268,10 @@ static long memblock_add_region(struct memblock_type *type, phys_addr_t base, ph
 			return 0;
 
 		adjacent = memblock_addrs_adjacent(base, size, rgnbase, rgnsize);
+		/* Check if arch allows coalescing */
+		if (adjacent != 0 && type == &memblock.memory &&
+		    !memblock_memory_can_coalesce(base, size, rgnbase, rgnsize))
+			break;
 		if (adjacent > 0) {
 			type->regions[i].base -= size;
 			type->regions[i].size += size;
@@ -274,7 +284,14 @@ static long memblock_add_region(struct memblock_type *type, phys_addr_t base, ph
 		}
 	}
 
-	if ((i < type->cnt - 1) && memblock_regions_adjacent(type, i, i+1)) {
+	/* If we plugged a hole, we may want to also coalesce with the
+	 * next region
+	 */
+	if ((i < type->cnt - 1) && memblock_regions_adjacent(type, i, i+1) &&
+	    ((type != &memblock.memory || memblock_memory_can_coalesce(type->regions[i].base,
+							     type->regions[i].size,
+							     type->regions[i+1].base,
+							     type->regions[i+1].size)))) {
 		memblock_coalesce_regions(type, i, i+1);
 		coalesced++;
 	}
-- 
cgit v1.2.3


From c196f76fd5ece716ee3b7fa5dda3576961c0cecc Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:39:16 -0700
Subject: memblock: NUMA allocate can now use early_pfn_map

We now provide a default (weak) implementation of memblock_nid_range()
which uses the early_pfn_map[] if CONFIG_ARCH_POPULATES_NODE_MAP
is set. Sparc still needs to use its own method due to the way
the pages can be scattered between nodes.

This implementation is inefficient due to our main algorithm and
callback construct wanting to work on an ascending addresses bases
while early_pfn_map[] would rather work with nid's (it's unsorted
at that stage). But it should work and we can look into improving
it subsequently, possibly using arch compile options to chose a
different algorithm alltogether.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h |  3 +++
 mm/memblock.c            | 28 +++++++++++++++++++++++++++-
 2 files changed, 30 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index e5e8f9db3a84..82b030244aa7 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -47,6 +47,9 @@ extern long memblock_remove(phys_addr_t base, phys_addr_t size);
 extern long __init memblock_free(phys_addr_t base, phys_addr_t size);
 extern long __init memblock_reserve(phys_addr_t base, phys_addr_t size);
 
+/* The numa aware allocator is only available if
+ * CONFIG_ARCH_POPULATES_NODE_MAP is set
+ */
 extern phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid);
 extern phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align);
 
diff --git a/mm/memblock.c b/mm/memblock.c
index 468ff43a72b4..af7e4d9cf400 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -15,6 +15,7 @@
 #include <linux/init.h>
 #include <linux/bitops.h>
 #include <linux/poison.h>
+#include <linux/pfn.h>
 #include <linux/memblock.h>
 
 struct memblock memblock;
@@ -451,11 +452,36 @@ phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align)
 /*
  * Additional node-local allocators. Search for node memory is bottom up
  * and walks memblock regions within that node bottom-up as well, but allocation
- * within an memblock region is top-down.
+ * within an memblock region is top-down. XXX I plan to fix that at some stage
+ *
+ * WARNING: Only available after early_node_map[] has been populated,
+ * on some architectures, that is after all the calls to add_active_range()
+ * have been done to populate it.
  */
 
 phys_addr_t __weak __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid)
 {
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+	/*
+	 * This code originates from sparc which really wants use to walk by addresses
+	 * and returns the nid. This is not very convenient for early_pfn_map[] users
+	 * as the map isn't sorted yet, and it really wants to be walked by nid.
+	 *
+	 * For now, I implement the inefficient method below which walks the early
+	 * map multiple times. Eventually we may want to use an ARCH config option
+	 * to implement a completely different method for both case.
+	 */
+	unsigned long start_pfn, end_pfn;
+	int i;
+
+	for (i = 0; i < MAX_NUMNODES; i++) {
+		get_pfn_range_for_nid(i, &start_pfn, &end_pfn);
+		if (start < PFN_PHYS(start_pfn) || start >= PFN_PHYS(end_pfn))
+			continue;
+		*nid = i;
+		return min(end, PFN_PHYS(end_pfn));
+	}
+#endif
 	*nid = 0;
 
 	return end;
-- 
cgit v1.2.3


From 9d1e24928e6a0728d1c7c76818ccbd11b93e7ac9 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 6 Jul 2010 15:39:17 -0700
Subject: memblock: Separate memblock_alloc_nid() and memblock_alloc_try_nid()

The former is now strict, it will fail if it cannot honor the allocation
within the node, while the later implements the previous semantic which
falls back to allocating anywhere.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/sparc/mm/init_64.c  |  4 ++--
 include/linux/memblock.h |  6 +++++-
 mm/memblock.c            | 14 ++++++++++++++
 3 files changed, 21 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index 0883113624b9..dc584d26d597 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -820,7 +820,7 @@ static void __init allocate_node_data(int nid)
 	struct pglist_data *p;
 
 #ifdef CONFIG_NEED_MULTIPLE_NODES
-	paddr = memblock_alloc_nid(sizeof(struct pglist_data), SMP_CACHE_BYTES, nid);
+	paddr = memblock_alloc_try_nid(sizeof(struct pglist_data), SMP_CACHE_BYTES, nid);
 	if (!paddr) {
 		prom_printf("Cannot allocate pglist_data for nid[%d]\n", nid);
 		prom_halt();
@@ -840,7 +840,7 @@ static void __init allocate_node_data(int nid)
 	if (p->node_spanned_pages) {
 		num_pages = bootmem_bootmap_pages(p->node_spanned_pages);
 
-		paddr = memblock_alloc_nid(num_pages << PAGE_SHIFT, PAGE_SIZE, nid);
+		paddr = memblock_alloc_try_nid(num_pages << PAGE_SHIFT, PAGE_SIZE, nid);
 		if (!paddr) {
 			prom_printf("Cannot allocate bootmap for nid[%d]\n",
 				  nid);
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 82b030244aa7..c8da03eb7ba3 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -50,7 +50,11 @@ extern long __init memblock_reserve(phys_addr_t base, phys_addr_t size);
 /* The numa aware allocator is only available if
  * CONFIG_ARCH_POPULATES_NODE_MAP is set
  */
-extern phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid);
+extern phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align,
+					int nid);
+extern phys_addr_t __init memblock_alloc_try_nid(phys_addr_t size, phys_addr_t align,
+					    int nid);
+
 extern phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align);
 
 /* Flags for memblock_alloc_base() amd __memblock_alloc_base() */
diff --git a/mm/memblock.c b/mm/memblock.c
index af7e4d9cf400..1802d97c7284 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -537,9 +537,23 @@ phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int n
 			return ret;
 	}
 
+	return 0;
+}
+
+phys_addr_t __init memblock_alloc_try_nid(phys_addr_t size, phys_addr_t align, int nid)
+{
+	phys_addr_t res = memblock_alloc_nid(size, align, nid);
+
+	if (res)
+		return res;
 	return memblock_alloc(size, align);
 }
 
+
+/*
+ * Remaining API functions
+ */
+
 /* You must call memblock_analyze() before this. */
 phys_addr_t __init memblock_phys_mem_size(void)
 {
-- 
cgit v1.2.3


From 5e63cf43af844ed30acc278b38b8c9bc51eba493 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 28 Jul 2010 15:07:21 +1000
Subject: memblock: Expose some memblock bits for use by x86

This exposes memblock_debug and associated memblock_dbg() macro,
along with memblock_can_resize so that x86 can use these when
ported to use memblock

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 5 +++++
 mm/memblock.c            | 3 ++-
 2 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index c8da03eb7ba3..eed0f9b8e526 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -39,6 +39,11 @@ struct memblock {
 };
 
 extern struct memblock memblock;
+extern int memblock_debug;
+extern int memblock_can_resize;
+
+#define memblock_dbg(fmt, ...) \
+	if (memblock_debug) printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__)
 
 extern void __init memblock_init(void);
 extern void __init memblock_analyze(void);
diff --git a/mm/memblock.c b/mm/memblock.c
index cc15be29fd0a..5499ab162b9d 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -22,7 +22,8 @@
 
 struct memblock memblock;
 
-static int memblock_debug, memblock_can_resize;
+int memblock_debug;
+int memblock_can_resize;
 static struct memblock_region memblock_memory_init_regions[INIT_MEMBLOCK_REGIONS + 1];
 static struct memblock_region memblock_reserved_init_regions[INIT_MEMBLOCK_REGIONS + 1];
 
-- 
cgit v1.2.3


From 37d8d4bf489e39eedc9537f8616fe87879b13cb0 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 28 Jul 2010 15:20:58 +1000
Subject: memblock: Export MEMBLOCK_ERROR

will used by x86 memblock_x86_find_in_range_node and nobootmem replacement

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 3 ++-
 mm/memblock.c            | 2 --
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index eed0f9b8e526..1a9c29cc92fa 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -18,7 +18,8 @@
 
 #include <asm/memblock.h>
 
-#define INIT_MEMBLOCK_REGIONS 128
+#define INIT_MEMBLOCK_REGIONS	128
+#define MEMBLOCK_ERROR		(~(phys_addr_t)0)
 
 struct memblock_region {
 	phys_addr_t base;
diff --git a/mm/memblock.c b/mm/memblock.c
index c3703abf057e..85cfa1d3ab28 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -27,8 +27,6 @@ int memblock_can_resize;
 static struct memblock_region memblock_memory_init_regions[INIT_MEMBLOCK_REGIONS + 1];
 static struct memblock_region memblock_reserved_init_regions[INIT_MEMBLOCK_REGIONS + 1];
 
-#define MEMBLOCK_ERROR	(~(phys_addr_t)0)
-
 /* inline so we don't get a warning when pr_debug is compiled out */
 static inline const char *memblock_type_name(struct memblock_type *type)
 {
-- 
cgit v1.2.3


From 25818f0f288cd5333ba5a90ad6dde3def4c4ff58 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 28 Jul 2010 15:25:10 +1000
Subject: memblock: Make MEMBLOCK_ERROR be 0

And ensure we don't hand out 0 as a valid allocation. We put the
low limit at PAGE_SIZE arbitrarily.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 2 +-
 mm/memblock.c            | 6 ++++++
 2 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 1a9c29cc92fa..dfa64494ced1 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -19,7 +19,7 @@
 #include <asm/memblock.h>
 
 #define INIT_MEMBLOCK_REGIONS	128
-#define MEMBLOCK_ERROR		(~(phys_addr_t)0)
+#define MEMBLOCK_ERROR		0
 
 struct memblock_region {
 	phys_addr_t base;
diff --git a/mm/memblock.c b/mm/memblock.c
index 85cfa1d3ab28..cb520df2a414 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -105,6 +105,12 @@ static phys_addr_t __init memblock_find_region(phys_addr_t start, phys_addr_t en
 	phys_addr_t base, res_base;
 	long j;
 
+	/* Prevent allocations returning 0 as it's also used to
+	 * indicate an allocation failure
+	 */
+	if (start == 0)
+		start = PAGE_SIZE;
+
 	base = memblock_align_down((end - size), align);
 	while (start <= base) {
 		j = memblock_overlaps_region(&memblock.reserved, base, size);
-- 
cgit v1.2.3


From f0b37fad9a63217c39997b2d2b31f44e3d8be727 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 28 Jul 2010 15:28:21 +1000
Subject: memblock: Protect memblock.h with CONFIG_HAVE_MEMBLOCK

This should make it easier to catch/debug incorrect use when
the CONFIG_ option isn't set.

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index dfa64494ced1..c24b27849096 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -2,6 +2,7 @@
 #define _LINUX_MEMBLOCK_H
 #ifdef __KERNEL__
 
+#ifdef CONFIG_HAVE_MEMBLOCK
 /*
  * Logical memory blocks.
  *
@@ -148,6 +149,8 @@ static inline unsigned long memblock_region_pages(const struct memblock_region *
 	     region++)
 
 
+#endif /* CONFIG_HAVE_MEMBLOCK */
+
 #endif /* __KERNEL__ */
 
 #endif /* _LINUX_MEMBLOCK_H */
-- 
cgit v1.2.3


From 10d0643988e976360eb3497dcafb55b393b8e480 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 28 Jul 2010 15:43:02 +1000
Subject: memblock: Option for the architecture to put memblock into the .init
 section

Arch code can define ARCH_DISCARD_MEMBLOCK in asm/memblock.h,
which in turns causes memblock code and data to go respectively
into the .init and .initdata sections. This will be used by the
x86 architecture.

If ARCH_DISCARD_MEMBLOCK is defined, the debugfs files to inspect
the memblock arrays after boot are not created.

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h |  8 ++++++++
 mm/memblock.c            | 48 ++++++++++++++++++++++++------------------------
 2 files changed, 32 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index c24b27849096..3978e6a8e824 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -149,6 +149,14 @@ static inline unsigned long memblock_region_pages(const struct memblock_region *
 	     region++)
 
 
+#ifdef ARCH_DISCARD_MEMBLOCK
+#define __init_memblock __init
+#define __initdata_memblock __initdata
+#else
+#define __init_memblock
+#define __initdata_memblock
+#endif
+
 #endif /* CONFIG_HAVE_MEMBLOCK */
 
 #endif /* __KERNEL__ */
diff --git a/mm/memblock.c b/mm/memblock.c
index cb520df2a414..a17faea37d47 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -20,12 +20,12 @@
 #include <linux/seq_file.h>
 #include <linux/memblock.h>
 
-struct memblock memblock;
+struct memblock memblock __initdata_memblock;
 
-int memblock_debug;
-int memblock_can_resize;
-static struct memblock_region memblock_memory_init_regions[INIT_MEMBLOCK_REGIONS + 1];
-static struct memblock_region memblock_reserved_init_regions[INIT_MEMBLOCK_REGIONS + 1];
+int memblock_debug __initdata_memblock;
+int memblock_can_resize __initdata_memblock;
+static struct memblock_region memblock_memory_init_regions[INIT_MEMBLOCK_REGIONS + 1] __initdata_memblock;
+static struct memblock_region memblock_reserved_init_regions[INIT_MEMBLOCK_REGIONS + 1] __initdata_memblock;
 
 /* inline so we don't get a warning when pr_debug is compiled out */
 static inline const char *memblock_type_name(struct memblock_type *type)
@@ -42,23 +42,23 @@ static inline const char *memblock_type_name(struct memblock_type *type)
  * Address comparison utilities
  */
 
-static phys_addr_t memblock_align_down(phys_addr_t addr, phys_addr_t size)
+static phys_addr_t __init_memblock memblock_align_down(phys_addr_t addr, phys_addr_t size)
 {
 	return addr & ~(size - 1);
 }
 
-static phys_addr_t memblock_align_up(phys_addr_t addr, phys_addr_t size)
+static phys_addr_t __init_memblock memblock_align_up(phys_addr_t addr, phys_addr_t size)
 {
 	return (addr + (size - 1)) & ~(size - 1);
 }
 
-static unsigned long memblock_addrs_overlap(phys_addr_t base1, phys_addr_t size1,
+static unsigned long __init_memblock memblock_addrs_overlap(phys_addr_t base1, phys_addr_t size1,
 				       phys_addr_t base2, phys_addr_t size2)
 {
 	return ((base1 < (base2 + size2)) && (base2 < (base1 + size1)));
 }
 
-static long memblock_addrs_adjacent(phys_addr_t base1, phys_addr_t size1,
+static long __init_memblock memblock_addrs_adjacent(phys_addr_t base1, phys_addr_t size1,
 			       phys_addr_t base2, phys_addr_t size2)
 {
 	if (base2 == base1 + size1)
@@ -69,7 +69,7 @@ static long memblock_addrs_adjacent(phys_addr_t base1, phys_addr_t size1,
 	return 0;
 }
 
-static long memblock_regions_adjacent(struct memblock_type *type,
+static long __init_memblock memblock_regions_adjacent(struct memblock_type *type,
 				 unsigned long r1, unsigned long r2)
 {
 	phys_addr_t base1 = type->regions[r1].base;
@@ -80,7 +80,7 @@ static long memblock_regions_adjacent(struct memblock_type *type,
 	return memblock_addrs_adjacent(base1, size1, base2, size2);
 }
 
-long memblock_overlaps_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
+long __init_memblock memblock_overlaps_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
 {
 	unsigned long i;
 
@@ -162,7 +162,7 @@ static phys_addr_t __init memblock_find_base(phys_addr_t size, phys_addr_t align
 	return MEMBLOCK_ERROR;
 }
 
-static void memblock_remove_region(struct memblock_type *type, unsigned long r)
+static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r)
 {
 	unsigned long i;
 
@@ -174,7 +174,7 @@ static void memblock_remove_region(struct memblock_type *type, unsigned long r)
 }
 
 /* Assumption: base addr of region 1 < base addr of region 2 */
-static void memblock_coalesce_regions(struct memblock_type *type,
+static void __init_memblock memblock_coalesce_regions(struct memblock_type *type,
 		unsigned long r1, unsigned long r2)
 {
 	type->regions[r1].size += type->regions[r2].size;
@@ -184,7 +184,7 @@ static void memblock_coalesce_regions(struct memblock_type *type,
 /* Defined below but needed now */
 static long memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size);
 
-static int memblock_double_array(struct memblock_type *type)
+static int __init_memblock memblock_double_array(struct memblock_type *type)
 {
 	struct memblock_region *new_array, *old_array;
 	phys_addr_t old_size, new_size, addr;
@@ -255,13 +255,13 @@ static int memblock_double_array(struct memblock_type *type)
 	return 0;
 }
 
-extern int __weak memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1,
+extern int __init_memblock __weak memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1,
 					  phys_addr_t addr2, phys_addr_t size2)
 {
 	return 1;
 }
 
-static long memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
+static long __init_memblock memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
 {
 	unsigned long coalesced = 0;
 	long adjacent, i;
@@ -348,13 +348,13 @@ static long memblock_add_region(struct memblock_type *type, phys_addr_t base, ph
 	return 0;
 }
 
-long memblock_add(phys_addr_t base, phys_addr_t size)
+long __init_memblock memblock_add(phys_addr_t base, phys_addr_t size)
 {
 	return memblock_add_region(&memblock.memory, base, size);
 
 }
 
-static long __memblock_remove(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
+static long __init_memblock __memblock_remove(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
 {
 	phys_addr_t rgnbegin, rgnend;
 	phys_addr_t end = base + size;
@@ -402,7 +402,7 @@ static long __memblock_remove(struct memblock_type *type, phys_addr_t base, phys
 	return memblock_add_region(type, end, rgnend - end);
 }
 
-long memblock_remove(phys_addr_t base, phys_addr_t size)
+long __init_memblock memblock_remove(phys_addr_t base, phys_addr_t size)
 {
 	return __memblock_remove(&memblock.memory, base, size);
 }
@@ -568,7 +568,7 @@ phys_addr_t __init memblock_phys_mem_size(void)
 	return memblock.memory_size;
 }
 
-phys_addr_t memblock_end_of_DRAM(void)
+phys_addr_t __init_memblock memblock_end_of_DRAM(void)
 {
 	int idx = memblock.memory.cnt - 1;
 
@@ -655,7 +655,7 @@ int memblock_is_region_memory(phys_addr_t base, phys_addr_t size)
 		 memblock.reserved.regions[idx].size) >= (base + size);
 }
 
-int memblock_is_region_reserved(phys_addr_t base, phys_addr_t size)
+int __init_memblock memblock_is_region_reserved(phys_addr_t base, phys_addr_t size)
 {
 	return memblock_overlaps_region(&memblock.reserved, base, size) >= 0;
 }
@@ -666,7 +666,7 @@ void __init memblock_set_current_limit(phys_addr_t limit)
 	memblock.current_limit = limit;
 }
 
-static void memblock_dump(struct memblock_type *region, char *name)
+static void __init_memblock memblock_dump(struct memblock_type *region, char *name)
 {
 	unsigned long long base, size;
 	int i;
@@ -682,7 +682,7 @@ static void memblock_dump(struct memblock_type *region, char *name)
 	}
 }
 
-void memblock_dump_all(void)
+void __init_memblock memblock_dump_all(void)
 {
 	if (!memblock_debug)
 		return;
@@ -748,7 +748,7 @@ static int __init early_memblock(char *p)
 }
 early_param("memblock", early_memblock);
 
-#ifdef CONFIG_DEBUG_FS
+#if defined(CONFIG_DEBUG_FS) && !defined(ARCH_DISCARD_MEMBLOCK)
 
 static int memblock_debug_show(struct seq_file *m, void *private)
 {
-- 
cgit v1.2.3


From 5303b68f57c227c27193a14e57dd12be27cd670f Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 28 Jul 2010 15:38:40 +1000
Subject: memblock: Add memblock_find_in_range()

This is a wrapper for memblock_find_base() using slightly different
arguments (start,end instead of start,size for example) in order to
make it easier to convert existing arch/x86 code.

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/memblock.h | 2 ++
 mm/memblock.c            | 8 ++++++++
 2 files changed, 10 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 3978e6a8e824..4df09bdcae42 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -47,6 +47,8 @@ extern int memblock_can_resize;
 #define memblock_dbg(fmt, ...) \
 	if (memblock_debug) printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__)
 
+u64 memblock_find_in_range(u64 start, u64 end, u64 size, u64 align);
+
 extern void __init memblock_init(void);
 extern void __init memblock_analyze(void);
 extern long memblock_add(phys_addr_t base, phys_addr_t size);
diff --git a/mm/memblock.c b/mm/memblock.c
index a17faea37d47..b7ab10a2ef46 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -162,6 +162,14 @@ static phys_addr_t __init memblock_find_base(phys_addr_t size, phys_addr_t align
 	return MEMBLOCK_ERROR;
 }
 
+/*
+ * Find a free area with specified alignment in a specific range.
+ */
+u64 __init_memblock memblock_find_in_range(u64 start, u64 end, u64 size, u64 align)
+{
+	return memblock_find_base(size, align, start, end);
+}
+
 static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r)
 {
 	unsigned long i;
-- 
cgit v1.2.3


From 04600794958f1833f5571c6cde40f260ab557f55 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 5 Aug 2010 17:45:15 +0200
Subject: cfg80211: support sysfs namespaces

Enable using network namespaces with
wireless devices even when sysfs is
enabled using the same infrastructure
that was built for netdevs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/netdevice.h | 2 ++
 net/core/net-sysfs.c      | 3 ++-
 net/wireless/core.c       | 7 ++++++-
 net/wireless/sysfs.c      | 9 +++++++++
 4 files changed, 19 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 46c36ffe20ee..a4b14fd81c6a 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2171,6 +2171,8 @@ extern void dev_seq_stop(struct seq_file *seq, void *v);
 extern int netdev_class_create_file(struct class_attribute *class_attr);
 extern void netdev_class_remove_file(struct class_attribute *class_attr);
 
+extern struct kobj_ns_type_operations net_ns_type_operations;
+
 extern char *netdev_drivername(const struct net_device *dev, char *buffer, int len);
 
 extern void linkwatch_run_queue(void);
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index af4dfbadf2a0..7d748542d97e 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -789,12 +789,13 @@ static const void *net_netlink_ns(struct sock *sk)
 	return sock_net(sk);
 }
 
-static struct kobj_ns_type_operations net_ns_type_operations = {
+struct kobj_ns_type_operations net_ns_type_operations = {
 	.type = KOBJ_NS_TYPE_NET,
 	.current_ns = net_current_ns,
 	.netlink_ns = net_netlink_ns,
 	.initial_ns = net_initial_ns,
 };
+EXPORT_SYMBOL_GPL(net_ns_type_operations);
 
 static void net_kobj_ns_exit(struct net *net)
 {
diff --git a/net/wireless/core.c b/net/wireless/core.c
index 541e2fff5e9c..c70909c3eae4 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -253,11 +253,16 @@ int cfg80211_switch_netns(struct cfg80211_registered_device *rdev,
 			WARN_ON(err);
 			wdev->netdev->features |= NETIF_F_NETNS_LOCAL;
 		}
+
+		return err;
 	}
 
 	wiphy_net_set(&rdev->wiphy, net);
 
-	return err;
+	err = device_rename(&rdev->wiphy.dev, dev_name(&rdev->wiphy.dev));
+	WARN_ON(err);
+
+	return 0;
 }
 
 static void cfg80211_rfkill_poll(struct rfkill *rfkill, void *data)
diff --git a/net/wireless/sysfs.c b/net/wireless/sysfs.c
index 9f2cef3e0ca0..74a9e3cce452 100644
--- a/net/wireless/sysfs.c
+++ b/net/wireless/sysfs.c
@@ -110,6 +110,13 @@ static int wiphy_resume(struct device *dev)
 	return ret;
 }
 
+static const void *wiphy_namespace(struct device *d)
+{
+	struct wiphy *wiphy = container_of(d, struct wiphy, dev);
+
+	return wiphy_net(wiphy);
+}
+
 struct class ieee80211_class = {
 	.name = "ieee80211",
 	.owner = THIS_MODULE,
@@ -120,6 +127,8 @@ struct class ieee80211_class = {
 #endif
 	.suspend = wiphy_suspend,
 	.resume = wiphy_resume,
+	.ns_type = &net_ns_type_operations,
+	.namespace = wiphy_namespace,
 };
 
 int wiphy_sysfs_init(void)
-- 
cgit v1.2.3


From bfb564e7391340638afe4ad67744a8f3858e7566 Mon Sep 17 00:00:00 2001
From: Krishna Kumar <krkumar2@in.ibm.com>
Date: Wed, 4 Aug 2010 06:15:52 +0000
Subject: core: Factor out flow calculation from get_rps_cpu

Factor out flow calculation code from get_rps_cpu, since other
functions can use the same code.

Revisions:

v2 (Ben): Separate flow calcuation out and use in select queue.
v3 (Arnd): Don't re-implement MIN.
v4 (Changli): skb->data points to ethernet header in macvtap, and
	make a fast path. Tested macvtap with this patch.
v5 (Changli):
	- Cache skb->rxhash in skb_get_rxhash
	- macvtap may not have pow(2) queues, so change code for
	  queue selection.
    (Arnd):
	- Use first available queue if all fails.

Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h |   9 +++++
 net/core/dev.c         | 106 +++++++++++++++++++++++++++++--------------------
 2 files changed, 71 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 77eb60d2b496..d8050382b189 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -558,6 +558,15 @@ extern unsigned int   skb_find_text(struct sk_buff *skb, unsigned int from,
 				    unsigned int to, struct ts_config *config,
 				    struct ts_state *state);
 
+extern __u32 __skb_get_rxhash(struct sk_buff *skb);
+static inline __u32 skb_get_rxhash(struct sk_buff *skb)
+{
+	if (!skb->rxhash)
+		skb->rxhash = __skb_get_rxhash(skb);
+
+	return skb->rxhash;
+}
+
 #ifdef NET_SKBUFF_DATA_USES_OFFSET
 static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
 {
diff --git a/net/core/dev.c b/net/core/dev.c
index 1ae654391442..586a11cb4398 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2259,69 +2259,41 @@ static inline void ____napi_schedule(struct softnet_data *sd,
 	__raise_softirq_irqoff(NET_RX_SOFTIRQ);
 }
 
-#ifdef CONFIG_RPS
-
-/* One global table that all flow-based protocols share. */
-struct rps_sock_flow_table *rps_sock_flow_table __read_mostly;
-EXPORT_SYMBOL(rps_sock_flow_table);
-
 /*
- * get_rps_cpu is called from netif_receive_skb and returns the target
- * CPU from the RPS map of the receiving queue for a given skb.
- * rcu_read_lock must be held on entry.
+ * __skb_get_rxhash: calculate a flow hash based on src/dst addresses
+ * and src/dst port numbers. Returns a non-zero hash number on success
+ * and 0 on failure.
  */
-static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb,
-		       struct rps_dev_flow **rflowp)
+__u32 __skb_get_rxhash(struct sk_buff *skb)
 {
+	int nhoff, hash = 0;
 	struct ipv6hdr *ip6;
 	struct iphdr *ip;
-	struct netdev_rx_queue *rxqueue;
-	struct rps_map *map;
-	struct rps_dev_flow_table *flow_table;
-	struct rps_sock_flow_table *sock_flow_table;
-	int cpu = -1;
 	u8 ip_proto;
-	u16 tcpu;
 	u32 addr1, addr2, ihl;
 	union {
 		u32 v32;
 		u16 v16[2];
 	} ports;
 
-	if (skb_rx_queue_recorded(skb)) {
-		u16 index = skb_get_rx_queue(skb);
-		if (unlikely(index >= dev->num_rx_queues)) {
-			WARN_ONCE(dev->num_rx_queues > 1, "%s received packet "
-				"on queue %u, but number of RX queues is %u\n",
-				dev->name, index, dev->num_rx_queues);
-			goto done;
-		}
-		rxqueue = dev->_rx + index;
-	} else
-		rxqueue = dev->_rx;
-
-	if (!rxqueue->rps_map && !rxqueue->rps_flow_table)
-		goto done;
-
-	if (skb->rxhash)
-		goto got_hash; /* Skip hash computation on packet header */
+	nhoff = skb_network_offset(skb);
 
 	switch (skb->protocol) {
 	case __constant_htons(ETH_P_IP):
-		if (!pskb_may_pull(skb, sizeof(*ip)))
+		if (!pskb_may_pull(skb, sizeof(*ip) + nhoff))
 			goto done;
 
-		ip = (struct iphdr *) skb->data;
+		ip = (struct iphdr *) skb->data + nhoff;
 		ip_proto = ip->protocol;
 		addr1 = (__force u32) ip->saddr;
 		addr2 = (__force u32) ip->daddr;
 		ihl = ip->ihl;
 		break;
 	case __constant_htons(ETH_P_IPV6):
-		if (!pskb_may_pull(skb, sizeof(*ip6)))
+		if (!pskb_may_pull(skb, sizeof(*ip6) + nhoff))
 			goto done;
 
-		ip6 = (struct ipv6hdr *) skb->data;
+		ip6 = (struct ipv6hdr *) skb->data + nhoff;
 		ip_proto = ip6->nexthdr;
 		addr1 = (__force u32) ip6->saddr.s6_addr32[3];
 		addr2 = (__force u32) ip6->daddr.s6_addr32[3];
@@ -2330,6 +2302,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb,
 	default:
 		goto done;
 	}
+
 	switch (ip_proto) {
 	case IPPROTO_TCP:
 	case IPPROTO_UDP:
@@ -2338,8 +2311,9 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb,
 	case IPPROTO_AH:
 	case IPPROTO_SCTP:
 	case IPPROTO_UDPLITE:
-		if (pskb_may_pull(skb, (ihl * 4) + 4)) {
-			ports.v32 = * (__force u32 *) (skb->data + (ihl * 4));
+		if (pskb_may_pull(skb, (ihl * 4) + 4 + nhoff)) {
+			ports.v32 = * (__force u32 *) (skb->data + nhoff +
+						       (ihl * 4));
 			if (ports.v16[1] < ports.v16[0])
 				swap(ports.v16[0], ports.v16[1]);
 			break;
@@ -2352,11 +2326,55 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb,
 	/* get a consistent hash (same value on both flow directions) */
 	if (addr2 < addr1)
 		swap(addr1, addr2);
-	skb->rxhash = jhash_3words(addr1, addr2, ports.v32, hashrnd);
-	if (!skb->rxhash)
-		skb->rxhash = 1;
 
-got_hash:
+	hash = jhash_3words(addr1, addr2, ports.v32, hashrnd);
+	if (!hash)
+		hash = 1;
+
+done:
+	return hash;
+}
+EXPORT_SYMBOL(__skb_get_rxhash);
+
+#ifdef CONFIG_RPS
+
+/* One global table that all flow-based protocols share. */
+struct rps_sock_flow_table *rps_sock_flow_table __read_mostly;
+EXPORT_SYMBOL(rps_sock_flow_table);
+
+/*
+ * get_rps_cpu is called from netif_receive_skb and returns the target
+ * CPU from the RPS map of the receiving queue for a given skb.
+ * rcu_read_lock must be held on entry.
+ */
+static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb,
+		       struct rps_dev_flow **rflowp)
+{
+	struct netdev_rx_queue *rxqueue;
+	struct rps_map *map;
+	struct rps_dev_flow_table *flow_table;
+	struct rps_sock_flow_table *sock_flow_table;
+	int cpu = -1;
+	u16 tcpu;
+
+	if (skb_rx_queue_recorded(skb)) {
+		u16 index = skb_get_rx_queue(skb);
+		if (unlikely(index >= dev->num_rx_queues)) {
+			WARN_ONCE(dev->num_rx_queues > 1, "%s received packet "
+				"on queue %u, but number of RX queues is %u\n",
+				dev->name, index, dev->num_rx_queues);
+			goto done;
+		}
+		rxqueue = dev->_rx + index;
+	} else
+		rxqueue = dev->_rx;
+
+	if (!rxqueue->rps_map && !rxqueue->rps_flow_table)
+		goto done;
+
+	if (!skb_get_rxhash(skb))
+		goto done;
+
 	flow_table = rcu_dereference(rxqueue->rps_flow_table);
 	sock_flow_table = rcu_dereference(rps_sock_flow_table);
 	if (flow_table && sock_flow_table) {
-- 
cgit v1.2.3


From 1565c7c1c4c8e931bdba66abc8aa6f141a406872 Mon Sep 17 00:00:00 2001
From: Krishna Kumar <krkumar2@in.ibm.com>
Date: Wed, 4 Aug 2010 06:15:59 +0000
Subject: macvtap: Implement multiqueue for macvtap driver

Implement multiqueue facility for macvtap driver. The idea is that
a macvtap device can be opened multiple times and the fd's can be
used to register eg, as backend for vhost.

Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/macvtap.c      | 99 ++++++++++++++++++++++++++++++++++++++--------
 include/linux/if_macvlan.h |  9 ++++-
 2 files changed, 90 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 3b1c54a9c6ef..42567279843e 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -84,26 +84,45 @@ static const struct proto_ops macvtap_socket_ops;
 static DEFINE_SPINLOCK(macvtap_lock);
 
 /*
- * Choose the next free queue, for now there is only one
+ * get_slot: return a [unused/occupied] slot in vlan->taps[]:
+ *	- if 'q' is NULL, return the first empty slot;
+ *	- otherwise, return the slot this pointer occupies.
  */
+static int get_slot(struct macvlan_dev *vlan, struct macvtap_queue *q)
+{
+	int i;
+
+	for (i = 0; i < MAX_MACVTAP_QUEUES; i++) {
+		if (rcu_dereference(vlan->taps[i]) == q)
+			return i;
+	}
+
+	/* Should never happen */
+	BUG_ON(1);
+}
+
 static int macvtap_set_queue(struct net_device *dev, struct file *file,
 				struct macvtap_queue *q)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
+	int index;
 	int err = -EBUSY;
 
 	spin_lock(&macvtap_lock);
-	if (rcu_dereference(vlan->tap))
+	if (vlan->numvtaps == MAX_MACVTAP_QUEUES)
 		goto out;
 
 	err = 0;
+	index = get_slot(vlan, NULL);
 	rcu_assign_pointer(q->vlan, vlan);
-	rcu_assign_pointer(vlan->tap, q);
+	rcu_assign_pointer(vlan->taps[index], q);
 	sock_hold(&q->sk);
 
 	q->file = file;
 	file->private_data = q;
 
+	vlan->numvtaps++;
+
 out:
 	spin_unlock(&macvtap_lock);
 	return err;
@@ -124,9 +143,12 @@ static void macvtap_put_queue(struct macvtap_queue *q)
 	spin_lock(&macvtap_lock);
 	vlan = rcu_dereference(q->vlan);
 	if (vlan) {
-		rcu_assign_pointer(vlan->tap, NULL);
+		int index = get_slot(vlan, q);
+
+		rcu_assign_pointer(vlan->taps[index], NULL);
 		rcu_assign_pointer(q->vlan, NULL);
 		sock_put(&q->sk);
+		--vlan->numvtaps;
 	}
 
 	spin_unlock(&macvtap_lock);
@@ -136,39 +158,82 @@ static void macvtap_put_queue(struct macvtap_queue *q)
 }
 
 /*
- * Since we only support one queue, just dereference the pointer.
+ * Select a queue based on the rxq of the device on which this packet
+ * arrived. If the incoming device is not mq, calculate a flow hash
+ * to select a queue. If all fails, find the first available queue.
+ * Cache vlan->numvtaps since it can become zero during the execution
+ * of this function.
  */
 static struct macvtap_queue *macvtap_get_queue(struct net_device *dev,
 					       struct sk_buff *skb)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
+	struct macvtap_queue *tap = NULL;
+	int numvtaps = vlan->numvtaps;
+	__u32 rxq;
+
+	if (!numvtaps)
+		goto out;
+
+	if (likely(skb_rx_queue_recorded(skb))) {
+		rxq = skb_get_rx_queue(skb);
+
+		while (unlikely(rxq >= numvtaps))
+			rxq -= numvtaps;
+
+		tap = rcu_dereference(vlan->taps[rxq]);
+		if (tap)
+			goto out;
+	}
+
+	/* Check if we can use flow to select a queue */
+	rxq = skb_get_rxhash(skb);
+	if (rxq) {
+		tap = rcu_dereference(vlan->taps[rxq % numvtaps]);
+		if (tap)
+			goto out;
+	}
 
-	return rcu_dereference(vlan->tap);
+	/* Everything failed - find first available queue */
+	for (rxq = 0; rxq < MAX_MACVTAP_QUEUES; rxq++) {
+		tap = rcu_dereference(vlan->taps[rxq]);
+		if (tap)
+			break;
+	}
+
+out:
+	return tap;
 }
 
 /*
  * The net_device is going away, give up the reference
- * that it holds on the queue (all the queues one day)
- * and safely set the pointer from the queues to NULL.
+ * that it holds on all queues and safely set the pointer
+ * from the queues to NULL.
  */
 static void macvtap_del_queues(struct net_device *dev)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
-	struct macvtap_queue *q;
+	struct macvtap_queue *q, *qlist[MAX_MACVTAP_QUEUES];
+	int i, j = 0;
 
+	/* macvtap_put_queue can free some slots, so go through all slots */
 	spin_lock(&macvtap_lock);
-	q = rcu_dereference(vlan->tap);
-	if (!q) {
-		spin_unlock(&macvtap_lock);
-		return;
+	for (i = 0; i < MAX_MACVTAP_QUEUES && vlan->numvtaps; i++) {
+		q = rcu_dereference(vlan->taps[i]);
+		if (q) {
+			qlist[j++] = q;
+			rcu_assign_pointer(vlan->taps[i], NULL);
+			rcu_assign_pointer(q->vlan, NULL);
+			vlan->numvtaps--;
+		}
 	}
-
-	rcu_assign_pointer(vlan->tap, NULL);
-	rcu_assign_pointer(q->vlan, NULL);
+	BUG_ON(vlan->numvtaps != 0);
 	spin_unlock(&macvtap_lock);
 
 	synchronize_rcu();
-	sock_put(&q->sk);
+
+	for (--j; j >= 0; j--)
+		sock_put(&qlist[j]->sk);
 }
 
 /*
diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h
index 35280b302290..8a2fd66a8b5f 100644
--- a/include/linux/if_macvlan.h
+++ b/include/linux/if_macvlan.h
@@ -40,6 +40,12 @@ struct macvlan_rx_stats {
 	unsigned long		rx_errors;
 };
 
+/*
+ * Maximum times a macvtap device can be opened. This can be used to
+ * configure the number of receive queue, e.g. for multiqueue virtio.
+ */
+#define MAX_MACVTAP_QUEUES	(NR_CPUS < 16 ? NR_CPUS : 16)
+
 struct macvlan_dev {
 	struct net_device	*dev;
 	struct list_head	list;
@@ -50,7 +56,8 @@ struct macvlan_dev {
 	enum macvlan_mode	mode;
 	int (*receive)(struct sk_buff *skb);
 	int (*forward)(struct net_device *dev, struct sk_buff *skb);
-	struct macvtap_queue	*tap;
+	struct macvtap_queue	*taps[MAX_MACVTAP_QUEUES];
+	int			numvtaps;
 };
 
 static inline void macvlan_count_rx(const struct macvlan_dev *vlan,
-- 
cgit v1.2.3


From f1b499f029c5dde85d46a8811353c62f29157541 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Thu, 5 Aug 2010 17:10:53 +0200
Subject: lockdep: Remove __debug_show_held_locks

There is no longer any functional difference between
__debug_show_held_locks() and debug_show_held_locks(),
so remove the former.

Signed-off-by: John Kacur <jkacur@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1281021054-4228-1-git-send-email-jkacur@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/debug_locks.h | 5 -----
 kernel/hung_task.c          | 2 +-
 kernel/lockdep.c            | 8 +-------
 3 files changed, 2 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/debug_locks.h b/include/linux/debug_locks.h
index 29b3ce3f2a1d..2833452ea01c 100644
--- a/include/linux/debug_locks.h
+++ b/include/linux/debug_locks.h
@@ -49,7 +49,6 @@ struct task_struct;
 
 #ifdef CONFIG_LOCKDEP
 extern void debug_show_all_locks(void);
-extern void __debug_show_held_locks(struct task_struct *task);
 extern void debug_show_held_locks(struct task_struct *task);
 extern void debug_check_no_locks_freed(const void *from, unsigned long len);
 extern void debug_check_no_locks_held(struct task_struct *task);
@@ -58,10 +57,6 @@ static inline void debug_show_all_locks(void)
 {
 }
 
-static inline void __debug_show_held_locks(struct task_struct *task)
-{
-}
-
 static inline void debug_show_held_locks(struct task_struct *task)
 {
 }
diff --git a/kernel/hung_task.c b/kernel/hung_task.c
index 0c642d51aac2..bca942379559 100644
--- a/kernel/hung_task.c
+++ b/kernel/hung_task.c
@@ -98,7 +98,7 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout)
 	printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\""
 			" disables this message.\n");
 	sched_show_task(t);
-	__debug_show_held_locks(t);
+	debug_show_held_locks(t);
 
 	touch_nmi_watchdog();
 
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index f2852a510232..84baa71cfda5 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -3775,7 +3775,7 @@ EXPORT_SYMBOL_GPL(debug_show_all_locks);
  * Careful: only use this function if you are sure that
  * the task cannot run in parallel!
  */
-void __debug_show_held_locks(struct task_struct *task)
+void debug_show_held_locks(struct task_struct *task)
 {
 	if (unlikely(!debug_locks)) {
 		printk("INFO: lockdep is turned off.\n");
@@ -3783,12 +3783,6 @@ void __debug_show_held_locks(struct task_struct *task)
 	}
 	lockdep_print_held_locks(task);
 }
-EXPORT_SYMBOL_GPL(__debug_show_held_locks);
-
-void debug_show_held_locks(struct task_struct *task)
-{
-		__debug_show_held_locks(task);
-}
 EXPORT_SYMBOL_GPL(debug_show_held_locks);
 
 void lockdep_sys_exit(void)
-- 
cgit v1.2.3


From 70791ce9ba68a5921c9905ef05d23f62a90bc10c Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Tue, 29 Jun 2010 19:34:05 +0200
Subject: perf: Generalize callchain_store()

callchain_store() is the same on every archs, inline it in
perf_event.h and rename it to perf_callchain_store() to avoid
any collision.

This removes repetitive code.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Tested-by: Will Deacon <will.deacon@arm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: David Miller <davem@davemloft.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Borislav Petkov <bp@amd64.org>
---
 arch/arm/kernel/perf_event.c         | 15 ++++----------
 arch/powerpc/kernel/perf_callchain.c | 40 +++++++++++++-----------------------
 arch/sh/kernel/perf_callchain.c      | 11 +++-------
 arch/sparc/kernel/perf_event.c       | 26 +++++++++--------------
 arch/x86/kernel/cpu/perf_event.c     | 20 +++++++-----------
 include/linux/perf_event.h           |  7 +++++++
 6 files changed, 45 insertions(+), 74 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index fdcb0be47df1..a07c3b1955f0 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -3001,13 +3001,6 @@ arch_initcall(init_hw_perf_events);
 /*
  * Callchain handling code.
  */
-static inline void
-callchain_store(struct perf_callchain_entry *entry,
-		u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
 
 /*
  * The registers we're interested in are at the end of the variable
@@ -3039,7 +3032,7 @@ user_backtrace(struct frame_tail *tail,
 	if (__copy_from_user_inatomic(&buftail, tail, sizeof(buftail)))
 		return NULL;
 
-	callchain_store(entry, buftail.lr);
+	perf_callchain_store(entry, buftail.lr);
 
 	/*
 	 * Frame pointers should strictly progress back up the stack
@@ -3057,7 +3050,7 @@ perf_callchain_user(struct pt_regs *regs,
 {
 	struct frame_tail *tail;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
+	perf_callchain_store(entry, PERF_CONTEXT_USER);
 
 	if (!user_mode(regs))
 		regs = task_pt_regs(current);
@@ -3078,7 +3071,7 @@ callchain_trace(struct stackframe *fr,
 		void *data)
 {
 	struct perf_callchain_entry *entry = data;
-	callchain_store(entry, fr->pc);
+	perf_callchain_store(entry, fr->pc);
 	return 0;
 }
 
@@ -3088,7 +3081,7 @@ perf_callchain_kernel(struct pt_regs *regs,
 {
 	struct stackframe fr;
 
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
+	perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
 	fr.fp = regs->ARM_fp;
 	fr.sp = regs->ARM_sp;
 	fr.lr = regs->ARM_lr;
diff --git a/arch/powerpc/kernel/perf_callchain.c b/arch/powerpc/kernel/perf_callchain.c
index 95ad9dad298e..a286c2e5a3ea 100644
--- a/arch/powerpc/kernel/perf_callchain.c
+++ b/arch/powerpc/kernel/perf_callchain.c
@@ -23,18 +23,6 @@
 #include "ppc32.h"
 #endif
 
-/*
- * Store another value in a callchain_entry.
- */
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	unsigned int nr = entry->nr;
-
-	if (nr < PERF_MAX_STACK_DEPTH) {
-		entry->ip[nr] = ip;
-		entry->nr = nr + 1;
-	}
-}
 
 /*
  * Is sp valid as the address of the next kernel stack frame after prev_sp?
@@ -69,8 +57,8 @@ static void perf_callchain_kernel(struct pt_regs *regs,
 
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->nip);
+	perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
+	perf_callchain_store(entry, regs->nip);
 
 	if (!validate_sp(sp, current, STACK_FRAME_OVERHEAD))
 		return;
@@ -89,7 +77,7 @@ static void perf_callchain_kernel(struct pt_regs *regs,
 			next_ip = regs->nip;
 			lr = regs->link;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_KERNEL);
+			perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
 
 		} else {
 			if (level == 0)
@@ -111,7 +99,7 @@ static void perf_callchain_kernel(struct pt_regs *regs,
 			++level;
 		}
 
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		if (!valid_next_sp(next_sp, sp))
 			return;
 		sp = next_sp;
@@ -246,8 +234,8 @@ static void perf_callchain_user_64(struct pt_regs *regs,
 	next_ip = regs->nip;
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, next_ip);
+	perf_callchain_store(entry, PERF_CONTEXT_USER);
+	perf_callchain_store(entry, next_ip);
 
 	for (;;) {
 		fp = (unsigned long __user *) sp;
@@ -276,14 +264,14 @@ static void perf_callchain_user_64(struct pt_regs *regs,
 			    read_user_stack_64(&uregs[PT_R1], &sp))
 				return;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_USER);
-			callchain_store(entry, next_ip);
+			perf_callchain_store(entry, PERF_CONTEXT_USER);
+			perf_callchain_store(entry, next_ip);
 			continue;
 		}
 
 		if (level == 0)
 			next_ip = lr;
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		++level;
 		sp = next_sp;
 	}
@@ -447,8 +435,8 @@ static void perf_callchain_user_32(struct pt_regs *regs,
 	next_ip = regs->nip;
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, next_ip);
+	perf_callchain_store(entry, PERF_CONTEXT_USER);
+	perf_callchain_store(entry, next_ip);
 
 	while (entry->nr < PERF_MAX_STACK_DEPTH) {
 		fp = (unsigned int __user *) (unsigned long) sp;
@@ -470,14 +458,14 @@ static void perf_callchain_user_32(struct pt_regs *regs,
 			    read_user_stack_32(&uregs[PT_R1], &sp))
 				return;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_USER);
-			callchain_store(entry, next_ip);
+			perf_callchain_store(entry, PERF_CONTEXT_USER);
+			perf_callchain_store(entry, next_ip);
 			continue;
 		}
 
 		if (level == 0)
 			next_ip = lr;
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		++level;
 		sp = next_sp;
 	}
diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c
index 1d6dbce7a3bc..00143f3dd196 100644
--- a/arch/sh/kernel/perf_callchain.c
+++ b/arch/sh/kernel/perf_callchain.c
@@ -14,11 +14,6 @@
 #include <asm/unwinder.h>
 #include <asm/ptrace.h>
 
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
 
 static void callchain_warning(void *data, char *msg)
 {
@@ -39,7 +34,7 @@ static void callchain_address(void *data, unsigned long addr, int reliable)
 	struct perf_callchain_entry *entry = data;
 
 	if (reliable)
-		callchain_store(entry, addr);
+		perf_callchain_store(entry, addr);
 }
 
 static const struct stacktrace_ops callchain_ops = {
@@ -52,8 +47,8 @@ static const struct stacktrace_ops callchain_ops = {
 static void
 perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
 {
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->pc);
+	perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
+	perf_callchain_store(entry, regs->pc);
 
 	unwind_stack(NULL, regs, NULL, &callchain_ops, entry);
 }
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 357ced3c33ff..2a95a9079862 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -1283,12 +1283,6 @@ void __init init_hw_perf_events(void)
 	register_die_notifier(&perf_event_nmi_notifier);
 }
 
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
-
 static void perf_callchain_kernel(struct pt_regs *regs,
 				  struct perf_callchain_entry *entry)
 {
@@ -1297,8 +1291,8 @@ static void perf_callchain_kernel(struct pt_regs *regs,
 	int graph = 0;
 #endif
 
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->tpc);
+	perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
+	perf_callchain_store(entry, regs->tpc);
 
 	ksp = regs->u_regs[UREG_I6];
 	fp = ksp + STACK_BIAS;
@@ -1322,13 +1316,13 @@ static void perf_callchain_kernel(struct pt_regs *regs,
 			pc = sf->callers_pc;
 			fp = (unsigned long)sf->fp + STACK_BIAS;
 		}
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 		if ((pc + 8UL) == (unsigned long) &return_to_handler) {
 			int index = current->curr_ret_stack;
 			if (current->ret_stack && index >= graph) {
 				pc = current->ret_stack[index - graph].ret;
-				callchain_store(entry, pc);
+				perf_callchain_store(entry, pc);
 				graph++;
 			}
 		}
@@ -1341,8 +1335,8 @@ static void perf_callchain_user_64(struct pt_regs *regs,
 {
 	unsigned long ufp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->tpc);
+	perf_callchain_store(entry, PERF_CONTEXT_USER);
+	perf_callchain_store(entry, regs->tpc);
 
 	ufp = regs->u_regs[UREG_I6] + STACK_BIAS;
 	do {
@@ -1355,7 +1349,7 @@ static void perf_callchain_user_64(struct pt_regs *regs,
 
 		pc = sf.callers_pc;
 		ufp = (unsigned long)sf.fp + STACK_BIAS;
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
@@ -1364,8 +1358,8 @@ static void perf_callchain_user_32(struct pt_regs *regs,
 {
 	unsigned long ufp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->tpc);
+	perf_callchain_store(entry, PERF_CONTEXT_USER);
+	perf_callchain_store(entry, regs->tpc);
 
 	ufp = regs->u_regs[UREG_I6] & 0xffffffffUL;
 	do {
@@ -1378,7 +1372,7 @@ static void perf_callchain_user_32(struct pt_regs *regs,
 
 		pc = sf.callers_pc;
 		ufp = (unsigned long)sf.fp;
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 4a4d191f9492..8af28caeafc1 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1571,12 +1571,6 @@ const struct pmu *hw_perf_event_init(struct perf_event *event)
  * callchain support
  */
 
-static inline
-void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
 
 static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
 static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry);
@@ -1602,7 +1596,7 @@ static void backtrace_address(void *data, unsigned long addr, int reliable)
 {
 	struct perf_callchain_entry *entry = data;
 
-	callchain_store(entry, addr);
+	perf_callchain_store(entry, addr);
 }
 
 static const struct stacktrace_ops backtrace_ops = {
@@ -1616,8 +1610,8 @@ static const struct stacktrace_ops backtrace_ops = {
 static void
 perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
 {
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->ip);
+	perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
+	perf_callchain_store(entry, regs->ip);
 
 	dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry);
 }
@@ -1646,7 +1640,7 @@ perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry)
 		if (fp < compat_ptr(regs->sp))
 			break;
 
-		callchain_store(entry, frame.return_address);
+		perf_callchain_store(entry, frame.return_address);
 		fp = compat_ptr(frame.next_frame);
 	}
 	return 1;
@@ -1670,8 +1664,8 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry)
 
 	fp = (void __user *)regs->bp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->ip);
+	perf_callchain_store(entry, PERF_CONTEXT_USER);
+	perf_callchain_store(entry, regs->ip);
 
 	if (perf_callchain_user32(regs, entry))
 		return;
@@ -1688,7 +1682,7 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry)
 		if ((unsigned long)fp < regs->sp)
 			break;
 
-		callchain_store(entry, frame.return_address);
+		perf_callchain_store(entry, frame.return_address);
 		fp = frame.next_frame;
 	}
 }
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 937495c25073..358880404b42 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -978,6 +978,13 @@ extern void perf_event_fork(struct task_struct *tsk);
 
 extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs);
 
+static inline void
+perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
+{
+	if (entry->nr < PERF_MAX_STACK_DEPTH)
+		entry->ip[entry->nr++] = ip;
+}
+
 extern int sysctl_perf_event_paranoid;
 extern int sysctl_perf_event_mlock;
 extern int sysctl_perf_event_sample_rate;
-- 
cgit v1.2.3


From 56962b4449af34070bb1994621ef4f0265eed4d8 Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Wed, 30 Jun 2010 23:03:51 +0200
Subject: perf: Generalize some arch callchain code

- Most archs use one callchain buffer per cpu, except x86 that needs
  to deal with NMIs. Provide a default perf_callchain_buffer()
  implementation that x86 overrides.

- Centralize all the kernel/user regs handling and invoke new arch
  handlers from there: perf_callchain_user() / perf_callchain_kernel()
  That avoid all the user_mode(), current->mm checks and so...

- Invert some parameters in perf_callchain_*() helpers: entry to the
  left, regs to the right, following the traditional (dst, src).

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Tested-by: Will Deacon <will.deacon@arm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: David Miller <davem@davemloft.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Borislav Petkov <bp@amd64.org>
---
 arch/arm/kernel/perf_event.c         | 43 +++----------------------------
 arch/powerpc/kernel/perf_callchain.c | 49 +++++++++++-------------------------
 arch/sh/kernel/perf_callchain.c      | 37 ++-------------------------
 arch/sparc/kernel/perf_event.c       | 46 +++++++++++----------------------
 arch/x86/kernel/cpu/perf_event.c     | 45 ++++++---------------------------
 include/linux/perf_event.h           | 10 +++++++-
 kernel/perf_event.c                  | 40 +++++++++++++++++++++++++++--
 7 files changed, 90 insertions(+), 180 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index a07c3b1955f0..0e3bbdb15927 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -3044,17 +3044,13 @@ user_backtrace(struct frame_tail *tail,
 	return buftail.fp - 1;
 }
 
-static void
-perf_callchain_user(struct pt_regs *regs,
-		    struct perf_callchain_entry *entry)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct frame_tail *tail;
 
 	perf_callchain_store(entry, PERF_CONTEXT_USER);
 
-	if (!user_mode(regs))
-		regs = task_pt_regs(current);
-
 	tail = (struct frame_tail *)regs->ARM_fp - 1;
 
 	while (tail && !((unsigned long)tail & 0x3))
@@ -3075,9 +3071,8 @@ callchain_trace(struct stackframe *fr,
 	return 0;
 }
 
-static void
-perf_callchain_kernel(struct pt_regs *regs,
-		      struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct stackframe fr;
 
@@ -3088,33 +3083,3 @@ perf_callchain_kernel(struct pt_regs *regs,
 	fr.pc = regs->ARM_pc;
 	walk_stackframe(&fr, callchain_trace, entry);
 }
-
-static void
-perf_do_callchain(struct pt_regs *regs,
-		  struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-
-	if (current->mm)
-		perf_callchain_user(regs, entry);
-}
-
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
-
-struct perf_callchain_entry *
-perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry = &__get_cpu_var(pmc_irq_entry);
-
-	entry->nr = 0;
-	perf_do_callchain(regs, entry);
-	return entry;
-}
diff --git a/arch/powerpc/kernel/perf_callchain.c b/arch/powerpc/kernel/perf_callchain.c
index a286c2e5a3ea..f7a85ede8407 100644
--- a/arch/powerpc/kernel/perf_callchain.c
+++ b/arch/powerpc/kernel/perf_callchain.c
@@ -46,8 +46,8 @@ static int valid_next_sp(unsigned long sp, unsigned long prev_sp)
 	return 0;
 }
 
-static void perf_callchain_kernel(struct pt_regs *regs,
-				  struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	unsigned long sp, next_sp;
 	unsigned long next_ip;
@@ -221,8 +221,8 @@ static int sane_signal_64_frame(unsigned long sp)
 		puc == (unsigned long) &sf->uc;
 }
 
-static void perf_callchain_user_64(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_64(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long sp, next_sp;
 	unsigned long next_ip;
@@ -303,8 +303,8 @@ static int read_user_stack_32(unsigned int __user *ptr, unsigned int *ret)
 	return __get_user_inatomic(*ret, ptr);
 }
 
-static inline void perf_callchain_user_64(struct pt_regs *regs,
-					  struct perf_callchain_entry *entry)
+static inline void perf_callchain_user_64(struct perf_callchain_entry *entry,
+					  struct pt_regs *regs)
 {
 }
 
@@ -423,8 +423,8 @@ static unsigned int __user *signal_frame_32_regs(unsigned int sp,
 	return mctx->mc_gregs;
 }
 
-static void perf_callchain_user_32(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_32(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned int sp, next_sp;
 	unsigned int next_ip;
@@ -471,32 +471,11 @@ static void perf_callchain_user_32(struct pt_regs *regs,
 	}
 }
 
-/*
- * Since we can't get PMU interrupts inside a PMU interrupt handler,
- * we don't need separate irq and nmi entries here.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, cpu_perf_callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	struct perf_callchain_entry *entry = &__get_cpu_var(cpu_perf_callchain);
-
-	entry->nr = 0;
-
-	if (!user_mode(regs)) {
-		perf_callchain_kernel(regs, entry);
-		if (current->mm)
-			regs = task_pt_regs(current);
-		else
-			regs = NULL;
-	}
-
-	if (regs) {
-		if (current_is_64bit())
-			perf_callchain_user_64(regs, entry);
-		else
-			perf_callchain_user_32(regs, entry);
-	}
-
-	return entry;
+	if (current_is_64bit())
+		perf_callchain_user_64(entry, regs);
+	else
+		perf_callchain_user_32(entry, regs);
 }
diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c
index 00143f3dd196..ef076a91292a 100644
--- a/arch/sh/kernel/perf_callchain.c
+++ b/arch/sh/kernel/perf_callchain.c
@@ -44,44 +44,11 @@ static const struct stacktrace_ops callchain_ops = {
 	.address	= callchain_address,
 };
 
-static void
-perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
 	perf_callchain_store(entry, regs->pc);
 
 	unwind_stack(NULL, regs, NULL, &callchain_ops, entry);
 }
-
-static void
-perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	/*
-	 * Only the kernel side is implemented for now.
-	 */
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-}
-
-/*
- * No need for separate IRQ and NMI entries.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry = &__get_cpu_var(callchain);
-
-	entry->nr = 0;
-
-	perf_do_callchain(regs, entry);
-
-	return entry;
-}
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 2a95a9079862..460162d74aba 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -1283,14 +1283,16 @@ void __init init_hw_perf_events(void)
 	register_die_notifier(&perf_event_nmi_notifier);
 }
 
-static void perf_callchain_kernel(struct pt_regs *regs,
-				  struct perf_callchain_entry *entry)
+void perf_callchain_kernel(struct perf_callchain_entry *entry,
+			   struct pt_regs *regs)
 {
 	unsigned long ksp, fp;
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 	int graph = 0;
 #endif
 
+	stack_trace_flush();
+
 	perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
 	perf_callchain_store(entry, regs->tpc);
 
@@ -1330,8 +1332,8 @@ static void perf_callchain_kernel(struct pt_regs *regs,
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-static void perf_callchain_user_64(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_64(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long ufp;
 
@@ -1353,8 +1355,8 @@ static void perf_callchain_user_64(struct pt_regs *regs,
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-static void perf_callchain_user_32(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_32(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long ufp;
 
@@ -1376,30 +1378,12 @@ static void perf_callchain_user_32(struct pt_regs *regs,
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-/* Like powerpc we can't get PMU interrupts within the PMU handler,
- * so no need for separate NMI and IRQ chains as on x86.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	struct perf_callchain_entry *entry = &__get_cpu_var(callchain);
-
-	entry->nr = 0;
-	if (!user_mode(regs)) {
-		stack_trace_flush();
-		perf_callchain_kernel(regs, entry);
-		if (current->mm)
-			regs = task_pt_regs(current);
-		else
-			regs = NULL;
-	}
-	if (regs) {
-		flushw_user();
-		if (test_thread_flag(TIF_32BIT))
-			perf_callchain_user_32(regs, entry);
-		else
-			perf_callchain_user_64(regs, entry);
-	}
-	return entry;
+	flushw_user();
+	if (test_thread_flag(TIF_32BIT))
+		perf_callchain_user_32(entry, regs);
+	else
+		perf_callchain_user_64(entry, regs);
 }
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 8af28caeafc1..39f8421b86e6 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1571,9 +1571,7 @@ const struct pmu *hw_perf_event_init(struct perf_event *event)
  * callchain support
  */
 
-
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry);
+static DEFINE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry_nmi);
 
 
 static void
@@ -1607,8 +1605,8 @@ static const struct stacktrace_ops backtrace_ops = {
 	.walk_stack		= print_context_stack_bp,
 };
 
-static void
-perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
 	perf_callchain_store(entry, regs->ip);
@@ -1653,14 +1651,12 @@ perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry)
 }
 #endif
 
-static void
-perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct stack_frame frame;
 	const void __user *fp;
 
-	if (!user_mode(regs))
-		regs = task_pt_regs(current);
 
 	fp = (void __user *)regs->bp;
 
@@ -1687,42 +1683,17 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry)
 	}
 }
 
-static void
-perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-
-	if (current->mm)
-		perf_callchain_user(regs, entry);
-}
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+struct perf_callchain_entry *perf_callchain_buffer(void)
 {
-	struct perf_callchain_entry *entry;
-
 	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
 		/* TODO: We don't support guest os callchain now */
 		return NULL;
 	}
 
 	if (in_nmi())
-		entry = &__get_cpu_var(pmc_nmi_entry);
-	else
-		entry = &__get_cpu_var(pmc_irq_entry);
-
-	entry->nr = 0;
-
-	perf_do_callchain(regs, entry);
+		return &__get_cpu_var(perf_callchain_entry_nmi);
 
-	return entry;
+	return &__get_cpu_var(perf_callchain_entry);
 }
 
 unsigned long perf_instruction_pointer(struct pt_regs *regs)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 358880404b42..4db61dded388 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -976,7 +976,15 @@ extern int perf_unregister_guest_info_callbacks(struct perf_guest_info_callbacks
 extern void perf_event_comm(struct task_struct *tsk);
 extern void perf_event_fork(struct task_struct *tsk);
 
-extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs);
+/* Callchains */
+DECLARE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry);
+
+extern void perf_callchain_user(struct perf_callchain_entry *entry,
+				struct pt_regs *regs);
+extern void perf_callchain_kernel(struct perf_callchain_entry *entry,
+				  struct pt_regs *regs);
+extern struct perf_callchain_entry *perf_callchain_buffer(void);
+
 
 static inline void
 perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index c772a3d4000d..02efde6c8798 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -2937,13 +2937,49 @@ void perf_event_do_pending(void)
 	__perf_pending_run();
 }
 
+DEFINE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry);
+
 /*
  * Callchain support -- arch specific
  */
 
-__weak struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+__weak struct perf_callchain_entry *perf_callchain_buffer(void)
 {
-	return NULL;
+	return &__get_cpu_var(perf_callchain_entry);
+}
+
+__weak void perf_callchain_kernel(struct perf_callchain_entry *entry,
+				  struct pt_regs *regs)
+{
+}
+
+__weak void perf_callchain_user(struct perf_callchain_entry *entry,
+				struct pt_regs *regs)
+{
+}
+
+static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+{
+	struct perf_callchain_entry *entry;
+
+	entry = perf_callchain_buffer();
+	if (!entry)
+		return NULL;
+
+	entry->nr = 0;
+
+	if (!user_mode(regs)) {
+		perf_callchain_kernel(entry, regs);
+		if (current->mm)
+			regs = task_pt_regs(current);
+		else
+			regs = NULL;
+	}
+
+	if (regs)
+		perf_callchain_user(entry, regs);
+
+	return entry;
 }
 
 
-- 
cgit v1.2.3


From 927c7a9e92c4f69097a6e9e086d11fc2f8a5b40b Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Thu, 1 Jul 2010 16:20:36 +0200
Subject: perf: Fix race in callchains

Now that software events don't have interrupt disabled anymore in
the event path, callchains can nest on any context. So seperating
nmi and others contexts in two buffers has become racy.

Fix this by providing one buffer per nesting level. Given the size
of the callchain entries (2040 bytes * 4), we now need to allocate
them dynamically.

v2: Fixed put_callchain_entry call after recursion.
    Fix the type of the recursion, it must be an array.

v3: Use a manual pr cpu allocation (temporary solution until NMIs
    can safely access vmalloc'ed memory).
    Do a better separation between callchain reference tracking and
    allocation. Make the "put" path lockless for non-release cases.

v4: Protect the callchain buffers with rcu.

v5: Do the cpu buffers allocations node affine.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Tested-by: Will Deacon <will.deacon@arm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: David Miller <davem@davemloft.net>
Cc: Borislav Petkov <bp@amd64.org>
---
 arch/x86/kernel/cpu/perf_event.c |  22 ++-
 include/linux/perf_event.h       |   1 -
 kernel/perf_event.c              | 298 ++++++++++++++++++++++++++++++---------
 3 files changed, 238 insertions(+), 83 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index a3c922288cc0..8e91cf34a9c8 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1608,6 +1608,11 @@ static const struct stacktrace_ops backtrace_ops = {
 void
 perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return NULL;
+	}
+
 	perf_callchain_store(entry, regs->ip);
 
 	dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry);
@@ -1656,6 +1661,10 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 	struct stack_frame frame;
 	const void __user *fp;
 
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return NULL;
+	}
 
 	fp = (void __user *)regs->bp;
 
@@ -1681,19 +1690,6 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 	}
 }
 
-struct perf_callchain_entry *perf_callchain_buffer(void)
-{
-	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
-		/* TODO: We don't support guest os callchain now */
-		return NULL;
-	}
-
-	if (in_nmi())
-		return &__get_cpu_var(perf_callchain_entry_nmi);
-
-	return &__get_cpu_var(perf_callchain_entry);
-}
-
 unsigned long perf_instruction_pointer(struct pt_regs *regs)
 {
 	unsigned long ip;
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 4db61dded388..d7e8ea690864 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -983,7 +983,6 @@ extern void perf_callchain_user(struct perf_callchain_entry *entry,
 				struct pt_regs *regs);
 extern void perf_callchain_kernel(struct perf_callchain_entry *entry,
 				  struct pt_regs *regs);
-extern struct perf_callchain_entry *perf_callchain_buffer(void);
 
 
 static inline void
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 615d024894cf..75ab8a2df6b2 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -1763,6 +1763,216 @@ static u64 perf_event_read(struct perf_event *event)
 	return perf_event_count(event);
 }
 
+/*
+ * Callchain support
+ */
+
+struct callchain_cpus_entries {
+	struct rcu_head			rcu_head;
+	struct perf_callchain_entry	*cpu_entries[0];
+};
+
+static DEFINE_PER_CPU(int, callchain_recursion[4]);
+static atomic_t nr_callchain_events;
+static DEFINE_MUTEX(callchain_mutex);
+struct callchain_cpus_entries *callchain_cpus_entries;
+
+
+__weak void perf_callchain_kernel(struct perf_callchain_entry *entry,
+				  struct pt_regs *regs)
+{
+}
+
+__weak void perf_callchain_user(struct perf_callchain_entry *entry,
+				struct pt_regs *regs)
+{
+}
+
+static void release_callchain_buffers_rcu(struct rcu_head *head)
+{
+	struct callchain_cpus_entries *entries;
+	int cpu;
+
+	entries = container_of(head, struct callchain_cpus_entries, rcu_head);
+
+	for_each_possible_cpu(cpu)
+		kfree(entries->cpu_entries[cpu]);
+
+	kfree(entries);
+}
+
+static void release_callchain_buffers(void)
+{
+	struct callchain_cpus_entries *entries;
+
+	entries = callchain_cpus_entries;
+	rcu_assign_pointer(callchain_cpus_entries, NULL);
+	call_rcu(&entries->rcu_head, release_callchain_buffers_rcu);
+}
+
+static int alloc_callchain_buffers(void)
+{
+	int cpu;
+	int size;
+	struct callchain_cpus_entries *entries;
+
+	/*
+	 * We can't use the percpu allocation API for data that can be
+	 * accessed from NMI. Use a temporary manual per cpu allocation
+	 * until that gets sorted out.
+	 */
+	size = sizeof(*entries) + sizeof(struct perf_callchain_entry *) *
+		num_possible_cpus();
+
+	entries = kzalloc(size, GFP_KERNEL);
+	if (!entries)
+		return -ENOMEM;
+
+	size = sizeof(struct perf_callchain_entry) * 4;
+
+	for_each_possible_cpu(cpu) {
+		entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL,
+							 cpu_to_node(cpu));
+		if (!entries->cpu_entries[cpu])
+			goto fail;
+	}
+
+	rcu_assign_pointer(callchain_cpus_entries, entries);
+
+	return 0;
+
+fail:
+	for_each_possible_cpu(cpu)
+		kfree(entries->cpu_entries[cpu]);
+	kfree(entries);
+
+	return -ENOMEM;
+}
+
+static int get_callchain_buffers(void)
+{
+	int err = 0;
+	int count;
+
+	mutex_lock(&callchain_mutex);
+
+	count = atomic_inc_return(&nr_callchain_events);
+	if (WARN_ON_ONCE(count < 1)) {
+		err = -EINVAL;
+		goto exit;
+	}
+
+	if (count > 1) {
+		/* If the allocation failed, give up */
+		if (!callchain_cpus_entries)
+			err = -ENOMEM;
+		goto exit;
+	}
+
+	err = alloc_callchain_buffers();
+	if (err)
+		release_callchain_buffers();
+exit:
+	mutex_unlock(&callchain_mutex);
+
+	return err;
+}
+
+static void put_callchain_buffers(void)
+{
+	if (atomic_dec_and_mutex_lock(&nr_callchain_events, &callchain_mutex)) {
+		release_callchain_buffers();
+		mutex_unlock(&callchain_mutex);
+	}
+}
+
+static int get_recursion_context(int *recursion)
+{
+	int rctx;
+
+	if (in_nmi())
+		rctx = 3;
+	else if (in_irq())
+		rctx = 2;
+	else if (in_softirq())
+		rctx = 1;
+	else
+		rctx = 0;
+
+	if (recursion[rctx])
+		return -1;
+
+	recursion[rctx]++;
+	barrier();
+
+	return rctx;
+}
+
+static inline void put_recursion_context(int *recursion, int rctx)
+{
+	barrier();
+	recursion[rctx]--;
+}
+
+static struct perf_callchain_entry *get_callchain_entry(int *rctx)
+{
+	int cpu;
+	struct callchain_cpus_entries *entries;
+
+	*rctx = get_recursion_context(__get_cpu_var(callchain_recursion));
+	if (*rctx == -1)
+		return NULL;
+
+	entries = rcu_dereference(callchain_cpus_entries);
+	if (!entries)
+		return NULL;
+
+	cpu = smp_processor_id();
+
+	return &entries->cpu_entries[cpu][*rctx];
+}
+
+static void
+put_callchain_entry(int rctx)
+{
+	put_recursion_context(__get_cpu_var(callchain_recursion), rctx);
+}
+
+static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+{
+	int rctx;
+	struct perf_callchain_entry *entry;
+
+
+	entry = get_callchain_entry(&rctx);
+	if (rctx == -1)
+		return NULL;
+
+	if (!entry)
+		goto exit_put;
+
+	entry->nr = 0;
+
+	if (!user_mode(regs)) {
+		perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
+		perf_callchain_kernel(entry, regs);
+		if (current->mm)
+			regs = task_pt_regs(current);
+		else
+			regs = NULL;
+	}
+
+	if (regs) {
+		perf_callchain_store(entry, PERF_CONTEXT_USER);
+		perf_callchain_user(entry, regs);
+	}
+
+exit_put:
+	put_callchain_entry(rctx);
+
+	return entry;
+}
+
 /*
  * Initialize the perf_event context in a task_struct:
  */
@@ -1895,6 +2105,8 @@ static void free_event(struct perf_event *event)
 			atomic_dec(&nr_comm_events);
 		if (event->attr.task)
 			atomic_dec(&nr_task_events);
+		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
+			put_callchain_buffers();
 	}
 
 	if (event->buffer) {
@@ -2937,55 +3149,6 @@ void perf_event_do_pending(void)
 	__perf_pending_run();
 }
 
-DEFINE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry);
-
-/*
- * Callchain support -- arch specific
- */
-
-__weak struct perf_callchain_entry *perf_callchain_buffer(void)
-{
-	return &__get_cpu_var(perf_callchain_entry);
-}
-
-__weak void perf_callchain_kernel(struct perf_callchain_entry *entry,
-				  struct pt_regs *regs)
-{
-}
-
-__weak void perf_callchain_user(struct perf_callchain_entry *entry,
-				struct pt_regs *regs)
-{
-}
-
-static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry;
-
-	entry = perf_callchain_buffer();
-	if (!entry)
-		return NULL;
-
-	entry->nr = 0;
-
-	if (!user_mode(regs)) {
-		perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
-		perf_callchain_kernel(entry, regs);
-		if (current->mm)
-			regs = task_pt_regs(current);
-		else
-			regs = NULL;
-	}
-
-	if (regs) {
-		perf_callchain_store(entry, PERF_CONTEXT_USER);
-		perf_callchain_user(entry, regs);
-	}
-
-	return entry;
-}
-
-
 /*
  * We assume there is only KVM supporting the callbacks.
  * Later on, we might change it to a list if there is
@@ -3480,14 +3643,20 @@ static void perf_event_output(struct perf_event *event, int nmi,
 	struct perf_output_handle handle;
 	struct perf_event_header header;
 
+	/* protect the callchain buffers */
+	rcu_read_lock();
+
 	perf_prepare_sample(&header, data, event, regs);
 
 	if (perf_output_begin(&handle, event, header.size, nmi, 1))
-		return;
+		goto exit;
 
 	perf_output_sample(&handle, &header, data, event);
 
 	perf_output_end(&handle);
+
+exit:
+	rcu_read_unlock();
 }
 
 /*
@@ -4243,32 +4412,16 @@ end:
 int perf_swevent_get_recursion_context(void)
 {
 	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	int rctx;
 
-	if (in_nmi())
-		rctx = 3;
-	else if (in_irq())
-		rctx = 2;
-	else if (in_softirq())
-		rctx = 1;
-	else
-		rctx = 0;
-
-	if (cpuctx->recursion[rctx])
-		return -1;
-
-	cpuctx->recursion[rctx]++;
-	barrier();
-
-	return rctx;
+	return get_recursion_context(cpuctx->recursion);
 }
 EXPORT_SYMBOL_GPL(perf_swevent_get_recursion_context);
 
 void inline perf_swevent_put_recursion_context(int rctx)
 {
 	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	barrier();
-	cpuctx->recursion[rctx]--;
+
+	put_recursion_context(cpuctx->recursion, rctx);
 }
 
 void __perf_sw_event(u32 event_id, u64 nr, int nmi,
@@ -4968,6 +5121,13 @@ done:
 			atomic_inc(&nr_comm_events);
 		if (event->attr.task)
 			atomic_inc(&nr_task_events);
+		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) {
+			err = get_callchain_buffers();
+			if (err) {
+				free_event(event);
+				return ERR_PTR(err);
+			}
+		}
 	}
 
 	return event;
-- 
cgit v1.2.3


From 7ae07ea3a48d30689ee037cb136bc21f0b37d8ae Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Sat, 14 Aug 2010 20:45:13 +0200
Subject: perf: Humanize the number of contexts

Instead of hardcoding the number of contexts for the recursions
barriers, define a cpp constant to make the code more
self-explanatory.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Stephane Eranian <eranian@google.com>
---
 include/linux/perf_event.h      | 14 ++++++++------
 kernel/perf_event.c             |  4 ++--
 kernel/trace/trace_event_perf.c |  8 ++++----
 3 files changed, 14 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index d7e8ea690864..ae6fa6050925 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -808,6 +808,12 @@ struct perf_event_context {
 	struct rcu_head			rcu_head;
 };
 
+/*
+ * Number of contexts where an event can trigger:
+ * 	task, softirq, hardirq, nmi.
+ */
+#define PERF_NR_CONTEXTS	4
+
 /**
  * struct perf_event_cpu_context - per cpu event context structure
  */
@@ -821,12 +827,8 @@ struct perf_cpu_context {
 	struct mutex			hlist_mutex;
 	int				hlist_refcount;
 
-	/*
-	 * Recursion avoidance:
-	 *
-	 * task, softirq, irq, nmi context
-	 */
-	int				recursion[4];
+	/* Recursion avoidance in each contexts */
+	int				recursion[PERF_NR_CONTEXTS];
 };
 
 struct perf_output_handle {
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 75ab8a2df6b2..f416aef242c3 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -1772,7 +1772,7 @@ struct callchain_cpus_entries {
 	struct perf_callchain_entry	*cpu_entries[0];
 };
 
-static DEFINE_PER_CPU(int, callchain_recursion[4]);
+static DEFINE_PER_CPU(int, callchain_recursion[PERF_NR_CONTEXTS]);
 static atomic_t nr_callchain_events;
 static DEFINE_MUTEX(callchain_mutex);
 struct callchain_cpus_entries *callchain_cpus_entries;
@@ -1828,7 +1828,7 @@ static int alloc_callchain_buffers(void)
 	if (!entries)
 		return -ENOMEM;
 
-	size = sizeof(struct perf_callchain_entry) * 4;
+	size = sizeof(struct perf_callchain_entry) * PERF_NR_CONTEXTS;
 
 	for_each_possible_cpu(cpu) {
 		entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL,
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 000e6e85b445..db2eae2efcf2 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -9,7 +9,7 @@
 #include <linux/kprobes.h>
 #include "trace.h"
 
-static char *perf_trace_buf[4];
+static char *perf_trace_buf[PERF_NR_CONTEXTS];
 
 /*
  * Force it to be aligned to unsigned long to avoid misaligned accesses
@@ -45,7 +45,7 @@ static int perf_trace_event_init(struct ftrace_event_call *tp_event,
 		char *buf;
 		int i;
 
-		for (i = 0; i < 4; i++) {
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
 			buf = (char *)alloc_percpu(perf_trace_t);
 			if (!buf)
 				goto fail;
@@ -65,7 +65,7 @@ fail:
 	if (!total_ref_count) {
 		int i;
 
-		for (i = 0; i < 4; i++) {
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
 			free_percpu(perf_trace_buf[i]);
 			perf_trace_buf[i] = NULL;
 		}
@@ -140,7 +140,7 @@ void perf_trace_destroy(struct perf_event *p_event)
 	tp_event->perf_events = NULL;
 
 	if (!--total_ref_count) {
-		for (i = 0; i < 4; i++) {
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
 			free_percpu(perf_trace_buf[i]);
 			perf_trace_buf[i] = NULL;
 		}
-- 
cgit v1.2.3


From 6016ee13db518ab1cd0cbf43fc2ad5712021e338 Mon Sep 17 00:00:00 2001
From: Namhyung Kim <namhyung@gmail.com>
Date: Wed, 11 Aug 2010 12:47:59 +0900
Subject: perf, tracing: add missing __percpu markups

ftrace_event_call->perf_events, perf_trace_buf,
fgraph_data->cpu_data and some local variables are percpu pointers
missing __percpu markups. Add them.

Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <1281498479-28551-1-git-send-email-namhyung@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 include/linux/ftrace_event.h         |  4 ++--
 kernel/trace/trace_event_perf.c      | 15 ++++++++-------
 kernel/trace/trace_functions_graph.c |  2 +-
 3 files changed, 11 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 02b8b24f8f51..5f8ad7bec636 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -191,8 +191,8 @@ struct ftrace_event_call {
 	unsigned int		flags;
 
 #ifdef CONFIG_PERF_EVENTS
-	int			perf_refcount;
-	struct hlist_head	*perf_events;
+	int				perf_refcount;
+	struct hlist_head __percpu	*perf_events;
 #endif
 };
 
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index db2eae2efcf2..92f5477a006a 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -9,7 +9,7 @@
 #include <linux/kprobes.h>
 #include "trace.h"
 
-static char *perf_trace_buf[PERF_NR_CONTEXTS];
+static char __percpu *perf_trace_buf[PERF_NR_CONTEXTS];
 
 /*
  * Force it to be aligned to unsigned long to avoid misaligned accesses
@@ -24,7 +24,7 @@ static int	total_ref_count;
 static int perf_trace_event_init(struct ftrace_event_call *tp_event,
 				 struct perf_event *p_event)
 {
-	struct hlist_head *list;
+	struct hlist_head __percpu *list;
 	int ret = -ENOMEM;
 	int cpu;
 
@@ -42,11 +42,11 @@ static int perf_trace_event_init(struct ftrace_event_call *tp_event,
 	tp_event->perf_events = list;
 
 	if (!total_ref_count) {
-		char *buf;
+		char __percpu *buf;
 		int i;
 
 		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
-			buf = (char *)alloc_percpu(perf_trace_t);
+			buf = (char __percpu *)alloc_percpu(perf_trace_t);
 			if (!buf)
 				goto fail;
 
@@ -102,13 +102,14 @@ int perf_trace_init(struct perf_event *p_event)
 int perf_trace_enable(struct perf_event *p_event)
 {
 	struct ftrace_event_call *tp_event = p_event->tp_event;
+	struct hlist_head __percpu *pcpu_list;
 	struct hlist_head *list;
 
-	list = tp_event->perf_events;
-	if (WARN_ON_ONCE(!list))
+	pcpu_list = tp_event->perf_events;
+	if (WARN_ON_ONCE(!pcpu_list))
 		return -EINVAL;
 
-	list = this_cpu_ptr(list);
+	list = this_cpu_ptr(pcpu_list);
 	hlist_add_head_rcu(&p_event->hlist_entry, list);
 
 	return 0;
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index 6bff23625781..fcb5a542cd21 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -23,7 +23,7 @@ struct fgraph_cpu_data {
 };
 
 struct fgraph_data {
-	struct fgraph_cpu_data		*cpu_data;
+	struct fgraph_cpu_data __percpu *cpu_data;
 
 	/* Place to preserve last processed entry. */
 	struct ftrace_graph_ent_entry	ent;
-- 
cgit v1.2.3


From 2244d07bfa2097cb00600da91c715a8aa547917e Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <socketcan@hartkopp.net>
Date: Tue, 17 Aug 2010 08:59:14 +0000
Subject: net: simplify flags for tx timestamping

This patch removes the abstraction introduced by the union skb_shared_tx in
the shared skb data.

The access of the different union elements at several places led to some
confusion about accessing the shared tx_flags e.g. in skb_orphan_try().

    http://marc.info/?l=linux-netdev&m=128084897415886&w=2

Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/timestamping.txt | 22 +++++++++-------
 drivers/net/bfin_mac.c                    | 10 +++----
 drivers/net/gianfar.c                     | 15 +++++------
 drivers/net/igb/igb.h                     |  2 +-
 drivers/net/igb/igb_main.c                | 11 ++++----
 include/linux/skbuff.h                    | 44 +++++++++++--------------------
 include/net/ip.h                          |  2 +-
 include/net/sock.h                        |  8 ++----
 net/can/raw.c                             |  4 +--
 net/core/dev.c                            |  6 ++---
 net/core/skbuff.c                         |  2 +-
 net/ipv4/icmp.c                           |  4 +--
 net/ipv4/ip_output.c                      |  6 ++---
 net/ipv4/raw.c                            |  2 +-
 net/ipv4/udp.c                            |  4 +--
 net/packet/af_packet.c                    |  4 +--
 net/socket.c                              |  9 +++----
 17 files changed, 68 insertions(+), 87 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/networking/timestamping.txt b/Documentation/networking/timestamping.txt
index e8c8f4f06c67..98097d8cb910 100644
--- a/Documentation/networking/timestamping.txt
+++ b/Documentation/networking/timestamping.txt
@@ -172,15 +172,19 @@ struct skb_shared_hwtstamps {
 };
 
 Time stamps for outgoing packets are to be generated as follows:
-- In hard_start_xmit(), check if skb_tx(skb)->hardware is set no-zero.
-  If yes, then the driver is expected to do hardware time stamping.
+- In hard_start_xmit(), check if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)
+  is set no-zero. If yes, then the driver is expected to do hardware time
+  stamping.
 - If this is possible for the skb and requested, then declare
-  that the driver is doing the time stamping by setting the field
-  skb_tx(skb)->in_progress non-zero. You might want to keep a pointer
-  to the associated skb for the next step and not free the skb. A driver
-  not supporting hardware time stamping doesn't do that. A driver must
-  never touch sk_buff::tstamp! It is used to store software generated
-  time stamps by the network subsystem.
+  that the driver is doing the time stamping by setting the flag
+  SKBTX_IN_PROGRESS in skb_shinfo(skb)->tx_flags , e.g. with
+
+      skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+
+  You might want to keep a pointer to the associated skb for the next step
+  and not free the skb. A driver not supporting hardware time stamping doesn't
+  do that. A driver must never touch sk_buff::tstamp! It is used to store
+  software generated time stamps by the network subsystem.
 - As soon as the driver has sent the packet and/or obtained a
   hardware time stamp for it, it passes the time stamp back by
   calling skb_hwtstamp_tx() with the original skb, the raw
@@ -191,6 +195,6 @@ Time stamps for outgoing packets are to be generated as follows:
   this would occur at a later time in the processing pipeline than other
   software time stamping and therefore could lead to unexpected deltas
   between time stamps.
-- If the driver did not call set skb_tx(skb)->in_progress, then
+- If the driver did not set the SKBTX_IN_PROGRESS flag (see above), then
   dev_hard_start_xmit() checks whether software time stamping
   is wanted as fallback and potentially generates the time stamp.
diff --git a/drivers/net/bfin_mac.c b/drivers/net/bfin_mac.c
index 012613fde3f4..7a0e4156fade 100644
--- a/drivers/net/bfin_mac.c
+++ b/drivers/net/bfin_mac.c
@@ -803,15 +803,14 @@ static void bfin_dump_hwtamp(char *s, ktime_t *hw, ktime_t *ts, struct timecompa
 static void bfin_tx_hwtstamp(struct net_device *netdev, struct sk_buff *skb)
 {
 	struct bfin_mac_local *lp = netdev_priv(netdev);
-	union skb_shared_tx *shtx = skb_tx(skb);
 
-	if (shtx->hardware) {
+	if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) {
 		int timeout_cnt = MAX_TIMEOUT_CNT;
 
 		/* When doing time stamping, keep the connection to the socket
 		 * a while longer
 		 */
-		shtx->in_progress = 1;
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 
 		/*
 		 * The timestamping is done at the EMAC module's MII/RMII interface
@@ -991,7 +990,6 @@ static int bfin_mac_hard_start_xmit(struct sk_buff *skb,
 	struct bfin_mac_local *lp = netdev_priv(dev);
 	u16 *data;
 	u32 data_align = (unsigned long)(skb->data) & 0x3;
-	union skb_shared_tx *shtx = skb_tx(skb);
 
 	current_tx_ptr->skb = skb;
 
@@ -1005,7 +1003,7 @@ static int bfin_mac_hard_start_xmit(struct sk_buff *skb,
 		 * of this field are the length of the packet payload in bytes and the higher
 		 * 4 bits are the timestamping enable field.
 		 */
-		if (shtx->hardware)
+		if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)
 			*data |= 0x1000;
 
 		current_tx_ptr->desc_a.start_addr = (u32)data;
@@ -1015,7 +1013,7 @@ static int bfin_mac_hard_start_xmit(struct sk_buff *skb,
 	} else {
 		*((u16 *)(current_tx_ptr->packet)) = (u16)(skb->len);
 		/* enable timestamping for the sent packet */
-		if (shtx->hardware)
+		if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)
 			*((u16 *)(current_tx_ptr->packet)) |= 0x1000;
 		memcpy((u8 *)(current_tx_ptr->packet + 2), skb->data,
 			skb->len);
diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c
index 3d9f958ebd2c..e6048d6ab0ea 100644
--- a/drivers/net/gianfar.c
+++ b/drivers/net/gianfar.c
@@ -2048,7 +2048,6 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	u32 bufaddr;
 	unsigned long flags;
 	unsigned int nr_frags, nr_txbds, length;
-	union skb_shared_tx *shtx;
 
 	/*
 	 * TOE=1 frames larger than 2500 bytes may see excess delays
@@ -2069,10 +2068,10 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	txq = netdev_get_tx_queue(dev, rq);
 	base = tx_queue->tx_bd_base;
 	regs = tx_queue->grp->regs;
-	shtx = skb_tx(skb);
 
 	/* check if time stamp should be generated */
-	if (unlikely(shtx->hardware && priv->hwts_tx_en))
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP &&
+		     priv->hwts_tx_en))
 		do_tstamp = 1;
 
 	/* make space for additional header when fcb is needed */
@@ -2174,7 +2173,7 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	/* Setup tx hardware time stamping if requested */
 	if (unlikely(do_tstamp)) {
-		shtx->in_progress = 1;
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 		if (fcb == NULL)
 			fcb = gfar_add_fcb(skb);
 		fcb->ptp = 1;
@@ -2446,7 +2445,6 @@ static int gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 	int howmany = 0;
 	u32 lstatus;
 	size_t buflen;
-	union skb_shared_tx *shtx;
 
 	rx_queue = priv->rx_queue[tx_queue->qindex];
 	bdp = tx_queue->dirty_tx;
@@ -2461,8 +2459,7 @@ static int gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 		 * When time stamping, one additional TxBD must be freed.
 		 * Also, we need to dma_unmap_single() the TxPAL.
 		 */
-		shtx = skb_tx(skb);
-		if (unlikely(shtx->in_progress))
+		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
 			nr_txbds = frags + 2;
 		else
 			nr_txbds = frags + 1;
@@ -2476,7 +2473,7 @@ static int gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 				(lstatus & BD_LENGTH_MASK))
 			break;
 
-		if (unlikely(shtx->in_progress)) {
+		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) {
 			next = next_txbd(bdp, base, tx_ring_size);
 			buflen = next->length + GMAC_FCB_LEN;
 		} else
@@ -2485,7 +2482,7 @@ static int gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue)
 		dma_unmap_single(&priv->ofdev->dev, bdp->bufPtr,
 				buflen, DMA_TO_DEVICE);
 
-		if (unlikely(shtx->in_progress)) {
+		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) {
 			struct skb_shared_hwtstamps shhwtstamps;
 			u64 *ns = (u64*) (((u32)skb->data + 0x10) & ~0x7);
 			memset(&shhwtstamps, 0, sizeof(shhwtstamps));
diff --git a/drivers/net/igb/igb.h b/drivers/net/igb/igb.h
index 6e63d9a7fc75..44e0ff1494e0 100644
--- a/drivers/net/igb/igb.h
+++ b/drivers/net/igb/igb.h
@@ -143,7 +143,7 @@ struct igb_buffer {
 			u16 next_to_watch;
 			unsigned int bytecount;
 			u16 gso_segs;
-			union skb_shared_tx shtx;
+			u8 tx_flags;
 			u8 mapped_as_page;
 		};
 		/* RX */
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 9b4e5895f5f9..985e37cf17b6 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -3954,7 +3954,7 @@ static inline int igb_tx_map_adv(struct igb_ring *tx_ring, struct sk_buff *skb,
 	}
 
 	tx_ring->buffer_info[i].skb = skb;
-	tx_ring->buffer_info[i].shtx = skb_shinfo(skb)->tx_flags;
+	tx_ring->buffer_info[i].tx_flags = skb_shinfo(skb)->tx_flags;
 	/* multiply data chunks by size of headers */
 	tx_ring->buffer_info[i].bytecount = ((gso_segs - 1) * hlen) + skb->len;
 	tx_ring->buffer_info[i].gso_segs = gso_segs;
@@ -4088,7 +4088,6 @@ netdev_tx_t igb_xmit_frame_ring_adv(struct sk_buff *skb,
 	u32 tx_flags = 0;
 	u16 first;
 	u8 hdr_len = 0;
-	union skb_shared_tx *shtx = skb_tx(skb);
 
 	/* need: 1 descriptor per page,
 	 *       + 2 desc gap to keep tail from touching head,
@@ -4100,8 +4099,8 @@ netdev_tx_t igb_xmit_frame_ring_adv(struct sk_buff *skb,
 		return NETDEV_TX_BUSY;
 	}
 
-	if (unlikely(shtx->hardware)) {
-		shtx->in_progress = 1;
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 		tx_flags |= IGB_TX_FLAGS_TSTAMP;
 	}
 
@@ -5319,7 +5318,7 @@ static void igb_tx_hwtstamp(struct igb_q_vector *q_vector, struct igb_buffer *bu
 	u64 regval;
 
 	/* if skb does not support hw timestamp or TX stamp not valid exit */
-	if (likely(!buffer_info->shtx.hardware) ||
+	if (likely(!(buffer_info->tx_flags & SKBTX_HW_TSTAMP)) ||
 	    !(rd32(E1000_TSYNCTXCTL) & E1000_TSYNCTXCTL_VALID))
 		return;
 
@@ -5500,7 +5499,7 @@ static void igb_rx_hwtstamp(struct igb_q_vector *q_vector, u32 staterr,
 	 * values must belong to this one here and therefore we don't need to
 	 * compare any of the additional attributes stored for it.
 	 *
-	 * If nothing went wrong, then it should have a skb_shared_tx that we
+	 * If nothing went wrong, then it should have a shared tx_flags that we
 	 * can turn into a skb_shared_hwtstamps.
 	 */
 	if (staterr & E1000_RXDADV_STAT_TSIP) {
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index d8050382b189..f067c95cf18a 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -163,26 +163,19 @@ struct skb_shared_hwtstamps {
 	ktime_t	syststamp;
 };
 
-/**
- * struct skb_shared_tx - instructions for time stamping of outgoing packets
- * @hardware:		generate hardware time stamp
- * @software:		generate software time stamp
- * @in_progress:	device driver is going to provide
- *			hardware time stamp
- * @prevent_sk_orphan:	make sk reference available on driver level
- * @flags:		all shared_tx flags
- *
- * These flags are attached to packets as part of the
- * &skb_shared_info. Use skb_tx() to get a pointer.
- */
-union skb_shared_tx {
-	struct {
-		__u8	hardware:1,
-			software:1,
-			in_progress:1,
-			prevent_sk_orphan:1;
-	};
-	__u8 flags;
+/* Definitions for tx_flags in struct skb_shared_info */
+enum {
+	/* generate hardware time stamp */
+	SKBTX_HW_TSTAMP = 1 << 0,
+
+	/* generate software time stamp */
+	SKBTX_SW_TSTAMP = 1 << 1,
+
+	/* device driver is going to provide hardware time stamp */
+	SKBTX_IN_PROGRESS = 1 << 2,
+
+	/* ensure the originating sk reference is available on driver level */
+	SKBTX_DRV_NEEDS_SK_REF = 1 << 3,
 };
 
 /* This data is invariant across clones and lives at
@@ -195,7 +188,7 @@ struct skb_shared_info {
 	unsigned short	gso_segs;
 	unsigned short  gso_type;
 	__be32          ip6_frag_id;
-	union skb_shared_tx tx_flags;
+	__u8		tx_flags;
 	struct sk_buff	*frag_list;
 	struct skb_shared_hwtstamps hwtstamps;
 
@@ -587,11 +580,6 @@ static inline struct skb_shared_hwtstamps *skb_hwtstamps(struct sk_buff *skb)
 	return &skb_shinfo(skb)->hwtstamps;
 }
 
-static inline union skb_shared_tx *skb_tx(struct sk_buff *skb)
-{
-	return &skb_shinfo(skb)->tx_flags;
-}
-
 /**
  *	skb_queue_empty - check if a queue is empty
  *	@list: queue head
@@ -1996,8 +1984,8 @@ extern void skb_tstamp_tx(struct sk_buff *orig_skb,
 
 static inline void sw_tx_timestamp(struct sk_buff *skb)
 {
-	union skb_shared_tx *shtx = skb_tx(skb);
-	if (shtx->software && !shtx->in_progress)
+	if (skb_shinfo(skb)->tx_flags & SKBTX_SW_TSTAMP &&
+	    !(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS))
 		skb_tstamp_tx(skb, NULL);
 }
 
diff --git a/include/net/ip.h b/include/net/ip.h
index 890f9725d681..7691aca133db 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -53,7 +53,7 @@ struct ipcm_cookie {
 	__be32			addr;
 	int			oif;
 	struct ip_options	*opt;
-	union skb_shared_tx	shtx;
+	__u8			tx_flags;
 };
 
 #define IPCB(skb) ((struct inet_skb_parm*)((skb)->cb))
diff --git a/include/net/sock.h b/include/net/sock.h
index ac53bfbdfe16..100e43bf95fb 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1669,17 +1669,13 @@ static inline void sock_recv_ts_and_drops(struct msghdr *msg, struct sock *sk,
 
 /**
  * sock_tx_timestamp - checks whether the outgoing packet is to be time stamped
- * @msg:	outgoing packet
  * @sk:		socket sending this packet
- * @shtx:	filled with instructions for time stamping
+ * @tx_flags:	filled with instructions for time stamping
  *
  * Currently only depends on SOCK_TIMESTAMPING* flags. Returns error code if
  * parameters are invalid.
  */
-extern int sock_tx_timestamp(struct msghdr *msg,
-			     struct sock *sk,
-			     union skb_shared_tx *shtx);
-
+extern int sock_tx_timestamp(struct sock *sk, __u8 *tx_flags);
 
 /**
  * sk_eat_skb - Release a skb if it is no longer needed
diff --git a/net/can/raw.c b/net/can/raw.c
index a10e3338f084..7d77e67e57af 100644
--- a/net/can/raw.c
+++ b/net/can/raw.c
@@ -647,12 +647,12 @@ static int raw_sendmsg(struct kiocb *iocb, struct socket *sock,
 	err = memcpy_fromiovec(skb_put(skb, size), msg->msg_iov, size);
 	if (err < 0)
 		goto free_skb;
-	err = sock_tx_timestamp(msg, sk, skb_tx(skb));
+	err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags);
 	if (err < 0)
 		goto free_skb;
 
 	/* to be able to check the received tx sock reference in raw_rcv() */
-	skb_tx(skb)->prevent_sk_orphan = 1;
+	skb_shinfo(skb)->tx_flags |= SKBTX_DRV_NEEDS_SK_REF;
 
 	skb->dev = dev;
 	skb->sk  = sk;
diff --git a/net/core/dev.c b/net/core/dev.c
index 586a11cb4398..c1dc8a95f6ff 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1902,14 +1902,14 @@ static int dev_gso_segment(struct sk_buff *skb)
 
 /*
  * Try to orphan skb early, right before transmission by the device.
- * We cannot orphan skb if tx timestamp is requested, since
- * drivers need to call skb_tstamp_tx() to send the timestamp.
+ * We cannot orphan skb if tx timestamp is requested or the sk-reference
+ * is needed on driver level for other reasons, e.g. see net/can/raw.c
  */
 static inline void skb_orphan_try(struct sk_buff *skb)
 {
 	struct sock *sk = skb->sk;
 
-	if (sk && !skb_tx(skb)->flags) {
+	if (sk && !skb_shinfo(skb)->tx_flags) {
 		/* skb_tx_hash() wont be able to get sk.
 		 * We copy sk_hash into skb->rxhash
 		 */
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 3a2513f0d0c3..99ef721f773d 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3016,7 +3016,7 @@ void skb_tstamp_tx(struct sk_buff *orig_skb,
 	} else {
 		/*
 		 * no hardware time stamps available,
-		 * so keep the skb_shared_tx and only
+		 * so keep the shared tx_flags and only
 		 * store software time stamp
 		 */
 		skb->tstamp = ktime_get_real();
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index a0d847c7cba5..96bc7f9475a3 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -379,7 +379,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 	inet->tos = ip_hdr(skb)->tos;
 	daddr = ipc.addr = rt->rt_src;
 	ipc.opt = NULL;
-	ipc.shtx.flags = 0;
+	ipc.tx_flags = 0;
 	if (icmp_param->replyopts.optlen) {
 		ipc.opt = &icmp_param->replyopts;
 		if (ipc.opt->srr)
@@ -538,7 +538,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 	inet_sk(sk)->tos = tos;
 	ipc.addr = iph->saddr;
 	ipc.opt = &icmp_param.replyopts;
-	ipc.shtx.flags = 0;
+	ipc.tx_flags = 0;
 
 	{
 		struct flowi fl = {
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 04b69896df5f..e807492f1777 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -953,7 +953,7 @@ alloc_new_skb:
 				else
 					/* only the initial fragment is
 					   time stamped */
-					ipc->shtx.flags = 0;
+					ipc->tx_flags = 0;
 			}
 			if (skb == NULL)
 				goto error;
@@ -964,7 +964,7 @@ alloc_new_skb:
 			skb->ip_summed = csummode;
 			skb->csum = 0;
 			skb_reserve(skb, hh_len);
-			*skb_tx(skb) = ipc->shtx;
+			skb_shinfo(skb)->tx_flags = ipc->tx_flags;
 
 			/*
 			 *	Find where to start putting bytes.
@@ -1384,7 +1384,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar
 
 	daddr = ipc.addr = rt->rt_src;
 	ipc.opt = NULL;
-	ipc.shtx.flags = 0;
+	ipc.tx_flags = 0;
 
 	if (replyopts.opt.optlen) {
 		ipc.opt = &replyopts.opt;
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 009a7b2aa1ef..1f85ef289895 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -505,7 +505,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 
 	ipc.addr = inet->inet_saddr;
 	ipc.opt = NULL;
-	ipc.shtx.flags = 0;
+	ipc.tx_flags = 0;
 	ipc.oif = sk->sk_bound_dev_if;
 
 	if (msg->msg_controllen) {
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 32e0bef60d0a..86e757e162ee 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -797,7 +797,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		return -EOPNOTSUPP;
 
 	ipc.opt = NULL;
-	ipc.shtx.flags = 0;
+	ipc.tx_flags = 0;
 
 	if (up->pending) {
 		/*
@@ -845,7 +845,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	ipc.addr = inet->inet_saddr;
 
 	ipc.oif = sk->sk_bound_dev_if;
-	err = sock_tx_timestamp(msg, sk, &ipc.shtx);
+	err = sock_tx_timestamp(sk, &ipc.tx_flags);
 	if (err)
 		return err;
 	if (msg->msg_controllen) {
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 9a17f28b1253..3616f27b9d46 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -488,7 +488,7 @@ retry:
 	skb->dev = dev;
 	skb->priority = sk->sk_priority;
 	skb->mark = sk->sk_mark;
-	err = sock_tx_timestamp(msg, sk, skb_tx(skb));
+	err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags);
 	if (err < 0)
 		goto out_unlock;
 
@@ -1209,7 +1209,7 @@ static int packet_snd(struct socket *sock,
 	err = skb_copy_datagram_from_iovec(skb, offset, msg->msg_iov, 0, len);
 	if (err)
 		goto out_free;
-	err = sock_tx_timestamp(msg, sk, skb_tx(skb));
+	err = sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags);
 	if (err < 0)
 		goto out_free;
 
diff --git a/net/socket.c b/net/socket.c
index 2270b941bcc7..7848d12f5e4d 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -535,14 +535,13 @@ void sock_release(struct socket *sock)
 }
 EXPORT_SYMBOL(sock_release);
 
-int sock_tx_timestamp(struct msghdr *msg, struct sock *sk,
-		      union skb_shared_tx *shtx)
+int sock_tx_timestamp(struct sock *sk, __u8 *tx_flags)
 {
-	shtx->flags = 0;
+	*tx_flags = 0;
 	if (sock_flag(sk, SOCK_TIMESTAMPING_TX_HARDWARE))
-		shtx->hardware = 1;
+		*tx_flags |= SKBTX_HW_TSTAMP;
 	if (sock_flag(sk, SOCK_TIMESTAMPING_TX_SOFTWARE))
-		shtx->software = 1;
+		*tx_flags |= SKBTX_SW_TSTAMP;
 	return 0;
 }
 EXPORT_SYMBOL(sock_tx_timestamp);
-- 
cgit v1.2.3


From e760702ed8333588f9f21e7bf6597873993006f1 Mon Sep 17 00:00:00 2001
From: Changli Gao <xiaosuo@gmail.com>
Date: Tue, 17 Aug 2010 19:03:44 +0000
Subject: net: introduce proto_ports_offset()

Introduce proto_ports_offset() for getting the position of the ports or
SPI in the message of a protocol.

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/in.h | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/in.h b/include/linux/in.h
index 41d88a4689af..beeb6dee2b49 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h
@@ -250,6 +250,25 @@ struct sockaddr_in {
 
 #ifdef __KERNEL__
 
+#include <linux/errno.h>
+
+static inline int proto_ports_offset(int proto)
+{
+	switch (proto) {
+	case IPPROTO_TCP:
+	case IPPROTO_UDP:
+	case IPPROTO_DCCP:
+	case IPPROTO_ESP:	/* SPI */
+	case IPPROTO_SCTP:
+	case IPPROTO_UDPLITE:
+		return 0;
+	case IPPROTO_AH:	/* SPI */
+		return 4;
+	default:
+		return -EINVAL;
+	}
+}
+
 static inline bool ipv4_is_loopback(__be32 addr)
 {
 	return (addr & htonl(0xff000000)) == htonl(0x7f000000);
-- 
cgit v1.2.3


From ca5ecddfa8fcbd948c95530e7e817cee9fb43a3d Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Wed, 28 Apr 2010 14:39:09 -0700
Subject: rcu: define __rcu address space modifier for sparse

This commit provides definitions for the __rcu annotation defined earlier.
This annotation permits sparse to check for correct use of RCU-protected
pointers.  If a pointer that is annotated with __rcu is accessed
directly (as opposed to via rcu_dereference(), rcu_assign_pointer(),
or one of their variants), sparse can be made to complain.  To enable
such complaints, use the new default-disabled CONFIG_SPARSE_RCU_POINTER
kernel configuration option.  Please note that these sparse complaints are
intended to be a debugging aid, -not- a code-style-enforcement mechanism.

There are special rcu_dereference_protected() and rcu_access_pointer()
accessors for use when RCU read-side protection is not required, for
example, when no other CPU has access to the data structure in question
or while the current CPU hold the update-side lock.

This patch also updates a number of docbook comments that were showing
their age.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Christopher Li <sparse@chrisli.org>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/compiler.h |   4 +
 include/linux/rcupdate.h | 352 ++++++++++++++++++++++++++++-------------------
 include/linux/srcu.h     |  27 +++-
 kernel/rcupdate.c        |   6 +-
 lib/Kconfig.debug        |  13 ++
 5 files changed, 257 insertions(+), 145 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index c1a62c56a660..320d6c94ff84 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -16,7 +16,11 @@
 # define __release(x)	__context__(x,-1)
 # define __cond_lock(x,c)	((c) ? ({ __acquire(x); 1; }) : 0)
 # define __percpu	__attribute__((noderef, address_space(3)))
+#ifdef CONFIG_SPARSE_RCU_POINTER
+# define __rcu		__attribute__((noderef, address_space(4)))
+#else
 # define __rcu
+#endif
 extern void __chk_user_ptr(const volatile void __user *);
 extern void __chk_io_ptr(const volatile void __iomem *);
 #else
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 9fbc54a2585d..b973dea2d6b0 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -41,6 +41,7 @@
 #include <linux/lockdep.h>
 #include <linux/completion.h>
 #include <linux/debugobjects.h>
+#include <linux/compiler.h>
 
 #ifdef CONFIG_RCU_TORTURE_TEST
 extern int rcutorture_runnable; /* for sysctl */
@@ -120,14 +121,15 @@ extern struct lockdep_map rcu_sched_lock_map;
 extern int debug_lockdep_rcu_enabled(void);
 
 /**
- * rcu_read_lock_held - might we be in RCU read-side critical section?
+ * rcu_read_lock_held() - might we be in RCU read-side critical section?
  *
  * If CONFIG_DEBUG_LOCK_ALLOC is selected, returns nonzero iff in an RCU
  * read-side critical section.  In absence of CONFIG_DEBUG_LOCK_ALLOC,
  * this assumes we are in an RCU read-side critical section unless it can
- * prove otherwise.
+ * prove otherwise.  This is useful for debug checks in functions that
+ * require that they be called within an RCU read-side critical section.
  *
- * Check debug_lockdep_rcu_enabled() to prevent false positives during boot
+ * Checks debug_lockdep_rcu_enabled() to prevent false positives during boot
  * and while lockdep is disabled.
  */
 static inline int rcu_read_lock_held(void)
@@ -144,14 +146,16 @@ static inline int rcu_read_lock_held(void)
 extern int rcu_read_lock_bh_held(void);
 
 /**
- * rcu_read_lock_sched_held - might we be in RCU-sched read-side critical section?
+ * rcu_read_lock_sched_held() - might we be in RCU-sched read-side critical section?
  *
  * If CONFIG_DEBUG_LOCK_ALLOC is selected, returns nonzero iff in an
  * RCU-sched read-side critical section.  In absence of
  * CONFIG_DEBUG_LOCK_ALLOC, this assumes we are in an RCU-sched read-side
  * critical section unless it can prove otherwise.  Note that disabling
  * of preemption (including disabling irqs) counts as an RCU-sched
- * read-side critical section.
+ * read-side critical section.  This is useful for debug checks in functions
+ * that required that they be called within an RCU-sched read-side
+ * critical section.
  *
  * Check debug_lockdep_rcu_enabled() to prevent false positives during boot
  * and while lockdep is disabled.
@@ -220,41 +224,155 @@ extern int rcu_my_thread_group_empty(void);
 		}							\
 	} while (0)
 
+#else /* #ifdef CONFIG_PROVE_RCU */
+
+#define __do_rcu_dereference_check(c) do { } while (0)
+
+#endif /* #else #ifdef CONFIG_PROVE_RCU */
+
+/*
+ * Helper functions for rcu_dereference_check(), rcu_dereference_protected()
+ * and rcu_assign_pointer().  Some of these could be folded into their
+ * callers, but they are left separate in order to ease introduction of
+ * multiple flavors of pointers to match the multiple flavors of RCU
+ * (e.g., __rcu_bh, * __rcu_sched, and __srcu), should this make sense in
+ * the future.
+ */
+#define __rcu_access_pointer(p, space) \
+	({ \
+		typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \
+		(void) (((typeof (*p) space *)p) == p); \
+		((typeof(*p) __force __kernel *)(_________p1)); \
+	})
+#define __rcu_dereference_check(p, c, space) \
+	({ \
+		typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \
+		__do_rcu_dereference_check(c); \
+		(void) (((typeof (*p) space *)p) == p); \
+		smp_read_barrier_depends(); \
+		((typeof(*p) __force __kernel *)(_________p1)); \
+	})
+#define __rcu_dereference_protected(p, c, space) \
+	({ \
+		__do_rcu_dereference_check(c); \
+		(void) (((typeof (*p) space *)p) == p); \
+		((typeof(*p) __force __kernel *)(p)); \
+	})
+
+#define __rcu_dereference_index_check(p, c) \
+	({ \
+		typeof(p) _________p1 = ACCESS_ONCE(p); \
+		__do_rcu_dereference_check(c); \
+		smp_read_barrier_depends(); \
+		(_________p1); \
+	})
+#define __rcu_assign_pointer(p, v, space) \
+	({ \
+		if (!__builtin_constant_p(v) || \
+		    ((v) != NULL)) \
+			smp_wmb(); \
+		(p) = (typeof(*v) __force space *)(v); \
+	})
+
+
+/**
+ * rcu_access_pointer() - fetch RCU pointer with no dereferencing
+ * @p: The pointer to read
+ *
+ * Return the value of the specified RCU-protected pointer, but omit the
+ * smp_read_barrier_depends() and keep the ACCESS_ONCE().  This is useful
+ * when the value of this pointer is accessed, but the pointer is not
+ * dereferenced, for example, when testing an RCU-protected pointer against
+ * NULL.  Although rcu_access_pointer() may also be used in cases where
+ * update-side locks prevent the value of the pointer from changing, you
+ * should instead use rcu_dereference_protected() for this use case.
+ */
+#define rcu_access_pointer(p) __rcu_access_pointer((p), __rcu)
+
 /**
- * rcu_dereference_check - rcu_dereference with debug checking
+ * rcu_dereference_check() - rcu_dereference with debug checking
  * @p: The pointer to read, prior to dereferencing
  * @c: The conditions under which the dereference will take place
  *
  * Do an rcu_dereference(), but check that the conditions under which the
- * dereference will take place are correct.  Typically the conditions indicate
- * the various locking conditions that should be held at that point.  The check
- * should return true if the conditions are satisfied.
+ * dereference will take place are correct.  Typically the conditions
+ * indicate the various locking conditions that should be held at that
+ * point.  The check should return true if the conditions are satisfied.
+ * An implicit check for being in an RCU read-side critical section
+ * (rcu_read_lock()) is included.
  *
  * For example:
  *
- *	bar = rcu_dereference_check(foo->bar, rcu_read_lock_held() ||
- *					      lockdep_is_held(&foo->lock));
+ *	bar = rcu_dereference_check(foo->bar, lockdep_is_held(&foo->lock));
  *
  * could be used to indicate to lockdep that foo->bar may only be dereferenced
- * if either the RCU read lock is held, or that the lock required to replace
+ * if either rcu_read_lock() is held, or that the lock required to replace
  * the bar struct at foo->bar is held.
  *
  * Note that the list of conditions may also include indications of when a lock
  * need not be held, for example during initialisation or destruction of the
  * target struct:
  *
- *	bar = rcu_dereference_check(foo->bar, rcu_read_lock_held() ||
- *					      lockdep_is_held(&foo->lock) ||
+ *	bar = rcu_dereference_check(foo->bar, lockdep_is_held(&foo->lock) ||
  *					      atomic_read(&foo->usage) == 0);
+ *
+ * Inserts memory barriers on architectures that require them
+ * (currently only the Alpha), prevents the compiler from refetching
+ * (and from merging fetches), and, more importantly, documents exactly
+ * which pointers are protected by RCU and checks that the pointer is
+ * annotated as __rcu.
  */
 #define rcu_dereference_check(p, c) \
-	({ \
-		__do_rcu_dereference_check(c); \
-		rcu_dereference_raw(p); \
-	})
+	__rcu_dereference_check((p), rcu_read_lock_held() || (c), __rcu)
+
+/**
+ * rcu_dereference_bh_check() - rcu_dereference_bh with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * This is the RCU-bh counterpart to rcu_dereference_check().
+ */
+#define rcu_dereference_bh_check(p, c) \
+	__rcu_dereference_check((p), rcu_read_lock_bh_held() || (c), __rcu)
 
 /**
- * rcu_dereference_protected - fetch RCU pointer when updates prevented
+ * rcu_dereference_sched_check() - rcu_dereference_sched with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * This is the RCU-sched counterpart to rcu_dereference_check().
+ */
+#define rcu_dereference_sched_check(p, c) \
+	__rcu_dereference_check((p), rcu_read_lock_sched_held() || (c), \
+				__rcu)
+
+#define rcu_dereference_raw(p) rcu_dereference_check(p, 1) /*@@@ needed? @@@*/
+
+/**
+ * rcu_dereference_index_check() - rcu_dereference for indices with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * Similar to rcu_dereference_check(), but omits the sparse checking.
+ * This allows rcu_dereference_index_check() to be used on integers,
+ * which can then be used as array indices.  Attempting to use
+ * rcu_dereference_check() on an integer will give compiler warnings
+ * because the sparse address-space mechanism relies on dereferencing
+ * the RCU-protected pointer.  Dereferencing integers is not something
+ * that even gcc will put up with.
+ *
+ * Note that this function does not implicitly check for RCU read-side
+ * critical sections.  If this function gains lots of uses, it might
+ * make sense to provide versions for each flavor of RCU, but it does
+ * not make sense as of early 2010.
+ */
+#define rcu_dereference_index_check(p, c) \
+	__rcu_dereference_index_check((p), (c))
+
+/**
+ * rcu_dereference_protected() - fetch RCU pointer when updates prevented
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
  *
  * Return the value of the specified RCU-protected pointer, but omit
  * both the smp_read_barrier_depends() and the ACCESS_ONCE().  This
@@ -263,35 +381,61 @@ extern int rcu_my_thread_group_empty(void);
  * prevent the compiler from repeating this reference or combining it
  * with other references, so it should not be used without protection
  * of appropriate locks.
+ *
+ * This function is only for update-side use.  Using this function
+ * when protected only by rcu_read_lock() will result in infrequent
+ * but very ugly failures.
  */
 #define rcu_dereference_protected(p, c) \
-	({ \
-		__do_rcu_dereference_check(c); \
-		(p); \
-	})
+	__rcu_dereference_protected((p), (c), __rcu)
 
-#else /* #ifdef CONFIG_PROVE_RCU */
+/**
+ * rcu_dereference_bh_protected() - fetch RCU-bh pointer when updates prevented
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * This is the RCU-bh counterpart to rcu_dereference_protected().
+ */
+#define rcu_dereference_bh_protected(p, c) \
+	__rcu_dereference_protected((p), (c), __rcu)
 
-#define rcu_dereference_check(p, c)	rcu_dereference_raw(p)
-#define rcu_dereference_protected(p, c) (p)
+/**
+ * rcu_dereference_sched_protected() - fetch RCU-sched pointer when updates prevented
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * This is the RCU-sched counterpart to rcu_dereference_protected().
+ */
+#define rcu_dereference_sched_protected(p, c) \
+	__rcu_dereference_protected((p), (c), __rcu)
 
-#endif /* #else #ifdef CONFIG_PROVE_RCU */
 
 /**
- * rcu_access_pointer - fetch RCU pointer with no dereferencing
+ * rcu_dereference() - fetch RCU-protected pointer for dereferencing
+ * @p: The pointer to read, prior to dereferencing
  *
- * Return the value of the specified RCU-protected pointer, but omit the
- * smp_read_barrier_depends() and keep the ACCESS_ONCE().  This is useful
- * when the value of this pointer is accessed, but the pointer is not
- * dereferenced, for example, when testing an RCU-protected pointer against
- * NULL.  This may also be used in cases where update-side locks prevent
- * the value of the pointer from changing, but rcu_dereference_protected()
- * is a lighter-weight primitive for this use case.
+ * This is a simple wrapper around rcu_dereference_check().
  */
-#define rcu_access_pointer(p)	ACCESS_ONCE(p)
+#define rcu_dereference(p) rcu_dereference_check(p, 0)
 
 /**
- * rcu_read_lock - mark the beginning of an RCU read-side critical section.
+ * rcu_dereference_bh() - fetch an RCU-bh-protected pointer for dereferencing
+ * @p: The pointer to read, prior to dereferencing
+ *
+ * Makes rcu_dereference_check() do the dirty work.
+ */
+#define rcu_dereference_bh(p) rcu_dereference_bh_check(p, 0)
+
+/**
+ * rcu_dereference_sched() - fetch RCU-sched-protected pointer for dereferencing
+ * @p: The pointer to read, prior to dereferencing
+ *
+ * Makes rcu_dereference_check() do the dirty work.
+ */
+#define rcu_dereference_sched(p) rcu_dereference_sched_check(p, 0)
+
+/**
+ * rcu_read_lock() - mark the beginning of an RCU read-side critical section
  *
  * When synchronize_rcu() is invoked on one CPU while other CPUs
  * are within RCU read-side critical sections, then the
@@ -337,7 +481,7 @@ static inline void rcu_read_lock(void)
  */
 
 /**
- * rcu_read_unlock - marks the end of an RCU read-side critical section.
+ * rcu_read_unlock() - marks the end of an RCU read-side critical section.
  *
  * See rcu_read_lock() for more information.
  */
@@ -349,15 +493,16 @@ static inline void rcu_read_unlock(void)
 }
 
 /**
- * rcu_read_lock_bh - mark the beginning of a softirq-only RCU critical section
+ * rcu_read_lock_bh() - mark the beginning of an RCU-bh critical section
  *
  * This is equivalent of rcu_read_lock(), but to be used when updates
- * are being done using call_rcu_bh(). Since call_rcu_bh() callbacks
- * consider completion of a softirq handler to be a quiescent state,
- * a process in RCU read-side critical section must be protected by
- * disabling softirqs. Read-side critical sections in interrupt context
- * can use just rcu_read_lock().
- *
+ * are being done using call_rcu_bh() or synchronize_rcu_bh(). Since
+ * both call_rcu_bh() and synchronize_rcu_bh() consider completion of a
+ * softirq handler to be a quiescent state, a process in RCU read-side
+ * critical section must be protected by disabling softirqs. Read-side
+ * critical sections in interrupt context can use just rcu_read_lock(),
+ * though this should at least be commented to avoid confusing people
+ * reading the code.
  */
 static inline void rcu_read_lock_bh(void)
 {
@@ -379,13 +524,12 @@ static inline void rcu_read_unlock_bh(void)
 }
 
 /**
- * rcu_read_lock_sched - mark the beginning of a RCU-classic critical section
+ * rcu_read_lock_sched() - mark the beginning of a RCU-sched critical section
  *
- * Should be used with either
- * - synchronize_sched()
- * or
- * - call_rcu_sched() and rcu_barrier_sched()
- * on the write-side to insure proper synchronization.
+ * This is equivalent of rcu_read_lock(), but to be used when updates
+ * are being done using call_rcu_sched() or synchronize_rcu_sched().
+ * Read-side critical sections can also be introduced by anything that
+ * disables preemption, including local_irq_disable() and friends.
  */
 static inline void rcu_read_lock_sched(void)
 {
@@ -420,54 +564,14 @@ static inline notrace void rcu_read_unlock_sched_notrace(void)
 	preempt_enable_notrace();
 }
 
-
 /**
- * rcu_dereference_raw - fetch an RCU-protected pointer
+ * rcu_assign_pointer() - assign to RCU-protected pointer
+ * @p: pointer to assign to
+ * @v: value to assign (publish)
  *
- * The caller must be within some flavor of RCU read-side critical
- * section, or must be otherwise preventing the pointer from changing,
- * for example, by holding an appropriate lock.  This pointer may later
- * be safely dereferenced.  It is the caller's responsibility to have
- * done the right thing, as this primitive does no checking of any kind.
- *
- * Inserts memory barriers on architectures that require them
- * (currently only the Alpha), and, more importantly, documents
- * exactly which pointers are protected by RCU.
- */
-#define rcu_dereference_raw(p)	({ \
-				typeof(p) _________p1 = ACCESS_ONCE(p); \
-				smp_read_barrier_depends(); \
-				(_________p1); \
-				})
-
-/**
- * rcu_dereference - fetch an RCU-protected pointer, checking for RCU
- *
- * Makes rcu_dereference_check() do the dirty work.
- */
-#define rcu_dereference(p) \
-	rcu_dereference_check(p, rcu_read_lock_held())
-
-/**
- * rcu_dereference_bh - fetch an RCU-protected pointer, checking for RCU-bh
- *
- * Makes rcu_dereference_check() do the dirty work.
- */
-#define rcu_dereference_bh(p) \
-		rcu_dereference_check(p, rcu_read_lock_bh_held())
-
-/**
- * rcu_dereference_sched - fetch RCU-protected pointer, checking for RCU-sched
- *
- * Makes rcu_dereference_check() do the dirty work.
- */
-#define rcu_dereference_sched(p) \
-		rcu_dereference_check(p, rcu_read_lock_sched_held())
-
-/**
- * rcu_assign_pointer - assign (publicize) a pointer to a newly
- * initialized structure that will be dereferenced by RCU read-side
- * critical sections.  Returns the value assigned.
+ * Assigns the specified value to the specified RCU-protected
+ * pointer, ensuring that any concurrent RCU readers will see
+ * any prior initialization.  Returns the value assigned.
  *
  * Inserts memory barriers on architectures that require them
  * (pretty much all of them other than x86), and also prevents
@@ -476,14 +580,17 @@ static inline notrace void rcu_read_unlock_sched_notrace(void)
  * call documents which pointers will be dereferenced by RCU read-side
  * code.
  */
-
 #define rcu_assign_pointer(p, v) \
-	({ \
-		if (!__builtin_constant_p(v) || \
-		    ((v) != NULL)) \
-			smp_wmb(); \
-		(p) = (v); \
-	})
+	__rcu_assign_pointer((p), (v), __rcu)
+
+/**
+ * RCU_INIT_POINTER() - initialize an RCU protected pointer
+ *
+ * Initialize an RCU-protected pointer in such a way to avoid RCU-lockdep
+ * splats.
+ */
+#define RCU_INIT_POINTER(p, v) \
+		p = (typeof(*v) __force __rcu *)(v)
 
 /* Infrastructure to implement the synchronize_() primitives. */
 
@@ -495,7 +602,7 @@ struct rcu_synchronize {
 extern void wakeme_after_rcu(struct rcu_head  *head);
 
 /**
- * call_rcu - Queue an RCU callback for invocation after a grace period.
+ * call_rcu() - Queue an RCU callback for invocation after a grace period.
  * @head: structure to be used for queueing the RCU updates.
  * @func: actual update function to be invoked after the grace period
  *
@@ -509,7 +616,7 @@ extern void call_rcu(struct rcu_head *head,
 			      void (*func)(struct rcu_head *head));
 
 /**
- * call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
+ * call_rcu_bh() - Queue an RCU for invocation after a quicker grace period.
  * @head: structure to be used for queueing the RCU updates.
  * @func: actual update function to be invoked after the grace period
  *
@@ -566,37 +673,4 @@ static inline void debug_rcu_head_unqueue(struct rcu_head *head)
 }
 #endif	/* #else !CONFIG_DEBUG_OBJECTS_RCU_HEAD */
 
-#ifndef CONFIG_PROVE_RCU
-#define __do_rcu_dereference_check(c) do { } while (0)
-#endif /* #ifdef CONFIG_PROVE_RCU */
-
-#define __rcu_dereference_index_check(p, c) \
-	({ \
-		typeof(p) _________p1 = ACCESS_ONCE(p); \
-		__do_rcu_dereference_check(c); \
-		smp_read_barrier_depends(); \
-		(_________p1); \
-	})
-
-/**
- * rcu_dereference_index_check() - rcu_dereference for indices with debug checking
- * @p: The pointer to read, prior to dereferencing
- * @c: The conditions under which the dereference will take place
- *
- * Similar to rcu_dereference_check(), but omits the sparse checking.
- * This allows rcu_dereference_index_check() to be used on integers,
- * which can then be used as array indices.  Attempting to use
- * rcu_dereference_check() on an integer will give compiler warnings
- * because the sparse address-space mechanism relies on dereferencing
- * the RCU-protected pointer.  Dereferencing integers is not something
- * that even gcc will put up with.
- *
- * Note that this function does not implicitly check for RCU read-side
- * critical sections.  If this function gains lots of uses, it might
- * make sense to provide versions for each flavor of RCU, but it does
- * not make sense as of early 2010.
- */
-#define rcu_dereference_index_check(p, c) \
-	__rcu_dereference_index_check((p), (c))
-
 #endif /* __LINUX_RCUPDATE_H */
diff --git a/include/linux/srcu.h b/include/linux/srcu.h
index 4d5d2f546dbf..6f456a720ff0 100644
--- a/include/linux/srcu.h
+++ b/include/linux/srcu.h
@@ -108,12 +108,31 @@ static inline int srcu_read_lock_held(struct srcu_struct *sp)
 #endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */
 
 /**
- * srcu_dereference - fetch SRCU-protected pointer with checking
+ * srcu_dereference_check - fetch SRCU-protected pointer for later dereferencing
+ * @p: the pointer to fetch and protect for later dereferencing
+ * @sp: pointer to the srcu_struct, which is used to check that we
+ *	really are in an SRCU read-side critical section.
+ * @c: condition to check for update-side use
  *
- * Makes rcu_dereference_check() do the dirty work.
+ * If PROVE_RCU is enabled, invoking this outside of an RCU read-side
+ * critical section will result in an RCU-lockdep splat, unless @c evaluates
+ * to 1.  The @c argument will normally be a logical expression containing
+ * lockdep_is_held() calls.
  */
-#define srcu_dereference(p, sp) \
-		rcu_dereference_check(p, srcu_read_lock_held(sp))
+#define srcu_dereference_check(p, sp, c) \
+	__rcu_dereference_check((p), srcu_read_lock_held(sp) || (c), __rcu)
+
+/**
+ * srcu_dereference - fetch SRCU-protected pointer for later dereferencing
+ * @p: the pointer to fetch and protect for later dereferencing
+ * @sp: pointer to the srcu_struct, which is used to check that we
+ *	really are in an SRCU read-side critical section.
+ *
+ * Makes rcu_dereference_check() do the dirty work.  If PROVE_RCU
+ * is enabled, invoking this outside of an RCU read-side critical
+ * section will result in an RCU-lockdep splat.
+ */
+#define srcu_dereference(p, sp) srcu_dereference_check((p), (sp), 0)
 
 /**
  * srcu_read_lock - register a new reader for an SRCU-protected structure.
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index 4d169835fb36..6c79e851521c 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -73,12 +73,14 @@ int debug_lockdep_rcu_enabled(void)
 EXPORT_SYMBOL_GPL(debug_lockdep_rcu_enabled);
 
 /**
- * rcu_read_lock_bh_held - might we be in RCU-bh read-side critical section?
+ * rcu_read_lock_bh_held() - might we be in RCU-bh read-side critical section?
  *
  * Check for bottom half being disabled, which covers both the
  * CONFIG_PROVE_RCU and not cases.  Note that if someone uses
  * rcu_read_lock_bh(), but then later enables BH, lockdep (if enabled)
- * will show the situation.
+ * will show the situation.  This is useful for debug checks in functions
+ * that require that they be called within an RCU read-side critical
+ * section.
  *
  * Check debug_lockdep_rcu_enabled() to prevent false positives during boot.
  */
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 1b4afd2e6ca0..12465f2ef766 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -539,6 +539,19 @@ config PROVE_RCU_REPEATEDLY
 	 disabling, allowing multiple RCU-lockdep warnings to be printed
 	 on a single reboot.
 
+config SPARSE_RCU_POINTER
+	bool "RCU debugging: sparse-based checks for pointer usage"
+	default n
+	help
+	 This feature enables the __rcu sparse annotation for
+	 RCU-protected pointers.  This annotation will cause sparse
+	 to flag any non-RCU used of annotated pointers.  This can be
+	 helpful when debugging RCU usage.  Please note that this feature
+	 is not intended to enforce code cleanliness; it is instead merely
+	 a debugging aid.
+
+	 Say Y to make sparse flag questionable use of RCU-protected pointers
+
 	 Say N if you are unsure.
 
 config LOCKDEP
-- 
cgit v1.2.3


From 67bdbffd696f29a0b68aa8daa285783a06651583 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 25 Feb 2010 16:55:13 +0100
Subject: rculist: avoid __rcu annotations

This avoids warnings from missing __rcu annotations
in the rculist implementation, making it possible to
use the same lists in both RCU and non-RCU cases.

We can add rculist annotations later, together with
lockdep support for rculist, which is missing as well,
but that may involve changing all the users.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Sukadev Bhattiprolu <sukadev@us.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/rculist.h       | 53 +++++++++++++++++++++++++++----------------
 include/linux/rculist_nulls.h | 16 +++++++++----
 kernel/pid.c                  |  2 +-
 3 files changed, 46 insertions(+), 25 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rculist.h b/include/linux/rculist.h
index 4ec3b38ce9c5..c10b1050dbe6 100644
--- a/include/linux/rculist.h
+++ b/include/linux/rculist.h
@@ -9,6 +9,12 @@
 #include <linux/list.h>
 #include <linux/rcupdate.h>
 
+/*
+ * return the ->next pointer of a list_head in an rcu safe
+ * way, we must not access it directly
+ */
+#define list_next_rcu(list)	(*((struct list_head __rcu **)(&(list)->next)))
+
 /*
  * Insert a new entry between two known consecutive entries.
  *
@@ -20,7 +26,7 @@ static inline void __list_add_rcu(struct list_head *new,
 {
 	new->next = next;
 	new->prev = prev;
-	rcu_assign_pointer(prev->next, new);
+	rcu_assign_pointer(list_next_rcu(prev), new);
 	next->prev = new;
 }
 
@@ -138,7 +144,7 @@ static inline void list_replace_rcu(struct list_head *old,
 {
 	new->next = old->next;
 	new->prev = old->prev;
-	rcu_assign_pointer(new->prev->next, new);
+	rcu_assign_pointer(list_next_rcu(new->prev), new);
 	new->next->prev = new;
 	old->prev = LIST_POISON2;
 }
@@ -193,7 +199,7 @@ static inline void list_splice_init_rcu(struct list_head *list,
 	 */
 
 	last->next = at;
-	rcu_assign_pointer(head->next, first);
+	rcu_assign_pointer(list_next_rcu(head), first);
 	first->prev = head;
 	at->prev = last;
 }
@@ -208,7 +214,9 @@ static inline void list_splice_init_rcu(struct list_head *list,
  * primitives such as list_add_rcu() as long as it's guarded by rcu_read_lock().
  */
 #define list_entry_rcu(ptr, type, member) \
-	container_of(rcu_dereference_raw(ptr), type, member)
+	({typeof (*ptr) __rcu *__ptr = (typeof (*ptr) __rcu __force *)ptr; \
+	 container_of((typeof(ptr))rcu_dereference_raw(__ptr), type, member); \
+	})
 
 /**
  * list_first_entry_rcu - get the first element from a list
@@ -225,9 +233,9 @@ static inline void list_splice_init_rcu(struct list_head *list,
 	list_entry_rcu((ptr)->next, type, member)
 
 #define __list_for_each_rcu(pos, head) \
-	for (pos = rcu_dereference_raw((head)->next); \
+	for (pos = rcu_dereference_raw(list_next_rcu(head)); \
 		pos != (head); \
-		pos = rcu_dereference_raw(pos->next))
+		pos = rcu_dereference_raw(list_next_rcu((pos)))
 
 /**
  * list_for_each_entry_rcu	-	iterate over rcu list of given type
@@ -257,9 +265,9 @@ static inline void list_splice_init_rcu(struct list_head *list,
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define list_for_each_continue_rcu(pos, head) \
-	for ((pos) = rcu_dereference_raw((pos)->next); \
+	for ((pos) = rcu_dereference_raw(list_next_rcu(pos)); \
 		prefetch((pos)->next), (pos) != (head); \
-		(pos) = rcu_dereference_raw((pos)->next))
+		(pos) = rcu_dereference_raw(list_next_rcu(pos)))
 
 /**
  * list_for_each_entry_continue_rcu - continue iteration over list of given type
@@ -314,12 +322,19 @@ static inline void hlist_replace_rcu(struct hlist_node *old,
 
 	new->next = next;
 	new->pprev = old->pprev;
-	rcu_assign_pointer(*new->pprev, new);
+	rcu_assign_pointer(*(struct hlist_node __rcu **)new->pprev, new);
 	if (next)
 		new->next->pprev = &new->next;
 	old->pprev = LIST_POISON2;
 }
 
+/*
+ * return the first or the next element in an RCU protected hlist
+ */
+#define hlist_first_rcu(head)	(*((struct hlist_node __rcu **)(&(head)->first)))
+#define hlist_next_rcu(node)	(*((struct hlist_node __rcu **)(&(node)->next)))
+#define hlist_pprev_rcu(node)	(*((struct hlist_node __rcu **)((node)->pprev)))
+
 /**
  * hlist_add_head_rcu
  * @n: the element to add to the hash list.
@@ -346,7 +361,7 @@ static inline void hlist_add_head_rcu(struct hlist_node *n,
 
 	n->next = first;
 	n->pprev = &h->first;
-	rcu_assign_pointer(h->first, n);
+	rcu_assign_pointer(hlist_first_rcu(h), n);
 	if (first)
 		first->pprev = &n->next;
 }
@@ -374,7 +389,7 @@ static inline void hlist_add_before_rcu(struct hlist_node *n,
 {
 	n->pprev = next->pprev;
 	n->next = next;
-	rcu_assign_pointer(*(n->pprev), n);
+	rcu_assign_pointer(hlist_pprev_rcu(n), n);
 	next->pprev = &n->next;
 }
 
@@ -401,15 +416,15 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev,
 {
 	n->next = prev->next;
 	n->pprev = &prev->next;
-	rcu_assign_pointer(prev->next, n);
+	rcu_assign_pointer(hlist_next_rcu(prev), n);
 	if (n->next)
 		n->next->pprev = &n->next;
 }
 
-#define __hlist_for_each_rcu(pos, head)			\
-	for (pos = rcu_dereference((head)->first);	\
-	     pos && ({ prefetch(pos->next); 1; });	\
-	     pos = rcu_dereference(pos->next))
+#define __hlist_for_each_rcu(pos, head)				\
+	for (pos = rcu_dereference(hlist_first_rcu(head));	\
+	     pos && ({ prefetch(pos->next); 1; });		\
+	     pos = rcu_dereference(hlist_next_rcu(pos)))
 
 /**
  * hlist_for_each_entry_rcu - iterate over rcu list of given type
@@ -422,11 +437,11 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev,
  * the _rcu list-mutation primitives such as hlist_add_head_rcu()
  * as long as the traversal is guarded by rcu_read_lock().
  */
-#define hlist_for_each_entry_rcu(tpos, pos, head, member)		 \
-	for (pos = rcu_dereference_raw((head)->first);			 \
+#define hlist_for_each_entry_rcu(tpos, pos, head, member)		\
+	for (pos = rcu_dereference_raw(hlist_first_rcu(head));		\
 		pos && ({ prefetch(pos->next); 1; }) &&			 \
 		({ tpos = hlist_entry(pos, typeof(*tpos), member); 1; }); \
-		pos = rcu_dereference_raw(pos->next))
+		pos = rcu_dereference_raw(hlist_next_rcu(pos)))
 
 /**
  * hlist_for_each_entry_rcu_bh - iterate over rcu list of given type
diff --git a/include/linux/rculist_nulls.h b/include/linux/rculist_nulls.h
index b70ffe53cb9f..2ae13714828b 100644
--- a/include/linux/rculist_nulls.h
+++ b/include/linux/rculist_nulls.h
@@ -37,6 +37,12 @@ static inline void hlist_nulls_del_init_rcu(struct hlist_nulls_node *n)
 	}
 }
 
+#define hlist_nulls_first_rcu(head) \
+	(*((struct hlist_nulls_node __rcu __force **)&(head)->first))
+
+#define hlist_nulls_next_rcu(node) \
+	(*((struct hlist_nulls_node __rcu __force **)&(node)->next))
+
 /**
  * hlist_nulls_del_rcu - deletes entry from hash list without re-initialization
  * @n: the element to delete from the hash list.
@@ -88,7 +94,7 @@ static inline void hlist_nulls_add_head_rcu(struct hlist_nulls_node *n,
 
 	n->next = first;
 	n->pprev = &h->first;
-	rcu_assign_pointer(h->first, n);
+	rcu_assign_pointer(hlist_nulls_first_rcu(h), n);
 	if (!is_a_nulls(first))
 		first->pprev = &n->next;
 }
@@ -100,11 +106,11 @@ static inline void hlist_nulls_add_head_rcu(struct hlist_nulls_node *n,
  * @member:	the name of the hlist_nulls_node within the struct.
  *
  */
-#define hlist_nulls_for_each_entry_rcu(tpos, pos, head, member) \
-	for (pos = rcu_dereference_raw((head)->first);			 \
-		(!is_a_nulls(pos)) &&			\
+#define hlist_nulls_for_each_entry_rcu(tpos, pos, head, member)			\
+	for (pos = rcu_dereference_raw(hlist_nulls_first_rcu(head));		\
+		(!is_a_nulls(pos)) &&						\
 		({ tpos = hlist_nulls_entry(pos, typeof(*tpos), member); 1; }); \
-		pos = rcu_dereference_raw(pos->next))
+		pos = rcu_dereference_raw(hlist_nulls_next_rcu(pos)))
 
 #endif
 #endif
diff --git a/kernel/pid.c b/kernel/pid.c
index d55c6fb8d087..0f90c2f713f1 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -401,7 +401,7 @@ struct task_struct *pid_task(struct pid *pid, enum pid_type type)
 	struct task_struct *result = NULL;
 	if (pid) {
 		struct hlist_node *first;
-		first = rcu_dereference_check(pid->tasks[type].first,
+		first = rcu_dereference_check(hlist_first_rcu(&pid->tasks[type]),
 					      rcu_read_lock_held() ||
 					      lockdep_tasklist_lock_is_held());
 		if (first)
-- 
cgit v1.2.3


From 2c392b8c3450ceb69ba1b93cb0cddb3998fb8cdc Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 24 Feb 2010 19:41:39 +0100
Subject: cgroups: __rcu annotations

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Paul Menage <menage@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/cgroup.h | 4 ++--
 include/linux/sched.h  | 2 +-
 kernel/cgroup.c        | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index ed3e92e41c6e..3cb7d04308cd 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -75,7 +75,7 @@ struct cgroup_subsys_state {
 
 	unsigned long flags;
 	/* ID for this css, if possible */
-	struct css_id *id;
+	struct css_id __rcu *id;
 };
 
 /* bits in struct cgroup_subsys_state flags field */
@@ -205,7 +205,7 @@ struct cgroup {
 	struct list_head children;	/* my children */
 
 	struct cgroup *parent;		/* my parent */
-	struct dentry *dentry;	  	/* cgroup fs entry, RCU protected */
+	struct dentry __rcu *dentry;	/* cgroup fs entry, RCU protected */
 
 	/* Private pointers for each registered subsystem */
 	struct cgroup_subsys_state *subsys[CGROUP_SUBSYS_COUNT];
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1e2a6db2d7dd..bbffd087476c 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1418,7 +1418,7 @@ struct task_struct {
 #endif
 #ifdef CONFIG_CGROUPS
 	/* Control Group info protected by css_set_lock */
-	struct css_set *cgroups;
+	struct css_set __rcu *cgroups;
 	/* cg_list protected by css_set_lock and tsk->alloc_lock */
 	struct list_head cg_list;
 #endif
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 192f88c5b0f9..e5c5497a7dca 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -138,7 +138,7 @@ struct css_id {
 	 * is called after synchronize_rcu(). But for safe use, css_is_removed()
 	 * css_tryget() should be used for avoiding race.
 	 */
-	struct cgroup_subsys_state *css;
+	struct cgroup_subsys_state __rcu *css;
 	/*
 	 * ID of this css.
 	 */
-- 
cgit v1.2.3


From 1b0ba1c9037b2265d6e5d0165d31e4c0269b603b Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 24 Feb 2010 19:45:09 +0100
Subject: credentials: rcu annotation

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Acked-by: David Howells <dhowells@redhat.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/sched.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index bbffd087476c..2c756666c111 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1288,9 +1288,9 @@ struct task_struct {
 	struct list_head cpu_timers[3];
 
 /* process credentials */
-	const struct cred *real_cred;	/* objective and real subjective task
+	const struct cred __rcu *real_cred; /* objective and real subjective task
 					 * credentials (COW) */
-	const struct cred *cred;	/* effective (overridable) subjective task
+	const struct cred __rcu *cred;	/* effective (overridable) subjective task
 					 * credentials (COW) */
 	struct mutex cred_guard_mutex;	/* guard against foreign influences on
 					 * credential calculations
-- 
cgit v1.2.3


From e63ba744a64d234c8a07c469ab1806443cb0a6ff Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Fri, 26 Feb 2010 18:01:20 +0100
Subject: keys: __rcu annotations

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: David Howells <dhowells@redhat.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/cred.h | 2 +-
 include/linux/key.h  | 3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cred.h b/include/linux/cred.h
index 4d2c39573f36..4aaeab376446 100644
--- a/include/linux/cred.h
+++ b/include/linux/cred.h
@@ -84,7 +84,7 @@ struct thread_group_cred {
 	atomic_t	usage;
 	pid_t		tgid;			/* thread group process ID */
 	spinlock_t	lock;
-	struct key	*session_keyring;	/* keyring inherited over fork */
+	struct key __rcu *session_keyring;	/* keyring inherited over fork */
 	struct key	*process_keyring;	/* keyring private to this process */
 	struct rcu_head	rcu;			/* RCU deletion hook */
 };
diff --git a/include/linux/key.h b/include/linux/key.h
index cd50dfa1d4c2..3db0adce1fda 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -178,8 +178,9 @@ struct key {
 	 */
 	union {
 		unsigned long		value;
+		void __rcu		*rcudata;
 		void			*data;
-		struct keyring_list	*subscriptions;
+		struct keyring_list __rcu *subscriptions;
 	} payload;
 };
 
-- 
cgit v1.2.3


From 5b22216e11f717adc344abc7f97b42e03127c6c0 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 3 Mar 2010 10:20:10 +0100
Subject: nfs: __rcu annotations

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/nfs_fs.h          | 2 +-
 include/linux/sunrpc/auth_gss.h | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 508f8cf6da37..d0edf7d823ae 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -185,7 +185,7 @@ struct nfs_inode {
 	struct nfs4_cached_acl	*nfs4_acl;
         /* NFSv4 state */
 	struct list_head	open_states;
-	struct nfs_delegation	*delegation;
+	struct nfs_delegation __rcu *delegation;
 	fmode_t			 delegation_state;
 	struct rw_semaphore	rwsem;
 #endif /* CONFIG_NFS_V4*/
diff --git a/include/linux/sunrpc/auth_gss.h b/include/linux/sunrpc/auth_gss.h
index 671538d25bc1..8eee9dbbfe7a 100644
--- a/include/linux/sunrpc/auth_gss.h
+++ b/include/linux/sunrpc/auth_gss.h
@@ -69,7 +69,7 @@ struct gss_cl_ctx {
 	enum rpc_gss_proc	gc_proc;
 	u32			gc_seq;
 	spinlock_t		gc_seq_lock;
-	struct gss_ctx		*gc_gss_ctx;
+	struct gss_ctx __rcu	*gc_gss_ctx;
 	struct xdr_netobj	gc_wire_ctx;
 	u32			gc_win;
 	unsigned long		gc_expiry;
@@ -80,7 +80,7 @@ struct gss_upcall_msg;
 struct gss_cred {
 	struct rpc_cred		gc_base;
 	enum rpc_gss_svc	gc_service;
-	struct gss_cl_ctx	*gc_ctx;
+	struct gss_cl_ctx __rcu	*gc_ctx;
 	struct gss_upcall_msg	*gc_upcall;
 	unsigned long		gc_upcall_timestamp;
 	unsigned char		gc_machine_cred : 1;
-- 
cgit v1.2.3


From 2be85279281bafe7de808ca99de59af4fd474c49 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 4 Mar 2010 15:50:28 +0100
Subject: input: __rcu annotations

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Acked-by: Dmitry Torokhov <dtor@mail.ru>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 drivers/input/evdev.c | 2 +-
 include/linux/input.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c
index c908c5f83645..5808731f72d2 100644
--- a/drivers/input/evdev.c
+++ b/drivers/input/evdev.c
@@ -28,7 +28,7 @@ struct evdev {
 	int minor;
 	struct input_handle handle;
 	wait_queue_head_t wait;
-	struct evdev_client *grab;
+	struct evdev_client __rcu *grab;
 	struct list_head client_list;
 	spinlock_t client_lock; /* protects client_list */
 	struct mutex mutex;
diff --git a/include/linux/input.h b/include/linux/input.h
index 896a92227bc4..d6ae1761be97 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -1196,7 +1196,7 @@ struct input_dev {
 	int (*flush)(struct input_dev *dev, struct file *file);
 	int (*event)(struct input_dev *dev, unsigned int type, unsigned int code, int value);
 
-	struct input_handle *grab;
+	struct input_handle __rcu *grab;
 
 	spinlock_t event_lock;
 	struct mutex mutex;
-- 
cgit v1.2.3


From 4b6a2872a2a00042ee50024822ab706e5456aad8 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 4 Mar 2010 15:59:23 +0100
Subject: kvm: add __rcu annotations

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/kvm_host.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index c13cc48697aa..ac740b26eb10 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -205,7 +205,7 @@ struct kvm {
 
 	struct mutex irq_lock;
 #ifdef CONFIG_HAVE_KVM_IRQCHIP
-	struct kvm_irq_routing_table *irq_routing;
+	struct kvm_irq_routing_table __rcu *irq_routing;
 	struct hlist_head mask_notifier_list;
 	struct hlist_head irq_ack_notifier_list;
 #endif
-- 
cgit v1.2.3


From 4221a9918e38b7494cee341dda7b7b4bb8c04bde Mon Sep 17 00:00:00 2001
From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Date: Sat, 26 Jun 2010 01:08:19 +0900
Subject: Add RCU check for find_task_by_vpid().

find_task_by_vpid() says "Must be called under rcu_read_lock().". But due to
commit 3120438 "rcu: Disable lockdep checking in RCU list-traversal primitives",
we are currently unable to catch "find_task_by_vpid() with tasklist_lock held
but RCU lock not held" errors due to the RCU-lockdep checks being
suppressed in the RCU variants of the struct list_head traversals.
This commit therefore places an explicit check for being in an RCU
read-side critical section in find_task_by_pid_ns().

  ===================================================
  [ INFO: suspicious rcu_dereference_check() usage. ]
  ---------------------------------------------------
  kernel/pid.c:386 invoked rcu_dereference_check() without protection!

  other info that might help us debug this:

  rcu_scheduler_active = 1, debug_locks = 1
  1 lock held by rc.sysinit/1102:
   #0:  (tasklist_lock){.+.+..}, at: [<c1048340>] sys_setpgid+0x40/0x160

  stack backtrace:
  Pid: 1102, comm: rc.sysinit Not tainted 2.6.35-rc3-dirty #1
  Call Trace:
   [<c105e714>] lockdep_rcu_dereference+0x94/0xb0
   [<c104b4cd>] find_task_by_pid_ns+0x6d/0x70
   [<c104b4e8>] find_task_by_vpid+0x18/0x20
   [<c1048347>] sys_setpgid+0x47/0x160
   [<c1002b50>] sysenter_do_call+0x12/0x36

Commit updated to use a new rcu_lockdep_assert() exported API rather than
the old internal __do_rcu_dereference().

Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/rcupdate.h | 14 +++++++++-----
 kernel/pid.c             |  1 +
 2 files changed, 10 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index b973dea2d6b0..b124bc6a75ad 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -215,7 +215,11 @@ static inline int rcu_read_lock_sched_held(void)
 
 extern int rcu_my_thread_group_empty(void);
 
-#define __do_rcu_dereference_check(c)					\
+/**
+ * rcu_lockdep_assert - emit lockdep splat if specified condition not met
+ * @c: condition to check
+ */
+#define rcu_lockdep_assert(c)						\
 	do {								\
 		static bool __warned;					\
 		if (debug_lockdep_rcu_enabled() && !__warned && !(c)) {	\
@@ -226,7 +230,7 @@ extern int rcu_my_thread_group_empty(void);
 
 #else /* #ifdef CONFIG_PROVE_RCU */
 
-#define __do_rcu_dereference_check(c) do { } while (0)
+#define rcu_lockdep_assert(c) do { } while (0)
 
 #endif /* #else #ifdef CONFIG_PROVE_RCU */
 
@@ -247,14 +251,14 @@ extern int rcu_my_thread_group_empty(void);
 #define __rcu_dereference_check(p, c, space) \
 	({ \
 		typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \
-		__do_rcu_dereference_check(c); \
+		rcu_lockdep_assert(c); \
 		(void) (((typeof (*p) space *)p) == p); \
 		smp_read_barrier_depends(); \
 		((typeof(*p) __force __kernel *)(_________p1)); \
 	})
 #define __rcu_dereference_protected(p, c, space) \
 	({ \
-		__do_rcu_dereference_check(c); \
+		rcu_lockdep_assert(c); \
 		(void) (((typeof (*p) space *)p) == p); \
 		((typeof(*p) __force __kernel *)(p)); \
 	})
@@ -262,7 +266,7 @@ extern int rcu_my_thread_group_empty(void);
 #define __rcu_dereference_index_check(p, c) \
 	({ \
 		typeof(p) _________p1 = ACCESS_ONCE(p); \
-		__do_rcu_dereference_check(c); \
+		rcu_lockdep_assert(c); \
 		smp_read_barrier_depends(); \
 		(_________p1); \
 	})
diff --git a/kernel/pid.c b/kernel/pid.c
index 0f90c2f713f1..39b65b69584f 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -416,6 +416,7 @@ EXPORT_SYMBOL(pid_task);
  */
 struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns)
 {
+	rcu_lockdep_assert(rcu_read_lock_held());
 	return pid_task(find_pid_ns(nr, ns), PIDTYPE_PID);
 }
 
-- 
cgit v1.2.3


From 77d8485a8b5416c615b6acd95f01bfcacd7d81ff Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Thu, 8 Jul 2010 17:38:59 -0700
Subject: rcu: improve kerneldoc for rcu_read_lock(), call_rcu(), and
 synchronize_rcu()

Make it explicit that new RCU read-side critical sections that start
after call_rcu() and synchronize_rcu() start might still be running
after the end of the relevant grace period.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/rcupdate.h | 16 +++++++++-------
 kernel/rcutree_plugin.h  |  8 +++++---
 2 files changed, 14 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index b124bc6a75ad..3e1b6625553b 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -450,7 +450,7 @@ extern int rcu_my_thread_group_empty(void);
  * until after the all the other CPUs exit their critical sections.
  *
  * Note, however, that RCU callbacks are permitted to run concurrently
- * with RCU read-side critical sections.  One way that this can happen
+ * with new RCU read-side critical sections.  One way that this can happen
  * is via the following sequence of events: (1) CPU 0 enters an RCU
  * read-side critical section, (2) CPU 1 invokes call_rcu() to register
  * an RCU callback, (3) CPU 0 exits the RCU read-side critical section,
@@ -608,11 +608,13 @@ extern void wakeme_after_rcu(struct rcu_head  *head);
 /**
  * call_rcu() - Queue an RCU callback for invocation after a grace period.
  * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
+ * @func: actual callback function to be invoked after the grace period
  *
- * The update function will be invoked some time after a full grace
- * period elapses, in other words after all currently executing RCU
- * read-side critical sections have completed.  RCU read-side critical
+ * The callback function will be invoked some time after a full grace
+ * period elapses, in other words after all pre-existing RCU read-side
+ * critical sections have completed.  However, the callback function
+ * might well execute concurrently with RCU read-side critical sections
+ * that started after call_rcu() was invoked.  RCU read-side critical
  * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
  * and may be nested.
  */
@@ -622,9 +624,9 @@ extern void call_rcu(struct rcu_head *head,
 /**
  * call_rcu_bh() - Queue an RCU for invocation after a quicker grace period.
  * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
+ * @func: actual callback function to be invoked after the grace period
  *
- * The update function will be invoked some time after a full grace
+ * The callback function will be invoked some time after a full grace
  * period elapses, in other words after all currently executing RCU
  * read-side critical sections have completed. call_rcu_bh() assumes
  * that the read-side critical sections end on completion of a softirq
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 9906f85c7780..63bb7714fdeb 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -546,9 +546,11 @@ EXPORT_SYMBOL_GPL(call_rcu);
  *
  * Control will return to the caller some time after a full grace
  * period has elapsed, in other words after all currently executing RCU
- * read-side critical sections have completed.  RCU read-side critical
- * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
- * and may be nested.
+ * read-side critical sections have completed.  Note, however, that
+ * upon return from synchronize_rcu(), the caller might well be executing
+ * concurrently with new RCU read-side critical sections that began while
+ * synchronize_rcu() was waiting.  RCU read-side critical sections are
+ * delimited by rcu_read_lock() and rcu_read_unlock(), and may be nested.
  */
 void synchronize_rcu(void)
 {
-- 
cgit v1.2.3


From 374a8e0dc33c984fac284de7d57d77af3cfdbfb7 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 24 Feb 2010 20:00:13 +0100
Subject: notifiers: __rcu annotations

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/notifier.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/notifier.h b/include/linux/notifier.h
index b2f1a4d83550..2026f9e1ceb8 100644
--- a/include/linux/notifier.h
+++ b/include/linux/notifier.h
@@ -49,28 +49,28 @@
 
 struct notifier_block {
 	int (*notifier_call)(struct notifier_block *, unsigned long, void *);
-	struct notifier_block *next;
+	struct notifier_block __rcu *next;
 	int priority;
 };
 
 struct atomic_notifier_head {
 	spinlock_t lock;
-	struct notifier_block *head;
+	struct notifier_block __rcu *head;
 };
 
 struct blocking_notifier_head {
 	struct rw_semaphore rwsem;
-	struct notifier_block *head;
+	struct notifier_block __rcu *head;
 };
 
 struct raw_notifier_head {
-	struct notifier_block *head;
+	struct notifier_block __rcu *head;
 };
 
 struct srcu_notifier_head {
 	struct mutex mutex;
 	struct srcu_struct srcu;
-	struct notifier_block *head;
+	struct notifier_block __rcu *head;
 };
 
 #define ATOMIC_INIT_NOTIFIER_HEAD(name) do {	\
-- 
cgit v1.2.3


From a1115570b31091f3e3ab9e6cf7ee8d320a42be84 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 25 Feb 2010 23:43:52 +0100
Subject: radix-tree: __rcu annotations

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Nick Piggin <npiggin@suse.de>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/radix-tree.h | 4 +++-
 lib/radix-tree.c           | 2 +-
 2 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h
index 634b8e674ac5..a39cbed9ee17 100644
--- a/include/linux/radix-tree.h
+++ b/include/linux/radix-tree.h
@@ -47,6 +47,8 @@ static inline void *radix_tree_indirect_to_ptr(void *ptr)
 {
 	return (void *)((unsigned long)ptr & ~RADIX_TREE_INDIRECT_PTR);
 }
+#define radix_tree_indirect_to_ptr(ptr) \
+	radix_tree_indirect_to_ptr((void __force *)(ptr))
 
 static inline int radix_tree_is_indirect_ptr(void *ptr)
 {
@@ -61,7 +63,7 @@ static inline int radix_tree_is_indirect_ptr(void *ptr)
 struct radix_tree_root {
 	unsigned int		height;
 	gfp_t			gfp_mask;
-	struct radix_tree_node	*rnode;
+	struct radix_tree_node	__rcu *rnode;
 };
 
 #define RADIX_TREE_INIT(mask)	{					\
diff --git a/lib/radix-tree.c b/lib/radix-tree.c
index e907858498a6..899fb750946f 100644
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -49,7 +49,7 @@ struct radix_tree_node {
 	unsigned int	height;		/* Height from the bottom */
 	unsigned int	count;
 	struct rcu_head	rcu_head;
-	void		*slots[RADIX_TREE_MAP_SIZE];
+	void __rcu	*slots[RADIX_TREE_MAP_SIZE];
 	unsigned long	tags[RADIX_TREE_MAX_TAGS][RADIX_TREE_TAG_LONGS];
 };
 
-- 
cgit v1.2.3


From d2c2486bc8e185548490e8edbc84d185de9eaff1 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Fri, 26 Feb 2010 14:53:26 +0100
Subject: idr: __rcu annotations

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/idr.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/idr.h b/include/linux/idr.h
index e968db71e33a..cdb715e58e3e 100644
--- a/include/linux/idr.h
+++ b/include/linux/idr.h
@@ -50,14 +50,14 @@
 
 struct idr_layer {
 	unsigned long		 bitmap; /* A zero bit means "space here" */
-	struct idr_layer	*ary[1<<IDR_BITS];
+	struct idr_layer __rcu	*ary[1<<IDR_BITS];
 	int			 count;	 /* When zero, we can release it */
 	int			 layer;	 /* distance from leaf */
 	struct rcu_head		 rcu_head;
 };
 
 struct idr {
-	struct idr_layer *top;
+	struct idr_layer __rcu *top;
 	struct idr_layer *id_free;
 	int		  layers; /* only valid without concurrent changes */
 	int		  id_free_cnt;
-- 
cgit v1.2.3


From 4d2deb40b20c2608486598364e63e37b09a9ac2f Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Wed, 24 Feb 2010 20:01:56 +0100
Subject: kernel: __rcu annotations

This adds annotations for RCU operations in core kernel components

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/fdtable.h   | 6 +++---
 include/linux/fs.h        | 2 +-
 include/linux/genhd.h     | 6 +++---
 include/linux/init_task.h | 4 ++--
 include/linux/iocontext.h | 2 +-
 include/linux/mm_types.h  | 2 +-
 6 files changed, 11 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/fdtable.h b/include/linux/fdtable.h
index f59ed297b661..133c0ba25e30 100644
--- a/include/linux/fdtable.h
+++ b/include/linux/fdtable.h
@@ -31,7 +31,7 @@ struct embedded_fd_set {
 
 struct fdtable {
 	unsigned int max_fds;
-	struct file ** fd;      /* current fd array */
+	struct file __rcu **fd;      /* current fd array */
 	fd_set *close_on_exec;
 	fd_set *open_fds;
 	struct rcu_head rcu;
@@ -46,7 +46,7 @@ struct files_struct {
    * read mostly part
    */
 	atomic_t count;
-	struct fdtable *fdt;
+	struct fdtable __rcu *fdt;
 	struct fdtable fdtab;
   /*
    * written part on a separate cache line in SMP
@@ -55,7 +55,7 @@ struct files_struct {
 	int next_fd;
 	struct embedded_fd_set close_on_exec_init;
 	struct embedded_fd_set open_fds_init;
-	struct file * fd_array[NR_OPEN_DEFAULT];
+	struct file __rcu * fd_array[NR_OPEN_DEFAULT];
 };
 
 #define rcu_dereference_check_fdtable(files, fdtfd) \
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 76041b614758..aa3dc8d20436 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1380,7 +1380,7 @@ struct super_block {
 	 * Saved mount options for lazy filesystems using
 	 * generic_show_options()
 	 */
-	char *s_options;
+	char __rcu *s_options;
 };
 
 extern struct timespec current_fs_time(struct super_block *sb);
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 5f2f4c4d8fb0..af3f06b41dc1 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -129,8 +129,8 @@ struct blk_scsi_cmd_filter {
 struct disk_part_tbl {
 	struct rcu_head rcu_head;
 	int len;
-	struct hd_struct *last_lookup;
-	struct hd_struct *part[];
+	struct hd_struct __rcu *last_lookup;
+	struct hd_struct __rcu *part[];
 };
 
 struct gendisk {
@@ -149,7 +149,7 @@ struct gendisk {
 	 * non-critical accesses use RCU.  Always access through
 	 * helpers.
 	 */
-	struct disk_part_tbl *part_tbl;
+	struct disk_part_tbl __rcu *part_tbl;
 	struct hd_struct part0;
 
 	const struct block_device_operations *fops;
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 1f43fa56f600..6460fc65ed6b 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -137,8 +137,8 @@ extern struct cred init_cred;
 	.children	= LIST_HEAD_INIT(tsk.children),			\
 	.sibling	= LIST_HEAD_INIT(tsk.sibling),			\
 	.group_leader	= &tsk,						\
-	.real_cred	= &init_cred,					\
-	.cred		= &init_cred,					\
+	RCU_INIT_POINTER(.real_cred, &init_cred),			\
+	RCU_INIT_POINTER(.cred, &init_cred),				\
 	.cred_guard_mutex =						\
 		 __MUTEX_INITIALIZER(tsk.cred_guard_mutex),		\
 	.comm		= "swapper",					\
diff --git a/include/linux/iocontext.h b/include/linux/iocontext.h
index 64d529133031..3e70b21884a9 100644
--- a/include/linux/iocontext.h
+++ b/include/linux/iocontext.h
@@ -53,7 +53,7 @@ struct io_context {
 
 	struct radix_tree_root radix_root;
 	struct hlist_head cic_list;
-	void *ioc_data;
+	void __rcu *ioc_data;
 };
 
 static inline struct io_context *ioc_task_link(struct io_context *ioc)
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index b8bb9a6a1f37..05537a5eb855 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -299,7 +299,7 @@ struct mm_struct {
 	 * new_owner->mm == mm
 	 * new_owner->alloc_lock is held
 	 */
-	struct task_struct *owner;
+	struct task_struct __rcu *owner;
 #endif
 
 #ifdef CONFIG_PROC_FS
-- 
cgit v1.2.3


From 5e8067adfdbaf97039a97540765b1e16eb8d61cc Mon Sep 17 00:00:00 2001
From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Date: Sat, 17 Apr 2010 08:48:41 -0400
Subject: rcu head remove init

RCU heads really don't need to be initialized. Their state before call_rcu()
really does not matter.

We need to keep init/destroy_rcu_head_on_stack() though, since we want
debugobjects to be able to keep track of these objects.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
CC: David S. Miller <davem@davemloft.net>
CC: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
CC: akpm@linux-foundation.org
CC: mingo@elte.hu
CC: laijs@cn.fujitsu.com
CC: dipankar@in.ibm.com
CC: josh@joshtriplett.org
CC: dvhltc@us.ibm.com
CC: niv@us.ibm.com
CC: tglx@linutronix.de
CC: peterz@infradead.org
CC: rostedt@goodmis.org
CC: Valdis.Kletnieks@vt.edu
CC: dhowells@redhat.com
CC: eric.dumazet@gmail.com
CC: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
---
 include/linux/rcupdate.h | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 3e1b6625553b..27b44b3e3024 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -75,12 +75,6 @@ extern void rcu_init(void);
 #error "Unknown RCU implementation specified to kernel configuration"
 #endif
 
-#define RCU_HEAD_INIT	{ .next = NULL, .func = NULL }
-#define RCU_HEAD(head) struct rcu_head head = RCU_HEAD_INIT
-#define INIT_RCU_HEAD(ptr) do { \
-       (ptr)->next = NULL; (ptr)->func = NULL; \
-} while (0)
-
 /*
  * init_rcu_head_on_stack()/destroy_rcu_head_on_stack() are needed for dynamic
  * initialization and destruction of rcu_head on the stack. rcu_head structures
-- 
cgit v1.2.3


From eb4d40654505e47aa9d2035bb97f631fa61d14b4 Mon Sep 17 00:00:00 2001
From: Grégoire Baron <baronchon@n7mm.org>
Date: Wed, 18 Aug 2010 13:10:35 +0000
Subject: net/sched: add ACT_CSUM action to update packets checksums

net/sched: add ACT_CSUM action to update packets checksums

ACT_CSUM can be called just after ACT_PEDIT in order to re-compute some
altered checksums in IPv4 and IPv6 packets. The following checksums are
supported by this patch:
 - IPv4: IPv4 header, ICMP, IGMP, TCP, UDP & UDPLite
 - IPv6: ICMPv6, TCP, UDP & UDPLite
It's possible to request in the same action to update different kind of
checksums, if the packets flow mix TCP, UDP and UDPLite, ...

An example of usage is done in the associated iproute2 patch.

Version 3 changes:
 - remove useless goto instructions
 - improve IPv6 hop options decoding

Version 2 changes:
 - coding style correction
 - remove useless arguments of some functions
 - use stack in tcf_csum_dump()
 - add tcf_csum_skb_nextlayer() to factor code

Signed-off-by: Gregoire Baron <baronchon@n7mm.org>
Acked-by: jamal <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tc_act/Kbuild    |   1 +
 include/linux/tc_act/tc_csum.h |  32 +++
 include/net/tc_act/tc_csum.h   |  15 ++
 net/sched/Kconfig              |  10 +
 net/sched/Makefile             |   1 +
 net/sched/act_csum.c           | 595 +++++++++++++++++++++++++++++++++++++++++
 6 files changed, 654 insertions(+)
 create mode 100644 include/linux/tc_act/tc_csum.h
 create mode 100644 include/net/tc_act/tc_csum.h
 create mode 100644 net/sched/act_csum.c

(limited to 'include/linux')

diff --git a/include/linux/tc_act/Kbuild b/include/linux/tc_act/Kbuild
index 76990937f4c9..67b501c302b2 100644
--- a/include/linux/tc_act/Kbuild
+++ b/include/linux/tc_act/Kbuild
@@ -4,3 +4,4 @@ header-y += tc_mirred.h
 header-y += tc_pedit.h
 header-y += tc_nat.h
 header-y += tc_skbedit.h
+header-y += tc_csum.h
diff --git a/include/linux/tc_act/tc_csum.h b/include/linux/tc_act/tc_csum.h
new file mode 100644
index 000000000000..a047c49a3153
--- /dev/null
+++ b/include/linux/tc_act/tc_csum.h
@@ -0,0 +1,32 @@
+#ifndef __LINUX_TC_CSUM_H
+#define __LINUX_TC_CSUM_H
+
+#include <linux/types.h>
+#include <linux/pkt_cls.h>
+
+#define TCA_ACT_CSUM 16
+
+enum {
+	TCA_CSUM_UNSPEC,
+	TCA_CSUM_PARMS,
+	TCA_CSUM_TM,
+	__TCA_CSUM_MAX
+};
+#define TCA_CSUM_MAX (__TCA_CSUM_MAX - 1)
+
+enum {
+	TCA_CSUM_UPDATE_FLAG_IPV4HDR = 1,
+	TCA_CSUM_UPDATE_FLAG_ICMP    = 2,
+	TCA_CSUM_UPDATE_FLAG_IGMP    = 4,
+	TCA_CSUM_UPDATE_FLAG_TCP     = 8,
+	TCA_CSUM_UPDATE_FLAG_UDP     = 16,
+	TCA_CSUM_UPDATE_FLAG_UDPLITE = 32
+};
+
+struct tc_csum {
+	tc_gen;
+
+	__u32 update_flags;
+};
+
+#endif /* __LINUX_TC_CSUM_H */
diff --git a/include/net/tc_act/tc_csum.h b/include/net/tc_act/tc_csum.h
new file mode 100644
index 000000000000..9e8710be7a04
--- /dev/null
+++ b/include/net/tc_act/tc_csum.h
@@ -0,0 +1,15 @@
+#ifndef __NET_TC_CSUM_H
+#define __NET_TC_CSUM_H
+
+#include <linux/types.h>
+#include <net/act_api.h>
+
+struct tcf_csum {
+	struct tcf_common common;
+
+	u32 update_flags;
+};
+#define to_tcf_csum(pc) \
+	container_of(pc,struct tcf_csum,common)
+
+#endif /* __NET_TC_CSUM_H */
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index 2f691fb180d1..522d5a9a2825 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -518,6 +518,16 @@ config NET_ACT_SKBEDIT
 	  To compile this code as a module, choose M here: the
 	  module will be called act_skbedit.
 
+config NET_ACT_CSUM
+        tristate "Checksum Updating"
+        depends on NET_CLS_ACT
+        ---help---
+	  Say Y here to update some common checksum after some direct
+	  packet alterations.
+
+	  To compile this code as a module, choose M here: the
+	  module will be called act_csum.
+
 config NET_CLS_IND
 	bool "Incoming device classification"
 	depends on NET_CLS_U32 || NET_CLS_FW
diff --git a/net/sched/Makefile b/net/sched/Makefile
index f14e71bfa58f..960f5dba6304 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -15,6 +15,7 @@ obj-$(CONFIG_NET_ACT_NAT)	+= act_nat.o
 obj-$(CONFIG_NET_ACT_PEDIT)	+= act_pedit.o
 obj-$(CONFIG_NET_ACT_SIMP)	+= act_simple.o
 obj-$(CONFIG_NET_ACT_SKBEDIT)	+= act_skbedit.o
+obj-$(CONFIG_NET_ACT_CSUM)	+= act_csum.o
 obj-$(CONFIG_NET_SCH_FIFO)	+= sch_fifo.o
 obj-$(CONFIG_NET_SCH_CBQ)	+= sch_cbq.o
 obj-$(CONFIG_NET_SCH_HTB)	+= sch_htb.o
diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c
new file mode 100644
index 000000000000..58d7f36949da
--- /dev/null
+++ b/net/sched/act_csum.c
@@ -0,0 +1,595 @@
+/*
+ * Checksum updating actions
+ *
+ * Copyright (c) 2010 Gregoire Baron <baronchon@n7mm.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 the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+
+#include <linux/netlink.h>
+#include <net/netlink.h>
+#include <linux/rtnetlink.h>
+
+#include <linux/skbuff.h>
+
+#include <net/ip.h>
+#include <net/ipv6.h>
+#include <net/icmp.h>
+#include <linux/icmpv6.h>
+#include <linux/igmp.h>
+#include <net/tcp.h>
+#include <net/udp.h>
+
+#include <net/act_api.h>
+
+#include <linux/tc_act/tc_csum.h>
+#include <net/tc_act/tc_csum.h>
+
+#define CSUM_TAB_MASK 15
+static struct tcf_common *tcf_csum_ht[CSUM_TAB_MASK + 1];
+static u32 csum_idx_gen;
+static DEFINE_RWLOCK(csum_lock);
+
+static struct tcf_hashinfo csum_hash_info = {
+	.htab	=	tcf_csum_ht,
+	.hmask	=	CSUM_TAB_MASK,
+	.lock	=	&csum_lock,
+};
+
+static const struct nla_policy csum_policy[TCA_CSUM_MAX + 1] = {
+	[TCA_CSUM_PARMS] = { .len = sizeof(struct tc_csum), },
+};
+
+static int tcf_csum_init(struct nlattr *nla, struct nlattr *est,
+			 struct tc_action *a, int ovr, int bind)
+{
+	struct nlattr *tb[TCA_CSUM_MAX + 1];
+	struct tc_csum *parm;
+	struct tcf_common *pc;
+	struct tcf_csum *p;
+	int ret = 0, err;
+
+	if (nla == NULL)
+		return -EINVAL;
+
+	err = nla_parse_nested(tb, TCA_CSUM_MAX, nla,csum_policy);
+	if (err < 0)
+		return err;
+
+	if (tb[TCA_CSUM_PARMS] == NULL)
+		return -EINVAL;
+	parm = nla_data(tb[TCA_CSUM_PARMS]);
+
+	pc = tcf_hash_check(parm->index, a, bind, &csum_hash_info);
+	if (!pc) {
+		pc = tcf_hash_create(parm->index, est, a, sizeof(*p), bind, &csum_idx_gen, &csum_hash_info);
+		if (IS_ERR(pc))
+			return PTR_ERR(pc);
+		p = to_tcf_csum(pc);
+		ret = ACT_P_CREATED;
+	} else {
+		p = to_tcf_csum(pc);
+		if (!ovr) {
+			tcf_hash_release(pc, bind, &csum_hash_info);
+			return -EEXIST;
+		}
+	}
+
+	spin_lock_bh(&p->tcf_lock);
+	p->tcf_action = parm->action;
+	p->update_flags = parm->update_flags;
+	spin_unlock_bh(&p->tcf_lock);
+
+	if (ret == ACT_P_CREATED)
+		tcf_hash_insert(pc, &csum_hash_info);
+
+	return ret;
+}
+
+static int tcf_csum_cleanup(struct tc_action *a, int bind)
+{
+	struct tcf_csum *p = a->priv;
+	return tcf_hash_release(&p->common, bind, &csum_hash_info);
+}
+
+/**
+ * tcf_csum_skb_nextlayer - Get next layer pointer
+ * @skb: sk_buff to use
+ * @ihl: previous summed headers length
+ * @ipl: complete packet length
+ * @jhl: next header length
+ *
+ * Check the expected next layer availability in the specified sk_buff.
+ * Return the next layer pointer if pass, NULL otherwise.
+ */
+static void *tcf_csum_skb_nextlayer(struct sk_buff *skb,
+				    unsigned int ihl, unsigned int ipl,
+				    unsigned int jhl)
+{
+	int ntkoff = skb_network_offset(skb);
+	int hl = ihl + jhl;
+
+	if (!pskb_may_pull(skb, ipl + ntkoff) || (ipl < hl) ||
+	    (skb_cloned(skb) &&
+	     !skb_clone_writable(skb, hl + ntkoff) &&
+	     pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
+		return NULL;
+	else
+		return (void *)(skb_network_header(skb) + ihl);
+}
+
+static int tcf_csum_ipv4_icmp(struct sk_buff *skb,
+			      unsigned int ihl, unsigned int ipl)
+{
+	struct icmphdr *icmph;
+
+	icmph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*icmph));
+	if (icmph == NULL)
+		return 0;
+
+	icmph->checksum = 0;
+	skb->csum = csum_partial(icmph, ipl - ihl, 0);
+	icmph->checksum = csum_fold(skb->csum);
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+	return 1;
+}
+
+static int tcf_csum_ipv4_igmp(struct sk_buff *skb,
+			      unsigned int ihl, unsigned int ipl)
+{
+	struct igmphdr *igmph;
+
+	igmph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*igmph));
+	if (igmph == NULL)
+		return 0;
+
+	igmph->csum = 0;
+	skb->csum = csum_partial(igmph, ipl - ihl, 0);
+	igmph->csum = csum_fold(skb->csum);
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+	return 1;
+}
+
+static int tcf_csum_ipv6_icmp(struct sk_buff *skb, struct ipv6hdr *ip6h,
+			      unsigned int ihl, unsigned int ipl)
+{
+	struct icmp6hdr *icmp6h;
+
+	icmp6h = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*icmp6h));
+	if (icmp6h == NULL)
+		return 0;
+
+	icmp6h->icmp6_cksum = 0;
+	skb->csum = csum_partial(icmp6h, ipl - ihl, 0);
+	icmp6h->icmp6_cksum = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr,
+					      ipl - ihl, IPPROTO_ICMPV6,
+					      skb->csum);
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+	return 1;
+}
+
+static int tcf_csum_ipv4_tcp(struct sk_buff *skb, struct iphdr *iph,
+			     unsigned int ihl, unsigned int ipl)
+{
+	struct tcphdr *tcph;
+
+	tcph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*tcph));
+	if (tcph == NULL)
+		return 0;
+
+	tcph->check = 0;
+	skb->csum = csum_partial(tcph, ipl - ihl, 0);
+	tcph->check = tcp_v4_check(ipl - ihl,
+				   iph->saddr, iph->daddr, skb->csum);
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+	return 1;
+}
+
+static int tcf_csum_ipv6_tcp(struct sk_buff *skb, struct ipv6hdr *ip6h,
+			     unsigned int ihl, unsigned int ipl)
+{
+	struct tcphdr *tcph;
+
+	tcph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*tcph));
+	if (tcph == NULL)
+		return 0;
+
+	tcph->check = 0;
+	skb->csum = csum_partial(tcph, ipl - ihl, 0);
+	tcph->check = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr,
+				      ipl - ihl, IPPROTO_TCP,
+				      skb->csum);
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+	return 1;
+}
+
+static int tcf_csum_ipv4_udp(struct sk_buff *skb, struct iphdr *iph,
+			     unsigned int ihl, unsigned int ipl, int udplite)
+{
+	struct udphdr *udph;
+	u16 ul;
+
+	/* Support both UDP and UDPLITE checksum algorithms,
+	 * Don't use udph->len to get the real length without any protocol check,
+	 * UDPLITE uses udph->len for another thing,
+	 * Use iph->tot_len, or just ipl.
+	 */
+
+	udph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*udph));
+	if (udph == NULL)
+		return 0;
+
+	ul = ntohs(udph->len);
+
+	if (udplite || udph->check) {
+
+		udph->check = 0;
+
+		if (udplite) {
+			if (ul == 0)
+				skb->csum = csum_partial(udph, ipl - ihl, 0);
+
+			else if ((ul >= sizeof(*udph)) && (ul <= ipl - ihl))
+				skb->csum = csum_partial(udph, ul, 0);
+
+			else
+				goto ignore_obscure_skb;
+		} else {
+			if (ul != ipl - ihl)
+				goto ignore_obscure_skb;
+
+			skb->csum = csum_partial(udph, ul, 0);
+		}
+
+		udph->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
+						ul, iph->protocol,
+						skb->csum);
+
+		if (!udph->check)
+			udph->check = CSUM_MANGLED_0;
+	}
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+ignore_obscure_skb:
+	return 1;
+}
+
+static int tcf_csum_ipv6_udp(struct sk_buff *skb, struct ipv6hdr *ip6h,
+			     unsigned int ihl, unsigned int ipl, int udplite)
+{
+	struct udphdr *udph;
+	u16 ul;
+
+	/* Support both UDP and UDPLITE checksum algorithms,
+	 * Don't use udph->len to get the real length without any protocol check,
+	 * UDPLITE uses udph->len for another thing,
+	 * Use ip6h->payload_len + sizeof(*ip6h) ... , or just ipl.
+	 */
+
+	udph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*udph));
+	if (udph == NULL)
+		return 0;
+
+	ul = ntohs(udph->len);
+
+	udph->check = 0;
+
+	if (udplite) {
+		if (ul == 0)
+			skb->csum = csum_partial(udph, ipl - ihl, 0);
+
+		else if ((ul >= sizeof(*udph)) && (ul <= ipl - ihl))
+			skb->csum = csum_partial(udph, ul, 0);
+
+		else
+			goto ignore_obscure_skb;
+	} else {
+		if (ul != ipl - ihl)
+			goto ignore_obscure_skb;
+
+		skb->csum = csum_partial(udph, ul, 0);
+	}
+
+	udph->check = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, ul,
+				      udplite ? IPPROTO_UDPLITE : IPPROTO_UDP,
+				      skb->csum);
+
+	if (!udph->check)
+		udph->check = CSUM_MANGLED_0;
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+ignore_obscure_skb:
+	return 1;
+}
+
+static int tcf_csum_ipv4(struct sk_buff *skb, u32 update_flags)
+{
+	struct iphdr *iph;
+	int ntkoff;
+
+	ntkoff = skb_network_offset(skb);
+
+	if (!pskb_may_pull(skb, sizeof(*iph) + ntkoff))
+		goto fail;
+
+	iph = ip_hdr(skb);
+
+	switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) {
+	case IPPROTO_ICMP:
+		if (update_flags & TCA_CSUM_UPDATE_FLAG_ICMP)
+			if (!tcf_csum_ipv4_icmp(skb,
+						iph->ihl * 4, ntohs(iph->tot_len)))
+				goto fail;
+		break;
+	case IPPROTO_IGMP:
+		if (update_flags & TCA_CSUM_UPDATE_FLAG_IGMP)
+			if (!tcf_csum_ipv4_igmp(skb,
+						iph->ihl * 4, ntohs(iph->tot_len)))
+				goto fail;
+		break;
+	case IPPROTO_TCP:
+		if (update_flags & TCA_CSUM_UPDATE_FLAG_TCP)
+			if (!tcf_csum_ipv4_tcp(skb, iph,
+					       iph->ihl * 4, ntohs(iph->tot_len)))
+				goto fail;
+		break;
+	case IPPROTO_UDP:
+		if (update_flags & TCA_CSUM_UPDATE_FLAG_UDP)
+			if (!tcf_csum_ipv4_udp(skb, iph,
+					       iph->ihl * 4, ntohs(iph->tot_len), 0))
+				goto fail;
+		break;
+	case IPPROTO_UDPLITE:
+		if (update_flags & TCA_CSUM_UPDATE_FLAG_UDPLITE)
+			if (!tcf_csum_ipv4_udp(skb, iph,
+					       iph->ihl * 4, ntohs(iph->tot_len), 1))
+				goto fail;
+		break;
+	}
+
+	if (update_flags & TCA_CSUM_UPDATE_FLAG_IPV4HDR) {
+		if (skb_cloned(skb) &&
+		    !skb_clone_writable(skb, sizeof(*iph) + ntkoff) &&
+		    pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
+			goto fail;
+
+		ip_send_check(iph);
+	}
+
+	return 1;
+
+fail:
+	return 0;
+}
+
+static int tcf_csum_ipv6_hopopts(struct ipv6_opt_hdr *ip6xh,
+				  unsigned int ixhl, unsigned int *pl)
+{
+	int off, len, optlen;
+	unsigned char *xh = (void *)ip6xh;
+
+	off = sizeof(*ip6xh);
+	len = ixhl - off;
+
+	while (len > 1) {
+		switch (xh[off])
+		{
+		case IPV6_TLV_PAD0:
+			optlen = 1;
+			break;
+		case IPV6_TLV_JUMBO:
+			optlen = xh[off + 1] + 2;
+			if (optlen != 6 || len < 6 || (off & 3) != 2)
+				/* wrong jumbo option length/alignment */
+				return 0;
+			*pl = ntohl(*(__be32 *)(xh + off + 2));
+			goto done;
+		default:
+			optlen = xh[off + 1] + 2;
+			if (optlen > len)
+				/* ignore obscure options */
+				goto done;
+			break;
+		}
+		off += optlen;
+		len -= optlen;
+	}
+
+done:
+	return 1;
+}
+
+static int tcf_csum_ipv6(struct sk_buff *skb, u32 update_flags)
+{
+	struct ipv6hdr *ip6h;
+	struct ipv6_opt_hdr *ip6xh;
+	unsigned int hl, ixhl;
+	unsigned int pl;
+	int ntkoff;
+	u8 nexthdr;
+
+	ntkoff = skb_network_offset(skb);
+
+	hl = sizeof(*ip6h);
+
+	if (!pskb_may_pull(skb, hl + ntkoff))
+		goto fail;
+
+	ip6h = ipv6_hdr(skb);
+
+	pl = ntohs(ip6h->payload_len);
+	nexthdr = ip6h->nexthdr;
+
+	do {
+		switch (nexthdr) {
+		case NEXTHDR_FRAGMENT:
+			goto ignore_skb;
+		case NEXTHDR_ROUTING:
+		case NEXTHDR_HOP:
+		case NEXTHDR_DEST:
+			if (!pskb_may_pull(skb, hl + sizeof(*ip6xh) + ntkoff))
+				goto fail;
+			ip6xh = (void *)(skb_network_header(skb) + hl);
+			ixhl = ipv6_optlen(ip6xh);
+			if (!pskb_may_pull(skb, hl + ixhl + ntkoff))
+				goto fail;
+			if ((nexthdr == NEXTHDR_HOP) &&
+			    !(tcf_csum_ipv6_hopopts(ip6xh, ixhl, &pl)))
+				goto fail;
+			nexthdr = ip6xh->nexthdr;
+			hl += ixhl;
+			break;
+		case IPPROTO_ICMPV6:
+			if (update_flags & TCA_CSUM_UPDATE_FLAG_ICMP)
+				if (!tcf_csum_ipv6_icmp(skb, ip6h,
+							hl, pl + sizeof(*ip6h)))
+					goto fail;
+			goto done;
+		case IPPROTO_TCP:
+			if (update_flags & TCA_CSUM_UPDATE_FLAG_TCP)
+				if (!tcf_csum_ipv6_tcp(skb, ip6h,
+						       hl, pl + sizeof(*ip6h)))
+					goto fail;
+			goto done;
+		case IPPROTO_UDP:
+			if (update_flags & TCA_CSUM_UPDATE_FLAG_UDP)
+				if (!tcf_csum_ipv6_udp(skb, ip6h,
+						       hl, pl + sizeof(*ip6h), 0))
+					goto fail;
+			goto done;
+		case IPPROTO_UDPLITE:
+			if (update_flags & TCA_CSUM_UPDATE_FLAG_UDPLITE)
+				if (!tcf_csum_ipv6_udp(skb, ip6h,
+						       hl, pl + sizeof(*ip6h), 1))
+					goto fail;
+			goto done;
+		default:
+			goto ignore_skb;
+		}
+	} while (pskb_may_pull(skb, hl + 1 + ntkoff));
+
+done:
+ignore_skb:
+	return 1;
+
+fail:
+	return 0;
+}
+
+static int tcf_csum(struct sk_buff *skb,
+		    struct tc_action *a, struct tcf_result *res)
+{
+	struct tcf_csum *p = a->priv;
+	int action;
+	u32 update_flags;
+
+	spin_lock(&p->tcf_lock);
+	p->tcf_tm.lastuse = jiffies;
+	p->tcf_bstats.bytes += qdisc_pkt_len(skb);
+	p->tcf_bstats.packets++;
+	action = p->tcf_action;
+	update_flags = p->update_flags;
+	spin_unlock(&p->tcf_lock);
+
+	if (unlikely(action == TC_ACT_SHOT))
+		goto drop;
+
+	switch (skb->protocol) {
+	case cpu_to_be16(ETH_P_IP):
+		if (!tcf_csum_ipv4(skb, update_flags))
+			goto drop;
+		break;
+	case cpu_to_be16(ETH_P_IPV6):
+		if (!tcf_csum_ipv6(skb, update_flags))
+			goto drop;
+		break;
+	}
+
+	return action;
+
+drop:
+	spin_lock(&p->tcf_lock);
+	p->tcf_qstats.drops++;
+	spin_unlock(&p->tcf_lock);
+	return TC_ACT_SHOT;
+}
+
+static int tcf_csum_dump(struct sk_buff *skb,
+			 struct tc_action *a, int bind, int ref)
+{
+	unsigned char *b = skb_tail_pointer(skb);
+	struct tcf_csum *p = a->priv;
+	struct tc_csum opt = {
+		.update_flags = p->update_flags,
+
+		.index   = p->tcf_index,
+		.action  = p->tcf_action,
+		.refcnt  = p->tcf_refcnt - ref,
+		.bindcnt = p->tcf_bindcnt - bind,
+	};
+	struct tcf_t t;
+
+	NLA_PUT(skb, TCA_CSUM_PARMS, sizeof(opt), &opt);
+	t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install);
+	t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse);
+	t.expires = jiffies_to_clock_t(p->tcf_tm.expires);
+	NLA_PUT(skb, TCA_CSUM_TM, sizeof(t), &t);
+
+	return skb->len;
+
+nla_put_failure:
+	nlmsg_trim(skb, b);
+	return -1;
+}
+
+static struct tc_action_ops act_csum_ops = {
+	.kind		=	"csum",
+	.hinfo		=	&csum_hash_info,
+	.type		=	TCA_ACT_CSUM,
+	.capab		=	TCA_CAP_NONE,
+	.owner		=	THIS_MODULE,
+	.act		=	tcf_csum,
+	.dump		=	tcf_csum_dump,
+	.cleanup	=	tcf_csum_cleanup,
+	.lookup		=	tcf_hash_search,
+	.init		=	tcf_csum_init,
+	.walk		=	tcf_generic_walker
+};
+
+MODULE_DESCRIPTION("Checksum updating actions");
+MODULE_LICENSE("GPL");
+
+static int __init csum_init_module(void)
+{
+	return tcf_register_action(&act_csum_ops);
+}
+
+static void __exit csum_cleanup_module(void)
+{
+	tcf_unregister_action(&act_csum_ops);
+}
+
+module_init(csum_init_module);
+module_exit(csum_cleanup_module);
-- 
cgit v1.2.3


From a57eb940d130477a799dfb24a570ee04979c0f7f Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Tue, 29 Jun 2010 16:49:16 -0700
Subject: rcu: Add a TINY_PREEMPT_RCU
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Implement a small-memory-footprint uniprocessor-only implementation of
preemptible RCU.  This implementation uses but a single blocked-tasks
list rather than the combinatorial number used per leaf rcu_node by
TREE_PREEMPT_RCU, which reduces memory consumption and greatly simplifies
processing.  This version also takes advantage of uniprocessor execution
to accelerate grace periods in the case where there are no readers.

The general design is otherwise broadly similar to that of TREE_PREEMPT_RCU.

This implementation is a step towards having RCU implementation driven
off of the SMP and PREEMPT kernel configuration variables, which can
happen once this implementation has accumulated sufficient experience.

Removed ACCESS_ONCE() from __rcu_read_unlock() and added barrier() as
suggested by Steve Rostedt in order to avoid the compiler-reordering
issue noted by Mathieu Desnoyers (http://lkml.org/lkml/2010/8/16/183).

As can be seen below, CONFIG_TINY_PREEMPT_RCU represents almost 5Kbyte
savings compared to CONFIG_TREE_PREEMPT_RCU.  Of course, for non-real-time
workloads, CONFIG_TINY_RCU is even better.

	CONFIG_TREE_PREEMPT_RCU

	   text	   data	    bss	    dec	   filename
	     13	      0	      0	     13	   kernel/rcupdate.o
	   6170	    825	     28	   7023	   kernel/rcutree.o
				   ----
				   7026    Total

	CONFIG_TINY_PREEMPT_RCU

	   text	   data	    bss	    dec	   filename
	     13	      0	      0	     13	   kernel/rcupdate.o
	   2081	     81	      8	   2170	   kernel/rcutiny.o
				   ----
				   2183    Total

	CONFIG_TINY_RCU (non-preemptible)

	   text	   data	    bss	    dec	   filename
	     13	      0	      0	     13	   kernel/rcupdate.o
	    719	     25	      0	    744	   kernel/rcutiny.o
				    ---
				    757    Total

Requested-by: Loïc Minier <loic.minier@canonical.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/hardirq.h   |   2 +-
 include/linux/init_task.h |  10 +-
 include/linux/rcupdate.h  |   3 +-
 include/linux/rcutiny.h   | 126 +++++++---
 include/linux/rcutree.h   |   2 +
 include/linux/sched.h     |  10 +-
 init/Kconfig              |  16 +-
 kernel/Makefile           |   1 +
 kernel/rcutiny.c          |  33 ++-
 kernel/rcutiny_plugin.h   | 582 +++++++++++++++++++++++++++++++++++++++++++++-
 10 files changed, 717 insertions(+), 68 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index d5b387669dab..1f4517d55b19 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -139,7 +139,7 @@ static inline void account_system_vtime(struct task_struct *tsk)
 #endif
 
 #if defined(CONFIG_NO_HZ)
-#if defined(CONFIG_TINY_RCU)
+#if defined(CONFIG_TINY_RCU) || defined(CONFIG_TINY_PREEMPT_RCU)
 extern void rcu_enter_nohz(void);
 extern void rcu_exit_nohz(void);
 
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 6460fc65ed6b..2fea6c8ef6ba 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -82,11 +82,17 @@ extern struct group_info init_groups;
 # define CAP_INIT_BSET  CAP_FULL_SET
 
 #ifdef CONFIG_TREE_PREEMPT_RCU
+#define INIT_TASK_RCU_TREE_PREEMPT()					\
+	.rcu_blocked_node = NULL,
+#else
+#define INIT_TASK_RCU_TREE_PREEMPT(tsk)
+#endif
+#ifdef CONFIG_PREEMPT_RCU
 #define INIT_TASK_RCU_PREEMPT(tsk)					\
 	.rcu_read_lock_nesting = 0,					\
 	.rcu_read_unlock_special = 0,					\
-	.rcu_blocked_node = NULL,					\
-	.rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry),
+	.rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry),		\
+	INIT_TASK_RCU_TREE_PREEMPT()
 #else
 #define INIT_TASK_RCU_PREEMPT(tsk)
 #endif
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 27b44b3e3024..24b896649384 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -58,7 +58,6 @@ struct rcu_head {
 };
 
 /* Exported common interfaces */
-extern void rcu_barrier(void);
 extern void rcu_barrier_bh(void);
 extern void rcu_barrier_sched(void);
 extern void synchronize_sched_expedited(void);
@@ -69,7 +68,7 @@ extern void rcu_init(void);
 
 #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU)
 #include <linux/rcutree.h>
-#elif defined(CONFIG_TINY_RCU)
+#elif defined(CONFIG_TINY_RCU) || defined(CONFIG_TINY_PREEMPT_RCU)
 #include <linux/rcutiny.h>
 #else
 #error "Unknown RCU implementation specified to kernel configuration"
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index e2e893144a84..4cc5eba41616 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -29,66 +29,51 @@
 
 void rcu_sched_qs(int cpu);
 void rcu_bh_qs(int cpu);
-static inline void rcu_note_context_switch(int cpu)
-{
-	rcu_sched_qs(cpu);
-}
 
+#ifdef CONFIG_TINY_RCU
 #define __rcu_read_lock()	preempt_disable()
 #define __rcu_read_unlock()	preempt_enable()
+#else /* #ifdef CONFIG_TINY_RCU */
+void __rcu_read_lock(void);
+void __rcu_read_unlock(void);
+#endif /* #else #ifdef CONFIG_TINY_RCU */
 #define __rcu_read_lock_bh()	local_bh_disable()
 #define __rcu_read_unlock_bh()	local_bh_enable()
-#define call_rcu_sched		call_rcu
+extern void call_rcu_sched(struct rcu_head *head,
+			   void (*func)(struct rcu_head *rcu));
 
 #define rcu_init_sched()	do { } while (0)
-extern void rcu_check_callbacks(int cpu, int user);
 
-static inline int rcu_needs_cpu(int cpu)
-{
-	return 0;
-}
+extern void synchronize_sched(void);
 
-/*
- * Return the number of grace periods.
- */
-static inline long rcu_batches_completed(void)
-{
-	return 0;
-}
+#ifdef CONFIG_TINY_RCU
 
-/*
- * Return the number of bottom-half grace periods.
- */
-static inline long rcu_batches_completed_bh(void)
-{
-	return 0;
-}
+#define call_rcu		call_rcu_sched
 
-static inline void rcu_force_quiescent_state(void)
+static inline void synchronize_rcu(void)
 {
+	synchronize_sched();
 }
 
-static inline void rcu_bh_force_quiescent_state(void)
+static inline void synchronize_rcu_expedited(void)
 {
+	synchronize_sched();	/* Only one CPU, so pretty fast anyway!!! */
 }
 
-static inline void rcu_sched_force_quiescent_state(void)
+static inline void rcu_barrier(void)
 {
+	rcu_barrier_sched();  /* Only one CPU, so only one list of callbacks! */
 }
 
-extern void synchronize_sched(void);
+#else /* #ifdef CONFIG_TINY_RCU */
 
-static inline void synchronize_rcu(void)
-{
-	synchronize_sched();
-}
+void synchronize_rcu(void);
+void rcu_barrier(void);
+void synchronize_rcu_expedited(void);
 
-static inline void synchronize_rcu_bh(void)
-{
-	synchronize_sched();
-}
+#endif /* #else #ifdef CONFIG_TINY_RCU */
 
-static inline void synchronize_rcu_expedited(void)
+static inline void synchronize_rcu_bh(void)
 {
 	synchronize_sched();
 }
@@ -117,15 +102,82 @@ static inline void rcu_exit_nohz(void)
 
 #endif /* #else #ifdef CONFIG_NO_HZ */
 
+#ifdef CONFIG_TINY_RCU
+
+static inline void rcu_preempt_note_context_switch(void)
+{
+}
+
 static inline void exit_rcu(void)
 {
 }
 
+static inline int rcu_needs_cpu(int cpu)
+{
+	return 0;
+}
+
 static inline int rcu_preempt_depth(void)
 {
 	return 0;
 }
 
+#else /* #ifdef CONFIG_TINY_RCU */
+
+void rcu_preempt_note_context_switch(void);
+extern void exit_rcu(void);
+int rcu_preempt_needs_cpu(void);
+
+static inline int rcu_needs_cpu(int cpu)
+{
+	return rcu_preempt_needs_cpu();
+}
+
+/*
+ * Defined as macro as it is a very low level header
+ * included from areas that don't even know about current
+ * FIXME: combine with include/linux/rcutree.h into rcupdate.h.
+ */
+#define rcu_preempt_depth() (current->rcu_read_lock_nesting)
+
+#endif /* #else #ifdef CONFIG_TINY_RCU */
+
+static inline void rcu_note_context_switch(int cpu)
+{
+	rcu_sched_qs(cpu);
+	rcu_preempt_note_context_switch();
+}
+
+extern void rcu_check_callbacks(int cpu, int user);
+
+/*
+ * Return the number of grace periods.
+ */
+static inline long rcu_batches_completed(void)
+{
+	return 0;
+}
+
+/*
+ * Return the number of bottom-half grace periods.
+ */
+static inline long rcu_batches_completed_bh(void)
+{
+	return 0;
+}
+
+static inline void rcu_force_quiescent_state(void)
+{
+}
+
+static inline void rcu_bh_force_quiescent_state(void)
+{
+}
+
+static inline void rcu_sched_force_quiescent_state(void)
+{
+}
+
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 
 extern int rcu_scheduler_active __read_mostly;
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index c0ed1c056f29..c13b85dd22bc 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -95,6 +95,8 @@ static inline void synchronize_rcu_bh_expedited(void)
 	synchronize_sched_expedited();
 }
 
+extern void rcu_barrier(void);
+
 extern void rcu_check_callbacks(int cpu, int user);
 
 extern long rcu_batches_completed(void);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2c756666c111..e18473f0eb78 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1202,11 +1202,13 @@ struct task_struct {
 	unsigned int policy;
 	cpumask_t cpus_allowed;
 
-#ifdef CONFIG_TREE_PREEMPT_RCU
+#ifdef CONFIG_PREEMPT_RCU
 	int rcu_read_lock_nesting;
 	char rcu_read_unlock_special;
-	struct rcu_node *rcu_blocked_node;
 	struct list_head rcu_node_entry;
+#endif /* #ifdef CONFIG_PREEMPT_RCU */
+#ifdef CONFIG_TREE_PREEMPT_RCU
+	struct rcu_node *rcu_blocked_node;
 #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */
 
 #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT)
@@ -1740,7 +1742,7 @@ extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *
 #define tsk_used_math(p) ((p)->flags & PF_USED_MATH)
 #define used_math() tsk_used_math(current)
 
-#ifdef CONFIG_TREE_PREEMPT_RCU
+#ifdef CONFIG_PREEMPT_RCU
 
 #define RCU_READ_UNLOCK_BLOCKED (1 << 0) /* blocked while in RCU read-side. */
 #define RCU_READ_UNLOCK_NEED_QS (1 << 1) /* RCU core needs CPU response. */
@@ -1749,7 +1751,9 @@ static inline void rcu_copy_process(struct task_struct *p)
 {
 	p->rcu_read_lock_nesting = 0;
 	p->rcu_read_unlock_special = 0;
+#ifdef CONFIG_TREE_PREEMPT_RCU
 	p->rcu_blocked_node = NULL;
+#endif
 	INIT_LIST_HEAD(&p->rcu_node_entry);
 }
 
diff --git a/init/Kconfig b/init/Kconfig
index dbc08baad77e..a619a1ac7f4c 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -348,7 +348,7 @@ config TREE_RCU
 	  smaller systems.
 
 config TREE_PREEMPT_RCU
-	bool "Preemptable tree-based hierarchical RCU"
+	bool "Preemptible tree-based hierarchical RCU"
 	depends on PREEMPT
 	help
 	  This option selects the RCU implementation that is
@@ -366,8 +366,22 @@ config TINY_RCU
 	  is not required.  This option greatly reduces the
 	  memory footprint of RCU.
 
+config TINY_PREEMPT_RCU
+	bool "Preemptible UP-only small-memory-footprint RCU"
+	depends on !SMP && PREEMPT
+	help
+	  This option selects the RCU implementation that is designed
+	  for real-time UP systems.  This option greatly reduces the
+	  memory footprint of RCU.
+
 endchoice
 
+config PREEMPT_RCU
+	def_bool ( TREE_PREEMPT_RCU || TINY_PREEMPT_RCU )
+	help
+	  This option enables preemptible-RCU code that is common between
+	  the TREE_PREEMPT_RCU and TINY_PREEMPT_RCU implementations.
+
 config RCU_TRACE
 	bool "Enable tracing for RCU"
 	depends on TREE_RCU || TREE_PREEMPT_RCU
diff --git a/kernel/Makefile b/kernel/Makefile
index 0b72d1a74be0..17046b6e7c90 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -86,6 +86,7 @@ obj-$(CONFIG_TREE_RCU) += rcutree.o
 obj-$(CONFIG_TREE_PREEMPT_RCU) += rcutree.o
 obj-$(CONFIG_TREE_RCU_TRACE) += rcutree_trace.o
 obj-$(CONFIG_TINY_RCU) += rcutiny.o
+obj-$(CONFIG_TINY_PREEMPT_RCU) += rcutiny.o
 obj-$(CONFIG_RELAY) += relay.o
 obj-$(CONFIG_SYSCTL) += utsname_sysctl.o
 obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c
index 196ec02f8be0..d806735342ac 100644
--- a/kernel/rcutiny.c
+++ b/kernel/rcutiny.c
@@ -59,6 +59,14 @@ int rcu_scheduler_active __read_mostly;
 EXPORT_SYMBOL_GPL(rcu_scheduler_active);
 #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
 
+/* Forward declarations for rcutiny_plugin.h. */
+static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp);
+static void __call_rcu(struct rcu_head *head,
+		       void (*func)(struct rcu_head *rcu),
+		       struct rcu_ctrlblk *rcp);
+
+#include "rcutiny_plugin.h"
+
 #ifdef CONFIG_NO_HZ
 
 static long rcu_dynticks_nesting = 1;
@@ -140,6 +148,7 @@ void rcu_check_callbacks(int cpu, int user)
 		rcu_sched_qs(cpu);
 	else if (!in_softirq())
 		rcu_bh_qs(cpu);
+	rcu_preempt_check_callbacks();
 }
 
 /*
@@ -162,6 +171,7 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
 	*rcp->donetail = NULL;
 	if (rcp->curtail == rcp->donetail)
 		rcp->curtail = &rcp->rcucblist;
+	rcu_preempt_remove_callbacks(rcp);
 	rcp->donetail = &rcp->rcucblist;
 	local_irq_restore(flags);
 
@@ -182,6 +192,7 @@ static void rcu_process_callbacks(struct softirq_action *unused)
 {
 	__rcu_process_callbacks(&rcu_sched_ctrlblk);
 	__rcu_process_callbacks(&rcu_bh_ctrlblk);
+	rcu_preempt_process_callbacks();
 }
 
 /*
@@ -223,15 +234,15 @@ static void __call_rcu(struct rcu_head *head,
 }
 
 /*
- * Post an RCU callback to be invoked after the end of an RCU grace
+ * Post an RCU callback to be invoked after the end of an RCU-sched grace
  * period.  But since we have but one CPU, that would be after any
  * quiescent state.
  */
-void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
+void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
 {
 	__call_rcu(head, func, &rcu_sched_ctrlblk);
 }
-EXPORT_SYMBOL_GPL(call_rcu);
+EXPORT_SYMBOL_GPL(call_rcu_sched);
 
 /*
  * Post an RCU bottom-half callback to be invoked after any subsequent
@@ -243,20 +254,6 @@ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
 }
 EXPORT_SYMBOL_GPL(call_rcu_bh);
 
-void rcu_barrier(void)
-{
-	struct rcu_synchronize rcu;
-
-	init_rcu_head_on_stack(&rcu.head);
-	init_completion(&rcu.completion);
-	/* Will wake me after RCU finished. */
-	call_rcu(&rcu.head, wakeme_after_rcu);
-	/* Wait for it. */
-	wait_for_completion(&rcu.completion);
-	destroy_rcu_head_on_stack(&rcu.head);
-}
-EXPORT_SYMBOL_GPL(rcu_barrier);
-
 void rcu_barrier_bh(void)
 {
 	struct rcu_synchronize rcu;
@@ -289,5 +286,3 @@ void __init rcu_init(void)
 {
 	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
 }
-
-#include "rcutiny_plugin.h"
diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h
index d223a92bc742..e6bc1b447c6c 100644
--- a/kernel/rcutiny_plugin.h
+++ b/kernel/rcutiny_plugin.h
@@ -1,7 +1,7 @@
 /*
- * Read-Copy Update mechanism for mutual exclusion (tree-based version)
+ * Read-Copy Update mechanism for mutual exclusion, the Bloatwatch edition
  * Internal non-public definitions that provide either classic
- * or preemptable semantics.
+ * or preemptible semantics.
  *
  * 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
@@ -17,11 +17,587 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  *
- * Copyright IBM Corporation, 2009
+ * Copyright (c) 2010 Linaro
  *
  * Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
  */
 
+#ifdef CONFIG_TINY_PREEMPT_RCU
+
+#include <linux/delay.h>
+
+/* FIXME: merge with definitions in kernel/rcutree.h. */
+#define ULONG_CMP_GE(a, b)	(ULONG_MAX / 2 >= (a) - (b))
+#define ULONG_CMP_LT(a, b)	(ULONG_MAX / 2 < (a) - (b))
+
+/* Global control variables for preemptible RCU. */
+struct rcu_preempt_ctrlblk {
+	struct rcu_ctrlblk rcb;	/* curtail: ->next ptr of last CB for GP. */
+	struct rcu_head **nexttail;
+				/* Tasks blocked in a preemptible RCU */
+				/*  read-side critical section while an */
+				/*  preemptible-RCU grace period is in */
+				/*  progress must wait for a later grace */
+				/*  period.  This pointer points to the */
+				/*  ->next pointer of the last task that */
+				/*  must wait for a later grace period, or */
+				/*  to &->rcb.rcucblist if there is no */
+				/*  such task. */
+	struct list_head blkd_tasks;
+				/* Tasks blocked in RCU read-side critical */
+				/*  section.  Tasks are placed at the head */
+				/*  of this list and age towards the tail. */
+	struct list_head *gp_tasks;
+				/* Pointer to the first task blocking the */
+				/*  current grace period, or NULL if there */
+				/*  is not such task. */
+	struct list_head *exp_tasks;
+				/* Pointer to first task blocking the */
+				/*  current expedited grace period, or NULL */
+				/*  if there is no such task.  If there */
+				/*  is no current expedited grace period, */
+				/*  then there cannot be any such task. */
+	u8 gpnum;		/* Current grace period. */
+	u8 gpcpu;		/* Last grace period blocked by the CPU. */
+	u8 completed;		/* Last grace period completed. */
+				/*  If all three are equal, RCU is idle. */
+};
+
+static struct rcu_preempt_ctrlblk rcu_preempt_ctrlblk = {
+	.rcb.donetail = &rcu_preempt_ctrlblk.rcb.rcucblist,
+	.rcb.curtail = &rcu_preempt_ctrlblk.rcb.rcucblist,
+	.nexttail = &rcu_preempt_ctrlblk.rcb.rcucblist,
+	.blkd_tasks = LIST_HEAD_INIT(rcu_preempt_ctrlblk.blkd_tasks),
+};
+
+static int rcu_preempted_readers_exp(void);
+static void rcu_report_exp_done(void);
+
+/*
+ * Return true if the CPU has not yet responded to the current grace period.
+ */
+static int rcu_cpu_cur_gp(void)
+{
+	return rcu_preempt_ctrlblk.gpcpu != rcu_preempt_ctrlblk.gpnum;
+}
+
+/*
+ * Check for a running RCU reader.  Because there is only one CPU,
+ * there can be but one running RCU reader at a time.  ;-)
+ */
+static int rcu_preempt_running_reader(void)
+{
+	return current->rcu_read_lock_nesting;
+}
+
+/*
+ * Check for preempted RCU readers blocking any grace period.
+ * If the caller needs a reliable answer, it must disable hard irqs.
+ */
+static int rcu_preempt_blocked_readers_any(void)
+{
+	return !list_empty(&rcu_preempt_ctrlblk.blkd_tasks);
+}
+
+/*
+ * Check for preempted RCU readers blocking the current grace period.
+ * If the caller needs a reliable answer, it must disable hard irqs.
+ */
+static int rcu_preempt_blocked_readers_cgp(void)
+{
+	return rcu_preempt_ctrlblk.gp_tasks != NULL;
+}
+
+/*
+ * Return true if another preemptible-RCU grace period is needed.
+ */
+static int rcu_preempt_needs_another_gp(void)
+{
+	return *rcu_preempt_ctrlblk.rcb.curtail != NULL;
+}
+
+/*
+ * Return true if a preemptible-RCU grace period is in progress.
+ * The caller must disable hardirqs.
+ */
+static int rcu_preempt_gp_in_progress(void)
+{
+	return rcu_preempt_ctrlblk.completed != rcu_preempt_ctrlblk.gpnum;
+}
+
+/*
+ * Record a preemptible-RCU quiescent state for the specified CPU.  Note
+ * that this just means that the task currently running on the CPU is
+ * in a quiescent state.  There might be any number of tasks blocked
+ * while in an RCU read-side critical section.
+ *
+ * Unlike the other rcu_*_qs() functions, callers to this function
+ * must disable irqs in order to protect the assignment to
+ * ->rcu_read_unlock_special.
+ *
+ * Because this is a single-CPU implementation, the only way a grace
+ * period can end is if the CPU is in a quiescent state.  The reason is
+ * that a blocked preemptible-RCU reader can exit its critical section
+ * only if the CPU is running it at the time.  Therefore, when the
+ * last task blocking the current grace period exits its RCU read-side
+ * critical section, neither the CPU nor blocked tasks will be stopping
+ * the current grace period.  (In contrast, SMP implementations
+ * might have CPUs running in RCU read-side critical sections that
+ * block later grace periods -- but this is not possible given only
+ * one CPU.)
+ */
+static void rcu_preempt_cpu_qs(void)
+{
+	/* Record both CPU and task as having responded to current GP. */
+	rcu_preempt_ctrlblk.gpcpu = rcu_preempt_ctrlblk.gpnum;
+	current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS;
+
+	/*
+	 * If there is no GP, or if blocked readers are still blocking GP,
+	 * then there is nothing more to do.
+	 */
+	if (!rcu_preempt_gp_in_progress() || rcu_preempt_blocked_readers_cgp())
+		return;
+
+	/* Advance callbacks. */
+	rcu_preempt_ctrlblk.completed = rcu_preempt_ctrlblk.gpnum;
+	rcu_preempt_ctrlblk.rcb.donetail = rcu_preempt_ctrlblk.rcb.curtail;
+	rcu_preempt_ctrlblk.rcb.curtail = rcu_preempt_ctrlblk.nexttail;
+
+	/* If there are no blocked readers, next GP is done instantly. */
+	if (!rcu_preempt_blocked_readers_any())
+		rcu_preempt_ctrlblk.rcb.donetail = rcu_preempt_ctrlblk.nexttail;
+
+	/* If there are done callbacks, make RCU_SOFTIRQ process them. */
+	if (*rcu_preempt_ctrlblk.rcb.donetail != NULL)
+		raise_softirq(RCU_SOFTIRQ);
+}
+
+/*
+ * Start a new RCU grace period if warranted.  Hard irqs must be disabled.
+ */
+static void rcu_preempt_start_gp(void)
+{
+	if (!rcu_preempt_gp_in_progress() && rcu_preempt_needs_another_gp()) {
+
+		/* Official start of GP. */
+		rcu_preempt_ctrlblk.gpnum++;
+
+		/* Any blocked RCU readers block new GP. */
+		if (rcu_preempt_blocked_readers_any())
+			rcu_preempt_ctrlblk.gp_tasks =
+				rcu_preempt_ctrlblk.blkd_tasks.next;
+
+		/* If there is no running reader, CPU is done with GP. */
+		if (!rcu_preempt_running_reader())
+			rcu_preempt_cpu_qs();
+	}
+}
+
+/*
+ * We have entered the scheduler, and the current task might soon be
+ * context-switched away from.  If this task is in an RCU read-side
+ * critical section, we will no longer be able to rely on the CPU to
+ * record that fact, so we enqueue the task on the blkd_tasks list.
+ * If the task started after the current grace period began, as recorded
+ * by ->gpcpu, we enqueue at the beginning of the list.  Otherwise
+ * before the element referenced by ->gp_tasks (or at the tail if
+ * ->gp_tasks is NULL) and point ->gp_tasks at the newly added element.
+ * The task will dequeue itself when it exits the outermost enclosing
+ * RCU read-side critical section.  Therefore, the current grace period
+ * cannot be permitted to complete until the ->gp_tasks pointer becomes
+ * NULL.
+ *
+ * Caller must disable preemption.
+ */
+void rcu_preempt_note_context_switch(void)
+{
+	struct task_struct *t = current;
+	unsigned long flags;
+
+	local_irq_save(flags); /* must exclude scheduler_tick(). */
+	if (rcu_preempt_running_reader() &&
+	    (t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) {
+
+		/* Possibly blocking in an RCU read-side critical section. */
+		t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED;
+
+		/*
+		 * If this CPU has already checked in, then this task
+		 * will hold up the next grace period rather than the
+		 * current grace period.  Queue the task accordingly.
+		 * If the task is queued for the current grace period
+		 * (i.e., this CPU has not yet passed through a quiescent
+		 * state for the current grace period), then as long
+		 * as that task remains queued, the current grace period
+		 * cannot end.
+		 */
+		list_add(&t->rcu_node_entry, &rcu_preempt_ctrlblk.blkd_tasks);
+		if (rcu_cpu_cur_gp())
+			rcu_preempt_ctrlblk.gp_tasks = &t->rcu_node_entry;
+	}
+
+	/*
+	 * Either we were not in an RCU read-side critical section to
+	 * begin with, or we have now recorded that critical section
+	 * globally.  Either way, we can now note a quiescent state
+	 * for this CPU.  Again, if we were in an RCU read-side critical
+	 * section, and if that critical section was blocking the current
+	 * grace period, then the fact that the task has been enqueued
+	 * means that current grace period continues to be blocked.
+	 */
+	rcu_preempt_cpu_qs();
+	local_irq_restore(flags);
+}
+
+/*
+ * Tiny-preemptible RCU implementation for rcu_read_lock().
+ * Just increment ->rcu_read_lock_nesting, shared state will be updated
+ * if we block.
+ */
+void __rcu_read_lock(void)
+{
+	current->rcu_read_lock_nesting++;
+	barrier();  /* needed if we ever invoke rcu_read_lock in rcutiny.c */
+}
+EXPORT_SYMBOL_GPL(__rcu_read_lock);
+
+/*
+ * Handle special cases during rcu_read_unlock(), such as needing to
+ * notify RCU core processing or task having blocked during the RCU
+ * read-side critical section.
+ */
+static void rcu_read_unlock_special(struct task_struct *t)
+{
+	int empty;
+	int empty_exp;
+	unsigned long flags;
+	struct list_head *np;
+	int special;
+
+	/*
+	 * NMI handlers cannot block and cannot safely manipulate state.
+	 * They therefore cannot possibly be special, so just leave.
+	 */
+	if (in_nmi())
+		return;
+
+	local_irq_save(flags);
+
+	/*
+	 * If RCU core is waiting for this CPU to exit critical section,
+	 * let it know that we have done so.
+	 */
+	special = t->rcu_read_unlock_special;
+	if (special & RCU_READ_UNLOCK_NEED_QS)
+		rcu_preempt_cpu_qs();
+
+	/* Hardware IRQ handlers cannot block. */
+	if (in_irq()) {
+		local_irq_restore(flags);
+		return;
+	}
+
+	/* Clean up if blocked during RCU read-side critical section. */
+	if (special & RCU_READ_UNLOCK_BLOCKED) {
+		t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BLOCKED;
+
+		/*
+		 * Remove this task from the ->blkd_tasks list and adjust
+		 * any pointers that might have been referencing it.
+		 */
+		empty = !rcu_preempt_blocked_readers_cgp();
+		empty_exp = rcu_preempt_ctrlblk.exp_tasks == NULL;
+		np = t->rcu_node_entry.next;
+		if (np == &rcu_preempt_ctrlblk.blkd_tasks)
+			np = NULL;
+		list_del(&t->rcu_node_entry);
+		if (&t->rcu_node_entry == rcu_preempt_ctrlblk.gp_tasks)
+			rcu_preempt_ctrlblk.gp_tasks = np;
+		if (&t->rcu_node_entry == rcu_preempt_ctrlblk.exp_tasks)
+			rcu_preempt_ctrlblk.exp_tasks = np;
+		INIT_LIST_HEAD(&t->rcu_node_entry);
+
+		/*
+		 * If this was the last task on the current list, and if
+		 * we aren't waiting on the CPU, report the quiescent state
+		 * and start a new grace period if needed.
+		 */
+		if (!empty && !rcu_preempt_blocked_readers_cgp()) {
+			rcu_preempt_cpu_qs();
+			rcu_preempt_start_gp();
+		}
+
+		/*
+		 * If this was the last task on the expedited lists,
+		 * then we need wake up the waiting task.
+		 */
+		if (!empty_exp && rcu_preempt_ctrlblk.exp_tasks == NULL)
+			rcu_report_exp_done();
+	}
+	local_irq_restore(flags);
+}
+
+/*
+ * Tiny-preemptible RCU implementation for rcu_read_unlock().
+ * Decrement ->rcu_read_lock_nesting.  If the result is zero (outermost
+ * rcu_read_unlock()) and ->rcu_read_unlock_special is non-zero, then
+ * invoke rcu_read_unlock_special() to clean up after a context switch
+ * in an RCU read-side critical section and other special cases.
+ */
+void __rcu_read_unlock(void)
+{
+	struct task_struct *t = current;
+
+	barrier();  /* needed if we ever invoke rcu_read_unlock in rcutiny.c */
+	--t->rcu_read_lock_nesting;
+	barrier();  /* decrement before load of ->rcu_read_unlock_special */
+	if (t->rcu_read_lock_nesting == 0 &&
+	    unlikely(ACCESS_ONCE(t->rcu_read_unlock_special)))
+		rcu_read_unlock_special(t);
+#ifdef CONFIG_PROVE_LOCKING
+	WARN_ON_ONCE(t->rcu_read_lock_nesting < 0);
+#endif /* #ifdef CONFIG_PROVE_LOCKING */
+}
+EXPORT_SYMBOL_GPL(__rcu_read_unlock);
+
+/*
+ * Check for a quiescent state from the current CPU.  When a task blocks,
+ * the task is recorded in the rcu_preempt_ctrlblk structure, which is
+ * checked elsewhere.  This is called from the scheduling-clock interrupt.
+ *
+ * Caller must disable hard irqs.
+ */
+static void rcu_preempt_check_callbacks(void)
+{
+	struct task_struct *t = current;
+
+	if (!rcu_preempt_running_reader() && rcu_preempt_gp_in_progress())
+		rcu_preempt_cpu_qs();
+	if (&rcu_preempt_ctrlblk.rcb.rcucblist !=
+	    rcu_preempt_ctrlblk.rcb.donetail)
+		raise_softirq(RCU_SOFTIRQ);
+	if (rcu_preempt_gp_in_progress() && rcu_preempt_running_reader())
+		t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS;
+}
+
+/*
+ * TINY_PREEMPT_RCU has an extra callback-list tail pointer to
+ * update, so this is invoked from __rcu_process_callbacks() to
+ * handle that case.  Of course, it is invoked for all flavors of
+ * RCU, but RCU callbacks can appear only on one of the lists, and
+ * neither ->nexttail nor ->donetail can possibly be NULL, so there
+ * is no need for an explicit check.
+ */
+static void rcu_preempt_remove_callbacks(struct rcu_ctrlblk *rcp)
+{
+	if (rcu_preempt_ctrlblk.nexttail == rcp->donetail)
+		rcu_preempt_ctrlblk.nexttail = &rcp->rcucblist;
+}
+
+/*
+ * Process callbacks for preemptible RCU.
+ */
+static void rcu_preempt_process_callbacks(void)
+{
+	__rcu_process_callbacks(&rcu_preempt_ctrlblk.rcb);
+}
+
+/*
+ * Queue a preemptible -RCU callback for invocation after a grace period.
+ */
+void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+
+	debug_rcu_head_queue(head);
+	head->func = func;
+	head->next = NULL;
+
+	local_irq_save(flags);
+	*rcu_preempt_ctrlblk.nexttail = head;
+	rcu_preempt_ctrlblk.nexttail = &head->next;
+	rcu_preempt_start_gp();  /* checks to see if GP needed. */
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(call_rcu);
+
+void rcu_barrier(void)
+{
+	struct rcu_synchronize rcu;
+
+	init_rcu_head_on_stack(&rcu.head);
+	init_completion(&rcu.completion);
+	/* Will wake me after RCU finished. */
+	call_rcu(&rcu.head, wakeme_after_rcu);
+	/* Wait for it. */
+	wait_for_completion(&rcu.completion);
+	destroy_rcu_head_on_stack(&rcu.head);
+}
+EXPORT_SYMBOL_GPL(rcu_barrier);
+
+/*
+ * synchronize_rcu - wait until a grace period has elapsed.
+ *
+ * Control will return to the caller some time after a full grace
+ * period has elapsed, in other words after all currently executing RCU
+ * read-side critical sections have completed.  RCU read-side critical
+ * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
+ * and may be nested.
+ */
+void synchronize_rcu(void)
+{
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	if (!rcu_scheduler_active)
+		return;
+#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+
+	WARN_ON_ONCE(rcu_preempt_running_reader());
+	if (!rcu_preempt_blocked_readers_any())
+		return;
+
+	/* Once we get past the fastpath checks, same code as rcu_barrier(). */
+	rcu_barrier();
+}
+EXPORT_SYMBOL_GPL(synchronize_rcu);
+
+static DECLARE_WAIT_QUEUE_HEAD(sync_rcu_preempt_exp_wq);
+static unsigned long sync_rcu_preempt_exp_count;
+static DEFINE_MUTEX(sync_rcu_preempt_exp_mutex);
+
+/*
+ * Return non-zero if there are any tasks in RCU read-side critical
+ * sections blocking the current preemptible-RCU expedited grace period.
+ * If there is no preemptible-RCU expedited grace period currently in
+ * progress, returns zero unconditionally.
+ */
+static int rcu_preempted_readers_exp(void)
+{
+	return rcu_preempt_ctrlblk.exp_tasks != NULL;
+}
+
+/*
+ * Report the exit from RCU read-side critical section for the last task
+ * that queued itself during or before the current expedited preemptible-RCU
+ * grace period.
+ */
+static void rcu_report_exp_done(void)
+{
+	wake_up(&sync_rcu_preempt_exp_wq);
+}
+
+/*
+ * Wait for an rcu-preempt grace period, but expedite it.  The basic idea
+ * is to rely in the fact that there is but one CPU, and that it is
+ * illegal for a task to invoke synchronize_rcu_expedited() while in a
+ * preemptible-RCU read-side critical section.  Therefore, any such
+ * critical sections must correspond to blocked tasks, which must therefore
+ * be on the ->blkd_tasks list.  So just record the current head of the
+ * list in the ->exp_tasks pointer, and wait for all tasks including and
+ * after the task pointed to by ->exp_tasks to drain.
+ */
+void synchronize_rcu_expedited(void)
+{
+	unsigned long flags;
+	struct rcu_preempt_ctrlblk *rpcp = &rcu_preempt_ctrlblk;
+	unsigned long snap;
+
+	barrier(); /* ensure prior action seen before grace period. */
+
+	WARN_ON_ONCE(rcu_preempt_running_reader());
+
+	/*
+	 * Acquire lock so that there is only one preemptible RCU grace
+	 * period in flight.  Of course, if someone does the expedited
+	 * grace period for us while we are acquiring the lock, just leave.
+	 */
+	snap = sync_rcu_preempt_exp_count + 1;
+	mutex_lock(&sync_rcu_preempt_exp_mutex);
+	if (ULONG_CMP_LT(snap, sync_rcu_preempt_exp_count))
+		goto unlock_mb_ret; /* Others did our work for us. */
+
+	local_irq_save(flags);
+
+	/*
+	 * All RCU readers have to already be on blkd_tasks because
+	 * we cannot legally be executing in an RCU read-side critical
+	 * section.
+	 */
+
+	/* Snapshot current head of ->blkd_tasks list. */
+	rpcp->exp_tasks = rpcp->blkd_tasks.next;
+	if (rpcp->exp_tasks == &rpcp->blkd_tasks)
+		rpcp->exp_tasks = NULL;
+	local_irq_restore(flags);
+
+	/* Wait for tail of ->blkd_tasks list to drain. */
+	if (rcu_preempted_readers_exp())
+		wait_event(sync_rcu_preempt_exp_wq,
+			   !rcu_preempted_readers_exp());
+
+	/* Clean up and exit. */
+	barrier(); /* ensure expedited GP seen before counter increment. */
+	sync_rcu_preempt_exp_count++;
+unlock_mb_ret:
+	mutex_unlock(&sync_rcu_preempt_exp_mutex);
+	barrier(); /* ensure subsequent action seen after grace period. */
+}
+EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
+
+/*
+ * Does preemptible RCU need the CPU to stay out of dynticks mode?
+ */
+int rcu_preempt_needs_cpu(void)
+{
+	if (!rcu_preempt_running_reader())
+		rcu_preempt_cpu_qs();
+	return rcu_preempt_ctrlblk.rcb.rcucblist != NULL;
+}
+
+/*
+ * Check for a task exiting while in a preemptible -RCU read-side
+ * critical section, clean up if so.  No need to issue warnings,
+ * as debug_check_no_locks_held() already does this if lockdep
+ * is enabled.
+ */
+void exit_rcu(void)
+{
+	struct task_struct *t = current;
+
+	if (t->rcu_read_lock_nesting == 0)
+		return;
+	t->rcu_read_lock_nesting = 1;
+	rcu_read_unlock();
+}
+
+#else /* #ifdef CONFIG_TINY_PREEMPT_RCU */
+
+/*
+ * Because preemptible RCU does not exist, it never has any callbacks
+ * to check.
+ */
+static void rcu_preempt_check_callbacks(void)
+{
+}
+
+/*
+ * Because preemptible RCU does not exist, it never has any callbacks
+ * to remove.
+ */
+static void rcu_preempt_remove_callbacks(struct rcu_ctrlblk *rcp)
+{
+}
+
+/*
+ * Because preemptible RCU does not exist, it never has any callbacks
+ * to process.
+ */
+static void rcu_preempt_process_callbacks(void)
+{
+}
+
+#endif /* #else #ifdef CONFIG_TINY_PREEMPT_RCU */
+
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 
 #include <linux/kernel_stat.h>
-- 
cgit v1.2.3


From 9079fd7c2e06a92cf27d05224a1f478581916c5b Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Sat, 7 Aug 2010 21:59:54 -0700
Subject: rcu: update obsolete rcu_read_lock() comment.

The comment says that blocking is illegal in rcu_read_lock()-style
RCU read-side critical sections, which is no longer entirely true
given preemptible RCU.  This commit provides a fix.

Suggested-by: David Miller <davem@davemloft.net>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/rcupdate.h | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 24b896649384..d7af96ef6fcf 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -458,7 +458,20 @@ extern int rcu_my_thread_group_empty(void);
  * will be deferred until the outermost RCU read-side critical section
  * completes.
  *
- * It is illegal to block while in an RCU read-side critical section.
+ * You can avoid reading and understanding the next paragraph by
+ * following this rule: don't put anything in an rcu_read_lock() RCU
+ * read-side critical section that would block in a !PREEMPT kernel.
+ * But if you want the full story, read on!
+ *
+ * In non-preemptible RCU implementations (TREE_RCU and TINY_RCU), it
+ * is illegal to block while in an RCU read-side critical section.  In
+ * preemptible RCU implementations (TREE_PREEMPT_RCU and TINY_PREEMPT_RCU)
+ * in CONFIG_PREEMPT kernel builds, RCU read-side critical sections may
+ * be preempted, but explicit blocking is illegal.  Finally, in preemptible
+ * RCU implementations in real-time (CONFIG_PREEMPT_RT) kernel builds,
+ * RCU read-side critical sections may be preempted and they may also
+ * block, but only when acquiring spinlocks that are subject to priority
+ * inheritance.
  */
 static inline void rcu_read_lock(void)
 {
-- 
cgit v1.2.3


From 53d84e004d5e8c018be395c4330dc72fd60bd13e Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Tue, 10 Aug 2010 14:28:53 -0700
Subject: rcu: permit suppressing current grace period's CPU stall warnings

When using a kernel debugger, a long sojourn in the debugger can get
you lots of RCU CPU stall warnings once you resume.  This might not be
helpful, especially if you are using the system console.  This patch
therefore allows RCU CPU stall warnings to be suppressed, but only for
the duration of the current set of grace periods.

This differs from Jason's original patch in that it adds support for
tiny RCU and preemptible RCU, and uses a slightly different method for
suppressing the RCU CPU stall warning messages.

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Jason Wessel <jason.wessel@windriver.com>
---
 include/linux/rcutiny.h |  4 ++++
 include/linux/rcutree.h |  1 +
 kernel/rcutree.c        | 20 ++++++++++++++++++++
 kernel/rcutree.h        |  1 +
 kernel/rcutree_plugin.h | 18 ++++++++++++++++++
 5 files changed, 44 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index 4cc5eba41616..3fa179784e18 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -178,6 +178,10 @@ static inline void rcu_sched_force_quiescent_state(void)
 {
 }
 
+static inline void rcu_cpu_stall_reset(void)
+{
+}
+
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 
 extern int rcu_scheduler_active __read_mostly;
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index c13b85dd22bc..0726809497ba 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -36,6 +36,7 @@ extern void rcu_sched_qs(int cpu);
 extern void rcu_bh_qs(int cpu);
 extern void rcu_note_context_switch(int cpu);
 extern int rcu_needs_cpu(int cpu);
+extern void rcu_cpu_stall_reset(void);
 
 #ifdef CONFIG_TREE_PREEMPT_RCU
 
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index ff214118e4b8..42140a860bb9 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -565,6 +565,22 @@ static int rcu_panic(struct notifier_block *this, unsigned long ev, void *ptr)
 	return NOTIFY_DONE;
 }
 
+/**
+ * rcu_cpu_stall_reset - prevent further stall warnings in current grace period
+ *
+ * Set the stall-warning timeout way off into the future, thus preventing
+ * any RCU CPU stall-warning messages from appearing in the current set of
+ * RCU grace periods.
+ *
+ * The caller must disable hard irqs.
+ */
+void rcu_cpu_stall_reset(void)
+{
+	rcu_sched_state.jiffies_stall = jiffies + ULONG_MAX / 2;
+	rcu_bh_state.jiffies_stall = jiffies + ULONG_MAX / 2;
+	rcu_preempt_stall_reset();
+}
+
 static struct notifier_block rcu_panic_block = {
 	.notifier_call = rcu_panic,
 };
@@ -584,6 +600,10 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp)
 {
 }
 
+void rcu_cpu_stall_reset(void)
+{
+}
+
 static void __init check_cpu_stall_init(void)
 {
 }
diff --git a/kernel/rcutree.h b/kernel/rcutree.h
index bb4d08695c45..7abd439a7573 100644
--- a/kernel/rcutree.h
+++ b/kernel/rcutree.h
@@ -372,6 +372,7 @@ static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp,
 #ifdef CONFIG_RCU_CPU_STALL_DETECTOR
 static void rcu_print_detail_task_stall(struct rcu_state *rsp);
 static void rcu_print_task_stall(struct rcu_node *rnp);
+static void rcu_preempt_stall_reset(void);
 #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
 static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp);
 #ifdef CONFIG_HOTPLUG_CPU
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 63bb7714fdeb..561410f70d4a 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -417,6 +417,16 @@ static void rcu_print_task_stall(struct rcu_node *rnp)
 	}
 }
 
+/*
+ * Suppress preemptible RCU's CPU stall warnings by pushing the
+ * time of the next stall-warning message comfortably far into the
+ * future.
+ */
+static void rcu_preempt_stall_reset(void)
+{
+	rcu_preempt_state.jiffies_stall = jiffies + ULONG_MAX / 2;
+}
+
 #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
 
 /*
@@ -867,6 +877,14 @@ static void rcu_print_task_stall(struct rcu_node *rnp)
 {
 }
 
+/*
+ * Because preemptible RCU does not exist, there is no need to suppress
+ * its CPU stall warnings.
+ */
+static void rcu_preempt_stall_reset(void)
+{
+}
+
 #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
 
 /*
-- 
cgit v1.2.3


From a3dc3fb161f9b4066c0fce22db72638af8baf83b Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Fri, 13 Aug 2010 16:16:25 -0700
Subject: rcu: repair code-duplication FIXMEs

Combine the duplicate definitions of ULONG_CMP_GE(), ULONG_CMP_LT(),
and rcu_preempt_depth() into include/linux/rcupdate.h.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/rcupdate.h | 15 +++++++++++++++
 include/linux/rcutiny.h  |  7 -------
 include/linux/rcutree.h  |  6 ------
 kernel/rcutiny_plugin.h  |  4 ----
 kernel/rcutree.h         |  3 ---
 5 files changed, 15 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index d7af96ef6fcf..325bad7bbca9 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -47,6 +47,9 @@
 extern int rcutorture_runnable; /* for sysctl */
 #endif /* #ifdef CONFIG_RCU_TORTURE_TEST */
 
+#define ULONG_CMP_GE(a, b)	(ULONG_MAX / 2 >= (a) - (b))
+#define ULONG_CMP_LT(a, b)	(ULONG_MAX / 2 < (a) - (b))
+
 /**
  * struct rcu_head - callback structure for use with RCU
  * @next: next update requests in a list
@@ -66,6 +69,18 @@ extern int sched_expedited_torture_stats(char *page);
 /* Internal to kernel */
 extern void rcu_init(void);
 
+#ifdef CONFIG_PREEMPT_RCU
+
+/*
+ * Defined as a macro as it is a very low level header included from
+ * areas that don't even know about current.  This gives the rcu_read_lock()
+ * nesting depth, but makes sense only if CONFIG_PREEMPT_RCU -- in other
+ * types of kernel builds, the rcu_read_lock() nesting depth is unknowable.
+ */
+#define rcu_preempt_depth() (current->rcu_read_lock_nesting)
+
+#endif /* #ifdef CONFIG_PREEMPT_RCU */
+
 #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU)
 #include <linux/rcutree.h>
 #elif defined(CONFIG_TINY_RCU) || defined(CONFIG_TINY_PREEMPT_RCU)
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index 3fa179784e18..c6b11dc5ba0a 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -133,13 +133,6 @@ static inline int rcu_needs_cpu(int cpu)
 	return rcu_preempt_needs_cpu();
 }
 
-/*
- * Defined as macro as it is a very low level header
- * included from areas that don't even know about current
- * FIXME: combine with include/linux/rcutree.h into rcupdate.h.
- */
-#define rcu_preempt_depth() (current->rcu_read_lock_nesting)
-
 #endif /* #else #ifdef CONFIG_TINY_RCU */
 
 static inline void rcu_note_context_switch(int cpu)
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index 0726809497ba..54a20c11f98d 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -45,12 +45,6 @@ extern void __rcu_read_unlock(void);
 extern void synchronize_rcu(void);
 extern void exit_rcu(void);
 
-/*
- * Defined as macro as it is a very low level header
- * included from areas that don't even know about current
- */
-#define rcu_preempt_depth() (current->rcu_read_lock_nesting)
-
 #else /* #ifdef CONFIG_TREE_PREEMPT_RCU */
 
 static inline void __rcu_read_lock(void)
diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h
index e6bc1b447c6c..c5bea1137dcb 100644
--- a/kernel/rcutiny_plugin.h
+++ b/kernel/rcutiny_plugin.h
@@ -26,10 +26,6 @@
 
 #include <linux/delay.h>
 
-/* FIXME: merge with definitions in kernel/rcutree.h. */
-#define ULONG_CMP_GE(a, b)	(ULONG_MAX / 2 >= (a) - (b))
-#define ULONG_CMP_LT(a, b)	(ULONG_MAX / 2 < (a) - (b))
-
 /* Global control variables for preemptible RCU. */
 struct rcu_preempt_ctrlblk {
 	struct rcu_ctrlblk rcb;	/* curtail: ->next ptr of last CB for GP. */
diff --git a/kernel/rcutree.h b/kernel/rcutree.h
index 7abd439a7573..7918ba61873f 100644
--- a/kernel/rcutree.h
+++ b/kernel/rcutree.h
@@ -272,9 +272,6 @@ struct rcu_data {
 
 #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
 
-#define ULONG_CMP_GE(a, b)	(ULONG_MAX / 2 >= (a) - (b))
-#define ULONG_CMP_LT(a, b)	(ULONG_MAX / 2 < (a) - (b))
-
 /*
  * RCU global state, including node hierarchy.  This hierarchy is
  * represented in "heap" form in a dense array.  The root (first level)
-- 
cgit v1.2.3


From 73d4da4d360136826b36f78f5cf72b29da82c8a6 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Mon, 16 Aug 2010 10:50:54 -0700
Subject: rcu: Upgrade srcu_read_lock() docbook about SRCU grace periods

It is illegal to wait for an SRCU grace period while within the
corresponding flavor of SRCU read-side critical section.  Therefore,
this commit updates the srcu_read_lock() docbook accordingly.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/srcu.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/srcu.h b/include/linux/srcu.h
index 6f456a720ff0..58971e891f48 100644
--- a/include/linux/srcu.h
+++ b/include/linux/srcu.h
@@ -139,7 +139,12 @@ static inline int srcu_read_lock_held(struct srcu_struct *sp)
  * @sp: srcu_struct in which to register the new reader.
  *
  * Enter an SRCU read-side critical section.  Note that SRCU read-side
- * critical sections may be nested.
+ * critical sections may be nested.  However, it is illegal to
+ * call anything that waits on an SRCU grace period for the same
+ * srcu_struct, whether directly or indirectly.  Please note that
+ * one way to indirectly wait on an SRCU grace period is to acquire
+ * a mutex that is held elsewhere while calling synchronize_srcu() or
+ * synchronize_srcu_expedited().
  */
 static inline int srcu_read_lock(struct srcu_struct *sp) __acquires(sp)
 {
-- 
cgit v1.2.3


From 7b0b759b65247cbc66384a912be9acf8d4800636 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Tue, 17 Aug 2010 14:18:46 -0700
Subject: rcu: combine duplicate code, courtesy of CONFIG_PREEMPT_RCU

The CONFIG_PREEMPT_RCU kernel configuration parameter was recently
re-introduced, but as an indication of the type of RCU (preemptible
vs. non-preemptible) instead of as selecting a given implementation.
This commit uses CONFIG_PREEMPT_RCU to combine duplicate code
from include/linux/rcutiny.h and include/linux/rcutree.h into
include/linux/rcupdate.h.  This commit also combines a few other pieces
of duplicate code that have accumulated.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/rcupdate.h | 75 ++++++++++++++++++++++++++++++++++++++++++++++--
 include/linux/rcutiny.h  | 51 --------------------------------
 include/linux/rcutree.h  | 50 --------------------------------
 kernel/rcutree_plugin.h  |  9 ------
 4 files changed, 72 insertions(+), 113 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 325bad7bbca9..89414d67d961 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -61,16 +61,30 @@ struct rcu_head {
 };
 
 /* Exported common interfaces */
+extern void call_rcu_sched(struct rcu_head *head,
+			   void (*func)(struct rcu_head *rcu));
+extern void synchronize_sched(void);
 extern void rcu_barrier_bh(void);
 extern void rcu_barrier_sched(void);
 extern void synchronize_sched_expedited(void);
 extern int sched_expedited_torture_stats(char *page);
 
-/* Internal to kernel */
-extern void rcu_init(void);
+static inline void __rcu_read_lock_bh(void)
+{
+	local_bh_disable();
+}
+
+static inline void __rcu_read_unlock_bh(void)
+{
+	local_bh_enable();
+}
 
 #ifdef CONFIG_PREEMPT_RCU
 
+extern void __rcu_read_lock(void);
+extern void __rcu_read_unlock(void);
+void synchronize_rcu(void);
+
 /*
  * Defined as a macro as it is a very low level header included from
  * areas that don't even know about current.  This gives the rcu_read_lock()
@@ -79,7 +93,53 @@ extern void rcu_init(void);
  */
 #define rcu_preempt_depth() (current->rcu_read_lock_nesting)
 
-#endif /* #ifdef CONFIG_PREEMPT_RCU */
+#else /* #ifdef CONFIG_PREEMPT_RCU */
+
+static inline void __rcu_read_lock(void)
+{
+	preempt_disable();
+}
+
+static inline void __rcu_read_unlock(void)
+{
+	preempt_enable();
+}
+
+static inline void synchronize_rcu(void)
+{
+	synchronize_sched();
+}
+
+static inline int rcu_preempt_depth(void)
+{
+	return 0;
+}
+
+#endif /* #else #ifdef CONFIG_PREEMPT_RCU */
+
+/* Internal to kernel */
+extern void rcu_init(void);
+extern void rcu_sched_qs(int cpu);
+extern void rcu_bh_qs(int cpu);
+extern void rcu_check_callbacks(int cpu, int user);
+struct notifier_block;
+
+#ifdef CONFIG_NO_HZ
+
+extern void rcu_enter_nohz(void);
+extern void rcu_exit_nohz(void);
+
+#else /* #ifdef CONFIG_NO_HZ */
+
+static inline void rcu_enter_nohz(void)
+{
+}
+
+static inline void rcu_exit_nohz(void)
+{
+}
+
+#endif /* #else #ifdef CONFIG_NO_HZ */
 
 #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU)
 #include <linux/rcutree.h>
@@ -626,6 +686,8 @@ struct rcu_synchronize {
 
 extern void wakeme_after_rcu(struct rcu_head  *head);
 
+#ifdef CONFIG_PREEMPT_RCU
+
 /**
  * call_rcu() - Queue an RCU callback for invocation after a grace period.
  * @head: structure to be used for queueing the RCU updates.
@@ -642,6 +704,13 @@ extern void wakeme_after_rcu(struct rcu_head  *head);
 extern void call_rcu(struct rcu_head *head,
 			      void (*func)(struct rcu_head *head));
 
+#else /* #ifdef CONFIG_PREEMPT_RCU */
+
+/* In classic RCU, call_rcu() is just call_rcu_sched(). */
+#define	call_rcu	call_rcu_sched
+
+#endif /* #else #ifdef CONFIG_PREEMPT_RCU */
+
 /**
  * call_rcu_bh() - Queue an RCU for invocation after a quicker grace period.
  * @head: structure to be used for queueing the RCU updates.
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index c6b11dc5ba0a..13877cb93a60 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -27,34 +27,10 @@
 
 #include <linux/cache.h>
 
-void rcu_sched_qs(int cpu);
-void rcu_bh_qs(int cpu);
-
-#ifdef CONFIG_TINY_RCU
-#define __rcu_read_lock()	preempt_disable()
-#define __rcu_read_unlock()	preempt_enable()
-#else /* #ifdef CONFIG_TINY_RCU */
-void __rcu_read_lock(void);
-void __rcu_read_unlock(void);
-#endif /* #else #ifdef CONFIG_TINY_RCU */
-#define __rcu_read_lock_bh()	local_bh_disable()
-#define __rcu_read_unlock_bh()	local_bh_enable()
-extern void call_rcu_sched(struct rcu_head *head,
-			   void (*func)(struct rcu_head *rcu));
-
 #define rcu_init_sched()	do { } while (0)
 
-extern void synchronize_sched(void);
-
 #ifdef CONFIG_TINY_RCU
 
-#define call_rcu		call_rcu_sched
-
-static inline void synchronize_rcu(void)
-{
-	synchronize_sched();
-}
-
 static inline void synchronize_rcu_expedited(void)
 {
 	synchronize_sched();	/* Only one CPU, so pretty fast anyway!!! */
@@ -67,7 +43,6 @@ static inline void rcu_barrier(void)
 
 #else /* #ifdef CONFIG_TINY_RCU */
 
-void synchronize_rcu(void);
 void rcu_barrier(void);
 void synchronize_rcu_expedited(void);
 
@@ -83,25 +58,6 @@ static inline void synchronize_rcu_bh_expedited(void)
 	synchronize_sched();
 }
 
-struct notifier_block;
-
-#ifdef CONFIG_NO_HZ
-
-extern void rcu_enter_nohz(void);
-extern void rcu_exit_nohz(void);
-
-#else /* #ifdef CONFIG_NO_HZ */
-
-static inline void rcu_enter_nohz(void)
-{
-}
-
-static inline void rcu_exit_nohz(void)
-{
-}
-
-#endif /* #else #ifdef CONFIG_NO_HZ */
-
 #ifdef CONFIG_TINY_RCU
 
 static inline void rcu_preempt_note_context_switch(void)
@@ -117,11 +73,6 @@ static inline int rcu_needs_cpu(int cpu)
 	return 0;
 }
 
-static inline int rcu_preempt_depth(void)
-{
-	return 0;
-}
-
 #else /* #ifdef CONFIG_TINY_RCU */
 
 void rcu_preempt_note_context_switch(void);
@@ -141,8 +92,6 @@ static inline void rcu_note_context_switch(int cpu)
 	rcu_preempt_note_context_switch();
 }
 
-extern void rcu_check_callbacks(int cpu, int user);
-
 /*
  * Return the number of grace periods.
  */
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index 54a20c11f98d..95518e628794 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -30,59 +30,23 @@
 #ifndef __LINUX_RCUTREE_H
 #define __LINUX_RCUTREE_H
 
-struct notifier_block;
-
-extern void rcu_sched_qs(int cpu);
-extern void rcu_bh_qs(int cpu);
 extern void rcu_note_context_switch(int cpu);
 extern int rcu_needs_cpu(int cpu);
 extern void rcu_cpu_stall_reset(void);
 
 #ifdef CONFIG_TREE_PREEMPT_RCU
 
-extern void __rcu_read_lock(void);
-extern void __rcu_read_unlock(void);
-extern void synchronize_rcu(void);
 extern void exit_rcu(void);
 
 #else /* #ifdef CONFIG_TREE_PREEMPT_RCU */
 
-static inline void __rcu_read_lock(void)
-{
-	preempt_disable();
-}
-
-static inline void __rcu_read_unlock(void)
-{
-	preempt_enable();
-}
-
-#define synchronize_rcu synchronize_sched
-
 static inline void exit_rcu(void)
 {
 }
 
-static inline int rcu_preempt_depth(void)
-{
-	return 0;
-}
-
 #endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */
 
-static inline void __rcu_read_lock_bh(void)
-{
-	local_bh_disable();
-}
-static inline void __rcu_read_unlock_bh(void)
-{
-	local_bh_enable();
-}
-
-extern void call_rcu_sched(struct rcu_head *head,
-			   void (*func)(struct rcu_head *rcu));
 extern void synchronize_rcu_bh(void);
-extern void synchronize_sched(void);
 extern void synchronize_rcu_expedited(void);
 
 static inline void synchronize_rcu_bh_expedited(void)
@@ -92,8 +56,6 @@ static inline void synchronize_rcu_bh_expedited(void)
 
 extern void rcu_barrier(void);
 
-extern void rcu_check_callbacks(int cpu, int user);
-
 extern long rcu_batches_completed(void);
 extern long rcu_batches_completed_bh(void);
 extern long rcu_batches_completed_sched(void);
@@ -101,18 +63,6 @@ extern void rcu_force_quiescent_state(void);
 extern void rcu_bh_force_quiescent_state(void);
 extern void rcu_sched_force_quiescent_state(void);
 
-#ifdef CONFIG_NO_HZ
-void rcu_enter_nohz(void);
-void rcu_exit_nohz(void);
-#else /* CONFIG_NO_HZ */
-static inline void rcu_enter_nohz(void)
-{
-}
-static inline void rcu_exit_nohz(void)
-{
-}
-#endif /* CONFIG_NO_HZ */
-
 /* A context switch is a grace period for RCU-sched and RCU-bh. */
 static inline int rcu_blocking_is_gp(void)
 {
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 561410f70d4a..87f60f06b18e 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -938,15 +938,6 @@ static void rcu_preempt_process_callbacks(void)
 {
 }
 
-/*
- * In classic RCU, call_rcu() is just call_rcu_sched().
- */
-void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
-{
-	call_rcu_sched(head, func);
-}
-EXPORT_SYMBOL_GPL(call_rcu);
-
 /*
  * Wait for an rcu-preempt grace period, but make it happen quickly.
  * But because preemptable RCU does not exist, map to rcu-sched.
-- 
cgit v1.2.3


From 65e6bf484c497f02d47a0faae69ee398cd59cfda Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Thu, 19 Aug 2010 21:43:09 -0700
Subject: rcu: add comment stating that list_empty() applies to RCU-protected
 lists

Because list_empty() does not dereference any RCU-protected pointers, and
further does not pass such pointers to the caller (so that the caller
does not dereference them either), it is safe to use list_empty() on
RCU-protected lists.  There is no need for a list_empty_rcu().  This
commit adds a comment stating this explicitly.

Requested-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/rculist.h | 9 +++++++++
 1 file changed, 9 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/rculist.h b/include/linux/rculist.h
index c10b1050dbe6..f31ef61f1c65 100644
--- a/include/linux/rculist.h
+++ b/include/linux/rculist.h
@@ -9,6 +9,15 @@
 #include <linux/list.h>
 #include <linux/rcupdate.h>
 
+/*
+ * Why is there no list_empty_rcu()?  Because list_empty() serves this
+ * purpose.  The list_empty() function fetches the RCU-protected pointer
+ * and compares it to the address of the list head, but neither dereferences
+ * this pointer itself nor provides this pointer to the caller.  Therefore,
+ * it is not necessary to use rcu_dereference(), so that list_empty() can
+ * be used anywhere you would want to use a list_empty_rcu().
+ */
+
 /*
  * return the ->next pointer of a list_head in an rcu safe
  * way, we must not access it directly
-- 
cgit v1.2.3


From 00959ade36acadc00e757f87060bf6e4501d545f Mon Sep 17 00:00:00 2001
From: Dmitry Kozlov <xeb@mail.ru>
Date: Sat, 21 Aug 2010 23:05:39 -0700
Subject: PPTP: PPP over IPv4 (Point-to-Point Tunneling Protocol)

PPP: introduce "pptp" module which implements point-to-point tunneling protocol using pppox framework
NET: introduce the "gre" module for demultiplexing GRE packets on version criteria
     (required to pptp and ip_gre may coexists)
NET: ip_gre: update to use the "gre" module

This patch introduces then pptp support to the linux kernel which
dramatically speeds up pptp vpn connections and decreases cpu usage in
comparison of existing user-space implementation
(poptop/pptpclient). There is accel-pptp project
(https://sourceforge.net/projects/accel-pptp/) to utilize this module,
it contains plugin for pppd to use pptp in client-mode and modified
pptpd (poptop) to build high-performance pptp NAS.

There was many changes from initial submitted patch, most important are:
1. using rcu instead of read-write locks
2. using static bitmap instead of dynamically allocated
3. using vmalloc for memory allocation instead of BITS_PER_LONG + __get_free_pages
4. fixed many coding style issues
Thanks to Eric Dumazet.

Signed-off-by: Dmitry Kozlov <xeb@mail.ru>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS              |  14 +
 drivers/net/Kconfig      |  11 +
 drivers/net/Makefile     |   1 +
 drivers/net/pptp.c       | 726 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/if_pppox.h |  52 ++--
 include/net/gre.h        |  18 ++
 net/ipv4/Kconfig         |   7 +
 net/ipv4/Makefile        |   1 +
 net/ipv4/gre.c           | 151 ++++++++++
 net/ipv4/ip_gre.c        |  14 +-
 10 files changed, 971 insertions(+), 24 deletions(-)
 create mode 100644 drivers/net/pptp.c
 create mode 100644 include/net/gre.h
 create mode 100644 net/ipv4/gre.c

(limited to 'include/linux')

diff --git a/MAINTAINERS b/MAINTAINERS
index b5b8baa1d70e..43c9efcd8e10 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6528,6 +6528,20 @@ M:	"Maciej W. Rozycki" <macro@linux-mips.org>
 S:	Maintained
 F:	drivers/serial/zs.*
 
+GRE DEMULTIPLEXER DRIVER
+M:	Dmitry Kozlov <xeb@mail.ru>
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	net/ipv4/gre.c
+F:	include/net/gre.h
+
+PPTP DRIVER
+M:	Dmitry Kozlov <xeb@mail.ru>
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	drivers/net/pptp.c
+W:	http://sourceforge.net/projects/accel-pptp
+
 THE REST
 M:	Linus Torvalds <torvalds@linux-foundation.org>
 L:	linux-kernel@vger.kernel.org
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 5a6895320b48..9b2a72089a68 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -3192,6 +3192,17 @@ config PPPOE
 	  which contains instruction on how to use this driver (under 
 	  the heading "Kernel mode PPPoE").
 
+config PPTP
+	tristate "PPP over IPv4 (PPTP) (EXPERIMENTAL)"
+	depends on EXPERIMENTAL && PPP && NET_IPGRE_DEMUX
+	help
+	  Support for PPP over IPv4.(Point-to-Point Tunneling Protocol)
+
+	  This driver requires pppd plugin to work in client mode or
+	  modified pptpd (poptop) to work in server mode.
+	  See http://accel-pptp.sourceforge.net/ for information how to
+	  utilize this module.
+
 config PPPOATM
 	tristate "PPP over ATM"
 	depends on ATM && PPP
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 56e8c27f77ce..0b371083d481 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -162,6 +162,7 @@ obj-$(CONFIG_PPP_BSDCOMP) += bsd_comp.o
 obj-$(CONFIG_PPP_MPPE) += ppp_mppe.o
 obj-$(CONFIG_PPPOE) += pppox.o pppoe.o
 obj-$(CONFIG_PPPOL2TP) += pppox.o
+obj-$(CONFIG_PPTP) += pppox.o pptp.o
 
 obj-$(CONFIG_SLIP) += slip.o
 obj-$(CONFIG_SLHC) += slhc.o
diff --git a/drivers/net/pptp.c b/drivers/net/pptp.c
new file mode 100644
index 000000000000..761f0eced724
--- /dev/null
+++ b/drivers/net/pptp.c
@@ -0,0 +1,726 @@
+/*
+ *  Point-to-Point Tunneling Protocol for Linux
+ *
+ *	Authors: Dmitry Kozlov <xeb@mail.ru>
+ *
+ *	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, or (at your option) any later version.
+ *
+ */
+
+#include <linux/string.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/errno.h>
+#include <linux/netdevice.h>
+#include <linux/net.h>
+#include <linux/skbuff.h>
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+#include <linux/ppp_channel.h>
+#include <linux/ppp_defs.h>
+#include <linux/if_pppox.h>
+#include <linux/if_ppp.h>
+#include <linux/notifier.h>
+#include <linux/file.h>
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/netfilter.h>
+#include <linux/netfilter_ipv4.h>
+#include <linux/version.h>
+#include <linux/rcupdate.h>
+#include <linux/spinlock.h>
+
+#include <net/sock.h>
+#include <net/protocol.h>
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/route.h>
+#include <net/gre.h>
+
+#include <linux/uaccess.h>
+
+#define PPTP_DRIVER_VERSION "0.8.5"
+
+#define MAX_CALLID 65535
+
+static DECLARE_BITMAP(callid_bitmap, MAX_CALLID + 1);
+static struct pppox_sock **callid_sock;
+
+static DEFINE_SPINLOCK(chan_lock);
+
+static struct proto pptp_sk_proto __read_mostly;
+static struct ppp_channel_ops pptp_chan_ops;
+static const struct proto_ops pptp_ops;
+
+#define PPP_LCP_ECHOREQ 0x09
+#define PPP_LCP_ECHOREP 0x0A
+#define SC_RCV_BITS	(SC_RCV_B7_1|SC_RCV_B7_0|SC_RCV_ODDP|SC_RCV_EVNP)
+
+#define MISSING_WINDOW 20
+#define WRAPPED(curseq, lastseq)\
+	((((curseq) & 0xffffff00) == 0) &&\
+	(((lastseq) & 0xffffff00) == 0xffffff00))
+
+#define PPTP_GRE_PROTO  0x880B
+#define PPTP_GRE_VER    0x1
+
+#define PPTP_GRE_FLAG_C	0x80
+#define PPTP_GRE_FLAG_R	0x40
+#define PPTP_GRE_FLAG_K	0x20
+#define PPTP_GRE_FLAG_S	0x10
+#define PPTP_GRE_FLAG_A	0x80
+
+#define PPTP_GRE_IS_C(f) ((f)&PPTP_GRE_FLAG_C)
+#define PPTP_GRE_IS_R(f) ((f)&PPTP_GRE_FLAG_R)
+#define PPTP_GRE_IS_K(f) ((f)&PPTP_GRE_FLAG_K)
+#define PPTP_GRE_IS_S(f) ((f)&PPTP_GRE_FLAG_S)
+#define PPTP_GRE_IS_A(f) ((f)&PPTP_GRE_FLAG_A)
+
+#define PPTP_HEADER_OVERHEAD (2+sizeof(struct pptp_gre_header))
+struct pptp_gre_header {
+	u8  flags;
+	u8  ver;
+	u16 protocol;
+	u16 payload_len;
+	u16 call_id;
+	u32 seq;
+	u32 ack;
+} __packed;
+
+static struct pppox_sock *lookup_chan(u16 call_id, __be32 s_addr)
+{
+	struct pppox_sock *sock;
+	struct pptp_opt *opt;
+
+	rcu_read_lock();
+	sock = rcu_dereference(callid_sock[call_id]);
+	if (sock) {
+		opt = &sock->proto.pptp;
+		if (opt->dst_addr.sin_addr.s_addr != s_addr)
+			sock = NULL;
+		else
+			sock_hold(sk_pppox(sock));
+	}
+	rcu_read_unlock();
+
+	return sock;
+}
+
+static int lookup_chan_dst(u16 call_id, __be32 d_addr)
+{
+	struct pppox_sock *sock;
+	struct pptp_opt *opt;
+	int i;
+
+	rcu_read_lock();
+	for (i = find_next_bit(callid_bitmap, MAX_CALLID, 1); i < MAX_CALLID;
+	     i = find_next_bit(callid_bitmap, MAX_CALLID, i + 1)) {
+		sock = rcu_dereference(callid_sock[i]);
+		if (!sock)
+			continue;
+		opt = &sock->proto.pptp;
+		if (opt->dst_addr.call_id == call_id &&
+			  opt->dst_addr.sin_addr.s_addr == d_addr)
+			break;
+	}
+	rcu_read_unlock();
+
+	return i < MAX_CALLID;
+}
+
+static int add_chan(struct pppox_sock *sock)
+{
+	static int call_id;
+
+	spin_lock(&chan_lock);
+	if (!sock->proto.pptp.src_addr.call_id)	{
+		call_id = find_next_zero_bit(callid_bitmap, MAX_CALLID, call_id + 1);
+		if (call_id == MAX_CALLID) {
+			call_id = find_next_zero_bit(callid_bitmap, MAX_CALLID, 1);
+			if (call_id == MAX_CALLID)
+				goto out_err;
+		}
+		sock->proto.pptp.src_addr.call_id = call_id;
+	} else if (test_bit(sock->proto.pptp.src_addr.call_id, callid_bitmap))
+		goto out_err;
+
+	set_bit(sock->proto.pptp.src_addr.call_id, callid_bitmap);
+	rcu_assign_pointer(callid_sock[sock->proto.pptp.src_addr.call_id], sock);
+	spin_unlock(&chan_lock);
+
+	return 0;
+
+out_err:
+	spin_unlock(&chan_lock);
+	return -1;
+}
+
+static void del_chan(struct pppox_sock *sock)
+{
+	spin_lock(&chan_lock);
+	clear_bit(sock->proto.pptp.src_addr.call_id, callid_bitmap);
+	rcu_assign_pointer(callid_sock[sock->proto.pptp.src_addr.call_id], NULL);
+	spin_unlock(&chan_lock);
+	synchronize_rcu();
+}
+
+static int pptp_xmit(struct ppp_channel *chan, struct sk_buff *skb)
+{
+	struct sock *sk = (struct sock *) chan->private;
+	struct pppox_sock *po = pppox_sk(sk);
+	struct pptp_opt *opt = &po->proto.pptp;
+	struct pptp_gre_header *hdr;
+	unsigned int header_len = sizeof(*hdr);
+	int err = 0;
+	int islcp;
+	int len;
+	unsigned char *data;
+	__u32 seq_recv;
+
+
+	struct rtable *rt;
+	struct net_device *tdev;
+	struct iphdr  *iph;
+	int    max_headroom;
+
+	if (sk_pppox(po)->sk_state & PPPOX_DEAD)
+		goto tx_error;
+
+	{
+		struct flowi fl = { .oif = 0,
+			.nl_u = {
+				.ip4_u = {
+					.daddr = opt->dst_addr.sin_addr.s_addr,
+					.saddr = opt->src_addr.sin_addr.s_addr,
+					.tos = RT_TOS(0) } },
+			.proto = IPPROTO_GRE };
+		err = ip_route_output_key(&init_net, &rt, &fl);
+		if (err)
+			goto tx_error;
+	}
+	tdev = rt->dst.dev;
+
+	max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(*iph) + sizeof(*hdr) + 2;
+
+	if (skb_headroom(skb) < max_headroom || skb_cloned(skb) || skb_shared(skb)) {
+		struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom);
+		if (!new_skb) {
+			ip_rt_put(rt);
+			goto tx_error;
+		}
+		if (skb->sk)
+			skb_set_owner_w(new_skb, skb->sk);
+		kfree_skb(skb);
+		skb = new_skb;
+	}
+
+	data = skb->data;
+	islcp = ((data[0] << 8) + data[1]) == PPP_LCP && 1 <= data[2] && data[2] <= 7;
+
+	/* compress protocol field */
+	if ((opt->ppp_flags & SC_COMP_PROT) && data[0] == 0 && !islcp)
+		skb_pull(skb, 1);
+
+	/* Put in the address/control bytes if necessary */
+	if ((opt->ppp_flags & SC_COMP_AC) == 0 || islcp) {
+		data = skb_push(skb, 2);
+		data[0] = PPP_ALLSTATIONS;
+		data[1] = PPP_UI;
+	}
+
+	len = skb->len;
+
+	seq_recv = opt->seq_recv;
+
+	if (opt->ack_sent == seq_recv)
+		header_len -= sizeof(hdr->ack);
+
+	/* Push down and install GRE header */
+	skb_push(skb, header_len);
+	hdr = (struct pptp_gre_header *)(skb->data);
+
+	hdr->flags       = PPTP_GRE_FLAG_K;
+	hdr->ver         = PPTP_GRE_VER;
+	hdr->protocol    = htons(PPTP_GRE_PROTO);
+	hdr->call_id     = htons(opt->dst_addr.call_id);
+
+	hdr->flags      |= PPTP_GRE_FLAG_S;
+	hdr->seq         = htonl(++opt->seq_sent);
+	if (opt->ack_sent != seq_recv)	{
+		/* send ack with this message */
+		hdr->ver |= PPTP_GRE_FLAG_A;
+		hdr->ack  = htonl(seq_recv);
+		opt->ack_sent = seq_recv;
+	}
+	hdr->payload_len = htons(len);
+
+	/*	Push down and install the IP header. */
+
+	skb_reset_transport_header(skb);
+	skb_push(skb, sizeof(*iph));
+	skb_reset_network_header(skb);
+	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED | IPSKB_REROUTED);
+
+	iph =	ip_hdr(skb);
+	iph->version =	4;
+	iph->ihl =	sizeof(struct iphdr) >> 2;
+	if (ip_dont_fragment(sk, &rt->dst))
+		iph->frag_off	=	htons(IP_DF);
+	else
+		iph->frag_off	=	0;
+	iph->protocol = IPPROTO_GRE;
+	iph->tos      = 0;
+	iph->daddr    = rt->rt_dst;
+	iph->saddr    = rt->rt_src;
+	iph->ttl      = dst_metric(&rt->dst, RTAX_HOPLIMIT);
+	iph->tot_len  = htons(skb->len);
+
+	skb_dst_drop(skb);
+	skb_dst_set(skb, &rt->dst);
+
+	nf_reset(skb);
+
+	skb->ip_summed = CHECKSUM_NONE;
+	ip_select_ident(iph, &rt->dst, NULL);
+	ip_send_check(iph);
+
+	ip_local_out(skb);
+
+tx_error:
+	return 1;
+}
+
+static int pptp_rcv_core(struct sock *sk, struct sk_buff *skb)
+{
+	struct pppox_sock *po = pppox_sk(sk);
+	struct pptp_opt *opt = &po->proto.pptp;
+	int headersize, payload_len, seq;
+	__u8 *payload;
+	struct pptp_gre_header *header;
+
+	if (!(sk->sk_state & PPPOX_CONNECTED)) {
+		if (sock_queue_rcv_skb(sk, skb))
+			goto drop;
+		return NET_RX_SUCCESS;
+	}
+
+	header = (struct pptp_gre_header *)(skb->data);
+
+	/* test if acknowledgement present */
+	if (PPTP_GRE_IS_A(header->ver)) {
+		__u32 ack = (PPTP_GRE_IS_S(header->flags)) ?
+				header->ack : header->seq; /* ack in different place if S = 0 */
+
+		ack = ntohl(ack);
+
+		if (ack > opt->ack_recv)
+			opt->ack_recv = ack;
+		/* also handle sequence number wrap-around  */
+		if (WRAPPED(ack, opt->ack_recv))
+			opt->ack_recv = ack;
+	}
+
+	/* test if payload present */
+	if (!PPTP_GRE_IS_S(header->flags))
+		goto drop;
+
+	headersize  = sizeof(*header);
+	payload_len = ntohs(header->payload_len);
+	seq         = ntohl(header->seq);
+
+	/* no ack present? */
+	if (!PPTP_GRE_IS_A(header->ver))
+		headersize -= sizeof(header->ack);
+	/* check for incomplete packet (length smaller than expected) */
+	if (skb->len - headersize < payload_len)
+		goto drop;
+
+	payload = skb->data + headersize;
+	/* check for expected sequence number */
+	if (seq < opt->seq_recv + 1 || WRAPPED(opt->seq_recv, seq)) {
+		if ((payload[0] == PPP_ALLSTATIONS) && (payload[1] == PPP_UI) &&
+				(PPP_PROTOCOL(payload) == PPP_LCP) &&
+				((payload[4] == PPP_LCP_ECHOREQ) || (payload[4] == PPP_LCP_ECHOREP)))
+			goto allow_packet;
+	} else {
+		opt->seq_recv = seq;
+allow_packet:
+		skb_pull(skb, headersize);
+
+		if (payload[0] == PPP_ALLSTATIONS && payload[1] == PPP_UI) {
+			/* chop off address/control */
+			if (skb->len < 3)
+				goto drop;
+			skb_pull(skb, 2);
+		}
+
+		if ((*skb->data) & 1) {
+			/* protocol is compressed */
+			skb_push(skb, 1)[0] = 0;
+		}
+
+		skb->ip_summed = CHECKSUM_NONE;
+		skb_set_network_header(skb, skb->head-skb->data);
+		ppp_input(&po->chan, skb);
+
+		return NET_RX_SUCCESS;
+	}
+drop:
+	kfree_skb(skb);
+	return NET_RX_DROP;
+}
+
+static int pptp_rcv(struct sk_buff *skb)
+{
+	struct pppox_sock *po;
+	struct pptp_gre_header *header;
+	struct iphdr *iph;
+
+	if (skb->pkt_type != PACKET_HOST)
+		goto drop;
+
+	if (!pskb_may_pull(skb, 12))
+		goto drop;
+
+	iph = ip_hdr(skb);
+
+	header = (struct pptp_gre_header *)skb->data;
+
+	if (ntohs(header->protocol) != PPTP_GRE_PROTO || /* PPTP-GRE protocol for PPTP */
+		PPTP_GRE_IS_C(header->flags) ||                /* flag C should be clear */
+		PPTP_GRE_IS_R(header->flags) ||                /* flag R should be clear */
+		!PPTP_GRE_IS_K(header->flags) ||               /* flag K should be set */
+		(header->flags&0xF) != 0)                      /* routing and recursion ctrl = 0 */
+		/* if invalid, discard this packet */
+		goto drop;
+
+	po = lookup_chan(htons(header->call_id), iph->saddr);
+	if (po) {
+		skb_dst_drop(skb);
+		nf_reset(skb);
+		return sk_receive_skb(sk_pppox(po), skb, 0);
+	}
+drop:
+	kfree_skb(skb);
+	return NET_RX_DROP;
+}
+
+static int pptp_bind(struct socket *sock, struct sockaddr *uservaddr,
+	int sockaddr_len)
+{
+	struct sock *sk = sock->sk;
+	struct sockaddr_pppox *sp = (struct sockaddr_pppox *) uservaddr;
+	struct pppox_sock *po = pppox_sk(sk);
+	struct pptp_opt *opt = &po->proto.pptp;
+	int error = 0;
+
+	lock_sock(sk);
+
+	opt->src_addr = sp->sa_addr.pptp;
+	if (add_chan(po)) {
+		release_sock(sk);
+		error = -EBUSY;
+	}
+
+	release_sock(sk);
+	return error;
+}
+
+static int pptp_connect(struct socket *sock, struct sockaddr *uservaddr,
+	int sockaddr_len, int flags)
+{
+	struct sock *sk = sock->sk;
+	struct sockaddr_pppox *sp = (struct sockaddr_pppox *) uservaddr;
+	struct pppox_sock *po = pppox_sk(sk);
+	struct pptp_opt *opt = &po->proto.pptp;
+	struct rtable *rt;
+	int error = 0;
+
+	if (sp->sa_protocol != PX_PROTO_PPTP)
+		return -EINVAL;
+
+	if (lookup_chan_dst(sp->sa_addr.pptp.call_id, sp->sa_addr.pptp.sin_addr.s_addr))
+		return -EALREADY;
+
+	lock_sock(sk);
+	/* Check for already bound sockets */
+	if (sk->sk_state & PPPOX_CONNECTED) {
+		error = -EBUSY;
+		goto end;
+	}
+
+	/* Check for already disconnected sockets, on attempts to disconnect */
+	if (sk->sk_state & PPPOX_DEAD) {
+		error = -EALREADY;
+		goto end;
+	}
+
+	if (!opt->src_addr.sin_addr.s_addr || !sp->sa_addr.pptp.sin_addr.s_addr) {
+		error = -EINVAL;
+		goto end;
+	}
+
+	po->chan.private = sk;
+	po->chan.ops = &pptp_chan_ops;
+
+	{
+		struct flowi fl = {
+			.nl_u = {
+				.ip4_u = {
+					.daddr = opt->dst_addr.sin_addr.s_addr,
+					.saddr = opt->src_addr.sin_addr.s_addr,
+					.tos = RT_CONN_FLAGS(sk) } },
+			.proto = IPPROTO_GRE };
+		security_sk_classify_flow(sk, &fl);
+		if (ip_route_output_key(&init_net, &rt, &fl)) {
+			error = -EHOSTUNREACH;
+			goto end;
+		}
+		sk_setup_caps(sk, &rt->dst);
+	}
+	po->chan.mtu = dst_mtu(&rt->dst);
+	if (!po->chan.mtu)
+		po->chan.mtu = PPP_MTU;
+	ip_rt_put(rt);
+	po->chan.mtu -= PPTP_HEADER_OVERHEAD;
+
+	po->chan.hdrlen = 2 + sizeof(struct pptp_gre_header);
+	error = ppp_register_channel(&po->chan);
+	if (error) {
+		pr_err("PPTP: failed to register PPP channel (%d)\n", error);
+		goto end;
+	}
+
+	opt->dst_addr = sp->sa_addr.pptp;
+	sk->sk_state = PPPOX_CONNECTED;
+
+ end:
+	release_sock(sk);
+	return error;
+}
+
+static int pptp_getname(struct socket *sock, struct sockaddr *uaddr,
+	int *usockaddr_len, int peer)
+{
+	int len = sizeof(struct sockaddr_pppox);
+	struct sockaddr_pppox sp;
+
+	sp.sa_family	  = AF_PPPOX;
+	sp.sa_protocol  = PX_PROTO_PPTP;
+	sp.sa_addr.pptp = pppox_sk(sock->sk)->proto.pptp.src_addr;
+
+	memcpy(uaddr, &sp, len);
+
+	*usockaddr_len = len;
+
+	return 0;
+}
+
+static int pptp_release(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+	struct pppox_sock *po;
+	struct pptp_opt *opt;
+	int error = 0;
+
+	if (!sk)
+		return 0;
+
+	lock_sock(sk);
+
+	if (sock_flag(sk, SOCK_DEAD)) {
+		release_sock(sk);
+		return -EBADF;
+	}
+
+	po = pppox_sk(sk);
+	opt = &po->proto.pptp;
+	del_chan(po);
+
+	pppox_unbind_sock(sk);
+	sk->sk_state = PPPOX_DEAD;
+
+	sock_orphan(sk);
+	sock->sk = NULL;
+
+	release_sock(sk);
+	sock_put(sk);
+
+	return error;
+}
+
+static void pptp_sock_destruct(struct sock *sk)
+{
+	if (!(sk->sk_state & PPPOX_DEAD)) {
+		del_chan(pppox_sk(sk));
+		pppox_unbind_sock(sk);
+	}
+	skb_queue_purge(&sk->sk_receive_queue);
+}
+
+static int pptp_create(struct net *net, struct socket *sock)
+{
+	int error = -ENOMEM;
+	struct sock *sk;
+	struct pppox_sock *po;
+	struct pptp_opt *opt;
+
+	sk = sk_alloc(net, PF_PPPOX, GFP_KERNEL, &pptp_sk_proto);
+	if (!sk)
+		goto out;
+
+	sock_init_data(sock, sk);
+
+	sock->state = SS_UNCONNECTED;
+	sock->ops   = &pptp_ops;
+
+	sk->sk_backlog_rcv = pptp_rcv_core;
+	sk->sk_state       = PPPOX_NONE;
+	sk->sk_type        = SOCK_STREAM;
+	sk->sk_family      = PF_PPPOX;
+	sk->sk_protocol    = PX_PROTO_PPTP;
+	sk->sk_destruct    = pptp_sock_destruct;
+
+	po = pppox_sk(sk);
+	opt = &po->proto.pptp;
+
+	opt->seq_sent = 0; opt->seq_recv = 0;
+	opt->ack_recv = 0; opt->ack_sent = 0;
+
+	error = 0;
+out:
+	return error;
+}
+
+static int pptp_ppp_ioctl(struct ppp_channel *chan, unsigned int cmd,
+	unsigned long arg)
+{
+	struct sock *sk = (struct sock *) chan->private;
+	struct pppox_sock *po = pppox_sk(sk);
+	struct pptp_opt *opt = &po->proto.pptp;
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
+	int err, val;
+
+	err = -EFAULT;
+	switch (cmd) {
+	case PPPIOCGFLAGS:
+		val = opt->ppp_flags;
+		if (put_user(val, p))
+			break;
+		err = 0;
+		break;
+	case PPPIOCSFLAGS:
+		if (get_user(val, p))
+			break;
+		opt->ppp_flags = val & ~SC_RCV_BITS;
+		err = 0;
+		break;
+	default:
+		err = -ENOTTY;
+	}
+
+	return err;
+}
+
+static struct ppp_channel_ops pptp_chan_ops = {
+	.start_xmit = pptp_xmit,
+	.ioctl      = pptp_ppp_ioctl,
+};
+
+static struct proto pptp_sk_proto __read_mostly = {
+	.name     = "PPTP",
+	.owner    = THIS_MODULE,
+	.obj_size = sizeof(struct pppox_sock),
+};
+
+static const struct proto_ops pptp_ops = {
+	.family     = AF_PPPOX,
+	.owner      = THIS_MODULE,
+	.release    = pptp_release,
+	.bind       = pptp_bind,
+	.connect    = pptp_connect,
+	.socketpair = sock_no_socketpair,
+	.accept     = sock_no_accept,
+	.getname    = pptp_getname,
+	.poll       = sock_no_poll,
+	.listen     = sock_no_listen,
+	.shutdown   = sock_no_shutdown,
+	.setsockopt = sock_no_setsockopt,
+	.getsockopt = sock_no_getsockopt,
+	.sendmsg    = sock_no_sendmsg,
+	.recvmsg    = sock_no_recvmsg,
+	.mmap       = sock_no_mmap,
+	.ioctl      = pppox_ioctl,
+};
+
+static struct pppox_proto pppox_pptp_proto = {
+	.create = pptp_create,
+	.owner  = THIS_MODULE,
+};
+
+static struct gre_protocol gre_pptp_protocol = {
+	.handler = pptp_rcv,
+};
+
+static int __init pptp_init_module(void)
+{
+	int err = 0;
+	pr_info("PPTP driver version " PPTP_DRIVER_VERSION "\n");
+
+	callid_sock = __vmalloc((MAX_CALLID + 1) * sizeof(void *),
+		GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL);
+	if (!callid_sock) {
+		pr_err("PPTP: cann't allocate memory\n");
+		return -ENOMEM;
+	}
+
+	err = gre_add_protocol(&gre_pptp_protocol, GREPROTO_PPTP);
+	if (err) {
+		pr_err("PPTP: can't add gre protocol\n");
+		goto out_mem_free;
+	}
+
+	err = proto_register(&pptp_sk_proto, 0);
+	if (err) {
+		pr_err("PPTP: can't register sk_proto\n");
+		goto out_gre_del_protocol;
+	}
+
+	err = register_pppox_proto(PX_PROTO_PPTP, &pppox_pptp_proto);
+	if (err) {
+		pr_err("PPTP: can't register pppox_proto\n");
+		goto out_unregister_sk_proto;
+	}
+
+	return 0;
+
+out_unregister_sk_proto:
+	proto_unregister(&pptp_sk_proto);
+out_gre_del_protocol:
+	gre_del_protocol(&gre_pptp_protocol, GREPROTO_PPTP);
+out_mem_free:
+	vfree(callid_sock);
+
+	return err;
+}
+
+static void __exit pptp_exit_module(void)
+{
+	unregister_pppox_proto(PX_PROTO_PPTP);
+	proto_unregister(&pptp_sk_proto);
+	gre_del_protocol(&gre_pptp_protocol, GREPROTO_PPTP);
+	vfree(callid_sock);
+}
+
+module_init(pptp_init_module);
+module_exit(pptp_exit_module);
+
+MODULE_DESCRIPTION("Point-to-Point Tunneling Protocol");
+MODULE_AUTHOR("D. Kozlov (xeb@mail.ru)");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/if_pppox.h b/include/linux/if_pppox.h
index 1925e0c3f162..1525b2156b2a 100644
--- a/include/linux/if_pppox.h
+++ b/include/linux/if_pppox.h
@@ -40,25 +40,35 @@
  * PPPoE addressing definition 
  */ 
 typedef __be16 sid_t;
-struct pppoe_addr{ 
-       sid_t           sid;                    /* Session identifier */ 
-       unsigned char   remote[ETH_ALEN];       /* Remote address */ 
-       char            dev[IFNAMSIZ];          /* Local device to use */ 
+struct pppoe_addr {
+	sid_t         sid;                    /* Session identifier */
+	unsigned char remote[ETH_ALEN];       /* Remote address */
+	char          dev[IFNAMSIZ];          /* Local device to use */
 }; 
  
 /************************************************************************ 
- * Protocols supported by AF_PPPOX 
- */ 
+ * PPTP addressing definition
+ */
+struct pptp_addr {
+	u16             call_id;
+	struct in_addr  sin_addr;
+};
+
+/************************************************************************
+ * Protocols supported by AF_PPPOX
+ */
 #define PX_PROTO_OE    0 /* Currently just PPPoE */
 #define PX_PROTO_OL2TP 1 /* Now L2TP also */
-#define PX_MAX_PROTO   2
-
-struct sockaddr_pppox { 
-       sa_family_t     sa_family;            /* address family, AF_PPPOX */ 
-       unsigned int    sa_protocol;          /* protocol identifier */ 
-       union{ 
-               struct pppoe_addr       pppoe; 
-       }sa_addr; 
+#define PX_PROTO_PPTP  2
+#define PX_MAX_PROTO   3
+
+struct sockaddr_pppox {
+	sa_family_t     sa_family;            /* address family, AF_PPPOX */
+	unsigned int    sa_protocol;          /* protocol identifier */
+	union {
+		struct pppoe_addr  pppoe;
+		struct pptp_addr   pptp;
+	} sa_addr;
 } __packed;
 
 /* The use of the above union isn't viable because the size of this
@@ -101,7 +111,7 @@ struct pppoe_tag {
 	__be16 tag_type;
 	__be16 tag_len;
 	char tag_data[0];
-} __attribute ((packed));
+} __packed;
 
 /* Tag identifiers */
 #define PTT_EOL		__cpu_to_be16(0x0000)
@@ -150,15 +160,23 @@ struct pppoe_opt {
 					     relayed to (PPPoE relaying) */
 };
 
+struct pptp_opt {
+	struct pptp_addr src_addr;
+	struct pptp_addr dst_addr;
+	u32 ack_sent, ack_recv;
+	u32 seq_sent, seq_recv;
+	int ppp_flags;
+};
 #include <net/sock.h>
 
 struct pppox_sock {
 	/* struct sock must be the first member of pppox_sock */
-	struct sock		sk;
-	struct ppp_channel	chan;
+	struct sock sk;
+	struct ppp_channel chan;
 	struct pppox_sock	*next;	  /* for hash table */
 	union {
 		struct pppoe_opt pppoe;
+		struct pptp_opt  pptp;
 	} proto;
 	__be16			num;
 };
diff --git a/include/net/gre.h b/include/net/gre.h
new file mode 100644
index 000000000000..82665474bcb7
--- /dev/null
+++ b/include/net/gre.h
@@ -0,0 +1,18 @@
+#ifndef __LINUX_GRE_H
+#define __LINUX_GRE_H
+
+#include <linux/skbuff.h>
+
+#define GREPROTO_CISCO		0
+#define GREPROTO_PPTP		1
+#define GREPROTO_MAX		2
+
+struct gre_protocol {
+	int  (*handler)(struct sk_buff *skb);
+	void (*err_handler)(struct sk_buff *skb, u32 info);
+};
+
+int gre_add_protocol(const struct gre_protocol *proto, u8 version);
+int gre_del_protocol(const struct gre_protocol *proto, u8 version);
+
+#endif
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 7c3a7d191249..7458bdae7e9f 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -215,8 +215,15 @@ config NET_IPIP
 	  be inserted in and removed from the running kernel whenever you
 	  want). Most people won't need this and can say N.
 
+config NET_IPGRE_DEMUX
+	tristate "IP: GRE demultiplexer"
+	help
+	 This is helper module to demultiplex GRE packets on GRE version field criteria.
+	 Required by ip_gre and pptp modules.
+
 config NET_IPGRE
 	tristate "IP: GRE tunnels over IP"
+	depends on NET_IPGRE_DEMUX
 	help
 	  Tunneling means encapsulating data of one protocol type within
 	  another protocol and sending it over a channel that understands the
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 80ff87ce43aa..4978d22f9a75 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_PROC_FS) += proc.o
 obj-$(CONFIG_IP_MULTIPLE_TABLES) += fib_rules.o
 obj-$(CONFIG_IP_MROUTE) += ipmr.o
 obj-$(CONFIG_NET_IPIP) += ipip.o
+obj-$(CONFIG_NET_IPGRE_DEMUX) += gre.o
 obj-$(CONFIG_NET_IPGRE) += ip_gre.o
 obj-$(CONFIG_SYN_COOKIES) += syncookies.o
 obj-$(CONFIG_INET_AH) += ah4.o
diff --git a/net/ipv4/gre.c b/net/ipv4/gre.c
new file mode 100644
index 000000000000..b546736da2e1
--- /dev/null
+++ b/net/ipv4/gre.c
@@ -0,0 +1,151 @@
+/*
+ *	GRE over IPv4 demultiplexer driver
+ *
+ *	Authors: Dmitry Kozlov (xeb@mail.ru)
+ *
+ *	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, or (at your option) any later version.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/kmod.h>
+#include <linux/skbuff.h>
+#include <linux/in.h>
+#include <linux/netdevice.h>
+#include <linux/version.h>
+#include <linux/spinlock.h>
+#include <net/protocol.h>
+#include <net/gre.h>
+
+
+const struct gre_protocol *gre_proto[GREPROTO_MAX] __read_mostly;
+static DEFINE_SPINLOCK(gre_proto_lock);
+
+int gre_add_protocol(const struct gre_protocol *proto, u8 version)
+{
+	if (version >= GREPROTO_MAX)
+		goto err_out;
+
+	spin_lock(&gre_proto_lock);
+	if (gre_proto[version])
+		goto err_out_unlock;
+
+	rcu_assign_pointer(gre_proto[version], proto);
+	spin_unlock(&gre_proto_lock);
+	return 0;
+
+err_out_unlock:
+	spin_unlock(&gre_proto_lock);
+err_out:
+	return -1;
+}
+EXPORT_SYMBOL_GPL(gre_add_protocol);
+
+int gre_del_protocol(const struct gre_protocol *proto, u8 version)
+{
+	if (version >= GREPROTO_MAX)
+		goto err_out;
+
+	spin_lock(&gre_proto_lock);
+	if (gre_proto[version] != proto)
+		goto err_out_unlock;
+	rcu_assign_pointer(gre_proto[version], NULL);
+	spin_unlock(&gre_proto_lock);
+	synchronize_rcu();
+	return 0;
+
+err_out_unlock:
+	spin_unlock(&gre_proto_lock);
+err_out:
+	return -1;
+}
+EXPORT_SYMBOL_GPL(gre_del_protocol);
+
+static int gre_rcv(struct sk_buff *skb)
+{
+	const struct gre_protocol *proto;
+	u8 ver;
+	int ret;
+
+	if (!pskb_may_pull(skb, 12))
+		goto drop;
+
+	ver = skb->data[1]&0x7f;
+	if (ver >= GREPROTO_MAX)
+		goto drop;
+
+	rcu_read_lock();
+	proto = rcu_dereference(gre_proto[ver]);
+	if (!proto || !proto->handler)
+		goto drop_unlock;
+	ret = proto->handler(skb);
+	rcu_read_unlock();
+	return ret;
+
+drop_unlock:
+	rcu_read_unlock();
+drop:
+	kfree_skb(skb);
+	return NET_RX_DROP;
+}
+
+static void gre_err(struct sk_buff *skb, u32 info)
+{
+	const struct gre_protocol *proto;
+	u8 ver;
+
+	if (!pskb_may_pull(skb, 12))
+		goto drop;
+
+	ver = skb->data[1]&0x7f;
+	if (ver >= GREPROTO_MAX)
+		goto drop;
+
+	rcu_read_lock();
+	proto = rcu_dereference(gre_proto[ver]);
+	if (!proto || !proto->err_handler)
+		goto drop_unlock;
+	proto->err_handler(skb, info);
+	rcu_read_unlock();
+	return;
+
+drop_unlock:
+	rcu_read_unlock();
+drop:
+	kfree_skb(skb);
+}
+
+static const struct net_protocol net_gre_protocol = {
+	.handler     = gre_rcv,
+	.err_handler = gre_err,
+	.netns_ok    = 1,
+};
+
+static int __init gre_init(void)
+{
+	pr_info("GRE over IPv4 demultiplexor driver");
+
+	if (inet_add_protocol(&net_gre_protocol, IPPROTO_GRE) < 0) {
+		pr_err("gre: can't add protocol\n");
+		return -EAGAIN;
+	}
+
+	return 0;
+}
+
+static void __exit gre_exit(void)
+{
+	inet_del_protocol(&net_gre_protocol, IPPROTO_GRE);
+}
+
+module_init(gre_init);
+module_exit(gre_exit);
+
+MODULE_DESCRIPTION("GRE over IPv4 demultiplexer driver");
+MODULE_AUTHOR("D. Kozlov (xeb@mail.ru)");
+MODULE_LICENSE("GPL");
+
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 945b20a5ad50..85176895495a 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -44,6 +44,7 @@
 #include <net/net_namespace.h>
 #include <net/netns/generic.h>
 #include <net/rtnetlink.h>
+#include <net/gre.h>
 
 #ifdef CONFIG_IPV6
 #include <net/ipv6.h>
@@ -1278,10 +1279,9 @@ static void ipgre_fb_tunnel_init(struct net_device *dev)
 }
 
 
-static const struct net_protocol ipgre_protocol = {
-	.handler	=	ipgre_rcv,
-	.err_handler	=	ipgre_err,
-	.netns_ok	=	1,
+static const struct gre_protocol ipgre_protocol = {
+	.handler     = ipgre_rcv,
+	.err_handler = ipgre_err,
 };
 
 static void ipgre_destroy_tunnels(struct ipgre_net *ign, struct list_head *head)
@@ -1663,7 +1663,7 @@ static int __init ipgre_init(void)
 	if (err < 0)
 		return err;
 
-	err = inet_add_protocol(&ipgre_protocol, IPPROTO_GRE);
+	err = gre_add_protocol(&ipgre_protocol, GREPROTO_CISCO);
 	if (err < 0) {
 		printk(KERN_INFO "ipgre init: can't add protocol\n");
 		goto add_proto_failed;
@@ -1683,7 +1683,7 @@ out:
 tap_ops_failed:
 	rtnl_link_unregister(&ipgre_link_ops);
 rtnl_link_failed:
-	inet_del_protocol(&ipgre_protocol, IPPROTO_GRE);
+	gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO);
 add_proto_failed:
 	unregister_pernet_device(&ipgre_net_ops);
 	goto out;
@@ -1693,7 +1693,7 @@ static void __exit ipgre_fini(void)
 {
 	rtnl_link_unregister(&ipgre_tap_ops);
 	rtnl_link_unregister(&ipgre_link_ops);
-	if (inet_del_protocol(&ipgre_protocol, IPPROTO_GRE) < 0)
+	if (gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO) < 0)
 		printk(KERN_INFO "ipgre close: can't remove protocol\n");
 	unregister_pernet_device(&ipgre_net_ops);
 }
-- 
cgit v1.2.3


From 739a91ef0625e0e4a40b835f4f891313c47915df Mon Sep 17 00:00:00 2001
From: Changli Gao <xiaosuo@gmail.com>
Date: Sat, 21 Aug 2010 06:23:15 +0000
Subject: net_sched: cls_flow: add key rxhash

We can use rxhash to classify the traffic into flows. As rxhash maybe
supplied by NIC or RPS, it is cheaper.

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Acked-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_cls.h | 1 +
 net/sched/cls_flow.c    | 7 +++++++
 2 files changed, 8 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h
index 7f6ba8658abe..defbde203d07 100644
--- a/include/linux/pkt_cls.h
+++ b/include/linux/pkt_cls.h
@@ -332,6 +332,7 @@ enum {
 	FLOW_KEY_SKUID,
 	FLOW_KEY_SKGID,
 	FLOW_KEY_VLAN_TAG,
+	FLOW_KEY_RXHASH,
 	__FLOW_KEY_MAX,
 };
 
diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c
index cd709f1294df..5b271a18bc3a 100644
--- a/net/sched/cls_flow.c
+++ b/net/sched/cls_flow.c
@@ -306,6 +306,11 @@ static u32 flow_get_vlan_tag(const struct sk_buff *skb)
 	return tag & VLAN_VID_MASK;
 }
 
+static u32 flow_get_rxhash(struct sk_buff *skb)
+{
+	return skb_get_rxhash(skb);
+}
+
 static u32 flow_key_get(struct sk_buff *skb, int key)
 {
 	switch (key) {
@@ -343,6 +348,8 @@ static u32 flow_key_get(struct sk_buff *skb, int key)
 		return flow_get_skgid(skb);
 	case FLOW_KEY_VLAN_TAG:
 		return flow_get_vlan_tag(skb);
+	case FLOW_KEY_RXHASH:
+		return flow_get_rxhash(skb);
 	default:
 		WARN_ON(1);
 		return 0;
-- 
cgit v1.2.3


From d8287fc864643beaf1623c92aceb1ab38eae0648 Mon Sep 17 00:00:00 2001
From: Changli Gao <xiaosuo@gmail.com>
Date: Sun, 22 Aug 2010 18:37:27 -0700
Subject: net: use __be16 instead of u16 for the userspace code

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_pppox.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_pppox.h b/include/linux/if_pppox.h
index 1525b2156b2a..770e8fa669d2 100644
--- a/include/linux/if_pppox.h
+++ b/include/linux/if_pppox.h
@@ -50,8 +50,8 @@ struct pppoe_addr {
  * PPTP addressing definition
  */
 struct pptp_addr {
-	u16             call_id;
-	struct in_addr  sin_addr;
+	__be16		call_id;
+	struct in_addr	sin_addr;
 };
 
 /************************************************************************
-- 
cgit v1.2.3


From 05532121da0728eaedac2a0a5c3cecad3a95d765 Mon Sep 17 00:00:00 2001
From: Changli Gao <xiaosuo@gmail.com>
Date: Sun, 22 Aug 2010 21:03:33 -0700
Subject: net: 802.1q: make vlan_hwaccel_do_receive() return void

vlan_hwaccel_do_receive() always returns 0, so make it return void.

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 5 ++---
 net/8021q/vlan_core.c   | 3 +--
 net/core/dev.c          | 4 ++--
 3 files changed, 5 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 3d870fda8c4f..a52320751bfc 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -119,7 +119,7 @@ extern u16 vlan_dev_vlan_id(const struct net_device *dev);
 
 extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 			     u16 vlan_tci, int polling);
-extern int vlan_hwaccel_do_receive(struct sk_buff *skb);
+extern void vlan_hwaccel_do_receive(struct sk_buff *skb);
 extern gro_result_t
 vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp,
 		 unsigned int vlan_tci, struct sk_buff *skb);
@@ -147,9 +147,8 @@ static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 	return NET_XMIT_SUCCESS;
 }
 
-static inline int vlan_hwaccel_do_receive(struct sk_buff *skb)
+static inline void vlan_hwaccel_do_receive(struct sk_buff *skb)
 {
-	return 0;
 }
 
 static inline gro_result_t
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 01ddb0472f86..07eeb5b99dce 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -35,7 +35,7 @@ drop:
 }
 EXPORT_SYMBOL(__vlan_hwaccel_rx);
 
-int vlan_hwaccel_do_receive(struct sk_buff *skb)
+void vlan_hwaccel_do_receive(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
 	struct vlan_rx_stats     *rx_stats;
@@ -69,7 +69,6 @@ int vlan_hwaccel_do_receive(struct sk_buff *skb)
 		break;
 	}
 	u64_stats_update_end(&rx_stats->syncp);
-	return 0;
 }
 
 struct net_device *vlan_dev_real_dev(const struct net_device *dev)
diff --git a/net/core/dev.c b/net/core/dev.c
index 7cd5237d9822..d569f88bcf80 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2841,8 +2841,8 @@ static int __netif_receive_skb(struct sk_buff *skb)
 	if (!netdev_tstamp_prequeue)
 		net_timestamp_check(skb);
 
-	if (vlan_tx_tag_present(skb) && vlan_hwaccel_do_receive(skb))
-		return NET_RX_SUCCESS;
+	if (vlan_tx_tag_present(skb))
+		vlan_hwaccel_do_receive(skb);
 
 	/* if we've gotten here through NAPI, check netpoll */
 	if (netpoll_receive_skb(skb))
-- 
cgit v1.2.3


From fcb12fd2236f49aa8fdc1568ed4ebdfe4fddc6b5 Mon Sep 17 00:00:00 2001
From: Changli Gao <xiaosuo@gmail.com>
Date: Sun, 22 Aug 2010 16:41:59 +0000
Subject: net: rds: remove duplication type definitions

__be* are defined in linux/types.h now, and in fact, rds.h isn't exported
to user space even.

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rds.h | 9 ---------
 1 file changed, 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rds.h b/include/linux/rds.h
index 24bce3ded9ea..7f3971d9fc5c 100644
--- a/include/linux/rds.h
+++ b/include/linux/rds.h
@@ -36,15 +36,6 @@
 
 #include <linux/types.h>
 
-/* These sparse annotated types shouldn't be in any user
- * visible header file. We should clean this up rather
- * than kludging around them. */
-#ifndef __KERNEL__
-#define __be16	u_int16_t
-#define __be32	u_int32_t
-#define __be64	u_int64_t
-#endif
-
 #define RDS_IB_ABI_VERSION		0x301
 
 /*
-- 
cgit v1.2.3


From 21dc330157454046dd7c494961277d76e1c957fe Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 23 Aug 2010 00:13:46 -0700
Subject: net: Rename skb_has_frags to skb_has_frag_list

SKBs can be "fragmented" in two ways, via a page array (called
skb_shinfo(skb)->frags[]) and via a list of SKBs (called
skb_shinfo(skb)->frag_list).

Since skb_has_frags() tests the latter, it's name is confusing
since it sounds more like it's testing the former.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h               |  2 +-
 include/linux/skbuff.h                  |  4 ++--
 net/core/dev.c                          |  4 ++--
 net/core/skbuff.c                       | 18 +++++++++---------
 net/ipv4/ip_fragment.c                  |  2 +-
 net/ipv4/ip_output.c                    |  2 +-
 net/ipv6/ip6_output.c                   |  2 +-
 net/ipv6/netfilter/nf_conntrack_reasm.c |  2 +-
 net/ipv6/reassembly.c                   |  2 +-
 9 files changed, 19 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 46c36ffe20ee..ce2de8b64083 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2191,7 +2191,7 @@ static inline int net_gso_ok(int features, int gso_type)
 static inline int skb_gso_ok(struct sk_buff *skb, int features)
 {
 	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
-	       (!skb_has_frags(skb) || (features & NETIF_F_FRAGLIST));
+	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
 }
 
 static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index f067c95cf18a..f900ffcd847e 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1120,7 +1120,7 @@ extern void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page,
 			    int off, int size);
 
 #define SKB_PAGE_ASSERT(skb) 	BUG_ON(skb_shinfo(skb)->nr_frags)
-#define SKB_FRAG_ASSERT(skb) 	BUG_ON(skb_has_frags(skb))
+#define SKB_FRAG_ASSERT(skb) 	BUG_ON(skb_has_frag_list(skb))
 #define SKB_LINEAR_ASSERT(skb)  BUG_ON(skb_is_nonlinear(skb))
 
 #ifdef NET_SKBUFF_DATA_USES_OFFSET
@@ -1784,7 +1784,7 @@ static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len)
 		     skb = skb->prev)
 
 
-static inline bool skb_has_frags(const struct sk_buff *skb)
+static inline bool skb_has_frag_list(const struct sk_buff *skb)
 {
 	return skb_shinfo(skb)->frag_list != NULL;
 }
diff --git a/net/core/dev.c b/net/core/dev.c
index d569f88bcf80..859e30ff044a 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1930,7 +1930,7 @@ static inline int skb_needs_linearize(struct sk_buff *skb,
 				      struct net_device *dev)
 {
 	return skb_is_nonlinear(skb) &&
-	       ((skb_has_frags(skb) && !(dev->features & NETIF_F_FRAGLIST)) ||
+	       ((skb_has_frag_list(skb) && !(dev->features & NETIF_F_FRAGLIST)) ||
 	        (skb_shinfo(skb)->nr_frags && (!(dev->features & NETIF_F_SG) ||
 					      illegal_highdma(dev, skb))));
 }
@@ -3090,7 +3090,7 @@ enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 	if (!(skb->dev->features & NETIF_F_GRO) || netpoll_rx_on(skb))
 		goto normal;
 
-	if (skb_is_gso(skb) || skb_has_frags(skb))
+	if (skb_is_gso(skb) || skb_has_frag_list(skb))
 		goto normal;
 
 	rcu_read_lock();
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 99ef721f773d..e2535fb4985d 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -340,7 +340,7 @@ static void skb_release_data(struct sk_buff *skb)
 				put_page(skb_shinfo(skb)->frags[i].page);
 		}
 
-		if (skb_has_frags(skb))
+		if (skb_has_frag_list(skb))
 			skb_drop_fraglist(skb);
 
 		kfree(skb->head);
@@ -759,7 +759,7 @@ struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask)
 		skb_shinfo(n)->nr_frags = i;
 	}
 
-	if (skb_has_frags(skb)) {
+	if (skb_has_frag_list(skb)) {
 		skb_shinfo(n)->frag_list = skb_shinfo(skb)->frag_list;
 		skb_clone_fraglist(n);
 	}
@@ -822,7 +822,7 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
 		get_page(skb_shinfo(skb)->frags[i].page);
 
-	if (skb_has_frags(skb))
+	if (skb_has_frag_list(skb))
 		skb_clone_fraglist(skb);
 
 	skb_release_data(skb);
@@ -1099,7 +1099,7 @@ drop_pages:
 		for (; i < nfrags; i++)
 			put_page(skb_shinfo(skb)->frags[i].page);
 
-		if (skb_has_frags(skb))
+		if (skb_has_frag_list(skb))
 			skb_drop_fraglist(skb);
 		goto done;
 	}
@@ -1194,7 +1194,7 @@ unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta)
 	/* Optimization: no fragments, no reasons to preestimate
 	 * size of pulled pages. Superb.
 	 */
-	if (!skb_has_frags(skb))
+	if (!skb_has_frag_list(skb))
 		goto pull_pages;
 
 	/* Estimate size of pulled pages. */
@@ -2323,7 +2323,7 @@ next_skb:
 		st->frag_data = NULL;
 	}
 
-	if (st->root_skb == st->cur_skb && skb_has_frags(st->root_skb)) {
+	if (st->root_skb == st->cur_skb && skb_has_frag_list(st->root_skb)) {
 		st->cur_skb = skb_shinfo(st->root_skb)->frag_list;
 		st->frag_idx = 0;
 		goto next_skb;
@@ -2889,7 +2889,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer)
 		return -ENOMEM;
 
 	/* Easy case. Most of packets will go this way. */
-	if (!skb_has_frags(skb)) {
+	if (!skb_has_frag_list(skb)) {
 		/* A little of trouble, not enough of space for trailer.
 		 * This should not happen, when stack is tuned to generate
 		 * good frames. OK, on miss we reallocate and reserve even more
@@ -2924,7 +2924,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer)
 
 		if (skb1->next == NULL && tailbits) {
 			if (skb_shinfo(skb1)->nr_frags ||
-			    skb_has_frags(skb1) ||
+			    skb_has_frag_list(skb1) ||
 			    skb_tailroom(skb1) < tailbits)
 				ntail = tailbits + 128;
 		}
@@ -2933,7 +2933,7 @@ int skb_cow_data(struct sk_buff *skb, int tailbits, struct sk_buff **trailer)
 		    skb_cloned(skb1) ||
 		    ntail ||
 		    skb_shinfo(skb1)->nr_frags ||
-		    skb_has_frags(skb1)) {
+		    skb_has_frag_list(skb1)) {
 			struct sk_buff *skb2;
 
 			/* Fuck, we are miserable poor guys... */
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index b7c41654dde5..f4dc879e258e 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -542,7 +542,7 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev,
 	/* If the first fragment is fragmented itself, we split
 	 * it to two chunks: the first with data and paged part
 	 * and the second, holding only fragments. */
-	if (skb_has_frags(head)) {
+	if (skb_has_frag_list(head)) {
 		struct sk_buff *clone;
 		int i, plen = 0;
 
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index e807492f1777..6d2753c7ffdd 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -487,7 +487,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 	 * LATER: this step can be merged to real generation of fragments,
 	 * we can switch to copy when see the first bad fragment.
 	 */
-	if (skb_has_frags(skb)) {
+	if (skb_has_frag_list(skb)) {
 		struct sk_buff *frag;
 		int first_len = skb_pagelen(skb);
 		int truesizes = 0;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index d40b330c0ee6..1838927a2243 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -637,7 +637,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 	}
 	mtu -= hlen + sizeof(struct frag_hdr);
 
-	if (skb_has_frags(skb)) {
+	if (skb_has_frag_list(skb)) {
 		int first_len = skb_pagelen(skb);
 		int truesizes = 0;
 
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index 13ef5bc05cf5..089c598773c7 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -413,7 +413,7 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev)
 	/* If the first fragment is fragmented itself, we split
 	 * it to two chunks: the first with data and paged part
 	 * and the second, holding only fragments. */
-	if (skb_has_frags(head)) {
+	if (skb_has_frag_list(head)) {
 		struct sk_buff *clone;
 		int i, plen = 0;
 
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 545c4141b755..8aea3f3f18d7 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -499,7 +499,7 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev,
 	/* If the first fragment is fragmented itself, we split
 	 * it to two chunks: the first with data and paged part
 	 * and the second, holding only fragments. */
-	if (skb_has_frags(head)) {
+	if (skb_has_frag_list(head)) {
 		struct sk_buff *clone;
 		int i, plen = 0;
 
-- 
cgit v1.2.3


From 8b230ed8ec96c933047dd0625cf95f739e4939a6 Mon Sep 17 00:00:00 2001
From: Rasesh Mody <rmody@brocade.com>
Date: Mon, 23 Aug 2010 20:24:12 -0700
Subject: bna: Brocade 10Gb Ethernet device driver

This is patch 1/6 which contains linux driver source for
Brocade's BR1010/BR1020 10Gb CEE capable ethernet adapter.

Signed-off-by: Debashis Dutt <ddutt@brocade.com>
Signed-off-by: Rasesh Mody <rmody@brocade.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS                         |    7 +
 drivers/net/Kconfig                 |   14 +
 drivers/net/Makefile                |    1 +
 drivers/net/bna/Makefile            |   11 +
 drivers/net/bna/bfa_cee.c           |  407 ++++
 drivers/net/bna/bfa_cee.h           |   72 +
 drivers/net/bna/bfa_defs.h          |  243 ++
 drivers/net/bna/bfa_defs_cna.h      |  223 ++
 drivers/net/bna/bfa_defs_mfg_comm.h |  244 ++
 drivers/net/bna/bfa_defs_status.h   |  216 ++
 drivers/net/bna/bfa_ioc.c           | 1839 +++++++++++++++
 drivers/net/bna/bfa_ioc.h           |  343 +++
 drivers/net/bna/bfa_ioc_ct.c        |  391 ++++
 drivers/net/bna/bfa_sm.h            |   88 +
 drivers/net/bna/bfa_wc.h            |   69 +
 drivers/net/bna/bfi.h               |  392 ++++
 drivers/net/bna/bfi_cna.h           |  199 ++
 drivers/net/bna/bfi_ctreg.h         |  637 ++++++
 drivers/net/bna/bfi_ll.h            |  438 ++++
 drivers/net/bna/bna.h               |  654 ++++++
 drivers/net/bna/bna_ctrl.c          | 3626 ++++++++++++++++++++++++++++++
 drivers/net/bna/bna_hw.h            | 1491 +++++++++++++
 drivers/net/bna/bna_txrx.c          | 4209 +++++++++++++++++++++++++++++++++++
 drivers/net/bna/bna_types.h         | 1128 ++++++++++
 drivers/net/bna/bnad.c              | 3270 +++++++++++++++++++++++++++
 drivers/net/bna/bnad.h              |  334 +++
 drivers/net/bna/bnad_ethtool.c      | 1282 +++++++++++
 drivers/net/bna/cna.h               |   81 +
 drivers/net/bna/cna_fwimg.c         |   64 +
 include/linux/pci_ids.h             |    3 +
 30 files changed, 21976 insertions(+)
 create mode 100644 drivers/net/bna/Makefile
 create mode 100644 drivers/net/bna/bfa_cee.c
 create mode 100644 drivers/net/bna/bfa_cee.h
 create mode 100644 drivers/net/bna/bfa_defs.h
 create mode 100644 drivers/net/bna/bfa_defs_cna.h
 create mode 100644 drivers/net/bna/bfa_defs_mfg_comm.h
 create mode 100644 drivers/net/bna/bfa_defs_status.h
 create mode 100644 drivers/net/bna/bfa_ioc.c
 create mode 100644 drivers/net/bna/bfa_ioc.h
 create mode 100644 drivers/net/bna/bfa_ioc_ct.c
 create mode 100644 drivers/net/bna/bfa_sm.h
 create mode 100644 drivers/net/bna/bfa_wc.h
 create mode 100644 drivers/net/bna/bfi.h
 create mode 100644 drivers/net/bna/bfi_cna.h
 create mode 100644 drivers/net/bna/bfi_ctreg.h
 create mode 100644 drivers/net/bna/bfi_ll.h
 create mode 100644 drivers/net/bna/bna.h
 create mode 100644 drivers/net/bna/bna_ctrl.c
 create mode 100644 drivers/net/bna/bna_hw.h
 create mode 100644 drivers/net/bna/bna_txrx.c
 create mode 100644 drivers/net/bna/bna_types.h
 create mode 100644 drivers/net/bna/bnad.c
 create mode 100644 drivers/net/bna/bnad.h
 create mode 100644 drivers/net/bna/bnad_ethtool.c
 create mode 100644 drivers/net/bna/cna.h
 create mode 100644 drivers/net/bna/cna_fwimg.c

(limited to 'include/linux')

diff --git a/MAINTAINERS b/MAINTAINERS
index 43c9efcd8e10..93198c1980a9 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1387,6 +1387,13 @@ L:	linux-scsi@vger.kernel.org
 S:	Supported
 F:	drivers/scsi/bfa/
 
+BROCADE BNA 10 GIGABIT ETHERNET DRIVER
+M:	Rasesh Mody <rmody@brocade.com>
+M:	Debashis Dutt <ddutt@brocade.com>
+L:	netdev@vger.kernel.org
+S:	Supported
+F:	drivers/net/bna/
+
 BSG (block layer generic sg v4 driver)
 M:	FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
 L:	linux-scsi@vger.kernel.org
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 85485287192a..53c4810b119e 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2869,6 +2869,20 @@ config QLGE
 	  To compile this driver as a module, choose M here: the module
 	  will be called qlge.
 
+config BNA
+        tristate "Brocade 1010/1020 10Gb Ethernet Driver support"
+        depends on PCI
+        ---help---
+          This driver supports Brocade 1010/1020 10Gb CEE capable Ethernet
+          cards.
+          To compile this driver as a module, choose M here: the module
+          will be called bna.
+
+          For general information and support, go to the Brocade support
+          website at:
+
+          <http://support.brocade.com>
+
 source "drivers/net/sfc/Kconfig"
 
 source "drivers/net/benet/Kconfig"
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index f34b3baf7ee6..18a277709a2a 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -34,6 +34,7 @@ obj-$(CONFIG_ENIC) += enic/
 obj-$(CONFIG_JME) += jme.o
 obj-$(CONFIG_BE2NET) += benet/
 obj-$(CONFIG_VMXNET3) += vmxnet3/
+obj-$(CONFIG_BNA) += bna/
 
 gianfar_driver-objs := gianfar.o \
 		gianfar_ethtool.o \
diff --git a/drivers/net/bna/Makefile b/drivers/net/bna/Makefile
new file mode 100644
index 000000000000..a5d604de7fea
--- /dev/null
+++ b/drivers/net/bna/Makefile
@@ -0,0 +1,11 @@
+#
+# Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+# All rights reserved.
+#
+
+obj-$(CONFIG_BNA) += bna.o
+
+bna-objs := bnad.o bnad_ethtool.o bna_ctrl.o bna_txrx.o
+bna-objs += bfa_ioc.o bfa_ioc_ct.o bfa_cee.o cna_fwimg.o
+
+EXTRA_CFLAGS := -Idrivers/net/bna
diff --git a/drivers/net/bna/bfa_cee.c b/drivers/net/bna/bfa_cee.c
new file mode 100644
index 000000000000..1545fc9720f8
--- /dev/null
+++ b/drivers/net/bna/bfa_cee.c
@@ -0,0 +1,407 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#include "bfa_defs_cna.h"
+#include "cna.h"
+#include "bfa_cee.h"
+#include "bfi_cna.h"
+#include "bfa_ioc.h"
+
+#define bfa_ioc_portid(__ioc) ((__ioc)->port_id)
+#define bfa_lpuid(__arg) bfa_ioc_portid(&(__arg)->ioc)
+
+static void bfa_cee_format_lldp_cfg(struct bfa_cee_lldp_cfg *lldp_cfg);
+static void bfa_cee_format_cee_cfg(void *buffer);
+
+static void
+bfa_cee_format_cee_cfg(void *buffer)
+{
+	struct bfa_cee_attr *cee_cfg = buffer;
+	bfa_cee_format_lldp_cfg(&cee_cfg->lldp_remote);
+}
+
+static void
+bfa_cee_stats_swap(struct bfa_cee_stats *stats)
+{
+	u32 *buffer = (u32 *)stats;
+	int i;
+
+	for (i = 0; i < (sizeof(struct bfa_cee_stats) / sizeof(u32));
+		i++) {
+		buffer[i] = ntohl(buffer[i]);
+	}
+}
+
+static void
+bfa_cee_format_lldp_cfg(struct bfa_cee_lldp_cfg *lldp_cfg)
+{
+	lldp_cfg->time_to_live =
+			ntohs(lldp_cfg->time_to_live);
+	lldp_cfg->enabled_system_cap =
+			ntohs(lldp_cfg->enabled_system_cap);
+}
+
+/**
+ * bfa_cee_attr_meminfo()
+ *
+ * @brief Returns the size of the DMA memory needed by CEE attributes
+ *
+ * @param[in] void
+ *
+ * @return Size of DMA region
+ */
+static u32
+bfa_cee_attr_meminfo(void)
+{
+	return roundup(sizeof(struct bfa_cee_attr), BFA_DMA_ALIGN_SZ);
+}
+/**
+ * bfa_cee_stats_meminfo()
+ *
+ * @brief Returns the size of the DMA memory needed by CEE stats
+ *
+ * @param[in] void
+ *
+ * @return Size of DMA region
+ */
+static u32
+bfa_cee_stats_meminfo(void)
+{
+	return roundup(sizeof(struct bfa_cee_stats), BFA_DMA_ALIGN_SZ);
+}
+
+/**
+ * bfa_cee_get_attr_isr()
+ *
+ * @brief CEE ISR for get-attributes responses from f/w
+ *
+ * @param[in] cee - Pointer to the CEE module
+ *            status - Return status from the f/w
+ *
+ * @return void
+ */
+static void
+bfa_cee_get_attr_isr(struct bfa_cee *cee, enum bfa_status status)
+{
+	cee->get_attr_status = status;
+	if (status == BFA_STATUS_OK) {
+		memcpy(cee->attr, cee->attr_dma.kva,
+		    sizeof(struct bfa_cee_attr));
+		bfa_cee_format_cee_cfg(cee->attr);
+	}
+	cee->get_attr_pending = false;
+	if (cee->cbfn.get_attr_cbfn)
+		cee->cbfn.get_attr_cbfn(cee->cbfn.get_attr_cbarg, status);
+}
+
+/**
+ * bfa_cee_get_attr_isr()
+ *
+ * @brief CEE ISR for get-stats responses from f/w
+ *
+ * @param[in] cee - Pointer to the CEE module
+ *            status - Return status from the f/w
+ *
+ * @return void
+ */
+static void
+bfa_cee_get_stats_isr(struct bfa_cee *cee, enum bfa_status status)
+{
+	cee->get_stats_status = status;
+	if (status == BFA_STATUS_OK) {
+		memcpy(cee->stats, cee->stats_dma.kva,
+			sizeof(struct bfa_cee_stats));
+		bfa_cee_stats_swap(cee->stats);
+	}
+	cee->get_stats_pending = false;
+	if (cee->cbfn.get_stats_cbfn)
+		cee->cbfn.get_stats_cbfn(cee->cbfn.get_stats_cbarg, status);
+}
+
+/**
+ * bfa_cee_get_attr_isr()
+ *
+ * @brief CEE ISR for reset-stats responses from f/w
+ *
+ * @param[in] cee - Pointer to the CEE module
+ *            status - Return status from the f/w
+ *
+ * @return void
+ */
+static void
+bfa_cee_reset_stats_isr(struct bfa_cee *cee, enum bfa_status status)
+{
+	cee->reset_stats_status = status;
+	cee->reset_stats_pending = false;
+	if (cee->cbfn.reset_stats_cbfn)
+		cee->cbfn.reset_stats_cbfn(cee->cbfn.reset_stats_cbarg, status);
+}
+/**
+ * bfa_cee_meminfo()
+ *
+ * @brief Returns the size of the DMA memory needed by CEE module
+ *
+ * @param[in] void
+ *
+ * @return Size of DMA region
+ */
+u32
+bfa_cee_meminfo(void)
+{
+	return bfa_cee_attr_meminfo() + bfa_cee_stats_meminfo();
+}
+
+/**
+ * bfa_cee_mem_claim()
+ *
+ * @brief Initialized CEE DMA Memory
+ *
+ * @param[in] cee CEE module pointer
+ *	      dma_kva Kernel Virtual Address of CEE DMA Memory
+ *	      dma_pa  Physical Address of CEE DMA Memory
+ *
+ * @return void
+ */
+void
+bfa_cee_mem_claim(struct bfa_cee *cee, u8 *dma_kva, u64 dma_pa)
+{
+	cee->attr_dma.kva = dma_kva;
+	cee->attr_dma.pa = dma_pa;
+	cee->stats_dma.kva = dma_kva + bfa_cee_attr_meminfo();
+	cee->stats_dma.pa = dma_pa + bfa_cee_attr_meminfo();
+	cee->attr = (struct bfa_cee_attr *) dma_kva;
+	cee->stats = (struct bfa_cee_stats *)
+		(dma_kva + bfa_cee_attr_meminfo());
+}
+
+/**
+ * bfa_cee_get_attr()
+ *
+ * @brief
+ *   Send the request to the f/w to fetch CEE attributes.
+ *
+ * @param[in] Pointer to the CEE module data structure.
+ *
+ * @return Status
+ */
+
+enum bfa_status
+bfa_cee_get_attr(struct bfa_cee *cee, struct bfa_cee_attr *attr,
+		     bfa_cee_get_attr_cbfn_t cbfn, void *cbarg)
+{
+	struct bfi_cee_get_req *cmd;
+
+	BUG_ON(!((cee != NULL) && (cee->ioc != NULL)));
+	if (!bfa_ioc_is_operational(cee->ioc))
+		return BFA_STATUS_IOC_FAILURE;
+	if (cee->get_attr_pending == true)
+		return 	BFA_STATUS_DEVBUSY;
+	cee->get_attr_pending = true;
+	cmd = (struct bfi_cee_get_req *) cee->get_cfg_mb.msg;
+	cee->attr = attr;
+	cee->cbfn.get_attr_cbfn = cbfn;
+	cee->cbfn.get_attr_cbarg = cbarg;
+	bfi_h2i_set(cmd->mh, BFI_MC_CEE, BFI_CEE_H2I_GET_CFG_REQ,
+	    bfa_ioc_portid(cee->ioc));
+	bfa_dma_be_addr_set(cmd->dma_addr, cee->attr_dma.pa);
+	bfa_ioc_mbox_queue(cee->ioc, &cee->get_cfg_mb);
+
+	return BFA_STATUS_OK;
+}
+
+/**
+ * bfa_cee_get_stats()
+ *
+ * @brief
+ *   Send the request to the f/w to fetch CEE statistics.
+ *
+ * @param[in] Pointer to the CEE module data structure.
+ *
+ * @return Status
+ */
+
+enum bfa_status
+bfa_cee_get_stats(struct bfa_cee *cee, struct bfa_cee_stats *stats,
+		      bfa_cee_get_stats_cbfn_t cbfn, void *cbarg)
+{
+	struct bfi_cee_get_req *cmd;
+
+	BUG_ON(!((cee != NULL) && (cee->ioc != NULL)));
+
+	if (!bfa_ioc_is_operational(cee->ioc))
+		return BFA_STATUS_IOC_FAILURE;
+	if (cee->get_stats_pending == true)
+		return 	BFA_STATUS_DEVBUSY;
+	cee->get_stats_pending = true;
+	cmd = (struct bfi_cee_get_req *) cee->get_stats_mb.msg;
+	cee->stats = stats;
+	cee->cbfn.get_stats_cbfn = cbfn;
+	cee->cbfn.get_stats_cbarg = cbarg;
+	bfi_h2i_set(cmd->mh, BFI_MC_CEE, BFI_CEE_H2I_GET_STATS_REQ,
+	    bfa_ioc_portid(cee->ioc));
+	bfa_dma_be_addr_set(cmd->dma_addr, cee->stats_dma.pa);
+	bfa_ioc_mbox_queue(cee->ioc, &cee->get_stats_mb);
+
+	return BFA_STATUS_OK;
+}
+
+/**
+ * bfa_cee_reset_stats()
+ *
+ * @brief Clears CEE Stats in the f/w.
+ *
+ * @param[in] Pointer to the CEE module data structure.
+ *
+ * @return Status
+ */
+
+enum bfa_status
+bfa_cee_reset_stats(struct bfa_cee *cee, bfa_cee_reset_stats_cbfn_t cbfn,
+			void *cbarg)
+{
+	struct bfi_cee_reset_stats *cmd;
+
+	BUG_ON(!((cee != NULL) && (cee->ioc != NULL)));
+	if (!bfa_ioc_is_operational(cee->ioc))
+		return BFA_STATUS_IOC_FAILURE;
+	if (cee->reset_stats_pending == true)
+		return 	BFA_STATUS_DEVBUSY;
+	cee->reset_stats_pending = true;
+	cmd = (struct bfi_cee_reset_stats *) cee->reset_stats_mb.msg;
+	cee->cbfn.reset_stats_cbfn = cbfn;
+	cee->cbfn.reset_stats_cbarg = cbarg;
+	bfi_h2i_set(cmd->mh, BFI_MC_CEE, BFI_CEE_H2I_RESET_STATS,
+	    bfa_ioc_portid(cee->ioc));
+	bfa_ioc_mbox_queue(cee->ioc, &cee->reset_stats_mb);
+	return BFA_STATUS_OK;
+}
+
+/**
+ * bfa_cee_isrs()
+ *
+ * @brief Handles Mail-box interrupts for CEE module.
+ *
+ * @param[in] Pointer to the CEE module data structure.
+ *
+ * @return void
+ */
+
+void
+bfa_cee_isr(void *cbarg, struct bfi_mbmsg *m)
+{
+	union bfi_cee_i2h_msg_u *msg;
+	struct bfi_cee_get_rsp *get_rsp;
+	struct bfa_cee *cee = (struct bfa_cee *) cbarg;
+	msg = (union bfi_cee_i2h_msg_u *) m;
+	get_rsp = (struct bfi_cee_get_rsp *) m;
+	switch (msg->mh.msg_id) {
+	case BFI_CEE_I2H_GET_CFG_RSP:
+		bfa_cee_get_attr_isr(cee, get_rsp->cmd_status);
+		break;
+	case BFI_CEE_I2H_GET_STATS_RSP:
+		bfa_cee_get_stats_isr(cee, get_rsp->cmd_status);
+		break;
+	case BFI_CEE_I2H_RESET_STATS_RSP:
+		bfa_cee_reset_stats_isr(cee, get_rsp->cmd_status);
+		break;
+	default:
+		BUG_ON(1);
+	}
+}
+
+/**
+ * bfa_cee_hbfail()
+ *
+ * @brief CEE module heart-beat failure handler.
+ *
+ * @param[in] Pointer to the CEE module data structure.
+ *
+ * @return void
+ */
+
+void
+bfa_cee_hbfail(void *arg)
+{
+	struct bfa_cee *cee;
+	cee = (struct bfa_cee *) arg;
+
+	if (cee->get_attr_pending == true) {
+		cee->get_attr_status = BFA_STATUS_FAILED;
+		cee->get_attr_pending  = false;
+		if (cee->cbfn.get_attr_cbfn) {
+			cee->cbfn.get_attr_cbfn(cee->cbfn.get_attr_cbarg,
+			    BFA_STATUS_FAILED);
+		}
+	}
+	if (cee->get_stats_pending == true) {
+		cee->get_stats_status = BFA_STATUS_FAILED;
+		cee->get_stats_pending  = false;
+		if (cee->cbfn.get_stats_cbfn) {
+			cee->cbfn.get_stats_cbfn(cee->cbfn.get_stats_cbarg,
+			    BFA_STATUS_FAILED);
+		}
+	}
+	if (cee->reset_stats_pending == true) {
+		cee->reset_stats_status = BFA_STATUS_FAILED;
+		cee->reset_stats_pending  = false;
+		if (cee->cbfn.reset_stats_cbfn) {
+			cee->cbfn.reset_stats_cbfn(cee->cbfn.reset_stats_cbarg,
+			    BFA_STATUS_FAILED);
+		}
+	}
+}
+
+/**
+ * bfa_cee_attach()
+ *
+ * @brief CEE module-attach API
+ *
+ * @param[in] cee - Pointer to the CEE module data structure
+ *            ioc - Pointer to the ioc module data structure
+ *            dev - Pointer to the device driver module data structure
+ *                  The device driver specific mbox ISR functions have
+ *                  this pointer as one of the parameters.
+ *
+ * @return void
+ */
+void
+bfa_cee_attach(struct bfa_cee *cee, struct bfa_ioc *ioc,
+		void *dev)
+{
+	BUG_ON(!(cee != NULL));
+	cee->dev = dev;
+	cee->ioc = ioc;
+
+	bfa_ioc_mbox_regisr(cee->ioc, BFI_MC_CEE, bfa_cee_isr, cee);
+	bfa_ioc_hbfail_init(&cee->hbfail, bfa_cee_hbfail, cee);
+	bfa_ioc_hbfail_register(cee->ioc, &cee->hbfail);
+}
+
+/**
+ * bfa_cee_detach()
+ *
+ * @brief CEE module-detach API
+ *
+ * @param[in] cee - Pointer to the CEE module data structure
+ *
+ * @return void
+ */
+void
+bfa_cee_detach(struct bfa_cee *cee)
+{
+}
diff --git a/drivers/net/bna/bfa_cee.h b/drivers/net/bna/bfa_cee.h
new file mode 100644
index 000000000000..1208cadeceed
--- /dev/null
+++ b/drivers/net/bna/bfa_cee.h
@@ -0,0 +1,72 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#ifndef __BFA_CEE_H__
+#define __BFA_CEE_H__
+
+#include "bfa_defs_cna.h"
+#include "bfa_ioc.h"
+
+typedef void (*bfa_cee_get_attr_cbfn_t) (void *dev, enum bfa_status status);
+typedef void (*bfa_cee_get_stats_cbfn_t) (void *dev, enum bfa_status status);
+typedef void (*bfa_cee_reset_stats_cbfn_t) (void *dev, enum bfa_status status);
+typedef void (*bfa_cee_hbfail_cbfn_t) (void *dev, enum bfa_status status);
+
+struct bfa_cee_cbfn {
+	bfa_cee_get_attr_cbfn_t    get_attr_cbfn;
+	void *get_attr_cbarg;
+	bfa_cee_get_stats_cbfn_t   get_stats_cbfn;
+	void *get_stats_cbarg;
+	bfa_cee_reset_stats_cbfn_t reset_stats_cbfn;
+	void *reset_stats_cbarg;
+};
+
+struct bfa_cee {
+	void *dev;
+	bool get_attr_pending;
+	bool get_stats_pending;
+	bool reset_stats_pending;
+	enum bfa_status get_attr_status;
+	enum bfa_status get_stats_status;
+	enum bfa_status reset_stats_status;
+	struct bfa_cee_cbfn cbfn;
+	struct bfa_ioc_hbfail_notify hbfail;
+	struct bfa_cee_attr *attr;
+	struct bfa_cee_stats *stats;
+	struct bfa_dma attr_dma;
+	struct bfa_dma stats_dma;
+	struct bfa_ioc *ioc;
+	struct bfa_mbox_cmd get_cfg_mb;
+	struct bfa_mbox_cmd get_stats_mb;
+	struct bfa_mbox_cmd reset_stats_mb;
+};
+
+u32 bfa_cee_meminfo(void);
+void bfa_cee_mem_claim(struct bfa_cee *cee, u8 *dma_kva,
+	u64 dma_pa);
+void bfa_cee_attach(struct bfa_cee *cee, struct bfa_ioc *ioc, void *dev);
+void bfa_cee_detach(struct bfa_cee *cee);
+enum bfa_status bfa_cee_get_attr(struct bfa_cee *cee,
+	struct bfa_cee_attr *attr, bfa_cee_get_attr_cbfn_t cbfn, void *cbarg);
+enum bfa_status bfa_cee_get_stats(struct bfa_cee *cee,
+	struct bfa_cee_stats *stats, bfa_cee_get_stats_cbfn_t cbfn,
+	void *cbarg);
+enum bfa_status bfa_cee_reset_stats(struct bfa_cee *cee,
+	bfa_cee_reset_stats_cbfn_t cbfn, void *cbarg);
+
+#endif /* __BFA_CEE_H__ */
diff --git a/drivers/net/bna/bfa_defs.h b/drivers/net/bna/bfa_defs.h
new file mode 100644
index 000000000000..29c1b8de2c2d
--- /dev/null
+++ b/drivers/net/bna/bfa_defs.h
@@ -0,0 +1,243 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#ifndef __BFA_DEFS_H__
+#define __BFA_DEFS_H__
+
+#include "cna.h"
+#include "bfa_defs_status.h"
+#include "bfa_defs_mfg_comm.h"
+
+#define BFA_STRING_32	32
+#define BFA_VERSION_LEN 64
+
+/**
+ * ---------------------- adapter definitions ------------
+ */
+
+/**
+ * BFA adapter level attributes.
+ */
+enum {
+	BFA_ADAPTER_SERIAL_NUM_LEN = STRSZ(BFA_MFG_SERIALNUM_SIZE),
+					/*
+					 *!< adapter serial num length
+					 */
+	BFA_ADAPTER_MODEL_NAME_LEN  = 16,  /*!< model name length */
+	BFA_ADAPTER_MODEL_DESCR_LEN = 128, /*!< model description length */
+	BFA_ADAPTER_MFG_NAME_LEN    = 8,   /*!< manufacturer name length */
+	BFA_ADAPTER_SYM_NAME_LEN    = 64,  /*!< adapter symbolic name length */
+	BFA_ADAPTER_OS_TYPE_LEN	    = 64,  /*!< adapter os type length */
+};
+
+struct bfa_adapter_attr {
+	char		manufacturer[BFA_ADAPTER_MFG_NAME_LEN];
+	char		serial_num[BFA_ADAPTER_SERIAL_NUM_LEN];
+	u32	card_type;
+	char		model[BFA_ADAPTER_MODEL_NAME_LEN];
+	char		model_descr[BFA_ADAPTER_MODEL_DESCR_LEN];
+	u64		pwwn;
+	char		node_symname[FC_SYMNAME_MAX];
+	char		hw_ver[BFA_VERSION_LEN];
+	char		fw_ver[BFA_VERSION_LEN];
+	char		optrom_ver[BFA_VERSION_LEN];
+	char		os_type[BFA_ADAPTER_OS_TYPE_LEN];
+	struct bfa_mfg_vpd vpd;
+	struct mac mac;
+
+	u8		nports;
+	u8		max_speed;
+	u8		prototype;
+	char	        asic_rev;
+
+	u8		pcie_gen;
+	u8		pcie_lanes_orig;
+	u8		pcie_lanes;
+	u8	        cna_capable;
+
+	u8		is_mezz;
+	u8		trunk_capable;
+};
+
+/**
+ * ---------------------- IOC definitions ------------
+ */
+
+enum {
+	BFA_IOC_DRIVER_LEN	= 16,
+	BFA_IOC_CHIP_REV_LEN 	= 8,
+};
+
+/**
+ * Driver and firmware versions.
+ */
+struct bfa_ioc_driver_attr {
+	char		driver[BFA_IOC_DRIVER_LEN];	/*!< driver name */
+	char		driver_ver[BFA_VERSION_LEN];	/*!< driver version */
+	char		fw_ver[BFA_VERSION_LEN];	/*!< firmware version */
+	char		bios_ver[BFA_VERSION_LEN];	/*!< bios version */
+	char		efi_ver[BFA_VERSION_LEN];	/*!< EFI version */
+	char		ob_ver[BFA_VERSION_LEN];	/*!< openboot version */
+};
+
+/**
+ * IOC PCI device attributes
+ */
+struct bfa_ioc_pci_attr {
+	u16	vendor_id;	/*!< PCI vendor ID */
+	u16	device_id;	/*!< PCI device ID */
+	u16	ssid;		/*!< subsystem ID */
+	u16	ssvid;		/*!< subsystem vendor ID */
+	u32	pcifn;		/*!< PCI device function */
+	u32	rsvd;		/* padding */
+	char		chip_rev[BFA_IOC_CHIP_REV_LEN];	 /*!< chip revision */
+};
+
+/**
+ * IOC states
+ */
+enum bfa_ioc_state {
+	BFA_IOC_RESET		= 1,	/*!< IOC is in reset state */
+	BFA_IOC_SEMWAIT		= 2,	/*!< Waiting for IOC h/w semaphore */
+	BFA_IOC_HWINIT		= 3,	/*!< IOC h/w is being initialized */
+	BFA_IOC_GETATTR		= 4,	/*!< IOC is being configured */
+	BFA_IOC_OPERATIONAL	= 5,	/*!< IOC is operational */
+	BFA_IOC_INITFAIL	= 6,	/*!< IOC hardware failure */
+	BFA_IOC_HBFAIL		= 7,	/*!< IOC heart-beat failure */
+	BFA_IOC_DISABLING	= 8,	/*!< IOC is being disabled */
+	BFA_IOC_DISABLED	= 9,	/*!< IOC is disabled */
+	BFA_IOC_FWMISMATCH	= 10,	/*!< IOC f/w different from drivers */
+};
+
+/**
+ * IOC firmware stats
+ */
+struct bfa_fw_ioc_stats {
+	u32	enable_reqs;
+	u32	disable_reqs;
+	u32	get_attr_reqs;
+	u32	dbg_sync;
+	u32	dbg_dump;
+	u32	unknown_reqs;
+};
+
+/**
+ * IOC driver stats
+ */
+struct bfa_ioc_drv_stats {
+	u32	ioc_isrs;
+	u32	ioc_enables;
+	u32	ioc_disables;
+	u32	ioc_hbfails;
+	u32	ioc_boots;
+	u32	stats_tmos;
+	u32	hb_count;
+	u32	disable_reqs;
+	u32	enable_reqs;
+	u32	disable_replies;
+	u32	enable_replies;
+};
+
+/**
+ * IOC statistics
+ */
+struct bfa_ioc_stats {
+	struct bfa_ioc_drv_stats drv_stats; /*!< driver IOC stats */
+	struct bfa_fw_ioc_stats fw_stats;  /*!< firmware IOC stats */
+};
+
+enum bfa_ioc_type {
+	BFA_IOC_TYPE_FC		= 1,
+	BFA_IOC_TYPE_FCoE	= 2,
+	BFA_IOC_TYPE_LL		= 3,
+};
+
+/**
+ * IOC attributes returned in queries
+ */
+struct bfa_ioc_attr {
+	enum bfa_ioc_type ioc_type;
+	enum bfa_ioc_state 		state;		/*!< IOC state      */
+	struct bfa_adapter_attr adapter_attr;	/*!< HBA attributes */
+	struct bfa_ioc_driver_attr driver_attr;	/*!< driver attr    */
+	struct bfa_ioc_pci_attr pci_attr;
+	u8				port_id;	/*!< port number    */
+	u8				rsvd[7];	/*!< 64bit align    */
+};
+
+/**
+ * ---------------------- mfg definitions ------------
+ */
+
+/**
+ * Checksum size
+ */
+#define BFA_MFG_CHKSUM_SIZE			16
+
+#define BFA_MFG_PARTNUM_SIZE			14
+#define BFA_MFG_SUPPLIER_ID_SIZE		10
+#define BFA_MFG_SUPPLIER_PARTNUM_SIZE		20
+#define BFA_MFG_SUPPLIER_SERIALNUM_SIZE		20
+#define BFA_MFG_SUPPLIER_REVISION_SIZE		4
+
+#pragma pack(1)
+
+/**
+ * @brief BFA adapter manufacturing block definition.
+ *
+ * All numerical fields are in big-endian format.
+ */
+struct bfa_mfg_block {
+	u8		version;	/*!< manufacturing block version */
+	u8		mfg_sig[3];	/*!< characters 'M', 'F', 'G' */
+	u16	mfgsize;	/*!< mfg block size */
+	u16	u16_chksum;	/*!< old u16 checksum */
+	char		brcd_serialnum[STRSZ(BFA_MFG_SERIALNUM_SIZE)];
+	char		brcd_partnum[STRSZ(BFA_MFG_PARTNUM_SIZE)];
+	u8		mfg_day;	/*!< manufacturing day */
+	u8		mfg_month;	/*!< manufacturing month */
+	u16	mfg_year;	/*!< manufacturing year */
+	u64		mfg_wwn;	/*!< wwn base for this adapter */
+	u8		num_wwn;	/*!< number of wwns assigned */
+	u8		mfg_speeds;	/*!< speeds allowed for this adapter */
+	u8		rsv[2];
+	char		supplier_id[STRSZ(BFA_MFG_SUPPLIER_ID_SIZE)];
+	char		supplier_partnum[STRSZ(BFA_MFG_SUPPLIER_PARTNUM_SIZE)];
+	char
+		supplier_serialnum[STRSZ(BFA_MFG_SUPPLIER_SERIALNUM_SIZE)];
+	char
+		supplier_revision[STRSZ(BFA_MFG_SUPPLIER_REVISION_SIZE)];
+	mac_t		mfg_mac;	/*!< mac address */
+	u8		num_mac;	/*!< number of mac addresses */
+	u8		rsv2;
+	u32	mfg_type;	/*!< card type */
+	u8		rsv3[108];
+	u8		md5_chksum[BFA_MFG_CHKSUM_SIZE]; /*!< md5 checksum */
+};
+
+#pragma pack()
+
+/**
+ * ---------------------- pci definitions ------------
+ */
+
+#define bfa_asic_id_ct(devid)			\
+	((devid) == PCI_DEVICE_ID_BROCADE_CT ||	\
+	(devid) == PCI_DEVICE_ID_BROCADE_CT_FC)
+
+#endif /* __BFA_DEFS_H__ */
diff --git a/drivers/net/bna/bfa_defs_cna.h b/drivers/net/bna/bfa_defs_cna.h
new file mode 100644
index 000000000000..7e0a9187bdd5
--- /dev/null
+++ b/drivers/net/bna/bfa_defs_cna.h
@@ -0,0 +1,223 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#ifndef __BFA_DEFS_CNA_H__
+#define __BFA_DEFS_CNA_H__
+
+#include "bfa_defs.h"
+
+/**
+ * @brief
+ * FC physical port statistics.
+ */
+struct bfa_port_fc_stats {
+	u64	secs_reset;	/*!< Seconds since stats is reset */
+	u64	tx_frames;	/*!< Tx frames			*/
+	u64	tx_words;	/*!< Tx words			*/
+	u64	tx_lip;		/*!< Tx LIP			*/
+	u64	tx_nos;		/*!< Tx NOS			*/
+	u64	tx_ols;		/*!< Tx OLS			*/
+	u64	tx_lr;		/*!< Tx LR			*/
+	u64	tx_lrr;		/*!< Tx LRR			*/
+	u64	rx_frames;	/*!< Rx frames			*/
+	u64	rx_words;	/*!< Rx words			*/
+	u64	lip_count;	/*!< Rx LIP			*/
+	u64	nos_count;	/*!< Rx NOS			*/
+	u64	ols_count;	/*!< Rx OLS			*/
+	u64	lr_count;	/*!< Rx LR			*/
+	u64	lrr_count;	/*!< Rx LRR			*/
+	u64	invalid_crcs;	/*!< Rx CRC err frames		*/
+	u64	invalid_crc_gd_eof; /*!< Rx CRC err good EOF frames */
+	u64	undersized_frm; /*!< Rx undersized frames	*/
+	u64	oversized_frm;	/*!< Rx oversized frames	*/
+	u64	bad_eof_frm;	/*!< Rx frames with bad EOF	*/
+	u64	error_frames;	/*!< Errored frames		*/
+	u64	dropped_frames;	/*!< Dropped frames		*/
+	u64	link_failures;	/*!< Link Failure (LF) count	*/
+	u64	loss_of_syncs;	/*!< Loss of sync count		*/
+	u64	loss_of_signals; /*!< Loss of signal count	*/
+	u64	primseq_errs;	/*!< Primitive sequence protocol err. */
+	u64	bad_os_count;	/*!< Invalid ordered sets	*/
+	u64	err_enc_out;	/*!< Encoding err nonframe_8b10b */
+	u64	err_enc;	/*!< Encoding err frame_8b10b	*/
+};
+
+/**
+ * @brief
+ * Eth Physical Port statistics.
+ */
+struct bfa_port_eth_stats {
+	u64	secs_reset;	/*!< Seconds since stats is reset */
+	u64	frame_64;	/*!< Frames 64 bytes		*/
+	u64	frame_65_127;	/*!< Frames 65-127 bytes	*/
+	u64	frame_128_255;	/*!< Frames 128-255 bytes	*/
+	u64	frame_256_511;	/*!< Frames 256-511 bytes	*/
+	u64	frame_512_1023;	/*!< Frames 512-1023 bytes	*/
+	u64	frame_1024_1518; /*!< Frames 1024-1518 bytes	*/
+	u64	frame_1519_1522; /*!< Frames 1519-1522 bytes	*/
+	u64	tx_bytes;	/*!< Tx bytes			*/
+	u64	tx_packets;	 /*!< Tx packets		*/
+	u64	tx_mcast_packets; /*!< Tx multicast packets	*/
+	u64	tx_bcast_packets; /*!< Tx broadcast packets	*/
+	u64	tx_control_frame; /*!< Tx control frame		*/
+	u64	tx_drop;	/*!< Tx drops			*/
+	u64	tx_jabber;	/*!< Tx jabber			*/
+	u64	tx_fcs_error;	/*!< Tx FCS errors		*/
+	u64	tx_fragments;	/*!< Tx fragments		*/
+	u64	rx_bytes;	/*!< Rx bytes			*/
+	u64	rx_packets;	/*!< Rx packets			*/
+	u64	rx_mcast_packets; /*!< Rx multicast packets	*/
+	u64	rx_bcast_packets; /*!< Rx broadcast packets	*/
+	u64	rx_control_frames; /*!< Rx control frames	*/
+	u64	rx_unknown_opcode; /*!< Rx unknown opcode	*/
+	u64	rx_drop;	/*!< Rx drops			*/
+	u64	rx_jabber;	/*!< Rx jabber			*/
+	u64	rx_fcs_error;	/*!< Rx FCS errors		*/
+	u64	rx_alignment_error; /*!< Rx alignment errors	*/
+	u64	rx_frame_length_error; /*!< Rx frame len errors	*/
+	u64	rx_code_error;	/*!< Rx code errors		*/
+	u64	rx_fragments;	/*!< Rx fragments		*/
+	u64	rx_pause;	/*!< Rx pause			*/
+	u64	rx_zero_pause;	/*!< Rx zero pause		*/
+	u64	tx_pause;	/*!< Tx pause			*/
+	u64	tx_zero_pause;	/*!< Tx zero pause		*/
+	u64	rx_fcoe_pause;	/*!< Rx FCoE pause		*/
+	u64	rx_fcoe_zero_pause; /*!< Rx FCoE zero pause	*/
+	u64	tx_fcoe_pause;	/*!< Tx FCoE pause		*/
+	u64	tx_fcoe_zero_pause; /*!< Tx FCoE zero pause	*/
+};
+
+/**
+ * @brief
+ *		Port statistics.
+ */
+union bfa_port_stats_u {
+	struct bfa_port_fc_stats fc;
+	struct bfa_port_eth_stats eth;
+};
+
+#pragma pack(1)
+
+#define BFA_CEE_LLDP_MAX_STRING_LEN (128)
+#define BFA_CEE_DCBX_MAX_PRIORITY	(8)
+#define BFA_CEE_DCBX_MAX_PGID		(8)
+
+#define BFA_CEE_LLDP_SYS_CAP_OTHER	0x0001
+#define BFA_CEE_LLDP_SYS_CAP_REPEATER	0x0002
+#define BFA_CEE_LLDP_SYS_CAP_MAC_BRIDGE	0x0004
+#define BFA_CEE_LLDP_SYS_CAP_WLAN_AP	0x0008
+#define BFA_CEE_LLDP_SYS_CAP_ROUTER	0x0010
+#define BFA_CEE_LLDP_SYS_CAP_TELEPHONE	0x0020
+#define BFA_CEE_LLDP_SYS_CAP_DOCSIS_CD	0x0040
+#define BFA_CEE_LLDP_SYS_CAP_STATION	0x0080
+#define BFA_CEE_LLDP_SYS_CAP_CVLAN	0x0100
+#define BFA_CEE_LLDP_SYS_CAP_SVLAN	0x0200
+#define BFA_CEE_LLDP_SYS_CAP_TPMR	0x0400
+
+/* LLDP string type */
+struct bfa_cee_lldp_str {
+	u8 sub_type;
+	u8 len;
+	u8 rsvd[2];
+	u8 value[BFA_CEE_LLDP_MAX_STRING_LEN];
+};
+
+/* LLDP paramters */
+struct bfa_cee_lldp_cfg {
+	struct bfa_cee_lldp_str chassis_id;
+	struct bfa_cee_lldp_str port_id;
+	struct bfa_cee_lldp_str port_desc;
+	struct bfa_cee_lldp_str sys_name;
+	struct bfa_cee_lldp_str sys_desc;
+	struct bfa_cee_lldp_str mgmt_addr;
+	u16 time_to_live;
+	u16 enabled_system_cap;
+};
+
+enum bfa_cee_dcbx_version {
+	DCBX_PROTOCOL_PRECEE	= 1,
+	DCBX_PROTOCOL_CEE	= 2,
+};
+
+enum bfa_cee_lls {
+	/* LLS is down because the TLV not sent by the peer */
+	CEE_LLS_DOWN_NO_TLV = 0,
+	/* LLS is down as advertised by the peer */
+	CEE_LLS_DOWN	= 1,
+	CEE_LLS_UP	= 2,
+};
+
+/* CEE/DCBX parameters */
+struct bfa_cee_dcbx_cfg {
+	u8 pgid[BFA_CEE_DCBX_MAX_PRIORITY];
+	u8 pg_percentage[BFA_CEE_DCBX_MAX_PGID];
+	u8 pfc_primap; /* bitmap of priorties with PFC enabled */
+	u8 fcoe_primap; /* bitmap of priorities used for FcoE traffic */
+	u8 iscsi_primap; /* bitmap of priorities used for iSCSI traffic */
+	u8 dcbx_version; /* operating version:CEE or preCEE */
+	u8 lls_fcoe; /* FCoE Logical Link Status */
+	u8 lls_lan; /* LAN Logical Link Status */
+	u8 rsvd[2];
+};
+
+/* CEE status */
+/* Making this to tri-state for the benefit of port list command */
+enum bfa_cee_status {
+	CEE_UP = 0,
+	CEE_PHY_UP = 1,
+	CEE_LOOPBACK = 2,
+	CEE_PHY_DOWN = 3,
+};
+
+/* CEE Query */
+struct bfa_cee_attr {
+	u8	cee_status;
+	u8 error_reason;
+	struct bfa_cee_lldp_cfg lldp_remote;
+	struct bfa_cee_dcbx_cfg dcbx_remote;
+	mac_t src_mac;
+	u8 link_speed;
+	u8 nw_priority;
+	u8 filler[2];
+};
+
+/* LLDP/DCBX/CEE Statistics */
+struct bfa_cee_stats {
+	u32	lldp_tx_frames;		/*!< LLDP Tx Frames */
+	u32	lldp_rx_frames;		/*!< LLDP Rx Frames */
+	u32	lldp_rx_frames_invalid;	/*!< LLDP Rx Frames invalid */
+	u32	lldp_rx_frames_new;	/*!< LLDP Rx Frames new */
+	u32	lldp_tlvs_unrecognized;	/*!< LLDP Rx unrecognized TLVs */
+	u32	lldp_rx_shutdown_tlvs;	/*!< LLDP Rx shutdown TLVs */
+	u32	lldp_info_aged_out;	/*!< LLDP remote info aged out */
+	u32	dcbx_phylink_ups;	/*!< DCBX phy link ups */
+	u32	dcbx_phylink_downs;	/*!< DCBX phy link downs */
+	u32	dcbx_rx_tlvs;		/*!< DCBX Rx TLVs */
+	u32	dcbx_rx_tlvs_invalid;	/*!< DCBX Rx TLVs invalid */
+	u32	dcbx_control_tlv_error;	/*!< DCBX control TLV errors */
+	u32	dcbx_feature_tlv_error;	/*!< DCBX feature TLV errors */
+	u32	dcbx_cee_cfg_new;	/*!< DCBX new CEE cfg rcvd */
+	u32	cee_status_down;	/*!< CEE status down */
+	u32	cee_status_up;		/*!< CEE status up */
+	u32	cee_hw_cfg_changed;	/*!< CEE hw cfg changed */
+	u32	cee_rx_invalid_cfg;	/*!< CEE invalid cfg */
+};
+
+#pragma pack()
+
+#endif	/* __BFA_DEFS_CNA_H__ */
diff --git a/drivers/net/bna/bfa_defs_mfg_comm.h b/drivers/net/bna/bfa_defs_mfg_comm.h
new file mode 100644
index 000000000000..987978fcb3fe
--- /dev/null
+++ b/drivers/net/bna/bfa_defs_mfg_comm.h
@@ -0,0 +1,244 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#ifndef __BFA_DEFS_MFG_COMM_H__
+#define __BFA_DEFS_MFG_COMM_H__
+
+#include "cna.h"
+
+/**
+ * Manufacturing block version
+ */
+#define BFA_MFG_VERSION				2
+#define BFA_MFG_VERSION_UNINIT			0xFF
+
+/**
+ * Manufacturing block encrypted version
+ */
+#define BFA_MFG_ENC_VER				2
+
+/**
+ * Manufacturing block version 1 length
+ */
+#define BFA_MFG_VER1_LEN			128
+
+/**
+ * Manufacturing block header length
+ */
+#define BFA_MFG_HDR_LEN				4
+
+#define BFA_MFG_SERIALNUM_SIZE			11
+#define STRSZ(_n)				(((_n) + 4) & ~3)
+
+/**
+ * Manufacturing card type
+ */
+enum {
+	BFA_MFG_TYPE_CB_MAX  = 825,      /*!< Crossbow card type max	*/
+	BFA_MFG_TYPE_FC8P2   = 825,      /*!< 8G 2port FC card		*/
+	BFA_MFG_TYPE_FC8P1   = 815,      /*!< 8G 1port FC card		*/
+	BFA_MFG_TYPE_FC4P2   = 425,      /*!< 4G 2port FC card		*/
+	BFA_MFG_TYPE_FC4P1   = 415,      /*!< 4G 1port FC card		*/
+	BFA_MFG_TYPE_CNA10P2 = 1020,     /*!< 10G 2port CNA card	*/
+	BFA_MFG_TYPE_CNA10P1 = 1010,     /*!< 10G 1port CNA card	*/
+	BFA_MFG_TYPE_JAYHAWK = 804,	 /*!< Jayhawk mezz card		*/
+	BFA_MFG_TYPE_WANCHESE = 1007,	 /*!< Wanchese mezz card	*/
+	BFA_MFG_TYPE_ASTRA    = 807,	 /*!< Astra mezz card		*/
+	BFA_MFG_TYPE_LIGHTNING_P0 = 902, /*!< Lightning mezz card - old	*/
+	BFA_MFG_TYPE_LIGHTNING = 1741,	 /*!< Lightning mezz card	*/
+	BFA_MFG_TYPE_INVALID = 0,	 /*!< Invalid card type		*/
+};
+
+#pragma pack(1)
+
+/**
+ * Check if 1-port card
+ */
+#define bfa_mfg_is_1port(type) (( \
+	(type) == BFA_MFG_TYPE_FC8P1 || \
+	(type) == BFA_MFG_TYPE_FC4P1 || \
+	(type) == BFA_MFG_TYPE_CNA10P1))
+
+/**
+ * Check if Mezz card
+ */
+#define bfa_mfg_is_mezz(type) (( \
+	(type) == BFA_MFG_TYPE_JAYHAWK || \
+	(type) == BFA_MFG_TYPE_WANCHESE || \
+	(type) == BFA_MFG_TYPE_ASTRA || \
+	(type) == BFA_MFG_TYPE_LIGHTNING_P0 || \
+	(type) == BFA_MFG_TYPE_LIGHTNING))
+
+/**
+ * Check if card type valid
+ */
+#define bfa_mfg_is_card_type_valid(type) (( \
+	(type) == BFA_MFG_TYPE_FC8P2 || \
+	(type) == BFA_MFG_TYPE_FC8P1 || \
+	(type) == BFA_MFG_TYPE_FC4P2 || \
+	(type) == BFA_MFG_TYPE_FC4P1 || \
+	(type) == BFA_MFG_TYPE_CNA10P2 || \
+	(type) == BFA_MFG_TYPE_CNA10P1 || \
+	bfa_mfg_is_mezz(type)))
+
+/**
+ * Check if the card having old wwn/mac handling
+ */
+#define bfa_mfg_is_old_wwn_mac_model(type) (( \
+	(type) == BFA_MFG_TYPE_FC8P2 || \
+	(type) == BFA_MFG_TYPE_FC8P1 || \
+	(type) == BFA_MFG_TYPE_FC4P2 || \
+	(type) == BFA_MFG_TYPE_FC4P1 || \
+	(type) == BFA_MFG_TYPE_CNA10P2 || \
+	(type) == BFA_MFG_TYPE_CNA10P1 || \
+	(type) == BFA_MFG_TYPE_JAYHAWK || \
+	(type) == BFA_MFG_TYPE_WANCHESE))
+
+#define bfa_mfg_increment_wwn_mac(m, i)				\
+do {								\
+	u32 t = ((m)[0] << 16) | ((m)[1] << 8) | (m)[2];	\
+	t += (i);						\
+	(m)[0] = (t >> 16) & 0xFF;				\
+	(m)[1] = (t >> 8) & 0xFF;				\
+	(m)[2] = t & 0xFF;					\
+} while (0)
+
+#define bfa_mfg_adapter_prop_init_flash(card_type, prop)	\
+do {								\
+	switch ((card_type)) {					\
+	case BFA_MFG_TYPE_FC8P2:				\
+	case BFA_MFG_TYPE_JAYHAWK:				\
+	case BFA_MFG_TYPE_ASTRA:				\
+		(prop) = BFI_ADAPTER_SETP(NPORTS, 2) |		\
+			BFI_ADAPTER_SETP(SPEED, 8);		\
+		break;						\
+	case BFA_MFG_TYPE_FC8P1:				\
+		(prop) = BFI_ADAPTER_SETP(NPORTS, 1) |		\
+			BFI_ADAPTER_SETP(SPEED, 8);		\
+		break;						\
+	case BFA_MFG_TYPE_FC4P2:				\
+		(prop) = BFI_ADAPTER_SETP(NPORTS, 2) |		\
+			BFI_ADAPTER_SETP(SPEED, 4);		\
+		break;						\
+	case BFA_MFG_TYPE_FC4P1:				\
+		(prop) = BFI_ADAPTER_SETP(NPORTS, 1) |		\
+			BFI_ADAPTER_SETP(SPEED, 4);		\
+		break;						\
+	case BFA_MFG_TYPE_CNA10P2:				\
+	case BFA_MFG_TYPE_WANCHESE:				\
+	case BFA_MFG_TYPE_LIGHTNING_P0:				\
+	case BFA_MFG_TYPE_LIGHTNING:				\
+		(prop) = BFI_ADAPTER_SETP(NPORTS, 2);		\
+		(prop) |= BFI_ADAPTER_SETP(SPEED, 10);		\
+		break;						\
+	case BFA_MFG_TYPE_CNA10P1:				\
+		(prop) = BFI_ADAPTER_SETP(NPORTS, 1);		\
+		(prop) |= BFI_ADAPTER_SETP(SPEED, 10);		\
+		break;						\
+	default:						\
+		(prop) = BFI_ADAPTER_UNSUPP;			\
+	}							\
+} while (0)
+
+enum {
+	CB_GPIO_TTV	= (1),		/*!< TTV debug capable cards	*/
+	CB_GPIO_FC8P2   = (2),		/*!< 8G 2port FC card		*/
+	CB_GPIO_FC8P1   = (3),		/*!< 8G 1port FC card		*/
+	CB_GPIO_FC4P2   = (4),		/*!< 4G 2port FC card		*/
+	CB_GPIO_FC4P1   = (5),		/*!< 4G 1port FC card		*/
+	CB_GPIO_DFLY    = (6),		/*!< 8G 2port FC mezzanine card	*/
+	CB_GPIO_PROTO   = (1 << 7)	/*!< 8G 2port FC prototypes	*/
+};
+
+#define bfa_mfg_adapter_prop_init_gpio(gpio, card_type, prop)	\
+do {								\
+	if ((gpio) & CB_GPIO_PROTO) {				\
+		(prop) |= BFI_ADAPTER_PROTO;			\
+		(gpio) &= ~CB_GPIO_PROTO;			\
+	}							\
+	switch ((gpio)) {					\
+	case CB_GPIO_TTV:					\
+		(prop) |= BFI_ADAPTER_TTV;			\
+	case CB_GPIO_DFLY:					\
+	case CB_GPIO_FC8P2:					\
+		(prop) |= BFI_ADAPTER_SETP(NPORTS, 2);		\
+		(prop) |= BFI_ADAPTER_SETP(SPEED, 8);		\
+		(card_type) = BFA_MFG_TYPE_FC8P2;		\
+		break;						\
+	case CB_GPIO_FC8P1:					\
+		(prop) |= BFI_ADAPTER_SETP(NPORTS, 1);		\
+		(prop) |= BFI_ADAPTER_SETP(SPEED, 8);		\
+		(card_type) = BFA_MFG_TYPE_FC8P1;		\
+		break;						\
+	case CB_GPIO_FC4P2:					\
+		(prop) |= BFI_ADAPTER_SETP(NPORTS, 2);		\
+		(prop) |= BFI_ADAPTER_SETP(SPEED, 4);		\
+		(card_type) = BFA_MFG_TYPE_FC4P2;		\
+		break;						\
+	case CB_GPIO_FC4P1:					\
+		(prop) |= BFI_ADAPTER_SETP(NPORTS, 1);		\
+		(prop) |= BFI_ADAPTER_SETP(SPEED, 4);		\
+		(card_type) = BFA_MFG_TYPE_FC4P1;		\
+		break;						\
+	default:						\
+		(prop) |= BFI_ADAPTER_UNSUPP;			\
+		(card_type) = BFA_MFG_TYPE_INVALID;		\
+	}							\
+} while (0)
+
+/**
+ * VPD data length
+ */
+#define BFA_MFG_VPD_LEN			512
+#define BFA_MFG_VPD_LEN_INVALID		0
+
+#define BFA_MFG_VPD_PCI_HDR_OFF		137
+#define BFA_MFG_VPD_PCI_VER_MASK	0x07	/*!< version mask 3 bits */
+#define BFA_MFG_VPD_PCI_VDR_MASK	0xf8	/*!< vendor mask 5 bits */
+
+/**
+ * VPD vendor tag
+ */
+enum {
+	BFA_MFG_VPD_UNKNOWN	= 0,     /*!< vendor unknown 		*/
+	BFA_MFG_VPD_IBM 	= 1,     /*!< vendor IBM 		*/
+	BFA_MFG_VPD_HP  	= 2,     /*!< vendor HP  		*/
+	BFA_MFG_VPD_DELL  	= 3,     /*!< vendor DELL  		*/
+	BFA_MFG_VPD_PCI_IBM 	= 0x08,  /*!< PCI VPD IBM     		*/
+	BFA_MFG_VPD_PCI_HP  	= 0x10,  /*!< PCI VPD HP		*/
+	BFA_MFG_VPD_PCI_DELL  	= 0x20,  /*!< PCI VPD DELL		*/
+	BFA_MFG_VPD_PCI_BRCD 	= 0xf8,  /*!< PCI VPD Brocade 		*/
+};
+
+/**
+ * @brief BFA adapter flash vpd data definition.
+ *
+ * All numerical fields are in big-endian format.
+ */
+struct bfa_mfg_vpd {
+	u8		version;	/*!< vpd data version */
+	u8		vpd_sig[3];	/*!< characters 'V', 'P', 'D' */
+	u8		chksum;		/*!< u8 checksum */
+	u8		vendor;		/*!< vendor */
+	u8 	len;		/*!< vpd data length excluding header */
+	u8 	rsv;
+	u8		data[BFA_MFG_VPD_LEN];	/*!< vpd data */
+};
+
+#pragma pack()
+
+#endif /* __BFA_DEFS_MFG_H__ */
diff --git a/drivers/net/bna/bfa_defs_status.h b/drivers/net/bna/bfa_defs_status.h
new file mode 100644
index 000000000000..af951126375c
--- /dev/null
+++ b/drivers/net/bna/bfa_defs_status.h
@@ -0,0 +1,216 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#ifndef __BFA_DEFS_STATUS_H__
+#define __BFA_DEFS_STATUS_H__
+
+/**
+ * API status return values
+ *
+ * NOTE: The error msgs are auto generated from the comments. Only singe line
+ * comments are supported
+ */
+enum bfa_status {
+	BFA_STATUS_OK 		= 0,
+	BFA_STATUS_FAILED 	= 1,
+	BFA_STATUS_EINVAL 	= 2,
+	BFA_STATUS_ENOMEM 	= 3,
+	BFA_STATUS_ENOSYS 	= 4,
+	BFA_STATUS_ETIMER 	= 5,
+	BFA_STATUS_EPROTOCOL 	= 6,
+	BFA_STATUS_ENOFCPORTS 	= 7,
+	BFA_STATUS_NOFLASH 	= 8,
+	BFA_STATUS_BADFLASH 	= 9,
+	BFA_STATUS_SFP_UNSUPP 	= 10,
+	BFA_STATUS_UNKNOWN_VFID = 11,
+	BFA_STATUS_DATACORRUPTED = 12,
+	BFA_STATUS_DEVBUSY 	= 13,
+	BFA_STATUS_ABORTED 	= 14,
+	BFA_STATUS_NODEV 	= 15,
+	BFA_STATUS_HDMA_FAILED 	= 16,
+	BFA_STATUS_FLASH_BAD_LEN = 17,
+	BFA_STATUS_UNKNOWN_LWWN = 18,
+	BFA_STATUS_UNKNOWN_RWWN = 19,
+	BFA_STATUS_FCPT_LS_RJT 	= 20,
+	BFA_STATUS_VPORT_EXISTS = 21,
+	BFA_STATUS_VPORT_MAX 	= 22,
+	BFA_STATUS_UNSUPP_SPEED = 23,
+	BFA_STATUS_INVLD_DFSZ 	= 24,
+	BFA_STATUS_CNFG_FAILED 	= 25,
+	BFA_STATUS_CMD_NOTSUPP 	= 26,
+	BFA_STATUS_NO_ADAPTER 	= 27,
+	BFA_STATUS_LINKDOWN 	= 28,
+	BFA_STATUS_FABRIC_RJT 	= 29,
+	BFA_STATUS_UNKNOWN_VWWN = 30,
+	BFA_STATUS_NSLOGIN_FAILED = 31,
+	BFA_STATUS_NO_RPORTS 	= 32,
+	BFA_STATUS_NSQUERY_FAILED = 33,
+	BFA_STATUS_PORT_OFFLINE = 34,
+	BFA_STATUS_RPORT_OFFLINE = 35,
+	BFA_STATUS_TGTOPEN_FAILED = 36,
+	BFA_STATUS_BAD_LUNS 	= 37,
+	BFA_STATUS_IO_FAILURE 	= 38,
+	BFA_STATUS_NO_FABRIC 	= 39,
+	BFA_STATUS_EBADF 	= 40,
+	BFA_STATUS_EINTR 	= 41,
+	BFA_STATUS_EIO 		= 42,
+	BFA_STATUS_ENOTTY 	= 43,
+	BFA_STATUS_ENXIO 	= 44,
+	BFA_STATUS_EFOPEN 	= 45,
+	BFA_STATUS_VPORT_WWN_BP = 46,
+	BFA_STATUS_PORT_NOT_DISABLED = 47,
+	BFA_STATUS_BADFRMHDR 	= 48,
+	BFA_STATUS_BADFRMSZ 	= 49,
+	BFA_STATUS_MISSINGFRM 	= 50,
+	BFA_STATUS_LINKTIMEOUT 	= 51,
+	BFA_STATUS_NO_FCPIM_NEXUS = 52,
+	BFA_STATUS_CHECKSUM_FAIL = 53,
+	BFA_STATUS_GZME_FAILED 	= 54,
+	BFA_STATUS_SCSISTART_REQD = 55,
+	BFA_STATUS_IOC_FAILURE 	= 56,
+	BFA_STATUS_INVALID_WWN 	= 57,
+	BFA_STATUS_MISMATCH 	= 58,
+	BFA_STATUS_IOC_ENABLED 	= 59,
+	BFA_STATUS_ADAPTER_ENABLED = 60,
+	BFA_STATUS_IOC_NON_OP 	= 61,
+	BFA_STATUS_ADDR_MAP_FAILURE = 62,
+	BFA_STATUS_SAME_NAME 	= 63,
+	BFA_STATUS_PENDING      = 64,
+	BFA_STATUS_8G_SPD	= 65,
+	BFA_STATUS_4G_SPD	= 66,
+	BFA_STATUS_AD_IS_ENABLE = 67,
+	BFA_STATUS_EINVAL_TOV 	= 68,
+	BFA_STATUS_EINVAL_QDEPTH = 69,
+	BFA_STATUS_VERSION_FAIL = 70,
+	BFA_STATUS_DIAG_BUSY    = 71,
+	BFA_STATUS_BEACON_ON	= 72,
+	BFA_STATUS_BEACON_OFF	= 73,
+	BFA_STATUS_LBEACON_ON   = 74,
+	BFA_STATUS_LBEACON_OFF	= 75,
+	BFA_STATUS_PORT_NOT_INITED = 76,
+	BFA_STATUS_RPSC_ENABLED = 77,
+	BFA_STATUS_ENOFSAVE 	= 78,
+	BFA_STATUS_BAD_FILE		= 79,
+	BFA_STATUS_RLIM_EN		= 80,
+	BFA_STATUS_RLIM_DIS		= 81,
+	BFA_STATUS_IOC_DISABLED  = 82,
+	BFA_STATUS_ADAPTER_DISABLED  = 83,
+	BFA_STATUS_BIOS_DISABLED  = 84,
+	BFA_STATUS_AUTH_ENABLED  = 85,
+	BFA_STATUS_AUTH_DISABLED  = 86,
+	BFA_STATUS_ERROR_TRL_ENABLED  = 87,
+	BFA_STATUS_ERROR_QOS_ENABLED  = 88,
+	BFA_STATUS_NO_SFP_DEV = 89,
+	BFA_STATUS_MEMTEST_FAILED = 90,
+	BFA_STATUS_INVALID_DEVID = 91,
+	BFA_STATUS_QOS_ENABLED = 92,
+	BFA_STATUS_QOS_DISABLED = 93,
+	BFA_STATUS_INCORRECT_DRV_CONFIG = 94,
+	BFA_STATUS_REG_FAIL = 95,
+	BFA_STATUS_IM_INV_CODE = 96,
+	BFA_STATUS_IM_INV_VLAN = 97,
+	BFA_STATUS_IM_INV_ADAPT_NAME = 98,
+	BFA_STATUS_IM_LOW_RESOURCES = 99,
+	BFA_STATUS_IM_VLANID_IS_PVID = 100,
+	BFA_STATUS_IM_VLANID_EXISTS = 101,
+	BFA_STATUS_IM_FW_UPDATE_FAIL = 102,
+	BFA_STATUS_PORTLOG_ENABLED = 103,
+	BFA_STATUS_PORTLOG_DISABLED = 104,
+	BFA_STATUS_FILE_NOT_FOUND = 105,
+	BFA_STATUS_QOS_FC_ONLY = 106,
+	BFA_STATUS_RLIM_FC_ONLY = 107,
+	BFA_STATUS_CT_SPD = 108,
+	BFA_STATUS_LEDTEST_OP = 109,
+	BFA_STATUS_CEE_NOT_DN = 110,
+	BFA_STATUS_10G_SPD = 111,
+	BFA_STATUS_IM_INV_TEAM_NAME = 112,
+	BFA_STATUS_IM_DUP_TEAM_NAME = 113,
+	BFA_STATUS_IM_ADAPT_ALREADY_IN_TEAM = 114,
+	BFA_STATUS_IM_ADAPT_HAS_VLANS = 115,
+	BFA_STATUS_IM_PVID_MISMATCH = 116,
+	BFA_STATUS_IM_LINK_SPEED_MISMATCH = 117,
+	BFA_STATUS_IM_MTU_MISMATCH = 118,
+	BFA_STATUS_IM_RSS_MISMATCH = 119,
+	BFA_STATUS_IM_HDS_MISMATCH = 120,
+	BFA_STATUS_IM_OFFLOAD_MISMATCH = 121,
+	BFA_STATUS_IM_PORT_PARAMS = 122,
+	BFA_STATUS_IM_PORT_NOT_IN_TEAM = 123,
+	BFA_STATUS_IM_CANNOT_REM_PRI = 124,
+	BFA_STATUS_IM_MAX_PORTS_REACHED = 125,
+	BFA_STATUS_IM_LAST_PORT_DELETE = 126,
+	BFA_STATUS_IM_NO_DRIVER = 127,
+	BFA_STATUS_IM_MAX_VLANS_REACHED = 128,
+	BFA_STATUS_TOMCAT_SPD_NOT_ALLOWED = 129,
+	BFA_STATUS_NO_MINPORT_DRIVER = 130,
+	BFA_STATUS_CARD_TYPE_MISMATCH = 131,
+	BFA_STATUS_BAD_ASICBLK = 132,
+	BFA_STATUS_NO_DRIVER = 133,
+	BFA_STATUS_INVALID_MAC = 134,
+	BFA_STATUS_IM_NO_VLAN = 135,
+	BFA_STATUS_IM_ETH_LB_FAILED = 136,
+	BFA_STATUS_IM_PVID_REMOVE = 137,
+	BFA_STATUS_IM_PVID_EDIT = 138,
+	BFA_STATUS_CNA_NO_BOOT = 139,
+	BFA_STATUS_IM_PVID_NON_ZERO = 140,
+	BFA_STATUS_IM_INETCFG_LOCK_FAILED = 141,
+	BFA_STATUS_IM_GET_INETCFG_FAILED = 142,
+	BFA_STATUS_IM_NOT_BOUND = 143,
+	BFA_STATUS_INSUFFICIENT_PERMS = 144,
+	BFA_STATUS_IM_INV_VLAN_NAME = 145,
+	BFA_STATUS_CMD_NOTSUPP_CNA = 146,
+	BFA_STATUS_IM_PASSTHRU_EDIT = 147,
+	BFA_STATUS_IM_BIND_FAILED = 148,
+	BFA_STATUS_IM_UNBIND_FAILED = 149,
+	BFA_STATUS_IM_PORT_IN_TEAM = 150,
+	BFA_STATUS_IM_VLAN_NOT_FOUND = 151,
+	BFA_STATUS_IM_TEAM_NOT_FOUND = 152,
+	BFA_STATUS_IM_TEAM_CFG_NOT_ALLOWED = 153,
+	BFA_STATUS_PBC = 154,
+	BFA_STATUS_DEVID_MISSING = 155,
+	BFA_STATUS_BAD_FWCFG = 156,
+	BFA_STATUS_CREATE_FILE = 157,
+	BFA_STATUS_INVALID_VENDOR = 158,
+	BFA_STATUS_SFP_NOT_READY = 159,
+	BFA_STATUS_FLASH_UNINIT = 160,
+	BFA_STATUS_FLASH_EMPTY = 161,
+	BFA_STATUS_FLASH_CKFAIL = 162,
+	BFA_STATUS_TRUNK_UNSUPP = 163,
+	BFA_STATUS_TRUNK_ENABLED = 164,
+	BFA_STATUS_TRUNK_DISABLED  = 165,
+	BFA_STATUS_TRUNK_ERROR_TRL_ENABLED = 166,
+	BFA_STATUS_BOOT_CODE_UPDATED = 167,
+	BFA_STATUS_BOOT_VERSION = 168,
+	BFA_STATUS_CARDTYPE_MISSING = 169,
+	BFA_STATUS_INVALID_CARDTYPE = 170,
+	BFA_STATUS_NO_TOPOLOGY_FOR_CNA = 171,
+	BFA_STATUS_IM_VLAN_OVER_TEAM_DELETE_FAILED = 172,
+	BFA_STATUS_ETHBOOT_ENABLED  = 173,
+	BFA_STATUS_ETHBOOT_DISABLED  = 174,
+	BFA_STATUS_IOPROFILE_OFF = 175,
+	BFA_STATUS_NO_PORT_INSTANCE = 176,
+	BFA_STATUS_BOOT_CODE_TIMEDOUT = 177,
+	BFA_STATUS_NO_VPORT_LOCK = 178,
+	BFA_STATUS_VPORT_NO_CNFG = 179,
+	BFA_STATUS_MAX_VAL
+};
+
+enum bfa_eproto_status {
+	BFA_EPROTO_BAD_ACCEPT = 0,
+	BFA_EPROTO_UNKNOWN_RSP = 1
+};
+
+#endif /* __BFA_DEFS_STATUS_H__ */
diff --git a/drivers/net/bna/bfa_ioc.c b/drivers/net/bna/bfa_ioc.c
new file mode 100644
index 000000000000..cdc2cb1597ec
--- /dev/null
+++ b/drivers/net/bna/bfa_ioc.c
@@ -0,0 +1,1839 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#include "bfa_ioc.h"
+#include "cna.h"
+#include "bfi.h"
+#include "bfi_ctreg.h"
+#include "bfa_defs.h"
+
+/**
+ * IOC local definitions
+ */
+
+#define bfa_ioc_timer_start(__ioc)					\
+	mod_timer(&(__ioc)->ioc_timer, jiffies +	\
+			msecs_to_jiffies(BFA_IOC_TOV))
+#define bfa_ioc_timer_stop(__ioc)   del_timer(&(__ioc)->ioc_timer)
+
+#define bfa_ioc_recovery_timer_start(__ioc)				\
+	mod_timer(&(__ioc)->ioc_timer, jiffies +	\
+			msecs_to_jiffies(BFA_IOC_TOV_RECOVER))
+
+#define bfa_sem_timer_start(__ioc)					\
+	mod_timer(&(__ioc)->sem_timer, jiffies +	\
+			msecs_to_jiffies(BFA_IOC_HWSEM_TOV))
+#define bfa_sem_timer_stop(__ioc)	del_timer(&(__ioc)->sem_timer)
+
+#define bfa_hb_timer_start(__ioc)					\
+	mod_timer(&(__ioc)->hb_timer, jiffies +		\
+			msecs_to_jiffies(BFA_IOC_HB_TOV))
+#define bfa_hb_timer_stop(__ioc)	del_timer(&(__ioc)->hb_timer)
+
+/**
+ * Asic specific macros : see bfa_hw_cb.c and bfa_hw_ct.c for details.
+ */
+
+#define bfa_ioc_firmware_lock(__ioc)			\
+			((__ioc)->ioc_hwif->ioc_firmware_lock(__ioc))
+#define bfa_ioc_firmware_unlock(__ioc)			\
+			((__ioc)->ioc_hwif->ioc_firmware_unlock(__ioc))
+#define bfa_ioc_reg_init(__ioc) ((__ioc)->ioc_hwif->ioc_reg_init(__ioc))
+#define bfa_ioc_map_port(__ioc) ((__ioc)->ioc_hwif->ioc_map_port(__ioc))
+#define bfa_ioc_notify_hbfail(__ioc)			\
+			((__ioc)->ioc_hwif->ioc_notify_hbfail(__ioc))
+
+#define bfa_ioc_is_optrom(__ioc)	\
+	(bfa_cb_image_get_size(BFA_IOC_FWIMG_TYPE(__ioc)) < BFA_IOC_FWIMG_MINSZ)
+
+#define bfa_ioc_mbox_cmd_pending(__ioc)		\
+			(!list_empty(&((__ioc)->mbox_mod.cmd_q)) || \
+			readl((__ioc)->ioc_regs.hfn_mbox_cmd))
+
+bool bfa_auto_recover = true;
+
+/*
+ * forward declarations
+ */
+static void bfa_ioc_hw_sem_get(struct bfa_ioc *ioc);
+static void bfa_ioc_hw_sem_get_cancel(struct bfa_ioc *ioc);
+static void bfa_ioc_hwinit(struct bfa_ioc *ioc, bool force);
+static void bfa_ioc_send_enable(struct bfa_ioc *ioc);
+static void bfa_ioc_send_disable(struct bfa_ioc *ioc);
+static void bfa_ioc_send_getattr(struct bfa_ioc *ioc);
+static void bfa_ioc_hb_monitor(struct bfa_ioc *ioc);
+static void bfa_ioc_hb_stop(struct bfa_ioc *ioc);
+static void bfa_ioc_reset(struct bfa_ioc *ioc, bool force);
+static void bfa_ioc_mbox_poll(struct bfa_ioc *ioc);
+static void bfa_ioc_mbox_hbfail(struct bfa_ioc *ioc);
+static void bfa_ioc_recover(struct bfa_ioc *ioc);
+static void bfa_ioc_check_attr_wwns(struct bfa_ioc *ioc);
+static void bfa_ioc_disable_comp(struct bfa_ioc *ioc);
+static void bfa_ioc_lpu_stop(struct bfa_ioc *ioc);
+
+/**
+ * IOC state machine events
+ */
+enum ioc_event {
+	IOC_E_ENABLE		= 1,	/*!< IOC enable request		*/
+	IOC_E_DISABLE		= 2,	/*!< IOC disable request	*/
+	IOC_E_TIMEOUT		= 3,	/*!< f/w response timeout	*/
+	IOC_E_FWREADY		= 4,	/*!< f/w initialization done	*/
+	IOC_E_FWRSP_GETATTR	= 5,	/*!< IOC get attribute response	*/
+	IOC_E_FWRSP_ENABLE	= 6,	/*!< enable f/w response	*/
+	IOC_E_FWRSP_DISABLE	= 7,	/*!< disable f/w response	*/
+	IOC_E_HBFAIL		= 8,	/*!< heartbeat failure		*/
+	IOC_E_HWERROR		= 9,	/*!< hardware error interrupt	*/
+	IOC_E_SEMLOCKED		= 10,	/*!< h/w semaphore is locked	*/
+	IOC_E_DETACH		= 11,	/*!< driver detach cleanup	*/
+};
+
+bfa_fsm_state_decl(bfa_ioc, reset, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, fwcheck, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, mismatch, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, semwait, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, hwinit, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, enabling, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, getattr, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, op, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, initfail, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, hbfail, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, disabling, struct bfa_ioc, enum ioc_event);
+bfa_fsm_state_decl(bfa_ioc, disabled, struct bfa_ioc, enum ioc_event);
+
+static struct bfa_sm_table ioc_sm_table[] = {
+	{BFA_SM(bfa_ioc_sm_reset), BFA_IOC_RESET},
+	{BFA_SM(bfa_ioc_sm_fwcheck), BFA_IOC_FWMISMATCH},
+	{BFA_SM(bfa_ioc_sm_mismatch), BFA_IOC_FWMISMATCH},
+	{BFA_SM(bfa_ioc_sm_semwait), BFA_IOC_SEMWAIT},
+	{BFA_SM(bfa_ioc_sm_hwinit), BFA_IOC_HWINIT},
+	{BFA_SM(bfa_ioc_sm_enabling), BFA_IOC_HWINIT},
+	{BFA_SM(bfa_ioc_sm_getattr), BFA_IOC_GETATTR},
+	{BFA_SM(bfa_ioc_sm_op), BFA_IOC_OPERATIONAL},
+	{BFA_SM(bfa_ioc_sm_initfail), BFA_IOC_INITFAIL},
+	{BFA_SM(bfa_ioc_sm_hbfail), BFA_IOC_HBFAIL},
+	{BFA_SM(bfa_ioc_sm_disabling), BFA_IOC_DISABLING},
+	{BFA_SM(bfa_ioc_sm_disabled), BFA_IOC_DISABLED},
+};
+
+/**
+ * Reset entry actions -- initialize state machine
+ */
+static void
+bfa_ioc_sm_reset_entry(struct bfa_ioc *ioc)
+{
+	ioc->retry_count = 0;
+	ioc->auto_recover = bfa_auto_recover;
+}
+
+/**
+ * Beginning state. IOC is in reset state.
+ */
+static void
+bfa_ioc_sm_reset(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_ENABLE:
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_fwcheck);
+		break;
+
+	case IOC_E_DISABLE:
+		bfa_ioc_disable_comp(ioc);
+		break;
+
+	case IOC_E_DETACH:
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+/**
+ * Semaphore should be acquired for version check.
+ */
+static void
+bfa_ioc_sm_fwcheck_entry(struct bfa_ioc *ioc)
+{
+	bfa_ioc_hw_sem_get(ioc);
+}
+
+/**
+ * Awaiting h/w semaphore to continue with version check.
+ */
+static void
+bfa_ioc_sm_fwcheck(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_SEMLOCKED:
+		if (bfa_ioc_firmware_lock(ioc)) {
+			ioc->retry_count = 0;
+			bfa_fsm_set_state(ioc, bfa_ioc_sm_hwinit);
+		} else {
+			bfa_ioc_hw_sem_release(ioc);
+			bfa_fsm_set_state(ioc, bfa_ioc_sm_mismatch);
+		}
+		break;
+
+	case IOC_E_DISABLE:
+		bfa_ioc_disable_comp(ioc);
+		/* fall through */
+
+	case IOC_E_DETACH:
+		bfa_ioc_hw_sem_get_cancel(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_reset);
+		break;
+
+	case IOC_E_FWREADY:
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+/**
+ * Notify enable completion callback and generate mismatch AEN.
+ */
+static void
+bfa_ioc_sm_mismatch_entry(struct bfa_ioc *ioc)
+{
+	/**
+	 * Provide enable completion callback and AEN notification only once.
+	 */
+	if (ioc->retry_count == 0)
+		ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE);
+	ioc->retry_count++;
+	bfa_ioc_timer_start(ioc);
+}
+
+/**
+ * Awaiting firmware version match.
+ */
+static void
+bfa_ioc_sm_mismatch(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_TIMEOUT:
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_fwcheck);
+		break;
+
+	case IOC_E_DISABLE:
+		bfa_ioc_disable_comp(ioc);
+		/* fall through */
+
+	case IOC_E_DETACH:
+		bfa_ioc_timer_stop(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_reset);
+		break;
+
+	case IOC_E_FWREADY:
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+/**
+ * Request for semaphore.
+ */
+static void
+bfa_ioc_sm_semwait_entry(struct bfa_ioc *ioc)
+{
+	bfa_ioc_hw_sem_get(ioc);
+}
+
+/**
+ * Awaiting semaphore for h/w initialzation.
+ */
+static void
+bfa_ioc_sm_semwait(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_SEMLOCKED:
+		ioc->retry_count = 0;
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_hwinit);
+		break;
+
+	case IOC_E_DISABLE:
+		bfa_ioc_hw_sem_get_cancel(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabled);
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+static void
+bfa_ioc_sm_hwinit_entry(struct bfa_ioc *ioc)
+{
+	bfa_ioc_timer_start(ioc);
+	bfa_ioc_reset(ioc, false);
+}
+
+/**
+ * @brief
+ * Hardware is being initialized. Interrupts are enabled.
+ * Holding hardware semaphore lock.
+ */
+static void
+bfa_ioc_sm_hwinit(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_FWREADY:
+		bfa_ioc_timer_stop(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_enabling);
+		break;
+
+	case IOC_E_HWERROR:
+		bfa_ioc_timer_stop(ioc);
+		/* fall through */
+
+	case IOC_E_TIMEOUT:
+		ioc->retry_count++;
+		if (ioc->retry_count < BFA_IOC_HWINIT_MAX) {
+			bfa_ioc_timer_start(ioc);
+			bfa_ioc_reset(ioc, true);
+			break;
+		}
+
+		bfa_ioc_hw_sem_release(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_initfail);
+		break;
+
+	case IOC_E_DISABLE:
+		bfa_ioc_hw_sem_release(ioc);
+		bfa_ioc_timer_stop(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabled);
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+static void
+bfa_ioc_sm_enabling_entry(struct bfa_ioc *ioc)
+{
+	bfa_ioc_timer_start(ioc);
+	bfa_ioc_send_enable(ioc);
+}
+
+/**
+ * Host IOC function is being enabled, awaiting response from firmware.
+ * Semaphore is acquired.
+ */
+static void
+bfa_ioc_sm_enabling(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_FWRSP_ENABLE:
+		bfa_ioc_timer_stop(ioc);
+		bfa_ioc_hw_sem_release(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_getattr);
+		break;
+
+	case IOC_E_HWERROR:
+		bfa_ioc_timer_stop(ioc);
+		/* fall through */
+
+	case IOC_E_TIMEOUT:
+		ioc->retry_count++;
+		if (ioc->retry_count < BFA_IOC_HWINIT_MAX) {
+			writel(BFI_IOC_UNINIT,
+				      ioc->ioc_regs.ioc_fwstate);
+			bfa_fsm_set_state(ioc, bfa_ioc_sm_hwinit);
+			break;
+		}
+
+		bfa_ioc_hw_sem_release(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_initfail);
+		break;
+
+	case IOC_E_DISABLE:
+		bfa_ioc_timer_stop(ioc);
+		bfa_ioc_hw_sem_release(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabled);
+		break;
+
+	case IOC_E_FWREADY:
+		bfa_ioc_send_enable(ioc);
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+static void
+bfa_ioc_sm_getattr_entry(struct bfa_ioc *ioc)
+{
+	bfa_ioc_timer_start(ioc);
+	bfa_ioc_send_getattr(ioc);
+}
+
+/**
+ * @brief
+ * IOC configuration in progress. Timer is active.
+ */
+static void
+bfa_ioc_sm_getattr(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_FWRSP_GETATTR:
+		bfa_ioc_timer_stop(ioc);
+		bfa_ioc_check_attr_wwns(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_op);
+		break;
+
+	case IOC_E_HWERROR:
+		bfa_ioc_timer_stop(ioc);
+		/* fall through */
+
+	case IOC_E_TIMEOUT:
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_initfail);
+		break;
+
+	case IOC_E_DISABLE:
+		bfa_ioc_timer_stop(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabled);
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+static void
+bfa_ioc_sm_op_entry(struct bfa_ioc *ioc)
+{
+	ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_OK);
+	bfa_ioc_hb_monitor(ioc);
+}
+
+static void
+bfa_ioc_sm_op(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_ENABLE:
+		break;
+
+	case IOC_E_DISABLE:
+		bfa_ioc_hb_stop(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabling);
+		break;
+
+	case IOC_E_HWERROR:
+	case IOC_E_FWREADY:
+		/**
+		 * Hard error or IOC recovery by other function.
+		 * Treat it same as heartbeat failure.
+		 */
+		bfa_ioc_hb_stop(ioc);
+		/* !!! fall through !!! */
+
+	case IOC_E_HBFAIL:
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_hbfail);
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+static void
+bfa_ioc_sm_disabling_entry(struct bfa_ioc *ioc)
+{
+	bfa_ioc_timer_start(ioc);
+	bfa_ioc_send_disable(ioc);
+}
+
+/**
+ * IOC is being disabled
+ */
+static void
+bfa_ioc_sm_disabling(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_FWRSP_DISABLE:
+		bfa_ioc_timer_stop(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabled);
+		break;
+
+	case IOC_E_HWERROR:
+		bfa_ioc_timer_stop(ioc);
+		/*
+		 * !!! fall through !!!
+		 */
+
+	case IOC_E_TIMEOUT:
+		writel(BFI_IOC_FAIL, ioc->ioc_regs.ioc_fwstate);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabled);
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+/**
+ * IOC disable completion entry.
+ */
+static void
+bfa_ioc_sm_disabled_entry(struct bfa_ioc *ioc)
+{
+	bfa_ioc_disable_comp(ioc);
+}
+
+static void
+bfa_ioc_sm_disabled(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_ENABLE:
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_semwait);
+		break;
+
+	case IOC_E_DISABLE:
+		ioc->cbfn->disable_cbfn(ioc->bfa);
+		break;
+
+	case IOC_E_FWREADY:
+		break;
+
+	case IOC_E_DETACH:
+		bfa_ioc_firmware_unlock(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_reset);
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+static void
+bfa_ioc_sm_initfail_entry(struct bfa_ioc *ioc)
+{
+	ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE);
+	bfa_ioc_timer_start(ioc);
+}
+
+/**
+ * @brief
+ * Hardware initialization failed.
+ */
+static void
+bfa_ioc_sm_initfail(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+	case IOC_E_DISABLE:
+		bfa_ioc_timer_stop(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabled);
+		break;
+
+	case IOC_E_DETACH:
+		bfa_ioc_timer_stop(ioc);
+		bfa_ioc_firmware_unlock(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_reset);
+		break;
+
+	case IOC_E_TIMEOUT:
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_semwait);
+		break;
+
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+static void
+bfa_ioc_sm_hbfail_entry(struct bfa_ioc *ioc)
+{
+	struct list_head			*qe;
+	struct bfa_ioc_hbfail_notify *notify;
+
+	/**
+	 * Mark IOC as failed in hardware and stop firmware.
+	 */
+	bfa_ioc_lpu_stop(ioc);
+	writel(BFI_IOC_FAIL, ioc->ioc_regs.ioc_fwstate);
+
+	/**
+	 * Notify other functions on HB failure.
+	 */
+	bfa_ioc_notify_hbfail(ioc);
+
+	/**
+	 * Notify driver and common modules registered for notification.
+	 */
+	ioc->cbfn->hbfail_cbfn(ioc->bfa);
+	list_for_each(qe, &ioc->hb_notify_q) {
+		notify = (struct bfa_ioc_hbfail_notify *) qe;
+		notify->cbfn(notify->cbarg);
+	}
+
+	/**
+	 * Flush any queued up mailbox requests.
+	 */
+	bfa_ioc_mbox_hbfail(ioc);
+
+	/**
+	 * Trigger auto-recovery after a delay.
+	 */
+	if (ioc->auto_recover)
+		mod_timer(&ioc->ioc_timer, jiffies +
+			msecs_to_jiffies(BFA_IOC_TOV_RECOVER));
+}
+
+/**
+ * @brief
+ * IOC heartbeat failure.
+ */
+static void
+bfa_ioc_sm_hbfail(struct bfa_ioc *ioc, enum ioc_event event)
+{
+	switch (event) {
+
+	case IOC_E_ENABLE:
+		ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE);
+		break;
+
+	case IOC_E_DISABLE:
+		if (ioc->auto_recover)
+			bfa_ioc_timer_stop(ioc);
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_disabled);
+		break;
+
+	case IOC_E_TIMEOUT:
+		bfa_fsm_set_state(ioc, bfa_ioc_sm_semwait);
+		break;
+
+	case IOC_E_FWREADY:
+		/**
+		 * Recovery is already initiated by other function.
+		 */
+		break;
+
+	case IOC_E_HWERROR:
+		/*
+		 * HB failure notification, ignore.
+		 */
+		break;
+	default:
+		bfa_sm_fault(ioc, event);
+	}
+}
+
+/**
+ * BFA IOC private functions
+ */
+
+static void
+bfa_ioc_disable_comp(struct bfa_ioc *ioc)
+{
+	struct list_head			*qe;
+	struct bfa_ioc_hbfail_notify *notify;
+
+	ioc->cbfn->disable_cbfn(ioc->bfa);
+
+	/**
+	 * Notify common modules registered for notification.
+	 */
+	list_for_each(qe, &ioc->hb_notify_q) {
+		notify = (struct bfa_ioc_hbfail_notify *) qe;
+		notify->cbfn(notify->cbarg);
+	}
+}
+
+void
+bfa_ioc_sem_timeout(void *ioc_arg)
+{
+	struct bfa_ioc *ioc = (struct bfa_ioc *) ioc_arg;
+
+	bfa_ioc_hw_sem_get(ioc);
+}
+
+bool
+bfa_ioc_sem_get(void __iomem *sem_reg)
+{
+	u32 r32;
+	int cnt = 0;
+#define BFA_SEM_SPINCNT	3000
+
+	r32 = readl(sem_reg);
+
+	while (r32 && (cnt < BFA_SEM_SPINCNT)) {
+		cnt++;
+		udelay(2);
+		r32 = readl(sem_reg);
+	}
+
+	if (r32 == 0)
+		return true;
+
+	BUG_ON(!(cnt < BFA_SEM_SPINCNT));
+	return false;
+}
+
+void
+bfa_ioc_sem_release(void __iomem *sem_reg)
+{
+	writel(1, sem_reg);
+}
+
+static void
+bfa_ioc_hw_sem_get(struct bfa_ioc *ioc)
+{
+	u32	r32;
+
+	/**
+	 * First read to the semaphore register will return 0, subsequent reads
+	 * will return 1. Semaphore is released by writing 1 to the register
+	 */
+	r32 = readl(ioc->ioc_regs.ioc_sem_reg);
+	if (r32 == 0) {
+		bfa_fsm_send_event(ioc, IOC_E_SEMLOCKED);
+		return;
+	}
+
+	mod_timer(&ioc->sem_timer, jiffies +
+		msecs_to_jiffies(BFA_IOC_HWSEM_TOV));
+}
+
+void
+bfa_ioc_hw_sem_release(struct bfa_ioc *ioc)
+{
+	writel(1, ioc->ioc_regs.ioc_sem_reg);
+}
+
+static void
+bfa_ioc_hw_sem_get_cancel(struct bfa_ioc *ioc)
+{
+	del_timer(&ioc->sem_timer);
+}
+
+/**
+ * @brief
+ * Initialize LPU local memory (aka secondary memory / SRAM)
+ */
+static void
+bfa_ioc_lmem_init(struct bfa_ioc *ioc)
+{
+	u32	pss_ctl;
+	int		i;
+#define PSS_LMEM_INIT_TIME  10000
+
+	pss_ctl = readl(ioc->ioc_regs.pss_ctl_reg);
+	pss_ctl &= ~__PSS_LMEM_RESET;
+	pss_ctl |= __PSS_LMEM_INIT_EN;
+
+	/*
+	 * i2c workaround 12.5khz clock
+	 */
+	pss_ctl |= __PSS_I2C_CLK_DIV(3UL);
+	writel(pss_ctl, ioc->ioc_regs.pss_ctl_reg);
+
+	/**
+	 * wait for memory initialization to be complete
+	 */
+	i = 0;
+	do {
+		pss_ctl = readl(ioc->ioc_regs.pss_ctl_reg);
+		i++;
+	} while (!(pss_ctl & __PSS_LMEM_INIT_DONE) && (i < PSS_LMEM_INIT_TIME));
+
+	/**
+	 * If memory initialization is not successful, IOC timeout will catch
+	 * such failures.
+	 */
+	BUG_ON(!(pss_ctl & __PSS_LMEM_INIT_DONE));
+
+	pss_ctl &= ~(__PSS_LMEM_INIT_DONE | __PSS_LMEM_INIT_EN);
+	writel(pss_ctl, ioc->ioc_regs.pss_ctl_reg);
+}
+
+static void
+bfa_ioc_lpu_start(struct bfa_ioc *ioc)
+{
+	u32	pss_ctl;
+
+	/**
+	 * Take processor out of reset.
+	 */
+	pss_ctl = readl(ioc->ioc_regs.pss_ctl_reg);
+	pss_ctl &= ~__PSS_LPU0_RESET;
+
+	writel(pss_ctl, ioc->ioc_regs.pss_ctl_reg);
+}
+
+static void
+bfa_ioc_lpu_stop(struct bfa_ioc *ioc)
+{
+	u32	pss_ctl;
+
+	/**
+	 * Put processors in reset.
+	 */
+	pss_ctl = readl(ioc->ioc_regs.pss_ctl_reg);
+	pss_ctl |= (__PSS_LPU0_RESET | __PSS_LPU1_RESET);
+
+	writel(pss_ctl, ioc->ioc_regs.pss_ctl_reg);
+}
+
+/**
+ * Get driver and firmware versions.
+ */
+void
+bfa_ioc_fwver_get(struct bfa_ioc *ioc, struct bfi_ioc_image_hdr *fwhdr)
+{
+	u32	pgnum, pgoff;
+	u32	loff = 0;
+	int		i;
+	u32	*fwsig = (u32 *) fwhdr;
+
+	pgnum = bfa_ioc_smem_pgnum(ioc, loff);
+	pgoff = bfa_ioc_smem_pgoff(ioc, loff);
+	writel(pgnum, ioc->ioc_regs.host_page_num_fn);
+
+	for (i = 0; i < (sizeof(struct bfi_ioc_image_hdr) / sizeof(u32));
+	     i++) {
+		fwsig[i] =
+			swab32(readl((loff) + (ioc->ioc_regs.smem_page_start)));
+		loff += sizeof(u32);
+	}
+}
+
+/**
+ * Returns TRUE if same.
+ */
+bool
+bfa_ioc_fwver_cmp(struct bfa_ioc *ioc, struct bfi_ioc_image_hdr *fwhdr)
+{
+	struct bfi_ioc_image_hdr *drv_fwhdr;
+	int i;
+
+	drv_fwhdr = (struct bfi_ioc_image_hdr *)
+		bfa_cb_image_get_chunk(BFA_IOC_FWIMG_TYPE(ioc), 0);
+
+	for (i = 0; i < BFI_IOC_MD5SUM_SZ; i++) {
+		if (fwhdr->md5sum[i] != drv_fwhdr->md5sum[i])
+			return false;
+	}
+
+	return true;
+}
+
+/**
+ * Return true if current running version is valid. Firmware signature and
+ * execution context (driver/bios) must match.
+ */
+static bool
+bfa_ioc_fwver_valid(struct bfa_ioc *ioc)
+{
+	struct bfi_ioc_image_hdr fwhdr, *drv_fwhdr;
+
+	/**
+	 * If bios/efi boot (flash based) -- return true
+	 */
+	if (bfa_ioc_is_optrom(ioc))
+		return true;
+
+	bfa_ioc_fwver_get(ioc, &fwhdr);
+	drv_fwhdr = (struct bfi_ioc_image_hdr *)
+		bfa_cb_image_get_chunk(BFA_IOC_FWIMG_TYPE(ioc), 0);
+
+	if (fwhdr.signature != drv_fwhdr->signature)
+		return false;
+
+	if (fwhdr.exec != drv_fwhdr->exec)
+		return false;
+
+	return bfa_ioc_fwver_cmp(ioc, &fwhdr);
+}
+
+/**
+ * Conditionally flush any pending message from firmware at start.
+ */
+static void
+bfa_ioc_msgflush(struct bfa_ioc *ioc)
+{
+	u32	r32;
+
+	r32 = readl(ioc->ioc_regs.lpu_mbox_cmd);
+	if (r32)
+		writel(1, ioc->ioc_regs.lpu_mbox_cmd);
+}
+
+/**
+ * @img ioc_init_logic.jpg
+ */
+static void
+bfa_ioc_hwinit(struct bfa_ioc *ioc, bool force)
+{
+	enum bfi_ioc_state ioc_fwstate;
+	bool fwvalid;
+
+	ioc_fwstate = readl(ioc->ioc_regs.ioc_fwstate);
+
+	if (force)
+		ioc_fwstate = BFI_IOC_UNINIT;
+
+	/**
+	 * check if firmware is valid
+	 */
+	fwvalid = (ioc_fwstate == BFI_IOC_UNINIT) ?
+		false : bfa_ioc_fwver_valid(ioc);
+
+	if (!fwvalid) {
+		bfa_ioc_boot(ioc, BFI_BOOT_TYPE_NORMAL, ioc->pcidev.device_id);
+		return;
+	}
+
+	/**
+	 * If hardware initialization is in progress (initialized by other IOC),
+	 * just wait for an initialization completion interrupt.
+	 */
+	if (ioc_fwstate == BFI_IOC_INITING) {
+		ioc->cbfn->reset_cbfn(ioc->bfa);
+		return;
+	}
+
+	/**
+	 * If IOC function is disabled and firmware version is same,
+	 * just re-enable IOC.
+	 *
+	 * If option rom, IOC must not be in operational state. With
+	 * convergence, IOC will be in operational state when 2nd driver
+	 * is loaded.
+	 */
+	if (ioc_fwstate == BFI_IOC_DISABLED ||
+	    (!bfa_ioc_is_optrom(ioc) && ioc_fwstate == BFI_IOC_OP)) {
+		/**
+		 * When using MSI-X any pending firmware ready event should
+		 * be flushed. Otherwise MSI-X interrupts are not delivered.
+		 */
+		bfa_ioc_msgflush(ioc);
+		ioc->cbfn->reset_cbfn(ioc->bfa);
+		bfa_fsm_send_event(ioc, IOC_E_FWREADY);
+		return;
+	}
+
+	/**
+	 * Initialize the h/w for any other states.
+	 */
+	bfa_ioc_boot(ioc, BFI_BOOT_TYPE_NORMAL, ioc->pcidev.device_id);
+}
+
+void
+bfa_ioc_timeout(void *ioc_arg)
+{
+	struct bfa_ioc *ioc = (struct bfa_ioc *) ioc_arg;
+
+	bfa_fsm_send_event(ioc, IOC_E_TIMEOUT);
+}
+
+void
+bfa_ioc_mbox_send(struct bfa_ioc *ioc, void *ioc_msg, int len)
+{
+	u32 *msgp = (u32 *) ioc_msg;
+	u32 i;
+
+	BUG_ON(!(len <= BFI_IOC_MSGLEN_MAX));
+
+	/*
+	 * first write msg to mailbox registers
+	 */
+	for (i = 0; i < len / sizeof(u32); i++)
+		writel(cpu_to_le32(msgp[i]),
+			      ioc->ioc_regs.hfn_mbox + i * sizeof(u32));
+
+	for (; i < BFI_IOC_MSGLEN_MAX / sizeof(u32); i++)
+		writel(0, ioc->ioc_regs.hfn_mbox + i * sizeof(u32));
+
+	/*
+	 * write 1 to mailbox CMD to trigger LPU event
+	 */
+	writel(1, ioc->ioc_regs.hfn_mbox_cmd);
+	(void) readl(ioc->ioc_regs.hfn_mbox_cmd);
+}
+
+static void
+bfa_ioc_send_enable(struct bfa_ioc *ioc)
+{
+	struct bfi_ioc_ctrl_req enable_req;
+	struct timeval tv;
+
+	bfi_h2i_set(enable_req.mh, BFI_MC_IOC, BFI_IOC_H2I_ENABLE_REQ,
+		    bfa_ioc_portid(ioc));
+	enable_req.ioc_class = ioc->ioc_mc;
+	do_gettimeofday(&tv);
+	enable_req.tv_sec = ntohl(tv.tv_sec);
+	bfa_ioc_mbox_send(ioc, &enable_req, sizeof(struct bfi_ioc_ctrl_req));
+}
+
+static void
+bfa_ioc_send_disable(struct bfa_ioc *ioc)
+{
+	struct bfi_ioc_ctrl_req disable_req;
+
+	bfi_h2i_set(disable_req.mh, BFI_MC_IOC, BFI_IOC_H2I_DISABLE_REQ,
+		    bfa_ioc_portid(ioc));
+	bfa_ioc_mbox_send(ioc, &disable_req, sizeof(struct bfi_ioc_ctrl_req));
+}
+
+static void
+bfa_ioc_send_getattr(struct bfa_ioc *ioc)
+{
+	struct bfi_ioc_getattr_req attr_req;
+
+	bfi_h2i_set(attr_req.mh, BFI_MC_IOC, BFI_IOC_H2I_GETATTR_REQ,
+		    bfa_ioc_portid(ioc));
+	bfa_dma_be_addr_set(attr_req.attr_addr, ioc->attr_dma.pa);
+	bfa_ioc_mbox_send(ioc, &attr_req, sizeof(attr_req));
+}
+
+void
+bfa_ioc_hb_check(void *cbarg)
+{
+	struct bfa_ioc *ioc = cbarg;
+	u32	hb_count;
+
+	hb_count = readl(ioc->ioc_regs.heartbeat);
+	if (ioc->hb_count == hb_count) {
+		pr_crit("Firmware heartbeat failure at %d", hb_count);
+		bfa_ioc_recover(ioc);
+		return;
+	} else {
+		ioc->hb_count = hb_count;
+	}
+
+	bfa_ioc_mbox_poll(ioc);
+	mod_timer(&ioc->hb_timer, jiffies +
+		msecs_to_jiffies(BFA_IOC_HB_TOV));
+}
+
+static void
+bfa_ioc_hb_monitor(struct bfa_ioc *ioc)
+{
+	ioc->hb_count = readl(ioc->ioc_regs.heartbeat);
+	mod_timer(&ioc->hb_timer, jiffies +
+		msecs_to_jiffies(BFA_IOC_HB_TOV));
+}
+
+static void
+bfa_ioc_hb_stop(struct bfa_ioc *ioc)
+{
+	del_timer(&ioc->hb_timer);
+}
+
+/**
+ * @brief
+ *	Initiate a full firmware download.
+ */
+static void
+bfa_ioc_download_fw(struct bfa_ioc *ioc, u32 boot_type,
+		    u32 boot_param)
+{
+	u32 *fwimg;
+	u32 pgnum, pgoff;
+	u32 loff = 0;
+	u32 chunkno = 0;
+	u32 i;
+
+	/**
+	 * Initialize LMEM first before code download
+	 */
+	bfa_ioc_lmem_init(ioc);
+
+	/**
+	 * Flash based firmware boot
+	 */
+	if (bfa_ioc_is_optrom(ioc))
+		boot_type = BFI_BOOT_TYPE_FLASH;
+	fwimg = bfa_cb_image_get_chunk(BFA_IOC_FWIMG_TYPE(ioc), chunkno);
+
+	pgnum = bfa_ioc_smem_pgnum(ioc, loff);
+	pgoff = bfa_ioc_smem_pgoff(ioc, loff);
+
+	writel(pgnum, ioc->ioc_regs.host_page_num_fn);
+
+	for (i = 0; i < bfa_cb_image_get_size(BFA_IOC_FWIMG_TYPE(ioc)); i++) {
+		if (BFA_IOC_FLASH_CHUNK_NO(i) != chunkno) {
+			chunkno = BFA_IOC_FLASH_CHUNK_NO(i);
+			fwimg = bfa_cb_image_get_chunk(BFA_IOC_FWIMG_TYPE(ioc),
+					BFA_IOC_FLASH_CHUNK_ADDR(chunkno));
+		}
+
+		/**
+		 * write smem
+		 */
+		writel((swab32(fwimg[BFA_IOC_FLASH_OFFSET_IN_CHUNK(i)])),
+			      ((ioc->ioc_regs.smem_page_start) + (loff)));
+
+		loff += sizeof(u32);
+
+		/**
+		 * handle page offset wrap around
+		 */
+		loff = PSS_SMEM_PGOFF(loff);
+		if (loff == 0) {
+			pgnum++;
+			writel(pgnum,
+				      ioc->ioc_regs.host_page_num_fn);
+		}
+	}
+
+	writel(bfa_ioc_smem_pgnum(ioc, 0),
+		      ioc->ioc_regs.host_page_num_fn);
+
+	/*
+	 * Set boot type and boot param at the end.
+	*/
+	writel((swab32(swab32(boot_type))), ((ioc->ioc_regs.smem_page_start)
+			+ (BFI_BOOT_TYPE_OFF)));
+	writel((swab32(swab32(boot_param))), ((ioc->ioc_regs.smem_page_start)
+			+ (BFI_BOOT_PARAM_OFF)));
+}
+
+static void
+bfa_ioc_reset(struct bfa_ioc *ioc, bool force)
+{
+	bfa_ioc_hwinit(ioc, force);
+}
+
+/**
+ * @brief
+ * Update BFA configuration from firmware configuration.
+ */
+static void
+bfa_ioc_getattr_reply(struct bfa_ioc *ioc)
+{
+	struct bfi_ioc_attr *attr = ioc->attr;
+
+	attr->adapter_prop  = ntohl(attr->adapter_prop);
+	attr->card_type     = ntohl(attr->card_type);
+	attr->maxfrsize	    = ntohs(attr->maxfrsize);
+
+	bfa_fsm_send_event(ioc, IOC_E_FWRSP_GETATTR);
+}
+
+/**
+ * Attach time initialization of mbox logic.
+ */
+static void
+bfa_ioc_mbox_attach(struct bfa_ioc *ioc)
+{
+	struct bfa_ioc_mbox_mod *mod = &ioc->mbox_mod;
+	int	mc;
+
+	INIT_LIST_HEAD(&mod->cmd_q);
+	for (mc = 0; mc < BFI_MC_MAX; mc++) {
+		mod->mbhdlr[mc].cbfn = NULL;
+		mod->mbhdlr[mc].cbarg = ioc->bfa;
+	}
+}
+
+/**
+ * Mbox poll timer -- restarts any pending mailbox requests.
+ */
+static void
+bfa_ioc_mbox_poll(struct bfa_ioc *ioc)
+{
+	struct bfa_ioc_mbox_mod *mod = &ioc->mbox_mod;
+	struct bfa_mbox_cmd *cmd;
+	u32			stat;
+
+	/**
+	 * If no command pending, do nothing
+	 */
+	if (list_empty(&mod->cmd_q))
+		return;
+
+	/**
+	 * If previous command is not yet fetched by firmware, do nothing
+	 */
+	stat = readl(ioc->ioc_regs.hfn_mbox_cmd);
+	if (stat)
+		return;
+
+	/**
+	 * Enqueue command to firmware.
+	 */
+	bfa_q_deq(&mod->cmd_q, &cmd);
+	bfa_ioc_mbox_send(ioc, cmd->msg, sizeof(cmd->msg));
+}
+
+/**
+ * Cleanup any pending requests.
+ */
+static void
+bfa_ioc_mbox_hbfail(struct bfa_ioc *ioc)
+{
+	struct bfa_ioc_mbox_mod *mod = &ioc->mbox_mod;
+	struct bfa_mbox_cmd *cmd;
+
+	while (!list_empty(&mod->cmd_q))
+		bfa_q_deq(&mod->cmd_q, &cmd);
+}
+
+/**
+ * IOC public
+ */
+enum bfa_status
+bfa_ioc_pll_init(struct bfa_ioc *ioc)
+{
+	/*
+	 *  Hold semaphore so that nobody can access the chip during init.
+	 */
+	bfa_ioc_sem_get(ioc->ioc_regs.ioc_init_sem_reg);
+
+	bfa_ioc_pll_init_asic(ioc);
+
+	ioc->pllinit = true;
+	/*
+	 *  release semaphore.
+	 */
+	bfa_ioc_sem_release(ioc->ioc_regs.ioc_init_sem_reg);
+
+	return BFA_STATUS_OK;
+}
+
+/**
+ * Interface used by diag module to do firmware boot with memory test
+ * as the entry vector.
+ */
+void
+bfa_ioc_boot(struct bfa_ioc *ioc, u32 boot_type, u32 boot_param)
+{
+	void __iomem *rb;
+
+	bfa_ioc_stats(ioc, ioc_boots);
+
+	if (bfa_ioc_pll_init(ioc) != BFA_STATUS_OK)
+		return;
+
+	/**
+	 * Initialize IOC state of all functions on a chip reset.
+	 */
+	rb = ioc->pcidev.pci_bar_kva;
+	if (boot_param == BFI_BOOT_TYPE_MEMTEST) {
+		writel(BFI_IOC_MEMTEST, (rb + BFA_IOC0_STATE_REG));
+		writel(BFI_IOC_MEMTEST, (rb + BFA_IOC1_STATE_REG));
+	} else {
+		writel(BFI_IOC_INITING, (rb + BFA_IOC0_STATE_REG));
+		writel(BFI_IOC_INITING, (rb + BFA_IOC1_STATE_REG));
+	}
+
+	bfa_ioc_msgflush(ioc);
+	bfa_ioc_download_fw(ioc, boot_type, boot_param);
+
+	/**
+	 * Enable interrupts just before starting LPU
+	 */
+	ioc->cbfn->reset_cbfn(ioc->bfa);
+	bfa_ioc_lpu_start(ioc);
+}
+
+/**
+ * Enable/disable IOC failure auto recovery.
+ */
+void
+bfa_ioc_auto_recover(bool auto_recover)
+{
+	bfa_auto_recover = auto_recover;
+}
+
+bool
+bfa_ioc_is_operational(struct bfa_ioc *ioc)
+{
+	return bfa_fsm_cmp_state(ioc, bfa_ioc_sm_op);
+}
+
+bool
+bfa_ioc_is_initialized(struct bfa_ioc *ioc)
+{
+	u32 r32 = readl(ioc->ioc_regs.ioc_fwstate);
+
+	return ((r32 != BFI_IOC_UNINIT) &&
+		(r32 != BFI_IOC_INITING) &&
+		(r32 != BFI_IOC_MEMTEST));
+}
+
+void
+bfa_ioc_msgget(struct bfa_ioc *ioc, void *mbmsg)
+{
+	u32	*msgp = mbmsg;
+	u32	r32;
+	int		i;
+
+	/**
+	 * read the MBOX msg
+	 */
+	for (i = 0; i < (sizeof(union bfi_ioc_i2h_msg_u) / sizeof(u32));
+	     i++) {
+		r32 = readl(ioc->ioc_regs.lpu_mbox +
+				   i * sizeof(u32));
+		msgp[i] = htonl(r32);
+	}
+
+	/**
+	 * turn off mailbox interrupt by clearing mailbox status
+	 */
+	writel(1, ioc->ioc_regs.lpu_mbox_cmd);
+	readl(ioc->ioc_regs.lpu_mbox_cmd);
+}
+
+void
+bfa_ioc_isr(struct bfa_ioc *ioc, struct bfi_mbmsg *m)
+{
+	union bfi_ioc_i2h_msg_u	*msg;
+
+	msg = (union bfi_ioc_i2h_msg_u *) m;
+
+	bfa_ioc_stats(ioc, ioc_isrs);
+
+	switch (msg->mh.msg_id) {
+	case BFI_IOC_I2H_HBEAT:
+		break;
+
+	case BFI_IOC_I2H_READY_EVENT:
+		bfa_fsm_send_event(ioc, IOC_E_FWREADY);
+		break;
+
+	case BFI_IOC_I2H_ENABLE_REPLY:
+		bfa_fsm_send_event(ioc, IOC_E_FWRSP_ENABLE);
+		break;
+
+	case BFI_IOC_I2H_DISABLE_REPLY:
+		bfa_fsm_send_event(ioc, IOC_E_FWRSP_DISABLE);
+		break;
+
+	case BFI_IOC_I2H_GETATTR_REPLY:
+		bfa_ioc_getattr_reply(ioc);
+		break;
+
+	default:
+		BUG_ON(1);
+	}
+}
+
+/**
+ * IOC attach time initialization and setup.
+ *
+ * @param[in]	ioc	memory for IOC
+ * @param[in]	bfa	driver instance structure
+ */
+void
+bfa_ioc_attach(struct bfa_ioc *ioc, void *bfa, struct bfa_ioc_cbfn *cbfn)
+{
+	ioc->bfa	= bfa;
+	ioc->cbfn	= cbfn;
+	ioc->fcmode	= false;
+	ioc->pllinit	= false;
+	ioc->dbg_fwsave_once = true;
+
+	bfa_ioc_mbox_attach(ioc);
+	INIT_LIST_HEAD(&ioc->hb_notify_q);
+
+	bfa_fsm_set_state(ioc, bfa_ioc_sm_reset);
+}
+
+/**
+ * Driver detach time IOC cleanup.
+ */
+void
+bfa_ioc_detach(struct bfa_ioc *ioc)
+{
+	bfa_fsm_send_event(ioc, IOC_E_DETACH);
+}
+
+/**
+ * Setup IOC PCI properties.
+ *
+ * @param[in]	pcidev	PCI device information for this IOC
+ */
+void
+bfa_ioc_pci_init(struct bfa_ioc *ioc, struct bfa_pcidev *pcidev,
+		 enum bfi_mclass mc)
+{
+	ioc->ioc_mc	= mc;
+	ioc->pcidev	= *pcidev;
+	ioc->ctdev	= bfa_asic_id_ct(ioc->pcidev.device_id);
+	ioc->cna	= ioc->ctdev && !ioc->fcmode;
+
+	bfa_ioc_set_ct_hwif(ioc);
+
+	bfa_ioc_map_port(ioc);
+	bfa_ioc_reg_init(ioc);
+}
+
+/**
+ * Initialize IOC dma memory
+ *
+ * @param[in]	dm_kva	kernel virtual address of IOC dma memory
+ * @param[in]	dm_pa	physical address of IOC dma memory
+ */
+void
+bfa_ioc_mem_claim(struct bfa_ioc *ioc,  u8 *dm_kva, u64 dm_pa)
+{
+	/**
+	 * dma memory for firmware attribute
+	 */
+	ioc->attr_dma.kva = dm_kva;
+	ioc->attr_dma.pa = dm_pa;
+	ioc->attr = (struct bfi_ioc_attr *) dm_kva;
+}
+
+/**
+ * Return size of dma memory required.
+ */
+u32
+bfa_ioc_meminfo(void)
+{
+	return roundup(sizeof(struct bfi_ioc_attr), BFA_DMA_ALIGN_SZ);
+}
+
+void
+bfa_ioc_enable(struct bfa_ioc *ioc)
+{
+	bfa_ioc_stats(ioc, ioc_enables);
+	ioc->dbg_fwsave_once = true;
+
+	bfa_fsm_send_event(ioc, IOC_E_ENABLE);
+}
+
+void
+bfa_ioc_disable(struct bfa_ioc *ioc)
+{
+	bfa_ioc_stats(ioc, ioc_disables);
+	bfa_fsm_send_event(ioc, IOC_E_DISABLE);
+}
+
+u32
+bfa_ioc_smem_pgnum(struct bfa_ioc *ioc, u32 fmaddr)
+{
+	return PSS_SMEM_PGNUM(ioc->ioc_regs.smem_pg0, fmaddr);
+}
+
+u32
+bfa_ioc_smem_pgoff(struct bfa_ioc *ioc, u32 fmaddr)
+{
+	return PSS_SMEM_PGOFF(fmaddr);
+}
+
+/**
+ * Register mailbox message handler functions
+ *
+ * @param[in]	ioc		IOC instance
+ * @param[in]	mcfuncs		message class handler functions
+ */
+void
+bfa_ioc_mbox_register(struct bfa_ioc *ioc, bfa_ioc_mbox_mcfunc_t *mcfuncs)
+{
+	struct bfa_ioc_mbox_mod *mod = &ioc->mbox_mod;
+	int				mc;
+
+	for (mc = 0; mc < BFI_MC_MAX; mc++)
+		mod->mbhdlr[mc].cbfn = mcfuncs[mc];
+}
+
+/**
+ * Register mailbox message handler function, to be called by common modules
+ */
+void
+bfa_ioc_mbox_regisr(struct bfa_ioc *ioc, enum bfi_mclass mc,
+		    bfa_ioc_mbox_mcfunc_t cbfn, void *cbarg)
+{
+	struct bfa_ioc_mbox_mod *mod = &ioc->mbox_mod;
+
+	mod->mbhdlr[mc].cbfn	= cbfn;
+	mod->mbhdlr[mc].cbarg = cbarg;
+}
+
+/**
+ * Queue a mailbox command request to firmware. Waits if mailbox is busy.
+ * Responsibility of caller to serialize
+ *
+ * @param[in]	ioc	IOC instance
+ * @param[i]	cmd	Mailbox command
+ */
+void
+bfa_ioc_mbox_queue(struct bfa_ioc *ioc, struct bfa_mbox_cmd *cmd)
+{
+	struct bfa_ioc_mbox_mod *mod = &ioc->mbox_mod;
+	u32			stat;
+
+	/**
+	 * If a previous command is pending, queue new command
+	 */
+	if (!list_empty(&mod->cmd_q)) {
+		list_add_tail(&cmd->qe, &mod->cmd_q);
+		return;
+	}
+
+	/**
+	 * If mailbox is busy, queue command for poll timer
+	 */
+	stat = readl(ioc->ioc_regs.hfn_mbox_cmd);
+	if (stat) {
+		list_add_tail(&cmd->qe, &mod->cmd_q);
+		return;
+	}
+
+	/**
+	 * mailbox is free -- queue command to firmware
+	 */
+	bfa_ioc_mbox_send(ioc, cmd->msg, sizeof(cmd->msg));
+}
+
+/**
+ * Handle mailbox interrupts
+ */
+void
+bfa_ioc_mbox_isr(struct bfa_ioc *ioc)
+{
+	struct bfa_ioc_mbox_mod *mod = &ioc->mbox_mod;
+	struct bfi_mbmsg m;
+	int				mc;
+
+	bfa_ioc_msgget(ioc, &m);
+
+	/**
+	 * Treat IOC message class as special.
+	 */
+	mc = m.mh.msg_class;
+	if (mc == BFI_MC_IOC) {
+		bfa_ioc_isr(ioc, &m);
+		return;
+	}
+
+	if ((mc > BFI_MC_MAX) || (mod->mbhdlr[mc].cbfn == NULL))
+		return;
+
+	mod->mbhdlr[mc].cbfn(mod->mbhdlr[mc].cbarg, &m);
+}
+
+void
+bfa_ioc_error_isr(struct bfa_ioc *ioc)
+{
+	bfa_fsm_send_event(ioc, IOC_E_HWERROR);
+}
+
+void
+bfa_ioc_set_fcmode(struct bfa_ioc *ioc)
+{
+	ioc->fcmode  = true;
+	ioc->port_id = bfa_ioc_pcifn(ioc);
+}
+
+/**
+ * return true if IOC is disabled
+ */
+bool
+bfa_ioc_is_disabled(struct bfa_ioc *ioc)
+{
+	return bfa_fsm_cmp_state(ioc, bfa_ioc_sm_disabling) ||
+		bfa_fsm_cmp_state(ioc, bfa_ioc_sm_disabled);
+}
+
+/**
+ * return true if IOC firmware is different.
+ */
+bool
+bfa_ioc_fw_mismatch(struct bfa_ioc *ioc)
+{
+	return bfa_fsm_cmp_state(ioc, bfa_ioc_sm_reset) ||
+		bfa_fsm_cmp_state(ioc, bfa_ioc_sm_fwcheck) ||
+		bfa_fsm_cmp_state(ioc, bfa_ioc_sm_mismatch);
+}
+
+#define bfa_ioc_state_disabled(__sm)		\
+	(((__sm) == BFI_IOC_UNINIT) ||		\
+	 ((__sm) == BFI_IOC_INITING) ||		\
+	 ((__sm) == BFI_IOC_HWINIT) ||		\
+	 ((__sm) == BFI_IOC_DISABLED) ||	\
+	 ((__sm) == BFI_IOC_FAIL) ||		\
+	 ((__sm) == BFI_IOC_CFG_DISABLED))
+
+/**
+ * Check if adapter is disabled -- both IOCs should be in a disabled
+ * state.
+ */
+bool
+bfa_ioc_adapter_is_disabled(struct bfa_ioc *ioc)
+{
+	u32	ioc_state;
+	void __iomem *rb = ioc->pcidev.pci_bar_kva;
+
+	if (!bfa_fsm_cmp_state(ioc, bfa_ioc_sm_disabled))
+		return false;
+
+	ioc_state = readl(rb + BFA_IOC0_STATE_REG);
+	if (!bfa_ioc_state_disabled(ioc_state))
+		return false;
+
+	if (ioc->pcidev.device_id != PCI_DEVICE_ID_BROCADE_FC_8G1P) {
+		ioc_state = readl(rb + BFA_IOC1_STATE_REG);
+		if (!bfa_ioc_state_disabled(ioc_state))
+			return false;
+	}
+
+	return true;
+}
+
+/**
+ * Add to IOC heartbeat failure notification queue. To be used by common
+ * modules such as cee, port, diag.
+ */
+void
+bfa_ioc_hbfail_register(struct bfa_ioc *ioc,
+			struct bfa_ioc_hbfail_notify *notify)
+{
+	list_add_tail(&notify->qe, &ioc->hb_notify_q);
+}
+
+#define BFA_MFG_NAME "Brocade"
+void
+bfa_ioc_get_adapter_attr(struct bfa_ioc *ioc,
+			 struct bfa_adapter_attr *ad_attr)
+{
+	struct bfi_ioc_attr *ioc_attr;
+
+	ioc_attr = ioc->attr;
+
+	bfa_ioc_get_adapter_serial_num(ioc, ad_attr->serial_num);
+	bfa_ioc_get_adapter_fw_ver(ioc, ad_attr->fw_ver);
+	bfa_ioc_get_adapter_optrom_ver(ioc, ad_attr->optrom_ver);
+	bfa_ioc_get_adapter_manufacturer(ioc, ad_attr->manufacturer);
+	memcpy(&ad_attr->vpd, &ioc_attr->vpd,
+		      sizeof(struct bfa_mfg_vpd));
+
+	ad_attr->nports = bfa_ioc_get_nports(ioc);
+	ad_attr->max_speed = bfa_ioc_speed_sup(ioc);
+
+	bfa_ioc_get_adapter_model(ioc, ad_attr->model);
+	/* For now, model descr uses same model string */
+	bfa_ioc_get_adapter_model(ioc, ad_attr->model_descr);
+
+	ad_attr->card_type = ioc_attr->card_type;
+	ad_attr->is_mezz = bfa_mfg_is_mezz(ioc_attr->card_type);
+
+	if (BFI_ADAPTER_IS_SPECIAL(ioc_attr->adapter_prop))
+		ad_attr->prototype = 1;
+	else
+		ad_attr->prototype = 0;
+
+	ad_attr->pwwn = bfa_ioc_get_pwwn(ioc);
+	ad_attr->mac  = bfa_ioc_get_mac(ioc);
+
+	ad_attr->pcie_gen = ioc_attr->pcie_gen;
+	ad_attr->pcie_lanes = ioc_attr->pcie_lanes;
+	ad_attr->pcie_lanes_orig = ioc_attr->pcie_lanes_orig;
+	ad_attr->asic_rev = ioc_attr->asic_rev;
+
+	bfa_ioc_get_pci_chip_rev(ioc, ad_attr->hw_ver);
+
+	ad_attr->cna_capable = ioc->cna;
+	ad_attr->trunk_capable = (ad_attr->nports > 1) && !ioc->cna;
+}
+
+enum bfa_ioc_type
+bfa_ioc_get_type(struct bfa_ioc *ioc)
+{
+	if (!ioc->ctdev || ioc->fcmode)
+		return BFA_IOC_TYPE_FC;
+	else if (ioc->ioc_mc == BFI_MC_IOCFC)
+		return BFA_IOC_TYPE_FCoE;
+	else if (ioc->ioc_mc == BFI_MC_LL)
+		return BFA_IOC_TYPE_LL;
+	else {
+		BUG_ON(!(ioc->ioc_mc == BFI_MC_LL));
+		return BFA_IOC_TYPE_LL;
+	}
+}
+
+void
+bfa_ioc_get_adapter_serial_num(struct bfa_ioc *ioc, char *serial_num)
+{
+	memset(serial_num, 0, BFA_ADAPTER_SERIAL_NUM_LEN);
+	memcpy(serial_num,
+			(void *)ioc->attr->brcd_serialnum,
+			BFA_ADAPTER_SERIAL_NUM_LEN);
+}
+
+void
+bfa_ioc_get_adapter_fw_ver(struct bfa_ioc *ioc, char *fw_ver)
+{
+	memset(fw_ver, 0, BFA_VERSION_LEN);
+	memcpy(fw_ver, ioc->attr->fw_version, BFA_VERSION_LEN);
+}
+
+void
+bfa_ioc_get_pci_chip_rev(struct bfa_ioc *ioc, char *chip_rev)
+{
+	BUG_ON(!(chip_rev));
+
+	memset(chip_rev, 0, BFA_IOC_CHIP_REV_LEN);
+
+	chip_rev[0] = 'R';
+	chip_rev[1] = 'e';
+	chip_rev[2] = 'v';
+	chip_rev[3] = '-';
+	chip_rev[4] = ioc->attr->asic_rev;
+	chip_rev[5] = '\0';
+}
+
+void
+bfa_ioc_get_adapter_optrom_ver(struct bfa_ioc *ioc, char *optrom_ver)
+{
+	memset(optrom_ver, 0, BFA_VERSION_LEN);
+	memcpy(optrom_ver, ioc->attr->optrom_version,
+		      BFA_VERSION_LEN);
+}
+
+void
+bfa_ioc_get_adapter_manufacturer(struct bfa_ioc *ioc, char *manufacturer)
+{
+	memset(manufacturer, 0, BFA_ADAPTER_MFG_NAME_LEN);
+	memcpy(manufacturer, BFA_MFG_NAME, BFA_ADAPTER_MFG_NAME_LEN);
+}
+
+void
+bfa_ioc_get_adapter_model(struct bfa_ioc *ioc, char *model)
+{
+	struct bfi_ioc_attr *ioc_attr;
+
+	BUG_ON(!(model));
+	memset(model, 0, BFA_ADAPTER_MODEL_NAME_LEN);
+
+	ioc_attr = ioc->attr;
+
+	/**
+	 * model name
+	 */
+	snprintf(model, BFA_ADAPTER_MODEL_NAME_LEN, "%s-%u",
+		BFA_MFG_NAME, ioc_attr->card_type);
+}
+
+enum bfa_ioc_state
+bfa_ioc_get_state(struct bfa_ioc *ioc)
+{
+	return bfa_sm_to_state(ioc_sm_table, ioc->fsm);
+}
+
+void
+bfa_ioc_get_attr(struct bfa_ioc *ioc, struct bfa_ioc_attr *ioc_attr)
+{
+	memset((void *)ioc_attr, 0, sizeof(struct bfa_ioc_attr));
+
+	ioc_attr->state = bfa_ioc_get_state(ioc);
+	ioc_attr->port_id = ioc->port_id;
+
+	ioc_attr->ioc_type = bfa_ioc_get_type(ioc);
+
+	bfa_ioc_get_adapter_attr(ioc, &ioc_attr->adapter_attr);
+
+	ioc_attr->pci_attr.device_id = ioc->pcidev.device_id;
+	ioc_attr->pci_attr.pcifn = ioc->pcidev.pci_func;
+	bfa_ioc_get_pci_chip_rev(ioc, ioc_attr->pci_attr.chip_rev);
+}
+
+/**
+ * WWN public
+ */
+u64
+bfa_ioc_get_pwwn(struct bfa_ioc *ioc)
+{
+	return ioc->attr->pwwn;
+}
+
+u64
+bfa_ioc_get_nwwn(struct bfa_ioc *ioc)
+{
+	return ioc->attr->nwwn;
+}
+
+u64
+bfa_ioc_get_adid(struct bfa_ioc *ioc)
+{
+	return ioc->attr->mfg_pwwn;
+}
+
+mac_t
+bfa_ioc_get_mac(struct bfa_ioc *ioc)
+{
+	/*
+	 * Currently mfg mac is used as FCoE enode mac (not configured by PBC)
+	 */
+	if (bfa_ioc_get_type(ioc) == BFA_IOC_TYPE_FCoE)
+		return bfa_ioc_get_mfg_mac(ioc);
+	else
+		return ioc->attr->mac;
+}
+
+u64
+bfa_ioc_get_mfg_pwwn(struct bfa_ioc *ioc)
+{
+	return ioc->attr->mfg_pwwn;
+}
+
+u64
+bfa_ioc_get_mfg_nwwn(struct bfa_ioc *ioc)
+{
+	return ioc->attr->mfg_nwwn;
+}
+
+mac_t
+bfa_ioc_get_mfg_mac(struct bfa_ioc *ioc)
+{
+	mac_t	m;
+
+	m = ioc->attr->mfg_mac;
+	if (bfa_mfg_is_old_wwn_mac_model(ioc->attr->card_type))
+		m.mac[MAC_ADDRLEN - 1] += bfa_ioc_pcifn(ioc);
+	else
+		bfa_mfg_increment_wwn_mac(&(m.mac[MAC_ADDRLEN-3]),
+			bfa_ioc_pcifn(ioc));
+
+	return m;
+}
+
+bool
+bfa_ioc_get_fcmode(struct bfa_ioc *ioc)
+{
+	return ioc->fcmode || !bfa_asic_id_ct(ioc->pcidev.device_id);
+}
+
+/**
+ * Firmware failure detected. Start recovery actions.
+ */
+static void
+bfa_ioc_recover(struct bfa_ioc *ioc)
+{
+	bfa_ioc_stats(ioc, ioc_hbfails);
+	bfa_fsm_send_event(ioc, IOC_E_HBFAIL);
+}
+
+static void
+bfa_ioc_check_attr_wwns(struct bfa_ioc *ioc)
+{
+	if (bfa_ioc_get_type(ioc) == BFA_IOC_TYPE_LL)
+		return;
+
+}
diff --git a/drivers/net/bna/bfa_ioc.h b/drivers/net/bna/bfa_ioc.h
new file mode 100644
index 000000000000..2e5c0adef899
--- /dev/null
+++ b/drivers/net/bna/bfa_ioc.h
@@ -0,0 +1,343 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#ifndef __BFA_IOC_H__
+#define __BFA_IOC_H__
+
+#include "bfa_sm.h"
+#include "bfi.h"
+#include "cna.h"
+
+#define BFA_IOC_TOV		3000	/* msecs */
+#define BFA_IOC_HWSEM_TOV	500	/* msecs */
+#define BFA_IOC_HB_TOV		500	/* msecs */
+#define BFA_IOC_HWINIT_MAX	2
+#define BFA_IOC_TOV_RECOVER	BFA_IOC_HB_TOV
+
+/**
+ * Generic Scatter Gather Element used by driver
+ */
+struct bfa_sge {
+	u32	sg_len;
+	void	*sg_addr;
+};
+
+/**
+ * PCI device information required by IOC
+ */
+struct bfa_pcidev {
+	int	pci_slot;
+	u8	pci_func;
+	u16	device_id;
+	void	__iomem *pci_bar_kva;
+};
+
+/**
+ * Structure used to remember the DMA-able memory block's KVA and Physical
+ * Address
+ */
+struct bfa_dma {
+	void	*kva;	/* ! Kernel virtual address	*/
+	u64	pa;	/* ! Physical address		*/
+};
+
+#define BFA_DMA_ALIGN_SZ	256
+
+/**
+ * smem size for Crossbow and Catapult
+ */
+#define BFI_SMEM_CB_SIZE	0x200000U	/* ! 2MB for crossbow	*/
+#define BFI_SMEM_CT_SIZE	0x280000U	/* ! 2.5MB for catapult	*/
+
+/**
+ * @brief BFA dma address assignment macro
+ */
+#define bfa_dma_addr_set(dma_addr, pa)	\
+		__bfa_dma_addr_set(&dma_addr, (u64)pa)
+
+static inline void
+__bfa_dma_addr_set(union bfi_addr_u *dma_addr, u64 pa)
+{
+	dma_addr->a32.addr_lo = (u32) pa;
+	dma_addr->a32.addr_hi = (u32) (upper_32_bits(pa));
+}
+
+/**
+ * @brief BFA dma address assignment macro. (big endian format)
+ */
+#define bfa_dma_be_addr_set(dma_addr, pa)	\
+		__bfa_dma_be_addr_set(&dma_addr, (u64)pa)
+static inline void
+__bfa_dma_be_addr_set(union bfi_addr_u *dma_addr, u64 pa)
+{
+	dma_addr->a32.addr_lo = (u32) htonl(pa);
+	dma_addr->a32.addr_hi = (u32) htonl(upper_32_bits(pa));
+}
+
+struct bfa_ioc_regs {
+	void __iomem *hfn_mbox_cmd;
+	void __iomem *hfn_mbox;
+	void __iomem *lpu_mbox_cmd;
+	void __iomem *lpu_mbox;
+	void __iomem *pss_ctl_reg;
+	void __iomem *pss_err_status_reg;
+	void __iomem *app_pll_fast_ctl_reg;
+	void __iomem *app_pll_slow_ctl_reg;
+	void __iomem *ioc_sem_reg;
+	void __iomem *ioc_usage_sem_reg;
+	void __iomem *ioc_init_sem_reg;
+	void __iomem *ioc_usage_reg;
+	void __iomem *host_page_num_fn;
+	void __iomem *heartbeat;
+	void __iomem *ioc_fwstate;
+	void __iomem *ll_halt;
+	void __iomem *err_set;
+	void __iomem *shirq_isr_next;
+	void __iomem *shirq_msk_next;
+	void __iomem *smem_page_start;
+	u32	smem_pg0;
+};
+
+/**
+ * IOC Mailbox structures
+ */
+struct bfa_mbox_cmd {
+	struct list_head	qe;
+	u32			msg[BFI_IOC_MSGSZ];
+};
+
+/**
+ * IOC mailbox module
+ */
+typedef void (*bfa_ioc_mbox_mcfunc_t)(void *cbarg, struct bfi_mbmsg *m);
+struct bfa_ioc_mbox_mod {
+	struct list_head	cmd_q;		/*!< pending mbox queue	*/
+	int			nmclass;	/*!< number of handlers */
+	struct {
+		bfa_ioc_mbox_mcfunc_t	cbfn;	/*!< message handlers	*/
+		void			*cbarg;
+	} mbhdlr[BFI_MC_MAX];
+};
+
+/**
+ * IOC callback function interfaces
+ */
+typedef void (*bfa_ioc_enable_cbfn_t)(void *bfa, enum bfa_status status);
+typedef void (*bfa_ioc_disable_cbfn_t)(void *bfa);
+typedef void (*bfa_ioc_hbfail_cbfn_t)(void *bfa);
+typedef void (*bfa_ioc_reset_cbfn_t)(void *bfa);
+struct bfa_ioc_cbfn {
+	bfa_ioc_enable_cbfn_t	enable_cbfn;
+	bfa_ioc_disable_cbfn_t	disable_cbfn;
+	bfa_ioc_hbfail_cbfn_t	hbfail_cbfn;
+	bfa_ioc_reset_cbfn_t	reset_cbfn;
+};
+
+/**
+ * Heartbeat failure notification queue element.
+ */
+struct bfa_ioc_hbfail_notify {
+	struct list_head	qe;
+	bfa_ioc_hbfail_cbfn_t	cbfn;
+	void			*cbarg;
+};
+
+/**
+ * Initialize a heartbeat failure notification structure
+ */
+#define bfa_ioc_hbfail_init(__notify, __cbfn, __cbarg) do {	\
+	(__notify)->cbfn = (__cbfn);				\
+	(__notify)->cbarg = (__cbarg);				\
+} while (0)
+
+struct bfa_ioc {
+	bfa_fsm_t		fsm;
+	struct bfa 		*bfa;
+	struct bfa_pcidev 	pcidev;
+	struct bfa_timer_mod	*timer_mod;
+	struct timer_list 	ioc_timer;
+	struct timer_list 	sem_timer;
+	struct timer_list	hb_timer;
+	u32			hb_count;
+	u32			retry_count;
+	struct list_head	hb_notify_q;
+	void			*dbg_fwsave;
+	int			dbg_fwsave_len;
+	bool			dbg_fwsave_once;
+	enum bfi_mclass		ioc_mc;
+	struct bfa_ioc_regs 	ioc_regs;
+	struct bfa_ioc_drv_stats stats;
+	bool			auto_recover;
+	bool			fcmode;
+	bool			ctdev;
+	bool			cna;
+	bool			pllinit;
+	bool   			stats_busy;	/*!< outstanding stats */
+	u8			port_id;
+
+	struct bfa_dma		attr_dma;
+	struct bfi_ioc_attr	*attr;
+	struct bfa_ioc_cbfn	*cbfn;
+	struct bfa_ioc_mbox_mod	mbox_mod;
+	struct bfa_ioc_hwif	*ioc_hwif;
+};
+
+struct bfa_ioc_hwif {
+	enum bfa_status (*ioc_pll_init) (void __iomem *rb, bool fcmode);
+	bool		(*ioc_firmware_lock)	(struct bfa_ioc *ioc);
+	void		(*ioc_firmware_unlock)	(struct bfa_ioc *ioc);
+	void		(*ioc_reg_init)	(struct bfa_ioc *ioc);
+	void		(*ioc_map_port)	(struct bfa_ioc *ioc);
+	void		(*ioc_isr_mode_set)	(struct bfa_ioc *ioc,
+					bool msix);
+	void		(*ioc_notify_hbfail)	(struct bfa_ioc *ioc);
+	void		(*ioc_ownership_reset)	(struct bfa_ioc *ioc);
+};
+
+#define bfa_ioc_pcifn(__ioc)		((__ioc)->pcidev.pci_func)
+#define bfa_ioc_devid(__ioc)		((__ioc)->pcidev.device_id)
+#define bfa_ioc_bar0(__ioc)		((__ioc)->pcidev.pci_bar_kva)
+#define bfa_ioc_portid(__ioc)		((__ioc)->port_id)
+#define bfa_ioc_fetch_stats(__ioc, __stats) \
+		(((__stats)->drv_stats) = (__ioc)->stats)
+#define bfa_ioc_clr_stats(__ioc)	\
+		memset(&(__ioc)->stats, 0, sizeof((__ioc)->stats))
+#define bfa_ioc_maxfrsize(__ioc)	((__ioc)->attr->maxfrsize)
+#define bfa_ioc_rx_bbcredit(__ioc)	((__ioc)->attr->rx_bbcredit)
+#define bfa_ioc_speed_sup(__ioc)	\
+	BFI_ADAPTER_GETP(SPEED, (__ioc)->attr->adapter_prop)
+#define bfa_ioc_get_nports(__ioc)	\
+	BFI_ADAPTER_GETP(NPORTS, (__ioc)->attr->adapter_prop)
+
+#define bfa_ioc_stats(_ioc, _stats)	((_ioc)->stats._stats++)
+#define BFA_IOC_FWIMG_MINSZ	(16 * 1024)
+#define BFA_IOC_FWIMG_TYPE(__ioc)					\
+	(((__ioc)->ctdev) ? 						\
+	 (((__ioc)->fcmode) ? BFI_IMAGE_CT_FC : BFI_IMAGE_CT_CNA) :	\
+	 BFI_IMAGE_CB_FC)
+#define BFA_IOC_FW_SMEM_SIZE(__ioc)					\
+	(((__ioc)->ctdev) ? BFI_SMEM_CT_SIZE : BFI_SMEM_CB_SIZE)
+#define BFA_IOC_FLASH_CHUNK_NO(off)		(off / BFI_FLASH_CHUNK_SZ_WORDS)
+#define BFA_IOC_FLASH_OFFSET_IN_CHUNK(off)	(off % BFI_FLASH_CHUNK_SZ_WORDS)
+#define BFA_IOC_FLASH_CHUNK_ADDR(chunkno)  (chunkno * BFI_FLASH_CHUNK_SZ_WORDS)
+
+/**
+ * IOC mailbox interface
+ */
+void bfa_ioc_mbox_queue(struct bfa_ioc *ioc, struct bfa_mbox_cmd *cmd);
+void bfa_ioc_mbox_register(struct bfa_ioc *ioc,
+		bfa_ioc_mbox_mcfunc_t *mcfuncs);
+void bfa_ioc_mbox_isr(struct bfa_ioc *ioc);
+void bfa_ioc_mbox_send(struct bfa_ioc *ioc, void *ioc_msg, int len);
+void bfa_ioc_msgget(struct bfa_ioc *ioc, void *mbmsg);
+void bfa_ioc_mbox_regisr(struct bfa_ioc *ioc, enum bfi_mclass mc,
+		bfa_ioc_mbox_mcfunc_t cbfn, void *cbarg);
+
+/**
+ * IOC interfaces
+ */
+
+#define bfa_ioc_pll_init_asic(__ioc) \
+	((__ioc)->ioc_hwif->ioc_pll_init((__ioc)->pcidev.pci_bar_kva, \
+			   (__ioc)->fcmode))
+
+enum bfa_status bfa_ioc_pll_init(struct bfa_ioc *ioc);
+enum bfa_status bfa_ioc_cb_pll_init(void __iomem *rb, bool fcmode);
+enum bfa_status bfa_ioc_ct_pll_init(void __iomem *rb, bool fcmode);
+
+#define	bfa_ioc_isr_mode_set(__ioc, __msix)			\
+			((__ioc)->ioc_hwif->ioc_isr_mode_set(__ioc, __msix))
+#define	bfa_ioc_ownership_reset(__ioc)				\
+			((__ioc)->ioc_hwif->ioc_ownership_reset(__ioc))
+
+void bfa_ioc_set_ct_hwif(struct bfa_ioc *ioc);
+
+void bfa_ioc_attach(struct bfa_ioc *ioc, void *bfa,
+		struct bfa_ioc_cbfn *cbfn);
+void bfa_ioc_auto_recover(bool auto_recover);
+void bfa_ioc_detach(struct bfa_ioc *ioc);
+void bfa_ioc_pci_init(struct bfa_ioc *ioc, struct bfa_pcidev *pcidev,
+		enum bfi_mclass mc);
+u32 bfa_ioc_meminfo(void);
+void bfa_ioc_mem_claim(struct bfa_ioc *ioc,  u8 *dm_kva, u64 dm_pa);
+void bfa_ioc_enable(struct bfa_ioc *ioc);
+void bfa_ioc_disable(struct bfa_ioc *ioc);
+bool bfa_ioc_intx_claim(struct bfa_ioc *ioc);
+
+void bfa_ioc_boot(struct bfa_ioc *ioc, u32 boot_type,
+		u32 boot_param);
+void bfa_ioc_isr(struct bfa_ioc *ioc, struct bfi_mbmsg *msg);
+void bfa_ioc_error_isr(struct bfa_ioc *ioc);
+bool bfa_ioc_is_operational(struct bfa_ioc *ioc);
+bool bfa_ioc_is_initialized(struct bfa_ioc *ioc);
+bool bfa_ioc_is_disabled(struct bfa_ioc *ioc);
+bool bfa_ioc_fw_mismatch(struct bfa_ioc *ioc);
+bool bfa_ioc_adapter_is_disabled(struct bfa_ioc *ioc);
+void bfa_ioc_cfg_complete(struct bfa_ioc *ioc);
+enum bfa_ioc_type bfa_ioc_get_type(struct bfa_ioc *ioc);
+void bfa_ioc_get_adapter_serial_num(struct bfa_ioc *ioc, char *serial_num);
+void bfa_ioc_get_adapter_fw_ver(struct bfa_ioc *ioc, char *fw_ver);
+void bfa_ioc_get_adapter_optrom_ver(struct bfa_ioc *ioc, char *optrom_ver);
+void bfa_ioc_get_adapter_model(struct bfa_ioc *ioc, char *model);
+void bfa_ioc_get_adapter_manufacturer(struct bfa_ioc *ioc,
+		char *manufacturer);
+void bfa_ioc_get_pci_chip_rev(struct bfa_ioc *ioc, char *chip_rev);
+enum bfa_ioc_state bfa_ioc_get_state(struct bfa_ioc *ioc);
+
+void bfa_ioc_get_attr(struct bfa_ioc *ioc, struct bfa_ioc_attr *ioc_attr);
+void bfa_ioc_get_adapter_attr(struct bfa_ioc *ioc,
+		struct bfa_adapter_attr *ad_attr);
+u32 bfa_ioc_smem_pgnum(struct bfa_ioc *ioc, u32 fmaddr);
+u32 bfa_ioc_smem_pgoff(struct bfa_ioc *ioc, u32 fmaddr);
+void bfa_ioc_set_fcmode(struct bfa_ioc *ioc);
+bool bfa_ioc_get_fcmode(struct bfa_ioc *ioc);
+void bfa_ioc_hbfail_register(struct bfa_ioc *ioc,
+	struct bfa_ioc_hbfail_notify *notify);
+bool bfa_ioc_sem_get(void __iomem *sem_reg);
+void bfa_ioc_sem_release(void __iomem *sem_reg);
+void bfa_ioc_hw_sem_release(struct bfa_ioc *ioc);
+void bfa_ioc_fwver_get(struct bfa_ioc *ioc,
+			struct bfi_ioc_image_hdr *fwhdr);
+bool bfa_ioc_fwver_cmp(struct bfa_ioc *ioc,
+			struct bfi_ioc_image_hdr *fwhdr);
+
+/*
+ * Timeout APIs
+ */
+void bfa_ioc_timeout(void *ioc);
+void bfa_ioc_hb_check(void *ioc);
+void bfa_ioc_sem_timeout(void *ioc);
+
+/*
+ * bfa mfg wwn API functions
+ */
+u64 bfa_ioc_get_pwwn(struct bfa_ioc *ioc);
+u64 bfa_ioc_get_nwwn(struct bfa_ioc *ioc);
+mac_t bfa_ioc_get_mac(struct bfa_ioc *ioc);
+u64 bfa_ioc_get_mfg_pwwn(struct bfa_ioc *ioc);
+u64 bfa_ioc_get_mfg_nwwn(struct bfa_ioc *ioc);
+mac_t bfa_ioc_get_mfg_mac(struct bfa_ioc *ioc);
+u64 bfa_ioc_get_adid(struct bfa_ioc *ioc);
+
+/*
+ * F/W Image Size & Chunk
+ */
+u32 *bfa_cb_image_get_chunk(int type, u32 off);
+u32 bfa_cb_image_get_size(int type);
+
+#endif /* __BFA_IOC_H__ */
diff --git a/drivers/net/bna/bfa_ioc_ct.c b/drivers/net/bna/bfa_ioc_ct.c
new file mode 100644
index 000000000000..870046e32c8d
--- /dev/null
+++ b/drivers/net/bna/bfa_ioc_ct.c
@@ -0,0 +1,391 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#include "bfa_ioc.h"
+#include "cna.h"
+#include "bfi.h"
+#include "bfi_ctreg.h"
+#include "bfa_defs.h"
+
+/*
+ * forward declarations
+ */
+static bool bfa_ioc_ct_firmware_lock(struct bfa_ioc *ioc);
+static void bfa_ioc_ct_firmware_unlock(struct bfa_ioc *ioc);
+static void bfa_ioc_ct_reg_init(struct bfa_ioc *ioc);
+static void bfa_ioc_ct_map_port(struct bfa_ioc *ioc);
+static void bfa_ioc_ct_isr_mode_set(struct bfa_ioc *ioc, bool msix);
+static void bfa_ioc_ct_notify_hbfail(struct bfa_ioc *ioc);
+static void bfa_ioc_ct_ownership_reset(struct bfa_ioc *ioc);
+
+struct bfa_ioc_hwif hwif_ct;
+
+/**
+ * Called from bfa_ioc_attach() to map asic specific calls.
+ */
+void
+bfa_ioc_set_ct_hwif(struct bfa_ioc *ioc)
+{
+	hwif_ct.ioc_pll_init = bfa_ioc_ct_pll_init;
+	hwif_ct.ioc_firmware_lock = bfa_ioc_ct_firmware_lock;
+	hwif_ct.ioc_firmware_unlock = bfa_ioc_ct_firmware_unlock;
+	hwif_ct.ioc_reg_init = bfa_ioc_ct_reg_init;
+	hwif_ct.ioc_map_port = bfa_ioc_ct_map_port;
+	hwif_ct.ioc_isr_mode_set = bfa_ioc_ct_isr_mode_set;
+	hwif_ct.ioc_notify_hbfail = bfa_ioc_ct_notify_hbfail;
+	hwif_ct.ioc_ownership_reset = bfa_ioc_ct_ownership_reset;
+
+	ioc->ioc_hwif = &hwif_ct;
+}
+
+/**
+ * Return true if firmware of current driver matches the running firmware.
+ */
+static bool
+bfa_ioc_ct_firmware_lock(struct bfa_ioc *ioc)
+{
+	enum bfi_ioc_state ioc_fwstate;
+	u32 usecnt;
+	struct bfi_ioc_image_hdr fwhdr;
+
+	/**
+	 * Firmware match check is relevant only for CNA.
+	 */
+	if (!ioc->cna)
+		return true;
+
+	/**
+	 * If bios boot (flash based) -- do not increment usage count
+	 */
+	if (bfa_cb_image_get_size(BFA_IOC_FWIMG_TYPE(ioc)) <
+						BFA_IOC_FWIMG_MINSZ)
+		return true;
+
+	bfa_ioc_sem_get(ioc->ioc_regs.ioc_usage_sem_reg);
+	usecnt = readl(ioc->ioc_regs.ioc_usage_reg);
+
+	/**
+	 * If usage count is 0, always return TRUE.
+	 */
+	if (usecnt == 0) {
+		writel(1, ioc->ioc_regs.ioc_usage_reg);
+		bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
+		return true;
+	}
+
+	ioc_fwstate = readl(ioc->ioc_regs.ioc_fwstate);
+
+	/**
+	 * Use count cannot be non-zero and chip in uninitialized state.
+	 */
+	BUG_ON(!(ioc_fwstate != BFI_IOC_UNINIT));
+
+	/**
+	 * Check if another driver with a different firmware is active
+	 */
+	bfa_ioc_fwver_get(ioc, &fwhdr);
+	if (!bfa_ioc_fwver_cmp(ioc, &fwhdr)) {
+		bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
+		return false;
+	}
+
+	/**
+	 * Same firmware version. Increment the reference count.
+	 */
+	usecnt++;
+	writel(usecnt, ioc->ioc_regs.ioc_usage_reg);
+	bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
+	return true;
+}
+
+static void
+bfa_ioc_ct_firmware_unlock(struct bfa_ioc *ioc)
+{
+	u32 usecnt;
+
+	/**
+	 * Firmware lock is relevant only for CNA.
+	 */
+	if (!ioc->cna)
+		return;
+
+	/**
+	 * If bios boot (flash based) -- do not decrement usage count
+	 */
+	if (bfa_cb_image_get_size(BFA_IOC_FWIMG_TYPE(ioc)) <
+						BFA_IOC_FWIMG_MINSZ)
+		return;
+
+	/**
+	 * decrement usage count
+	 */
+	bfa_ioc_sem_get(ioc->ioc_regs.ioc_usage_sem_reg);
+	usecnt = readl(ioc->ioc_regs.ioc_usage_reg);
+	BUG_ON(!(usecnt > 0));
+
+	usecnt--;
+	writel(usecnt, ioc->ioc_regs.ioc_usage_reg);
+
+	bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
+}
+
+/**
+ * Notify other functions on HB failure.
+ */
+static void
+bfa_ioc_ct_notify_hbfail(struct bfa_ioc *ioc)
+{
+	if (ioc->cna) {
+		writel(__FW_INIT_HALT_P, ioc->ioc_regs.ll_halt);
+		/* Wait for halt to take effect */
+		readl(ioc->ioc_regs.ll_halt);
+	} else {
+		writel(__PSS_ERR_STATUS_SET, ioc->ioc_regs.err_set);
+		readl(ioc->ioc_regs.err_set);
+	}
+}
+
+/**
+ * Host to LPU mailbox message addresses
+ */
+static struct { u32 hfn_mbox, lpu_mbox, hfn_pgn; } iocreg_fnreg[] = {
+	{ HOSTFN0_LPU_MBOX0_0, LPU_HOSTFN0_MBOX0_0, HOST_PAGE_NUM_FN0 },
+	{ HOSTFN1_LPU_MBOX0_8, LPU_HOSTFN1_MBOX0_8, HOST_PAGE_NUM_FN1 },
+	{ HOSTFN2_LPU_MBOX0_0, LPU_HOSTFN2_MBOX0_0, HOST_PAGE_NUM_FN2 },
+	{ HOSTFN3_LPU_MBOX0_8, LPU_HOSTFN3_MBOX0_8, HOST_PAGE_NUM_FN3 }
+};
+
+/**
+ * Host <-> LPU mailbox command/status registers - port 0
+ */
+static struct { u32 hfn, lpu; } iocreg_mbcmd_p0[] = {
+	{ HOSTFN0_LPU0_MBOX0_CMD_STAT, LPU0_HOSTFN0_MBOX0_CMD_STAT },
+	{ HOSTFN1_LPU0_MBOX0_CMD_STAT, LPU0_HOSTFN1_MBOX0_CMD_STAT },
+	{ HOSTFN2_LPU0_MBOX0_CMD_STAT, LPU0_HOSTFN2_MBOX0_CMD_STAT },
+	{ HOSTFN3_LPU0_MBOX0_CMD_STAT, LPU0_HOSTFN3_MBOX0_CMD_STAT }
+};
+
+/**
+ * Host <-> LPU mailbox command/status registers - port 1
+ */
+static struct { u32 hfn, lpu; } iocreg_mbcmd_p1[] = {
+	{ HOSTFN0_LPU1_MBOX0_CMD_STAT, LPU1_HOSTFN0_MBOX0_CMD_STAT },
+	{ HOSTFN1_LPU1_MBOX0_CMD_STAT, LPU1_HOSTFN1_MBOX0_CMD_STAT },
+	{ HOSTFN2_LPU1_MBOX0_CMD_STAT, LPU1_HOSTFN2_MBOX0_CMD_STAT },
+	{ HOSTFN3_LPU1_MBOX0_CMD_STAT, LPU1_HOSTFN3_MBOX0_CMD_STAT }
+};
+
+static void
+bfa_ioc_ct_reg_init(struct bfa_ioc *ioc)
+{
+	void __iomem *rb;
+	int		pcifn = bfa_ioc_pcifn(ioc);
+
+	rb = bfa_ioc_bar0(ioc);
+
+	ioc->ioc_regs.hfn_mbox = rb + iocreg_fnreg[pcifn].hfn_mbox;
+	ioc->ioc_regs.lpu_mbox = rb + iocreg_fnreg[pcifn].lpu_mbox;
+	ioc->ioc_regs.host_page_num_fn = rb + iocreg_fnreg[pcifn].hfn_pgn;
+
+	if (ioc->port_id == 0) {
+		ioc->ioc_regs.heartbeat = rb + BFA_IOC0_HBEAT_REG;
+		ioc->ioc_regs.ioc_fwstate = rb + BFA_IOC0_STATE_REG;
+		ioc->ioc_regs.hfn_mbox_cmd = rb + iocreg_mbcmd_p0[pcifn].hfn;
+		ioc->ioc_regs.lpu_mbox_cmd = rb + iocreg_mbcmd_p0[pcifn].lpu;
+		ioc->ioc_regs.ll_halt = rb + FW_INIT_HALT_P0;
+	} else {
+		ioc->ioc_regs.heartbeat = (rb + BFA_IOC1_HBEAT_REG);
+		ioc->ioc_regs.ioc_fwstate = (rb + BFA_IOC1_STATE_REG);
+		ioc->ioc_regs.hfn_mbox_cmd = rb + iocreg_mbcmd_p1[pcifn].hfn;
+		ioc->ioc_regs.lpu_mbox_cmd = rb + iocreg_mbcmd_p1[pcifn].lpu;
+		ioc->ioc_regs.ll_halt = rb + FW_INIT_HALT_P1;
+	}
+
+	/*
+	 * PSS control registers
+	 */
+	ioc->ioc_regs.pss_ctl_reg = (rb + PSS_CTL_REG);
+	ioc->ioc_regs.pss_err_status_reg = (rb + PSS_ERR_STATUS_REG);
+	ioc->ioc_regs.app_pll_fast_ctl_reg = (rb + APP_PLL_425_CTL_REG);
+	ioc->ioc_regs.app_pll_slow_ctl_reg = (rb + APP_PLL_312_CTL_REG);
+
+	/*
+	 * IOC semaphore registers and serialization
+	 */
+	ioc->ioc_regs.ioc_sem_reg = (rb + HOST_SEM0_REG);
+	ioc->ioc_regs.ioc_usage_sem_reg = (rb + HOST_SEM1_REG);
+	ioc->ioc_regs.ioc_init_sem_reg = (rb + HOST_SEM2_REG);
+	ioc->ioc_regs.ioc_usage_reg = (rb + BFA_FW_USE_COUNT);
+
+	/**
+	 * sram memory access
+	 */
+	ioc->ioc_regs.smem_page_start = (rb + PSS_SMEM_PAGE_START);
+	ioc->ioc_regs.smem_pg0 = BFI_IOC_SMEM_PG0_CT;
+
+	/*
+	 * err set reg : for notification of hb failure in fcmode
+	 */
+	ioc->ioc_regs.err_set = (rb + ERR_SET_REG);
+}
+
+/**
+ * Initialize IOC to port mapping.
+ */
+
+#define FNC_PERS_FN_SHIFT(__fn)	((__fn) * 8)
+static void
+bfa_ioc_ct_map_port(struct bfa_ioc *ioc)
+{
+	void __iomem *rb = ioc->pcidev.pci_bar_kva;
+	u32	r32;
+
+	/**
+	 * For catapult, base port id on personality register and IOC type
+	 */
+	r32 = readl(rb + FNC_PERS_REG);
+	r32 >>= FNC_PERS_FN_SHIFT(bfa_ioc_pcifn(ioc));
+	ioc->port_id = (r32 & __F0_PORT_MAP_MK) >> __F0_PORT_MAP_SH;
+
+}
+
+/**
+ * Set interrupt mode for a function: INTX or MSIX
+ */
+static void
+bfa_ioc_ct_isr_mode_set(struct bfa_ioc *ioc, bool msix)
+{
+	void __iomem *rb = ioc->pcidev.pci_bar_kva;
+	u32	r32, mode;
+
+	r32 = readl(rb + FNC_PERS_REG);
+
+	mode = (r32 >> FNC_PERS_FN_SHIFT(bfa_ioc_pcifn(ioc))) &
+		__F0_INTX_STATUS;
+
+	/**
+	 * If already in desired mode, do not change anything
+	 */
+	if (!msix && mode)
+		return;
+
+	if (msix)
+		mode = __F0_INTX_STATUS_MSIX;
+	else
+		mode = __F0_INTX_STATUS_INTA;
+
+	r32 &= ~(__F0_INTX_STATUS << FNC_PERS_FN_SHIFT(bfa_ioc_pcifn(ioc)));
+	r32 |= (mode << FNC_PERS_FN_SHIFT(bfa_ioc_pcifn(ioc)));
+
+	writel(r32, rb + FNC_PERS_REG);
+}
+
+/**
+ * Cleanup hw semaphore and usecnt registers
+ */
+static void
+bfa_ioc_ct_ownership_reset(struct bfa_ioc *ioc)
+{
+	if (ioc->cna) {
+		bfa_ioc_sem_get(ioc->ioc_regs.ioc_usage_sem_reg);
+		writel(0, ioc->ioc_regs.ioc_usage_reg);
+		bfa_ioc_sem_release(ioc->ioc_regs.ioc_usage_sem_reg);
+	}
+
+	/*
+	 * Read the hw sem reg to make sure that it is locked
+	 * before we clear it. If it is not locked, writing 1
+	 * will lock it instead of clearing it.
+	 */
+	readl(ioc->ioc_regs.ioc_sem_reg);
+	bfa_ioc_hw_sem_release(ioc);
+}
+
+enum bfa_status
+bfa_ioc_ct_pll_init(void __iomem *rb, bool fcmode)
+{
+	u32	pll_sclk, pll_fclk, r32;
+
+	pll_sclk = __APP_PLL_312_LRESETN | __APP_PLL_312_ENARST |
+		__APP_PLL_312_RSEL200500 | __APP_PLL_312_P0_1(3U) |
+		__APP_PLL_312_JITLMT0_1(3U) |
+		__APP_PLL_312_CNTLMT0_1(1U);
+	pll_fclk = __APP_PLL_425_LRESETN | __APP_PLL_425_ENARST |
+		__APP_PLL_425_RSEL200500 | __APP_PLL_425_P0_1(3U) |
+		__APP_PLL_425_JITLMT0_1(3U) |
+		__APP_PLL_425_CNTLMT0_1(1U);
+	if (fcmode) {
+		writel(0, (rb + OP_MODE));
+		writel(__APP_EMS_CMLCKSEL |
+				__APP_EMS_REFCKBUFEN2 |
+				__APP_EMS_CHANNEL_SEL,
+				(rb + ETH_MAC_SER_REG));
+	} else {
+		writel(__GLOBAL_FCOE_MODE, (rb + OP_MODE));
+		writel(__APP_EMS_REFCKBUFEN1,
+				(rb + ETH_MAC_SER_REG));
+	}
+	writel(BFI_IOC_UNINIT, (rb + BFA_IOC0_STATE_REG));
+	writel(BFI_IOC_UNINIT, (rb + BFA_IOC1_STATE_REG));
+	writel(0xffffffffU, (rb + HOSTFN0_INT_MSK));
+	writel(0xffffffffU, (rb + HOSTFN1_INT_MSK));
+	writel(0xffffffffU, (rb + HOSTFN0_INT_STATUS));
+	writel(0xffffffffU, (rb + HOSTFN1_INT_STATUS));
+	writel(0xffffffffU, (rb + HOSTFN0_INT_MSK));
+	writel(0xffffffffU, (rb + HOSTFN1_INT_MSK));
+	writel(pll_sclk |
+		__APP_PLL_312_LOGIC_SOFT_RESET,
+		rb + APP_PLL_312_CTL_REG);
+	writel(pll_fclk |
+		__APP_PLL_425_LOGIC_SOFT_RESET,
+		rb + APP_PLL_425_CTL_REG);
+	writel(pll_sclk |
+		__APP_PLL_312_LOGIC_SOFT_RESET | __APP_PLL_312_ENABLE,
+		rb + APP_PLL_312_CTL_REG);
+	writel(pll_fclk |
+		__APP_PLL_425_LOGIC_SOFT_RESET | __APP_PLL_425_ENABLE,
+		rb + APP_PLL_425_CTL_REG);
+	readl(rb + HOSTFN0_INT_MSK);
+	udelay(2000);
+	writel(0xffffffffU, (rb + HOSTFN0_INT_STATUS));
+	writel(0xffffffffU, (rb + HOSTFN1_INT_STATUS));
+	writel(pll_sclk |
+		__APP_PLL_312_ENABLE,
+		rb + APP_PLL_312_CTL_REG);
+	writel(pll_fclk |
+		__APP_PLL_425_ENABLE,
+		rb + APP_PLL_425_CTL_REG);
+	if (!fcmode) {
+		writel(__PMM_1T_RESET_P, (rb + PMM_1T_RESET_REG_P0));
+		writel(__PMM_1T_RESET_P, (rb + PMM_1T_RESET_REG_P1));
+	}
+	r32 = readl((rb + PSS_CTL_REG));
+	r32 &= ~__PSS_LMEM_RESET;
+	writel(r32, (rb + PSS_CTL_REG));
+	udelay(1000);
+	if (!fcmode) {
+		writel(0, (rb + PMM_1T_RESET_REG_P0));
+		writel(0, (rb + PMM_1T_RESET_REG_P1));
+	}
+
+	writel(__EDRAM_BISTR_START, (rb + MBIST_CTL_REG));
+	udelay(1000);
+	r32 = readl((rb + MBIST_STAT_REG));
+	writel(0, (rb + MBIST_CTL_REG));
+	return BFA_STATUS_OK;
+}
diff --git a/drivers/net/bna/bfa_sm.h b/drivers/net/bna/bfa_sm.h
new file mode 100644
index 000000000000..1d3d975d6f68
--- /dev/null
+++ b/drivers/net/bna/bfa_sm.h
@@ -0,0 +1,88 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+/**
+ * @file bfasm.h State machine defines
+ */
+
+#ifndef __BFA_SM_H__
+#define __BFA_SM_H__
+
+#include "cna.h"
+
+typedef void (*bfa_sm_t)(void *sm, int event);
+
+/**
+ * oc - object class eg. bfa_ioc
+ * st - state, eg. reset
+ * otype - object type, eg. struct bfa_ioc
+ * etype - object type, eg. enum ioc_event
+ */
+#define bfa_sm_state_decl(oc, st, otype, etype)		\
+	static void oc ## _sm_ ## st(otype * fsm, etype event)
+
+#define bfa_sm_set_state(_sm, _state)	((_sm)->sm = (bfa_sm_t)(_state))
+#define bfa_sm_send_event(_sm, _event)	((_sm)->sm((_sm), (_event)))
+#define bfa_sm_get_state(_sm)		((_sm)->sm)
+#define bfa_sm_cmp_state(_sm, _state)	((_sm)->sm == (bfa_sm_t)(_state))
+
+/**
+ * For converting from state machine function to state encoding.
+ */
+struct bfa_sm_table {
+	bfa_sm_t	sm;	/*!< state machine function	*/
+	int		state;	/*!< state machine encoding	*/
+	char		*name;	/*!< state name for display	*/
+};
+#define BFA_SM(_sm)	((bfa_sm_t)(_sm))
+
+/**
+ * State machine with entry actions.
+ */
+typedef void (*bfa_fsm_t)(void *fsm, int event);
+
+/**
+ * oc - object class eg. bfa_ioc
+ * st - state, eg. reset
+ * otype - object type, eg. struct bfa_ioc
+ * etype - object type, eg. enum ioc_event
+ */
+#define bfa_fsm_state_decl(oc, st, otype, etype)		\
+	static void oc ## _sm_ ## st(otype * fsm, etype event);	\
+	static void oc ## _sm_ ## st ## _entry(otype * fsm)
+
+#define bfa_fsm_set_state(_fsm, _state) do {	\
+	(_fsm)->fsm = (bfa_fsm_t)(_state);	\
+	_state ## _entry(_fsm);			\
+} while (0)
+
+#define bfa_fsm_send_event(_fsm, _event)	((_fsm)->fsm((_fsm), (_event)))
+#define bfa_fsm_get_state(_fsm)			((_fsm)->fsm)
+#define bfa_fsm_cmp_state(_fsm, _state)		\
+	((_fsm)->fsm == (bfa_fsm_t)(_state))
+
+static inline int
+bfa_sm_to_state(struct bfa_sm_table *smt, bfa_sm_t sm)
+{
+	int	i = 0;
+
+	while (smt[i].sm && smt[i].sm != sm)
+		i++;
+	return smt[i].state;
+}
+#endif
diff --git a/drivers/net/bna/bfa_wc.h b/drivers/net/bna/bfa_wc.h
new file mode 100644
index 000000000000..d0e4caee67b0
--- /dev/null
+++ b/drivers/net/bna/bfa_wc.h
@@ -0,0 +1,69 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+/**
+ * @file bfa_wc.h Generic wait counter.
+ */
+
+#ifndef __BFA_WC_H__
+#define __BFA_WC_H__
+
+typedef void (*bfa_wc_resume_t) (void *cbarg);
+
+struct bfa_wc {
+	bfa_wc_resume_t wc_resume;
+	void		*wc_cbarg;
+	int		wc_count;
+};
+
+static inline void
+bfa_wc_up(struct bfa_wc *wc)
+{
+	wc->wc_count++;
+}
+
+static inline void
+bfa_wc_down(struct bfa_wc *wc)
+{
+	wc->wc_count--;
+	if (wc->wc_count == 0)
+		wc->wc_resume(wc->wc_cbarg);
+}
+
+/**
+ * Initialize a waiting counter.
+ */
+static inline void
+bfa_wc_init(struct bfa_wc *wc, bfa_wc_resume_t wc_resume, void *wc_cbarg)
+{
+	wc->wc_resume = wc_resume;
+	wc->wc_cbarg = wc_cbarg;
+	wc->wc_count = 0;
+	bfa_wc_up(wc);
+}
+
+/**
+ * Wait for counter to reach zero
+ */
+static inline void
+bfa_wc_wait(struct bfa_wc *wc)
+{
+	bfa_wc_down(wc);
+}
+
+#endif
diff --git a/drivers/net/bna/bfi.h b/drivers/net/bna/bfi.h
new file mode 100644
index 000000000000..a97396811050
--- /dev/null
+++ b/drivers/net/bna/bfi.h
@@ -0,0 +1,392 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#ifndef __BFI_H__
+#define __BFI_H__
+
+#include "bfa_defs.h"
+
+#pragma pack(1)
+
+/**
+ * BFI FW image type
+ */
+#define	BFI_FLASH_CHUNK_SZ			256	/*!< Flash chunk size */
+#define	BFI_FLASH_CHUNK_SZ_WORDS	(BFI_FLASH_CHUNK_SZ/sizeof(u32))
+enum {
+	BFI_IMAGE_CB_FC,
+	BFI_IMAGE_CT_FC,
+	BFI_IMAGE_CT_CNA,
+	BFI_IMAGE_MAX,
+};
+
+/**
+ * Msg header common to all msgs
+ */
+struct bfi_mhdr {
+	u8		msg_class;	/*!< @ref enum bfi_mclass	    */
+	u8		msg_id;		/*!< msg opcode with in the class   */
+	union {
+		struct {
+			u8	rsvd;
+			u8	lpu_id;	/*!< msg destination		    */
+		} h2i;
+		u16	i2htok;	/*!< token in msgs to host	    */
+	} mtag;
+};
+
+#define bfi_h2i_set(_mh, _mc, _op, _lpuid) do {		\
+	(_mh).msg_class 		= (_mc);		\
+	(_mh).msg_id			= (_op);		\
+	(_mh).mtag.h2i.lpu_id	= (_lpuid);			\
+} while (0)
+
+#define bfi_i2h_set(_mh, _mc, _op, _i2htok) do {		\
+	(_mh).msg_class 		= (_mc);		\
+	(_mh).msg_id			= (_op);		\
+	(_mh).mtag.i2htok		= (_i2htok);		\
+} while (0)
+
+/*
+ * Message opcodes: 0-127 to firmware, 128-255 to host
+ */
+#define BFI_I2H_OPCODE_BASE	128
+#define BFA_I2HM(_x) 			((_x) + BFI_I2H_OPCODE_BASE)
+
+/**
+ ****************************************************************************
+ *
+ * Scatter Gather Element and Page definition
+ *
+ ****************************************************************************
+ */
+
+#define BFI_SGE_INLINE	1
+#define BFI_SGE_INLINE_MAX	(BFI_SGE_INLINE + 1)
+
+/**
+ * SG Flags
+ */
+enum {
+	BFI_SGE_DATA		= 0,	/*!< data address, not last	     */
+	BFI_SGE_DATA_CPL	= 1,	/*!< data addr, last in current page */
+	BFI_SGE_DATA_LAST	= 3,	/*!< data address, last		     */
+	BFI_SGE_LINK		= 2,	/*!< link address		     */
+	BFI_SGE_PGDLEN		= 2,	/*!< cumulative data length for page */
+};
+
+/**
+ * DMA addresses
+ */
+union bfi_addr_u {
+	struct {
+		u32	addr_lo;
+		u32	addr_hi;
+	} a32;
+};
+
+/**
+ * Scatter Gather Element
+ */
+struct bfi_sge {
+#ifdef __BIGENDIAN
+	u32	flags:2,
+			rsvd:2,
+			sg_len:28;
+#else
+	u32	sg_len:28,
+			rsvd:2,
+			flags:2;
+#endif
+	union bfi_addr_u sga;
+};
+
+/**
+ * Scatter Gather Page
+ */
+#define BFI_SGPG_DATA_SGES		7
+#define BFI_SGPG_SGES_MAX		(BFI_SGPG_DATA_SGES + 1)
+#define BFI_SGPG_RSVD_WD_LEN	8
+struct bfi_sgpg {
+	struct bfi_sge sges[BFI_SGPG_SGES_MAX];
+	u32	rsvd[BFI_SGPG_RSVD_WD_LEN];
+};
+
+/*
+ * Large Message structure - 128 Bytes size Msgs
+ */
+#define BFI_LMSG_SZ		128
+#define BFI_LMSG_PL_WSZ	\
+			((BFI_LMSG_SZ - sizeof(struct bfi_mhdr)) / 4)
+
+struct bfi_msg {
+	struct bfi_mhdr mhdr;
+	u32	pl[BFI_LMSG_PL_WSZ];
+};
+
+/**
+ * Mailbox message structure
+ */
+#define BFI_MBMSG_SZ		7
+struct bfi_mbmsg {
+	struct bfi_mhdr mh;
+	u32		pl[BFI_MBMSG_SZ];
+};
+
+/**
+ * Message Classes
+ */
+enum bfi_mclass {
+	BFI_MC_IOC		= 1,	/*!< IO Controller (IOC)	    */
+	BFI_MC_DIAG		= 2,	/*!< Diagnostic Msgs		    */
+	BFI_MC_FLASH		= 3,	/*!< Flash message class	    */
+	BFI_MC_CEE		= 4,	/*!< CEE			    */
+	BFI_MC_FCPORT		= 5,	/*!< FC port			    */
+	BFI_MC_IOCFC		= 6,	/*!< FC - IO Controller (IOC)	    */
+	BFI_MC_LL		= 7,	/*!< Link Layer			    */
+	BFI_MC_UF		= 8,	/*!< Unsolicited frame receive	    */
+	BFI_MC_FCXP		= 9,	/*!< FC Transport		    */
+	BFI_MC_LPS		= 10,	/*!< lport fc login services	    */
+	BFI_MC_RPORT		= 11,	/*!< Remote port		    */
+	BFI_MC_ITNIM		= 12,	/*!< I-T nexus (Initiator mode)	    */
+	BFI_MC_IOIM_READ	= 13,	/*!< read IO (Initiator mode)	    */
+	BFI_MC_IOIM_WRITE	= 14,	/*!< write IO (Initiator mode)	    */
+	BFI_MC_IOIM_IO		= 15,	/*!< IO (Initiator mode)	    */
+	BFI_MC_IOIM		= 16,	/*!< IO (Initiator mode)	    */
+	BFI_MC_IOIM_IOCOM	= 17,	/*!< good IO completion		    */
+	BFI_MC_TSKIM		= 18,	/*!< Initiator Task management	    */
+	BFI_MC_SBOOT		= 19,	/*!< SAN boot services		    */
+	BFI_MC_IPFC		= 20,	/*!< IP over FC Msgs		    */
+	BFI_MC_PORT		= 21,	/*!< Physical port		    */
+	BFI_MC_SFP		= 22,	/*!< SFP module			    */
+	BFI_MC_MSGQ		= 23,	/*!< MSGQ			    */
+	BFI_MC_ENET		= 24,	/*!< ENET commands/responses	    */
+	BFI_MC_MAX		= 32
+};
+
+#define BFI_IOC_MAX_CQS		4
+#define BFI_IOC_MAX_CQS_ASIC	8
+#define BFI_IOC_MSGLEN_MAX	32	/* 32 bytes */
+
+#define BFI_BOOT_TYPE_OFF		8
+#define BFI_BOOT_PARAM_OFF		12
+
+#define BFI_BOOT_TYPE_NORMAL 		0	/* param is device id */
+#define	BFI_BOOT_TYPE_FLASH		1
+#define	BFI_BOOT_TYPE_MEMTEST		2
+
+#define BFI_BOOT_MEMTEST_RES_ADDR   0x900
+#define BFI_BOOT_MEMTEST_RES_SIG    0xA0A1A2A3
+
+/**
+ *----------------------------------------------------------------------
+ *				IOC
+ *----------------------------------------------------------------------
+ */
+
+enum bfi_ioc_h2i_msgs {
+	BFI_IOC_H2I_ENABLE_REQ		= 1,
+	BFI_IOC_H2I_DISABLE_REQ		= 2,
+	BFI_IOC_H2I_GETATTR_REQ		= 3,
+	BFI_IOC_H2I_DBG_SYNC		= 4,
+	BFI_IOC_H2I_DBG_DUMP		= 5,
+};
+
+enum bfi_ioc_i2h_msgs {
+	BFI_IOC_I2H_ENABLE_REPLY	= BFA_I2HM(1),
+	BFI_IOC_I2H_DISABLE_REPLY 	= BFA_I2HM(2),
+	BFI_IOC_I2H_GETATTR_REPLY 	= BFA_I2HM(3),
+	BFI_IOC_I2H_READY_EVENT 	= BFA_I2HM(4),
+	BFI_IOC_I2H_HBEAT		= BFA_I2HM(5),
+};
+
+/**
+ * BFI_IOC_H2I_GETATTR_REQ message
+ */
+struct bfi_ioc_getattr_req {
+	struct bfi_mhdr mh;
+	union bfi_addr_u	attr_addr;
+};
+
+struct bfi_ioc_attr {
+	u64		mfg_pwwn;	/*!< Mfg port wwn	   */
+	u64		mfg_nwwn;	/*!< Mfg node wwn	   */
+	mac_t		mfg_mac;	/*!< Mfg mac		   */
+	u16	rsvd_a;
+	u64		pwwn;
+	u64		nwwn;
+	mac_t		mac;		/*!< PBC or Mfg mac	   */
+	u16	rsvd_b;
+	mac_t		fcoe_mac;
+	u16	rsvd_c;
+	char		brcd_serialnum[STRSZ(BFA_MFG_SERIALNUM_SIZE)];
+	u8		pcie_gen;
+	u8		pcie_lanes_orig;
+	u8		pcie_lanes;
+	u8		rx_bbcredit;	/*!< receive buffer credits */
+	u32	adapter_prop;	/*!< adapter properties     */
+	u16	maxfrsize;	/*!< max receive frame size */
+	char		asic_rev;
+	u8		rsvd_d;
+	char		fw_version[BFA_VERSION_LEN];
+	char		optrom_version[BFA_VERSION_LEN];
+	struct bfa_mfg_vpd vpd;
+	u32	card_type;	/*!< card type			*/
+};
+
+/**
+ * BFI_IOC_I2H_GETATTR_REPLY message
+ */
+struct bfi_ioc_getattr_reply {
+	struct bfi_mhdr mh;	/*!< Common msg header		*/
+	u8			status;	/*!< cfg reply status		*/
+	u8			rsvd[3];
+};
+
+/**
+ * Firmware memory page offsets
+ */
+#define BFI_IOC_SMEM_PG0_CB	(0x40)
+#define BFI_IOC_SMEM_PG0_CT	(0x180)
+
+/**
+ * Firmware statistic offset
+ */
+#define BFI_IOC_FWSTATS_OFF	(0x6B40)
+#define BFI_IOC_FWSTATS_SZ	(4096)
+
+/**
+ * Firmware trace offset
+ */
+#define BFI_IOC_TRC_OFF		(0x4b00)
+#define BFI_IOC_TRC_ENTS	256
+
+#define BFI_IOC_FW_SIGNATURE	(0xbfadbfad)
+#define BFI_IOC_MD5SUM_SZ	4
+struct bfi_ioc_image_hdr {
+	u32	signature;	/*!< constant signature */
+	u32	rsvd_a;
+	u32	exec;		/*!< exec vector	*/
+	u32	param;		/*!< parameters		*/
+	u32	rsvd_b[4];
+	u32	md5sum[BFI_IOC_MD5SUM_SZ];
+};
+
+/**
+ *  BFI_IOC_I2H_READY_EVENT message
+ */
+struct bfi_ioc_rdy_event {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u8			init_status;	/*!< init event status */
+	u8			rsvd[3];
+};
+
+struct bfi_ioc_hbeat {
+	struct bfi_mhdr mh;		/*!< common msg header		*/
+	u32	   hb_count;	/*!< current heart beat count	*/
+};
+
+/**
+ * IOC hardware/firmware state
+ */
+enum bfi_ioc_state {
+	BFI_IOC_UNINIT		= 0,	/*!< not initialized		     */
+	BFI_IOC_INITING		= 1,	/*!< h/w is being initialized	     */
+	BFI_IOC_HWINIT		= 2,	/*!< h/w is initialized		     */
+	BFI_IOC_CFG		= 3,	/*!< IOC configuration in progress   */
+	BFI_IOC_OP		= 4,	/*!< IOC is operational		     */
+	BFI_IOC_DISABLING	= 5,	/*!< IOC is being disabled	     */
+	BFI_IOC_DISABLED	= 6,	/*!< IOC is disabled		     */
+	BFI_IOC_CFG_DISABLED	= 7,	/*!< IOC is being disabled;transient */
+	BFI_IOC_FAIL		= 8,	/*!< IOC heart-beat failure	     */
+	BFI_IOC_MEMTEST		= 9,	/*!< IOC is doing memtest	     */
+};
+
+#define BFI_IOC_ENDIAN_SIG  0x12345678
+
+enum {
+	BFI_ADAPTER_TYPE_FC	= 0x01,		/*!< FC adapters	   */
+	BFI_ADAPTER_TYPE_MK	= 0x0f0000,	/*!< adapter type mask     */
+	BFI_ADAPTER_TYPE_SH	= 16,	        /*!< adapter type shift    */
+	BFI_ADAPTER_NPORTS_MK	= 0xff00,	/*!< number of ports mask  */
+	BFI_ADAPTER_NPORTS_SH	= 8,	        /*!< number of ports shift */
+	BFI_ADAPTER_SPEED_MK	= 0xff,		/*!< adapter speed mask    */
+	BFI_ADAPTER_SPEED_SH	= 0,	        /*!< adapter speed shift   */
+	BFI_ADAPTER_PROTO	= 0x100000,	/*!< prototype adapaters   */
+	BFI_ADAPTER_TTV		= 0x200000,	/*!< TTV debug capable     */
+	BFI_ADAPTER_UNSUPP	= 0x400000,	/*!< unknown adapter type  */
+};
+
+#define BFI_ADAPTER_GETP(__prop, __adap_prop)			\
+	(((__adap_prop) & BFI_ADAPTER_ ## __prop ## _MK) >>	\
+		BFI_ADAPTER_ ## __prop ## _SH)
+#define BFI_ADAPTER_SETP(__prop, __val)				\
+	((__val) << BFI_ADAPTER_ ## __prop ## _SH)
+#define BFI_ADAPTER_IS_PROTO(__adap_type)			\
+	((__adap_type) & BFI_ADAPTER_PROTO)
+#define BFI_ADAPTER_IS_TTV(__adap_type)				\
+	((__adap_type) & BFI_ADAPTER_TTV)
+#define BFI_ADAPTER_IS_UNSUPP(__adap_type)			\
+	((__adap_type) & BFI_ADAPTER_UNSUPP)
+#define BFI_ADAPTER_IS_SPECIAL(__adap_type)			\
+	((__adap_type) & (BFI_ADAPTER_TTV | BFI_ADAPTER_PROTO |	\
+			BFI_ADAPTER_UNSUPP))
+
+/**
+ * BFI_IOC_H2I_ENABLE_REQ & BFI_IOC_H2I_DISABLE_REQ messages
+ */
+struct bfi_ioc_ctrl_req {
+	struct bfi_mhdr mh;
+	u8			ioc_class;
+	u8			rsvd[3];
+	u32		tv_sec;
+};
+
+/**
+ * BFI_IOC_I2H_ENABLE_REPLY & BFI_IOC_I2H_DISABLE_REPLY messages
+ */
+struct bfi_ioc_ctrl_reply {
+	struct bfi_mhdr mh;		/*!< Common msg header     */
+	u8			status;		/*!< enable/disable status */
+	u8			rsvd[3];
+};
+
+#define BFI_IOC_MSGSZ   8
+/**
+ * H2I Messages
+ */
+union bfi_ioc_h2i_msg_u {
+	struct bfi_mhdr mh;
+	struct bfi_ioc_ctrl_req enable_req;
+	struct bfi_ioc_ctrl_req disable_req;
+	struct bfi_ioc_getattr_req getattr_req;
+	u32			mboxmsg[BFI_IOC_MSGSZ];
+};
+
+/**
+ * I2H Messages
+ */
+union bfi_ioc_i2h_msg_u {
+	struct bfi_mhdr mh;
+	struct bfi_ioc_rdy_event rdy_event;
+	u32			mboxmsg[BFI_IOC_MSGSZ];
+};
+
+#pragma pack()
+
+#endif /* __BFI_H__ */
diff --git a/drivers/net/bna/bfi_cna.h b/drivers/net/bna/bfi_cna.h
new file mode 100644
index 000000000000..4eecabea397b
--- /dev/null
+++ b/drivers/net/bna/bfi_cna.h
@@ -0,0 +1,199 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#ifndef __BFI_CNA_H__
+#define __BFI_CNA_H__
+
+#include "bfi.h"
+#include "bfa_defs_cna.h"
+
+#pragma pack(1)
+
+enum bfi_port_h2i {
+	BFI_PORT_H2I_ENABLE_REQ		= (1),
+	BFI_PORT_H2I_DISABLE_REQ	= (2),
+	BFI_PORT_H2I_GET_STATS_REQ	= (3),
+	BFI_PORT_H2I_CLEAR_STATS_REQ	= (4),
+};
+
+enum bfi_port_i2h {
+	BFI_PORT_I2H_ENABLE_RSP		= BFA_I2HM(1),
+	BFI_PORT_I2H_DISABLE_RSP	= BFA_I2HM(2),
+	BFI_PORT_I2H_GET_STATS_RSP	= BFA_I2HM(3),
+	BFI_PORT_I2H_CLEAR_STATS_RSP	= BFA_I2HM(4),
+};
+
+/**
+ * Generic REQ type
+ */
+struct bfi_port_generic_req {
+	struct bfi_mhdr mh;		/*!< msg header			    */
+	u32	msgtag;		/*!< msgtag for reply		    */
+	u32	rsvd;
+};
+
+/**
+ * Generic RSP type
+ */
+struct bfi_port_generic_rsp {
+	struct bfi_mhdr mh;		/*!< common msg header		    */
+	u8		status;		/*!< port enable status		    */
+	u8		rsvd[3];
+	u32	msgtag;		/*!< msgtag for reply		    */
+};
+
+/**
+ * @todo
+ * BFI_PORT_H2I_ENABLE_REQ
+ */
+
+/**
+ * @todo
+ * BFI_PORT_I2H_ENABLE_RSP
+ */
+
+/**
+ * BFI_PORT_H2I_DISABLE_REQ
+ */
+
+/**
+ * BFI_PORT_I2H_DISABLE_RSP
+ */
+
+/**
+ * BFI_PORT_H2I_GET_STATS_REQ
+ */
+struct bfi_port_get_stats_req {
+	struct bfi_mhdr mh;		/*!< common msg header		    */
+	union bfi_addr_u   dma_addr;
+};
+
+/**
+ * BFI_PORT_I2H_GET_STATS_RSP
+ */
+
+/**
+ * BFI_PORT_H2I_CLEAR_STATS_REQ
+ */
+
+/**
+ * BFI_PORT_I2H_CLEAR_STATS_RSP
+ */
+
+union bfi_port_h2i_msg_u {
+	struct bfi_mhdr mh;
+	struct bfi_port_generic_req enable_req;
+	struct bfi_port_generic_req disable_req;
+	struct bfi_port_get_stats_req getstats_req;
+	struct bfi_port_generic_req clearstats_req;
+};
+
+union bfi_port_i2h_msg_u {
+	struct bfi_mhdr mh;
+	struct bfi_port_generic_rsp enable_rsp;
+	struct bfi_port_generic_rsp disable_rsp;
+	struct bfi_port_generic_rsp getstats_rsp;
+	struct bfi_port_generic_rsp clearstats_rsp;
+};
+
+/* @brief Mailbox commands from host to (DCBX/LLDP) firmware */
+enum bfi_cee_h2i_msgs {
+	BFI_CEE_H2I_GET_CFG_REQ = 1,
+	BFI_CEE_H2I_RESET_STATS = 2,
+	BFI_CEE_H2I_GET_STATS_REQ = 3,
+};
+
+/* @brief Mailbox reply and AEN messages from DCBX/LLDP firmware to host */
+enum bfi_cee_i2h_msgs {
+	BFI_CEE_I2H_GET_CFG_RSP = BFA_I2HM(1),
+	BFI_CEE_I2H_RESET_STATS_RSP = BFA_I2HM(2),
+	BFI_CEE_I2H_GET_STATS_RSP = BFA_I2HM(3),
+};
+
+/* Data structures */
+
+/*
+ * @brief H2I command structure for resetting the stats.
+ * BFI_CEE_H2I_RESET_STATS
+ */
+struct bfi_lldp_reset_stats {
+	struct bfi_mhdr mh;
+};
+
+/*
+ * @brief H2I command structure for resetting the stats.
+ * BFI_CEE_H2I_RESET_STATS
+ */
+struct bfi_cee_reset_stats {
+	struct bfi_mhdr mh;
+};
+
+/*
+ * @brief  get configuration  command from host
+ * BFI_CEE_H2I_GET_CFG_REQ
+ */
+struct bfi_cee_get_req {
+	struct bfi_mhdr mh;
+	union bfi_addr_u   dma_addr;
+};
+
+/*
+ * @brief reply message from firmware
+ * BFI_CEE_I2H_GET_CFG_RSP
+ */
+struct bfi_cee_get_rsp {
+	struct bfi_mhdr mh;
+	u8			cmd_status;
+	u8			rsvd[3];
+};
+
+/*
+ * @brief  get configuration  command from host
+ * BFI_CEE_H2I_GET_STATS_REQ
+ */
+struct bfi_cee_stats_req {
+	struct bfi_mhdr mh;
+	union bfi_addr_u   dma_addr;
+};
+
+/*
+ * @brief reply message from firmware
+ * BFI_CEE_I2H_GET_STATS_RSP
+ */
+struct bfi_cee_stats_rsp {
+	struct bfi_mhdr mh;
+	u8			cmd_status;
+	u8			rsvd[3];
+};
+
+/* @brief mailbox command structures from host to firmware */
+union bfi_cee_h2i_msg_u {
+	struct bfi_mhdr mh;
+	struct bfi_cee_get_req get_req;
+	struct bfi_cee_stats_req stats_req;
+};
+
+/* @brief mailbox message structures from firmware to host	*/
+union bfi_cee_i2h_msg_u {
+	struct bfi_mhdr mh;
+	struct bfi_cee_get_rsp get_rsp;
+	struct bfi_cee_stats_rsp stats_rsp;
+};
+
+#pragma pack()
+
+#endif /* __BFI_CNA_H__ */
diff --git a/drivers/net/bna/bfi_ctreg.h b/drivers/net/bna/bfi_ctreg.h
new file mode 100644
index 000000000000..404ea351d4a1
--- /dev/null
+++ b/drivers/net/bna/bfi_ctreg.h
@@ -0,0 +1,637 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+/*
+ * bfi_ctreg.h catapult host block register definitions
+ *
+ * !!! Do not edit. Auto generated. !!!
+ */
+
+#ifndef __BFI_CTREG_H__
+#define __BFI_CTREG_H__
+
+#define HOSTFN0_LPU_MBOX0_0		0x00019200
+#define HOSTFN1_LPU_MBOX0_8		0x00019260
+#define LPU_HOSTFN0_MBOX0_0		0x00019280
+#define LPU_HOSTFN1_MBOX0_8		0x000192e0
+#define HOSTFN2_LPU_MBOX0_0		0x00019400
+#define HOSTFN3_LPU_MBOX0_8		0x00019460
+#define LPU_HOSTFN2_MBOX0_0		0x00019480
+#define LPU_HOSTFN3_MBOX0_8		0x000194e0
+#define HOSTFN0_INT_STATUS		0x00014000
+#define __HOSTFN0_HALT_OCCURRED		0x01000000
+#define __HOSTFN0_INT_STATUS_LVL_MK	0x00f00000
+#define __HOSTFN0_INT_STATUS_LVL_SH	20
+#define __HOSTFN0_INT_STATUS_LVL(_v)	((_v) << __HOSTFN0_INT_STATUS_LVL_SH)
+#define __HOSTFN0_INT_STATUS_P_MK	0x000f0000
+#define __HOSTFN0_INT_STATUS_P_SH	16
+#define __HOSTFN0_INT_STATUS_P(_v)	((_v) << __HOSTFN0_INT_STATUS_P_SH)
+#define __HOSTFN0_INT_STATUS_F		0x0000ffff
+#define HOSTFN0_INT_MSK			0x00014004
+#define HOST_PAGE_NUM_FN0		0x00014008
+#define __HOST_PAGE_NUM_FN		0x000001ff
+#define HOST_MSIX_ERR_INDEX_FN0		0x0001400c
+#define __MSIX_ERR_INDEX_FN		0x000001ff
+#define HOSTFN1_INT_STATUS		0x00014100
+#define __HOSTFN1_HALT_OCCURRED		0x01000000
+#define __HOSTFN1_INT_STATUS_LVL_MK	0x00f00000
+#define __HOSTFN1_INT_STATUS_LVL_SH	20
+#define __HOSTFN1_INT_STATUS_LVL(_v)	((_v) << __HOSTFN1_INT_STATUS_LVL_SH)
+#define __HOSTFN1_INT_STATUS_P_MK	0x000f0000
+#define __HOSTFN1_INT_STATUS_P_SH	16
+#define __HOSTFN1_INT_STATUS_P(_v)	((_v) << __HOSTFN1_INT_STATUS_P_SH)
+#define __HOSTFN1_INT_STATUS_F		0x0000ffff
+#define HOSTFN1_INT_MSK			0x00014104
+#define HOST_PAGE_NUM_FN1		0x00014108
+#define HOST_MSIX_ERR_INDEX_FN1		0x0001410c
+#define APP_PLL_425_CTL_REG		0x00014204
+#define __P_425_PLL_LOCK		0x80000000
+#define __APP_PLL_425_SRAM_USE_100MHZ	0x00100000
+#define __APP_PLL_425_RESET_TIMER_MK	0x000e0000
+#define __APP_PLL_425_RESET_TIMER_SH	17
+#define __APP_PLL_425_RESET_TIMER(_v)	((_v) << __APP_PLL_425_RESET_TIMER_SH)
+#define __APP_PLL_425_LOGIC_SOFT_RESET	0x00010000
+#define __APP_PLL_425_CNTLMT0_1_MK	0x0000c000
+#define __APP_PLL_425_CNTLMT0_1_SH	14
+#define __APP_PLL_425_CNTLMT0_1(_v)	((_v) << __APP_PLL_425_CNTLMT0_1_SH)
+#define __APP_PLL_425_JITLMT0_1_MK	0x00003000
+#define __APP_PLL_425_JITLMT0_1_SH	12
+#define __APP_PLL_425_JITLMT0_1(_v)	((_v) << __APP_PLL_425_JITLMT0_1_SH)
+#define __APP_PLL_425_HREF		0x00000800
+#define __APP_PLL_425_HDIV		0x00000400
+#define __APP_PLL_425_P0_1_MK		0x00000300
+#define __APP_PLL_425_P0_1_SH		8
+#define __APP_PLL_425_P0_1(_v)		((_v) << __APP_PLL_425_P0_1_SH)
+#define __APP_PLL_425_Z0_2_MK		0x000000e0
+#define __APP_PLL_425_Z0_2_SH		5
+#define __APP_PLL_425_Z0_2(_v)		((_v) << __APP_PLL_425_Z0_2_SH)
+#define __APP_PLL_425_RSEL200500	0x00000010
+#define __APP_PLL_425_ENARST		0x00000008
+#define __APP_PLL_425_BYPASS		0x00000004
+#define __APP_PLL_425_LRESETN		0x00000002
+#define __APP_PLL_425_ENABLE		0x00000001
+#define APP_PLL_312_CTL_REG		0x00014208
+#define __P_312_PLL_LOCK		0x80000000
+#define __ENABLE_MAC_AHB_1		0x00800000
+#define __ENABLE_MAC_AHB_0		0x00400000
+#define __ENABLE_MAC_1			0x00200000
+#define __ENABLE_MAC_0			0x00100000
+#define __APP_PLL_312_RESET_TIMER_MK	0x000e0000
+#define __APP_PLL_312_RESET_TIMER_SH	17
+#define __APP_PLL_312_RESET_TIMER(_v)	((_v) << __APP_PLL_312_RESET_TIMER_SH)
+#define __APP_PLL_312_LOGIC_SOFT_RESET	0x00010000
+#define __APP_PLL_312_CNTLMT0_1_MK	0x0000c000
+#define __APP_PLL_312_CNTLMT0_1_SH	14
+#define __APP_PLL_312_CNTLMT0_1(_v)	((_v) << __APP_PLL_312_CNTLMT0_1_SH)
+#define __APP_PLL_312_JITLMT0_1_MK	0x00003000
+#define __APP_PLL_312_JITLMT0_1_SH	12
+#define __APP_PLL_312_JITLMT0_1(_v)	((_v) << __APP_PLL_312_JITLMT0_1_SH)
+#define __APP_PLL_312_HREF		0x00000800
+#define __APP_PLL_312_HDIV		0x00000400
+#define __APP_PLL_312_P0_1_MK		0x00000300
+#define __APP_PLL_312_P0_1_SH		8
+#define __APP_PLL_312_P0_1(_v)		((_v) << __APP_PLL_312_P0_1_SH)
+#define __APP_PLL_312_Z0_2_MK		0x000000e0
+#define __APP_PLL_312_Z0_2_SH		5
+#define __APP_PLL_312_Z0_2(_v)		((_v) << __APP_PLL_312_Z0_2_SH)
+#define __APP_PLL_312_RSEL200500	0x00000010
+#define __APP_PLL_312_ENARST		0x00000008
+#define __APP_PLL_312_BYPASS		0x00000004
+#define __APP_PLL_312_LRESETN		0x00000002
+#define __APP_PLL_312_ENABLE		0x00000001
+#define MBIST_CTL_REG			0x00014220
+#define __EDRAM_BISTR_START		0x00000004
+#define __MBIST_RESET			0x00000002
+#define __MBIST_START			0x00000001
+#define MBIST_STAT_REG			0x00014224
+#define __EDRAM_BISTR_STATUS		0x00000008
+#define __EDRAM_BISTR_DONE		0x00000004
+#define __MEM_BIT_STATUS		0x00000002
+#define __MBIST_DONE			0x00000001
+#define HOST_SEM0_REG			0x00014230
+#define __HOST_SEMAPHORE		0x00000001
+#define HOST_SEM1_REG			0x00014234
+#define HOST_SEM2_REG			0x00014238
+#define HOST_SEM3_REG			0x0001423c
+#define HOST_SEM0_INFO_REG		0x00014240
+#define HOST_SEM1_INFO_REG		0x00014244
+#define HOST_SEM2_INFO_REG		0x00014248
+#define HOST_SEM3_INFO_REG		0x0001424c
+#define ETH_MAC_SER_REG			0x00014288
+#define __APP_EMS_CKBUFAMPIN		0x00000020
+#define __APP_EMS_REFCLKSEL		0x00000010
+#define __APP_EMS_CMLCKSEL		0x00000008
+#define __APP_EMS_REFCKBUFEN2		0x00000004
+#define __APP_EMS_REFCKBUFEN1		0x00000002
+#define __APP_EMS_CHANNEL_SEL		0x00000001
+#define HOSTFN2_INT_STATUS		0x00014300
+#define __HOSTFN2_HALT_OCCURRED		0x01000000
+#define __HOSTFN2_INT_STATUS_LVL_MK	0x00f00000
+#define __HOSTFN2_INT_STATUS_LVL_SH	20
+#define __HOSTFN2_INT_STATUS_LVL(_v)	((_v) << __HOSTFN2_INT_STATUS_LVL_SH)
+#define __HOSTFN2_INT_STATUS_P_MK	0x000f0000
+#define __HOSTFN2_INT_STATUS_P_SH	16
+#define __HOSTFN2_INT_STATUS_P(_v)	((_v) << __HOSTFN2_INT_STATUS_P_SH)
+#define __HOSTFN2_INT_STATUS_F		0x0000ffff
+#define HOSTFN2_INT_MSK			0x00014304
+#define HOST_PAGE_NUM_FN2		0x00014308
+#define HOST_MSIX_ERR_INDEX_FN2		0x0001430c
+#define HOSTFN3_INT_STATUS		0x00014400
+#define __HALT_OCCURRED			0x01000000
+#define __HOSTFN3_INT_STATUS_LVL_MK	0x00f00000
+#define __HOSTFN3_INT_STATUS_LVL_SH	20
+#define __HOSTFN3_INT_STATUS_LVL(_v)	((_v) << __HOSTFN3_INT_STATUS_LVL_SH)
+#define __HOSTFN3_INT_STATUS_P_MK	0x000f0000
+#define __HOSTFN3_INT_STATUS_P_SH	16
+#define __HOSTFN3_INT_STATUS_P(_v)	((_v) << __HOSTFN3_INT_STATUS_P_SH)
+#define __HOSTFN3_INT_STATUS_F		0x0000ffff
+#define HOSTFN3_INT_MSK			0x00014404
+#define HOST_PAGE_NUM_FN3		0x00014408
+#define HOST_MSIX_ERR_INDEX_FN3		0x0001440c
+#define FNC_ID_REG			0x00014600
+#define __FUNCTION_NUMBER		0x00000007
+#define FNC_PERS_REG			0x00014604
+#define __F3_FUNCTION_ACTIVE		0x80000000
+#define __F3_FUNCTION_MODE		0x40000000
+#define __F3_PORT_MAP_MK		0x30000000
+#define __F3_PORT_MAP_SH		28
+#define __F3_PORT_MAP(_v)		((_v) << __F3_PORT_MAP_SH)
+#define __F3_VM_MODE			0x08000000
+#define __F3_INTX_STATUS_MK		0x07000000
+#define __F3_INTX_STATUS_SH		24
+#define __F3_INTX_STATUS(_v)		((_v) << __F3_INTX_STATUS_SH)
+#define __F2_FUNCTION_ACTIVE		0x00800000
+#define __F2_FUNCTION_MODE		0x00400000
+#define __F2_PORT_MAP_MK		0x00300000
+#define __F2_PORT_MAP_SH		20
+#define __F2_PORT_MAP(_v)		((_v) << __F2_PORT_MAP_SH)
+#define __F2_VM_MODE			0x00080000
+#define __F2_INTX_STATUS_MK		0x00070000
+#define __F2_INTX_STATUS_SH		16
+#define __F2_INTX_STATUS(_v)		((_v) << __F2_INTX_STATUS_SH)
+#define __F1_FUNCTION_ACTIVE		0x00008000
+#define __F1_FUNCTION_MODE		0x00004000
+#define __F1_PORT_MAP_MK		0x00003000
+#define __F1_PORT_MAP_SH		12
+#define __F1_PORT_MAP(_v)		((_v) << __F1_PORT_MAP_SH)
+#define __F1_VM_MODE			0x00000800
+#define __F1_INTX_STATUS_MK		0x00000700
+#define __F1_INTX_STATUS_SH		8
+#define __F1_INTX_STATUS(_v)		((_v) << __F1_INTX_STATUS_SH)
+#define __F0_FUNCTION_ACTIVE		0x00000080
+#define __F0_FUNCTION_MODE		0x00000040
+#define __F0_PORT_MAP_MK		0x00000030
+#define __F0_PORT_MAP_SH		4
+#define __F0_PORT_MAP(_v)		((_v) << __F0_PORT_MAP_SH)
+#define __F0_VM_MODE		0x00000008
+#define __F0_INTX_STATUS		0x00000007
+enum {
+	__F0_INTX_STATUS_MSIX		= 0x0,
+	__F0_INTX_STATUS_INTA		= 0x1,
+	__F0_INTX_STATUS_INTB		= 0x2,
+	__F0_INTX_STATUS_INTC		= 0x3,
+	__F0_INTX_STATUS_INTD		= 0x4,
+};
+#define OP_MODE				0x0001460c
+#define __APP_ETH_CLK_LOWSPEED		0x00000004
+#define __GLOBAL_CORECLK_HALFSPEED	0x00000002
+#define __GLOBAL_FCOE_MODE		0x00000001
+#define HOST_SEM4_REG			0x00014610
+#define HOST_SEM5_REG			0x00014614
+#define HOST_SEM6_REG			0x00014618
+#define HOST_SEM7_REG			0x0001461c
+#define HOST_SEM4_INFO_REG		0x00014620
+#define HOST_SEM5_INFO_REG		0x00014624
+#define HOST_SEM6_INFO_REG		0x00014628
+#define HOST_SEM7_INFO_REG		0x0001462c
+#define HOSTFN0_LPU0_MBOX0_CMD_STAT	0x00019000
+#define __HOSTFN0_LPU0_MBOX0_INFO_MK	0xfffffffe
+#define __HOSTFN0_LPU0_MBOX0_INFO_SH	1
+#define __HOSTFN0_LPU0_MBOX0_INFO(_v)	((_v) << __HOSTFN0_LPU0_MBOX0_INFO_SH)
+#define __HOSTFN0_LPU0_MBOX0_CMD_STATUS 0x00000001
+#define HOSTFN0_LPU1_MBOX0_CMD_STAT	0x00019004
+#define __HOSTFN0_LPU1_MBOX0_INFO_MK	0xfffffffe
+#define __HOSTFN0_LPU1_MBOX0_INFO_SH	1
+#define __HOSTFN0_LPU1_MBOX0_INFO(_v)	((_v) << __HOSTFN0_LPU1_MBOX0_INFO_SH)
+#define __HOSTFN0_LPU1_MBOX0_CMD_STATUS 0x00000001
+#define LPU0_HOSTFN0_MBOX0_CMD_STAT	0x00019008
+#define __LPU0_HOSTFN0_MBOX0_INFO_MK	0xfffffffe
+#define __LPU0_HOSTFN0_MBOX0_INFO_SH	1
+#define __LPU0_HOSTFN0_MBOX0_INFO(_v)	((_v) << __LPU0_HOSTFN0_MBOX0_INFO_SH)
+#define __LPU0_HOSTFN0_MBOX0_CMD_STATUS 0x00000001
+#define LPU1_HOSTFN0_MBOX0_CMD_STAT	0x0001900c
+#define __LPU1_HOSTFN0_MBOX0_INFO_MK	0xfffffffe
+#define __LPU1_HOSTFN0_MBOX0_INFO_SH	1
+#define __LPU1_HOSTFN0_MBOX0_INFO(_v)	((_v) << __LPU1_HOSTFN0_MBOX0_INFO_SH)
+#define __LPU1_HOSTFN0_MBOX0_CMD_STATUS 0x00000001
+#define HOSTFN1_LPU0_MBOX0_CMD_STAT	0x00019010
+#define __HOSTFN1_LPU0_MBOX0_INFO_MK	0xfffffffe
+#define __HOSTFN1_LPU0_MBOX0_INFO_SH	1
+#define __HOSTFN1_LPU0_MBOX0_INFO(_v)	((_v) << __HOSTFN1_LPU0_MBOX0_INFO_SH)
+#define __HOSTFN1_LPU0_MBOX0_CMD_STATUS 0x00000001
+#define HOSTFN1_LPU1_MBOX0_CMD_STAT	0x00019014
+#define __HOSTFN1_LPU1_MBOX0_INFO_MK	0xfffffffe
+#define __HOSTFN1_LPU1_MBOX0_INFO_SH	1
+#define __HOSTFN1_LPU1_MBOX0_INFO(_v)	((_v) << __HOSTFN1_LPU1_MBOX0_INFO_SH)
+#define __HOSTFN1_LPU1_MBOX0_CMD_STATUS 0x00000001
+#define LPU0_HOSTFN1_MBOX0_CMD_STAT	0x00019018
+#define __LPU0_HOSTFN1_MBOX0_INFO_MK	0xfffffffe
+#define __LPU0_HOSTFN1_MBOX0_INFO_SH	1
+#define __LPU0_HOSTFN1_MBOX0_INFO(_v)	((_v) << __LPU0_HOSTFN1_MBOX0_INFO_SH)
+#define __LPU0_HOSTFN1_MBOX0_CMD_STATUS 0x00000001
+#define LPU1_HOSTFN1_MBOX0_CMD_STAT	0x0001901c
+#define __LPU1_HOSTFN1_MBOX0_INFO_MK	0xfffffffe
+#define __LPU1_HOSTFN1_MBOX0_INFO_SH	1
+#define __LPU1_HOSTFN1_MBOX0_INFO(_v)	((_v) << __LPU1_HOSTFN1_MBOX0_INFO_SH)
+#define __LPU1_HOSTFN1_MBOX0_CMD_STATUS 0x00000001
+#define HOSTFN2_LPU0_MBOX0_CMD_STAT	0x00019150
+#define __HOSTFN2_LPU0_MBOX0_INFO_MK	0xfffffffe
+#define __HOSTFN2_LPU0_MBOX0_INFO_SH	1
+#define __HOSTFN2_LPU0_MBOX0_INFO(_v)	((_v) << __HOSTFN2_LPU0_MBOX0_INFO_SH)
+#define __HOSTFN2_LPU0_MBOX0_CMD_STATUS 0x00000001
+#define HOSTFN2_LPU1_MBOX0_CMD_STAT	0x00019154
+#define __HOSTFN2_LPU1_MBOX0_INFO_MK	0xfffffffe
+#define __HOSTFN2_LPU1_MBOX0_INFO_SH	1
+#define __HOSTFN2_LPU1_MBOX0_INFO(_v)	((_v) << __HOSTFN2_LPU1_MBOX0_INFO_SH)
+#define __HOSTFN2_LPU1_MBOX0BOX0_CMD_STATUS 0x00000001
+#define LPU0_HOSTFN2_MBOX0_CMD_STAT	0x00019158
+#define __LPU0_HOSTFN2_MBOX0_INFO_MK	0xfffffffe
+#define __LPU0_HOSTFN2_MBOX0_INFO_SH	1
+#define __LPU0_HOSTFN2_MBOX0_INFO(_v)	((_v) << __LPU0_HOSTFN2_MBOX0_INFO_SH)
+#define __LPU0_HOSTFN2_MBOX0_CMD_STATUS 0x00000001
+#define LPU1_HOSTFN2_MBOX0_CMD_STAT	0x0001915c
+#define __LPU1_HOSTFN2_MBOX0_INFO_MK	0xfffffffe
+#define __LPU1_HOSTFN2_MBOX0_INFO_SH	1
+#define __LPU1_HOSTFN2_MBOX0_INFO(_v)	((_v) << __LPU1_HOSTFN2_MBOX0_INFO_SH)
+#define __LPU1_HOSTFN2_MBOX0_CMD_STATUS 0x00000001
+#define HOSTFN3_LPU0_MBOX0_CMD_STAT	0x00019160
+#define __HOSTFN3_LPU0_MBOX0_INFO_MK	0xfffffffe
+#define __HOSTFN3_LPU0_MBOX0_INFO_SH	1
+#define __HOSTFN3_LPU0_MBOX0_INFO(_v)	((_v) << __HOSTFN3_LPU0_MBOX0_INFO_SH)
+#define __HOSTFN3_LPU0_MBOX0_CMD_STATUS 0x00000001
+#define HOSTFN3_LPU1_MBOX0_CMD_STAT	0x00019164
+#define __HOSTFN3_LPU1_MBOX0_INFO_MK	0xfffffffe
+#define __HOSTFN3_LPU1_MBOX0_INFO_SH	1
+#define __HOSTFN3_LPU1_MBOX0_INFO(_v)	((_v) << __HOSTFN3_LPU1_MBOX0_INFO_SH)
+#define __HOSTFN3_LPU1_MBOX0_CMD_STATUS 0x00000001
+#define LPU0_HOSTFN3_MBOX0_CMD_STAT	0x00019168
+#define __LPU0_HOSTFN3_MBOX0_INFO_MK	0xfffffffe
+#define __LPU0_HOSTFN3_MBOX0_INFO_SH	1
+#define __LPU0_HOSTFN3_MBOX0_INFO(_v)	((_v) << __LPU0_HOSTFN3_MBOX0_INFO_SH)
+#define __LPU0_HOSTFN3_MBOX0_CMD_STATUS 0x00000001
+#define LPU1_HOSTFN3_MBOX0_CMD_STAT	0x0001916c
+#define __LPU1_HOSTFN3_MBOX0_INFO_MK	0xfffffffe
+#define __LPU1_HOSTFN3_MBOX0_INFO_SH	1
+#define __LPU1_HOSTFN3_MBOX0_INFO(_v)	((_v) << __LPU1_HOSTFN3_MBOX0_INFO_SH)
+#define __LPU1_HOSTFN3_MBOX0_CMD_STATUS	0x00000001
+#define FW_INIT_HALT_P0			0x000191ac
+#define __FW_INIT_HALT_P		0x00000001
+#define FW_INIT_HALT_P1			0x000191bc
+#define CPE_PI_PTR_Q0			0x00038000
+#define __CPE_PI_UNUSED_MK		0xffff0000
+#define __CPE_PI_UNUSED_SH		16
+#define __CPE_PI_UNUSED(_v)		((_v) << __CPE_PI_UNUSED_SH)
+#define __CPE_PI_PTR			0x0000ffff
+#define CPE_PI_PTR_Q1			0x00038040
+#define CPE_CI_PTR_Q0			0x00038004
+#define __CPE_CI_UNUSED_MK		0xffff0000
+#define __CPE_CI_UNUSED_SH		16
+#define __CPE_CI_UNUSED(_v)		((_v) << __CPE_CI_UNUSED_SH)
+#define __CPE_CI_PTR			0x0000ffff
+#define CPE_CI_PTR_Q1			0x00038044
+#define CPE_DEPTH_Q0			0x00038008
+#define __CPE_DEPTH_UNUSED_MK		0xf8000000
+#define __CPE_DEPTH_UNUSED_SH		27
+#define __CPE_DEPTH_UNUSED(_v)		((_v) << __CPE_DEPTH_UNUSED_SH)
+#define __CPE_MSIX_VEC_INDEX_MK		0x07ff0000
+#define __CPE_MSIX_VEC_INDEX_SH		16
+#define __CPE_MSIX_VEC_INDEX(_v)	((_v) << __CPE_MSIX_VEC_INDEX_SH)
+#define __CPE_DEPTH			0x0000ffff
+#define CPE_DEPTH_Q1			0x00038048
+#define CPE_QCTRL_Q0			0x0003800c
+#define __CPE_CTRL_UNUSED30_MK		0xfc000000
+#define __CPE_CTRL_UNUSED30_SH		26
+#define __CPE_CTRL_UNUSED30(_v)		((_v) << __CPE_CTRL_UNUSED30_SH)
+#define __CPE_FUNC_INT_CTRL_MK		0x03000000
+#define __CPE_FUNC_INT_CTRL_SH		24
+#define __CPE_FUNC_INT_CTRL(_v)		((_v) << __CPE_FUNC_INT_CTRL_SH)
+enum {
+	__CPE_FUNC_INT_CTRL_DISABLE		= 0x0,
+	__CPE_FUNC_INT_CTRL_F2NF		= 0x1,
+	__CPE_FUNC_INT_CTRL_3QUART		= 0x2,
+	__CPE_FUNC_INT_CTRL_HALF		= 0x3,
+};
+#define __CPE_CTRL_UNUSED20_MK		0x00f00000
+#define __CPE_CTRL_UNUSED20_SH		20
+#define __CPE_CTRL_UNUSED20(_v)		((_v) << __CPE_CTRL_UNUSED20_SH)
+#define __CPE_SCI_TH_MK			0x000f0000
+#define __CPE_SCI_TH_SH			16
+#define __CPE_SCI_TH(_v)		((_v) << __CPE_SCI_TH_SH)
+#define __CPE_CTRL_UNUSED10_MK		0x0000c000
+#define __CPE_CTRL_UNUSED10_SH		14
+#define __CPE_CTRL_UNUSED10(_v)		((_v) << __CPE_CTRL_UNUSED10_SH)
+#define __CPE_ACK_PENDING		0x00002000
+#define __CPE_CTRL_UNUSED40_MK		0x00001c00
+#define __CPE_CTRL_UNUSED40_SH		10
+#define __CPE_CTRL_UNUSED40(_v)		((_v) << __CPE_CTRL_UNUSED40_SH)
+#define __CPE_PCIEID_MK			0x00000300
+#define __CPE_PCIEID_SH			8
+#define __CPE_PCIEID(_v)		((_v) << __CPE_PCIEID_SH)
+#define __CPE_CTRL_UNUSED00_MK		0x000000fe
+#define __CPE_CTRL_UNUSED00_SH		1
+#define __CPE_CTRL_UNUSED00(_v)		((_v) << __CPE_CTRL_UNUSED00_SH)
+#define __CPE_ESIZE			0x00000001
+#define CPE_QCTRL_Q1			0x0003804c
+#define __CPE_CTRL_UNUSED31_MK		0xfc000000
+#define __CPE_CTRL_UNUSED31_SH		26
+#define __CPE_CTRL_UNUSED31(_v)		((_v) << __CPE_CTRL_UNUSED31_SH)
+#define __CPE_CTRL_UNUSED21_MK		0x00f00000
+#define __CPE_CTRL_UNUSED21_SH		20
+#define __CPE_CTRL_UNUSED21(_v)		((_v) << __CPE_CTRL_UNUSED21_SH)
+#define __CPE_CTRL_UNUSED11_MK		0x0000c000
+#define __CPE_CTRL_UNUSED11_SH		14
+#define __CPE_CTRL_UNUSED11(_v)		((_v) << __CPE_CTRL_UNUSED11_SH)
+#define __CPE_CTRL_UNUSED41_MK		0x00001c00
+#define __CPE_CTRL_UNUSED41_SH		10
+#define __CPE_CTRL_UNUSED41(_v)		((_v) << __CPE_CTRL_UNUSED41_SH)
+#define __CPE_CTRL_UNUSED01_MK		0x000000fe
+#define __CPE_CTRL_UNUSED01_SH		1
+#define __CPE_CTRL_UNUSED01(_v)		((_v) << __CPE_CTRL_UNUSED01_SH)
+#define RME_PI_PTR_Q0			0x00038020
+#define __LATENCY_TIME_STAMP_MK		0xffff0000
+#define __LATENCY_TIME_STAMP_SH		16
+#define __LATENCY_TIME_STAMP(_v)	((_v) << __LATENCY_TIME_STAMP_SH)
+#define __RME_PI_PTR			0x0000ffff
+#define RME_PI_PTR_Q1			0x00038060
+#define RME_CI_PTR_Q0			0x00038024
+#define __DELAY_TIME_STAMP_MK		0xffff0000
+#define __DELAY_TIME_STAMP_SH		16
+#define __DELAY_TIME_STAMP(_v)		((_v) << __DELAY_TIME_STAMP_SH)
+#define __RME_CI_PTR			0x0000ffff
+#define RME_CI_PTR_Q1			0x00038064
+#define RME_DEPTH_Q0			0x00038028
+#define __RME_DEPTH_UNUSED_MK		0xf8000000
+#define __RME_DEPTH_UNUSED_SH		27
+#define __RME_DEPTH_UNUSED(_v)		((_v) << __RME_DEPTH_UNUSED_SH)
+#define __RME_MSIX_VEC_INDEX_MK		0x07ff0000
+#define __RME_MSIX_VEC_INDEX_SH		16
+#define __RME_MSIX_VEC_INDEX(_v)	((_v) << __RME_MSIX_VEC_INDEX_SH)
+#define __RME_DEPTH			0x0000ffff
+#define RME_DEPTH_Q1			0x00038068
+#define RME_QCTRL_Q0			0x0003802c
+#define __RME_INT_LATENCY_TIMER_MK	0xff000000
+#define __RME_INT_LATENCY_TIMER_SH	24
+#define __RME_INT_LATENCY_TIMER(_v)	((_v) << __RME_INT_LATENCY_TIMER_SH)
+#define __RME_INT_DELAY_TIMER_MK	0x00ff0000
+#define __RME_INT_DELAY_TIMER_SH	16
+#define __RME_INT_DELAY_TIMER(_v)	((_v) << __RME_INT_DELAY_TIMER_SH)
+#define __RME_INT_DELAY_DISABLE		0x00008000
+#define __RME_DLY_DELAY_DISABLE		0x00004000
+#define __RME_ACK_PENDING		0x00002000
+#define __RME_FULL_INTERRUPT_DISABLE	0x00001000
+#define __RME_CTRL_UNUSED10_MK		0x00000c00
+#define __RME_CTRL_UNUSED10_SH		10
+#define __RME_CTRL_UNUSED10(_v)		((_v) << __RME_CTRL_UNUSED10_SH)
+#define __RME_PCIEID_MK			0x00000300
+#define __RME_PCIEID_SH			8
+#define __RME_PCIEID(_v)		((_v) << __RME_PCIEID_SH)
+#define __RME_CTRL_UNUSED00_MK		0x000000fe
+#define __RME_CTRL_UNUSED00_SH		1
+#define __RME_CTRL_UNUSED00(_v)		((_v) << __RME_CTRL_UNUSED00_SH)
+#define __RME_ESIZE			0x00000001
+#define RME_QCTRL_Q1			0x0003806c
+#define __RME_CTRL_UNUSED11_MK		0x00000c00
+#define __RME_CTRL_UNUSED11_SH		10
+#define __RME_CTRL_UNUSED11(_v)		((_v) << __RME_CTRL_UNUSED11_SH)
+#define __RME_CTRL_UNUSED01_MK		0x000000fe
+#define __RME_CTRL_UNUSED01_SH		1
+#define __RME_CTRL_UNUSED01(_v)		((_v) << __RME_CTRL_UNUSED01_SH)
+#define PSS_CTL_REG			0x00018800
+#define __PSS_I2C_CLK_DIV_MK		0x007f0000
+#define __PSS_I2C_CLK_DIV_SH		16
+#define __PSS_I2C_CLK_DIV(_v)		((_v) << __PSS_I2C_CLK_DIV_SH)
+#define __PSS_LMEM_INIT_DONE		0x00001000
+#define __PSS_LMEM_RESET		0x00000200
+#define __PSS_LMEM_INIT_EN		0x00000100
+#define __PSS_LPU1_RESET		0x00000002
+#define __PSS_LPU0_RESET		0x00000001
+#define PSS_ERR_STATUS_REG		0x00018810
+#define __PSS_LPU1_TCM_READ_ERR		0x00200000
+#define __PSS_LPU0_TCM_READ_ERR		0x00100000
+#define __PSS_LMEM5_CORR_ERR		0x00080000
+#define __PSS_LMEM4_CORR_ERR		0x00040000
+#define __PSS_LMEM3_CORR_ERR		0x00020000
+#define __PSS_LMEM2_CORR_ERR		0x00010000
+#define __PSS_LMEM1_CORR_ERR		0x00008000
+#define __PSS_LMEM0_CORR_ERR		0x00004000
+#define __PSS_LMEM5_UNCORR_ERR		0x00002000
+#define __PSS_LMEM4_UNCORR_ERR		0x00001000
+#define __PSS_LMEM3_UNCORR_ERR		0x00000800
+#define __PSS_LMEM2_UNCORR_ERR		0x00000400
+#define __PSS_LMEM1_UNCORR_ERR		0x00000200
+#define __PSS_LMEM0_UNCORR_ERR		0x00000100
+#define __PSS_BAL_PERR			0x00000080
+#define __PSS_DIP_IF_ERR		0x00000040
+#define __PSS_IOH_IF_ERR		0x00000020
+#define __PSS_TDS_IF_ERR		0x00000010
+#define __PSS_RDS_IF_ERR		0x00000008
+#define __PSS_SGM_IF_ERR		0x00000004
+#define __PSS_LPU1_RAM_ERR		0x00000002
+#define __PSS_LPU0_RAM_ERR		0x00000001
+#define ERR_SET_REG			0x00018818
+#define __PSS_ERR_STATUS_SET		0x003fffff
+#define PMM_1T_RESET_REG_P0		0x0002381c
+#define __PMM_1T_RESET_P		0x00000001
+#define PMM_1T_RESET_REG_P1		0x00023c1c
+#define HQM_QSET0_RXQ_DRBL_P0		0x00038000
+#define __RXQ0_ADD_VECTORS_P		0x80000000
+#define __RXQ0_STOP_P			0x40000000
+#define __RXQ0_PRD_PTR_P		0x0000ffff
+#define HQM_QSET1_RXQ_DRBL_P0		0x00038080
+#define __RXQ1_ADD_VECTORS_P		0x80000000
+#define __RXQ1_STOP_P			0x40000000
+#define __RXQ1_PRD_PTR_P		0x0000ffff
+#define HQM_QSET0_RXQ_DRBL_P1		0x0003c000
+#define HQM_QSET1_RXQ_DRBL_P1		0x0003c080
+#define HQM_QSET0_TXQ_DRBL_P0		0x00038020
+#define __TXQ0_ADD_VECTORS_P		0x80000000
+#define __TXQ0_STOP_P			0x40000000
+#define __TXQ0_PRD_PTR_P		0x0000ffff
+#define HQM_QSET1_TXQ_DRBL_P0		0x000380a0
+#define __TXQ1_ADD_VECTORS_P		0x80000000
+#define __TXQ1_STOP_P			0x40000000
+#define __TXQ1_PRD_PTR_P		0x0000ffff
+#define HQM_QSET0_TXQ_DRBL_P1		0x0003c020
+#define HQM_QSET1_TXQ_DRBL_P1		0x0003c0a0
+#define HQM_QSET0_IB_DRBL_1_P0		0x00038040
+#define __IB1_0_ACK_P			0x80000000
+#define __IB1_0_DISABLE_P		0x40000000
+#define __IB1_0_COALESCING_CFG_P_MK	0x00ff0000
+#define __IB1_0_COALESCING_CFG_P_SH	16
+#define __IB1_0_COALESCING_CFG_P(_v)	((_v) << __IB1_0_COALESCING_CFG_P_SH)
+#define __IB1_0_NUM_OF_ACKED_EVENTS_P	0x0000ffff
+#define HQM_QSET1_IB_DRBL_1_P0		0x000380c0
+#define __IB1_1_ACK_P			0x80000000
+#define __IB1_1_DISABLE_P		0x40000000
+#define __IB1_1_COALESCING_CFG_P_MK	0x00ff0000
+#define __IB1_1_COALESCING_CFG_P_SH	16
+#define __IB1_1_COALESCING_CFG_P(_v)	((_v) << __IB1_1_COALESCING_CFG_P_SH)
+#define __IB1_1_NUM_OF_ACKED_EVENTS_P	0x0000ffff
+#define HQM_QSET0_IB_DRBL_1_P1		0x0003c040
+#define HQM_QSET1_IB_DRBL_1_P1		0x0003c0c0
+#define HQM_QSET0_IB_DRBL_2_P0		0x00038060
+#define __IB2_0_ACK_P			0x80000000
+#define __IB2_0_DISABLE_P		0x40000000
+#define __IB2_0_COALESCING_CFG_P_MK	0x00ff0000
+#define __IB2_0_COALESCING_CFG_P_SH	16
+#define __IB2_0_COALESCING_CFG_P(_v)	((_v) << __IB2_0_COALESCING_CFG_P_SH)
+#define __IB2_0_NUM_OF_ACKED_EVENTS_P	0x0000ffff
+#define HQM_QSET1_IB_DRBL_2_P0		0x000380e0
+#define __IB2_1_ACK_P			0x80000000
+#define __IB2_1_DISABLE_P		0x40000000
+#define __IB2_1_COALESCING_CFG_P_MK	0x00ff0000
+#define __IB2_1_COALESCING_CFG_P_SH	16
+#define __IB2_1_COALESCING_CFG_P(_v)	((_v) << __IB2_1_COALESCING_CFG_P_SH)
+#define __IB2_1_NUM_OF_ACKED_EVENTS_P	0x0000ffff
+#define HQM_QSET0_IB_DRBL_2_P1		0x0003c060
+#define HQM_QSET1_IB_DRBL_2_P1		0x0003c0e0
+
+/*
+ * These definitions are either in error/missing in spec. Its auto-generated
+ * from hard coded values in regparse.pl.
+ */
+#define __EMPHPOST_AT_4G_MK_FIX		0x0000001c
+#define __EMPHPOST_AT_4G_SH_FIX		0x00000002
+#define __EMPHPRE_AT_4G_FIX		0x00000003
+#define __SFP_TXRATE_EN_FIX		0x00000100
+#define __SFP_RXRATE_EN_FIX		0x00000080
+
+/*
+ * These register definitions are auto-generated from hard coded values
+ * in regparse.pl.
+ */
+
+/*
+ * These register mapping definitions are auto-generated from mapping tables
+ * in regparse.pl.
+ */
+#define BFA_IOC0_HBEAT_REG		HOST_SEM0_INFO_REG
+#define BFA_IOC0_STATE_REG		HOST_SEM1_INFO_REG
+#define BFA_IOC1_HBEAT_REG		HOST_SEM2_INFO_REG
+#define BFA_IOC1_STATE_REG		HOST_SEM3_INFO_REG
+#define BFA_FW_USE_COUNT		 HOST_SEM4_INFO_REG
+
+#define CPE_DEPTH_Q(__n) \
+	(CPE_DEPTH_Q0 + (__n) * (CPE_DEPTH_Q1 - CPE_DEPTH_Q0))
+#define CPE_QCTRL_Q(__n) \
+	(CPE_QCTRL_Q0 + (__n) * (CPE_QCTRL_Q1 - CPE_QCTRL_Q0))
+#define CPE_PI_PTR_Q(__n) \
+	(CPE_PI_PTR_Q0 + (__n) * (CPE_PI_PTR_Q1 - CPE_PI_PTR_Q0))
+#define CPE_CI_PTR_Q(__n) \
+	(CPE_CI_PTR_Q0 + (__n) * (CPE_CI_PTR_Q1 - CPE_CI_PTR_Q0))
+#define RME_DEPTH_Q(__n) \
+	(RME_DEPTH_Q0 + (__n) * (RME_DEPTH_Q1 - RME_DEPTH_Q0))
+#define RME_QCTRL_Q(__n) \
+	(RME_QCTRL_Q0 + (__n) * (RME_QCTRL_Q1 - RME_QCTRL_Q0))
+#define RME_PI_PTR_Q(__n) \
+	(RME_PI_PTR_Q0 + (__n) * (RME_PI_PTR_Q1 - RME_PI_PTR_Q0))
+#define RME_CI_PTR_Q(__n) \
+	(RME_CI_PTR_Q0 + (__n) * (RME_CI_PTR_Q1 - RME_CI_PTR_Q0))
+#define HQM_QSET_RXQ_DRBL_P0(__n) (HQM_QSET0_RXQ_DRBL_P0 + (__n) \
+	* (HQM_QSET1_RXQ_DRBL_P0 - HQM_QSET0_RXQ_DRBL_P0))
+#define HQM_QSET_TXQ_DRBL_P0(__n) (HQM_QSET0_TXQ_DRBL_P0 + (__n) \
+	* (HQM_QSET1_TXQ_DRBL_P0 - HQM_QSET0_TXQ_DRBL_P0))
+#define HQM_QSET_IB_DRBL_1_P0(__n) (HQM_QSET0_IB_DRBL_1_P0 + (__n) \
+	* (HQM_QSET1_IB_DRBL_1_P0 - HQM_QSET0_IB_DRBL_1_P0))
+#define HQM_QSET_IB_DRBL_2_P0(__n) (HQM_QSET0_IB_DRBL_2_P0 + (__n) \
+	* (HQM_QSET1_IB_DRBL_2_P0 - HQM_QSET0_IB_DRBL_2_P0))
+#define HQM_QSET_RXQ_DRBL_P1(__n) (HQM_QSET0_RXQ_DRBL_P1 + (__n) \
+	* (HQM_QSET1_RXQ_DRBL_P1 - HQM_QSET0_RXQ_DRBL_P1))
+#define HQM_QSET_TXQ_DRBL_P1(__n) (HQM_QSET0_TXQ_DRBL_P1 + (__n) \
+	* (HQM_QSET1_TXQ_DRBL_P1 - HQM_QSET0_TXQ_DRBL_P1))
+#define HQM_QSET_IB_DRBL_1_P1(__n) (HQM_QSET0_IB_DRBL_1_P1 + (__n) \
+	* (HQM_QSET1_IB_DRBL_1_P1 - HQM_QSET0_IB_DRBL_1_P1))
+#define HQM_QSET_IB_DRBL_2_P1(__n) (HQM_QSET0_IB_DRBL_2_P1 + (__n) \
+	* (HQM_QSET1_IB_DRBL_2_P1 - HQM_QSET0_IB_DRBL_2_P1))
+
+#define CPE_Q_NUM(__fn, __q) (((__fn) << 2) + (__q))
+#define RME_Q_NUM(__fn, __q) (((__fn) << 2) + (__q))
+#define CPE_Q_MASK(__q) ((__q) & 0x3)
+#define RME_Q_MASK(__q) ((__q) & 0x3)
+
+/*
+ * PCI MSI-X vector defines
+ */
+enum {
+	BFA_MSIX_CPE_Q0 = 0,
+	BFA_MSIX_CPE_Q1 = 1,
+	BFA_MSIX_CPE_Q2 = 2,
+	BFA_MSIX_CPE_Q3 = 3,
+	BFA_MSIX_RME_Q0 = 4,
+	BFA_MSIX_RME_Q1 = 5,
+	BFA_MSIX_RME_Q2 = 6,
+	BFA_MSIX_RME_Q3 = 7,
+	BFA_MSIX_LPU_ERR = 8,
+	BFA_MSIX_CT_MAX = 9,
+};
+
+/*
+ * And corresponding host interrupt status bit field defines
+ */
+#define __HFN_INT_CPE_Q0		0x00000001U
+#define __HFN_INT_CPE_Q1		0x00000002U
+#define __HFN_INT_CPE_Q2		0x00000004U
+#define __HFN_INT_CPE_Q3		0x00000008U
+#define __HFN_INT_CPE_Q4		0x00000010U
+#define __HFN_INT_CPE_Q5		0x00000020U
+#define __HFN_INT_CPE_Q6		0x00000040U
+#define __HFN_INT_CPE_Q7		0x00000080U
+#define __HFN_INT_RME_Q0		0x00000100U
+#define __HFN_INT_RME_Q1		0x00000200U
+#define __HFN_INT_RME_Q2		0x00000400U
+#define __HFN_INT_RME_Q3		0x00000800U
+#define __HFN_INT_RME_Q4		0x00001000U
+#define __HFN_INT_RME_Q5		0x00002000U
+#define __HFN_INT_RME_Q6		0x00004000U
+#define __HFN_INT_RME_Q7		0x00008000U
+#define __HFN_INT_ERR_EMC		0x00010000U
+#define __HFN_INT_ERR_LPU0		0x00020000U
+#define __HFN_INT_ERR_LPU1		0x00040000U
+#define __HFN_INT_ERR_PSS		0x00080000U
+#define __HFN_INT_MBOX_LPU0		0x00100000U
+#define __HFN_INT_MBOX_LPU1		0x00200000U
+#define __HFN_INT_MBOX1_LPU0		0x00400000U
+#define __HFN_INT_MBOX1_LPU1		0x00800000U
+#define __HFN_INT_LL_HALT		0x01000000U
+#define __HFN_INT_CPE_MASK		0x000000ffU
+#define __HFN_INT_RME_MASK		0x0000ff00U
+
+/*
+ * catapult memory map.
+ */
+#define LL_PGN_HQM0		0x0096
+#define LL_PGN_HQM1		0x0097
+#define PSS_SMEM_PAGE_START	0x8000
+#define PSS_SMEM_PGNUM(_pg0, _ma)	((_pg0) + ((_ma) >> 15))
+#define PSS_SMEM_PGOFF(_ma)	((_ma) & 0x7fff)
+
+/*
+ * End of catapult memory map
+ */
+
+#endif /* __BFI_CTREG_H__ */
diff --git a/drivers/net/bna/bfi_ll.h b/drivers/net/bna/bfi_ll.h
new file mode 100644
index 000000000000..bee4d054066a
--- /dev/null
+++ b/drivers/net/bna/bfi_ll.h
@@ -0,0 +1,438 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#ifndef __BFI_LL_H__
+#define __BFI_LL_H__
+
+#include "bfi.h"
+
+#pragma pack(1)
+
+/**
+ * @brief
+ *	"enums" for all LL mailbox messages other than IOC
+ */
+enum {
+	BFI_LL_H2I_MAC_UCAST_SET_REQ = 1,
+	BFI_LL_H2I_MAC_UCAST_ADD_REQ = 2,
+	BFI_LL_H2I_MAC_UCAST_DEL_REQ = 3,
+
+	BFI_LL_H2I_MAC_MCAST_ADD_REQ = 4,
+	BFI_LL_H2I_MAC_MCAST_DEL_REQ = 5,
+	BFI_LL_H2I_MAC_MCAST_FILTER_REQ = 6,
+	BFI_LL_H2I_MAC_MCAST_DEL_ALL_REQ = 7,
+
+	BFI_LL_H2I_PORT_ADMIN_REQ = 8,
+	BFI_LL_H2I_STATS_GET_REQ = 9,
+	BFI_LL_H2I_STATS_CLEAR_REQ = 10,
+
+	BFI_LL_H2I_RXF_PROMISCUOUS_SET_REQ = 11,
+	BFI_LL_H2I_RXF_DEFAULT_SET_REQ = 12,
+
+	BFI_LL_H2I_TXQ_STOP_REQ = 13,
+	BFI_LL_H2I_RXQ_STOP_REQ = 14,
+
+	BFI_LL_H2I_DIAG_LOOPBACK_REQ = 15,
+
+	BFI_LL_H2I_SET_PAUSE_REQ = 16,
+	BFI_LL_H2I_MTU_INFO_REQ = 17,
+
+	BFI_LL_H2I_RX_REQ = 18,
+} ;
+
+enum {
+	BFI_LL_I2H_MAC_UCAST_SET_RSP = BFA_I2HM(1),
+	BFI_LL_I2H_MAC_UCAST_ADD_RSP = BFA_I2HM(2),
+	BFI_LL_I2H_MAC_UCAST_DEL_RSP = BFA_I2HM(3),
+
+	BFI_LL_I2H_MAC_MCAST_ADD_RSP = BFA_I2HM(4),
+	BFI_LL_I2H_MAC_MCAST_DEL_RSP = BFA_I2HM(5),
+	BFI_LL_I2H_MAC_MCAST_FILTER_RSP = BFA_I2HM(6),
+	BFI_LL_I2H_MAC_MCAST_DEL_ALL_RSP = BFA_I2HM(7),
+
+	BFI_LL_I2H_PORT_ADMIN_RSP = BFA_I2HM(8),
+	BFI_LL_I2H_STATS_GET_RSP = BFA_I2HM(9),
+	BFI_LL_I2H_STATS_CLEAR_RSP = BFA_I2HM(10),
+
+	BFI_LL_I2H_RXF_PROMISCUOUS_SET_RSP = BFA_I2HM(11),
+	BFI_LL_I2H_RXF_DEFAULT_SET_RSP = BFA_I2HM(12),
+
+	BFI_LL_I2H_TXQ_STOP_RSP = BFA_I2HM(13),
+	BFI_LL_I2H_RXQ_STOP_RSP = BFA_I2HM(14),
+
+	BFI_LL_I2H_DIAG_LOOPBACK_RSP = BFA_I2HM(15),
+
+	BFI_LL_I2H_SET_PAUSE_RSP = BFA_I2HM(16),
+
+	BFI_LL_I2H_MTU_INFO_RSP = BFA_I2HM(17),
+	BFI_LL_I2H_RX_RSP = BFA_I2HM(18),
+
+	BFI_LL_I2H_LINK_DOWN_AEN = BFA_I2HM(19),
+	BFI_LL_I2H_LINK_UP_AEN = BFA_I2HM(20),
+
+	BFI_LL_I2H_PORT_ENABLE_AEN = BFA_I2HM(21),
+	BFI_LL_I2H_PORT_DISABLE_AEN = BFA_I2HM(22),
+} ;
+
+/**
+ * @brief bfi_ll_mac_addr_req is used by:
+ *        BFI_LL_H2I_MAC_UCAST_SET_REQ
+ *        BFI_LL_H2I_MAC_UCAST_ADD_REQ
+ *        BFI_LL_H2I_MAC_UCAST_DEL_REQ
+ *        BFI_LL_H2I_MAC_MCAST_ADD_REQ
+ *        BFI_LL_H2I_MAC_MCAST_DEL_REQ
+ */
+struct bfi_ll_mac_addr_req {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u8		rxf_id;
+	u8		rsvd1[3];
+	mac_t		mac_addr;
+	u8		rsvd2[2];
+};
+
+/**
+ * @brief bfi_ll_mcast_filter_req is used by:
+ *	  BFI_LL_H2I_MAC_MCAST_FILTER_REQ
+ */
+struct bfi_ll_mcast_filter_req {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u8		rxf_id;
+	u8		enable;
+	u8		rsvd[2];
+};
+
+/**
+ * @brief bfi_ll_mcast_del_all is used by:
+ *	  BFI_LL_H2I_MAC_MCAST_DEL_ALL_REQ
+ */
+struct bfi_ll_mcast_del_all_req {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u8		   rxf_id;
+	u8		   rsvd[3];
+};
+
+/**
+ * @brief bfi_ll_q_stop_req is used by:
+ *	BFI_LL_H2I_TXQ_STOP_REQ
+ *	BFI_LL_H2I_RXQ_STOP_REQ
+ */
+struct bfi_ll_q_stop_req {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u32	q_id_mask[2];	/* !< bit-mask for queue ids */
+};
+
+/**
+ * @brief bfi_ll_stats_req is used by:
+ *    BFI_LL_I2H_STATS_GET_REQ
+ *    BFI_LL_I2H_STATS_CLEAR_REQ
+ */
+struct bfi_ll_stats_req {
+	struct bfi_mhdr mh;	/*!< common msg header */
+	u16 stats_mask;	/* !< bit-mask for non-function statistics */
+	u8	rsvd[2];
+	u32 rxf_id_mask[2];	/* !< bit-mask for RxF Statistics */
+	u32 txf_id_mask[2];	/* !< bit-mask for TxF Statistics */
+	union bfi_addr_u  host_buffer;	/* !< where statistics are returned */
+};
+
+/**
+ * @brief defines for "stats_mask" above.
+ */
+#define BFI_LL_STATS_MAC	(1 << 0)	/* !< MAC Statistics */
+#define BFI_LL_STATS_BPC	(1 << 1)	/* !< Pause Stats from BPC */
+#define BFI_LL_STATS_RAD	(1 << 2)	/* !< Rx Admission Statistics */
+#define BFI_LL_STATS_RX_FC	(1 << 3)	/* !< Rx FC Stats from RxA */
+#define BFI_LL_STATS_TX_FC	(1 << 4)	/* !< Tx FC Stats from TxA */
+
+#define BFI_LL_STATS_ALL	0x1f
+
+/**
+ * @brief bfi_ll_port_admin_req
+ */
+struct bfi_ll_port_admin_req {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u8		 up;
+	u8		 rsvd[3];
+};
+
+/**
+ * @brief bfi_ll_rxf_req is used by:
+ *      BFI_LL_H2I_RXF_PROMISCUOUS_SET_REQ
+ *      BFI_LL_H2I_RXF_DEFAULT_SET_REQ
+ */
+struct bfi_ll_rxf_req {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u8		rxf_id;
+	u8		enable;
+	u8		rsvd[2];
+};
+
+/**
+ * @brief bfi_ll_rxf_multi_req is used by:
+ *	BFI_LL_H2I_RX_REQ
+ */
+struct bfi_ll_rxf_multi_req {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u32	rxf_id_mask[2];
+	u8		enable;
+	u8		rsvd[3];
+};
+
+/**
+ * @brief enum for Loopback opmodes
+ */
+enum {
+	BFI_LL_DIAG_LB_OPMODE_EXT = 0,
+	BFI_LL_DIAG_LB_OPMODE_CBL = 1,
+};
+
+/**
+ * @brief bfi_ll_set_pause_req is used by:
+ *	BFI_LL_H2I_SET_PAUSE_REQ
+ */
+struct bfi_ll_set_pause_req {
+	struct bfi_mhdr mh;
+	u8		tx_pause; /* 1 = enable, 0 =  disable */
+	u8		rx_pause; /* 1 = enable, 0 =  disable */
+	u8		rsvd[2];
+};
+
+/**
+ * @brief bfi_ll_mtu_info_req is used by:
+ *	BFI_LL_H2I_MTU_INFO_REQ
+ */
+struct bfi_ll_mtu_info_req {
+	struct bfi_mhdr mh;
+	u16	mtu;
+	u8		rsvd[2];
+};
+
+/**
+ * @brief
+ *	  Response header format used by all responses
+ *	  For both responses and asynchronous notifications
+ */
+struct bfi_ll_rsp {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u8		error;
+	u8		rsvd[3];
+};
+
+/**
+ * @brief bfi_ll_cee_aen is used by:
+ *	BFI_LL_I2H_LINK_DOWN_AEN
+ *	BFI_LL_I2H_LINK_UP_AEN
+ */
+struct bfi_ll_aen {
+	struct bfi_mhdr mh;		/*!< common msg header */
+	u32	reason;
+	u8		cee_linkup;
+	u8		prio_map;    /*!< LL priority bit-map */
+	u8		rsvd[2];
+};
+
+/**
+ * @brief
+ * 	The following error codes can be returned
+ *	by the mbox commands
+ */
+enum {
+	BFI_LL_CMD_OK 		= 0,
+	BFI_LL_CMD_FAIL 	= 1,
+	BFI_LL_CMD_DUP_ENTRY	= 2,	/* !< Duplicate entry in CAM */
+	BFI_LL_CMD_CAM_FULL	= 3,	/* !< CAM is full */
+	BFI_LL_CMD_NOT_OWNER	= 4,   	/* !< Not permitted, b'cos not owner */
+	BFI_LL_CMD_NOT_EXEC	= 5,   	/* !< Was not sent to f/w at all */
+	BFI_LL_CMD_WAITING	= 6,	/* !< Waiting for completion (VMware) */
+	BFI_LL_CMD_PORT_DISABLED	= 7,	/* !< port in disabled state */
+} ;
+
+/* Statistics */
+#define BFI_LL_TXF_ID_MAX  	64
+#define BFI_LL_RXF_ID_MAX  	64
+
+/* TxF Frame Statistics */
+struct bfi_ll_stats_txf {
+	u64 ucast_octets;
+	u64 ucast;
+	u64 ucast_vlan;
+
+	u64 mcast_octets;
+	u64 mcast;
+	u64 mcast_vlan;
+
+	u64 bcast_octets;
+	u64 bcast;
+	u64 bcast_vlan;
+
+	u64 errors;
+	u64 filter_vlan;      /* frames filtered due to VLAN */
+	u64 filter_mac_sa;    /* frames filtered due to SA check */
+};
+
+/* RxF Frame Statistics */
+struct bfi_ll_stats_rxf {
+	u64 ucast_octets;
+	u64 ucast;
+	u64 ucast_vlan;
+
+	u64 mcast_octets;
+	u64 mcast;
+	u64 mcast_vlan;
+
+	u64 bcast_octets;
+	u64 bcast;
+	u64 bcast_vlan;
+	u64 frame_drops;
+};
+
+/* FC Tx Frame Statistics */
+struct bfi_ll_stats_fc_tx {
+	u64 txf_ucast_octets;
+	u64 txf_ucast;
+	u64 txf_ucast_vlan;
+
+	u64 txf_mcast_octets;
+	u64 txf_mcast;
+	u64 txf_mcast_vlan;
+
+	u64 txf_bcast_octets;
+	u64 txf_bcast;
+	u64 txf_bcast_vlan;
+
+	u64 txf_parity_errors;
+	u64 txf_timeout;
+	u64 txf_fid_parity_errors;
+};
+
+/* FC Rx Frame Statistics */
+struct bfi_ll_stats_fc_rx {
+	u64 rxf_ucast_octets;
+	u64 rxf_ucast;
+	u64 rxf_ucast_vlan;
+
+	u64 rxf_mcast_octets;
+	u64 rxf_mcast;
+	u64 rxf_mcast_vlan;
+
+	u64 rxf_bcast_octets;
+	u64 rxf_bcast;
+	u64 rxf_bcast_vlan;
+};
+
+/* RAD Frame Statistics */
+struct bfi_ll_stats_rad {
+	u64 rx_frames;
+	u64 rx_octets;
+	u64 rx_vlan_frames;
+
+	u64 rx_ucast;
+	u64 rx_ucast_octets;
+	u64 rx_ucast_vlan;
+
+	u64 rx_mcast;
+	u64 rx_mcast_octets;
+	u64 rx_mcast_vlan;
+
+	u64 rx_bcast;
+	u64 rx_bcast_octets;
+	u64 rx_bcast_vlan;
+
+	u64 rx_drops;
+};
+
+/* BPC Tx Registers */
+struct bfi_ll_stats_bpc {
+	/* transmit stats */
+	u64 tx_pause[8];
+	u64 tx_zero_pause[8];	/*!< Pause cancellation */
+	/*!<Pause initiation rather than retention */
+	u64 tx_first_pause[8];
+
+	/* receive stats */
+	u64 rx_pause[8];
+	u64 rx_zero_pause[8];	/*!< Pause cancellation */
+	/*!<Pause initiation rather than retention */
+	u64 rx_first_pause[8];
+};
+
+/* MAC Rx Statistics */
+struct bfi_ll_stats_mac {
+	u64 frame_64;		/* both rx and tx counter */
+	u64 frame_65_127;		/* both rx and tx counter */
+	u64 frame_128_255;		/* both rx and tx counter */
+	u64 frame_256_511;		/* both rx and tx counter */
+	u64 frame_512_1023;	/* both rx and tx counter */
+	u64 frame_1024_1518;	/* both rx and tx counter */
+	u64 frame_1519_1522;	/* both rx and tx counter */
+
+	/* receive stats */
+	u64 rx_bytes;
+	u64 rx_packets;
+	u64 rx_fcs_error;
+	u64 rx_multicast;
+	u64 rx_broadcast;
+	u64 rx_control_frames;
+	u64 rx_pause;
+	u64 rx_unknown_opcode;
+	u64 rx_alignment_error;
+	u64 rx_frame_length_error;
+	u64 rx_code_error;
+	u64 rx_carrier_sense_error;
+	u64 rx_undersize;
+	u64 rx_oversize;
+	u64 rx_fragments;
+	u64 rx_jabber;
+	u64 rx_drop;
+
+	/* transmit stats */
+	u64 tx_bytes;
+	u64 tx_packets;
+	u64 tx_multicast;
+	u64 tx_broadcast;
+	u64 tx_pause;
+	u64 tx_deferral;
+	u64 tx_excessive_deferral;
+	u64 tx_single_collision;
+	u64 tx_muliple_collision;
+	u64 tx_late_collision;
+	u64 tx_excessive_collision;
+	u64 tx_total_collision;
+	u64 tx_pause_honored;
+	u64 tx_drop;
+	u64 tx_jabber;
+	u64 tx_fcs_error;
+	u64 tx_control_frame;
+	u64 tx_oversize;
+	u64 tx_undersize;
+	u64 tx_fragments;
+};
+
+/* Complete statistics */
+struct bfi_ll_stats {
+	struct bfi_ll_stats_mac		mac_stats;
+	struct bfi_ll_stats_bpc		bpc_stats;
+	struct bfi_ll_stats_rad		rad_stats;
+	struct bfi_ll_stats_fc_rx	fc_rx_stats;
+	struct bfi_ll_stats_fc_tx	fc_tx_stats;
+	struct bfi_ll_stats_rxf	rxf_stats[BFI_LL_RXF_ID_MAX];
+	struct bfi_ll_stats_txf	txf_stats[BFI_LL_TXF_ID_MAX];
+};
+
+#pragma pack()
+
+#endif  /* __BFI_LL_H__ */
diff --git a/drivers/net/bna/bna.h b/drivers/net/bna/bna.h
new file mode 100644
index 000000000000..6a2b3291c190
--- /dev/null
+++ b/drivers/net/bna/bna.h
@@ -0,0 +1,654 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+#ifndef __BNA_H__
+#define __BNA_H__
+
+#include "bfa_wc.h"
+#include "bfa_ioc.h"
+#include "cna.h"
+#include "bfi_ll.h"
+#include "bna_types.h"
+
+extern u32 bna_dim_vector[][BNA_BIAS_T_MAX];
+extern u32 bna_napi_dim_vector[][BNA_BIAS_T_MAX];
+
+/**
+ *
+ *  Macros and constants
+ *
+ */
+
+#define BNA_IOC_TIMER_FREQ		200
+
+/* Log string size */
+#define BNA_MESSAGE_SIZE		256
+
+#define bna_device_timer(_dev)		bfa_timer_beat(&((_dev)->timer_mod))
+
+/* MBOX API for PORT, TX, RX */
+#define bna_mbox_qe_fill(_qe, _cmd, _cmd_len, _cbfn, _cbarg)		\
+do {									\
+	memcpy(&((_qe)->cmd.msg[0]), (_cmd), (_cmd_len));	\
+	(_qe)->cbfn = (_cbfn);						\
+	(_qe)->cbarg = (_cbarg);					\
+} while (0)
+
+#define bna_is_small_rxq(rcb) ((rcb)->id == 1)
+
+#define BNA_MAC_IS_EQUAL(_mac1, _mac2)					\
+	(!memcmp((_mac1), (_mac2), sizeof(mac_t)))
+
+#define BNA_POWER_OF_2(x) (((x) & ((x) - 1)) == 0)
+
+#define BNA_TO_POWER_OF_2(x)						\
+do {									\
+	int _shift = 0;							\
+	while ((x) && (x) != 1) {					\
+		(x) >>= 1;						\
+		_shift++;						\
+	}								\
+	(x) <<= _shift;							\
+} while (0)
+
+#define BNA_TO_POWER_OF_2_HIGH(x)					\
+do {									\
+	int n = 1;							\
+	while (n < (x))							\
+		n <<= 1;						\
+	(x) = n;							\
+} while (0)
+
+/*
+ * input : _addr-> os dma addr in host endian format,
+ * output : _bna_dma_addr-> pointer to hw dma addr
+ */
+#define BNA_SET_DMA_ADDR(_addr, _bna_dma_addr)				\
+do {									\
+	u64 tmp_addr =						\
+	cpu_to_be64((u64)(_addr));				\
+	(_bna_dma_addr)->msb = ((struct bna_dma_addr *)&tmp_addr)->msb; \
+	(_bna_dma_addr)->lsb = ((struct bna_dma_addr *)&tmp_addr)->lsb; \
+} while (0)
+
+/*
+ * input : _bna_dma_addr-> pointer to hw dma addr
+ * output : _addr-> os dma addr in host endian format
+ */
+#define BNA_GET_DMA_ADDR(_bna_dma_addr, _addr)			\
+do {								\
+	(_addr) = ((((u64)ntohl((_bna_dma_addr)->msb))) << 32)		\
+	| ((ntohl((_bna_dma_addr)->lsb) & 0xffffffff));	\
+} while (0)
+
+#define	containing_rec(addr, type, field)				\
+	((type *)((unsigned char *)(addr) - 				\
+	(unsigned char *)(&((type *)0)->field)))
+
+#define BNA_TXQ_WI_NEEDED(_vectors)	(((_vectors) + 3) >> 2)
+
+/* TxQ element is 64 bytes */
+#define BNA_TXQ_PAGE_INDEX_MAX		(PAGE_SIZE >> 6)
+#define BNA_TXQ_PAGE_INDEX_MAX_SHIFT	(PAGE_SHIFT - 6)
+
+#define BNA_TXQ_QPGE_PTR_GET(_qe_idx, _qpt_ptr, _qe_ptr, _qe_ptr_range) \
+{									\
+	unsigned int page_index;	/* index within a page */	\
+	void *page_addr;						\
+	page_index = (_qe_idx) & (BNA_TXQ_PAGE_INDEX_MAX - 1); 		\
+	(_qe_ptr_range) = (BNA_TXQ_PAGE_INDEX_MAX - page_index); 	\
+	page_addr = (_qpt_ptr)[((_qe_idx) >>  BNA_TXQ_PAGE_INDEX_MAX_SHIFT)];\
+	(_qe_ptr) = &((struct bna_txq_entry *)(page_addr))[page_index]; \
+}
+
+/* RxQ element is 8 bytes */
+#define BNA_RXQ_PAGE_INDEX_MAX		(PAGE_SIZE >> 3)
+#define BNA_RXQ_PAGE_INDEX_MAX_SHIFT	(PAGE_SHIFT - 3)
+
+#define BNA_RXQ_QPGE_PTR_GET(_qe_idx, _qpt_ptr, _qe_ptr, _qe_ptr_range) \
+{									\
+	unsigned int page_index;	/* index within a page */	\
+	void *page_addr;						\
+	page_index = (_qe_idx) & (BNA_RXQ_PAGE_INDEX_MAX - 1);		\
+	(_qe_ptr_range) = (BNA_RXQ_PAGE_INDEX_MAX - page_index);	\
+	page_addr = (_qpt_ptr)[((_qe_idx) >>				\
+				BNA_RXQ_PAGE_INDEX_MAX_SHIFT)];		\
+	(_qe_ptr) = &((struct bna_rxq_entry *)(page_addr))[page_index]; \
+}
+
+/* CQ element is 16 bytes */
+#define BNA_CQ_PAGE_INDEX_MAX		(PAGE_SIZE >> 4)
+#define BNA_CQ_PAGE_INDEX_MAX_SHIFT	(PAGE_SHIFT - 4)
+
+#define BNA_CQ_QPGE_PTR_GET(_qe_idx, _qpt_ptr, _qe_ptr, _qe_ptr_range)	\
+{									\
+	unsigned int page_index;	  /* index within a page */	\
+	void *page_addr;						\
+									\
+	page_index = (_qe_idx) & (BNA_CQ_PAGE_INDEX_MAX - 1);		\
+	(_qe_ptr_range) = (BNA_CQ_PAGE_INDEX_MAX - page_index);		\
+	page_addr = (_qpt_ptr)[((_qe_idx) >>				\
+				    BNA_CQ_PAGE_INDEX_MAX_SHIFT)];	\
+	(_qe_ptr) = &((struct bna_cq_entry *)(page_addr))[page_index];\
+}
+
+#define BNA_QE_INDX_2_PTR(_cast, _qe_idx, _q_base)			\
+	(&((_cast *)(_q_base))[(_qe_idx)])
+
+#define BNA_QE_INDX_RANGE(_qe_idx, _q_depth) ((_q_depth) - (_qe_idx))
+
+#define BNA_QE_INDX_ADD(_qe_idx, _qe_num, _q_depth)			\
+	((_qe_idx) = ((_qe_idx) + (_qe_num)) & ((_q_depth) - 1))
+
+#define BNA_Q_INDEX_CHANGE(_old_idx, _updated_idx, _q_depth)		\
+	(((_updated_idx) - (_old_idx)) & ((_q_depth) - 1))
+
+#define BNA_QE_FREE_CNT(_q_ptr, _q_depth)				\
+	(((_q_ptr)->consumer_index - (_q_ptr)->producer_index - 1) &	\
+	 ((_q_depth) - 1))
+
+#define BNA_QE_IN_USE_CNT(_q_ptr, _q_depth)				\
+	((((_q_ptr)->producer_index - (_q_ptr)->consumer_index)) &	\
+	 (_q_depth - 1))
+
+#define BNA_Q_GET_CI(_q_ptr)		((_q_ptr)->q.consumer_index)
+
+#define BNA_Q_GET_PI(_q_ptr)		((_q_ptr)->q.producer_index)
+
+#define BNA_Q_PI_ADD(_q_ptr, _num)					\
+	(_q_ptr)->q.producer_index =					\
+		(((_q_ptr)->q.producer_index + (_num)) &		\
+		((_q_ptr)->q.q_depth - 1))
+
+#define BNA_Q_CI_ADD(_q_ptr, _num) 					\
+	(_q_ptr)->q.consumer_index =					\
+		(((_q_ptr)->q.consumer_index + (_num))  		\
+		& ((_q_ptr)->q.q_depth - 1))
+
+#define BNA_Q_FREE_COUNT(_q_ptr)					\
+	(BNA_QE_FREE_CNT(&((_q_ptr)->q), (_q_ptr)->q.q_depth))
+
+#define BNA_Q_IN_USE_COUNT(_q_ptr)  					\
+	(BNA_QE_IN_USE_CNT(&(_q_ptr)->q, (_q_ptr)->q.q_depth))
+
+/* These macros build the data portion of the TxQ/RxQ doorbell */
+#define BNA_DOORBELL_Q_PRD_IDX(_pi) 	(0x80000000 | (_pi))
+#define BNA_DOORBELL_Q_STOP		(0x40000000)
+
+/* These macros build the data portion of the IB doorbell */
+#define BNA_DOORBELL_IB_INT_ACK(_timeout, _events) \
+	(0x80000000 | ((_timeout) << 16) | (_events))
+#define BNA_DOORBELL_IB_INT_DISABLE 	(0x40000000)
+
+/* Set the coalescing timer for the given ib */
+#define bna_ib_coalescing_timer_set(_i_dbell, _cls_timer)		\
+	((_i_dbell)->doorbell_ack = BNA_DOORBELL_IB_INT_ACK((_cls_timer), 0));
+
+/* Acks 'events' # of events for a given ib */
+#define bna_ib_ack(_i_dbell, _events)					\
+	(writel(((_i_dbell)->doorbell_ack | (_events)), \
+		(_i_dbell)->doorbell_addr));
+
+#define bna_txq_prod_indx_doorbell(_tcb)				\
+	(writel(BNA_DOORBELL_Q_PRD_IDX((_tcb)->producer_index), \
+		(_tcb)->q_dbell));
+
+#define bna_rxq_prod_indx_doorbell(_rcb)				\
+	(writel(BNA_DOORBELL_Q_PRD_IDX((_rcb)->producer_index), \
+		(_rcb)->q_dbell));
+
+#define BNA_LARGE_PKT_SIZE		1000
+
+#define BNA_UPDATE_PKT_CNT(_pkt, _len)					\
+do {									\
+	if ((_len) > BNA_LARGE_PKT_SIZE) {				\
+		(_pkt)->large_pkt_cnt++;				\
+	} else {							\
+		(_pkt)->small_pkt_cnt++;				\
+	}								\
+} while (0)
+
+#define	call_rxf_stop_cbfn(rxf, status)					\
+	if ((rxf)->stop_cbfn) {						\
+		(*(rxf)->stop_cbfn)((rxf)->stop_cbarg, (status));	\
+		(rxf)->stop_cbfn = NULL;				\
+		(rxf)->stop_cbarg = NULL;				\
+	}
+
+#define	call_rxf_start_cbfn(rxf, status)				\
+	if ((rxf)->start_cbfn) {					\
+		(*(rxf)->start_cbfn)((rxf)->start_cbarg, (status));	\
+		(rxf)->start_cbfn = NULL;				\
+		(rxf)->start_cbarg = NULL;				\
+	}
+
+#define	call_rxf_cam_fltr_cbfn(rxf, status)				\
+	if ((rxf)->cam_fltr_cbfn) {					\
+		(*(rxf)->cam_fltr_cbfn)((rxf)->cam_fltr_cbarg, rxf->rx,	\
+					(status));			\
+		(rxf)->cam_fltr_cbfn = NULL;				\
+		(rxf)->cam_fltr_cbarg = NULL;				\
+	}
+
+#define	call_rxf_pause_cbfn(rxf, status)				\
+	if ((rxf)->oper_state_cbfn) {					\
+		(*(rxf)->oper_state_cbfn)((rxf)->oper_state_cbarg, rxf->rx,\
+					(status));			\
+		(rxf)->rxf_flags &= ~BNA_RXF_FL_OPERSTATE_CHANGED;	\
+		(rxf)->oper_state_cbfn = NULL;				\
+		(rxf)->oper_state_cbarg = NULL;				\
+	}
+
+#define	call_rxf_resume_cbfn(rxf, status) call_rxf_pause_cbfn(rxf, status)
+
+#define is_xxx_enable(mode, bitmask, xxx) ((bitmask & xxx) && (mode & xxx))
+
+#define is_xxx_disable(mode, bitmask, xxx) ((bitmask & xxx) && !(mode & xxx))
+
+#define xxx_enable(mode, bitmask, xxx)					\
+do {									\
+	bitmask |= xxx;							\
+	mode |= xxx;							\
+} while (0)
+
+#define xxx_disable(mode, bitmask, xxx)					\
+do {									\
+	bitmask |= xxx;							\
+	mode &= ~xxx;							\
+} while (0)
+
+#define xxx_inactive(mode, bitmask, xxx)				\
+do {									\
+	bitmask &= ~xxx;						\
+	mode &= ~xxx;							\
+} while (0)
+
+#define is_promisc_enable(mode, bitmask)				\
+	is_xxx_enable(mode, bitmask, BNA_RXMODE_PROMISC)
+
+#define is_promisc_disable(mode, bitmask)				\
+	is_xxx_disable(mode, bitmask, BNA_RXMODE_PROMISC)
+
+#define promisc_enable(mode, bitmask)					\
+	xxx_enable(mode, bitmask, BNA_RXMODE_PROMISC)
+
+#define promisc_disable(mode, bitmask)					\
+	xxx_disable(mode, bitmask, BNA_RXMODE_PROMISC)
+
+#define promisc_inactive(mode, bitmask)					\
+	xxx_inactive(mode, bitmask, BNA_RXMODE_PROMISC)
+
+#define is_default_enable(mode, bitmask)				\
+	is_xxx_enable(mode, bitmask, BNA_RXMODE_DEFAULT)
+
+#define is_default_disable(mode, bitmask)				\
+	is_xxx_disable(mode, bitmask, BNA_RXMODE_DEFAULT)
+
+#define default_enable(mode, bitmask)					\
+	xxx_enable(mode, bitmask, BNA_RXMODE_DEFAULT)
+
+#define default_disable(mode, bitmask)					\
+	xxx_disable(mode, bitmask, BNA_RXMODE_DEFAULT)
+
+#define default_inactive(mode, bitmask)					\
+	xxx_inactive(mode, bitmask, BNA_RXMODE_DEFAULT)
+
+#define is_allmulti_enable(mode, bitmask)				\
+	is_xxx_enable(mode, bitmask, BNA_RXMODE_ALLMULTI)
+
+#define is_allmulti_disable(mode, bitmask)				\
+	is_xxx_disable(mode, bitmask, BNA_RXMODE_ALLMULTI)
+
+#define allmulti_enable(mode, bitmask)					\
+	xxx_enable(mode, bitmask, BNA_RXMODE_ALLMULTI)
+
+#define allmulti_disable(mode, bitmask)					\
+	xxx_disable(mode, bitmask, BNA_RXMODE_ALLMULTI)
+
+#define allmulti_inactive(mode, bitmask)				\
+	xxx_inactive(mode, bitmask, BNA_RXMODE_ALLMULTI)
+
+#define	GET_RXQS(rxp, q0, q1)	do {					\
+	switch ((rxp)->type) {						\
+	case BNA_RXP_SINGLE:						\
+		(q0) = rxp->rxq.single.only;				\
+		(q1) = NULL;						\
+		break;							\
+	case BNA_RXP_SLR:						\
+		(q0) = rxp->rxq.slr.large;				\
+		(q1) = rxp->rxq.slr.small;				\
+		break;							\
+	case BNA_RXP_HDS:						\
+		(q0) = rxp->rxq.hds.data;				\
+		(q1) = rxp->rxq.hds.hdr;				\
+		break;							\
+	}								\
+} while (0)
+
+/**
+ *
+ * Function prototypes
+ *
+ */
+
+/**
+ * BNA
+ */
+
+/* Internal APIs */
+void bna_adv_res_req(struct bna_res_info *res_info);
+
+/* APIs for BNAD */
+void bna_res_req(struct bna_res_info *res_info);
+void bna_init(struct bna *bna, struct bnad *bnad,
+			struct bfa_pcidev *pcidev,
+			struct bna_res_info *res_info);
+void bna_uninit(struct bna *bna);
+void bna_stats_get(struct bna *bna);
+void bna_stats_clr(struct bna *bna);
+void bna_get_perm_mac(struct bna *bna, u8 *mac);
+
+/* APIs for Rx */
+int bna_rit_mod_can_satisfy(struct bna_rit_mod *rit_mod, int seg_size);
+
+/* APIs for RxF */
+struct bna_mac *bna_ucam_mod_mac_get(struct bna_ucam_mod *ucam_mod);
+void bna_ucam_mod_mac_put(struct bna_ucam_mod *ucam_mod,
+			  struct bna_mac *mac);
+struct bna_mac *bna_mcam_mod_mac_get(struct bna_mcam_mod *mcam_mod);
+void bna_mcam_mod_mac_put(struct bna_mcam_mod *mcam_mod,
+			  struct bna_mac *mac);
+struct bna_rit_segment *
+bna_rit_mod_seg_get(struct bna_rit_mod *rit_mod, int seg_size);
+void bna_rit_mod_seg_put(struct bna_rit_mod *rit_mod,
+			struct bna_rit_segment *seg);
+
+/**
+ * DEVICE
+ */
+
+/* Interanl APIs */
+void bna_adv_device_init(struct bna_device *device, struct bna *bna,
+			struct bna_res_info *res_info);
+
+/* APIs for BNA */
+void bna_device_init(struct bna_device *device, struct bna *bna,
+		     struct bna_res_info *res_info);
+void bna_device_uninit(struct bna_device *device);
+void bna_device_cb_port_stopped(void *arg, enum bna_cb_status status);
+int bna_device_status_get(struct bna_device *device);
+int bna_device_state_get(struct bna_device *device);
+
+/* APIs for BNAD */
+void bna_device_enable(struct bna_device *device);
+void bna_device_disable(struct bna_device *device,
+			enum bna_cleanup_type type);
+
+/**
+ * MBOX
+ */
+
+/* APIs for DEVICE */
+void bna_mbox_mod_init(struct bna_mbox_mod *mbox_mod, struct bna *bna);
+void bna_mbox_mod_uninit(struct bna_mbox_mod *mbox_mod);
+void bna_mbox_mod_start(struct bna_mbox_mod *mbox_mod);
+void bna_mbox_mod_stop(struct bna_mbox_mod *mbox_mod);
+
+/* APIs for PORT, TX, RX */
+void bna_mbox_handler(struct bna *bna, u32 intr_status);
+void bna_mbox_send(struct bna *bna, struct bna_mbox_qe *mbox_qe);
+
+/**
+ * PORT
+ */
+
+/* APIs for BNA */
+void bna_port_init(struct bna_port *port, struct bna *bna);
+void bna_port_uninit(struct bna_port *port);
+int bna_port_state_get(struct bna_port *port);
+int bna_llport_state_get(struct bna_llport *llport);
+
+/* APIs for DEVICE */
+void bna_port_start(struct bna_port *port);
+void bna_port_stop(struct bna_port *port);
+void bna_port_fail(struct bna_port *port);
+
+/* API for RX */
+int bna_port_mtu_get(struct bna_port *port);
+void bna_llport_admin_up(struct bna_llport *llport);
+void bna_llport_admin_down(struct bna_llport *llport);
+
+/* API for BNAD */
+void bna_port_enable(struct bna_port *port);
+void bna_port_disable(struct bna_port *port, enum bna_cleanup_type type,
+		      void (*cbfn)(void *, enum bna_cb_status));
+void bna_port_pause_config(struct bna_port *port,
+			   struct bna_pause_config *pause_config,
+			   void (*cbfn)(struct bnad *, enum bna_cb_status));
+void bna_port_mtu_set(struct bna_port *port, int mtu,
+		      void (*cbfn)(struct bnad *, enum bna_cb_status));
+void bna_port_mac_get(struct bna_port *port, mac_t *mac);
+void bna_port_type_set(struct bna_port *port, enum bna_port_type type);
+void bna_port_linkcbfn_set(struct bna_port *port,
+			   void (*linkcbfn)(struct bnad *,
+					    enum bna_link_status));
+void bna_port_admin_up(struct bna_port *port);
+void bna_port_admin_down(struct bna_port *port);
+
+/* Callbacks for TX, RX */
+void bna_port_cb_tx_stopped(struct bna_port *port,
+			    enum bna_cb_status status);
+void bna_port_cb_rx_stopped(struct bna_port *port,
+			    enum bna_cb_status status);
+
+/* Callbacks for MBOX */
+void bna_port_cb_link_up(struct bna_port *port, struct bfi_ll_aen *aen,
+			 int status);
+void bna_port_cb_link_down(struct bna_port *port, int status);
+
+/**
+ * IB
+ */
+
+/* APIs for BNA */
+void bna_ib_mod_init(struct bna_ib_mod *ib_mod, struct bna *bna,
+		     struct bna_res_info *res_info);
+void bna_ib_mod_uninit(struct bna_ib_mod *ib_mod);
+
+/* APIs for TX, RX */
+struct bna_ib *bna_ib_get(struct bna_ib_mod *ib_mod,
+			    enum bna_intr_type intr_type, int vector);
+void bna_ib_put(struct bna_ib_mod *ib_mod, struct bna_ib *ib);
+int bna_ib_reserve_idx(struct bna_ib *ib);
+void bna_ib_release_idx(struct bna_ib *ib, int idx);
+int bna_ib_config(struct bna_ib *ib, struct bna_ib_config *ib_config);
+void bna_ib_start(struct bna_ib *ib);
+void bna_ib_stop(struct bna_ib *ib);
+void bna_ib_fail(struct bna_ib *ib);
+void bna_ib_coalescing_timeo_set(struct bna_ib *ib, u8 coalescing_timeo);
+
+/**
+ * TX MODULE AND TX
+ */
+
+/* Internal APIs */
+void bna_tx_prio_changed(struct bna_tx *tx, int prio);
+
+/* APIs for BNA */
+void bna_tx_mod_init(struct bna_tx_mod *tx_mod, struct bna *bna,
+		     struct bna_res_info *res_info);
+void bna_tx_mod_uninit(struct bna_tx_mod *tx_mod);
+int bna_tx_state_get(struct bna_tx *tx);
+
+/* APIs for PORT */
+void bna_tx_mod_start(struct bna_tx_mod *tx_mod, enum bna_tx_type type);
+void bna_tx_mod_stop(struct bna_tx_mod *tx_mod, enum bna_tx_type type);
+void bna_tx_mod_fail(struct bna_tx_mod *tx_mod);
+void bna_tx_mod_prio_changed(struct bna_tx_mod *tx_mod, int prio);
+void bna_tx_mod_cee_link_status(struct bna_tx_mod *tx_mod, int cee_link);
+
+/* APIs for BNAD */
+void bna_tx_res_req(int num_txq, int txq_depth,
+		    struct bna_res_info *res_info);
+struct bna_tx *bna_tx_create(struct bna *bna, struct bnad *bnad,
+			       struct bna_tx_config *tx_cfg,
+			       struct bna_tx_event_cbfn *tx_cbfn,
+			       struct bna_res_info *res_info, void *priv);
+void bna_tx_destroy(struct bna_tx *tx);
+void bna_tx_enable(struct bna_tx *tx);
+void bna_tx_disable(struct bna_tx *tx, enum bna_cleanup_type type,
+		    void (*cbfn)(void *, struct bna_tx *,
+				 enum bna_cb_status));
+enum bna_cb_status
+bna_tx_prio_set(struct bna_tx *tx, int prio,
+		void (*cbfn)(struct bnad *, struct bna_tx *,
+			     enum bna_cb_status));
+void bna_tx_coalescing_timeo_set(struct bna_tx *tx, int coalescing_timeo);
+
+/**
+ * RX MODULE, RX, RXF
+ */
+
+/* Internal APIs */
+void rxf_cb_cam_fltr_mbox_cmd(void *arg, int status);
+void rxf_cam_mbox_cmd(struct bna_rxf *rxf, u8 cmd,
+		const struct bna_mac *mac_addr);
+void __rxf_vlan_filter_set(struct bna_rxf *rxf, enum bna_status status);
+void bna_rxf_adv_init(struct bna_rxf *rxf,
+		struct bna_rx *rx,
+		struct bna_rx_config *q_config);
+int rxf_process_packet_filter_ucast(struct bna_rxf *rxf);
+int rxf_process_packet_filter_promisc(struct bna_rxf *rxf);
+int rxf_process_packet_filter_default(struct bna_rxf *rxf);
+int rxf_process_packet_filter_allmulti(struct bna_rxf *rxf);
+int rxf_clear_packet_filter_ucast(struct bna_rxf *rxf);
+int rxf_clear_packet_filter_promisc(struct bna_rxf *rxf);
+int rxf_clear_packet_filter_default(struct bna_rxf *rxf);
+int rxf_clear_packet_filter_allmulti(struct bna_rxf *rxf);
+void rxf_reset_packet_filter_ucast(struct bna_rxf *rxf);
+void rxf_reset_packet_filter_promisc(struct bna_rxf *rxf);
+void rxf_reset_packet_filter_default(struct bna_rxf *rxf);
+void rxf_reset_packet_filter_allmulti(struct bna_rxf *rxf);
+
+/* APIs for BNA */
+void bna_rx_mod_init(struct bna_rx_mod *rx_mod, struct bna *bna,
+		     struct bna_res_info *res_info);
+void bna_rx_mod_uninit(struct bna_rx_mod *rx_mod);
+int bna_rx_state_get(struct bna_rx *rx);
+int bna_rxf_state_get(struct bna_rxf *rxf);
+
+/* APIs for PORT */
+void bna_rx_mod_start(struct bna_rx_mod *rx_mod, enum bna_rx_type type);
+void bna_rx_mod_stop(struct bna_rx_mod *rx_mod, enum bna_rx_type type);
+void bna_rx_mod_fail(struct bna_rx_mod *rx_mod);
+
+/* APIs for BNAD */
+void bna_rx_res_req(struct bna_rx_config *rx_config,
+		    struct bna_res_info *res_info);
+struct bna_rx *bna_rx_create(struct bna *bna, struct bnad *bnad,
+			       struct bna_rx_config *rx_cfg,
+			       struct bna_rx_event_cbfn *rx_cbfn,
+			       struct bna_res_info *res_info, void *priv);
+void bna_rx_destroy(struct bna_rx *rx);
+void bna_rx_enable(struct bna_rx *rx);
+void bna_rx_disable(struct bna_rx *rx, enum bna_cleanup_type type,
+		    void (*cbfn)(void *, struct bna_rx *,
+				 enum bna_cb_status));
+void bna_rx_coalescing_timeo_set(struct bna_rx *rx, int coalescing_timeo);
+void bna_rx_dim_reconfig(struct bna *bna, u32 vector[][BNA_BIAS_T_MAX]);
+void bna_rx_dim_update(struct bna_ccb *ccb);
+enum bna_cb_status
+bna_rx_ucast_set(struct bna_rx *rx, u8 *ucmac,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status));
+enum bna_cb_status
+bna_rx_ucast_add(struct bna_rx *rx, u8* ucmac,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status));
+enum bna_cb_status
+bna_rx_ucast_del(struct bna_rx *rx, u8 *ucmac,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status));
+enum bna_cb_status
+bna_rx_mcast_add(struct bna_rx *rx, u8 *mcmac,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status));
+enum bna_cb_status
+bna_rx_mcast_del(struct bna_rx *rx, u8 *mcmac,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status));
+enum bna_cb_status
+bna_rx_mcast_listset(struct bna_rx *rx, int count, u8 *mcmac,
+		     void (*cbfn)(struct bnad *, struct bna_rx *,
+				  enum bna_cb_status));
+void bna_rx_mcast_delall(struct bna_rx *rx,
+			 void (*cbfn)(struct bnad *, struct bna_rx *,
+				      enum bna_cb_status));
+enum bna_cb_status
+bna_rx_mode_set(struct bna_rx *rx, enum bna_rxmode rxmode,
+		enum bna_rxmode bitmask,
+		void (*cbfn)(struct bnad *, struct bna_rx *,
+			     enum bna_cb_status));
+void bna_rx_vlan_add(struct bna_rx *rx, int vlan_id);
+void bna_rx_vlan_del(struct bna_rx *rx, int vlan_id);
+void bna_rx_vlanfilter_enable(struct bna_rx *rx);
+void bna_rx_vlanfilter_disable(struct bna_rx *rx);
+void bna_rx_rss_enable(struct bna_rx *rx);
+void bna_rx_rss_disable(struct bna_rx *rx);
+void bna_rx_rss_reconfig(struct bna_rx *rx, struct bna_rxf_rss *rss_config);
+void bna_rx_rss_rit_set(struct bna_rx *rx, unsigned int *vectors,
+			int nvectors);
+void bna_rx_hds_enable(struct bna_rx *rx, struct bna_rxf_hds *hds_config,
+		       void (*cbfn)(struct bnad *, struct bna_rx *,
+				    enum bna_cb_status));
+void bna_rx_hds_disable(struct bna_rx *rx,
+			void (*cbfn)(struct bnad *, struct bna_rx *,
+				     enum bna_cb_status));
+void bna_rx_receive_pause(struct bna_rx *rx,
+			  void (*cbfn)(struct bnad *, struct bna_rx *,
+				       enum bna_cb_status));
+void bna_rx_receive_resume(struct bna_rx *rx,
+			   void (*cbfn)(struct bnad *, struct bna_rx *,
+					enum bna_cb_status));
+
+/* RxF APIs for RX */
+void bna_rxf_start(struct bna_rxf *rxf);
+void bna_rxf_stop(struct bna_rxf *rxf);
+void bna_rxf_fail(struct bna_rxf *rxf);
+void bna_rxf_init(struct bna_rxf *rxf, struct bna_rx *rx,
+		  struct bna_rx_config *q_config);
+void bna_rxf_uninit(struct bna_rxf *rxf);
+
+/* Callback from RXF to RX */
+void bna_rx_cb_rxf_stopped(struct bna_rx *rx, enum bna_cb_status);
+void bna_rx_cb_rxf_started(struct bna_rx *rx, enum bna_cb_status);
+
+/**
+ * BNAD
+ */
+
+/* Callbacks for BNA */
+void bnad_cb_stats_get(struct bnad *bnad, enum bna_cb_status status,
+		       struct bna_stats *stats);
+void bnad_cb_stats_clr(struct bnad *bnad);
+
+/* Callbacks for DEVICE */
+void bnad_cb_device_enabled(struct bnad *bnad, enum bna_cb_status status);
+void bnad_cb_device_disabled(struct bnad *bnad, enum bna_cb_status status);
+void bnad_cb_device_enable_mbox_intr(struct bnad *bnad);
+void bnad_cb_device_disable_mbox_intr(struct bnad *bnad);
+
+/* Callbacks for port */
+void bnad_cb_port_link_status(struct bnad *bnad,
+			      enum bna_link_status status);
+
+#endif  /* __BNA_H__ */
diff --git a/drivers/net/bna/bna_ctrl.c b/drivers/net/bna/bna_ctrl.c
new file mode 100644
index 000000000000..9d41ebf41cf4
--- /dev/null
+++ b/drivers/net/bna/bna_ctrl.c
@@ -0,0 +1,3626 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#include "bna.h"
+#include "bfa_sm.h"
+#include "bfa_wc.h"
+
+/**
+ * MBOX
+ */
+static int
+bna_is_aen(u8 msg_id)
+{
+	return (msg_id == BFI_LL_I2H_LINK_DOWN_AEN ||
+		msg_id == BFI_LL_I2H_LINK_UP_AEN);
+}
+
+static void
+bna_mbox_aen_callback(struct bna *bna, struct bfi_mbmsg *msg)
+{
+	struct bfi_ll_aen *aen = (struct bfi_ll_aen *)(msg);
+
+	switch (aen->mh.msg_id) {
+	case BFI_LL_I2H_LINK_UP_AEN:
+		bna_port_cb_link_up(&bna->port, aen, aen->reason);
+		break;
+	case BFI_LL_I2H_LINK_DOWN_AEN:
+		bna_port_cb_link_down(&bna->port, aen->reason);
+		break;
+	default:
+		break;
+	}
+}
+
+static void
+bna_ll_isr(void *llarg, struct bfi_mbmsg *msg)
+{
+	struct bna *bna = (struct bna *)(llarg);
+	struct bfi_ll_rsp *mb_rsp = (struct bfi_ll_rsp *)(msg);
+	struct bfi_mhdr *cmd_h, *rsp_h;
+	struct bna_mbox_qe *mb_qe = NULL;
+	int to_post = 0;
+	u8 aen = 0;
+	char message[BNA_MESSAGE_SIZE];
+
+	aen = bna_is_aen(mb_rsp->mh.msg_id);
+
+	if (!aen) {
+		mb_qe = bfa_q_first(&bna->mbox_mod.posted_q);
+		cmd_h = (struct bfi_mhdr *)(&mb_qe->cmd.msg[0]);
+		rsp_h = (struct bfi_mhdr *)(&mb_rsp->mh);
+
+		if ((BFA_I2HM(cmd_h->msg_id) == rsp_h->msg_id) &&
+		    (cmd_h->mtag.i2htok == rsp_h->mtag.i2htok)) {
+			/* Remove the request from posted_q, update state  */
+			list_del(&mb_qe->qe);
+			bna->mbox_mod.msg_pending--;
+			if (list_empty(&bna->mbox_mod.posted_q))
+				bna->mbox_mod.state = BNA_MBOX_FREE;
+			else
+				to_post = 1;
+
+			/* Dispatch the cbfn */
+			if (mb_qe->cbfn)
+				mb_qe->cbfn(mb_qe->cbarg, mb_rsp->error);
+
+			/* Post the next entry, if needed */
+			if (to_post) {
+				mb_qe = bfa_q_first(&bna->mbox_mod.posted_q);
+				bfa_ioc_mbox_queue(&bna->device.ioc,
+							&mb_qe->cmd);
+			}
+		} else {
+			snprintf(message, BNA_MESSAGE_SIZE,
+				       "No matching rsp for [%d:%d:%d]\n",
+				       mb_rsp->mh.msg_class, mb_rsp->mh.msg_id,
+				       mb_rsp->mh.mtag.i2htok);
+		pr_info("%s", message);
+		}
+
+	} else
+		bna_mbox_aen_callback(bna, msg);
+}
+
+void
+bna_err_handler(struct bna *bna, u32 intr_status)
+{
+	u32 init_halt;
+
+	if (intr_status & __HALT_STATUS_BITS) {
+		init_halt = readl(bna->device.ioc.ioc_regs.ll_halt);
+		init_halt &= ~__FW_INIT_HALT_P;
+		writel(init_halt, bna->device.ioc.ioc_regs.ll_halt);
+	}
+
+	bfa_ioc_error_isr(&bna->device.ioc);
+}
+
+void
+bna_mbox_handler(struct bna *bna, u32 intr_status)
+{
+	if (BNA_IS_ERR_INTR(intr_status)) {
+		bna_err_handler(bna, intr_status);
+		return;
+	}
+	if (BNA_IS_MBOX_INTR(intr_status))
+		bfa_ioc_mbox_isr(&bna->device.ioc);
+}
+
+void
+bna_mbox_send(struct bna *bna, struct bna_mbox_qe *mbox_qe)
+{
+	struct bfi_mhdr *mh;
+
+	mh = (struct bfi_mhdr *)(&mbox_qe->cmd.msg[0]);
+
+	mh->mtag.i2htok = htons(bna->mbox_mod.msg_ctr);
+	bna->mbox_mod.msg_ctr++;
+	bna->mbox_mod.msg_pending++;
+	if (bna->mbox_mod.state == BNA_MBOX_FREE) {
+		list_add_tail(&mbox_qe->qe, &bna->mbox_mod.posted_q);
+		bfa_ioc_mbox_queue(&bna->device.ioc, &mbox_qe->cmd);
+		bna->mbox_mod.state = BNA_MBOX_POSTED;
+	} else {
+		list_add_tail(&mbox_qe->qe, &bna->mbox_mod.posted_q);
+	}
+}
+
+void
+bna_mbox_flush_q(struct bna *bna, struct list_head *q)
+{
+	struct bna_mbox_qe *mb_qe = NULL;
+	struct bfi_mhdr *cmd_h;
+	struct list_head			*mb_q;
+	void 			(*cbfn)(void *arg, int status);
+	void 			*cbarg;
+
+	mb_q = &bna->mbox_mod.posted_q;
+
+	while (!list_empty(mb_q)) {
+		bfa_q_deq(mb_q, &mb_qe);
+		cbfn = mb_qe->cbfn;
+		cbarg = mb_qe->cbarg;
+		bfa_q_qe_init(mb_qe);
+		bna->mbox_mod.msg_pending--;
+
+		cmd_h = (struct bfi_mhdr *)(&mb_qe->cmd.msg[0]);
+		if (cbfn)
+			cbfn(cbarg, BNA_CB_NOT_EXEC);
+	}
+
+	bna->mbox_mod.state = BNA_MBOX_FREE;
+}
+
+void
+bna_mbox_mod_start(struct bna_mbox_mod *mbox_mod)
+{
+}
+
+void
+bna_mbox_mod_stop(struct bna_mbox_mod *mbox_mod)
+{
+	bna_mbox_flush_q(mbox_mod->bna, &mbox_mod->posted_q);
+}
+
+void
+bna_mbox_mod_init(struct bna_mbox_mod *mbox_mod, struct bna *bna)
+{
+	bfa_ioc_mbox_regisr(&bna->device.ioc, BFI_MC_LL, bna_ll_isr, bna);
+	mbox_mod->state = BNA_MBOX_FREE;
+	mbox_mod->msg_ctr = mbox_mod->msg_pending = 0;
+	INIT_LIST_HEAD(&mbox_mod->posted_q);
+	mbox_mod->bna = bna;
+}
+
+void
+bna_mbox_mod_uninit(struct bna_mbox_mod *mbox_mod)
+{
+	mbox_mod->bna = NULL;
+}
+
+/**
+ * LLPORT
+ */
+#define call_llport_stop_cbfn(llport, status)\
+do {\
+	if ((llport)->stop_cbfn)\
+		(llport)->stop_cbfn(&(llport)->bna->port, status);\
+	(llport)->stop_cbfn = NULL;\
+} while (0)
+
+static void bna_fw_llport_up(struct bna_llport *llport);
+static void bna_fw_cb_llport_up(void *arg, int status);
+static void bna_fw_llport_down(struct bna_llport *llport);
+static void bna_fw_cb_llport_down(void *arg, int status);
+static void bna_llport_start(struct bna_llport *llport);
+static void bna_llport_stop(struct bna_llport *llport);
+static void bna_llport_fail(struct bna_llport *llport);
+
+enum bna_llport_event {
+	LLPORT_E_START			= 1,
+	LLPORT_E_STOP			= 2,
+	LLPORT_E_FAIL			= 3,
+	LLPORT_E_UP			= 4,
+	LLPORT_E_DOWN			= 5,
+	LLPORT_E_FWRESP_UP		= 6,
+	LLPORT_E_FWRESP_DOWN		= 7
+};
+
+enum bna_llport_state {
+	BNA_LLPORT_STOPPED		= 1,
+	BNA_LLPORT_DOWN			= 2,
+	BNA_LLPORT_UP_RESP_WAIT		= 3,
+	BNA_LLPORT_DOWN_RESP_WAIT	= 4,
+	BNA_LLPORT_UP			= 5,
+	BNA_LLPORT_LAST_RESP_WAIT 	= 6
+};
+
+bfa_fsm_state_decl(bna_llport, stopped, struct bna_llport,
+			enum bna_llport_event);
+bfa_fsm_state_decl(bna_llport, down, struct bna_llport,
+			enum bna_llport_event);
+bfa_fsm_state_decl(bna_llport, up_resp_wait, struct bna_llport,
+			enum bna_llport_event);
+bfa_fsm_state_decl(bna_llport, down_resp_wait, struct bna_llport,
+			enum bna_llport_event);
+bfa_fsm_state_decl(bna_llport, up, struct bna_llport,
+			enum bna_llport_event);
+bfa_fsm_state_decl(bna_llport, last_resp_wait, struct bna_llport,
+			enum bna_llport_event);
+
+static struct bfa_sm_table llport_sm_table[] = {
+	{BFA_SM(bna_llport_sm_stopped), BNA_LLPORT_STOPPED},
+	{BFA_SM(bna_llport_sm_down), BNA_LLPORT_DOWN},
+	{BFA_SM(bna_llport_sm_up_resp_wait), BNA_LLPORT_UP_RESP_WAIT},
+	{BFA_SM(bna_llport_sm_down_resp_wait), BNA_LLPORT_DOWN_RESP_WAIT},
+	{BFA_SM(bna_llport_sm_up), BNA_LLPORT_UP},
+	{BFA_SM(bna_llport_sm_last_resp_wait), BNA_LLPORT_LAST_RESP_WAIT}
+};
+
+static void
+bna_llport_sm_stopped_entry(struct bna_llport *llport)
+{
+	llport->bna->port.link_cbfn((llport)->bna->bnad, BNA_LINK_DOWN);
+	call_llport_stop_cbfn(llport, BNA_CB_SUCCESS);
+}
+
+static void
+bna_llport_sm_stopped(struct bna_llport *llport,
+			enum bna_llport_event event)
+{
+	switch (event) {
+	case LLPORT_E_START:
+		bfa_fsm_set_state(llport, bna_llport_sm_down);
+		break;
+
+	case LLPORT_E_STOP:
+		call_llport_stop_cbfn(llport, BNA_CB_SUCCESS);
+		break;
+
+	case LLPORT_E_FAIL:
+		break;
+
+	case LLPORT_E_DOWN:
+		/* This event is received due to Rx objects failing */
+		/* No-op */
+		break;
+
+	case LLPORT_E_FWRESP_UP:
+	case LLPORT_E_FWRESP_DOWN:
+		/**
+		 * These events are received due to flushing of mbox when
+		 * device fails
+		 */
+		/* No-op */
+		break;
+
+	default:
+		bfa_sm_fault(llport->bna, event);
+	}
+}
+
+static void
+bna_llport_sm_down_entry(struct bna_llport *llport)
+{
+	bnad_cb_port_link_status((llport)->bna->bnad, BNA_LINK_DOWN);
+}
+
+static void
+bna_llport_sm_down(struct bna_llport *llport,
+			enum bna_llport_event event)
+{
+	switch (event) {
+	case LLPORT_E_STOP:
+		bfa_fsm_set_state(llport, bna_llport_sm_stopped);
+		break;
+
+	case LLPORT_E_FAIL:
+		bfa_fsm_set_state(llport, bna_llport_sm_stopped);
+		break;
+
+	case LLPORT_E_UP:
+		bfa_fsm_set_state(llport, bna_llport_sm_up_resp_wait);
+		bna_fw_llport_up(llport);
+		break;
+
+	default:
+		bfa_sm_fault(llport->bna, event);
+	}
+}
+
+static void
+bna_llport_sm_up_resp_wait_entry(struct bna_llport *llport)
+{
+	/**
+	 * NOTE: Do not call bna_fw_llport_up() here. That will over step
+	 * mbox due to down_resp_wait -> up_resp_wait transition on event
+	 * LLPORT_E_UP
+	 */
+}
+
+static void
+bna_llport_sm_up_resp_wait(struct bna_llport *llport,
+			enum bna_llport_event event)
+{
+	switch (event) {
+	case LLPORT_E_STOP:
+		bfa_fsm_set_state(llport, bna_llport_sm_last_resp_wait);
+		break;
+
+	case LLPORT_E_FAIL:
+		bfa_fsm_set_state(llport, bna_llport_sm_stopped);
+		break;
+
+	case LLPORT_E_DOWN:
+		bfa_fsm_set_state(llport, bna_llport_sm_down_resp_wait);
+		break;
+
+	case LLPORT_E_FWRESP_UP:
+		bfa_fsm_set_state(llport, bna_llport_sm_up);
+		break;
+
+	case LLPORT_E_FWRESP_DOWN:
+		/* down_resp_wait -> up_resp_wait transition on LLPORT_E_UP */
+		bna_fw_llport_up(llport);
+		break;
+
+	default:
+		bfa_sm_fault(llport->bna, event);
+	}
+}
+
+static void
+bna_llport_sm_down_resp_wait_entry(struct bna_llport *llport)
+{
+	/**
+	 * NOTE: Do not call bna_fw_llport_down() here. That will over step
+	 * mbox due to up_resp_wait -> down_resp_wait transition on event
+	 * LLPORT_E_DOWN
+	 */
+}
+
+static void
+bna_llport_sm_down_resp_wait(struct bna_llport *llport,
+			enum bna_llport_event event)
+{
+	switch (event) {
+	case LLPORT_E_STOP:
+		bfa_fsm_set_state(llport, bna_llport_sm_last_resp_wait);
+		break;
+
+	case LLPORT_E_FAIL:
+		bfa_fsm_set_state(llport, bna_llport_sm_stopped);
+		break;
+
+	case LLPORT_E_UP:
+		bfa_fsm_set_state(llport, bna_llport_sm_up_resp_wait);
+		break;
+
+	case LLPORT_E_FWRESP_UP:
+		/* up_resp_wait->down_resp_wait transition on LLPORT_E_DOWN */
+		bna_fw_llport_down(llport);
+		break;
+
+	case LLPORT_E_FWRESP_DOWN:
+		bfa_fsm_set_state(llport, bna_llport_sm_down);
+		break;
+
+	default:
+		bfa_sm_fault(llport->bna, event);
+	}
+}
+
+static void
+bna_llport_sm_up_entry(struct bna_llport *llport)
+{
+}
+
+static void
+bna_llport_sm_up(struct bna_llport *llport,
+			enum bna_llport_event event)
+{
+	switch (event) {
+	case LLPORT_E_STOP:
+		bfa_fsm_set_state(llport, bna_llport_sm_last_resp_wait);
+		bna_fw_llport_down(llport);
+		break;
+
+	case LLPORT_E_FAIL:
+		bfa_fsm_set_state(llport, bna_llport_sm_stopped);
+		break;
+
+	case LLPORT_E_DOWN:
+		bfa_fsm_set_state(llport, bna_llport_sm_down_resp_wait);
+		bna_fw_llport_down(llport);
+		break;
+
+	default:
+		bfa_sm_fault(llport->bna, event);
+	}
+}
+
+static void
+bna_llport_sm_last_resp_wait_entry(struct bna_llport *llport)
+{
+}
+
+static void
+bna_llport_sm_last_resp_wait(struct bna_llport *llport,
+			enum bna_llport_event event)
+{
+	switch (event) {
+	case LLPORT_E_FAIL:
+		bfa_fsm_set_state(llport, bna_llport_sm_stopped);
+		break;
+
+	case LLPORT_E_DOWN:
+		/**
+		 * This event is received due to Rx objects stopping in
+		 * parallel to llport
+		 */
+		/* No-op */
+		break;
+
+	case LLPORT_E_FWRESP_UP:
+		/* up_resp_wait->last_resp_wait transition on LLPORT_T_STOP */
+		bna_fw_llport_down(llport);
+		break;
+
+	case LLPORT_E_FWRESP_DOWN:
+		bfa_fsm_set_state(llport, bna_llport_sm_stopped);
+		break;
+
+	default:
+		bfa_sm_fault(llport->bna, event);
+	}
+}
+
+static void
+bna_fw_llport_admin_up(struct bna_llport *llport)
+{
+	struct bfi_ll_port_admin_req ll_req;
+
+	memset(&ll_req, 0, sizeof(ll_req));
+	ll_req.mh.msg_class = BFI_MC_LL;
+	ll_req.mh.msg_id = BFI_LL_H2I_PORT_ADMIN_REQ;
+	ll_req.mh.mtag.h2i.lpu_id = 0;
+
+	ll_req.up = BNA_STATUS_T_ENABLED;
+
+	bna_mbox_qe_fill(&llport->mbox_qe, &ll_req, sizeof(ll_req),
+			bna_fw_cb_llport_up, llport);
+
+	bna_mbox_send(llport->bna, &llport->mbox_qe);
+}
+
+static void
+bna_fw_llport_up(struct bna_llport *llport)
+{
+	if (llport->type == BNA_PORT_T_REGULAR)
+		bna_fw_llport_admin_up(llport);
+}
+
+static void
+bna_fw_cb_llport_up(void *arg, int status)
+{
+	struct bna_llport *llport = (struct bna_llport *)arg;
+
+	bfa_q_qe_init(&llport->mbox_qe.qe);
+	bfa_fsm_send_event(llport, LLPORT_E_FWRESP_UP);
+}
+
+static void
+bna_fw_llport_admin_down(struct bna_llport *llport)
+{
+	struct bfi_ll_port_admin_req ll_req;
+
+	memset(&ll_req, 0, sizeof(ll_req));
+	ll_req.mh.msg_class = BFI_MC_LL;
+	ll_req.mh.msg_id = BFI_LL_H2I_PORT_ADMIN_REQ;
+	ll_req.mh.mtag.h2i.lpu_id = 0;
+
+	ll_req.up = BNA_STATUS_T_DISABLED;
+
+	bna_mbox_qe_fill(&llport->mbox_qe, &ll_req, sizeof(ll_req),
+			bna_fw_cb_llport_down, llport);
+
+	bna_mbox_send(llport->bna, &llport->mbox_qe);
+}
+
+static void
+bna_fw_llport_down(struct bna_llport *llport)
+{
+	if (llport->type == BNA_PORT_T_REGULAR)
+		bna_fw_llport_admin_down(llport);
+}
+
+static void
+bna_fw_cb_llport_down(void *arg, int status)
+{
+	struct bna_llport *llport = (struct bna_llport *)arg;
+
+	bfa_q_qe_init(&llport->mbox_qe.qe);
+	bfa_fsm_send_event(llport, LLPORT_E_FWRESP_DOWN);
+}
+
+void
+bna_port_cb_llport_stopped(struct bna_port *port,
+				enum bna_cb_status status)
+{
+	bfa_wc_down(&port->chld_stop_wc);
+}
+
+static void
+bna_llport_init(struct bna_llport *llport, struct bna *bna)
+{
+	llport->flags |= BNA_LLPORT_F_ENABLED;
+	llport->type = BNA_PORT_T_REGULAR;
+	llport->bna = bna;
+
+	llport->link_status = BNA_LINK_DOWN;
+
+	llport->admin_up_count = 0;
+
+	llport->stop_cbfn = NULL;
+
+	bfa_q_qe_init(&llport->mbox_qe.qe);
+
+	bfa_fsm_set_state(llport, bna_llport_sm_stopped);
+}
+
+static void
+bna_llport_uninit(struct bna_llport *llport)
+{
+	llport->flags &= ~BNA_LLPORT_F_ENABLED;
+
+	llport->bna = NULL;
+}
+
+static void
+bna_llport_start(struct bna_llport *llport)
+{
+	bfa_fsm_send_event(llport, LLPORT_E_START);
+}
+
+static void
+bna_llport_stop(struct bna_llport *llport)
+{
+	llport->stop_cbfn = bna_port_cb_llport_stopped;
+
+	bfa_fsm_send_event(llport, LLPORT_E_STOP);
+}
+
+static void
+bna_llport_fail(struct bna_llport *llport)
+{
+	bfa_fsm_send_event(llport, LLPORT_E_FAIL);
+}
+
+int
+bna_llport_state_get(struct bna_llport *llport)
+{
+	return bfa_sm_to_state(llport_sm_table, llport->fsm);
+}
+
+void
+bna_llport_admin_up(struct bna_llport *llport)
+{
+	llport->admin_up_count++;
+
+	if (llport->admin_up_count == 1) {
+		llport->flags |= BNA_LLPORT_F_RX_ENABLED;
+		if (llport->flags & BNA_LLPORT_F_ENABLED)
+			bfa_fsm_send_event(llport, LLPORT_E_UP);
+	}
+}
+
+void
+bna_llport_admin_down(struct bna_llport *llport)
+{
+	llport->admin_up_count--;
+
+	if (llport->admin_up_count == 0) {
+		llport->flags &= ~BNA_LLPORT_F_RX_ENABLED;
+		if (llport->flags & BNA_LLPORT_F_ENABLED)
+			bfa_fsm_send_event(llport, LLPORT_E_DOWN);
+	}
+}
+
+/**
+ * PORT
+ */
+#define bna_port_chld_start(port)\
+do {\
+	enum bna_tx_type tx_type = ((port)->type == BNA_PORT_T_REGULAR) ?\
+					BNA_TX_T_REGULAR : BNA_TX_T_LOOPBACK;\
+	enum bna_rx_type rx_type = ((port)->type == BNA_PORT_T_REGULAR) ?\
+					BNA_RX_T_REGULAR : BNA_RX_T_LOOPBACK;\
+	bna_llport_start(&(port)->llport);\
+	bna_tx_mod_start(&(port)->bna->tx_mod, tx_type);\
+	bna_rx_mod_start(&(port)->bna->rx_mod, rx_type);\
+} while (0)
+
+#define bna_port_chld_stop(port)\
+do {\
+	enum bna_tx_type tx_type = ((port)->type == BNA_PORT_T_REGULAR) ?\
+					BNA_TX_T_REGULAR : BNA_TX_T_LOOPBACK;\
+	enum bna_rx_type rx_type = ((port)->type == BNA_PORT_T_REGULAR) ?\
+					BNA_RX_T_REGULAR : BNA_RX_T_LOOPBACK;\
+	bfa_wc_up(&(port)->chld_stop_wc);\
+	bfa_wc_up(&(port)->chld_stop_wc);\
+	bfa_wc_up(&(port)->chld_stop_wc);\
+	bna_llport_stop(&(port)->llport);\
+	bna_tx_mod_stop(&(port)->bna->tx_mod, tx_type);\
+	bna_rx_mod_stop(&(port)->bna->rx_mod, rx_type);\
+} while (0)
+
+#define bna_port_chld_fail(port)\
+do {\
+	bna_llport_fail(&(port)->llport);\
+	bna_tx_mod_fail(&(port)->bna->tx_mod);\
+	bna_rx_mod_fail(&(port)->bna->rx_mod);\
+} while (0)
+
+#define bna_port_rx_start(port)\
+do {\
+	enum bna_rx_type rx_type = ((port)->type == BNA_PORT_T_REGULAR) ?\
+					BNA_RX_T_REGULAR : BNA_RX_T_LOOPBACK;\
+	bna_rx_mod_start(&(port)->bna->rx_mod, rx_type);\
+} while (0)
+
+#define bna_port_rx_stop(port)\
+do {\
+	enum bna_rx_type rx_type = ((port)->type == BNA_PORT_T_REGULAR) ?\
+					BNA_RX_T_REGULAR : BNA_RX_T_LOOPBACK;\
+	bfa_wc_up(&(port)->chld_stop_wc);\
+	bna_rx_mod_stop(&(port)->bna->rx_mod, rx_type);\
+} while (0)
+
+#define call_port_stop_cbfn(port, status)\
+do {\
+	if ((port)->stop_cbfn)\
+		(port)->stop_cbfn((port)->stop_cbarg, status);\
+	(port)->stop_cbfn = NULL;\
+	(port)->stop_cbarg = NULL;\
+} while (0)
+
+#define call_port_pause_cbfn(port, status)\
+do {\
+	if ((port)->pause_cbfn)\
+		(port)->pause_cbfn((port)->bna->bnad, status);\
+	(port)->pause_cbfn = NULL;\
+} while (0)
+
+#define call_port_mtu_cbfn(port, status)\
+do {\
+	if ((port)->mtu_cbfn)\
+		(port)->mtu_cbfn((port)->bna->bnad, status);\
+	(port)->mtu_cbfn = NULL;\
+} while (0)
+
+static void bna_fw_pause_set(struct bna_port *port);
+static void bna_fw_cb_pause_set(void *arg, int status);
+static void bna_fw_mtu_set(struct bna_port *port);
+static void bna_fw_cb_mtu_set(void *arg, int status);
+
+enum bna_port_event {
+	PORT_E_START			= 1,
+	PORT_E_STOP			= 2,
+	PORT_E_FAIL			= 3,
+	PORT_E_PAUSE_CFG		= 4,
+	PORT_E_MTU_CFG			= 5,
+	PORT_E_CHLD_STOPPED		= 6,
+	PORT_E_FWRESP_PAUSE		= 7,
+	PORT_E_FWRESP_MTU		= 8
+};
+
+enum bna_port_state {
+	BNA_PORT_STOPPED		= 1,
+	BNA_PORT_MTU_INIT_WAIT		= 2,
+	BNA_PORT_PAUSE_INIT_WAIT	= 3,
+	BNA_PORT_LAST_RESP_WAIT		= 4,
+	BNA_PORT_STARTED		= 5,
+	BNA_PORT_PAUSE_CFG_WAIT		= 6,
+	BNA_PORT_RX_STOP_WAIT		= 7,
+	BNA_PORT_MTU_CFG_WAIT 		= 8,
+	BNA_PORT_CHLD_STOP_WAIT		= 9
+};
+
+bfa_fsm_state_decl(bna_port, stopped, struct bna_port,
+			enum bna_port_event);
+bfa_fsm_state_decl(bna_port, mtu_init_wait, struct bna_port,
+			enum bna_port_event);
+bfa_fsm_state_decl(bna_port, pause_init_wait, struct bna_port,
+			enum bna_port_event);
+bfa_fsm_state_decl(bna_port, last_resp_wait, struct bna_port,
+			enum bna_port_event);
+bfa_fsm_state_decl(bna_port, started, struct bna_port,
+			enum bna_port_event);
+bfa_fsm_state_decl(bna_port, pause_cfg_wait, struct bna_port,
+			enum bna_port_event);
+bfa_fsm_state_decl(bna_port, rx_stop_wait, struct bna_port,
+			enum bna_port_event);
+bfa_fsm_state_decl(bna_port, mtu_cfg_wait, struct bna_port,
+			enum bna_port_event);
+bfa_fsm_state_decl(bna_port, chld_stop_wait, struct bna_port,
+			enum bna_port_event);
+
+static struct bfa_sm_table port_sm_table[] = {
+	{BFA_SM(bna_port_sm_stopped), BNA_PORT_STOPPED},
+	{BFA_SM(bna_port_sm_mtu_init_wait), BNA_PORT_MTU_INIT_WAIT},
+	{BFA_SM(bna_port_sm_pause_init_wait), BNA_PORT_PAUSE_INIT_WAIT},
+	{BFA_SM(bna_port_sm_last_resp_wait), BNA_PORT_LAST_RESP_WAIT},
+	{BFA_SM(bna_port_sm_started), BNA_PORT_STARTED},
+	{BFA_SM(bna_port_sm_pause_cfg_wait), BNA_PORT_PAUSE_CFG_WAIT},
+	{BFA_SM(bna_port_sm_rx_stop_wait), BNA_PORT_RX_STOP_WAIT},
+	{BFA_SM(bna_port_sm_mtu_cfg_wait), BNA_PORT_MTU_CFG_WAIT},
+	{BFA_SM(bna_port_sm_chld_stop_wait), BNA_PORT_CHLD_STOP_WAIT}
+};
+
+static void
+bna_port_sm_stopped_entry(struct bna_port *port)
+{
+	call_port_pause_cbfn(port, BNA_CB_SUCCESS);
+	call_port_mtu_cbfn(port, BNA_CB_SUCCESS);
+	call_port_stop_cbfn(port, BNA_CB_SUCCESS);
+}
+
+static void
+bna_port_sm_stopped(struct bna_port *port, enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_START:
+		bfa_fsm_set_state(port, bna_port_sm_mtu_init_wait);
+		break;
+
+	case PORT_E_STOP:
+		call_port_stop_cbfn(port, BNA_CB_SUCCESS);
+		break;
+
+	case PORT_E_FAIL:
+		/* No-op */
+		break;
+
+	case PORT_E_PAUSE_CFG:
+		call_port_pause_cbfn(port, BNA_CB_SUCCESS);
+		break;
+
+	case PORT_E_MTU_CFG:
+		call_port_mtu_cbfn(port, BNA_CB_SUCCESS);
+		break;
+
+	case PORT_E_CHLD_STOPPED:
+		/**
+		 * This event is received due to LLPort, Tx and Rx objects
+		 * failing
+		 */
+		/* No-op */
+		break;
+
+	case PORT_E_FWRESP_PAUSE:
+	case PORT_E_FWRESP_MTU:
+		/**
+		 * These events are received due to flushing of mbox when
+		 * device fails
+		 */
+		/* No-op */
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_port_sm_mtu_init_wait_entry(struct bna_port *port)
+{
+	bna_fw_mtu_set(port);
+}
+
+static void
+bna_port_sm_mtu_init_wait(struct bna_port *port, enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_STOP:
+		bfa_fsm_set_state(port, bna_port_sm_last_resp_wait);
+		break;
+
+	case PORT_E_FAIL:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		break;
+
+	case PORT_E_PAUSE_CFG:
+		/* No-op */
+		break;
+
+	case PORT_E_MTU_CFG:
+		port->flags |= BNA_PORT_F_MTU_CHANGED;
+		break;
+
+	case PORT_E_FWRESP_MTU:
+		if (port->flags & BNA_PORT_F_MTU_CHANGED) {
+			port->flags &= ~BNA_PORT_F_MTU_CHANGED;
+			bna_fw_mtu_set(port);
+		} else {
+			bfa_fsm_set_state(port, bna_port_sm_pause_init_wait);
+		}
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_port_sm_pause_init_wait_entry(struct bna_port *port)
+{
+	bna_fw_pause_set(port);
+}
+
+static void
+bna_port_sm_pause_init_wait(struct bna_port *port,
+				enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_STOP:
+		bfa_fsm_set_state(port, bna_port_sm_last_resp_wait);
+		break;
+
+	case PORT_E_FAIL:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		break;
+
+	case PORT_E_PAUSE_CFG:
+		port->flags |= BNA_PORT_F_PAUSE_CHANGED;
+		break;
+
+	case PORT_E_MTU_CFG:
+		port->flags |= BNA_PORT_F_MTU_CHANGED;
+		break;
+
+	case PORT_E_FWRESP_PAUSE:
+		if (port->flags & BNA_PORT_F_PAUSE_CHANGED) {
+			port->flags &= ~BNA_PORT_F_PAUSE_CHANGED;
+			bna_fw_pause_set(port);
+		} else if (port->flags & BNA_PORT_F_MTU_CHANGED) {
+			port->flags &= ~BNA_PORT_F_MTU_CHANGED;
+			bfa_fsm_set_state(port, bna_port_sm_mtu_init_wait);
+		} else {
+			bfa_fsm_set_state(port, bna_port_sm_started);
+			bna_port_chld_start(port);
+		}
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_port_sm_last_resp_wait_entry(struct bna_port *port)
+{
+}
+
+static void
+bna_port_sm_last_resp_wait(struct bna_port *port,
+				enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_FAIL:
+	case PORT_E_FWRESP_PAUSE:
+	case PORT_E_FWRESP_MTU:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_port_sm_started_entry(struct bna_port *port)
+{
+	/**
+	 * NOTE: Do not call bna_port_chld_start() here, since it will be
+	 * inadvertently called during pause_cfg_wait->started transition
+	 * as well
+	 */
+	call_port_pause_cbfn(port, BNA_CB_SUCCESS);
+	call_port_mtu_cbfn(port, BNA_CB_SUCCESS);
+}
+
+static void
+bna_port_sm_started(struct bna_port *port,
+			enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_STOP:
+		bfa_fsm_set_state(port, bna_port_sm_chld_stop_wait);
+		break;
+
+	case PORT_E_FAIL:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		bna_port_chld_fail(port);
+		break;
+
+	case PORT_E_PAUSE_CFG:
+		bfa_fsm_set_state(port, bna_port_sm_pause_cfg_wait);
+		break;
+
+	case PORT_E_MTU_CFG:
+		bfa_fsm_set_state(port, bna_port_sm_rx_stop_wait);
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_port_sm_pause_cfg_wait_entry(struct bna_port *port)
+{
+	bna_fw_pause_set(port);
+}
+
+static void
+bna_port_sm_pause_cfg_wait(struct bna_port *port,
+				enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_FAIL:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		bna_port_chld_fail(port);
+		break;
+
+	case PORT_E_FWRESP_PAUSE:
+		bfa_fsm_set_state(port, bna_port_sm_started);
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_port_sm_rx_stop_wait_entry(struct bna_port *port)
+{
+	bna_port_rx_stop(port);
+}
+
+static void
+bna_port_sm_rx_stop_wait(struct bna_port *port,
+				enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_FAIL:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		bna_port_chld_fail(port);
+		break;
+
+	case PORT_E_CHLD_STOPPED:
+		bfa_fsm_set_state(port, bna_port_sm_mtu_cfg_wait);
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_port_sm_mtu_cfg_wait_entry(struct bna_port *port)
+{
+	bna_fw_mtu_set(port);
+}
+
+static void
+bna_port_sm_mtu_cfg_wait(struct bna_port *port, enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_FAIL:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		bna_port_chld_fail(port);
+		break;
+
+	case PORT_E_FWRESP_MTU:
+		bfa_fsm_set_state(port, bna_port_sm_started);
+		bna_port_rx_start(port);
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_port_sm_chld_stop_wait_entry(struct bna_port *port)
+{
+	bna_port_chld_stop(port);
+}
+
+static void
+bna_port_sm_chld_stop_wait(struct bna_port *port,
+				enum bna_port_event event)
+{
+	switch (event) {
+	case PORT_E_FAIL:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		bna_port_chld_fail(port);
+		break;
+
+	case PORT_E_CHLD_STOPPED:
+		bfa_fsm_set_state(port, bna_port_sm_stopped);
+		break;
+
+	default:
+		bfa_sm_fault(port->bna, event);
+	}
+}
+
+static void
+bna_fw_pause_set(struct bna_port *port)
+{
+	struct bfi_ll_set_pause_req ll_req;
+
+	memset(&ll_req, 0, sizeof(ll_req));
+	ll_req.mh.msg_class = BFI_MC_LL;
+	ll_req.mh.msg_id = BFI_LL_H2I_SET_PAUSE_REQ;
+	ll_req.mh.mtag.h2i.lpu_id = 0;
+
+	ll_req.tx_pause = port->pause_config.tx_pause;
+	ll_req.rx_pause = port->pause_config.rx_pause;
+
+	bna_mbox_qe_fill(&port->mbox_qe, &ll_req, sizeof(ll_req),
+			bna_fw_cb_pause_set, port);
+
+	bna_mbox_send(port->bna, &port->mbox_qe);
+}
+
+static void
+bna_fw_cb_pause_set(void *arg, int status)
+{
+	struct bna_port *port = (struct bna_port *)arg;
+
+	bfa_q_qe_init(&port->mbox_qe.qe);
+	bfa_fsm_send_event(port, PORT_E_FWRESP_PAUSE);
+}
+
+void
+bna_fw_mtu_set(struct bna_port *port)
+{
+	struct bfi_ll_mtu_info_req ll_req;
+
+	bfi_h2i_set(ll_req.mh, BFI_MC_LL, BFI_LL_H2I_MTU_INFO_REQ, 0);
+	ll_req.mtu = htons((u16)port->mtu);
+
+	bna_mbox_qe_fill(&port->mbox_qe, &ll_req, sizeof(ll_req),
+				bna_fw_cb_mtu_set, port);
+	bna_mbox_send(port->bna, &port->mbox_qe);
+}
+
+void
+bna_fw_cb_mtu_set(void *arg, int status)
+{
+	struct bna_port *port = (struct bna_port *)arg;
+
+	bfa_q_qe_init(&port->mbox_qe.qe);
+	bfa_fsm_send_event(port, PORT_E_FWRESP_MTU);
+}
+
+static void
+bna_port_cb_chld_stopped(void *arg)
+{
+	struct bna_port *port = (struct bna_port *)arg;
+
+	bfa_fsm_send_event(port, PORT_E_CHLD_STOPPED);
+}
+
+void
+bna_port_init(struct bna_port *port, struct bna *bna)
+{
+	port->bna = bna;
+	port->flags = 0;
+	port->mtu = 0;
+	port->type = BNA_PORT_T_REGULAR;
+
+	port->link_cbfn = bnad_cb_port_link_status;
+
+	port->chld_stop_wc.wc_resume = bna_port_cb_chld_stopped;
+	port->chld_stop_wc.wc_cbarg = port;
+	port->chld_stop_wc.wc_count = 0;
+
+	port->stop_cbfn = NULL;
+	port->stop_cbarg = NULL;
+
+	port->pause_cbfn = NULL;
+
+	port->mtu_cbfn = NULL;
+
+	bfa_q_qe_init(&port->mbox_qe.qe);
+
+	bfa_fsm_set_state(port, bna_port_sm_stopped);
+
+	bna_llport_init(&port->llport, bna);
+}
+
+void
+bna_port_uninit(struct bna_port *port)
+{
+	bna_llport_uninit(&port->llport);
+
+	port->flags = 0;
+
+	port->bna = NULL;
+}
+
+int
+bna_port_state_get(struct bna_port *port)
+{
+	return bfa_sm_to_state(port_sm_table, port->fsm);
+}
+
+void
+bna_port_start(struct bna_port *port)
+{
+	port->flags |= BNA_PORT_F_DEVICE_READY;
+	if (port->flags & BNA_PORT_F_ENABLED)
+		bfa_fsm_send_event(port, PORT_E_START);
+}
+
+void
+bna_port_stop(struct bna_port *port)
+{
+	port->stop_cbfn = bna_device_cb_port_stopped;
+	port->stop_cbarg = &port->bna->device;
+
+	port->flags &= ~BNA_PORT_F_DEVICE_READY;
+	bfa_fsm_send_event(port, PORT_E_STOP);
+}
+
+void
+bna_port_fail(struct bna_port *port)
+{
+	port->flags &= ~BNA_PORT_F_DEVICE_READY;
+	bfa_fsm_send_event(port, PORT_E_FAIL);
+}
+
+void
+bna_port_cb_tx_stopped(struct bna_port *port, enum bna_cb_status status)
+{
+	bfa_wc_down(&port->chld_stop_wc);
+}
+
+void
+bna_port_cb_rx_stopped(struct bna_port *port, enum bna_cb_status status)
+{
+	bfa_wc_down(&port->chld_stop_wc);
+}
+
+void
+bna_port_cb_link_up(struct bna_port *port, struct bfi_ll_aen *aen,
+			int status)
+{
+	int i;
+	u8 prio_map;
+
+	port->llport.link_status = BNA_LINK_UP;
+	if (aen->cee_linkup)
+		port->llport.link_status = BNA_CEE_UP;
+
+	/* Compute the priority */
+	prio_map = aen->prio_map;
+	if (prio_map) {
+		for (i = 0; i < 8; i++) {
+			if ((prio_map >> i) & 0x1)
+				break;
+		}
+		port->priority = i;
+	} else
+		port->priority = 0;
+
+	/* Dispatch events */
+	bna_tx_mod_cee_link_status(&port->bna->tx_mod, aen->cee_linkup);
+	bna_tx_mod_prio_changed(&port->bna->tx_mod, port->priority);
+	port->link_cbfn(port->bna->bnad, port->llport.link_status);
+}
+
+void
+bna_port_cb_link_down(struct bna_port *port, int status)
+{
+	port->llport.link_status = BNA_LINK_DOWN;
+
+	/* Dispatch events */
+	bna_tx_mod_cee_link_status(&port->bna->tx_mod, BNA_LINK_DOWN);
+	port->link_cbfn(port->bna->bnad, BNA_LINK_DOWN);
+}
+
+int
+bna_port_mtu_get(struct bna_port *port)
+{
+	return port->mtu;
+}
+
+void
+bna_port_enable(struct bna_port *port)
+{
+	if (port->fsm != (bfa_sm_t)bna_port_sm_stopped)
+		return;
+
+	port->flags |= BNA_PORT_F_ENABLED;
+
+	if (port->flags & BNA_PORT_F_DEVICE_READY)
+		bfa_fsm_send_event(port, PORT_E_START);
+}
+
+void
+bna_port_disable(struct bna_port *port, enum bna_cleanup_type type,
+		 void (*cbfn)(void *, enum bna_cb_status))
+{
+	if (type == BNA_SOFT_CLEANUP) {
+		(*cbfn)(port->bna->bnad, BNA_CB_SUCCESS);
+		return;
+	}
+
+	port->stop_cbfn = cbfn;
+	port->stop_cbarg = port->bna->bnad;
+
+	port->flags &= ~BNA_PORT_F_ENABLED;
+
+	bfa_fsm_send_event(port, PORT_E_STOP);
+}
+
+void
+bna_port_pause_config(struct bna_port *port,
+		      struct bna_pause_config *pause_config,
+		      void (*cbfn)(struct bnad *, enum bna_cb_status))
+{
+	port->pause_config = *pause_config;
+
+	port->pause_cbfn = cbfn;
+
+	bfa_fsm_send_event(port, PORT_E_PAUSE_CFG);
+}
+
+void
+bna_port_mtu_set(struct bna_port *port, int mtu,
+		 void (*cbfn)(struct bnad *, enum bna_cb_status))
+{
+	port->mtu = mtu;
+
+	port->mtu_cbfn = cbfn;
+
+	bfa_fsm_send_event(port, PORT_E_MTU_CFG);
+}
+
+void
+bna_port_mac_get(struct bna_port *port, mac_t *mac)
+{
+	*mac = bfa_ioc_get_mac(&port->bna->device.ioc);
+}
+
+/**
+ * Should be called only when port is disabled
+ */
+void
+bna_port_type_set(struct bna_port *port, enum bna_port_type type)
+{
+	port->type = type;
+	port->llport.type = type;
+}
+
+/**
+ * Should be called only when port is disabled
+ */
+void
+bna_port_linkcbfn_set(struct bna_port *port,
+		      void (*linkcbfn)(struct bnad *, enum bna_link_status))
+{
+	port->link_cbfn = linkcbfn;
+}
+
+void
+bna_port_admin_up(struct bna_port *port)
+{
+	struct bna_llport *llport = &port->llport;
+
+	if (llport->flags & BNA_LLPORT_F_ENABLED)
+		return;
+
+	llport->flags |= BNA_LLPORT_F_ENABLED;
+
+	if (llport->flags & BNA_LLPORT_F_RX_ENABLED)
+		bfa_fsm_send_event(llport, LLPORT_E_UP);
+}
+
+void
+bna_port_admin_down(struct bna_port *port)
+{
+	struct bna_llport *llport = &port->llport;
+
+	if (!(llport->flags & BNA_LLPORT_F_ENABLED))
+		return;
+
+	llport->flags &= ~BNA_LLPORT_F_ENABLED;
+
+	if (llport->flags & BNA_LLPORT_F_RX_ENABLED)
+		bfa_fsm_send_event(llport, LLPORT_E_DOWN);
+}
+
+/**
+ * DEVICE
+ */
+#define enable_mbox_intr(_device)\
+do {\
+	u32 intr_status;\
+	bna_intr_status_get((_device)->bna, intr_status);\
+	bnad_cb_device_enable_mbox_intr((_device)->bna->bnad);\
+	bna_mbox_intr_enable((_device)->bna);\
+} while (0)
+
+#define disable_mbox_intr(_device)\
+do {\
+	bna_mbox_intr_disable((_device)->bna);\
+	bnad_cb_device_disable_mbox_intr((_device)->bna->bnad);\
+} while (0)
+
+const struct bna_chip_regs_offset reg_offset[] =
+{{HOST_PAGE_NUM_FN0, HOSTFN0_INT_STATUS,
+	HOSTFN0_INT_MASK, HOST_MSIX_ERR_INDEX_FN0},
+{HOST_PAGE_NUM_FN1, HOSTFN1_INT_STATUS,
+	HOSTFN1_INT_MASK, HOST_MSIX_ERR_INDEX_FN1},
+{HOST_PAGE_NUM_FN2, HOSTFN2_INT_STATUS,
+	HOSTFN2_INT_MASK, HOST_MSIX_ERR_INDEX_FN2},
+{HOST_PAGE_NUM_FN3, HOSTFN3_INT_STATUS,
+	HOSTFN3_INT_MASK, HOST_MSIX_ERR_INDEX_FN3},
+};
+
+enum bna_device_event {
+	DEVICE_E_ENABLE			= 1,
+	DEVICE_E_DISABLE		= 2,
+	DEVICE_E_IOC_READY		= 3,
+	DEVICE_E_IOC_FAILED		= 4,
+	DEVICE_E_IOC_DISABLED		= 5,
+	DEVICE_E_IOC_RESET		= 6,
+	DEVICE_E_PORT_STOPPED		= 7,
+};
+
+enum bna_device_state {
+	BNA_DEVICE_STOPPED		= 1,
+	BNA_DEVICE_IOC_READY_WAIT 	= 2,
+	BNA_DEVICE_READY		= 3,
+	BNA_DEVICE_PORT_STOP_WAIT 	= 4,
+	BNA_DEVICE_IOC_DISABLE_WAIT 	= 5,
+	BNA_DEVICE_FAILED		= 6
+};
+
+bfa_fsm_state_decl(bna_device, stopped, struct bna_device,
+			enum bna_device_event);
+bfa_fsm_state_decl(bna_device, ioc_ready_wait, struct bna_device,
+			enum bna_device_event);
+bfa_fsm_state_decl(bna_device, ready, struct bna_device,
+			enum bna_device_event);
+bfa_fsm_state_decl(bna_device, port_stop_wait, struct bna_device,
+			enum bna_device_event);
+bfa_fsm_state_decl(bna_device, ioc_disable_wait, struct bna_device,
+			enum bna_device_event);
+bfa_fsm_state_decl(bna_device, failed, struct bna_device,
+			enum bna_device_event);
+
+static struct bfa_sm_table device_sm_table[] = {
+	{BFA_SM(bna_device_sm_stopped), BNA_DEVICE_STOPPED},
+	{BFA_SM(bna_device_sm_ioc_ready_wait), BNA_DEVICE_IOC_READY_WAIT},
+	{BFA_SM(bna_device_sm_ready), BNA_DEVICE_READY},
+	{BFA_SM(bna_device_sm_port_stop_wait), BNA_DEVICE_PORT_STOP_WAIT},
+	{BFA_SM(bna_device_sm_ioc_disable_wait), BNA_DEVICE_IOC_DISABLE_WAIT},
+	{BFA_SM(bna_device_sm_failed), BNA_DEVICE_FAILED},
+};
+
+static void
+bna_device_sm_stopped_entry(struct bna_device *device)
+{
+	if (device->stop_cbfn)
+		device->stop_cbfn(device->stop_cbarg, BNA_CB_SUCCESS);
+
+	device->stop_cbfn = NULL;
+	device->stop_cbarg = NULL;
+}
+
+static void
+bna_device_sm_stopped(struct bna_device *device,
+			enum bna_device_event event)
+{
+	switch (event) {
+	case DEVICE_E_ENABLE:
+		if (device->intr_type == BNA_INTR_T_MSIX)
+			bna_mbox_msix_idx_set(device);
+		bfa_ioc_enable(&device->ioc);
+		bfa_fsm_set_state(device, bna_device_sm_ioc_ready_wait);
+		break;
+
+	case DEVICE_E_DISABLE:
+		bfa_fsm_set_state(device, bna_device_sm_stopped);
+		break;
+
+	case DEVICE_E_IOC_RESET:
+		enable_mbox_intr(device);
+		break;
+
+	case DEVICE_E_IOC_FAILED:
+		bfa_fsm_set_state(device, bna_device_sm_failed);
+		break;
+
+	default:
+		bfa_sm_fault(device->bna, event);
+	}
+}
+
+static void
+bna_device_sm_ioc_ready_wait_entry(struct bna_device *device)
+{
+	/**
+	 * Do not call bfa_ioc_enable() here. It must be called in the
+	 * previous state due to failed -> ioc_ready_wait transition.
+	 */
+}
+
+static void
+bna_device_sm_ioc_ready_wait(struct bna_device *device,
+				enum bna_device_event event)
+{
+	switch (event) {
+	case DEVICE_E_DISABLE:
+		if (device->ready_cbfn)
+			device->ready_cbfn(device->ready_cbarg,
+						BNA_CB_INTERRUPT);
+		device->ready_cbfn = NULL;
+		device->ready_cbarg = NULL;
+		bfa_fsm_set_state(device, bna_device_sm_ioc_disable_wait);
+		break;
+
+	case DEVICE_E_IOC_READY:
+		bfa_fsm_set_state(device, bna_device_sm_ready);
+		break;
+
+	case DEVICE_E_IOC_FAILED:
+		bfa_fsm_set_state(device, bna_device_sm_failed);
+		break;
+
+	case DEVICE_E_IOC_RESET:
+		enable_mbox_intr(device);
+		break;
+
+	default:
+		bfa_sm_fault(device->bna, event);
+	}
+}
+
+static void
+bna_device_sm_ready_entry(struct bna_device *device)
+{
+	bna_mbox_mod_start(&device->bna->mbox_mod);
+	bna_port_start(&device->bna->port);
+
+	if (device->ready_cbfn)
+		device->ready_cbfn(device->ready_cbarg,
+					BNA_CB_SUCCESS);
+	device->ready_cbfn = NULL;
+	device->ready_cbarg = NULL;
+}
+
+static void
+bna_device_sm_ready(struct bna_device *device, enum bna_device_event event)
+{
+	switch (event) {
+	case DEVICE_E_DISABLE:
+		bfa_fsm_set_state(device, bna_device_sm_port_stop_wait);
+		break;
+
+	case DEVICE_E_IOC_FAILED:
+		bfa_fsm_set_state(device, bna_device_sm_failed);
+		break;
+
+	default:
+		bfa_sm_fault(device->bna, event);
+	}
+}
+
+static void
+bna_device_sm_port_stop_wait_entry(struct bna_device *device)
+{
+	bna_port_stop(&device->bna->port);
+}
+
+static void
+bna_device_sm_port_stop_wait(struct bna_device *device,
+				enum bna_device_event event)
+{
+	switch (event) {
+	case DEVICE_E_PORT_STOPPED:
+		bna_mbox_mod_stop(&device->bna->mbox_mod);
+		bfa_fsm_set_state(device, bna_device_sm_ioc_disable_wait);
+		break;
+
+	case DEVICE_E_IOC_FAILED:
+		disable_mbox_intr(device);
+		bna_port_fail(&device->bna->port);
+		break;
+
+	default:
+		bfa_sm_fault(device->bna, event);
+	}
+}
+
+static void
+bna_device_sm_ioc_disable_wait_entry(struct bna_device *device)
+{
+	bfa_ioc_disable(&device->ioc);
+}
+
+static void
+bna_device_sm_ioc_disable_wait(struct bna_device *device,
+				enum bna_device_event event)
+{
+	switch (event) {
+	case DEVICE_E_IOC_DISABLED:
+		disable_mbox_intr(device);
+		bfa_fsm_set_state(device, bna_device_sm_stopped);
+		break;
+
+	default:
+		bfa_sm_fault(device->bna, event);
+	}
+}
+
+static void
+bna_device_sm_failed_entry(struct bna_device *device)
+{
+	disable_mbox_intr(device);
+	bna_port_fail(&device->bna->port);
+	bna_mbox_mod_stop(&device->bna->mbox_mod);
+
+	if (device->ready_cbfn)
+		device->ready_cbfn(device->ready_cbarg,
+					BNA_CB_FAIL);
+	device->ready_cbfn = NULL;
+	device->ready_cbarg = NULL;
+}
+
+static void
+bna_device_sm_failed(struct bna_device *device,
+			enum bna_device_event event)
+{
+	switch (event) {
+	case DEVICE_E_DISABLE:
+		bfa_fsm_set_state(device, bna_device_sm_ioc_disable_wait);
+		break;
+
+	case DEVICE_E_IOC_RESET:
+		enable_mbox_intr(device);
+		bfa_fsm_set_state(device, bna_device_sm_ioc_ready_wait);
+		break;
+
+	default:
+		bfa_sm_fault(device->bna, event);
+	}
+}
+
+/* IOC callback functions */
+
+static void
+bna_device_cb_iocll_ready(void *dev, enum bfa_status error)
+{
+	struct bna_device *device = (struct bna_device *)dev;
+
+	if (error)
+		bfa_fsm_send_event(device, DEVICE_E_IOC_FAILED);
+	else
+		bfa_fsm_send_event(device, DEVICE_E_IOC_READY);
+}
+
+static void
+bna_device_cb_iocll_disabled(void *dev)
+{
+	struct bna_device *device = (struct bna_device *)dev;
+
+	bfa_fsm_send_event(device, DEVICE_E_IOC_DISABLED);
+}
+
+static void
+bna_device_cb_iocll_failed(void *dev)
+{
+	struct bna_device *device = (struct bna_device *)dev;
+
+	bfa_fsm_send_event(device, DEVICE_E_IOC_FAILED);
+}
+
+static void
+bna_device_cb_iocll_reset(void *dev)
+{
+	struct bna_device *device = (struct bna_device *)dev;
+
+	bfa_fsm_send_event(device, DEVICE_E_IOC_RESET);
+}
+
+static struct bfa_ioc_cbfn bfa_iocll_cbfn = {
+	bna_device_cb_iocll_ready,
+	bna_device_cb_iocll_disabled,
+	bna_device_cb_iocll_failed,
+	bna_device_cb_iocll_reset
+};
+
+void
+bna_device_init(struct bna_device *device, struct bna *bna,
+		struct bna_res_info *res_info)
+{
+	u64 dma;
+
+	device->bna = bna;
+
+	/**
+	 * Attach IOC and claim:
+	 *	1. DMA memory for IOC attributes
+	 *	2. Kernel memory for FW trace
+	 */
+	bfa_ioc_attach(&device->ioc, device, &bfa_iocll_cbfn);
+	bfa_ioc_pci_init(&device->ioc, &bna->pcidev, BFI_MC_LL);
+
+	BNA_GET_DMA_ADDR(
+		&res_info[BNA_RES_MEM_T_ATTR].res_u.mem_info.mdl[0].dma, dma);
+	bfa_ioc_mem_claim(&device->ioc,
+		res_info[BNA_RES_MEM_T_ATTR].res_u.mem_info.mdl[0].kva,
+			  dma);
+
+	bna_adv_device_init(device, bna, res_info);
+	/*
+	 * Initialize mbox_mod only after IOC, so that mbox handler
+	 * registration goes through
+	 */
+	device->intr_type =
+		res_info[BNA_RES_INTR_T_MBOX].res_u.intr_info.intr_type;
+	device->vector =
+		res_info[BNA_RES_INTR_T_MBOX].res_u.intr_info.idl[0].vector;
+	bna_mbox_mod_init(&bna->mbox_mod, bna);
+
+	device->ready_cbfn = device->stop_cbfn = NULL;
+	device->ready_cbarg = device->stop_cbarg = NULL;
+
+	bfa_fsm_set_state(device, bna_device_sm_stopped);
+}
+
+void
+bna_device_uninit(struct bna_device *device)
+{
+	bna_mbox_mod_uninit(&device->bna->mbox_mod);
+
+	bfa_cee_detach(&device->bna->cee);
+
+	bfa_ioc_detach(&device->ioc);
+
+	device->bna = NULL;
+}
+
+void
+bna_device_cb_port_stopped(void *arg, enum bna_cb_status status)
+{
+	struct bna_device *device = (struct bna_device *)arg;
+
+	bfa_fsm_send_event(device, DEVICE_E_PORT_STOPPED);
+}
+
+int
+bna_device_status_get(struct bna_device *device)
+{
+	return (device->fsm == (bfa_fsm_t)bna_device_sm_ready);
+}
+
+void
+bna_device_enable(struct bna_device *device)
+{
+	if (device->fsm != (bfa_fsm_t)bna_device_sm_stopped) {
+		bnad_cb_device_enabled(device->bna->bnad, BNA_CB_BUSY);
+		return;
+	}
+
+	device->ready_cbfn = bnad_cb_device_enabled;
+	device->ready_cbarg = device->bna->bnad;
+
+	bfa_fsm_send_event(device, DEVICE_E_ENABLE);
+}
+
+void
+bna_device_disable(struct bna_device *device, enum bna_cleanup_type type)
+{
+	if (type == BNA_SOFT_CLEANUP) {
+		bnad_cb_device_disabled(device->bna->bnad, BNA_CB_SUCCESS);
+		return;
+	}
+
+	device->stop_cbfn = bnad_cb_device_disabled;
+	device->stop_cbarg = device->bna->bnad;
+
+	bfa_fsm_send_event(device, DEVICE_E_DISABLE);
+}
+
+int
+bna_device_state_get(struct bna_device *device)
+{
+	return bfa_sm_to_state(device_sm_table, device->fsm);
+}
+
+u32 bna_dim_vector[BNA_LOAD_T_MAX][BNA_BIAS_T_MAX] = {
+	{12, 20},
+	{10, 18},
+	{8, 16},
+	{6, 12},
+	{4, 8},
+	{3, 6},
+	{2, 4},
+	{1, 2},
+};
+
+u32 bna_napi_dim_vector[BNA_LOAD_T_MAX][BNA_BIAS_T_MAX] = {
+	{12, 12},
+	{6, 10},
+	{5, 10},
+	{4, 8},
+	{3, 6},
+	{3, 6},
+	{2, 4},
+	{1, 2},
+};
+
+/* device */
+void
+bna_adv_device_init(struct bna_device *device, struct bna *bna,
+		struct bna_res_info *res_info)
+{
+	u8 *kva;
+	u64 dma;
+
+	device->bna = bna;
+
+	kva = res_info[BNA_RES_MEM_T_FWTRC].res_u.mem_info.mdl[0].kva;
+
+	/**
+	 * Attach common modules (Diag, SFP, CEE, Port) and claim respective
+	 * DMA memory.
+	 */
+	BNA_GET_DMA_ADDR(
+		&res_info[BNA_RES_MEM_T_COM].res_u.mem_info.mdl[0].dma, dma);
+	kva = res_info[BNA_RES_MEM_T_COM].res_u.mem_info.mdl[0].kva;
+
+	bfa_cee_attach(&bna->cee, &device->ioc, bna);
+	bfa_cee_mem_claim(&bna->cee, kva, dma);
+	kva += bfa_cee_meminfo();
+	dma += bfa_cee_meminfo();
+
+}
+
+/* utils */
+
+void
+bna_adv_res_req(struct bna_res_info *res_info)
+{
+	/* DMA memory for COMMON_MODULE */
+	res_info[BNA_RES_MEM_T_COM].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_COM].res_u.mem_info.mem_type = BNA_MEM_T_DMA;
+	res_info[BNA_RES_MEM_T_COM].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_COM].res_u.mem_info.len = ALIGN(
+				bfa_cee_meminfo(), PAGE_SIZE);
+
+	/* Virtual memory for retreiving fw_trc */
+	res_info[BNA_RES_MEM_T_FWTRC].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_FWTRC].res_u.mem_info.mem_type = BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_FWTRC].res_u.mem_info.num = 0;
+	res_info[BNA_RES_MEM_T_FWTRC].res_u.mem_info.len = 0;
+
+	/* DMA memory for retreiving stats */
+	res_info[BNA_RES_MEM_T_STATS].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_STATS].res_u.mem_info.mem_type = BNA_MEM_T_DMA;
+	res_info[BNA_RES_MEM_T_STATS].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_STATS].res_u.mem_info.len =
+				ALIGN(BFI_HW_STATS_SIZE, PAGE_SIZE);
+
+	/* Virtual memory for soft stats */
+	res_info[BNA_RES_MEM_T_SWSTATS].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_SWSTATS].res_u.mem_info.mem_type = BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_SWSTATS].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_SWSTATS].res_u.mem_info.len =
+				sizeof(struct bna_sw_stats);
+}
+
+static void
+bna_sw_stats_get(struct bna *bna, struct bna_sw_stats *sw_stats)
+{
+	struct bna_tx *tx;
+	struct bna_txq *txq;
+	struct bna_rx *rx;
+	struct bna_rxp *rxp;
+	struct list_head *qe;
+	struct list_head *txq_qe;
+	struct list_head *rxp_qe;
+	struct list_head *mac_qe;
+	int i;
+
+	sw_stats->device_state = bna_device_state_get(&bna->device);
+	sw_stats->port_state = bna_port_state_get(&bna->port);
+	sw_stats->port_flags = bna->port.flags;
+	sw_stats->llport_state = bna_llport_state_get(&bna->port.llport);
+	sw_stats->priority = bna->port.priority;
+
+	i = 0;
+	list_for_each(qe, &bna->tx_mod.tx_active_q) {
+		tx = (struct bna_tx *)qe;
+		sw_stats->tx_stats[i].tx_state = bna_tx_state_get(tx);
+		sw_stats->tx_stats[i].tx_flags = tx->flags;
+
+		sw_stats->tx_stats[i].num_txqs = 0;
+		sw_stats->tx_stats[i].txq_bmap[0] = 0;
+		sw_stats->tx_stats[i].txq_bmap[1] = 0;
+		list_for_each(txq_qe, &tx->txq_q) {
+			txq = (struct bna_txq *)txq_qe;
+			if (txq->txq_id < 32)
+				sw_stats->tx_stats[i].txq_bmap[0] |=
+						((u32)1 << txq->txq_id);
+			else
+				sw_stats->tx_stats[i].txq_bmap[1] |=
+						((u32)
+						 1 << (txq->txq_id - 32));
+			sw_stats->tx_stats[i].num_txqs++;
+		}
+
+		sw_stats->tx_stats[i].txf_id = tx->txf.txf_id;
+
+		i++;
+	}
+	sw_stats->num_active_tx = i;
+
+	i = 0;
+	list_for_each(qe, &bna->rx_mod.rx_active_q) {
+		rx = (struct bna_rx *)qe;
+		sw_stats->rx_stats[i].rx_state = bna_rx_state_get(rx);
+		sw_stats->rx_stats[i].rx_flags = rx->rx_flags;
+
+		sw_stats->rx_stats[i].num_rxps = 0;
+		sw_stats->rx_stats[i].num_rxqs = 0;
+		sw_stats->rx_stats[i].rxq_bmap[0] = 0;
+		sw_stats->rx_stats[i].rxq_bmap[1] = 0;
+		sw_stats->rx_stats[i].cq_bmap[0] = 0;
+		sw_stats->rx_stats[i].cq_bmap[1] = 0;
+		list_for_each(rxp_qe, &rx->rxp_q) {
+			rxp = (struct bna_rxp *)rxp_qe;
+
+			sw_stats->rx_stats[i].num_rxqs += 1;
+
+			if (rxp->type == BNA_RXP_SINGLE) {
+				if (rxp->rxq.single.only->rxq_id < 32) {
+					sw_stats->rx_stats[i].rxq_bmap[0] |=
+					((u32)1 <<
+					rxp->rxq.single.only->rxq_id);
+				} else {
+					sw_stats->rx_stats[i].rxq_bmap[1] |=
+					((u32)1 <<
+					(rxp->rxq.single.only->rxq_id - 32));
+				}
+			} else {
+				if (rxp->rxq.slr.large->rxq_id < 32) {
+					sw_stats->rx_stats[i].rxq_bmap[0] |=
+					((u32)1 <<
+					rxp->rxq.slr.large->rxq_id);
+				} else {
+					sw_stats->rx_stats[i].rxq_bmap[1] |=
+					((u32)1 <<
+					(rxp->rxq.slr.large->rxq_id - 32));
+				}
+
+				if (rxp->rxq.slr.small->rxq_id < 32) {
+					sw_stats->rx_stats[i].rxq_bmap[0] |=
+					((u32)1 <<
+					rxp->rxq.slr.small->rxq_id);
+				} else {
+					sw_stats->rx_stats[i].rxq_bmap[1] |=
+				((u32)1 <<
+				 (rxp->rxq.slr.small->rxq_id - 32));
+				}
+				sw_stats->rx_stats[i].num_rxqs += 1;
+			}
+
+			if (rxp->cq.cq_id < 32)
+				sw_stats->rx_stats[i].cq_bmap[0] |=
+					(1 << rxp->cq.cq_id);
+			else
+				sw_stats->rx_stats[i].cq_bmap[1] |=
+					(1 << (rxp->cq.cq_id - 32));
+
+			sw_stats->rx_stats[i].num_rxps++;
+		}
+
+		sw_stats->rx_stats[i].rxf_id = rx->rxf.rxf_id;
+		sw_stats->rx_stats[i].rxf_state = bna_rxf_state_get(&rx->rxf);
+		sw_stats->rx_stats[i].rxf_oper_state = rx->rxf.rxf_oper_state;
+
+		sw_stats->rx_stats[i].num_active_ucast = 0;
+		if (rx->rxf.ucast_active_mac)
+			sw_stats->rx_stats[i].num_active_ucast++;
+		list_for_each(mac_qe, &rx->rxf.ucast_active_q)
+			sw_stats->rx_stats[i].num_active_ucast++;
+
+		sw_stats->rx_stats[i].num_active_mcast = 0;
+		list_for_each(mac_qe, &rx->rxf.mcast_active_q)
+			sw_stats->rx_stats[i].num_active_mcast++;
+
+		sw_stats->rx_stats[i].rxmode_active = rx->rxf.rxmode_active;
+		sw_stats->rx_stats[i].vlan_filter_status =
+						rx->rxf.vlan_filter_status;
+		memcpy(sw_stats->rx_stats[i].vlan_filter_table,
+				rx->rxf.vlan_filter_table,
+				sizeof(u32) * ((BFI_MAX_VLAN + 1) / 32));
+
+		sw_stats->rx_stats[i].rss_status = rx->rxf.rss_status;
+		sw_stats->rx_stats[i].hds_status = rx->rxf.hds_status;
+
+		i++;
+	}
+	sw_stats->num_active_rx = i;
+}
+
+static void
+bna_fw_cb_stats_get(void *arg, int status)
+{
+	struct bna *bna = (struct bna *)arg;
+	u64 *p_stats;
+	int i, count;
+	int rxf_count, txf_count;
+	u64 rxf_bmap, txf_bmap;
+
+	bfa_q_qe_init(&bna->mbox_qe.qe);
+
+	if (status == 0) {
+		p_stats = (u64 *)bna->stats.hw_stats;
+		count = sizeof(struct bfi_ll_stats) / sizeof(u64);
+		for (i = 0; i < count; i++)
+			p_stats[i] = cpu_to_be64(p_stats[i]);
+
+		rxf_count = 0;
+		rxf_bmap = (u64)bna->stats.rxf_bmap[0] |
+			((u64)bna->stats.rxf_bmap[1] << 32);
+		for (i = 0; i < BFI_LL_RXF_ID_MAX; i++)
+			if (rxf_bmap & ((u64)1 << i))
+				rxf_count++;
+
+		txf_count = 0;
+		txf_bmap = (u64)bna->stats.txf_bmap[0] |
+			((u64)bna->stats.txf_bmap[1] << 32);
+		for (i = 0; i < BFI_LL_TXF_ID_MAX; i++)
+			if (txf_bmap & ((u64)1 << i))
+				txf_count++;
+
+		p_stats = (u64 *)&bna->stats.hw_stats->rxf_stats[0] +
+				((rxf_count * sizeof(struct bfi_ll_stats_rxf) +
+				txf_count * sizeof(struct bfi_ll_stats_txf))/
+				sizeof(u64));
+
+		/* Populate the TXF stats from the firmware DMAed copy */
+		for (i = (BFI_LL_TXF_ID_MAX - 1); i >= 0; i--)
+			if (txf_bmap & ((u64)1 << i)) {
+				p_stats -= sizeof(struct bfi_ll_stats_txf)/
+						sizeof(u64);
+				memcpy(&bna->stats.hw_stats->txf_stats[i],
+					p_stats,
+					sizeof(struct bfi_ll_stats_txf));
+			}
+
+		/* Populate the RXF stats from the firmware DMAed copy */
+		for (i = (BFI_LL_RXF_ID_MAX - 1); i >= 0; i--)
+			if (rxf_bmap & ((u64)1 << i)) {
+				p_stats -= sizeof(struct bfi_ll_stats_rxf)/
+						sizeof(u64);
+				memcpy(&bna->stats.hw_stats->rxf_stats[i],
+					p_stats,
+					sizeof(struct bfi_ll_stats_rxf));
+			}
+
+		bna_sw_stats_get(bna, bna->stats.sw_stats);
+		bnad_cb_stats_get(bna->bnad, BNA_CB_SUCCESS, &bna->stats);
+	} else
+		bnad_cb_stats_get(bna->bnad, BNA_CB_FAIL, &bna->stats);
+}
+
+static void
+bna_fw_stats_get(struct bna *bna)
+{
+	struct bfi_ll_stats_req ll_req;
+
+	bfi_h2i_set(ll_req.mh, BFI_MC_LL, BFI_LL_H2I_STATS_GET_REQ, 0);
+	ll_req.stats_mask = htons(BFI_LL_STATS_ALL);
+
+	ll_req.rxf_id_mask[0] = htonl(bna->rx_mod.rxf_bmap[0]);
+	ll_req.rxf_id_mask[1] =	htonl(bna->rx_mod.rxf_bmap[1]);
+	ll_req.txf_id_mask[0] =	htonl(bna->tx_mod.txf_bmap[0]);
+	ll_req.txf_id_mask[1] =	htonl(bna->tx_mod.txf_bmap[1]);
+
+	ll_req.host_buffer.a32.addr_hi = bna->hw_stats_dma.msb;
+	ll_req.host_buffer.a32.addr_lo = bna->hw_stats_dma.lsb;
+
+	bna_mbox_qe_fill(&bna->mbox_qe, &ll_req, sizeof(ll_req),
+				bna_fw_cb_stats_get, bna);
+	bna_mbox_send(bna, &bna->mbox_qe);
+
+	bna->stats.rxf_bmap[0] = bna->rx_mod.rxf_bmap[0];
+	bna->stats.rxf_bmap[1] = bna->rx_mod.rxf_bmap[1];
+	bna->stats.txf_bmap[0] = bna->tx_mod.txf_bmap[0];
+	bna->stats.txf_bmap[1] = bna->tx_mod.txf_bmap[1];
+}
+
+static void
+bna_fw_cb_stats_clr(void *arg, int status)
+{
+	struct bna *bna = (struct bna *)arg;
+
+	bfa_q_qe_init(&bna->mbox_qe.qe);
+
+	memset(bna->stats.sw_stats, 0, sizeof(struct bna_sw_stats));
+	memset(bna->stats.hw_stats, 0, sizeof(struct bfi_ll_stats));
+
+	bnad_cb_stats_clr(bna->bnad);
+}
+
+static void
+bna_fw_stats_clr(struct bna *bna)
+{
+	struct bfi_ll_stats_req ll_req;
+
+	bfi_h2i_set(ll_req.mh, BFI_MC_LL, BFI_LL_H2I_STATS_CLEAR_REQ, 0);
+	ll_req.stats_mask = htons(BFI_LL_STATS_ALL);
+	ll_req.rxf_id_mask[0] = htonl(0xffffffff);
+	ll_req.rxf_id_mask[1] =	htonl(0xffffffff);
+	ll_req.txf_id_mask[0] =	htonl(0xffffffff);
+	ll_req.txf_id_mask[1] =	htonl(0xffffffff);
+
+	bna_mbox_qe_fill(&bna->mbox_qe, &ll_req, sizeof(ll_req),
+				bna_fw_cb_stats_clr, bna);
+	bna_mbox_send(bna, &bna->mbox_qe);
+}
+
+void
+bna_stats_get(struct bna *bna)
+{
+	if (bna_device_status_get(&bna->device))
+		bna_fw_stats_get(bna);
+	else
+		bnad_cb_stats_get(bna->bnad, BNA_CB_FAIL, &bna->stats);
+}
+
+void
+bna_stats_clr(struct bna *bna)
+{
+	if (bna_device_status_get(&bna->device))
+		bna_fw_stats_clr(bna);
+	else {
+		memset(&bna->stats.sw_stats, 0,
+				sizeof(struct bna_sw_stats));
+		memset(bna->stats.hw_stats, 0,
+				sizeof(struct bfi_ll_stats));
+		bnad_cb_stats_clr(bna->bnad);
+	}
+}
+
+/* IB */
+void
+bna_ib_coalescing_timeo_set(struct bna_ib *ib, u8 coalescing_timeo)
+{
+	ib->ib_config.coalescing_timeo = coalescing_timeo;
+
+	if (ib->start_count)
+		ib->door_bell.doorbell_ack = BNA_DOORBELL_IB_INT_ACK(
+				(u32)ib->ib_config.coalescing_timeo, 0);
+}
+
+/* RxF */
+void
+bna_rxf_adv_init(struct bna_rxf *rxf,
+		struct bna_rx *rx,
+		struct bna_rx_config *q_config)
+{
+	switch (q_config->rxp_type) {
+	case BNA_RXP_SINGLE:
+		/* No-op */
+		break;
+	case BNA_RXP_SLR:
+		rxf->ctrl_flags |= BNA_RXF_CF_SM_LG_RXQ;
+		break;
+	case BNA_RXP_HDS:
+		rxf->hds_cfg.hdr_type = q_config->hds_config.hdr_type;
+		rxf->hds_cfg.header_size =
+				q_config->hds_config.header_size;
+		rxf->forced_offset = 0;
+		break;
+	default:
+		break;
+	}
+
+	if (q_config->rss_status == BNA_STATUS_T_ENABLED) {
+		rxf->ctrl_flags |= BNA_RXF_CF_RSS_ENABLE;
+		rxf->rss_cfg.hash_type = q_config->rss_config.hash_type;
+		rxf->rss_cfg.hash_mask = q_config->rss_config.hash_mask;
+		memcpy(&rxf->rss_cfg.toeplitz_hash_key[0],
+			&q_config->rss_config.toeplitz_hash_key[0],
+			sizeof(rxf->rss_cfg.toeplitz_hash_key));
+	}
+}
+
+static void
+rxf_fltr_mbox_cmd(struct bna_rxf *rxf, u8 cmd, enum bna_status status)
+{
+	struct bfi_ll_rxf_req req;
+
+	bfi_h2i_set(req.mh, BFI_MC_LL, cmd, 0);
+
+	req.rxf_id = rxf->rxf_id;
+	req.enable = status;
+
+	bna_mbox_qe_fill(&rxf->mbox_qe, &req, sizeof(req),
+			rxf_cb_cam_fltr_mbox_cmd, rxf);
+
+	bna_mbox_send(rxf->rx->bna, &rxf->mbox_qe);
+}
+
+void
+__rxf_default_function_config(struct bna_rxf *rxf, enum bna_status status)
+{
+	struct bna_rx_fndb_ram *rx_fndb_ram;
+	u32 ctrl_flags;
+	int i;
+
+	rx_fndb_ram = (struct bna_rx_fndb_ram *)
+			BNA_GET_MEM_BASE_ADDR(rxf->rx->bna->pcidev.pci_bar_kva,
+			RX_FNDB_RAM_BASE_OFFSET);
+
+	for (i = 0; i < BFI_MAX_RXF; i++) {
+		if (status == BNA_STATUS_T_ENABLED) {
+			if (i == rxf->rxf_id)
+				continue;
+
+			ctrl_flags =
+				readl(&rx_fndb_ram[i].control_flags);
+			ctrl_flags |= BNA_RXF_CF_DEFAULT_FUNCTION_ENABLE;
+			writel(ctrl_flags,
+						&rx_fndb_ram[i].control_flags);
+		} else {
+			ctrl_flags =
+				readl(&rx_fndb_ram[i].control_flags);
+			ctrl_flags &= ~BNA_RXF_CF_DEFAULT_FUNCTION_ENABLE;
+			writel(ctrl_flags,
+						&rx_fndb_ram[i].control_flags);
+		}
+	}
+}
+
+int
+rxf_process_packet_filter_ucast(struct bna_rxf *rxf)
+{
+	struct bna_mac *mac = NULL;
+	struct list_head *qe;
+
+	/* Add additional MAC entries */
+	if (!list_empty(&rxf->ucast_pending_add_q)) {
+		bfa_q_deq(&rxf->ucast_pending_add_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_UCAST_ADD_REQ, mac);
+		list_add_tail(&mac->qe, &rxf->ucast_active_q);
+		return 1;
+	}
+
+	/* Delete MAC addresses previousely added */
+	if (!list_empty(&rxf->ucast_pending_del_q)) {
+		bfa_q_deq(&rxf->ucast_pending_del_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_UCAST_DEL_REQ, mac);
+		bna_ucam_mod_mac_put(&rxf->rx->bna->ucam_mod, mac);
+		return 1;
+	}
+
+	return 0;
+}
+
+int
+rxf_process_packet_filter_promisc(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+
+	/* Enable/disable promiscuous mode */
+	if (is_promisc_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* move promisc configuration from pending -> active */
+		promisc_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active |= BNA_RXMODE_PROMISC;
+
+		/* Disable VLAN filter to allow all VLANs */
+		__rxf_vlan_filter_set(rxf, BNA_STATUS_T_DISABLED);
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_RXF_PROMISCUOUS_SET_REQ,
+				BNA_STATUS_T_ENABLED);
+		return 1;
+	} else if (is_promisc_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* move promisc configuration from pending -> active */
+		promisc_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_PROMISC;
+		bna->rxf_promisc_id = BFI_MAX_RXF;
+
+		/* Revert VLAN filter */
+		__rxf_vlan_filter_set(rxf, rxf->vlan_filter_status);
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_RXF_PROMISCUOUS_SET_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	return 0;
+}
+
+int
+rxf_process_packet_filter_default(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+
+	/* Enable/disable default mode */
+	if (is_default_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* move default configuration from pending -> active */
+		default_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active |= BNA_RXMODE_DEFAULT;
+
+		/* Disable VLAN filter to allow all VLANs */
+		__rxf_vlan_filter_set(rxf, BNA_STATUS_T_DISABLED);
+		/* Redirect all other RxF vlan filtering to this one */
+		__rxf_default_function_config(rxf, BNA_STATUS_T_ENABLED);
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_RXF_DEFAULT_SET_REQ,
+				BNA_STATUS_T_ENABLED);
+		return 1;
+	} else if (is_default_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* move default configuration from pending -> active */
+		default_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_DEFAULT;
+		bna->rxf_default_id = BFI_MAX_RXF;
+
+		/* Revert VLAN filter */
+		__rxf_vlan_filter_set(rxf, rxf->vlan_filter_status);
+		/* Stop RxF vlan filter table redirection */
+		__rxf_default_function_config(rxf, BNA_STATUS_T_DISABLED);
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_RXF_DEFAULT_SET_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	return 0;
+}
+
+int
+rxf_process_packet_filter_allmulti(struct bna_rxf *rxf)
+{
+	/* Enable/disable allmulti mode */
+	if (is_allmulti_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* move allmulti configuration from pending -> active */
+		allmulti_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active |= BNA_RXMODE_ALLMULTI;
+
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_MAC_MCAST_FILTER_REQ,
+				BNA_STATUS_T_ENABLED);
+		return 1;
+	} else if (is_allmulti_disable(rxf->rxmode_pending,
+					rxf->rxmode_pending_bitmask)) {
+		/* move allmulti configuration from pending -> active */
+		allmulti_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_ALLMULTI;
+
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_MAC_MCAST_FILTER_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	return 0;
+}
+
+int
+rxf_clear_packet_filter_ucast(struct bna_rxf *rxf)
+{
+	struct bna_mac *mac = NULL;
+	struct list_head *qe;
+
+	/* 1. delete pending ucast entries */
+	if (!list_empty(&rxf->ucast_pending_del_q)) {
+		bfa_q_deq(&rxf->ucast_pending_del_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_UCAST_DEL_REQ, mac);
+		bna_ucam_mod_mac_put(&rxf->rx->bna->ucam_mod, mac);
+		return 1;
+	}
+
+	/* 2. clear active ucast entries; move them to pending_add_q */
+	if (!list_empty(&rxf->ucast_active_q)) {
+		bfa_q_deq(&rxf->ucast_active_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_UCAST_DEL_REQ, mac);
+		list_add_tail(&mac->qe, &rxf->ucast_pending_add_q);
+		return 1;
+	}
+
+	return 0;
+}
+
+int
+rxf_clear_packet_filter_promisc(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+
+	/* 6. Execute pending promisc mode disable command */
+	if (is_promisc_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* move promisc configuration from pending -> active */
+		promisc_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_PROMISC;
+		bna->rxf_promisc_id = BFI_MAX_RXF;
+
+		/* Revert VLAN filter */
+		__rxf_vlan_filter_set(rxf, rxf->vlan_filter_status);
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_RXF_PROMISCUOUS_SET_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	/* 7. Clear active promisc mode; move it to pending enable */
+	if (rxf->rxmode_active & BNA_RXMODE_PROMISC) {
+		/* move promisc configuration from active -> pending */
+		promisc_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_PROMISC;
+
+		/* Revert VLAN filter */
+		__rxf_vlan_filter_set(rxf, rxf->vlan_filter_status);
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_RXF_PROMISCUOUS_SET_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	return 0;
+}
+
+int
+rxf_clear_packet_filter_default(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+
+	/* 8. Execute pending default mode disable command */
+	if (is_default_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* move default configuration from pending -> active */
+		default_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_DEFAULT;
+		bna->rxf_default_id = BFI_MAX_RXF;
+
+		/* Revert VLAN filter */
+		__rxf_vlan_filter_set(rxf, rxf->vlan_filter_status);
+		/* Stop RxF vlan filter table redirection */
+		__rxf_default_function_config(rxf, BNA_STATUS_T_DISABLED);
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_RXF_DEFAULT_SET_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	/* 9. Clear active default mode; move it to pending enable */
+	if (rxf->rxmode_active & BNA_RXMODE_DEFAULT) {
+		/* move default configuration from active -> pending */
+		default_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_DEFAULT;
+
+		/* Revert VLAN filter */
+		__rxf_vlan_filter_set(rxf, rxf->vlan_filter_status);
+		/* Stop RxF vlan filter table redirection */
+		__rxf_default_function_config(rxf, BNA_STATUS_T_DISABLED);
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_RXF_DEFAULT_SET_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	return 0;
+}
+
+int
+rxf_clear_packet_filter_allmulti(struct bna_rxf *rxf)
+{
+	/* 10. Execute pending allmulti mode disable command */
+	if (is_allmulti_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* move allmulti configuration from pending -> active */
+		allmulti_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_ALLMULTI;
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_MAC_MCAST_FILTER_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	/* 11. Clear active allmulti mode; move it to pending enable */
+	if (rxf->rxmode_active & BNA_RXMODE_ALLMULTI) {
+		/* move allmulti configuration from active -> pending */
+		allmulti_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_ALLMULTI;
+		rxf_fltr_mbox_cmd(rxf, BFI_LL_H2I_MAC_MCAST_FILTER_REQ,
+				BNA_STATUS_T_DISABLED);
+		return 1;
+	}
+
+	return 0;
+}
+
+void
+rxf_reset_packet_filter_ucast(struct bna_rxf *rxf)
+{
+	struct list_head *qe;
+	struct bna_mac *mac;
+
+	/* 1. Move active ucast entries to pending_add_q */
+	while (!list_empty(&rxf->ucast_active_q)) {
+		bfa_q_deq(&rxf->ucast_active_q, &qe);
+		bfa_q_qe_init(qe);
+		list_add_tail(qe, &rxf->ucast_pending_add_q);
+	}
+
+	/* 2. Throw away delete pending ucast entries */
+	while (!list_empty(&rxf->ucast_pending_del_q)) {
+		bfa_q_deq(&rxf->ucast_pending_del_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		bna_ucam_mod_mac_put(&rxf->rx->bna->ucam_mod, mac);
+	}
+}
+
+void
+rxf_reset_packet_filter_promisc(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+
+	/* 6. Clear pending promisc mode disable */
+	if (is_promisc_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		promisc_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_PROMISC;
+		bna->rxf_promisc_id = BFI_MAX_RXF;
+	}
+
+	/* 7. Move promisc mode config from active -> pending */
+	if (rxf->rxmode_active & BNA_RXMODE_PROMISC) {
+		promisc_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_PROMISC;
+	}
+
+}
+
+void
+rxf_reset_packet_filter_default(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+
+	/* 8. Clear pending default mode disable */
+	if (is_default_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		default_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_DEFAULT;
+		bna->rxf_default_id = BFI_MAX_RXF;
+	}
+
+	/* 9. Move default mode config from active -> pending */
+	if (rxf->rxmode_active & BNA_RXMODE_DEFAULT) {
+		default_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_DEFAULT;
+	}
+}
+
+void
+rxf_reset_packet_filter_allmulti(struct bna_rxf *rxf)
+{
+	/* 10. Clear pending allmulti mode disable */
+	if (is_allmulti_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		allmulti_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_ALLMULTI;
+	}
+
+	/* 11. Move allmulti mode config from active -> pending */
+	if (rxf->rxmode_active & BNA_RXMODE_ALLMULTI) {
+		allmulti_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		rxf->rxmode_active &= ~BNA_RXMODE_ALLMULTI;
+	}
+}
+
+/**
+ * Should only be called by bna_rxf_mode_set.
+ * Helps deciding if h/w configuration is needed or not.
+ *  Returns:
+ *	0 = no h/w change
+ *	1 = need h/w change
+ */
+int
+rxf_promisc_enable(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+	int ret = 0;
+
+	/* There can not be any pending disable command */
+
+	/* Do nothing if pending enable or already enabled */
+	if (is_promisc_enable(rxf->rxmode_pending,
+			rxf->rxmode_pending_bitmask) ||
+			(rxf->rxmode_active & BNA_RXMODE_PROMISC)) {
+		/* Schedule enable */
+	} else {
+		/* Promisc mode should not be active in the system */
+		promisc_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		bna->rxf_promisc_id = rxf->rxf_id;
+		ret = 1;
+	}
+
+	return ret;
+}
+
+/**
+ * Should only be called by bna_rxf_mode_set.
+ * Helps deciding if h/w configuration is needed or not.
+ *  Returns:
+ *	0 = no h/w change
+ *	1 = need h/w change
+ */
+int
+rxf_promisc_disable(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+	int ret = 0;
+
+	/* There can not be any pending disable */
+
+	/* Turn off pending enable command , if any */
+	if (is_promisc_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* Promisc mode should not be active */
+		/* system promisc state should be pending */
+		promisc_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		/* Remove the promisc state from the system */
+		bna->rxf_promisc_id = BFI_MAX_RXF;
+
+		/* Schedule disable */
+	} else if (rxf->rxmode_active & BNA_RXMODE_PROMISC) {
+		/* Promisc mode should be active in the system */
+		promisc_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		ret = 1;
+
+	/* Do nothing if already disabled */
+	} else {
+	}
+
+	return ret;
+}
+
+/**
+ * Should only be called by bna_rxf_mode_set.
+ * Helps deciding if h/w configuration is needed or not.
+ *  Returns:
+ *	0 = no h/w change
+ *	1 = need h/w change
+ */
+int
+rxf_default_enable(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+	int ret = 0;
+
+	/* There can not be any pending disable command */
+
+	/* Do nothing if pending enable or already enabled */
+	if (is_default_enable(rxf->rxmode_pending,
+		rxf->rxmode_pending_bitmask) ||
+		(rxf->rxmode_active & BNA_RXMODE_DEFAULT)) {
+		/* Schedule enable */
+	} else {
+		/* Default mode should not be active in the system */
+		default_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		bna->rxf_default_id = rxf->rxf_id;
+		ret = 1;
+	}
+
+	return ret;
+}
+
+/**
+ * Should only be called by bna_rxf_mode_set.
+ * Helps deciding if h/w configuration is needed or not.
+ *  Returns:
+ *	0 = no h/w change
+ *	1 = need h/w change
+ */
+int
+rxf_default_disable(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+	int ret = 0;
+
+	/* There can not be any pending disable */
+
+	/* Turn off pending enable command , if any */
+	if (is_default_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* Promisc mode should not be active */
+		/* system default state should be pending */
+		default_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		/* Remove the default state from the system */
+		bna->rxf_default_id = BFI_MAX_RXF;
+
+	/* Schedule disable */
+	} else if (rxf->rxmode_active & BNA_RXMODE_DEFAULT) {
+		/* Default mode should be active in the system */
+		default_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		ret = 1;
+
+	/* Do nothing if already disabled */
+	} else {
+	}
+
+	return ret;
+}
+
+/**
+ * Should only be called by bna_rxf_mode_set.
+ * Helps deciding if h/w configuration is needed or not.
+ *  Returns:
+ *	0 = no h/w change
+ *	1 = need h/w change
+ */
+int
+rxf_allmulti_enable(struct bna_rxf *rxf)
+{
+	int ret = 0;
+
+	/* There can not be any pending disable command */
+
+	/* Do nothing if pending enable or already enabled */
+	if (is_allmulti_enable(rxf->rxmode_pending,
+			rxf->rxmode_pending_bitmask) ||
+			(rxf->rxmode_active & BNA_RXMODE_ALLMULTI)) {
+		/* Schedule enable */
+	} else {
+		allmulti_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		ret = 1;
+	}
+
+	return ret;
+}
+
+/**
+ * Should only be called by bna_rxf_mode_set.
+ * Helps deciding if h/w configuration is needed or not.
+ *  Returns:
+ *	0 = no h/w change
+ *	1 = need h/w change
+ */
+int
+rxf_allmulti_disable(struct bna_rxf *rxf)
+{
+	int ret = 0;
+
+	/* There can not be any pending disable */
+
+	/* Turn off pending enable command , if any */
+	if (is_allmulti_enable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask)) {
+		/* Allmulti mode should not be active */
+		allmulti_inactive(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+
+	/* Schedule disable */
+	} else if (rxf->rxmode_active & BNA_RXMODE_ALLMULTI) {
+		allmulti_disable(rxf->rxmode_pending,
+				rxf->rxmode_pending_bitmask);
+		ret = 1;
+	}
+
+	return ret;
+}
+
+/* RxF <- bnad */
+void
+bna_rx_mcast_delall(struct bna_rx *rx,
+		    void (*cbfn)(struct bnad *, struct bna_rx *,
+				 enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	struct list_head *qe;
+	struct bna_mac *mac;
+	int need_hw_config = 0;
+
+	/* Purge all entries from pending_add_q */
+	while (!list_empty(&rxf->mcast_pending_add_q)) {
+		bfa_q_deq(&rxf->mcast_pending_add_q, &qe);
+		mac = (struct bna_mac *)qe;
+		bfa_q_qe_init(&mac->qe);
+		bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod, mac);
+	}
+
+	/* Schedule all entries in active_q for deletion */
+	while (!list_empty(&rxf->mcast_active_q)) {
+		bfa_q_deq(&rxf->mcast_active_q, &qe);
+		mac = (struct bna_mac *)qe;
+		bfa_q_qe_init(&mac->qe);
+		list_add_tail(&mac->qe, &rxf->mcast_pending_del_q);
+		need_hw_config = 1;
+	}
+
+	if (need_hw_config) {
+		rxf->cam_fltr_cbfn = cbfn;
+		rxf->cam_fltr_cbarg = rx->bna->bnad;
+		bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+		return;
+	}
+
+	if (cbfn)
+		(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+}
+
+/* RxF <- Rx */
+void
+bna_rx_receive_resume(struct bna_rx *rx,
+		      void (*cbfn)(struct bnad *, struct bna_rx *,
+				   enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+
+	if (rxf->rxf_oper_state == BNA_RXF_OPER_STATE_PAUSED) {
+		rxf->oper_state_cbfn = cbfn;
+		rxf->oper_state_cbarg = rx->bna->bnad;
+		bfa_fsm_send_event(rxf, RXF_E_RESUME);
+	} else if (cbfn)
+		(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+}
+
+void
+bna_rx_receive_pause(struct bna_rx *rx,
+		     void (*cbfn)(struct bnad *, struct bna_rx *,
+				  enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+
+	if (rxf->rxf_oper_state == BNA_RXF_OPER_STATE_RUNNING) {
+		rxf->oper_state_cbfn = cbfn;
+		rxf->oper_state_cbarg = rx->bna->bnad;
+		bfa_fsm_send_event(rxf, RXF_E_PAUSE);
+	} else if (cbfn)
+		(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+}
+
+/* RxF <- bnad */
+enum bna_cb_status
+bna_rx_ucast_add(struct bna_rx *rx, u8 *addr,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	struct list_head *qe;
+	struct bna_mac *mac;
+
+	/* Check if already added */
+	list_for_each(qe, &rxf->ucast_active_q) {
+		mac = (struct bna_mac *)qe;
+		if (BNA_MAC_IS_EQUAL(mac->addr, addr)) {
+			if (cbfn)
+				(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+			return BNA_CB_SUCCESS;
+		}
+	}
+
+	/* Check if pending addition */
+	list_for_each(qe, &rxf->ucast_pending_add_q) {
+		mac = (struct bna_mac *)qe;
+		if (BNA_MAC_IS_EQUAL(mac->addr, addr)) {
+			if (cbfn)
+				(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+			return BNA_CB_SUCCESS;
+		}
+	}
+
+	mac = bna_ucam_mod_mac_get(&rxf->rx->bna->ucam_mod);
+	if (mac == NULL)
+		return BNA_CB_UCAST_CAM_FULL;
+	bfa_q_qe_init(&mac->qe);
+	memcpy(mac->addr, addr, ETH_ALEN);
+	list_add_tail(&mac->qe, &rxf->ucast_pending_add_q);
+
+	rxf->cam_fltr_cbfn = cbfn;
+	rxf->cam_fltr_cbarg = rx->bna->bnad;
+
+	bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+
+	return BNA_CB_SUCCESS;
+}
+
+/* RxF <- bnad */
+enum bna_cb_status
+bna_rx_ucast_del(struct bna_rx *rx, u8 *addr,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	struct list_head *qe;
+	struct bna_mac *mac;
+
+	list_for_each(qe, &rxf->ucast_pending_add_q) {
+		mac = (struct bna_mac *)qe;
+		if (BNA_MAC_IS_EQUAL(mac->addr, addr)) {
+			list_del(qe);
+			bfa_q_qe_init(qe);
+			bna_ucam_mod_mac_put(&rxf->rx->bna->ucam_mod, mac);
+			if (cbfn)
+				(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+			return BNA_CB_SUCCESS;
+		}
+	}
+
+	list_for_each(qe, &rxf->ucast_active_q) {
+		mac = (struct bna_mac *)qe;
+		if (BNA_MAC_IS_EQUAL(mac->addr, addr)) {
+			list_del(qe);
+			bfa_q_qe_init(qe);
+			list_add_tail(qe, &rxf->ucast_pending_del_q);
+			rxf->cam_fltr_cbfn = cbfn;
+			rxf->cam_fltr_cbarg = rx->bna->bnad;
+			bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+			return BNA_CB_SUCCESS;
+		}
+	}
+
+	return BNA_CB_INVALID_MAC;
+}
+
+/* RxF <- bnad */
+enum bna_cb_status
+bna_rx_mode_set(struct bna_rx *rx, enum bna_rxmode new_mode,
+		enum bna_rxmode bitmask,
+		void (*cbfn)(struct bnad *, struct bna_rx *,
+			     enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	int need_hw_config = 0;
+
+	/* Error checks */
+
+	if (is_promisc_enable(new_mode, bitmask)) {
+		/* If promisc mode is already enabled elsewhere in the system */
+		if ((rx->bna->rxf_promisc_id != BFI_MAX_RXF) &&
+			(rx->bna->rxf_promisc_id != rxf->rxf_id))
+			goto err_return;
+
+		/* If default mode is already enabled in the system */
+		if (rx->bna->rxf_default_id != BFI_MAX_RXF)
+			goto err_return;
+
+		/* Trying to enable promiscuous and default mode together */
+		if (is_default_enable(new_mode, bitmask))
+			goto err_return;
+	}
+
+	if (is_default_enable(new_mode, bitmask)) {
+		/* If default mode is already enabled elsewhere in the system */
+		if ((rx->bna->rxf_default_id != BFI_MAX_RXF) &&
+			(rx->bna->rxf_default_id != rxf->rxf_id)) {
+				goto err_return;
+		}
+
+		/* If promiscuous mode is already enabled in the system */
+		if (rx->bna->rxf_promisc_id != BFI_MAX_RXF)
+			goto err_return;
+	}
+
+	/* Process the commands */
+
+	if (is_promisc_enable(new_mode, bitmask)) {
+		if (rxf_promisc_enable(rxf))
+			need_hw_config = 1;
+	} else if (is_promisc_disable(new_mode, bitmask)) {
+		if (rxf_promisc_disable(rxf))
+			need_hw_config = 1;
+	}
+
+	if (is_default_enable(new_mode, bitmask)) {
+		if (rxf_default_enable(rxf))
+			need_hw_config = 1;
+	} else if (is_default_disable(new_mode, bitmask)) {
+		if (rxf_default_disable(rxf))
+			need_hw_config = 1;
+	}
+
+	if (is_allmulti_enable(new_mode, bitmask)) {
+		if (rxf_allmulti_enable(rxf))
+			need_hw_config = 1;
+	} else if (is_allmulti_disable(new_mode, bitmask)) {
+		if (rxf_allmulti_disable(rxf))
+			need_hw_config = 1;
+	}
+
+	/* Trigger h/w if needed */
+
+	if (need_hw_config) {
+		rxf->cam_fltr_cbfn = cbfn;
+		rxf->cam_fltr_cbarg = rx->bna->bnad;
+		bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+	} else if (cbfn)
+		(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+
+	return BNA_CB_SUCCESS;
+
+err_return:
+	return BNA_CB_FAIL;
+}
+
+/* RxF <- bnad */
+void
+bna_rx_rss_enable(struct bna_rx *rx)
+{
+	struct bna_rxf *rxf = &rx->rxf;
+
+	rxf->rxf_flags |= BNA_RXF_FL_RSS_CONFIG_PENDING;
+	rxf->rss_status = BNA_STATUS_T_ENABLED;
+	bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+}
+
+/* RxF <- bnad */
+void
+bna_rx_rss_disable(struct bna_rx *rx)
+{
+	struct bna_rxf *rxf = &rx->rxf;
+
+	rxf->rxf_flags |= BNA_RXF_FL_RSS_CONFIG_PENDING;
+	rxf->rss_status = BNA_STATUS_T_DISABLED;
+	bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+}
+
+/* RxF <- bnad */
+void
+bna_rx_rss_reconfig(struct bna_rx *rx, struct bna_rxf_rss *rss_config)
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	rxf->rxf_flags |= BNA_RXF_FL_RSS_CONFIG_PENDING;
+	rxf->rss_status = BNA_STATUS_T_ENABLED;
+	rxf->rss_cfg = *rss_config;
+	bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+}
+
+void
+/* RxF <- bnad */
+bna_rx_vlanfilter_enable(struct bna_rx *rx)
+{
+	struct bna_rxf *rxf = &rx->rxf;
+
+	if (rxf->vlan_filter_status == BNA_STATUS_T_DISABLED) {
+		rxf->rxf_flags |= BNA_RXF_FL_VLAN_CONFIG_PENDING;
+		rxf->vlan_filter_status = BNA_STATUS_T_ENABLED;
+		bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+	}
+}
+
+/* RxF <- bnad */
+void
+bna_rx_vlanfilter_disable(struct bna_rx *rx)
+{
+	struct bna_rxf *rxf = &rx->rxf;
+
+	if (rxf->vlan_filter_status == BNA_STATUS_T_ENABLED) {
+		rxf->rxf_flags |= BNA_RXF_FL_VLAN_CONFIG_PENDING;
+		rxf->vlan_filter_status = BNA_STATUS_T_DISABLED;
+		bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+	}
+}
+
+/* Rx */
+
+struct bna_rxp *
+bna_rx_get_rxp(struct bna_rx *rx, int vector)
+{
+	struct bna_rxp *rxp;
+	struct list_head *qe;
+
+	list_for_each(qe, &rx->rxp_q) {
+		rxp = (struct bna_rxp *)qe;
+		if (rxp->vector == vector)
+			return rxp;
+	}
+	return NULL;
+}
+
+/*
+ * bna_rx_rss_rit_set()
+ * Sets the Q ids for the specified msi-x vectors in the RIT.
+ * Maximum rit size supported is 64, which should be the max size of the
+ * vectors array.
+ */
+
+void
+bna_rx_rss_rit_set(struct bna_rx *rx, unsigned int *vectors, int nvectors)
+{
+	int i;
+	struct bna_rxp *rxp;
+	struct bna_rxq *q0 = NULL, *q1 = NULL;
+	struct bna *bna;
+	struct bna_rxf *rxf;
+
+	/* Build the RIT contents for this RX */
+	bna = rx->bna;
+
+	rxf = &rx->rxf;
+	for (i = 0; i < nvectors; i++) {
+		rxp = bna_rx_get_rxp(rx, vectors[i]);
+
+		GET_RXQS(rxp, q0, q1);
+		rxf->rit_segment->rit[i].large_rxq_id = q0->rxq_id;
+		rxf->rit_segment->rit[i].small_rxq_id = (q1 ? q1->rxq_id : 0);
+	}
+
+	rxf->rit_segment->rit_size = nvectors;
+
+	/* Subsequent call to enable/reconfig RSS will update the RIT in h/w */
+}
+
+/* Rx <- bnad */
+void
+bna_rx_coalescing_timeo_set(struct bna_rx *rx, int coalescing_timeo)
+{
+	struct bna_rxp *rxp;
+	struct list_head *qe;
+
+	list_for_each(qe, &rx->rxp_q) {
+		rxp = (struct bna_rxp *)qe;
+		rxp->cq.ccb->rx_coalescing_timeo = coalescing_timeo;
+		bna_ib_coalescing_timeo_set(rxp->cq.ib, coalescing_timeo);
+	}
+}
+
+/* Rx <- bnad */
+void
+bna_rx_dim_reconfig(struct bna *bna, u32 vector[][BNA_BIAS_T_MAX])
+{
+	int i, j;
+
+	for (i = 0; i < BNA_LOAD_T_MAX; i++)
+		for (j = 0; j < BNA_BIAS_T_MAX; j++)
+			bna->rx_mod.dim_vector[i][j] = vector[i][j];
+}
+
+/* Rx <- bnad */
+void
+bna_rx_dim_update(struct bna_ccb *ccb)
+{
+	struct bna *bna = ccb->cq->rx->bna;
+	u32 load, bias;
+	u32 pkt_rt, small_rt, large_rt;
+	u8 coalescing_timeo;
+
+	if ((ccb->pkt_rate.small_pkt_cnt == 0) &&
+		(ccb->pkt_rate.large_pkt_cnt == 0))
+		return;
+
+	/* Arrive at preconfigured coalescing timeo value based on pkt rate */
+
+	small_rt = ccb->pkt_rate.small_pkt_cnt;
+	large_rt = ccb->pkt_rate.large_pkt_cnt;
+
+	pkt_rt = small_rt + large_rt;
+
+	if (pkt_rt < BNA_PKT_RATE_10K)
+		load = BNA_LOAD_T_LOW_4;
+	else if (pkt_rt < BNA_PKT_RATE_20K)
+		load = BNA_LOAD_T_LOW_3;
+	else if (pkt_rt < BNA_PKT_RATE_30K)
+		load = BNA_LOAD_T_LOW_2;
+	else if (pkt_rt < BNA_PKT_RATE_40K)
+		load = BNA_LOAD_T_LOW_1;
+	else if (pkt_rt < BNA_PKT_RATE_50K)
+		load = BNA_LOAD_T_HIGH_1;
+	else if (pkt_rt < BNA_PKT_RATE_60K)
+		load = BNA_LOAD_T_HIGH_2;
+	else if (pkt_rt < BNA_PKT_RATE_80K)
+		load = BNA_LOAD_T_HIGH_3;
+	else
+		load = BNA_LOAD_T_HIGH_4;
+
+	if (small_rt > (large_rt << 1))
+		bias = 0;
+	else
+		bias = 1;
+
+	ccb->pkt_rate.small_pkt_cnt = 0;
+	ccb->pkt_rate.large_pkt_cnt = 0;
+
+	coalescing_timeo = bna->rx_mod.dim_vector[load][bias];
+	ccb->rx_coalescing_timeo = coalescing_timeo;
+
+	/* Set it to IB */
+	bna_ib_coalescing_timeo_set(ccb->cq->ib, coalescing_timeo);
+}
+
+/* Tx */
+/* TX <- bnad */
+enum bna_cb_status
+bna_tx_prio_set(struct bna_tx *tx, int prio,
+		void (*cbfn)(struct bnad *, struct bna_tx *,
+			     enum bna_cb_status))
+{
+	if (tx->flags & BNA_TX_F_PRIO_LOCK)
+		return BNA_CB_FAIL;
+	else {
+		tx->prio_change_cbfn = cbfn;
+		bna_tx_prio_changed(tx, prio);
+	}
+
+	return BNA_CB_SUCCESS;
+}
+
+/* TX <- bnad */
+void
+bna_tx_coalescing_timeo_set(struct bna_tx *tx, int coalescing_timeo)
+{
+	struct bna_txq *txq;
+	struct list_head *qe;
+
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		bna_ib_coalescing_timeo_set(txq->ib, coalescing_timeo);
+	}
+}
+
+/*
+ * Private data
+ */
+
+struct bna_ritseg_pool_cfg {
+	u32	pool_size;
+	u32	pool_entry_size;
+};
+init_ritseg_pool(ritseg_pool_cfg);
+
+/*
+ * Private functions
+ */
+static void
+bna_ucam_mod_init(struct bna_ucam_mod *ucam_mod, struct bna *bna,
+		  struct bna_res_info *res_info)
+{
+	int i;
+
+	ucam_mod->ucmac = (struct bna_mac *)
+		res_info[BNA_RES_MEM_T_UCMAC_ARRAY].res_u.mem_info.mdl[0].kva;
+
+	INIT_LIST_HEAD(&ucam_mod->free_q);
+	for (i = 0; i < BFI_MAX_UCMAC; i++) {
+		bfa_q_qe_init(&ucam_mod->ucmac[i].qe);
+		list_add_tail(&ucam_mod->ucmac[i].qe, &ucam_mod->free_q);
+	}
+
+	ucam_mod->bna = bna;
+}
+
+static void
+bna_ucam_mod_uninit(struct bna_ucam_mod *ucam_mod)
+{
+	struct list_head *qe;
+	int i = 0;
+
+	list_for_each(qe, &ucam_mod->free_q)
+		i++;
+
+	ucam_mod->bna = NULL;
+}
+
+static void
+bna_mcam_mod_init(struct bna_mcam_mod *mcam_mod, struct bna *bna,
+		  struct bna_res_info *res_info)
+{
+	int i;
+
+	mcam_mod->mcmac = (struct bna_mac *)
+		res_info[BNA_RES_MEM_T_MCMAC_ARRAY].res_u.mem_info.mdl[0].kva;
+
+	INIT_LIST_HEAD(&mcam_mod->free_q);
+	for (i = 0; i < BFI_MAX_MCMAC; i++) {
+		bfa_q_qe_init(&mcam_mod->mcmac[i].qe);
+		list_add_tail(&mcam_mod->mcmac[i].qe, &mcam_mod->free_q);
+	}
+
+	mcam_mod->bna = bna;
+}
+
+static void
+bna_mcam_mod_uninit(struct bna_mcam_mod *mcam_mod)
+{
+	struct list_head *qe;
+	int i = 0;
+
+	list_for_each(qe, &mcam_mod->free_q)
+		i++;
+
+	mcam_mod->bna = NULL;
+}
+
+static void
+bna_rit_mod_init(struct bna_rit_mod *rit_mod,
+		struct bna_res_info *res_info)
+{
+	int i;
+	int j;
+	int count;
+	int offset;
+
+	rit_mod->rit = (struct bna_rit_entry *)
+		res_info[BNA_RES_MEM_T_RIT_ENTRY].res_u.mem_info.mdl[0].kva;
+	rit_mod->rit_segment = (struct bna_rit_segment *)
+		res_info[BNA_RES_MEM_T_RIT_SEGMENT].res_u.mem_info.mdl[0].kva;
+
+	count = 0;
+	offset = 0;
+	for (i = 0; i < BFI_RIT_SEG_TOTAL_POOLS; i++) {
+		INIT_LIST_HEAD(&rit_mod->rit_seg_pool[i]);
+		for (j = 0; j < ritseg_pool_cfg[i].pool_size; j++) {
+			bfa_q_qe_init(&rit_mod->rit_segment[count].qe);
+			rit_mod->rit_segment[count].max_rit_size =
+					ritseg_pool_cfg[i].pool_entry_size;
+			rit_mod->rit_segment[count].rit_offset = offset;
+			rit_mod->rit_segment[count].rit =
+					&rit_mod->rit[offset];
+			list_add_tail(&rit_mod->rit_segment[count].qe,
+				&rit_mod->rit_seg_pool[i]);
+			count++;
+			offset += ritseg_pool_cfg[i].pool_entry_size;
+		}
+	}
+}
+
+static void
+bna_rit_mod_uninit(struct bna_rit_mod *rit_mod)
+{
+	struct bna_rit_segment *rit_segment;
+	struct list_head *qe;
+	int i;
+	int j;
+
+	for (i = 0; i < BFI_RIT_SEG_TOTAL_POOLS; i++) {
+		j = 0;
+		list_for_each(qe, &rit_mod->rit_seg_pool[i]) {
+			rit_segment = (struct bna_rit_segment *)qe;
+			j++;
+		}
+	}
+}
+
+/*
+ * Public functions
+ */
+
+/* Called during probe(), before calling bna_init() */
+void
+bna_res_req(struct bna_res_info *res_info)
+{
+	bna_adv_res_req(res_info);
+
+	/* DMA memory for retrieving IOC attributes */
+	res_info[BNA_RES_MEM_T_ATTR].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_ATTR].res_u.mem_info.mem_type = BNA_MEM_T_DMA;
+	res_info[BNA_RES_MEM_T_ATTR].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_ATTR].res_u.mem_info.len =
+				ALIGN(bfa_ioc_meminfo(), PAGE_SIZE);
+
+	/* DMA memory for index segment of an IB */
+	res_info[BNA_RES_MEM_T_IBIDX].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_IBIDX].res_u.mem_info.mem_type = BNA_MEM_T_DMA;
+	res_info[BNA_RES_MEM_T_IBIDX].res_u.mem_info.len =
+				BFI_IBIDX_SIZE * BFI_IBIDX_MAX_SEGSIZE;
+	res_info[BNA_RES_MEM_T_IBIDX].res_u.mem_info.num = BFI_MAX_IB;
+
+	/* Virtual memory for IB objects - stored by IB module */
+	res_info[BNA_RES_MEM_T_IB_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_IB_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_IB_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_IB_ARRAY].res_u.mem_info.len =
+				BFI_MAX_IB * sizeof(struct bna_ib);
+
+	/* Virtual memory for intr objects - stored by IB module */
+	res_info[BNA_RES_MEM_T_INTR_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_INTR_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_INTR_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_INTR_ARRAY].res_u.mem_info.len =
+				BFI_MAX_IB * sizeof(struct bna_intr);
+
+	/* Virtual memory for idx_seg objects - stored by IB module */
+	res_info[BNA_RES_MEM_T_IDXSEG_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_IDXSEG_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_IDXSEG_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_IDXSEG_ARRAY].res_u.mem_info.len =
+			BFI_IBIDX_TOTAL_SEGS * sizeof(struct bna_ibidx_seg);
+
+	/* Virtual memory for Tx objects - stored by Tx module */
+	res_info[BNA_RES_MEM_T_TX_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_TX_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_TX_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_TX_ARRAY].res_u.mem_info.len =
+			BFI_MAX_TXQ * sizeof(struct bna_tx);
+
+	/* Virtual memory for TxQ - stored by Tx module */
+	res_info[BNA_RES_MEM_T_TXQ_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_TXQ_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_TXQ_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_TXQ_ARRAY].res_u.mem_info.len =
+			BFI_MAX_TXQ * sizeof(struct bna_txq);
+
+	/* Virtual memory for Rx objects - stored by Rx module */
+	res_info[BNA_RES_MEM_T_RX_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_RX_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_RX_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_RX_ARRAY].res_u.mem_info.len =
+			BFI_MAX_RXQ * sizeof(struct bna_rx);
+
+	/* Virtual memory for RxPath - stored by Rx module */
+	res_info[BNA_RES_MEM_T_RXP_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_RXP_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_RXP_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_RXP_ARRAY].res_u.mem_info.len =
+			BFI_MAX_RXQ * sizeof(struct bna_rxp);
+
+	/* Virtual memory for RxQ - stored by Rx module */
+	res_info[BNA_RES_MEM_T_RXQ_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_RXQ_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_RXQ_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_RXQ_ARRAY].res_u.mem_info.len =
+			BFI_MAX_RXQ * sizeof(struct bna_rxq);
+
+	/* Virtual memory for Unicast MAC address - stored by ucam module */
+	res_info[BNA_RES_MEM_T_UCMAC_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_UCMAC_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_UCMAC_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_UCMAC_ARRAY].res_u.mem_info.len =
+			BFI_MAX_UCMAC * sizeof(struct bna_mac);
+
+	/* Virtual memory for Multicast MAC address - stored by mcam module */
+	res_info[BNA_RES_MEM_T_MCMAC_ARRAY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_MCMAC_ARRAY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_MCMAC_ARRAY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_MCMAC_ARRAY].res_u.mem_info.len =
+			BFI_MAX_MCMAC * sizeof(struct bna_mac);
+
+	/* Virtual memory for RIT entries */
+	res_info[BNA_RES_MEM_T_RIT_ENTRY].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_RIT_ENTRY].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_RIT_ENTRY].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_RIT_ENTRY].res_u.mem_info.len =
+			BFI_MAX_RIT_SIZE * sizeof(struct bna_rit_entry);
+
+	/* Virtual memory for RIT segment table */
+	res_info[BNA_RES_MEM_T_RIT_SEGMENT].res_type = BNA_RES_T_MEM;
+	res_info[BNA_RES_MEM_T_RIT_SEGMENT].res_u.mem_info.mem_type =
+								BNA_MEM_T_KVA;
+	res_info[BNA_RES_MEM_T_RIT_SEGMENT].res_u.mem_info.num = 1;
+	res_info[BNA_RES_MEM_T_RIT_SEGMENT].res_u.mem_info.len =
+			BFI_RIT_TOTAL_SEGS * sizeof(struct bna_rit_segment);
+
+	/* Interrupt resource for mailbox interrupt */
+	res_info[BNA_RES_INTR_T_MBOX].res_type = BNA_RES_T_INTR;
+	res_info[BNA_RES_INTR_T_MBOX].res_u.intr_info.intr_type =
+							BNA_INTR_T_MSIX;
+	res_info[BNA_RES_INTR_T_MBOX].res_u.intr_info.num = 1;
+}
+
+/* Called during probe() */
+void
+bna_init(struct bna *bna, struct bnad *bnad, struct bfa_pcidev *pcidev,
+		struct bna_res_info *res_info)
+{
+	bna->bnad = bnad;
+	bna->pcidev = *pcidev;
+
+	bna->stats.hw_stats = (struct bfi_ll_stats *)
+		res_info[BNA_RES_MEM_T_STATS].res_u.mem_info.mdl[0].kva;
+	bna->hw_stats_dma.msb =
+		res_info[BNA_RES_MEM_T_STATS].res_u.mem_info.mdl[0].dma.msb;
+	bna->hw_stats_dma.lsb =
+		res_info[BNA_RES_MEM_T_STATS].res_u.mem_info.mdl[0].dma.lsb;
+	bna->stats.sw_stats = (struct bna_sw_stats *)
+		res_info[BNA_RES_MEM_T_SWSTATS].res_u.mem_info.mdl[0].kva;
+
+	bna->regs.page_addr = bna->pcidev.pci_bar_kva +
+				reg_offset[bna->pcidev.pci_func].page_addr;
+	bna->regs.fn_int_status = bna->pcidev.pci_bar_kva +
+				reg_offset[bna->pcidev.pci_func].fn_int_status;
+	bna->regs.fn_int_mask = bna->pcidev.pci_bar_kva +
+				reg_offset[bna->pcidev.pci_func].fn_int_mask;
+
+	if (bna->pcidev.pci_func < 3)
+		bna->port_num = 0;
+	else
+		bna->port_num = 1;
+
+	/* Also initializes diag, cee, sfp, phy_port and mbox_mod */
+	bna_device_init(&bna->device, bna, res_info);
+
+	bna_port_init(&bna->port, bna);
+
+	bna_tx_mod_init(&bna->tx_mod, bna, res_info);
+
+	bna_rx_mod_init(&bna->rx_mod, bna, res_info);
+
+	bna_ib_mod_init(&bna->ib_mod, bna, res_info);
+
+	bna_rit_mod_init(&bna->rit_mod, res_info);
+
+	bna_ucam_mod_init(&bna->ucam_mod, bna, res_info);
+
+	bna_mcam_mod_init(&bna->mcam_mod, bna, res_info);
+
+	bna->rxf_default_id = BFI_MAX_RXF;
+	bna->rxf_promisc_id = BFI_MAX_RXF;
+
+	/* Mbox q element for posting stat request to f/w */
+	bfa_q_qe_init(&bna->mbox_qe.qe);
+}
+
+void
+bna_uninit(struct bna *bna)
+{
+	bna_mcam_mod_uninit(&bna->mcam_mod);
+
+	bna_ucam_mod_uninit(&bna->ucam_mod);
+
+	bna_rit_mod_uninit(&bna->rit_mod);
+
+	bna_ib_mod_uninit(&bna->ib_mod);
+
+	bna_rx_mod_uninit(&bna->rx_mod);
+
+	bna_tx_mod_uninit(&bna->tx_mod);
+
+	bna_port_uninit(&bna->port);
+
+	bna_device_uninit(&bna->device);
+
+	bna->bnad = NULL;
+}
+
+struct bna_mac *
+bna_ucam_mod_mac_get(struct bna_ucam_mod *ucam_mod)
+{
+	struct list_head *qe;
+
+	if (list_empty(&ucam_mod->free_q))
+		return NULL;
+
+	bfa_q_deq(&ucam_mod->free_q, &qe);
+
+	return (struct bna_mac *)qe;
+}
+
+void
+bna_ucam_mod_mac_put(struct bna_ucam_mod *ucam_mod, struct bna_mac *mac)
+{
+	list_add_tail(&mac->qe, &ucam_mod->free_q);
+}
+
+struct bna_mac *
+bna_mcam_mod_mac_get(struct bna_mcam_mod *mcam_mod)
+{
+	struct list_head *qe;
+
+	if (list_empty(&mcam_mod->free_q))
+		return NULL;
+
+	bfa_q_deq(&mcam_mod->free_q, &qe);
+
+	return (struct bna_mac *)qe;
+}
+
+void
+bna_mcam_mod_mac_put(struct bna_mcam_mod *mcam_mod, struct bna_mac *mac)
+{
+	list_add_tail(&mac->qe, &mcam_mod->free_q);
+}
+
+/**
+ * Note: This should be called in the same locking context as the call to
+ * bna_rit_mod_seg_get()
+ */
+int
+bna_rit_mod_can_satisfy(struct bna_rit_mod *rit_mod, int seg_size)
+{
+	int i;
+
+	/* Select the pool for seg_size */
+	for (i = 0; i < BFI_RIT_SEG_TOTAL_POOLS; i++) {
+		if (seg_size <= ritseg_pool_cfg[i].pool_entry_size)
+			break;
+	}
+
+	if (i == BFI_RIT_SEG_TOTAL_POOLS)
+		return 0;
+
+	if (list_empty(&rit_mod->rit_seg_pool[i]))
+		return 0;
+
+	return 1;
+}
+
+struct bna_rit_segment *
+bna_rit_mod_seg_get(struct bna_rit_mod *rit_mod, int seg_size)
+{
+	struct bna_rit_segment *seg;
+	struct list_head *qe;
+	int i;
+
+	/* Select the pool for seg_size */
+	for (i = 0; i < BFI_RIT_SEG_TOTAL_POOLS; i++) {
+		if (seg_size <= ritseg_pool_cfg[i].pool_entry_size)
+			break;
+	}
+
+	if (i == BFI_RIT_SEG_TOTAL_POOLS)
+		return NULL;
+
+	if (list_empty(&rit_mod->rit_seg_pool[i]))
+		return NULL;
+
+	bfa_q_deq(&rit_mod->rit_seg_pool[i], &qe);
+	seg = (struct bna_rit_segment *)qe;
+	bfa_q_qe_init(&seg->qe);
+	seg->rit_size = seg_size;
+
+	return seg;
+}
+
+void
+bna_rit_mod_seg_put(struct bna_rit_mod *rit_mod,
+			struct bna_rit_segment *seg)
+{
+	int i;
+
+	/* Select the pool for seg->max_rit_size */
+	for (i = 0; i < BFI_RIT_SEG_TOTAL_POOLS; i++) {
+		if (seg->max_rit_size == ritseg_pool_cfg[i].pool_entry_size)
+			break;
+	}
+
+	seg->rit_size = 0;
+	list_add_tail(&seg->qe, &rit_mod->rit_seg_pool[i]);
+}
diff --git a/drivers/net/bna/bna_hw.h b/drivers/net/bna/bna_hw.h
new file mode 100644
index 000000000000..67eb376c5c7e
--- /dev/null
+++ b/drivers/net/bna/bna_hw.h
@@ -0,0 +1,1491 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ *
+ * File for interrupt macros and functions
+ */
+
+#ifndef __BNA_HW_H__
+#define __BNA_HW_H__
+
+#include "bfi_ctreg.h"
+
+/**
+ *
+ * SW imposed limits
+ *
+ */
+
+#ifndef BNA_BIOS_BUILD
+
+#define BFI_MAX_TXQ			64
+#define BFI_MAX_RXQ			64
+#define	BFI_MAX_RXF			64
+#define BFI_MAX_IB			128
+#define	BFI_MAX_RIT_SIZE		256
+#define	BFI_RSS_RIT_SIZE		64
+#define	BFI_NONRSS_RIT_SIZE		1
+#define BFI_MAX_UCMAC			256
+#define BFI_MAX_MCMAC			512
+#define BFI_IBIDX_SIZE			4
+#define BFI_MAX_VLAN			4095
+
+/**
+ * There are 2 free IB index pools:
+ *	pool1: 120 segments of 1 index each
+ *	pool8: 1 segment of 8 indexes
+ */
+#define BFI_IBIDX_POOL1_SIZE		116
+#define	BFI_IBIDX_POOL1_ENTRY_SIZE	1
+#define BFI_IBIDX_POOL2_SIZE		2
+#define	BFI_IBIDX_POOL2_ENTRY_SIZE	2
+#define	BFI_IBIDX_POOL8_SIZE		1
+#define	BFI_IBIDX_POOL8_ENTRY_SIZE	8
+#define	BFI_IBIDX_TOTAL_POOLS		3
+#define	BFI_IBIDX_TOTAL_SEGS		119 /* (POOL1 + POOL2 + POOL8)_SIZE */
+#define	BFI_IBIDX_MAX_SEGSIZE		8
+#define init_ibidx_pool(name)						\
+static struct bna_ibidx_pool name[BFI_IBIDX_TOTAL_POOLS] =		\
+{									\
+	{ BFI_IBIDX_POOL1_SIZE, BFI_IBIDX_POOL1_ENTRY_SIZE },		\
+	{ BFI_IBIDX_POOL2_SIZE, BFI_IBIDX_POOL2_ENTRY_SIZE },		\
+	{ BFI_IBIDX_POOL8_SIZE, BFI_IBIDX_POOL8_ENTRY_SIZE }		\
+}
+
+/**
+ * There are 2 free RIT segment pools:
+ * 	Pool1: 192 segments of 1 RIT entry each
+ *	Pool2: 1 segment of 64 RIT entry
+ */
+#define BFI_RIT_SEG_POOL1_SIZE		192
+#define BFI_RIT_SEG_POOL1_ENTRY_SIZE	1
+#define BFI_RIT_SEG_POOLRSS_SIZE	1
+#define BFI_RIT_SEG_POOLRSS_ENTRY_SIZE	64
+#define BFI_RIT_SEG_TOTAL_POOLS		2
+#define BFI_RIT_TOTAL_SEGS		193 /* POOL1_SIZE + POOLRSS_SIZE */
+#define init_ritseg_pool(name)						\
+static struct bna_ritseg_pool_cfg name[BFI_RIT_SEG_TOTAL_POOLS] =	\
+{									\
+	{ BFI_RIT_SEG_POOL1_SIZE, BFI_RIT_SEG_POOL1_ENTRY_SIZE },	\
+	{ BFI_RIT_SEG_POOLRSS_SIZE, BFI_RIT_SEG_POOLRSS_ENTRY_SIZE }	\
+}
+
+#else /* BNA_BIOS_BUILD */
+
+#define BFI_MAX_TXQ			1
+#define BFI_MAX_RXQ			1
+#define	BFI_MAX_RXF			1
+#define BFI_MAX_IB			2
+#define	BFI_MAX_RIT_SIZE		2
+#define	BFI_RSS_RIT_SIZE		64
+#define	BFI_NONRSS_RIT_SIZE		1
+#define BFI_MAX_UCMAC			1
+#define BFI_MAX_MCMAC			8
+#define BFI_IBIDX_SIZE			4
+#define BFI_MAX_VLAN			4095
+/* There is one free pool: 2 segments of 1 index each */
+#define BFI_IBIDX_POOL1_SIZE		2
+#define	BFI_IBIDX_POOL1_ENTRY_SIZE	1
+#define	BFI_IBIDX_TOTAL_POOLS		1
+#define	BFI_IBIDX_TOTAL_SEGS		2 /* POOL1_SIZE */
+#define	BFI_IBIDX_MAX_SEGSIZE		1
+#define init_ibidx_pool(name)						\
+static struct bna_ibidx_pool name[BFI_IBIDX_TOTAL_POOLS] =		\
+{									\
+	{ BFI_IBIDX_POOL1_SIZE, BFI_IBIDX_POOL1_ENTRY_SIZE }		\
+}
+
+#define BFI_RIT_SEG_POOL1_SIZE		1
+#define BFI_RIT_SEG_POOL1_ENTRY_SIZE	1
+#define BFI_RIT_SEG_TOTAL_POOLS		1
+#define BFI_RIT_TOTAL_SEGS		1 /* POOL1_SIZE */
+#define init_ritseg_pool(name)						\
+static struct bna_ritseg_pool_cfg name[BFI_RIT_SEG_TOTAL_POOLS] =	\
+{									\
+	{ BFI_RIT_SEG_POOL1_SIZE, BFI_RIT_SEG_POOL1_ENTRY_SIZE }	\
+}
+
+#endif /* BNA_BIOS_BUILD */
+
+#define BFI_RSS_HASH_KEY_LEN		10
+
+#define BFI_COALESCING_TIMER_UNIT	5	/* 5us */
+#define BFI_MAX_COALESCING_TIMEO	0xFF	/* in 5us units */
+#define BFI_MAX_INTERPKT_COUNT		0xFF
+#define BFI_MAX_INTERPKT_TIMEO		0xF	/* in 0.5us units */
+#define BFI_TX_COALESCING_TIMEO		20	/* 20 * 5 = 100us */
+#define BFI_TX_INTERPKT_COUNT		32
+#define	BFI_RX_COALESCING_TIMEO		12	/* 12 * 5 = 60us */
+#define	BFI_RX_INTERPKT_COUNT		6	/* Pkt Cnt = 6 */
+#define	BFI_RX_INTERPKT_TIMEO		3	/* 3 * 0.5 = 1.5us */
+
+#define BFI_TXQ_WI_SIZE			64	/* bytes */
+#define BFI_RXQ_WI_SIZE			8	/* bytes */
+#define BFI_CQ_WI_SIZE			16	/* bytes */
+#define BFI_TX_MAX_WRR_QUOTA		0xFFF
+
+#define BFI_TX_MAX_VECTORS_PER_WI	4
+#define BFI_TX_MAX_VECTORS_PER_PKT	0xFF
+#define BFI_TX_MAX_DATA_PER_VECTOR	0xFFFF
+#define BFI_TX_MAX_DATA_PER_PKT		0xFFFFFF
+
+/* Small Q buffer size */
+#define BFI_SMALL_RXBUF_SIZE		128
+
+/* Defined separately since BFA_FLASH_DMA_BUF_SZ is in bfa_flash.c */
+#define BFI_FLASH_DMA_BUF_SZ		0x010000 /* 64K DMA */
+#define BFI_HW_STATS_SIZE		0x4000 /* 16K DMA */
+
+/**
+ *
+ * HW register offsets, macros
+ *
+ */
+
+/* DMA Block Register Host Window Start Address */
+#define DMA_BLK_REG_ADDR		0x00013000
+
+/* DMA Block Internal Registers */
+#define DMA_CTRL_REG0			(DMA_BLK_REG_ADDR + 0x000)
+#define DMA_CTRL_REG1			(DMA_BLK_REG_ADDR + 0x004)
+#define DMA_ERR_INT_STATUS		(DMA_BLK_REG_ADDR + 0x008)
+#define DMA_ERR_INT_ENABLE		(DMA_BLK_REG_ADDR + 0x00c)
+#define DMA_ERR_INT_STATUS_SET		(DMA_BLK_REG_ADDR + 0x010)
+
+/* APP Block Register Address Offset from BAR0 */
+#define APP_BLK_REG_ADDR		0x00014000
+
+/* Host Function Interrupt Mask Registers */
+#define HOSTFN0_INT_MASK		(APP_BLK_REG_ADDR + 0x004)
+#define HOSTFN1_INT_MASK		(APP_BLK_REG_ADDR + 0x104)
+#define HOSTFN2_INT_MASK		(APP_BLK_REG_ADDR + 0x304)
+#define HOSTFN3_INT_MASK		(APP_BLK_REG_ADDR + 0x404)
+
+/**
+ * Host Function PCIe Error Registers
+ * Duplicates "Correctable" & "Uncorrectable"
+ * registers in PCIe Config space.
+ */
+#define FN0_PCIE_ERR_REG		(APP_BLK_REG_ADDR + 0x014)
+#define FN1_PCIE_ERR_REG		(APP_BLK_REG_ADDR + 0x114)
+#define FN2_PCIE_ERR_REG		(APP_BLK_REG_ADDR + 0x314)
+#define FN3_PCIE_ERR_REG		(APP_BLK_REG_ADDR + 0x414)
+
+/* Host Function Error Type Status Registers */
+#define FN0_ERR_TYPE_STATUS_REG		(APP_BLK_REG_ADDR + 0x018)
+#define FN1_ERR_TYPE_STATUS_REG		(APP_BLK_REG_ADDR + 0x118)
+#define FN2_ERR_TYPE_STATUS_REG		(APP_BLK_REG_ADDR + 0x318)
+#define FN3_ERR_TYPE_STATUS_REG		(APP_BLK_REG_ADDR + 0x418)
+
+/* Host Function Error Type Mask Registers */
+#define FN0_ERR_TYPE_MSK_STATUS_REG	(APP_BLK_REG_ADDR + 0x01c)
+#define FN1_ERR_TYPE_MSK_STATUS_REG	(APP_BLK_REG_ADDR + 0x11c)
+#define FN2_ERR_TYPE_MSK_STATUS_REG	(APP_BLK_REG_ADDR + 0x31c)
+#define FN3_ERR_TYPE_MSK_STATUS_REG	(APP_BLK_REG_ADDR + 0x41c)
+
+/* Catapult Host Semaphore Status Registers (App block) */
+#define HOST_SEM_STS0_REG		(APP_BLK_REG_ADDR + 0x630)
+#define HOST_SEM_STS1_REG		(APP_BLK_REG_ADDR + 0x634)
+#define HOST_SEM_STS2_REG		(APP_BLK_REG_ADDR + 0x638)
+#define HOST_SEM_STS3_REG		(APP_BLK_REG_ADDR + 0x63c)
+#define HOST_SEM_STS4_REG		(APP_BLK_REG_ADDR + 0x640)
+#define HOST_SEM_STS5_REG		(APP_BLK_REG_ADDR + 0x644)
+#define HOST_SEM_STS6_REG		(APP_BLK_REG_ADDR + 0x648)
+#define HOST_SEM_STS7_REG		(APP_BLK_REG_ADDR + 0x64c)
+
+/* PCIe Misc Register */
+#define PCIE_MISC_REG			(APP_BLK_REG_ADDR + 0x200)
+
+/* Temp Sensor Control Registers */
+#define TEMPSENSE_CNTL_REG		(APP_BLK_REG_ADDR + 0x250)
+#define TEMPSENSE_STAT_REG		(APP_BLK_REG_ADDR + 0x254)
+
+/* APP Block local error registers */
+#define APP_LOCAL_ERR_STAT		(APP_BLK_REG_ADDR + 0x258)
+#define APP_LOCAL_ERR_MSK		(APP_BLK_REG_ADDR + 0x25c)
+
+/* PCIe Link Error registers */
+#define PCIE_LNK_ERR_STAT		(APP_BLK_REG_ADDR + 0x260)
+#define PCIE_LNK_ERR_MSK		(APP_BLK_REG_ADDR + 0x264)
+
+/**
+ * FCoE/FIP Ethertype Register
+ * 31:16 -- Chip wide value for FIP type
+ * 15:0  -- Chip wide value for FCoE type
+ */
+#define FCOE_FIP_ETH_TYPE		(APP_BLK_REG_ADDR + 0x280)
+
+/**
+ * Reserved Ethertype Register
+ * 31:16 -- Reserved
+ * 15:0  -- Other ethertype
+ */
+#define RESV_ETH_TYPE			(APP_BLK_REG_ADDR + 0x284)
+
+/**
+ * Host Command Status Registers
+ * Each set consists of 3 registers :
+ * clear, set, cmd
+ * 16 such register sets in all
+ * See catapult_spec.pdf for detailed functionality
+ * Put each type in a single macro accessed by _num ?
+ */
+#define HOST_CMDSTS0_CLR_REG		(APP_BLK_REG_ADDR + 0x500)
+#define HOST_CMDSTS0_SET_REG		(APP_BLK_REG_ADDR + 0x504)
+#define HOST_CMDSTS0_REG		(APP_BLK_REG_ADDR + 0x508)
+#define HOST_CMDSTS1_CLR_REG		(APP_BLK_REG_ADDR + 0x510)
+#define HOST_CMDSTS1_SET_REG		(APP_BLK_REG_ADDR + 0x514)
+#define HOST_CMDSTS1_REG		(APP_BLK_REG_ADDR + 0x518)
+#define HOST_CMDSTS2_CLR_REG		(APP_BLK_REG_ADDR + 0x520)
+#define HOST_CMDSTS2_SET_REG		(APP_BLK_REG_ADDR + 0x524)
+#define HOST_CMDSTS2_REG		(APP_BLK_REG_ADDR + 0x528)
+#define HOST_CMDSTS3_CLR_REG		(APP_BLK_REG_ADDR + 0x530)
+#define HOST_CMDSTS3_SET_REG		(APP_BLK_REG_ADDR + 0x534)
+#define HOST_CMDSTS3_REG		(APP_BLK_REG_ADDR + 0x538)
+#define HOST_CMDSTS4_CLR_REG		(APP_BLK_REG_ADDR + 0x540)
+#define HOST_CMDSTS4_SET_REG		(APP_BLK_REG_ADDR + 0x544)
+#define HOST_CMDSTS4_REG		(APP_BLK_REG_ADDR + 0x548)
+#define HOST_CMDSTS5_CLR_REG		(APP_BLK_REG_ADDR + 0x550)
+#define HOST_CMDSTS5_SET_REG		(APP_BLK_REG_ADDR + 0x554)
+#define HOST_CMDSTS5_REG		(APP_BLK_REG_ADDR + 0x558)
+#define HOST_CMDSTS6_CLR_REG		(APP_BLK_REG_ADDR + 0x560)
+#define HOST_CMDSTS6_SET_REG		(APP_BLK_REG_ADDR + 0x564)
+#define HOST_CMDSTS6_REG		(APP_BLK_REG_ADDR + 0x568)
+#define HOST_CMDSTS7_CLR_REG		(APP_BLK_REG_ADDR + 0x570)
+#define HOST_CMDSTS7_SET_REG		(APP_BLK_REG_ADDR + 0x574)
+#define HOST_CMDSTS7_REG		(APP_BLK_REG_ADDR + 0x578)
+#define HOST_CMDSTS8_CLR_REG		(APP_BLK_REG_ADDR + 0x580)
+#define HOST_CMDSTS8_SET_REG		(APP_BLK_REG_ADDR + 0x584)
+#define HOST_CMDSTS8_REG		(APP_BLK_REG_ADDR + 0x588)
+#define HOST_CMDSTS9_CLR_REG		(APP_BLK_REG_ADDR + 0x590)
+#define HOST_CMDSTS9_SET_REG		(APP_BLK_REG_ADDR + 0x594)
+#define HOST_CMDSTS9_REG		(APP_BLK_REG_ADDR + 0x598)
+#define HOST_CMDSTS10_CLR_REG		(APP_BLK_REG_ADDR + 0x5A0)
+#define HOST_CMDSTS10_SET_REG		(APP_BLK_REG_ADDR + 0x5A4)
+#define HOST_CMDSTS10_REG		(APP_BLK_REG_ADDR + 0x5A8)
+#define HOST_CMDSTS11_CLR_REG		(APP_BLK_REG_ADDR + 0x5B0)
+#define HOST_CMDSTS11_SET_REG		(APP_BLK_REG_ADDR + 0x5B4)
+#define HOST_CMDSTS11_REG		(APP_BLK_REG_ADDR + 0x5B8)
+#define HOST_CMDSTS12_CLR_REG		(APP_BLK_REG_ADDR + 0x5C0)
+#define HOST_CMDSTS12_SET_REG		(APP_BLK_REG_ADDR + 0x5C4)
+#define HOST_CMDSTS12_REG		(APP_BLK_REG_ADDR + 0x5C8)
+#define HOST_CMDSTS13_CLR_REG		(APP_BLK_REG_ADDR + 0x5D0)
+#define HOST_CMDSTS13_SET_REG		(APP_BLK_REG_ADDR + 0x5D4)
+#define HOST_CMDSTS13_REG		(APP_BLK_REG_ADDR + 0x5D8)
+#define HOST_CMDSTS14_CLR_REG		(APP_BLK_REG_ADDR + 0x5E0)
+#define HOST_CMDSTS14_SET_REG		(APP_BLK_REG_ADDR + 0x5E4)
+#define HOST_CMDSTS14_REG		(APP_BLK_REG_ADDR + 0x5E8)
+#define HOST_CMDSTS15_CLR_REG		(APP_BLK_REG_ADDR + 0x5F0)
+#define HOST_CMDSTS15_SET_REG		(APP_BLK_REG_ADDR + 0x5F4)
+#define HOST_CMDSTS15_REG		(APP_BLK_REG_ADDR + 0x5F8)
+
+/**
+ * LPU0 Block Register Address Offset from BAR0
+ * Range 0x18000 - 0x18033
+ */
+#define LPU0_BLK_REG_ADDR		0x00018000
+
+/**
+ * LPU0 Registers
+ * Should they be directly used from host,
+ * except for diagnostics ?
+ * CTL_REG : Control register
+ * CMD_REG : Triggers exec. of cmd. in
+ *           Mailbox memory
+ */
+#define LPU0_MBOX_CTL_REG		(LPU0_BLK_REG_ADDR + 0x000)
+#define LPU0_MBOX_CMD_REG		(LPU0_BLK_REG_ADDR + 0x004)
+#define LPU0_MBOX_LINK_0REG		(LPU0_BLK_REG_ADDR + 0x008)
+#define LPU1_MBOX_LINK_0REG		(LPU0_BLK_REG_ADDR + 0x00c)
+#define LPU0_MBOX_STATUS_0REG		(LPU0_BLK_REG_ADDR + 0x010)
+#define LPU1_MBOX_STATUS_0REG		(LPU0_BLK_REG_ADDR + 0x014)
+#define LPU0_ERR_STATUS_REG		(LPU0_BLK_REG_ADDR + 0x018)
+#define LPU0_ERR_SET_REG		(LPU0_BLK_REG_ADDR + 0x020)
+
+/**
+ * LPU1 Block Register Address Offset from BAR0
+ * Range 0x18400 - 0x18433
+ */
+#define LPU1_BLK_REG_ADDR		0x00018400
+
+/**
+ * LPU1 Registers
+ * Same as LPU0 registers above
+ */
+#define LPU1_MBOX_CTL_REG		(LPU1_BLK_REG_ADDR + 0x000)
+#define LPU1_MBOX_CMD_REG		(LPU1_BLK_REG_ADDR + 0x004)
+#define LPU0_MBOX_LINK_1REG		(LPU1_BLK_REG_ADDR + 0x008)
+#define LPU1_MBOX_LINK_1REG		(LPU1_BLK_REG_ADDR + 0x00c)
+#define LPU0_MBOX_STATUS_1REG		(LPU1_BLK_REG_ADDR + 0x010)
+#define LPU1_MBOX_STATUS_1REG		(LPU1_BLK_REG_ADDR + 0x014)
+#define LPU1_ERR_STATUS_REG		(LPU1_BLK_REG_ADDR + 0x018)
+#define LPU1_ERR_SET_REG		(LPU1_BLK_REG_ADDR + 0x020)
+
+/**
+ * PSS Block Register Address Offset from BAR0
+ * Range 0x18800 - 0x188DB
+ */
+#define PSS_BLK_REG_ADDR		0x00018800
+
+/**
+ * PSS Registers
+ * For details, see catapult_spec.pdf
+ * ERR_STATUS_REG : Indicates error in PSS module
+ * RAM_ERR_STATUS_REG : Indicates RAM module that detected error
+ */
+#define ERR_STATUS_SET			(PSS_BLK_REG_ADDR + 0x018)
+#define PSS_RAM_ERR_STATUS_REG		(PSS_BLK_REG_ADDR + 0x01C)
+
+/**
+ * PSS Semaphore Lock Registers, total 16
+ * First read when unlocked returns 0,
+ * and is set to 1, atomically.
+ * Subsequent reads returns 1.
+ * To clear set the value to 0.
+ * Range : 0x20 to 0x5c
+ */
+#define PSS_SEM_LOCK_REG(_num) 		\
+	(PSS_BLK_REG_ADDR + 0x020 + ((_num) << 2))
+
+/**
+ * PSS Semaphore Status Registers,
+ * corresponding to the lock registers above
+ */
+#define PSS_SEM_STATUS_REG(_num) 		\
+	(PSS_BLK_REG_ADDR + 0x060 + ((_num) << 2))
+
+/**
+ * Catapult CPQ Registers
+ * Defines for Mailbox Registers
+ * Used to send mailbox commands to firmware from
+ * host. The data part is written to the MBox
+ * memory, registers are used to indicate that
+ * a commnad is resident in memory.
+ *
+ * Note : LPU0<->LPU1 mailboxes are not listed here
+ */
+#define CPQ_BLK_REG_ADDR		0x00019000
+
+#define HOSTFN0_LPU0_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x130)
+#define HOSTFN0_LPU1_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x134)
+#define LPU0_HOSTFN0_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x138)
+#define LPU1_HOSTFN0_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x13C)
+
+#define HOSTFN1_LPU0_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x140)
+#define HOSTFN1_LPU1_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x144)
+#define LPU0_HOSTFN1_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x148)
+#define LPU1_HOSTFN1_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x14C)
+
+#define HOSTFN2_LPU0_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x170)
+#define HOSTFN2_LPU1_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x174)
+#define LPU0_HOSTFN2_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x178)
+#define LPU1_HOSTFN2_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x17C)
+
+#define HOSTFN3_LPU0_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x180)
+#define HOSTFN3_LPU1_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x184)
+#define LPU0_HOSTFN3_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x188)
+#define LPU1_HOSTFN3_MBOX1_CMD_STAT	(CPQ_BLK_REG_ADDR + 0x18C)
+
+/* Host Function Force Parity Error Registers */
+#define HOSTFN0_LPU_FORCE_PERR		(CPQ_BLK_REG_ADDR + 0x120)
+#define HOSTFN1_LPU_FORCE_PERR		(CPQ_BLK_REG_ADDR + 0x124)
+#define HOSTFN2_LPU_FORCE_PERR		(CPQ_BLK_REG_ADDR + 0x128)
+#define HOSTFN3_LPU_FORCE_PERR		(CPQ_BLK_REG_ADDR + 0x12C)
+
+/* LL Port[0|1] Halt Mask Registers */
+#define LL_HALT_MSK_P0			(CPQ_BLK_REG_ADDR + 0x1A0)
+#define LL_HALT_MSK_P1			(CPQ_BLK_REG_ADDR + 0x1B0)
+
+/* LL Port[0|1] Error Mask Registers */
+#define LL_ERR_MSK_P0			(CPQ_BLK_REG_ADDR + 0x1D0)
+#define LL_ERR_MSK_P1			(CPQ_BLK_REG_ADDR + 0x1D4)
+
+/* EMC FLI (Flash Controller) Block Register Address Offset from BAR0 */
+#define FLI_BLK_REG_ADDR		0x0001D000
+
+/* EMC FLI Registers */
+#define FLI_CMD_REG			(FLI_BLK_REG_ADDR + 0x000)
+#define FLI_ADDR_REG			(FLI_BLK_REG_ADDR + 0x004)
+#define FLI_CTL_REG			(FLI_BLK_REG_ADDR + 0x008)
+#define FLI_WRDATA_REG			(FLI_BLK_REG_ADDR + 0x00C)
+#define FLI_RDDATA_REG			(FLI_BLK_REG_ADDR + 0x010)
+#define FLI_DEV_STATUS_REG		(FLI_BLK_REG_ADDR + 0x014)
+#define FLI_SIG_WD_REG			(FLI_BLK_REG_ADDR + 0x018)
+
+/**
+ * RO register
+ * 31:16 -- Vendor Id
+ * 15:0  -- Device Id
+ */
+#define FLI_DEV_VENDOR_REG		(FLI_BLK_REG_ADDR + 0x01C)
+#define FLI_ERR_STATUS_REG		(FLI_BLK_REG_ADDR + 0x020)
+
+/**
+ * RAD (RxAdm) Block Register Address Offset from BAR0
+ * RAD0 Range : 0x20000 - 0x203FF
+ * RAD1 Range : 0x20400 - 0x207FF
+ */
+#define RAD0_BLK_REG_ADDR		0x00020000
+#define RAD1_BLK_REG_ADDR		0x00020400
+
+/* RAD0 Registers */
+#define RAD0_CTL_REG			(RAD0_BLK_REG_ADDR + 0x000)
+#define RAD0_PE_PARM_REG		(RAD0_BLK_REG_ADDR + 0x004)
+#define RAD0_BCN_REG			(RAD0_BLK_REG_ADDR + 0x008)
+
+/* Default function ID register */
+#define RAD0_DEFAULT_REG		(RAD0_BLK_REG_ADDR + 0x00C)
+
+/* Default promiscuous ID register */
+#define RAD0_PROMISC_REG		(RAD0_BLK_REG_ADDR + 0x010)
+
+#define RAD0_BCNQ_REG			(RAD0_BLK_REG_ADDR + 0x014)
+
+/*
+ * This register selects 1 of 8 PM Q's using
+ * VLAN pri, for non-BCN packets without a VLAN tag
+ */
+#define RAD0_DEFAULTQ_REG		(RAD0_BLK_REG_ADDR + 0x018)
+
+#define RAD0_ERR_STS			(RAD0_BLK_REG_ADDR + 0x01C)
+#define RAD0_SET_ERR_STS		(RAD0_BLK_REG_ADDR + 0x020)
+#define RAD0_ERR_INT_EN			(RAD0_BLK_REG_ADDR + 0x024)
+#define RAD0_FIRST_ERR			(RAD0_BLK_REG_ADDR + 0x028)
+#define RAD0_FORCE_ERR			(RAD0_BLK_REG_ADDR + 0x02C)
+
+#define RAD0_IF_RCVD			(RAD0_BLK_REG_ADDR + 0x030)
+#define RAD0_IF_RCVD_OCTETS_HIGH	(RAD0_BLK_REG_ADDR + 0x034)
+#define RAD0_IF_RCVD_OCTETS_LOW		(RAD0_BLK_REG_ADDR + 0x038)
+#define RAD0_IF_RCVD_VLAN		(RAD0_BLK_REG_ADDR + 0x03C)
+#define RAD0_IF_RCVD_UCAST		(RAD0_BLK_REG_ADDR + 0x040)
+#define RAD0_IF_RCVD_UCAST_OCTETS_HIGH	(RAD0_BLK_REG_ADDR + 0x044)
+#define RAD0_IF_RCVD_UCAST_OCTETS_LOW   (RAD0_BLK_REG_ADDR + 0x048)
+#define RAD0_IF_RCVD_UCAST_VLAN		(RAD0_BLK_REG_ADDR + 0x04C)
+#define RAD0_IF_RCVD_MCAST		(RAD0_BLK_REG_ADDR + 0x050)
+#define RAD0_IF_RCVD_MCAST_OCTETS_HIGH  (RAD0_BLK_REG_ADDR + 0x054)
+#define RAD0_IF_RCVD_MCAST_OCTETS_LOW   (RAD0_BLK_REG_ADDR + 0x058)
+#define RAD0_IF_RCVD_MCAST_VLAN		(RAD0_BLK_REG_ADDR + 0x05C)
+#define RAD0_IF_RCVD_BCAST		(RAD0_BLK_REG_ADDR + 0x060)
+#define RAD0_IF_RCVD_BCAST_OCTETS_HIGH  (RAD0_BLK_REG_ADDR + 0x064)
+#define RAD0_IF_RCVD_BCAST_OCTETS_LOW   (RAD0_BLK_REG_ADDR + 0x068)
+#define RAD0_IF_RCVD_BCAST_VLAN		(RAD0_BLK_REG_ADDR + 0x06C)
+#define RAD0_DROPPED_FRAMES		(RAD0_BLK_REG_ADDR + 0x070)
+
+#define RAD0_MAC_MAN_1H			(RAD0_BLK_REG_ADDR + 0x080)
+#define RAD0_MAC_MAN_1L			(RAD0_BLK_REG_ADDR + 0x084)
+#define RAD0_MAC_MAN_2H			(RAD0_BLK_REG_ADDR + 0x088)
+#define RAD0_MAC_MAN_2L			(RAD0_BLK_REG_ADDR + 0x08C)
+#define RAD0_MAC_MAN_3H			(RAD0_BLK_REG_ADDR + 0x090)
+#define RAD0_MAC_MAN_3L			(RAD0_BLK_REG_ADDR + 0x094)
+#define RAD0_MAC_MAN_4H			(RAD0_BLK_REG_ADDR + 0x098)
+#define RAD0_MAC_MAN_4L			(RAD0_BLK_REG_ADDR + 0x09C)
+
+#define RAD0_LAST4_IP			(RAD0_BLK_REG_ADDR + 0x100)
+
+/* RAD1 Registers */
+#define RAD1_CTL_REG			(RAD1_BLK_REG_ADDR + 0x000)
+#define RAD1_PE_PARM_REG		(RAD1_BLK_REG_ADDR + 0x004)
+#define RAD1_BCN_REG			(RAD1_BLK_REG_ADDR + 0x008)
+
+/* Default function ID register */
+#define RAD1_DEFAULT_REG		(RAD1_BLK_REG_ADDR + 0x00C)
+
+/* Promiscuous function ID register */
+#define RAD1_PROMISC_REG		(RAD1_BLK_REG_ADDR + 0x010)
+
+#define RAD1_BCNQ_REG			(RAD1_BLK_REG_ADDR + 0x014)
+
+/*
+ * This register selects 1 of 8 PM Q's using
+ * VLAN pri, for non-BCN packets without a VLAN tag
+ */
+#define RAD1_DEFAULTQ_REG		(RAD1_BLK_REG_ADDR + 0x018)
+
+#define RAD1_ERR_STS			(RAD1_BLK_REG_ADDR + 0x01C)
+#define RAD1_SET_ERR_STS		(RAD1_BLK_REG_ADDR + 0x020)
+#define RAD1_ERR_INT_EN			(RAD1_BLK_REG_ADDR + 0x024)
+
+/**
+ * TXA Block Register Address Offset from BAR0
+ * TXA0 Range : 0x21000 - 0x213FF
+ * TXA1 Range : 0x21400 - 0x217FF
+ */
+#define TXA0_BLK_REG_ADDR		0x00021000
+#define TXA1_BLK_REG_ADDR		0x00021400
+
+/* TXA Registers */
+#define TXA0_CTRL_REG			(TXA0_BLK_REG_ADDR + 0x000)
+#define TXA1_CTRL_REG			(TXA1_BLK_REG_ADDR + 0x000)
+
+/**
+ * TSO Sequence # Registers (RO)
+ * Total 8 (for 8 queues)
+ * Holds the last seq.# for TSO frames
+ * See catapult_spec.pdf for more details
+ */
+#define TXA0_TSO_TCP_SEQ_REG(_num)		\
+	(TXA0_BLK_REG_ADDR + 0x020 + ((_num) << 2))
+
+#define TXA1_TSO_TCP_SEQ_REG(_num)		\
+	(TXA1_BLK_REG_ADDR + 0x020 + ((_num) << 2))
+
+/**
+ * TSO IP ID # Registers (RO)
+ * Total 8 (for 8 queues)
+ * Holds the last IP ID for TSO frames
+ * See catapult_spec.pdf for more details
+ */
+#define TXA0_TSO_IP_INFO_REG(_num)		\
+	(TXA0_BLK_REG_ADDR + 0x040 + ((_num) << 2))
+
+#define TXA1_TSO_IP_INFO_REG(_num)		\
+	(TXA1_BLK_REG_ADDR + 0x040 + ((_num) << 2))
+
+/**
+ * RXA Block Register Address Offset from BAR0
+ * RXA0 Range : 0x21800 - 0x21BFF
+ * RXA1 Range : 0x21C00 - 0x21FFF
+ */
+#define RXA0_BLK_REG_ADDR		0x00021800
+#define RXA1_BLK_REG_ADDR		0x00021C00
+
+/* RXA Registers */
+#define RXA0_CTL_REG			(RXA0_BLK_REG_ADDR + 0x040)
+#define RXA1_CTL_REG			(RXA1_BLK_REG_ADDR + 0x040)
+
+/**
+ * PPLB Block Register Address Offset from BAR0
+ * PPLB0 Range : 0x22000 - 0x223FF
+ * PPLB1 Range : 0x22400 - 0x227FF
+ */
+#define PLB0_BLK_REG_ADDR		0x00022000
+#define PLB1_BLK_REG_ADDR		0x00022400
+
+/**
+ * PLB Registers
+ * Holds RL timer used time stamps in RLT tagged frames
+ */
+#define PLB0_ECM_TIMER_REG		(PLB0_BLK_REG_ADDR + 0x05C)
+#define PLB1_ECM_TIMER_REG		(PLB1_BLK_REG_ADDR + 0x05C)
+
+/* Controls the rate-limiter on each of the priority class */
+#define PLB0_RL_CTL			(PLB0_BLK_REG_ADDR + 0x060)
+#define PLB1_RL_CTL			(PLB1_BLK_REG_ADDR + 0x060)
+
+/**
+ * Max byte register, total 8, 0-7
+ * see catapult_spec.pdf for details
+ */
+#define PLB0_RL_MAX_BC(_num)			\
+	(PLB0_BLK_REG_ADDR + 0x064 + ((_num) << 2))
+#define PLB1_RL_MAX_BC(_num)			\
+	(PLB1_BLK_REG_ADDR + 0x064 + ((_num) << 2))
+
+/**
+ * RL Time Unit Register for priority 0-7
+ * 4 bits per priority
+ * (2^rl_unit)*1us is the actual time period
+ */
+#define PLB0_RL_TU_PRIO			(PLB0_BLK_REG_ADDR + 0x084)
+#define PLB1_RL_TU_PRIO			(PLB1_BLK_REG_ADDR + 0x084)
+
+/**
+ * RL byte count register,
+ * bytes transmitted in (rl_unit*1)us time period
+ * 1 per priority, 8 in all, 0-7.
+ */
+#define PLB0_RL_BYTE_CNT(_num)			\
+	(PLB0_BLK_REG_ADDR + 0x088 + ((_num) << 2))
+#define PLB1_RL_BYTE_CNT(_num)			\
+	(PLB1_BLK_REG_ADDR + 0x088 + ((_num) << 2))
+
+/**
+ * RL Min factor register
+ * 2 bits per priority,
+ * 4 factors possible: 1, 0.5, 0.25, 0
+ * 2'b00 - 0; 2'b01 - 0.25; 2'b10 - 0.5; 2'b11 - 1
+ */
+#define PLB0_RL_MIN_REG			(PLB0_BLK_REG_ADDR + 0x0A8)
+#define PLB1_RL_MIN_REG			(PLB1_BLK_REG_ADDR + 0x0A8)
+
+/**
+ * RL Max factor register
+ * 2 bits per priority,
+ * 4 factors possible: 1, 0.5, 0.25, 0
+ * 2'b00 - 0; 2'b01 - 0.25; 2'b10 - 0.5; 2'b11 - 1
+ */
+#define PLB0_RL_MAX_REG			(PLB0_BLK_REG_ADDR + 0x0AC)
+#define PLB1_RL_MAX_REG			(PLB1_BLK_REG_ADDR + 0x0AC)
+
+/* MAC SERDES Address Paging register */
+#define PLB0_EMS_ADD_REG		(PLB0_BLK_REG_ADDR + 0xD0)
+#define PLB1_EMS_ADD_REG		(PLB1_BLK_REG_ADDR + 0xD0)
+
+/* LL EMS Registers */
+#define LL_EMS0_BLK_REG_ADDR		0x00026800
+#define LL_EMS1_BLK_REG_ADDR		0x00026C00
+
+/**
+ * BPC Block Register Address Offset from BAR0
+ * BPC0 Range : 0x23000 - 0x233FF
+ * BPC1 Range : 0x23400 - 0x237FF
+ */
+#define BPC0_BLK_REG_ADDR		0x00023000
+#define BPC1_BLK_REG_ADDR		0x00023400
+
+/**
+ * PMM Block Register Address Offset from BAR0
+ * PMM0 Range : 0x23800 - 0x23BFF
+ * PMM1 Range : 0x23C00 - 0x23FFF
+ */
+#define PMM0_BLK_REG_ADDR		0x00023800
+#define PMM1_BLK_REG_ADDR		0x00023C00
+
+/**
+ * HQM Block Register Address Offset from BAR0
+ * HQM0 Range : 0x24000 - 0x243FF
+ * HQM1 Range : 0x24400 - 0x247FF
+ */
+#define HQM0_BLK_REG_ADDR		0x00024000
+#define HQM1_BLK_REG_ADDR		0x00024400
+
+/**
+ * HQM Control Register
+ * Controls some aspects of IB
+ * See catapult_spec.pdf for details
+ */
+#define HQM0_CTL_REG			(HQM0_BLK_REG_ADDR + 0x000)
+#define HQM1_CTL_REG			(HQM1_BLK_REG_ADDR + 0x000)
+
+/**
+ * HQM Stop Q Semaphore Registers.
+ * Only one Queue resource can be stopped at
+ * any given time. This register controls access
+ * to the single stop Q resource.
+ * See catapult_spec.pdf for details
+ */
+#define HQM0_RXQ_STOP_SEM		(HQM0_BLK_REG_ADDR + 0x028)
+#define HQM0_TXQ_STOP_SEM		(HQM0_BLK_REG_ADDR + 0x02C)
+#define HQM1_RXQ_STOP_SEM		(HQM1_BLK_REG_ADDR + 0x028)
+#define HQM1_TXQ_STOP_SEM		(HQM1_BLK_REG_ADDR + 0x02C)
+
+/**
+ * LUT Block Register Address Offset from BAR0
+ * LUT0 Range : 0x25800 - 0x25BFF
+ * LUT1 Range : 0x25C00 - 0x25FFF
+ */
+#define LUT0_BLK_REG_ADDR		0x00025800
+#define LUT1_BLK_REG_ADDR		0x00025C00
+
+/**
+ * LUT Registers
+ * See catapult_spec.pdf for details
+ */
+#define LUT0_ERR_STS			(LUT0_BLK_REG_ADDR + 0x000)
+#define LUT1_ERR_STS			(LUT1_BLK_REG_ADDR + 0x000)
+#define LUT0_SET_ERR_STS		(LUT0_BLK_REG_ADDR + 0x004)
+#define LUT1_SET_ERR_STS		(LUT1_BLK_REG_ADDR + 0x004)
+
+/**
+ * TRC (Debug/Trace) Register Offset from BAR0
+ * Range : 0x26000 -- 0x263FFF
+ */
+#define TRC_BLK_REG_ADDR		0x00026000
+
+/**
+ * TRC Registers
+ * See catapult_spec.pdf for details of each
+ */
+#define TRC_CTL_REG			(TRC_BLK_REG_ADDR + 0x000)
+#define TRC_MODS_REG			(TRC_BLK_REG_ADDR + 0x004)
+#define TRC_TRGC_REG			(TRC_BLK_REG_ADDR + 0x008)
+#define TRC_CNT1_REG			(TRC_BLK_REG_ADDR + 0x010)
+#define TRC_CNT2_REG			(TRC_BLK_REG_ADDR + 0x014)
+#define TRC_NXTS_REG			(TRC_BLK_REG_ADDR + 0x018)
+#define TRC_DIRR_REG			(TRC_BLK_REG_ADDR + 0x01C)
+
+/**
+ * TRC Trigger match filters, total 10
+ * Determines the trigger condition
+ */
+#define TRC_TRGM_REG(_num)		\
+	(TRC_BLK_REG_ADDR + 0x040 + ((_num) << 2))
+
+/**
+ * TRC Next State filters, total 10
+ * Determines the next state conditions
+ */
+#define TRC_NXTM_REG(_num)		\
+	(TRC_BLK_REG_ADDR + 0x080 + ((_num) << 2))
+
+/**
+ * TRC Store Match filters, total 10
+ * Determines the store conditions
+ */
+#define TRC_STRM_REG(_num)		\
+	(TRC_BLK_REG_ADDR + 0x0C0 + ((_num) << 2))
+
+/* DOORBELLS ACCESS */
+
+/**
+ * Catapult doorbells
+ * Each doorbell-queue set has
+ * 1 RxQ, 1 TxQ, 2 IBs in that order
+ * Size of each entry in 32 bytes, even though only 1 word
+ * is used. For Non-VM case each doorbell-q set is
+ * separated by 128 bytes, for VM case it is separated
+ * by 4K bytes
+ * Non VM case Range : 0x38000 - 0x39FFF
+ * VM case Range     : 0x100000 - 0x11FFFF
+ * The range applies to both HQMs
+ */
+#define HQM_DOORBELL_BLK_BASE_ADDR	0x00038000
+#define HQM_DOORBELL_VM_BLK_BASE_ADDR	0x00100000
+
+/* MEMORY ACCESS */
+
+/**
+ * Catapult H/W Block Memory Access Address
+ * To the host a memory space of 32K (page) is visible
+ * at a time. The address range is from 0x08000 to 0x0FFFF
+ */
+#define HW_BLK_HOST_MEM_ADDR		0x08000
+
+/**
+ * Catapult LUT Memory Access Page Numbers
+ * Range : LUT0 0xa0-0xa1
+ *         LUT1 0xa2-0xa3
+ */
+#define LUT0_MEM_BLK_BASE_PG_NUM	0x000000A0
+#define LUT1_MEM_BLK_BASE_PG_NUM	0x000000A2
+
+/**
+ * Catapult RxFn Database Memory Block Base Offset
+ *
+ * The Rx function database exists in LUT block.
+ * In PCIe space this is accessible as a 256x32
+ * bit block. Each entry in this database is 4
+ * (4 byte) words. Max. entries is 64.
+ * Address of an entry corresponding to a function
+ * = base_addr + (function_no. * 16)
+ */
+#define RX_FNDB_RAM_BASE_OFFSET		0x0000B400
+
+/**
+ * Catapult TxFn Database Memory Block Base Offset Address
+ *
+ * The Tx function database exists in LUT block.
+ * In PCIe space this is accessible as a 64x32
+ * bit block. Each entry in this database is 1
+ * (4 byte) word. Max. entries is 64.
+ * Address of an entry corresponding to a function
+ * = base_addr + (function_no. * 4)
+ */
+#define TX_FNDB_RAM_BASE_OFFSET		0x0000B800
+
+/**
+ * Catapult Unicast CAM Base Offset Address
+ *
+ * Exists in LUT memory space.
+ * Shared by both the LL & FCoE driver.
+ * Size is 256x48 bits; mapped to PCIe space
+ * 512x32 bit blocks. For each address, bits
+ * are written in the order : [47:32] and then
+ * [31:0].
+ */
+#define UCAST_CAM_BASE_OFFSET		0x0000A800
+
+/**
+ * Catapult Unicast RAM Base Offset Address
+ *
+ * Exists in LUT memory space.
+ * Shared by both the LL & FCoE driver.
+ * Size is 256x9 bits.
+ */
+#define UCAST_RAM_BASE_OFFSET		0x0000B000
+
+/**
+ * Catapult Mulicast CAM Base Offset Address
+ *
+ * Exists in LUT memory space.
+ * Shared by both the LL & FCoE driver.
+ * Size is 256x48 bits; mapped to PCIe space
+ * 512x32 bit blocks. For each address, bits
+ * are written in the order : [47:32] and then
+ * [31:0].
+ */
+#define MCAST_CAM_BASE_OFFSET		0x0000A000
+
+/**
+ * Catapult VLAN RAM Base Offset Address
+ *
+ * Exists in LUT memory space.
+ * Size is 4096x66 bits; mapped to PCIe space as
+ * 8192x32 bit blocks.
+ * All the 4K entries are within the address range
+ * 0x0000 to 0x8000, so in the first LUT page.
+ */
+#define VLAN_RAM_BASE_OFFSET		0x00000000
+
+/**
+ * Catapult Tx Stats RAM Base Offset Address
+ *
+ * Exists in LUT memory space.
+ * Size is 1024x33 bits;
+ * Each Tx function has 64 bytes of space
+ */
+#define TX_STATS_RAM_BASE_OFFSET	0x00009000
+
+/**
+ * Catapult Rx Stats RAM Base Offset Address
+ *
+ * Exists in LUT memory space.
+ * Size is 1024x33 bits;
+ * Each Rx function has 64 bytes of space
+ */
+#define RX_STATS_RAM_BASE_OFFSET	0x00008000
+
+/* Catapult RXA Memory Access Page Numbers */
+#define RXA0_MEM_BLK_BASE_PG_NUM	0x0000008C
+#define RXA1_MEM_BLK_BASE_PG_NUM	0x0000008D
+
+/**
+ * Catapult Multicast Vector Table Base Offset Address
+ *
+ * Exists in RxA memory space.
+ * Organized as 512x65 bit block.
+ * However for each entry 16 bytes allocated (power of 2)
+ * Total size 512*16 bytes.
+ * There are two logical divisions, 256 entries each :
+ * a) Entries 0x00 to 0xff (256) -- Approx. MVT
+ *    Offset 0x000 to 0xFFF
+ * b) Entries 0x100 to 0x1ff (256) -- Exact MVT
+ *    Offsets 0x1000 to 0x1FFF
+ */
+#define MCAST_APPROX_MVT_BASE_OFFSET	0x00000000
+#define MCAST_EXACT_MVT_BASE_OFFSET	0x00001000
+
+/**
+ * Catapult RxQ Translate Table (RIT) Base Offset Address
+ *
+ * Exists in RxA memory space
+ * Total no. of entries 64
+ * Each entry is 1 (4 byte) word.
+ * 31:12 -- Reserved
+ * 11:0  -- Two 6 bit RxQ Ids
+ */
+#define FUNCTION_TO_RXQ_TRANSLATE	0x00002000
+
+/* Catapult RxAdm (RAD) Memory Access Page Numbers */
+#define RAD0_MEM_BLK_BASE_PG_NUM	0x00000086
+#define RAD1_MEM_BLK_BASE_PG_NUM	0x00000087
+
+/**
+ * Catapult RSS Table Base Offset Address
+ *
+ * Exists in RAD memory space.
+ * Each entry is 352 bits, but alligned on
+ * 64 byte (512 bit) boundary. Accessed
+ * 4 byte words, the whole entry can be
+ * broken into 11 word accesses.
+ */
+#define RSS_TABLE_BASE_OFFSET		0x00000800
+
+/**
+ * Catapult CPQ Block Page Number
+ * This value is written to the page number registers
+ * to access the memory associated with the mailboxes.
+ */
+#define CPQ_BLK_PG_NUM			0x00000005
+
+/**
+ * Clarification :
+ * LL functions are 2 & 3; can HostFn0/HostFn1
+ * <-> LPU0/LPU1 memories be used ?
+ */
+/**
+ * Catapult HostFn0/HostFn1 to LPU0/LPU1 Mbox memory
+ * Per catapult_spec.pdf, the offset of the mbox
+ * memory is in the register space at an offset of 0x200
+ */
+#define CPQ_BLK_REG_MBOX_ADDR		(CPQ_BLK_REG_ADDR + 0x200)
+
+#define HOSTFN_LPU_MBOX			(CPQ_BLK_REG_MBOX_ADDR + 0x000)
+
+/* Catapult LPU0/LPU1 to HostFn0/HostFn1 Mbox memory */
+#define LPU_HOSTFN_MBOX			(CPQ_BLK_REG_MBOX_ADDR + 0x080)
+
+/**
+ * Catapult HQM Block Page Number
+ * This is written to the page number register for
+ * the appropriate function to access the memory
+ * associated with HQM
+ */
+#define HQM0_BLK_PG_NUM			0x00000096
+#define HQM1_BLK_PG_NUM			0x00000097
+
+/**
+ * Note that TxQ and RxQ entries are interlaced
+ * the HQM memory, i.e RXQ0, TXQ0, RXQ1, TXQ1.. etc.
+ */
+
+#define HQM_RXTX_Q_RAM_BASE_OFFSET	0x00004000
+
+/**
+ * CQ Memory
+ * Exists in HQM Memory space
+ * Each entry is 16 (4 byte) words of which
+ * only 12 words are used for configuration
+ * Total 64 entries per HQM memory space
+ */
+#define HQM_CQ_RAM_BASE_OFFSET		0x00006000
+
+/**
+ * Interrupt Block (IB) Memory
+ * Exists in HQM Memory space
+ * Each entry is 8 (4 byte) words of which
+ * only 5 words are used for configuration
+ * Total 128 entries per HQM memory space
+ */
+#define HQM_IB_RAM_BASE_OFFSET		0x00001000
+
+/**
+ * Index Table (IT) Memory
+ * Exists in HQM Memory space
+ * Each entry is 1 (4 byte) word which
+ * is used for configuration
+ * Total 128 entries per HQM memory space
+ */
+#define HQM_INDX_TBL_RAM_BASE_OFFSET	0x00002000
+
+/**
+ * PSS Block Memory Page Number
+ * This is written to the appropriate page number
+ * register to access the CPU memory.
+ * Also known as the PSS secondary memory (SMEM).
+ * Range : 0x180 to 0x1CF
+ * See catapult_spec.pdf for details
+ */
+#define PSS_BLK_PG_NUM			0x00000180
+
+/**
+ * Offsets of different instances of PSS SMEM
+ * 2.5M of continuous 1T memory space : 2 blocks
+ * of 1M each (32 pages each, page=32KB) and 4 smaller
+ * blocks of 128K each (4 pages each, page=32KB)
+ * PSS_LMEM_INST0 is used for firmware download
+ */
+#define PSS_LMEM_INST0			0x00000000
+#define PSS_LMEM_INST1			0x00100000
+#define PSS_LMEM_INST2			0x00200000
+#define PSS_LMEM_INST3			0x00220000
+#define PSS_LMEM_INST4			0x00240000
+#define PSS_LMEM_INST5			0x00260000
+
+#define BNA_PCI_REG_CT_ADDRSZ		(0x40000)
+
+#define BNA_GET_PAGE_NUM(_base_page, _offset)   \
+	((_base_page) + ((_offset) >> 15))
+
+#define BNA_GET_PAGE_OFFSET(_offset)    \
+	((_offset) & 0x7fff)
+
+#define BNA_GET_MEM_BASE_ADDR(_bar0, _base_offset)	\
+	((_bar0) + HW_BLK_HOST_MEM_ADDR		\
+	  + BNA_GET_PAGE_OFFSET((_base_offset)))
+
+#define BNA_GET_VLAN_MEM_ENTRY_ADDR(_bar0, _fn_id, _vlan_id)\
+	(_bar0 + (HW_BLK_HOST_MEM_ADDR)  \
+	+ (BNA_GET_PAGE_OFFSET(VLAN_RAM_BASE_OFFSET))	\
+	+ (((_fn_id) & 0x3f) << 9)	  \
+	+ (((_vlan_id) & 0xfe0) >> 3))
+
+/**
+ *
+ *  Interrupt related bits, flags and macros
+ *
+ */
+
+#define __LPU02HOST_MBOX0_STATUS_BITS 0x00100000
+#define __LPU12HOST_MBOX0_STATUS_BITS 0x00200000
+#define __LPU02HOST_MBOX1_STATUS_BITS 0x00400000
+#define __LPU12HOST_MBOX1_STATUS_BITS 0x00800000
+
+#define __LPU02HOST_MBOX0_MASK_BITS	0x00100000
+#define __LPU12HOST_MBOX0_MASK_BITS	0x00200000
+#define __LPU02HOST_MBOX1_MASK_BITS	0x00400000
+#define __LPU12HOST_MBOX1_MASK_BITS	0x00800000
+
+#define __LPU2HOST_MBOX_MASK_BITS			 \
+	(__LPU02HOST_MBOX0_MASK_BITS | __LPU02HOST_MBOX1_MASK_BITS |	\
+	  __LPU12HOST_MBOX0_MASK_BITS | __LPU12HOST_MBOX1_MASK_BITS)
+
+#define __LPU2HOST_IB_STATUS_BITS	0x0000ffff
+
+#define BNA_IS_LPU0_MBOX_INTR(_intr_status) \
+	((_intr_status) & (__LPU02HOST_MBOX0_STATUS_BITS | \
+			__LPU02HOST_MBOX1_STATUS_BITS))
+
+#define BNA_IS_LPU1_MBOX_INTR(_intr_status) \
+	((_intr_status) & (__LPU12HOST_MBOX0_STATUS_BITS | \
+		__LPU12HOST_MBOX1_STATUS_BITS))
+
+#define BNA_IS_MBOX_INTR(_intr_status)		\
+	((_intr_status) &  			\
+	(__LPU02HOST_MBOX0_STATUS_BITS |	\
+	 __LPU02HOST_MBOX1_STATUS_BITS |	\
+	 __LPU12HOST_MBOX0_STATUS_BITS |	\
+	 __LPU12HOST_MBOX1_STATUS_BITS))
+
+#define __EMC_ERROR_STATUS_BITS		0x00010000
+#define __LPU0_ERROR_STATUS_BITS	0x00020000
+#define __LPU1_ERROR_STATUS_BITS	0x00040000
+#define __PSS_ERROR_STATUS_BITS		0x00080000
+
+#define __HALT_STATUS_BITS		0x01000000
+
+#define __EMC_ERROR_MASK_BITS		0x00010000
+#define __LPU0_ERROR_MASK_BITS		0x00020000
+#define __LPU1_ERROR_MASK_BITS		0x00040000
+#define __PSS_ERROR_MASK_BITS		0x00080000
+
+#define __HALT_MASK_BITS		0x01000000
+
+#define __ERROR_MASK_BITS		\
+	(__EMC_ERROR_MASK_BITS | __LPU0_ERROR_MASK_BITS | \
+	  __LPU1_ERROR_MASK_BITS | __PSS_ERROR_MASK_BITS | \
+	  __HALT_MASK_BITS)
+
+#define BNA_IS_ERR_INTR(_intr_status)	\
+	((_intr_status) &  		\
+	(__EMC_ERROR_STATUS_BITS |  	\
+	 __LPU0_ERROR_STATUS_BITS | 	\
+	 __LPU1_ERROR_STATUS_BITS | 	\
+	 __PSS_ERROR_STATUS_BITS  | 	\
+	 __HALT_STATUS_BITS))
+
+#define BNA_IS_MBOX_ERR_INTR(_intr_status)	\
+	(BNA_IS_MBOX_INTR((_intr_status)) |	\
+	 BNA_IS_ERR_INTR((_intr_status)))
+
+#define BNA_IS_INTX_DATA_INTR(_intr_status)	\
+	((_intr_status) & __LPU2HOST_IB_STATUS_BITS)
+
+#define BNA_INTR_STATUS_MBOX_CLR(_intr_status)			\
+do {								\
+	(_intr_status) &= ~(__LPU02HOST_MBOX0_STATUS_BITS |	\
+			__LPU02HOST_MBOX1_STATUS_BITS | 	\
+			__LPU12HOST_MBOX0_STATUS_BITS | 	\
+			__LPU12HOST_MBOX1_STATUS_BITS); 	\
+} while (0)
+
+#define BNA_INTR_STATUS_ERR_CLR(_intr_status)		\
+do {							\
+	(_intr_status) &= ~(__EMC_ERROR_STATUS_BITS |	\
+		__LPU0_ERROR_STATUS_BITS |		\
+		__LPU1_ERROR_STATUS_BITS |		\
+		__PSS_ERROR_STATUS_BITS  |		\
+		__HALT_STATUS_BITS);			\
+} while (0)
+
+#define bna_intx_disable(_bna, _cur_mask)		\
+{							\
+	(_cur_mask) = readl((_bna)->regs.fn_int_mask);\
+	writel(0xffffffff, (_bna)->regs.fn_int_mask);\
+}
+
+#define bna_intx_enable(bna, new_mask) 			\
+	writel((new_mask), (bna)->regs.fn_int_mask)
+
+#define bna_mbox_intr_disable(bna)		\
+	writel((readl((bna)->regs.fn_int_mask) | \
+	     (__LPU2HOST_MBOX_MASK_BITS | __ERROR_MASK_BITS)), \
+	     (bna)->regs.fn_int_mask)
+
+#define bna_mbox_intr_enable(bna)		\
+	writel((readl((bna)->regs.fn_int_mask) & \
+	     ~(__LPU2HOST_MBOX_MASK_BITS | __ERROR_MASK_BITS)), \
+	     (bna)->regs.fn_int_mask)
+
+#define bna_intr_status_get(_bna, _status)				\
+{									\
+	(_status) = readl((_bna)->regs.fn_int_status);		\
+	if ((_status)) {						\
+		writel((_status) & ~(__LPU02HOST_MBOX0_STATUS_BITS |\
+					  __LPU02HOST_MBOX1_STATUS_BITS |\
+					  __LPU12HOST_MBOX0_STATUS_BITS |\
+					  __LPU12HOST_MBOX1_STATUS_BITS), \
+			      (_bna)->regs.fn_int_status);\
+	}								\
+}
+
+#define bna_intr_status_get_no_clr(_bna, _status)		\
+	(_status) = readl((_bna)->regs.fn_int_status)
+
+#define bna_intr_mask_get(bna, mask)		\
+	(*mask) = readl((bna)->regs.fn_int_mask)
+
+#define bna_intr_ack(bna, intr_bmap)		\
+	writel((intr_bmap), (bna)->regs.fn_int_status)
+
+#define bna_ib_intx_disable(bna, ib_id)		\
+	writel(readl((bna)->regs.fn_int_mask) | \
+	    (1 << (ib_id)), \
+	    (bna)->regs.fn_int_mask)
+
+#define bna_ib_intx_enable(bna, ib_id)		\
+	writel(readl((bna)->regs.fn_int_mask) & \
+	    ~(1 << (ib_id)), \
+	    (bna)->regs.fn_int_mask)
+
+#define bna_mbox_msix_idx_set(_device) \
+do {\
+	writel(((_device)->vector & 0x000001FF), \
+		(_device)->bna->pcidev.pci_bar_kva + \
+		reg_offset[(_device)->bna->pcidev.pci_func].msix_idx);\
+} while (0)
+
+/**
+ *
+ * TxQ, RxQ, CQ related bits, offsets, macros
+ *
+ */
+
+#define	BNA_Q_IDLE_STATE	0x00008001
+
+#define BNA_GET_DOORBELL_BASE_ADDR(_bar0)	\
+	((_bar0) + HQM_DOORBELL_BLK_BASE_ADDR)
+
+#define BNA_GET_DOORBELL_ENTRY_OFFSET(_entry)		\
+	((HQM_DOORBELL_BLK_BASE_ADDR)		\
+	+ (_entry << 7))
+
+#define BNA_DOORBELL_IB_INT_ACK(_timeout, _events) \
+		(0x80000000 | ((_timeout) << 16) | (_events))
+
+#define BNA_DOORBELL_IB_INT_DISABLE		(0x40000000)
+
+/* TxQ Entry Opcodes */
+#define BNA_TXQ_WI_SEND 		(0x402)	/* Single Frame Transmission */
+#define BNA_TXQ_WI_SEND_LSO 		(0x403)	/* Multi-Frame Transmission */
+#define BNA_TXQ_WI_EXTENSION		(0x104)	/* Extension WI */
+
+/* TxQ Entry Control Flags */
+#define BNA_TXQ_WI_CF_FCOE_CRC  	(1 << 8)
+#define BNA_TXQ_WI_CF_IPID_MODE 	(1 << 5)
+#define BNA_TXQ_WI_CF_INS_PRIO  	(1 << 4)
+#define BNA_TXQ_WI_CF_INS_VLAN  	(1 << 3)
+#define BNA_TXQ_WI_CF_UDP_CKSUM 	(1 << 2)
+#define BNA_TXQ_WI_CF_TCP_CKSUM 	(1 << 1)
+#define BNA_TXQ_WI_CF_IP_CKSUM  	(1 << 0)
+
+#define BNA_TXQ_WI_L4_HDR_N_OFFSET(_hdr_size, _offset) \
+		(((_hdr_size) << 10) | ((_offset) & 0x3FF))
+
+/*
+ * Completion Q defines
+ */
+/* CQ Entry Flags */
+#define	BNA_CQ_EF_MAC_ERROR 	(1 <<  0)
+#define	BNA_CQ_EF_FCS_ERROR 	(1 <<  1)
+#define	BNA_CQ_EF_TOO_LONG  	(1 <<  2)
+#define	BNA_CQ_EF_FC_CRC_OK 	(1 <<  3)
+
+#define	BNA_CQ_EF_RSVD1 	(1 <<  4)
+#define	BNA_CQ_EF_L4_CKSUM_OK	(1 <<  5)
+#define	BNA_CQ_EF_L3_CKSUM_OK	(1 <<  6)
+#define	BNA_CQ_EF_HDS_HEADER	(1 <<  7)
+
+#define	BNA_CQ_EF_UDP   	(1 <<  8)
+#define	BNA_CQ_EF_TCP   	(1 <<  9)
+#define	BNA_CQ_EF_IP_OPTIONS	(1 << 10)
+#define	BNA_CQ_EF_IPV6  	(1 << 11)
+
+#define	BNA_CQ_EF_IPV4  	(1 << 12)
+#define	BNA_CQ_EF_VLAN  	(1 << 13)
+#define	BNA_CQ_EF_RSS   	(1 << 14)
+#define	BNA_CQ_EF_RSVD2 	(1 << 15)
+
+#define	BNA_CQ_EF_MCAST_MATCH   (1 << 16)
+#define	BNA_CQ_EF_MCAST 	(1 << 17)
+#define BNA_CQ_EF_BCAST 	(1 << 18)
+#define	BNA_CQ_EF_REMOTE 	(1 << 19)
+
+#define	BNA_CQ_EF_LOCAL		(1 << 20)
+
+/**
+ *
+ * Data structures
+ *
+ */
+
+enum txf_flags {
+	BFI_TXF_CF_ENABLE		= 1 << 0,
+	BFI_TXF_CF_VLAN_FILTER		= 1 << 8,
+	BFI_TXF_CF_VLAN_ADMIT		= 1 << 9,
+	BFI_TXF_CF_VLAN_INSERT		= 1 << 10,
+	BFI_TXF_CF_RSVD1		= 1 << 11,
+	BFI_TXF_CF_MAC_SA_CHECK		= 1 << 12,
+	BFI_TXF_CF_VLAN_WI_BASED	= 1 << 13,
+	BFI_TXF_CF_VSWITCH_MCAST	= 1 << 14,
+	BFI_TXF_CF_VSWITCH_UCAST	= 1 << 15,
+	BFI_TXF_CF_RSVD2		= 0x7F << 1
+};
+
+enum ib_flags {
+	BFI_IB_CF_MASTER_ENABLE		= (1 << 0),
+	BFI_IB_CF_MSIX_MODE		= (1 << 1),
+	BFI_IB_CF_COALESCING_MODE	= (1 << 2),
+	BFI_IB_CF_INTER_PKT_ENABLE	= (1 << 3),
+	BFI_IB_CF_INT_ENABLE		= (1 << 4),
+	BFI_IB_CF_INTER_PKT_DMA		= (1 << 5),
+	BFI_IB_CF_ACK_PENDING		= (1 << 6),
+	BFI_IB_CF_RESERVED1		= (1 << 7)
+};
+
+enum rss_hash_type {
+	BFI_RSS_T_V4_TCP    		= (1 << 11),
+	BFI_RSS_T_V4_IP     		= (1 << 10),
+	BFI_RSS_T_V6_TCP    		= (1 <<  9),
+	BFI_RSS_T_V6_IP     		= (1 <<  8)
+};
+enum hds_header_type {
+	BNA_HDS_T_V4_TCP	= (1 << 11),
+	BNA_HDS_T_V4_UDP	= (1 << 10),
+	BNA_HDS_T_V6_TCP	= (1 << 9),
+	BNA_HDS_T_V6_UDP	= (1 << 8),
+	BNA_HDS_FORCED		= (1 << 7),
+};
+enum rxf_flags {
+	BNA_RXF_CF_SM_LG_RXQ			= (1 << 15),
+	BNA_RXF_CF_DEFAULT_VLAN			= (1 << 14),
+	BNA_RXF_CF_DEFAULT_FUNCTION_ENABLE	= (1 << 13),
+	BNA_RXF_CF_VLAN_STRIP			= (1 << 12),
+	BNA_RXF_CF_RSS_ENABLE			= (1 <<  8)
+};
+struct bna_chip_regs_offset {
+	u32 page_addr;
+	u32 fn_int_status;
+	u32 fn_int_mask;
+	u32 msix_idx;
+};
+extern const struct bna_chip_regs_offset reg_offset[];
+
+struct bna_chip_regs {
+	void __iomem *page_addr;
+	void __iomem *fn_int_status;
+	void __iomem *fn_int_mask;
+};
+
+struct bna_txq_mem {
+	u32 pg_tbl_addr_lo;
+	u32 pg_tbl_addr_hi;
+	u32 cur_q_entry_lo;
+	u32 cur_q_entry_hi;
+	u32 reserved1;
+	u32 reserved2;
+	u32 pg_cnt_n_prd_ptr;	/* 31:16->total page count */
+					/* 15:0 ->producer pointer (index?) */
+	u32 entry_n_pg_size; 	/* 31:16->entry size */
+					/* 15:0 ->page size */
+	u32 int_blk_n_cns_ptr;	/* 31:24->Int Blk Id;  */
+					/* 23:16->Int Blk Offset */
+					/* 15:0 ->consumer pointer(index?) */
+	u32 cns_ptr2_n_q_state;	/* 31:16->cons. ptr 2; 15:0-> Q state */
+	u32 nxt_qid_n_fid_n_pri;	/* 17:10->next */
+					/* QId;9:3->FID;2:0->Priority */
+	u32 wvc_n_cquota_n_rquota; /* 31:24->WI Vector Count; */
+					/* 23:12->Cfg Quota; */
+					/* 11:0 ->Run Quota */
+	u32 reserved3[4];
+};
+
+struct bna_rxq_mem {
+	u32 pg_tbl_addr_lo;
+	u32 pg_tbl_addr_hi;
+	u32 cur_q_entry_lo;
+	u32 cur_q_entry_hi;
+	u32 reserved1;
+	u32 reserved2;
+	u32 pg_cnt_n_prd_ptr;	/* 31:16->total page count */
+					/* 15:0 ->producer pointer (index?) */
+	u32 entry_n_pg_size;	/* 31:16->entry size */
+					/* 15:0 ->page size */
+	u32 sg_n_cq_n_cns_ptr;	/* 31:28->reserved; 27:24->sg count */
+					/* 23:16->CQ; */
+					/* 15:0->consumer pointer(index?) */
+	u32 buf_sz_n_q_state; 	/* 31:16->buffer size; 15:0-> Q state */
+	u32 next_qid;		/* 17:10->next QId */
+	u32 reserved3;
+	u32 reserved4[4];
+};
+
+struct bna_rxtx_q_mem {
+	struct bna_rxq_mem rxq;
+	struct bna_txq_mem txq;
+};
+
+struct bna_cq_mem {
+	u32 pg_tbl_addr_lo;
+	u32 pg_tbl_addr_hi;
+	u32 cur_q_entry_lo;
+	u32 cur_q_entry_hi;
+
+	u32 reserved1;
+	u32 reserved2;
+	u32 pg_cnt_n_prd_ptr;	/* 31:16->total page count */
+					/* 15:0 ->producer pointer (index?) */
+	u32 entry_n_pg_size;	/* 31:16->entry size */
+					/* 15:0 ->page size */
+	u32 int_blk_n_cns_ptr;	/* 31:24->Int Blk Id; */
+					/* 23:16->Int Blk Offset */
+					/* 15:0 ->consumer pointer(index?) */
+	u32 q_state;		/* 31:16->reserved; 15:0-> Q state */
+	u32 reserved3[2];
+	u32 reserved4[4];
+};
+
+struct bna_ib_blk_mem {
+	u32 host_addr_lo;
+	u32 host_addr_hi;
+	u32 clsc_n_ctrl_n_msix;	/* 31:24->coalescing; */
+					/* 23:16->coalescing cfg; */
+					/* 15:8 ->control; */
+					/* 7:0 ->msix; */
+	u32 ipkt_n_ent_n_idxof;
+	u32 ipkt_cnt_cfg_n_unacked;
+
+	u32 reserved[3];
+};
+
+struct bna_idx_tbl_mem {
+	u32 idx;	  /* !< 31:16->res;15:0->idx; */
+};
+
+struct bna_doorbell_qset {
+	u32 rxq[0x20 >> 2];
+	u32 txq[0x20 >> 2];
+	u32 ib0[0x20 >> 2];
+	u32 ib1[0x20 >> 2];
+};
+
+struct bna_rx_fndb_ram {
+	u32 rss_prop;
+	u32 size_routing_props;
+	u32 rit_hds_mcastq;
+	u32 control_flags;
+};
+
+struct bna_tx_fndb_ram {
+	u32 vlan_n_ctrl_flags;
+};
+
+/**
+ * @brief
+ *  Structure which maps to RxFn Indirection Table (RIT)
+ *  Size : 1 word
+ *  See catapult_spec.pdf, RxA for details
+ */
+struct bna_rit_mem {
+	u32 rxq_ids;	/* !< 31:12->res;11:0->two 6 bit RxQ Ids */
+};
+
+/**
+ * @brief
+ *  Structure which maps to RSS Table entry
+ *  Size : 16 words
+ *  See catapult_spec.pdf, RAD for details
+ */
+struct bna_rss_mem {
+	/*
+	 * 31:12-> res
+	 * 11:8 -> protocol type
+	 *  7:0 -> hash index
+	 */
+	u32 type_n_hash;
+	u32 hash_key[10];  /* !< 40 byte Toeplitz hash key */
+	u32 reserved[5];
+};
+
+/* TxQ Vector (a.k.a. Tx-Buffer Descriptor) */
+struct bna_dma_addr {
+	u32		msb;
+	u32		lsb;
+};
+
+struct bna_txq_wi_vector {
+	u16 		reserved;
+	u16 		length;		/* Only 14 LSB are valid */
+	struct bna_dma_addr host_addr; /* Tx-Buf DMA addr */
+};
+
+typedef u16 bna_txq_wi_opcode_t;
+
+typedef u16 bna_txq_wi_ctrl_flag_t;
+
+/**
+ *  TxQ Entry Structure
+ *
+ *  BEWARE:  Load values into this structure with correct endianess.
+ */
+struct bna_txq_entry {
+	union {
+		struct {
+			u8 reserved;
+			u8 num_vectors;	/* number of vectors present */
+			bna_txq_wi_opcode_t opcode; /* Either */
+						    /* BNA_TXQ_WI_SEND or */
+						    /* BNA_TXQ_WI_SEND_LSO */
+			bna_txq_wi_ctrl_flag_t flags; /* OR of all the flags */
+			u16 l4_hdr_size_n_offset;
+			u16 vlan_tag;
+			u16 lso_mss;	/* Only 14 LSB are valid */
+			u32 frame_length;	/* Only 24 LSB are valid */
+		} wi;
+
+		struct {
+			u16 reserved;
+			bna_txq_wi_opcode_t opcode; /* Must be */
+						    /* BNA_TXQ_WI_EXTENSION */
+			u32 reserved2[3];	/* Place holder for */
+						/* removed vector (12 bytes) */
+		} wi_ext;
+	} hdr;
+	struct bna_txq_wi_vector vector[4];
+};
+#define wi_hdr  	hdr.wi
+#define wi_ext_hdr  hdr.wi_ext
+
+/* RxQ Entry Structure */
+struct bna_rxq_entry {		/* Rx-Buffer */
+	struct bna_dma_addr host_addr; /* Rx-Buffer DMA address */
+};
+
+typedef u32 bna_cq_e_flag_t;
+
+/* CQ Entry Structure */
+struct bna_cq_entry {
+	bna_cq_e_flag_t flags;
+	u16 vlan_tag;
+	u16 length;
+	u32 rss_hash;
+	u8 valid;
+	u8 reserved1;
+	u8 reserved2;
+	u8 rxq_id;
+};
+
+#endif /* __BNA_HW_H__ */
diff --git a/drivers/net/bna/bna_txrx.c b/drivers/net/bna/bna_txrx.c
new file mode 100644
index 000000000000..890846d55502
--- /dev/null
+++ b/drivers/net/bna/bna_txrx.c
@@ -0,0 +1,4209 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+  */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#include "bna.h"
+#include "bfa_sm.h"
+#include "bfi.h"
+
+/**
+ * IB
+ */
+#define bna_ib_find_free_ibidx(_mask, _pos)\
+do {\
+	(_pos) = 0;\
+	while (((_pos) < (BFI_IBIDX_MAX_SEGSIZE)) &&\
+		((1 << (_pos)) & (_mask)))\
+		(_pos)++;\
+} while (0)
+
+#define bna_ib_count_ibidx(_mask, _count)\
+do {\
+	int pos = 0;\
+	(_count) = 0;\
+	while (pos < (BFI_IBIDX_MAX_SEGSIZE)) {\
+		if ((1 << pos) & (_mask))\
+			(_count) = pos + 1;\
+		pos++;\
+	} \
+} while (0)
+
+#define bna_ib_select_segpool(_count, _q_idx)\
+do {\
+	int i;\
+	(_q_idx) = -1;\
+	for (i = 0; i < BFI_IBIDX_TOTAL_POOLS; i++) {\
+		if ((_count <= ibidx_pool[i].pool_entry_size)) {\
+			(_q_idx) = i;\
+			break;\
+		} \
+	} \
+} while (0)
+
+struct bna_ibidx_pool {
+	int	pool_size;
+	int	pool_entry_size;
+};
+init_ibidx_pool(ibidx_pool);
+
+static struct bna_intr *
+bna_intr_get(struct bna_ib_mod *ib_mod, enum bna_intr_type intr_type,
+		int vector)
+{
+	struct bna_intr *intr;
+	struct list_head *qe;
+
+	list_for_each(qe, &ib_mod->intr_active_q) {
+		intr = (struct bna_intr *)qe;
+
+		if ((intr->intr_type == intr_type) &&
+			(intr->vector == vector)) {
+			intr->ref_count++;
+			return intr;
+		}
+	}
+
+	if (list_empty(&ib_mod->intr_free_q))
+		return NULL;
+
+	bfa_q_deq(&ib_mod->intr_free_q, &intr);
+	bfa_q_qe_init(&intr->qe);
+
+	intr->ref_count = 1;
+	intr->intr_type = intr_type;
+	intr->vector = vector;
+
+	list_add_tail(&intr->qe, &ib_mod->intr_active_q);
+
+	return intr;
+}
+
+static void
+bna_intr_put(struct bna_ib_mod *ib_mod,
+		struct bna_intr *intr)
+{
+	intr->ref_count--;
+
+	if (intr->ref_count == 0) {
+		intr->ib = NULL;
+		list_del(&intr->qe);
+		bfa_q_qe_init(&intr->qe);
+		list_add_tail(&intr->qe, &ib_mod->intr_free_q);
+	}
+}
+
+void
+bna_ib_mod_init(struct bna_ib_mod *ib_mod, struct bna *bna,
+		struct bna_res_info *res_info)
+{
+	int i;
+	int j;
+	int count;
+	u8 offset;
+	struct bna_doorbell_qset *qset;
+	unsigned long off;
+
+	ib_mod->bna = bna;
+
+	ib_mod->ib = (struct bna_ib *)
+		res_info[BNA_RES_MEM_T_IB_ARRAY].res_u.mem_info.mdl[0].kva;
+	ib_mod->intr = (struct bna_intr *)
+		res_info[BNA_RES_MEM_T_INTR_ARRAY].res_u.mem_info.mdl[0].kva;
+	ib_mod->idx_seg = (struct bna_ibidx_seg *)
+		res_info[BNA_RES_MEM_T_IDXSEG_ARRAY].res_u.mem_info.mdl[0].kva;
+
+	INIT_LIST_HEAD(&ib_mod->ib_free_q);
+	INIT_LIST_HEAD(&ib_mod->intr_free_q);
+	INIT_LIST_HEAD(&ib_mod->intr_active_q);
+
+	for (i = 0; i < BFI_IBIDX_TOTAL_POOLS; i++)
+		INIT_LIST_HEAD(&ib_mod->ibidx_seg_pool[i]);
+
+	for (i = 0; i < BFI_MAX_IB; i++) {
+		ib_mod->ib[i].ib_id = i;
+
+		ib_mod->ib[i].ib_seg_host_addr_kva =
+		res_info[BNA_RES_MEM_T_IBIDX].res_u.mem_info.mdl[i].kva;
+		ib_mod->ib[i].ib_seg_host_addr.lsb =
+		res_info[BNA_RES_MEM_T_IBIDX].res_u.mem_info.mdl[i].dma.lsb;
+		ib_mod->ib[i].ib_seg_host_addr.msb =
+		res_info[BNA_RES_MEM_T_IBIDX].res_u.mem_info.mdl[i].dma.msb;
+
+		qset = (struct bna_doorbell_qset *)0;
+		off = (unsigned long)(&qset[i >> 1].ib0[(i & 0x1)
+					* (0x20 >> 2)]);
+		ib_mod->ib[i].door_bell.doorbell_addr = off +
+			BNA_GET_DOORBELL_BASE_ADDR(bna->pcidev.pci_bar_kva);
+
+		bfa_q_qe_init(&ib_mod->ib[i].qe);
+		list_add_tail(&ib_mod->ib[i].qe, &ib_mod->ib_free_q);
+
+		bfa_q_qe_init(&ib_mod->intr[i].qe);
+		list_add_tail(&ib_mod->intr[i].qe, &ib_mod->intr_free_q);
+	}
+
+	count = 0;
+	offset = 0;
+	for (i = 0; i < BFI_IBIDX_TOTAL_POOLS; i++) {
+		for (j = 0; j < ibidx_pool[i].pool_size; j++) {
+			bfa_q_qe_init(&ib_mod->idx_seg[count]);
+			ib_mod->idx_seg[count].ib_seg_size =
+					ibidx_pool[i].pool_entry_size;
+			ib_mod->idx_seg[count].ib_idx_tbl_offset = offset;
+			list_add_tail(&ib_mod->idx_seg[count].qe,
+				&ib_mod->ibidx_seg_pool[i]);
+			count++;
+			offset += ibidx_pool[i].pool_entry_size;
+		}
+	}
+}
+
+void
+bna_ib_mod_uninit(struct bna_ib_mod *ib_mod)
+{
+	int i;
+	int j;
+	struct list_head *qe;
+
+	i = 0;
+	list_for_each(qe, &ib_mod->ib_free_q)
+		i++;
+
+	i = 0;
+	list_for_each(qe, &ib_mod->intr_free_q)
+		i++;
+
+	for (i = 0; i < BFI_IBIDX_TOTAL_POOLS; i++) {
+		j = 0;
+		list_for_each(qe, &ib_mod->ibidx_seg_pool[i])
+			j++;
+	}
+
+	ib_mod->bna = NULL;
+}
+
+struct bna_ib *
+bna_ib_get(struct bna_ib_mod *ib_mod,
+		enum bna_intr_type intr_type,
+		int vector)
+{
+	struct bna_ib *ib;
+	struct bna_intr *intr;
+
+	if (intr_type == BNA_INTR_T_INTX)
+		vector = (1 << vector);
+
+	intr = bna_intr_get(ib_mod, intr_type, vector);
+	if (intr == NULL)
+		return NULL;
+
+	if (intr->ib) {
+		if (intr->ib->ref_count == BFI_IBIDX_MAX_SEGSIZE) {
+			bna_intr_put(ib_mod, intr);
+			return NULL;
+		}
+		intr->ib->ref_count++;
+		return intr->ib;
+	}
+
+	if (list_empty(&ib_mod->ib_free_q)) {
+		bna_intr_put(ib_mod, intr);
+		return NULL;
+	}
+
+	bfa_q_deq(&ib_mod->ib_free_q, &ib);
+	bfa_q_qe_init(&ib->qe);
+
+	ib->ref_count = 1;
+	ib->start_count = 0;
+	ib->idx_mask = 0;
+
+	ib->intr = intr;
+	ib->idx_seg = NULL;
+	intr->ib = ib;
+
+	ib->bna = ib_mod->bna;
+
+	return ib;
+}
+
+void
+bna_ib_put(struct bna_ib_mod *ib_mod, struct bna_ib *ib)
+{
+	bna_intr_put(ib_mod, ib->intr);
+
+	ib->ref_count--;
+
+	if (ib->ref_count == 0) {
+		ib->intr = NULL;
+		ib->bna = NULL;
+		list_add_tail(&ib->qe, &ib_mod->ib_free_q);
+	}
+}
+
+/* Returns index offset - starting from 0 */
+int
+bna_ib_reserve_idx(struct bna_ib *ib)
+{
+	struct bna_ib_mod *ib_mod = &ib->bna->ib_mod;
+	struct bna_ibidx_seg *idx_seg;
+	int idx;
+	int num_idx;
+	int q_idx;
+
+	/* Find the first free index position */
+	bna_ib_find_free_ibidx(ib->idx_mask, idx);
+	if (idx == BFI_IBIDX_MAX_SEGSIZE)
+		return -1;
+
+	/*
+	 * Calculate the total number of indexes held by this IB,
+	 * including the index newly reserved above.
+	 */
+	bna_ib_count_ibidx((ib->idx_mask | (1 << idx)), num_idx);
+
+	/* See if there is a free space in the index segment held by this IB */
+	if (ib->idx_seg && (num_idx <= ib->idx_seg->ib_seg_size)) {
+		ib->idx_mask |= (1 << idx);
+		return idx;
+	}
+
+	if (ib->start_count)
+		return -1;
+
+	/* Allocate a new segment */
+	bna_ib_select_segpool(num_idx, q_idx);
+	while (1) {
+		if (q_idx == BFI_IBIDX_TOTAL_POOLS)
+			return -1;
+		if (!list_empty(&ib_mod->ibidx_seg_pool[q_idx]))
+			break;
+		q_idx++;
+	}
+	bfa_q_deq(&ib_mod->ibidx_seg_pool[q_idx], &idx_seg);
+	bfa_q_qe_init(&idx_seg->qe);
+
+	/* Free the old segment */
+	if (ib->idx_seg) {
+		bna_ib_select_segpool(ib->idx_seg->ib_seg_size, q_idx);
+		list_add_tail(&ib->idx_seg->qe, &ib_mod->ibidx_seg_pool[q_idx]);
+	}
+
+	ib->idx_seg = idx_seg;
+
+	ib->idx_mask |= (1 << idx);
+
+	return idx;
+}
+
+void
+bna_ib_release_idx(struct bna_ib *ib, int idx)
+{
+	struct bna_ib_mod *ib_mod = &ib->bna->ib_mod;
+	struct bna_ibidx_seg *idx_seg;
+	int num_idx;
+	int cur_q_idx;
+	int new_q_idx;
+
+	ib->idx_mask &= ~(1 << idx);
+
+	if (ib->start_count)
+		return;
+
+	bna_ib_count_ibidx(ib->idx_mask, num_idx);
+
+	/*
+	 * Free the segment, if there are no more indexes in the segment
+	 * held by this IB
+	 */
+	if (!num_idx) {
+		bna_ib_select_segpool(ib->idx_seg->ib_seg_size, cur_q_idx);
+		list_add_tail(&ib->idx_seg->qe,
+			&ib_mod->ibidx_seg_pool[cur_q_idx]);
+		ib->idx_seg = NULL;
+		return;
+	}
+
+	/* See if we can move to a smaller segment */
+	bna_ib_select_segpool(num_idx, new_q_idx);
+	bna_ib_select_segpool(ib->idx_seg->ib_seg_size, cur_q_idx);
+	while (new_q_idx < cur_q_idx) {
+		if (!list_empty(&ib_mod->ibidx_seg_pool[new_q_idx]))
+			break;
+		new_q_idx++;
+	}
+	if (new_q_idx < cur_q_idx) {
+		/* Select the new smaller segment */
+		bfa_q_deq(&ib_mod->ibidx_seg_pool[new_q_idx], &idx_seg);
+		bfa_q_qe_init(&idx_seg->qe);
+		/* Free the old segment */
+		list_add_tail(&ib->idx_seg->qe,
+			&ib_mod->ibidx_seg_pool[cur_q_idx]);
+		ib->idx_seg = idx_seg;
+	}
+}
+
+int
+bna_ib_config(struct bna_ib *ib, struct bna_ib_config *ib_config)
+{
+	if (ib->start_count)
+		return -1;
+
+	ib->ib_config.coalescing_timeo = ib_config->coalescing_timeo;
+	ib->ib_config.interpkt_timeo = ib_config->interpkt_timeo;
+	ib->ib_config.interpkt_count = ib_config->interpkt_count;
+	ib->ib_config.ctrl_flags = ib_config->ctrl_flags;
+
+	ib->ib_config.ctrl_flags |= BFI_IB_CF_MASTER_ENABLE;
+	if (ib->intr->intr_type == BNA_INTR_T_MSIX)
+		ib->ib_config.ctrl_flags |= BFI_IB_CF_MSIX_MODE;
+
+	return 0;
+}
+
+void
+bna_ib_start(struct bna_ib *ib)
+{
+	struct bna_ib_blk_mem ib_cfg;
+	struct bna_ib_blk_mem *ib_mem;
+	u32 pg_num;
+	u32 intx_mask;
+	int i;
+	void __iomem *base_addr;
+	unsigned long off;
+
+	ib->start_count++;
+
+	if (ib->start_count > 1)
+		return;
+
+	ib_cfg.host_addr_lo = (u32)(ib->ib_seg_host_addr.lsb);
+	ib_cfg.host_addr_hi = (u32)(ib->ib_seg_host_addr.msb);
+
+	ib_cfg.clsc_n_ctrl_n_msix = (((u32)
+				     ib->ib_config.coalescing_timeo << 16) |
+				((u32)ib->ib_config.ctrl_flags << 8) |
+				(ib->intr->vector));
+	ib_cfg.ipkt_n_ent_n_idxof =
+				((u32)
+				 (ib->ib_config.interpkt_timeo & 0xf) << 16) |
+				((u32)ib->idx_seg->ib_seg_size << 8) |
+				(ib->idx_seg->ib_idx_tbl_offset);
+	ib_cfg.ipkt_cnt_cfg_n_unacked = ((u32)
+					 ib->ib_config.interpkt_count << 24);
+
+	pg_num = BNA_GET_PAGE_NUM(HQM0_BLK_PG_NUM + ib->bna->port_num,
+				HQM_IB_RAM_BASE_OFFSET);
+	writel(pg_num, ib->bna->regs.page_addr);
+
+	base_addr = BNA_GET_MEM_BASE_ADDR(ib->bna->pcidev.pci_bar_kva,
+					HQM_IB_RAM_BASE_OFFSET);
+
+	ib_mem = (struct bna_ib_blk_mem *)0;
+	off = (unsigned long)&ib_mem[ib->ib_id].host_addr_lo;
+	writel(htonl(ib_cfg.host_addr_lo), base_addr + off);
+
+	off = (unsigned long)&ib_mem[ib->ib_id].host_addr_hi;
+	writel(htonl(ib_cfg.host_addr_hi), base_addr + off);
+
+	off = (unsigned long)&ib_mem[ib->ib_id].clsc_n_ctrl_n_msix;
+	writel(ib_cfg.clsc_n_ctrl_n_msix, base_addr + off);
+
+	off = (unsigned long)&ib_mem[ib->ib_id].ipkt_n_ent_n_idxof;
+	writel(ib_cfg.ipkt_n_ent_n_idxof, base_addr + off);
+
+	off = (unsigned long)&ib_mem[ib->ib_id].ipkt_cnt_cfg_n_unacked;
+	writel(ib_cfg.ipkt_cnt_cfg_n_unacked, base_addr + off);
+
+	ib->door_bell.doorbell_ack = BNA_DOORBELL_IB_INT_ACK(
+				(u32)ib->ib_config.coalescing_timeo, 0);
+
+	pg_num = BNA_GET_PAGE_NUM(HQM0_BLK_PG_NUM + ib->bna->port_num,
+				HQM_INDX_TBL_RAM_BASE_OFFSET);
+	writel(pg_num, ib->bna->regs.page_addr);
+
+	base_addr = BNA_GET_MEM_BASE_ADDR(ib->bna->pcidev.pci_bar_kva,
+					HQM_INDX_TBL_RAM_BASE_OFFSET);
+	for (i = 0; i < ib->idx_seg->ib_seg_size; i++) {
+		off = (unsigned long)
+		((ib->idx_seg->ib_idx_tbl_offset + i) * BFI_IBIDX_SIZE);
+		writel(0, base_addr + off);
+	}
+
+	if (ib->intr->intr_type == BNA_INTR_T_INTX) {
+		bna_intx_disable(ib->bna, intx_mask);
+		intx_mask &= ~(ib->intr->vector);
+		bna_intx_enable(ib->bna, intx_mask);
+	}
+}
+
+void
+bna_ib_stop(struct bna_ib *ib)
+{
+	u32 intx_mask;
+
+	ib->start_count--;
+
+	if (ib->start_count == 0) {
+		writel(BNA_DOORBELL_IB_INT_DISABLE,
+				ib->door_bell.doorbell_addr);
+		if (ib->intr->intr_type == BNA_INTR_T_INTX) {
+			bna_intx_disable(ib->bna, intx_mask);
+			intx_mask |= (ib->intr->vector);
+			bna_intx_enable(ib->bna, intx_mask);
+		}
+	}
+}
+
+void
+bna_ib_fail(struct bna_ib *ib)
+{
+	ib->start_count = 0;
+}
+
+/**
+ * RXF
+ */
+static void rxf_enable(struct bna_rxf *rxf);
+static void rxf_disable(struct bna_rxf *rxf);
+static void __rxf_config_set(struct bna_rxf *rxf);
+static void __rxf_rit_set(struct bna_rxf *rxf);
+static void __bna_rxf_stat_clr(struct bna_rxf *rxf);
+static int rxf_process_packet_filter(struct bna_rxf *rxf);
+static int rxf_clear_packet_filter(struct bna_rxf *rxf);
+static void rxf_reset_packet_filter(struct bna_rxf *rxf);
+static void rxf_cb_enabled(void *arg, int status);
+static void rxf_cb_disabled(void *arg, int status);
+static void bna_rxf_cb_stats_cleared(void *arg, int status);
+static void __rxf_enable(struct bna_rxf *rxf);
+static void __rxf_disable(struct bna_rxf *rxf);
+
+bfa_fsm_state_decl(bna_rxf, stopped, struct bna_rxf,
+			enum bna_rxf_event);
+bfa_fsm_state_decl(bna_rxf, start_wait, struct bna_rxf,
+			enum bna_rxf_event);
+bfa_fsm_state_decl(bna_rxf, cam_fltr_mod_wait, struct bna_rxf,
+			enum bna_rxf_event);
+bfa_fsm_state_decl(bna_rxf, started, struct bna_rxf,
+			enum bna_rxf_event);
+bfa_fsm_state_decl(bna_rxf, cam_fltr_clr_wait, struct bna_rxf,
+			enum bna_rxf_event);
+bfa_fsm_state_decl(bna_rxf, stop_wait, struct bna_rxf,
+			enum bna_rxf_event);
+bfa_fsm_state_decl(bna_rxf, pause_wait, struct bna_rxf,
+			enum bna_rxf_event);
+bfa_fsm_state_decl(bna_rxf, resume_wait, struct bna_rxf,
+			enum bna_rxf_event);
+bfa_fsm_state_decl(bna_rxf, stat_clr_wait, struct bna_rxf,
+			enum bna_rxf_event);
+
+static struct bfa_sm_table rxf_sm_table[] = {
+	{BFA_SM(bna_rxf_sm_stopped), BNA_RXF_STOPPED},
+	{BFA_SM(bna_rxf_sm_start_wait), BNA_RXF_START_WAIT},
+	{BFA_SM(bna_rxf_sm_cam_fltr_mod_wait), BNA_RXF_CAM_FLTR_MOD_WAIT},
+	{BFA_SM(bna_rxf_sm_started), BNA_RXF_STARTED},
+	{BFA_SM(bna_rxf_sm_cam_fltr_clr_wait), BNA_RXF_CAM_FLTR_CLR_WAIT},
+	{BFA_SM(bna_rxf_sm_stop_wait), BNA_RXF_STOP_WAIT},
+	{BFA_SM(bna_rxf_sm_pause_wait), BNA_RXF_PAUSE_WAIT},
+	{BFA_SM(bna_rxf_sm_resume_wait), BNA_RXF_RESUME_WAIT},
+	{BFA_SM(bna_rxf_sm_stat_clr_wait), BNA_RXF_STAT_CLR_WAIT}
+};
+
+static void
+bna_rxf_sm_stopped_entry(struct bna_rxf *rxf)
+{
+	call_rxf_stop_cbfn(rxf, BNA_CB_SUCCESS);
+}
+
+static void
+bna_rxf_sm_stopped(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_START:
+		bfa_fsm_set_state(rxf, bna_rxf_sm_start_wait);
+		break;
+
+	case RXF_E_STOP:
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	case RXF_E_FAIL:
+		/* No-op */
+		break;
+
+	case RXF_E_CAM_FLTR_MOD:
+		call_rxf_cam_fltr_cbfn(rxf, BNA_CB_SUCCESS);
+		break;
+
+	case RXF_E_STARTED:
+	case RXF_E_STOPPED:
+	case RXF_E_CAM_FLTR_RESP:
+		/**
+		 * These events are received due to flushing of mbox
+		 * when device fails
+		 */
+		/* No-op */
+		break;
+
+	case RXF_E_PAUSE:
+		rxf->rxf_oper_state = BNA_RXF_OPER_STATE_PAUSED;
+		call_rxf_pause_cbfn(rxf, BNA_CB_SUCCESS);
+		break;
+
+	case RXF_E_RESUME:
+		rxf->rxf_oper_state = BNA_RXF_OPER_STATE_RUNNING;
+		call_rxf_resume_cbfn(rxf, BNA_CB_SUCCESS);
+		break;
+
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+bna_rxf_sm_start_wait_entry(struct bna_rxf *rxf)
+{
+	__rxf_config_set(rxf);
+	__rxf_rit_set(rxf);
+	rxf_enable(rxf);
+}
+
+static void
+bna_rxf_sm_start_wait(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_STOP:
+		/**
+		 * STOP is originated from bnad. When this happens,
+		 * it can not be waiting for filter update
+		 */
+		call_rxf_start_cbfn(rxf, BNA_CB_INTERRUPT);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stop_wait);
+		break;
+
+	case RXF_E_FAIL:
+		call_rxf_cam_fltr_cbfn(rxf, BNA_CB_SUCCESS);
+		call_rxf_start_cbfn(rxf, BNA_CB_FAIL);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	case RXF_E_CAM_FLTR_MOD:
+		/* No-op */
+		break;
+
+	case RXF_E_STARTED:
+		/**
+		 * Force rxf_process_filter() to go through initial
+		 * config
+		 */
+		if ((rxf->ucast_active_mac != NULL) &&
+			(rxf->ucast_pending_set == 0))
+			rxf->ucast_pending_set = 1;
+
+		if (rxf->rss_status == BNA_STATUS_T_ENABLED)
+			rxf->rxf_flags |= BNA_RXF_FL_RSS_CONFIG_PENDING;
+
+		rxf->rxf_flags |= BNA_RXF_FL_VLAN_CONFIG_PENDING;
+
+		bfa_fsm_set_state(rxf, bna_rxf_sm_cam_fltr_mod_wait);
+		break;
+
+	case RXF_E_PAUSE:
+	case RXF_E_RESUME:
+		rxf->rxf_flags |= BNA_RXF_FL_OPERSTATE_CHANGED;
+		break;
+
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+bna_rxf_sm_cam_fltr_mod_wait_entry(struct bna_rxf *rxf)
+{
+	if (!rxf_process_packet_filter(rxf)) {
+		/* No more pending CAM entries to update */
+		bfa_fsm_set_state(rxf, bna_rxf_sm_started);
+	}
+}
+
+static void
+bna_rxf_sm_cam_fltr_mod_wait(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_STOP:
+		/**
+		 * STOP is originated from bnad. When this happens,
+		 * it can not be waiting for filter update
+		 */
+		call_rxf_start_cbfn(rxf, BNA_CB_INTERRUPT);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_cam_fltr_clr_wait);
+		break;
+
+	case RXF_E_FAIL:
+		rxf_reset_packet_filter(rxf);
+		call_rxf_cam_fltr_cbfn(rxf, BNA_CB_SUCCESS);
+		call_rxf_start_cbfn(rxf, BNA_CB_FAIL);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	case RXF_E_CAM_FLTR_MOD:
+		/* No-op */
+		break;
+
+	case RXF_E_CAM_FLTR_RESP:
+		if (!rxf_process_packet_filter(rxf)) {
+			/* No more pending CAM entries to update */
+			call_rxf_cam_fltr_cbfn(rxf, BNA_CB_SUCCESS);
+			bfa_fsm_set_state(rxf, bna_rxf_sm_started);
+		}
+		break;
+
+	case RXF_E_PAUSE:
+	case RXF_E_RESUME:
+		rxf->rxf_flags |= BNA_RXF_FL_OPERSTATE_CHANGED;
+		break;
+
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+bna_rxf_sm_started_entry(struct bna_rxf *rxf)
+{
+	call_rxf_start_cbfn(rxf, BNA_CB_SUCCESS);
+
+	if (rxf->rxf_flags & BNA_RXF_FL_OPERSTATE_CHANGED) {
+		if (rxf->rxf_oper_state == BNA_RXF_OPER_STATE_PAUSED)
+			bfa_fsm_send_event(rxf, RXF_E_PAUSE);
+		else
+			bfa_fsm_send_event(rxf, RXF_E_RESUME);
+	}
+
+}
+
+static void
+bna_rxf_sm_started(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_STOP:
+		bfa_fsm_set_state(rxf, bna_rxf_sm_cam_fltr_clr_wait);
+		/* Hack to get FSM start clearing CAM entries */
+		bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_RESP);
+		break;
+
+	case RXF_E_FAIL:
+		rxf_reset_packet_filter(rxf);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	case RXF_E_CAM_FLTR_MOD:
+		bfa_fsm_set_state(rxf, bna_rxf_sm_cam_fltr_mod_wait);
+		break;
+
+	case RXF_E_PAUSE:
+		bfa_fsm_set_state(rxf, bna_rxf_sm_pause_wait);
+		break;
+
+	case RXF_E_RESUME:
+		bfa_fsm_set_state(rxf, bna_rxf_sm_resume_wait);
+		break;
+
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+bna_rxf_sm_cam_fltr_clr_wait_entry(struct bna_rxf *rxf)
+{
+	/**
+	 *  Note: Do not add rxf_clear_packet_filter here.
+	 * It will overstep mbox when this transition happens:
+	 * 	cam_fltr_mod_wait -> cam_fltr_clr_wait on RXF_E_STOP event
+	 */
+}
+
+static void
+bna_rxf_sm_cam_fltr_clr_wait(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_FAIL:
+		/**
+		 * FSM was in the process of stopping, initiated by
+		 * bnad. When this happens, no one can be waiting for
+		 * start or filter update
+		 */
+		rxf_reset_packet_filter(rxf);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	case RXF_E_CAM_FLTR_RESP:
+		if (!rxf_clear_packet_filter(rxf)) {
+			/* No more pending CAM entries to clear */
+			bfa_fsm_set_state(rxf, bna_rxf_sm_stop_wait);
+			rxf_disable(rxf);
+		}
+		break;
+
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+bna_rxf_sm_stop_wait_entry(struct bna_rxf *rxf)
+{
+	/**
+	 * NOTE: Do not add  rxf_disable here.
+	 * It will overstep mbox when this transition happens:
+	 * 	start_wait -> stop_wait on RXF_E_STOP event
+	 */
+}
+
+static void
+bna_rxf_sm_stop_wait(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_FAIL:
+		/**
+		 * FSM was in the process of stopping, initiated by
+		 * bnad. When this happens, no one can be waiting for
+		 * start or filter update
+		 */
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	case RXF_E_STARTED:
+		/**
+		 * This event is received due to abrupt transition from
+		 * bna_rxf_sm_start_wait state on receiving
+		 * RXF_E_STOP event
+		 */
+		rxf_disable(rxf);
+		break;
+
+	case RXF_E_STOPPED:
+		/**
+		 * FSM was in the process of stopping, initiated by
+		 * bnad. When this happens, no one can be waiting for
+		 * start or filter update
+		 */
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stat_clr_wait);
+		break;
+
+	case RXF_E_PAUSE:
+		rxf->rxf_oper_state = BNA_RXF_OPER_STATE_PAUSED;
+		break;
+
+	case RXF_E_RESUME:
+		rxf->rxf_oper_state = BNA_RXF_OPER_STATE_RUNNING;
+		break;
+
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+bna_rxf_sm_pause_wait_entry(struct bna_rxf *rxf)
+{
+	rxf->rxf_flags &=
+		~(BNA_RXF_FL_OPERSTATE_CHANGED | BNA_RXF_FL_RXF_ENABLED);
+	__rxf_disable(rxf);
+}
+
+static void
+bna_rxf_sm_pause_wait(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_FAIL:
+		/**
+		 * FSM was in the process of disabling rxf, initiated by
+		 * bnad.
+		 */
+		call_rxf_pause_cbfn(rxf, BNA_CB_FAIL);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	case RXF_E_STOPPED:
+		rxf->rxf_oper_state = BNA_RXF_OPER_STATE_PAUSED;
+		call_rxf_pause_cbfn(rxf, BNA_CB_SUCCESS);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_started);
+		break;
+
+	/*
+	 * Since PAUSE/RESUME can only be sent by bnad, we don't expect
+	 * any other event during these states
+	 */
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+bna_rxf_sm_resume_wait_entry(struct bna_rxf *rxf)
+{
+	rxf->rxf_flags &= ~(BNA_RXF_FL_OPERSTATE_CHANGED);
+	rxf->rxf_flags |= BNA_RXF_FL_RXF_ENABLED;
+	__rxf_enable(rxf);
+}
+
+static void
+bna_rxf_sm_resume_wait(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_FAIL:
+		/**
+		 * FSM was in the process of disabling rxf, initiated by
+		 * bnad.
+		 */
+		call_rxf_resume_cbfn(rxf, BNA_CB_FAIL);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	case RXF_E_STARTED:
+		rxf->rxf_oper_state = BNA_RXF_OPER_STATE_RUNNING;
+		call_rxf_resume_cbfn(rxf, BNA_CB_SUCCESS);
+		bfa_fsm_set_state(rxf, bna_rxf_sm_started);
+		break;
+
+	/*
+	 * Since PAUSE/RESUME can only be sent by bnad, we don't expect
+	 * any other event during these states
+	 */
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+bna_rxf_sm_stat_clr_wait_entry(struct bna_rxf *rxf)
+{
+	__bna_rxf_stat_clr(rxf);
+}
+
+static void
+bna_rxf_sm_stat_clr_wait(struct bna_rxf *rxf, enum bna_rxf_event event)
+{
+	switch (event) {
+	case RXF_E_FAIL:
+	case RXF_E_STAT_CLEARED:
+		bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+		break;
+
+	default:
+		bfa_sm_fault(rxf->rx->bna, event);
+	}
+}
+
+static void
+__rxf_enable(struct bna_rxf *rxf)
+{
+	struct bfi_ll_rxf_multi_req ll_req;
+	u32 bm[2] = {0, 0};
+
+	if (rxf->rxf_id < 32)
+		bm[0] = 1 << rxf->rxf_id;
+	else
+		bm[1] = 1 << (rxf->rxf_id - 32);
+
+	bfi_h2i_set(ll_req.mh, BFI_MC_LL, BFI_LL_H2I_RX_REQ, 0);
+	ll_req.rxf_id_mask[0] = htonl(bm[0]);
+	ll_req.rxf_id_mask[1] = htonl(bm[1]);
+	ll_req.enable = 1;
+
+	bna_mbox_qe_fill(&rxf->mbox_qe, &ll_req, sizeof(ll_req),
+			rxf_cb_enabled, rxf);
+
+	bna_mbox_send(rxf->rx->bna, &rxf->mbox_qe);
+}
+
+static void
+__rxf_disable(struct bna_rxf *rxf)
+{
+	struct bfi_ll_rxf_multi_req ll_req;
+	u32 bm[2] = {0, 0};
+
+	if (rxf->rxf_id < 32)
+		bm[0] = 1 << rxf->rxf_id;
+	else
+		bm[1] = 1 << (rxf->rxf_id - 32);
+
+	bfi_h2i_set(ll_req.mh, BFI_MC_LL, BFI_LL_H2I_RX_REQ, 0);
+	ll_req.rxf_id_mask[0] = htonl(bm[0]);
+	ll_req.rxf_id_mask[1] = htonl(bm[1]);
+	ll_req.enable = 0;
+
+	bna_mbox_qe_fill(&rxf->mbox_qe, &ll_req, sizeof(ll_req),
+			rxf_cb_disabled, rxf);
+
+	bna_mbox_send(rxf->rx->bna, &rxf->mbox_qe);
+}
+
+static void
+__rxf_config_set(struct bna_rxf *rxf)
+{
+	u32 i;
+	struct bna_rss_mem *rss_mem;
+	struct bna_rx_fndb_ram *rx_fndb_ram;
+	struct bna *bna = rxf->rx->bna;
+	void __iomem *base_addr;
+	unsigned long off;
+
+	base_addr = BNA_GET_MEM_BASE_ADDR(bna->pcidev.pci_bar_kva,
+			RSS_TABLE_BASE_OFFSET);
+
+	rss_mem = (struct bna_rss_mem *)0;
+
+	/* Configure RSS if required */
+	if (rxf->ctrl_flags & BNA_RXF_CF_RSS_ENABLE) {
+		/* configure RSS Table */
+		writel(BNA_GET_PAGE_NUM(RAD0_MEM_BLK_BASE_PG_NUM +
+			bna->port_num, RSS_TABLE_BASE_OFFSET),
+					bna->regs.page_addr);
+
+		/* temporarily disable RSS, while hash value is written */
+		off = (unsigned long)&rss_mem[0].type_n_hash;
+		writel(0, base_addr + off);
+
+		for (i = 0; i < BFI_RSS_HASH_KEY_LEN; i++) {
+			off = (unsigned long)
+			&rss_mem[0].hash_key[(BFI_RSS_HASH_KEY_LEN - 1) - i];
+			writel(htonl(rxf->rss_cfg.toeplitz_hash_key[i]),
+			base_addr + off);
+		}
+
+		off = (unsigned long)&rss_mem[0].type_n_hash;
+		writel(rxf->rss_cfg.hash_type | rxf->rss_cfg.hash_mask,
+			base_addr + off);
+	}
+
+	/* Configure RxF */
+	writel(BNA_GET_PAGE_NUM(
+		LUT0_MEM_BLK_BASE_PG_NUM + (bna->port_num * 2),
+		RX_FNDB_RAM_BASE_OFFSET),
+		bna->regs.page_addr);
+
+	base_addr = BNA_GET_MEM_BASE_ADDR(bna->pcidev.pci_bar_kva,
+		RX_FNDB_RAM_BASE_OFFSET);
+
+	rx_fndb_ram = (struct bna_rx_fndb_ram *)0;
+
+	/* We always use RSS table 0 */
+	off = (unsigned long)&rx_fndb_ram[rxf->rxf_id].rss_prop;
+	writel(rxf->ctrl_flags & BNA_RXF_CF_RSS_ENABLE,
+		base_addr + off);
+
+	/* small large buffer enable/disable */
+	off = (unsigned long)&rx_fndb_ram[rxf->rxf_id].size_routing_props;
+	writel((rxf->ctrl_flags & BNA_RXF_CF_SM_LG_RXQ) | 0x80,
+		base_addr + off);
+
+	/* RIT offset,  HDS forced offset, multicast RxQ Id */
+	off = (unsigned long)&rx_fndb_ram[rxf->rxf_id].rit_hds_mcastq;
+	writel((rxf->rit_segment->rit_offset << 16) |
+		(rxf->forced_offset << 8) |
+		(rxf->hds_cfg.hdr_type & BNA_HDS_FORCED) | rxf->mcast_rxq_id,
+		base_addr + off);
+
+	/*
+	 * default vlan tag, default function enable, strip vlan bytes,
+	 * HDS type, header size
+	 */
+
+	off = (unsigned long)&rx_fndb_ram[rxf->rxf_id].control_flags;
+	 writel(((u32)rxf->default_vlan_tag << 16) |
+		(rxf->ctrl_flags &
+			(BNA_RXF_CF_DEFAULT_VLAN |
+			BNA_RXF_CF_DEFAULT_FUNCTION_ENABLE |
+			BNA_RXF_CF_VLAN_STRIP)) |
+		(rxf->hds_cfg.hdr_type & ~BNA_HDS_FORCED) |
+		rxf->hds_cfg.header_size,
+		base_addr + off);
+}
+
+void
+__rxf_vlan_filter_set(struct bna_rxf *rxf, enum bna_status status)
+{
+	struct bna *bna = rxf->rx->bna;
+	int i;
+
+	writel(BNA_GET_PAGE_NUM(LUT0_MEM_BLK_BASE_PG_NUM +
+			(bna->port_num * 2), VLAN_RAM_BASE_OFFSET),
+			bna->regs.page_addr);
+
+	if (status == BNA_STATUS_T_ENABLED) {
+		/* enable VLAN filtering on this function */
+		for (i = 0; i <= BFI_MAX_VLAN / 32; i++) {
+			writel(rxf->vlan_filter_table[i],
+					BNA_GET_VLAN_MEM_ENTRY_ADDR
+					(bna->pcidev.pci_bar_kva, rxf->rxf_id,
+						i * 32));
+		}
+	} else {
+		/* disable VLAN filtering on this function */
+		for (i = 0; i <= BFI_MAX_VLAN / 32; i++) {
+			writel(0xffffffff,
+					BNA_GET_VLAN_MEM_ENTRY_ADDR
+					(bna->pcidev.pci_bar_kva, rxf->rxf_id,
+						i * 32));
+		}
+	}
+}
+
+static void
+__rxf_rit_set(struct bna_rxf *rxf)
+{
+	struct bna *bna = rxf->rx->bna;
+	struct bna_rit_mem *rit_mem;
+	int i;
+	void __iomem *base_addr;
+	unsigned long off;
+
+	base_addr = BNA_GET_MEM_BASE_ADDR(bna->pcidev.pci_bar_kva,
+			FUNCTION_TO_RXQ_TRANSLATE);
+
+	rit_mem = (struct bna_rit_mem *)0;
+
+	writel(BNA_GET_PAGE_NUM(RXA0_MEM_BLK_BASE_PG_NUM + bna->port_num,
+		FUNCTION_TO_RXQ_TRANSLATE),
+		bna->regs.page_addr);
+
+	for (i = 0; i < rxf->rit_segment->rit_size; i++) {
+		off = (unsigned long)&rit_mem[i + rxf->rit_segment->rit_offset];
+		writel(rxf->rit_segment->rit[i].large_rxq_id << 6 |
+			rxf->rit_segment->rit[i].small_rxq_id,
+			base_addr + off);
+	}
+}
+
+static void
+__bna_rxf_stat_clr(struct bna_rxf *rxf)
+{
+	struct bfi_ll_stats_req ll_req;
+	u32 bm[2] = {0, 0};
+
+	if (rxf->rxf_id < 32)
+		bm[0] = 1 << rxf->rxf_id;
+	else
+		bm[1] = 1 << (rxf->rxf_id - 32);
+
+	bfi_h2i_set(ll_req.mh, BFI_MC_LL, BFI_LL_H2I_STATS_CLEAR_REQ, 0);
+	ll_req.stats_mask = 0;
+	ll_req.txf_id_mask[0] = 0;
+	ll_req.txf_id_mask[1] =	0;
+
+	ll_req.rxf_id_mask[0] = htonl(bm[0]);
+	ll_req.rxf_id_mask[1] = htonl(bm[1]);
+
+	bna_mbox_qe_fill(&rxf->mbox_qe, &ll_req, sizeof(ll_req),
+			bna_rxf_cb_stats_cleared, rxf);
+	bna_mbox_send(rxf->rx->bna, &rxf->mbox_qe);
+}
+
+static void
+rxf_enable(struct bna_rxf *rxf)
+{
+	if (rxf->rxf_oper_state == BNA_RXF_OPER_STATE_PAUSED)
+		bfa_fsm_send_event(rxf, RXF_E_STARTED);
+	else {
+		rxf->rxf_flags |= BNA_RXF_FL_RXF_ENABLED;
+		__rxf_enable(rxf);
+	}
+}
+
+static void
+rxf_cb_enabled(void *arg, int status)
+{
+	struct bna_rxf *rxf = (struct bna_rxf *)arg;
+
+	bfa_q_qe_init(&rxf->mbox_qe.qe);
+	bfa_fsm_send_event(rxf, RXF_E_STARTED);
+}
+
+static void
+rxf_disable(struct bna_rxf *rxf)
+{
+	if (rxf->rxf_oper_state == BNA_RXF_OPER_STATE_PAUSED)
+		bfa_fsm_send_event(rxf, RXF_E_STOPPED);
+	else
+		rxf->rxf_flags &= ~BNA_RXF_FL_RXF_ENABLED;
+		__rxf_disable(rxf);
+}
+
+static void
+rxf_cb_disabled(void *arg, int status)
+{
+	struct bna_rxf *rxf = (struct bna_rxf *)arg;
+
+	bfa_q_qe_init(&rxf->mbox_qe.qe);
+	bfa_fsm_send_event(rxf, RXF_E_STOPPED);
+}
+
+void
+rxf_cb_cam_fltr_mbox_cmd(void *arg, int status)
+{
+	struct bna_rxf *rxf = (struct bna_rxf *)arg;
+
+	bfa_q_qe_init(&rxf->mbox_qe.qe);
+
+	bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_RESP);
+}
+
+static void
+bna_rxf_cb_stats_cleared(void *arg, int status)
+{
+	struct bna_rxf *rxf = (struct bna_rxf *)arg;
+
+	bfa_q_qe_init(&rxf->mbox_qe.qe);
+	bfa_fsm_send_event(rxf, RXF_E_STAT_CLEARED);
+}
+
+void
+rxf_cam_mbox_cmd(struct bna_rxf *rxf, u8 cmd,
+		const struct bna_mac *mac_addr)
+{
+	struct bfi_ll_mac_addr_req req;
+
+	bfi_h2i_set(req.mh, BFI_MC_LL, cmd, 0);
+
+	req.rxf_id = rxf->rxf_id;
+	memcpy(&req.mac_addr, (void *)&mac_addr->addr, ETH_ALEN);
+
+	bna_mbox_qe_fill(&rxf->mbox_qe, &req, sizeof(req),
+				rxf_cb_cam_fltr_mbox_cmd, rxf);
+
+	bna_mbox_send(rxf->rx->bna, &rxf->mbox_qe);
+}
+
+static int
+rxf_process_packet_filter_mcast(struct bna_rxf *rxf)
+{
+	struct bna_mac *mac = NULL;
+	struct list_head *qe;
+
+	/* Add multicast entries */
+	if (!list_empty(&rxf->mcast_pending_add_q)) {
+		bfa_q_deq(&rxf->mcast_pending_add_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_MCAST_ADD_REQ, mac);
+		list_add_tail(&mac->qe, &rxf->mcast_active_q);
+		return 1;
+	}
+
+	/* Delete multicast entries previousely added */
+	if (!list_empty(&rxf->mcast_pending_del_q)) {
+		bfa_q_deq(&rxf->mcast_pending_del_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_MCAST_DEL_REQ, mac);
+		bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod, mac);
+		return 1;
+	}
+
+	return 0;
+}
+
+static int
+rxf_process_packet_filter_vlan(struct bna_rxf *rxf)
+{
+	/* Apply the VLAN filter */
+	if (rxf->rxf_flags & BNA_RXF_FL_VLAN_CONFIG_PENDING) {
+		rxf->rxf_flags &= ~BNA_RXF_FL_VLAN_CONFIG_PENDING;
+		if (!(rxf->rxmode_active & BNA_RXMODE_PROMISC) &&
+			!(rxf->rxmode_active & BNA_RXMODE_DEFAULT))
+			__rxf_vlan_filter_set(rxf, rxf->vlan_filter_status);
+	}
+
+	/* Apply RSS configuration */
+	if (rxf->rxf_flags & BNA_RXF_FL_RSS_CONFIG_PENDING) {
+		rxf->rxf_flags &= ~BNA_RXF_FL_RSS_CONFIG_PENDING;
+		if (rxf->rss_status == BNA_STATUS_T_DISABLED) {
+			/* RSS is being disabled */
+			rxf->ctrl_flags &= ~BNA_RXF_CF_RSS_ENABLE;
+			__rxf_rit_set(rxf);
+			__rxf_config_set(rxf);
+		} else {
+			/* RSS is being enabled or reconfigured */
+			rxf->ctrl_flags |= BNA_RXF_CF_RSS_ENABLE;
+			__rxf_rit_set(rxf);
+			__rxf_config_set(rxf);
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * Processes pending ucast, mcast entry addition/deletion and issues mailbox
+ * command. Also processes pending filter configuration - promiscuous mode,
+ * default mode, allmutli mode and issues mailbox command or directly applies
+ * to h/w
+ */
+static int
+rxf_process_packet_filter(struct bna_rxf *rxf)
+{
+	/* Set the default MAC first */
+	if (rxf->ucast_pending_set > 0) {
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_UCAST_SET_REQ,
+				rxf->ucast_active_mac);
+		rxf->ucast_pending_set--;
+		return 1;
+	}
+
+	if (rxf_process_packet_filter_ucast(rxf))
+		return 1;
+
+	if (rxf_process_packet_filter_mcast(rxf))
+		return 1;
+
+	if (rxf_process_packet_filter_promisc(rxf))
+		return 1;
+
+	if (rxf_process_packet_filter_default(rxf))
+		return 1;
+
+	if (rxf_process_packet_filter_allmulti(rxf))
+		return 1;
+
+	if (rxf_process_packet_filter_vlan(rxf))
+		return 1;
+
+	return 0;
+}
+
+static int
+rxf_clear_packet_filter_mcast(struct bna_rxf *rxf)
+{
+	struct bna_mac *mac = NULL;
+	struct list_head *qe;
+
+	/* 3. delete pending mcast entries */
+	if (!list_empty(&rxf->mcast_pending_del_q)) {
+		bfa_q_deq(&rxf->mcast_pending_del_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_MCAST_DEL_REQ, mac);
+		bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod, mac);
+		return 1;
+	}
+
+	/* 4. clear active mcast entries; move them to pending_add_q */
+	if (!list_empty(&rxf->mcast_active_q)) {
+		bfa_q_deq(&rxf->mcast_active_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		rxf_cam_mbox_cmd(rxf, BFI_LL_H2I_MAC_MCAST_DEL_REQ, mac);
+		list_add_tail(&mac->qe, &rxf->mcast_pending_add_q);
+		return 1;
+	}
+
+	return 0;
+}
+
+/**
+ * In the rxf stop path, processes pending ucast/mcast delete queue and issues
+ * the mailbox command. Moves the active ucast/mcast entries to pending add q,
+ * so that they are added to CAM again in the rxf start path. Moves the current
+ * filter settings - promiscuous, default, allmutli - to pending filter
+ * configuration
+ */
+static int
+rxf_clear_packet_filter(struct bna_rxf *rxf)
+{
+	if (rxf_clear_packet_filter_ucast(rxf))
+		return 1;
+
+	if (rxf_clear_packet_filter_mcast(rxf))
+		return 1;
+
+	/* 5. clear active default MAC in the CAM */
+	if (rxf->ucast_pending_set > 0)
+		rxf->ucast_pending_set = 0;
+
+	if (rxf_clear_packet_filter_promisc(rxf))
+		return 1;
+
+	if (rxf_clear_packet_filter_default(rxf))
+		return 1;
+
+	if (rxf_clear_packet_filter_allmulti(rxf))
+		return 1;
+
+	return 0;
+}
+
+static void
+rxf_reset_packet_filter_mcast(struct bna_rxf *rxf)
+{
+	struct list_head *qe;
+	struct bna_mac *mac;
+
+	/* 3. Move active mcast entries to pending_add_q */
+	while (!list_empty(&rxf->mcast_active_q)) {
+		bfa_q_deq(&rxf->mcast_active_q, &qe);
+		bfa_q_qe_init(qe);
+		list_add_tail(qe, &rxf->mcast_pending_add_q);
+	}
+
+	/* 4. Throw away delete pending mcast entries */
+	while (!list_empty(&rxf->mcast_pending_del_q)) {
+		bfa_q_deq(&rxf->mcast_pending_del_q, &qe);
+		bfa_q_qe_init(qe);
+		mac = (struct bna_mac *)qe;
+		bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod, mac);
+	}
+}
+
+/**
+ * In the rxf fail path, throws away the ucast/mcast entries pending for
+ * deletion, moves all active ucast/mcast entries to pending queue so that
+ * they are added back to CAM in the rxf start path. Also moves the current
+ * filter configuration to pending filter configuration.
+ */
+static void
+rxf_reset_packet_filter(struct bna_rxf *rxf)
+{
+	rxf_reset_packet_filter_ucast(rxf);
+
+	rxf_reset_packet_filter_mcast(rxf);
+
+	/* 5. Turn off ucast set flag */
+	rxf->ucast_pending_set = 0;
+
+	rxf_reset_packet_filter_promisc(rxf);
+
+	rxf_reset_packet_filter_default(rxf);
+
+	rxf_reset_packet_filter_allmulti(rxf);
+}
+
+void
+bna_rxf_init(struct bna_rxf *rxf,
+		struct bna_rx *rx,
+		struct bna_rx_config *q_config)
+{
+	struct list_head *qe;
+	struct bna_rxp *rxp;
+
+	/* rxf_id is initialized during rx_mod init */
+	rxf->rx = rx;
+
+	INIT_LIST_HEAD(&rxf->ucast_pending_add_q);
+	INIT_LIST_HEAD(&rxf->ucast_pending_del_q);
+	rxf->ucast_pending_set = 0;
+	INIT_LIST_HEAD(&rxf->ucast_active_q);
+	rxf->ucast_active_mac = NULL;
+
+	INIT_LIST_HEAD(&rxf->mcast_pending_add_q);
+	INIT_LIST_HEAD(&rxf->mcast_pending_del_q);
+	INIT_LIST_HEAD(&rxf->mcast_active_q);
+
+	bfa_q_qe_init(&rxf->mbox_qe.qe);
+
+	if (q_config->vlan_strip_status == BNA_STATUS_T_ENABLED)
+		rxf->ctrl_flags |= BNA_RXF_CF_VLAN_STRIP;
+
+	rxf->rxf_oper_state = (q_config->paused) ?
+		BNA_RXF_OPER_STATE_PAUSED : BNA_RXF_OPER_STATE_RUNNING;
+
+	bna_rxf_adv_init(rxf, rx, q_config);
+
+	rxf->rit_segment = bna_rit_mod_seg_get(&rxf->rx->bna->rit_mod,
+					q_config->num_paths);
+
+	list_for_each(qe, &rx->rxp_q) {
+		rxp = (struct bna_rxp *)qe;
+		if (q_config->rxp_type == BNA_RXP_SINGLE)
+			rxf->mcast_rxq_id = rxp->rxq.single.only->rxq_id;
+		else
+			rxf->mcast_rxq_id = rxp->rxq.slr.large->rxq_id;
+		break;
+	}
+
+	rxf->vlan_filter_status = BNA_STATUS_T_DISABLED;
+	memset(rxf->vlan_filter_table, 0,
+			(sizeof(u32) * ((BFI_MAX_VLAN + 1) / 32)));
+
+	bfa_fsm_set_state(rxf, bna_rxf_sm_stopped);
+}
+
+void
+bna_rxf_uninit(struct bna_rxf *rxf)
+{
+	struct bna_mac *mac;
+
+	bna_rit_mod_seg_put(&rxf->rx->bna->rit_mod, rxf->rit_segment);
+	rxf->rit_segment = NULL;
+
+	rxf->ucast_pending_set = 0;
+
+	while (!list_empty(&rxf->ucast_pending_add_q)) {
+		bfa_q_deq(&rxf->ucast_pending_add_q, &mac);
+		bfa_q_qe_init(&mac->qe);
+		bna_ucam_mod_mac_put(&rxf->rx->bna->ucam_mod, mac);
+	}
+
+	if (rxf->ucast_active_mac) {
+		bfa_q_qe_init(&rxf->ucast_active_mac->qe);
+		bna_ucam_mod_mac_put(&rxf->rx->bna->ucam_mod,
+			rxf->ucast_active_mac);
+		rxf->ucast_active_mac = NULL;
+	}
+
+	while (!list_empty(&rxf->mcast_pending_add_q)) {
+		bfa_q_deq(&rxf->mcast_pending_add_q, &mac);
+		bfa_q_qe_init(&mac->qe);
+		bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod, mac);
+	}
+
+	rxf->rx = NULL;
+}
+
+void
+bna_rxf_start(struct bna_rxf *rxf)
+{
+	rxf->start_cbfn = bna_rx_cb_rxf_started;
+	rxf->start_cbarg = rxf->rx;
+	rxf->rxf_flags &= ~BNA_RXF_FL_FAILED;
+	bfa_fsm_send_event(rxf, RXF_E_START);
+}
+
+void
+bna_rxf_stop(struct bna_rxf *rxf)
+{
+	rxf->stop_cbfn = bna_rx_cb_rxf_stopped;
+	rxf->stop_cbarg = rxf->rx;
+	bfa_fsm_send_event(rxf, RXF_E_STOP);
+}
+
+void
+bna_rxf_fail(struct bna_rxf *rxf)
+{
+	rxf->rxf_flags |= BNA_RXF_FL_FAILED;
+	bfa_fsm_send_event(rxf, RXF_E_FAIL);
+}
+
+int
+bna_rxf_state_get(struct bna_rxf *rxf)
+{
+	return bfa_sm_to_state(rxf_sm_table, rxf->fsm);
+}
+
+enum bna_cb_status
+bna_rx_ucast_set(struct bna_rx *rx, u8 *ucmac,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+
+	if (rxf->ucast_active_mac == NULL) {
+		rxf->ucast_active_mac =
+				bna_ucam_mod_mac_get(&rxf->rx->bna->ucam_mod);
+		if (rxf->ucast_active_mac == NULL)
+			return BNA_CB_UCAST_CAM_FULL;
+		bfa_q_qe_init(&rxf->ucast_active_mac->qe);
+	}
+
+	memcpy(rxf->ucast_active_mac->addr, ucmac, ETH_ALEN);
+	rxf->ucast_pending_set++;
+	rxf->cam_fltr_cbfn = cbfn;
+	rxf->cam_fltr_cbarg = rx->bna->bnad;
+
+	bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+
+	return BNA_CB_SUCCESS;
+}
+
+enum bna_cb_status
+bna_rx_mcast_add(struct bna_rx *rx, u8 *addr,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	struct list_head	*qe;
+	struct bna_mac *mac;
+
+	/* Check if already added */
+	list_for_each(qe, &rxf->mcast_active_q) {
+		mac = (struct bna_mac *)qe;
+		if (BNA_MAC_IS_EQUAL(mac->addr, addr)) {
+			if (cbfn)
+				(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+			return BNA_CB_SUCCESS;
+		}
+	}
+
+	/* Check if pending addition */
+	list_for_each(qe, &rxf->mcast_pending_add_q) {
+		mac = (struct bna_mac *)qe;
+		if (BNA_MAC_IS_EQUAL(mac->addr, addr)) {
+			if (cbfn)
+				(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+			return BNA_CB_SUCCESS;
+		}
+	}
+
+	mac = bna_mcam_mod_mac_get(&rxf->rx->bna->mcam_mod);
+	if (mac == NULL)
+		return BNA_CB_MCAST_LIST_FULL;
+	bfa_q_qe_init(&mac->qe);
+	memcpy(mac->addr, addr, ETH_ALEN);
+	list_add_tail(&mac->qe, &rxf->mcast_pending_add_q);
+
+	rxf->cam_fltr_cbfn = cbfn;
+	rxf->cam_fltr_cbarg = rx->bna->bnad;
+
+	bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+
+	return BNA_CB_SUCCESS;
+}
+
+enum bna_cb_status
+bna_rx_mcast_del(struct bna_rx *rx, u8 *addr,
+		 void (*cbfn)(struct bnad *, struct bna_rx *,
+			      enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	struct list_head *qe;
+	struct bna_mac *mac;
+
+	list_for_each(qe, &rxf->mcast_pending_add_q) {
+		mac = (struct bna_mac *)qe;
+		if (BNA_MAC_IS_EQUAL(mac->addr, addr)) {
+			list_del(qe);
+			bfa_q_qe_init(qe);
+			bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod, mac);
+			if (cbfn)
+				(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+			return BNA_CB_SUCCESS;
+		}
+	}
+
+	list_for_each(qe, &rxf->mcast_active_q) {
+		mac = (struct bna_mac *)qe;
+		if (BNA_MAC_IS_EQUAL(mac->addr, addr)) {
+			list_del(qe);
+			bfa_q_qe_init(qe);
+			list_add_tail(qe, &rxf->mcast_pending_del_q);
+			rxf->cam_fltr_cbfn = cbfn;
+			rxf->cam_fltr_cbarg = rx->bna->bnad;
+			bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+			return BNA_CB_SUCCESS;
+		}
+	}
+
+	return BNA_CB_INVALID_MAC;
+}
+
+enum bna_cb_status
+bna_rx_mcast_listset(struct bna_rx *rx, int count, u8 *mclist,
+		     void (*cbfn)(struct bnad *, struct bna_rx *,
+				  enum bna_cb_status))
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	struct list_head list_head;
+	struct list_head *qe;
+	u8 *mcaddr;
+	struct bna_mac *mac;
+	struct bna_mac *mac1;
+	int skip;
+	int delete;
+	int need_hw_config = 0;
+	int i;
+
+	/* Allocate nodes */
+	INIT_LIST_HEAD(&list_head);
+	for (i = 0, mcaddr = mclist; i < count; i++) {
+		mac = bna_mcam_mod_mac_get(&rxf->rx->bna->mcam_mod);
+		if (mac == NULL)
+			goto err_return;
+		bfa_q_qe_init(&mac->qe);
+		memcpy(mac->addr, mcaddr, ETH_ALEN);
+		list_add_tail(&mac->qe, &list_head);
+
+		mcaddr += ETH_ALEN;
+	}
+
+	/* Schedule for addition */
+	while (!list_empty(&list_head)) {
+		bfa_q_deq(&list_head, &qe);
+		mac = (struct bna_mac *)qe;
+		bfa_q_qe_init(&mac->qe);
+
+		skip = 0;
+
+		/* Skip if already added */
+		list_for_each(qe, &rxf->mcast_active_q) {
+			mac1 = (struct bna_mac *)qe;
+			if (BNA_MAC_IS_EQUAL(mac1->addr, mac->addr)) {
+				bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod,
+							mac);
+				skip = 1;
+				break;
+			}
+		}
+
+		if (skip)
+			continue;
+
+		/* Skip if pending addition */
+		list_for_each(qe, &rxf->mcast_pending_add_q) {
+			mac1 = (struct bna_mac *)qe;
+			if (BNA_MAC_IS_EQUAL(mac1->addr, mac->addr)) {
+				bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod,
+							mac);
+				skip = 1;
+				break;
+			}
+		}
+
+		if (skip)
+			continue;
+
+		need_hw_config = 1;
+		list_add_tail(&mac->qe, &rxf->mcast_pending_add_q);
+	}
+
+	/**
+	 * Delete the entries that are in the pending_add_q but not
+	 * in the new list
+	 */
+	while (!list_empty(&rxf->mcast_pending_add_q)) {
+		bfa_q_deq(&rxf->mcast_pending_add_q, &qe);
+		mac = (struct bna_mac *)qe;
+		bfa_q_qe_init(&mac->qe);
+		for (i = 0, mcaddr = mclist, delete = 1; i < count; i++) {
+			if (BNA_MAC_IS_EQUAL(mcaddr, mac->addr)) {
+				delete = 0;
+				break;
+			}
+			mcaddr += ETH_ALEN;
+		}
+		if (delete)
+			bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod, mac);
+		else
+			list_add_tail(&mac->qe, &list_head);
+	}
+	while (!list_empty(&list_head)) {
+		bfa_q_deq(&list_head, &qe);
+		mac = (struct bna_mac *)qe;
+		bfa_q_qe_init(&mac->qe);
+		list_add_tail(&mac->qe, &rxf->mcast_pending_add_q);
+	}
+
+	/**
+	 * Schedule entries for deletion that are in the active_q but not
+	 * in the new list
+	 */
+	while (!list_empty(&rxf->mcast_active_q)) {
+		bfa_q_deq(&rxf->mcast_active_q, &qe);
+		mac = (struct bna_mac *)qe;
+		bfa_q_qe_init(&mac->qe);
+		for (i = 0, mcaddr = mclist, delete = 1; i < count; i++) {
+			if (BNA_MAC_IS_EQUAL(mcaddr, mac->addr)) {
+				delete = 0;
+				break;
+			}
+			mcaddr += ETH_ALEN;
+		}
+		if (delete) {
+			list_add_tail(&mac->qe, &rxf->mcast_pending_del_q);
+			need_hw_config = 1;
+		} else {
+			list_add_tail(&mac->qe, &list_head);
+		}
+	}
+	while (!list_empty(&list_head)) {
+		bfa_q_deq(&list_head, &qe);
+		mac = (struct bna_mac *)qe;
+		bfa_q_qe_init(&mac->qe);
+		list_add_tail(&mac->qe, &rxf->mcast_active_q);
+	}
+
+	if (need_hw_config) {
+		rxf->cam_fltr_cbfn = cbfn;
+		rxf->cam_fltr_cbarg = rx->bna->bnad;
+		bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+	} else if (cbfn)
+		(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+
+	return BNA_CB_SUCCESS;
+
+err_return:
+	while (!list_empty(&list_head)) {
+		bfa_q_deq(&list_head, &qe);
+		mac = (struct bna_mac *)qe;
+		bfa_q_qe_init(&mac->qe);
+		bna_mcam_mod_mac_put(&rxf->rx->bna->mcam_mod, mac);
+	}
+
+	return BNA_CB_MCAST_LIST_FULL;
+}
+
+void
+bna_rx_vlan_add(struct bna_rx *rx, int vlan_id)
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	int index = (vlan_id >> 5);
+	int bit = (1 << (vlan_id & 0x1F));
+
+	rxf->vlan_filter_table[index] |= bit;
+	if (rxf->vlan_filter_status == BNA_STATUS_T_ENABLED) {
+		rxf->rxf_flags |= BNA_RXF_FL_VLAN_CONFIG_PENDING;
+		bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+	}
+}
+
+void
+bna_rx_vlan_del(struct bna_rx *rx, int vlan_id)
+{
+	struct bna_rxf *rxf = &rx->rxf;
+	int index = (vlan_id >> 5);
+	int bit = (1 << (vlan_id & 0x1F));
+
+	rxf->vlan_filter_table[index] &= ~bit;
+	if (rxf->vlan_filter_status == BNA_STATUS_T_ENABLED) {
+		rxf->rxf_flags |= BNA_RXF_FL_VLAN_CONFIG_PENDING;
+		bfa_fsm_send_event(rxf, RXF_E_CAM_FLTR_MOD);
+	}
+}
+
+/**
+ * RX
+ */
+#define	RXQ_RCB_INIT(q, rxp, qdepth, bna, _id, unmapq_mem)	do {	\
+	struct bna_doorbell_qset *_qset;				\
+	unsigned long off;						\
+	(q)->rcb->producer_index = (q)->rcb->consumer_index = 0;	\
+	(q)->rcb->q_depth = (qdepth);					\
+	(q)->rcb->unmap_q = unmapq_mem;					\
+	(q)->rcb->rxq = (q);						\
+	(q)->rcb->cq = &(rxp)->cq;					\
+	(q)->rcb->bnad = (bna)->bnad;					\
+	_qset = (struct bna_doorbell_qset *)0;			\
+	off = (unsigned long)&_qset[(q)->rxq_id].rxq[0];		\
+	(q)->rcb->q_dbell = off +					\
+		BNA_GET_DOORBELL_BASE_ADDR((bna)->pcidev.pci_bar_kva);	\
+	(q)->rcb->id = _id;						\
+} while (0)
+
+#define	BNA_GET_RXQS(qcfg)	(((qcfg)->rxp_type == BNA_RXP_SINGLE) ?	\
+	(qcfg)->num_paths : ((qcfg)->num_paths * 2))
+
+#define	SIZE_TO_PAGES(size)	(((size) >> PAGE_SHIFT) + ((((size) &\
+	(PAGE_SIZE - 1)) + (PAGE_SIZE - 1)) >> PAGE_SHIFT))
+
+#define	call_rx_stop_callback(rx, status)				\
+	if ((rx)->stop_cbfn) {						\
+		(*(rx)->stop_cbfn)((rx)->stop_cbarg, rx, (status));	\
+		(rx)->stop_cbfn = NULL;					\
+		(rx)->stop_cbarg = NULL;				\
+	}
+
+/*
+ * Since rx_enable is synchronous callback, there is no start_cbfn required.
+ * Instead, we'll call bnad_rx_post(rxp) so that bnad can post the buffers
+ * for each rxpath.
+ */
+
+#define	call_rx_disable_cbfn(rx, status)				\
+		if ((rx)->disable_cbfn)	{				\
+			(*(rx)->disable_cbfn)((rx)->disable_cbarg,	\
+					status);			\
+			(rx)->disable_cbfn = NULL;			\
+			(rx)->disable_cbarg = NULL;			\
+		}							\
+
+#define	rxqs_reqd(type, num_rxqs)					\
+	(((type) == BNA_RXP_SINGLE) ? (num_rxqs) : ((num_rxqs) * 2))
+
+#define rx_ib_fail(rx)						\
+do {								\
+	struct bna_rxp *rxp;					\
+	struct list_head *qe;						\
+	list_for_each(qe, &(rx)->rxp_q) {				\
+		rxp = (struct bna_rxp *)qe;			\
+		bna_ib_fail(rxp->cq.ib);			\
+	}							\
+} while (0)
+
+static void __bna_multi_rxq_stop(struct bna_rxp *, u32 *);
+static void __bna_rxq_start(struct bna_rxq *rxq);
+static void __bna_cq_start(struct bna_cq *cq);
+static void bna_rit_create(struct bna_rx *rx);
+static void bna_rx_cb_multi_rxq_stopped(void *arg, int status);
+static void bna_rx_cb_rxq_stopped_all(void *arg);
+
+bfa_fsm_state_decl(bna_rx, stopped,
+	struct bna_rx, enum bna_rx_event);
+bfa_fsm_state_decl(bna_rx, rxf_start_wait,
+	struct bna_rx, enum bna_rx_event);
+bfa_fsm_state_decl(bna_rx, started,
+	struct bna_rx, enum bna_rx_event);
+bfa_fsm_state_decl(bna_rx, rxf_stop_wait,
+	struct bna_rx, enum bna_rx_event);
+bfa_fsm_state_decl(bna_rx, rxq_stop_wait,
+	struct bna_rx, enum bna_rx_event);
+
+static struct bfa_sm_table rx_sm_table[] = {
+	{BFA_SM(bna_rx_sm_stopped), BNA_RX_STOPPED},
+	{BFA_SM(bna_rx_sm_rxf_start_wait), BNA_RX_RXF_START_WAIT},
+	{BFA_SM(bna_rx_sm_started), BNA_RX_STARTED},
+	{BFA_SM(bna_rx_sm_rxf_stop_wait), BNA_RX_RXF_STOP_WAIT},
+	{BFA_SM(bna_rx_sm_rxq_stop_wait), BNA_RX_RXQ_STOP_WAIT},
+};
+
+static void bna_rx_sm_stopped_entry(struct bna_rx *rx)
+{
+	struct bna_rxp *rxp;
+	struct list_head *qe_rxp;
+
+	list_for_each(qe_rxp, &rx->rxp_q) {
+		rxp = (struct bna_rxp *)qe_rxp;
+		rx->rx_cleanup_cbfn(rx->bna->bnad, rxp->cq.ccb);
+	}
+
+	call_rx_stop_callback(rx, BNA_CB_SUCCESS);
+}
+
+static void bna_rx_sm_stopped(struct bna_rx *rx,
+				enum bna_rx_event event)
+{
+	switch (event) {
+	case RX_E_START:
+		bfa_fsm_set_state(rx, bna_rx_sm_rxf_start_wait);
+		break;
+	case RX_E_STOP:
+		call_rx_stop_callback(rx, BNA_CB_SUCCESS);
+		break;
+	case RX_E_FAIL:
+		/* no-op */
+		break;
+	default:
+		bfa_sm_fault(rx->bna, event);
+		break;
+	}
+
+}
+
+static void bna_rx_sm_rxf_start_wait_entry(struct bna_rx *rx)
+{
+	struct bna_rxp *rxp;
+	struct list_head *qe_rxp;
+	struct bna_rxq *q0 = NULL, *q1 = NULL;
+
+	/* Setup the RIT */
+	bna_rit_create(rx);
+
+	list_for_each(qe_rxp, &rx->rxp_q) {
+		rxp = (struct bna_rxp *)qe_rxp;
+		bna_ib_start(rxp->cq.ib);
+		GET_RXQS(rxp, q0, q1);
+		q0->buffer_size = bna_port_mtu_get(&rx->bna->port);
+		__bna_rxq_start(q0);
+		rx->rx_post_cbfn(rx->bna->bnad, q0->rcb);
+		if (q1)  {
+			__bna_rxq_start(q1);
+			rx->rx_post_cbfn(rx->bna->bnad, q1->rcb);
+		}
+		__bna_cq_start(&rxp->cq);
+	}
+
+	bna_rxf_start(&rx->rxf);
+}
+
+static void bna_rx_sm_rxf_start_wait(struct bna_rx *rx,
+				enum bna_rx_event event)
+{
+	switch (event) {
+	case RX_E_STOP:
+		bfa_fsm_set_state(rx, bna_rx_sm_rxf_stop_wait);
+		break;
+	case RX_E_FAIL:
+		bfa_fsm_set_state(rx, bna_rx_sm_stopped);
+		rx_ib_fail(rx);
+		bna_rxf_fail(&rx->rxf);
+		break;
+	case RX_E_RXF_STARTED:
+		bfa_fsm_set_state(rx, bna_rx_sm_started);
+		break;
+	default:
+		bfa_sm_fault(rx->bna, event);
+		break;
+	}
+}
+
+void
+bna_rx_sm_started_entry(struct bna_rx *rx)
+{
+	struct bna_rxp *rxp;
+	struct list_head *qe_rxp;
+
+	/* Start IB */
+	list_for_each(qe_rxp, &rx->rxp_q) {
+		rxp = (struct bna_rxp *)qe_rxp;
+		bna_ib_ack(&rxp->cq.ib->door_bell, 0);
+	}
+
+	bna_llport_admin_up(&rx->bna->port.llport);
+}
+
+void
+bna_rx_sm_started(struct bna_rx *rx, enum bna_rx_event event)
+{
+	switch (event) {
+	case RX_E_FAIL:
+		bna_llport_admin_down(&rx->bna->port.llport);
+		bfa_fsm_set_state(rx, bna_rx_sm_stopped);
+		rx_ib_fail(rx);
+		bna_rxf_fail(&rx->rxf);
+		break;
+	case RX_E_STOP:
+		bna_llport_admin_down(&rx->bna->port.llport);
+		bfa_fsm_set_state(rx, bna_rx_sm_rxf_stop_wait);
+		break;
+	default:
+		bfa_sm_fault(rx->bna, event);
+		break;
+	}
+}
+
+void
+bna_rx_sm_rxf_stop_wait_entry(struct bna_rx *rx)
+{
+	bna_rxf_stop(&rx->rxf);
+}
+
+void
+bna_rx_sm_rxf_stop_wait(struct bna_rx *rx, enum bna_rx_event event)
+{
+	switch (event) {
+	case RX_E_RXF_STOPPED:
+		bfa_fsm_set_state(rx, bna_rx_sm_rxq_stop_wait);
+		break;
+	case RX_E_RXF_STARTED:
+		/**
+		 * RxF was in the process of starting up when
+		 * RXF_E_STOP was issued. Ignore this event
+		 */
+		break;
+	case RX_E_FAIL:
+		bfa_fsm_set_state(rx, bna_rx_sm_stopped);
+		rx_ib_fail(rx);
+		bna_rxf_fail(&rx->rxf);
+		break;
+	default:
+		bfa_sm_fault(rx->bna, event);
+		break;
+	}
+
+}
+
+void
+bna_rx_sm_rxq_stop_wait_entry(struct bna_rx *rx)
+{
+	struct bna_rxp *rxp = NULL;
+	struct bna_rxq *q0 = NULL;
+	struct bna_rxq *q1 = NULL;
+	struct list_head	*qe;
+	u32 rxq_mask[2] = {0, 0};
+
+	/* Only one call to multi-rxq-stop for all RXPs in this RX */
+	bfa_wc_up(&rx->rxq_stop_wc);
+	list_for_each(qe, &rx->rxp_q) {
+		rxp = (struct bna_rxp *)qe;
+		GET_RXQS(rxp, q0, q1);
+		if (q0->rxq_id < 32)
+			rxq_mask[0] |= ((u32)1 << q0->rxq_id);
+		else
+			rxq_mask[1] |= ((u32)1 << (q0->rxq_id - 32));
+		if (q1) {
+			if (q1->rxq_id < 32)
+				rxq_mask[0] |= ((u32)1 << q1->rxq_id);
+			else
+				rxq_mask[1] |= ((u32)
+						1 << (q1->rxq_id - 32));
+		}
+	}
+
+	__bna_multi_rxq_stop(rxp, rxq_mask);
+}
+
+void
+bna_rx_sm_rxq_stop_wait(struct bna_rx *rx, enum bna_rx_event event)
+{
+	struct bna_rxp *rxp = NULL;
+	struct list_head	*qe;
+
+	switch (event) {
+	case RX_E_RXQ_STOPPED:
+		list_for_each(qe, &rx->rxp_q) {
+			rxp = (struct bna_rxp *)qe;
+			bna_ib_stop(rxp->cq.ib);
+		}
+		/* Fall through */
+	case RX_E_FAIL:
+		bfa_fsm_set_state(rx, bna_rx_sm_stopped);
+		break;
+	default:
+		bfa_sm_fault(rx->bna, event);
+		break;
+	}
+}
+
+void
+__bna_multi_rxq_stop(struct bna_rxp *rxp, u32 * rxq_id_mask)
+{
+	struct bfi_ll_q_stop_req ll_req;
+
+	bfi_h2i_set(ll_req.mh, BFI_MC_LL, BFI_LL_H2I_RXQ_STOP_REQ, 0);
+	ll_req.q_id_mask[0] = htonl(rxq_id_mask[0]);
+	ll_req.q_id_mask[1] = htonl(rxq_id_mask[1]);
+	bna_mbox_qe_fill(&rxp->mbox_qe, &ll_req, sizeof(ll_req),
+		bna_rx_cb_multi_rxq_stopped, rxp);
+	bna_mbox_send(rxp->rx->bna, &rxp->mbox_qe);
+}
+
+void
+__bna_rxq_start(struct bna_rxq *rxq)
+{
+	struct bna_rxtx_q_mem *q_mem;
+	struct bna_rxq_mem rxq_cfg, *rxq_mem;
+	struct bna_dma_addr cur_q_addr;
+	/* struct bna_doorbell_qset *qset; */
+	struct bna_qpt *qpt;
+	u32 pg_num;
+	struct bna *bna = rxq->rx->bna;
+	void __iomem *base_addr;
+	unsigned long off;
+
+	qpt = &rxq->qpt;
+	cur_q_addr = *((struct bna_dma_addr *)(qpt->kv_qpt_ptr));
+
+	rxq_cfg.pg_tbl_addr_lo = qpt->hw_qpt_ptr.lsb;
+	rxq_cfg.pg_tbl_addr_hi = qpt->hw_qpt_ptr.msb;
+	rxq_cfg.cur_q_entry_lo = cur_q_addr.lsb;
+	rxq_cfg.cur_q_entry_hi = cur_q_addr.msb;
+
+	rxq_cfg.pg_cnt_n_prd_ptr = ((u32)qpt->page_count << 16) | 0x0;
+	rxq_cfg.entry_n_pg_size = ((u32)(BFI_RXQ_WI_SIZE >> 2) << 16) |
+		(qpt->page_size >> 2);
+	rxq_cfg.sg_n_cq_n_cns_ptr =
+		((u32)(rxq->rxp->cq.cq_id & 0xff) << 16) | 0x0;
+	rxq_cfg.buf_sz_n_q_state = ((u32)rxq->buffer_size << 16) |
+		BNA_Q_IDLE_STATE;
+	rxq_cfg.next_qid = 0x0 | (0x3 << 8);
+
+	/* Write the page number register */
+	pg_num = BNA_GET_PAGE_NUM(HQM0_BLK_PG_NUM + bna->port_num,
+			HQM_RXTX_Q_RAM_BASE_OFFSET);
+	writel(pg_num, bna->regs.page_addr);
+
+	/* Write to h/w */
+	base_addr = BNA_GET_MEM_BASE_ADDR(bna->pcidev.pci_bar_kva,
+					HQM_RXTX_Q_RAM_BASE_OFFSET);
+
+	q_mem = (struct bna_rxtx_q_mem *)0;
+	rxq_mem = &q_mem[rxq->rxq_id].rxq;
+
+	off = (unsigned long)&rxq_mem->pg_tbl_addr_lo;
+	writel(htonl(rxq_cfg.pg_tbl_addr_lo), base_addr + off);
+
+	off = (unsigned long)&rxq_mem->pg_tbl_addr_hi;
+	writel(htonl(rxq_cfg.pg_tbl_addr_hi), base_addr + off);
+
+	off = (unsigned long)&rxq_mem->cur_q_entry_lo;
+	writel(htonl(rxq_cfg.cur_q_entry_lo), base_addr + off);
+
+	off = (unsigned long)&rxq_mem->cur_q_entry_hi;
+	writel(htonl(rxq_cfg.cur_q_entry_hi), base_addr + off);
+
+	off = (unsigned long)&rxq_mem->pg_cnt_n_prd_ptr;
+	writel(rxq_cfg.pg_cnt_n_prd_ptr, base_addr + off);
+
+	off = (unsigned long)&rxq_mem->entry_n_pg_size;
+	writel(rxq_cfg.entry_n_pg_size, base_addr + off);
+
+	off = (unsigned long)&rxq_mem->sg_n_cq_n_cns_ptr;
+	writel(rxq_cfg.sg_n_cq_n_cns_ptr, base_addr + off);
+
+	off = (unsigned long)&rxq_mem->buf_sz_n_q_state;
+	writel(rxq_cfg.buf_sz_n_q_state, base_addr + off);
+
+	off = (unsigned long)&rxq_mem->next_qid;
+	writel(rxq_cfg.next_qid, base_addr + off);
+
+	rxq->rcb->producer_index = 0;
+	rxq->rcb->consumer_index = 0;
+}
+
+void
+__bna_cq_start(struct bna_cq *cq)
+{
+	struct bna_cq_mem cq_cfg, *cq_mem;
+	const struct bna_qpt *qpt;
+	struct bna_dma_addr cur_q_addr;
+	u32 pg_num;
+	struct bna *bna = cq->rx->bna;
+	void __iomem *base_addr;
+	unsigned long off;
+
+	qpt = &cq->qpt;
+	cur_q_addr = *((struct bna_dma_addr *)(qpt->kv_qpt_ptr));
+
+	/*
+	 * Fill out structure, to be subsequently written
+	 * to hardware
+	 */
+	cq_cfg.pg_tbl_addr_lo = qpt->hw_qpt_ptr.lsb;
+	cq_cfg.pg_tbl_addr_hi = qpt->hw_qpt_ptr.msb;
+	cq_cfg.cur_q_entry_lo = cur_q_addr.lsb;
+	cq_cfg.cur_q_entry_hi = cur_q_addr.msb;
+
+	cq_cfg.pg_cnt_n_prd_ptr = (qpt->page_count << 16) | 0x0;
+	cq_cfg.entry_n_pg_size =
+		((u32)(BFI_CQ_WI_SIZE >> 2) << 16) | (qpt->page_size >> 2);
+	cq_cfg.int_blk_n_cns_ptr = ((((u32)cq->ib_seg_offset) << 24) |
+			((u32)(cq->ib->ib_id & 0xff)  << 16) | 0x0);
+	cq_cfg.q_state = BNA_Q_IDLE_STATE;
+
+	/* Write the page number register */
+	pg_num = BNA_GET_PAGE_NUM(HQM0_BLK_PG_NUM + bna->port_num,
+				  HQM_CQ_RAM_BASE_OFFSET);
+
+	writel(pg_num, bna->regs.page_addr);
+
+	/* H/W write */
+	base_addr = BNA_GET_MEM_BASE_ADDR(bna->pcidev.pci_bar_kva,
+					HQM_CQ_RAM_BASE_OFFSET);
+
+	cq_mem = (struct bna_cq_mem *)0;
+
+	off = (unsigned long)&cq_mem[cq->cq_id].pg_tbl_addr_lo;
+	writel(htonl(cq_cfg.pg_tbl_addr_lo), base_addr + off);
+
+	off = (unsigned long)&cq_mem[cq->cq_id].pg_tbl_addr_hi;
+	writel(htonl(cq_cfg.pg_tbl_addr_hi), base_addr + off);
+
+	off = (unsigned long)&cq_mem[cq->cq_id].cur_q_entry_lo;
+	writel(htonl(cq_cfg.cur_q_entry_lo), base_addr + off);
+
+	off = (unsigned long)&cq_mem[cq->cq_id].cur_q_entry_hi;
+	writel(htonl(cq_cfg.cur_q_entry_hi), base_addr + off);
+
+	off = (unsigned long)&cq_mem[cq->cq_id].pg_cnt_n_prd_ptr;
+	writel(cq_cfg.pg_cnt_n_prd_ptr, base_addr + off);
+
+	off = (unsigned long)&cq_mem[cq->cq_id].entry_n_pg_size;
+	writel(cq_cfg.entry_n_pg_size, base_addr + off);
+
+	off = (unsigned long)&cq_mem[cq->cq_id].int_blk_n_cns_ptr;
+	writel(cq_cfg.int_blk_n_cns_ptr, base_addr + off);
+
+	off = (unsigned long)&cq_mem[cq->cq_id].q_state;
+	writel(cq_cfg.q_state, base_addr + off);
+
+	cq->ccb->producer_index = 0;
+	*(cq->ccb->hw_producer_index) = 0;
+}
+
+void
+bna_rit_create(struct bna_rx *rx)
+{
+	struct list_head	*qe_rxp;
+	struct bna *bna;
+	struct bna_rxp *rxp;
+	struct bna_rxq *q0 = NULL;
+	struct bna_rxq *q1 = NULL;
+	int offset;
+
+	bna = rx->bna;
+
+	offset = 0;
+	list_for_each(qe_rxp, &rx->rxp_q) {
+		rxp = (struct bna_rxp *)qe_rxp;
+		GET_RXQS(rxp, q0, q1);
+		rx->rxf.rit_segment->rit[offset].large_rxq_id = q0->rxq_id;
+		rx->rxf.rit_segment->rit[offset].small_rxq_id =
+						(q1 ? q1->rxq_id : 0);
+		offset++;
+	}
+}
+
+int
+_rx_can_satisfy(struct bna_rx_mod *rx_mod,
+		struct bna_rx_config *rx_cfg)
+{
+	if ((rx_mod->rx_free_count == 0) ||
+		(rx_mod->rxp_free_count == 0) ||
+		(rx_mod->rxq_free_count == 0))
+		return 0;
+
+	if (rx_cfg->rxp_type == BNA_RXP_SINGLE) {
+		if ((rx_mod->rxp_free_count < rx_cfg->num_paths) ||
+			(rx_mod->rxq_free_count < rx_cfg->num_paths))
+				return 0;
+	} else {
+		if ((rx_mod->rxp_free_count < rx_cfg->num_paths) ||
+			(rx_mod->rxq_free_count < (2 * rx_cfg->num_paths)))
+			return 0;
+	}
+
+	if (!bna_rit_mod_can_satisfy(&rx_mod->bna->rit_mod, rx_cfg->num_paths))
+		return 0;
+
+	return 1;
+}
+
+struct bna_rxq *
+_get_free_rxq(struct bna_rx_mod *rx_mod)
+{
+	struct bna_rxq *rxq = NULL;
+	struct list_head	*qe = NULL;
+
+	bfa_q_deq(&rx_mod->rxq_free_q, &qe);
+	if (qe) {
+		rx_mod->rxq_free_count--;
+		rxq = (struct bna_rxq *)qe;
+	}
+	return rxq;
+}
+
+void
+_put_free_rxq(struct bna_rx_mod *rx_mod, struct bna_rxq *rxq)
+{
+	bfa_q_qe_init(&rxq->qe);
+	list_add_tail(&rxq->qe, &rx_mod->rxq_free_q);
+	rx_mod->rxq_free_count++;
+}
+
+struct bna_rxp *
+_get_free_rxp(struct bna_rx_mod *rx_mod)
+{
+	struct list_head	*qe = NULL;
+	struct bna_rxp *rxp = NULL;
+
+	bfa_q_deq(&rx_mod->rxp_free_q, &qe);
+	if (qe) {
+		rx_mod->rxp_free_count--;
+
+		rxp = (struct bna_rxp *)qe;
+	}
+
+	return rxp;
+}
+
+void
+_put_free_rxp(struct bna_rx_mod *rx_mod, struct bna_rxp *rxp)
+{
+	bfa_q_qe_init(&rxp->qe);
+	list_add_tail(&rxp->qe, &rx_mod->rxp_free_q);
+	rx_mod->rxp_free_count++;
+}
+
+struct bna_rx *
+_get_free_rx(struct bna_rx_mod *rx_mod)
+{
+	struct list_head	*qe = NULL;
+	struct bna_rx *rx = NULL;
+
+	bfa_q_deq(&rx_mod->rx_free_q, &qe);
+	if (qe) {
+		rx_mod->rx_free_count--;
+
+		rx = (struct bna_rx *)qe;
+		bfa_q_qe_init(qe);
+		list_add_tail(&rx->qe, &rx_mod->rx_active_q);
+	}
+
+	return rx;
+}
+
+void
+_put_free_rx(struct bna_rx_mod *rx_mod, struct bna_rx *rx)
+{
+	bfa_q_qe_init(&rx->qe);
+	list_add_tail(&rx->qe, &rx_mod->rx_free_q);
+	rx_mod->rx_free_count++;
+}
+
+void
+_rx_init(struct bna_rx *rx, struct bna *bna)
+{
+	rx->bna = bna;
+	rx->rx_flags = 0;
+
+	INIT_LIST_HEAD(&rx->rxp_q);
+
+	rx->rxq_stop_wc.wc_resume = bna_rx_cb_rxq_stopped_all;
+	rx->rxq_stop_wc.wc_cbarg = rx;
+	rx->rxq_stop_wc.wc_count = 0;
+
+	rx->stop_cbfn = NULL;
+	rx->stop_cbarg = NULL;
+}
+
+void
+_rxp_add_rxqs(struct bna_rxp *rxp,
+		struct bna_rxq *q0,
+		struct bna_rxq *q1)
+{
+	switch (rxp->type) {
+	case BNA_RXP_SINGLE:
+		rxp->rxq.single.only = q0;
+		rxp->rxq.single.reserved = NULL;
+		break;
+	case BNA_RXP_SLR:
+		rxp->rxq.slr.large = q0;
+		rxp->rxq.slr.small = q1;
+		break;
+	case BNA_RXP_HDS:
+		rxp->rxq.hds.data = q0;
+		rxp->rxq.hds.hdr = q1;
+		break;
+	default:
+		break;
+	}
+}
+
+void
+_rxq_qpt_init(struct bna_rxq *rxq,
+		struct bna_rxp *rxp,
+		u32 page_count,
+		u32 page_size,
+		struct bna_mem_descr *qpt_mem,
+		struct bna_mem_descr *swqpt_mem,
+		struct bna_mem_descr *page_mem)
+{
+	int	i;
+
+	rxq->qpt.hw_qpt_ptr.lsb = qpt_mem->dma.lsb;
+	rxq->qpt.hw_qpt_ptr.msb = qpt_mem->dma.msb;
+	rxq->qpt.kv_qpt_ptr = qpt_mem->kva;
+	rxq->qpt.page_count = page_count;
+	rxq->qpt.page_size = page_size;
+
+	rxq->rcb->sw_qpt = (void **) swqpt_mem->kva;
+
+	for (i = 0; i < rxq->qpt.page_count; i++) {
+		rxq->rcb->sw_qpt[i] = page_mem[i].kva;
+		((struct bna_dma_addr *)rxq->qpt.kv_qpt_ptr)[i].lsb =
+			page_mem[i].dma.lsb;
+		((struct bna_dma_addr *)rxq->qpt.kv_qpt_ptr)[i].msb =
+			page_mem[i].dma.msb;
+
+	}
+}
+
+void
+_rxp_cqpt_setup(struct bna_rxp *rxp,
+		u32 page_count,
+		u32 page_size,
+		struct bna_mem_descr *qpt_mem,
+		struct bna_mem_descr *swqpt_mem,
+		struct bna_mem_descr *page_mem)
+{
+	int	i;
+
+	rxp->cq.qpt.hw_qpt_ptr.lsb = qpt_mem->dma.lsb;
+	rxp->cq.qpt.hw_qpt_ptr.msb = qpt_mem->dma.msb;
+	rxp->cq.qpt.kv_qpt_ptr = qpt_mem->kva;
+	rxp->cq.qpt.page_count = page_count;
+	rxp->cq.qpt.page_size = page_size;
+
+	rxp->cq.ccb->sw_qpt = (void **) swqpt_mem->kva;
+
+	for (i = 0; i < rxp->cq.qpt.page_count; i++) {
+		rxp->cq.ccb->sw_qpt[i] = page_mem[i].kva;
+
+		((struct bna_dma_addr *)rxp->cq.qpt.kv_qpt_ptr)[i].lsb =
+			page_mem[i].dma.lsb;
+		((struct bna_dma_addr *)rxp->cq.qpt.kv_qpt_ptr)[i].msb =
+			page_mem[i].dma.msb;
+
+	}
+}
+
+void
+_rx_add_rxp(struct bna_rx *rx, struct bna_rxp *rxp)
+{
+	list_add_tail(&rxp->qe, &rx->rxp_q);
+}
+
+void
+_init_rxmod_queues(struct bna_rx_mod *rx_mod)
+{
+	INIT_LIST_HEAD(&rx_mod->rx_free_q);
+	INIT_LIST_HEAD(&rx_mod->rxq_free_q);
+	INIT_LIST_HEAD(&rx_mod->rxp_free_q);
+	INIT_LIST_HEAD(&rx_mod->rx_active_q);
+
+	rx_mod->rx_free_count = 0;
+	rx_mod->rxq_free_count = 0;
+	rx_mod->rxp_free_count = 0;
+}
+
+void
+_rx_ctor(struct bna_rx *rx, int id)
+{
+	bfa_q_qe_init(&rx->qe);
+	INIT_LIST_HEAD(&rx->rxp_q);
+	rx->bna = NULL;
+
+	rx->rxf.rxf_id = id;
+
+	/* FIXME: mbox_qe ctor()?? */
+	bfa_q_qe_init(&rx->mbox_qe.qe);
+
+	rx->stop_cbfn = NULL;
+	rx->stop_cbarg = NULL;
+}
+
+void
+bna_rx_cb_multi_rxq_stopped(void *arg, int status)
+{
+	struct bna_rxp *rxp = (struct bna_rxp *)arg;
+
+	bfa_wc_down(&rxp->rx->rxq_stop_wc);
+}
+
+void
+bna_rx_cb_rxq_stopped_all(void *arg)
+{
+	struct bna_rx *rx = (struct bna_rx *)arg;
+
+	bfa_fsm_send_event(rx, RX_E_RXQ_STOPPED);
+}
+
+void
+bna_rx_mod_cb_rx_stopped(void *arg, struct bna_rx *rx,
+			 enum bna_cb_status status)
+{
+	struct bna_rx_mod *rx_mod = (struct bna_rx_mod *)arg;
+
+	bfa_wc_down(&rx_mod->rx_stop_wc);
+}
+
+void
+bna_rx_mod_cb_rx_stopped_all(void *arg)
+{
+	struct bna_rx_mod *rx_mod = (struct bna_rx_mod *)arg;
+
+	if (rx_mod->stop_cbfn)
+		rx_mod->stop_cbfn(&rx_mod->bna->port, BNA_CB_SUCCESS);
+	rx_mod->stop_cbfn = NULL;
+}
+
+void
+bna_rx_start(struct bna_rx *rx)
+{
+	rx->rx_flags |= BNA_RX_F_PORT_ENABLED;
+	if (rx->rx_flags & BNA_RX_F_ENABLE)
+		bfa_fsm_send_event(rx, RX_E_START);
+}
+
+void
+bna_rx_stop(struct bna_rx *rx)
+{
+	rx->rx_flags &= ~BNA_RX_F_PORT_ENABLED;
+	if (rx->fsm == (bfa_fsm_t) bna_rx_sm_stopped)
+		bna_rx_mod_cb_rx_stopped(&rx->bna->rx_mod, rx, BNA_CB_SUCCESS);
+	else {
+		rx->stop_cbfn = bna_rx_mod_cb_rx_stopped;
+		rx->stop_cbarg = &rx->bna->rx_mod;
+		bfa_fsm_send_event(rx, RX_E_STOP);
+	}
+}
+
+void
+bna_rx_fail(struct bna_rx *rx)
+{
+	/* Indicate port is not enabled, and failed */
+	rx->rx_flags &= ~BNA_RX_F_PORT_ENABLED;
+	rx->rx_flags |= BNA_RX_F_PORT_FAILED;
+	bfa_fsm_send_event(rx, RX_E_FAIL);
+}
+
+void
+bna_rx_cb_rxf_started(struct bna_rx *rx, enum bna_cb_status status)
+{
+	bfa_fsm_send_event(rx, RX_E_RXF_STARTED);
+	if (rx->rxf.rxf_id < 32)
+		rx->bna->rx_mod.rxf_bmap[0] |= ((u32)1 << rx->rxf.rxf_id);
+	else
+		rx->bna->rx_mod.rxf_bmap[1] |= ((u32)
+				1 << (rx->rxf.rxf_id - 32));
+}
+
+void
+bna_rx_cb_rxf_stopped(struct bna_rx *rx, enum bna_cb_status status)
+{
+	bfa_fsm_send_event(rx, RX_E_RXF_STOPPED);
+	if (rx->rxf.rxf_id < 32)
+		rx->bna->rx_mod.rxf_bmap[0] &= ~(u32)1 << rx->rxf.rxf_id;
+	else
+		rx->bna->rx_mod.rxf_bmap[1] &= ~(u32)
+				1 << (rx->rxf.rxf_id - 32);
+}
+
+void
+bna_rx_mod_start(struct bna_rx_mod *rx_mod, enum bna_rx_type type)
+{
+	struct bna_rx *rx;
+	struct list_head *qe;
+
+	rx_mod->flags |= BNA_RX_MOD_F_PORT_STARTED;
+	if (type == BNA_RX_T_LOOPBACK)
+		rx_mod->flags |= BNA_RX_MOD_F_PORT_LOOPBACK;
+
+	list_for_each(qe, &rx_mod->rx_active_q) {
+		rx = (struct bna_rx *)qe;
+		if (rx->type == type)
+			bna_rx_start(rx);
+	}
+}
+
+void
+bna_rx_mod_stop(struct bna_rx_mod *rx_mod, enum bna_rx_type type)
+{
+	struct bna_rx *rx;
+	struct list_head *qe;
+
+	rx_mod->flags &= ~BNA_RX_MOD_F_PORT_STARTED;
+	rx_mod->flags &= ~BNA_RX_MOD_F_PORT_LOOPBACK;
+
+	rx_mod->stop_cbfn = bna_port_cb_rx_stopped;
+
+	/**
+	 * Before calling bna_rx_stop(), increment rx_stop_wc as many times
+	 * as we are going to call bna_rx_stop
+	 */
+	list_for_each(qe, &rx_mod->rx_active_q) {
+		rx = (struct bna_rx *)qe;
+		if (rx->type == type)
+			bfa_wc_up(&rx_mod->rx_stop_wc);
+	}
+
+	if (rx_mod->rx_stop_wc.wc_count == 0) {
+		rx_mod->stop_cbfn(&rx_mod->bna->port, BNA_CB_SUCCESS);
+		rx_mod->stop_cbfn = NULL;
+		return;
+	}
+
+	list_for_each(qe, &rx_mod->rx_active_q) {
+		rx = (struct bna_rx *)qe;
+		if (rx->type == type)
+			bna_rx_stop(rx);
+	}
+}
+
+void
+bna_rx_mod_fail(struct bna_rx_mod *rx_mod)
+{
+	struct bna_rx *rx;
+	struct list_head *qe;
+
+	rx_mod->flags &= ~BNA_RX_MOD_F_PORT_STARTED;
+	rx_mod->flags &= ~BNA_RX_MOD_F_PORT_LOOPBACK;
+
+	list_for_each(qe, &rx_mod->rx_active_q) {
+		rx = (struct bna_rx *)qe;
+		bna_rx_fail(rx);
+	}
+}
+
+void bna_rx_mod_init(struct bna_rx_mod *rx_mod, struct bna *bna,
+			struct bna_res_info *res_info)
+{
+	int	index;
+	struct bna_rx *rx_ptr;
+	struct bna_rxp *rxp_ptr;
+	struct bna_rxq *rxq_ptr;
+
+	rx_mod->bna = bna;
+	rx_mod->flags = 0;
+
+	rx_mod->rx = (struct bna_rx *)
+		res_info[BNA_RES_MEM_T_RX_ARRAY].res_u.mem_info.mdl[0].kva;
+	rx_mod->rxp = (struct bna_rxp *)
+		res_info[BNA_RES_MEM_T_RXP_ARRAY].res_u.mem_info.mdl[0].kva;
+	rx_mod->rxq = (struct bna_rxq *)
+		res_info[BNA_RES_MEM_T_RXQ_ARRAY].res_u.mem_info.mdl[0].kva;
+
+	/* Initialize the queues */
+	_init_rxmod_queues(rx_mod);
+
+	/* Build RX queues */
+	for (index = 0; index < BFI_MAX_RXQ; index++) {
+		rx_ptr = &rx_mod->rx[index];
+		_rx_ctor(rx_ptr, index);
+		list_add_tail(&rx_ptr->qe, &rx_mod->rx_free_q);
+		rx_mod->rx_free_count++;
+	}
+
+	/* build RX-path queue */
+	for (index = 0; index < BFI_MAX_RXQ; index++) {
+		rxp_ptr = &rx_mod->rxp[index];
+		rxp_ptr->cq.cq_id = index;
+		bfa_q_qe_init(&rxp_ptr->qe);
+		list_add_tail(&rxp_ptr->qe, &rx_mod->rxp_free_q);
+		rx_mod->rxp_free_count++;
+	}
+
+	/* build RXQ queue */
+	for (index = 0; index < BFI_MAX_RXQ; index++) {
+		rxq_ptr = &rx_mod->rxq[index];
+		rxq_ptr->rxq_id = index;
+
+		bfa_q_qe_init(&rxq_ptr->qe);
+		list_add_tail(&rxq_ptr->qe, &rx_mod->rxq_free_q);
+		rx_mod->rxq_free_count++;
+	}
+
+	rx_mod->rx_stop_wc.wc_resume = bna_rx_mod_cb_rx_stopped_all;
+	rx_mod->rx_stop_wc.wc_cbarg = rx_mod;
+	rx_mod->rx_stop_wc.wc_count = 0;
+}
+
+void
+bna_rx_mod_uninit(struct bna_rx_mod *rx_mod)
+{
+	struct list_head		*qe;
+	int i;
+
+	i = 0;
+	list_for_each(qe, &rx_mod->rx_free_q)
+		i++;
+
+	i = 0;
+	list_for_each(qe, &rx_mod->rxp_free_q)
+		i++;
+
+	i = 0;
+	list_for_each(qe, &rx_mod->rxq_free_q)
+		i++;
+
+	rx_mod->bna = NULL;
+}
+
+int
+bna_rx_state_get(struct bna_rx *rx)
+{
+	return bfa_sm_to_state(rx_sm_table, rx->fsm);
+}
+
+void
+bna_rx_res_req(struct bna_rx_config *q_cfg, struct bna_res_info *res_info)
+{
+	u32 cq_size, hq_size, dq_size;
+	u32 cpage_count, hpage_count, dpage_count;
+	struct bna_mem_info *mem_info;
+	u32 cq_depth;
+	u32 hq_depth;
+	u32 dq_depth;
+
+	dq_depth = q_cfg->q_depth;
+	hq_depth = ((q_cfg->rxp_type == BNA_RXP_SINGLE) ? 0 : q_cfg->q_depth);
+	cq_depth = dq_depth + hq_depth;
+
+	BNA_TO_POWER_OF_2_HIGH(cq_depth);
+	cq_size = cq_depth * BFI_CQ_WI_SIZE;
+	cq_size = ALIGN(cq_size, PAGE_SIZE);
+	cpage_count = SIZE_TO_PAGES(cq_size);
+
+	BNA_TO_POWER_OF_2_HIGH(dq_depth);
+	dq_size = dq_depth * BFI_RXQ_WI_SIZE;
+	dq_size = ALIGN(dq_size, PAGE_SIZE);
+	dpage_count = SIZE_TO_PAGES(dq_size);
+
+	if (BNA_RXP_SINGLE != q_cfg->rxp_type) {
+		BNA_TO_POWER_OF_2_HIGH(hq_depth);
+		hq_size = hq_depth * BFI_RXQ_WI_SIZE;
+		hq_size = ALIGN(hq_size, PAGE_SIZE);
+		hpage_count = SIZE_TO_PAGES(hq_size);
+	} else {
+		hpage_count = 0;
+	}
+
+	/* CCB structures */
+	res_info[BNA_RX_RES_MEM_T_CCB].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_CCB].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_KVA;
+	mem_info->len = sizeof(struct bna_ccb);
+	mem_info->num = q_cfg->num_paths;
+
+	/* RCB structures */
+	res_info[BNA_RX_RES_MEM_T_RCB].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_RCB].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_KVA;
+	mem_info->len = sizeof(struct bna_rcb);
+	mem_info->num = BNA_GET_RXQS(q_cfg);
+
+	/* Completion QPT */
+	res_info[BNA_RX_RES_MEM_T_CQPT].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_CQPT].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_DMA;
+	mem_info->len = cpage_count * sizeof(struct bna_dma_addr);
+	mem_info->num = q_cfg->num_paths;
+
+	/* Completion s/w QPT */
+	res_info[BNA_RX_RES_MEM_T_CSWQPT].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_CSWQPT].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_KVA;
+	mem_info->len = cpage_count * sizeof(void *);
+	mem_info->num = q_cfg->num_paths;
+
+	/* Completion QPT pages */
+	res_info[BNA_RX_RES_MEM_T_CQPT_PAGE].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_CQPT_PAGE].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_DMA;
+	mem_info->len = PAGE_SIZE;
+	mem_info->num = cpage_count * q_cfg->num_paths;
+
+	/* Data QPTs */
+	res_info[BNA_RX_RES_MEM_T_DQPT].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_DQPT].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_DMA;
+	mem_info->len = dpage_count * sizeof(struct bna_dma_addr);
+	mem_info->num = q_cfg->num_paths;
+
+	/* Data s/w QPTs */
+	res_info[BNA_RX_RES_MEM_T_DSWQPT].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_DSWQPT].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_KVA;
+	mem_info->len = dpage_count * sizeof(void *);
+	mem_info->num = q_cfg->num_paths;
+
+	/* Data QPT pages */
+	res_info[BNA_RX_RES_MEM_T_DPAGE].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_DPAGE].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_DMA;
+	mem_info->len = PAGE_SIZE;
+	mem_info->num = dpage_count * q_cfg->num_paths;
+
+	/* Hdr QPTs */
+	res_info[BNA_RX_RES_MEM_T_HQPT].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_HQPT].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_DMA;
+	mem_info->len = hpage_count * sizeof(struct bna_dma_addr);
+	mem_info->num = (hpage_count ? q_cfg->num_paths : 0);
+
+	/* Hdr s/w QPTs */
+	res_info[BNA_RX_RES_MEM_T_HSWQPT].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_HSWQPT].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_KVA;
+	mem_info->len = hpage_count * sizeof(void *);
+	mem_info->num = (hpage_count ? q_cfg->num_paths : 0);
+
+	/* Hdr QPT pages */
+	res_info[BNA_RX_RES_MEM_T_HPAGE].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_RX_RES_MEM_T_HPAGE].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_DMA;
+	mem_info->len = (hpage_count ? PAGE_SIZE : 0);
+	mem_info->num = (hpage_count ? (hpage_count * q_cfg->num_paths) : 0);
+
+	/* RX Interrupts */
+	res_info[BNA_RX_RES_T_INTR].res_type = BNA_RES_T_INTR;
+	res_info[BNA_RX_RES_T_INTR].res_u.intr_info.intr_type = BNA_INTR_T_MSIX;
+	res_info[BNA_RX_RES_T_INTR].res_u.intr_info.num = q_cfg->num_paths;
+}
+
+struct bna_rx *
+bna_rx_create(struct bna *bna, struct bnad *bnad,
+		struct bna_rx_config *rx_cfg,
+		struct bna_rx_event_cbfn *rx_cbfn,
+		struct bna_res_info *res_info,
+		void *priv)
+{
+	struct bna_rx_mod *rx_mod = &bna->rx_mod;
+	struct bna_rx *rx;
+	struct bna_rxp *rxp;
+	struct bna_rxq *q0;
+	struct bna_rxq *q1;
+	struct bna_intr_info *intr_info;
+	u32 page_count;
+	struct bna_mem_descr *ccb_mem;
+	struct bna_mem_descr *rcb_mem;
+	struct bna_mem_descr *unmapq_mem;
+	struct bna_mem_descr *cqpt_mem;
+	struct bna_mem_descr *cswqpt_mem;
+	struct bna_mem_descr *cpage_mem;
+	struct bna_mem_descr *hqpt_mem;	/* Header/Small Q qpt */
+	struct bna_mem_descr *dqpt_mem;	/* Data/Large Q qpt */
+	struct bna_mem_descr *hsqpt_mem;	/* s/w qpt for hdr */
+	struct bna_mem_descr *dsqpt_mem;	/* s/w qpt for data */
+	struct bna_mem_descr *hpage_mem;	/* hdr page mem */
+	struct bna_mem_descr *dpage_mem;	/* data page mem */
+	int i, cpage_idx = 0, dpage_idx = 0, hpage_idx = 0, ret;
+	int dpage_count, hpage_count, rcb_idx;
+	struct bna_ib_config ibcfg;
+	/* Fail if we don't have enough RXPs, RXQs */
+	if (!_rx_can_satisfy(rx_mod, rx_cfg))
+		return NULL;
+
+	/* Initialize resource pointers */
+	intr_info = &res_info[BNA_RX_RES_T_INTR].res_u.intr_info;
+	ccb_mem = &res_info[BNA_RX_RES_MEM_T_CCB].res_u.mem_info.mdl[0];
+	rcb_mem = &res_info[BNA_RX_RES_MEM_T_RCB].res_u.mem_info.mdl[0];
+	unmapq_mem = &res_info[BNA_RX_RES_MEM_T_UNMAPQ].res_u.mem_info.mdl[0];
+	cqpt_mem = &res_info[BNA_RX_RES_MEM_T_CQPT].res_u.mem_info.mdl[0];
+	cswqpt_mem = &res_info[BNA_RX_RES_MEM_T_CSWQPT].res_u.mem_info.mdl[0];
+	cpage_mem = &res_info[BNA_RX_RES_MEM_T_CQPT_PAGE].res_u.mem_info.mdl[0];
+	hqpt_mem = &res_info[BNA_RX_RES_MEM_T_HQPT].res_u.mem_info.mdl[0];
+	dqpt_mem = &res_info[BNA_RX_RES_MEM_T_DQPT].res_u.mem_info.mdl[0];
+	hsqpt_mem = &res_info[BNA_RX_RES_MEM_T_HSWQPT].res_u.mem_info.mdl[0];
+	dsqpt_mem = &res_info[BNA_RX_RES_MEM_T_DSWQPT].res_u.mem_info.mdl[0];
+	hpage_mem = &res_info[BNA_RX_RES_MEM_T_HPAGE].res_u.mem_info.mdl[0];
+	dpage_mem = &res_info[BNA_RX_RES_MEM_T_DPAGE].res_u.mem_info.mdl[0];
+
+	/* Compute q depth & page count */
+	page_count = res_info[BNA_RX_RES_MEM_T_CQPT_PAGE].res_u.mem_info.num /
+			rx_cfg->num_paths;
+
+	dpage_count = res_info[BNA_RX_RES_MEM_T_DPAGE].res_u.mem_info.num /
+			rx_cfg->num_paths;
+
+	hpage_count = res_info[BNA_RX_RES_MEM_T_HPAGE].res_u.mem_info.num /
+			rx_cfg->num_paths;
+	/* Get RX pointer */
+	rx = _get_free_rx(rx_mod);
+	_rx_init(rx, bna);
+	rx->priv = priv;
+	rx->type = rx_cfg->rx_type;
+
+	rx->rcb_setup_cbfn = rx_cbfn->rcb_setup_cbfn;
+	rx->rcb_destroy_cbfn = rx_cbfn->rcb_destroy_cbfn;
+	rx->ccb_setup_cbfn = rx_cbfn->ccb_setup_cbfn;
+	rx->ccb_destroy_cbfn = rx_cbfn->ccb_destroy_cbfn;
+	/* Following callbacks are mandatory */
+	rx->rx_cleanup_cbfn = rx_cbfn->rx_cleanup_cbfn;
+	rx->rx_post_cbfn = rx_cbfn->rx_post_cbfn;
+
+	if (rx->bna->rx_mod.flags & BNA_RX_MOD_F_PORT_STARTED) {
+		switch (rx->type) {
+		case BNA_RX_T_REGULAR:
+			if (!(rx->bna->rx_mod.flags &
+				BNA_RX_MOD_F_PORT_LOOPBACK))
+				rx->rx_flags |= BNA_RX_F_PORT_ENABLED;
+			break;
+		case BNA_RX_T_LOOPBACK:
+			if (rx->bna->rx_mod.flags & BNA_RX_MOD_F_PORT_LOOPBACK)
+				rx->rx_flags |= BNA_RX_F_PORT_ENABLED;
+			break;
+		}
+	}
+
+	for (i = 0, rcb_idx = 0; i < rx_cfg->num_paths; i++) {
+		rxp = _get_free_rxp(rx_mod);
+		rxp->type = rx_cfg->rxp_type;
+		rxp->rx = rx;
+		rxp->cq.rx = rx;
+
+		/* Get required RXQs, and queue them to rx-path */
+		q0 = _get_free_rxq(rx_mod);
+		if (BNA_RXP_SINGLE == rx_cfg->rxp_type)
+			q1 = NULL;
+		else
+			q1 = _get_free_rxq(rx_mod);
+
+		/* Initialize IB */
+		if (1 == intr_info->num) {
+			rxp->cq.ib = bna_ib_get(&bna->ib_mod,
+					intr_info->intr_type,
+					intr_info->idl[0].vector);
+			rxp->vector = intr_info->idl[0].vector;
+		} else {
+			rxp->cq.ib = bna_ib_get(&bna->ib_mod,
+					intr_info->intr_type,
+					intr_info->idl[i].vector);
+
+			/* Map the MSI-x vector used for this RXP */
+			rxp->vector = intr_info->idl[i].vector;
+		}
+
+		rxp->cq.ib_seg_offset = bna_ib_reserve_idx(rxp->cq.ib);
+
+		ibcfg.coalescing_timeo = BFI_RX_COALESCING_TIMEO;
+		ibcfg.interpkt_count = BFI_RX_INTERPKT_COUNT;
+		ibcfg.interpkt_timeo = BFI_RX_INTERPKT_TIMEO;
+		ibcfg.ctrl_flags = BFI_IB_CF_INT_ENABLE;
+
+		ret = bna_ib_config(rxp->cq.ib, &ibcfg);
+
+		/* Link rxqs to rxp */
+		_rxp_add_rxqs(rxp, q0, q1);
+
+		/* Link rxp to rx */
+		_rx_add_rxp(rx, rxp);
+
+		q0->rx = rx;
+		q0->rxp = rxp;
+
+		/* Initialize RCB for the large / data q */
+		q0->rcb = (struct bna_rcb *) rcb_mem[rcb_idx].kva;
+		RXQ_RCB_INIT(q0, rxp, rx_cfg->q_depth, bna, 0,
+			(void *)unmapq_mem[rcb_idx].kva);
+		rcb_idx++;
+		(q0)->rx_packets = (q0)->rx_bytes = 0;
+		(q0)->rx_packets_with_error = (q0)->rxbuf_alloc_failed = 0;
+
+		/* Initialize RXQs */
+		_rxq_qpt_init(q0, rxp, dpage_count, PAGE_SIZE,
+			&dqpt_mem[i], &dsqpt_mem[i], &dpage_mem[dpage_idx]);
+		q0->rcb->page_idx = dpage_idx;
+		q0->rcb->page_count = dpage_count;
+		dpage_idx += dpage_count;
+
+		/* Call bnad to complete rcb setup */
+		if (rx->rcb_setup_cbfn)
+			rx->rcb_setup_cbfn(bnad, q0->rcb);
+
+		if (q1) {
+			q1->rx = rx;
+			q1->rxp = rxp;
+
+			q1->rcb = (struct bna_rcb *) rcb_mem[rcb_idx].kva;
+			RXQ_RCB_INIT(q1, rxp, rx_cfg->q_depth, bna, 1,
+				(void *)unmapq_mem[rcb_idx].kva);
+			rcb_idx++;
+			(q1)->buffer_size = (rx_cfg)->small_buff_size;
+			(q1)->rx_packets = (q1)->rx_bytes = 0;
+			(q1)->rx_packets_with_error =
+				(q1)->rxbuf_alloc_failed = 0;
+
+			_rxq_qpt_init(q1, rxp, hpage_count, PAGE_SIZE,
+				&hqpt_mem[i], &hsqpt_mem[i],
+				&hpage_mem[hpage_idx]);
+			q1->rcb->page_idx = hpage_idx;
+			q1->rcb->page_count = hpage_count;
+			hpage_idx += hpage_count;
+
+			/* Call bnad to complete rcb setup */
+			if (rx->rcb_setup_cbfn)
+				rx->rcb_setup_cbfn(bnad, q1->rcb);
+		}
+		/* Setup RXP::CQ */
+		rxp->cq.ccb = (struct bna_ccb *) ccb_mem[i].kva;
+		_rxp_cqpt_setup(rxp, page_count, PAGE_SIZE,
+			&cqpt_mem[i], &cswqpt_mem[i], &cpage_mem[cpage_idx]);
+		rxp->cq.ccb->page_idx = cpage_idx;
+		rxp->cq.ccb->page_count = page_count;
+		cpage_idx += page_count;
+
+		rxp->cq.ccb->pkt_rate.small_pkt_cnt = 0;
+		rxp->cq.ccb->pkt_rate.large_pkt_cnt = 0;
+
+		rxp->cq.ccb->producer_index = 0;
+		rxp->cq.ccb->q_depth =	rx_cfg->q_depth +
+					((rx_cfg->rxp_type == BNA_RXP_SINGLE) ?
+					0 : rx_cfg->q_depth);
+		rxp->cq.ccb->i_dbell = &rxp->cq.ib->door_bell;
+		rxp->cq.ccb->rcb[0] = q0->rcb;
+		if (q1)
+			rxp->cq.ccb->rcb[1] = q1->rcb;
+		rxp->cq.ccb->cq = &rxp->cq;
+		rxp->cq.ccb->bnad = bna->bnad;
+		rxp->cq.ccb->hw_producer_index =
+			((volatile u32 *)rxp->cq.ib->ib_seg_host_addr_kva +
+				      (rxp->cq.ib_seg_offset * BFI_IBIDX_SIZE));
+		*(rxp->cq.ccb->hw_producer_index) = 0;
+		rxp->cq.ccb->intr_type = intr_info->intr_type;
+		rxp->cq.ccb->intr_vector = (intr_info->num == 1) ?
+						intr_info->idl[0].vector :
+						intr_info->idl[i].vector;
+		rxp->cq.ccb->rx_coalescing_timeo =
+					rxp->cq.ib->ib_config.coalescing_timeo;
+		rxp->cq.ccb->id = i;
+
+		/* Call bnad to complete CCB setup */
+		if (rx->ccb_setup_cbfn)
+			rx->ccb_setup_cbfn(bnad, rxp->cq.ccb);
+
+	} /* for each rx-path */
+
+	bna_rxf_init(&rx->rxf, rx, rx_cfg);
+
+	bfa_fsm_set_state(rx, bna_rx_sm_stopped);
+
+	return rx;
+}
+
+void
+bna_rx_destroy(struct bna_rx *rx)
+{
+	struct bna_rx_mod *rx_mod = &rx->bna->rx_mod;
+	struct bna_ib_mod *ib_mod = &rx->bna->ib_mod;
+	struct bna_rxq *q0 = NULL;
+	struct bna_rxq *q1 = NULL;
+	struct bna_rxp *rxp;
+	struct list_head *qe;
+
+	bna_rxf_uninit(&rx->rxf);
+
+	while (!list_empty(&rx->rxp_q)) {
+		bfa_q_deq(&rx->rxp_q, &rxp);
+		GET_RXQS(rxp, q0, q1);
+		/* Callback to bnad for destroying RCB */
+		if (rx->rcb_destroy_cbfn)
+			rx->rcb_destroy_cbfn(rx->bna->bnad, q0->rcb);
+		q0->rcb = NULL;
+		q0->rxp = NULL;
+		q0->rx = NULL;
+		_put_free_rxq(rx_mod, q0);
+		if (q1) {
+			/* Callback to bnad for destroying RCB */
+			if (rx->rcb_destroy_cbfn)
+				rx->rcb_destroy_cbfn(rx->bna->bnad, q1->rcb);
+			q1->rcb = NULL;
+			q1->rxp = NULL;
+			q1->rx = NULL;
+			_put_free_rxq(rx_mod, q1);
+		}
+		rxp->rxq.slr.large = NULL;
+		rxp->rxq.slr.small = NULL;
+		if (rxp->cq.ib) {
+			if (rxp->cq.ib_seg_offset != 0xff)
+				bna_ib_release_idx(rxp->cq.ib,
+						rxp->cq.ib_seg_offset);
+			bna_ib_put(ib_mod, rxp->cq.ib);
+			rxp->cq.ib = NULL;
+		}
+		/* Callback to bnad for destroying CCB */
+		if (rx->ccb_destroy_cbfn)
+			rx->ccb_destroy_cbfn(rx->bna->bnad, rxp->cq.ccb);
+		rxp->cq.ccb = NULL;
+		rxp->rx = NULL;
+		_put_free_rxp(rx_mod, rxp);
+	}
+
+	list_for_each(qe, &rx_mod->rx_active_q) {
+		if (qe == &rx->qe) {
+			list_del(&rx->qe);
+			bfa_q_qe_init(&rx->qe);
+			break;
+		}
+	}
+
+	rx->bna = NULL;
+	rx->priv = NULL;
+	_put_free_rx(rx_mod, rx);
+}
+
+void
+bna_rx_enable(struct bna_rx *rx)
+{
+	if (rx->fsm != (bfa_sm_t)bna_rx_sm_stopped)
+		return;
+
+	rx->rx_flags |= BNA_RX_F_ENABLE;
+	if (rx->rx_flags & BNA_RX_F_PORT_ENABLED)
+		bfa_fsm_send_event(rx, RX_E_START);
+}
+
+void
+bna_rx_disable(struct bna_rx *rx, enum bna_cleanup_type type,
+		void (*cbfn)(void *, struct bna_rx *,
+				enum bna_cb_status))
+{
+	if (type == BNA_SOFT_CLEANUP) {
+		/* h/w should not be accessed. Treat we're stopped */
+		(*cbfn)(rx->bna->bnad, rx, BNA_CB_SUCCESS);
+	} else {
+		rx->stop_cbfn = cbfn;
+		rx->stop_cbarg = rx->bna->bnad;
+
+		rx->rx_flags &= ~BNA_RX_F_ENABLE;
+
+		bfa_fsm_send_event(rx, RX_E_STOP);
+	}
+}
+
+/**
+ * TX
+ */
+#define call_tx_stop_cbfn(tx, status)\
+do {\
+	if ((tx)->stop_cbfn)\
+		(tx)->stop_cbfn((tx)->stop_cbarg, (tx), status);\
+	(tx)->stop_cbfn = NULL;\
+	(tx)->stop_cbarg = NULL;\
+} while (0)
+
+#define call_tx_prio_change_cbfn(tx, status)\
+do {\
+	if ((tx)->prio_change_cbfn)\
+		(tx)->prio_change_cbfn((tx)->bna->bnad, (tx), status);\
+	(tx)->prio_change_cbfn = NULL;\
+} while (0)
+
+static void bna_tx_mod_cb_tx_stopped(void *tx_mod, struct bna_tx *tx,
+					enum bna_cb_status status);
+static void bna_tx_cb_txq_stopped(void *arg, int status);
+static void bna_tx_cb_stats_cleared(void *arg, int status);
+static void __bna_tx_stop(struct bna_tx *tx);
+static void __bna_tx_start(struct bna_tx *tx);
+static void __bna_txf_stat_clr(struct bna_tx *tx);
+
+enum bna_tx_event {
+	TX_E_START			= 1,
+	TX_E_STOP			= 2,
+	TX_E_FAIL			= 3,
+	TX_E_TXQ_STOPPED		= 4,
+	TX_E_PRIO_CHANGE		= 5,
+	TX_E_STAT_CLEARED		= 6,
+};
+
+enum bna_tx_state {
+	BNA_TX_STOPPED			= 1,
+	BNA_TX_STARTED			= 2,
+	BNA_TX_TXQ_STOP_WAIT		= 3,
+	BNA_TX_PRIO_STOP_WAIT		= 4,
+	BNA_TX_STAT_CLR_WAIT		= 5,
+};
+
+bfa_fsm_state_decl(bna_tx, stopped, struct bna_tx,
+			enum bna_tx_event);
+bfa_fsm_state_decl(bna_tx, started, struct bna_tx,
+			enum bna_tx_event);
+bfa_fsm_state_decl(bna_tx, txq_stop_wait, struct bna_tx,
+			enum bna_tx_event);
+bfa_fsm_state_decl(bna_tx, prio_stop_wait, struct bna_tx,
+			enum bna_tx_event);
+bfa_fsm_state_decl(bna_tx, stat_clr_wait, struct bna_tx,
+			enum bna_tx_event);
+
+static struct bfa_sm_table tx_sm_table[] = {
+	{BFA_SM(bna_tx_sm_stopped), BNA_TX_STOPPED},
+	{BFA_SM(bna_tx_sm_started), BNA_TX_STARTED},
+	{BFA_SM(bna_tx_sm_txq_stop_wait), BNA_TX_TXQ_STOP_WAIT},
+	{BFA_SM(bna_tx_sm_prio_stop_wait), BNA_TX_PRIO_STOP_WAIT},
+	{BFA_SM(bna_tx_sm_stat_clr_wait), BNA_TX_STAT_CLR_WAIT},
+};
+
+static void
+bna_tx_sm_stopped_entry(struct bna_tx *tx)
+{
+	struct bna_txq *txq;
+	struct list_head		 *qe;
+
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		(tx->tx_cleanup_cbfn)(tx->bna->bnad, txq->tcb);
+	}
+
+	call_tx_stop_cbfn(tx, BNA_CB_SUCCESS);
+}
+
+static void
+bna_tx_sm_stopped(struct bna_tx *tx, enum bna_tx_event event)
+{
+	switch (event) {
+	case TX_E_START:
+		bfa_fsm_set_state(tx, bna_tx_sm_started);
+		break;
+
+	case TX_E_STOP:
+		bfa_fsm_set_state(tx, bna_tx_sm_stopped);
+		break;
+
+	case TX_E_FAIL:
+		/* No-op */
+		break;
+
+	case TX_E_PRIO_CHANGE:
+		call_tx_prio_change_cbfn(tx, BNA_CB_SUCCESS);
+		break;
+
+	case TX_E_TXQ_STOPPED:
+		/**
+		 * This event is received due to flushing of mbox when
+		 * device fails
+		 */
+		/* No-op */
+		break;
+
+	default:
+		bfa_sm_fault(tx->bna, event);
+	}
+}
+
+static void
+bna_tx_sm_started_entry(struct bna_tx *tx)
+{
+	struct bna_txq *txq;
+	struct list_head		 *qe;
+
+	__bna_tx_start(tx);
+
+	/* Start IB */
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		bna_ib_ack(&txq->ib->door_bell, 0);
+	}
+}
+
+static void
+bna_tx_sm_started(struct bna_tx *tx, enum bna_tx_event event)
+{
+	struct bna_txq *txq;
+	struct list_head		 *qe;
+
+	switch (event) {
+	case TX_E_STOP:
+		bfa_fsm_set_state(tx, bna_tx_sm_txq_stop_wait);
+		__bna_tx_stop(tx);
+		break;
+
+	case TX_E_FAIL:
+		list_for_each(qe, &tx->txq_q) {
+			txq = (struct bna_txq *)qe;
+			bna_ib_fail(txq->ib);
+			(tx->tx_stall_cbfn)(tx->bna->bnad, txq->tcb);
+		}
+		bfa_fsm_set_state(tx, bna_tx_sm_stopped);
+		break;
+
+	case TX_E_PRIO_CHANGE:
+		bfa_fsm_set_state(tx, bna_tx_sm_prio_stop_wait);
+		break;
+
+	default:
+		bfa_sm_fault(tx->bna, event);
+	}
+}
+
+static void
+bna_tx_sm_txq_stop_wait_entry(struct bna_tx *tx)
+{
+}
+
+static void
+bna_tx_sm_txq_stop_wait(struct bna_tx *tx, enum bna_tx_event event)
+{
+	struct bna_txq *txq;
+	struct list_head		 *qe;
+
+	switch (event) {
+	case TX_E_FAIL:
+		bfa_fsm_set_state(tx, bna_tx_sm_stopped);
+		break;
+
+	case TX_E_TXQ_STOPPED:
+		list_for_each(qe, &tx->txq_q) {
+			txq = (struct bna_txq *)qe;
+			bna_ib_stop(txq->ib);
+		}
+		bfa_fsm_set_state(tx, bna_tx_sm_stat_clr_wait);
+		break;
+
+	case TX_E_PRIO_CHANGE:
+		/* No-op */
+		break;
+
+	default:
+		bfa_sm_fault(tx->bna, event);
+	}
+}
+
+static void
+bna_tx_sm_prio_stop_wait_entry(struct bna_tx *tx)
+{
+	__bna_tx_stop(tx);
+}
+
+static void
+bna_tx_sm_prio_stop_wait(struct bna_tx *tx, enum bna_tx_event event)
+{
+	struct bna_txq *txq;
+	struct list_head		 *qe;
+
+	switch (event) {
+	case TX_E_STOP:
+		bfa_fsm_set_state(tx, bna_tx_sm_txq_stop_wait);
+		break;
+
+	case TX_E_FAIL:
+		call_tx_prio_change_cbfn(tx, BNA_CB_FAIL);
+		bfa_fsm_set_state(tx, bna_tx_sm_stopped);
+		break;
+
+	case TX_E_TXQ_STOPPED:
+		list_for_each(qe, &tx->txq_q) {
+			txq = (struct bna_txq *)qe;
+			bna_ib_stop(txq->ib);
+			(tx->tx_cleanup_cbfn)(tx->bna->bnad, txq->tcb);
+		}
+		call_tx_prio_change_cbfn(tx, BNA_CB_SUCCESS);
+		bfa_fsm_set_state(tx, bna_tx_sm_started);
+		break;
+
+	case TX_E_PRIO_CHANGE:
+		/* No-op */
+		break;
+
+	default:
+		bfa_sm_fault(tx->bna, event);
+	}
+}
+
+static void
+bna_tx_sm_stat_clr_wait_entry(struct bna_tx *tx)
+{
+	__bna_txf_stat_clr(tx);
+}
+
+static void
+bna_tx_sm_stat_clr_wait(struct bna_tx *tx, enum bna_tx_event event)
+{
+	switch (event) {
+	case TX_E_FAIL:
+	case TX_E_STAT_CLEARED:
+		bfa_fsm_set_state(tx, bna_tx_sm_stopped);
+		break;
+
+	default:
+		bfa_sm_fault(tx->bna, event);
+	}
+}
+
+static void
+__bna_txq_start(struct bna_tx *tx, struct bna_txq *txq)
+{
+	struct bna_rxtx_q_mem *q_mem;
+	struct bna_txq_mem txq_cfg;
+	struct bna_txq_mem *txq_mem;
+	struct bna_dma_addr cur_q_addr;
+	u32 pg_num;
+	void __iomem *base_addr;
+	unsigned long off;
+
+	/* Fill out structure, to be subsequently written to hardware */
+	txq_cfg.pg_tbl_addr_lo = txq->qpt.hw_qpt_ptr.lsb;
+	txq_cfg.pg_tbl_addr_hi = txq->qpt.hw_qpt_ptr.msb;
+	cur_q_addr = *((struct bna_dma_addr *)(txq->qpt.kv_qpt_ptr));
+	txq_cfg.cur_q_entry_lo = cur_q_addr.lsb;
+	txq_cfg.cur_q_entry_hi = cur_q_addr.msb;
+
+	txq_cfg.pg_cnt_n_prd_ptr = (txq->qpt.page_count << 16) | 0x0;
+
+	txq_cfg.entry_n_pg_size = ((u32)(BFI_TXQ_WI_SIZE >> 2) << 16) |
+			(txq->qpt.page_size >> 2);
+	txq_cfg.int_blk_n_cns_ptr = ((((u32)txq->ib_seg_offset) << 24) |
+			((u32)(txq->ib->ib_id & 0xff) << 16) | 0x0);
+
+	txq_cfg.cns_ptr2_n_q_state = BNA_Q_IDLE_STATE;
+	txq_cfg.nxt_qid_n_fid_n_pri = (((tx->txf.txf_id & 0x3f) << 3) |
+			(txq->priority & 0x3));
+	txq_cfg.wvc_n_cquota_n_rquota =
+			((((u32)BFI_TX_MAX_WRR_QUOTA & 0xfff) << 12) |
+			(BFI_TX_MAX_WRR_QUOTA & 0xfff));
+
+	/* Setup the page and write to H/W */
+
+	pg_num = BNA_GET_PAGE_NUM(HQM0_BLK_PG_NUM + tx->bna->port_num,
+			HQM_RXTX_Q_RAM_BASE_OFFSET);
+	writel(pg_num, tx->bna->regs.page_addr);
+
+	base_addr = BNA_GET_MEM_BASE_ADDR(tx->bna->pcidev.pci_bar_kva,
+					HQM_RXTX_Q_RAM_BASE_OFFSET);
+	q_mem = (struct bna_rxtx_q_mem *)0;
+	txq_mem = &q_mem[txq->txq_id].txq;
+
+	/*
+	 * The following 4 lines, is a hack b'cos the H/W needs to read
+	 * these DMA addresses as little endian
+	 */
+
+	off = (unsigned long)&txq_mem->pg_tbl_addr_lo;
+	writel(htonl(txq_cfg.pg_tbl_addr_lo), base_addr + off);
+
+	off = (unsigned long)&txq_mem->pg_tbl_addr_hi;
+	writel(htonl(txq_cfg.pg_tbl_addr_hi), base_addr + off);
+
+	off = (unsigned long)&txq_mem->cur_q_entry_lo;
+	writel(htonl(txq_cfg.cur_q_entry_lo), base_addr + off);
+
+	off = (unsigned long)&txq_mem->cur_q_entry_hi;
+	writel(htonl(txq_cfg.cur_q_entry_hi), base_addr + off);
+
+	off = (unsigned long)&txq_mem->pg_cnt_n_prd_ptr;
+	writel(txq_cfg.pg_cnt_n_prd_ptr, base_addr + off);
+
+	off = (unsigned long)&txq_mem->entry_n_pg_size;
+	writel(txq_cfg.entry_n_pg_size, base_addr + off);
+
+	off = (unsigned long)&txq_mem->int_blk_n_cns_ptr;
+	writel(txq_cfg.int_blk_n_cns_ptr, base_addr + off);
+
+	off = (unsigned long)&txq_mem->cns_ptr2_n_q_state;
+	writel(txq_cfg.cns_ptr2_n_q_state, base_addr + off);
+
+	off = (unsigned long)&txq_mem->nxt_qid_n_fid_n_pri;
+	writel(txq_cfg.nxt_qid_n_fid_n_pri, base_addr + off);
+
+	off = (unsigned long)&txq_mem->wvc_n_cquota_n_rquota;
+	writel(txq_cfg.wvc_n_cquota_n_rquota, base_addr + off);
+
+	txq->tcb->producer_index = 0;
+	txq->tcb->consumer_index = 0;
+	*(txq->tcb->hw_consumer_index) = 0;
+
+}
+
+static void
+__bna_txq_stop(struct bna_tx *tx, struct bna_txq *txq)
+{
+	struct bfi_ll_q_stop_req ll_req;
+	u32 bit_mask[2] = {0, 0};
+	if (txq->txq_id < 32)
+		bit_mask[0] = (u32)1 << txq->txq_id;
+	else
+		bit_mask[1] = (u32)1 << (txq->txq_id - 32);
+
+	memset(&ll_req, 0, sizeof(ll_req));
+	ll_req.mh.msg_class = BFI_MC_LL;
+	ll_req.mh.msg_id = BFI_LL_H2I_TXQ_STOP_REQ;
+	ll_req.mh.mtag.h2i.lpu_id = 0;
+	ll_req.q_id_mask[0] = htonl(bit_mask[0]);
+	ll_req.q_id_mask[1] = htonl(bit_mask[1]);
+
+	bna_mbox_qe_fill(&tx->mbox_qe, &ll_req, sizeof(ll_req),
+			bna_tx_cb_txq_stopped, tx);
+
+	bna_mbox_send(tx->bna, &tx->mbox_qe);
+}
+
+static void
+__bna_txf_start(struct bna_tx *tx)
+{
+	struct bna_tx_fndb_ram *tx_fndb;
+	struct bna_txf *txf = &tx->txf;
+	void __iomem *base_addr;
+	unsigned long off;
+
+	writel(BNA_GET_PAGE_NUM(LUT0_MEM_BLK_BASE_PG_NUM +
+			(tx->bna->port_num * 2), TX_FNDB_RAM_BASE_OFFSET),
+			tx->bna->regs.page_addr);
+
+	base_addr = BNA_GET_MEM_BASE_ADDR(tx->bna->pcidev.pci_bar_kva,
+					TX_FNDB_RAM_BASE_OFFSET);
+
+	tx_fndb = (struct bna_tx_fndb_ram *)0;
+	off = (unsigned long)&tx_fndb[txf->txf_id].vlan_n_ctrl_flags;
+
+	writel(((u32)txf->vlan << 16) | txf->ctrl_flags,
+			base_addr + off);
+
+	if (tx->txf.txf_id < 32)
+		tx->bna->tx_mod.txf_bmap[0] |= ((u32)1 << tx->txf.txf_id);
+	else
+		tx->bna->tx_mod.txf_bmap[1] |= ((u32)
+						 1 << (tx->txf.txf_id - 32));
+}
+
+static void
+__bna_txf_stop(struct bna_tx *tx)
+{
+	struct bna_tx_fndb_ram *tx_fndb;
+	u32 page_num;
+	u32 ctl_flags;
+	struct bna_txf *txf = &tx->txf;
+	void __iomem *base_addr;
+	unsigned long off;
+
+	/* retrieve the running txf_flags & turn off enable bit */
+	page_num = BNA_GET_PAGE_NUM(LUT0_MEM_BLK_BASE_PG_NUM +
+			(tx->bna->port_num * 2), TX_FNDB_RAM_BASE_OFFSET);
+	writel(page_num, tx->bna->regs.page_addr);
+
+	base_addr = BNA_GET_MEM_BASE_ADDR(tx->bna->pcidev.pci_bar_kva,
+					TX_FNDB_RAM_BASE_OFFSET);
+	tx_fndb = (struct bna_tx_fndb_ram *)0;
+	off = (unsigned long)&tx_fndb[txf->txf_id].vlan_n_ctrl_flags;
+
+	ctl_flags = readl(base_addr + off);
+	ctl_flags &= ~BFI_TXF_CF_ENABLE;
+
+	writel(ctl_flags, base_addr + off);
+
+	if (tx->txf.txf_id < 32)
+		tx->bna->tx_mod.txf_bmap[0] &= ~((u32)1 << tx->txf.txf_id);
+	else
+		tx->bna->tx_mod.txf_bmap[0] &= ~((u32)
+						 1 << (tx->txf.txf_id - 32));
+}
+
+static void
+__bna_txf_stat_clr(struct bna_tx *tx)
+{
+	struct bfi_ll_stats_req ll_req;
+	u32 txf_bmap[2] = {0, 0};
+	if (tx->txf.txf_id < 32)
+		txf_bmap[0] = ((u32)1 << tx->txf.txf_id);
+	else
+		txf_bmap[1] = ((u32)1 << (tx->txf.txf_id - 32));
+	bfi_h2i_set(ll_req.mh, BFI_MC_LL, BFI_LL_H2I_STATS_CLEAR_REQ, 0);
+	ll_req.stats_mask = 0;
+	ll_req.rxf_id_mask[0] = 0;
+	ll_req.rxf_id_mask[1] =	0;
+	ll_req.txf_id_mask[0] =	htonl(txf_bmap[0]);
+	ll_req.txf_id_mask[1] =	htonl(txf_bmap[1]);
+
+	bna_mbox_qe_fill(&tx->mbox_qe, &ll_req, sizeof(ll_req),
+			bna_tx_cb_stats_cleared, tx);
+	bna_mbox_send(tx->bna, &tx->mbox_qe);
+}
+
+static void
+__bna_tx_start(struct bna_tx *tx)
+{
+	struct bna_txq *txq;
+	struct list_head		 *qe;
+
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		bna_ib_start(txq->ib);
+		__bna_txq_start(tx, txq);
+	}
+
+	__bna_txf_start(tx);
+
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		txq->tcb->priority = txq->priority;
+		(tx->tx_resume_cbfn)(tx->bna->bnad, txq->tcb);
+	}
+}
+
+static void
+__bna_tx_stop(struct bna_tx *tx)
+{
+	struct bna_txq *txq;
+	struct list_head		 *qe;
+
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		(tx->tx_stall_cbfn)(tx->bna->bnad, txq->tcb);
+	}
+
+	__bna_txf_stop(tx);
+
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		bfa_wc_up(&tx->txq_stop_wc);
+	}
+
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		__bna_txq_stop(tx, txq);
+	}
+}
+
+static void
+bna_txq_qpt_setup(struct bna_txq *txq, int page_count, int page_size,
+		struct bna_mem_descr *qpt_mem,
+		struct bna_mem_descr *swqpt_mem,
+		struct bna_mem_descr *page_mem)
+{
+	int i;
+
+	txq->qpt.hw_qpt_ptr.lsb = qpt_mem->dma.lsb;
+	txq->qpt.hw_qpt_ptr.msb = qpt_mem->dma.msb;
+	txq->qpt.kv_qpt_ptr = qpt_mem->kva;
+	txq->qpt.page_count = page_count;
+	txq->qpt.page_size = page_size;
+
+	txq->tcb->sw_qpt = (void **) swqpt_mem->kva;
+
+	for (i = 0; i < page_count; i++) {
+		txq->tcb->sw_qpt[i] = page_mem[i].kva;
+
+		((struct bna_dma_addr *)txq->qpt.kv_qpt_ptr)[i].lsb =
+			page_mem[i].dma.lsb;
+		((struct bna_dma_addr *)txq->qpt.kv_qpt_ptr)[i].msb =
+			page_mem[i].dma.msb;
+
+	}
+}
+
+static void
+bna_tx_free(struct bna_tx *tx)
+{
+	struct bna_tx_mod *tx_mod = &tx->bna->tx_mod;
+	struct bna_txq *txq;
+	struct bna_ib_mod *ib_mod = &tx->bna->ib_mod;
+	struct list_head *qe;
+
+	while (!list_empty(&tx->txq_q)) {
+		bfa_q_deq(&tx->txq_q, &txq);
+		bfa_q_qe_init(&txq->qe);
+		if (txq->ib) {
+			if (txq->ib_seg_offset != -1)
+				bna_ib_release_idx(txq->ib,
+						txq->ib_seg_offset);
+			bna_ib_put(ib_mod, txq->ib);
+			txq->ib = NULL;
+		}
+		txq->tcb = NULL;
+		txq->tx = NULL;
+		list_add_tail(&txq->qe, &tx_mod->txq_free_q);
+	}
+
+	list_for_each(qe, &tx_mod->tx_active_q) {
+		if (qe == &tx->qe) {
+			list_del(&tx->qe);
+			bfa_q_qe_init(&tx->qe);
+			break;
+		}
+	}
+
+	tx->bna = NULL;
+	tx->priv = NULL;
+	list_add_tail(&tx->qe, &tx_mod->tx_free_q);
+}
+
+static void
+bna_tx_cb_txq_stopped(void *arg, int status)
+{
+	struct bna_tx *tx = (struct bna_tx *)arg;
+
+	bfa_q_qe_init(&tx->mbox_qe.qe);
+	bfa_wc_down(&tx->txq_stop_wc);
+}
+
+static void
+bna_tx_cb_txq_stopped_all(void *arg)
+{
+	struct bna_tx *tx = (struct bna_tx *)arg;
+
+	bfa_fsm_send_event(tx, TX_E_TXQ_STOPPED);
+}
+
+static void
+bna_tx_cb_stats_cleared(void *arg, int status)
+{
+	struct bna_tx *tx = (struct bna_tx *)arg;
+
+	bfa_q_qe_init(&tx->mbox_qe.qe);
+
+	bfa_fsm_send_event(tx, TX_E_STAT_CLEARED);
+}
+
+static void
+bna_tx_start(struct bna_tx *tx)
+{
+	tx->flags |= BNA_TX_F_PORT_STARTED;
+	if (tx->flags & BNA_TX_F_ENABLED)
+		bfa_fsm_send_event(tx, TX_E_START);
+}
+
+static void
+bna_tx_stop(struct bna_tx *tx)
+{
+	tx->stop_cbfn = bna_tx_mod_cb_tx_stopped;
+	tx->stop_cbarg = &tx->bna->tx_mod;
+
+	tx->flags &= ~BNA_TX_F_PORT_STARTED;
+	bfa_fsm_send_event(tx, TX_E_STOP);
+}
+
+static void
+bna_tx_fail(struct bna_tx *tx)
+{
+	tx->flags &= ~BNA_TX_F_PORT_STARTED;
+	bfa_fsm_send_event(tx, TX_E_FAIL);
+}
+
+void
+bna_tx_prio_changed(struct bna_tx *tx, int prio)
+{
+	struct bna_txq *txq;
+	struct list_head		 *qe;
+
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		txq->priority = prio;
+	}
+
+	bfa_fsm_send_event(tx, TX_E_PRIO_CHANGE);
+}
+
+static void
+bna_tx_cee_link_status(struct bna_tx *tx, int cee_link)
+{
+	if (cee_link)
+		tx->flags |= BNA_TX_F_PRIO_LOCK;
+	else
+		tx->flags &= ~BNA_TX_F_PRIO_LOCK;
+}
+
+static void
+bna_tx_mod_cb_tx_stopped(void *arg, struct bna_tx *tx,
+			enum bna_cb_status status)
+{
+	struct bna_tx_mod *tx_mod = (struct bna_tx_mod *)arg;
+
+	bfa_wc_down(&tx_mod->tx_stop_wc);
+}
+
+static void
+bna_tx_mod_cb_tx_stopped_all(void *arg)
+{
+	struct bna_tx_mod *tx_mod = (struct bna_tx_mod *)arg;
+
+	if (tx_mod->stop_cbfn)
+		tx_mod->stop_cbfn(&tx_mod->bna->port, BNA_CB_SUCCESS);
+	tx_mod->stop_cbfn = NULL;
+}
+
+void
+bna_tx_res_req(int num_txq, int txq_depth, struct bna_res_info *res_info)
+{
+	u32 q_size;
+	u32 page_count;
+	struct bna_mem_info *mem_info;
+
+	res_info[BNA_TX_RES_MEM_T_TCB].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_TX_RES_MEM_T_TCB].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_KVA;
+	mem_info->len = sizeof(struct bna_tcb);
+	mem_info->num = num_txq;
+
+	q_size = txq_depth * BFI_TXQ_WI_SIZE;
+	q_size = ALIGN(q_size, PAGE_SIZE);
+	page_count = q_size >> PAGE_SHIFT;
+
+	res_info[BNA_TX_RES_MEM_T_QPT].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_TX_RES_MEM_T_QPT].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_DMA;
+	mem_info->len = page_count * sizeof(struct bna_dma_addr);
+	mem_info->num = num_txq;
+
+	res_info[BNA_TX_RES_MEM_T_SWQPT].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_TX_RES_MEM_T_SWQPT].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_KVA;
+	mem_info->len = page_count * sizeof(void *);
+	mem_info->num = num_txq;
+
+	res_info[BNA_TX_RES_MEM_T_PAGE].res_type = BNA_RES_T_MEM;
+	mem_info = &res_info[BNA_TX_RES_MEM_T_PAGE].res_u.mem_info;
+	mem_info->mem_type = BNA_MEM_T_DMA;
+	mem_info->len = PAGE_SIZE;
+	mem_info->num = num_txq * page_count;
+
+	res_info[BNA_TX_RES_INTR_T_TXCMPL].res_type = BNA_RES_T_INTR;
+	res_info[BNA_TX_RES_INTR_T_TXCMPL].res_u.intr_info.intr_type =
+			BNA_INTR_T_MSIX;
+	res_info[BNA_TX_RES_INTR_T_TXCMPL].res_u.intr_info.num = num_txq;
+}
+
+struct bna_tx *
+bna_tx_create(struct bna *bna, struct bnad *bnad,
+		struct bna_tx_config *tx_cfg,
+		struct bna_tx_event_cbfn *tx_cbfn,
+		struct bna_res_info *res_info, void *priv)
+{
+	struct bna_intr_info *intr_info;
+	struct bna_tx_mod *tx_mod = &bna->tx_mod;
+	struct bna_tx *tx;
+	struct bna_txq *txq;
+	struct list_head *qe;
+	struct bna_ib_mod *ib_mod = &bna->ib_mod;
+	struct bna_doorbell_qset *qset;
+	struct bna_ib_config ib_config;
+	int page_count;
+	int page_size;
+	int page_idx;
+	int i;
+	unsigned long off;
+
+	intr_info = &res_info[BNA_TX_RES_INTR_T_TXCMPL].res_u.intr_info;
+	page_count = (res_info[BNA_TX_RES_MEM_T_PAGE].res_u.mem_info.num) /
+			tx_cfg->num_txq;
+	page_size = res_info[BNA_TX_RES_MEM_T_PAGE].res_u.mem_info.len;
+
+	/**
+	 * Get resources
+	 */
+
+	if ((intr_info->num != 1) && (intr_info->num != tx_cfg->num_txq))
+		return NULL;
+
+	/* Tx */
+
+	if (list_empty(&tx_mod->tx_free_q))
+		return NULL;
+	bfa_q_deq(&tx_mod->tx_free_q, &tx);
+	bfa_q_qe_init(&tx->qe);
+
+	/* TxQs */
+
+	INIT_LIST_HEAD(&tx->txq_q);
+	for (i = 0; i < tx_cfg->num_txq; i++) {
+		if (list_empty(&tx_mod->txq_free_q))
+			goto err_return;
+
+		bfa_q_deq(&tx_mod->txq_free_q, &txq);
+		bfa_q_qe_init(&txq->qe);
+		list_add_tail(&txq->qe, &tx->txq_q);
+		txq->ib = NULL;
+		txq->ib_seg_offset = -1;
+		txq->tx = tx;
+	}
+
+	/* IBs */
+	i = 0;
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+
+		if (intr_info->num == 1)
+			txq->ib = bna_ib_get(ib_mod, intr_info->intr_type,
+						intr_info->idl[0].vector);
+		else
+			txq->ib = bna_ib_get(ib_mod, intr_info->intr_type,
+						intr_info->idl[i].vector);
+
+		if (txq->ib == NULL)
+			goto err_return;
+
+		txq->ib_seg_offset = bna_ib_reserve_idx(txq->ib);
+		if (txq->ib_seg_offset == -1)
+			goto err_return;
+
+		i++;
+	}
+
+	/*
+	 * Initialize
+	 */
+
+	/* Tx */
+
+	tx->tcb_setup_cbfn = tx_cbfn->tcb_setup_cbfn;
+	tx->tcb_destroy_cbfn = tx_cbfn->tcb_destroy_cbfn;
+	/* Following callbacks are mandatory */
+	tx->tx_stall_cbfn = tx_cbfn->tx_stall_cbfn;
+	tx->tx_resume_cbfn = tx_cbfn->tx_resume_cbfn;
+	tx->tx_cleanup_cbfn = tx_cbfn->tx_cleanup_cbfn;
+
+	list_add_tail(&tx->qe, &tx_mod->tx_active_q);
+	tx->bna = bna;
+	tx->priv = priv;
+	tx->txq_stop_wc.wc_resume = bna_tx_cb_txq_stopped_all;
+	tx->txq_stop_wc.wc_cbarg = tx;
+	tx->txq_stop_wc.wc_count = 0;
+
+	tx->type = tx_cfg->tx_type;
+
+	tx->flags = 0;
+	if (tx->bna->tx_mod.flags & BNA_TX_MOD_F_PORT_STARTED) {
+		switch (tx->type) {
+		case BNA_TX_T_REGULAR:
+			if (!(tx->bna->tx_mod.flags &
+				BNA_TX_MOD_F_PORT_LOOPBACK))
+				tx->flags |= BNA_TX_F_PORT_STARTED;
+			break;
+		case BNA_TX_T_LOOPBACK:
+			if (tx->bna->tx_mod.flags & BNA_TX_MOD_F_PORT_LOOPBACK)
+				tx->flags |= BNA_TX_F_PORT_STARTED;
+			break;
+		}
+	}
+	if (tx->bna->tx_mod.cee_link)
+		tx->flags |= BNA_TX_F_PRIO_LOCK;
+
+	/* TxQ */
+
+	i = 0;
+	page_idx = 0;
+	list_for_each(qe, &tx->txq_q) {
+		txq = (struct bna_txq *)qe;
+		txq->priority = tx_mod->priority;
+		txq->tcb = (struct bna_tcb *)
+		  res_info[BNA_TX_RES_MEM_T_TCB].res_u.mem_info.mdl[i].kva;
+		txq->tx_packets = 0;
+		txq->tx_bytes = 0;
+
+		/* IB */
+
+		ib_config.coalescing_timeo = BFI_TX_COALESCING_TIMEO;
+		ib_config.interpkt_timeo = 0; /* Not used */
+		ib_config.interpkt_count = BFI_TX_INTERPKT_COUNT;
+		ib_config.ctrl_flags = (BFI_IB_CF_INTER_PKT_DMA |
+					BFI_IB_CF_INT_ENABLE |
+					BFI_IB_CF_COALESCING_MODE);
+		bna_ib_config(txq->ib, &ib_config);
+
+		/* TCB */
+
+		txq->tcb->producer_index = 0;
+		txq->tcb->consumer_index = 0;
+		txq->tcb->hw_consumer_index = (volatile u32 *)
+			((volatile u8 *)txq->ib->ib_seg_host_addr_kva +
+			 (txq->ib_seg_offset * BFI_IBIDX_SIZE));
+		*(txq->tcb->hw_consumer_index) = 0;
+		txq->tcb->q_depth = tx_cfg->txq_depth;
+		txq->tcb->unmap_q = (void *)
+		res_info[BNA_TX_RES_MEM_T_UNMAPQ].res_u.mem_info.mdl[i].kva;
+		qset = (struct bna_doorbell_qset *)0;
+		off = (unsigned long)&qset[txq->txq_id].txq[0];
+		txq->tcb->q_dbell = off +
+			BNA_GET_DOORBELL_BASE_ADDR(bna->pcidev.pci_bar_kva);
+		txq->tcb->i_dbell = &txq->ib->door_bell;
+		txq->tcb->intr_type = intr_info->intr_type;
+		txq->tcb->intr_vector = (intr_info->num == 1) ?
+					intr_info->idl[0].vector :
+					intr_info->idl[i].vector;
+		txq->tcb->txq = txq;
+		txq->tcb->bnad = bnad;
+		txq->tcb->id = i;
+
+		/* QPT, SWQPT, Pages */
+		bna_txq_qpt_setup(txq, page_count, page_size,
+			&res_info[BNA_TX_RES_MEM_T_QPT].res_u.mem_info.mdl[i],
+			&res_info[BNA_TX_RES_MEM_T_SWQPT].res_u.mem_info.mdl[i],
+			&res_info[BNA_TX_RES_MEM_T_PAGE].
+				  res_u.mem_info.mdl[page_idx]);
+		txq->tcb->page_idx = page_idx;
+		txq->tcb->page_count = page_count;
+		page_idx += page_count;
+
+		/* Callback to bnad for setting up TCB */
+		if (tx->tcb_setup_cbfn)
+			(tx->tcb_setup_cbfn)(bna->bnad, txq->tcb);
+
+		i++;
+	}
+
+	/* TxF */
+
+	tx->txf.ctrl_flags = BFI_TXF_CF_ENABLE | BFI_TXF_CF_VLAN_WI_BASED;
+	tx->txf.vlan = 0;
+
+	/* Mbox element */
+	bfa_q_qe_init(&tx->mbox_qe.qe);
+
+	bfa_fsm_set_state(tx, bna_tx_sm_stopped);
+
+	return tx;
+
+err_return:
+	bna_tx_free(tx);
+	return NULL;
+}
+
+void
+bna_tx_destroy(struct bna_tx *tx)
+{
+	/* Callback to bnad for destroying TCB */
+	if (tx->tcb_destroy_cbfn) {
+		struct bna_txq *txq;
+		struct list_head *qe;
+
+		list_for_each(qe, &tx->txq_q) {
+			txq = (struct bna_txq *)qe;
+			(tx->tcb_destroy_cbfn)(tx->bna->bnad, txq->tcb);
+		}
+	}
+
+	bna_tx_free(tx);
+}
+
+void
+bna_tx_enable(struct bna_tx *tx)
+{
+	if (tx->fsm != (bfa_sm_t)bna_tx_sm_stopped)
+		return;
+
+	tx->flags |= BNA_TX_F_ENABLED;
+
+	if (tx->flags & BNA_TX_F_PORT_STARTED)
+		bfa_fsm_send_event(tx, TX_E_START);
+}
+
+void
+bna_tx_disable(struct bna_tx *tx, enum bna_cleanup_type type,
+		void (*cbfn)(void *, struct bna_tx *, enum bna_cb_status))
+{
+	if (type == BNA_SOFT_CLEANUP) {
+		(*cbfn)(tx->bna->bnad, tx, BNA_CB_SUCCESS);
+		return;
+	}
+
+	tx->stop_cbfn = cbfn;
+	tx->stop_cbarg = tx->bna->bnad;
+
+	tx->flags &= ~BNA_TX_F_ENABLED;
+
+	bfa_fsm_send_event(tx, TX_E_STOP);
+}
+
+int
+bna_tx_state_get(struct bna_tx *tx)
+{
+	return bfa_sm_to_state(tx_sm_table, tx->fsm);
+}
+
+void
+bna_tx_mod_init(struct bna_tx_mod *tx_mod, struct bna *bna,
+		struct bna_res_info *res_info)
+{
+	int i;
+
+	tx_mod->bna = bna;
+	tx_mod->flags = 0;
+
+	tx_mod->tx = (struct bna_tx *)
+		res_info[BNA_RES_MEM_T_TX_ARRAY].res_u.mem_info.mdl[0].kva;
+	tx_mod->txq = (struct bna_txq *)
+		res_info[BNA_RES_MEM_T_TXQ_ARRAY].res_u.mem_info.mdl[0].kva;
+
+	INIT_LIST_HEAD(&tx_mod->tx_free_q);
+	INIT_LIST_HEAD(&tx_mod->tx_active_q);
+
+	INIT_LIST_HEAD(&tx_mod->txq_free_q);
+
+	for (i = 0; i < BFI_MAX_TXQ; i++) {
+		tx_mod->tx[i].txf.txf_id = i;
+		bfa_q_qe_init(&tx_mod->tx[i].qe);
+		list_add_tail(&tx_mod->tx[i].qe, &tx_mod->tx_free_q);
+
+		tx_mod->txq[i].txq_id = i;
+		bfa_q_qe_init(&tx_mod->txq[i].qe);
+		list_add_tail(&tx_mod->txq[i].qe, &tx_mod->txq_free_q);
+	}
+
+	tx_mod->tx_stop_wc.wc_resume = bna_tx_mod_cb_tx_stopped_all;
+	tx_mod->tx_stop_wc.wc_cbarg = tx_mod;
+	tx_mod->tx_stop_wc.wc_count = 0;
+}
+
+void
+bna_tx_mod_uninit(struct bna_tx_mod *tx_mod)
+{
+	struct list_head		*qe;
+	int i;
+
+	i = 0;
+	list_for_each(qe, &tx_mod->tx_free_q)
+		i++;
+
+	i = 0;
+	list_for_each(qe, &tx_mod->txq_free_q)
+		i++;
+
+	tx_mod->bna = NULL;
+}
+
+void
+bna_tx_mod_start(struct bna_tx_mod *tx_mod, enum bna_tx_type type)
+{
+	struct bna_tx *tx;
+	struct list_head		*qe;
+
+	tx_mod->flags |= BNA_TX_MOD_F_PORT_STARTED;
+	if (type == BNA_TX_T_LOOPBACK)
+		tx_mod->flags |= BNA_TX_MOD_F_PORT_LOOPBACK;
+
+	list_for_each(qe, &tx_mod->tx_active_q) {
+		tx = (struct bna_tx *)qe;
+		if (tx->type == type)
+			bna_tx_start(tx);
+	}
+}
+
+void
+bna_tx_mod_stop(struct bna_tx_mod *tx_mod, enum bna_tx_type type)
+{
+	struct bna_tx *tx;
+	struct list_head		*qe;
+
+	tx_mod->flags &= ~BNA_TX_MOD_F_PORT_STARTED;
+	tx_mod->flags &= ~BNA_TX_MOD_F_PORT_LOOPBACK;
+
+	tx_mod->stop_cbfn = bna_port_cb_tx_stopped;
+
+	/**
+	 * Before calling bna_tx_stop(), increment tx_stop_wc as many times
+	 * as we are going to call bna_tx_stop
+	 */
+	list_for_each(qe, &tx_mod->tx_active_q) {
+		tx = (struct bna_tx *)qe;
+		if (tx->type == type)
+			bfa_wc_up(&tx_mod->tx_stop_wc);
+	}
+
+	if (tx_mod->tx_stop_wc.wc_count == 0) {
+		tx_mod->stop_cbfn(&tx_mod->bna->port, BNA_CB_SUCCESS);
+		tx_mod->stop_cbfn = NULL;
+		return;
+	}
+
+	list_for_each(qe, &tx_mod->tx_active_q) {
+		tx = (struct bna_tx *)qe;
+		if (tx->type == type)
+			bna_tx_stop(tx);
+	}
+}
+
+void
+bna_tx_mod_fail(struct bna_tx_mod *tx_mod)
+{
+	struct bna_tx *tx;
+	struct list_head		*qe;
+
+	tx_mod->flags &= ~BNA_TX_MOD_F_PORT_STARTED;
+	tx_mod->flags &= ~BNA_TX_MOD_F_PORT_LOOPBACK;
+
+	list_for_each(qe, &tx_mod->tx_active_q) {
+		tx = (struct bna_tx *)qe;
+		bna_tx_fail(tx);
+	}
+}
+
+void
+bna_tx_mod_prio_changed(struct bna_tx_mod *tx_mod, int prio)
+{
+	struct bna_tx *tx;
+	struct list_head		*qe;
+
+	if (prio != tx_mod->priority) {
+		tx_mod->priority = prio;
+
+		list_for_each(qe, &tx_mod->tx_active_q) {
+			tx = (struct bna_tx *)qe;
+			bna_tx_prio_changed(tx, prio);
+		}
+	}
+}
+
+void
+bna_tx_mod_cee_link_status(struct bna_tx_mod *tx_mod, int cee_link)
+{
+	struct bna_tx *tx;
+	struct list_head		*qe;
+
+	tx_mod->cee_link = cee_link;
+
+	list_for_each(qe, &tx_mod->tx_active_q) {
+		tx = (struct bna_tx *)qe;
+		bna_tx_cee_link_status(tx, cee_link);
+	}
+}
diff --git a/drivers/net/bna/bna_types.h b/drivers/net/bna/bna_types.h
new file mode 100644
index 000000000000..6877310f6ef4
--- /dev/null
+++ b/drivers/net/bna/bna_types.h
@@ -0,0 +1,1128 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#ifndef __BNA_TYPES_H__
+#define __BNA_TYPES_H__
+
+#include "cna.h"
+#include "bna_hw.h"
+#include "bfa_cee.h"
+
+/**
+ *
+ * Forward declarations
+ *
+ */
+
+struct bna_txq;
+struct bna_tx;
+struct bna_rxq;
+struct bna_cq;
+struct bna_rx;
+struct bna_rxf;
+struct bna_port;
+struct bna;
+struct bnad;
+
+/**
+ *
+ * Enums, primitive data types
+ *
+ */
+
+enum bna_status {
+	BNA_STATUS_T_DISABLED	= 0,
+	BNA_STATUS_T_ENABLED	= 1
+};
+
+enum bna_cleanup_type {
+	BNA_HARD_CLEANUP 	= 0,
+	BNA_SOFT_CLEANUP 	= 1
+};
+
+enum bna_cb_status {
+	BNA_CB_SUCCESS 		= 0,
+	BNA_CB_FAIL		= 1,
+	BNA_CB_INTERRUPT	= 2,
+	BNA_CB_BUSY		= 3,
+	BNA_CB_INVALID_MAC	= 4,
+	BNA_CB_MCAST_LIST_FULL	= 5,
+	BNA_CB_UCAST_CAM_FULL	= 6,
+	BNA_CB_WAITING		= 7,
+	BNA_CB_NOT_EXEC		= 8
+};
+
+enum bna_res_type {
+	BNA_RES_T_MEM		= 1,
+	BNA_RES_T_INTR		= 2
+};
+
+enum bna_mem_type {
+	BNA_MEM_T_KVA 		= 1,
+	BNA_MEM_T_DMA 		= 2
+};
+
+enum bna_intr_type {
+	BNA_INTR_T_INTX		= 1,
+	BNA_INTR_T_MSIX		= 2
+};
+
+enum bna_res_req_type {
+	BNA_RES_MEM_T_COM 		= 0,
+	BNA_RES_MEM_T_ATTR 		= 1,
+	BNA_RES_MEM_T_FWTRC 		= 2,
+	BNA_RES_MEM_T_STATS 		= 3,
+	BNA_RES_MEM_T_SWSTATS		= 4,
+	BNA_RES_MEM_T_IBIDX		= 5,
+	BNA_RES_MEM_T_IB_ARRAY		= 6,
+	BNA_RES_MEM_T_INTR_ARRAY	= 7,
+	BNA_RES_MEM_T_IDXSEG_ARRAY	= 8,
+	BNA_RES_MEM_T_TX_ARRAY		= 9,
+	BNA_RES_MEM_T_TXQ_ARRAY		= 10,
+	BNA_RES_MEM_T_RX_ARRAY		= 11,
+	BNA_RES_MEM_T_RXP_ARRAY		= 12,
+	BNA_RES_MEM_T_RXQ_ARRAY		= 13,
+	BNA_RES_MEM_T_UCMAC_ARRAY	= 14,
+	BNA_RES_MEM_T_MCMAC_ARRAY	= 15,
+	BNA_RES_MEM_T_RIT_ENTRY		= 16,
+	BNA_RES_MEM_T_RIT_SEGMENT	= 17,
+	BNA_RES_INTR_T_MBOX		= 18,
+	BNA_RES_T_MAX
+};
+
+enum bna_tx_res_req_type {
+	BNA_TX_RES_MEM_T_TCB	= 0,
+	BNA_TX_RES_MEM_T_UNMAPQ	= 1,
+	BNA_TX_RES_MEM_T_QPT 	= 2,
+	BNA_TX_RES_MEM_T_SWQPT	= 3,
+	BNA_TX_RES_MEM_T_PAGE 	= 4,
+	BNA_TX_RES_INTR_T_TXCMPL = 5,
+	BNA_TX_RES_T_MAX,
+};
+
+enum bna_rx_mem_type {
+	BNA_RX_RES_MEM_T_CCB		= 0,	/* CQ context */
+	BNA_RX_RES_MEM_T_RCB		= 1,	/* CQ context */
+	BNA_RX_RES_MEM_T_UNMAPQ		= 2,	/* UnmapQ for RxQs */
+	BNA_RX_RES_MEM_T_CQPT		= 3,	/* CQ QPT */
+	BNA_RX_RES_MEM_T_CSWQPT		= 4,	/* S/W QPT */
+	BNA_RX_RES_MEM_T_CQPT_PAGE	= 5,	/* CQPT page */
+	BNA_RX_RES_MEM_T_HQPT		= 6,	/* RX QPT */
+	BNA_RX_RES_MEM_T_DQPT		= 7,	/* RX QPT */
+	BNA_RX_RES_MEM_T_HSWQPT		= 8,	/* RX s/w QPT */
+	BNA_RX_RES_MEM_T_DSWQPT		= 9,	/* RX s/w QPT */
+	BNA_RX_RES_MEM_T_DPAGE		= 10,	/* RX s/w QPT */
+	BNA_RX_RES_MEM_T_HPAGE		= 11,	/* RX s/w QPT */
+	BNA_RX_RES_T_INTR		= 12,	/* Rx interrupts */
+	BNA_RX_RES_T_MAX		= 13
+};
+
+enum bna_mbox_state {
+	BNA_MBOX_FREE		= 0,
+	BNA_MBOX_POSTED		= 1
+};
+
+enum bna_tx_type {
+	BNA_TX_T_REGULAR	= 0,
+	BNA_TX_T_LOOPBACK	= 1,
+};
+
+enum bna_tx_flags {
+	BNA_TX_F_PORT_STARTED	= 1,
+	BNA_TX_F_ENABLED	= 2,
+	BNA_TX_F_PRIO_LOCK	= 4,
+};
+
+enum bna_tx_mod_flags {
+	BNA_TX_MOD_F_PORT_STARTED	= 1,
+	BNA_TX_MOD_F_PORT_LOOPBACK	= 2,
+};
+
+enum bna_rx_type {
+	BNA_RX_T_REGULAR	= 0,
+	BNA_RX_T_LOOPBACK	= 1,
+};
+
+enum bna_rxp_type {
+	BNA_RXP_SINGLE 		= 1,
+	BNA_RXP_SLR 		= 2,
+	BNA_RXP_HDS 		= 3
+};
+
+enum bna_rxmode {
+	BNA_RXMODE_PROMISC 	= 1,
+	BNA_RXMODE_DEFAULT 	= 2,
+	BNA_RXMODE_ALLMULTI 	= 4
+};
+
+enum bna_rx_event {
+	RX_E_START			= 1,
+	RX_E_STOP			= 2,
+	RX_E_FAIL			= 3,
+	RX_E_RXF_STARTED		= 4,
+	RX_E_RXF_STOPPED		= 5,
+	RX_E_RXQ_STOPPED		= 6,
+};
+
+enum bna_rx_state {
+	BNA_RX_STOPPED			= 1,
+	BNA_RX_RXF_START_WAIT		= 2,
+	BNA_RX_STARTED			= 3,
+	BNA_RX_RXF_STOP_WAIT		= 4,
+	BNA_RX_RXQ_STOP_WAIT		= 5,
+};
+
+enum bna_rx_flags {
+	BNA_RX_F_ENABLE		= 0x01,		/* bnad enabled rxf */
+	BNA_RX_F_PORT_ENABLED	= 0x02,		/* Port object is enabled */
+	BNA_RX_F_PORT_FAILED	= 0x04,		/* Port in failed state */
+};
+
+enum bna_rx_mod_flags {
+	BNA_RX_MOD_F_PORT_STARTED	= 1,
+	BNA_RX_MOD_F_PORT_LOOPBACK	= 2,
+};
+
+enum bna_rxf_oper_state {
+	BNA_RXF_OPER_STATE_RUNNING	= 0x01, /* rxf operational */
+	BNA_RXF_OPER_STATE_PAUSED	= 0x02,	/* rxf in PAUSED state */
+};
+
+enum bna_rxf_flags {
+	BNA_RXF_FL_STOP_PENDING 	= 0x01,
+	BNA_RXF_FL_FAILED		= 0x02,
+	BNA_RXF_FL_RSS_CONFIG_PENDING	= 0x04,
+	BNA_RXF_FL_OPERSTATE_CHANGED	= 0x08,
+	BNA_RXF_FL_RXF_ENABLED		= 0x10,
+	BNA_RXF_FL_VLAN_CONFIG_PENDING	= 0x20,
+};
+
+enum bna_rxf_event {
+	RXF_E_START			= 1,
+	RXF_E_STOP			= 2,
+	RXF_E_FAIL			= 3,
+	RXF_E_CAM_FLTR_MOD		= 4,
+	RXF_E_STARTED			= 5,
+	RXF_E_STOPPED			= 6,
+	RXF_E_CAM_FLTR_RESP		= 7,
+	RXF_E_PAUSE			= 8,
+	RXF_E_RESUME			= 9,
+	RXF_E_STAT_CLEARED		= 10,
+};
+
+enum bna_rxf_state {
+	BNA_RXF_STOPPED			= 1,
+	BNA_RXF_START_WAIT		= 2,
+	BNA_RXF_CAM_FLTR_MOD_WAIT	= 3,
+	BNA_RXF_STARTED			= 4,
+	BNA_RXF_CAM_FLTR_CLR_WAIT	= 5,
+	BNA_RXF_STOP_WAIT		= 6,
+	BNA_RXF_PAUSE_WAIT		= 7,
+	BNA_RXF_RESUME_WAIT		= 8,
+	BNA_RXF_STAT_CLR_WAIT		= 9,
+};
+
+enum bna_port_type {
+	BNA_PORT_T_REGULAR		= 0,
+	BNA_PORT_T_LOOPBACK_INTERNAL	= 1,
+	BNA_PORT_T_LOOPBACK_EXTERNAL	= 2,
+};
+
+enum bna_link_status {
+	BNA_LINK_DOWN		= 0,
+	BNA_LINK_UP		= 1,
+	BNA_CEE_UP 		= 2
+};
+
+enum bna_llport_flags {
+	BNA_LLPORT_F_ENABLED 	= 1,
+	BNA_LLPORT_F_RX_ENABLED	= 2
+};
+
+enum bna_port_flags {
+	BNA_PORT_F_DEVICE_READY	= 1,
+	BNA_PORT_F_ENABLED	= 2,
+	BNA_PORT_F_PAUSE_CHANGED = 4,
+	BNA_PORT_F_MTU_CHANGED	= 8
+};
+
+enum bna_pkt_rates {
+	BNA_PKT_RATE_10K		= 10000,
+	BNA_PKT_RATE_20K		= 20000,
+	BNA_PKT_RATE_30K		= 30000,
+	BNA_PKT_RATE_40K		= 40000,
+	BNA_PKT_RATE_50K		= 50000,
+	BNA_PKT_RATE_60K		= 60000,
+	BNA_PKT_RATE_70K		= 70000,
+	BNA_PKT_RATE_80K		= 80000,
+};
+
+enum bna_dim_load_types {
+	BNA_LOAD_T_HIGH_4		= 0, /* 80K <= r */
+	BNA_LOAD_T_HIGH_3		= 1, /* 60K <= r < 80K */
+	BNA_LOAD_T_HIGH_2		= 2, /* 50K <= r < 60K */
+	BNA_LOAD_T_HIGH_1		= 3, /* 40K <= r < 50K */
+	BNA_LOAD_T_LOW_1		= 4, /* 30K <= r < 40K */
+	BNA_LOAD_T_LOW_2		= 5, /* 20K <= r < 30K */
+	BNA_LOAD_T_LOW_3		= 6, /* 10K <= r < 20K */
+	BNA_LOAD_T_LOW_4		= 7, /* r < 10K */
+	BNA_LOAD_T_MAX			= 8
+};
+
+enum bna_dim_bias_types {
+	BNA_BIAS_T_SMALL		= 0, /* small pkts > (large pkts * 2) */
+	BNA_BIAS_T_LARGE		= 1, /* Not BNA_BIAS_T_SMALL */
+	BNA_BIAS_T_MAX			= 2
+};
+
+struct bna_mac {
+	/* This should be the first one */
+	struct list_head			qe;
+	u8			addr[ETH_ALEN];
+};
+
+struct bna_mem_descr {
+	u32		len;
+	void		*kva;
+	struct bna_dma_addr dma;
+};
+
+struct bna_mem_info {
+	enum bna_mem_type mem_type;
+	u32		len;
+	u32 		num;
+	u32		align_sz; /* 0/1 = no alignment */
+	struct bna_mem_descr *mdl;
+	void			*cookie; /* For bnad to unmap dma later */
+};
+
+struct bna_intr_descr {
+	int			vector;
+};
+
+struct bna_intr_info {
+	enum bna_intr_type intr_type;
+	int			num;
+	struct bna_intr_descr *idl;
+};
+
+union bna_res_u {
+	struct bna_mem_info mem_info;
+	struct bna_intr_info intr_info;
+};
+
+struct bna_res_info {
+	enum bna_res_type res_type;
+	union bna_res_u		res_u;
+};
+
+/* HW QPT */
+struct bna_qpt {
+	struct bna_dma_addr hw_qpt_ptr;
+	void		*kv_qpt_ptr;
+	u32		page_count;
+	u32		page_size;
+};
+
+/**
+ *
+ * Device
+ *
+ */
+
+struct bna_device {
+	bfa_fsm_t		fsm;
+	struct bfa_ioc ioc;
+
+	enum bna_intr_type intr_type;
+	int			vector;
+
+	void (*ready_cbfn)(struct bnad *bnad, enum bna_cb_status status);
+	struct bnad *ready_cbarg;
+
+	void (*stop_cbfn)(struct bnad *bnad, enum bna_cb_status status);
+	struct bnad *stop_cbarg;
+
+	struct bna *bna;
+};
+
+/**
+ *
+ * Mail box
+ *
+ */
+
+struct bna_mbox_qe {
+	/* This should be the first one */
+	struct list_head			qe;
+
+	struct bfa_mbox_cmd cmd;
+	u32 		cmd_len;
+	/* Callback for port, tx, rx, rxf */
+	void (*cbfn)(void *arg, int status);
+	void 			*cbarg;
+};
+
+struct bna_mbox_mod {
+	enum bna_mbox_state state;
+	struct list_head			posted_q;
+	u32		msg_pending;
+	u32		msg_ctr;
+	struct bna *bna;
+};
+
+/**
+ *
+ * Port
+ *
+ */
+
+/* Pause configuration */
+struct bna_pause_config {
+	enum bna_status tx_pause;
+	enum bna_status rx_pause;
+};
+
+struct bna_llport {
+	bfa_fsm_t		fsm;
+	enum bna_llport_flags flags;
+
+	enum bna_port_type type;
+
+	enum bna_link_status link_status;
+
+	int			admin_up_count;
+
+	void (*stop_cbfn)(struct bna_port *, enum bna_cb_status);
+
+	struct bna_mbox_qe mbox_qe;
+
+	struct bna *bna;
+};
+
+struct bna_port {
+	bfa_fsm_t		fsm;
+	enum bna_port_flags flags;
+
+	enum bna_port_type type;
+
+	struct bna_llport llport;
+
+	struct bna_pause_config pause_config;
+	u8			priority;
+	int			mtu;
+
+	/* Callback for bna_port_disable(), port_stop() */
+	void (*stop_cbfn)(void *, enum bna_cb_status);
+	void			*stop_cbarg;
+
+	/* Callback for bna_port_pause_config() */
+	void (*pause_cbfn)(struct bnad *, enum bna_cb_status);
+
+	/* Callback for bna_port_mtu_set() */
+	void (*mtu_cbfn)(struct bnad *, enum bna_cb_status);
+
+	void (*link_cbfn)(struct bnad *, enum bna_link_status);
+
+	struct bfa_wc		chld_stop_wc;
+
+	struct bna_mbox_qe mbox_qe;
+
+	struct bna *bna;
+};
+
+/**
+ *
+ * Interrupt Block
+ *
+ */
+
+/* IB index segment structure */
+struct bna_ibidx_seg {
+	/* This should be the first one */
+	struct list_head			qe;
+
+	u8			ib_seg_size;
+	u8			ib_idx_tbl_offset;
+};
+
+/* Interrupt structure */
+struct bna_intr {
+	/* This should be the first one */
+	struct list_head			qe;
+	int			ref_count;
+
+	enum bna_intr_type intr_type;
+	int			vector;
+
+	struct bna_ib *ib;
+};
+
+/* Doorbell structure */
+struct bna_ib_dbell {
+	void *__iomem doorbell_addr;
+	u32		doorbell_ack;
+};
+
+/* Interrupt timer configuration */
+struct bna_ib_config {
+	u8 		coalescing_timeo;    /* Unit is 5usec. */
+
+	int			interpkt_count;
+	int			interpkt_timeo;
+
+	enum ib_flags ctrl_flags;
+};
+
+/* IB structure */
+struct bna_ib {
+	/* This should be the first one */
+	struct list_head			qe;
+
+	int			ib_id;
+
+	int			ref_count;
+	int			start_count;
+
+	struct bna_dma_addr ib_seg_host_addr;
+	void		*ib_seg_host_addr_kva;
+	u32		idx_mask; /* Size >= BNA_IBIDX_MAX_SEGSIZE */
+
+	struct bna_ibidx_seg *idx_seg;
+
+	struct bna_ib_dbell door_bell;
+
+	struct bna_intr *intr;
+
+	struct bna_ib_config ib_config;
+
+	struct bna *bna;
+};
+
+/* IB module - keeps track of IBs and interrupts */
+struct bna_ib_mod {
+	struct bna_ib *ib;		/* BFI_MAX_IB entries */
+	struct bna_intr *intr;		/* BFI_MAX_IB entries */
+	struct bna_ibidx_seg *idx_seg;	/* BNA_IBIDX_TOTAL_SEGS */
+
+	struct list_head			ib_free_q;
+
+	struct list_head		ibidx_seg_pool[BFI_IBIDX_TOTAL_POOLS];
+
+	struct list_head			intr_free_q;
+	struct list_head			intr_active_q;
+
+	struct bna *bna;
+};
+
+/**
+ *
+ * Tx object
+ *
+ */
+
+/* Tx datapath control structure */
+#define BNA_Q_NAME_SIZE		16
+struct bna_tcb {
+	/* Fast path */
+	void			**sw_qpt;
+	void			*unmap_q;
+	u32		producer_index;
+	u32		consumer_index;
+	volatile u32	*hw_consumer_index;
+	u32		q_depth;
+	void *__iomem q_dbell;
+	struct bna_ib_dbell *i_dbell;
+	int			page_idx;
+	int			page_count;
+	/* Control path */
+	struct bna_txq *txq;
+	struct bnad *bnad;
+	enum bna_intr_type intr_type;
+	int			intr_vector;
+	u8			priority; /* Current priority */
+	unsigned long		flags; /* Used by bnad as required */
+	int			id;
+	char			name[BNA_Q_NAME_SIZE];
+};
+
+/* TxQ QPT and configuration */
+struct bna_txq {
+	/* This should be the first one */
+	struct list_head			qe;
+
+	int			txq_id;
+
+	u8			priority;
+
+	struct bna_qpt qpt;
+	struct bna_tcb *tcb;
+	struct bna_ib *ib;
+	int			ib_seg_offset;
+
+	struct bna_tx *tx;
+
+	u64 		tx_packets;
+	u64 		tx_bytes;
+};
+
+/* TxF structure (hardware Tx Function) */
+struct bna_txf {
+	int			txf_id;
+	enum txf_flags ctrl_flags;
+	u16		vlan;
+};
+
+/* Tx object */
+struct bna_tx {
+	/* This should be the first one */
+	struct list_head			qe;
+
+	bfa_fsm_t		fsm;
+	enum bna_tx_flags flags;
+
+	enum bna_tx_type type;
+
+	struct list_head			txq_q;
+	struct bna_txf txf;
+
+	/* Tx event handlers */
+	void (*tcb_setup_cbfn)(struct bnad *, struct bna_tcb *);
+	void (*tcb_destroy_cbfn)(struct bnad *, struct bna_tcb *);
+	void (*tx_stall_cbfn)(struct bnad *, struct bna_tcb *);
+	void (*tx_resume_cbfn)(struct bnad *, struct bna_tcb *);
+	void (*tx_cleanup_cbfn)(struct bnad *, struct bna_tcb *);
+
+	/* callback for bna_tx_disable(), bna_tx_stop() */
+	void (*stop_cbfn)(void *arg, struct bna_tx *tx,
+				enum bna_cb_status status);
+	void			*stop_cbarg;
+
+	/* callback for bna_tx_prio_set() */
+	void (*prio_change_cbfn)(struct bnad *bnad, struct bna_tx *tx,
+				enum bna_cb_status status);
+
+	struct bfa_wc		txq_stop_wc;
+
+	struct bna_mbox_qe mbox_qe;
+
+	struct bna *bna;
+	void			*priv;	/* bnad's cookie */
+};
+
+struct bna_tx_config {
+	int			num_txq;
+	int			txq_depth;
+	enum bna_tx_type tx_type;
+};
+
+struct bna_tx_event_cbfn {
+	/* Optional */
+	void (*tcb_setup_cbfn)(struct bnad *, struct bna_tcb *);
+	void (*tcb_destroy_cbfn)(struct bnad *, struct bna_tcb *);
+	/* Mandatory */
+	void (*tx_stall_cbfn)(struct bnad *, struct bna_tcb *);
+	void (*tx_resume_cbfn)(struct bnad *, struct bna_tcb *);
+	void (*tx_cleanup_cbfn)(struct bnad *, struct bna_tcb *);
+};
+
+/* Tx module - keeps track of free, active tx objects */
+struct bna_tx_mod {
+	struct bna_tx *tx;		/* BFI_MAX_TXQ entries */
+	struct bna_txq *txq;		/* BFI_MAX_TXQ entries */
+
+	struct list_head			tx_free_q;
+	struct list_head			tx_active_q;
+
+	struct list_head			txq_free_q;
+
+	/* callback for bna_tx_mod_stop() */
+	void (*stop_cbfn)(struct bna_port *port,
+				enum bna_cb_status status);
+
+	struct bfa_wc		tx_stop_wc;
+
+	enum bna_tx_mod_flags flags;
+
+	int			priority;
+	int			cee_link;
+
+	u32		txf_bmap[2];
+
+	struct bna *bna;
+};
+
+/**
+ *
+ * Receive Indirection Table
+ *
+ */
+
+/* One row of RIT table */
+struct bna_rit_entry {
+	u8 large_rxq_id;	/* used for either large or data buffers */
+	u8 small_rxq_id;	/* used for either small or header buffers */
+};
+
+/* RIT segment */
+struct bna_rit_segment {
+	struct list_head			qe;
+
+	u32		rit_offset;
+	u32		rit_size;
+	/**
+	 * max_rit_size: Varies per RIT segment depending on how RIT is
+	 * partitioned
+	 */
+	u32		max_rit_size;
+
+	struct bna_rit_entry *rit;
+};
+
+struct bna_rit_mod {
+	struct bna_rit_entry *rit;
+	struct bna_rit_segment *rit_segment;
+
+	struct list_head		rit_seg_pool[BFI_RIT_SEG_TOTAL_POOLS];
+};
+
+/**
+ *
+ * Rx object
+ *
+ */
+
+/* Rx datapath control structure */
+struct bna_rcb {
+	/* Fast path */
+	void			**sw_qpt;
+	void			*unmap_q;
+	u32		producer_index;
+	u32		consumer_index;
+	u32		q_depth;
+	void *__iomem q_dbell;
+	int			page_idx;
+	int			page_count;
+	/* Control path */
+	struct bna_rxq *rxq;
+	struct bna_cq *cq;
+	struct bnad *bnad;
+	unsigned long		flags;
+	int			id;
+};
+
+/* RxQ structure - QPT, configuration */
+struct bna_rxq {
+	struct list_head			qe;
+	int			rxq_id;
+
+	int			buffer_size;
+	int			q_depth;
+
+	struct bna_qpt qpt;
+	struct bna_rcb *rcb;
+
+	struct bna_rxp *rxp;
+	struct bna_rx *rx;
+
+	u64 		rx_packets;
+	u64		rx_bytes;
+	u64 		rx_packets_with_error;
+	u64 		rxbuf_alloc_failed;
+};
+
+/* RxQ pair */
+union bna_rxq_u {
+	struct {
+		struct bna_rxq *hdr;
+		struct bna_rxq *data;
+	} hds;
+	struct {
+		struct bna_rxq *small;
+		struct bna_rxq *large;
+	} slr;
+	struct {
+		struct bna_rxq *only;
+		struct bna_rxq *reserved;
+	} single;
+};
+
+/* Packet rate for Dynamic Interrupt Moderation */
+struct bna_pkt_rate {
+	u32		small_pkt_cnt;
+	u32		large_pkt_cnt;
+};
+
+/* Completion control structure */
+struct bna_ccb {
+	/* Fast path */
+	void			**sw_qpt;
+	u32		producer_index;
+	volatile u32	*hw_producer_index;
+	u32		q_depth;
+	struct bna_ib_dbell *i_dbell;
+	struct bna_rcb *rcb[2];
+	void			*ctrl; /* For bnad */
+	struct bna_pkt_rate pkt_rate;
+	int			page_idx;
+	int			page_count;
+
+	/* Control path */
+	struct bna_cq *cq;
+	struct bnad *bnad;
+	enum bna_intr_type intr_type;
+	int			intr_vector;
+	u8			rx_coalescing_timeo; /* For NAPI */
+	int			id;
+	char			name[BNA_Q_NAME_SIZE];
+};
+
+/* CQ QPT, configuration  */
+struct bna_cq {
+	int			cq_id;
+
+	struct bna_qpt qpt;
+	struct bna_ccb *ccb;
+
+	struct bna_ib *ib;
+	u8			ib_seg_offset;
+
+	struct bna_rx *rx;
+};
+
+struct bna_rss_config {
+	enum rss_hash_type hash_type;
+	u8			hash_mask;
+	u32		toeplitz_hash_key[BFI_RSS_HASH_KEY_LEN];
+};
+
+struct bna_hds_config {
+	enum hds_header_type hdr_type;
+	int			header_size;
+};
+
+/* This structure is used during RX creation */
+struct bna_rx_config {
+	enum bna_rx_type rx_type;
+	int			num_paths;
+	enum bna_rxp_type rxp_type;
+	int			paused;
+	int			q_depth;
+	/*
+	 * Small/Large (or Header/Data) buffer size to be configured
+	 * for SLR and HDS queue type. Large buffer size comes from
+	 * port->mtu.
+	 */
+	int			small_buff_size;
+
+	enum bna_status rss_status;
+	struct bna_rss_config rss_config;
+
+	enum bna_status hds_status;
+	struct bna_hds_config hds_config;
+
+	enum bna_status vlan_strip_status;
+};
+
+/* Rx Path structure - one per MSIX vector/CPU */
+struct bna_rxp {
+	/* This should be the first one */
+	struct list_head			qe;
+
+	enum bna_rxp_type type;
+	union	bna_rxq_u	rxq;
+	struct bna_cq cq;
+
+	struct bna_rx *rx;
+
+	/* MSI-x vector number for configuring RSS */
+	int			vector;
+
+	struct bna_mbox_qe mbox_qe;
+};
+
+/* HDS configuration structure */
+struct bna_rxf_hds {
+	enum hds_header_type hdr_type;
+	int			header_size;
+};
+
+/* RSS configuration structure */
+struct bna_rxf_rss {
+	enum rss_hash_type hash_type;
+	u8			hash_mask;
+	u32		toeplitz_hash_key[BFI_RSS_HASH_KEY_LEN];
+};
+
+/* RxF structure (hardware Rx Function) */
+struct bna_rxf {
+	bfa_fsm_t		fsm;
+	int			rxf_id;
+	enum rxf_flags ctrl_flags;
+	u16		default_vlan_tag;
+	enum bna_rxf_oper_state rxf_oper_state;
+	enum bna_status hds_status;
+	struct bna_rxf_hds hds_cfg;
+	enum bna_status rss_status;
+	struct bna_rxf_rss rss_cfg;
+	struct bna_rit_segment *rit_segment;
+	struct bna_rx *rx;
+	u32		forced_offset;
+	struct bna_mbox_qe mbox_qe;
+	int			mcast_rxq_id;
+
+	/* callback for bna_rxf_start() */
+	void (*start_cbfn) (struct bna_rx *rx, enum bna_cb_status status);
+	struct bna_rx *start_cbarg;
+
+	/* callback for bna_rxf_stop() */
+	void (*stop_cbfn) (struct bna_rx *rx, enum bna_cb_status status);
+	struct bna_rx *stop_cbarg;
+
+	/* callback for bna_rxf_receive_enable() / bna_rxf_receive_disable() */
+	void (*oper_state_cbfn) (struct bnad *bnad, struct bna_rx *rx,
+			enum bna_cb_status status);
+	struct bnad *oper_state_cbarg;
+
+	/**
+	 * callback for:
+	 *	bna_rxf_ucast_set()
+	 *	bna_rxf_{ucast/mcast}_add(),
+	 * 	bna_rxf_{ucast/mcast}_del(),
+	 *	bna_rxf_mode_set()
+	 */
+	void (*cam_fltr_cbfn)(struct bnad *bnad, struct bna_rx *rx,
+				enum bna_cb_status status);
+	struct bnad *cam_fltr_cbarg;
+
+	enum bna_rxf_flags rxf_flags;
+
+	/* List of unicast addresses yet to be applied to h/w */
+	struct list_head			ucast_pending_add_q;
+	struct list_head			ucast_pending_del_q;
+	int			ucast_pending_set;
+	/* ucast addresses applied to the h/w */
+	struct list_head			ucast_active_q;
+	struct bna_mac *ucast_active_mac;
+
+	/* List of multicast addresses yet to be applied to h/w */
+	struct list_head			mcast_pending_add_q;
+	struct list_head			mcast_pending_del_q;
+	/* multicast addresses applied to the h/w */
+	struct list_head			mcast_active_q;
+
+	/* Rx modes yet to be applied to h/w */
+	enum bna_rxmode rxmode_pending;
+	enum bna_rxmode rxmode_pending_bitmask;
+	/* Rx modes applied to h/w */
+	enum bna_rxmode rxmode_active;
+
+	enum bna_status vlan_filter_status;
+	u32		vlan_filter_table[(BFI_MAX_VLAN + 1) / 32];
+};
+
+/* Rx object */
+struct bna_rx {
+	/* This should be the first one */
+	struct list_head			qe;
+
+	bfa_fsm_t		fsm;
+
+	enum bna_rx_type type;
+
+	/* list-head for RX path objects */
+	struct list_head			rxp_q;
+
+	struct bna_rxf rxf;
+
+	enum bna_rx_flags rx_flags;
+
+	struct bna_mbox_qe mbox_qe;
+
+	struct bfa_wc		rxq_stop_wc;
+
+	/* Rx event handlers */
+	void (*rcb_setup_cbfn)(struct bnad *, struct bna_rcb *);
+	void (*rcb_destroy_cbfn)(struct bnad *, struct bna_rcb *);
+	void (*ccb_setup_cbfn)(struct bnad *, struct bna_ccb *);
+	void (*ccb_destroy_cbfn)(struct bnad *, struct bna_ccb *);
+	void (*rx_cleanup_cbfn)(struct bnad *, struct bna_ccb *);
+	void (*rx_post_cbfn)(struct bnad *, struct bna_rcb *);
+
+	/* callback for bna_rx_disable(), bna_rx_stop() */
+	void (*stop_cbfn)(void *arg, struct bna_rx *rx,
+				enum bna_cb_status status);
+	void			*stop_cbarg;
+
+	struct bna *bna;
+	void			*priv; /* bnad's cookie */
+};
+
+struct bna_rx_event_cbfn {
+	/* Optional */
+	void (*rcb_setup_cbfn)(struct bnad *, struct bna_rcb *);
+	void (*rcb_destroy_cbfn)(struct bnad *, struct bna_rcb *);
+	void (*ccb_setup_cbfn)(struct bnad *, struct bna_ccb *);
+	void (*ccb_destroy_cbfn)(struct bnad *, struct bna_ccb *);
+	/* Mandatory */
+	void (*rx_cleanup_cbfn)(struct bnad *, struct bna_ccb *);
+	void (*rx_post_cbfn)(struct bnad *, struct bna_rcb *);
+};
+
+/* Rx module - keeps track of free, active rx objects */
+struct bna_rx_mod {
+	struct bna *bna;		/* back pointer to parent */
+	struct bna_rx *rx;		/* BFI_MAX_RXQ entries */
+	struct bna_rxp *rxp;		/* BFI_MAX_RXQ entries */
+	struct bna_rxq *rxq;		/* BFI_MAX_RXQ entries */
+
+	struct list_head			rx_free_q;
+	struct list_head			rx_active_q;
+	int			rx_free_count;
+
+	struct list_head			rxp_free_q;
+	int			rxp_free_count;
+
+	struct list_head			rxq_free_q;
+	int			rxq_free_count;
+
+	enum bna_rx_mod_flags flags;
+
+	/* callback for bna_rx_mod_stop() */
+	void (*stop_cbfn)(struct bna_port *port,
+				enum bna_cb_status status);
+
+	struct bfa_wc		rx_stop_wc;
+	u32		dim_vector[BNA_LOAD_T_MAX][BNA_BIAS_T_MAX];
+	u32		rxf_bmap[2];
+};
+
+/**
+ *
+ * CAM
+ *
+ */
+
+struct bna_ucam_mod {
+	struct bna_mac *ucmac;		/* BFI_MAX_UCMAC entries */
+	struct list_head			free_q;
+
+	struct bna *bna;
+};
+
+struct bna_mcam_mod {
+	struct bna_mac *mcmac;		/* BFI_MAX_MCMAC entries */
+	struct list_head			free_q;
+
+	struct bna *bna;
+};
+
+/**
+ *
+ * Statistics
+ *
+ */
+
+struct bna_tx_stats {
+	int			tx_state;
+	int			tx_flags;
+	int			num_txqs;
+	u32		txq_bmap[2];
+	int			txf_id;
+};
+
+struct bna_rx_stats {
+	int			rx_state;
+	int			rx_flags;
+	int			num_rxps;
+	int			num_rxqs;
+	u32		rxq_bmap[2];
+	u32		cq_bmap[2];
+	int			rxf_id;
+	int			rxf_state;
+	int			rxf_oper_state;
+	int			num_active_ucast;
+	int			num_active_mcast;
+	int			rxmode_active;
+	int			vlan_filter_status;
+	u32		vlan_filter_table[(BFI_MAX_VLAN + 1) / 32];
+	int			rss_status;
+	int			hds_status;
+};
+
+struct bna_sw_stats {
+	int			device_state;
+	int			port_state;
+	int			port_flags;
+	int			llport_state;
+	int			priority;
+	int			num_active_tx;
+	int			num_active_rx;
+	struct bna_tx_stats tx_stats[BFI_MAX_TXQ];
+	struct bna_rx_stats rx_stats[BFI_MAX_RXQ];
+};
+
+struct bna_stats {
+	u32		txf_bmap[2];
+	u32		rxf_bmap[2];
+	struct bfi_ll_stats	*hw_stats;
+	struct bna_sw_stats *sw_stats;
+};
+
+/**
+ *
+ * BNA
+ *
+ */
+
+struct bna {
+	struct bfa_pcidev pcidev;
+
+	int			port_num;
+
+	struct bna_chip_regs regs;
+
+	struct bna_dma_addr hw_stats_dma;
+	struct bna_stats stats;
+
+	struct bna_device device;
+	struct bfa_cee cee;
+
+	struct bna_mbox_mod mbox_mod;
+
+	struct bna_port port;
+
+	struct bna_tx_mod tx_mod;
+
+	struct bna_rx_mod rx_mod;
+
+	struct bna_ib_mod ib_mod;
+
+	struct bna_ucam_mod ucam_mod;
+	struct bna_mcam_mod mcam_mod;
+
+	struct bna_rit_mod rit_mod;
+
+	int			rxf_default_id;
+	int			rxf_promisc_id;
+
+	struct bna_mbox_qe mbox_qe;
+
+	struct bnad *bnad;
+};
+
+#endif	/* __BNA_TYPES_H__ */
diff --git a/drivers/net/bna/bnad.c b/drivers/net/bna/bnad.c
new file mode 100644
index 000000000000..491d148f88ae
--- /dev/null
+++ b/drivers/net/bna/bnad.c
@@ -0,0 +1,3270 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+#include <linux/etherdevice.h>
+#include <linux/in.h>
+#include <linux/ethtool.h>
+#include <linux/if_vlan.h>
+#include <linux/if_ether.h>
+#include <linux/ip.h>
+
+#include "bnad.h"
+#include "bna.h"
+#include "cna.h"
+
+DEFINE_MUTEX(bnad_fwimg_mutex);
+
+/*
+ * Module params
+ */
+static uint bnad_msix_disable;
+module_param(bnad_msix_disable, uint, 0444);
+MODULE_PARM_DESC(bnad_msix_disable, "Disable MSIX mode");
+
+static uint bnad_ioc_auto_recover = 1;
+module_param(bnad_ioc_auto_recover, uint, 0444);
+MODULE_PARM_DESC(bnad_ioc_auto_recover, "Enable / Disable auto recovery");
+
+/*
+ * Global variables
+ */
+u32 bnad_rxqs_per_cq = 2;
+
+const u8 bnad_bcast_addr[] =  {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
+
+/*
+ * Local MACROS
+ */
+#define BNAD_TX_UNMAPQ_DEPTH (bnad->txq_depth * 2)
+
+#define BNAD_RX_UNMAPQ_DEPTH (bnad->rxq_depth)
+
+#define BNAD_GET_MBOX_IRQ(_bnad)				\
+	(((_bnad)->cfg_flags & BNAD_CF_MSIX) ?			\
+	 ((_bnad)->msix_table[(_bnad)->msix_num - 1].vector) : 	\
+	 ((_bnad)->pcidev->irq))
+
+#define BNAD_FILL_UNMAPQ_MEM_REQ(_res_info, _num, _depth)	\
+do {								\
+	(_res_info)->res_type = BNA_RES_T_MEM;			\
+	(_res_info)->res_u.mem_info.mem_type = BNA_MEM_T_KVA;	\
+	(_res_info)->res_u.mem_info.num = (_num);		\
+	(_res_info)->res_u.mem_info.len =			\
+	sizeof(struct bnad_unmap_q) +				\
+	(sizeof(struct bnad_skb_unmap) * ((_depth) - 1));	\
+} while (0)
+
+/*
+ * Reinitialize completions in CQ, once Rx is taken down
+ */
+static void
+bnad_cq_cmpl_init(struct bnad *bnad, struct bna_ccb *ccb)
+{
+	struct bna_cq_entry *cmpl, *next_cmpl;
+	unsigned int wi_range, wis = 0, ccb_prod = 0;
+	int i;
+
+	BNA_CQ_QPGE_PTR_GET(ccb_prod, ccb->sw_qpt, cmpl,
+			    wi_range);
+
+	for (i = 0; i < ccb->q_depth; i++) {
+		wis++;
+		if (likely(--wi_range))
+			next_cmpl = cmpl + 1;
+		else {
+			BNA_QE_INDX_ADD(ccb_prod, wis, ccb->q_depth);
+			wis = 0;
+			BNA_CQ_QPGE_PTR_GET(ccb_prod, ccb->sw_qpt,
+						next_cmpl, wi_range);
+		}
+		cmpl->valid = 0;
+		cmpl = next_cmpl;
+	}
+}
+
+/*
+ * Frees all pending Tx Bufs
+ * At this point no activity is expected on the Q,
+ * so DMA unmap & freeing is fine.
+ */
+static void
+bnad_free_all_txbufs(struct bnad *bnad,
+		 struct bna_tcb *tcb)
+{
+	u16 		unmap_cons;
+	struct bnad_unmap_q *unmap_q = tcb->unmap_q;
+	struct bnad_skb_unmap *unmap_array;
+	struct sk_buff 		*skb = NULL;
+	int			i;
+
+	unmap_array = unmap_q->unmap_array;
+
+	unmap_cons = 0;
+	while (unmap_cons < unmap_q->q_depth) {
+		skb = unmap_array[unmap_cons].skb;
+		if (!skb) {
+			unmap_cons++;
+			continue;
+		}
+		unmap_array[unmap_cons].skb = NULL;
+
+		pci_unmap_single(bnad->pcidev,
+				 pci_unmap_addr(&unmap_array[unmap_cons],
+						dma_addr), skb_headlen(skb),
+						PCI_DMA_TODEVICE);
+
+		pci_unmap_addr_set(&unmap_array[unmap_cons], dma_addr, 0);
+		unmap_cons++;
+		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+			pci_unmap_page(bnad->pcidev,
+				       pci_unmap_addr(&unmap_array[unmap_cons],
+						      dma_addr),
+				       skb_shinfo(skb)->frags[i].size,
+				       PCI_DMA_TODEVICE);
+			pci_unmap_addr_set(&unmap_array[unmap_cons], dma_addr,
+					   0);
+			unmap_cons++;
+		}
+		dev_kfree_skb_any(skb);
+	}
+}
+
+/* Data Path Handlers */
+
+/*
+ * bnad_free_txbufs : Frees the Tx bufs on Tx completion
+ * Can be called in a) Interrupt context
+ *		    b) Sending context
+ *		    c) Tasklet context
+ */
+static u32
+bnad_free_txbufs(struct bnad *bnad,
+		 struct bna_tcb *tcb)
+{
+	u32 		sent_packets = 0, sent_bytes = 0;
+	u16 		wis, unmap_cons, updated_hw_cons;
+	struct bnad_unmap_q *unmap_q = tcb->unmap_q;
+	struct bnad_skb_unmap *unmap_array;
+	struct sk_buff 		*skb;
+	int i;
+
+	/*
+	 * Just return if TX is stopped. This check is useful
+	 * when bnad_free_txbufs() runs out of a tasklet scheduled
+	 * before bnad_cb_tx_cleanup() cleared BNAD_RF_TX_STARTED bit
+	 * but this routine runs actually after the cleanup has been
+	 * executed.
+	 */
+	if (!test_bit(BNAD_RF_TX_STARTED, &bnad->run_flags))
+		return 0;
+
+	updated_hw_cons = *(tcb->hw_consumer_index);
+
+	wis = BNA_Q_INDEX_CHANGE(tcb->consumer_index,
+				  updated_hw_cons, tcb->q_depth);
+
+	BUG_ON(!(wis <= BNA_QE_IN_USE_CNT(tcb, tcb->q_depth)));
+
+	unmap_array = unmap_q->unmap_array;
+	unmap_cons = unmap_q->consumer_index;
+
+	prefetch(&unmap_array[unmap_cons + 1]);
+	while (wis) {
+		skb = unmap_array[unmap_cons].skb;
+
+		unmap_array[unmap_cons].skb = NULL;
+
+		sent_packets++;
+		sent_bytes += skb->len;
+		wis -= BNA_TXQ_WI_NEEDED(1 + skb_shinfo(skb)->nr_frags);
+
+		pci_unmap_single(bnad->pcidev,
+				 pci_unmap_addr(&unmap_array[unmap_cons],
+						dma_addr), skb_headlen(skb),
+				 PCI_DMA_TODEVICE);
+		pci_unmap_addr_set(&unmap_array[unmap_cons], dma_addr, 0);
+		BNA_QE_INDX_ADD(unmap_cons, 1, unmap_q->q_depth);
+
+		prefetch(&unmap_array[unmap_cons + 1]);
+		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+			prefetch(&unmap_array[unmap_cons + 1]);
+
+			pci_unmap_page(bnad->pcidev,
+				       pci_unmap_addr(&unmap_array[unmap_cons],
+						      dma_addr),
+				       skb_shinfo(skb)->frags[i].size,
+				       PCI_DMA_TODEVICE);
+			pci_unmap_addr_set(&unmap_array[unmap_cons], dma_addr,
+					   0);
+			BNA_QE_INDX_ADD(unmap_cons, 1, unmap_q->q_depth);
+		}
+		dev_kfree_skb_any(skb);
+	}
+
+	/* Update consumer pointers. */
+	tcb->consumer_index = updated_hw_cons;
+	unmap_q->consumer_index = unmap_cons;
+
+	tcb->txq->tx_packets += sent_packets;
+	tcb->txq->tx_bytes += sent_bytes;
+
+	return sent_packets;
+}
+
+/* Tx Free Tasklet function */
+/* Frees for all the tcb's in all the Tx's */
+/*
+ * Scheduled from sending context, so that
+ * the fat Tx lock is not held for too long
+ * in the sending context.
+ */
+static void
+bnad_tx_free_tasklet(unsigned long bnad_ptr)
+{
+	struct bnad *bnad = (struct bnad *)bnad_ptr;
+	struct bna_tcb *tcb;
+	u32 		acked;
+	int			i, j;
+
+	for (i = 0; i < bnad->num_tx; i++) {
+		for (j = 0; j < bnad->num_txq_per_tx; j++) {
+			tcb = bnad->tx_info[i].tcb[j];
+			if (!tcb)
+				continue;
+			if (((u16) (*tcb->hw_consumer_index) !=
+				tcb->consumer_index) &&
+				(!test_and_set_bit(BNAD_TXQ_FREE_SENT,
+						  &tcb->flags))) {
+				acked = bnad_free_txbufs(bnad, tcb);
+				bna_ib_ack(tcb->i_dbell, acked);
+				smp_mb__before_clear_bit();
+				clear_bit(BNAD_TXQ_FREE_SENT, &tcb->flags);
+			}
+		}
+	}
+}
+
+static u32
+bnad_tx(struct bnad *bnad, struct bna_tcb *tcb)
+{
+	struct net_device *netdev = bnad->netdev;
+	u32 sent;
+
+	if (test_and_set_bit(BNAD_TXQ_FREE_SENT, &tcb->flags))
+		return 0;
+
+	sent = bnad_free_txbufs(bnad, tcb);
+	if (sent) {
+		if (netif_queue_stopped(netdev) &&
+		    netif_carrier_ok(netdev) &&
+		    BNA_QE_FREE_CNT(tcb, tcb->q_depth) >=
+				    BNAD_NETIF_WAKE_THRESHOLD) {
+			netif_wake_queue(netdev);
+			BNAD_UPDATE_CTR(bnad, netif_queue_wakeup);
+		}
+		bna_ib_ack(tcb->i_dbell, sent);
+	} else
+		bna_ib_ack(tcb->i_dbell, 0);
+
+	smp_mb__before_clear_bit();
+	clear_bit(BNAD_TXQ_FREE_SENT, &tcb->flags);
+
+	return sent;
+}
+
+/* MSIX Tx Completion Handler */
+static irqreturn_t
+bnad_msix_tx(int irq, void *data)
+{
+	struct bna_tcb *tcb = (struct bna_tcb *)data;
+	struct bnad *bnad = tcb->bnad;
+
+	bnad_tx(bnad, tcb);
+
+	return IRQ_HANDLED;
+}
+
+static void
+bnad_reset_rcb(struct bnad *bnad, struct bna_rcb *rcb)
+{
+	struct bnad_unmap_q *unmap_q = rcb->unmap_q;
+
+	rcb->producer_index = 0;
+	rcb->consumer_index = 0;
+
+	unmap_q->producer_index = 0;
+	unmap_q->consumer_index = 0;
+}
+
+static void
+bnad_free_rxbufs(struct bnad *bnad, struct bna_rcb *rcb)
+{
+	struct bnad_unmap_q *unmap_q;
+	struct sk_buff *skb;
+
+	unmap_q = rcb->unmap_q;
+	while (BNA_QE_IN_USE_CNT(unmap_q, unmap_q->q_depth)) {
+		skb = unmap_q->unmap_array[unmap_q->consumer_index].skb;
+		BUG_ON(!(skb));
+		unmap_q->unmap_array[unmap_q->consumer_index].skb = NULL;
+		pci_unmap_single(bnad->pcidev, pci_unmap_addr(&unmap_q->
+					unmap_array[unmap_q->consumer_index],
+					dma_addr), rcb->rxq->buffer_size +
+					NET_IP_ALIGN, PCI_DMA_FROMDEVICE);
+		dev_kfree_skb(skb);
+		BNA_QE_INDX_ADD(unmap_q->consumer_index, 1, unmap_q->q_depth);
+		BNA_QE_INDX_ADD(rcb->consumer_index, 1, rcb->q_depth);
+	}
+
+	bnad_reset_rcb(bnad, rcb);
+}
+
+static void
+bnad_alloc_n_post_rxbufs(struct bnad *bnad, struct bna_rcb *rcb)
+{
+	u16 to_alloc, alloced, unmap_prod, wi_range;
+	struct bnad_unmap_q *unmap_q = rcb->unmap_q;
+	struct bnad_skb_unmap *unmap_array;
+	struct bna_rxq_entry *rxent;
+	struct sk_buff *skb;
+	dma_addr_t dma_addr;
+
+	alloced = 0;
+	to_alloc =
+		BNA_QE_FREE_CNT(unmap_q, unmap_q->q_depth);
+
+	unmap_array = unmap_q->unmap_array;
+	unmap_prod = unmap_q->producer_index;
+
+	BNA_RXQ_QPGE_PTR_GET(unmap_prod, rcb->sw_qpt, rxent, wi_range);
+
+	while (to_alloc--) {
+		if (!wi_range) {
+			BNA_RXQ_QPGE_PTR_GET(unmap_prod, rcb->sw_qpt, rxent,
+					     wi_range);
+		}
+		skb = alloc_skb(rcb->rxq->buffer_size + NET_IP_ALIGN,
+				     GFP_ATOMIC);
+		if (unlikely(!skb)) {
+			BNAD_UPDATE_CTR(bnad, rxbuf_alloc_failed);
+			goto finishing;
+		}
+		skb->dev = bnad->netdev;
+		skb_reserve(skb, NET_IP_ALIGN);
+		unmap_array[unmap_prod].skb = skb;
+		dma_addr = pci_map_single(bnad->pcidev, skb->data,
+			rcb->rxq->buffer_size, PCI_DMA_FROMDEVICE);
+		pci_unmap_addr_set(&unmap_array[unmap_prod], dma_addr,
+				   dma_addr);
+		BNA_SET_DMA_ADDR(dma_addr, &rxent->host_addr);
+		BNA_QE_INDX_ADD(unmap_prod, 1, unmap_q->q_depth);
+
+		rxent++;
+		wi_range--;
+		alloced++;
+	}
+
+finishing:
+	if (likely(alloced)) {
+		unmap_q->producer_index = unmap_prod;
+		rcb->producer_index = unmap_prod;
+		smp_mb();
+		bna_rxq_prod_indx_doorbell(rcb);
+	}
+}
+
+/*
+ * Locking is required in the enable path
+ * because it is called from a napi poll
+ * context, where the bna_lock is not held
+ * unlike the IRQ context.
+ */
+static void
+bnad_enable_txrx_irqs(struct bnad *bnad)
+{
+	struct bna_tcb *tcb;
+	struct bna_ccb *ccb;
+	int i, j;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	for (i = 0; i < bnad->num_tx; i++) {
+		for (j = 0; j < bnad->num_txq_per_tx; j++) {
+			tcb = bnad->tx_info[i].tcb[j];
+			bna_ib_coalescing_timer_set(tcb->i_dbell,
+				tcb->txq->ib->ib_config.coalescing_timeo);
+			bna_ib_ack(tcb->i_dbell, 0);
+		}
+	}
+
+	for (i = 0; i < bnad->num_rx; i++) {
+		for (j = 0; j < bnad->num_rxp_per_rx; j++) {
+			ccb = bnad->rx_info[i].rx_ctrl[j].ccb;
+			bnad_enable_rx_irq_unsafe(ccb);
+		}
+	}
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+}
+
+static inline void
+bnad_refill_rxq(struct bnad *bnad, struct bna_rcb *rcb)
+{
+	struct bnad_unmap_q *unmap_q = rcb->unmap_q;
+
+	if (!test_and_set_bit(BNAD_RXQ_REFILL, &rcb->flags)) {
+		if (BNA_QE_FREE_CNT(unmap_q, unmap_q->q_depth)
+			 >> BNAD_RXQ_REFILL_THRESHOLD_SHIFT)
+			bnad_alloc_n_post_rxbufs(bnad, rcb);
+		smp_mb__before_clear_bit();
+		clear_bit(BNAD_RXQ_REFILL, &rcb->flags);
+	}
+}
+
+static u32
+bnad_poll_cq(struct bnad *bnad, struct bna_ccb *ccb, int budget)
+{
+	struct bna_cq_entry *cmpl, *next_cmpl;
+	struct bna_rcb *rcb = NULL;
+	unsigned int wi_range, packets = 0, wis = 0;
+	struct bnad_unmap_q *unmap_q;
+	struct sk_buff *skb;
+	u32 flags;
+	u32 qid0 = ccb->rcb[0]->rxq->rxq_id;
+	struct bna_pkt_rate *pkt_rt = &ccb->pkt_rate;
+
+	prefetch(bnad->netdev);
+	BNA_CQ_QPGE_PTR_GET(ccb->producer_index, ccb->sw_qpt, cmpl,
+			    wi_range);
+	BUG_ON(!(wi_range <= ccb->q_depth));
+	while (cmpl->valid && packets < budget) {
+		packets++;
+		BNA_UPDATE_PKT_CNT(pkt_rt, ntohs(cmpl->length));
+
+		if (qid0 == cmpl->rxq_id)
+			rcb = ccb->rcb[0];
+		else
+			rcb = ccb->rcb[1];
+
+		unmap_q = rcb->unmap_q;
+
+		skb = unmap_q->unmap_array[unmap_q->consumer_index].skb;
+		BUG_ON(!(skb));
+		unmap_q->unmap_array[unmap_q->consumer_index].skb = NULL;
+		pci_unmap_single(bnad->pcidev,
+				 pci_unmap_addr(&unmap_q->
+						unmap_array[unmap_q->
+							    consumer_index],
+						dma_addr),
+						rcb->rxq->buffer_size,
+						PCI_DMA_FROMDEVICE);
+		BNA_QE_INDX_ADD(unmap_q->consumer_index, 1, unmap_q->q_depth);
+
+		/* Should be more efficient ? Performance ? */
+		BNA_QE_INDX_ADD(rcb->consumer_index, 1, rcb->q_depth);
+
+		wis++;
+		if (likely(--wi_range))
+			next_cmpl = cmpl + 1;
+		else {
+			BNA_QE_INDX_ADD(ccb->producer_index, wis, ccb->q_depth);
+			wis = 0;
+			BNA_CQ_QPGE_PTR_GET(ccb->producer_index, ccb->sw_qpt,
+						next_cmpl, wi_range);
+			BUG_ON(!(wi_range <= ccb->q_depth));
+		}
+		prefetch(next_cmpl);
+
+		flags = ntohl(cmpl->flags);
+		if (unlikely
+		    (flags &
+		     (BNA_CQ_EF_MAC_ERROR | BNA_CQ_EF_FCS_ERROR |
+		      BNA_CQ_EF_TOO_LONG))) {
+			dev_kfree_skb_any(skb);
+			rcb->rxq->rx_packets_with_error++;
+			goto next;
+		}
+
+		skb_put(skb, ntohs(cmpl->length));
+		if (likely
+		    (bnad->rx_csum &&
+		     (((flags & BNA_CQ_EF_IPV4) &&
+		      (flags & BNA_CQ_EF_L3_CKSUM_OK)) ||
+		      (flags & BNA_CQ_EF_IPV6)) &&
+		      (flags & (BNA_CQ_EF_TCP | BNA_CQ_EF_UDP)) &&
+		      (flags & BNA_CQ_EF_L4_CKSUM_OK)))
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+		else
+			skb->ip_summed = CHECKSUM_NONE;
+
+		rcb->rxq->rx_packets++;
+		rcb->rxq->rx_bytes += skb->len;
+		skb->protocol = eth_type_trans(skb, bnad->netdev);
+
+		if (bnad->vlan_grp && (flags & BNA_CQ_EF_VLAN)) {
+			struct bnad_rx_ctrl *rx_ctrl =
+				(struct bnad_rx_ctrl *)ccb->ctrl;
+			if (skb->ip_summed == CHECKSUM_UNNECESSARY)
+				vlan_gro_receive(&rx_ctrl->napi, bnad->vlan_grp,
+						ntohs(cmpl->vlan_tag), skb);
+			else
+				vlan_hwaccel_receive_skb(skb,
+							 bnad->vlan_grp,
+							 ntohs(cmpl->vlan_tag));
+
+		} else { /* Not VLAN tagged/stripped */
+			struct bnad_rx_ctrl *rx_ctrl =
+				(struct bnad_rx_ctrl *)ccb->ctrl;
+			if (skb->ip_summed == CHECKSUM_UNNECESSARY)
+				napi_gro_receive(&rx_ctrl->napi, skb);
+			else
+				netif_receive_skb(skb);
+		}
+
+next:
+		cmpl->valid = 0;
+		cmpl = next_cmpl;
+	}
+
+	BNA_QE_INDX_ADD(ccb->producer_index, wis, ccb->q_depth);
+
+	if (likely(ccb)) {
+		bna_ib_ack(ccb->i_dbell, packets);
+		bnad_refill_rxq(bnad, ccb->rcb[0]);
+		if (ccb->rcb[1])
+			bnad_refill_rxq(bnad, ccb->rcb[1]);
+	} else
+		bna_ib_ack(ccb->i_dbell, 0);
+
+	return packets;
+}
+
+static void
+bnad_disable_rx_irq(struct bnad *bnad, struct bna_ccb *ccb)
+{
+	bna_ib_coalescing_timer_set(ccb->i_dbell, 0);
+	bna_ib_ack(ccb->i_dbell, 0);
+}
+
+static void
+bnad_enable_rx_irq(struct bnad *bnad, struct bna_ccb *ccb)
+{
+	spin_lock_irq(&bnad->bna_lock); /* Because of polling context */
+	bnad_enable_rx_irq_unsafe(ccb);
+	spin_unlock_irq(&bnad->bna_lock);
+}
+
+static void
+bnad_netif_rx_schedule_poll(struct bnad *bnad, struct bna_ccb *ccb)
+{
+	struct bnad_rx_ctrl *rx_ctrl = (struct bnad_rx_ctrl *)(ccb->ctrl);
+	if (likely(napi_schedule_prep((&rx_ctrl->napi)))) {
+		bnad_disable_rx_irq(bnad, ccb);
+		__napi_schedule((&rx_ctrl->napi));
+	}
+	BNAD_UPDATE_CTR(bnad, netif_rx_schedule);
+}
+
+/* MSIX Rx Path Handler */
+static irqreturn_t
+bnad_msix_rx(int irq, void *data)
+{
+	struct bna_ccb *ccb = (struct bna_ccb *)data;
+	struct bnad *bnad = ccb->bnad;
+
+	bnad_netif_rx_schedule_poll(bnad, ccb);
+
+	return IRQ_HANDLED;
+}
+
+/* Interrupt handlers */
+
+/* Mbox Interrupt Handlers */
+static irqreturn_t
+bnad_msix_mbox_handler(int irq, void *data)
+{
+	u32 intr_status;
+	unsigned long  flags;
+	struct net_device *netdev = data;
+	struct bnad *bnad;
+
+	bnad = netdev_priv(netdev);
+
+	/* BNA_ISR_GET(bnad); Inc Ref count */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+
+	bna_intr_status_get(&bnad->bna, intr_status);
+
+	if (BNA_IS_MBOX_ERR_INTR(intr_status))
+		bna_mbox_handler(&bnad->bna, intr_status);
+
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	/* BNAD_ISR_PUT(bnad); Dec Ref count */
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t
+bnad_isr(int irq, void *data)
+{
+	int i, j;
+	u32 intr_status;
+	unsigned long flags;
+	struct net_device *netdev = data;
+	struct bnad *bnad = netdev_priv(netdev);
+	struct bnad_rx_info *rx_info;
+	struct bnad_rx_ctrl *rx_ctrl;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+
+	bna_intr_status_get(&bnad->bna, intr_status);
+	if (!intr_status) {
+		spin_unlock_irqrestore(&bnad->bna_lock, flags);
+		return IRQ_NONE;
+	}
+
+	if (BNA_IS_MBOX_ERR_INTR(intr_status)) {
+		bna_mbox_handler(&bnad->bna, intr_status);
+		if (!BNA_IS_INTX_DATA_INTR(intr_status)) {
+			spin_unlock_irqrestore(&bnad->bna_lock, flags);
+			goto done;
+		}
+	}
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	/* Process data interrupts */
+	for (i = 0; i < bnad->num_rx; i++) {
+		rx_info = &bnad->rx_info[i];
+		if (!rx_info->rx)
+			continue;
+		for (j = 0; j < bnad->num_rxp_per_rx; j++) {
+			rx_ctrl = &rx_info->rx_ctrl[j];
+			if (rx_ctrl->ccb)
+				bnad_netif_rx_schedule_poll(bnad,
+							    rx_ctrl->ccb);
+		}
+	}
+done:
+	return IRQ_HANDLED;
+}
+
+/*
+ * Called in interrupt / callback context
+ * with bna_lock held, so cfg_flags access is OK
+ */
+static void
+bnad_enable_mbox_irq(struct bnad *bnad)
+{
+	int irq = BNAD_GET_MBOX_IRQ(bnad);
+
+	if (!(bnad->cfg_flags & BNAD_CF_MSIX))
+		return;
+
+	if (test_and_clear_bit(BNAD_RF_MBOX_IRQ_DISABLED, &bnad->run_flags))
+		enable_irq(irq);
+	BNAD_UPDATE_CTR(bnad, mbox_intr_enabled);
+}
+
+/*
+ * Called with bnad->bna_lock held b'cos of
+ * bnad->cfg_flags access.
+ */
+void
+bnad_disable_mbox_irq(struct bnad *bnad)
+{
+	int irq = BNAD_GET_MBOX_IRQ(bnad);
+
+	if (!(bnad->cfg_flags & BNAD_CF_MSIX))
+		return;
+
+	if (!test_and_set_bit(BNAD_RF_MBOX_IRQ_DISABLED, &bnad->run_flags))
+		disable_irq_nosync(irq);
+	BNAD_UPDATE_CTR(bnad, mbox_intr_disabled);
+}
+
+/* Control Path Handlers */
+
+/* Callbacks */
+void
+bnad_cb_device_enable_mbox_intr(struct bnad *bnad)
+{
+	bnad_enable_mbox_irq(bnad);
+}
+
+void
+bnad_cb_device_disable_mbox_intr(struct bnad *bnad)
+{
+	bnad_disable_mbox_irq(bnad);
+}
+
+void
+bnad_cb_device_enabled(struct bnad *bnad, enum bna_cb_status status)
+{
+	complete(&bnad->bnad_completions.ioc_comp);
+	bnad->bnad_completions.ioc_comp_status = status;
+}
+
+void
+bnad_cb_device_disabled(struct bnad *bnad, enum bna_cb_status status)
+{
+	complete(&bnad->bnad_completions.ioc_comp);
+	bnad->bnad_completions.ioc_comp_status = status;
+}
+
+static void
+bnad_cb_port_disabled(void *arg, enum bna_cb_status status)
+{
+	struct bnad *bnad = (struct bnad *)arg;
+
+	complete(&bnad->bnad_completions.port_comp);
+
+	netif_carrier_off(bnad->netdev);
+}
+
+void
+bnad_cb_port_link_status(struct bnad *bnad,
+			enum bna_link_status link_status)
+{
+	bool link_up = 0;
+
+	link_up = (link_status == BNA_LINK_UP) || (link_status == BNA_CEE_UP);
+
+	if (link_status == BNA_CEE_UP) {
+		set_bit(BNAD_RF_CEE_RUNNING, &bnad->run_flags);
+		BNAD_UPDATE_CTR(bnad, cee_up);
+	} else
+		clear_bit(BNAD_RF_CEE_RUNNING, &bnad->run_flags);
+
+	if (link_up) {
+		if (!netif_carrier_ok(bnad->netdev)) {
+			pr_warn("bna: %s link up\n",
+				bnad->netdev->name);
+			netif_carrier_on(bnad->netdev);
+			BNAD_UPDATE_CTR(bnad, link_toggle);
+			if (test_bit(BNAD_RF_TX_STARTED, &bnad->run_flags)) {
+				/* Force an immediate Transmit Schedule */
+				pr_info("bna: %s TX_STARTED\n",
+					bnad->netdev->name);
+				netif_wake_queue(bnad->netdev);
+				BNAD_UPDATE_CTR(bnad, netif_queue_wakeup);
+			} else {
+				netif_stop_queue(bnad->netdev);
+				BNAD_UPDATE_CTR(bnad, netif_queue_stop);
+			}
+		}
+	} else {
+		if (netif_carrier_ok(bnad->netdev)) {
+			pr_warn("bna: %s link down\n",
+				bnad->netdev->name);
+			netif_carrier_off(bnad->netdev);
+			BNAD_UPDATE_CTR(bnad, link_toggle);
+		}
+	}
+}
+
+static void
+bnad_cb_tx_disabled(void *arg, struct bna_tx *tx,
+			enum bna_cb_status status)
+{
+	struct bnad *bnad = (struct bnad *)arg;
+
+	complete(&bnad->bnad_completions.tx_comp);
+}
+
+static void
+bnad_cb_tcb_setup(struct bnad *bnad, struct bna_tcb *tcb)
+{
+	struct bnad_tx_info *tx_info =
+			(struct bnad_tx_info *)tcb->txq->tx->priv;
+	struct bnad_unmap_q *unmap_q = tcb->unmap_q;
+
+	tx_info->tcb[tcb->id] = tcb;
+	unmap_q->producer_index = 0;
+	unmap_q->consumer_index = 0;
+	unmap_q->q_depth = BNAD_TX_UNMAPQ_DEPTH;
+}
+
+static void
+bnad_cb_tcb_destroy(struct bnad *bnad, struct bna_tcb *tcb)
+{
+	struct bnad_tx_info *tx_info =
+			(struct bnad_tx_info *)tcb->txq->tx->priv;
+
+	tx_info->tcb[tcb->id] = NULL;
+}
+
+static void
+bnad_cb_rcb_setup(struct bnad *bnad, struct bna_rcb *rcb)
+{
+	struct bnad_unmap_q *unmap_q = rcb->unmap_q;
+
+	unmap_q->producer_index = 0;
+	unmap_q->consumer_index = 0;
+	unmap_q->q_depth = BNAD_RX_UNMAPQ_DEPTH;
+}
+
+static void
+bnad_cb_ccb_setup(struct bnad *bnad, struct bna_ccb *ccb)
+{
+	struct bnad_rx_info *rx_info =
+			(struct bnad_rx_info *)ccb->cq->rx->priv;
+
+	rx_info->rx_ctrl[ccb->id].ccb = ccb;
+	ccb->ctrl = &rx_info->rx_ctrl[ccb->id];
+}
+
+static void
+bnad_cb_ccb_destroy(struct bnad *bnad, struct bna_ccb *ccb)
+{
+	struct bnad_rx_info *rx_info =
+			(struct bnad_rx_info *)ccb->cq->rx->priv;
+
+	rx_info->rx_ctrl[ccb->id].ccb = NULL;
+}
+
+static void
+bnad_cb_tx_stall(struct bnad *bnad, struct bna_tcb *tcb)
+{
+	struct bnad_tx_info *tx_info =
+			(struct bnad_tx_info *)tcb->txq->tx->priv;
+
+	if (tx_info != &bnad->tx_info[0])
+		return;
+
+	clear_bit(BNAD_RF_TX_STARTED, &bnad->run_flags);
+	netif_stop_queue(bnad->netdev);
+	pr_info("bna: %s TX_STOPPED\n", bnad->netdev->name);
+}
+
+static void
+bnad_cb_tx_resume(struct bnad *bnad, struct bna_tcb *tcb)
+{
+	if (test_and_set_bit(BNAD_RF_TX_STARTED, &bnad->run_flags))
+		return;
+
+	if (netif_carrier_ok(bnad->netdev)) {
+		pr_info("bna: %s TX_STARTED\n", bnad->netdev->name);
+		netif_wake_queue(bnad->netdev);
+		BNAD_UPDATE_CTR(bnad, netif_queue_wakeup);
+	}
+}
+
+static void
+bnad_cb_tx_cleanup(struct bnad *bnad, struct bna_tcb *tcb)
+{
+	struct bnad_unmap_q *unmap_q = tcb->unmap_q;
+
+	if (!tcb || (!tcb->unmap_q))
+		return;
+
+	if (!unmap_q->unmap_array)
+		return;
+
+	if (test_and_set_bit(BNAD_TXQ_FREE_SENT, &tcb->flags))
+		return;
+
+	bnad_free_all_txbufs(bnad, tcb);
+
+	unmap_q->producer_index = 0;
+	unmap_q->consumer_index = 0;
+
+	smp_mb__before_clear_bit();
+	clear_bit(BNAD_TXQ_FREE_SENT, &tcb->flags);
+}
+
+static void
+bnad_cb_rx_cleanup(struct bnad *bnad,
+			struct bna_ccb *ccb)
+{
+	bnad_cq_cmpl_init(bnad, ccb);
+
+	bnad_free_rxbufs(bnad, ccb->rcb[0]);
+	clear_bit(BNAD_RXQ_STARTED, &ccb->rcb[0]->flags);
+
+	if (ccb->rcb[1]) {
+		bnad_free_rxbufs(bnad, ccb->rcb[1]);
+		clear_bit(BNAD_RXQ_STARTED, &ccb->rcb[1]->flags);
+	}
+}
+
+static void
+bnad_cb_rx_post(struct bnad *bnad, struct bna_rcb *rcb)
+{
+	struct bnad_unmap_q *unmap_q = rcb->unmap_q;
+
+	set_bit(BNAD_RXQ_STARTED, &rcb->flags);
+
+	/* Now allocate & post buffers for this RCB */
+	/* !!Allocation in callback context */
+	if (!test_and_set_bit(BNAD_RXQ_REFILL, &rcb->flags)) {
+		if (BNA_QE_FREE_CNT(unmap_q, unmap_q->q_depth)
+			 >> BNAD_RXQ_REFILL_THRESHOLD_SHIFT)
+			bnad_alloc_n_post_rxbufs(bnad, rcb);
+		smp_mb__before_clear_bit();
+		clear_bit(BNAD_RXQ_REFILL, &rcb->flags);
+	}
+}
+
+static void
+bnad_cb_rx_disabled(void *arg, struct bna_rx *rx,
+			enum bna_cb_status status)
+{
+	struct bnad *bnad = (struct bnad *)arg;
+
+	complete(&bnad->bnad_completions.rx_comp);
+}
+
+static void
+bnad_cb_rx_mcast_add(struct bnad *bnad, struct bna_rx *rx,
+				enum bna_cb_status status)
+{
+	bnad->bnad_completions.mcast_comp_status = status;
+	complete(&bnad->bnad_completions.mcast_comp);
+}
+
+void
+bnad_cb_stats_get(struct bnad *bnad, enum bna_cb_status status,
+		       struct bna_stats *stats)
+{
+	if (status == BNA_CB_SUCCESS)
+		BNAD_UPDATE_CTR(bnad, hw_stats_updates);
+
+	if (!netif_running(bnad->netdev) ||
+		!test_bit(BNAD_RF_STATS_TIMER_RUNNING, &bnad->run_flags))
+		return;
+
+	mod_timer(&bnad->stats_timer,
+		  jiffies + msecs_to_jiffies(BNAD_STATS_TIMER_FREQ));
+}
+
+void
+bnad_cb_stats_clr(struct bnad *bnad)
+{
+}
+
+/* Resource allocation, free functions */
+
+static void
+bnad_mem_free(struct bnad *bnad,
+	      struct bna_mem_info *mem_info)
+{
+	int i;
+	dma_addr_t dma_pa;
+
+	if (mem_info->mdl == NULL)
+		return;
+
+	for (i = 0; i < mem_info->num; i++) {
+		if (mem_info->mdl[i].kva != NULL) {
+			if (mem_info->mem_type == BNA_MEM_T_DMA) {
+				BNA_GET_DMA_ADDR(&(mem_info->mdl[i].dma),
+						dma_pa);
+				pci_free_consistent(bnad->pcidev,
+						mem_info->mdl[i].len,
+						mem_info->mdl[i].kva, dma_pa);
+			} else
+				kfree(mem_info->mdl[i].kva);
+		}
+	}
+	kfree(mem_info->mdl);
+	mem_info->mdl = NULL;
+}
+
+static int
+bnad_mem_alloc(struct bnad *bnad,
+	       struct bna_mem_info *mem_info)
+{
+	int i;
+	dma_addr_t dma_pa;
+
+	if ((mem_info->num == 0) || (mem_info->len == 0)) {
+		mem_info->mdl = NULL;
+		return 0;
+	}
+
+	mem_info->mdl = kcalloc(mem_info->num, sizeof(struct bna_mem_descr),
+				GFP_KERNEL);
+	if (mem_info->mdl == NULL)
+		return -ENOMEM;
+
+	if (mem_info->mem_type == BNA_MEM_T_DMA) {
+		for (i = 0; i < mem_info->num; i++) {
+			mem_info->mdl[i].len = mem_info->len;
+			mem_info->mdl[i].kva =
+				pci_alloc_consistent(bnad->pcidev,
+						mem_info->len, &dma_pa);
+
+			if (mem_info->mdl[i].kva == NULL)
+				goto err_return;
+
+			BNA_SET_DMA_ADDR(dma_pa,
+					 &(mem_info->mdl[i].dma));
+		}
+	} else {
+		for (i = 0; i < mem_info->num; i++) {
+			mem_info->mdl[i].len = mem_info->len;
+			mem_info->mdl[i].kva = kzalloc(mem_info->len,
+							GFP_KERNEL);
+			if (mem_info->mdl[i].kva == NULL)
+				goto err_return;
+		}
+	}
+
+	return 0;
+
+err_return:
+	bnad_mem_free(bnad, mem_info);
+	return -ENOMEM;
+}
+
+/* Free IRQ for Mailbox */
+static void
+bnad_mbox_irq_free(struct bnad *bnad,
+		   struct bna_intr_info *intr_info)
+{
+	int irq;
+	unsigned long flags;
+
+	if (intr_info->idl == NULL)
+		return;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+
+	bnad_disable_mbox_irq(bnad);
+
+	irq = BNAD_GET_MBOX_IRQ(bnad);
+	free_irq(irq, bnad->netdev);
+
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	kfree(intr_info->idl);
+}
+
+/*
+ * Allocates IRQ for Mailbox, but keep it disabled
+ * This will be enabled once we get the mbox enable callback
+ * from bna
+ */
+static int
+bnad_mbox_irq_alloc(struct bnad *bnad,
+		    struct bna_intr_info *intr_info)
+{
+	int 		err;
+	unsigned long 	flags;
+	u32	irq;
+	irq_handler_t 	irq_handler;
+
+	/* Mbox should use only 1 vector */
+
+	intr_info->idl = kzalloc(sizeof(*(intr_info->idl)), GFP_KERNEL);
+	if (!intr_info->idl)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (bnad->cfg_flags & BNAD_CF_MSIX) {
+		irq_handler = (irq_handler_t)bnad_msix_mbox_handler;
+		irq = bnad->msix_table[bnad->msix_num - 1].vector;
+		flags = 0;
+		intr_info->intr_type = BNA_INTR_T_MSIX;
+		intr_info->idl[0].vector = bnad->msix_num - 1;
+	} else {
+		irq_handler = (irq_handler_t)bnad_isr;
+		irq = bnad->pcidev->irq;
+		flags = IRQF_SHARED;
+		intr_info->intr_type = BNA_INTR_T_INTX;
+		/* intr_info->idl.vector = 0 ? */
+	}
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	sprintf(bnad->mbox_irq_name, "%s", BNAD_NAME);
+
+	err = request_irq(irq, irq_handler, flags,
+			  bnad->mbox_irq_name, bnad->netdev);
+	if (err) {
+		kfree(intr_info->idl);
+		intr_info->idl = NULL;
+		return err;
+	}
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bnad_disable_mbox_irq(bnad);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	return 0;
+}
+
+static void
+bnad_txrx_irq_free(struct bnad *bnad, struct bna_intr_info *intr_info)
+{
+	kfree(intr_info->idl);
+	intr_info->idl = NULL;
+}
+
+/* Allocates Interrupt Descriptor List for MSIX/INT-X vectors */
+static int
+bnad_txrx_irq_alloc(struct bnad *bnad, enum bnad_intr_source src,
+		    uint txrx_id, struct bna_intr_info *intr_info)
+{
+	int i, vector_start = 0;
+	u32 cfg_flags;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	cfg_flags = bnad->cfg_flags;
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	if (cfg_flags & BNAD_CF_MSIX) {
+		intr_info->intr_type = BNA_INTR_T_MSIX;
+		intr_info->idl = kcalloc(intr_info->num,
+					sizeof(struct bna_intr_descr),
+					GFP_KERNEL);
+		if (!intr_info->idl)
+			return -ENOMEM;
+
+		switch (src) {
+		case BNAD_INTR_TX:
+			vector_start = txrx_id;
+			break;
+
+		case BNAD_INTR_RX:
+			vector_start = bnad->num_tx * bnad->num_txq_per_tx +
+					txrx_id;
+			break;
+
+		default:
+			BUG();
+		}
+
+		for (i = 0; i < intr_info->num; i++)
+			intr_info->idl[i].vector = vector_start + i;
+	} else {
+		intr_info->intr_type = BNA_INTR_T_INTX;
+		intr_info->num = 1;
+		intr_info->idl = kcalloc(intr_info->num,
+					sizeof(struct bna_intr_descr),
+					GFP_KERNEL);
+		if (!intr_info->idl)
+			return -ENOMEM;
+
+		switch (src) {
+		case BNAD_INTR_TX:
+			intr_info->idl[0].vector = 0x1; /* Bit mask : Tx IB */
+			break;
+
+		case BNAD_INTR_RX:
+			intr_info->idl[0].vector = 0x2; /* Bit mask : Rx IB */
+			break;
+		}
+	}
+	return 0;
+}
+
+/**
+ * NOTE: Should be called for MSIX only
+ * Unregisters Tx MSIX vector(s) from the kernel
+ */
+static void
+bnad_tx_msix_unregister(struct bnad *bnad, struct bnad_tx_info *tx_info,
+			int num_txqs)
+{
+	int i;
+	int vector_num;
+
+	for (i = 0; i < num_txqs; i++) {
+		if (tx_info->tcb[i] == NULL)
+			continue;
+
+		vector_num = tx_info->tcb[i]->intr_vector;
+		free_irq(bnad->msix_table[vector_num].vector, tx_info->tcb[i]);
+	}
+}
+
+/**
+ * NOTE: Should be called for MSIX only
+ * Registers Tx MSIX vector(s) and ISR(s), cookie with the kernel
+ */
+static int
+bnad_tx_msix_register(struct bnad *bnad, struct bnad_tx_info *tx_info,
+			uint tx_id, int num_txqs)
+{
+	int i;
+	int err;
+	int vector_num;
+
+	for (i = 0; i < num_txqs; i++) {
+		vector_num = tx_info->tcb[i]->intr_vector;
+		sprintf(tx_info->tcb[i]->name, "%s TXQ %d", bnad->netdev->name,
+				tx_id + tx_info->tcb[i]->id);
+		err = request_irq(bnad->msix_table[vector_num].vector,
+				  (irq_handler_t)bnad_msix_tx, 0,
+				  tx_info->tcb[i]->name,
+				  tx_info->tcb[i]);
+		if (err)
+			goto err_return;
+	}
+
+	return 0;
+
+err_return:
+	if (i > 0)
+		bnad_tx_msix_unregister(bnad, tx_info, (i - 1));
+	return -1;
+}
+
+/**
+ * NOTE: Should be called for MSIX only
+ * Unregisters Rx MSIX vector(s) from the kernel
+ */
+static void
+bnad_rx_msix_unregister(struct bnad *bnad, struct bnad_rx_info *rx_info,
+			int num_rxps)
+{
+	int i;
+	int vector_num;
+
+	for (i = 0; i < num_rxps; i++) {
+		if (rx_info->rx_ctrl[i].ccb == NULL)
+			continue;
+
+		vector_num = rx_info->rx_ctrl[i].ccb->intr_vector;
+		free_irq(bnad->msix_table[vector_num].vector,
+			 rx_info->rx_ctrl[i].ccb);
+	}
+}
+
+/**
+ * NOTE: Should be called for MSIX only
+ * Registers Tx MSIX vector(s) and ISR(s), cookie with the kernel
+ */
+static int
+bnad_rx_msix_register(struct bnad *bnad, struct bnad_rx_info *rx_info,
+			uint rx_id, int num_rxps)
+{
+	int i;
+	int err;
+	int vector_num;
+
+	for (i = 0; i < num_rxps; i++) {
+		vector_num = rx_info->rx_ctrl[i].ccb->intr_vector;
+		sprintf(rx_info->rx_ctrl[i].ccb->name, "%s CQ %d",
+			bnad->netdev->name,
+			rx_id + rx_info->rx_ctrl[i].ccb->id);
+		err = request_irq(bnad->msix_table[vector_num].vector,
+				  (irq_handler_t)bnad_msix_rx, 0,
+				  rx_info->rx_ctrl[i].ccb->name,
+				  rx_info->rx_ctrl[i].ccb);
+		if (err)
+			goto err_return;
+	}
+
+	return 0;
+
+err_return:
+	if (i > 0)
+		bnad_rx_msix_unregister(bnad, rx_info, (i - 1));
+	return -1;
+}
+
+/* Free Tx object Resources */
+static void
+bnad_tx_res_free(struct bnad *bnad, struct bna_res_info *res_info)
+{
+	int i;
+
+	for (i = 0; i < BNA_TX_RES_T_MAX; i++) {
+		if (res_info[i].res_type == BNA_RES_T_MEM)
+			bnad_mem_free(bnad, &res_info[i].res_u.mem_info);
+		else if (res_info[i].res_type == BNA_RES_T_INTR)
+			bnad_txrx_irq_free(bnad, &res_info[i].res_u.intr_info);
+	}
+}
+
+/* Allocates memory and interrupt resources for Tx object */
+static int
+bnad_tx_res_alloc(struct bnad *bnad, struct bna_res_info *res_info,
+		  uint tx_id)
+{
+	int i, err = 0;
+
+	for (i = 0; i < BNA_TX_RES_T_MAX; i++) {
+		if (res_info[i].res_type == BNA_RES_T_MEM)
+			err = bnad_mem_alloc(bnad,
+					&res_info[i].res_u.mem_info);
+		else if (res_info[i].res_type == BNA_RES_T_INTR)
+			err = bnad_txrx_irq_alloc(bnad, BNAD_INTR_TX, tx_id,
+					&res_info[i].res_u.intr_info);
+		if (err)
+			goto err_return;
+	}
+	return 0;
+
+err_return:
+	bnad_tx_res_free(bnad, res_info);
+	return err;
+}
+
+/* Free Rx object Resources */
+static void
+bnad_rx_res_free(struct bnad *bnad, struct bna_res_info *res_info)
+{
+	int i;
+
+	for (i = 0; i < BNA_RX_RES_T_MAX; i++) {
+		if (res_info[i].res_type == BNA_RES_T_MEM)
+			bnad_mem_free(bnad, &res_info[i].res_u.mem_info);
+		else if (res_info[i].res_type == BNA_RES_T_INTR)
+			bnad_txrx_irq_free(bnad, &res_info[i].res_u.intr_info);
+	}
+}
+
+/* Allocates memory and interrupt resources for Rx object */
+static int
+bnad_rx_res_alloc(struct bnad *bnad, struct bna_res_info *res_info,
+		  uint rx_id)
+{
+	int i, err = 0;
+
+	/* All memory needs to be allocated before setup_ccbs */
+	for (i = 0; i < BNA_RX_RES_T_MAX; i++) {
+		if (res_info[i].res_type == BNA_RES_T_MEM)
+			err = bnad_mem_alloc(bnad,
+					&res_info[i].res_u.mem_info);
+		else if (res_info[i].res_type == BNA_RES_T_INTR)
+			err = bnad_txrx_irq_alloc(bnad, BNAD_INTR_RX, rx_id,
+					&res_info[i].res_u.intr_info);
+		if (err)
+			goto err_return;
+	}
+	return 0;
+
+err_return:
+	bnad_rx_res_free(bnad, res_info);
+	return err;
+}
+
+/* Timer callbacks */
+/* a) IOC timer */
+static void
+bnad_ioc_timeout(unsigned long data)
+{
+	struct bnad *bnad = (struct bnad *)data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bfa_ioc_timeout((void *) &bnad->bna.device.ioc);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+}
+
+static void
+bnad_ioc_hb_check(unsigned long data)
+{
+	struct bnad *bnad = (struct bnad *)data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bfa_ioc_hb_check((void *) &bnad->bna.device.ioc);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+}
+
+static void
+bnad_ioc_sem_timeout(unsigned long data)
+{
+	struct bnad *bnad = (struct bnad *)data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bfa_ioc_sem_timeout((void *) &bnad->bna.device.ioc);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+}
+
+/*
+ * All timer routines use bnad->bna_lock to protect against
+ * the following race, which may occur in case of no locking:
+ * 	Time	CPU m  		CPU n
+ *	0       1 = test_bit
+ *	1			clear_bit
+ *	2			del_timer_sync
+ *	3	mod_timer
+ */
+
+/* b) Dynamic Interrupt Moderation Timer */
+static void
+bnad_dim_timeout(unsigned long data)
+{
+	struct bnad *bnad = (struct bnad *)data;
+	struct bnad_rx_info *rx_info;
+	struct bnad_rx_ctrl *rx_ctrl;
+	int i, j;
+	unsigned long flags;
+
+	if (!netif_carrier_ok(bnad->netdev))
+		return;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	for (i = 0; i < bnad->num_rx; i++) {
+		rx_info = &bnad->rx_info[i];
+		if (!rx_info->rx)
+			continue;
+		for (j = 0; j < bnad->num_rxp_per_rx; j++) {
+			rx_ctrl = &rx_info->rx_ctrl[j];
+			if (!rx_ctrl->ccb)
+				continue;
+			bna_rx_dim_update(rx_ctrl->ccb);
+		}
+	}
+
+	/* Check for BNAD_CF_DIM_ENABLED, does not eleminate a race */
+	if (test_bit(BNAD_RF_DIM_TIMER_RUNNING, &bnad->run_flags))
+		mod_timer(&bnad->dim_timer,
+			  jiffies + msecs_to_jiffies(BNAD_DIM_TIMER_FREQ));
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+}
+
+/* c)  Statistics Timer */
+static void
+bnad_stats_timeout(unsigned long data)
+{
+	struct bnad *bnad = (struct bnad *)data;
+	unsigned long flags;
+
+	if (!netif_running(bnad->netdev) ||
+		!test_bit(BNAD_RF_STATS_TIMER_RUNNING, &bnad->run_flags))
+		return;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_stats_get(&bnad->bna);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+}
+
+/*
+ * Set up timer for DIM
+ * Called with bnad->bna_lock held
+ */
+void
+bnad_dim_timer_start(struct bnad *bnad)
+{
+	if (bnad->cfg_flags & BNAD_CF_DIM_ENABLED &&
+	    !test_bit(BNAD_RF_DIM_TIMER_RUNNING, &bnad->run_flags)) {
+		setup_timer(&bnad->dim_timer, bnad_dim_timeout,
+			    (unsigned long)bnad);
+		set_bit(BNAD_RF_DIM_TIMER_RUNNING, &bnad->run_flags);
+		mod_timer(&bnad->dim_timer,
+			  jiffies + msecs_to_jiffies(BNAD_DIM_TIMER_FREQ));
+	}
+}
+
+/*
+ * Set up timer for statistics
+ * Called with mutex_lock(&bnad->conf_mutex) held
+ */
+static void
+bnad_stats_timer_start(struct bnad *bnad)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (!test_and_set_bit(BNAD_RF_STATS_TIMER_RUNNING, &bnad->run_flags)) {
+		setup_timer(&bnad->stats_timer, bnad_stats_timeout,
+			    (unsigned long)bnad);
+		mod_timer(&bnad->stats_timer,
+			  jiffies + msecs_to_jiffies(BNAD_STATS_TIMER_FREQ));
+	}
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+}
+
+/*
+ * Stops the stats timer
+ * Called with mutex_lock(&bnad->conf_mutex) held
+ */
+static void
+bnad_stats_timer_stop(struct bnad *bnad)
+{
+	int to_del = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (test_and_clear_bit(BNAD_RF_STATS_TIMER_RUNNING, &bnad->run_flags))
+		to_del = 1;
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	if (to_del)
+		del_timer_sync(&bnad->stats_timer);
+}
+
+/* Utilities */
+
+static void
+bnad_netdev_mc_list_get(struct net_device *netdev, u8 *mc_list)
+{
+	int i = 1; /* Index 0 has broadcast address */
+	struct netdev_hw_addr *mc_addr;
+
+	netdev_for_each_mc_addr(mc_addr, netdev) {
+		memcpy(&mc_list[i * ETH_ALEN], &mc_addr->addr[0],
+							ETH_ALEN);
+		i++;
+	}
+}
+
+static int
+bnad_napi_poll_rx(struct napi_struct *napi, int budget)
+{
+	struct bnad_rx_ctrl *rx_ctrl =
+		container_of(napi, struct bnad_rx_ctrl, napi);
+	struct bna_ccb *ccb;
+	struct bnad *bnad;
+	int rcvd = 0;
+
+	ccb = rx_ctrl->ccb;
+
+	bnad = ccb->bnad;
+
+	if (!netif_carrier_ok(bnad->netdev))
+		goto poll_exit;
+
+	rcvd = bnad_poll_cq(bnad, ccb, budget);
+	if (rcvd == budget)
+		return rcvd;
+
+poll_exit:
+	napi_complete((napi));
+
+	BNAD_UPDATE_CTR(bnad, netif_rx_complete);
+
+	bnad_enable_rx_irq(bnad, ccb);
+	return rcvd;
+}
+
+static int
+bnad_napi_poll_txrx(struct napi_struct *napi, int budget)
+{
+	struct bnad_rx_ctrl *rx_ctrl =
+		container_of(napi, struct bnad_rx_ctrl, napi);
+	struct bna_ccb *ccb;
+	struct bnad *bnad;
+	int 			rcvd = 0;
+	int			i, j;
+
+	ccb = rx_ctrl->ccb;
+
+	bnad = ccb->bnad;
+
+	if (!netif_carrier_ok(bnad->netdev))
+		goto poll_exit;
+
+	/* Handle Tx Completions, if any */
+	for (i = 0; i < bnad->num_tx; i++) {
+		for (j = 0; j < bnad->num_txq_per_tx; j++)
+			bnad_tx(bnad, bnad->tx_info[i].tcb[j]);
+	}
+
+	/* Handle Rx Completions */
+	rcvd = bnad_poll_cq(bnad, ccb, budget);
+	if (rcvd == budget)
+		return rcvd;
+poll_exit:
+	napi_complete((napi));
+
+	BNAD_UPDATE_CTR(bnad, netif_rx_complete);
+
+	bnad_enable_txrx_irqs(bnad);
+	return rcvd;
+}
+
+static void
+bnad_napi_enable(struct bnad *bnad, u32 rx_id)
+{
+	int (*napi_poll) (struct napi_struct *, int);
+	struct bnad_rx_ctrl *rx_ctrl;
+	int i;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (bnad->cfg_flags & BNAD_CF_MSIX)
+		napi_poll = bnad_napi_poll_rx;
+	else
+		napi_poll = bnad_napi_poll_txrx;
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	/* Initialize & enable NAPI */
+	for (i = 0; i <	bnad->num_rxp_per_rx; i++) {
+		rx_ctrl = &bnad->rx_info[rx_id].rx_ctrl[i];
+		netif_napi_add(bnad->netdev, &rx_ctrl->napi,
+			       napi_poll, 64);
+		napi_enable(&rx_ctrl->napi);
+	}
+}
+
+static void
+bnad_napi_disable(struct bnad *bnad, u32 rx_id)
+{
+	int i;
+
+	/* First disable and then clean up */
+	for (i = 0; i < bnad->num_rxp_per_rx; i++) {
+		napi_disable(&bnad->rx_info[rx_id].rx_ctrl[i].napi);
+		netif_napi_del(&bnad->rx_info[rx_id].rx_ctrl[i].napi);
+	}
+}
+
+/* Should be held with conf_lock held */
+void
+bnad_cleanup_tx(struct bnad *bnad, uint tx_id)
+{
+	struct bnad_tx_info *tx_info = &bnad->tx_info[tx_id];
+	struct bna_res_info *res_info = &bnad->tx_res_info[tx_id].res_info[0];
+	unsigned long flags;
+
+	if (!tx_info->tx)
+		return;
+
+	init_completion(&bnad->bnad_completions.tx_comp);
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_tx_disable(tx_info->tx, BNA_HARD_CLEANUP, bnad_cb_tx_disabled);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	wait_for_completion(&bnad->bnad_completions.tx_comp);
+
+	if (tx_info->tcb[0]->intr_type == BNA_INTR_T_MSIX)
+		bnad_tx_msix_unregister(bnad, tx_info,
+			bnad->num_txq_per_tx);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_tx_destroy(tx_info->tx);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	tx_info->tx = NULL;
+
+	if (0 == tx_id)
+		tasklet_kill(&bnad->tx_free_tasklet);
+
+	bnad_tx_res_free(bnad, res_info);
+}
+
+/* Should be held with conf_lock held */
+int
+bnad_setup_tx(struct bnad *bnad, uint tx_id)
+{
+	int err;
+	struct bnad_tx_info *tx_info = &bnad->tx_info[tx_id];
+	struct bna_res_info *res_info = &bnad->tx_res_info[tx_id].res_info[0];
+	struct bna_intr_info *intr_info =
+			&res_info[BNA_TX_RES_INTR_T_TXCMPL].res_u.intr_info;
+	struct bna_tx_config *tx_config = &bnad->tx_config[tx_id];
+	struct bna_tx_event_cbfn tx_cbfn;
+	struct bna_tx *tx;
+	unsigned long flags;
+
+	/* Initialize the Tx object configuration */
+	tx_config->num_txq = bnad->num_txq_per_tx;
+	tx_config->txq_depth = bnad->txq_depth;
+	tx_config->tx_type = BNA_TX_T_REGULAR;
+
+	/* Initialize the tx event handlers */
+	tx_cbfn.tcb_setup_cbfn = bnad_cb_tcb_setup;
+	tx_cbfn.tcb_destroy_cbfn = bnad_cb_tcb_destroy;
+	tx_cbfn.tx_stall_cbfn = bnad_cb_tx_stall;
+	tx_cbfn.tx_resume_cbfn = bnad_cb_tx_resume;
+	tx_cbfn.tx_cleanup_cbfn = bnad_cb_tx_cleanup;
+
+	/* Get BNA's resource requirement for one tx object */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_tx_res_req(bnad->num_txq_per_tx,
+		bnad->txq_depth, res_info);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	/* Fill Unmap Q memory requirements */
+	BNAD_FILL_UNMAPQ_MEM_REQ(
+			&res_info[BNA_TX_RES_MEM_T_UNMAPQ],
+			bnad->num_txq_per_tx,
+			BNAD_TX_UNMAPQ_DEPTH);
+
+	/* Allocate resources */
+	err = bnad_tx_res_alloc(bnad, res_info, tx_id);
+	if (err)
+		return err;
+
+	/* Ask BNA to create one Tx object, supplying required resources */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	tx = bna_tx_create(&bnad->bna, bnad, tx_config, &tx_cbfn, res_info,
+			tx_info);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	if (!tx)
+		goto err_return;
+	tx_info->tx = tx;
+
+	/* Register ISR for the Tx object */
+	if (intr_info->intr_type == BNA_INTR_T_MSIX) {
+		err = bnad_tx_msix_register(bnad, tx_info,
+			tx_id, bnad->num_txq_per_tx);
+		if (err)
+			goto err_return;
+	}
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_tx_enable(tx);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	return 0;
+
+err_return:
+	bnad_tx_res_free(bnad, res_info);
+	return err;
+}
+
+/* Setup the rx config for bna_rx_create */
+/* bnad decides the configuration */
+static void
+bnad_init_rx_config(struct bnad *bnad, struct bna_rx_config *rx_config)
+{
+	rx_config->rx_type = BNA_RX_T_REGULAR;
+	rx_config->num_paths = bnad->num_rxp_per_rx;
+
+	if (bnad->num_rxp_per_rx > 1) {
+		rx_config->rss_status = BNA_STATUS_T_ENABLED;
+		rx_config->rss_config.hash_type =
+				(BFI_RSS_T_V4_TCP |
+				 BFI_RSS_T_V6_TCP |
+				 BFI_RSS_T_V4_IP  |
+				 BFI_RSS_T_V6_IP);
+		rx_config->rss_config.hash_mask =
+				bnad->num_rxp_per_rx - 1;
+		get_random_bytes(rx_config->rss_config.toeplitz_hash_key,
+			sizeof(rx_config->rss_config.toeplitz_hash_key));
+	} else {
+		rx_config->rss_status = BNA_STATUS_T_DISABLED;
+		memset(&rx_config->rss_config, 0,
+		       sizeof(rx_config->rss_config));
+	}
+	rx_config->rxp_type = BNA_RXP_SLR;
+	rx_config->q_depth = bnad->rxq_depth;
+
+	rx_config->small_buff_size = BFI_SMALL_RXBUF_SIZE;
+
+	rx_config->vlan_strip_status = BNA_STATUS_T_ENABLED;
+}
+
+/* Called with mutex_lock(&bnad->conf_mutex) held */
+void
+bnad_cleanup_rx(struct bnad *bnad, uint rx_id)
+{
+	struct bnad_rx_info *rx_info = &bnad->rx_info[rx_id];
+	struct bna_rx_config *rx_config = &bnad->rx_config[rx_id];
+	struct bna_res_info *res_info = &bnad->rx_res_info[rx_id].res_info[0];
+	unsigned long flags;
+	int dim_timer_del = 0;
+
+	if (!rx_info->rx)
+		return;
+
+	if (0 == rx_id) {
+		spin_lock_irqsave(&bnad->bna_lock, flags);
+		dim_timer_del = bnad_dim_timer_running(bnad);
+		if (dim_timer_del)
+			clear_bit(BNAD_RF_DIM_TIMER_RUNNING, &bnad->run_flags);
+		spin_unlock_irqrestore(&bnad->bna_lock, flags);
+		if (dim_timer_del)
+			del_timer_sync(&bnad->dim_timer);
+	}
+
+	bnad_napi_disable(bnad, rx_id);
+
+	init_completion(&bnad->bnad_completions.rx_comp);
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_rx_disable(rx_info->rx, BNA_HARD_CLEANUP, bnad_cb_rx_disabled);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	wait_for_completion(&bnad->bnad_completions.rx_comp);
+
+	if (rx_info->rx_ctrl[0].ccb->intr_type == BNA_INTR_T_MSIX)
+		bnad_rx_msix_unregister(bnad, rx_info, rx_config->num_paths);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_rx_destroy(rx_info->rx);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	rx_info->rx = NULL;
+
+	bnad_rx_res_free(bnad, res_info);
+}
+
+/* Called with mutex_lock(&bnad->conf_mutex) held */
+int
+bnad_setup_rx(struct bnad *bnad, uint rx_id)
+{
+	int err;
+	struct bnad_rx_info *rx_info = &bnad->rx_info[rx_id];
+	struct bna_res_info *res_info = &bnad->rx_res_info[rx_id].res_info[0];
+	struct bna_intr_info *intr_info =
+			&res_info[BNA_RX_RES_T_INTR].res_u.intr_info;
+	struct bna_rx_config *rx_config = &bnad->rx_config[rx_id];
+	struct bna_rx_event_cbfn rx_cbfn;
+	struct bna_rx *rx;
+	unsigned long flags;
+
+	/* Initialize the Rx object configuration */
+	bnad_init_rx_config(bnad, rx_config);
+
+	/* Initialize the Rx event handlers */
+	rx_cbfn.rcb_setup_cbfn = bnad_cb_rcb_setup;
+	rx_cbfn.rcb_destroy_cbfn = NULL;
+	rx_cbfn.ccb_setup_cbfn = bnad_cb_ccb_setup;
+	rx_cbfn.ccb_destroy_cbfn = bnad_cb_ccb_destroy;
+	rx_cbfn.rx_cleanup_cbfn = bnad_cb_rx_cleanup;
+	rx_cbfn.rx_post_cbfn = bnad_cb_rx_post;
+
+	/* Get BNA's resource requirement for one Rx object */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_rx_res_req(rx_config, res_info);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	/* Fill Unmap Q memory requirements */
+	BNAD_FILL_UNMAPQ_MEM_REQ(
+			&res_info[BNA_RX_RES_MEM_T_UNMAPQ],
+			rx_config->num_paths +
+			((rx_config->rxp_type == BNA_RXP_SINGLE) ? 0 :
+				rx_config->num_paths), BNAD_RX_UNMAPQ_DEPTH);
+
+	/* Allocate resource */
+	err = bnad_rx_res_alloc(bnad, res_info, rx_id);
+	if (err)
+		return err;
+
+	/* Ask BNA to create one Rx object, supplying required resources */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	rx = bna_rx_create(&bnad->bna, bnad, rx_config, &rx_cbfn, res_info,
+			rx_info);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	if (!rx)
+		goto err_return;
+	rx_info->rx = rx;
+
+	/* Register ISR for the Rx object */
+	if (intr_info->intr_type == BNA_INTR_T_MSIX) {
+		err = bnad_rx_msix_register(bnad, rx_info, rx_id,
+						rx_config->num_paths);
+		if (err)
+			goto err_return;
+	}
+
+	/* Enable NAPI */
+	bnad_napi_enable(bnad, rx_id);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (0 == rx_id) {
+		/* Set up Dynamic Interrupt Moderation Vector */
+		if (bnad->cfg_flags & BNAD_CF_DIM_ENABLED)
+			bna_rx_dim_reconfig(&bnad->bna, bna_napi_dim_vector);
+
+		/* Enable VLAN filtering only on the default Rx */
+		bna_rx_vlanfilter_enable(rx);
+
+		/* Start the DIM timer */
+		bnad_dim_timer_start(bnad);
+	}
+
+	bna_rx_enable(rx);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	return 0;
+
+err_return:
+	bnad_cleanup_rx(bnad, rx_id);
+	return err;
+}
+
+/* Called with conf_lock & bnad->bna_lock held */
+void
+bnad_tx_coalescing_timeo_set(struct bnad *bnad)
+{
+	struct bnad_tx_info *tx_info;
+
+	tx_info = &bnad->tx_info[0];
+	if (!tx_info->tx)
+		return;
+
+	bna_tx_coalescing_timeo_set(tx_info->tx, bnad->tx_coalescing_timeo);
+}
+
+/* Called with conf_lock & bnad->bna_lock held */
+void
+bnad_rx_coalescing_timeo_set(struct bnad *bnad)
+{
+	struct bnad_rx_info *rx_info;
+	int 	i;
+
+	for (i = 0; i < bnad->num_rx; i++) {
+		rx_info = &bnad->rx_info[i];
+		if (!rx_info->rx)
+			continue;
+		bna_rx_coalescing_timeo_set(rx_info->rx,
+				bnad->rx_coalescing_timeo);
+	}
+}
+
+/*
+ * Called with bnad->bna_lock held
+ */
+static int
+bnad_mac_addr_set_locked(struct bnad *bnad, u8 *mac_addr)
+{
+	int ret;
+
+	if (!is_valid_ether_addr(mac_addr))
+		return -EADDRNOTAVAIL;
+
+	/* If datapath is down, pretend everything went through */
+	if (!bnad->rx_info[0].rx)
+		return 0;
+
+	ret = bna_rx_ucast_set(bnad->rx_info[0].rx, mac_addr, NULL);
+	if (ret != BNA_CB_SUCCESS)
+		return -EADDRNOTAVAIL;
+
+	return 0;
+}
+
+/* Should be called with conf_lock held */
+static int
+bnad_enable_default_bcast(struct bnad *bnad)
+{
+	struct bnad_rx_info *rx_info = &bnad->rx_info[0];
+	int ret;
+	unsigned long flags;
+
+	init_completion(&bnad->bnad_completions.mcast_comp);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	ret = bna_rx_mcast_add(rx_info->rx, (u8 *)bnad_bcast_addr,
+				bnad_cb_rx_mcast_add);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	if (ret == BNA_CB_SUCCESS)
+		wait_for_completion(&bnad->bnad_completions.mcast_comp);
+	else
+		return -ENODEV;
+
+	if (bnad->bnad_completions.mcast_comp_status != BNA_CB_SUCCESS)
+		return -ENODEV;
+
+	return 0;
+}
+
+/* Statistics utilities */
+void
+bnad_netdev_qstats_fill(struct bnad *bnad)
+{
+	struct net_device_stats *net_stats = &bnad->net_stats;
+	int i, j;
+
+	for (i = 0; i < bnad->num_rx; i++) {
+		for (j = 0; j < bnad->num_rxp_per_rx; j++) {
+			if (bnad->rx_info[i].rx_ctrl[j].ccb) {
+				net_stats->rx_packets += bnad->rx_info[i].
+				rx_ctrl[j].ccb->rcb[0]->rxq->rx_packets;
+				net_stats->rx_bytes += bnad->rx_info[i].
+					rx_ctrl[j].ccb->rcb[0]->rxq->rx_bytes;
+				if (bnad->rx_info[i].rx_ctrl[j].ccb->rcb[1] &&
+					bnad->rx_info[i].rx_ctrl[j].ccb->
+					rcb[1]->rxq) {
+					net_stats->rx_packets +=
+						bnad->rx_info[i].rx_ctrl[j].
+						ccb->rcb[1]->rxq->rx_packets;
+					net_stats->rx_bytes +=
+						bnad->rx_info[i].rx_ctrl[j].
+						ccb->rcb[1]->rxq->rx_bytes;
+				}
+			}
+		}
+	}
+	for (i = 0; i < bnad->num_tx; i++) {
+		for (j = 0; j < bnad->num_txq_per_tx; j++) {
+			if (bnad->tx_info[i].tcb[j]) {
+				net_stats->tx_packets +=
+				bnad->tx_info[i].tcb[j]->txq->tx_packets;
+				net_stats->tx_bytes +=
+					bnad->tx_info[i].tcb[j]->txq->tx_bytes;
+			}
+		}
+	}
+}
+
+/*
+ * Must be called with the bna_lock held.
+ */
+void
+bnad_netdev_hwstats_fill(struct bnad *bnad)
+{
+	struct bfi_ll_stats_mac *mac_stats;
+	struct net_device_stats *net_stats = &bnad->net_stats;
+	u64 bmap;
+	int i;
+
+	mac_stats = &bnad->stats.bna_stats->hw_stats->mac_stats;
+	net_stats->rx_errors =
+		mac_stats->rx_fcs_error + mac_stats->rx_alignment_error +
+		mac_stats->rx_frame_length_error + mac_stats->rx_code_error +
+		mac_stats->rx_undersize;
+	net_stats->tx_errors = mac_stats->tx_fcs_error +
+					mac_stats->tx_undersize;
+	net_stats->rx_dropped = mac_stats->rx_drop;
+	net_stats->tx_dropped = mac_stats->tx_drop;
+	net_stats->multicast = mac_stats->rx_multicast;
+	net_stats->collisions = mac_stats->tx_total_collision;
+
+	net_stats->rx_length_errors = mac_stats->rx_frame_length_error;
+
+	/* receive ring buffer overflow  ?? */
+
+	net_stats->rx_crc_errors = mac_stats->rx_fcs_error;
+	net_stats->rx_frame_errors = mac_stats->rx_alignment_error;
+	/* recv'r fifo overrun */
+	bmap = (u64)bnad->stats.bna_stats->rxf_bmap[0] |
+		((u64)bnad->stats.bna_stats->rxf_bmap[1] << 32);
+	for (i = 0; bmap && (i < BFI_LL_RXF_ID_MAX); i++) {
+		if (bmap & 1) {
+			net_stats->rx_fifo_errors =
+				bnad->stats.bna_stats->
+					hw_stats->rxf_stats[i].frame_drops;
+			break;
+		}
+		bmap >>= 1;
+	}
+}
+
+static void
+bnad_mbox_irq_sync(struct bnad *bnad)
+{
+	u32 irq;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (bnad->cfg_flags & BNAD_CF_MSIX)
+		irq = bnad->msix_table[bnad->msix_num - 1].vector;
+	else
+		irq = bnad->pcidev->irq;
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	synchronize_irq(irq);
+}
+
+/* Utility used by bnad_start_xmit, for doing TSO */
+static int
+bnad_tso_prepare(struct bnad *bnad, struct sk_buff *skb)
+{
+	int err;
+
+	/* SKB_GSO_TCPV4 and SKB_GSO_TCPV6 is defined since 2.6.18. */
+	BUG_ON(!(skb_shinfo(skb)->gso_type == SKB_GSO_TCPV4 ||
+		   skb_shinfo(skb)->gso_type == SKB_GSO_TCPV6));
+	if (skb_header_cloned(skb)) {
+		err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC);
+		if (err) {
+			BNAD_UPDATE_CTR(bnad, tso_err);
+			return err;
+		}
+	}
+
+	/*
+	 * For TSO, the TCP checksum field is seeded with pseudo-header sum
+	 * excluding the length field.
+	 */
+	if (skb->protocol == htons(ETH_P_IP)) {
+		struct iphdr *iph = ip_hdr(skb);
+
+		/* Do we really need these? */
+		iph->tot_len = 0;
+		iph->check = 0;
+
+		tcp_hdr(skb)->check =
+			~csum_tcpudp_magic(iph->saddr, iph->daddr, 0,
+					   IPPROTO_TCP, 0);
+		BNAD_UPDATE_CTR(bnad, tso4);
+	} else {
+		struct ipv6hdr *ipv6h = ipv6_hdr(skb);
+
+		BUG_ON(!(skb->protocol == htons(ETH_P_IPV6)));
+		ipv6h->payload_len = 0;
+		tcp_hdr(skb)->check =
+			~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr, 0,
+					 IPPROTO_TCP, 0);
+		BNAD_UPDATE_CTR(bnad, tso6);
+	}
+
+	return 0;
+}
+
+/*
+ * Initialize Q numbers depending on Rx Paths
+ * Called with bnad->bna_lock held, because of cfg_flags
+ * access.
+ */
+static void
+bnad_q_num_init(struct bnad *bnad)
+{
+	int rxps;
+
+	rxps = min((uint)num_online_cpus(),
+			(uint)(BNAD_MAX_RXS * BNAD_MAX_RXPS_PER_RX));
+
+	if (!(bnad->cfg_flags & BNAD_CF_MSIX))
+		rxps = 1;	/* INTx */
+
+	bnad->num_rx = 1;
+	bnad->num_tx = 1;
+	bnad->num_rxp_per_rx = rxps;
+	bnad->num_txq_per_tx = BNAD_TXQ_NUM;
+}
+
+/*
+ * Adjusts the Q numbers, given a number of msix vectors
+ * Give preference to RSS as opposed to Tx priority Queues,
+ * in such a case, just use 1 Tx Q
+ * Called with bnad->bna_lock held b'cos of cfg_flags access
+ */
+static void
+bnad_q_num_adjust(struct bnad *bnad, int msix_vectors)
+{
+	bnad->num_txq_per_tx = 1;
+	if ((msix_vectors >= (bnad->num_tx * bnad->num_txq_per_tx)  +
+	     bnad_rxqs_per_cq + BNAD_MAILBOX_MSIX_VECTORS) &&
+	    (bnad->cfg_flags & BNAD_CF_MSIX)) {
+		bnad->num_rxp_per_rx = msix_vectors -
+			(bnad->num_tx * bnad->num_txq_per_tx) -
+			BNAD_MAILBOX_MSIX_VECTORS;
+	} else
+		bnad->num_rxp_per_rx = 1;
+}
+
+static void
+bnad_set_netdev_perm_addr(struct bnad *bnad)
+{
+	struct net_device *netdev = bnad->netdev;
+
+	memcpy(netdev->perm_addr, &bnad->perm_addr, netdev->addr_len);
+	if (is_zero_ether_addr(netdev->dev_addr))
+		memcpy(netdev->dev_addr, &bnad->perm_addr, netdev->addr_len);
+}
+
+/* Enable / disable device */
+static void
+bnad_device_disable(struct bnad *bnad)
+{
+	unsigned long flags;
+
+	init_completion(&bnad->bnad_completions.ioc_comp);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_device_disable(&bnad->bna.device, BNA_HARD_CLEANUP);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	wait_for_completion(&bnad->bnad_completions.ioc_comp);
+
+}
+
+static int
+bnad_device_enable(struct bnad *bnad)
+{
+	int err = 0;
+	unsigned long flags;
+
+	init_completion(&bnad->bnad_completions.ioc_comp);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_device_enable(&bnad->bna.device);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	wait_for_completion(&bnad->bnad_completions.ioc_comp);
+
+	if (bnad->bnad_completions.ioc_comp_status)
+		err = bnad->bnad_completions.ioc_comp_status;
+
+	return err;
+}
+
+/* Free BNA resources */
+static void
+bnad_res_free(struct bnad *bnad)
+{
+	int i;
+	struct bna_res_info *res_info = &bnad->res_info[0];
+
+	for (i = 0; i < BNA_RES_T_MAX; i++) {
+		if (res_info[i].res_type == BNA_RES_T_MEM)
+			bnad_mem_free(bnad, &res_info[i].res_u.mem_info);
+		else
+			bnad_mbox_irq_free(bnad, &res_info[i].res_u.intr_info);
+	}
+}
+
+/* Allocates memory and interrupt resources for BNA */
+static int
+bnad_res_alloc(struct bnad *bnad)
+{
+	int i, err;
+	struct bna_res_info *res_info = &bnad->res_info[0];
+
+	for (i = 0; i < BNA_RES_T_MAX; i++) {
+		if (res_info[i].res_type == BNA_RES_T_MEM)
+			err = bnad_mem_alloc(bnad, &res_info[i].res_u.mem_info);
+		else
+			err = bnad_mbox_irq_alloc(bnad,
+						  &res_info[i].res_u.intr_info);
+		if (err)
+			goto err_return;
+	}
+	return 0;
+
+err_return:
+	bnad_res_free(bnad);
+	return err;
+}
+
+/* Interrupt enable / disable */
+static void
+bnad_enable_msix(struct bnad *bnad)
+{
+	int i, ret;
+	u32 tot_msix_num;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (!(bnad->cfg_flags & BNAD_CF_MSIX)) {
+		spin_unlock_irqrestore(&bnad->bna_lock, flags);
+		return;
+	}
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	if (bnad->msix_table)
+		return;
+
+	tot_msix_num = bnad->msix_num + bnad->msix_diag_num;
+
+	bnad->msix_table =
+		kcalloc(tot_msix_num, sizeof(struct msix_entry), GFP_KERNEL);
+
+	if (!bnad->msix_table)
+		goto intx_mode;
+
+	for (i = 0; i < tot_msix_num; i++)
+		bnad->msix_table[i].entry = i;
+
+	ret = pci_enable_msix(bnad->pcidev, bnad->msix_table, tot_msix_num);
+	if (ret > 0) {
+		/* Not enough MSI-X vectors. */
+
+		spin_lock_irqsave(&bnad->bna_lock, flags);
+		/* ret = #of vectors that we got */
+		bnad_q_num_adjust(bnad, ret);
+		spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+		bnad->msix_num = (bnad->num_tx * bnad->num_txq_per_tx)
+			+ (bnad->num_rx
+			* bnad->num_rxp_per_rx) +
+			 BNAD_MAILBOX_MSIX_VECTORS;
+		tot_msix_num = bnad->msix_num + bnad->msix_diag_num;
+
+		/* Try once more with adjusted numbers */
+		/* If this fails, fall back to INTx */
+		ret = pci_enable_msix(bnad->pcidev, bnad->msix_table,
+				      tot_msix_num);
+		if (ret)
+			goto intx_mode;
+
+	} else if (ret < 0)
+		goto intx_mode;
+	return;
+
+intx_mode:
+
+	kfree(bnad->msix_table);
+	bnad->msix_table = NULL;
+	bnad->msix_num = 0;
+	bnad->msix_diag_num = 0;
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bnad->cfg_flags &= ~BNAD_CF_MSIX;
+	bnad_q_num_init(bnad);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+}
+
+static void
+bnad_disable_msix(struct bnad *bnad)
+{
+	u32 cfg_flags;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	cfg_flags = bnad->cfg_flags;
+	if (bnad->cfg_flags & BNAD_CF_MSIX)
+		bnad->cfg_flags &= ~BNAD_CF_MSIX;
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	if (cfg_flags & BNAD_CF_MSIX) {
+		pci_disable_msix(bnad->pcidev);
+		kfree(bnad->msix_table);
+		bnad->msix_table = NULL;
+	}
+}
+
+/* Netdev entry points */
+static int
+bnad_open(struct net_device *netdev)
+{
+	int err;
+	struct bnad *bnad = netdev_priv(netdev);
+	struct bna_pause_config pause_config;
+	int mtu;
+	unsigned long flags;
+
+	mutex_lock(&bnad->conf_mutex);
+
+	/* Tx */
+	err = bnad_setup_tx(bnad, 0);
+	if (err)
+		goto err_return;
+
+	/* Rx */
+	err = bnad_setup_rx(bnad, 0);
+	if (err)
+		goto cleanup_tx;
+
+	/* Port */
+	pause_config.tx_pause = 0;
+	pause_config.rx_pause = 0;
+
+	mtu = ETH_HLEN + bnad->netdev->mtu + ETH_FCS_LEN;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_port_mtu_set(&bnad->bna.port, mtu, NULL);
+	bna_port_pause_config(&bnad->bna.port, &pause_config, NULL);
+	bna_port_enable(&bnad->bna.port);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	/* Enable broadcast */
+	bnad_enable_default_bcast(bnad);
+
+	/* Set the UCAST address */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bnad_mac_addr_set_locked(bnad, netdev->dev_addr);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	/* Start the stats timer */
+	bnad_stats_timer_start(bnad);
+
+	mutex_unlock(&bnad->conf_mutex);
+
+	return 0;
+
+cleanup_tx:
+	bnad_cleanup_tx(bnad, 0);
+
+err_return:
+	mutex_unlock(&bnad->conf_mutex);
+	return err;
+}
+
+static int
+bnad_stop(struct net_device *netdev)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	unsigned long flags;
+
+	mutex_lock(&bnad->conf_mutex);
+
+	/* Stop the stats timer */
+	bnad_stats_timer_stop(bnad);
+
+	init_completion(&bnad->bnad_completions.port_comp);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_port_disable(&bnad->bna.port, BNA_HARD_CLEANUP,
+			bnad_cb_port_disabled);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	wait_for_completion(&bnad->bnad_completions.port_comp);
+
+	bnad_cleanup_tx(bnad, 0);
+	bnad_cleanup_rx(bnad, 0);
+
+	/* Synchronize mailbox IRQ */
+	bnad_mbox_irq_sync(bnad);
+
+	mutex_unlock(&bnad->conf_mutex);
+
+	return 0;
+}
+
+/* TX */
+/*
+ * bnad_start_xmit : Netdev entry point for Transmit
+ *		     Called under lock held by net_device
+ */
+static netdev_tx_t
+bnad_start_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+
+	u16 		txq_prod, vlan_tag = 0;
+	u32 		unmap_prod, wis, wis_used, wi_range;
+	u32 		vectors, vect_id, i, acked;
+	u32		tx_id;
+	int 			err;
+
+	struct bnad_tx_info *tx_info;
+	struct bna_tcb *tcb;
+	struct bnad_unmap_q *unmap_q;
+	dma_addr_t 		dma_addr;
+	struct bna_txq_entry *txqent;
+	bna_txq_wi_ctrl_flag_t 	flags;
+
+	if (unlikely
+	    (skb->len <= ETH_HLEN || skb->len > BFI_TX_MAX_DATA_PER_PKT)) {
+		dev_kfree_skb(skb);
+		return NETDEV_TX_OK;
+	}
+
+	/*
+	 * Takes care of the Tx that is scheduled between clearing the flag
+	 * and the netif_stop_queue() call.
+	 */
+	if (unlikely(!test_bit(BNAD_RF_TX_STARTED, &bnad->run_flags))) {
+		dev_kfree_skb(skb);
+		return NETDEV_TX_OK;
+	}
+
+	tx_id = 0;
+
+	tx_info = &bnad->tx_info[tx_id];
+	tcb = tx_info->tcb[tx_id];
+	unmap_q = tcb->unmap_q;
+
+	vectors = 1 + skb_shinfo(skb)->nr_frags;
+	if (vectors > BFI_TX_MAX_VECTORS_PER_PKT) {
+		dev_kfree_skb(skb);
+		return NETDEV_TX_OK;
+	}
+	wis = BNA_TXQ_WI_NEEDED(vectors);	/* 4 vectors per work item */
+	acked = 0;
+	if (unlikely
+	    (wis > BNA_QE_FREE_CNT(tcb, tcb->q_depth) ||
+	     vectors > BNA_QE_FREE_CNT(unmap_q, unmap_q->q_depth))) {
+		if ((u16) (*tcb->hw_consumer_index) !=
+		    tcb->consumer_index &&
+		    !test_and_set_bit(BNAD_TXQ_FREE_SENT, &tcb->flags)) {
+			acked = bnad_free_txbufs(bnad, tcb);
+			bna_ib_ack(tcb->i_dbell, acked);
+			smp_mb__before_clear_bit();
+			clear_bit(BNAD_TXQ_FREE_SENT, &tcb->flags);
+		} else {
+			netif_stop_queue(netdev);
+			BNAD_UPDATE_CTR(bnad, netif_queue_stop);
+		}
+
+		smp_mb();
+		/*
+		 * Check again to deal with race condition between
+		 * netif_stop_queue here, and netif_wake_queue in
+		 * interrupt handler which is not inside netif tx lock.
+		 */
+		if (likely
+		    (wis > BNA_QE_FREE_CNT(tcb, tcb->q_depth) ||
+		     vectors > BNA_QE_FREE_CNT(unmap_q, unmap_q->q_depth))) {
+			BNAD_UPDATE_CTR(bnad, netif_queue_stop);
+			return NETDEV_TX_BUSY;
+		} else {
+			netif_wake_queue(netdev);
+			BNAD_UPDATE_CTR(bnad, netif_queue_wakeup);
+		}
+	}
+
+	unmap_prod = unmap_q->producer_index;
+	wis_used = 1;
+	vect_id = 0;
+	flags = 0;
+
+	txq_prod = tcb->producer_index;
+	BNA_TXQ_QPGE_PTR_GET(txq_prod, tcb->sw_qpt, txqent, wi_range);
+	BUG_ON(!(wi_range <= tcb->q_depth));
+	txqent->hdr.wi.reserved = 0;
+	txqent->hdr.wi.num_vectors = vectors;
+	txqent->hdr.wi.opcode =
+		htons((skb_is_gso(skb) ? BNA_TXQ_WI_SEND_LSO :
+		       BNA_TXQ_WI_SEND));
+
+	if (bnad->vlan_grp && vlan_tx_tag_present(skb)) {
+		vlan_tag = (u16) vlan_tx_tag_get(skb);
+		flags |= (BNA_TXQ_WI_CF_INS_PRIO | BNA_TXQ_WI_CF_INS_VLAN);
+	}
+	if (test_bit(BNAD_RF_CEE_RUNNING, &bnad->run_flags)) {
+		vlan_tag =
+			(tcb->priority & 0x7) << 13 | (vlan_tag & 0x1fff);
+		flags |= (BNA_TXQ_WI_CF_INS_PRIO | BNA_TXQ_WI_CF_INS_VLAN);
+	}
+
+	txqent->hdr.wi.vlan_tag = htons(vlan_tag);
+
+	if (skb_is_gso(skb)) {
+		err = bnad_tso_prepare(bnad, skb);
+		if (err) {
+			dev_kfree_skb(skb);
+			return NETDEV_TX_OK;
+		}
+		txqent->hdr.wi.lso_mss = htons(skb_is_gso(skb));
+		flags |= (BNA_TXQ_WI_CF_IP_CKSUM | BNA_TXQ_WI_CF_TCP_CKSUM);
+		txqent->hdr.wi.l4_hdr_size_n_offset =
+			htons(BNA_TXQ_WI_L4_HDR_N_OFFSET
+			      (tcp_hdrlen(skb) >> 2,
+			       skb_transport_offset(skb)));
+	} else if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		u8 proto = 0;
+
+		txqent->hdr.wi.lso_mss = 0;
+
+		if (skb->protocol == htons(ETH_P_IP))
+			proto = ip_hdr(skb)->protocol;
+		else if (skb->protocol == htons(ETH_P_IPV6)) {
+			/* nexthdr may not be TCP immediately. */
+			proto = ipv6_hdr(skb)->nexthdr;
+		}
+		if (proto == IPPROTO_TCP) {
+			flags |= BNA_TXQ_WI_CF_TCP_CKSUM;
+			txqent->hdr.wi.l4_hdr_size_n_offset =
+				htons(BNA_TXQ_WI_L4_HDR_N_OFFSET
+				      (0, skb_transport_offset(skb)));
+
+			BNAD_UPDATE_CTR(bnad, tcpcsum_offload);
+
+			BUG_ON(!(skb_headlen(skb) >=
+				skb_transport_offset(skb) + tcp_hdrlen(skb)));
+
+		} else if (proto == IPPROTO_UDP) {
+			flags |= BNA_TXQ_WI_CF_UDP_CKSUM;
+			txqent->hdr.wi.l4_hdr_size_n_offset =
+				htons(BNA_TXQ_WI_L4_HDR_N_OFFSET
+				      (0, skb_transport_offset(skb)));
+
+			BNAD_UPDATE_CTR(bnad, udpcsum_offload);
+
+			BUG_ON(!(skb_headlen(skb) >=
+				   skb_transport_offset(skb) +
+				   sizeof(struct udphdr)));
+		} else {
+			err = skb_checksum_help(skb);
+			BNAD_UPDATE_CTR(bnad, csum_help);
+			if (err) {
+				dev_kfree_skb(skb);
+				BNAD_UPDATE_CTR(bnad, csum_help_err);
+				return NETDEV_TX_OK;
+			}
+		}
+	} else {
+		txqent->hdr.wi.lso_mss = 0;
+		txqent->hdr.wi.l4_hdr_size_n_offset = 0;
+	}
+
+	txqent->hdr.wi.flags = htons(flags);
+
+	txqent->hdr.wi.frame_length = htonl(skb->len);
+
+	unmap_q->unmap_array[unmap_prod].skb = skb;
+	BUG_ON(!(skb_headlen(skb) <= BFI_TX_MAX_DATA_PER_VECTOR));
+	txqent->vector[vect_id].length = htons(skb_headlen(skb));
+	dma_addr = pci_map_single(bnad->pcidev, skb->data, skb_headlen(skb),
+		PCI_DMA_TODEVICE);
+	pci_unmap_addr_set(&unmap_q->unmap_array[unmap_prod], dma_addr,
+			   dma_addr);
+
+	BNA_SET_DMA_ADDR(dma_addr, &txqent->vector[vect_id].host_addr);
+	BNA_QE_INDX_ADD(unmap_prod, 1, unmap_q->q_depth);
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i];
+		u32		size = frag->size;
+
+		if (++vect_id == BFI_TX_MAX_VECTORS_PER_WI) {
+			vect_id = 0;
+			if (--wi_range)
+				txqent++;
+			else {
+				BNA_QE_INDX_ADD(txq_prod, wis_used,
+						tcb->q_depth);
+				wis_used = 0;
+				BNA_TXQ_QPGE_PTR_GET(txq_prod, tcb->sw_qpt,
+						     txqent, wi_range);
+				BUG_ON(!(wi_range <= tcb->q_depth));
+			}
+			wis_used++;
+			txqent->hdr.wi_ext.opcode = htons(BNA_TXQ_WI_EXTENSION);
+		}
+
+		BUG_ON(!(size <= BFI_TX_MAX_DATA_PER_VECTOR));
+		txqent->vector[vect_id].length = htons(size);
+		dma_addr =
+			pci_map_page(bnad->pcidev, frag->page,
+				     frag->page_offset, size,
+				     PCI_DMA_TODEVICE);
+		pci_unmap_addr_set(&unmap_q->unmap_array[unmap_prod], dma_addr,
+				   dma_addr);
+		BNA_SET_DMA_ADDR(dma_addr, &txqent->vector[vect_id].host_addr);
+		BNA_QE_INDX_ADD(unmap_prod, 1, unmap_q->q_depth);
+	}
+
+	unmap_q->producer_index = unmap_prod;
+	BNA_QE_INDX_ADD(txq_prod, wis_used, tcb->q_depth);
+	tcb->producer_index = txq_prod;
+
+	smp_mb();
+	bna_txq_prod_indx_doorbell(tcb);
+
+	if ((u16) (*tcb->hw_consumer_index) != tcb->consumer_index)
+		tasklet_schedule(&bnad->tx_free_tasklet);
+
+	return NETDEV_TX_OK;
+}
+
+/*
+ * Used spin_lock to synchronize reading of stats structures, which
+ * is written by BNA under the same lock.
+ */
+static struct net_device_stats *
+bnad_get_netdev_stats(struct net_device *netdev)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+
+	memset(&bnad->net_stats, 0, sizeof(struct net_device_stats));
+
+	bnad_netdev_qstats_fill(bnad);
+	bnad_netdev_hwstats_fill(bnad);
+
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	return &bnad->net_stats;
+}
+
+static void
+bnad_set_rx_mode(struct net_device *netdev)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	u32	new_mask, valid_mask;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+
+	new_mask = valid_mask = 0;
+
+	if (netdev->flags & IFF_PROMISC) {
+		if (!(bnad->cfg_flags & BNAD_CF_PROMISC)) {
+			new_mask = BNAD_RXMODE_PROMISC_DEFAULT;
+			valid_mask = BNAD_RXMODE_PROMISC_DEFAULT;
+			bnad->cfg_flags |= BNAD_CF_PROMISC;
+		}
+	} else {
+		if (bnad->cfg_flags & BNAD_CF_PROMISC) {
+			new_mask = ~BNAD_RXMODE_PROMISC_DEFAULT;
+			valid_mask = BNAD_RXMODE_PROMISC_DEFAULT;
+			bnad->cfg_flags &= ~BNAD_CF_PROMISC;
+		}
+	}
+
+	if (netdev->flags & IFF_ALLMULTI) {
+		if (!(bnad->cfg_flags & BNAD_CF_ALLMULTI)) {
+			new_mask |= BNA_RXMODE_ALLMULTI;
+			valid_mask |= BNA_RXMODE_ALLMULTI;
+			bnad->cfg_flags |= BNAD_CF_ALLMULTI;
+		}
+	} else {
+		if (bnad->cfg_flags & BNAD_CF_ALLMULTI) {
+			new_mask &= ~BNA_RXMODE_ALLMULTI;
+			valid_mask |= BNA_RXMODE_ALLMULTI;
+			bnad->cfg_flags &= ~BNAD_CF_ALLMULTI;
+		}
+	}
+
+	bna_rx_mode_set(bnad->rx_info[0].rx, new_mask, valid_mask, NULL);
+
+	if (!netdev_mc_empty(netdev)) {
+		u8 *mcaddr_list;
+		int mc_count = netdev_mc_count(netdev);
+
+		/* Index 0 holds the broadcast address */
+		mcaddr_list =
+			kzalloc((mc_count + 1) * ETH_ALEN,
+				GFP_ATOMIC);
+		if (!mcaddr_list)
+			return;
+
+		memcpy(&mcaddr_list[0], &bnad_bcast_addr[0], ETH_ALEN);
+
+		/* Copy rest of the MC addresses */
+		bnad_netdev_mc_list_get(netdev, mcaddr_list);
+
+		bna_rx_mcast_listset(bnad->rx_info[0].rx, mc_count + 1,
+					mcaddr_list, NULL);
+
+		/* Should we enable BNAD_CF_ALLMULTI for err != 0 ? */
+		kfree(mcaddr_list);
+	}
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+}
+
+/*
+ * bna_lock is used to sync writes to netdev->addr
+ * conf_lock cannot be used since this call may be made
+ * in a non-blocking context.
+ */
+static int
+bnad_set_mac_address(struct net_device *netdev, void *mac_addr)
+{
+	int err;
+	struct bnad *bnad = netdev_priv(netdev);
+	struct sockaddr *sa = (struct sockaddr *)mac_addr;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+
+	err = bnad_mac_addr_set_locked(bnad, sa->sa_data);
+
+	if (!err)
+		memcpy(netdev->dev_addr, sa->sa_data, netdev->addr_len);
+
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	return err;
+}
+
+static int
+bnad_change_mtu(struct net_device *netdev, int new_mtu)
+{
+	int mtu, err = 0;
+	unsigned long flags;
+
+	struct bnad *bnad = netdev_priv(netdev);
+
+	if (new_mtu + ETH_HLEN < ETH_ZLEN || new_mtu > BNAD_JUMBO_MTU)
+		return -EINVAL;
+
+	mutex_lock(&bnad->conf_mutex);
+
+	netdev->mtu = new_mtu;
+
+	mtu = ETH_HLEN + new_mtu + ETH_FCS_LEN;
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_port_mtu_set(&bnad->bna.port, mtu, NULL);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	mutex_unlock(&bnad->conf_mutex);
+	return err;
+}
+
+static void
+bnad_vlan_rx_register(struct net_device *netdev,
+				  struct vlan_group *vlan_grp)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+
+	mutex_lock(&bnad->conf_mutex);
+	bnad->vlan_grp = vlan_grp;
+	mutex_unlock(&bnad->conf_mutex);
+}
+
+static void
+bnad_vlan_rx_add_vid(struct net_device *netdev,
+				 unsigned short vid)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	unsigned long flags;
+
+	if (!bnad->rx_info[0].rx)
+		return;
+
+	mutex_lock(&bnad->conf_mutex);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_rx_vlan_add(bnad->rx_info[0].rx, vid);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	mutex_unlock(&bnad->conf_mutex);
+}
+
+static void
+bnad_vlan_rx_kill_vid(struct net_device *netdev,
+				  unsigned short vid)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	unsigned long flags;
+
+	if (!bnad->rx_info[0].rx)
+		return;
+
+	mutex_lock(&bnad->conf_mutex);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_rx_vlan_del(bnad->rx_info[0].rx, vid);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	mutex_unlock(&bnad->conf_mutex);
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void
+bnad_netpoll(struct net_device *netdev)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	struct bnad_rx_info *rx_info;
+	struct bnad_rx_ctrl *rx_ctrl;
+	u32 curr_mask;
+	int i, j;
+
+	if (!(bnad->cfg_flags & BNAD_CF_MSIX)) {
+		bna_intx_disable(&bnad->bna, curr_mask);
+		bnad_isr(bnad->pcidev->irq, netdev);
+		bna_intx_enable(&bnad->bna, curr_mask);
+	} else {
+		for (i = 0; i < bnad->num_rx; i++) {
+			rx_info = &bnad->rx_info[i];
+			if (!rx_info->rx)
+				continue;
+			for (j = 0; j < bnad->num_rxp_per_rx; j++) {
+				rx_ctrl = &rx_info->rx_ctrl[j];
+				if (rx_ctrl->ccb) {
+					bnad_disable_rx_irq(bnad,
+							    rx_ctrl->ccb);
+					bnad_netif_rx_schedule_poll(bnad,
+							    rx_ctrl->ccb);
+				}
+			}
+		}
+	}
+}
+#endif
+
+static const struct net_device_ops bnad_netdev_ops = {
+	.ndo_open		= bnad_open,
+	.ndo_stop		= bnad_stop,
+	.ndo_start_xmit		= bnad_start_xmit,
+	.ndo_get_stats		= bnad_get_netdev_stats,
+	.ndo_set_rx_mode	= bnad_set_rx_mode,
+	.ndo_set_multicast_list = bnad_set_rx_mode,
+	.ndo_validate_addr      = eth_validate_addr,
+	.ndo_set_mac_address    = bnad_set_mac_address,
+	.ndo_change_mtu		= bnad_change_mtu,
+	.ndo_vlan_rx_register   = bnad_vlan_rx_register,
+	.ndo_vlan_rx_add_vid    = bnad_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid   = bnad_vlan_rx_kill_vid,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller    = bnad_netpoll
+#endif
+};
+
+static void
+bnad_netdev_init(struct bnad *bnad, bool using_dac)
+{
+	struct net_device *netdev = bnad->netdev;
+
+	netdev->features |= NETIF_F_IPV6_CSUM;
+	netdev->features |= NETIF_F_TSO;
+	netdev->features |= NETIF_F_TSO6;
+
+	netdev->features |= NETIF_F_GRO;
+	pr_warn("bna: GRO enabled, using kernel stack GRO\n");
+
+	netdev->features |= NETIF_F_SG | NETIF_F_IP_CSUM;
+
+	if (using_dac)
+		netdev->features |= NETIF_F_HIGHDMA;
+
+	netdev->features |=
+		NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX |
+		NETIF_F_HW_VLAN_FILTER;
+
+	netdev->vlan_features = netdev->features;
+	netdev->mem_start = bnad->mmio_start;
+	netdev->mem_end = bnad->mmio_start + bnad->mmio_len - 1;
+
+	netdev->netdev_ops = &bnad_netdev_ops;
+	bnad_set_ethtool_ops(netdev);
+}
+
+/*
+ * 1. Initialize the bnad structure
+ * 2. Setup netdev pointer in pci_dev
+ * 3. Initialze Tx free tasklet
+ * 4. Initialize no. of TxQ & CQs & MSIX vectors
+ */
+static int
+bnad_init(struct bnad *bnad,
+	  struct pci_dev *pdev, struct net_device *netdev)
+{
+	unsigned long flags;
+
+	SET_NETDEV_DEV(netdev, &pdev->dev);
+	pci_set_drvdata(pdev, netdev);
+
+	bnad->netdev = netdev;
+	bnad->pcidev = pdev;
+	bnad->mmio_start = pci_resource_start(pdev, 0);
+	bnad->mmio_len = pci_resource_len(pdev, 0);
+	bnad->bar0 = ioremap_nocache(bnad->mmio_start, bnad->mmio_len);
+	if (!bnad->bar0) {
+		dev_err(&pdev->dev, "ioremap for bar0 failed\n");
+		pci_set_drvdata(pdev, NULL);
+		return -ENOMEM;
+	}
+	pr_info("bar0 mapped to %p, len %llu\n", bnad->bar0,
+	       (unsigned long long) bnad->mmio_len);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (!bnad_msix_disable)
+		bnad->cfg_flags = BNAD_CF_MSIX;
+
+	bnad->cfg_flags |= BNAD_CF_DIM_ENABLED;
+
+	bnad_q_num_init(bnad);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	bnad->msix_num = (bnad->num_tx * bnad->num_txq_per_tx) +
+		(bnad->num_rx * bnad->num_rxp_per_rx) +
+			 BNAD_MAILBOX_MSIX_VECTORS;
+	bnad->msix_diag_num = 2;	/* 1 for Tx, 1 for Rx */
+
+	bnad->txq_depth = BNAD_TXQ_DEPTH;
+	bnad->rxq_depth = BNAD_RXQ_DEPTH;
+	bnad->rx_csum = true;
+
+	bnad->tx_coalescing_timeo = BFI_TX_COALESCING_TIMEO;
+	bnad->rx_coalescing_timeo = BFI_RX_COALESCING_TIMEO;
+
+	tasklet_init(&bnad->tx_free_tasklet, bnad_tx_free_tasklet,
+		     (unsigned long)bnad);
+
+	return 0;
+}
+
+/*
+ * Must be called after bnad_pci_uninit()
+ * so that iounmap() and pci_set_drvdata(NULL)
+ * happens only after PCI uninitialization.
+ */
+static void
+bnad_uninit(struct bnad *bnad)
+{
+	if (bnad->bar0)
+		iounmap(bnad->bar0);
+	pci_set_drvdata(bnad->pcidev, NULL);
+}
+
+/*
+ * Initialize locks
+	a) Per device mutes used for serializing configuration
+	   changes from OS interface
+	b) spin lock used to protect bna state machine
+ */
+static void
+bnad_lock_init(struct bnad *bnad)
+{
+	spin_lock_init(&bnad->bna_lock);
+	mutex_init(&bnad->conf_mutex);
+}
+
+static void
+bnad_lock_uninit(struct bnad *bnad)
+{
+	mutex_destroy(&bnad->conf_mutex);
+}
+
+/* PCI Initialization */
+static int
+bnad_pci_init(struct bnad *bnad,
+	      struct pci_dev *pdev, bool *using_dac)
+{
+	int err;
+
+	err = pci_enable_device(pdev);
+	if (err)
+		return err;
+	err = pci_request_regions(pdev, BNAD_NAME);
+	if (err)
+		goto disable_device;
+	if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64)) &&
+	    !pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64))) {
+		*using_dac = 1;
+	} else {
+		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+		if (err) {
+			err = pci_set_consistent_dma_mask(pdev,
+						DMA_BIT_MASK(32));
+			if (err)
+				goto release_regions;
+		}
+		*using_dac = 0;
+	}
+	pci_set_master(pdev);
+	return 0;
+
+release_regions:
+	pci_release_regions(pdev);
+disable_device:
+	pci_disable_device(pdev);
+
+	return err;
+}
+
+static void
+bnad_pci_uninit(struct pci_dev *pdev)
+{
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+}
+
+static int __devinit
+bnad_pci_probe(struct pci_dev *pdev,
+		const struct pci_device_id *pcidev_id)
+{
+	bool 	using_dac;
+	int 	err;
+	struct bnad *bnad;
+	struct bna *bna;
+	struct net_device *netdev;
+	struct bfa_pcidev pcidev_info;
+	unsigned long flags;
+
+	pr_info("bnad_pci_probe : (0x%p, 0x%p) PCI Func : (%d)\n",
+	       pdev, pcidev_id, PCI_FUNC(pdev->devfn));
+
+	mutex_lock(&bnad_fwimg_mutex);
+	if (!cna_get_firmware_buf(pdev)) {
+		mutex_unlock(&bnad_fwimg_mutex);
+		pr_warn("Failed to load Firmware Image!\n");
+		return -ENODEV;
+	}
+	mutex_unlock(&bnad_fwimg_mutex);
+
+	/*
+	 * Allocates sizeof(struct net_device + struct bnad)
+	 * bnad = netdev->priv
+	 */
+	netdev = alloc_etherdev(sizeof(struct bnad));
+	if (!netdev) {
+		dev_err(&pdev->dev, "alloc_etherdev failed\n");
+		err = -ENOMEM;
+		return err;
+	}
+	bnad = netdev_priv(netdev);
+
+
+	/*
+	 * PCI initialization
+	 * 	Output : using_dac = 1 for 64 bit DMA
+	 *		           = 0 for 32 bit DMA
+	 */
+	err = bnad_pci_init(bnad, pdev, &using_dac);
+	if (err)
+		goto free_netdev;
+
+	bnad_lock_init(bnad);
+	/*
+	 * Initialize bnad structure
+	 * Setup relation between pci_dev & netdev
+	 * Init Tx free tasklet
+	 */
+	err = bnad_init(bnad, pdev, netdev);
+	if (err)
+		goto pci_uninit;
+	/* Initialize netdev structure, set up ethtool ops */
+	bnad_netdev_init(bnad, using_dac);
+
+	bnad_enable_msix(bnad);
+
+	/* Get resource requirement form bna */
+	bna_res_req(&bnad->res_info[0]);
+
+	/* Allocate resources from bna */
+	err = bnad_res_alloc(bnad);
+	if (err)
+		goto free_netdev;
+
+	bna = &bnad->bna;
+
+	/* Setup pcidev_info for bna_init() */
+	pcidev_info.pci_slot = PCI_SLOT(bnad->pcidev->devfn);
+	pcidev_info.pci_func = PCI_FUNC(bnad->pcidev->devfn);
+	pcidev_info.device_id = bnad->pcidev->device;
+	pcidev_info.pci_bar_kva = bnad->bar0;
+
+	mutex_lock(&bnad->conf_mutex);
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_init(bna, bnad, &pcidev_info, &bnad->res_info[0]);
+
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	bnad->stats.bna_stats = &bna->stats;
+
+	/* Set up timers */
+	setup_timer(&bnad->bna.device.ioc.ioc_timer, bnad_ioc_timeout,
+				((unsigned long)bnad));
+	setup_timer(&bnad->bna.device.ioc.hb_timer, bnad_ioc_hb_check,
+				((unsigned long)bnad));
+	setup_timer(&bnad->bna.device.ioc.sem_timer, bnad_ioc_sem_timeout,
+				((unsigned long)bnad));
+
+	/* Now start the timer before calling IOC */
+	mod_timer(&bnad->bna.device.ioc.ioc_timer,
+		  jiffies + msecs_to_jiffies(BNA_IOC_TIMER_FREQ));
+
+	/*
+	 * Start the chip
+	 * Don't care even if err != 0, bna state machine will
+	 * deal with it
+	 */
+	err = bnad_device_enable(bnad);
+
+	/* Get the burnt-in mac */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_port_mac_get(&bna->port, &bnad->perm_addr);
+	bnad_set_netdev_perm_addr(bnad);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	mutex_unlock(&bnad->conf_mutex);
+
+	/*
+	 * Make sure the link appears down to the stack
+	 */
+	netif_carrier_off(netdev);
+
+	/* Finally, reguister with net_device layer */
+	err = register_netdev(netdev);
+	if (err) {
+		pr_err("BNA : Registering with netdev failed\n");
+		goto disable_device;
+	}
+
+	return 0;
+
+disable_device:
+	mutex_lock(&bnad->conf_mutex);
+	bnad_device_disable(bnad);
+	del_timer_sync(&bnad->bna.device.ioc.ioc_timer);
+	del_timer_sync(&bnad->bna.device.ioc.sem_timer);
+	del_timer_sync(&bnad->bna.device.ioc.hb_timer);
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_uninit(bna);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	mutex_unlock(&bnad->conf_mutex);
+
+	bnad_res_free(bnad);
+	bnad_disable_msix(bnad);
+pci_uninit:
+	bnad_pci_uninit(pdev);
+	bnad_lock_uninit(bnad);
+	bnad_uninit(bnad);
+free_netdev:
+	free_netdev(netdev);
+	return err;
+}
+
+static void __devexit
+bnad_pci_remove(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct bnad *bnad;
+	struct bna *bna;
+	unsigned long flags;
+
+	if (!netdev)
+		return;
+
+	pr_info("%s bnad_pci_remove\n", netdev->name);
+	bnad = netdev_priv(netdev);
+	bna = &bnad->bna;
+
+	unregister_netdev(netdev);
+
+	mutex_lock(&bnad->conf_mutex);
+	bnad_device_disable(bnad);
+	del_timer_sync(&bnad->bna.device.ioc.ioc_timer);
+	del_timer_sync(&bnad->bna.device.ioc.sem_timer);
+	del_timer_sync(&bnad->bna.device.ioc.hb_timer);
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bna_uninit(bna);
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	mutex_unlock(&bnad->conf_mutex);
+
+	bnad_res_free(bnad);
+	bnad_disable_msix(bnad);
+	bnad_pci_uninit(pdev);
+	bnad_lock_uninit(bnad);
+	bnad_uninit(bnad);
+	free_netdev(netdev);
+}
+
+const struct pci_device_id bnad_pci_id_table[] = {
+	{
+		PCI_DEVICE(PCI_VENDOR_ID_BROCADE,
+			PCI_DEVICE_ID_BROCADE_CT),
+		.class = PCI_CLASS_NETWORK_ETHERNET << 8,
+		.class_mask =  0xffff00
+	}, {0,  }
+};
+
+MODULE_DEVICE_TABLE(pci, bnad_pci_id_table);
+
+static struct pci_driver bnad_pci_driver = {
+	.name = BNAD_NAME,
+	.id_table = bnad_pci_id_table,
+	.probe = bnad_pci_probe,
+	.remove = __devexit_p(bnad_pci_remove),
+};
+
+static int __init
+bnad_module_init(void)
+{
+	int err;
+
+	pr_info("Brocade 10G Ethernet driver\n");
+
+	bfa_ioc_auto_recover(bnad_ioc_auto_recover);
+
+	err = pci_register_driver(&bnad_pci_driver);
+	if (err < 0) {
+		pr_err("bna : PCI registration failed in module init "
+		       "(%d)\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+static void __exit
+bnad_module_exit(void)
+{
+	pci_unregister_driver(&bnad_pci_driver);
+
+	if (bfi_fw)
+		release_firmware(bfi_fw);
+}
+
+module_init(bnad_module_init);
+module_exit(bnad_module_exit);
+
+MODULE_AUTHOR("Brocade");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Brocade 10G PCIe Ethernet driver");
+MODULE_VERSION(BNAD_VERSION);
+MODULE_FIRMWARE(CNA_FW_FILE_CT);
diff --git a/drivers/net/bna/bnad.h b/drivers/net/bna/bnad.h
new file mode 100644
index 000000000000..3261401e35cb
--- /dev/null
+++ b/drivers/net/bna/bnad.h
@@ -0,0 +1,334 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#ifndef __BNAD_H__
+#define __BNAD_H__
+
+#include <linux/rtnetlink.h>
+#include <linux/workqueue.h>
+#include <linux/ipv6.h>
+#include <linux/etherdevice.h>
+#include <linux/mutex.h>
+#include <linux/firmware.h>
+
+/* Fix for IA64 */
+#include <asm/checksum.h>
+#include <net/ip6_checksum.h>
+
+#include <net/ip.h>
+#include <net/tcp.h>
+
+#include "bna.h"
+
+#define BNAD_TXQ_DEPTH		2048
+#define BNAD_RXQ_DEPTH		2048
+
+#define BNAD_MAX_TXS		1
+#define BNAD_MAX_TXQ_PER_TX	8	/* 8 priority queues */
+#define BNAD_TXQ_NUM		1
+
+#define BNAD_MAX_RXS		1
+#define BNAD_MAX_RXPS_PER_RX	16
+
+/*
+ * Control structure pointed to ccb->ctrl, which
+ * determines the NAPI / LRO behavior CCB
+ * There is 1:1 corres. between ccb & ctrl
+ */
+struct bnad_rx_ctrl {
+	struct bna_ccb *ccb;
+	struct napi_struct	napi;
+};
+
+#define BNAD_RXMODE_PROMISC_DEFAULT	BNA_RXMODE_PROMISC
+
+#define BNAD_GET_TX_ID(_skb)	(0)
+
+/*
+ * GLOBAL #defines (CONSTANTS)
+ */
+#define BNAD_NAME			"bna"
+#define BNAD_NAME_LEN			64
+
+#define BNAD_VERSION			"2.3.2.0"
+
+#define BNAD_MAILBOX_MSIX_VECTORS	1
+
+#define BNAD_STATS_TIMER_FREQ		1000 	/* in msecs */
+#define BNAD_DIM_TIMER_FREQ		1000 	/* in msecs */
+
+#define BNAD_MAX_Q_DEPTH		0x10000
+#define BNAD_MIN_Q_DEPTH		0x200
+
+#define BNAD_JUMBO_MTU			9000
+
+#define BNAD_NETIF_WAKE_THRESHOLD	8
+
+#define BNAD_RXQ_REFILL_THRESHOLD_SHIFT	3
+
+/* Bit positions for tcb->flags */
+#define BNAD_TXQ_FREE_SENT		0
+
+/* Bit positions for rcb->flags */
+#define BNAD_RXQ_REFILL			0
+#define BNAD_RXQ_STARTED		1
+
+/*
+ * DATA STRUCTURES
+ */
+
+/* enums */
+enum bnad_intr_source {
+	BNAD_INTR_TX		= 1,
+	BNAD_INTR_RX		= 2
+};
+
+enum bnad_link_state {
+	BNAD_LS_DOWN		= 0,
+	BNAD_LS_UP 		= 1
+};
+
+struct bnad_completion {
+	struct completion 	ioc_comp;
+	struct completion 	ucast_comp;
+	struct completion	mcast_comp;
+	struct completion	tx_comp;
+	struct completion	rx_comp;
+	struct completion	stats_comp;
+	struct completion	port_comp;
+
+	u8			ioc_comp_status;
+	u8			ucast_comp_status;
+	u8			mcast_comp_status;
+	u8			tx_comp_status;
+	u8			rx_comp_status;
+	u8			stats_comp_status;
+	u8			port_comp_status;
+};
+
+/* Tx Rx Control Stats */
+struct bnad_drv_stats {
+	u64 		netif_queue_stop;
+	u64		netif_queue_wakeup;
+	u64		tso4;
+	u64		tso6;
+	u64		tso_err;
+	u64		tcpcsum_offload;
+	u64		udpcsum_offload;
+	u64		csum_help;
+	u64		csum_help_err;
+
+	u64		hw_stats_updates;
+	u64		netif_rx_schedule;
+	u64		netif_rx_complete;
+	u64		netif_rx_dropped;
+
+	u64		link_toggle;
+	u64		cee_up;
+
+	u64		rxp_info_alloc_failed;
+	u64		mbox_intr_disabled;
+	u64		mbox_intr_enabled;
+	u64		tx_unmap_q_alloc_failed;
+	u64		rx_unmap_q_alloc_failed;
+
+	u64		rxbuf_alloc_failed;
+};
+
+/* Complete driver stats */
+struct bnad_stats {
+	struct bnad_drv_stats drv_stats;
+	struct bna_stats *bna_stats;
+};
+
+/* Tx / Rx Resources */
+struct bnad_tx_res_info {
+	struct bna_res_info res_info[BNA_TX_RES_T_MAX];
+};
+
+struct bnad_rx_res_info {
+	struct bna_res_info res_info[BNA_RX_RES_T_MAX];
+};
+
+struct bnad_tx_info {
+	struct bna_tx *tx; /* 1:1 between tx_info & tx */
+	struct bna_tcb *tcb[BNAD_MAX_TXQ_PER_TX];
+} ____cacheline_aligned;
+
+struct bnad_rx_info {
+	struct bna_rx *rx; /* 1:1 between rx_info & rx */
+
+	struct bnad_rx_ctrl rx_ctrl[BNAD_MAX_RXPS_PER_RX];
+} ____cacheline_aligned;
+
+/* Unmap queues for Tx / Rx cleanup */
+struct bnad_skb_unmap {
+	struct sk_buff		*skb;
+	DECLARE_PCI_UNMAP_ADDR(dma_addr)
+};
+
+struct bnad_unmap_q {
+	u32		producer_index;
+	u32		consumer_index;
+	u32 		q_depth;
+	/* This should be the last one */
+	struct bnad_skb_unmap unmap_array[1];
+};
+
+/* Bit mask values for bnad->cfg_flags */
+#define	BNAD_CF_DIM_ENABLED		0x01	/* DIM */
+#define	BNAD_CF_PROMISC			0x02
+#define BNAD_CF_ALLMULTI		0x04
+#define	BNAD_CF_MSIX			0x08	/* If in MSIx mode */
+
+/* Defines for run_flags bit-mask */
+/* Set, tested & cleared using xxx_bit() functions */
+/* Values indicated bit positions */
+#define	BNAD_RF_CEE_RUNNING		1
+#define BNAD_RF_HW_ERROR 		2
+#define BNAD_RF_MBOX_IRQ_DISABLED	3
+#define BNAD_RF_TX_STARTED		4
+#define BNAD_RF_RX_STARTED		5
+#define BNAD_RF_DIM_TIMER_RUNNING	6
+#define BNAD_RF_STATS_TIMER_RUNNING	7
+
+struct bnad {
+	struct net_device 	*netdev;
+
+	/* Data path */
+	struct bnad_tx_info tx_info[BNAD_MAX_TXS];
+	struct bnad_rx_info rx_info[BNAD_MAX_RXS];
+
+	struct vlan_group	*vlan_grp;
+	/*
+	 * These q numbers are global only because
+	 * they are used to calculate MSIx vectors.
+	 * Actually the exact # of queues are per Tx/Rx
+	 * object.
+	 */
+	u32		num_tx;
+	u32		num_rx;
+	u32		num_txq_per_tx;
+	u32		num_rxp_per_rx;
+
+	u32		txq_depth;
+	u32		rxq_depth;
+
+	u8			tx_coalescing_timeo;
+	u8			rx_coalescing_timeo;
+
+	struct bna_rx_config rx_config[BNAD_MAX_RXS];
+	struct bna_tx_config tx_config[BNAD_MAX_TXS];
+
+	u32		rx_csum;
+
+	void __iomem		*bar0;	/* BAR0 address */
+
+	struct bna bna;
+
+	u32		cfg_flags;
+	unsigned long		run_flags;
+
+	struct pci_dev 		*pcidev;
+	u64		mmio_start;
+	u64		mmio_len;
+
+	u32		msix_num;
+	u32		msix_diag_num;
+	struct msix_entry	*msix_table;
+
+	struct mutex		conf_mutex;
+	spinlock_t		bna_lock ____cacheline_aligned;
+
+	/* Timers */
+	struct timer_list	ioc_timer;
+	struct timer_list	dim_timer;
+	struct timer_list	stats_timer;
+
+	/* Control path resources, memory & irq */
+	struct bna_res_info res_info[BNA_RES_T_MAX];
+	struct bnad_tx_res_info tx_res_info[BNAD_MAX_TXS];
+	struct bnad_rx_res_info rx_res_info[BNAD_MAX_RXS];
+
+	struct bnad_completion bnad_completions;
+
+	/* Burnt in MAC address */
+	mac_t			perm_addr;
+
+	struct tasklet_struct	tx_free_tasklet;
+
+	/* Statistics */
+	struct bnad_stats stats;
+	struct net_device_stats net_stats;
+
+	struct bnad_diag *diag;
+
+	char			adapter_name[BNAD_NAME_LEN];
+	char 			port_name[BNAD_NAME_LEN];
+	char			mbox_irq_name[BNAD_NAME_LEN];
+};
+
+/*
+ * EXTERN VARIABLES
+ */
+extern struct firmware *bfi_fw;
+extern u32 		bnad_rxqs_per_cq;
+
+/*
+ * EXTERN PROTOTYPES
+ */
+extern u32 *cna_get_firmware_buf(struct pci_dev *pdev);
+/* Netdev entry point prototypes */
+extern void bnad_set_ethtool_ops(struct net_device *netdev);
+
+/* Configuration & setup */
+extern void bnad_tx_coalescing_timeo_set(struct bnad *bnad);
+extern void bnad_rx_coalescing_timeo_set(struct bnad *bnad);
+
+extern int bnad_setup_rx(struct bnad *bnad, uint rx_id);
+extern int bnad_setup_tx(struct bnad *bnad, uint tx_id);
+extern void bnad_cleanup_tx(struct bnad *bnad, uint tx_id);
+extern void bnad_cleanup_rx(struct bnad *bnad, uint rx_id);
+
+/* Timer start/stop protos */
+extern void bnad_dim_timer_start(struct bnad *bnad);
+
+/* Statistics */
+extern void bnad_netdev_qstats_fill(struct bnad *bnad);
+extern void bnad_netdev_hwstats_fill(struct bnad *bnad);
+
+/**
+ * MACROS
+ */
+/* To set & get the stats counters */
+#define BNAD_UPDATE_CTR(_bnad, _ctr)				\
+				(((_bnad)->stats.drv_stats._ctr)++)
+
+#define BNAD_GET_CTR(_bnad, _ctr) ((_bnad)->stats.drv_stats._ctr)
+
+#define bnad_enable_rx_irq_unsafe(_ccb)			\
+{							\
+	bna_ib_coalescing_timer_set((_ccb)->i_dbell,	\
+		(_ccb)->rx_coalescing_timeo);		\
+	bna_ib_ack((_ccb)->i_dbell, 0);			\
+}
+
+#define bnad_dim_timer_running(_bnad)				\
+	(((_bnad)->cfg_flags & BNAD_CF_DIM_ENABLED) && 		\
+	(test_bit(BNAD_RF_DIM_TIMER_RUNNING, &((_bnad)->run_flags))))
+
+#endif /* __BNAD_H__ */
diff --git a/drivers/net/bna/bnad_ethtool.c b/drivers/net/bna/bnad_ethtool.c
new file mode 100644
index 000000000000..e982785b6b25
--- /dev/null
+++ b/drivers/net/bna/bnad_ethtool.c
@@ -0,0 +1,1282 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#include "cna.h"
+
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+#include <linux/ethtool.h>
+#include <linux/rtnetlink.h>
+
+#include "bna.h"
+
+#include "bnad.h"
+
+#define BNAD_NUM_TXF_COUNTERS 12
+#define BNAD_NUM_RXF_COUNTERS 10
+#define BNAD_NUM_CQ_COUNTERS 3
+#define BNAD_NUM_RXQ_COUNTERS 6
+#define BNAD_NUM_TXQ_COUNTERS 5
+
+#define BNAD_ETHTOOL_STATS_NUM						\
+	(sizeof(struct net_device_stats) / sizeof(unsigned long) +	\
+	sizeof(struct bnad_drv_stats) / sizeof(u64) +		\
+	offsetof(struct bfi_ll_stats, rxf_stats[0]) / sizeof(u64))
+
+static char *bnad_net_stats_strings[BNAD_ETHTOOL_STATS_NUM] = {
+	"rx_packets",
+	"tx_packets",
+	"rx_bytes",
+	"tx_bytes",
+	"rx_errors",
+	"tx_errors",
+	"rx_dropped",
+	"tx_dropped",
+	"multicast",
+	"collisions",
+
+	"rx_length_errors",
+	"rx_over_errors",
+	"rx_crc_errors",
+	"rx_frame_errors",
+	"rx_fifo_errors",
+	"rx_missed_errors",
+
+	"tx_aborted_errors",
+	"tx_carrier_errors",
+	"tx_fifo_errors",
+	"tx_heartbeat_errors",
+	"tx_window_errors",
+
+	"rx_compressed",
+	"tx_compressed",
+
+	"netif_queue_stop",
+	"netif_queue_wakeup",
+	"tso4",
+	"tso6",
+	"tso_err",
+	"tcpcsum_offload",
+	"udpcsum_offload",
+	"csum_help",
+	"csum_help_err",
+	"hw_stats_updates",
+	"netif_rx_schedule",
+	"netif_rx_complete",
+	"netif_rx_dropped",
+
+	"link_toggle",
+	"cee_up",
+
+	"rxp_info_alloc_failed",
+	"mbox_intr_disabled",
+	"mbox_intr_enabled",
+	"tx_unmap_q_alloc_failed",
+	"rx_unmap_q_alloc_failed",
+	"rxbuf_alloc_failed",
+
+	"mac_frame_64",
+	"mac_frame_65_127",
+	"mac_frame_128_255",
+	"mac_frame_256_511",
+	"mac_frame_512_1023",
+	"mac_frame_1024_1518",
+	"mac_frame_1518_1522",
+	"mac_rx_bytes",
+	"mac_rx_packets",
+	"mac_rx_fcs_error",
+	"mac_rx_multicast",
+	"mac_rx_broadcast",
+	"mac_rx_control_frames",
+	"mac_rx_pause",
+	"mac_rx_unknown_opcode",
+	"mac_rx_alignment_error",
+	"mac_rx_frame_length_error",
+	"mac_rx_code_error",
+	"mac_rx_carrier_sense_error",
+	"mac_rx_undersize",
+	"mac_rx_oversize",
+	"mac_rx_fragments",
+	"mac_rx_jabber",
+	"mac_rx_drop",
+
+	"mac_tx_bytes",
+	"mac_tx_packets",
+	"mac_tx_multicast",
+	"mac_tx_broadcast",
+	"mac_tx_pause",
+	"mac_tx_deferral",
+	"mac_tx_excessive_deferral",
+	"mac_tx_single_collision",
+	"mac_tx_muliple_collision",
+	"mac_tx_late_collision",
+	"mac_tx_excessive_collision",
+	"mac_tx_total_collision",
+	"mac_tx_pause_honored",
+	"mac_tx_drop",
+	"mac_tx_jabber",
+	"mac_tx_fcs_error",
+	"mac_tx_control_frame",
+	"mac_tx_oversize",
+	"mac_tx_undersize",
+	"mac_tx_fragments",
+
+	"bpc_tx_pause_0",
+	"bpc_tx_pause_1",
+	"bpc_tx_pause_2",
+	"bpc_tx_pause_3",
+	"bpc_tx_pause_4",
+	"bpc_tx_pause_5",
+	"bpc_tx_pause_6",
+	"bpc_tx_pause_7",
+	"bpc_tx_zero_pause_0",
+	"bpc_tx_zero_pause_1",
+	"bpc_tx_zero_pause_2",
+	"bpc_tx_zero_pause_3",
+	"bpc_tx_zero_pause_4",
+	"bpc_tx_zero_pause_5",
+	"bpc_tx_zero_pause_6",
+	"bpc_tx_zero_pause_7",
+	"bpc_tx_first_pause_0",
+	"bpc_tx_first_pause_1",
+	"bpc_tx_first_pause_2",
+	"bpc_tx_first_pause_3",
+	"bpc_tx_first_pause_4",
+	"bpc_tx_first_pause_5",
+	"bpc_tx_first_pause_6",
+	"bpc_tx_first_pause_7",
+
+	"bpc_rx_pause_0",
+	"bpc_rx_pause_1",
+	"bpc_rx_pause_2",
+	"bpc_rx_pause_3",
+	"bpc_rx_pause_4",
+	"bpc_rx_pause_5",
+	"bpc_rx_pause_6",
+	"bpc_rx_pause_7",
+	"bpc_rx_zero_pause_0",
+	"bpc_rx_zero_pause_1",
+	"bpc_rx_zero_pause_2",
+	"bpc_rx_zero_pause_3",
+	"bpc_rx_zero_pause_4",
+	"bpc_rx_zero_pause_5",
+	"bpc_rx_zero_pause_6",
+	"bpc_rx_zero_pause_7",
+	"bpc_rx_first_pause_0",
+	"bpc_rx_first_pause_1",
+	"bpc_rx_first_pause_2",
+	"bpc_rx_first_pause_3",
+	"bpc_rx_first_pause_4",
+	"bpc_rx_first_pause_5",
+	"bpc_rx_first_pause_6",
+	"bpc_rx_first_pause_7",
+
+	"rad_rx_frames",
+	"rad_rx_octets",
+	"rad_rx_vlan_frames",
+	"rad_rx_ucast",
+	"rad_rx_ucast_octets",
+	"rad_rx_ucast_vlan",
+	"rad_rx_mcast",
+	"rad_rx_mcast_octets",
+	"rad_rx_mcast_vlan",
+	"rad_rx_bcast",
+	"rad_rx_bcast_octets",
+	"rad_rx_bcast_vlan",
+	"rad_rx_drops",
+
+	"fc_rx_ucast_octets",
+	"fc_rx_ucast",
+	"fc_rx_ucast_vlan",
+	"fc_rx_mcast_octets",
+	"fc_rx_mcast",
+	"fc_rx_mcast_vlan",
+	"fc_rx_bcast_octets",
+	"fc_rx_bcast",
+	"fc_rx_bcast_vlan",
+
+	"fc_tx_ucast_octets",
+	"fc_tx_ucast",
+	"fc_tx_ucast_vlan",
+	"fc_tx_mcast_octets",
+	"fc_tx_mcast",
+	"fc_tx_mcast_vlan",
+	"fc_tx_bcast_octets",
+	"fc_tx_bcast",
+	"fc_tx_bcast_vlan",
+	"fc_tx_parity_errors",
+	"fc_tx_timeout",
+	"fc_tx_fid_parity_errors",
+};
+
+static int
+bnad_get_settings(struct net_device *netdev, struct ethtool_cmd *cmd)
+{
+	cmd->supported = SUPPORTED_10000baseT_Full;
+	cmd->advertising = ADVERTISED_10000baseT_Full;
+	cmd->autoneg = AUTONEG_DISABLE;
+	cmd->supported |= SUPPORTED_FIBRE;
+	cmd->advertising |= ADVERTISED_FIBRE;
+	cmd->port = PORT_FIBRE;
+	cmd->phy_address = 0;
+
+	if (netif_carrier_ok(netdev)) {
+		cmd->speed = SPEED_10000;
+		cmd->duplex = DUPLEX_FULL;
+	} else {
+		cmd->speed = -1;
+		cmd->duplex = -1;
+	}
+	cmd->transceiver = XCVR_EXTERNAL;
+	cmd->maxtxpkt = 0;
+	cmd->maxrxpkt = 0;
+
+	return 0;
+}
+
+static int
+bnad_set_settings(struct net_device *netdev, struct ethtool_cmd *cmd)
+{
+	/* 10G full duplex setting supported only */
+	if (cmd->autoneg == AUTONEG_ENABLE)
+		return -EOPNOTSUPP; else {
+		if ((cmd->speed == SPEED_10000) && (cmd->duplex == DUPLEX_FULL))
+			return 0;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static void
+bnad_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	struct bfa_ioc_attr *ioc_attr;
+	unsigned long flags;
+
+	strcpy(drvinfo->driver, BNAD_NAME);
+	strcpy(drvinfo->version, BNAD_VERSION);
+
+	ioc_attr = kzalloc(sizeof(*ioc_attr), GFP_KERNEL);
+	if (ioc_attr) {
+		memset(ioc_attr, 0, sizeof(*ioc_attr));
+		spin_lock_irqsave(&bnad->bna_lock, flags);
+		bfa_ioc_get_attr(&bnad->bna.device.ioc, ioc_attr);
+		spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+		strncpy(drvinfo->fw_version, ioc_attr->adapter_attr.fw_ver,
+			sizeof(drvinfo->fw_version) - 1);
+		kfree(ioc_attr);
+	}
+
+	strncpy(drvinfo->bus_info, pci_name(bnad->pcidev), ETHTOOL_BUSINFO_LEN);
+}
+
+static int
+get_regs(struct bnad *bnad, u32 * regs)
+{
+	int num = 0, i;
+	u32 reg_addr;
+	unsigned long flags;
+
+#define BNAD_GET_REG(addr) 					\
+do {								\
+	if (regs)						\
+		regs[num++] = readl(bnad->bar0 + (addr));	\
+	else							\
+		num++;						\
+} while (0)
+
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+
+	/* DMA Block Internal Registers */
+	BNAD_GET_REG(DMA_CTRL_REG0);
+	BNAD_GET_REG(DMA_CTRL_REG1);
+	BNAD_GET_REG(DMA_ERR_INT_STATUS);
+	BNAD_GET_REG(DMA_ERR_INT_ENABLE);
+	BNAD_GET_REG(DMA_ERR_INT_STATUS_SET);
+
+	/* APP Block Register Address Offset from BAR0 */
+	BNAD_GET_REG(HOSTFN0_INT_STATUS);
+	BNAD_GET_REG(HOSTFN0_INT_MASK);
+	BNAD_GET_REG(HOST_PAGE_NUM_FN0);
+	BNAD_GET_REG(HOST_MSIX_ERR_INDEX_FN0);
+	BNAD_GET_REG(FN0_PCIE_ERR_REG);
+	BNAD_GET_REG(FN0_ERR_TYPE_STATUS_REG);
+	BNAD_GET_REG(FN0_ERR_TYPE_MSK_STATUS_REG);
+
+	BNAD_GET_REG(HOSTFN1_INT_STATUS);
+	BNAD_GET_REG(HOSTFN1_INT_MASK);
+	BNAD_GET_REG(HOST_PAGE_NUM_FN1);
+	BNAD_GET_REG(HOST_MSIX_ERR_INDEX_FN1);
+	BNAD_GET_REG(FN1_PCIE_ERR_REG);
+	BNAD_GET_REG(FN1_ERR_TYPE_STATUS_REG);
+	BNAD_GET_REG(FN1_ERR_TYPE_MSK_STATUS_REG);
+
+	BNAD_GET_REG(PCIE_MISC_REG);
+
+	BNAD_GET_REG(HOST_SEM0_REG);
+	BNAD_GET_REG(HOST_SEM1_REG);
+	BNAD_GET_REG(HOST_SEM2_REG);
+	BNAD_GET_REG(HOST_SEM3_REG);
+	BNAD_GET_REG(HOST_SEM0_INFO_REG);
+	BNAD_GET_REG(HOST_SEM1_INFO_REG);
+	BNAD_GET_REG(HOST_SEM2_INFO_REG);
+	BNAD_GET_REG(HOST_SEM3_INFO_REG);
+
+	BNAD_GET_REG(TEMPSENSE_CNTL_REG);
+	BNAD_GET_REG(TEMPSENSE_STAT_REG);
+
+	BNAD_GET_REG(APP_LOCAL_ERR_STAT);
+	BNAD_GET_REG(APP_LOCAL_ERR_MSK);
+
+	BNAD_GET_REG(PCIE_LNK_ERR_STAT);
+	BNAD_GET_REG(PCIE_LNK_ERR_MSK);
+
+	BNAD_GET_REG(FCOE_FIP_ETH_TYPE);
+	BNAD_GET_REG(RESV_ETH_TYPE);
+
+	BNAD_GET_REG(HOSTFN2_INT_STATUS);
+	BNAD_GET_REG(HOSTFN2_INT_MASK);
+	BNAD_GET_REG(HOST_PAGE_NUM_FN2);
+	BNAD_GET_REG(HOST_MSIX_ERR_INDEX_FN2);
+	BNAD_GET_REG(FN2_PCIE_ERR_REG);
+	BNAD_GET_REG(FN2_ERR_TYPE_STATUS_REG);
+	BNAD_GET_REG(FN2_ERR_TYPE_MSK_STATUS_REG);
+
+	BNAD_GET_REG(HOSTFN3_INT_STATUS);
+	BNAD_GET_REG(HOSTFN3_INT_MASK);
+	BNAD_GET_REG(HOST_PAGE_NUM_FN3);
+	BNAD_GET_REG(HOST_MSIX_ERR_INDEX_FN3);
+	BNAD_GET_REG(FN3_PCIE_ERR_REG);
+	BNAD_GET_REG(FN3_ERR_TYPE_STATUS_REG);
+	BNAD_GET_REG(FN3_ERR_TYPE_MSK_STATUS_REG);
+
+	/* Host Command Status Registers */
+	reg_addr = HOST_CMDSTS0_CLR_REG;
+	for (i = 0; i < 16; i++) {
+		BNAD_GET_REG(reg_addr);
+		BNAD_GET_REG(reg_addr + 4);
+		BNAD_GET_REG(reg_addr + 8);
+		reg_addr += 0x10;
+	}
+
+	/* Function ID register */
+	BNAD_GET_REG(FNC_ID_REG);
+
+	/* Function personality register */
+	BNAD_GET_REG(FNC_PERS_REG);
+
+	/* Operation mode register */
+	BNAD_GET_REG(OP_MODE);
+
+	/* LPU0 Registers */
+	BNAD_GET_REG(LPU0_MBOX_CTL_REG);
+	BNAD_GET_REG(LPU0_MBOX_CMD_REG);
+	BNAD_GET_REG(LPU0_MBOX_LINK_0REG);
+	BNAD_GET_REG(LPU1_MBOX_LINK_0REG);
+	BNAD_GET_REG(LPU0_MBOX_STATUS_0REG);
+	BNAD_GET_REG(LPU1_MBOX_STATUS_0REG);
+	BNAD_GET_REG(LPU0_ERR_STATUS_REG);
+	BNAD_GET_REG(LPU0_ERR_SET_REG);
+
+	/* LPU1 Registers */
+	BNAD_GET_REG(LPU1_MBOX_CTL_REG);
+	BNAD_GET_REG(LPU1_MBOX_CMD_REG);
+	BNAD_GET_REG(LPU0_MBOX_LINK_1REG);
+	BNAD_GET_REG(LPU1_MBOX_LINK_1REG);
+	BNAD_GET_REG(LPU0_MBOX_STATUS_1REG);
+	BNAD_GET_REG(LPU1_MBOX_STATUS_1REG);
+	BNAD_GET_REG(LPU1_ERR_STATUS_REG);
+	BNAD_GET_REG(LPU1_ERR_SET_REG);
+
+	/* PSS Registers */
+	BNAD_GET_REG(PSS_CTL_REG);
+	BNAD_GET_REG(PSS_ERR_STATUS_REG);
+	BNAD_GET_REG(ERR_STATUS_SET);
+	BNAD_GET_REG(PSS_RAM_ERR_STATUS_REG);
+
+	/* Catapult CPQ Registers */
+	BNAD_GET_REG(HOSTFN0_LPU0_MBOX0_CMD_STAT);
+	BNAD_GET_REG(HOSTFN0_LPU1_MBOX0_CMD_STAT);
+	BNAD_GET_REG(LPU0_HOSTFN0_MBOX0_CMD_STAT);
+	BNAD_GET_REG(LPU1_HOSTFN0_MBOX0_CMD_STAT);
+
+	BNAD_GET_REG(HOSTFN0_LPU0_MBOX1_CMD_STAT);
+	BNAD_GET_REG(HOSTFN0_LPU1_MBOX1_CMD_STAT);
+	BNAD_GET_REG(LPU0_HOSTFN0_MBOX1_CMD_STAT);
+	BNAD_GET_REG(LPU1_HOSTFN0_MBOX1_CMD_STAT);
+
+	BNAD_GET_REG(HOSTFN1_LPU0_MBOX0_CMD_STAT);
+	BNAD_GET_REG(HOSTFN1_LPU1_MBOX0_CMD_STAT);
+	BNAD_GET_REG(LPU0_HOSTFN1_MBOX0_CMD_STAT);
+	BNAD_GET_REG(LPU1_HOSTFN1_MBOX0_CMD_STAT);
+
+	BNAD_GET_REG(HOSTFN1_LPU0_MBOX1_CMD_STAT);
+	BNAD_GET_REG(HOSTFN1_LPU1_MBOX1_CMD_STAT);
+	BNAD_GET_REG(LPU0_HOSTFN1_MBOX1_CMD_STAT);
+	BNAD_GET_REG(LPU1_HOSTFN1_MBOX1_CMD_STAT);
+
+	BNAD_GET_REG(HOSTFN2_LPU0_MBOX0_CMD_STAT);
+	BNAD_GET_REG(HOSTFN2_LPU1_MBOX0_CMD_STAT);
+	BNAD_GET_REG(LPU0_HOSTFN2_MBOX0_CMD_STAT);
+	BNAD_GET_REG(LPU1_HOSTFN2_MBOX0_CMD_STAT);
+
+	BNAD_GET_REG(HOSTFN2_LPU0_MBOX1_CMD_STAT);
+	BNAD_GET_REG(HOSTFN2_LPU1_MBOX1_CMD_STAT);
+	BNAD_GET_REG(LPU0_HOSTFN2_MBOX1_CMD_STAT);
+	BNAD_GET_REG(LPU1_HOSTFN2_MBOX1_CMD_STAT);
+
+	BNAD_GET_REG(HOSTFN3_LPU0_MBOX0_CMD_STAT);
+	BNAD_GET_REG(HOSTFN3_LPU1_MBOX0_CMD_STAT);
+	BNAD_GET_REG(LPU0_HOSTFN3_MBOX0_CMD_STAT);
+	BNAD_GET_REG(LPU1_HOSTFN3_MBOX0_CMD_STAT);
+
+	BNAD_GET_REG(HOSTFN3_LPU0_MBOX1_CMD_STAT);
+	BNAD_GET_REG(HOSTFN3_LPU1_MBOX1_CMD_STAT);
+	BNAD_GET_REG(LPU0_HOSTFN3_MBOX1_CMD_STAT);
+	BNAD_GET_REG(LPU1_HOSTFN3_MBOX1_CMD_STAT);
+
+	/* Host Function Force Parity Error Registers */
+	BNAD_GET_REG(HOSTFN0_LPU_FORCE_PERR);
+	BNAD_GET_REG(HOSTFN1_LPU_FORCE_PERR);
+	BNAD_GET_REG(HOSTFN2_LPU_FORCE_PERR);
+	BNAD_GET_REG(HOSTFN3_LPU_FORCE_PERR);
+
+	/* LL Port[0|1] Halt Mask Registers */
+	BNAD_GET_REG(LL_HALT_MSK_P0);
+	BNAD_GET_REG(LL_HALT_MSK_P1);
+
+	/* LL Port[0|1] Error Mask Registers */
+	BNAD_GET_REG(LL_ERR_MSK_P0);
+	BNAD_GET_REG(LL_ERR_MSK_P1);
+
+	/* EMC FLI Registers */
+	BNAD_GET_REG(FLI_CMD_REG);
+	BNAD_GET_REG(FLI_ADDR_REG);
+	BNAD_GET_REG(FLI_CTL_REG);
+	BNAD_GET_REG(FLI_WRDATA_REG);
+	BNAD_GET_REG(FLI_RDDATA_REG);
+	BNAD_GET_REG(FLI_DEV_STATUS_REG);
+	BNAD_GET_REG(FLI_SIG_WD_REG);
+
+	BNAD_GET_REG(FLI_DEV_VENDOR_REG);
+	BNAD_GET_REG(FLI_ERR_STATUS_REG);
+
+	/* RxAdm 0 Registers */
+	BNAD_GET_REG(RAD0_CTL_REG);
+	BNAD_GET_REG(RAD0_PE_PARM_REG);
+	BNAD_GET_REG(RAD0_BCN_REG);
+	BNAD_GET_REG(RAD0_DEFAULT_REG);
+	BNAD_GET_REG(RAD0_PROMISC_REG);
+	BNAD_GET_REG(RAD0_BCNQ_REG);
+	BNAD_GET_REG(RAD0_DEFAULTQ_REG);
+
+	BNAD_GET_REG(RAD0_ERR_STS);
+	BNAD_GET_REG(RAD0_SET_ERR_STS);
+	BNAD_GET_REG(RAD0_ERR_INT_EN);
+	BNAD_GET_REG(RAD0_FIRST_ERR);
+	BNAD_GET_REG(RAD0_FORCE_ERR);
+
+	BNAD_GET_REG(RAD0_MAC_MAN_1H);
+	BNAD_GET_REG(RAD0_MAC_MAN_1L);
+	BNAD_GET_REG(RAD0_MAC_MAN_2H);
+	BNAD_GET_REG(RAD0_MAC_MAN_2L);
+	BNAD_GET_REG(RAD0_MAC_MAN_3H);
+	BNAD_GET_REG(RAD0_MAC_MAN_3L);
+	BNAD_GET_REG(RAD0_MAC_MAN_4H);
+	BNAD_GET_REG(RAD0_MAC_MAN_4L);
+
+	BNAD_GET_REG(RAD0_LAST4_IP);
+
+	/* RxAdm 1 Registers */
+	BNAD_GET_REG(RAD1_CTL_REG);
+	BNAD_GET_REG(RAD1_PE_PARM_REG);
+	BNAD_GET_REG(RAD1_BCN_REG);
+	BNAD_GET_REG(RAD1_DEFAULT_REG);
+	BNAD_GET_REG(RAD1_PROMISC_REG);
+	BNAD_GET_REG(RAD1_BCNQ_REG);
+	BNAD_GET_REG(RAD1_DEFAULTQ_REG);
+
+	BNAD_GET_REG(RAD1_ERR_STS);
+	BNAD_GET_REG(RAD1_SET_ERR_STS);
+	BNAD_GET_REG(RAD1_ERR_INT_EN);
+
+	/* TxA0 Registers */
+	BNAD_GET_REG(TXA0_CTRL_REG);
+	/* TxA0 TSO Sequence # Registers (RO) */
+	for (i = 0; i < 8; i++) {
+		BNAD_GET_REG(TXA0_TSO_TCP_SEQ_REG(i));
+		BNAD_GET_REG(TXA0_TSO_IP_INFO_REG(i));
+	}
+
+	/* TxA1 Registers */
+	BNAD_GET_REG(TXA1_CTRL_REG);
+	/* TxA1 TSO Sequence # Registers (RO) */
+	for (i = 0; i < 8; i++) {
+		BNAD_GET_REG(TXA1_TSO_TCP_SEQ_REG(i));
+		BNAD_GET_REG(TXA1_TSO_IP_INFO_REG(i));
+	}
+
+	/* RxA Registers */
+	BNAD_GET_REG(RXA0_CTL_REG);
+	BNAD_GET_REG(RXA1_CTL_REG);
+
+	/* PLB0 Registers */
+	BNAD_GET_REG(PLB0_ECM_TIMER_REG);
+	BNAD_GET_REG(PLB0_RL_CTL);
+	for (i = 0; i < 8; i++)
+		BNAD_GET_REG(PLB0_RL_MAX_BC(i));
+	BNAD_GET_REG(PLB0_RL_TU_PRIO);
+	for (i = 0; i < 8; i++)
+		BNAD_GET_REG(PLB0_RL_BYTE_CNT(i));
+	BNAD_GET_REG(PLB0_RL_MIN_REG);
+	BNAD_GET_REG(PLB0_RL_MAX_REG);
+	BNAD_GET_REG(PLB0_EMS_ADD_REG);
+
+	/* PLB1 Registers */
+	BNAD_GET_REG(PLB1_ECM_TIMER_REG);
+	BNAD_GET_REG(PLB1_RL_CTL);
+	for (i = 0; i < 8; i++)
+		BNAD_GET_REG(PLB1_RL_MAX_BC(i));
+	BNAD_GET_REG(PLB1_RL_TU_PRIO);
+	for (i = 0; i < 8; i++)
+		BNAD_GET_REG(PLB1_RL_BYTE_CNT(i));
+	BNAD_GET_REG(PLB1_RL_MIN_REG);
+	BNAD_GET_REG(PLB1_RL_MAX_REG);
+	BNAD_GET_REG(PLB1_EMS_ADD_REG);
+
+	/* HQM Control Register */
+	BNAD_GET_REG(HQM0_CTL_REG);
+	BNAD_GET_REG(HQM0_RXQ_STOP_SEM);
+	BNAD_GET_REG(HQM0_TXQ_STOP_SEM);
+	BNAD_GET_REG(HQM1_CTL_REG);
+	BNAD_GET_REG(HQM1_RXQ_STOP_SEM);
+	BNAD_GET_REG(HQM1_TXQ_STOP_SEM);
+
+	/* LUT Registers */
+	BNAD_GET_REG(LUT0_ERR_STS);
+	BNAD_GET_REG(LUT0_SET_ERR_STS);
+	BNAD_GET_REG(LUT1_ERR_STS);
+	BNAD_GET_REG(LUT1_SET_ERR_STS);
+
+	/* TRC Registers */
+	BNAD_GET_REG(TRC_CTL_REG);
+	BNAD_GET_REG(TRC_MODS_REG);
+	BNAD_GET_REG(TRC_TRGC_REG);
+	BNAD_GET_REG(TRC_CNT1_REG);
+	BNAD_GET_REG(TRC_CNT2_REG);
+	BNAD_GET_REG(TRC_NXTS_REG);
+	BNAD_GET_REG(TRC_DIRR_REG);
+	for (i = 0; i < 10; i++)
+		BNAD_GET_REG(TRC_TRGM_REG(i));
+	for (i = 0; i < 10; i++)
+		BNAD_GET_REG(TRC_NXTM_REG(i));
+	for (i = 0; i < 10; i++)
+		BNAD_GET_REG(TRC_STRM_REG(i));
+
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+#undef BNAD_GET_REG
+	return num;
+}
+static int
+bnad_get_regs_len(struct net_device *netdev)
+{
+	int ret = get_regs(netdev_priv(netdev), NULL) * sizeof(u32);
+	return ret;
+}
+
+static void
+bnad_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *buf)
+{
+	memset(buf, 0, bnad_get_regs_len(netdev));
+	get_regs(netdev_priv(netdev), buf);
+}
+
+static void
+bnad_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wolinfo)
+{
+	wolinfo->supported = 0;
+	wolinfo->wolopts = 0;
+}
+
+static int
+bnad_get_coalesce(struct net_device *netdev, struct ethtool_coalesce *coalesce)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	unsigned long flags;
+
+	/* Lock rqd. to access bnad->bna_lock */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	coalesce->use_adaptive_rx_coalesce =
+		(bnad->cfg_flags & BNAD_CF_DIM_ENABLED) ? true : false;
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	coalesce->rx_coalesce_usecs = bnad->rx_coalescing_timeo *
+					BFI_COALESCING_TIMER_UNIT;
+	coalesce->tx_coalesce_usecs = bnad->tx_coalescing_timeo *
+					BFI_COALESCING_TIMER_UNIT;
+	coalesce->tx_max_coalesced_frames = BFI_TX_INTERPKT_COUNT;
+
+	return 0;
+}
+
+static int
+bnad_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *coalesce)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	unsigned long flags;
+	int dim_timer_del = 0;
+
+	if (coalesce->rx_coalesce_usecs == 0 ||
+	    coalesce->rx_coalesce_usecs >
+	    BFI_MAX_COALESCING_TIMEO * BFI_COALESCING_TIMER_UNIT)
+		return -EINVAL;
+
+	if (coalesce->tx_coalesce_usecs == 0 ||
+	    coalesce->tx_coalesce_usecs >
+	    BFI_MAX_COALESCING_TIMEO * BFI_COALESCING_TIMER_UNIT)
+		return -EINVAL;
+
+	mutex_lock(&bnad->conf_mutex);
+	/*
+	 * Do not need to store rx_coalesce_usecs here
+	 * Every time DIM is disabled, we can get it from the
+	 * stack.
+	 */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	if (coalesce->use_adaptive_rx_coalesce) {
+		if (!(bnad->cfg_flags & BNAD_CF_DIM_ENABLED)) {
+			bnad->cfg_flags |= BNAD_CF_DIM_ENABLED;
+			bnad_dim_timer_start(bnad);
+		}
+	} else {
+		if (bnad->cfg_flags & BNAD_CF_DIM_ENABLED) {
+			bnad->cfg_flags &= ~BNAD_CF_DIM_ENABLED;
+			dim_timer_del = bnad_dim_timer_running(bnad);
+			if (dim_timer_del) {
+				clear_bit(BNAD_RF_DIM_TIMER_RUNNING,
+							&bnad->run_flags);
+				spin_unlock_irqrestore(&bnad->bna_lock, flags);
+				del_timer_sync(&bnad->dim_timer);
+				spin_lock_irqsave(&bnad->bna_lock, flags);
+			}
+			bnad_rx_coalescing_timeo_set(bnad);
+		}
+	}
+	if (bnad->tx_coalescing_timeo != coalesce->tx_coalesce_usecs /
+					BFI_COALESCING_TIMER_UNIT) {
+		bnad->tx_coalescing_timeo = coalesce->tx_coalesce_usecs /
+						BFI_COALESCING_TIMER_UNIT;
+		bnad_tx_coalescing_timeo_set(bnad);
+	}
+
+	if (bnad->rx_coalescing_timeo != coalesce->rx_coalesce_usecs /
+					BFI_COALESCING_TIMER_UNIT) {
+		bnad->rx_coalescing_timeo = coalesce->rx_coalesce_usecs /
+						BFI_COALESCING_TIMER_UNIT;
+
+		if (!(bnad->cfg_flags & BNAD_CF_DIM_ENABLED))
+			bnad_rx_coalescing_timeo_set(bnad);
+
+	}
+
+	/* Add Tx Inter-pkt DMA count?  */
+
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	mutex_unlock(&bnad->conf_mutex);
+	return 0;
+}
+
+static void
+bnad_get_ringparam(struct net_device *netdev,
+		   struct ethtool_ringparam *ringparam)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+
+	ringparam->rx_max_pending = BNAD_MAX_Q_DEPTH / bnad_rxqs_per_cq;
+	ringparam->rx_mini_max_pending = 0;
+	ringparam->rx_jumbo_max_pending = 0;
+	ringparam->tx_max_pending = BNAD_MAX_Q_DEPTH;
+
+	ringparam->rx_pending = bnad->rxq_depth;
+	ringparam->rx_mini_max_pending = 0;
+	ringparam->rx_jumbo_max_pending = 0;
+	ringparam->tx_pending = bnad->txq_depth;
+}
+
+static int
+bnad_set_ringparam(struct net_device *netdev,
+		   struct ethtool_ringparam *ringparam)
+{
+	int i, current_err, err = 0;
+	struct bnad *bnad = netdev_priv(netdev);
+
+	mutex_lock(&bnad->conf_mutex);
+	if (ringparam->rx_pending == bnad->rxq_depth &&
+	    ringparam->tx_pending == bnad->txq_depth) {
+		mutex_unlock(&bnad->conf_mutex);
+		return 0;
+	}
+
+	if (ringparam->rx_pending < BNAD_MIN_Q_DEPTH ||
+	    ringparam->rx_pending > BNAD_MAX_Q_DEPTH / bnad_rxqs_per_cq ||
+	    !BNA_POWER_OF_2(ringparam->rx_pending)) {
+		mutex_unlock(&bnad->conf_mutex);
+		return -EINVAL;
+	}
+	if (ringparam->tx_pending < BNAD_MIN_Q_DEPTH ||
+	    ringparam->tx_pending > BNAD_MAX_Q_DEPTH ||
+	    !BNA_POWER_OF_2(ringparam->tx_pending)) {
+		mutex_unlock(&bnad->conf_mutex);
+		return -EINVAL;
+	}
+
+	if (ringparam->rx_pending != bnad->rxq_depth) {
+		bnad->rxq_depth = ringparam->rx_pending;
+		for (i = 0; i < bnad->num_rx; i++) {
+			if (!bnad->rx_info[i].rx)
+				continue;
+			bnad_cleanup_rx(bnad, i);
+			current_err = bnad_setup_rx(bnad, i);
+			if (current_err && !err)
+				err = current_err;
+		}
+	}
+	if (ringparam->tx_pending != bnad->txq_depth) {
+		bnad->txq_depth = ringparam->tx_pending;
+		for (i = 0; i < bnad->num_tx; i++) {
+			if (!bnad->tx_info[i].tx)
+				continue;
+			bnad_cleanup_tx(bnad, i);
+			current_err = bnad_setup_tx(bnad, i);
+			if (current_err && !err)
+				err = current_err;
+		}
+	}
+
+	mutex_unlock(&bnad->conf_mutex);
+	return err;
+}
+
+static void
+bnad_get_pauseparam(struct net_device *netdev,
+		    struct ethtool_pauseparam *pauseparam)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+
+	pauseparam->autoneg = 0;
+	pauseparam->rx_pause = bnad->bna.port.pause_config.rx_pause;
+	pauseparam->tx_pause = bnad->bna.port.pause_config.tx_pause;
+}
+
+static int
+bnad_set_pauseparam(struct net_device *netdev,
+		    struct ethtool_pauseparam *pauseparam)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	struct bna_pause_config pause_config;
+	unsigned long flags;
+
+	if (pauseparam->autoneg == AUTONEG_ENABLE)
+		return -EINVAL;
+
+	mutex_lock(&bnad->conf_mutex);
+	if (pauseparam->rx_pause != bnad->bna.port.pause_config.rx_pause ||
+	    pauseparam->tx_pause != bnad->bna.port.pause_config.tx_pause) {
+		pause_config.rx_pause = pauseparam->rx_pause;
+		pause_config.tx_pause = pauseparam->tx_pause;
+		spin_lock_irqsave(&bnad->bna_lock, flags);
+		bna_port_pause_config(&bnad->bna.port, &pause_config, NULL);
+		spin_unlock_irqrestore(&bnad->bna_lock, flags);
+	}
+	mutex_unlock(&bnad->conf_mutex);
+	return 0;
+}
+
+static u32
+bnad_get_rx_csum(struct net_device *netdev)
+{
+	u32 rx_csum;
+	struct bnad *bnad = netdev_priv(netdev);
+
+	rx_csum = bnad->rx_csum;
+	return rx_csum;
+}
+
+static int
+bnad_set_rx_csum(struct net_device *netdev, u32 rx_csum)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+
+	mutex_lock(&bnad->conf_mutex);
+	bnad->rx_csum = rx_csum;
+	mutex_unlock(&bnad->conf_mutex);
+	return 0;
+}
+
+static int
+bnad_set_tx_csum(struct net_device *netdev, u32 tx_csum)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+
+	mutex_lock(&bnad->conf_mutex);
+	if (tx_csum) {
+		netdev->features |= NETIF_F_IP_CSUM;
+		netdev->features |= NETIF_F_IPV6_CSUM;
+	} else {
+		netdev->features &= ~NETIF_F_IP_CSUM;
+		netdev->features &= ~NETIF_F_IPV6_CSUM;
+	}
+	mutex_unlock(&bnad->conf_mutex);
+	return 0;
+}
+
+static int
+bnad_set_tso(struct net_device *netdev, u32 tso)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+
+	mutex_lock(&bnad->conf_mutex);
+	if (tso) {
+		netdev->features |= NETIF_F_TSO;
+		netdev->features |= NETIF_F_TSO6;
+	} else {
+		netdev->features &= ~NETIF_F_TSO;
+		netdev->features &= ~NETIF_F_TSO6;
+	}
+	mutex_unlock(&bnad->conf_mutex);
+	return 0;
+}
+
+static void
+bnad_get_strings(struct net_device *netdev, u32 stringset, u8 * string)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	int i, j, q_num;
+	u64 bmap;
+
+	mutex_lock(&bnad->conf_mutex);
+
+	switch (stringset) {
+	case ETH_SS_STATS:
+		for (i = 0; i < BNAD_ETHTOOL_STATS_NUM; i++) {
+			BUG_ON(!(strlen(bnad_net_stats_strings[i]) <
+				   ETH_GSTRING_LEN));
+			memcpy(string, bnad_net_stats_strings[i],
+			       ETH_GSTRING_LEN);
+			string += ETH_GSTRING_LEN;
+		}
+		bmap = (u64)bnad->bna.tx_mod.txf_bmap[0] |
+			((u64)bnad->bna.tx_mod.txf_bmap[1] << 32);
+		for (i = 0; bmap && (i < BFI_LL_TXF_ID_MAX); i++) {
+			if (bmap & 1) {
+				sprintf(string, "txf%d_ucast_octets", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_ucast", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_ucast_vlan", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_mcast_octets", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_mcast", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_mcast_vlan", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_bcast_octets", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_bcast", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_bcast_vlan", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_errors", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_filter_vlan", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txf%d_filter_mac_sa", i);
+				string += ETH_GSTRING_LEN;
+			}
+			bmap >>= 1;
+		}
+
+		bmap = (u64)bnad->bna.rx_mod.rxf_bmap[0] |
+			((u64)bnad->bna.rx_mod.rxf_bmap[1] << 32);
+		for (i = 0; bmap && (i < BFI_LL_RXF_ID_MAX); i++) {
+			if (bmap & 1) {
+				sprintf(string, "rxf%d_ucast_octets", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_ucast", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_ucast_vlan", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_mcast_octets", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_mcast", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_mcast_vlan", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_bcast_octets", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_bcast", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_bcast_vlan", i);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxf%d_frame_drops", i);
+				string += ETH_GSTRING_LEN;
+			}
+			bmap >>= 1;
+		}
+
+		q_num = 0;
+		for (i = 0; i < bnad->num_rx; i++) {
+			if (!bnad->rx_info[i].rx)
+				continue;
+			for (j = 0; j < bnad->num_rxp_per_rx; j++) {
+				sprintf(string, "cq%d_producer_index", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "cq%d_consumer_index", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "cq%d_hw_producer_index",
+					q_num);
+				string += ETH_GSTRING_LEN;
+				q_num++;
+			}
+		}
+
+		q_num = 0;
+		for (i = 0; i < bnad->num_rx; i++) {
+			if (!bnad->rx_info[i].rx)
+				continue;
+			for (j = 0; j < bnad->num_rxp_per_rx; j++) {
+				sprintf(string, "rxq%d_packets", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxq%d_bytes", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxq%d_packets_with_error",
+								q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxq%d_allocbuf_failed", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxq%d_producer_index", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "rxq%d_consumer_index", q_num);
+				string += ETH_GSTRING_LEN;
+				q_num++;
+				if (bnad->rx_info[i].rx_ctrl[j].ccb &&
+					bnad->rx_info[i].rx_ctrl[j].ccb->
+					rcb[1] &&
+					bnad->rx_info[i].rx_ctrl[j].ccb->
+					rcb[1]->rxq) {
+					sprintf(string, "rxq%d_packets", q_num);
+					string += ETH_GSTRING_LEN;
+					sprintf(string, "rxq%d_bytes", q_num);
+					string += ETH_GSTRING_LEN;
+					sprintf(string,
+					"rxq%d_packets_with_error", q_num);
+					string += ETH_GSTRING_LEN;
+					sprintf(string, "rxq%d_allocbuf_failed",
+								q_num);
+					string += ETH_GSTRING_LEN;
+					sprintf(string, "rxq%d_producer_index",
+								q_num);
+					string += ETH_GSTRING_LEN;
+					sprintf(string, "rxq%d_consumer_index",
+								q_num);
+					string += ETH_GSTRING_LEN;
+					q_num++;
+				}
+			}
+		}
+
+		q_num = 0;
+		for (i = 0; i < bnad->num_tx; i++) {
+			if (!bnad->tx_info[i].tx)
+				continue;
+			for (j = 0; j < bnad->num_txq_per_tx; j++) {
+				sprintf(string, "txq%d_packets", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txq%d_bytes", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txq%d_producer_index", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txq%d_consumer_index", q_num);
+				string += ETH_GSTRING_LEN;
+				sprintf(string, "txq%d_hw_consumer_index",
+									q_num);
+				string += ETH_GSTRING_LEN;
+				q_num++;
+			}
+		}
+
+		break;
+
+	default:
+		break;
+	}
+
+	mutex_unlock(&bnad->conf_mutex);
+}
+
+static int
+bnad_get_stats_count_locked(struct net_device *netdev)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	int i, j, count, rxf_active_num = 0, txf_active_num = 0;
+	u64 bmap;
+
+	bmap = (u64)bnad->bna.tx_mod.txf_bmap[0] |
+			((u64)bnad->bna.tx_mod.txf_bmap[1] << 32);
+	for (i = 0; bmap && (i < BFI_LL_TXF_ID_MAX); i++) {
+		if (bmap & 1)
+			txf_active_num++;
+		bmap >>= 1;
+	}
+	bmap = (u64)bnad->bna.rx_mod.rxf_bmap[0] |
+			((u64)bnad->bna.rx_mod.rxf_bmap[1] << 32);
+	for (i = 0; bmap && (i < BFI_LL_RXF_ID_MAX); i++) {
+		if (bmap & 1)
+			rxf_active_num++;
+		bmap >>= 1;
+	}
+	count = BNAD_ETHTOOL_STATS_NUM +
+		txf_active_num * BNAD_NUM_TXF_COUNTERS +
+		rxf_active_num * BNAD_NUM_RXF_COUNTERS;
+
+	for (i = 0; i < bnad->num_rx; i++) {
+		if (!bnad->rx_info[i].rx)
+			continue;
+		count += bnad->num_rxp_per_rx * BNAD_NUM_CQ_COUNTERS;
+		count += bnad->num_rxp_per_rx * BNAD_NUM_RXQ_COUNTERS;
+		for (j = 0; j < bnad->num_rxp_per_rx; j++)
+			if (bnad->rx_info[i].rx_ctrl[j].ccb &&
+				bnad->rx_info[i].rx_ctrl[j].ccb->rcb[1] &&
+				bnad->rx_info[i].rx_ctrl[j].ccb->rcb[1]->rxq)
+				count +=  BNAD_NUM_RXQ_COUNTERS;
+	}
+
+	for (i = 0; i < bnad->num_tx; i++) {
+		if (!bnad->tx_info[i].tx)
+			continue;
+		count += bnad->num_txq_per_tx * BNAD_NUM_TXQ_COUNTERS;
+	}
+	return count;
+}
+
+static int
+bnad_per_q_stats_fill(struct bnad *bnad, u64 *buf, int bi)
+{
+	int i, j;
+	struct bna_rcb *rcb = NULL;
+	struct bna_tcb *tcb = NULL;
+
+	for (i = 0; i < bnad->num_rx; i++) {
+		if (!bnad->rx_info[i].rx)
+			continue;
+		for (j = 0; j < bnad->num_rxp_per_rx; j++)
+			if (bnad->rx_info[i].rx_ctrl[j].ccb &&
+				bnad->rx_info[i].rx_ctrl[j].ccb->rcb[0] &&
+				bnad->rx_info[i].rx_ctrl[j].ccb->rcb[0]->rxq) {
+				buf[bi++] = bnad->rx_info[i].rx_ctrl[j].
+						ccb->producer_index;
+				buf[bi++] = 0; /* ccb->consumer_index */
+				buf[bi++] = *(bnad->rx_info[i].rx_ctrl[j].
+						ccb->hw_producer_index);
+			}
+	}
+	for (i = 0; i < bnad->num_rx; i++) {
+		if (!bnad->rx_info[i].rx)
+			continue;
+		for (j = 0; j < bnad->num_rxp_per_rx; j++)
+			if (bnad->rx_info[i].rx_ctrl[j].ccb) {
+				if (bnad->rx_info[i].rx_ctrl[j].ccb->rcb[0] &&
+					bnad->rx_info[i].rx_ctrl[j].ccb->
+					rcb[0]->rxq) {
+					rcb = bnad->rx_info[i].rx_ctrl[j].
+							ccb->rcb[0];
+					buf[bi++] = rcb->rxq->rx_packets;
+					buf[bi++] = rcb->rxq->rx_bytes;
+					buf[bi++] = rcb->rxq->
+							rx_packets_with_error;
+					buf[bi++] = rcb->rxq->
+							rxbuf_alloc_failed;
+					buf[bi++] = rcb->producer_index;
+					buf[bi++] = rcb->consumer_index;
+				}
+				if (bnad->rx_info[i].rx_ctrl[j].ccb->rcb[1] &&
+					bnad->rx_info[i].rx_ctrl[j].ccb->
+					rcb[1]->rxq) {
+					rcb = bnad->rx_info[i].rx_ctrl[j].
+								ccb->rcb[1];
+					buf[bi++] = rcb->rxq->rx_packets;
+					buf[bi++] = rcb->rxq->rx_bytes;
+					buf[bi++] = rcb->rxq->
+							rx_packets_with_error;
+					buf[bi++] = rcb->rxq->
+							rxbuf_alloc_failed;
+					buf[bi++] = rcb->producer_index;
+					buf[bi++] = rcb->consumer_index;
+				}
+			}
+	}
+
+	for (i = 0; i < bnad->num_tx; i++) {
+		if (!bnad->tx_info[i].tx)
+			continue;
+		for (j = 0; j < bnad->num_txq_per_tx; j++)
+			if (bnad->tx_info[i].tcb[j] &&
+				bnad->tx_info[i].tcb[j]->txq) {
+				tcb = bnad->tx_info[i].tcb[j];
+				buf[bi++] = tcb->txq->tx_packets;
+				buf[bi++] = tcb->txq->tx_bytes;
+				buf[bi++] = tcb->producer_index;
+				buf[bi++] = tcb->consumer_index;
+				buf[bi++] = *(tcb->hw_consumer_index);
+			}
+	}
+
+	return bi;
+}
+
+static void
+bnad_get_ethtool_stats(struct net_device *netdev, struct ethtool_stats *stats,
+		       u64 *buf)
+{
+	struct bnad *bnad = netdev_priv(netdev);
+	int i, j, bi;
+	unsigned long *net_stats, flags;
+	u64 *stats64;
+	u64 bmap;
+
+	mutex_lock(&bnad->conf_mutex);
+	if (bnad_get_stats_count_locked(netdev) != stats->n_stats) {
+		mutex_unlock(&bnad->conf_mutex);
+		return;
+	}
+
+	/*
+	 * Used bna_lock to sync reads from bna_stats, which is written
+	 * under the same lock
+	 */
+	spin_lock_irqsave(&bnad->bna_lock, flags);
+	bi = 0;
+	memset(buf, 0, stats->n_stats * sizeof(u64));
+	memset(&bnad->net_stats, 0, sizeof(struct net_device_stats));
+
+	bnad_netdev_qstats_fill(bnad);
+	bnad_netdev_hwstats_fill(bnad);
+
+	/* Fill net_stats into ethtool buffers */
+	net_stats = (unsigned long *)&bnad->net_stats;
+	for (i = 0; i < sizeof(struct net_device_stats) / sizeof(unsigned long);
+	     i++)
+		buf[bi++] = net_stats[i];
+
+	/* Fill driver stats into ethtool buffers */
+	stats64 = (u64 *)&bnad->stats.drv_stats;
+	for (i = 0; i < sizeof(struct bnad_drv_stats) / sizeof(u64); i++)
+		buf[bi++] = stats64[i];
+
+	/* Fill hardware stats excluding the rxf/txf into ethtool bufs */
+	stats64 = (u64 *) bnad->stats.bna_stats->hw_stats;
+	for (i = 0;
+	     i < offsetof(struct bfi_ll_stats, rxf_stats[0]) / sizeof(u64);
+	     i++)
+		buf[bi++] = stats64[i];
+
+	/* Fill txf stats into ethtool buffers */
+	bmap = (u64)bnad->bna.tx_mod.txf_bmap[0] |
+			((u64)bnad->bna.tx_mod.txf_bmap[1] << 32);
+	for (i = 0; bmap && (i < BFI_LL_TXF_ID_MAX); i++) {
+		if (bmap & 1) {
+			stats64 = (u64 *)&bnad->stats.bna_stats->
+						hw_stats->txf_stats[i];
+			for (j = 0; j < sizeof(struct bfi_ll_stats_txf) /
+					sizeof(u64); j++)
+				buf[bi++] = stats64[j];
+		}
+		bmap >>= 1;
+	}
+
+	/*  Fill rxf stats into ethtool buffers */
+	bmap = (u64)bnad->bna.rx_mod.rxf_bmap[0] |
+			((u64)bnad->bna.rx_mod.rxf_bmap[1] << 32);
+	for (i = 0; bmap && (i < BFI_LL_RXF_ID_MAX); i++) {
+		if (bmap & 1) {
+			stats64 = (u64 *)&bnad->stats.bna_stats->
+						hw_stats->rxf_stats[i];
+			for (j = 0; j < sizeof(struct bfi_ll_stats_rxf) /
+					sizeof(u64); j++)
+				buf[bi++] = stats64[j];
+		}
+		bmap >>= 1;
+	}
+
+	/* Fill per Q stats into ethtool buffers */
+	bi = bnad_per_q_stats_fill(bnad, buf, bi);
+
+	spin_unlock_irqrestore(&bnad->bna_lock, flags);
+
+	mutex_unlock(&bnad->conf_mutex);
+}
+
+static int
+bnad_get_sset_count(struct net_device *netdev, int sset)
+{
+	switch (sset) {
+	case ETH_SS_STATS:
+		return bnad_get_stats_count_locked(netdev);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static struct ethtool_ops bnad_ethtool_ops = {
+	.get_settings = bnad_get_settings,
+	.set_settings = bnad_set_settings,
+	.get_drvinfo = bnad_get_drvinfo,
+	.get_regs_len = bnad_get_regs_len,
+	.get_regs = bnad_get_regs,
+	.get_wol = bnad_get_wol,
+	.get_link = ethtool_op_get_link,
+	.get_coalesce = bnad_get_coalesce,
+	.set_coalesce = bnad_set_coalesce,
+	.get_ringparam = bnad_get_ringparam,
+	.set_ringparam = bnad_set_ringparam,
+	.get_pauseparam = bnad_get_pauseparam,
+	.set_pauseparam = bnad_set_pauseparam,
+	.get_rx_csum = bnad_get_rx_csum,
+	.set_rx_csum = bnad_set_rx_csum,
+	.get_tx_csum = ethtool_op_get_tx_csum,
+	.set_tx_csum = bnad_set_tx_csum,
+	.get_sg = ethtool_op_get_sg,
+	.set_sg = ethtool_op_set_sg,
+	.get_tso = ethtool_op_get_tso,
+	.set_tso = bnad_set_tso,
+	.get_flags = ethtool_op_get_flags,
+	.set_flags = ethtool_op_set_flags,
+	.get_strings = bnad_get_strings,
+	.get_ethtool_stats = bnad_get_ethtool_stats,
+	.get_sset_count = bnad_get_sset_count
+};
+
+void
+bnad_set_ethtool_ops(struct net_device *netdev)
+{
+	SET_ETHTOOL_OPS(netdev, &bnad_ethtool_ops);
+}
diff --git a/drivers/net/bna/cna.h b/drivers/net/bna/cna.h
new file mode 100644
index 000000000000..bbd39dc65972
--- /dev/null
+++ b/drivers/net/bna/cna.h
@@ -0,0 +1,81 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2006-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+
+#ifndef __CNA_H__
+#define __CNA_H__
+
+#include <linux/version.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/bitops.h>
+#include <linux/timer.h>
+#include <linux/interrupt.h>
+#include <linux/if_ether.h>
+#include <asm/page.h>
+#include <asm/io.h>
+#include <asm/string.h>
+
+#include <linux/list.h>
+
+#define bfa_sm_fault(__mod, __event)    do {                            \
+	pr_err("SM Assertion failure: %s: %d: event = %d", __FILE__, __LINE__, \
+		__event); \
+} while (0)
+
+extern char bfa_version[];
+
+#define	CNA_FW_FILE_CT	"ctfw_cna.bin"
+#define FC_SYMNAME_MAX	256	/*!< max name server symbolic name size */
+
+#pragma pack(1)
+
+#define MAC_ADDRLEN	(6)
+typedef struct mac { u8 mac[MAC_ADDRLEN]; } mac_t;
+
+#pragma pack()
+
+#define bfa_q_first(_q) ((void *)(((struct list_head *) (_q))->next))
+#define bfa_q_next(_qe)	(((struct list_head *) (_qe))->next)
+#define bfa_q_prev(_qe) (((struct list_head *) (_qe))->prev)
+
+/*
+ * bfa_q_qe_init - to initialize a queue element
+ */
+#define bfa_q_qe_init(_qe) {						\
+	bfa_q_next(_qe) = (struct list_head *) NULL;			\
+	bfa_q_prev(_qe) = (struct list_head *) NULL;			\
+}
+
+/*
+ * bfa_q_deq - dequeue an element from head of the queue
+ */
+#define bfa_q_deq(_q, _qe) {						\
+	if (!list_empty(_q)) {						\
+		(*((struct list_head **) (_qe))) = bfa_q_next(_q);	\
+		bfa_q_prev(bfa_q_next(*((struct list_head **) _qe))) =	\
+						(struct list_head *) (_q); \
+		bfa_q_next(_q) = bfa_q_next(*((struct list_head **) _qe)); \
+		bfa_q_qe_init(*((struct list_head **) _qe));		\
+	} else {							\
+		*((struct list_head **) (_qe)) = (struct list_head *) NULL; \
+	}								\
+}
+
+#endif /* __CNA_H__ */
diff --git a/drivers/net/bna/cna_fwimg.c b/drivers/net/bna/cna_fwimg.c
new file mode 100644
index 000000000000..0bd1d3790a27
--- /dev/null
+++ b/drivers/net/bna/cna_fwimg.c
@@ -0,0 +1,64 @@
+/*
+ * Linux network driver for Brocade Converged Network Adapter.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License (GPL) Version 2 as
+ * published by the Free Software Foundation
+ *
+ * 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.
+ */
+/*
+ * Copyright (c) 2005-2010 Brocade Communications Systems, Inc.
+ * All rights reserved
+ * www.brocade.com
+ */
+#include <linux/firmware.h>
+#include "cna.h"
+
+const struct firmware *bfi_fw;
+static u32 *bfi_image_ct_cna;
+static u32 bfi_image_ct_cna_size;
+
+u32 *
+cna_read_firmware(struct pci_dev *pdev, u32 **bfi_image,
+			u32 *bfi_image_size, char *fw_name)
+{
+	const struct firmware *fw;
+
+	if (request_firmware(&fw, fw_name, &pdev->dev)) {
+		pr_alert("Can't locate firmware %s\n", fw_name);
+		goto error;
+	}
+
+	*bfi_image = (u32 *)fw->data;
+	*bfi_image_size = fw->size/sizeof(u32);
+	bfi_fw = fw;
+
+	return *bfi_image;
+error:
+	return NULL;
+}
+
+u32 *
+cna_get_firmware_buf(struct pci_dev *pdev)
+{
+	if (bfi_image_ct_cna_size == 0)
+		cna_read_firmware(pdev, &bfi_image_ct_cna,
+			&bfi_image_ct_cna_size, CNA_FW_FILE_CT);
+	return bfi_image_ct_cna;
+}
+
+u32 *
+bfa_cb_image_get_chunk(int type, u32 off)
+{
+	return (u32 *)(bfi_image_ct_cna + off);
+}
+
+u32
+bfa_cb_image_get_size(int type)
+{
+	return bfi_image_ct_cna_size;
+}
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index f6a3b2d36cad..1f730de0df06 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2189,6 +2189,9 @@
 #define PCI_VENDOR_ID_ARIMA		0x161f
 
 #define PCI_VENDOR_ID_BROCADE		0x1657
+#define PCI_DEVICE_ID_BROCADE_CT	0x0014
+#define PCI_DEVICE_ID_BROCADE_FC_8G1P	0x0017
+#define PCI_DEVICE_ID_BROCADE_CT_FC	0x0021
 
 #define PCI_VENDOR_ID_SIBYTE		0x166d
 #define PCI_DEVICE_ID_BCM1250_PCI	0x0001
-- 
cgit v1.2.3


From 1726442e115a9e58f40747d009a5b4f303e0840a Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@verge.net.au>
Date: Mon, 23 Aug 2010 16:26:41 +0000
Subject: net: increase the size of priv_flags and add IFF_OVS_DATAPATH

IFF_OVS_DATAPATH is a place-holder for the Open vSwitch datapath
which I am preparing to submit for merging.

As all 16 bits of priv_flags are already assigned flags, also increase
the size of priv_flags to 32 bits.

Unfortunately, by my calculations this increases the size of
struct net_device by 4 bytes on 32bit architectures and
8 bytes on 64 bit architectures. I couldn't see an obvious
way to avoid that.

Cc: Jesse Gross <jesse@nicira.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if.h        | 2 ++
 include/linux/netdevice.h | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/if.h b/include/linux/if.h
index 53558ec59e1b..6ed43c1f07ab 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -75,6 +75,8 @@
 #define IFF_DISABLE_NETPOLL	0x2000	/* disable netpoll at run-time */
 #define IFF_MACVLAN_PORT	0x4000	/* device used as macvlan port */
 #define IFF_BRIDGE_PORT	0x8000		/* device used as bridge port */
+#define IFF_OVS_DATAPATH	0x10000	/* device used as Open vSwitch
+					 * dapath port */
 
 #define IF_GET_IFACE	0x0001		/* for querying only */
 #define IF_GET_PROTO	0x0002
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ce2de8b64083..59962dbc2758 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -901,7 +901,7 @@ struct net_device {
 
 	unsigned int		flags;	/* interface flags (a la BSD)	*/
 	unsigned short		gflags;
-        unsigned short          priv_flags; /* Like 'flags' but invisible to userspace. */
+        unsigned int            priv_flags; /* Like 'flags' but invisible to userspace. */
 	unsigned short		padded;	/* How much padding added by alloc_netdev() */
 
 	unsigned char		operstate; /* RFC2863 operstate */
-- 
cgit v1.2.3


From 2e161f78e5f63a7f9fd25a766bb7f816a01eb14a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 12 Aug 2010 15:38:38 +0200
Subject: cfg80211/mac80211: extensible frame processing

Allow userspace to register for more than just
action frames by giving the frame subtype, and
make it possible to use this in various modes
as well.

With some tweaks and some added functionality
this will, in the future, also be usable in AP
mode and be able to replace the cooked monitor
interface currently used in that case.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h    |  93 +++++++++++++++++++++++------
 include/net/cfg80211.h     |  56 +++++++++++-------
 net/mac80211/cfg.c         |  12 ++--
 net/mac80211/ieee80211_i.h |   1 +
 net/mac80211/iface.c       |   6 +-
 net/mac80211/main.c        |  37 ++++++++++++
 net/mac80211/rx.c          | 137 +++++++++++++++++++++++++++++-------------
 net/mac80211/status.c      |   2 +-
 net/mac80211/util.c        |   6 +-
 net/wireless/core.c        |   8 +--
 net/wireless/core.h        |  21 +++----
 net/wireless/mlme.c        | 144 +++++++++++++++++++++++++++++----------------
 net/wireless/nl80211.c     | 108 +++++++++++++++++++++++++---------
 net/wireless/nl80211.h     |  14 ++---
 net/wireless/util.c        |   2 +-
 15 files changed, 452 insertions(+), 195 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 2c8701687336..8af1e66c3cf9 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -39,6 +39,43 @@
  * TODO: need more info?
  */
 
+/**
+ * DOC: Frame transmission/registration support
+ *
+ * Frame transmission and registration support exists to allow userspace
+ * management entities such as wpa_supplicant react to management frames
+ * that are not being handled by the kernel. This includes, for example,
+ * certain classes of action frames that cannot be handled in the kernel
+ * for various reasons.
+ *
+ * Frame registration is done on a per-interface basis and registrations
+ * cannot be removed other than by closing the socket. It is possible to
+ * specify a registration filter to register, for example, only for a
+ * certain type of action frame. In particular with action frames, those
+ * that userspace registers for will not be returned as unhandled by the
+ * driver, so that the registered application has to take responsibility
+ * for doing that.
+ *
+ * The type of frame that can be registered for is also dependent on the
+ * driver and interface type. The frame types are advertised in wiphy
+ * attributes so applications know what to expect.
+ *
+ * NOTE: When an interface changes type while registrations are active,
+ *       these registrations are ignored until the interface type is
+ *       changed again. This means that changing the interface type can
+ *       lead to a situation that couldn't otherwise be produced, but
+ *       any such registrations will be dormant in the sense that they
+ *       will not be serviced, i.e. they will not receive any frames.
+ *
+ * Frame transmission allows userspace to send for example the required
+ * responses to action frames. It is subject to some sanity checking,
+ * but many frames can be transmitted. When a frame was transmitted, its
+ * status is indicated to the sending socket.
+ *
+ * For more technical details, see the corresponding command descriptions
+ * below.
+ */
+
 /**
  * enum nl80211_commands - supported nl80211 commands
  *
@@ -301,16 +338,18 @@
  *	rate selection. %NL80211_ATTR_IFINDEX is used to specify the interface
  *	and @NL80211_ATTR_TX_RATES the set of allowed rates.
  *
- * @NL80211_CMD_REGISTER_ACTION: Register for receiving certain action frames
- *	(via @NL80211_CMD_ACTION) for processing in userspace. This command
- *	requires an interface index and a match attribute containing the first
- *	few bytes of the frame that should match, e.g. a single byte for only
- *	a category match or four bytes for vendor frames including the OUI.
- *	The registration cannot be dropped, but is removed automatically
- *	when the netlink socket is closed. Multiple registrations can be made.
- * @NL80211_CMD_ACTION: Action frame TX request and RX notification. This
- *	command is used both as a request to transmit an Action frame and as an
- *	event indicating reception of an Action frame that was not processed in
+ * @NL80211_CMD_REGISTER_FRAME: Register for receiving certain mgmt frames
+ *	(via @NL80211_CMD_FRAME) for processing in userspace. This command
+ *	requires an interface index, a frame type attribute (optional for
+ *	backward compatibility reasons, if not given assumes action frames)
+ *	and a match attribute containing the first few bytes of the frame
+ *	that should match, e.g. a single byte for only a category match or
+ *	four bytes for vendor frames including the OUI. The registration
+ *	cannot be dropped, but is removed automatically when the netlink
+ *	socket is closed. Multiple registrations can be made.
+ * @NL80211_CMD_FRAME: Management frame TX request and RX notification. This
+ *	command is used both as a request to transmit a management frame and
+ *	as an event indicating reception of a frame that was not processed in
  *	kernel code, but is for us (i.e., which may need to be processed in a
  *	user space application). %NL80211_ATTR_FRAME is used to specify the
  *	frame contents (including header). %NL80211_ATTR_WIPHY_FREQ (and
@@ -320,8 +359,8 @@
  *	operational channel). When called, this operation returns a cookie
  *	(%NL80211_ATTR_COOKIE) that will be included with the TX status event
  *	pertaining to the TX request.
- * @NL80211_CMD_ACTION_TX_STATUS: Report TX status of an Action frame
- *	transmitted with %NL80211_CMD_ACTION. %NL80211_ATTR_COOKIE identifies
+ * @NL80211_CMD_FRAME_TX_STATUS: Report TX status of a management frame
+ *	transmitted with %NL80211_CMD_FRAME. %NL80211_ATTR_COOKIE identifies
  *	the TX command and %NL80211_ATTR_FRAME includes the contents of the
  *	frame. %NL80211_ATTR_ACK flag is included if the recipient acknowledged
  *	the frame.
@@ -429,9 +468,12 @@ enum nl80211_commands {
 
 	NL80211_CMD_SET_TX_BITRATE_MASK,
 
-	NL80211_CMD_REGISTER_ACTION,
-	NL80211_CMD_ACTION,
-	NL80211_CMD_ACTION_TX_STATUS,
+	NL80211_CMD_REGISTER_FRAME,
+	NL80211_CMD_REGISTER_ACTION = NL80211_CMD_REGISTER_FRAME,
+	NL80211_CMD_FRAME,
+	NL80211_CMD_ACTION = NL80211_CMD_FRAME,
+	NL80211_CMD_FRAME_TX_STATUS,
+	NL80211_CMD_ACTION_TX_STATUS = NL80211_CMD_FRAME_TX_STATUS,
 
 	NL80211_CMD_SET_POWER_SAVE,
 	NL80211_CMD_GET_POWER_SAVE,
@@ -708,7 +750,16 @@ enum nl80211_commands {
  *	is used with %NL80211_CMD_SET_TX_BITRATE_MASK.
  *
  * @NL80211_ATTR_FRAME_MATCH: A binary attribute which typically must contain
- *	at least one byte, currently used with @NL80211_CMD_REGISTER_ACTION.
+ *	at least one byte, currently used with @NL80211_CMD_REGISTER_FRAME.
+ * @NL80211_ATTR_FRAME_TYPE: A u16 indicating the frame type/subtype for the
+ *	@NL80211_CMD_REGISTER_FRAME command.
+ * @NL80211_ATTR_TX_FRAME_TYPES: wiphy capability attribute, which is a
+ *	nested attribute of %NL80211_ATTR_FRAME_TYPE attributes, containing
+ *	information about which frame types can be transmitted with
+ *	%NL80211_CMD_FRAME.
+ * @NL80211_ATTR_RX_FRAME_TYPES: wiphy capability attribute, which is a
+ *	nested attribute of %NL80211_ATTR_FRAME_TYPE attributes, containing
+ *	information about which frame types can be registered for RX.
  *
  * @NL80211_ATTR_ACK: Flag attribute indicating that the frame was
  *	acknowledged by the recipient.
@@ -891,6 +942,10 @@ enum nl80211_attrs {
 	NL80211_ATTR_WIPHY_TX_POWER_SETTING,
 	NL80211_ATTR_WIPHY_TX_POWER_LEVEL,
 
+	NL80211_ATTR_TX_FRAME_TYPES,
+	NL80211_ATTR_RX_FRAME_TYPES,
+	NL80211_ATTR_FRAME_TYPE,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
@@ -947,7 +1002,7 @@ enum nl80211_attrs {
  * @NL80211_IFTYPE_MONITOR: monitor interface receiving all frames
  * @NL80211_IFTYPE_MESH_POINT: mesh point
  * @NL80211_IFTYPE_MAX: highest interface type number currently defined
- * @__NL80211_IFTYPE_AFTER_LAST: internal use
+ * @NUM_NL80211_IFTYPES: number of defined interface types
  *
  * These values are used with the %NL80211_ATTR_IFTYPE
  * to set the type of an interface.
@@ -964,8 +1019,8 @@ enum nl80211_iftype {
 	NL80211_IFTYPE_MESH_POINT,
 
 	/* keep last */
-	__NL80211_IFTYPE_AFTER_LAST,
-	NL80211_IFTYPE_MAX = __NL80211_IFTYPE_AFTER_LAST - 1
+	NUM_NL80211_IFTYPES,
+	NL80211_IFTYPE_MAX = NUM_NL80211_IFTYPES - 1
 };
 
 /**
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 2b403c7ee32e..6a98b1b3bfde 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -1020,7 +1020,7 @@ struct cfg80211_pmksa {
  * @cancel_remain_on_channel: Cancel an on-going remain-on-channel operation.
  *	This allows the operation to be terminated prior to timeout based on
  *	the duration value.
- * @action: Transmit an action frame
+ * @mgmt_tx: Transmit a management frame
  *
  * @testmode_cmd: run a test mode command
  *
@@ -1172,7 +1172,7 @@ struct cfg80211_ops {
 					    struct net_device *dev,
 					    u64 cookie);
 
-	int	(*action)(struct wiphy *wiphy, struct net_device *dev,
+	int	(*mgmt_tx)(struct wiphy *wiphy, struct net_device *dev,
 			  struct ieee80211_channel *chan,
 			  enum nl80211_channel_type channel_type,
 			  bool channel_type_valid,
@@ -1236,6 +1236,10 @@ struct mac_address {
 	u8 addr[ETH_ALEN];
 };
 
+struct ieee80211_txrx_stypes {
+	u16 tx, rx;
+};
+
 /**
  * struct wiphy - wireless hardware description
  * @reg_notifier: the driver's regulatory notification callback
@@ -1286,6 +1290,10 @@ struct mac_address {
  * @privid: a pointer that drivers can use to identify if an arbitrary
  *	wiphy is theirs, e.g. in global notifiers
  * @bands: information about bands/channels supported by this device
+ *
+ * @mgmt_stypes: bitmasks of frame subtypes that can be subscribed to or
+ *	transmitted through nl80211, points to an array indexed by interface
+ *	type
  */
 struct wiphy {
 	/* assign these fields before you register the wiphy */
@@ -1294,9 +1302,12 @@ struct wiphy {
 	u8 perm_addr[ETH_ALEN];
 	u8 addr_mask[ETH_ALEN];
 
-	u16 n_addresses;
 	struct mac_address *addresses;
 
+	const struct ieee80211_txrx_stypes *mgmt_stypes;
+
+	u16 n_addresses;
+
 	/* Supported interface modes, OR together BIT(NL80211_IFTYPE_...) */
 	u16 interface_modes;
 
@@ -1492,8 +1503,8 @@ struct cfg80211_cached_keys;
  *	set by driver (if supported) on add_interface BEFORE registering the
  *	netdev and may otherwise be used by driver read-only, will be update
  *	by cfg80211 on change_interface
- * @action_registrations: list of registrations for action frames
- * @action_registrations_lock: lock for the list
+ * @mgmt_registrations: list of registrations for management frames
+ * @mgmt_registrations_lock: lock for the list
  * @mtx: mutex used to lock data in this struct
  * @cleanup_work: work struct used for cleanup that can't be done directly
  */
@@ -1505,8 +1516,8 @@ struct wireless_dev {
 	struct list_head list;
 	struct net_device *netdev;
 
-	struct list_head action_registrations;
-	spinlock_t action_registrations_lock;
+	struct list_head mgmt_registrations;
+	spinlock_t mgmt_registrations_lock;
 
 	struct mutex mtx;
 
@@ -2373,38 +2384,39 @@ void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr,
 		      struct station_info *sinfo, gfp_t gfp);
 
 /**
- * cfg80211_rx_action - notification of received, unprocessed Action frame
+ * cfg80211_rx_mgmt - notification of received, unprocessed management frame
  * @dev: network device
  * @freq: Frequency on which the frame was received in MHz
- * @buf: Action frame (header + body)
+ * @buf: Management frame (header + body)
  * @len: length of the frame data
  * @gfp: context flags
- * Returns %true if a user space application is responsible for rejecting the
- *	unrecognized Action frame; %false if no such application is registered
- *	(i.e., the driver is responsible for rejecting the unrecognized Action
- *	frame)
+ *
+ * Returns %true if a user space application has registered for this frame.
+ * For action frames, that makes it responsible for rejecting unrecognized
+ * action frames; %false otherwise, in which case for action frames the
+ * driver is responsible for rejecting the frame.
  *
  * This function is called whenever an Action frame is received for a station
  * mode interface, but is not processed in kernel.
  */
-bool cfg80211_rx_action(struct net_device *dev, int freq, const u8 *buf,
-			size_t len, gfp_t gfp);
+bool cfg80211_rx_mgmt(struct net_device *dev, int freq, const u8 *buf,
+		      size_t len, gfp_t gfp);
 
 /**
- * cfg80211_action_tx_status - notification of TX status for Action frame
+ * cfg80211_mgmt_tx_status - notification of TX status for management frame
  * @dev: network device
- * @cookie: Cookie returned by cfg80211_ops::action()
- * @buf: Action frame (header + body)
+ * @cookie: Cookie returned by cfg80211_ops::mgmt_tx()
+ * @buf: Management frame (header + body)
  * @len: length of the frame data
  * @ack: Whether frame was acknowledged
  * @gfp: context flags
  *
- * This function is called whenever an Action frame was requested to be
- * transmitted with cfg80211_ops::action() to report the TX status of the
+ * This function is called whenever a management frame was requested to be
+ * transmitted with cfg80211_ops::mgmt_tx() to report the TX status of the
  * transmission attempt.
  */
-void cfg80211_action_tx_status(struct net_device *dev, u64 cookie,
-			       const u8 *buf, size_t len, bool ack, gfp_t gfp);
+void cfg80211_mgmt_tx_status(struct net_device *dev, u64 cookie,
+			     const u8 *buf, size_t len, bool ack, gfp_t gfp);
 
 
 /**
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index f9a317766136..94787d21282c 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1521,11 +1521,11 @@ static int ieee80211_cancel_remain_on_channel(struct wiphy *wiphy,
 	return ieee80211_wk_cancel_remain_on_channel(sdata, cookie);
 }
 
-static int ieee80211_action(struct wiphy *wiphy, struct net_device *dev,
-			    struct ieee80211_channel *chan,
-			    enum nl80211_channel_type channel_type,
-			    bool channel_type_valid,
-			    const u8 *buf, size_t len, u64 *cookie)
+static int ieee80211_mgmt_tx(struct wiphy *wiphy, struct net_device *dev,
+			     struct ieee80211_channel *chan,
+			     enum nl80211_channel_type channel_type,
+			     bool channel_type_valid,
+			     const u8 *buf, size_t len, u64 *cookie)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct ieee80211_local *local = sdata->local;
@@ -1625,6 +1625,6 @@ struct cfg80211_ops mac80211_config_ops = {
 	.set_bitrate_mask = ieee80211_set_bitrate_mask,
 	.remain_on_channel = ieee80211_remain_on_channel,
 	.cancel_remain_on_channel = ieee80211_cancel_remain_on_channel,
-	.action = ieee80211_action,
+	.mgmt_tx = ieee80211_mgmt_tx,
 	.set_cqm_rssi_config = ieee80211_set_cqm_rssi_config,
 };
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 1bf05bfd149d..e73ae51dc036 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -170,6 +170,7 @@ typedef unsigned __bitwise__ ieee80211_rx_result;
 #define IEEE80211_RX_RA_MATCH		BIT(1)
 #define IEEE80211_RX_AMSDU		BIT(2)
 #define IEEE80211_RX_FRAGMENTED		BIT(3)
+#define IEEE80211_MALFORMED_ACTION_FRM	BIT(4)
 /* only add flags here that do not change with subframes of an aMPDU */
 
 struct ieee80211_rx_data {
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 9459aeee0ddc..86f434f234ae 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -177,7 +177,7 @@ static int ieee80211_open(struct net_device *dev)
 		/* no special treatment */
 		break;
 	case NL80211_IFTYPE_UNSPECIFIED:
-	case __NL80211_IFTYPE_AFTER_LAST:
+	case NUM_NL80211_IFTYPES:
 		/* cannot happen */
 		WARN_ON(1);
 		break;
@@ -634,7 +634,7 @@ static void ieee80211_teardown_sdata(struct net_device *dev)
 	case NL80211_IFTYPE_MONITOR:
 		break;
 	case NL80211_IFTYPE_UNSPECIFIED:
-	case __NL80211_IFTYPE_AFTER_LAST:
+	case NUM_NL80211_IFTYPES:
 		BUG();
 		break;
 	}
@@ -886,7 +886,7 @@ static void ieee80211_setup_sdata(struct ieee80211_sub_if_data *sdata,
 	case NL80211_IFTYPE_AP_VLAN:
 		break;
 	case NL80211_IFTYPE_UNSPECIFIED:
-	case __NL80211_IFTYPE_AFTER_LAST:
+	case NUM_NL80211_IFTYPES:
 		BUG();
 		break;
 	}
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 0afccda42a24..a53feac4618c 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -417,6 +417,41 @@ void ieee80211_napi_complete(struct ieee80211_hw *hw)
 }
 EXPORT_SYMBOL(ieee80211_napi_complete);
 
+/* There isn't a lot of sense in it, but you can transmit anything you like */
+static const struct ieee80211_txrx_stypes
+ieee80211_default_mgmt_stypes[NUM_NL80211_IFTYPES] = {
+	[NL80211_IFTYPE_ADHOC] = {
+		.tx = 0xffff,
+		.rx = BIT(IEEE80211_STYPE_ACTION >> 4),
+	},
+	[NL80211_IFTYPE_STATION] = {
+		.tx = 0xffff,
+		.rx = BIT(IEEE80211_STYPE_ACTION >> 4) |
+			BIT(IEEE80211_STYPE_PROBE_REQ >> 4),
+	},
+	[NL80211_IFTYPE_AP] = {
+		.tx = 0xffff,
+		.rx = BIT(IEEE80211_STYPE_ASSOC_REQ >> 4) |
+			BIT(IEEE80211_STYPE_REASSOC_REQ >> 4) |
+			BIT(IEEE80211_STYPE_PROBE_REQ >> 4) |
+			BIT(IEEE80211_STYPE_DISASSOC >> 4) |
+			BIT(IEEE80211_STYPE_AUTH >> 4) |
+			BIT(IEEE80211_STYPE_DEAUTH >> 4) |
+			BIT(IEEE80211_STYPE_ACTION >> 4),
+	},
+	[NL80211_IFTYPE_AP_VLAN] = {
+		/* copy AP */
+		.tx = 0xffff,
+		.rx = BIT(IEEE80211_STYPE_ASSOC_REQ >> 4) |
+			BIT(IEEE80211_STYPE_REASSOC_REQ >> 4) |
+			BIT(IEEE80211_STYPE_PROBE_REQ >> 4) |
+			BIT(IEEE80211_STYPE_DISASSOC >> 4) |
+			BIT(IEEE80211_STYPE_AUTH >> 4) |
+			BIT(IEEE80211_STYPE_DEAUTH >> 4) |
+			BIT(IEEE80211_STYPE_ACTION >> 4),
+	},
+};
+
 struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len,
 					const struct ieee80211_ops *ops)
 {
@@ -446,6 +481,8 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len,
 	if (!wiphy)
 		return NULL;
 
+	wiphy->mgmt_stypes = ieee80211_default_mgmt_stypes;
+
 	wiphy->flags |= WIPHY_FLAG_NETNS_OK |
 			WIPHY_FLAG_4ADDR_AP |
 			WIPHY_FLAG_4ADDR_STATION;
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 4fdbed58ca2f..aa41e382bbb3 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1939,14 +1939,37 @@ static void ieee80211_process_sa_query_req(struct ieee80211_sub_if_data *sdata,
 	ieee80211_tx_skb(sdata, skb);
 }
 
+static ieee80211_rx_result debug_noinline
+ieee80211_rx_h_mgmt_check(struct ieee80211_rx_data *rx)
+{
+	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *) rx->skb->data;
+
+	/*
+	 * From here on, look only at management frames.
+	 * Data and control frames are already handled,
+	 * and unknown (reserved) frames are useless.
+	 */
+	if (rx->skb->len < 24)
+		return RX_DROP_MONITOR;
+
+	if (!ieee80211_is_mgmt(mgmt->frame_control))
+		return RX_DROP_MONITOR;
+
+	if (!(rx->flags & IEEE80211_RX_RA_MATCH))
+		return RX_DROP_MONITOR;
+
+	if (ieee80211_drop_unencrypted_mgmt(rx))
+		return RX_DROP_UNUSABLE;
+
+	return RX_CONTINUE;
+}
+
 static ieee80211_rx_result debug_noinline
 ieee80211_rx_h_action(struct ieee80211_rx_data *rx)
 {
 	struct ieee80211_local *local = rx->local;
 	struct ieee80211_sub_if_data *sdata = rx->sdata;
 	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *) rx->skb->data;
-	struct sk_buff *nskb;
-	struct ieee80211_rx_status *status;
 	int len = rx->skb->len;
 
 	if (!ieee80211_is_action(mgmt->frame_control))
@@ -1962,9 +1985,6 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx)
 	if (!(rx->flags & IEEE80211_RX_RA_MATCH))
 		return RX_DROP_UNUSABLE;
 
-	if (ieee80211_drop_unencrypted_mgmt(rx))
-		return RX_DROP_UNUSABLE;
-
 	switch (mgmt->u.action.category) {
 	case WLAN_CATEGORY_BACK:
 		/*
@@ -2055,17 +2075,36 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx)
 		goto queue;
 	}
 
+	return RX_CONTINUE;
+
  invalid:
-	/*
-	 * For AP mode, hostapd is responsible for handling any action
-	 * frames that we didn't handle, including returning unknown
-	 * ones. For all other modes we will return them to the sender,
-	 * setting the 0x80 bit in the action category, as required by
-	 * 802.11-2007 7.3.1.11.
-	 */
-	if (sdata->vif.type == NL80211_IFTYPE_AP ||
-	    sdata->vif.type == NL80211_IFTYPE_AP_VLAN)
-		return RX_DROP_MONITOR;
+	rx->flags |= IEEE80211_MALFORMED_ACTION_FRM;
+	/* will return in the next handlers */
+	return RX_CONTINUE;
+
+ handled:
+	if (rx->sta)
+		rx->sta->rx_packets++;
+	dev_kfree_skb(rx->skb);
+	return RX_QUEUED;
+
+ queue:
+	rx->skb->pkt_type = IEEE80211_SDATA_QUEUE_TYPE_FRAME;
+	skb_queue_tail(&sdata->skb_queue, rx->skb);
+	ieee80211_queue_work(&local->hw, &sdata->work);
+	if (rx->sta)
+		rx->sta->rx_packets++;
+	return RX_QUEUED;
+}
+
+static ieee80211_rx_result debug_noinline
+ieee80211_rx_h_userspace_mgmt(struct ieee80211_rx_data *rx)
+{
+	struct ieee80211_rx_status *status;
+
+	/* skip known-bad action frames and return them in the next handler */
+	if (rx->flags & IEEE80211_MALFORMED_ACTION_FRM)
+		return RX_CONTINUE;
 
 	/*
 	 * Getting here means the kernel doesn't know how to handle
@@ -2075,10 +2114,44 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx)
 	 */
 	status = IEEE80211_SKB_RXCB(rx->skb);
 
-	if (cfg80211_rx_action(rx->sdata->dev, status->freq,
-			       rx->skb->data, rx->skb->len,
-			       GFP_ATOMIC))
-		goto handled;
+	if (cfg80211_rx_mgmt(rx->sdata->dev, status->freq,
+			     rx->skb->data, rx->skb->len,
+			     GFP_ATOMIC)) {
+		if (rx->sta)
+			rx->sta->rx_packets++;
+		dev_kfree_skb(rx->skb);
+		return RX_QUEUED;
+	}
+
+
+	return RX_CONTINUE;
+}
+
+static ieee80211_rx_result debug_noinline
+ieee80211_rx_h_action_return(struct ieee80211_rx_data *rx)
+{
+	struct ieee80211_local *local = rx->local;
+	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *) rx->skb->data;
+	struct sk_buff *nskb;
+	struct ieee80211_sub_if_data *sdata = rx->sdata;
+
+	if (!ieee80211_is_action(mgmt->frame_control))
+		return RX_CONTINUE;
+
+	/*
+	 * For AP mode, hostapd is responsible for handling any action
+	 * frames that we didn't handle, including returning unknown
+	 * ones. For all other modes we will return them to the sender,
+	 * setting the 0x80 bit in the action category, as required by
+	 * 802.11-2007 7.3.1.11.
+	 * Newer versions of hostapd shall also use the management frame
+	 * registration mechanisms, but older ones still use cooked
+	 * monitor interfaces so push all frames there.
+	 */
+	if (!(rx->flags & IEEE80211_MALFORMED_ACTION_FRM) &&
+	    (sdata->vif.type == NL80211_IFTYPE_AP ||
+	     sdata->vif.type == NL80211_IFTYPE_AP_VLAN))
+		return RX_DROP_MONITOR;
 
 	/* do not return rejected action frames */
 	if (mgmt->u.action.category & 0x80)
@@ -2097,20 +2170,8 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx)
 
 		ieee80211_tx_skb(rx->sdata, nskb);
 	}
-
- handled:
-	if (rx->sta)
-		rx->sta->rx_packets++;
 	dev_kfree_skb(rx->skb);
 	return RX_QUEUED;
-
- queue:
-	rx->skb->pkt_type = IEEE80211_SDATA_QUEUE_TYPE_FRAME;
-	skb_queue_tail(&sdata->skb_queue, rx->skb);
-	ieee80211_queue_work(&local->hw, &sdata->work);
-	if (rx->sta)
-		rx->sta->rx_packets++;
-	return RX_QUEUED;
 }
 
 static ieee80211_rx_result debug_noinline
@@ -2121,15 +2182,6 @@ ieee80211_rx_h_mgmt(struct ieee80211_rx_data *rx)
 	struct ieee80211_mgmt *mgmt = (void *)rx->skb->data;
 	__le16 stype;
 
-	if (!(rx->flags & IEEE80211_RX_RA_MATCH))
-		return RX_DROP_MONITOR;
-
-	if (rx->skb->len < 24)
-		return RX_DROP_MONITOR;
-
-	if (ieee80211_drop_unencrypted_mgmt(rx))
-		return RX_DROP_UNUSABLE;
-
 	rxs = ieee80211_work_rx_mgmt(rx->sdata, rx->skb);
 	if (rxs != RX_CONTINUE)
 		return rxs;
@@ -2374,7 +2426,10 @@ static void ieee80211_rx_handlers(struct ieee80211_rx_data *rx,
 		if (res != RX_CONTINUE)
 			goto rxh_next;
 
+		CALL_RXH(ieee80211_rx_h_mgmt_check)
 		CALL_RXH(ieee80211_rx_h_action)
+		CALL_RXH(ieee80211_rx_h_userspace_mgmt)
+		CALL_RXH(ieee80211_rx_h_action_return)
 		CALL_RXH(ieee80211_rx_h_mgmt)
 
  rxh_next:
@@ -2527,7 +2582,7 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata,
 		break;
 	case NL80211_IFTYPE_MONITOR:
 	case NL80211_IFTYPE_UNSPECIFIED:
-	case __NL80211_IFTYPE_AFTER_LAST:
+	case NUM_NL80211_IFTYPES:
 		/* should never get here */
 		WARN_ON(1);
 		break;
diff --git a/net/mac80211/status.c b/net/mac80211/status.c
index 10caec5ea8fa..67a35841bef0 100644
--- a/net/mac80211/status.c
+++ b/net/mac80211/status.c
@@ -296,7 +296,7 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb)
 	}
 
 	if (info->flags & IEEE80211_TX_INTFL_NL80211_FRAME_TX)
-		cfg80211_action_tx_status(
+		cfg80211_mgmt_tx_status(
 			skb->dev, (unsigned long) skb, skb->data, skb->len,
 			!!(info->flags & IEEE80211_TX_STAT_ACK), GFP_ATOMIC);
 
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 748387d45bc0..cd2b485fed4f 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -471,7 +471,7 @@ void ieee80211_iterate_active_interfaces(
 
 	list_for_each_entry(sdata, &local->interfaces, list) {
 		switch (sdata->vif.type) {
-		case __NL80211_IFTYPE_AFTER_LAST:
+		case NUM_NL80211_IFTYPES:
 		case NL80211_IFTYPE_UNSPECIFIED:
 		case NL80211_IFTYPE_MONITOR:
 		case NL80211_IFTYPE_AP_VLAN:
@@ -505,7 +505,7 @@ void ieee80211_iterate_active_interfaces_atomic(
 
 	list_for_each_entry_rcu(sdata, &local->interfaces, list) {
 		switch (sdata->vif.type) {
-		case __NL80211_IFTYPE_AFTER_LAST:
+		case NUM_NL80211_IFTYPES:
 		case NL80211_IFTYPE_UNSPECIFIED:
 		case NL80211_IFTYPE_MONITOR:
 		case NL80211_IFTYPE_AP_VLAN:
@@ -1189,7 +1189,7 @@ int ieee80211_reconfig(struct ieee80211_local *local)
 			/* ignore virtual */
 			break;
 		case NL80211_IFTYPE_UNSPECIFIED:
-		case __NL80211_IFTYPE_AFTER_LAST:
+		case NUM_NL80211_IFTYPES:
 			WARN_ON(1);
 			break;
 		}
diff --git a/net/wireless/core.c b/net/wireless/core.c
index c70909c3eae4..d52630bbab04 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -433,7 +433,7 @@ int wiphy_register(struct wiphy *wiphy)
 
 	/* sanity check ifmodes */
 	WARN_ON(!ifmodes);
-	ifmodes &= ((1 << __NL80211_IFTYPE_AFTER_LAST) - 1) & ~1;
+	ifmodes &= ((1 << NUM_NL80211_IFTYPES) - 1) & ~1;
 	if (WARN_ON(ifmodes != wiphy->interface_modes))
 		wiphy->interface_modes = ifmodes;
 
@@ -685,8 +685,8 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb,
 		INIT_WORK(&wdev->cleanup_work, wdev_cleanup_work);
 		INIT_LIST_HEAD(&wdev->event_list);
 		spin_lock_init(&wdev->event_lock);
-		INIT_LIST_HEAD(&wdev->action_registrations);
-		spin_lock_init(&wdev->action_registrations_lock);
+		INIT_LIST_HEAD(&wdev->mgmt_registrations);
+		spin_lock_init(&wdev->mgmt_registrations_lock);
 
 		mutex_lock(&rdev->devlist_mtx);
 		list_add_rcu(&wdev->list, &rdev->netdev_list);
@@ -806,7 +806,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb,
 			sysfs_remove_link(&dev->dev.kobj, "phy80211");
 			list_del_rcu(&wdev->list);
 			rdev->devlist_generation++;
-			cfg80211_mlme_purge_actions(wdev);
+			cfg80211_mlme_purge_registrations(wdev);
 #ifdef CONFIG_CFG80211_WEXT
 			kfree(wdev->wext.keys);
 #endif
diff --git a/net/wireless/core.h b/net/wireless/core.h
index 63d57ae399c3..58ab2c791d28 100644
--- a/net/wireless/core.h
+++ b/net/wireless/core.h
@@ -331,16 +331,17 @@ void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid,
 			       const u8 *resp_ie, size_t resp_ie_len,
 			       u16 status, bool wextev,
 			       struct cfg80211_bss *bss);
-int cfg80211_mlme_register_action(struct wireless_dev *wdev, u32 snd_pid,
-				  const u8 *match_data, int match_len);
-void cfg80211_mlme_unregister_actions(struct wireless_dev *wdev, u32 nlpid);
-void cfg80211_mlme_purge_actions(struct wireless_dev *wdev);
-int cfg80211_mlme_action(struct cfg80211_registered_device *rdev,
-			 struct net_device *dev,
-			 struct ieee80211_channel *chan,
-			 enum nl80211_channel_type channel_type,
-			 bool channel_type_valid,
-			 const u8 *buf, size_t len, u64 *cookie);
+int cfg80211_mlme_register_mgmt(struct wireless_dev *wdev, u32 snd_pid,
+				u16 frame_type, const u8 *match_data,
+				int match_len);
+void cfg80211_mlme_unregister_socket(struct wireless_dev *wdev, u32 nlpid);
+void cfg80211_mlme_purge_registrations(struct wireless_dev *wdev);
+int cfg80211_mlme_mgmt_tx(struct cfg80211_registered_device *rdev,
+			  struct net_device *dev,
+			  struct ieee80211_channel *chan,
+			  enum nl80211_channel_type channel_type,
+			  bool channel_type_valid,
+			  const u8 *buf, size_t len, u64 *cookie);
 
 /* SME */
 int __cfg80211_connect(struct cfg80211_registered_device *rdev,
diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c
index ee0af32ed59e..8515b1e5c578 100644
--- a/net/wireless/mlme.c
+++ b/net/wireless/mlme.c
@@ -748,31 +748,51 @@ void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr,
 }
 EXPORT_SYMBOL(cfg80211_new_sta);
 
-struct cfg80211_action_registration {
+struct cfg80211_mgmt_registration {
 	struct list_head list;
 
 	u32 nlpid;
 
 	int match_len;
 
+	__le16 frame_type;
+
 	u8 match[];
 };
 
-int cfg80211_mlme_register_action(struct wireless_dev *wdev, u32 snd_pid,
-				  const u8 *match_data, int match_len)
+int cfg80211_mlme_register_mgmt(struct wireless_dev *wdev, u32 snd_pid,
+				u16 frame_type, const u8 *match_data,
+				int match_len)
 {
-	struct cfg80211_action_registration *reg, *nreg;
+	struct cfg80211_mgmt_registration *reg, *nreg;
 	int err = 0;
+	u16 mgmt_type;
+
+	if (!wdev->wiphy->mgmt_stypes)
+		return -EOPNOTSUPP;
+
+	if ((frame_type & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_MGMT)
+		return -EINVAL;
+
+	if (frame_type & ~(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE))
+		return -EINVAL;
+
+	mgmt_type = (frame_type & IEEE80211_FCTL_STYPE) >> 4;
+	if (!(wdev->wiphy->mgmt_stypes[wdev->iftype].rx & BIT(mgmt_type)))
+		return -EINVAL;
 
 	nreg = kzalloc(sizeof(*reg) + match_len, GFP_KERNEL);
 	if (!nreg)
 		return -ENOMEM;
 
-	spin_lock_bh(&wdev->action_registrations_lock);
+	spin_lock_bh(&wdev->mgmt_registrations_lock);
 
-	list_for_each_entry(reg, &wdev->action_registrations, list) {
+	list_for_each_entry(reg, &wdev->mgmt_registrations, list) {
 		int mlen = min(match_len, reg->match_len);
 
+		if (frame_type != le16_to_cpu(reg->frame_type))
+			continue;
+
 		if (memcmp(reg->match, match_data, mlen) == 0) {
 			err = -EALREADY;
 			break;
@@ -787,62 +807,75 @@ int cfg80211_mlme_register_action(struct wireless_dev *wdev, u32 snd_pid,
 	memcpy(nreg->match, match_data, match_len);
 	nreg->match_len = match_len;
 	nreg->nlpid = snd_pid;
-	list_add(&nreg->list, &wdev->action_registrations);
+	nreg->frame_type = cpu_to_le16(frame_type);
+	list_add(&nreg->list, &wdev->mgmt_registrations);
 
  out:
-	spin_unlock_bh(&wdev->action_registrations_lock);
+	spin_unlock_bh(&wdev->mgmt_registrations_lock);
 	return err;
 }
 
-void cfg80211_mlme_unregister_actions(struct wireless_dev *wdev, u32 nlpid)
+void cfg80211_mlme_unregister_socket(struct wireless_dev *wdev, u32 nlpid)
 {
-	struct cfg80211_action_registration *reg, *tmp;
+	struct cfg80211_mgmt_registration *reg, *tmp;
 
-	spin_lock_bh(&wdev->action_registrations_lock);
+	spin_lock_bh(&wdev->mgmt_registrations_lock);
 
-	list_for_each_entry_safe(reg, tmp, &wdev->action_registrations, list) {
+	list_for_each_entry_safe(reg, tmp, &wdev->mgmt_registrations, list) {
 		if (reg->nlpid == nlpid) {
 			list_del(&reg->list);
 			kfree(reg);
 		}
 	}
 
-	spin_unlock_bh(&wdev->action_registrations_lock);
+	spin_unlock_bh(&wdev->mgmt_registrations_lock);
 }
 
-void cfg80211_mlme_purge_actions(struct wireless_dev *wdev)
+void cfg80211_mlme_purge_registrations(struct wireless_dev *wdev)
 {
-	struct cfg80211_action_registration *reg, *tmp;
+	struct cfg80211_mgmt_registration *reg, *tmp;
 
-	spin_lock_bh(&wdev->action_registrations_lock);
+	spin_lock_bh(&wdev->mgmt_registrations_lock);
 
-	list_for_each_entry_safe(reg, tmp, &wdev->action_registrations, list) {
+	list_for_each_entry_safe(reg, tmp, &wdev->mgmt_registrations, list) {
 		list_del(&reg->list);
 		kfree(reg);
 	}
 
-	spin_unlock_bh(&wdev->action_registrations_lock);
+	spin_unlock_bh(&wdev->mgmt_registrations_lock);
 }
 
-int cfg80211_mlme_action(struct cfg80211_registered_device *rdev,
-			 struct net_device *dev,
-			 struct ieee80211_channel *chan,
-			 enum nl80211_channel_type channel_type,
-			 bool channel_type_valid,
-			 const u8 *buf, size_t len, u64 *cookie)
+int cfg80211_mlme_mgmt_tx(struct cfg80211_registered_device *rdev,
+			  struct net_device *dev,
+			  struct ieee80211_channel *chan,
+			  enum nl80211_channel_type channel_type,
+			  bool channel_type_valid,
+			  const u8 *buf, size_t len, u64 *cookie)
 {
 	struct wireless_dev *wdev = dev->ieee80211_ptr;
 	const struct ieee80211_mgmt *mgmt;
+	u16 stype;
+
+	if (!wdev->wiphy->mgmt_stypes)
+		return -EOPNOTSUPP;
 
-	if (rdev->ops->action == NULL)
+	if (!rdev->ops->mgmt_tx)
 		return -EOPNOTSUPP;
+
 	if (len < 24 + 1)
 		return -EINVAL;
 
 	mgmt = (const struct ieee80211_mgmt *) buf;
-	if (!ieee80211_is_action(mgmt->frame_control))
+
+	if (!ieee80211_is_mgmt(mgmt->frame_control))
 		return -EINVAL;
-	if (mgmt->u.action.category != WLAN_CATEGORY_PUBLIC) {
+
+	stype = le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_STYPE;
+	if (!(wdev->wiphy->mgmt_stypes[wdev->iftype].tx & BIT(stype >> 4)))
+		return -EINVAL;
+
+	if (ieee80211_is_action(mgmt->frame_control) &&
+	    mgmt->u.action.category != WLAN_CATEGORY_PUBLIC) {
 		/* Verify that we are associated with the destination AP */
 		wdev_lock(wdev);
 
@@ -863,64 +896,75 @@ int cfg80211_mlme_action(struct cfg80211_registered_device *rdev,
 		return -EINVAL;
 
 	/* Transmit the Action frame as requested by user space */
-	return rdev->ops->action(&rdev->wiphy, dev, chan, channel_type,
-				 channel_type_valid, buf, len, cookie);
+	return rdev->ops->mgmt_tx(&rdev->wiphy, dev, chan, channel_type,
+				  channel_type_valid, buf, len, cookie);
 }
 
-bool cfg80211_rx_action(struct net_device *dev, int freq, const u8 *buf,
-			size_t len, gfp_t gfp)
+bool cfg80211_rx_mgmt(struct net_device *dev, int freq, const u8 *buf,
+		      size_t len, gfp_t gfp)
 {
 	struct wireless_dev *wdev = dev->ieee80211_ptr;
 	struct wiphy *wiphy = wdev->wiphy;
 	struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy);
-	struct cfg80211_action_registration *reg;
-	const u8 *action_data;
-	int action_data_len;
+	struct cfg80211_mgmt_registration *reg;
+	const struct ieee80211_txrx_stypes *stypes =
+		&wiphy->mgmt_stypes[wdev->iftype];
+	struct ieee80211_mgmt *mgmt = (void *)buf;
+	const u8 *data;
+	int data_len;
 	bool result = false;
+	__le16 ftype = mgmt->frame_control &
+		cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE);
+	u16 stype;
 
-	/* frame length - min size excluding category */
-	action_data_len = len - (IEEE80211_MIN_ACTION_SIZE - 1);
+	stype = (le16_to_cpu(mgmt->frame_control) & IEEE80211_FCTL_STYPE) >> 4;
 
-	/* action data starts with category */
-	action_data = buf + IEEE80211_MIN_ACTION_SIZE - 1;
+	if (!(stypes->rx & BIT(stype)))
+		return false;
 
-	spin_lock_bh(&wdev->action_registrations_lock);
+	data = buf + ieee80211_hdrlen(mgmt->frame_control);
+	data_len = len - ieee80211_hdrlen(mgmt->frame_control);
+
+	spin_lock_bh(&wdev->mgmt_registrations_lock);
+
+	list_for_each_entry(reg, &wdev->mgmt_registrations, list) {
+		if (reg->frame_type != ftype)
+			continue;
 
-	list_for_each_entry(reg, &wdev->action_registrations, list) {
-		if (reg->match_len > action_data_len)
+		if (reg->match_len > data_len)
 			continue;
 
-		if (memcmp(reg->match, action_data, reg->match_len))
+		if (memcmp(reg->match, data, reg->match_len))
 			continue;
 
 		/* found match! */
 
 		/* Indicate the received Action frame to user space */
-		if (nl80211_send_action(rdev, dev, reg->nlpid, freq,
-					buf, len, gfp))
+		if (nl80211_send_mgmt(rdev, dev, reg->nlpid, freq,
+				      buf, len, gfp))
 			continue;
 
 		result = true;
 		break;
 	}
 
-	spin_unlock_bh(&wdev->action_registrations_lock);
+	spin_unlock_bh(&wdev->mgmt_registrations_lock);
 
 	return result;
 }
-EXPORT_SYMBOL(cfg80211_rx_action);
+EXPORT_SYMBOL(cfg80211_rx_mgmt);
 
-void cfg80211_action_tx_status(struct net_device *dev, u64 cookie,
-			       const u8 *buf, size_t len, bool ack, gfp_t gfp)
+void cfg80211_mgmt_tx_status(struct net_device *dev, u64 cookie,
+			     const u8 *buf, size_t len, bool ack, gfp_t gfp)
 {
 	struct wireless_dev *wdev = dev->ieee80211_ptr;
 	struct wiphy *wiphy = wdev->wiphy;
 	struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy);
 
 	/* Indicate TX status of the Action frame to user space */
-	nl80211_send_action_tx_status(rdev, dev, cookie, buf, len, ack, gfp);
+	nl80211_send_mgmt_tx_status(rdev, dev, cookie, buf, len, ack, gfp);
 }
-EXPORT_SYMBOL(cfg80211_action_tx_status);
+EXPORT_SYMBOL(cfg80211_mgmt_tx_status);
 
 void cfg80211_cqm_rssi_notify(struct net_device *dev,
 			      enum nl80211_cqm_rssi_threshold_event rssi_event,
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index bb5b78eebeb2..927ffbd2aebc 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -156,6 +156,7 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = {
 
 	[NL80211_ATTR_WIPHY_TX_POWER_SETTING] = { .type = NLA_U32 },
 	[NL80211_ATTR_WIPHY_TX_POWER_LEVEL] = { .type = NLA_U32 },
+	[NL80211_ATTR_FRAME_TYPE] = { .type = NLA_U16 },
 };
 
 /* policy for the attributes */
@@ -437,6 +438,8 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 	struct ieee80211_rate *rate;
 	int i;
 	u16 ifmodes = dev->wiphy.interface_modes;
+	const struct ieee80211_txrx_stypes *mgmt_stypes =
+				dev->wiphy.mgmt_stypes;
 
 	hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_WIPHY);
 	if (!hdr)
@@ -587,7 +590,7 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 	CMD(flush_pmksa, FLUSH_PMKSA);
 	CMD(remain_on_channel, REMAIN_ON_CHANNEL);
 	CMD(set_bitrate_mask, SET_TX_BITRATE_MASK);
-	CMD(action, ACTION);
+	CMD(mgmt_tx, FRAME);
 	if (dev->wiphy.flags & WIPHY_FLAG_NETNS_OK) {
 		i++;
 		NLA_PUT_U32(msg, i, NL80211_CMD_SET_WIPHY_NETNS);
@@ -608,6 +611,53 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 
 	nla_nest_end(msg, nl_cmds);
 
+	if (mgmt_stypes) {
+		u16 stypes;
+		struct nlattr *nl_ftypes, *nl_ifs;
+		enum nl80211_iftype ift;
+
+		nl_ifs = nla_nest_start(msg, NL80211_ATTR_TX_FRAME_TYPES);
+		if (!nl_ifs)
+			goto nla_put_failure;
+
+		for (ift = 0; ift < NUM_NL80211_IFTYPES; ift++) {
+			nl_ftypes = nla_nest_start(msg, ift);
+			if (!nl_ftypes)
+				goto nla_put_failure;
+			i = 0;
+			stypes = mgmt_stypes[ift].tx;
+			while (stypes) {
+				if (stypes & 1)
+					NLA_PUT_U16(msg, NL80211_ATTR_FRAME_TYPE,
+						    (i << 4) | IEEE80211_FTYPE_MGMT);
+				stypes >>= 1;
+				i++;
+			}
+			nla_nest_end(msg, nl_ftypes);
+		}
+
+		nl_ifs = nla_nest_start(msg, NL80211_ATTR_RX_FRAME_TYPES);
+		if (!nl_ifs)
+			goto nla_put_failure;
+
+		for (ift = 0; ift < NUM_NL80211_IFTYPES; ift++) {
+			nl_ftypes = nla_nest_start(msg, ift);
+			if (!nl_ftypes)
+				goto nla_put_failure;
+			i = 0;
+			stypes = mgmt_stypes[ift].rx;
+			while (stypes) {
+				if (stypes & 1)
+					NLA_PUT_U16(msg, NL80211_ATTR_FRAME_TYPE,
+						    (i << 4) | IEEE80211_FTYPE_MGMT);
+				stypes >>= 1;
+				i++;
+			}
+			nla_nest_end(msg, nl_ftypes);
+		}
+		nla_nest_end(msg, nl_ifs);
+	}
+
 	return genlmsg_end(msg, hdr);
 
  nla_put_failure:
@@ -4732,17 +4782,18 @@ static int nl80211_set_tx_bitrate_mask(struct sk_buff *skb,
 	return err;
 }
 
-static int nl80211_register_action(struct sk_buff *skb, struct genl_info *info)
+static int nl80211_register_mgmt(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *rdev;
 	struct net_device *dev;
+	u16 frame_type = IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ACTION;
 	int err;
 
 	if (!info->attrs[NL80211_ATTR_FRAME_MATCH])
 		return -EINVAL;
 
-	if (nla_len(info->attrs[NL80211_ATTR_FRAME_MATCH]) < 1)
-		return -EINVAL;
+	if (info->attrs[NL80211_ATTR_FRAME_TYPE])
+		frame_type = nla_get_u16(info->attrs[NL80211_ATTR_FRAME_TYPE]);
 
 	rtnl_lock();
 
@@ -4757,12 +4808,13 @@ static int nl80211_register_action(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	/* not much point in registering if we can't reply */
-	if (!rdev->ops->action) {
+	if (!rdev->ops->mgmt_tx) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
 
-	err = cfg80211_mlme_register_action(dev->ieee80211_ptr, info->snd_pid,
+	err = cfg80211_mlme_register_mgmt(dev->ieee80211_ptr, info->snd_pid,
+			frame_type,
 			nla_data(info->attrs[NL80211_ATTR_FRAME_MATCH]),
 			nla_len(info->attrs[NL80211_ATTR_FRAME_MATCH]));
  out:
@@ -4773,7 +4825,7 @@ static int nl80211_register_action(struct sk_buff *skb, struct genl_info *info)
 	return err;
 }
 
-static int nl80211_action(struct sk_buff *skb, struct genl_info *info)
+static int nl80211_tx_mgmt(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *rdev;
 	struct net_device *dev;
@@ -4796,7 +4848,7 @@ static int nl80211_action(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto unlock_rtnl;
 
-	if (!rdev->ops->action) {
+	if (!rdev->ops->mgmt_tx) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -4839,17 +4891,17 @@ static int nl80211_action(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	hdr = nl80211hdr_put(msg, info->snd_pid, info->snd_seq, 0,
-			     NL80211_CMD_ACTION);
+			     NL80211_CMD_FRAME);
 
 	if (IS_ERR(hdr)) {
 		err = PTR_ERR(hdr);
 		goto free_msg;
 	}
-	err = cfg80211_mlme_action(rdev, dev, chan, channel_type,
-				   channel_type_valid,
-				   nla_data(info->attrs[NL80211_ATTR_FRAME]),
-				   nla_len(info->attrs[NL80211_ATTR_FRAME]),
-				   &cookie);
+	err = cfg80211_mlme_mgmt_tx(rdev, dev, chan, channel_type,
+				    channel_type_valid,
+				    nla_data(info->attrs[NL80211_ATTR_FRAME]),
+				    nla_len(info->attrs[NL80211_ATTR_FRAME]),
+				    &cookie);
 	if (err)
 		goto free_msg;
 
@@ -5348,14 +5400,14 @@ static struct genl_ops nl80211_ops[] = {
 		.flags = GENL_ADMIN_PERM,
 	},
 	{
-		.cmd = NL80211_CMD_REGISTER_ACTION,
-		.doit = nl80211_register_action,
+		.cmd = NL80211_CMD_REGISTER_FRAME,
+		.doit = nl80211_register_mgmt,
 		.policy = nl80211_policy,
 		.flags = GENL_ADMIN_PERM,
 	},
 	{
-		.cmd = NL80211_CMD_ACTION,
-		.doit = nl80211_action,
+		.cmd = NL80211_CMD_FRAME,
+		.doit = nl80211_tx_mgmt,
 		.policy = nl80211_policy,
 		.flags = GENL_ADMIN_PERM,
 	},
@@ -6055,9 +6107,9 @@ void nl80211_send_sta_event(struct cfg80211_registered_device *rdev,
 				nl80211_mlme_mcgrp.id, gfp);
 }
 
-int nl80211_send_action(struct cfg80211_registered_device *rdev,
-			struct net_device *netdev, u32 nlpid,
-			int freq, const u8 *buf, size_t len, gfp_t gfp)
+int nl80211_send_mgmt(struct cfg80211_registered_device *rdev,
+		      struct net_device *netdev, u32 nlpid,
+		      int freq, const u8 *buf, size_t len, gfp_t gfp)
 {
 	struct sk_buff *msg;
 	void *hdr;
@@ -6067,7 +6119,7 @@ int nl80211_send_action(struct cfg80211_registered_device *rdev,
 	if (!msg)
 		return -ENOMEM;
 
-	hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_ACTION);
+	hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_FRAME);
 	if (!hdr) {
 		nlmsg_free(msg);
 		return -ENOMEM;
@@ -6095,10 +6147,10 @@ int nl80211_send_action(struct cfg80211_registered_device *rdev,
 	return -ENOBUFS;
 }
 
-void nl80211_send_action_tx_status(struct cfg80211_registered_device *rdev,
-				   struct net_device *netdev, u64 cookie,
-				   const u8 *buf, size_t len, bool ack,
-				   gfp_t gfp)
+void nl80211_send_mgmt_tx_status(struct cfg80211_registered_device *rdev,
+				 struct net_device *netdev, u64 cookie,
+				 const u8 *buf, size_t len, bool ack,
+				 gfp_t gfp)
 {
 	struct sk_buff *msg;
 	void *hdr;
@@ -6107,7 +6159,7 @@ void nl80211_send_action_tx_status(struct cfg80211_registered_device *rdev,
 	if (!msg)
 		return;
 
-	hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_ACTION_TX_STATUS);
+	hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_FRAME_TX_STATUS);
 	if (!hdr) {
 		nlmsg_free(msg);
 		return;
@@ -6194,7 +6246,7 @@ static int nl80211_netlink_notify(struct notifier_block * nb,
 
 	list_for_each_entry_rcu(rdev, &cfg80211_rdev_list, list)
 		list_for_each_entry_rcu(wdev, &rdev->netdev_list, list)
-			cfg80211_mlme_unregister_actions(wdev, notify->pid);
+			cfg80211_mlme_unregister_socket(wdev, notify->pid);
 
 	rcu_read_unlock();
 
diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h
index 2ad7fbc7d9f1..30d2f939150d 100644
--- a/net/wireless/nl80211.h
+++ b/net/wireless/nl80211.h
@@ -74,13 +74,13 @@ void nl80211_send_sta_event(struct cfg80211_registered_device *rdev,
 			    struct net_device *dev, const u8 *mac_addr,
 			    struct station_info *sinfo, gfp_t gfp);
 
-int nl80211_send_action(struct cfg80211_registered_device *rdev,
-			struct net_device *netdev, u32 nlpid, int freq,
-			const u8 *buf, size_t len, gfp_t gfp);
-void nl80211_send_action_tx_status(struct cfg80211_registered_device *rdev,
-				   struct net_device *netdev, u64 cookie,
-				   const u8 *buf, size_t len, bool ack,
-				   gfp_t gfp);
+int nl80211_send_mgmt(struct cfg80211_registered_device *rdev,
+		      struct net_device *netdev, u32 nlpid, int freq,
+		      const u8 *buf, size_t len, gfp_t gfp);
+void nl80211_send_mgmt_tx_status(struct cfg80211_registered_device *rdev,
+				 struct net_device *netdev, u64 cookie,
+				 const u8 *buf, size_t len, bool ack,
+				 gfp_t gfp);
 
 void
 nl80211_send_cqm_rssi_notify(struct cfg80211_registered_device *rdev,
diff --git a/net/wireless/util.c b/net/wireless/util.c
index 1eb24162be61..8d961cc4ae98 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -823,7 +823,7 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev,
 			/* monitor can't bridge anyway */
 			break;
 		case NL80211_IFTYPE_UNSPECIFIED:
-		case __NL80211_IFTYPE_AFTER_LAST:
+		case NUM_NL80211_IFTYPES:
 			/* not happening */
 			break;
 		}
-- 
cgit v1.2.3


From d2730b2a6a019d14455556019d744ab051e6554b Mon Sep 17 00:00:00 2001
From: Gábor Stefanik <netrolller.3d@gmail.com>
Date: Mon, 16 Aug 2010 22:39:16 +0200
Subject: b43: N-PHY: Implement MAC PHY clock set
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Gábor Stefanik <netrolller.3d@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/b43/phy_n.c | 13 ++++++++++++-
 include/linux/ssb/ssb_regs.h     |  1 +
 2 files changed, 13 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/net/wireless/b43/phy_n.c b/drivers/net/wireless/b43/phy_n.c
index f1b57070ada9..d212726d509b 100644
--- a/drivers/net/wireless/b43/phy_n.c
+++ b/drivers/net/wireless/b43/phy_n.c
@@ -3074,6 +3074,17 @@ static int b43_nphy_cal_rx_iq(struct b43_wldev *dev,
 		return b43_nphy_rev2_cal_rx_iq(dev, target, type, debug);
 }
 
+/* http://bcm-v4.sipsolutions.net/802.11/PHY/N/MacPhyClkSet */
+static void b43_nphy_mac_phy_clock_set(struct b43_wldev *dev, bool on)
+{
+	u32 tmslow = ssb_read32(dev->dev, SSB_TMSLOW);
+	if (on)
+		tmslow |= SSB_TMSLOW_PHYCLK;
+	else
+		tmslow &= ~SSB_TMSLOW_PHYCLK;
+	ssb_write32(dev->dev, SSB_TMSLOW, tmslow);
+}
+
 /*
  * Init N-PHY
  * http://bcm-v4.sipsolutions.net/802.11/PHY/Init/N
@@ -3174,7 +3185,7 @@ int b43_phy_initn(struct b43_wldev *dev)
 	b43_phy_write(dev, B43_NPHY_BBCFG, tmp & ~B43_NPHY_BBCFG_RSTCCA);
 	b43_nphy_bmac_clock_fgc(dev, 0);
 
-	/* TODO N PHY MAC PHY Clock Set with argument 1 */
+	b43_nphy_mac_phy_clock_set(dev, true);
 
 	b43_nphy_pa_override(dev, false);
 	b43_nphy_force_rf_sequence(dev, B43_RFSEQ_RX2TX);
diff --git a/include/linux/ssb/ssb_regs.h b/include/linux/ssb/ssb_regs.h
index a6d5225b9275..11daf9c140e7 100644
--- a/include/linux/ssb/ssb_regs.h
+++ b/include/linux/ssb/ssb_regs.h
@@ -97,6 +97,7 @@
 #define  SSB_TMSLOW_RESET	0x00000001 /* Reset */
 #define  SSB_TMSLOW_REJECT_22	0x00000002 /* Reject (Backplane rev 2.2) */
 #define  SSB_TMSLOW_REJECT_23	0x00000004 /* Reject (Backplane rev 2.3) */
+#define  SSB_TMSLOW_PHYCLK	0x00000010 /* MAC PHY Clock Control Enable */
 #define  SSB_TMSLOW_CLOCK	0x00010000 /* Clock Enable */
 #define  SSB_TMSLOW_FGC		0x00020000 /* Force Gated Clocks On */
 #define  SSB_TMSLOW_PE		0x40000000 /* Power Management Enable */
-- 
cgit v1.2.3


From 2a5fb7b088f8418958775774dda9427d6c73c522 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 18 Aug 2010 17:44:36 +0200
Subject: nl80211: some documentation fixes

The nl80211 documentation is currently never
generated, so problems have accumulated. Fix
most of the trivial ones.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 57 +++++++++++++++++++++++++++++++++++++++----------
 1 file changed, 46 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 8af1e66c3cf9..ec1690da7845 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -347,6 +347,8 @@
  *	four bytes for vendor frames including the OUI. The registration
  *	cannot be dropped, but is removed automatically when the netlink
  *	socket is closed. Multiple registrations can be made.
+ * @NL80211_CMD_REGISTER_ACTION: Alias for @NL80211_CMD_REGISTER_FRAME for
+ *	backward compatibility
  * @NL80211_CMD_FRAME: Management frame TX request and RX notification. This
  *	command is used both as a request to transmit a management frame and
  *	as an event indicating reception of a frame that was not processed in
@@ -359,11 +361,14 @@
  *	operational channel). When called, this operation returns a cookie
  *	(%NL80211_ATTR_COOKIE) that will be included with the TX status event
  *	pertaining to the TX request.
+ * @NL80211_CMD_ACTION: Alias for @NL80211_CMD_FRAME for backward compatibility.
  * @NL80211_CMD_FRAME_TX_STATUS: Report TX status of a management frame
  *	transmitted with %NL80211_CMD_FRAME. %NL80211_ATTR_COOKIE identifies
  *	the TX command and %NL80211_ATTR_FRAME includes the contents of the
  *	frame. %NL80211_ATTR_ACK flag is included if the recipient acknowledged
  *	the frame.
+ * @NL80211_CMD_ACTION_TX_STATUS: Alias for @NL80211_CMD_FRAME_TX_STATUS for
+ *	backward compatibility.
  * @NL80211_CMD_SET_CQM: Connection quality monitor configuration. This command
  *	is used to configure connection quality monitoring notification trigger
  *	levels.
@@ -1029,11 +1034,14 @@ enum nl80211_iftype {
  * Station flags. When a station is added to an AP interface, it is
  * assumed to be already associated (and hence authenticated.)
  *
+ * @__NL80211_STA_FLAG_INVALID: attribute number 0 is reserved
  * @NL80211_STA_FLAG_AUTHORIZED: station is authorized (802.1X)
  * @NL80211_STA_FLAG_SHORT_PREAMBLE: station is capable of receiving frames
  *	with short barker preamble
  * @NL80211_STA_FLAG_WME: station is WME/QoS capable
  * @NL80211_STA_FLAG_MFP: station uses management frame protection
+ * @NL80211_STA_FLAG_MAX: highest station flag number currently defined
+ * @__NL80211_STA_FLAG_AFTER_LAST: internal use
  */
 enum nl80211_sta_flags {
 	__NL80211_STA_FLAG_INVALID,
@@ -1146,14 +1154,17 @@ enum nl80211_mpath_flags {
  * information about a mesh path.
  *
  * @__NL80211_MPATH_INFO_INVALID: attribute number 0 is reserved
- * @NL80211_ATTR_MPATH_FRAME_QLEN: number of queued frames for this destination
- * @NL80211_ATTR_MPATH_SN: destination sequence number
- * @NL80211_ATTR_MPATH_METRIC: metric (cost) of this mesh path
- * @NL80211_ATTR_MPATH_EXPTIME: expiration time for the path, in msec from now
- * @NL80211_ATTR_MPATH_FLAGS: mesh path flags, enumerated in
+ * @NL80211_MPATH_INFO_FRAME_QLEN: number of queued frames for this destination
+ * @NL80211_MPATH_INFO_SN: destination sequence number
+ * @NL80211_MPATH_INFO_METRIC: metric (cost) of this mesh path
+ * @NL80211_MPATH_INFO_EXPTIME: expiration time for the path, in msec from now
+ * @NL80211_MPATH_INFO_FLAGS: mesh path flags, enumerated in
  * 	&enum nl80211_mpath_flags;
- * @NL80211_ATTR_MPATH_DISCOVERY_TIMEOUT: total path discovery timeout, in msec
- * @NL80211_ATTR_MPATH_DISCOVERY_RETRIES: mesh path discovery retries
+ * @NL80211_MPATH_INFO_DISCOVERY_TIMEOUT: total path discovery timeout, in msec
+ * @NL80211_MPATH_INFO_DISCOVERY_RETRIES: mesh path discovery retries
+ * @NL80211_MPATH_INFO_MAX: highest mesh path information attribute number
+ *	currently defind
+ * @__NL80211_MPATH_INFO_AFTER_LAST: internal use
  */
 enum nl80211_mpath_info {
 	__NL80211_MPATH_INFO_INVALID,
@@ -1182,6 +1193,8 @@ enum nl80211_mpath_info {
  * @NL80211_BAND_ATTR_HT_CAPA: HT capabilities, as in the HT information IE
  * @NL80211_BAND_ATTR_HT_AMPDU_FACTOR: A-MPDU factor, as in 11n
  * @NL80211_BAND_ATTR_HT_AMPDU_DENSITY: A-MPDU density, as in 11n
+ * @NL80211_BAND_ATTR_MAX: highest band attribute currently defined
+ * @__NL80211_BAND_ATTR_AFTER_LAST: internal use
  */
 enum nl80211_band_attr {
 	__NL80211_BAND_ATTR_INVALID,
@@ -1202,6 +1215,7 @@ enum nl80211_band_attr {
 
 /**
  * enum nl80211_frequency_attr - frequency attributes
+ * @__NL80211_FREQUENCY_ATTR_INVALID: attribute number 0 is reserved
  * @NL80211_FREQUENCY_ATTR_FREQ: Frequency in MHz
  * @NL80211_FREQUENCY_ATTR_DISABLED: Channel is disabled in current
  *	regulatory domain.
@@ -1213,6 +1227,9 @@ enum nl80211_band_attr {
  *	on this channel in current regulatory domain.
  * @NL80211_FREQUENCY_ATTR_MAX_TX_POWER: Maximum transmission power in mBm
  *	(100 * dBm).
+ * @NL80211_FREQUENCY_ATTR_MAX: highest frequency attribute number
+ *	currently defined
+ * @__NL80211_FREQUENCY_ATTR_AFTER_LAST: internal use
  */
 enum nl80211_frequency_attr {
 	__NL80211_FREQUENCY_ATTR_INVALID,
@@ -1232,9 +1249,13 @@ enum nl80211_frequency_attr {
 
 /**
  * enum nl80211_bitrate_attr - bitrate attributes
+ * @__NL80211_BITRATE_ATTR_INVALID: attribute number 0 is reserved
  * @NL80211_BITRATE_ATTR_RATE: Bitrate in units of 100 kbps
  * @NL80211_BITRATE_ATTR_2GHZ_SHORTPREAMBLE: Short preamble supported
  *	in 2.4 GHz band.
+ * @NL80211_BITRATE_ATTR_MAX: highest bitrate attribute number
+ *	currently defined
+ * @__NL80211_BITRATE_ATTR_AFTER_LAST: internal use
  */
 enum nl80211_bitrate_attr {
 	__NL80211_BITRATE_ATTR_INVALID,
@@ -1290,6 +1311,7 @@ enum nl80211_reg_type {
 
 /**
  * enum nl80211_reg_rule_attr - regulatory rule attributes
+ * @__NL80211_REG_RULE_ATTR_INVALID: attribute number 0 is reserved
  * @NL80211_ATTR_REG_RULE_FLAGS: a set of flags which specify additional
  * 	considerations for a given frequency range. These are the
  * 	&enum nl80211_reg_rule_flags.
@@ -1306,6 +1328,9 @@ enum nl80211_reg_type {
  * 	If you don't have one then don't send this.
  * @NL80211_ATTR_POWER_RULE_MAX_EIRP: the maximum allowed EIRP for
  * 	a given frequency range. The value is in mBm (100 * dBm).
+ * @NL80211_REG_RULE_ATTR_MAX: highest regulatory rule attribute number
+ *	currently defined
+ * @__NL80211_REG_RULE_ATTR_AFTER_LAST: internal use
  */
 enum nl80211_reg_rule_attr {
 	__NL80211_REG_RULE_ATTR_INVALID,
@@ -1357,6 +1382,9 @@ enum nl80211_reg_rule_flags {
  * @__NL80211_SURVEY_INFO_INVALID: attribute number 0 is reserved
  * @NL80211_SURVEY_INFO_FREQUENCY: center frequency of channel
  * @NL80211_SURVEY_INFO_NOISE: noise level of channel (u8, dBm)
+ * @NL80211_SURVEY_INFO_MAX: highest survey info attribute number
+ *	currently defined
+ * @__NL80211_SURVEY_INFO_AFTER_LAST: internal use
  */
 enum nl80211_survey_info {
 	__NL80211_SURVEY_INFO_INVALID,
@@ -1521,6 +1549,7 @@ enum nl80211_channel_type {
  * enum nl80211_bss - netlink attributes for a BSS
  *
  * @__NL80211_BSS_INVALID: invalid
+ * @NL80211_BSS_BSSID: BSSID of the BSS (6 octets)
  * @NL80211_BSS_FREQUENCY: frequency in MHz (u32)
  * @NL80211_BSS_TSF: TSF of the received probe response/beacon (u64)
  * @NL80211_BSS_BEACON_INTERVAL: beacon interval of the (I)BSS (u16)
@@ -1564,6 +1593,12 @@ enum nl80211_bss {
 
 /**
  * enum nl80211_bss_status - BSS "status"
+ * @NL80211_BSS_STATUS_AUTHENTICATED: Authenticated with this BSS.
+ * @NL80211_BSS_STATUS_ASSOCIATED: Associated with this BSS.
+ * @NL80211_BSS_STATUS_IBSS_JOINED: Joined to this IBSS.
+ *
+ * The BSS status is a BSS attribute in scan dumps, which
+ * indicates the status the interface has wrt. this BSS.
  */
 enum nl80211_bss_status {
 	NL80211_BSS_STATUS_AUTHENTICATED,
@@ -1674,8 +1709,8 @@ enum nl80211_tx_rate_attributes {
 
 /**
  * enum nl80211_band - Frequency band
- * @NL80211_BAND_2GHZ - 2.4 GHz ISM band
- * @NL80211_BAND_5GHZ - around 5 GHz band (4.9 - 5.7 GHz)
+ * @NL80211_BAND_2GHZ: 2.4 GHz ISM band
+ * @NL80211_BAND_5GHZ: around 5 GHz band (4.9 - 5.7 GHz)
  */
 enum nl80211_band {
 	NL80211_BAND_2GHZ,
@@ -1713,9 +1748,9 @@ enum nl80211_attr_cqm {
 
 /**
  * enum nl80211_cqm_rssi_threshold_event - RSSI threshold event
- * @NL80211_CQM_RSSI_THRESHOLD_EVENT_LOW - The RSSI level is lower than the
+ * @NL80211_CQM_RSSI_THRESHOLD_EVENT_LOW: The RSSI level is lower than the
  *      configured threshold
- * @NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH - The RSSI is higher than the
+ * @NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH: The RSSI is higher than the
  *      configured threshold
  */
 enum nl80211_cqm_rssi_threshold_event {
-- 
cgit v1.2.3


From 5a46790ca4c40fdb6ed5336d7d6b593c96326b31 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Tue, 24 Aug 2010 14:46:53 -0700
Subject: include/linux/if_ether.h: Remove unused #define MAC_FMT

Last use was removed, so remove the #define.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_ether.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_ether.h b/include/linux/if_ether.h
index bed7a4682b90..f9c3df03db0f 100644
--- a/include/linux/if_ether.h
+++ b/include/linux/if_ether.h
@@ -137,8 +137,6 @@ extern struct ctl_table ether_table[];
 
 extern ssize_t sysfs_format_mac(char *buf, const unsigned char *addr, int len);
 
-#define MAC_FMT "%02x:%02x:%02x:%02x:%02x:%02x"
-
 #endif
 
 #endif	/* _LINUX_IF_ETHER_H */
-- 
cgit v1.2.3


From c2e3143e3c46ede22336316b3ff4746727c0d93a Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Tue, 24 Aug 2010 14:48:10 -0700
Subject: tc: add meta match on receive hash

Trivial extension to existing meta data match rules to allow
matching on skb receive hash value.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tc_ematch/tc_em_meta.h | 1 +
 net/sched/em_meta.c                  | 6 ++++++
 2 files changed, 7 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h
index 0864206ec1a3..7138962664f8 100644
--- a/include/linux/tc_ematch/tc_em_meta.h
+++ b/include/linux/tc_ematch/tc_em_meta.h
@@ -79,6 +79,7 @@ enum {
  	TCF_META_ID_SK_SENDMSG_OFF,
  	TCF_META_ID_SK_WRITE_PENDING,
 	TCF_META_ID_VLAN_TAG,
+	TCF_META_ID_RXHASH,
 	__TCF_META_ID_MAX
 };
 #define TCF_META_ID_MAX (__TCF_META_ID_MAX - 1)
diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c
index 3bcac8aa333c..34da5e29ea1a 100644
--- a/net/sched/em_meta.c
+++ b/net/sched/em_meta.c
@@ -223,6 +223,11 @@ META_COLLECTOR(int_maclen)
 	dst->value = skb->mac_len;
 }
 
+META_COLLECTOR(int_rxhash)
+{
+	dst->value = skb_get_rxhash(skb);
+}
+
 /**************************************************************************
  * Netfilter
  **************************************************************************/
@@ -541,6 +546,7 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = {
 		[META_ID(SK_SENDMSG_OFF)]	= META_FUNC(int_sk_sendmsg_off),
 		[META_ID(SK_WRITE_PENDING)]	= META_FUNC(int_sk_write_pend),
 		[META_ID(VLAN_TAG)]		= META_FUNC(int_vlan_tag),
+		[META_ID(RXHASH)]		= META_FUNC(int_rxhash),
 	}
 };
 
-- 
cgit v1.2.3


From e7c1c2c46201e46f8ce817196507d2ffd3dafd8e Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Tue, 24 Aug 2010 03:46:18 +0000
Subject: mlx4_en: Added self diagnostics test implementation

The selftest includes 5 features:
1. Interrupt test: Executing commands and receiving command completion
   on all our interrupt vectors.
2. Link test: Verifying we are connected to valid link partner.
3. Speed test: Check that we negotiated link speed correctly.
4. Registers test: Activate HW health check command.
5. Loopback test: Send a packet on loopback interface and catch it on RX side.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/Makefile      |   2 +-
 drivers/net/mlx4/en_ethtool.c  |  79 ++++++++++++------
 drivers/net/mlx4/en_netdev.c   |   2 +-
 drivers/net/mlx4/en_port.c     |  32 ++++++++
 drivers/net/mlx4/en_port.h     |  14 ++++
 drivers/net/mlx4/en_rx.c       |  20 +++++
 drivers/net/mlx4/en_selftest.c | 179 +++++++++++++++++++++++++++++++++++++++++
 drivers/net/mlx4/en_tx.c       |  16 ++++
 drivers/net/mlx4/eq.c          |  44 ++++++++++
 drivers/net/mlx4/fw.c          |   3 +
 drivers/net/mlx4/fw.h          |   1 +
 drivers/net/mlx4/main.c        |   1 +
 drivers/net/mlx4/mlx4_en.h     |  19 +++++
 include/linux/mlx4/cmd.h       |   1 +
 include/linux/mlx4/device.h    |   2 +
 15 files changed, 388 insertions(+), 27 deletions(-)
 create mode 100644 drivers/net/mlx4/en_selftest.c

(limited to 'include/linux')

diff --git a/drivers/net/mlx4/Makefile b/drivers/net/mlx4/Makefile
index 1fd068e1d930..d1aa45a15854 100644
--- a/drivers/net/mlx4/Makefile
+++ b/drivers/net/mlx4/Makefile
@@ -6,4 +6,4 @@ mlx4_core-y :=	alloc.o catas.o cmd.o cq.o eq.o fw.o icm.o intf.o main.o mcg.o \
 obj-$(CONFIG_MLX4_EN)               += mlx4_en.o
 
 mlx4_en-y := 	en_main.o en_tx.o en_rx.o en_ethtool.o en_port.o en_cq.o \
-		en_resources.o en_netdev.o
+		en_resources.o en_netdev.o en_selftest.o
diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c
index 398d54136967..f7d72d7a8704 100644
--- a/drivers/net/mlx4/en_ethtool.c
+++ b/drivers/net/mlx4/en_ethtool.c
@@ -125,6 +125,14 @@ static const char main_strings[][ETH_GSTRING_LEN] = {
 #define NUM_MAIN_STATS	21
 #define NUM_ALL_STATS	(NUM_MAIN_STATS + NUM_PORT_STATS + NUM_PKT_STATS + NUM_PERF_STATS)
 
+static const char mlx4_en_test_names[][ETH_GSTRING_LEN]= {
+	"Interupt Test",
+	"Link Test",
+	"Speed Test",
+	"Register Test",
+	"Loopback Test",
+};
+
 static u32 mlx4_en_get_msglevel(struct net_device *dev)
 {
 	return ((struct mlx4_en_priv *) netdev_priv(dev))->msg_enable;
@@ -146,10 +154,15 @@ static int mlx4_en_get_sset_count(struct net_device *dev, int sset)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 
-	if (sset != ETH_SS_STATS)
+	switch (sset) {
+	case ETH_SS_STATS:
+		return NUM_ALL_STATS +
+			(priv->tx_ring_num + priv->rx_ring_num) * 2;
+	case ETH_SS_TEST:
+		return MLX4_EN_NUM_SELF_TEST - !(priv->mdev->dev->caps.loopback_support) * 2;
+	default:
 		return -EOPNOTSUPP;
-
-	return NUM_ALL_STATS + (priv->tx_ring_num + priv->rx_ring_num) * 2;
+	}
 }
 
 static void mlx4_en_get_ethtool_stats(struct net_device *dev,
@@ -181,6 +194,12 @@ static void mlx4_en_get_ethtool_stats(struct net_device *dev,
 
 }
 
+static void mlx4_en_self_test(struct net_device *dev,
+			      struct ethtool_test *etest, u64 *buf)
+{
+	mlx4_en_ex_selftest(dev, &etest->flags, buf);
+}
+
 static void mlx4_en_get_strings(struct net_device *dev,
 				uint32_t stringset, uint8_t *data)
 {
@@ -188,30 +207,39 @@ static void mlx4_en_get_strings(struct net_device *dev,
 	int index = 0;
 	int i;
 
-	if (stringset != ETH_SS_STATS)
-		return;
-
-	/* Add main counters */
-	for (i = 0; i < NUM_MAIN_STATS; i++)
-		strcpy(data + (index++) * ETH_GSTRING_LEN, main_strings[i]);
-	for (i = 0; i < NUM_PORT_STATS; i++)
-		strcpy(data + (index++) * ETH_GSTRING_LEN,
+	switch (stringset) {
+	case ETH_SS_TEST:
+		for (i = 0; i < MLX4_EN_NUM_SELF_TEST - 2; i++)
+			strcpy(data + i * ETH_GSTRING_LEN, mlx4_en_test_names[i]);
+		if (priv->mdev->dev->caps.loopback_support)
+			for (; i < MLX4_EN_NUM_SELF_TEST; i++)
+				strcpy(data + i * ETH_GSTRING_LEN, mlx4_en_test_names[i]);
+		break;
+
+	case ETH_SS_STATS:
+		/* Add main counters */
+		for (i = 0; i < NUM_MAIN_STATS; i++)
+			strcpy(data + (index++) * ETH_GSTRING_LEN, main_strings[i]);
+		for (i = 0; i< NUM_PORT_STATS; i++)
+			strcpy(data + (index++) * ETH_GSTRING_LEN,
 			main_strings[i + NUM_MAIN_STATS]);
-	for (i = 0; i < priv->tx_ring_num; i++) {
-		sprintf(data + (index++) * ETH_GSTRING_LEN,
-			"tx%d_packets", i);
-		sprintf(data + (index++) * ETH_GSTRING_LEN,
-			"tx%d_bytes", i);
-	}
-	for (i = 0; i < priv->rx_ring_num; i++) {
-		sprintf(data + (index++) * ETH_GSTRING_LEN,
-			"rx%d_packets", i);
-		sprintf(data + (index++) * ETH_GSTRING_LEN,
-			"rx%d_bytes", i);
-	}
-	for (i = 0; i < NUM_PKT_STATS; i++)
-		strcpy(data + (index++) * ETH_GSTRING_LEN,
+		for (i = 0; i < priv->tx_ring_num; i++) {
+			sprintf(data + (index++) * ETH_GSTRING_LEN,
+				"tx%d_packets", i);
+			sprintf(data + (index++) * ETH_GSTRING_LEN,
+				"tx%d_bytes", i);
+		}
+		for (i = 0; i < priv->rx_ring_num; i++) {
+			sprintf(data + (index++) * ETH_GSTRING_LEN,
+				"rx%d_packets", i);
+			sprintf(data + (index++) * ETH_GSTRING_LEN,
+				"rx%d_bytes", i);
+		}
+		for (i = 0; i< NUM_PKT_STATS; i++)
+			strcpy(data + (index++) * ETH_GSTRING_LEN,
 			main_strings[i + NUM_MAIN_STATS + NUM_PORT_STATS]);
+		break;
+	}
 }
 
 static int mlx4_en_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
@@ -439,6 +467,7 @@ const struct ethtool_ops mlx4_en_ethtool_ops = {
 	.get_strings = mlx4_en_get_strings,
 	.get_sset_count = mlx4_en_get_sset_count,
 	.get_ethtool_stats = mlx4_en_get_ethtool_stats,
+	.self_test = mlx4_en_self_test,
 	.get_wol = mlx4_en_get_wol,
 	.get_msglevel = mlx4_en_get_msglevel,
 	.set_msglevel = mlx4_en_set_msglevel,
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index 34cfa3cfbdae..968e75b7acf4 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -109,7 +109,7 @@ static void mlx4_en_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid)
 	mutex_unlock(&mdev->state_lock);
 }
 
-static u64 mlx4_en_mac_to_u64(u8 *addr)
+u64 mlx4_en_mac_to_u64(u8 *addr)
 {
 	u64 mac = 0;
 	int i;
diff --git a/drivers/net/mlx4/en_port.c b/drivers/net/mlx4/en_port.c
index a29abe845d2e..aa3ef2aee5bf 100644
--- a/drivers/net/mlx4/en_port.c
+++ b/drivers/net/mlx4/en_port.c
@@ -142,6 +142,38 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn,
 	return err;
 }
 
+int mlx4_en_QUERY_PORT(struct mlx4_en_dev *mdev, u8 port)
+{
+	struct mlx4_en_query_port_context *qport_context;
+	struct mlx4_en_priv *priv = netdev_priv(mdev->pndev[port]);
+	struct mlx4_en_port_state *state = &priv->port_state;
+	struct mlx4_cmd_mailbox *mailbox;
+	int err;
+
+	mailbox = mlx4_alloc_cmd_mailbox(mdev->dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+	memset(mailbox->buf, 0, sizeof(*qport_context));
+	err = mlx4_cmd_box(mdev->dev, 0, mailbox->dma, port, 0,
+			   MLX4_CMD_QUERY_PORT, MLX4_CMD_TIME_CLASS_B);
+	if (err)
+		goto out;
+	qport_context = mailbox->buf;
+
+	/* This command is always accessed from Ethtool context
+	 * already synchronized, no need in locking */
+	state->link_state = !!(qport_context->link_up & MLX4_EN_LINK_UP_MASK);
+	if ((qport_context->link_speed & MLX4_EN_SPEED_MASK) ==
+	    MLX4_EN_1G_SPEED)
+		state->link_speed = 1000;
+	else
+		state->link_speed = 10000;
+	state->transciver = qport_context->transceiver;
+
+out:
+	mlx4_free_cmd_mailbox(mdev->dev, mailbox);
+	return err;
+}
 
 int mlx4_en_DUMP_ETH_STATS(struct mlx4_en_dev *mdev, u8 port, u8 reset)
 {
diff --git a/drivers/net/mlx4/en_port.h b/drivers/net/mlx4/en_port.h
index e6477f12beb5..f6511aa2b7df 100644
--- a/drivers/net/mlx4/en_port.h
+++ b/drivers/net/mlx4/en_port.h
@@ -84,6 +84,20 @@ enum {
 	MLX4_MCAST_ENABLE       = 2,
 };
 
+struct mlx4_en_query_port_context {
+	u8 link_up;
+#define MLX4_EN_LINK_UP_MASK	0x80
+	u8 reserved;
+	__be16 mtu;
+	u8 reserved2;
+	u8 link_speed;
+#define MLX4_EN_SPEED_MASK	0x3
+#define MLX4_EN_1G_SPEED	0x2
+	u16 reserved3[5];
+	__be64 mac;
+	u8 transceiver;
+};
+
 
 struct mlx4_en_stat_out_mbox {
 	/* Received frames with a length of 64 octets */
diff --git a/drivers/net/mlx4/en_rx.c b/drivers/net/mlx4/en_rx.c
index 7a5123c4a366..f421a3d42723 100644
--- a/drivers/net/mlx4/en_rx.c
+++ b/drivers/net/mlx4/en_rx.c
@@ -541,6 +541,21 @@ static struct sk_buff *mlx4_en_rx_skb(struct mlx4_en_priv *priv,
 	return skb;
 }
 
+static void validate_loopback(struct mlx4_en_priv *priv, struct sk_buff *skb)
+{
+	int i;
+	int offset = ETH_HLEN;
+
+	for (i = 0; i < MLX4_LOOPBACK_TEST_PAYLOAD; i++, offset++) {
+		if (*(skb->data + offset) != (unsigned char) (i & 0xff))
+			goto out_loopback;
+	}
+	/* Loopback found */
+	priv->loopback_ok = 1;
+
+out_loopback:
+	dev_kfree_skb_any(skb);
+}
 
 int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int budget)
 {
@@ -655,6 +670,11 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
 			goto next;
 		}
 
+                if (unlikely(priv->validate_loopback)) {
+			validate_loopback(priv, skb);
+			goto next;
+		}
+
 		skb->ip_summed = ip_summed;
 		skb->protocol = eth_type_trans(skb, dev);
 		skb_record_rx_queue(skb, cq->ring);
diff --git a/drivers/net/mlx4/en_selftest.c b/drivers/net/mlx4/en_selftest.c
new file mode 100644
index 000000000000..43357d35616a
--- /dev/null
+++ b/drivers/net/mlx4/en_selftest.c
@@ -0,0 +1,179 @@
+/*
+ * Copyright (c) 2007 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+#include <linux/delay.h>
+#include <linux/mlx4/driver.h>
+
+#include "mlx4_en.h"
+
+
+static int mlx4_en_test_registers(struct mlx4_en_priv *priv)
+{
+	return mlx4_cmd(priv->mdev->dev, 0, 0, 0, MLX4_CMD_HW_HEALTH_CHECK,
+			MLX4_CMD_TIME_CLASS_A);
+}
+
+static int mlx4_en_test_loopback_xmit(struct mlx4_en_priv *priv)
+{
+	struct sk_buff *skb;
+	struct ethhdr *ethh;
+	unsigned char *packet;
+	unsigned int packet_size = MLX4_LOOPBACK_TEST_PAYLOAD;
+	unsigned int i;
+	int err;
+
+
+	/* build the pkt before xmit */
+	skb = netdev_alloc_skb(priv->dev, MLX4_LOOPBACK_TEST_PAYLOAD + ETH_HLEN + NET_IP_ALIGN);
+	if (!skb) {
+		en_err(priv, "-LOOPBACK_TEST_XMIT- failed to create skb for xmit\n");
+		return -ENOMEM;
+	}
+	skb_reserve(skb, NET_IP_ALIGN);
+
+	ethh = (struct ethhdr *)skb_put(skb, sizeof(struct ethhdr));
+	packet	= (unsigned char *)skb_put(skb, packet_size);
+	memcpy(ethh->h_dest, priv->dev->dev_addr, ETH_ALEN);
+	memset(ethh->h_source, 0, ETH_ALEN);
+	ethh->h_proto = htons(ETH_P_ARP);
+	skb_set_mac_header(skb, 0);
+	for (i = 0; i < packet_size; ++i)	/* fill our packet */
+		packet[i] = (unsigned char)(i & 0xff);
+
+	/* xmit the pkt */
+	err = mlx4_en_xmit(skb, priv->dev);
+	return err;
+}
+
+static int mlx4_en_test_loopback(struct mlx4_en_priv *priv)
+{
+	u32 loopback_ok = 0;
+	int i;
+
+
+        priv->loopback_ok = 0;
+	priv->validate_loopback = 1;
+
+	/* xmit */
+	if (mlx4_en_test_loopback_xmit(priv)) {
+		en_err(priv, "Transmitting loopback packet failed\n");
+		goto mlx4_en_test_loopback_exit;
+	}
+
+	/* polling for result */
+	for (i = 0; i < MLX4_EN_LOOPBACK_RETRIES; ++i) {
+		msleep(MLX4_EN_LOOPBACK_TIMEOUT);
+		if (priv->loopback_ok) {
+			loopback_ok = 1;
+			break;
+		}
+	}
+	if (!loopback_ok)
+		en_err(priv, "Loopback packet didn't arrive\n");
+
+mlx4_en_test_loopback_exit:
+
+	priv->validate_loopback = 0;
+	return (!loopback_ok);
+}
+
+
+static int mlx4_en_test_link(struct mlx4_en_priv *priv)
+{
+	if (mlx4_en_QUERY_PORT(priv->mdev, priv->port))
+		return -ENOMEM;
+	if (priv->port_state.link_state == 1)
+		return 0;
+	else
+		return 1;
+}
+
+static int mlx4_en_test_speed(struct mlx4_en_priv *priv)
+{
+
+	if (mlx4_en_QUERY_PORT(priv->mdev, priv->port))
+		return -ENOMEM;
+
+	/* The device currently only supports 10G speed */
+	if (priv->port_state.link_speed != SPEED_10000)
+		return priv->port_state.link_speed;
+	return 0;
+}
+
+
+void mlx4_en_ex_selftest(struct net_device *dev, u32 *flags, u64 *buf)
+{
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+	struct mlx4_en_dev *mdev = priv->mdev;
+	struct mlx4_en_tx_ring *tx_ring;
+	int i, carrier_ok;
+
+	memset(buf, 0, sizeof(u64) * MLX4_EN_NUM_SELF_TEST);
+
+	if (*flags & ETH_TEST_FL_OFFLINE) {
+		/* disable the interface */
+		carrier_ok = netif_carrier_ok(dev);
+
+		netif_carrier_off(dev);
+retry_tx:
+		/* Wait untill all tx queues are empty.
+		 * there should not be any additional incoming traffic
+		 * since we turned the carrier off */
+		msleep(200);
+		for (i = 0; i < priv->tx_ring_num && carrier_ok; i++) {
+			tx_ring = &priv->tx_ring[i];
+			if (tx_ring->prod != (tx_ring->cons + tx_ring->last_nr_txbb))
+				goto retry_tx;
+		}
+
+		if (priv->mdev->dev->caps.loopback_support){
+			buf[3] = mlx4_en_test_registers(priv);
+			buf[4] = mlx4_en_test_loopback(priv);
+		}
+
+		if (carrier_ok)
+			netif_carrier_on(dev);
+
+	}
+	buf[0] = mlx4_test_interrupts(mdev->dev);
+	buf[1] = mlx4_en_test_link(priv);
+	buf[2] = mlx4_en_test_speed(priv);
+
+	for (i = 0; i < MLX4_EN_NUM_SELF_TEST; i++) {
+		if (buf[i])
+			*flags |= ETH_TEST_FL_FAILED;
+	}
+}
diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c
index 3deabd1d0357..b875f9c38848 100644
--- a/drivers/net/mlx4/en_tx.c
+++ b/drivers/net/mlx4/en_tx.c
@@ -600,6 +600,9 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct mlx4_wqe_data_seg *data;
 	struct skb_frag_struct *frag;
 	struct mlx4_en_tx_info *tx_info;
+	struct ethhdr *ethh;
+	u64 mac;
+	u32 mac_l, mac_h;
 	int tx_ind = 0;
 	int nr_txbb;
 	int desc_size;
@@ -679,6 +682,19 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 		priv->port_stats.tx_chksum_offload++;
 	}
 
+	if (unlikely(priv->validate_loopback)) {
+		/* Copy dst mac address to wqe */
+		skb_reset_mac_header(skb);
+		ethh = eth_hdr(skb);
+		if (ethh && ethh->h_dest) {
+			mac = mlx4_en_mac_to_u64(ethh->h_dest);
+			mac_h = (u32) ((mac & 0xffff00000000ULL) >> 16);
+			mac_l = (u32) (mac & 0xffffffff);
+			tx_desc->ctrl.srcrb_flags |= cpu_to_be32(mac_h);
+			tx_desc->ctrl.imm = cpu_to_be32(mac_l);
+		}
+	}
+
 	/* Handle LSO (TSO) packets */
 	if (lso_header_size) {
 		/* Mark opcode as LSO */
diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c
index 6d7b2bf210ce..552d0fce6f67 100644
--- a/drivers/net/mlx4/eq.c
+++ b/drivers/net/mlx4/eq.c
@@ -699,3 +699,47 @@ void mlx4_cleanup_eq_table(struct mlx4_dev *dev)
 
 	kfree(priv->eq_table.uar_map);
 }
+
+/* A test that verifies that we can accept interrupts on all
+ * the irq vectors of the device.
+ * Interrupts are checked using the NOP command.
+ */
+int mlx4_test_interrupts(struct mlx4_dev *dev)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	int i;
+	int err;
+
+	err = mlx4_NOP(dev);
+	/* When not in MSI_X, there is only one irq to check */
+	if (!(dev->flags & MLX4_FLAG_MSI_X))
+		return err;
+
+	/* A loop over all completion vectors, for each vector we will check
+	 * whether it works by mapping command completions to that vector
+	 * and performing a NOP command
+	 */
+	for(i = 0; !err && (i < dev->caps.num_comp_vectors); ++i) {
+		/* Temporary use polling for command completions */
+		mlx4_cmd_use_polling(dev);
+
+		/* Map the new eq to handle all asyncronous events */
+		err = mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 0,
+				  priv->eq_table.eq[i].eqn);
+		if (err) {
+			mlx4_warn(dev, "Failed mapping eq for interrupt test\n");
+			mlx4_cmd_use_events(dev);
+			break;
+		}
+
+		/* Go back to using events */
+		mlx4_cmd_use_events(dev);
+		err = mlx4_NOP(dev);
+	}
+
+	/* Return to default */
+	mlx4_MAP_EQ(dev, MLX4_ASYNC_EVENT_MASK, 0,
+		    priv->eq_table.eq[dev->caps.num_comp_vectors].eqn);
+	return err;
+}
+EXPORT_SYMBOL(mlx4_test_interrupts);
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 04f42ae1eda0..a87bf3c97055 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -178,6 +178,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 #define QUERY_DEV_CAP_MAX_GID_OFFSET		0x3b
 #define QUERY_DEV_CAP_RATE_SUPPORT_OFFSET	0x3c
 #define QUERY_DEV_CAP_MAX_PKEY_OFFSET		0x3f
+#define QUERY_DEV_CAP_ETH_UC_LOOPBACK_OFFSET	0x43
 #define QUERY_DEV_CAP_FLAGS_OFFSET		0x44
 #define QUERY_DEV_CAP_RSVD_UAR_OFFSET		0x48
 #define QUERY_DEV_CAP_UAR_SZ_OFFSET		0x49
@@ -268,6 +269,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev_cap->max_msg_sz = 1 << (field & 0x1f);
 	MLX4_GET(stat_rate, outbox, QUERY_DEV_CAP_RATE_SUPPORT_OFFSET);
 	dev_cap->stat_rate_support = stat_rate;
+	MLX4_GET(field, outbox, QUERY_DEV_CAP_ETH_UC_LOOPBACK_OFFSET);
+	dev_cap->loopback_support = field & 0x1;
 	MLX4_GET(dev_cap->flags, outbox, QUERY_DEV_CAP_FLAGS_OFFSET);
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_RSVD_UAR_OFFSET);
 	dev_cap->reserved_uars = field >> 4;
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index 526d7f30c041..2cc1ba5452e3 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -74,6 +74,7 @@ struct mlx4_dev_cap {
 	u64 def_mac[MLX4_MAX_PORTS + 1];
 	u16 eth_mtu[MLX4_MAX_PORTS + 1];
 	u16 stat_rate_support;
+	int loopback_support;
 	u32 flags;
 	int reserved_uars;
 	int uar_size;
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 5102ab1ac561..7390cdb19414 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -221,6 +221,7 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.bmme_flags	     = dev_cap->bmme_flags;
 	dev->caps.reserved_lkey	     = dev_cap->reserved_lkey;
 	dev->caps.stat_rate_support  = dev_cap->stat_rate_support;
+	dev->caps.loopback_support   = dev_cap->loopback_support;
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 
 	dev->caps.log_num_macs  = log_num_mac;
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index edf6523702c0..a09598b2b12e 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -45,6 +45,7 @@
 #include <linux/mlx4/cq.h>
 #include <linux/mlx4/srq.h>
 #include <linux/mlx4/doorbell.h>
+#include <linux/mlx4/cmd.h>
 
 #include "en_port.h"
 
@@ -139,10 +140,14 @@ enum {
 
 #define SMALL_PACKET_SIZE      (256 - NET_IP_ALIGN)
 #define HEADER_COPY_SIZE       (128 - NET_IP_ALIGN)
+#define MLX4_LOOPBACK_TEST_PAYLOAD (HEADER_COPY_SIZE - ETH_HLEN)
 
 #define MLX4_EN_MIN_MTU		46
 #define ETH_BCAST		0xffffffffffffULL
 
+#define MLX4_EN_LOOPBACK_RETRIES	5
+#define MLX4_EN_LOOPBACK_TIMEOUT	100
+
 #ifdef MLX4_EN_PERF_STAT
 /* Number of samples to 'average' */
 #define AVG_SIZE			128
@@ -356,6 +361,12 @@ struct mlx4_en_rss_context {
 	__be32 rss_key[10];
 };
 
+struct mlx4_en_port_state {
+	int link_state;
+	int link_speed;
+	int transciver;
+};
+
 struct mlx4_en_pkt_stats {
 	unsigned long broadcast;
 	unsigned long rx_prio[8];
@@ -404,6 +415,7 @@ struct mlx4_en_priv {
 	struct vlan_group *vlgrp;
 	struct net_device_stats stats;
 	struct net_device_stats ret_stats;
+	struct mlx4_en_port_state port_state;
 	spinlock_t stats_lock;
 
 	unsigned long last_moder_packets;
@@ -422,6 +434,8 @@ struct mlx4_en_priv {
 	u16 sample_interval;
 	u16 adaptive_rx_coal;
 	u32 msg_enable;
+	u32 loopback_ok;
+	u32 validate_loopback;
 
 	struct mlx4_hwq_resources res;
 	int link_state;
@@ -530,6 +544,11 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn,
 			   u8 promisc);
 
 int mlx4_en_DUMP_ETH_STATS(struct mlx4_en_dev *mdev, u8 port, u8 reset);
+int mlx4_en_QUERY_PORT(struct mlx4_en_dev *mdev, u8 port);
+
+#define MLX4_EN_NUM_SELF_TEST	5
+void mlx4_en_ex_selftest(struct net_device *dev, u32 *flags, u64 *buf);
+u64 mlx4_en_mac_to_u64(u8 *addr);
 
 /*
  * Globals
diff --git a/include/linux/mlx4/cmd.h b/include/linux/mlx4/cmd.h
index 0f82293a82ed..78a1b9671752 100644
--- a/include/linux/mlx4/cmd.h
+++ b/include/linux/mlx4/cmd.h
@@ -56,6 +56,7 @@ enum {
 	MLX4_CMD_QUERY_HCA	 = 0xb,
 	MLX4_CMD_QUERY_PORT	 = 0x43,
 	MLX4_CMD_SENSE_PORT	 = 0x4d,
+	MLX4_CMD_HW_HEALTH_CHECK = 0x50,
 	MLX4_CMD_SET_PORT	 = 0xc,
 	MLX4_CMD_ACCESS_DDR	 = 0x2e,
 	MLX4_CMD_MAP_ICM	 = 0xffa,
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 7a7f9c1e679a..2cec58722738 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -229,6 +229,7 @@ struct mlx4_caps {
 	u32			bmme_flags;
 	u32			reserved_lkey;
 	u16			stat_rate_support;
+	int			loopback_support;
 	u8			port_width_cap[MLX4_MAX_PORTS + 1];
 	int			max_gso_sz;
 	int                     reserved_qps_cnt[MLX4_NUM_QP_REGION];
@@ -480,5 +481,6 @@ void mlx4_fmr_unmap(struct mlx4_dev *dev, struct mlx4_fmr *fmr,
 		    u32 *lkey, u32 *rkey);
 int mlx4_fmr_free(struct mlx4_dev *dev, struct mlx4_fmr *fmr);
 int mlx4_SYNC_TPT(struct mlx4_dev *dev);
+int mlx4_test_interrupts(struct mlx4_dev *dev);
 
 #endif /* MLX4_DEVICE_H */
-- 
cgit v1.2.3


From 7699517db435fd24143bd32dd644275e3eeb4c86 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Tue, 24 Aug 2010 03:46:23 +0000
Subject: mlx4_en: Fixing report in Ethtool get_settings

The report now based on query from FW, giving the correct tranciever type
and link speed.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_ethtool.c | 27 +++++++++++++++++++++++++--
 drivers/net/mlx4/fw.c         |  9 +++++++++
 drivers/net/mlx4/fw.h         |  4 ++++
 drivers/net/mlx4/main.c       |  4 ++++
 include/linux/mlx4/device.h   |  4 ++++
 5 files changed, 46 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c
index f7d72d7a8704..fa2f2f43ab48 100644
--- a/drivers/net/mlx4/en_ethtool.c
+++ b/drivers/net/mlx4/en_ethtool.c
@@ -244,16 +244,39 @@ static void mlx4_en_get_strings(struct net_device *dev,
 
 static int mlx4_en_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 {
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+	int trans_type;
+
 	cmd->autoneg = AUTONEG_DISABLE;
 	cmd->supported = SUPPORTED_10000baseT_Full;
-	cmd->advertising = ADVERTISED_1000baseT_Full;
+	cmd->advertising = ADVERTISED_10000baseT_Full;
+
+	if (mlx4_en_QUERY_PORT(priv->mdev, priv->port))
+		return -ENOMEM;
+
+	trans_type = priv->port_state.transciver;
 	if (netif_carrier_ok(dev)) {
-		cmd->speed = SPEED_10000;
+		cmd->speed = priv->port_state.link_speed;
 		cmd->duplex = DUPLEX_FULL;
 	} else {
 		cmd->speed = -1;
 		cmd->duplex = -1;
 	}
+
+	if (trans_type > 0 && trans_type <= 0xC) {
+		cmd->port = PORT_FIBRE;
+		cmd->transceiver = XCVR_EXTERNAL;
+		cmd->supported |= SUPPORTED_FIBRE;
+		cmd->advertising |= ADVERTISED_FIBRE;
+	} else if (trans_type == 0x80 || trans_type == 0) {
+		cmd->port = PORT_TP;
+		cmd->transceiver = XCVR_INTERNAL;
+		cmd->supported |= SUPPORTED_TP;
+		cmd->advertising |= ADVERTISED_TP;
+	} else  {
+		cmd->port = -1;
+		cmd->transceiver = -1;
+	}
 	return 0;
 }
 
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index a87bf3c97055..515c6348f32b 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -141,6 +141,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	struct mlx4_cmd_mailbox *mailbox;
 	u32 *outbox;
 	u8 field;
+	u32 field32;
 	u16 size;
 	u16 stat_rate;
 	int err;
@@ -368,6 +369,9 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 #define QUERY_PORT_MAX_MACVLAN_OFFSET		0x0a
 #define QUERY_PORT_MAX_VL_OFFSET		0x0b
 #define QUERY_PORT_MAC_OFFSET			0x10
+#define QUERY_PORT_TRANS_VENDOR_OFFSET		0x18
+#define QUERY_PORT_WAVELENGTH_OFFSET		0x1c
+#define QUERY_PORT_TRANS_CODE_OFFSET		0x20
 
 		for (i = 1; i <= dev_cap->num_ports; ++i) {
 			err = mlx4_cmd_box(dev, 0, mailbox->dma, i, 0, MLX4_CMD_QUERY_PORT,
@@ -391,6 +395,11 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 			dev_cap->log_max_vlans[i] = field >> 4;
 			MLX4_GET(dev_cap->eth_mtu[i], outbox, QUERY_PORT_ETH_MTU_OFFSET);
 			MLX4_GET(dev_cap->def_mac[i], outbox, QUERY_PORT_MAC_OFFSET);
+			MLX4_GET(field32, outbox, QUERY_PORT_TRANS_VENDOR_OFFSET);
+			dev_cap->trans_type[i] = field32 >> 24;
+			dev_cap->vendor_oui[i] = field32 & 0xffffff;
+			MLX4_GET(dev_cap->wavelength[i], outbox, QUERY_PORT_WAVELENGTH_OFFSET);
+			MLX4_GET(dev_cap->trans_code[i], outbox, QUERY_PORT_TRANS_CODE_OFFSET);
 		}
 	}
 
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index 2cc1ba5452e3..443e456c9dab 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -73,6 +73,10 @@ struct mlx4_dev_cap {
 	int max_pkeys[MLX4_MAX_PORTS + 1];
 	u64 def_mac[MLX4_MAX_PORTS + 1];
 	u16 eth_mtu[MLX4_MAX_PORTS + 1];
+	int trans_type[MLX4_MAX_PORTS + 1];
+	int vendor_oui[MLX4_MAX_PORTS + 1];
+	u16 wavelength[MLX4_MAX_PORTS + 1];
+	u64 trans_code[MLX4_MAX_PORTS + 1];
 	u16 stat_rate_support;
 	int loopback_support;
 	u32 flags;
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 7390cdb19414..f4791fa2472f 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -184,6 +184,10 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		dev->caps.eth_mtu_cap[i]    = dev_cap->eth_mtu[i];
 		dev->caps.def_mac[i]        = dev_cap->def_mac[i];
 		dev->caps.supported_type[i] = dev_cap->supported_port_types[i];
+		dev->caps.trans_type[i]	    = dev_cap->trans_type[i];
+		dev->caps.vendor_oui[i]     = dev_cap->vendor_oui[i];
+		dev->caps.wavelength[i]     = dev_cap->wavelength[i];
+		dev->caps.trans_code[i]     = dev_cap->trans_code[i];
 	}
 
 	dev->caps.num_uars	     = dev_cap->uar_size / PAGE_SIZE;
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 2cec58722738..2a36a344fb3d 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -186,6 +186,10 @@ struct mlx4_caps {
 	int			eth_mtu_cap[MLX4_MAX_PORTS + 1];
 	int			gid_table_len[MLX4_MAX_PORTS + 1];
 	int			pkey_table_len[MLX4_MAX_PORTS + 1];
+	int			trans_type[MLX4_MAX_PORTS + 1];
+	int			vendor_oui[MLX4_MAX_PORTS + 1];
+	int			wavelength[MLX4_MAX_PORTS + 1];
+	u64			trans_code[MLX4_MAX_PORTS + 1];
 	int			local_ca_ack_delay;
 	int			num_uars;
 	int			bf_reg_size;
-- 
cgit v1.2.3


From 0533943c5c45cce2e26432bf0a6b8e114757c897 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Tue, 24 Aug 2010 03:46:42 +0000
Subject: mlx4_en: UDP RSS support

Adding capability for RSS for UDP traffic, hashing is done based on
IP addresses and UDP port number.
The support depends on HW/FW capabilities.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_main.c  | 19 ++++++++++++-------
 drivers/net/mlx4/en_rx.c    |  8 ++++----
 drivers/net/mlx4/fw.c       |  3 +++
 drivers/net/mlx4/fw.h       |  1 +
 drivers/net/mlx4/main.c     |  1 +
 drivers/net/mlx4/mlx4_en.h  |  3 +++
 include/linux/mlx4/device.h |  1 +
 7 files changed, 25 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/mlx4/en_main.c b/drivers/net/mlx4/en_main.c
index cacac4e966ee..143906417048 100644
--- a/drivers/net/mlx4/en_main.c
+++ b/drivers/net/mlx4/en_main.c
@@ -63,11 +63,12 @@ static const char mlx4_en_version[] =
  */
 
 
-/* Use a XOR rathern than Toeplitz hash function for RSS */
-MLX4_EN_PARM_INT(rss_xor, 0, "Use XOR hash function for RSS");
-
-/* RSS hash type mask - default to <saddr, daddr, sport, dport> */
-MLX4_EN_PARM_INT(rss_mask, 0xf, "RSS hash type bitmask");
+/* Enable RSS TCP traffic */
+MLX4_EN_PARM_INT(tcp_rss, 1,
+		 "Enable RSS for incomming TCP traffic or disabled (0)");
+/* Enable RSS UDP traffic */
+MLX4_EN_PARM_INT(udp_rss, 1,
+		 "Enable RSS for incomming UDP traffic or disabled (0)");
 
 /* Priority pausing */
 MLX4_EN_PARM_INT(pfctx, 0, "Priority based Flow Control policy on TX[7:0]."
@@ -103,8 +104,12 @@ static int mlx4_en_get_profile(struct mlx4_en_dev *mdev)
 	struct mlx4_en_profile *params = &mdev->profile;
 	int i;
 
-	params->rss_xor = (rss_xor != 0);
-	params->rss_mask = rss_mask & 0x1f;
+	params->tcp_rss = tcp_rss;
+	params->udp_rss = udp_rss;
+	if (params->udp_rss && !mdev->dev->caps.udp_rss) {
+		mlx4_warn(mdev, "UDP RSS is not supported on this device.\n");
+		params->udp_rss = 0;
+	}
 	for (i = 1; i <= MLX4_MAX_PORTS; i++) {
 		params->prof[i].rx_pause = 1;
 		params->prof[i].rx_ppp = pfcrx;
diff --git a/drivers/net/mlx4/en_rx.c b/drivers/net/mlx4/en_rx.c
index f421a3d42723..e2126c76d1dc 100644
--- a/drivers/net/mlx4/en_rx.c
+++ b/drivers/net/mlx4/en_rx.c
@@ -859,8 +859,7 @@ int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv)
 	struct mlx4_qp_context context;
 	struct mlx4_en_rss_context *rss_context;
 	void *ptr;
-	int rss_xor = mdev->profile.rss_xor;
-	u8 rss_mask = mdev->profile.rss_mask;
+	u8 rss_mask = 0x3f;
 	int i, qpn;
 	int err = 0;
 	int good_qps = 0;
@@ -906,9 +905,10 @@ int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv)
 	rss_context->base_qpn = cpu_to_be32(ilog2(priv->rx_ring_num) << 24 |
 					    (rss_map->base_qpn));
 	rss_context->default_qpn = cpu_to_be32(rss_map->base_qpn);
-	rss_context->hash_fn = rss_xor & 0x3;
-	rss_context->flags = rss_mask << 2;
+	rss_context->flags = rss_mask;
 
+	if (priv->mdev->profile.udp_rss)
+		rss_context->base_qpn_udp = rss_context->default_qpn;
 	err = mlx4_qp_to_ready(mdev->dev, &priv->res.mtt, &context,
 			       &rss_map->indir_qp, &rss_map->indir_state);
 	if (err)
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 515c6348f32b..b716e1a1b298 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -179,6 +179,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 #define QUERY_DEV_CAP_MAX_GID_OFFSET		0x3b
 #define QUERY_DEV_CAP_RATE_SUPPORT_OFFSET	0x3c
 #define QUERY_DEV_CAP_MAX_PKEY_OFFSET		0x3f
+#define QUERY_DEV_CAP_UDP_RSS_OFFSET		0x42
 #define QUERY_DEV_CAP_ETH_UC_LOOPBACK_OFFSET	0x43
 #define QUERY_DEV_CAP_FLAGS_OFFSET		0x44
 #define QUERY_DEV_CAP_RSVD_UAR_OFFSET		0x48
@@ -270,6 +271,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev_cap->max_msg_sz = 1 << (field & 0x1f);
 	MLX4_GET(stat_rate, outbox, QUERY_DEV_CAP_RATE_SUPPORT_OFFSET);
 	dev_cap->stat_rate_support = stat_rate;
+	MLX4_GET(field, outbox, QUERY_DEV_CAP_UDP_RSS_OFFSET);
+	dev_cap->udp_rss = field & 0x1;
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_ETH_UC_LOOPBACK_OFFSET);
 	dev_cap->loopback_support = field & 0x1;
 	MLX4_GET(dev_cap->flags, outbox, QUERY_DEV_CAP_FLAGS_OFFSET);
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index 443e456c9dab..65cc72eb899d 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -78,6 +78,7 @@ struct mlx4_dev_cap {
 	u16 wavelength[MLX4_MAX_PORTS + 1];
 	u64 trans_code[MLX4_MAX_PORTS + 1];
 	u16 stat_rate_support;
+	int udp_rss;
 	int loopback_support;
 	u32 flags;
 	int reserved_uars;
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index f4791fa2472f..569fa3df381f 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -225,6 +225,7 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.bmme_flags	     = dev_cap->bmme_flags;
 	dev->caps.reserved_lkey	     = dev_cap->reserved_lkey;
 	dev->caps.stat_rate_support  = dev_cap->stat_rate_support;
+	dev->caps.udp_rss	     = dev_cap->udp_rss;
 	dev->caps.loopback_support   = dev_cap->loopback_support;
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 5d8f097d7e06..4036a053ee32 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -318,6 +318,8 @@ struct mlx4_en_port_profile {
 
 struct mlx4_en_profile {
 	int rss_xor;
+	int tcp_rss;
+	int udp_rss;
 	u8 rss_mask;
 	u32 active_ports;
 	u32 small_pkt_int;
@@ -360,6 +362,7 @@ struct mlx4_en_rss_context {
 	u8 hash_fn;
 	u8 flags;
 	__be32 rss_key[10];
+	__be32 base_qpn_udp;
 };
 
 struct mlx4_en_port_state {
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 2a36a344fb3d..7338654c02b4 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -233,6 +233,7 @@ struct mlx4_caps {
 	u32			bmme_flags;
 	u32			reserved_lkey;
 	u16			stat_rate_support;
+	int			udp_rss;
 	int			loopback_support;
 	u8			port_width_cap[MLX4_MAX_PORTS + 1];
 	int			max_gso_sz;
-- 
cgit v1.2.3


From 4c5f7d7a1e6cf20ad515dad8a63c0813fac5bcea Mon Sep 17 00:00:00 2001
From: Kalle Valo <kvalo@adurom.com>
Date: Sun, 22 Aug 2010 22:46:28 +0300
Subject: wl12xx: change contact person for the include file

Luciano should be the contact person for the include/linux/spi/wl12xx.h file.

Signed-off-by: Kalle Valo <kvalo@adurom.com>
Acked-by: Luciano Coelho <luciano.coelho@nokia.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/spi/wl12xx.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/spi/wl12xx.h b/include/linux/spi/wl12xx.h
index a223ecbc71ef..a20bccf0b5c2 100644
--- a/include/linux/spi/wl12xx.h
+++ b/include/linux/spi/wl12xx.h
@@ -3,7 +3,7 @@
  *
  * Copyright (C) 2009 Nokia Corporation
  *
- * Contact: Kalle Valo <kalle.valo@nokia.com>
+ * Contact: Luciano Coelho <luciano.coelho@nokia.com>
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
-- 
cgit v1.2.3


From ad01b7d480a4a135f974afd5c617c417e0b0542f Mon Sep 17 00:00:00 2001
From: Giuseppe CAVALLARO <peppe.cavallaro@st.com>
Date: Mon, 23 Aug 2010 20:40:42 +0000
Subject: stmmac: make ioaddr 'void __iomem *' rather than unsigned long

This avoids unnecessary casting and adds the ioaddr in the
private structure.
This patch also removes many warning when compile the driver.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/stmmac/common.h         |  50 +++++++--------
 drivers/net/stmmac/dwmac1000_core.c |  18 +++---
 drivers/net/stmmac/dwmac1000_dma.c  |   8 +--
 drivers/net/stmmac/dwmac100_core.c  |  20 +++---
 drivers/net/stmmac/dwmac100_dma.c   |   8 +--
 drivers/net/stmmac/dwmac_dma.h      |  16 ++---
 drivers/net/stmmac/dwmac_lib.c      |  22 +++----
 drivers/net/stmmac/enh_desc.c       |   2 +-
 drivers/net/stmmac/norm_desc.c      |   2 +-
 drivers/net/stmmac/stmmac.h         |   3 +-
 drivers/net/stmmac/stmmac_ethtool.c |  21 +++---
 drivers/net/stmmac/stmmac_main.c    | 124 +++++++++++++++++-------------------
 drivers/net/stmmac/stmmac_mdio.c    |  21 +++---
 include/linux/stmmac.h              |   2 +-
 14 files changed, 153 insertions(+), 164 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/stmmac/common.h b/drivers/net/stmmac/common.h
index 66b9da0260fe..e8cbcb5c206e 100644
--- a/drivers/net/stmmac/common.h
+++ b/drivers/net/stmmac/common.h
@@ -167,7 +167,7 @@ struct stmmac_desc_ops {
 	int (*get_tx_ls) (struct dma_desc *p);
 	/* Return the transmit status looking at the TDES1 */
 	int (*tx_status) (void *data, struct stmmac_extra_stats *x,
-			  struct dma_desc *p, unsigned long ioaddr);
+			  struct dma_desc *p, void __iomem *ioaddr);
 	/* Get the buffer size from the descriptor */
 	int (*get_tx_len) (struct dma_desc *p);
 	/* Handle extra events on specific interrupts hw dependent */
@@ -182,44 +182,44 @@ struct stmmac_desc_ops {
 
 struct stmmac_dma_ops {
 	/* DMA core initialization */
-	int (*init) (unsigned long ioaddr, int pbl, u32 dma_tx, u32 dma_rx);
+	int (*init) (void __iomem *ioaddr, int pbl, u32 dma_tx, u32 dma_rx);
 	/* Dump DMA registers */
-	void (*dump_regs) (unsigned long ioaddr);
+	void (*dump_regs) (void __iomem *ioaddr);
 	/* Set tx/rx threshold in the csr6 register
 	 * An invalid value enables the store-and-forward mode */
-	void (*dma_mode) (unsigned long ioaddr, int txmode, int rxmode);
+	void (*dma_mode) (void __iomem *ioaddr, int txmode, int rxmode);
 	/* To track extra statistic (if supported) */
 	void (*dma_diagnostic_fr) (void *data, struct stmmac_extra_stats *x,
-				   unsigned long ioaddr);
-	void (*enable_dma_transmission) (unsigned long ioaddr);
-	void (*enable_dma_irq) (unsigned long ioaddr);
-	void (*disable_dma_irq) (unsigned long ioaddr);
-	void (*start_tx) (unsigned long ioaddr);
-	void (*stop_tx) (unsigned long ioaddr);
-	void (*start_rx) (unsigned long ioaddr);
-	void (*stop_rx) (unsigned long ioaddr);
-	int (*dma_interrupt) (unsigned long ioaddr,
+				   void __iomem *ioaddr);
+	void (*enable_dma_transmission) (void __iomem *ioaddr);
+	void (*enable_dma_irq) (void __iomem *ioaddr);
+	void (*disable_dma_irq) (void __iomem *ioaddr);
+	void (*start_tx) (void __iomem *ioaddr);
+	void (*stop_tx) (void __iomem *ioaddr);
+	void (*start_rx) (void __iomem *ioaddr);
+	void (*stop_rx) (void __iomem *ioaddr);
+	int (*dma_interrupt) (void __iomem *ioaddr,
 			      struct stmmac_extra_stats *x);
 };
 
 struct stmmac_ops {
 	/* MAC core initialization */
-	void (*core_init) (unsigned long ioaddr) ____cacheline_aligned;
+	void (*core_init) (void __iomem *ioaddr) ____cacheline_aligned;
 	/* Dump MAC registers */
-	void (*dump_regs) (unsigned long ioaddr);
+	void (*dump_regs) (void __iomem *ioaddr);
 	/* Handle extra events on specific interrupts hw dependent */
-	void (*host_irq_status) (unsigned long ioaddr);
+	void (*host_irq_status) (void __iomem *ioaddr);
 	/* Multicast filter setting */
 	void (*set_filter) (struct net_device *dev);
 	/* Flow control setting */
-	void (*flow_ctrl) (unsigned long ioaddr, unsigned int duplex,
+	void (*flow_ctrl) (void __iomem *ioaddr, unsigned int duplex,
 			   unsigned int fc, unsigned int pause_time);
 	/* Set power management mode (e.g. magic frame) */
-	void (*pmt) (unsigned long ioaddr, unsigned long mode);
+	void (*pmt) (void __iomem *ioaddr, unsigned long mode);
 	/* Set/Get Unicast MAC addresses */
-	void (*set_umac_addr) (unsigned long ioaddr, unsigned char *addr,
+	void (*set_umac_addr) (void __iomem *ioaddr, unsigned char *addr,
 			       unsigned int reg_n);
-	void (*get_umac_addr) (unsigned long ioaddr, unsigned char *addr,
+	void (*get_umac_addr) (void __iomem *ioaddr, unsigned char *addr,
 			       unsigned int reg_n);
 };
 
@@ -243,11 +243,11 @@ struct mac_device_info {
 	struct mac_link link;
 };
 
-struct mac_device_info *dwmac1000_setup(unsigned long addr);
-struct mac_device_info *dwmac100_setup(unsigned long addr);
+struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr);
+struct mac_device_info *dwmac100_setup(void __iomem *ioaddr);
 
-extern void stmmac_set_mac_addr(unsigned long ioaddr, u8 addr[6],
+extern void stmmac_set_mac_addr(void __iomem *ioaddr, u8 addr[6],
 				unsigned int high, unsigned int low);
-extern void stmmac_get_mac_addr(unsigned long ioaddr, unsigned char *addr,
+extern void stmmac_get_mac_addr(void __iomem *ioaddr, unsigned char *addr,
 				unsigned int high, unsigned int low);
-extern void dwmac_dma_flush_tx_fifo(unsigned long ioaddr);
+extern void dwmac_dma_flush_tx_fifo(void __iomem *ioaddr);
diff --git a/drivers/net/stmmac/dwmac1000_core.c b/drivers/net/stmmac/dwmac1000_core.c
index 2b2f5c8caf1c..8bbfc0f48dea 100644
--- a/drivers/net/stmmac/dwmac1000_core.c
+++ b/drivers/net/stmmac/dwmac1000_core.c
@@ -30,7 +30,7 @@
 #include <linux/slab.h>
 #include "dwmac1000.h"
 
-static void dwmac1000_core_init(unsigned long ioaddr)
+static void dwmac1000_core_init(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + GMAC_CONTROL);
 	value |= GMAC_CORE_INIT;
@@ -50,7 +50,7 @@ static void dwmac1000_core_init(unsigned long ioaddr)
 #endif
 }
 
-static void dwmac1000_dump_regs(unsigned long ioaddr)
+static void dwmac1000_dump_regs(void __iomem *ioaddr)
 {
 	int i;
 	pr_info("\tDWMAC1000 regs (base addr = 0x%8x)\n", (unsigned int)ioaddr);
@@ -62,14 +62,14 @@ static void dwmac1000_dump_regs(unsigned long ioaddr)
 	}
 }
 
-static void dwmac1000_set_umac_addr(unsigned long ioaddr, unsigned char *addr,
+static void dwmac1000_set_umac_addr(void __iomem *ioaddr, unsigned char *addr,
 				unsigned int reg_n)
 {
 	stmmac_set_mac_addr(ioaddr, addr, GMAC_ADDR_HIGH(reg_n),
 				GMAC_ADDR_LOW(reg_n));
 }
 
-static void dwmac1000_get_umac_addr(unsigned long ioaddr, unsigned char *addr,
+static void dwmac1000_get_umac_addr(void __iomem *ioaddr, unsigned char *addr,
 				unsigned int reg_n)
 {
 	stmmac_get_mac_addr(ioaddr, addr, GMAC_ADDR_HIGH(reg_n),
@@ -78,7 +78,7 @@ static void dwmac1000_get_umac_addr(unsigned long ioaddr, unsigned char *addr,
 
 static void dwmac1000_set_filter(struct net_device *dev)
 {
-	unsigned long ioaddr = dev->base_addr;
+	void __iomem *ioaddr = (void __iomem *) dev->base_addr;
 	unsigned int value = 0;
 
 	CHIP_DBG(KERN_INFO "%s: # mcasts %d, # unicast %d\n",
@@ -139,7 +139,7 @@ static void dwmac1000_set_filter(struct net_device *dev)
 	    readl(ioaddr + GMAC_HASH_HIGH), readl(ioaddr + GMAC_HASH_LOW));
 }
 
-static void dwmac1000_flow_ctrl(unsigned long ioaddr, unsigned int duplex,
+static void dwmac1000_flow_ctrl(void __iomem *ioaddr, unsigned int duplex,
 			   unsigned int fc, unsigned int pause_time)
 {
 	unsigned int flow = 0;
@@ -162,7 +162,7 @@ static void dwmac1000_flow_ctrl(unsigned long ioaddr, unsigned int duplex,
 	writel(flow, ioaddr + GMAC_FLOW_CTRL);
 }
 
-static void dwmac1000_pmt(unsigned long ioaddr, unsigned long mode)
+static void dwmac1000_pmt(void __iomem *ioaddr, unsigned long mode)
 {
 	unsigned int pmt = 0;
 
@@ -178,7 +178,7 @@ static void dwmac1000_pmt(unsigned long ioaddr, unsigned long mode)
 }
 
 
-static void dwmac1000_irq_status(unsigned long ioaddr)
+static void dwmac1000_irq_status(void __iomem *ioaddr)
 {
 	u32 intr_status = readl(ioaddr + GMAC_INT_STATUS);
 
@@ -211,7 +211,7 @@ struct stmmac_ops dwmac1000_ops = {
 	.get_umac_addr = dwmac1000_get_umac_addr,
 };
 
-struct mac_device_info *dwmac1000_setup(unsigned long ioaddr)
+struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr)
 {
 	struct mac_device_info *mac;
 	u32 uid = readl(ioaddr + GMAC_VERSION);
diff --git a/drivers/net/stmmac/dwmac1000_dma.c b/drivers/net/stmmac/dwmac1000_dma.c
index 415805057cb0..2ef5a56370e9 100644
--- a/drivers/net/stmmac/dwmac1000_dma.c
+++ b/drivers/net/stmmac/dwmac1000_dma.c
@@ -29,7 +29,7 @@
 #include "dwmac1000.h"
 #include "dwmac_dma.h"
 
-static int dwmac1000_dma_init(unsigned long ioaddr, int pbl, u32 dma_tx,
+static int dwmac1000_dma_init(void __iomem *ioaddr, int pbl, u32 dma_tx,
 			      u32 dma_rx)
 {
 	u32 value = readl(ioaddr + DMA_BUS_MODE);
@@ -58,7 +58,7 @@ static int dwmac1000_dma_init(unsigned long ioaddr, int pbl, u32 dma_tx,
 	return 0;
 }
 
-static void dwmac1000_dma_operation_mode(unsigned long ioaddr, int txmode,
+static void dwmac1000_dma_operation_mode(void __iomem *ioaddr, int txmode,
 				    int rxmode)
 {
 	u32 csr6 = readl(ioaddr + DMA_CONTROL);
@@ -111,12 +111,12 @@ static void dwmac1000_dma_operation_mode(unsigned long ioaddr, int txmode,
 
 /* Not yet implemented --- no RMON module */
 static void dwmac1000_dma_diagnostic_fr(void *data,
-		  struct stmmac_extra_stats *x, unsigned long ioaddr)
+		  struct stmmac_extra_stats *x, void __iomem *ioaddr)
 {
 	return;
 }
 
-static void dwmac1000_dump_dma_regs(unsigned long ioaddr)
+static void dwmac1000_dump_dma_regs(void __iomem *ioaddr)
 {
 	int i;
 	pr_info(" DMA registers\n");
diff --git a/drivers/net/stmmac/dwmac100_core.c b/drivers/net/stmmac/dwmac100_core.c
index 2fb165fa2ba0..135a8082816e 100644
--- a/drivers/net/stmmac/dwmac100_core.c
+++ b/drivers/net/stmmac/dwmac100_core.c
@@ -31,7 +31,7 @@
 #include <linux/crc32.h>
 #include "dwmac100.h"
 
-static void dwmac100_core_init(unsigned long ioaddr)
+static void dwmac100_core_init(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + MAC_CONTROL);
 
@@ -42,12 +42,12 @@ static void dwmac100_core_init(unsigned long ioaddr)
 #endif
 }
 
-static void dwmac100_dump_mac_regs(unsigned long ioaddr)
+static void dwmac100_dump_mac_regs(void __iomem *ioaddr)
 {
 	pr_info("\t----------------------------------------------\n"
 		"\t  DWMAC 100 CSR (base addr = 0x%8x)\n"
 		"\t----------------------------------------------\n",
-		(unsigned int)ioaddr);
+		(unsigned int) ioaddr);
 	pr_info("\tcontrol reg (offset 0x%x): 0x%08x\n", MAC_CONTROL,
 		readl(ioaddr + MAC_CONTROL));
 	pr_info("\taddr HI (offset 0x%x): 0x%08x\n ", MAC_ADDR_HIGH,
@@ -77,18 +77,18 @@ static void dwmac100_dump_mac_regs(unsigned long ioaddr)
 		MMC_LOW_INTR_MASK, readl(ioaddr + MMC_LOW_INTR_MASK));
 }
 
-static void dwmac100_irq_status(unsigned long ioaddr)
+static void dwmac100_irq_status(void __iomem *ioaddr)
 {
 	return;
 }
 
-static void dwmac100_set_umac_addr(unsigned long ioaddr, unsigned char *addr,
+static void dwmac100_set_umac_addr(void __iomem *ioaddr, unsigned char *addr,
 				   unsigned int reg_n)
 {
 	stmmac_set_mac_addr(ioaddr, addr, MAC_ADDR_HIGH, MAC_ADDR_LOW);
 }
 
-static void dwmac100_get_umac_addr(unsigned long ioaddr, unsigned char *addr,
+static void dwmac100_get_umac_addr(void __iomem *ioaddr, unsigned char *addr,
 				   unsigned int reg_n)
 {
 	stmmac_get_mac_addr(ioaddr, addr, MAC_ADDR_HIGH, MAC_ADDR_LOW);
@@ -96,7 +96,7 @@ static void dwmac100_get_umac_addr(unsigned long ioaddr, unsigned char *addr,
 
 static void dwmac100_set_filter(struct net_device *dev)
 {
-	unsigned long ioaddr = dev->base_addr;
+	void __iomem *ioaddr = (void __iomem *) dev->base_addr;
 	u32 value = readl(ioaddr + MAC_CONTROL);
 
 	if (dev->flags & IFF_PROMISC) {
@@ -145,7 +145,7 @@ static void dwmac100_set_filter(struct net_device *dev)
 	    readl(ioaddr + MAC_HASH_HIGH), readl(ioaddr + MAC_HASH_LOW));
 }
 
-static void dwmac100_flow_ctrl(unsigned long ioaddr, unsigned int duplex,
+static void dwmac100_flow_ctrl(void __iomem *ioaddr, unsigned int duplex,
 			       unsigned int fc, unsigned int pause_time)
 {
 	unsigned int flow = MAC_FLOW_CTRL_ENABLE;
@@ -158,7 +158,7 @@ static void dwmac100_flow_ctrl(unsigned long ioaddr, unsigned int duplex,
 /* No PMT module supported for this Ethernet Controller.
  * Tested on ST platforms only.
  */
-static void dwmac100_pmt(unsigned long ioaddr, unsigned long mode)
+static void dwmac100_pmt(void __iomem *ioaddr, unsigned long mode)
 {
 	return;
 }
@@ -174,7 +174,7 @@ struct stmmac_ops dwmac100_ops = {
 	.get_umac_addr = dwmac100_get_umac_addr,
 };
 
-struct mac_device_info *dwmac100_setup(unsigned long ioaddr)
+struct mac_device_info *dwmac100_setup(void __iomem *ioaddr)
 {
 	struct mac_device_info *mac;
 
diff --git a/drivers/net/stmmac/dwmac100_dma.c b/drivers/net/stmmac/dwmac100_dma.c
index 2fece7b72727..c7279d2b946b 100644
--- a/drivers/net/stmmac/dwmac100_dma.c
+++ b/drivers/net/stmmac/dwmac100_dma.c
@@ -31,7 +31,7 @@
 #include "dwmac100.h"
 #include "dwmac_dma.h"
 
-static int dwmac100_dma_init(unsigned long ioaddr, int pbl, u32 dma_tx,
+static int dwmac100_dma_init(void __iomem *ioaddr, int pbl, u32 dma_tx,
 			     u32 dma_rx)
 {
 	u32 value = readl(ioaddr + DMA_BUS_MODE);
@@ -58,7 +58,7 @@ static int dwmac100_dma_init(unsigned long ioaddr, int pbl, u32 dma_tx,
 /* Store and Forward capability is not used at all..
  * The transmit threshold can be programmed by
  * setting the TTC bits in the DMA control register.*/
-static void dwmac100_dma_operation_mode(unsigned long ioaddr, int txmode,
+static void dwmac100_dma_operation_mode(void __iomem *ioaddr, int txmode,
 					int rxmode)
 {
 	u32 csr6 = readl(ioaddr + DMA_CONTROL);
@@ -73,7 +73,7 @@ static void dwmac100_dma_operation_mode(unsigned long ioaddr, int txmode,
 	writel(csr6, ioaddr + DMA_CONTROL);
 }
 
-static void dwmac100_dump_dma_regs(unsigned long ioaddr)
+static void dwmac100_dump_dma_regs(void __iomem *ioaddr)
 {
 	int i;
 
@@ -91,7 +91,7 @@ static void dwmac100_dump_dma_regs(unsigned long ioaddr)
 /* DMA controller has two counters to track the number of
  * the receive missed frames. */
 static void dwmac100_dma_diagnostic_fr(void *data, struct stmmac_extra_stats *x,
-				       unsigned long ioaddr)
+				       void __iomem *ioaddr)
 {
 	struct net_device_stats *stats = (struct net_device_stats *)data;
 	u32 csr8 = readl(ioaddr + DMA_MISSED_FRAME_CTR);
diff --git a/drivers/net/stmmac/dwmac_dma.h b/drivers/net/stmmac/dwmac_dma.h
index 7b815a1b7b8c..da3f5ccf83d3 100644
--- a/drivers/net/stmmac/dwmac_dma.h
+++ b/drivers/net/stmmac/dwmac_dma.h
@@ -97,12 +97,12 @@
 #define DMA_STATUS_TI	0x00000001	/* Transmit Interrupt */
 #define DMA_CONTROL_FTF		0x00100000 /* Flush transmit FIFO */
 
-extern void dwmac_enable_dma_transmission(unsigned long ioaddr);
-extern void dwmac_enable_dma_irq(unsigned long ioaddr);
-extern void dwmac_disable_dma_irq(unsigned long ioaddr);
-extern void dwmac_dma_start_tx(unsigned long ioaddr);
-extern void dwmac_dma_stop_tx(unsigned long ioaddr);
-extern void dwmac_dma_start_rx(unsigned long ioaddr);
-extern void dwmac_dma_stop_rx(unsigned long ioaddr);
-extern int dwmac_dma_interrupt(unsigned long ioaddr,
+extern void dwmac_enable_dma_transmission(void __iomem *ioaddr);
+extern void dwmac_enable_dma_irq(void __iomem *ioaddr);
+extern void dwmac_disable_dma_irq(void __iomem *ioaddr);
+extern void dwmac_dma_start_tx(void __iomem *ioaddr);
+extern void dwmac_dma_stop_tx(void __iomem *ioaddr);
+extern void dwmac_dma_start_rx(void __iomem *ioaddr);
+extern void dwmac_dma_stop_rx(void __iomem *ioaddr);
+extern int dwmac_dma_interrupt(void __iomem *ioaddr,
 				struct stmmac_extra_stats *x);
diff --git a/drivers/net/stmmac/dwmac_lib.c b/drivers/net/stmmac/dwmac_lib.c
index a85415216ef4..d65fab1ba790 100644
--- a/drivers/net/stmmac/dwmac_lib.c
+++ b/drivers/net/stmmac/dwmac_lib.c
@@ -32,43 +32,43 @@
 #endif
 
 /* CSR1 enables the transmit DMA to check for new descriptor */
-void dwmac_enable_dma_transmission(unsigned long ioaddr)
+void dwmac_enable_dma_transmission(void __iomem *ioaddr)
 {
 	writel(1, ioaddr + DMA_XMT_POLL_DEMAND);
 }
 
-void dwmac_enable_dma_irq(unsigned long ioaddr)
+void dwmac_enable_dma_irq(void __iomem *ioaddr)
 {
 	writel(DMA_INTR_DEFAULT_MASK, ioaddr + DMA_INTR_ENA);
 }
 
-void dwmac_disable_dma_irq(unsigned long ioaddr)
+void dwmac_disable_dma_irq(void __iomem *ioaddr)
 {
 	writel(0, ioaddr + DMA_INTR_ENA);
 }
 
-void dwmac_dma_start_tx(unsigned long ioaddr)
+void dwmac_dma_start_tx(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + DMA_CONTROL);
 	value |= DMA_CONTROL_ST;
 	writel(value, ioaddr + DMA_CONTROL);
 }
 
-void dwmac_dma_stop_tx(unsigned long ioaddr)
+void dwmac_dma_stop_tx(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + DMA_CONTROL);
 	value &= ~DMA_CONTROL_ST;
 	writel(value, ioaddr + DMA_CONTROL);
 }
 
-void dwmac_dma_start_rx(unsigned long ioaddr)
+void dwmac_dma_start_rx(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + DMA_CONTROL);
 	value |= DMA_CONTROL_SR;
 	writel(value, ioaddr + DMA_CONTROL);
 }
 
-void dwmac_dma_stop_rx(unsigned long ioaddr)
+void dwmac_dma_stop_rx(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + DMA_CONTROL);
 	value &= ~DMA_CONTROL_SR;
@@ -145,7 +145,7 @@ static void show_rx_process_state(unsigned int status)
 }
 #endif
 
-int dwmac_dma_interrupt(unsigned long ioaddr,
+int dwmac_dma_interrupt(void __iomem *ioaddr,
 			struct stmmac_extra_stats *x)
 {
 	int ret = 0;
@@ -219,7 +219,7 @@ int dwmac_dma_interrupt(unsigned long ioaddr,
 	return ret;
 }
 
-void dwmac_dma_flush_tx_fifo(unsigned long ioaddr)
+void dwmac_dma_flush_tx_fifo(void __iomem *ioaddr)
 {
 	u32 csr6 = readl(ioaddr + DMA_CONTROL);
 	writel((csr6 | DMA_CONTROL_FTF), ioaddr + DMA_CONTROL);
@@ -227,7 +227,7 @@ void dwmac_dma_flush_tx_fifo(unsigned long ioaddr)
 	do {} while ((readl(ioaddr + DMA_CONTROL) & DMA_CONTROL_FTF));
 }
 
-void stmmac_set_mac_addr(unsigned long ioaddr, u8 addr[6],
+void stmmac_set_mac_addr(void __iomem *ioaddr, u8 addr[6],
 			 unsigned int high, unsigned int low)
 {
 	unsigned long data;
@@ -238,7 +238,7 @@ void stmmac_set_mac_addr(unsigned long ioaddr, u8 addr[6],
 	writel(data, ioaddr + low);
 }
 
-void stmmac_get_mac_addr(unsigned long ioaddr, unsigned char *addr,
+void stmmac_get_mac_addr(void __iomem *ioaddr, unsigned char *addr,
 			 unsigned int high, unsigned int low)
 {
 	unsigned int hi_addr, lo_addr;
diff --git a/drivers/net/stmmac/enh_desc.c b/drivers/net/stmmac/enh_desc.c
index f612f986a7e1..77ff88c3958b 100644
--- a/drivers/net/stmmac/enh_desc.c
+++ b/drivers/net/stmmac/enh_desc.c
@@ -25,7 +25,7 @@
 #include "common.h"
 
 static int enh_desc_get_tx_status(void *data, struct stmmac_extra_stats *x,
-				  struct dma_desc *p, unsigned long ioaddr)
+				  struct dma_desc *p, void __iomem *ioaddr)
 {
 	int ret = 0;
 	struct net_device_stats *stats = (struct net_device_stats *)data;
diff --git a/drivers/net/stmmac/norm_desc.c b/drivers/net/stmmac/norm_desc.c
index 31ad53643792..51f4440ab98b 100644
--- a/drivers/net/stmmac/norm_desc.c
+++ b/drivers/net/stmmac/norm_desc.c
@@ -25,7 +25,7 @@
 #include "common.h"
 
 static int ndesc_get_tx_status(void *data, struct stmmac_extra_stats *x,
-			       struct dma_desc *p, unsigned long ioaddr)
+			       struct dma_desc *p, void __iomem *ioaddr)
 {
 	int ret = 0;
 	struct net_device_stats *stats = (struct net_device_stats *)data;
diff --git a/drivers/net/stmmac/stmmac.h b/drivers/net/stmmac/stmmac.h
index ebebc644b1b8..cca53dbac361 100644
--- a/drivers/net/stmmac/stmmac.h
+++ b/drivers/net/stmmac/stmmac.h
@@ -54,6 +54,7 @@ struct stmmac_priv {
 	unsigned int dma_buf_sz;
 	struct device *device;
 	struct mac_device_info *hw;
+	void __iomem *ioaddr;
 
 	struct stmmac_extra_stats xstats;
 	struct napi_struct napi;
@@ -65,7 +66,7 @@ struct stmmac_priv {
 	int phy_mask;
 	int (*phy_reset) (void *priv);
 	void (*fix_mac_speed) (void *priv, unsigned int speed);
-	void (*bus_setup)(unsigned long ioaddr);
+	void (*bus_setup)(void __iomem *ioaddr);
 	void *bsp_priv;
 
 	int phy_irq;
diff --git a/drivers/net/stmmac/stmmac_ethtool.c b/drivers/net/stmmac/stmmac_ethtool.c
index f080509923f0..63b68e61afce 100644
--- a/drivers/net/stmmac/stmmac_ethtool.c
+++ b/drivers/net/stmmac/stmmac_ethtool.c
@@ -177,21 +177,21 @@ void stmmac_ethtool_gregs(struct net_device *dev,
 	if (!priv->is_gmac) {
 		/* MAC registers */
 		for (i = 0; i < 12; i++)
-			reg_space[i] = readl(dev->base_addr + (i * 4));
+			reg_space[i] = readl(priv->ioaddr + (i * 4));
 		/* DMA registers */
 		for (i = 0; i < 9; i++)
 			reg_space[i + 12] =
-			    readl(dev->base_addr + (DMA_BUS_MODE + (i * 4)));
-		reg_space[22] = readl(dev->base_addr + DMA_CUR_TX_BUF_ADDR);
-		reg_space[23] = readl(dev->base_addr + DMA_CUR_RX_BUF_ADDR);
+			    readl(priv->ioaddr + (DMA_BUS_MODE + (i * 4)));
+		reg_space[22] = readl(priv->ioaddr + DMA_CUR_TX_BUF_ADDR);
+		reg_space[23] = readl(priv->ioaddr + DMA_CUR_RX_BUF_ADDR);
 	} else {
 		/* MAC registers */
 		for (i = 0; i < 55; i++)
-			reg_space[i] = readl(dev->base_addr + (i * 4));
+			reg_space[i] = readl(priv->ioaddr + (i * 4));
 		/* DMA registers */
 		for (i = 0; i < 22; i++)
 			reg_space[i + 55] =
-			    readl(dev->base_addr + (DMA_BUS_MODE + (i * 4)));
+			    readl(priv->ioaddr + (DMA_BUS_MODE + (i * 4)));
 	}
 }
 
@@ -263,11 +263,9 @@ stmmac_set_pauseparam(struct net_device *netdev,
 			cmd.phy_address = phy->addr;
 			ret = phy_ethtool_sset(phy, &cmd);
 		}
-	} else {
-		unsigned long ioaddr = netdev->base_addr;
-		priv->hw->mac->flow_ctrl(ioaddr, phy->duplex,
+	} else
+		priv->hw->mac->flow_ctrl(priv->ioaddr, phy->duplex,
 					 priv->flow_ctrl, priv->pause);
-	}
 	spin_unlock(&priv->lock);
 	return ret;
 }
@@ -276,12 +274,11 @@ static void stmmac_get_ethtool_stats(struct net_device *dev,
 				 struct ethtool_stats *dummy, u64 *data)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
-	unsigned long ioaddr = dev->base_addr;
 	int i;
 
 	/* Update HW stats if supported */
 	priv->hw->dma->dma_diagnostic_fr(&dev->stats, (void *) &priv->xstats,
-					 ioaddr);
+					 priv->ioaddr);
 
 	for (i = 0; i < STMMAC_STATS_LEN; i++) {
 		char *p = (char *)priv + stmmac_gstrings_stats[i].stat_offset;
diff --git a/drivers/net/stmmac/stmmac_main.c b/drivers/net/stmmac/stmmac_main.c
index 86b6c69c068c..c59c1061252a 100644
--- a/drivers/net/stmmac/stmmac_main.c
+++ b/drivers/net/stmmac/stmmac_main.c
@@ -202,7 +202,6 @@ static void stmmac_adjust_link(struct net_device *dev)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
 	struct phy_device *phydev = priv->phydev;
-	unsigned long ioaddr = dev->base_addr;
 	unsigned long flags;
 	int new_state = 0;
 	unsigned int fc = priv->flow_ctrl, pause_time = priv->pause;
@@ -215,7 +214,7 @@ static void stmmac_adjust_link(struct net_device *dev)
 
 	spin_lock_irqsave(&priv->lock, flags);
 	if (phydev->link) {
-		u32 ctrl = readl(ioaddr + MAC_CTRL_REG);
+		u32 ctrl = readl(priv->ioaddr + MAC_CTRL_REG);
 
 		/* Now we make sure that we can be in full duplex mode.
 		 * If not, we operate in half-duplex mode. */
@@ -229,7 +228,7 @@ static void stmmac_adjust_link(struct net_device *dev)
 		}
 		/* Flow Control operation */
 		if (phydev->pause)
-			priv->hw->mac->flow_ctrl(ioaddr, phydev->duplex,
+			priv->hw->mac->flow_ctrl(priv->ioaddr, phydev->duplex,
 						 fc, pause_time);
 
 		if (phydev->speed != priv->speed) {
@@ -268,7 +267,7 @@ static void stmmac_adjust_link(struct net_device *dev)
 			priv->speed = phydev->speed;
 		}
 
-		writel(ctrl, ioaddr + MAC_CTRL_REG);
+		writel(ctrl, priv->ioaddr + MAC_CTRL_REG);
 
 		if (!priv->oldlink) {
 			new_state = 1;
@@ -345,7 +344,7 @@ static int stmmac_init_phy(struct net_device *dev)
 	return 0;
 }
 
-static inline void stmmac_mac_enable_rx(unsigned long ioaddr)
+static inline void stmmac_mac_enable_rx(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + MAC_CTRL_REG);
 	value |= MAC_RNABLE_RX;
@@ -353,7 +352,7 @@ static inline void stmmac_mac_enable_rx(unsigned long ioaddr)
 	writel(value, ioaddr + MAC_CTRL_REG);
 }
 
-static inline void stmmac_mac_enable_tx(unsigned long ioaddr)
+static inline void stmmac_mac_enable_tx(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + MAC_CTRL_REG);
 	value |= MAC_ENABLE_TX;
@@ -361,14 +360,14 @@ static inline void stmmac_mac_enable_tx(unsigned long ioaddr)
 	writel(value, ioaddr + MAC_CTRL_REG);
 }
 
-static inline void stmmac_mac_disable_rx(unsigned long ioaddr)
+static inline void stmmac_mac_disable_rx(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + MAC_CTRL_REG);
 	value &= ~MAC_RNABLE_RX;
 	writel(value, ioaddr + MAC_CTRL_REG);
 }
 
-static inline void stmmac_mac_disable_tx(unsigned long ioaddr)
+static inline void stmmac_mac_disable_tx(void __iomem *ioaddr)
 {
 	u32 value = readl(ioaddr + MAC_CTRL_REG);
 	value &= ~MAC_ENABLE_TX;
@@ -577,17 +576,17 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
 {
 	if (!priv->is_gmac) {
 		/* MAC 10/100 */
-		priv->hw->dma->dma_mode(priv->dev->base_addr, tc, 0);
+		priv->hw->dma->dma_mode(priv->ioaddr, tc, 0);
 		priv->tx_coe = NO_HW_CSUM;
 	} else {
 		if ((priv->dev->mtu <= ETH_DATA_LEN) && (tx_coe)) {
-			priv->hw->dma->dma_mode(priv->dev->base_addr,
+			priv->hw->dma->dma_mode(priv->ioaddr,
 						SF_DMA_MODE, SF_DMA_MODE);
 			tc = SF_DMA_MODE;
 			priv->tx_coe = HW_CSUM;
 		} else {
 			/* Checksum computation is performed in software. */
-			priv->hw->dma->dma_mode(priv->dev->base_addr, tc,
+			priv->hw->dma->dma_mode(priv->ioaddr, tc,
 						SF_DMA_MODE);
 			priv->tx_coe = NO_HW_CSUM;
 		}
@@ -603,7 +602,6 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
 static void stmmac_tx(struct stmmac_priv *priv)
 {
 	unsigned int txsize = priv->dma_tx_size;
-	unsigned long ioaddr = priv->dev->base_addr;
 
 	while (priv->dirty_tx != priv->cur_tx) {
 		int last;
@@ -621,7 +619,7 @@ static void stmmac_tx(struct stmmac_priv *priv)
 			int tx_error =
 				priv->hw->desc->tx_status(&priv->dev->stats,
 							  &priv->xstats, p,
-							  ioaddr);
+							  priv->ioaddr);
 			if (likely(tx_error == 0)) {
 				priv->dev->stats.tx_packets++;
 				priv->xstats.tx_pkt_n++;
@@ -677,7 +675,7 @@ static inline void stmmac_enable_irq(struct stmmac_priv *priv)
 		priv->tm->timer_start(tmrate);
 	else
 #endif
-		priv->hw->dma->enable_dma_irq(priv->dev->base_addr);
+		priv->hw->dma->enable_dma_irq(priv->ioaddr);
 }
 
 static inline void stmmac_disable_irq(struct stmmac_priv *priv)
@@ -687,7 +685,7 @@ static inline void stmmac_disable_irq(struct stmmac_priv *priv)
 		priv->tm->timer_stop();
 	else
 #endif
-		priv->hw->dma->disable_dma_irq(priv->dev->base_addr);
+		priv->hw->dma->disable_dma_irq(priv->ioaddr);
 }
 
 static int stmmac_has_work(struct stmmac_priv *priv)
@@ -742,14 +740,15 @@ static void stmmac_no_timer_stopped(void)
  */
 static void stmmac_tx_err(struct stmmac_priv *priv)
 {
+
 	netif_stop_queue(priv->dev);
 
-	priv->hw->dma->stop_tx(priv->dev->base_addr);
+	priv->hw->dma->stop_tx(priv->ioaddr);
 	dma_free_tx_skbufs(priv);
 	priv->hw->desc->init_tx_desc(priv->dma_tx, priv->dma_tx_size);
 	priv->dirty_tx = 0;
 	priv->cur_tx = 0;
-	priv->hw->dma->start_tx(priv->dev->base_addr);
+	priv->hw->dma->start_tx(priv->ioaddr);
 
 	priv->dev->stats.tx_errors++;
 	netif_wake_queue(priv->dev);
@@ -758,11 +757,9 @@ static void stmmac_tx_err(struct stmmac_priv *priv)
 
 static void stmmac_dma_interrupt(struct stmmac_priv *priv)
 {
-	unsigned long ioaddr = priv->dev->base_addr;
 	int status;
 
-	status = priv->hw->dma->dma_interrupt(priv->dev->base_addr,
-					      &priv->xstats);
+	status = priv->hw->dma->dma_interrupt(priv->ioaddr, &priv->xstats);
 	if (likely(status == handle_tx_rx))
 		_stmmac_schedule(priv);
 
@@ -770,7 +767,7 @@ static void stmmac_dma_interrupt(struct stmmac_priv *priv)
 		/* Try to bump up the dma threshold on this failure */
 		if (unlikely(tc != SF_DMA_MODE) && (tc <= 256)) {
 			tc += 64;
-			priv->hw->dma->dma_mode(ioaddr, tc, SF_DMA_MODE);
+			priv->hw->dma->dma_mode(priv->ioaddr, tc, SF_DMA_MODE);
 			priv->xstats.threshold = tc;
 		}
 		stmmac_tx_err(priv);
@@ -790,7 +787,6 @@ static void stmmac_dma_interrupt(struct stmmac_priv *priv)
 static int stmmac_open(struct net_device *dev)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
-	unsigned long ioaddr = dev->base_addr;
 	int ret;
 
 	/* Check that the MAC address is valid.  If its not, refuse
@@ -846,7 +842,8 @@ static int stmmac_open(struct net_device *dev)
 	init_dma_desc_rings(dev);
 
 	/* DMA initialization and SW reset */
-	if (unlikely(priv->hw->dma->init(ioaddr, priv->pbl, priv->dma_tx_phy,
+	if (unlikely(priv->hw->dma->init(priv->ioaddr, priv->pbl,
+					 priv->dma_tx_phy,
 					 priv->dma_rx_phy) < 0)) {
 
 		pr_err("%s: DMA initialization failed\n", __func__);
@@ -854,22 +851,22 @@ static int stmmac_open(struct net_device *dev)
 	}
 
 	/* Copy the MAC addr into the HW  */
-	priv->hw->mac->set_umac_addr(ioaddr, dev->dev_addr, 0);
+	priv->hw->mac->set_umac_addr(priv->ioaddr, dev->dev_addr, 0);
 	/* If required, perform hw setup of the bus. */
 	if (priv->bus_setup)
-		priv->bus_setup(ioaddr);
+		priv->bus_setup(priv->ioaddr);
 	/* Initialize the MAC Core */
-	priv->hw->mac->core_init(ioaddr);
+	priv->hw->mac->core_init(priv->ioaddr);
 
 	priv->shutdown = 0;
 
 	/* Initialise the MMC (if present) to disable all interrupts. */
-	writel(0xffffffff, ioaddr + MMC_HIGH_INTR_MASK);
-	writel(0xffffffff, ioaddr + MMC_LOW_INTR_MASK);
+	writel(0xffffffff, priv->ioaddr + MMC_HIGH_INTR_MASK);
+	writel(0xffffffff, priv->ioaddr + MMC_LOW_INTR_MASK);
 
 	/* Enable the MAC Rx/Tx */
-	stmmac_mac_enable_rx(ioaddr);
-	stmmac_mac_enable_tx(ioaddr);
+	stmmac_mac_enable_rx(priv->ioaddr);
+	stmmac_mac_enable_tx(priv->ioaddr);
 
 	/* Set the HW DMA mode and the COE */
 	stmmac_dma_operation_mode(priv);
@@ -880,16 +877,16 @@ static int stmmac_open(struct net_device *dev)
 
 	/* Start the ball rolling... */
 	DBG(probe, DEBUG, "%s: DMA RX/TX processes started...\n", dev->name);
-	priv->hw->dma->start_tx(ioaddr);
-	priv->hw->dma->start_rx(ioaddr);
+	priv->hw->dma->start_tx(priv->ioaddr);
+	priv->hw->dma->start_rx(priv->ioaddr);
 
 #ifdef CONFIG_STMMAC_TIMER
 	priv->tm->timer_start(tmrate);
 #endif
 	/* Dump DMA/MAC registers */
 	if (netif_msg_hw(priv)) {
-		priv->hw->mac->dump_regs(ioaddr);
-		priv->hw->dma->dump_regs(ioaddr);
+		priv->hw->mac->dump_regs(priv->ioaddr);
+		priv->hw->dma->dump_regs(priv->ioaddr);
 	}
 
 	if (priv->phydev)
@@ -933,15 +930,15 @@ static int stmmac_release(struct net_device *dev)
 	free_irq(dev->irq, dev);
 
 	/* Stop TX/RX DMA and clear the descriptors */
-	priv->hw->dma->stop_tx(dev->base_addr);
-	priv->hw->dma->stop_rx(dev->base_addr);
+	priv->hw->dma->stop_tx(priv->ioaddr);
+	priv->hw->dma->stop_rx(priv->ioaddr);
 
 	/* Release and free the Rx/Tx resources */
 	free_dma_desc_resources(priv);
 
 	/* Disable the MAC core */
-	stmmac_mac_disable_tx(dev->base_addr);
-	stmmac_mac_disable_rx(dev->base_addr);
+	stmmac_mac_disable_tx(priv->ioaddr);
+	stmmac_mac_disable_rx(priv->ioaddr);
 
 	netif_carrier_off(dev);
 
@@ -1143,7 +1140,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	dev->stats.tx_bytes += skb->len;
 
-	priv->hw->dma->enable_dma_transmission(dev->base_addr);
+	priv->hw->dma->enable_dma_transmission(priv->ioaddr);
 
 	return NETDEV_TX_OK;
 }
@@ -1408,11 +1405,9 @@ static irqreturn_t stmmac_interrupt(int irq, void *dev_id)
 		return IRQ_NONE;
 	}
 
-	if (priv->is_gmac) {
-		unsigned long ioaddr = dev->base_addr;
+	if (priv->is_gmac)
 		/* To handle GMAC own interrupts */
-		priv->hw->mac->host_irq_status(ioaddr);
-	}
+		priv->hw->mac->host_irq_status((void __iomem *) dev->base_addr);
 
 	stmmac_dma_interrupt(priv);
 
@@ -1525,7 +1520,8 @@ static int stmmac_probe(struct net_device *dev)
 	netif_napi_add(dev, &priv->napi, stmmac_poll, 64);
 
 	/* Get the MAC address */
-	priv->hw->mac->get_umac_addr(dev->base_addr, dev->dev_addr, 0);
+	priv->hw->mac->get_umac_addr((void __iomem *) dev->base_addr,
+				     dev->dev_addr, 0);
 
 	if (!is_valid_ether_addr(dev->dev_addr))
 		pr_warning("\tno valid MAC address;"
@@ -1555,14 +1551,13 @@ static int stmmac_probe(struct net_device *dev)
 static int stmmac_mac_device_setup(struct net_device *dev)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
-	unsigned long ioaddr = dev->base_addr;
 
 	struct mac_device_info *device;
 
 	if (priv->is_gmac)
-		device = dwmac1000_setup(ioaddr);
+		device = dwmac1000_setup(priv->ioaddr);
 	else
-		device = dwmac100_setup(ioaddr);
+		device = dwmac100_setup(priv->ioaddr);
 
 	if (!device)
 		return -ENOMEM;
@@ -1656,7 +1651,7 @@ static int stmmac_dvr_probe(struct platform_device *pdev)
 {
 	int ret = 0;
 	struct resource *res;
-	unsigned int *addr = NULL;
+	void __iomem *addr = NULL;
 	struct net_device *ndev = NULL;
 	struct stmmac_priv *priv;
 	struct plat_stmmacenet_data *plat_dat;
@@ -1711,6 +1706,7 @@ static int stmmac_dvr_probe(struct platform_device *pdev)
 	priv->pbl = plat_dat->pbl;	/* TLI */
 	priv->is_gmac = plat_dat->has_gmac;	/* GMAC is on board */
 	priv->enh_desc = plat_dat->enh_desc;
+	priv->ioaddr = addr;
 
 	platform_set_drvdata(pdev, ndev);
 
@@ -1782,11 +1778,11 @@ static int stmmac_dvr_remove(struct platform_device *pdev)
 
 	pr_info("%s:\n\tremoving driver", __func__);
 
-	priv->hw->dma->stop_rx(ndev->base_addr);
-	priv->hw->dma->stop_tx(ndev->base_addr);
+	priv->hw->dma->stop_rx(priv->ioaddr);
+	priv->hw->dma->stop_tx(priv->ioaddr);
 
-	stmmac_mac_disable_rx(ndev->base_addr);
-	stmmac_mac_disable_tx(ndev->base_addr);
+	stmmac_mac_disable_rx(priv->ioaddr);
+	stmmac_mac_disable_tx(priv->ioaddr);
 
 	netif_carrier_off(ndev);
 
@@ -1795,7 +1791,7 @@ static int stmmac_dvr_remove(struct platform_device *pdev)
 	platform_set_drvdata(pdev, NULL);
 	unregister_netdev(ndev);
 
-	iounmap((void *)ndev->base_addr);
+	iounmap((void *)priv->ioaddr);
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	release_mem_region(res->start, resource_size(res));
 
@@ -1830,22 +1826,21 @@ static int stmmac_suspend(struct platform_device *pdev, pm_message_t state)
 		napi_disable(&priv->napi);
 
 		/* Stop TX/RX DMA */
-		priv->hw->dma->stop_tx(dev->base_addr);
-		priv->hw->dma->stop_rx(dev->base_addr);
+		priv->hw->dma->stop_tx(priv->ioaddr);
+		priv->hw->dma->stop_rx(priv->ioaddr);
 		/* Clear the Rx/Tx descriptors */
 		priv->hw->desc->init_rx_desc(priv->dma_rx, priv->dma_rx_size,
 					     dis_ic);
 		priv->hw->desc->init_tx_desc(priv->dma_tx, priv->dma_tx_size);
 
-		stmmac_mac_disable_tx(dev->base_addr);
+		stmmac_mac_disable_tx(priv->ioaddr);
 
 		if (device_may_wakeup(&(pdev->dev))) {
 			/* Enable Power down mode by programming the PMT regs */
 			if (priv->wolenabled == PMT_SUPPORTED)
-				priv->hw->mac->pmt(dev->base_addr,
-						   priv->wolopts);
+				priv->hw->mac->pmt(priv->ioaddr, priv->wolopts);
 		} else {
-			stmmac_mac_disable_rx(dev->base_addr);
+			stmmac_mac_disable_rx(priv->ioaddr);
 		}
 	} else {
 		priv->shutdown = 1;
@@ -1863,7 +1858,6 @@ static int stmmac_resume(struct platform_device *pdev)
 {
 	struct net_device *dev = platform_get_drvdata(pdev);
 	struct stmmac_priv *priv = netdev_priv(dev);
-	unsigned long ioaddr = dev->base_addr;
 
 	if (!netif_running(dev))
 		return 0;
@@ -1884,15 +1878,15 @@ static int stmmac_resume(struct platform_device *pdev)
 	 * from another devices (e.g. serial console). */
 	if (device_may_wakeup(&(pdev->dev)))
 		if (priv->wolenabled == PMT_SUPPORTED)
-			priv->hw->mac->pmt(dev->base_addr, 0);
+			priv->hw->mac->pmt(priv->ioaddr, 0);
 
 	netif_device_attach(dev);
 
 	/* Enable the MAC and DMA */
-	stmmac_mac_enable_rx(ioaddr);
-	stmmac_mac_enable_tx(ioaddr);
-	priv->hw->dma->start_tx(ioaddr);
-	priv->hw->dma->start_rx(ioaddr);
+	stmmac_mac_enable_rx(priv->ioaddr);
+	stmmac_mac_enable_tx(priv->ioaddr);
+	priv->hw->dma->start_tx(priv->ioaddr);
+	priv->hw->dma->start_rx(priv->ioaddr);
 
 #ifdef CONFIG_STMMAC_TIMER
 	priv->tm->timer_start(tmrate);
diff --git a/drivers/net/stmmac/stmmac_mdio.c b/drivers/net/stmmac/stmmac_mdio.c
index 40b2c7929719..03dea1401571 100644
--- a/drivers/net/stmmac/stmmac_mdio.c
+++ b/drivers/net/stmmac/stmmac_mdio.c
@@ -47,7 +47,6 @@ static int stmmac_mdio_read(struct mii_bus *bus, int phyaddr, int phyreg)
 {
 	struct net_device *ndev = bus->priv;
 	struct stmmac_priv *priv = netdev_priv(ndev);
-	unsigned long ioaddr = ndev->base_addr;
 	unsigned int mii_address = priv->hw->mii.addr;
 	unsigned int mii_data = priv->hw->mii.data;
 
@@ -56,12 +55,12 @@ static int stmmac_mdio_read(struct mii_bus *bus, int phyaddr, int phyreg)
 			((phyreg << 6) & (0x000007C0)));
 	regValue |= MII_BUSY;	/* in case of GMAC */
 
-	do {} while (((readl(ioaddr + mii_address)) & MII_BUSY) == 1);
-	writel(regValue, ioaddr + mii_address);
-	do {} while (((readl(ioaddr + mii_address)) & MII_BUSY) == 1);
+	do {} while (((readl(priv->ioaddr + mii_address)) & MII_BUSY) == 1);
+	writel(regValue, priv->ioaddr + mii_address);
+	do {} while (((readl(priv->ioaddr + mii_address)) & MII_BUSY) == 1);
 
 	/* Read the data from the MII data register */
-	data = (int)readl(ioaddr + mii_data);
+	data = (int)readl(priv->ioaddr + mii_data);
 
 	return data;
 }
@@ -79,7 +78,6 @@ static int stmmac_mdio_write(struct mii_bus *bus, int phyaddr, int phyreg,
 {
 	struct net_device *ndev = bus->priv;
 	struct stmmac_priv *priv = netdev_priv(ndev);
-	unsigned long ioaddr = ndev->base_addr;
 	unsigned int mii_address = priv->hw->mii.addr;
 	unsigned int mii_data = priv->hw->mii.data;
 
@@ -90,14 +88,14 @@ static int stmmac_mdio_write(struct mii_bus *bus, int phyaddr, int phyreg,
 	value |= MII_BUSY;
 
 	/* Wait until any existing MII operation is complete */
-	do {} while (((readl(ioaddr + mii_address)) & MII_BUSY) == 1);
+	do {} while (((readl(priv->ioaddr + mii_address)) & MII_BUSY) == 1);
 
 	/* Set the MII address register to write */
-	writel(phydata, ioaddr + mii_data);
-	writel(value, ioaddr + mii_address);
+	writel(phydata, priv->ioaddr + mii_data);
+	writel(value, priv->ioaddr + mii_address);
 
 	/* Wait until any existing MII operation is complete */
-	do {} while (((readl(ioaddr + mii_address)) & MII_BUSY) == 1);
+	do {} while (((readl(priv->ioaddr + mii_address)) & MII_BUSY) == 1);
 
 	return 0;
 }
@@ -111,7 +109,6 @@ static int stmmac_mdio_reset(struct mii_bus *bus)
 {
 	struct net_device *ndev = bus->priv;
 	struct stmmac_priv *priv = netdev_priv(ndev);
-	unsigned long ioaddr = ndev->base_addr;
 	unsigned int mii_address = priv->hw->mii.addr;
 
 	if (priv->phy_reset) {
@@ -123,7 +120,7 @@ static int stmmac_mdio_reset(struct mii_bus *bus)
 	 * It doesn't complete its reset until at least one clock cycle
 	 * on MDC, so perform a dummy mdio read.
 	 */
-	writel(0, ioaddr + mii_address);
+	writel(0, priv->ioaddr + mii_address);
 
 	return 0;
 }
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index 632ff7c03280..a4adf0de6ed6 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -35,7 +35,7 @@ struct plat_stmmacenet_data {
 	int has_gmac;
 	int enh_desc;
 	void (*fix_mac_speed)(void *priv, unsigned int speed);
-	void (*bus_setup)(unsigned long ioaddr);
+	void (*bus_setup)(void __iomem *ioaddr);
 #ifdef CONFIG_STM_DRIVERS
 	struct stm_pad_config *pad_config;
 #endif
-- 
cgit v1.2.3


From 2971944582ff43b7dedbb460777052243ac9915a Mon Sep 17 00:00:00 2001
From: Rabin Vincent <rabin.vincent@stericsson.com>
Date: Mon, 9 Aug 2010 12:54:43 +0100
Subject: ARM: 6307/1: mmci: allow the card detect GPIO value not to be
 inverted

On some platforms, the GPIO value from the gpio_cd pin doesn't need to
be inverted to get it active high.  Add a cd_invert platform data
parameter and change existing platforms using GPIO for CD (only
Realview) to enable it.

Acked-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-realview/core.c | 2 ++
 drivers/mmc/host/mmci.c       | 5 +++--
 include/linux/amba/mmci.h     | 2 ++
 3 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/mach-realview/core.c b/arch/arm/mach-realview/core.c
index 2fa38df28414..07c08151dfe6 100644
--- a/arch/arm/mach-realview/core.c
+++ b/arch/arm/mach-realview/core.c
@@ -259,6 +259,7 @@ struct mmci_platform_data realview_mmc0_plat_data = {
 	.status		= realview_mmc_status,
 	.gpio_wp	= 17,
 	.gpio_cd	= 16,
+	.cd_invert	= true,
 };
 
 struct mmci_platform_data realview_mmc1_plat_data = {
@@ -266,6 +267,7 @@ struct mmci_platform_data realview_mmc1_plat_data = {
 	.status		= realview_mmc_status,
 	.gpio_wp	= 19,
 	.gpio_cd	= 18,
+	.cd_invert	= true,
 };
 
 /*
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index 840b301b5671..9a9aeac50a6c 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -570,12 +570,13 @@ static int mmci_get_ro(struct mmc_host *mmc)
 static int mmci_get_cd(struct mmc_host *mmc)
 {
 	struct mmci_host *host = mmc_priv(mmc);
+	struct mmci_platform_data *plat = host->plat;
 	unsigned int status;
 
 	if (host->gpio_cd == -ENOSYS)
-		status = host->plat->status(mmc_dev(host->mmc));
+		status = plat->status(mmc_dev(host->mmc));
 	else
-		status = !gpio_get_value(host->gpio_cd);
+		status = !!gpio_get_value(host->gpio_cd) ^ plat->cd_invert;
 
 	/*
 	 * Use positive logic throughout - status is zero for no card,
diff --git a/include/linux/amba/mmci.h b/include/linux/amba/mmci.h
index ca84ce70d5d5..f4ee9acc9721 100644
--- a/include/linux/amba/mmci.h
+++ b/include/linux/amba/mmci.h
@@ -24,6 +24,7 @@
  * whether a card is present in the MMC slot or not
  * @gpio_wp: read this GPIO pin to see if the card is write protected
  * @gpio_cd: read this GPIO pin to detect card insertion
+ * @cd_invert: true if the gpio_cd pin value is active low
  * @capabilities: the capabilities of the block as implemented in
  * this platform, signify anything MMC_CAP_* from mmc/host.h
  */
@@ -35,6 +36,7 @@ struct mmci_platform_data {
 	unsigned int (*status)(struct device *);
 	int	gpio_wp;
 	int	gpio_cd;
+	bool	cd_invert;
 	unsigned long capabilities;
 };
 
-- 
cgit v1.2.3


From 480125ba49ba62be93beea37770f266846e077ab Mon Sep 17 00:00:00 2001
From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Date: Thu, 26 Aug 2010 13:57:57 -0400
Subject: x86, iommu: Make all IOMMU's detection routines return a value.

We return 1 if the IOMMU has been detected. Zero or an error number
if we failed to find it. This is in preperation of using the IOMMU_INIT
so that we can detect whether an IOMMU is present. I have not
tested this for regression on Calgary, nor on AMD Vi chipsets as
I don't have that hardware.

CC: Muli Ben-Yehuda <muli@il.ibm.com>
CC: "Jon D. Mason" <jdmason@kudzu.us>
CC: "Darrick J. Wong" <djwong@us.ibm.com>
CC: Jesse Barnes <jbarnes@virtuousgeek.org>
CC: David Woodhouse <David.Woodhouse@intel.com>
CC: Chris Wright <chrisw@sous-sol.org>
CC: Yinghai Lu <yinghai@kernel.org>
CC: Joerg Roedel <joerg.roedel@amd.com>
CC: H. Peter Anvin <hpa@zytor.com>
CC: Fujita Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
LKML-Reference: <1282845485-8991-3-git-send-email-konrad.wilk@oracle.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 arch/x86/include/asm/amd_iommu.h |  4 ++--
 arch/x86/include/asm/calgary.h   |  4 ++--
 arch/x86/include/asm/gart.h      |  5 +++--
 arch/x86/kernel/amd_iommu_init.c |  8 +++++---
 arch/x86/kernel/aperture_64.c    | 11 +++++++----
 arch/x86/kernel/pci-calgary_64.c | 15 ++++++++-------
 drivers/pci/dmar.c               |  4 +++-
 include/linux/dmar.h             |  6 +++---
 8 files changed, 33 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/include/asm/amd_iommu.h b/arch/x86/include/asm/amd_iommu.h
index 5af2982133b5..2798142cdb49 100644
--- a/arch/x86/include/asm/amd_iommu.h
+++ b/arch/x86/include/asm/amd_iommu.h
@@ -24,11 +24,11 @@
 
 #ifdef CONFIG_AMD_IOMMU
 
-extern void amd_iommu_detect(void);
+extern int amd_iommu_detect(void);
 
 #else
 
-static inline void amd_iommu_detect(void) { }
+static inline int amd_iommu_detect(void) { return -ENODEV; }
 
 #endif
 
diff --git a/arch/x86/include/asm/calgary.h b/arch/x86/include/asm/calgary.h
index 0918654305af..0d467b338835 100644
--- a/arch/x86/include/asm/calgary.h
+++ b/arch/x86/include/asm/calgary.h
@@ -62,9 +62,9 @@ struct cal_chipset_ops {
 extern int use_calgary;
 
 #ifdef CONFIG_CALGARY_IOMMU
-extern void detect_calgary(void);
+extern int detect_calgary(void);
 #else
-static inline void detect_calgary(void) { return; }
+static inline int detect_calgary(void) { return -ENODEV; }
 #endif
 
 #endif /* _ASM_X86_CALGARY_H */
diff --git a/arch/x86/include/asm/gart.h b/arch/x86/include/asm/gart.h
index 4ac5b0f33fc1..d7d1d4c438a4 100644
--- a/arch/x86/include/asm/gart.h
+++ b/arch/x86/include/asm/gart.h
@@ -37,7 +37,7 @@ extern int gart_iommu_aperture_disabled;
 extern void early_gart_iommu_check(void);
 extern int gart_iommu_init(void);
 extern void __init gart_parse_options(char *);
-extern void gart_iommu_hole_init(void);
+extern int gart_iommu_hole_init(void);
 
 #else
 #define gart_iommu_aperture            0
@@ -50,8 +50,9 @@ static inline void early_gart_iommu_check(void)
 static inline void gart_parse_options(char *options)
 {
 }
-static inline void gart_iommu_hole_init(void)
+static inline int gart_iommu_hole_init(void)
 {
+	return -ENODEV;
 }
 #endif
 
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index 3cc63e2b8dd4..0b9e2dc4fc9a 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -1382,13 +1382,13 @@ static int __init early_amd_iommu_detect(struct acpi_table_header *table)
 	return 0;
 }
 
-void __init amd_iommu_detect(void)
+int __init amd_iommu_detect(void)
 {
 	if (no_iommu || (iommu_detected && !gart_iommu_aperture))
-		return;
+		return -ENODEV;
 
 	if (amd_iommu_disabled)
-		return;
+		return -ENODEV;
 
 	if (acpi_table_parse("IVRS", early_amd_iommu_detect) == 0) {
 		iommu_detected = 1;
@@ -1397,7 +1397,9 @@ void __init amd_iommu_detect(void)
 
 		/* Make sure ACS will be enabled */
 		pci_request_acs();
+		return 1;
 	}
+	return -ENODEV;
 }
 
 /****************************************************************************
diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c
index a2e0caf26e17..afa0dab3302f 100644
--- a/arch/x86/kernel/aperture_64.c
+++ b/arch/x86/kernel/aperture_64.c
@@ -371,7 +371,7 @@ void __init early_gart_iommu_check(void)
 
 static int __initdata printed_gart_size_msg;
 
-void __init gart_iommu_hole_init(void)
+int __init gart_iommu_hole_init(void)
 {
 	u32 agp_aper_base = 0, agp_aper_order = 0;
 	u32 aper_size, aper_alloc = 0, aper_order = 0, last_aper_order = 0;
@@ -381,7 +381,7 @@ void __init gart_iommu_hole_init(void)
 
 	if (gart_iommu_aperture_disabled || !fix_aperture ||
 	    !early_pci_allowed())
-		return;
+		return -ENODEV;
 
 	printk(KERN_INFO  "Checking aperture...\n");
 
@@ -463,8 +463,9 @@ out:
 			unsigned long n = (32 * 1024 * 1024) << last_aper_order;
 
 			insert_aperture_resource((u32)last_aper_base, n);
+			return 1;
 		}
-		return;
+		return 0;
 	}
 
 	if (!fallback_aper_force) {
@@ -500,7 +501,7 @@ out:
 			panic("Not enough memory for aperture");
 		}
 	} else {
-		return;
+		return 0;
 	}
 
 	/* Fix up the north bridges */
@@ -524,4 +525,6 @@ out:
 	}
 
 	set_up_gart_resume(aper_order, aper_alloc);
+
+	return 1;
 }
diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 078d4ec1a9d9..28c6b389fee6 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -1364,7 +1364,7 @@ static int __init calgary_iommu_init(void)
 	return 0;
 }
 
-void __init detect_calgary(void)
+int __init detect_calgary(void)
 {
 	int bus;
 	void *tbl;
@@ -1378,13 +1378,13 @@ void __init detect_calgary(void)
 	 * another HW IOMMU already, bail out.
 	 */
 	if (no_iommu || iommu_detected)
-		return;
+		return -ENODEV;
 
 	if (!use_calgary)
-		return;
+		return -ENODEV;
 
 	if (!early_pci_allowed())
-		return;
+		return -ENODEV;
 
 	printk(KERN_DEBUG "Calgary: detecting Calgary via BIOS EBDA area\n");
 
@@ -1410,13 +1410,13 @@ void __init detect_calgary(void)
 	if (!rio_table_hdr) {
 		printk(KERN_DEBUG "Calgary: Unable to locate Rio Grande table "
 		       "in EBDA - bailing!\n");
-		return;
+		return -ENODEV;
 	}
 
 	ret = build_detail_arrays();
 	if (ret) {
 		printk(KERN_DEBUG "Calgary: build_detail_arrays ret %d\n", ret);
-		return;
+		return -ENOMEM;
 	}
 
 	specified_table_size = determine_tce_table_size((is_kdump_kernel() ?
@@ -1464,7 +1464,7 @@ void __init detect_calgary(void)
 
 		x86_init.iommu.iommu_init = calgary_iommu_init;
 	}
-	return;
+	return calgary_found;
 
 cleanup:
 	for (--bus; bus >= 0; --bus) {
@@ -1473,6 +1473,7 @@ cleanup:
 		if (info->tce_space)
 			free_tce_table(info->tce_space);
 	}
+	return -ENOMEM;
 }
 
 static int __init calgary_parse_options(char *p)
diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 0a19708074c2..5fa64ea5416f 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -687,7 +687,7 @@ failed:
 	return 0;
 }
 
-void __init detect_intel_iommu(void)
+int __init detect_intel_iommu(void)
 {
 	int ret;
 
@@ -723,6 +723,8 @@ void __init detect_intel_iommu(void)
 	}
 	early_acpi_os_unmap_memory(dmar_tbl, dmar_tbl_size);
 	dmar_tbl = NULL;
+
+	return (ret ? 1 : -ENODEV);
 }
 
 
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index d7cecc90ed34..a20602041511 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -57,15 +57,15 @@ extern int dmar_table_init(void);
 extern int dmar_dev_scope_init(void);
 
 /* Intel IOMMU detection */
-extern void detect_intel_iommu(void);
+extern int detect_intel_iommu(void);
 extern int enable_drhd_fault_handling(void);
 
 extern int parse_ioapics_under_ir(void);
 extern int alloc_iommu(struct dmar_drhd_unit *);
 #else
-static inline void detect_intel_iommu(void)
+static inline int detect_intel_iommu(void)
 {
-	return;
+	return -ENODEV;
 }
 
 static inline int dmar_table_init(void)
-- 
cgit v1.2.3


From 40d0802b3eb47d57e2d57a5244a18cbbe9632e13 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 26 Aug 2010 22:03:08 -0700
Subject: gro: __napi_gro_receive() optimizations

compare_ether_header() can have a special implementation on 64 bit
arches if CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS is defined.

__napi_gro_receive() and vlan_gro_common() can avoid a conditional
branch to perform device match.

On x86_64, __napi_gro_receive() has now 38 instructions instead of 53

As gcc-4.4.3 still choose to not inline it, add inline keyword to this
performance critical function.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/etherdevice.h | 18 +++++++++++++++++-
 net/8021q/vlan_core.c       |  9 ++++++---
 net/core/dev.c              | 10 ++++++----
 3 files changed, 29 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h
index 2308fbb4523a..fb6aa6070921 100644
--- a/include/linux/etherdevice.h
+++ b/include/linux/etherdevice.h
@@ -237,13 +237,29 @@ static inline bool is_etherdev_addr(const struct net_device *dev,
  * entry points.
  */
 
-static inline int compare_ether_header(const void *a, const void *b)
+static inline unsigned long compare_ether_header(const void *a, const void *b)
 {
+#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64
+	unsigned long fold;
+
+	/*
+	 * We want to compare 14 bytes:
+	 *  [a0 ... a13] ^ [b0 ... b13]
+	 * Use two long XOR, ORed together, with an overlap of two bytes.
+	 *  [a0  a1  a2  a3  a4  a5  a6  a7 ] ^ [b0  b1  b2  b3  b4  b5  b6  b7 ] |
+	 *  [a6  a7  a8  a9  a10 a11 a12 a13] ^ [b6  b7  b8  b9  b10 b11 b12 b13]
+	 * This means the [a6 a7] ^ [b6 b7] part is done two times.
+	*/
+	fold = *(unsigned long *)a ^ *(unsigned long *)b;
+	fold |= *(unsigned long *)(a + 6) ^ *(unsigned long *)(b + 6);
+	return fold;
+#else
 	u32 *a32 = (u32 *)((u8 *)a + 2);
 	u32 *b32 = (u32 *)((u8 *)b + 2);
 
 	return (*(u16 *)a ^ *(u16 *)b) | (a32[0] ^ b32[0]) |
 	       (a32[1] ^ b32[1]) | (a32[2] ^ b32[2]);
+#endif
 }
 
 #endif	/* _LINUX_ETHERDEVICE_H */
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 07eeb5b99dce..3438c01bbacf 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -105,9 +105,12 @@ vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp,
 		goto drop;
 
 	for (p = napi->gro_list; p; p = p->next) {
-		NAPI_GRO_CB(p)->same_flow =
-			p->dev == skb->dev && !compare_ether_header(
-				skb_mac_header(p), skb_gro_mac_header(skb));
+		unsigned long diffs;
+
+		diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev;
+		diffs |= compare_ether_header(skb_mac_header(p),
+					      skb_gro_mac_header(skb));
+		NAPI_GRO_CB(p)->same_flow = !diffs;
 		NAPI_GRO_CB(p)->flush = 0;
 	}
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 859e30ff044a..63bd20a75929 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3169,16 +3169,18 @@ normal:
 }
 EXPORT_SYMBOL(dev_gro_receive);
 
-static gro_result_t
+static inline gro_result_t
 __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 {
 	struct sk_buff *p;
 
 	for (p = napi->gro_list; p; p = p->next) {
-		NAPI_GRO_CB(p)->same_flow =
-			(p->dev == skb->dev) &&
-			!compare_ether_header(skb_mac_header(p),
+		unsigned long diffs;
+
+		diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev;
+		diffs |= compare_ether_header(skb_mac_header(p),
 					      skb_gro_mac_header(skb));
+		NAPI_GRO_CB(p)->same_flow = !diffs;
 		NAPI_GRO_CB(p)->flush = 0;
 	}
 
-- 
cgit v1.2.3


From c0692b8fe29fb4d4dad33487aabf3ed7e1e880c0 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 27 Aug 2010 14:26:53 +0300
Subject: cfg80211: allow changing port control protocol

Some vendor specified mechanisms for 802.1X-style
functionality use a different protocol than EAP
(even if EAP is vendor-extensible). Allow setting
the ethertype for the protocol when a driver has
support for this. The default if unspecified is
EAP, of course.

Note: This is suitable only for station mode, not
      for AP implementation.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 16 +++++++++++++++-
 include/net/cfg80211.h  | 24 +++++++++++++++++-------
 net/wireless/nl80211.c  | 25 ++++++++++++++++++++++---
 net/wireless/wext-sme.c |  2 ++
 4 files changed, 56 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index ec1690da7845..31603e8b5581 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -295,7 +295,9 @@
  *	auth and assoc steps. For this, you need to specify the SSID in a
  *	%NL80211_ATTR_SSID attribute, and can optionally specify the association
  *	IEs in %NL80211_ATTR_IE, %NL80211_ATTR_AUTH_TYPE, %NL80211_ATTR_MAC,
- *	%NL80211_ATTR_WIPHY_FREQ and %NL80211_ATTR_CONTROL_PORT.
+ *	%NL80211_ATTR_WIPHY_FREQ, %NL80211_ATTR_CONTROL_PORT,
+ *	%NL80211_ATTR_CONTROL_PORT_ETHERTYPE and
+ *	%NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT.
  *	It is also sent as an event, with the BSSID and response IEs when the
  *	connection is established or failed to be established. This can be
  *	determined by the STATUS_CODE attribute.
@@ -686,6 +688,15 @@ enum nl80211_commands {
  *	request, the driver will assume that the port is unauthorized until
  *	authorized by user space. Otherwise, port is marked authorized by
  *	default in station mode.
+ * @NL80211_ATTR_CONTROL_PORT_ETHERTYPE: A 16-bit value indicating the
+ *	ethertype that will be used for key negotiation. It can be
+ *	specified with the associate and connect commands. If it is not
+ *	specified, the value defaults to 0x888E (PAE, 802.1X). This
+ *	attribute is also used as a flag in the wiphy information to
+ *	indicate that protocols other than PAE are supported.
+ * @NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT: When included along with
+ *	%NL80211_ATTR_CONTROL_PORT_ETHERTYPE, indicates that the custom
+ *	ethertype frames used for key negotiation must not be encrypted.
  *
  * @NL80211_ATTR_TESTDATA: Testmode data blob, passed through to the driver.
  *	We recommend using nested, driver-specific attributes within this.
@@ -951,6 +962,9 @@ enum nl80211_attrs {
 	NL80211_ATTR_RX_FRAME_TYPES,
 	NL80211_ATTR_FRAME_TYPE,
 
+	NL80211_ATTR_CONTROL_PORT_ETHERTYPE,
+	NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index f2740537b5d6..4c8c727d0cca 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -763,6 +763,10 @@ const u8 *ieee80211_bss_get_ie(struct cfg80211_bss *bss, u8 ie);
  *	sets/clears %NL80211_STA_FLAG_AUTHORIZED. If true, the driver is
  *	required to assume that the port is unauthorized until authorized by
  *	user space. Otherwise, port is marked authorized by default.
+ * @control_port_ethertype: the control port protocol that should be
+ *	allowed through even on unauthorized ports
+ * @control_port_no_encrypt: TRUE to prevent encryption of control port
+ *	protocol frames.
  */
 struct cfg80211_crypto_settings {
 	u32 wpa_versions;
@@ -772,6 +776,8 @@ struct cfg80211_crypto_settings {
 	int n_akm_suites;
 	u32 akm_suites[NL80211_MAX_NR_AKM_SUITES];
 	bool control_port;
+	__be16 control_port_ethertype;
+	bool control_port_no_encrypt;
 };
 
 /**
@@ -1293,15 +1299,19 @@ struct cfg80211_ops {
  * @WIPHY_FLAG_4ADDR_AP: supports 4addr mode even on AP (with a single station
  *	on a VLAN interface)
  * @WIPHY_FLAG_4ADDR_STATION: supports 4addr mode even as a station
+ * @WIPHY_FLAG_CONTROL_PORT_PROTOCOL: This device supports setting the
+ *	control port protocol ethertype. The device also honours the
+ *	control_port_no_encrypt flag.
  */
 enum wiphy_flags {
-	WIPHY_FLAG_CUSTOM_REGULATORY	= BIT(0),
-	WIPHY_FLAG_STRICT_REGULATORY	= BIT(1),
-	WIPHY_FLAG_DISABLE_BEACON_HINTS	= BIT(2),
-	WIPHY_FLAG_NETNS_OK		= BIT(3),
-	WIPHY_FLAG_PS_ON_BY_DEFAULT	= BIT(4),
-	WIPHY_FLAG_4ADDR_AP		= BIT(5),
-	WIPHY_FLAG_4ADDR_STATION	= BIT(6),
+	WIPHY_FLAG_CUSTOM_REGULATORY		= BIT(0),
+	WIPHY_FLAG_STRICT_REGULATORY		= BIT(1),
+	WIPHY_FLAG_DISABLE_BEACON_HINTS		= BIT(2),
+	WIPHY_FLAG_NETNS_OK			= BIT(3),
+	WIPHY_FLAG_PS_ON_BY_DEFAULT		= BIT(4),
+	WIPHY_FLAG_4ADDR_AP			= BIT(5),
+	WIPHY_FLAG_4ADDR_STATION		= BIT(6),
+	WIPHY_FLAG_CONTROL_PORT_PROTOCOL	= BIT(7),
 };
 
 struct mac_address {
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 49f5ca35e787..85a23de7bff3 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -136,6 +136,8 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = {
 		.len = sizeof(struct nl80211_sta_flag_update),
 	},
 	[NL80211_ATTR_CONTROL_PORT] = { .type = NLA_FLAG },
+	[NL80211_ATTR_CONTROL_PORT_ETHERTYPE] = { .type = NLA_U16 },
+	[NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT] = { .type = NLA_FLAG },
 	[NL80211_ATTR_PRIVACY] = { .type = NLA_FLAG },
 	[NL80211_ATTR_CIPHER_SUITE_GROUP] = { .type = NLA_U32 },
 	[NL80211_ATTR_WPA_VERSIONS] = { .type = NLA_U32 },
@@ -474,6 +476,9 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 	NLA_PUT_U8(msg, NL80211_ATTR_MAX_NUM_PMKIDS,
 		   dev->wiphy.max_num_pmkids);
 
+	if (dev->wiphy.flags & WIPHY_FLAG_CONTROL_PORT_PROTOCOL)
+		NLA_PUT_FLAG(msg, NL80211_ATTR_CONTROL_PORT_ETHERTYPE);
+
 	nl_modes = nla_nest_start(msg, NL80211_ATTR_SUPPORTED_IFTYPES);
 	if (!nl_modes)
 		goto nla_put_failure;
@@ -3691,7 +3696,8 @@ unlock_rtnl:
 	return err;
 }
 
-static int nl80211_crypto_settings(struct genl_info *info,
+static int nl80211_crypto_settings(struct cfg80211_registered_device *rdev,
+				   struct genl_info *info,
 				   struct cfg80211_crypto_settings *settings,
 				   int cipher_limit)
 {
@@ -3699,6 +3705,19 @@ static int nl80211_crypto_settings(struct genl_info *info,
 
 	settings->control_port = info->attrs[NL80211_ATTR_CONTROL_PORT];
 
+	if (info->attrs[NL80211_ATTR_CONTROL_PORT_ETHERTYPE]) {
+		u16 proto;
+		proto = nla_get_u16(
+			info->attrs[NL80211_ATTR_CONTROL_PORT_ETHERTYPE]);
+		settings->control_port_ethertype = cpu_to_be16(proto);
+		if (!(rdev->wiphy.flags & WIPHY_FLAG_CONTROL_PORT_PROTOCOL) &&
+		    proto != ETH_P_PAE)
+			return -EINVAL;
+		if (info->attrs[NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT])
+			settings->control_port_no_encrypt = true;
+	} else
+		settings->control_port_ethertype = cpu_to_be16(ETH_P_PAE);
+
 	if (info->attrs[NL80211_ATTR_CIPHER_SUITES_PAIRWISE]) {
 		void *data;
 		int len, i;
@@ -3826,7 +3845,7 @@ static int nl80211_associate(struct sk_buff *skb, struct genl_info *info)
 	if (info->attrs[NL80211_ATTR_PREV_BSSID])
 		prev_bssid = nla_data(info->attrs[NL80211_ATTR_PREV_BSSID]);
 
-	err = nl80211_crypto_settings(info, &crypto, 1);
+	err = nl80211_crypto_settings(rdev, info, &crypto, 1);
 	if (!err)
 		err = cfg80211_mlme_assoc(rdev, dev, chan, bssid, prev_bssid,
 					  ssid, ssid_len, ie, ie_len, use_mfp,
@@ -4303,7 +4322,7 @@ static int nl80211_connect(struct sk_buff *skb, struct genl_info *info)
 
 	connect.privacy = info->attrs[NL80211_ATTR_PRIVACY];
 
-	err = nl80211_crypto_settings(info, &connect.crypto,
+	err = nl80211_crypto_settings(rdev, info, &connect.crypto,
 				      NL80211_MAX_NR_CIPHER_SUITES);
 	if (err)
 		return err;
diff --git a/net/wireless/wext-sme.c b/net/wireless/wext-sme.c
index 9818198add8a..6fffe62d7c25 100644
--- a/net/wireless/wext-sme.c
+++ b/net/wireless/wext-sme.c
@@ -197,6 +197,8 @@ int cfg80211_mgd_wext_siwessid(struct net_device *dev,
 	wdev->wext.connect.ssid_len = len;
 
 	wdev->wext.connect.crypto.control_port = false;
+	wdev->wext.connect.crypto.control_port_ethertype =
+					cpu_to_be16(ETH_P_PAE);
 
 	err = cfg80211_mgd_wext_connect(rdev, wdev);
  out:
-- 
cgit v1.2.3


From 7950c407c0288b223a200c1bba8198941599ca37 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 25 Aug 2010 13:39:14 -0700
Subject: memblock: Add memblock_free/reserve_reserved_regions()

So we can avoid export memblock_reserved_init_regions()
Suggested by Ben.

-v2: use __init_memblock attribute

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 include/linux/memblock.h |  2 ++
 mm/memblock.c            | 24 ++++++++++++++++++++++++
 2 files changed, 26 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 4df09bdcae42..7d285271130d 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -48,6 +48,8 @@ extern int memblock_can_resize;
 	if (memblock_debug) printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__)
 
 u64 memblock_find_in_range(u64 start, u64 end, u64 size, u64 align);
+int memblock_free_reserved_regions(void);
+int memblock_reserve_reserved_regions(void);
 
 extern void __init memblock_init(void);
 extern void __init memblock_analyze(void);
diff --git a/mm/memblock.c b/mm/memblock.c
index b7ab10a2ef46..65e3ba8d09fb 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -170,6 +170,30 @@ u64 __init_memblock memblock_find_in_range(u64 start, u64 end, u64 size, u64 ali
 	return memblock_find_base(size, align, start, end);
 }
 
+/*
+ * Free memblock.reserved.regions
+ */
+int __init_memblock memblock_free_reserved_regions(void)
+{
+	if (memblock.reserved.regions == memblock_reserved_init_regions)
+		return 0;
+
+	return memblock_free(__pa(memblock.reserved.regions),
+		 sizeof(struct memblock_region) * memblock.reserved.max);
+}
+
+/*
+ * Reserve memblock.reserved.regions
+ */
+int __init_memblock memblock_reserve_reserved_regions(void)
+{
+	if (memblock.reserved.regions == memblock_reserved_init_regions)
+		return 0;
+
+	return memblock_reserve(__pa(memblock.reserved.regions),
+		 sizeof(struct memblock_region) * memblock.reserved.max);
+}
+
 static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r)
 {
 	unsigned long i;
-- 
cgit v1.2.3


From edbe7d23b4482e7f33179290bcff3b1feae1c5f3 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 25 Aug 2010 13:39:16 -0700
Subject: memblock: Add find_memory_core_early()

According to node range in early_node_map[] with __memblock_find_in_range
to find free range.

Will be used by memblock_x86_find_in_range_node()

memblock_x86_find_in_range_node will be used to find right buffer for NODE_DATA

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 include/linux/mm.h |  2 ++
 mm/page_alloc.c    | 36 ++++++++++++++++++++++++++++++++++++
 2 files changed, 38 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index a2b48041b910..993e85f0afcb 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1164,6 +1164,8 @@ extern void free_bootmem_with_active_regions(int nid,
 						unsigned long max_low_pfn);
 int add_from_early_node_map(struct range *range, int az,
 				   int nr_range, int nid);
+u64 __init find_memory_core_early(int nid, u64 size, u64 align,
+					u64 goal, u64 limit);
 void *__alloc_memory_core_early(int nodeid, u64 size, u64 align,
 				 u64 goal, u64 limit);
 typedef int (*work_fn_t)(unsigned long, unsigned long, void *);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9bd339eb04c6..8c9b34674d83 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -21,6 +21,7 @@
 #include <linux/pagemap.h>
 #include <linux/jiffies.h>
 #include <linux/bootmem.h>
+#include <linux/memblock.h>
 #include <linux/compiler.h>
 #include <linux/kernel.h>
 #include <linux/kmemcheck.h>
@@ -3612,6 +3613,41 @@ void __init free_bootmem_with_active_regions(int nid,
 	}
 }
 
+#ifdef CONFIG_HAVE_MEMBLOCK
+u64 __init find_memory_core_early(int nid, u64 size, u64 align,
+					u64 goal, u64 limit)
+{
+	int i;
+
+	/* Need to go over early_node_map to find out good range for node */
+	for_each_active_range_index_in_nid(i, nid) {
+		u64 addr;
+		u64 ei_start, ei_last;
+		u64 final_start, final_end;
+
+		ei_last = early_node_map[i].end_pfn;
+		ei_last <<= PAGE_SHIFT;
+		ei_start = early_node_map[i].start_pfn;
+		ei_start <<= PAGE_SHIFT;
+
+		final_start = max(ei_start, goal);
+		final_end = min(ei_last, limit);
+
+		if (final_start >= final_end)
+			continue;
+
+		addr = memblock_find_in_range(final_start, final_end, size, align);
+
+		if (addr == MEMBLOCK_ERROR)
+			continue;
+
+		return addr;
+	}
+
+	return MEMBLOCK_ERROR;
+}
+#endif
+
 int __init add_from_early_node_map(struct range *range, int az,
 				   int nr_range, int nid)
 {
-- 
cgit v1.2.3


From a587d2daebcd2bc159d4348b6a7b028950a6d803 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 25 Aug 2010 13:39:18 -0700
Subject: x86: Remove not used early_res code

and some functions in e820.c that are not used anymore

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/e820.h |  14 --
 arch/x86/kernel/e820.c      |  52 ----
 include/linux/early_res.h   |  23 --
 kernel/Makefile             |   1 -
 kernel/early_res.c          | 590 --------------------------------------------
 5 files changed, 680 deletions(-)
 delete mode 100644 include/linux/early_res.h
 delete mode 100644 kernel/early_res.c

(limited to 'include/linux')

diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h
index 388fed291467..718646384e03 100644
--- a/arch/x86/include/asm/e820.h
+++ b/arch/x86/include/asm/e820.h
@@ -112,31 +112,17 @@ static inline void early_memtest(unsigned long start, unsigned long end)
 }
 #endif
 
-extern unsigned long end_user_pfn;
-
-extern u64 find_e820_area(u64 start, u64 end, u64 size, u64 align);
-extern u64 find_e820_area_size(u64 start, u64 *sizep, u64 align);
-extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align);
-
 extern unsigned long e820_end_of_ram_pfn(void);
 extern unsigned long e820_end_of_low_ram_pfn(void);
-extern void e820_register_active_regions(int nid, unsigned long start_pfn,
-					 unsigned long end_pfn);
-extern u64 e820_hole_size(u64 start, u64 end);
-
 extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align);
 
 void memblock_x86_fill(void);
-
 extern void finish_e820_parsing(void);
 extern void e820_reserve_resources(void);
 extern void e820_reserve_resources_late(void);
 extern void setup_memory_map(void);
 extern char *default_machine_specific_memory_setup(void);
 
-void reserve_early(u64 start, u64 end, char *name);
-void free_early(u64 start, u64 end);
-
 /*
  * Returns true iff the specified range [s,e) is completely contained inside
  * the ISA region.
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index a9221d18a5ed..d5fd89462d79 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -738,32 +738,6 @@ static int __init e820_mark_nvs_memory(void)
 core_initcall(e820_mark_nvs_memory);
 #endif
 
-/*
- * Find a free area with specified alignment in a specific range.
- */
-u64 __init find_e820_area(u64 start, u64 end, u64 size, u64 align)
-{
-	u64 mem = memblock_find_in_range(start, end, size, align);
-
-	if (mem == MEMBLOCK_ERROR)
-		return -1ULL;
-
-	return mem;
-}
-
-/*
- * Find next free range after *start
- */
-u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align)
-{
-	u64 mem = memblock_x86_find_in_range_size(start, sizep, align);
-
-	if (mem == MEMBLOCK_ERROR)
-		return -1ULL
-
-	return mem;
-}
-
 /*
  * pre allocated 4k and reserved it in memblock and e820_saved
  */
@@ -856,32 +830,6 @@ unsigned long __init e820_end_of_low_ram_pfn(void)
 	return e820_end_pfn(1UL<<(32 - PAGE_SHIFT), E820_RAM);
 }
 
-/* Walk the e820 map and register active regions within a node */
-void __init e820_register_active_regions(int nid, unsigned long start_pfn,
-					 unsigned long last_pfn)
-{
-	memblock_x86_register_active_regions(nid, start_pfn, last_pfn);
-}
-
-/*
- * Find the hole size (in bytes) in the memory range.
- * @start: starting address of the memory range to scan
- * @end: ending address of the memory range to scan
- */
-u64 __init e820_hole_size(u64 start, u64 end)
-{
-	return memblock_x86_hole_size(start, end);
-}
-
-void reserve_early(u64 start, u64 end, char *name)
-{
-	memblock_x86_reserve_range(start, end, name);
-}
-void free_early(u64 start, u64 end)
-{
-	memblock_x86_free_range(start, end);
-}
-
 static void early_panic(char *msg)
 {
 	early_printk(msg);
diff --git a/include/linux/early_res.h b/include/linux/early_res.h
deleted file mode 100644
index 29c09f57a13c..000000000000
--- a/include/linux/early_res.h
+++ /dev/null
@@ -1,23 +0,0 @@
-#ifndef _LINUX_EARLY_RES_H
-#define _LINUX_EARLY_RES_H
-#ifdef __KERNEL__
-
-extern void reserve_early(u64 start, u64 end, char *name);
-extern void reserve_early_overlap_ok(u64 start, u64 end, char *name);
-extern void free_early(u64 start, u64 end);
-void free_early_partial(u64 start, u64 end);
-extern void early_res_to_bootmem(u64 start, u64 end);
-
-void reserve_early_without_check(u64 start, u64 end, char *name);
-u64 find_early_area(u64 ei_start, u64 ei_last, u64 start, u64 end,
-			 u64 size, u64 align);
-u64 find_early_area_size(u64 ei_start, u64 ei_last, u64 start,
-			 u64 *sizep, u64 align);
-u64 find_fw_memmap_area(u64 start, u64 end, u64 size, u64 align);
-u64 get_max_mapped(void);
-#include <linux/range.h>
-int get_free_all_memory_range(struct range **rangep, int nodeid);
-
-#endif /* __KERNEL__ */
-
-#endif /* _LINUX_EARLY_RES_H */
diff --git a/kernel/Makefile b/kernel/Makefile
index 057472fbc272..80e61c3e44ae 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -11,7 +11,6 @@ obj-y     = sched.o fork.o exec_domain.o panic.o printk.o \
 	    hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \
 	    notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \
 	    async.o range.o
-obj-$(CONFIG_HAVE_EARLY_RES) += early_res.o
 obj-y += groups.o
 
 ifdef CONFIG_FUNCTION_TRACER
diff --git a/kernel/early_res.c b/kernel/early_res.c
deleted file mode 100644
index 7bfae887f211..000000000000
--- a/kernel/early_res.c
+++ /dev/null
@@ -1,590 +0,0 @@
-/*
- * early_res, could be used to replace bootmem
- */
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/bootmem.h>
-#include <linux/mm.h>
-#include <linux/early_res.h>
-#include <linux/slab.h>
-#include <linux/kmemleak.h>
-
-/*
- * Early reserved memory areas.
- */
-/*
- * need to make sure this one is bigger enough before
- * find_fw_memmap_area could be used
- */
-#define MAX_EARLY_RES_X 32
-
-struct early_res {
-	u64 start, end;
-	char name[15];
-	char overlap_ok;
-};
-static struct early_res early_res_x[MAX_EARLY_RES_X] __initdata;
-
-static int max_early_res __initdata = MAX_EARLY_RES_X;
-static struct early_res *early_res __initdata = &early_res_x[0];
-static int early_res_count __initdata;
-
-static int __init find_overlapped_early(u64 start, u64 end)
-{
-	int i;
-	struct early_res *r;
-
-	for (i = 0; i < max_early_res && early_res[i].end; i++) {
-		r = &early_res[i];
-		if (end > r->start && start < r->end)
-			break;
-	}
-
-	return i;
-}
-
-/*
- * Drop the i-th range from the early reservation map,
- * by copying any higher ranges down one over it, and
- * clearing what had been the last slot.
- */
-static void __init drop_range(int i)
-{
-	int j;
-
-	for (j = i + 1; j < max_early_res && early_res[j].end; j++)
-		;
-
-	memmove(&early_res[i], &early_res[i + 1],
-	       (j - 1 - i) * sizeof(struct early_res));
-
-	early_res[j - 1].end = 0;
-	early_res_count--;
-}
-
-static void __init drop_range_partial(int i, u64 start, u64 end)
-{
-	u64 common_start, common_end;
-	u64 old_start, old_end;
-
-	old_start = early_res[i].start;
-	old_end = early_res[i].end;
-	common_start = max(old_start, start);
-	common_end = min(old_end, end);
-
-	/* no overlap ? */
-	if (common_start >= common_end)
-		return;
-
-	if (old_start < common_start) {
-		/* make head segment */
-		early_res[i].end = common_start;
-		if (old_end > common_end) {
-			char name[15];
-
-			/*
-			 * Save a local copy of the name, since the
-			 * early_res array could get resized inside
-			 * reserve_early_without_check() ->
-			 * __check_and_double_early_res(), which would
-			 * make the current name pointer invalid.
-			 */
-			strncpy(name, early_res[i].name,
-					 sizeof(early_res[i].name) - 1);
-			/* add another for left over on tail */
-			reserve_early_without_check(common_end, old_end, name);
-		}
-		return;
-	} else {
-		if (old_end > common_end) {
-			/* reuse the entry for tail left */
-			early_res[i].start = common_end;
-			return;
-		}
-		/* all covered */
-		drop_range(i);
-	}
-}
-
-/*
- * Split any existing ranges that:
- *  1) are marked 'overlap_ok', and
- *  2) overlap with the stated range [start, end)
- * into whatever portion (if any) of the existing range is entirely
- * below or entirely above the stated range.  Drop the portion
- * of the existing range that overlaps with the stated range,
- * which will allow the caller of this routine to then add that
- * stated range without conflicting with any existing range.
- */
-static void __init drop_overlaps_that_are_ok(u64 start, u64 end)
-{
-	int i;
-	struct early_res *r;
-	u64 lower_start, lower_end;
-	u64 upper_start, upper_end;
-	char name[15];
-
-	for (i = 0; i < max_early_res && early_res[i].end; i++) {
-		r = &early_res[i];
-
-		/* Continue past non-overlapping ranges */
-		if (end <= r->start || start >= r->end)
-			continue;
-
-		/*
-		 * Leave non-ok overlaps as is; let caller
-		 * panic "Overlapping early reservations"
-		 * when it hits this overlap.
-		 */
-		if (!r->overlap_ok)
-			return;
-
-		/*
-		 * We have an ok overlap.  We will drop it from the early
-		 * reservation map, and add back in any non-overlapping
-		 * portions (lower or upper) as separate, overlap_ok,
-		 * non-overlapping ranges.
-		 */
-
-		/* 1. Note any non-overlapping (lower or upper) ranges. */
-		strncpy(name, r->name, sizeof(name) - 1);
-
-		lower_start = lower_end = 0;
-		upper_start = upper_end = 0;
-		if (r->start < start) {
-			lower_start = r->start;
-			lower_end = start;
-		}
-		if (r->end > end) {
-			upper_start = end;
-			upper_end = r->end;
-		}
-
-		/* 2. Drop the original ok overlapping range */
-		drop_range(i);
-
-		i--;		/* resume for-loop on copied down entry */
-
-		/* 3. Add back in any non-overlapping ranges. */
-		if (lower_end)
-			reserve_early_overlap_ok(lower_start, lower_end, name);
-		if (upper_end)
-			reserve_early_overlap_ok(upper_start, upper_end, name);
-	}
-}
-
-static void __init __reserve_early(u64 start, u64 end, char *name,
-						int overlap_ok)
-{
-	int i;
-	struct early_res *r;
-
-	i = find_overlapped_early(start, end);
-	if (i >= max_early_res)
-		panic("Too many early reservations");
-	r = &early_res[i];
-	if (r->end)
-		panic("Overlapping early reservations "
-		      "%llx-%llx %s to %llx-%llx %s\n",
-		      start, end - 1, name ? name : "", r->start,
-		      r->end - 1, r->name);
-	r->start = start;
-	r->end = end;
-	r->overlap_ok = overlap_ok;
-	if (name)
-		strncpy(r->name, name, sizeof(r->name) - 1);
-	early_res_count++;
-}
-
-/*
- * A few early reservtations come here.
- *
- * The 'overlap_ok' in the name of this routine does -not- mean it
- * is ok for these reservations to overlap an earlier reservation.
- * Rather it means that it is ok for subsequent reservations to
- * overlap this one.
- *
- * Use this entry point to reserve early ranges when you are doing
- * so out of "Paranoia", reserving perhaps more memory than you need,
- * just in case, and don't mind a subsequent overlapping reservation
- * that is known to be needed.
- *
- * The drop_overlaps_that_are_ok() call here isn't really needed.
- * It would be needed if we had two colliding 'overlap_ok'
- * reservations, so that the second such would not panic on the
- * overlap with the first.  We don't have any such as of this
- * writing, but might as well tolerate such if it happens in
- * the future.
- */
-void __init reserve_early_overlap_ok(u64 start, u64 end, char *name)
-{
-	drop_overlaps_that_are_ok(start, end);
-	__reserve_early(start, end, name, 1);
-}
-
-static void __init __check_and_double_early_res(u64 ex_start, u64 ex_end)
-{
-	u64 start, end, size, mem;
-	struct early_res *new;
-
-	/* do we have enough slots left ? */
-	if ((max_early_res - early_res_count) > max(max_early_res/8, 2))
-		return;
-
-	/* double it */
-	mem = -1ULL;
-	size = sizeof(struct early_res) * max_early_res * 2;
-	if (early_res == early_res_x)
-		start = 0;
-	else
-		start = early_res[0].end;
-	end = ex_start;
-	if (start + size < end)
-		mem = find_fw_memmap_area(start, end, size,
-					 sizeof(struct early_res));
-	if (mem == -1ULL) {
-		start = ex_end;
-		end = get_max_mapped();
-		if (start + size < end)
-			mem = find_fw_memmap_area(start, end, size,
-						 sizeof(struct early_res));
-	}
-	if (mem == -1ULL)
-		panic("can not find more space for early_res array");
-
-	new = __va(mem);
-	/* save the first one for own */
-	new[0].start = mem;
-	new[0].end = mem + size;
-	new[0].overlap_ok = 0;
-	/* copy old to new */
-	if (early_res == early_res_x) {
-		memcpy(&new[1], &early_res[0],
-			 sizeof(struct early_res) * max_early_res);
-		memset(&new[max_early_res+1], 0,
-			 sizeof(struct early_res) * (max_early_res - 1));
-		early_res_count++;
-	} else {
-		memcpy(&new[1], &early_res[1],
-			 sizeof(struct early_res) * (max_early_res - 1));
-		memset(&new[max_early_res], 0,
-			 sizeof(struct early_res) * max_early_res);
-	}
-	memset(&early_res[0], 0, sizeof(struct early_res) * max_early_res);
-	early_res = new;
-	max_early_res *= 2;
-	printk(KERN_DEBUG "early_res array is doubled to %d at [%llx - %llx]\n",
-		max_early_res, mem, mem + size - 1);
-}
-
-/*
- * Most early reservations come here.
- *
- * We first have drop_overlaps_that_are_ok() drop any pre-existing
- * 'overlap_ok' ranges, so that we can then reserve this memory
- * range without risk of panic'ing on an overlapping overlap_ok
- * early reservation.
- */
-void __init reserve_early(u64 start, u64 end, char *name)
-{
-	if (start >= end)
-		return;
-
-	__check_and_double_early_res(start, end);
-
-	drop_overlaps_that_are_ok(start, end);
-	__reserve_early(start, end, name, 0);
-}
-
-void __init reserve_early_without_check(u64 start, u64 end, char *name)
-{
-	struct early_res *r;
-
-	if (start >= end)
-		return;
-
-	__check_and_double_early_res(start, end);
-
-	r = &early_res[early_res_count];
-
-	r->start = start;
-	r->end = end;
-	r->overlap_ok = 0;
-	if (name)
-		strncpy(r->name, name, sizeof(r->name) - 1);
-	early_res_count++;
-}
-
-void __init free_early(u64 start, u64 end)
-{
-	struct early_res *r;
-	int i;
-
-	kmemleak_free_part(__va(start), end - start);
-
-	i = find_overlapped_early(start, end);
-	r = &early_res[i];
-	if (i >= max_early_res || r->end != end || r->start != start)
-		panic("free_early on not reserved area: %llx-%llx!",
-			 start, end - 1);
-
-	drop_range(i);
-}
-
-void __init free_early_partial(u64 start, u64 end)
-{
-	struct early_res *r;
-	int i;
-
-	kmemleak_free_part(__va(start), end - start);
-
-	if (start == end)
-		return;
-
-	if (WARN_ONCE(start > end, "  wrong range [%#llx, %#llx]\n", start, end))
-		return;
-
-try_next:
-	i = find_overlapped_early(start, end);
-	if (i >= max_early_res)
-		return;
-
-	r = &early_res[i];
-	/* hole ? */
-	if (r->end >= end && r->start <= start) {
-		drop_range_partial(i, start, end);
-		return;
-	}
-
-	drop_range_partial(i, start, end);
-	goto try_next;
-}
-
-#ifdef CONFIG_NO_BOOTMEM
-static void __init subtract_early_res(struct range *range, int az)
-{
-	int i, count;
-	u64 final_start, final_end;
-	int idx = 0;
-
-	count  = 0;
-	for (i = 0; i < max_early_res && early_res[i].end; i++)
-		count++;
-
-	/* need to skip first one ?*/
-	if (early_res != early_res_x)
-		idx = 1;
-
-#define DEBUG_PRINT_EARLY_RES 1
-
-#if DEBUG_PRINT_EARLY_RES
-	printk(KERN_INFO "Subtract (%d early reservations)\n", count);
-#endif
-	for (i = idx; i < count; i++) {
-		struct early_res *r = &early_res[i];
-#if DEBUG_PRINT_EARLY_RES
-		printk(KERN_INFO "  #%d [%010llx - %010llx] %15s\n", i,
-			r->start, r->end, r->name);
-#endif
-		final_start = PFN_DOWN(r->start);
-		final_end = PFN_UP(r->end);
-		if (final_start >= final_end)
-			continue;
-		subtract_range(range, az, final_start, final_end);
-	}
-
-}
-
-int __init get_free_all_memory_range(struct range **rangep, int nodeid)
-{
-	int i, count;
-	u64 start = 0, end;
-	u64 size;
-	u64 mem;
-	struct range *range;
-	int nr_range;
-
-	count  = 0;
-	for (i = 0; i < max_early_res && early_res[i].end; i++)
-		count++;
-
-	count *= 2;
-
-	size = sizeof(struct range) * count;
-	end = get_max_mapped();
-#ifdef MAX_DMA32_PFN
-	if (end > (MAX_DMA32_PFN << PAGE_SHIFT))
-		start = MAX_DMA32_PFN << PAGE_SHIFT;
-#endif
-	mem = find_fw_memmap_area(start, end, size, sizeof(struct range));
-	if (mem == -1ULL)
-		panic("can not find more space for range free");
-
-	range = __va(mem);
-	/* use early_node_map[] and early_res to get range array at first */
-	memset(range, 0, size);
-	nr_range = 0;
-
-	/* need to go over early_node_map to find out good range for node */
-	nr_range = add_from_early_node_map(range, count, nr_range, nodeid);
-#ifdef CONFIG_X86_32
-	subtract_range(range, count, max_low_pfn, -1ULL);
-#endif
-	subtract_early_res(range, count);
-	nr_range = clean_sort_range(range, count);
-
-	/* need to clear it ? */
-	if (nodeid == MAX_NUMNODES) {
-		memset(&early_res[0], 0,
-			 sizeof(struct early_res) * max_early_res);
-		early_res = NULL;
-		max_early_res = 0;
-	}
-
-	*rangep = range;
-	return nr_range;
-}
-#else
-void __init early_res_to_bootmem(u64 start, u64 end)
-{
-	int i, count;
-	u64 final_start, final_end;
-	int idx = 0;
-
-	count  = 0;
-	for (i = 0; i < max_early_res && early_res[i].end; i++)
-		count++;
-
-	/* need to skip first one ?*/
-	if (early_res != early_res_x)
-		idx = 1;
-
-	printk(KERN_INFO "(%d/%d early reservations) ==> bootmem [%010llx - %010llx]\n",
-			 count - idx, max_early_res, start, end);
-	for (i = idx; i < count; i++) {
-		struct early_res *r = &early_res[i];
-		printk(KERN_INFO "  #%d [%010llx - %010llx] %16s", i,
-			r->start, r->end, r->name);
-		final_start = max(start, r->start);
-		final_end = min(end, r->end);
-		if (final_start >= final_end) {
-			printk(KERN_CONT "\n");
-			continue;
-		}
-		printk(KERN_CONT " ==> [%010llx - %010llx]\n",
-			final_start, final_end);
-		reserve_bootmem_generic(final_start, final_end - final_start,
-				BOOTMEM_DEFAULT);
-	}
-	/* clear them */
-	memset(&early_res[0], 0, sizeof(struct early_res) * max_early_res);
-	early_res = NULL;
-	max_early_res = 0;
-	early_res_count = 0;
-}
-#endif
-
-/* Check for already reserved areas */
-static inline int __init bad_addr(u64 *addrp, u64 size, u64 align)
-{
-	int i;
-	u64 addr = *addrp;
-	int changed = 0;
-	struct early_res *r;
-again:
-	i = find_overlapped_early(addr, addr + size);
-	r = &early_res[i];
-	if (i < max_early_res && r->end) {
-		*addrp = addr = round_up(r->end, align);
-		changed = 1;
-		goto again;
-	}
-	return changed;
-}
-
-/* Check for already reserved areas */
-static inline int __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align)
-{
-	int i;
-	u64 addr = *addrp, last;
-	u64 size = *sizep;
-	int changed = 0;
-again:
-	last = addr + size;
-	for (i = 0; i < max_early_res && early_res[i].end; i++) {
-		struct early_res *r = &early_res[i];
-		if (last > r->start && addr < r->start) {
-			size = r->start - addr;
-			changed = 1;
-			goto again;
-		}
-		if (last > r->end && addr < r->end) {
-			addr = round_up(r->end, align);
-			size = last - addr;
-			changed = 1;
-			goto again;
-		}
-		if (last <= r->end && addr >= r->start) {
-			(*sizep)++;
-			return 0;
-		}
-	}
-	if (changed) {
-		*addrp = addr;
-		*sizep = size;
-	}
-	return changed;
-}
-
-/*
- * Find a free area with specified alignment in a specific range.
- * only with the area.between start to end is active range from early_node_map
- * so they are good as RAM
- */
-u64 __init find_early_area(u64 ei_start, u64 ei_last, u64 start, u64 end,
-			 u64 size, u64 align)
-{
-	u64 addr, last;
-
-	addr = round_up(ei_start, align);
-	if (addr < start)
-		addr = round_up(start, align);
-	if (addr >= ei_last)
-		goto out;
-	while (bad_addr(&addr, size, align) && addr+size <= ei_last)
-		;
-	last = addr + size;
-	if (last > ei_last)
-		goto out;
-	if (last > end)
-		goto out;
-
-	return addr;
-
-out:
-	return -1ULL;
-}
-
-u64 __init find_early_area_size(u64 ei_start, u64 ei_last, u64 start,
-			 u64 *sizep, u64 align)
-{
-	u64 addr, last;
-
-	addr = round_up(ei_start, align);
-	if (addr < start)
-		addr = round_up(start, align);
-	if (addr >= ei_last)
-		goto out;
-	*sizep = ei_last - addr;
-	while (bad_addr_size(&addr, sizep, align) && addr + *sizep <= ei_last)
-		;
-	last = addr + *sizep;
-	if (last > ei_last)
-		goto out;
-
-	return addr;
-
-out:
-	return -1ULL;
-}
-- 
cgit v1.2.3


From 409456b10f87b28303643fec37543103f9ada00c Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@verge.net.au>
Date: Sun, 29 Aug 2010 21:57:55 -0700
Subject: net: fix datapath typo

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/if.h b/include/linux/if.h
index 6ed43c1f07ab..123959927745 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -76,7 +76,7 @@
 #define IFF_MACVLAN_PORT	0x4000	/* device used as macvlan port */
 #define IFF_BRIDGE_PORT	0x8000		/* device used as bridge port */
 #define IFF_OVS_DATAPATH	0x10000	/* device used as Open vSwitch
-					 * dapath port */
+					 * datapath port */
 
 #define IF_GET_IFACE	0x0001		/* for querying only */
 #define IF_GET_PROTO	0x0002
-- 
cgit v1.2.3


From dca43c75e7e545694a9dd6288553f55c53e2a3a3 Mon Sep 17 00:00:00 2001
From: Jerry Chu <hkchu@google.com>
Date: Fri, 27 Aug 2010 19:13:28 +0000
Subject: tcp: Add TCP_USER_TIMEOUT socket option.

This patch provides a "user timeout" support as described in RFC793. The
socket option is also needed for the the local half of RFC5482 "TCP User
Timeout Option".

TCP_USER_TIMEOUT is a TCP level socket option that takes an unsigned int,
when > 0, to specify the maximum amount of time in ms that transmitted
data may remain unacknowledged before TCP will forcefully close the
corresponding connection and return ETIMEDOUT to the application. If
0 is given, TCP will continue to use the system default.

Increasing the user timeouts allows a TCP connection to survive extended
periods without end-to-end connectivity. Decreasing the user timeouts
allows applications to "fail fast" if so desired. Otherwise it may take
upto 20 minutes with the current system defaults in a normal WAN
environment.

The socket option can be made during any state of a TCP connection, but
is only effective during the synchronized states of a connection
(ESTABLISHED, FIN-WAIT-1, FIN-WAIT-2, CLOSE-WAIT, CLOSING, or LAST-ACK).
Moreover, when used with the TCP keepalive (SO_KEEPALIVE) option,
TCP_USER_TIMEOUT will overtake keepalive to determine when to close a
connection due to keepalive failure.

The option does not change in anyway when TCP retransmits a packet, nor
when a keepalive probe will be sent.

This option, like many others, will be inherited by an acceptor from its
listener.

Signed-off-by: H.K. Jerry Chu <hkchu@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h                |  1 +
 include/net/inet_connection_sock.h |  1 +
 net/ipv4/tcp.c                     | 11 ++++++++++-
 net/ipv4/tcp_timer.c               | 40 ++++++++++++++++++++++++--------------
 4 files changed, 37 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index a778ee024590..e64f4c67d0ef 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -105,6 +105,7 @@ enum {
 #define TCP_COOKIE_TRANSACTIONS	15	/* TCP Cookie Transactions */
 #define TCP_THIN_LINEAR_TIMEOUTS 16      /* Use linear timeouts for thin streams*/
 #define TCP_THIN_DUPACK         17      /* Fast retrans. after 1 dupack */
+#define TCP_USER_TIMEOUT	18	/* How long for loss retry before timeout */
 
 /* for TCP_INFO socket option */
 #define TCPI_OPT_TIMESTAMPS	1
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index b6d3b55da19b..e4f494b42e06 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -125,6 +125,7 @@ struct inet_connection_sock {
 		int		  probe_size;
 	} icsk_mtup;
 	u32			  icsk_ca_priv[16];
+	u32			  icsk_user_timeout;
 #define ICSK_CA_PRIV_SIZE	(16 * sizeof(u32))
 };
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 176e11aaea77..cf3254528753 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2391,7 +2391,12 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
 		err = tp->af_specific->md5_parse(sk, optval, optlen);
 		break;
 #endif
-
+	case TCP_USER_TIMEOUT:
+		/* Cap the max timeout in ms TCP will retry/retrans
+		 * before giving up and aborting (ETIMEDOUT) a connection.
+		 */
+		icsk->icsk_user_timeout = msecs_to_jiffies(val);
+		break;
 	default:
 		err = -ENOPROTOOPT;
 		break;
@@ -2610,6 +2615,10 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 	case TCP_THIN_DUPACK:
 		val = tp->thin_dupack;
 		break;
+
+	case TCP_USER_TIMEOUT:
+		val = jiffies_to_msecs(icsk->icsk_user_timeout);
+		break;
 	default:
 		return -ENOPROTOOPT;
 	}
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 808bb920c9f5..11569deccbea 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -138,10 +138,10 @@ static void tcp_mtu_probing(struct inet_connection_sock *icsk, struct sock *sk)
  * retransmissions with an initial RTO of TCP_RTO_MIN.
  */
 static bool retransmits_timed_out(struct sock *sk,
-				  unsigned int boundary)
+				  unsigned int boundary,
+				  unsigned int timeout)
 {
-	unsigned int timeout, linear_backoff_thresh;
-	unsigned int start_ts;
+	unsigned int linear_backoff_thresh, start_ts;
 
 	if (!inet_csk(sk)->icsk_retransmits)
 		return false;
@@ -151,14 +151,15 @@ static bool retransmits_timed_out(struct sock *sk,
 	else
 		start_ts = tcp_sk(sk)->retrans_stamp;
 
-	linear_backoff_thresh = ilog2(TCP_RTO_MAX/TCP_RTO_MIN);
-
-	if (boundary <= linear_backoff_thresh)
-		timeout = ((2 << boundary) - 1) * TCP_RTO_MIN;
-	else
-		timeout = ((2 << linear_backoff_thresh) - 1) * TCP_RTO_MIN +
-			  (boundary - linear_backoff_thresh) * TCP_RTO_MAX;
+	if (likely(timeout == 0)) {
+		linear_backoff_thresh = ilog2(TCP_RTO_MAX/TCP_RTO_MIN);
 
+		if (boundary <= linear_backoff_thresh)
+			timeout = ((2 << boundary) - 1) * TCP_RTO_MIN;
+		else
+			timeout = ((2 << linear_backoff_thresh) - 1) * TCP_RTO_MIN +
+				(boundary - linear_backoff_thresh) * TCP_RTO_MAX;
+	}
 	return (tcp_time_stamp - start_ts) >= timeout;
 }
 
@@ -174,7 +175,7 @@ static int tcp_write_timeout(struct sock *sk)
 			dst_negative_advice(sk);
 		retry_until = icsk->icsk_syn_retries ? : sysctl_tcp_syn_retries;
 	} else {
-		if (retransmits_timed_out(sk, sysctl_tcp_retries1)) {
+		if (retransmits_timed_out(sk, sysctl_tcp_retries1, 0)) {
 			/* Black hole detection */
 			tcp_mtu_probing(icsk, sk);
 
@@ -187,14 +188,16 @@ static int tcp_write_timeout(struct sock *sk)
 
 			retry_until = tcp_orphan_retries(sk, alive);
 			do_reset = alive ||
-				   !retransmits_timed_out(sk, retry_until);
+				   !retransmits_timed_out(sk, retry_until, 0);
 
 			if (tcp_out_of_resources(sk, do_reset))
 				return 1;
 		}
 	}
 
-	if (retransmits_timed_out(sk, retry_until)) {
+	if (retransmits_timed_out(sk, retry_until,
+	    (1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV) ? 0 :
+	    icsk->icsk_user_timeout)) {
 		/* Has it gone just too far? */
 		tcp_write_err(sk);
 		return 1;
@@ -436,7 +439,7 @@ out_reset_timer:
 		icsk->icsk_rto = min(icsk->icsk_rto << 1, TCP_RTO_MAX);
 	}
 	inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, TCP_RTO_MAX);
-	if (retransmits_timed_out(sk, sysctl_tcp_retries1 + 1))
+	if (retransmits_timed_out(sk, sysctl_tcp_retries1 + 1, 0))
 		__sk_dst_reset(sk);
 
 out:;
@@ -556,7 +559,14 @@ static void tcp_keepalive_timer (unsigned long data)
 	elapsed = keepalive_time_elapsed(tp);
 
 	if (elapsed >= keepalive_time_when(tp)) {
-		if (icsk->icsk_probes_out >= keepalive_probes(tp)) {
+		/* If the TCP_USER_TIMEOUT option is enabled, use that
+		 * to determine when to timeout instead.
+		 */
+		if ((icsk->icsk_user_timeout != 0 &&
+		    elapsed >= icsk->icsk_user_timeout &&
+		    icsk->icsk_probes_out > 0) ||
+		    (icsk->icsk_user_timeout == 0 &&
+		    icsk->icsk_probes_out >= keepalive_probes(tp))) {
 			tcp_send_active_reset(sk, GFP_ATOMIC);
 			tcp_write_err(sk);
 			goto out;
-- 
cgit v1.2.3


From 4dc89133f49b8cfd77ba7e83f5960aed63aaa99e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 31 Aug 2010 07:40:16 +0000
Subject: net: add a comment on netdev->last_rx

As some driver authors seem to reintroduce dev->last_rx use,
add a comment to strongly discourage this.

Since commit 6cf3f41e6c0 (bonding, net: Move last_rx update into bonding
recv logic), network drivers dont need to update last_rx themselves,
unless they use this field to implement a timeout.

Not updating last_rx helps not dirtying a cache line, improving
performance in SMP.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 0cf9448a32c4..c82220a9f3d5 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -953,7 +953,14 @@ struct net_device {
 /*
  * Cache line mostly used on receive path (including eth_type_trans())
  */
-	unsigned long		last_rx;	/* Time of last Rx	*/
+	unsigned long		last_rx;	/* Time of last Rx
+						 * This should not be set in
+						 * drivers, unless really needed,
+						 * because network stack (bonding)
+						 * use it if/when necessary, to
+						 * avoid dirtying this cache line.
+						 */
+
 	/* Interface address info used in eth_type_trans() */
 	unsigned char		*dev_addr;	/* hw address, (before bcast
 						   because most packets are
-- 
cgit v1.2.3


From 86cac58b71227cc34a3d0e78f19585c0eff49ea3 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 31 Aug 2010 18:25:32 +0000
Subject: skge: add GRO support

- napi_gro_flush() is exported from net/core/dev.c, to avoid
  an irq_save/irq_restore in the packet receive path.
- use napi_gro_receive() instead of netif_receive_skb()
- use napi_gro_flush() before calling __napi_complete()
- turn on NETIF_F_GRO by default
- Tested on a Marvell 88E8001 Gigabit NIC

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/skge.c        | 5 +++--
 include/linux/netdevice.h | 1 +
 net/core/dev.c            | 3 ++-
 3 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/skge.c b/drivers/net/skge.c
index 40e5c46e7571..a8a63581d63d 100644
--- a/drivers/net/skge.c
+++ b/drivers/net/skge.c
@@ -3178,8 +3178,7 @@ static int skge_poll(struct napi_struct *napi, int to_do)
 
 		skb = skge_rx_get(dev, e, control, rd->status, rd->csum2);
 		if (likely(skb)) {
-			netif_receive_skb(skb);
-
+			napi_gro_receive(napi, skb);
 			++work_done;
 		}
 	}
@@ -3192,6 +3191,7 @@ static int skge_poll(struct napi_struct *napi, int to_do)
 	if (work_done < to_do) {
 		unsigned long flags;
 
+		napi_gro_flush(napi);
 		spin_lock_irqsave(&hw->hw_lock, flags);
 		__napi_complete(napi);
 		hw->intr_mask |= napimask[skge->port];
@@ -3849,6 +3849,7 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port,
 		dev->features |= NETIF_F_IP_CSUM | NETIF_F_SG;
 		skge->rx_csum = 1;
 	}
+	dev->features |= NETIF_F_GRO;
 
 	/* read the mac address */
 	memcpy_fromio(dev->dev_addr, hw->regs + B2_MAC_1 + port*8, ETH_ALEN);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c82220a9f3d5..af05186d5b36 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1702,6 +1702,7 @@ extern gro_result_t	dev_gro_receive(struct napi_struct *napi,
 extern gro_result_t	napi_skb_finish(gro_result_t ret, struct sk_buff *skb);
 extern gro_result_t	napi_gro_receive(struct napi_struct *napi,
 					 struct sk_buff *skb);
+extern void		napi_gro_flush(struct napi_struct *napi);
 extern void		napi_reuse_skb(struct napi_struct *napi,
 				       struct sk_buff *skb);
 extern struct sk_buff *	napi_get_frags(struct napi_struct *napi);
diff --git a/net/core/dev.c b/net/core/dev.c
index 63bd20a75929..d8c43e73f0b7 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3063,7 +3063,7 @@ out:
 	return netif_receive_skb(skb);
 }
 
-static void napi_gro_flush(struct napi_struct *napi)
+inline void napi_gro_flush(struct napi_struct *napi)
 {
 	struct sk_buff *skb, *next;
 
@@ -3076,6 +3076,7 @@ static void napi_gro_flush(struct napi_struct *napi)
 	napi->gro_count = 0;
 	napi->gro_list = NULL;
 }
+EXPORT_SYMBOL(napi_gro_flush);
 
 enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 {
-- 
cgit v1.2.3


From c68839963426d42bdb2c915b435f9860d060e645 Mon Sep 17 00:00:00 2001
From: Peter Meerwald <pmeerw@pmeerw.net>
Date: Thu, 2 Sep 2010 04:06:24 +0000
Subject: net: Improve comments in include/linux/phy.h

Correct state range of PHY bus addresses (i.e. 0-31) in comment,
make spelling of PHY consistent in comments.

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/phy.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/phy.h b/include/linux/phy.h
index 6b0a782c6224..a6e047a04f79 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -116,7 +116,7 @@ struct mii_bus {
 	/* list of all PHYs on bus */
 	struct phy_device *phy_map[PHY_MAX_ADDR];
 
-	/* Phy addresses to be ignored when probing */
+	/* PHY addresses to be ignored when probing */
 	u32 phy_mask;
 
 	/*
@@ -283,7 +283,7 @@ struct phy_device {
 
 	phy_interface_t interface;
 
-	/* Bus address of the PHY (0-32) */
+	/* Bus address of the PHY (0-31) */
 	int addr;
 
 	/*
-- 
cgit v1.2.3


From bc8acf2c8c3e43fcc192762a9f964b3e9a17748b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 2 Sep 2010 13:07:41 -0700
Subject: drivers/net: avoid some skb->ip_summed initializations

fresh skbs have ip_summed set to CHECKSUM_NONE (0)

We can avoid setting again skb->ip_summed to CHECKSUM_NONE in drivers.

Introduce skb_checksum_none_assert() helper so that we keep this
assertion documented in driver sources.

Change most occurrences of :

skb->ip_summed = CHECKSUM_NONE;

by :

skb_checksum_none_assert(skb);

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/8139cp.c               |  2 +-
 drivers/net/acenic.c               |  2 +-
 drivers/net/atl1c/atl1c_main.c     |  2 +-
 drivers/net/atl1e/atl1e_main.c     |  2 +-
 drivers/net/atlx/atl1.c            |  2 +-
 drivers/net/b44.c                  |  2 +-
 drivers/net/benet/be_main.c        |  2 +-
 drivers/net/bna/bnad.c             |  2 +-
 drivers/net/bnx2.c                 |  2 +-
 drivers/net/bnx2x/bnx2x_cmn.c      |  2 +-
 drivers/net/cassini.c              |  2 +-
 drivers/net/chelsio/sge.c          |  2 +-
 drivers/net/cpmac.c                |  2 +-
 drivers/net/cxgb3/sge.c            |  2 +-
 drivers/net/cxgb4/sge.c            |  2 +-
 drivers/net/cxgb4vf/sge.c          |  2 +-
 drivers/net/dm9000.c               |  2 +-
 drivers/net/e1000/e1000_main.c     |  3 ++-
 drivers/net/e1000e/netdev.c        |  3 ++-
 drivers/net/gianfar.c              |  2 +-
 drivers/net/greth.c                |  2 +-
 drivers/net/ibmlana.c              |  2 +-
 drivers/net/igb/igb_main.c         |  2 +-
 drivers/net/igbvf/netdev.c         |  2 +-
 drivers/net/ipg.c                  |  6 +++---
 drivers/net/iseries_veth.c         |  2 +-
 drivers/net/ixgb/ixgb_main.c       |  4 ++--
 drivers/net/ixgbe/ixgbe_fcoe.c     |  5 +++--
 drivers/net/ixgbe/ixgbe_main.c     |  2 +-
 drivers/net/ixgbevf/ixgbevf_main.c |  2 +-
 drivers/net/jme.c                  |  2 +-
 drivers/net/ll_temac_main.c        |  2 +-
 drivers/net/macb.c                 |  2 +-
 drivers/net/niu.c                  |  2 +-
 drivers/net/ns83820.c              |  2 +-
 drivers/net/pasemi_mac.c           |  2 +-
 drivers/net/ps3_gelic_net.c        |  4 ++--
 drivers/net/qla3xxx.c              |  4 ++--
 drivers/net/qlcnic/qlcnic_init.c   |  2 +-
 drivers/net/qlge/qlge_main.c       |  6 +++---
 drivers/net/r8169.c                |  2 +-
 drivers/net/s2io.c                 |  4 ++--
 drivers/net/sb1250-mac.c           |  2 +-
 drivers/net/sfc/rx.c               |  2 +-
 drivers/net/sh_eth.c               |  2 +-
 drivers/net/smsc911x.c             |  2 +-
 drivers/net/spider_net.c           |  4 ++--
 drivers/net/stmmac/stmmac_main.c   |  2 +-
 drivers/net/tehuti.c               |  5 +++--
 drivers/net/tg3.c                  |  2 +-
 drivers/net/typhoon.c              |  2 +-
 drivers/net/via-velocity.c         |  2 +-
 drivers/net/vmxnet3/vmxnet3_drv.c  |  4 ++--
 drivers/net/vxge/vxge-main.c       |  2 +-
 drivers/net/xilinx_emaclite.c      |  2 +-
 include/linux/skbuff.h             | 15 +++++++++++++++
 56 files changed, 86 insertions(+), 67 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/8139cp.c b/drivers/net/8139cp.c
index 4a4f6b81e32d..237d4ea5a416 100644
--- a/drivers/net/8139cp.c
+++ b/drivers/net/8139cp.c
@@ -561,7 +561,7 @@ rx_status_loop:
 		if (cp_rx_csum_ok(status))
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		else
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 
 		skb_put(skb, len);
 
diff --git a/drivers/net/acenic.c b/drivers/net/acenic.c
index b9a591604e5b..41d9911202d0 100644
--- a/drivers/net/acenic.c
+++ b/drivers/net/acenic.c
@@ -2033,7 +2033,7 @@ static void ace_rx_int(struct net_device *dev, u32 rxretprd, u32 rxretcsm)
 			skb->csum = htons(csum);
 			skb->ip_summed = CHECKSUM_COMPLETE;
 		} else {
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 		}
 
 		/* send it up */
diff --git a/drivers/net/atl1c/atl1c_main.c b/drivers/net/atl1c/atl1c_main.c
index 61f1634ab3f8..553230eb365c 100644
--- a/drivers/net/atl1c/atl1c_main.c
+++ b/drivers/net/atl1c/atl1c_main.c
@@ -1719,7 +1719,7 @@ static inline void atl1c_rx_checksum(struct atl1c_adapter *adapter,
 	 * cannot figure out if the packet is fragmented or not,
 	 * so we tell the KERNEL CHECKSUM_NONE
 	 */
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 }
 
 static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter, const int ringid)
diff --git a/drivers/net/atl1e/atl1e_main.c b/drivers/net/atl1e/atl1e_main.c
index 1ae44bb26317..56ace3fbe40d 100644
--- a/drivers/net/atl1e/atl1e_main.c
+++ b/drivers/net/atl1e/atl1e_main.c
@@ -1331,7 +1331,7 @@ static inline void atl1e_rx_checksum(struct atl1e_adapter *adapter,
 	u16 pkt_flags;
 	u16 err_flags;
 
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 	pkt_flags = prrs->pkt_flag;
 	err_flags = prrs->err_flag;
 	if (((pkt_flags & RRS_IS_IPV4) || (pkt_flags & RRS_IS_IPV6)) &&
diff --git a/drivers/net/atlx/atl1.c b/drivers/net/atlx/atl1.c
index 5837b0184d4b..e1e0171d6e62 100644
--- a/drivers/net/atlx/atl1.c
+++ b/drivers/net/atlx/atl1.c
@@ -1805,7 +1805,7 @@ static void atl1_rx_checksum(struct atl1_adapter *adapter,
 	 * the higher layers and let it be sorted out there.
 	 */
 
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	if (unlikely(rrd->pkt_flg & PACKET_FLAG_ERR)) {
 		if (rrd->err_flg & (ERR_FLAG_CRC | ERR_FLAG_TRUNC |
diff --git a/drivers/net/b44.c b/drivers/net/b44.c
index 37617abc1647..7342308718a0 100644
--- a/drivers/net/b44.c
+++ b/drivers/net/b44.c
@@ -818,7 +818,7 @@ static int b44_rx(struct b44 *bp, int budget)
 							 copy_skb->data, len);
 			skb = copy_skb;
 		}
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 		skb->protocol = eth_type_trans(skb, bp->dev);
 		netif_receive_skb(skb);
 		received++;
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index 9db10fec8235..dcee7a20c0b9 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -1016,7 +1016,7 @@ static void be_rx_compl_process(struct be_adapter *adapter,
 	skb_fill_rx_data(adapter, skb, rxcp, num_rcvd);
 
 	if (do_pkt_csum(rxcp, adapter->rx_csum))
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 	else
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 
diff --git a/drivers/net/bna/bnad.c b/drivers/net/bna/bnad.c
index 79c4c2441449..44adc7aefddc 100644
--- a/drivers/net/bna/bnad.c
+++ b/drivers/net/bna/bnad.c
@@ -510,7 +510,7 @@ bnad_poll_cq(struct bnad *bnad, struct bna_ccb *ccb, int budget)
 		      (flags & BNA_CQ_EF_L4_CKSUM_OK)))
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		else
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 
 		rcb->rxq->rx_packets++;
 		rcb->rxq->rx_bytes += skb->len;
diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index a0e02aa42214..4ff76e38e788 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -3218,7 +3218,7 @@ bnx2_rx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
 
 		}
 
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 		if (bp->rx_csum &&
 			(status & (L2_FHDR_STATUS_TCP_SEGMENT |
 			L2_FHDR_STATUS_UDP_DATAGRAM))) {
diff --git a/drivers/net/bnx2x/bnx2x_cmn.c b/drivers/net/bnx2x/bnx2x_cmn.c
index da96d1a18c20..0e4caf411905 100644
--- a/drivers/net/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/bnx2x/bnx2x_cmn.c
@@ -623,7 +623,7 @@ reuse_rx:
 			/* Set Toeplitz hash for a none-LRO skb */
 			bnx2x_set_skb_rxhash(bp, cqe, skb);
 
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 			if (bp->rx_csum) {
 				if (likely(BNX2X_RX_CSUM_OK(cqe)))
 					skb->ip_summed = CHECKSUM_UNNECESSARY;
diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c
index 28c88eeec757..32aaadc4734f 100644
--- a/drivers/net/cassini.c
+++ b/drivers/net/cassini.c
@@ -2149,7 +2149,7 @@ end_copy_pkt:
 		skb->csum = csum_unfold(~csum);
 		skb->ip_summed = CHECKSUM_COMPLETE;
 	} else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 	return len;
 }
 
diff --git a/drivers/net/chelsio/sge.c b/drivers/net/chelsio/sge.c
index f01cfdb995de..1950b9a20ecd 100644
--- a/drivers/net/chelsio/sge.c
+++ b/drivers/net/chelsio/sge.c
@@ -1388,7 +1388,7 @@ static void sge_rx(struct sge *sge, struct freelQ *fl, unsigned int len)
 		++st->rx_cso_good;
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	} else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 
 	if (unlikely(adapter->vlan_grp && p->vlan_valid)) {
 		st->vlan_xtract++;
diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index 5a5af1ca7541..fec939f8f65f 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -391,7 +391,7 @@ static struct sk_buff *cpmac_rx_one(struct cpmac_priv *priv,
 	if (likely(skb)) {
 		skb_put(desc->skb, desc->datalen);
 		desc->skb->protocol = eth_type_trans(desc->skb, priv->dev);
-		desc->skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(desc->skb);
 		priv->dev->stats.rx_packets++;
 		priv->dev->stats.rx_bytes += desc->datalen;
 		result = desc->skb;
diff --git a/drivers/net/cxgb3/sge.c b/drivers/net/cxgb3/sge.c
index 8ff96c6f6de5..c5a142bea5e9 100644
--- a/drivers/net/cxgb3/sge.c
+++ b/drivers/net/cxgb3/sge.c
@@ -2022,7 +2022,7 @@ static void rx_eth(struct adapter *adap, struct sge_rspq *rq,
 		qs->port_stats[SGE_PSTAT_RX_CSUM_GOOD]++;
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	} else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 	skb_record_rx_queue(skb, qs - &adap->sge.qs[0]);
 
 	if (unlikely(p->vlan_valid)) {
diff --git a/drivers/net/cxgb4/sge.c b/drivers/net/cxgb4/sge.c
index 6ddb3bb0ce67..9967f3debce7 100644
--- a/drivers/net/cxgb4/sge.c
+++ b/drivers/net/cxgb4/sge.c
@@ -1605,7 +1605,7 @@ int t4_ethrx_handler(struct sge_rspq *q, const __be64 *rsp,
 			rxq->stats.rx_cso++;
 		}
 	} else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 
 	if (unlikely(pkt->vlan_ex)) {
 		struct vlan_group *grp = pi->vlan_grp;
diff --git a/drivers/net/cxgb4vf/sge.c b/drivers/net/cxgb4vf/sge.c
index e00fd9c36e8b..f10864ddafbe 100644
--- a/drivers/net/cxgb4vf/sge.c
+++ b/drivers/net/cxgb4vf/sge.c
@@ -1534,7 +1534,7 @@ int t4vf_ethrx_handler(struct sge_rspq *rspq, const __be64 *rsp,
 		}
 		rxq->stats.rx_cso++;
 	} else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 
 	if (unlikely(pkt->vlan_ex)) {
 		struct vlan_group *grp = pi->vlan_grp;
diff --git a/drivers/net/dm9000.c b/drivers/net/dm9000.c
index 4fd6b2b4554b..9f6aeefa06bf 100644
--- a/drivers/net/dm9000.c
+++ b/drivers/net/dm9000.c
@@ -1056,7 +1056,7 @@ dm9000_rx(struct net_device *dev)
 				if ((((rxbyte & 0x1c) << 3) & rxbyte) == 0)
 					skb->ip_summed = CHECKSUM_UNNECESSARY;
 				else
-					skb->ip_summed = CHECKSUM_NONE;
+					skb_checksum_none_assert(skb);
 			}
 			netif_rx(skb);
 			dev->stats.rx_packets++;
diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index 3e8ac4baae1b..17f5867b5d9b 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -3552,7 +3552,8 @@ static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err,
 	struct e1000_hw *hw = &adapter->hw;
 	u16 status = (u16)status_err;
 	u8 errors = (u8)(status_err >> 24);
-	skb->ip_summed = CHECKSUM_NONE;
+
+	skb_checksum_none_assert(skb);
 
 	/* 82543 or newer only */
 	if (unlikely(hw->mac_type < e1000_82543)) return;
diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index 5f3eac6432cb..c9b66f4727e4 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -475,7 +475,8 @@ static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err,
 {
 	u16 status = (u16)status_err;
 	u8 errors = (u8)(status_err >> 24);
-	skb->ip_summed = CHECKSUM_NONE;
+
+	skb_checksum_none_assert(skb);
 
 	/* Ignore Checksum bit is set */
 	if (status & E1000_RXD_STAT_IXSM)
diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c
index e6048d6ab0ea..f30adbf86bb2 100644
--- a/drivers/net/gianfar.c
+++ b/drivers/net/gianfar.c
@@ -2654,7 +2654,7 @@ static inline void gfar_rx_checksum(struct sk_buff *skb, struct rxfcb *fcb)
 	if ((fcb->flags & RXFCB_CSUM_MASK) == (RXFCB_CIP | RXFCB_CTU))
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 }
 
 
diff --git a/drivers/net/greth.c b/drivers/net/greth.c
index fbeaf70d1727..27d6960ce09e 100644
--- a/drivers/net/greth.c
+++ b/drivers/net/greth.c
@@ -893,7 +893,7 @@ static int greth_rx_gbit(struct net_device *dev, int limit)
 				if (greth->flags & GRETH_FLAG_RX_CSUM && hw_checksummed(status))
 					skb->ip_summed = CHECKSUM_UNNECESSARY;
 				else
-					skb->ip_summed = CHECKSUM_NONE;
+					skb_checksum_none_assert(skb);
 
 				skb->protocol = eth_type_trans(skb, dev);
 				dev->stats.rx_packets++;
diff --git a/drivers/net/ibmlana.c b/drivers/net/ibmlana.c
index 294ccfb427cf..0037a696cd0a 100644
--- a/drivers/net/ibmlana.c
+++ b/drivers/net/ibmlana.c
@@ -602,7 +602,7 @@ static void irqrx_handler(struct net_device *dev)
 				/* set up skb fields */
 
 				skb->protocol = eth_type_trans(skb, dev);
-				skb->ip_summed = CHECKSUM_NONE;
+				skb_checksum_none_assert(skb);
 
 				/* bookkeeping */
 				dev->stats.rx_packets++;
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index d35cc38bf8b2..c4d861b557ca 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -5455,7 +5455,7 @@ static void igb_receive_skb(struct igb_q_vector *q_vector,
 static inline void igb_rx_checksum_adv(struct igb_ring *ring,
 				       u32 status_err, struct sk_buff *skb)
 {
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	/* Ignore Checksum bit is set or checksum is disabled through ethtool */
 	if (!(ring->flags & IGB_RING_FLAG_RX_CSUM) ||
diff --git a/drivers/net/igbvf/netdev.c b/drivers/net/igbvf/netdev.c
index c539f7c9c3e0..c7fab80d2490 100644
--- a/drivers/net/igbvf/netdev.c
+++ b/drivers/net/igbvf/netdev.c
@@ -103,7 +103,7 @@ static void igbvf_receive_skb(struct igbvf_adapter *adapter,
 static inline void igbvf_rx_checksum_adv(struct igbvf_adapter *adapter,
                                          u32 status_err, struct sk_buff *skb)
 {
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	/* Ignore Checksum bit is set or checksum is disabled through ethtool */
 	if ((status_err & E1000_RXD_STAT_IXSM) ||
diff --git a/drivers/net/ipg.c b/drivers/net/ipg.c
index 72e3d2da9e9f..dc0198092343 100644
--- a/drivers/net/ipg.c
+++ b/drivers/net/ipg.c
@@ -1213,7 +1213,7 @@ static void ipg_nic_rx_with_start_and_end(struct net_device *dev,
 
 	skb_put(skb, framelen);
 	skb->protocol = eth_type_trans(skb, dev);
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 	netif_rx(skb);
 	sp->rx_buff[entry] = NULL;
 }
@@ -1278,7 +1278,7 @@ static void ipg_nic_rx_with_end(struct net_device *dev,
 				jumbo->skb->protocol =
 				    eth_type_trans(jumbo->skb, dev);
 
-				jumbo->skb->ip_summed = CHECKSUM_NONE;
+				skb_checksum_none_assert(jumbo->skb);
 				netif_rx(jumbo->skb);
 			}
 		}
@@ -1476,7 +1476,7 @@ static int ipg_nic_rx(struct net_device *dev)
 			 * IP/TCP/UDP frame was received. Let the
 			 * upper layer decide.
 			 */
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 
 			/* Hand off frame for higher layer processing.
 			 * The function netif_rx() releases the sk_buff
diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c
index ba1de5973fb2..8df645e78f2e 100644
--- a/drivers/net/iseries_veth.c
+++ b/drivers/net/iseries_veth.c
@@ -1524,7 +1524,7 @@ static void veth_receive(struct veth_lpar_connection *cnx,
 
 		skb_put(skb, length);
 		skb->protocol = eth_type_trans(skb, dev);
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 		netif_rx(skb);	/* send it up */
 		dev->stats.rx_packets++;
 		dev->stats.rx_bytes += length;
diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c
index 33c4ffe6e103..c2f6e71e1181 100644
--- a/drivers/net/ixgb/ixgb_main.c
+++ b/drivers/net/ixgb/ixgb_main.c
@@ -1905,7 +1905,7 @@ ixgb_rx_checksum(struct ixgb_adapter *adapter,
 	 */
 	if ((rx_desc->status & IXGB_RX_DESC_STATUS_IXSM) ||
 	   (!(rx_desc->status & IXGB_RX_DESC_STATUS_TCPCS))) {
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 		return;
 	}
 
@@ -1913,7 +1913,7 @@ ixgb_rx_checksum(struct ixgb_adapter *adapter,
 	/* now look at the TCP checksum error bit */
 	if (rx_desc->errors & IXGB_RX_DESC_ERRORS_TCPE) {
 		/* let the stack verify checksum errors */
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 		adapter->hw_csum_rx_error++;
 	} else {
 		/* TCP checksum is good */
diff --git a/drivers/net/ixgbe/ixgbe_fcoe.c b/drivers/net/ixgbe/ixgbe_fcoe.c
index 86fa07cb061d..2f1de8b90f9e 100644
--- a/drivers/net/ixgbe/ixgbe_fcoe.c
+++ b/drivers/net/ixgbe/ixgbe_fcoe.c
@@ -304,12 +304,13 @@ int ixgbe_fcoe_ddp(struct ixgbe_adapter *adapter,
 	if (!ixgbe_rx_is_fcoe(rx_desc))
 		goto ddp_out;
 
-	skb->ip_summed = CHECKSUM_UNNECESSARY;
 	sterr = le32_to_cpu(rx_desc->wb.upper.status_error);
 	fcerr = (sterr & IXGBE_RXDADV_ERR_FCERR);
 	fceofe = (sterr & IXGBE_RXDADV_ERR_FCEOFE);
 	if (fcerr == IXGBE_FCERR_BADCRC)
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
+	else
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
 
 	if (eth_hdr(skb)->h_proto == htons(ETH_P_8021Q))
 		fh = (struct fc_frame_header *)(skb->data +
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 5e4dc1b0a1bd..3aafe94741ba 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -980,7 +980,7 @@ static inline void ixgbe_rx_checksum(struct ixgbe_adapter *adapter,
 {
 	u32 status_err = le32_to_cpu(rx_desc->wb.upper.status_error);
 
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	/* Rx csum disabled */
 	if (!(adapter->flags & IXGBE_FLAG_RX_CSUM_ENABLED))
diff --git a/drivers/net/ixgbevf/ixgbevf_main.c b/drivers/net/ixgbevf/ixgbevf_main.c
index 5d3c869283a5..bdbd26c60ae6 100644
--- a/drivers/net/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ixgbevf/ixgbevf_main.c
@@ -356,7 +356,7 @@ static void ixgbevf_receive_skb(struct ixgbevf_q_vector *q_vector,
 static inline void ixgbevf_rx_checksum(struct ixgbevf_adapter *adapter,
 				       u32 status_err, struct sk_buff *skb)
 {
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	/* Rx csum disabled */
 	if (!(adapter->flags & IXGBE_FLAG_RX_CSUM_ENABLED))
diff --git a/drivers/net/jme.c b/drivers/net/jme.c
index 99f24f5cac53..1fcd533b1a61 100644
--- a/drivers/net/jme.c
+++ b/drivers/net/jme.c
@@ -936,7 +936,7 @@ jme_alloc_and_feed_skb(struct jme_adapter *jme, int idx)
 		if (jme_rxsum_ok(jme, le16_to_cpu(rxdesc->descwb.flags)))
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		else
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 
 		if (rxdesc->descwb.flags & cpu_to_le16(RXWBFLAG_TAGON)) {
 			if (jme->vlgrp) {
diff --git a/drivers/net/ll_temac_main.c b/drivers/net/ll_temac_main.c
index bdf2149e5296..874ee01e8d9d 100644
--- a/drivers/net/ll_temac_main.c
+++ b/drivers/net/ll_temac_main.c
@@ -760,7 +760,7 @@ static void ll_temac_recv(struct net_device *ndev)
 		skb_put(skb, length);
 		skb->dev = ndev;
 		skb->protocol = eth_type_trans(skb, ndev);
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 
 		/* if we're doing rx csum offload, set it up */
 		if (((lp->temac_features & TEMAC_FEATURE_RX_CSUM) != 0) &&
diff --git a/drivers/net/macb.c b/drivers/net/macb.c
index ff2f158ab0b9..4297f6e8c4bc 100644
--- a/drivers/net/macb.c
+++ b/drivers/net/macb.c
@@ -407,7 +407,7 @@ static int macb_rx_frame(struct macb *bp, unsigned int first_frag,
 	}
 
 	skb_reserve(skb, RX_OFFSET);
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 	skb_put(skb, len);
 
 	for (frag = first_frag; ; frag = NEXT_RX(frag)) {
diff --git a/drivers/net/niu.c b/drivers/net/niu.c
index b4cc61f1fc59..1f89f472cbfb 100644
--- a/drivers/net/niu.c
+++ b/drivers/net/niu.c
@@ -3484,7 +3484,7 @@ static int niu_process_rx_pkt(struct napi_struct *napi, struct niu *np,
 				     RCR_ENTRY_ERROR)))
 				skb->ip_summed = CHECKSUM_UNNECESSARY;
 			else
-				skb->ip_summed = CHECKSUM_NONE;
+				skb_checksum_none_assert(skb);
 		} else if (!(val & RCR_ENTRY_MULTI))
 			append_size = len - skb->len;
 
diff --git a/drivers/net/ns83820.c b/drivers/net/ns83820.c
index 447c2c43769a..4475ca97f031 100644
--- a/drivers/net/ns83820.c
+++ b/drivers/net/ns83820.c
@@ -923,7 +923,7 @@ static void rx_irq(struct net_device *ndev)
 			if ((extsts & 0x002a0000) && !(extsts & 0x00540000)) {
 				skb->ip_summed = CHECKSUM_UNNECESSARY;
 			} else {
-				skb->ip_summed = CHECKSUM_NONE;
+				skb_checksum_none_assert(skb);
 			}
 			skb->protocol = eth_type_trans(skb, ndev);
 #ifdef NS83820_VLAN_ACCEL_SUPPORT
diff --git a/drivers/net/pasemi_mac.c b/drivers/net/pasemi_mac.c
index 8ab6ae0a6107..828e97cacdbf 100644
--- a/drivers/net/pasemi_mac.c
+++ b/drivers/net/pasemi_mac.c
@@ -808,7 +808,7 @@ static int pasemi_mac_clean_rx(struct pasemi_mac_rxring *rx,
 			skb->csum = (macrx & XCT_MACRX_CSUM_M) >>
 					   XCT_MACRX_CSUM_S;
 		} else
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 
 		packets++;
 		tot_bytes += len;
diff --git a/drivers/net/ps3_gelic_net.c b/drivers/net/ps3_gelic_net.c
index 87d6b8f36304..5526ab4895e6 100644
--- a/drivers/net/ps3_gelic_net.c
+++ b/drivers/net/ps3_gelic_net.c
@@ -956,9 +956,9 @@ static void gelic_net_pass_skb_up(struct gelic_descr *descr,
 		    (!(data_error & GELIC_DESCR_DATA_ERROR_CHK_MASK)))
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		else
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 	} else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 
 	/* update netdevice statistics */
 	netdev->stats.rx_packets++;
diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c
index 6168a130f33f..7496ed2c34ab 100644
--- a/drivers/net/qla3xxx.c
+++ b/drivers/net/qla3xxx.c
@@ -2029,7 +2029,7 @@ static void ql_process_mac_rx_intr(struct ql3_adapter *qdev,
 			 dma_unmap_len(lrg_buf_cb2, maplen),
 			 PCI_DMA_FROMDEVICE);
 	prefetch(skb->data);
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 	skb->protocol = eth_type_trans(skb, qdev->ndev);
 
 	netif_receive_skb(skb);
@@ -2076,7 +2076,7 @@ static void ql_process_macip_rx_intr(struct ql3_adapter *qdev,
 			 PCI_DMA_FROMDEVICE);
 	prefetch(skb2->data);
 
-	skb2->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb2);
 	if (qdev->device_id == QL3022_DEVICE_ID) {
 		/*
 		 * Copy the ethhdr from first buffer to second. This
diff --git a/drivers/net/qlcnic/qlcnic_init.c b/drivers/net/qlcnic/qlcnic_init.c
index 8e47d7aea562..26a7d6bca5c7 100644
--- a/drivers/net/qlcnic/qlcnic_init.c
+++ b/drivers/net/qlcnic/qlcnic_init.c
@@ -1369,7 +1369,7 @@ static struct sk_buff *qlcnic_process_rxbuf(struct qlcnic_adapter *adapter,
 		adapter->stats.csummed++;
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	} else {
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 	}
 
 	skb->dev = adapter->netdev;
diff --git a/drivers/net/qlge/qlge_main.c b/drivers/net/qlge/qlge_main.c
index 5a245211fc53..e2d0e108b9aa 100644
--- a/drivers/net/qlge/qlge_main.c
+++ b/drivers/net/qlge/qlge_main.c
@@ -1566,7 +1566,7 @@ static void ql_process_mac_rx_page(struct ql_adapter *qdev,
 	rx_ring->rx_packets++;
 	rx_ring->rx_bytes += skb->len;
 	skb->protocol = eth_type_trans(skb, ndev);
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	if (qdev->rx_csum &&
 		!(ib_mac_rsp->flags1 & IB_MAC_CSUM_ERR_MASK)) {
@@ -1676,7 +1676,7 @@ static void ql_process_mac_rx_skb(struct ql_adapter *qdev,
 	rx_ring->rx_packets++;
 	rx_ring->rx_bytes += skb->len;
 	skb->protocol = eth_type_trans(skb, ndev);
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	/* If rx checksum is on, and there are no
 	 * csum or frame errors.
@@ -1996,7 +1996,7 @@ static void ql_process_mac_split_rx_intr(struct ql_adapter *qdev,
 	}
 
 	skb->protocol = eth_type_trans(skb, ndev);
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	/* If rx checksum is on, and there are no
 	 * csum or frame errors.
diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index 078bbf4e6f19..07b3fb5175e5 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -4460,7 +4460,7 @@ static inline void rtl8169_rx_csum(struct sk_buff *skb, struct RxDesc *desc)
 	    ((status == RxProtoIP) && !(opts1 & IPFail)))
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 }
 
 static inline bool rtl8169_try_rx_copy(struct sk_buff **sk_buff,
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 7061fc8e99c7..c70ad515383a 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -7603,10 +7603,10 @@ static int rx_osm_handler(struct ring_info *ring_data, struct RxD_t * rxdp)
 			 * Packet with erroneous checksum, let the
 			 * upper layers deal with it.
 			 */
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 		}
 	} else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 
 	swstats->mem_freed += skb->truesize;
 send_up:
diff --git a/drivers/net/sb1250-mac.c b/drivers/net/sb1250-mac.c
index 8e6bd45b9f31..d8249d7653c6 100644
--- a/drivers/net/sb1250-mac.c
+++ b/drivers/net/sb1250-mac.c
@@ -1170,7 +1170,7 @@ again:
 						sb->ip_summed = CHECKSUM_UNNECESSARY;
 						/* don't need to set sb->csum */
 					} else {
-						sb->ip_summed = CHECKSUM_NONE;
+						skb_checksum_none_assert(sb);
 					}
 				}
 				prefetch(sb->data);
diff --git a/drivers/net/sfc/rx.c b/drivers/net/sfc/rx.c
index 799c461ce7b8..acb372e841b2 100644
--- a/drivers/net/sfc/rx.c
+++ b/drivers/net/sfc/rx.c
@@ -615,7 +615,7 @@ void __efx_rx_packet(struct efx_channel *channel,
 	EFX_BUG_ON_PARANOID(!skb);
 
 	/* Set the SKB flags */
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	/* Pass the packet up */
 	netif_receive_skb(skb);
diff --git a/drivers/net/sh_eth.c b/drivers/net/sh_eth.c
index a812efc3632e..50259dfec583 100644
--- a/drivers/net/sh_eth.c
+++ b/drivers/net/sh_eth.c
@@ -798,7 +798,7 @@ static int sh_eth_rx(struct net_device *ndev)
 			skb->dev = ndev;
 			sh_eth_set_receive_align(skb);
 
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 			rxdesc->addr = virt_to_phys(PTR_ALIGN(skb->data, 4));
 		}
 		if (entry >= RX_RING_SIZE - 1)
diff --git a/drivers/net/smsc911x.c b/drivers/net/smsc911x.c
index 0909ae934ad0..13ddcd487200 100644
--- a/drivers/net/smsc911x.c
+++ b/drivers/net/smsc911x.c
@@ -1048,7 +1048,7 @@ static int smsc911x_poll(struct napi_struct *napi, int budget)
 		smsc911x_rx_readfifo(pdata, (unsigned int *)skb->head,
 				     pktwords);
 		skb->protocol = eth_type_trans(skb, dev);
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 		netif_receive_skb(skb);
 
 		/* Update counters */
diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c
index 1636a34d95dd..cb6bcca9d541 100644
--- a/drivers/net/spider_net.c
+++ b/drivers/net/spider_net.c
@@ -1000,9 +1000,9 @@ spider_net_pass_skb_up(struct spider_net_descr *descr,
 		     !(data_error & SPIDER_NET_DATA_ERR_CKSUM_MASK))
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		else
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 	} else
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 
 	if (data_status & SPIDER_NET_VLAN_PACKET) {
 		/* further enhancements: HW-accel VLAN
diff --git a/drivers/net/stmmac/stmmac_main.c b/drivers/net/stmmac/stmmac_main.c
index e3f002eba89a..1ccea76d89ae 100644
--- a/drivers/net/stmmac/stmmac_main.c
+++ b/drivers/net/stmmac/stmmac_main.c
@@ -1256,7 +1256,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
 
 			if (unlikely(status == csum_none)) {
 				/* always for the old mac 10/100 */
-				skb->ip_summed = CHECKSUM_NONE;
+				skb_checksum_none_assert(skb);
 				netif_receive_skb(skb);
 			} else {
 				skb->ip_summed = CHECKSUM_UNNECESSARY;
diff --git a/drivers/net/tehuti.c b/drivers/net/tehuti.c
index 3128d6a8e9ce..8b3dc1eb4015 100644
--- a/drivers/net/tehuti.c
+++ b/drivers/net/tehuti.c
@@ -1297,12 +1297,13 @@ static int bdx_rx_receive(struct bdx_priv *priv, struct rxd_fifo *f, int budget)
 		ndev->stats.rx_bytes += len;
 
 		skb_put(skb, len);
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
 		skb->protocol = eth_type_trans(skb, ndev);
 
 		/* Non-IP packets aren't checksum-offloaded */
 		if (GET_RXD_PKT_ID(rxd_val1) == 0)
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
+		else
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
 
 		NETIF_RX_MUX(priv, rxd_val1, rxd_vlan, skb);
 
diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index bc3af78a869f..9f6ffffc8376 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -4719,7 +4719,7 @@ static int tg3_rx(struct tg3_napi *tnapi, int budget)
 		      >> RXD_TCPCSUM_SHIFT) == 0xffff))
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		else
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 
 		skb->protocol = eth_type_trans(skb, tp->dev);
 
diff --git a/drivers/net/typhoon.c b/drivers/net/typhoon.c
index 3f4681f78262..5dfb39539b3e 100644
--- a/drivers/net/typhoon.c
+++ b/drivers/net/typhoon.c
@@ -1760,7 +1760,7 @@ typhoon_rx(struct typhoon *tp, struct basic_ring *rxRing, volatile __le32 * read
 		   (TYPHOON_RX_IP_CHK_GOOD | TYPHOON_RX_UDP_CHK_GOOD)) {
 			new_skb->ip_summed = CHECKSUM_UNNECESSARY;
 		} else
-			new_skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(new_skb);
 
 		spin_lock(&tp->state_lock);
 		if(tp->vlgrp != NULL && rx->rxStatus & TYPHOON_RX_VLAN)
diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c
index fd69095ef6e3..ed7f4f5c4062 100644
--- a/drivers/net/via-velocity.c
+++ b/drivers/net/via-velocity.c
@@ -1954,7 +1954,7 @@ static int velocity_tx_srv(struct velocity_info *vptr)
  */
 static inline void velocity_rx_csum(struct rx_desc *rd, struct sk_buff *skb)
 {
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	if (rd->rdesc1.CSM & CSM_IPKT) {
 		if (rd->rdesc1.CSM & CSM_IPOK) {
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index abe0ff53daf3..198ce92af0c3 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -1042,11 +1042,11 @@ vmxnet3_rx_csum(struct vmxnet3_adapter *adapter,
 				skb->csum = htons(gdesc->rcd.csum);
 				skb->ip_summed = CHECKSUM_PARTIAL;
 			} else {
-				skb->ip_summed = CHECKSUM_NONE;
+				skb_checksum_none_assert(skb);
 			}
 		}
 	} else {
-		skb->ip_summed = CHECKSUM_NONE;
+		skb_checksum_none_assert(skb);
 	}
 }
 
diff --git a/drivers/net/vxge/vxge-main.c b/drivers/net/vxge/vxge-main.c
index 01cdec712b64..5378b849f54f 100644
--- a/drivers/net/vxge/vxge-main.c
+++ b/drivers/net/vxge/vxge-main.c
@@ -501,7 +501,7 @@ vxge_rx_1b_compl(struct __vxge_hw_ring *ringh, void *dtr,
 		    ext_info.l4_cksum == VXGE_HW_L4_CKSUM_OK)
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		else
-			skb->ip_summed = CHECKSUM_NONE;
+			skb_checksum_none_assert(skb);
 
 		vxge_rx_complete(ring, skb, ext_info.vlan,
 			pkt_length, &ext_info);
diff --git a/drivers/net/xilinx_emaclite.c b/drivers/net/xilinx_emaclite.c
index 71122ee4e830..f3f8be5a35fa 100644
--- a/drivers/net/xilinx_emaclite.c
+++ b/drivers/net/xilinx_emaclite.c
@@ -641,7 +641,7 @@ static void xemaclite_rx_handler(struct net_device *dev)
 	skb_put(skb, len);	/* Tell the skb how much data we got */
 
 	skb->protocol = eth_type_trans(skb, dev);
-	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
 
 	dev->stats.rx_packets++;
 	dev->stats.rx_bytes += len;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index f900ffcd847e..9e8085a89589 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2206,6 +2206,21 @@ static inline void skb_forward_csum(struct sk_buff *skb)
 		skb->ip_summed = CHECKSUM_NONE;
 }
 
+/**
+ * skb_checksum_none_assert - make sure skb ip_summed is CHECKSUM_NONE
+ * @skb: skb to check
+ *
+ * fresh skbs have their ip_summed set to CHECKSUM_NONE.
+ * Instead of forcing ip_summed to CHECKSUM_NONE, we can
+ * use this helper, to document places where we make this assertion.
+ */
+static inline void skb_checksum_none_assert(struct sk_buff *skb)
+{
+#ifdef DEBUG
+	BUG_ON(skb->ip_summed != CHECKSUM_NONE);
+#endif
+}
+
 bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off);
 #endif	/* __KERNEL__ */
 #endif	/* _LINUX_SKBUFF_H */
-- 
cgit v1.2.3


From fe8e0c25cad28e8858ecfa5863333c70685a6811 Mon Sep 17 00:00:00 2001
From: Alexander van Heukelum <heukelum@fastmail.fm>
Date: Mon, 6 Sep 2010 20:53:42 +0200
Subject: x86, 32-bit: Align percpu area and irq stacks to THREAD_SIZE

The irq stacks, located in the percpu-area, need to be
THREAD_SIZE aligned. Add the infrastucture to align percpu
variables to larger-than-pagesize amounts within the percpu
area, and use it to specify the alignment for the irq stacks.
Also align the percpu area itself to THREAD_SIZE.

This should make irq stacks work with 8K THREAD_SIZE.

Signed-off-by: Alexander van Heukelum <heukelum@fastmail.fm>
Cc: Tejun Heo <tj@kernel.org>
Cc: hch@lst.de
LKML-Reference: <1283799222.15941.1393621887@webmail.messagingengine.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/irq_32.c      |  4 ++--
 arch/x86/kernel/vmlinux.lds.S |  2 +-
 include/linux/percpu-defs.h   | 12 ++++++++++++
 3 files changed, 15 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index 3b5609f54c4b..50fbbe60e507 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -60,8 +60,8 @@ union irq_ctx {
 static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx);
 static DEFINE_PER_CPU(union irq_ctx *, softirq_ctx);
 
-static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, hardirq_stack);
-static DEFINE_PER_CPU_PAGE_ALIGNED(union irq_ctx, softirq_stack);
+static DEFINE_PER_CPU_MULTIPAGE_ALIGNED(union irq_ctx, hardirq_stack, THREAD_SIZE);
+static DEFINE_PER_CPU_MULTIPAGE_ALIGNED(union irq_ctx, softirq_stack, THREAD_SIZE);
 
 static void call_on_stack(void *func, void *stack)
 {
diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S
index d0bb52296fa3..bb899475355d 100644
--- a/arch/x86/kernel/vmlinux.lds.S
+++ b/arch/x86/kernel/vmlinux.lds.S
@@ -273,7 +273,7 @@ SECTIONS
 	}
 
 #if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP)
-	PERCPU(PAGE_SIZE)
+	PERCPU(THREAD_SIZE)
 #endif
 
 	. = ALIGN(PAGE_SIZE);
diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h
index ce2dc655cd1d..ab20d119a85d 100644
--- a/include/linux/percpu-defs.h
+++ b/include/linux/percpu-defs.h
@@ -138,6 +138,18 @@
 	DEFINE_PER_CPU_SECTION(type, name, "..page_aligned")		\
 	__aligned(PAGE_SIZE)
 
+/*
+ * Declaration/definition used for large per-CPU variables that must be
+ * aligned to something larger than the pagesize.
+ */
+#define DECLARE_PER_CPU_MULTIPAGE_ALIGNED(type, name, size)		\
+	DECLARE_PER_CPU_SECTION(type, name, "..page_aligned")		\
+	__aligned(size)
+
+#define DEFINE_PER_CPU_MULTIPAGE_ALIGNED(type, name, size)		\
+	DEFINE_PER_CPU_SECTION(type, name, "..page_aligned")		\
+	__aligned(size)
+
 /*
  * Intermodule exports for per-CPU variables.  sparse forgets about
  * address space across EXPORT_SYMBOL(), change EXPORT_SYMBOL() to
-- 
cgit v1.2.3


From 2bf2160d8805de64308e2e7c3cd97813cb58ed2f Mon Sep 17 00:00:00 2001
From: Lai Jiangshan <laijs@cn.fujitsu.com>
Date: Mon, 23 Aug 2010 18:42:48 +0900
Subject: irq: Add tracepoint to softirq_raise

Add a tracepoint for tracing when softirq action is raised.

This and the existing tracepoints complete softirq's tracepoints:
softirq_raise, softirq_entry and softirq_exit.

And when this tracepoint is used in combination with
the softirq_entry tracepoint we can determine
the softirq raise latency.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Cc: David Miller <davem@davemloft.net>
Cc: Kaneshige Kenji <kaneshige.kenji@jp.fujitsu.com>
Cc: Izumo Taku <izumi.taku@jp.fujitsu.com>
Cc: Kosaki Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Scott Mcmillan <scott.a.mcmillan@intel.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
LKML-Reference: <4C724298.4050509@jp.fujitsu.com>
[ factorize softirq events with DECLARE_EVENT_CLASS ]
Signed-off-by: Koki Sanagi <sanagi.koki@jp.fujitsu.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 include/linux/interrupt.h  |  8 +++++++-
 include/trace/events/irq.h | 26 ++++++++++++++++++++++++--
 2 files changed, 31 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index a0384a4d1e6f..531495db1708 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -18,6 +18,7 @@
 #include <asm/atomic.h>
 #include <asm/ptrace.h>
 #include <asm/system.h>
+#include <trace/events/irq.h>
 
 /*
  * These correspond to the IORESOURCE_IRQ_* defines in
@@ -407,7 +408,12 @@ asmlinkage void do_softirq(void);
 asmlinkage void __do_softirq(void);
 extern void open_softirq(int nr, void (*action)(struct softirq_action *));
 extern void softirq_init(void);
-#define __raise_softirq_irqoff(nr) do { or_softirq_pending(1UL << (nr)); } while (0)
+static inline void __raise_softirq_irqoff(unsigned int nr)
+{
+	trace_softirq_raise((struct softirq_action *)(unsigned long)nr, NULL);
+	or_softirq_pending(1UL << nr);
+}
+
 extern void raise_softirq_irqoff(unsigned int nr);
 extern void raise_softirq(unsigned int nr);
 extern void wakeup_softirqd(void);
diff --git a/include/trace/events/irq.h b/include/trace/events/irq.h
index 0e4cfb694fe7..6fa7cbab7d93 100644
--- a/include/trace/events/irq.h
+++ b/include/trace/events/irq.h
@@ -5,7 +5,9 @@
 #define _TRACE_IRQ_H
 
 #include <linux/tracepoint.h>
-#include <linux/interrupt.h>
+
+struct irqaction;
+struct softirq_action;
 
 #define softirq_name(sirq) { sirq##_SOFTIRQ, #sirq }
 #define show_softirq_name(val)				\
@@ -93,7 +95,10 @@ DECLARE_EVENT_CLASS(softirq,
 	),
 
 	TP_fast_assign(
-		__entry->vec = (int)(h - vec);
+		if (vec)
+			__entry->vec = (int)(h - vec);
+		else
+			__entry->vec = (int)(long)h;
 	),
 
 	TP_printk("vec=%d [action=%s]", __entry->vec,
@@ -136,6 +141,23 @@ DEFINE_EVENT(softirq, softirq_exit,
 	TP_ARGS(h, vec)
 );
 
+/**
+ * softirq_raise - called immediately when a softirq is raised
+ * @h: pointer to struct softirq_action
+ * @vec: pointer to first struct softirq_action in softirq_vec array
+ *
+ * The @h parameter contains a pointer to the softirq vector number which is
+ * raised. @vec is NULL and it means @h includes vector number not
+ * softirq_action. When used in combination with the softirq_entry tracepoint
+ * we can determine the softirq raise latency.
+ */
+DEFINE_EVENT(softirq, softirq_raise,
+
+	TP_PROTO(struct softirq_action *h, struct softirq_action *vec),
+
+	TP_ARGS(h, vec)
+);
+
 #endif /*  _TRACE_IRQ_H */
 
 /* This part must be outside protection */
-- 
cgit v1.2.3


From 269cddd44e3588d1c50a7ec055b78de4d6c72cb6 Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Tue, 7 Sep 2010 14:17:10 -0500
Subject: dlm: Fix dlm lock status block comment in dlm.h

There is only one place in the dlm where the sb_status is set
and that is queue_cast(). Tracing back the callers of that
function shows that the listed set of return values is
out of date, so here are an updated set.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
---
 include/linux/dlm.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dlm.h b/include/linux/dlm.h
index 0b3518c42356..d4e02f5353a0 100644
--- a/include/linux/dlm.h
+++ b/include/linux/dlm.h
@@ -48,10 +48,10 @@ typedef void dlm_lockspace_t;
  *
  * 0 if lock request was successful
  * -EAGAIN if request would block and is flagged DLM_LKF_NOQUEUE
- * -ENOMEM if there is no memory to process request
- * -EINVAL if there are invalid parameters
  * -DLM_EUNLOCK if unlock request was successful
  * -DLM_ECANCEL if a cancel completed successfully
+ * -EDEADLK if a deadlock was detected
+ * -ETIMEDOUT if the lock request was canceled due to a timeout
  */
 
 #define DLM_SBF_DEMOTED		0x01
-- 
cgit v1.2.3


From 4f8b02b4e5c6896e073bed736136d420bd44b627 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 18:22:47 +0200
Subject: vmalloc: pcpu_get/free_vm_areas() aren't needed on UP

These functions are used only by percpu memory allocator on SMP.
Don't build them on UP.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Reviewed-by: Chrsitoph Lameter <cl@linux.com>
---
 include/linux/vmalloc.h | 2 ++
 mm/vmalloc.c            | 2 ++
 2 files changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index 01c2145118dc..63a4fe6d51bd 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -117,10 +117,12 @@ extern rwlock_t vmlist_lock;
 extern struct vm_struct *vmlist;
 extern __init void vm_area_register_early(struct vm_struct *vm, size_t align);
 
+#ifdef CONFIG_SMP
 struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
 				     const size_t *sizes, int nr_vms,
 				     size_t align, gfp_t gfp_mask);
 
 void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms);
+#endif
 
 #endif /* _LINUX_VMALLOC_H */
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 6b8889da69a6..c623e0ce3f00 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -2056,6 +2056,7 @@ void free_vm_area(struct vm_struct *area)
 }
 EXPORT_SYMBOL_GPL(free_vm_area);
 
+#ifdef CONFIG_SMP
 static struct vmap_area *node_to_va(struct rb_node *n)
 {
 	return n ? rb_entry(n, struct vmap_area, rb_node) : NULL;
@@ -2336,6 +2337,7 @@ void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms)
 		free_vm_area(vms[i]);
 	kfree(vms);
 }
+#endif	/* CONFIG_SMP */
 
 #ifdef CONFIG_PROC_FS
 static void *s_start(struct seq_file *m, loff_t *pos)
-- 
cgit v1.2.3


From 6abad5acac09921f4944af77d3860f82d49f528d Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 18:22:47 +0200
Subject: percpu: reduce PCPU_MIN_UNIT_SIZE to 32k

In preparation of enabling percpu allocator for UP, reduce
PCPU_MIN_UNIT_SIZE to 32k.  On UP, the first chunk doesn't have to
include static percpu variables and chunk size can be smaller which is
important as UP percpu allocator will use contiguous kernel memory to
populate chunks.

PCPU_MIN_UNIT_SIZE also determines the maximum supported allocation
size but 32k should still be enough.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Christoph Lameter <cl@linux.com>
---
 include/linux/percpu.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 49466b13c5c6..fc8130a7cac0 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -42,7 +42,7 @@
 #ifdef CONFIG_SMP
 
 /* minimum unit size, also is the maximum supported allocation size */
-#define PCPU_MIN_UNIT_SIZE		PFN_ALIGN(64 << 10)
+#define PCPU_MIN_UNIT_SIZE		PFN_ALIGN(32 << 10)
 
 /*
  * Percpu allocator can serve percpu allocations before slab is
-- 
cgit v1.2.3


From bbddff0545878a8649c091a9dd7c43ce91516734 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 18:22:48 +0200
Subject: percpu: use percpu allocator on UP too

On UP, percpu allocations were redirected to kmalloc.  This has the
following problems.

* For certain amount of allocations (determined by
  PERCPU_DYNAMIC_EARLY_SLOTS and PERCPU_DYNAMIC_EARLY_SIZE), percpu
  allocator can be used before the usual kernel memory allocator is
  brought online.  On SMP, this is used to initialize the kernel
  memory allocator.

* percpu allocator honors alignment upto PAGE_SIZE but kmalloc()
  doesn't.  For example, workqueue makes use of larger alignments for
  cpu_workqueues.

Currently, users of percpu allocators need to handle UP differently,
which is somewhat fragile and ugly.  Other than small amount of
memory, there isn't much to lose by enabling percpu allocator on UP.
It can simply use kernel memory based chunk allocation which was added
for SMP archs w/o MMUs.

This patch removes mm/percpu_up.c, builds mm/percpu.c on UP too and
makes UP build use percpu-km.  As percpu addresses and kernel
addresses are always identity mapped and static percpu variables don't
need any special treatment, nothing is arch dependent and mm/percpu.c
implements generic setup_per_cpu_areas() for UP.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Christoph Lameter <cl@linux-foundation.org>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
---
 include/linux/percpu.h | 29 +++++-------------------
 mm/Kconfig             |  8 +++++++
 mm/Makefile            |  7 +-----
 mm/percpu-km.c         |  2 +-
 mm/percpu.c            | 60 ++++++++++++++++++++++++++++++++++++++++++++++----
 mm/percpu_up.c         | 30 -------------------------
 6 files changed, 71 insertions(+), 65 deletions(-)
 delete mode 100644 mm/percpu_up.c

(limited to 'include/linux')

diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index fc8130a7cac0..aeeeef1093cd 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -39,8 +39,6 @@
 	preempt_enable();				\
 } while (0)
 
-#ifdef CONFIG_SMP
-
 /* minimum unit size, also is the maximum supported allocation size */
 #define PCPU_MIN_UNIT_SIZE		PFN_ALIGN(32 << 10)
 
@@ -137,37 +135,20 @@ extern int __init pcpu_page_first_chunk(size_t reserved_size,
  * dynamically allocated. Non-atomic access to the current CPU's
  * version should probably be combined with get_cpu()/put_cpu().
  */
+#ifdef CONFIG_SMP
 #define per_cpu_ptr(ptr, cpu)	SHIFT_PERCPU_PTR((ptr), per_cpu_offset((cpu)))
+#else
+#define per_cpu_ptr(ptr, cpu)	({ (void)(cpu); VERIFY_PERCPU_PTR((ptr)); })
+#endif
 
 extern void __percpu *__alloc_reserved_percpu(size_t size, size_t align);
 extern bool is_kernel_percpu_address(unsigned long addr);
 
-#ifndef CONFIG_HAVE_SETUP_PER_CPU_AREA
+#if !defined(CONFIG_SMP) || !defined(CONFIG_HAVE_SETUP_PER_CPU_AREA)
 extern void __init setup_per_cpu_areas(void);
 #endif
 extern void __init percpu_init_late(void);
 
-#else /* CONFIG_SMP */
-
-#define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR((ptr)); })
-
-/* can't distinguish from other static vars, always false */
-static inline bool is_kernel_percpu_address(unsigned long addr)
-{
-	return false;
-}
-
-static inline void __init setup_per_cpu_areas(void) { }
-
-static inline void __init percpu_init_late(void) { }
-
-static inline void *pcpu_lpage_remapped(void *kaddr)
-{
-	return NULL;
-}
-
-#endif /* CONFIG_SMP */
-
 extern void __percpu *__alloc_percpu(size_t size, size_t align);
 extern void free_percpu(void __percpu *__pdata);
 extern phys_addr_t per_cpu_ptr_to_phys(void *addr);
diff --git a/mm/Kconfig b/mm/Kconfig
index f4e516e9c37c..01a57447a410 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -301,3 +301,11 @@ config NOMMU_INITIAL_TRIM_EXCESS
 	  of 1 says that all excess pages should be trimmed.
 
 	  See Documentation/nommu-mmap.txt for more information.
+
+#
+# UP and nommu archs use km based percpu allocator
+#
+config NEED_PER_CPU_KM
+	depends on !SMP
+	bool
+	default y
diff --git a/mm/Makefile b/mm/Makefile
index 34b2546a9e37..f73f75a29f82 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -11,7 +11,7 @@ obj-y			:= bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \
 			   maccess.o page_alloc.o page-writeback.o \
 			   readahead.o swap.o truncate.o vmscan.o shmem.o \
 			   prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \
-			   page_isolation.o mm_init.o mmu_context.o \
+			   page_isolation.o mm_init.o mmu_context.o percpu.o \
 			   $(mmu-y)
 obj-y += init-mm.o
 
@@ -36,11 +36,6 @@ obj-$(CONFIG_FAILSLAB) += failslab.o
 obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o
 obj-$(CONFIG_FS_XIP) += filemap_xip.o
 obj-$(CONFIG_MIGRATION) += migrate.o
-ifdef CONFIG_SMP
-obj-y += percpu.o
-else
-obj-y += percpu_up.o
-endif
 obj-$(CONFIG_QUICKLIST) += quicklist.o
 obj-$(CONFIG_CGROUP_MEM_RES_CTLR) += memcontrol.o page_cgroup.o
 obj-$(CONFIG_MEMORY_FAILURE) += memory-failure.o
diff --git a/mm/percpu-km.c b/mm/percpu-km.c
index df680855540a..7037bc73bfa4 100644
--- a/mm/percpu-km.c
+++ b/mm/percpu-km.c
@@ -27,7 +27,7 @@
  *   chunk size is not aligned.  percpu-km code will whine about it.
  */
 
-#ifdef CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK
+#if defined(CONFIG_SMP) && defined(CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK)
 #error "contiguous percpu allocation is incompatible with paged first chunk"
 #endif
 
diff --git a/mm/percpu.c b/mm/percpu.c
index 58c572b18b07..fa70122dfdd0 100644
--- a/mm/percpu.c
+++ b/mm/percpu.c
@@ -76,6 +76,7 @@
 #define PCPU_SLOT_BASE_SHIFT		5	/* 1-31 shares the same slot */
 #define PCPU_DFL_MAP_ALLOC		16	/* start a map with 16 ents */
 
+#ifdef CONFIG_SMP
 /* default addr <-> pcpu_ptr mapping, override in asm/percpu.h if necessary */
 #ifndef __addr_to_pcpu_ptr
 #define __addr_to_pcpu_ptr(addr)					\
@@ -89,6 +90,11 @@
 			 (unsigned long)pcpu_base_addr -		\
 			 (unsigned long)__per_cpu_start)
 #endif
+#else	/* CONFIG_SMP */
+/* on UP, it's always identity mapped */
+#define __addr_to_pcpu_ptr(addr)	(void __percpu *)(addr)
+#define __pcpu_ptr_to_addr(ptr)		(void __force *)(ptr)
+#endif	/* CONFIG_SMP */
 
 struct pcpu_chunk {
 	struct list_head	list;		/* linked to pcpu_slot lists */
@@ -949,6 +955,7 @@ EXPORT_SYMBOL_GPL(free_percpu);
  */
 bool is_kernel_percpu_address(unsigned long addr)
 {
+#ifdef CONFIG_SMP
 	const size_t static_size = __per_cpu_end - __per_cpu_start;
 	void __percpu *base = __addr_to_pcpu_ptr(pcpu_base_addr);
 	unsigned int cpu;
@@ -959,6 +966,8 @@ bool is_kernel_percpu_address(unsigned long addr)
 		if ((void *)addr >= start && (void *)addr < start + static_size)
 			return true;
         }
+#endif
+	/* on UP, can't distinguish from other static vars, always false */
 	return false;
 }
 
@@ -1066,6 +1075,8 @@ void __init pcpu_free_alloc_info(struct pcpu_alloc_info *ai)
 	free_bootmem(__pa(ai), ai->__ai_size);
 }
 
+#if defined(CONFIG_SMP) && (defined(CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK) || \
+			    defined(CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK))
 /**
  * pcpu_build_alloc_info - build alloc_info considering distances between CPUs
  * @reserved_size: the size of reserved percpu area in bytes
@@ -1220,6 +1231,8 @@ static struct pcpu_alloc_info * __init pcpu_build_alloc_info(
 
 	return ai;
 }
+#endif	/* CONFIG_SMP && (CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK ||
+			  CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK) */
 
 /**
  * pcpu_dump_alloc_info - print out information about pcpu_alloc_info
@@ -1363,7 +1376,9 @@ int __init pcpu_setup_first_chunk(const struct pcpu_alloc_info *ai,
 
 	/* sanity checks */
 	PCPU_SETUP_BUG_ON(ai->nr_groups <= 0);
+#ifdef CONFIG_SMP
 	PCPU_SETUP_BUG_ON(!ai->static_size);
+#endif
 	PCPU_SETUP_BUG_ON(!base_addr);
 	PCPU_SETUP_BUG_ON(ai->unit_size < size_sum);
 	PCPU_SETUP_BUG_ON(ai->unit_size & ~PAGE_MASK);
@@ -1488,6 +1503,8 @@ int __init pcpu_setup_first_chunk(const struct pcpu_alloc_info *ai,
 	return 0;
 }
 
+#ifdef CONFIG_SMP
+
 const char *pcpu_fc_names[PCPU_FC_NR] __initdata = {
 	[PCPU_FC_AUTO]	= "auto",
 	[PCPU_FC_EMBED]	= "embed",
@@ -1758,8 +1775,9 @@ out_free_ar:
 }
 #endif /* CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK */
 
+#ifndef	CONFIG_HAVE_SETUP_PER_CPU_AREA
 /*
- * Generic percpu area setup.
+ * Generic SMP percpu area setup.
  *
  * The embedding helper is used because its behavior closely resembles
  * the original non-dynamic generic percpu area setup.  This is
@@ -1770,7 +1788,6 @@ out_free_ar:
  * on the physical linear memory mapping which uses large page
  * mappings on applicable archs.
  */
-#ifndef CONFIG_HAVE_SETUP_PER_CPU_AREA
 unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 EXPORT_SYMBOL(__per_cpu_offset);
 
@@ -1799,13 +1816,48 @@ void __init setup_per_cpu_areas(void)
 				    PERCPU_DYNAMIC_RESERVE, PAGE_SIZE, NULL,
 				    pcpu_dfl_fc_alloc, pcpu_dfl_fc_free);
 	if (rc < 0)
-		panic("Failed to initialized percpu areas.");
+		panic("Failed to initialize percpu areas.");
 
 	delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start;
 	for_each_possible_cpu(cpu)
 		__per_cpu_offset[cpu] = delta + pcpu_unit_offsets[cpu];
 }
-#endif /* CONFIG_HAVE_SETUP_PER_CPU_AREA */
+#endif	/* CONFIG_HAVE_SETUP_PER_CPU_AREA */
+
+#else	/* CONFIG_SMP */
+
+/*
+ * UP percpu area setup.
+ *
+ * UP always uses km-based percpu allocator with identity mapping.
+ * Static percpu variables are indistinguishable from the usual static
+ * variables and don't require any special preparation.
+ */
+void __init setup_per_cpu_areas(void)
+{
+	const size_t unit_size =
+		roundup_pow_of_two(max_t(size_t, PCPU_MIN_UNIT_SIZE,
+					 PERCPU_DYNAMIC_RESERVE));
+	struct pcpu_alloc_info *ai;
+	void *fc;
+
+	ai = pcpu_alloc_alloc_info(1, 1);
+	fc = __alloc_bootmem(unit_size, PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
+	if (!ai || !fc)
+		panic("Failed to allocate memory for percpu areas.");
+
+	ai->dyn_size = unit_size;
+	ai->unit_size = unit_size;
+	ai->atom_size = unit_size;
+	ai->alloc_size = unit_size;
+	ai->groups[0].nr_units = 1;
+	ai->groups[0].cpu_map[0] = 0;
+
+	if (pcpu_setup_first_chunk(ai, fc) < 0)
+		panic("Failed to initialize percpu areas.");
+}
+
+#endif	/* CONFIG_SMP */
 
 /*
  * First and reserved chunks are initialized with temporary allocation
diff --git a/mm/percpu_up.c b/mm/percpu_up.c
deleted file mode 100644
index db884fae5721..000000000000
--- a/mm/percpu_up.c
+++ /dev/null
@@ -1,30 +0,0 @@
-/*
- * mm/percpu_up.c - dummy percpu memory allocator implementation for UP
- */
-
-#include <linux/module.h>
-#include <linux/percpu.h>
-#include <linux/slab.h>
-
-void __percpu *__alloc_percpu(size_t size, size_t align)
-{
-	/*
-	 * Can't easily make larger alignment work with kmalloc.  WARN
-	 * on it.  Larger alignment should only be used for module
-	 * percpu sections on SMP for which this path isn't used.
-	 */
-	WARN_ON_ONCE(align > SMP_CACHE_BYTES);
-	return (void __percpu __force *)kzalloc(size, GFP_KERNEL);
-}
-EXPORT_SYMBOL_GPL(__alloc_percpu);
-
-void free_percpu(void __percpu *p)
-{
-	kfree(this_cpu_ptr(p));
-}
-EXPORT_SYMBOL_GPL(free_percpu);
-
-phys_addr_t per_cpu_ptr_to_phys(void *addr)
-{
-	return __pa(addr);
-}
-- 
cgit v1.2.3


From a6e0fc8514d41dfdd98b1d15cacc432cf040f8af Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Wed, 8 Sep 2010 14:15:32 -0700
Subject: net: introduce rcu_dereference_rtnl

We use rcu_dereference_check(p, rcu_read_lock_held() ||
lockdep_rtnl_is_held()) several times in network stack.

More usages to come too, so its time to create a helper.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 58d44491880f..263690d991a8 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -749,6 +749,17 @@ extern int rtnl_is_locked(void);
 extern int lockdep_rtnl_is_held(void);
 #endif /* #ifdef CONFIG_PROVE_LOCKING */
 
+/**
+ * rcu_dereference_rtnl - rcu_dereference with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ *
+ * Do an rcu_dereference(p), but check caller either holds rcu_read_lock()
+ * or RTNL
+ */
+#define rcu_dereference_rtnl(p)					\
+	rcu_dereference_check(p, rcu_read_lock_held() ||	\
+				 lockdep_rtnl_is_held())
+
 extern void rtnetlink_init(void);
 extern void __rtnl_unlock(void);
 
-- 
cgit v1.2.3


From 15133f6e67d8d646d0744336b4daa3135452cb0d Mon Sep 17 00:00:00 2001
From: Andy Grover <andy.grover@oracle.com>
Date: Tue, 12 Jan 2010 14:33:38 -0800
Subject: RDS: Implement atomic operations

Implement a CMSG-based interface to do FADD and CSWP ops.

Alter send routines to handle atomic ops.

Add atomic counters to stats.

Add xmit_atomic() to struct rds_transport

Inline rds_ib_send_unmap_rdma into unmap_rm

Signed-off-by: Andy Grover <andy.grover@oracle.com>
---
 include/linux/rds.h |  19 +++++++
 net/rds/ib.c        |   1 +
 net/rds/ib.h        |   1 +
 net/rds/ib_rdma.c   |   4 +-
 net/rds/ib_send.c   | 140 ++++++++++++++++++++++++++++++++++++++++++++++------
 net/rds/rdma.c      |  73 +++++++++++++++++++++++++++
 net/rds/rds.h       |  33 +++++++++++--
 net/rds/send.c      |  71 ++++++++++++++++++++++++--
 net/rds/stats.c     |   2 +
 9 files changed, 321 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rds.h b/include/linux/rds.h
index 7f3971d9fc5c..9239152abf7a 100644
--- a/include/linux/rds.h
+++ b/include/linux/rds.h
@@ -73,6 +73,8 @@
 #define RDS_CMSG_RDMA_MAP		3
 #define RDS_CMSG_RDMA_STATUS		4
 #define RDS_CMSG_CONG_UPDATE		5
+#define RDS_CMSG_ATOMIC_FADD		6
+#define RDS_CMSG_ATOMIC_CSWP		7
 
 #define RDS_INFO_FIRST			10000
 #define RDS_INFO_COUNTERS		10000
@@ -237,6 +239,23 @@ struct rds_rdma_args {
 	u_int64_t	user_token;
 };
 
+struct rds_atomic_args {
+	rds_rdma_cookie_t cookie;
+	uint64_t 	local_addr;
+	uint64_t 	remote_addr;
+	union {
+		struct {
+			uint64_t	compare;
+			uint64_t	swap;
+		} cswp;
+		struct {
+			uint64_t	add;
+		} fadd;
+	};
+	uint64_t	flags;
+	uint64_t	user_token;
+};
+
 struct rds_rdma_notify {
 	u_int64_t	user_token;
 	int32_t		status;
diff --git a/net/rds/ib.c b/net/rds/ib.c
index 8f2d6dd7700a..f0d29656baff 100644
--- a/net/rds/ib.c
+++ b/net/rds/ib.c
@@ -264,6 +264,7 @@ struct rds_transport rds_ib_transport = {
 	.xmit			= rds_ib_xmit,
 	.xmit_cong_map		= NULL,
 	.xmit_rdma		= rds_ib_xmit_rdma,
+	.xmit_atomic		= rds_ib_xmit_atomic,
 	.recv			= rds_ib_recv,
 	.conn_alloc		= rds_ib_conn_alloc,
 	.conn_free		= rds_ib_conn_free,
diff --git a/net/rds/ib.h b/net/rds/ib.h
index 64df4e79b29f..d2fd0aa4fde7 100644
--- a/net/rds/ib.h
+++ b/net/rds/ib.h
@@ -336,6 +336,7 @@ void rds_ib_send_add_credits(struct rds_connection *conn, unsigned int credits);
 void rds_ib_advertise_credits(struct rds_connection *conn, unsigned int posted);
 int rds_ib_send_grab_credits(struct rds_ib_connection *ic, u32 wanted,
 			     u32 *adv_credits, int need_posted, int max_posted);
+int rds_ib_xmit_atomic(struct rds_connection *conn, struct rm_atomic_op *op);
 
 /* ib_stats.c */
 DECLARE_PER_CPU(struct rds_ib_statistics, rds_ib_stats);
diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index 0f3b5a2f3fe0..242231f09464 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -298,7 +298,9 @@ static struct rds_ib_mr *rds_ib_alloc_fmr(struct rds_ib_device *rds_ibdev)
 	ibmr->fmr = ib_alloc_fmr(rds_ibdev->pd,
 			(IB_ACCESS_LOCAL_WRITE |
 			 IB_ACCESS_REMOTE_READ |
-			 IB_ACCESS_REMOTE_WRITE),
+			 IB_ACCESS_REMOTE_WRITE|
+			 IB_ACCESS_REMOTE_ATOMIC),
+
 			&pool->fmr_attr);
 	if (IS_ERR(ibmr->fmr)) {
 		err = PTR_ERR(ibmr->fmr);
diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c
index f0edfdb2866c..b2bd164434ad 100644
--- a/net/rds/ib_send.c
+++ b/net/rds/ib_send.c
@@ -62,15 +62,17 @@ static void rds_ib_send_rdma_complete(struct rds_message *rm,
 	rds_rdma_send_complete(rm, notify_status);
 }
 
-static void rds_ib_send_unmap_rdma(struct rds_ib_connection *ic,
-				   struct rds_rdma_op *op)
+static void rds_ib_send_atomic_complete(struct rds_message *rm,
+				      int wc_status)
 {
-	if (op->r_mapped) {
-		ib_dma_unmap_sg(ic->i_cm_id->device,
-			op->r_sg, op->r_nents,
-			op->r_write ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
-		op->r_mapped = 0;
-	}
+	int notify_status;
+
+	if (wc_status != IB_WC_SUCCESS)
+		notify_status = RDS_RDMA_OTHER_ERROR;
+	else
+		notify_status = RDS_RDMA_SUCCESS;
+
+	rds_atomic_send_complete(rm, notify_status);
 }
 
 static void rds_ib_send_unmap_rm(struct rds_ib_connection *ic,
@@ -86,7 +88,14 @@ static void rds_ib_send_unmap_rm(struct rds_ib_connection *ic,
 			DMA_TO_DEVICE);
 
 	if (rm->rdma.m_rdma_op.r_active) {
-		rds_ib_send_unmap_rdma(ic, &rm->rdma.m_rdma_op);
+		struct rds_rdma_op *op = &rm->rdma.m_rdma_op;
+
+		if (op->r_mapped) {
+			ib_dma_unmap_sg(ic->i_cm_id->device,
+					op->r_sg, op->r_nents,
+					op->r_write ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
+			op->r_mapped = 0;
+		}
 
 		/* If the user asked for a completion notification on this
 		 * message, we can implement three different semantics:
@@ -116,6 +125,24 @@ static void rds_ib_send_unmap_rm(struct rds_ib_connection *ic,
 			rds_stats_add(s_recv_rdma_bytes, rm->rdma.m_rdma_op.r_bytes);
 	}
 
+	if (rm->atomic.op_active) {
+		struct rm_atomic_op *op = &rm->atomic;
+
+		/* unmap atomic recvbuf */
+		if (op->op_mapped) {
+			ib_dma_unmap_sg(ic->i_cm_id->device, op->op_sg, 1,
+					DMA_FROM_DEVICE);
+			op->op_mapped = 0;
+		}
+
+		rds_ib_send_atomic_complete(rm, wc_status);
+
+		if (rm->atomic.op_type == RDS_ATOMIC_TYPE_CSWP)
+			rds_stats_inc(s_atomic_cswp);
+		else
+			rds_stats_inc(s_atomic_fadd);
+	}
+
 	/* If anyone waited for this message to get flushed out, wake
 	 * them up now */
 	rds_message_unmapped(rm);
@@ -158,12 +185,9 @@ void rds_ib_send_clear_ring(struct rds_ib_connection *ic)
 	u32 i;
 
 	for (i = 0, send = ic->i_sends; i < ic->i_send_ring.w_nr; i++, send++) {
-		if (send->s_wr.opcode == 0xdead)
+		if (!send->s_rm || send->s_wr.opcode == 0xdead)
 			continue;
-		if (send->s_rm)
-			rds_ib_send_unmap_rm(ic, send, IB_WC_WR_FLUSH_ERR);
-		if (send->s_op)
-			rds_ib_send_unmap_rdma(ic, send->s_op);
+		rds_ib_send_unmap_rm(ic, send, IB_WC_WR_FLUSH_ERR);
 	}
 }
 
@@ -218,6 +242,8 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context)
 				break;
 			case IB_WR_RDMA_WRITE:
 			case IB_WR_RDMA_READ:
+			case IB_WR_ATOMIC_FETCH_AND_ADD:
+			case IB_WR_ATOMIC_CMP_AND_SWP:
 				/* Nothing to be done - the SG list will be unmapped
 				 * when the SEND completes. */
 				break;
@@ -243,8 +269,7 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context)
 
 				rm = rds_send_get_message(conn, send->s_op);
 				if (rm) {
-					if (rm->rdma.m_rdma_op.r_active)
-						rds_ib_send_unmap_rdma(ic, &rm->rdma.m_rdma_op);
+					rds_ib_send_unmap_rm(ic, send, wc.status);
 					rds_ib_send_rdma_complete(rm, wc.status);
 					rds_message_put(rm);
 				}
@@ -736,6 +761,89 @@ out:
 	return ret;
 }
 
+/*
+ * Issue atomic operation.
+ * A simplified version of the rdma case, we always map 1 SG, and
+ * only 8 bytes, for the return value from the atomic operation.
+ */
+int rds_ib_xmit_atomic(struct rds_connection *conn, struct rm_atomic_op *op)
+{
+	struct rds_ib_connection *ic = conn->c_transport_data;
+	struct rds_ib_send_work *send = NULL;
+	struct ib_send_wr *failed_wr;
+	struct rds_ib_device *rds_ibdev;
+	u32 pos;
+	u32 work_alloc;
+	int ret;
+
+	rds_ibdev = ib_get_client_data(ic->i_cm_id->device, &rds_ib_client);
+
+	work_alloc = rds_ib_ring_alloc(&ic->i_send_ring, 1, &pos);
+	if (work_alloc != 1) {
+		rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc);
+		rds_ib_stats_inc(s_ib_tx_ring_full);
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	/* address of send request in ring */
+	send = &ic->i_sends[pos];
+	send->s_queued = jiffies;
+
+	if (op->op_type == RDS_ATOMIC_TYPE_CSWP) {
+		send->s_wr.opcode = IB_WR_ATOMIC_CMP_AND_SWP;
+		send->s_wr.wr.atomic.compare_add = op->op_compare;
+		send->s_wr.wr.atomic.swap = op->op_swap_add;
+	} else { /* FADD */
+		send->s_wr.opcode = IB_WR_ATOMIC_FETCH_AND_ADD;
+		send->s_wr.wr.atomic.compare_add = op->op_swap_add;
+		send->s_wr.wr.atomic.swap = 0;
+	}
+	send->s_wr.send_flags = IB_SEND_SIGNALED;
+	send->s_wr.num_sge = 1;
+	send->s_wr.next = NULL;
+	send->s_wr.wr.atomic.remote_addr = op->op_remote_addr;
+	send->s_wr.wr.atomic.rkey = op->op_rkey;
+
+	/* map 8 byte retval buffer to the device */
+	ret = ib_dma_map_sg(ic->i_cm_id->device, op->op_sg, 1, DMA_FROM_DEVICE);
+	rdsdebug("ic %p mapping atomic op %p. mapped %d pg\n", ic, op, ret);
+	if (ret != 1) {
+		rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc);
+		rds_ib_stats_inc(s_ib_tx_sg_mapping_failure);
+		ret = -ENOMEM; /* XXX ? */
+		goto out;
+	}
+
+	/* Convert our struct scatterlist to struct ib_sge */
+	send->s_sge[0].addr = ib_sg_dma_address(ic->i_cm_id->device, op->op_sg);
+	send->s_sge[0].length = ib_sg_dma_len(ic->i_cm_id->device, op->op_sg);
+	send->s_sge[0].lkey = ic->i_mr->lkey;
+
+	rdsdebug("rva %Lx rpa %Lx len %u\n", op->op_remote_addr,
+		 send->s_sge[0].addr, send->s_sge[0].length);
+
+	failed_wr = &send->s_wr;
+	ret = ib_post_send(ic->i_cm_id->qp, &send->s_wr, &failed_wr);
+	rdsdebug("ic %p send %p (wr %p) ret %d wr %p\n", ic,
+		 send, &send->s_wr, ret, failed_wr);
+	BUG_ON(failed_wr != &send->s_wr);
+	if (ret) {
+		printk(KERN_WARNING "RDS/IB: atomic ib_post_send to %pI4 "
+		       "returned %d\n", &conn->c_faddr, ret);
+		rds_ib_ring_unalloc(&ic->i_send_ring, work_alloc);
+		goto out;
+	}
+
+	if (unlikely(failed_wr != &send->s_wr)) {
+		printk(KERN_WARNING "RDS/IB: atomic ib_post_send() rc=%d, but failed_wqe updated!\n", ret);
+		BUG_ON(failed_wr != &send->s_wr);
+	}
+
+out:
+	return ret;
+}
+
 int rds_ib_xmit_rdma(struct rds_connection *conn, struct rds_rdma_op *op)
 {
 	struct rds_ib_connection *ic = conn->c_transport_data;
diff --git a/net/rds/rdma.c b/net/rds/rdma.c
index 4fda33045598..a7019df38c70 100644
--- a/net/rds/rdma.c
+++ b/net/rds/rdma.c
@@ -719,3 +719,76 @@ int rds_cmsg_rdma_map(struct rds_sock *rs, struct rds_message *rm,
 
 	return __rds_rdma_map(rs, CMSG_DATA(cmsg), &rm->m_rdma_cookie, &rm->rdma.m_rdma_mr);
 }
+
+/*
+ * Fill in rds_message for an atomic request.
+ */
+int rds_cmsg_atomic(struct rds_sock *rs, struct rds_message *rm,
+		    struct cmsghdr *cmsg)
+{
+	struct page *page = NULL;
+	struct rds_atomic_args *args;
+	int ret = 0;
+
+	if (cmsg->cmsg_len < CMSG_LEN(sizeof(struct rds_atomic_args))
+	 || rm->atomic.op_active)
+		return -EINVAL;
+
+	args = CMSG_DATA(cmsg);
+
+	if (cmsg->cmsg_type == RDS_CMSG_ATOMIC_CSWP) {
+		rm->atomic.op_type = RDS_ATOMIC_TYPE_CSWP;
+		rm->atomic.op_swap_add = args->cswp.swap;
+		rm->atomic.op_compare = args->cswp.compare;
+	} else {
+		rm->atomic.op_type = RDS_ATOMIC_TYPE_FADD;
+		rm->atomic.op_swap_add = args->fadd.add;
+	}
+
+	rm->m_rdma_cookie = args->cookie;
+	rm->atomic.op_notify = !!(args->flags & RDS_RDMA_NOTIFY_ME);
+	rm->atomic.op_recverr = rs->rs_recverr;
+	rm->atomic.op_sg = rds_message_alloc_sgs(rm, 1);
+
+	/* verify 8 byte-aligned */
+	if (args->local_addr & 0x7) {
+		ret = -EFAULT;
+		goto err;
+	}
+
+	ret = rds_pin_pages(args->local_addr, 1, &page, 1);
+	if (ret != 1)
+		goto err;
+	ret = 0;
+
+	sg_set_page(rm->atomic.op_sg, page, 8, offset_in_page(args->local_addr));
+
+	if (rm->atomic.op_notify || rm->atomic.op_recverr) {
+		/* We allocate an uninitialized notifier here, because
+		 * we don't want to do that in the completion handler. We
+		 * would have to use GFP_ATOMIC there, and don't want to deal
+		 * with failed allocations.
+		 */
+		rm->atomic.op_notifier = kmalloc(sizeof(*rm->atomic.op_notifier), GFP_KERNEL);
+		if (!rm->atomic.op_notifier) {
+			ret = -ENOMEM;
+			goto err;
+		}
+
+		rm->atomic.op_notifier->n_user_token = args->user_token;
+		rm->atomic.op_notifier->n_status = RDS_RDMA_SUCCESS;
+	}
+
+	rm->atomic.op_rkey = rds_rdma_cookie_key(rm->m_rdma_cookie);
+	rm->atomic.op_remote_addr = args->remote_addr + rds_rdma_cookie_offset(args->cookie);
+
+	rm->atomic.op_active = 1;
+
+	return ret;
+err:
+	if (page)
+		put_page(page);
+	kfree(rm->atomic.op_notifier);
+
+	return ret;
+}
diff --git a/net/rds/rds.h b/net/rds/rds.h
index 0bb4957e0cfc..830e2bbb3332 100644
--- a/net/rds/rds.h
+++ b/net/rds/rds.h
@@ -97,6 +97,7 @@ struct rds_connection {
 	unsigned int		c_xmit_hdr_off;
 	unsigned int		c_xmit_data_off;
 	unsigned int		c_xmit_rdma_sent;
+	unsigned int		c_xmit_atomic_sent;
 
 	spinlock_t		c_lock;		/* protect msg queues */
 	u64			c_next_tx_seq;
@@ -260,6 +261,10 @@ static inline u32 rds_rdma_cookie_offset(rds_rdma_cookie_t cookie)
 	return cookie >> 32;
 }
 
+/* atomic operation types */
+#define RDS_ATOMIC_TYPE_CSWP		0
+#define RDS_ATOMIC_TYPE_FADD		1
+
 /*
  * m_sock_item and m_conn_item are on lists that are serialized under
  * conn->c_lock.  m_sock_item has additional meaning in that once it is empty
@@ -315,11 +320,27 @@ struct rds_message {
 	struct rds_sock		*m_rs;
 	rds_rdma_cookie_t	m_rdma_cookie;
 	struct {
-		struct {
+		struct rm_atomic_op {
+			int			op_type;
+			uint64_t		op_swap_add;
+			uint64_t		op_compare;
+
+			u32			op_rkey;
+			u64			op_remote_addr;
+			unsigned int		op_notify:1;
+			unsigned int		op_recverr:1;
+			unsigned int		op_mapped:1;
+			unsigned int		op_active:1;
+			struct rds_notifier	*op_notifier;
+			struct scatterlist	*op_sg;
+
+			struct rds_mr		*op_rdma_mr;
+		} atomic;
+		struct rm_rdma_op {
 			struct rds_rdma_op	m_rdma_op;
 			struct rds_mr		*m_rdma_mr;
 		} rdma;
-		struct {
+		struct rm_data_op {
 			unsigned int		m_nents;
 			unsigned int		m_count;
 			struct scatterlist	*m_sg;
@@ -397,6 +418,7 @@ struct rds_transport {
 	int (*xmit_cong_map)(struct rds_connection *conn,
 			     struct rds_cong_map *map, unsigned long offset);
 	int (*xmit_rdma)(struct rds_connection *conn, struct rds_rdma_op *op);
+	int (*xmit_atomic)(struct rds_connection *conn, struct rm_atomic_op *op);
 	int (*recv)(struct rds_connection *conn);
 	int (*inc_copy_to_user)(struct rds_incoming *inc, struct iovec *iov,
 				size_t size);
@@ -546,6 +568,8 @@ struct rds_statistics {
 	uint64_t	s_cong_update_received;
 	uint64_t	s_cong_send_error;
 	uint64_t	s_cong_send_blocked;
+	uint64_t	s_atomic_cswp;
+	uint64_t	s_atomic_fadd;
 };
 
 /* af_rds.c */
@@ -722,7 +746,10 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm,
 int rds_cmsg_rdma_map(struct rds_sock *rs, struct rds_message *rm,
 			  struct cmsghdr *cmsg);
 void rds_rdma_free_op(struct rds_rdma_op *ro);
-void rds_rdma_send_complete(struct rds_message *rm, int);
+void rds_rdma_send_complete(struct rds_message *rm, int wc_status);
+void rds_atomic_send_complete(struct rds_message *rm, int wc_status);
+int rds_cmsg_atomic(struct rds_sock *rs, struct rds_message *rm,
+		    struct cmsghdr *cmsg);
 
 extern void __rds_put_mr_final(struct rds_mr *mr);
 static inline void rds_mr_put(struct rds_mr *mr)
diff --git a/net/rds/send.c b/net/rds/send.c
index b751a8e77c41..f3f4e79274bf 100644
--- a/net/rds/send.c
+++ b/net/rds/send.c
@@ -73,6 +73,7 @@ void rds_send_reset(struct rds_connection *conn)
 	conn->c_xmit_hdr_off = 0;
 	conn->c_xmit_data_off = 0;
 	conn->c_xmit_rdma_sent = 0;
+	conn->c_xmit_atomic_sent = 0;
 
 	conn->c_map_queued = 0;
 
@@ -171,6 +172,7 @@ int rds_send_xmit(struct rds_connection *conn)
 			conn->c_xmit_hdr_off = 0;
 			conn->c_xmit_data_off = 0;
 			conn->c_xmit_rdma_sent = 0;
+			conn->c_xmit_atomic_sent = 0;
 
 			/* Release the reference to the previous message. */
 			rds_message_put(rm);
@@ -262,6 +264,17 @@ int rds_send_xmit(struct rds_connection *conn)
 			conn->c_xmit_rm = rm;
 		}
 
+
+		if (rm->atomic.op_active && !conn->c_xmit_atomic_sent) {
+			ret = conn->c_trans->xmit_atomic(conn, &rm->atomic);
+			if (ret)
+				break;
+			conn->c_xmit_atomic_sent = 1;
+			/* The transport owns the mapped memory for now.
+			 * You can't unmap it while it's on the send queue */
+			set_bit(RDS_MSG_MAPPED, &rm->m_flags);
+		}
+
 		/*
 		 * Try and send an rdma message.  Let's see if we can
 		 * keep this simple and require that the transport either
@@ -442,6 +455,41 @@ void rds_rdma_send_complete(struct rds_message *rm, int status)
 }
 EXPORT_SYMBOL_GPL(rds_rdma_send_complete);
 
+/*
+ * Just like above, except looks at atomic op
+ */
+void rds_atomic_send_complete(struct rds_message *rm, int status)
+{
+	struct rds_sock *rs = NULL;
+	struct rm_atomic_op *ao;
+	struct rds_notifier *notifier;
+
+	spin_lock(&rm->m_rs_lock);
+
+	ao = &rm->atomic;
+	if (test_bit(RDS_MSG_ON_SOCK, &rm->m_flags)
+	    && ao->op_active && ao->op_notify && ao->op_notifier) {
+		notifier = ao->op_notifier;
+		rs = rm->m_rs;
+		sock_hold(rds_rs_to_sk(rs));
+
+		notifier->n_status = status;
+		spin_lock(&rs->rs_lock);
+		list_add_tail(&notifier->n_list, &rs->rs_notify_queue);
+		spin_unlock(&rs->rs_lock);
+
+		ao->op_notifier = NULL;
+	}
+
+	spin_unlock(&rm->m_rs_lock);
+
+	if (rs) {
+		rds_wake_sk_sleep(rs);
+		sock_put(rds_rs_to_sk(rs));
+	}
+}
+EXPORT_SYMBOL_GPL(rds_atomic_send_complete);
+
 /*
  * This is the same as rds_rdma_send_complete except we
  * don't do any locking - we have all the ingredients (message,
@@ -788,6 +836,11 @@ static int rds_rm_size(struct msghdr *msg, int data_len)
 			/* these are valid but do no add any size */
 			break;
 
+		case RDS_CMSG_ATOMIC_CSWP:
+		case RDS_CMSG_ATOMIC_FADD:
+			size += sizeof(struct scatterlist);
+			break;
+
 		default:
 			return -EINVAL;
 		}
@@ -813,7 +866,7 @@ static int rds_cmsg_send(struct rds_sock *rs, struct rds_message *rm,
 			continue;
 
 		/* As a side effect, RDMA_DEST and RDMA_MAP will set
-		 * rm->m_rdma_cookie and rm->m_rdma_mr.
+		 * rm->rdma.m_rdma_cookie and rm->rdma.m_rdma_mr.
 		 */
 		switch (cmsg->cmsg_type) {
 		case RDS_CMSG_RDMA_ARGS:
@@ -829,6 +882,10 @@ static int rds_cmsg_send(struct rds_sock *rs, struct rds_message *rm,
 			if (!ret)
 				*allocated_mr = 1;
 			break;
+		case RDS_CMSG_ATOMIC_CSWP:
+		case RDS_CMSG_ATOMIC_FADD:
+			ret = rds_cmsg_atomic(rs, rm, cmsg);
+			break;
 
 		default:
 			return -EINVAL;
@@ -926,10 +983,18 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
 		goto out;
 
 	if ((rm->m_rdma_cookie || rm->rdma.m_rdma_op.r_active) &&
-	    !conn->c_trans->xmit_rdma) {
+	       !conn->c_trans->xmit_rdma) {
 		if (printk_ratelimit())
 			printk(KERN_NOTICE "rdma_op %p conn xmit_rdma %p\n",
-				&rm->rdma.m_rdma_op, conn->c_trans->xmit_rdma);
+			       &rm->rdma.m_rdma_op, conn->c_trans->xmit_rdma);
+		ret = -EOPNOTSUPP;
+		goto out;
+	}
+
+	if (rm->atomic.op_active && !conn->c_trans->xmit_atomic) {
+		if (printk_ratelimit())
+			printk(KERN_NOTICE "atomic_op %p conn xmit_atomic %p\n",
+			       &rm->atomic, conn->c_trans->xmit_atomic);
 		ret = -EOPNOTSUPP;
 		goto out;
 	}
diff --git a/net/rds/stats.c b/net/rds/stats.c
index 7598eb07cfb1..c66d95d9c262 100644
--- a/net/rds/stats.c
+++ b/net/rds/stats.c
@@ -75,6 +75,8 @@ static const char *const rds_stat_names[] = {
 	"cong_update_received",
 	"cong_send_error",
 	"cong_send_blocked",
+	"s_atomic_cswp",
+	"s_atomic_fadd",
 };
 
 void rds_stats_info_copy(struct rds_info_iterator *iter,
-- 
cgit v1.2.3


From 2c3a5f9abb1dc5efdab8ba9a568b1661c65fd1e3 Mon Sep 17 00:00:00 2001
From: Andy Grover <andy.grover@oracle.com>
Date: Mon, 1 Mar 2010 16:10:40 -0800
Subject: RDS: Add flag for silent ops. Do atomic op before RDMA

Add a flag to the API so users can indicate they want
silent operations. This is needed because silent ops
cannot be used with USE_ONCE MRs, so we can't just
assume silent.

Also, change send_xmit to do atomic op before rdma op if
both are present, and centralize the hairy logic to determine if
we want to attempt silent, or not.

Signed-off-by: Andy Grover <andy.grover@oracle.com>
---
 include/linux/rds.h |  1 +
 net/rds/rdma.c      |  2 ++
 net/rds/rds.h       |  2 ++
 net/rds/send.c      | 55 ++++++++++++++++++++++++++++++-----------------------
 4 files changed, 36 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rds.h b/include/linux/rds.h
index 9239152abf7a..109f1d343318 100644
--- a/include/linux/rds.h
+++ b/include/linux/rds.h
@@ -276,5 +276,6 @@ struct rds_rdma_notify {
 #define RDS_RDMA_USE_ONCE	0x0008	/* free MR after use */
 #define RDS_RDMA_DONTWAIT	0x0010	/* Don't wait in SET_BARRIER */
 #define RDS_RDMA_NOTIFY_ME	0x0020	/* Notify when operation completes */
+#define RDS_RDMA_SILENT		0x0040	/* Do not interrupt remote */
 
 #endif /* IB_RDS_H */
diff --git a/net/rds/rdma.c b/net/rds/rdma.c
index 5ba514684431..48781fe4431c 100644
--- a/net/rds/rdma.c
+++ b/net/rds/rdma.c
@@ -559,6 +559,7 @@ int rds_cmsg_rdma_args(struct rds_sock *rs, struct rds_message *rm,
 	op->op_write = !!(args->flags & RDS_RDMA_READWRITE);
 	op->op_fence = !!(args->flags & RDS_RDMA_FENCE);
 	op->op_notify = !!(args->flags & RDS_RDMA_NOTIFY_ME);
+	op->op_silent = !!(args->flags & RDS_RDMA_SILENT);
 	op->op_active = 1;
 	op->op_recverr = rs->rs_recverr;
 	WARN_ON(!nr_pages);
@@ -747,6 +748,7 @@ int rds_cmsg_atomic(struct rds_sock *rs, struct rds_message *rm,
 	}
 
 	rm->atomic.op_notify = !!(args->flags & RDS_RDMA_NOTIFY_ME);
+	rm->atomic.op_silent = !!(args->flags & RDS_RDMA_SILENT);
 	rm->atomic.op_active = 1;
 	rm->atomic.op_recverr = rs->rs_recverr;
 	rm->atomic.op_sg = rds_message_alloc_sgs(rm, 1);
diff --git a/net/rds/rds.h b/net/rds/rds.h
index 46d190d08549..23b921000e74 100644
--- a/net/rds/rds.h
+++ b/net/rds/rds.h
@@ -319,6 +319,7 @@ struct rds_message {
 			unsigned int		op_notify:1;
 			unsigned int		op_recverr:1;
 			unsigned int		op_mapped:1;
+			unsigned int		op_silent:1;
 			unsigned int		op_active:1;
 			struct scatterlist	*op_sg;
 			struct rds_notifier	*op_notifier;
@@ -333,6 +334,7 @@ struct rds_message {
 			unsigned int		op_notify:1;
 			unsigned int		op_recverr:1;
 			unsigned int		op_mapped:1;
+			unsigned int		op_silent:1;
 			unsigned int		op_active:1;
 			unsigned int		op_bytes;
 			unsigned int		op_nents;
diff --git a/net/rds/send.c b/net/rds/send.c
index cdca9747fcbc..38567f3ee7e8 100644
--- a/net/rds/send.c
+++ b/net/rds/send.c
@@ -250,42 +250,50 @@ int rds_send_xmit(struct rds_connection *conn)
 			conn->c_xmit_rm = rm;
 		}
 
-		if (rm->atomic.op_active && !conn->c_xmit_atomic_sent) {
-			ret = conn->c_trans->xmit_atomic(conn, rm);
+		/* The transport either sends the whole rdma or none of it */
+		if (rm->rdma.op_active && !conn->c_xmit_rdma_sent) {
+			ret = conn->c_trans->xmit_rdma(conn, &rm->rdma);
 			if (ret)
 				break;
-			conn->c_xmit_atomic_sent = 1;
+			conn->c_xmit_rdma_sent = 1;
+
 			/* The transport owns the mapped memory for now.
 			 * You can't unmap it while it's on the send queue */
 			set_bit(RDS_MSG_MAPPED, &rm->m_flags);
-
-			/*
-			 * This is evil, muahaha.
-			 * We permit 0-byte sends. (rds-ping depends on this.)
-			 * BUT if there is an atomic op and no sent data,
-			 * we turn off sending the header, to achieve
-			 * "silent" atomics.
-			 * But see below; RDMA op might toggle this back on!
-			 */
-			if (rm->data.op_nents == 0)
-				rm->data.op_active = 0;
 		}
 
-		/* The transport either sends the whole rdma or none of it */
-		if (rm->rdma.op_active && !conn->c_xmit_rdma_sent) {
-			ret = conn->c_trans->xmit_rdma(conn, &rm->rdma);
+		if (rm->atomic.op_active && !conn->c_xmit_atomic_sent) {
+			ret = conn->c_trans->xmit_atomic(conn, rm);
 			if (ret)
 				break;
-			conn->c_xmit_rdma_sent = 1;
-
-			/* rdmas need data sent, even if just the header */
-			rm->data.op_active = 1;
-
+			conn->c_xmit_atomic_sent = 1;
 			/* The transport owns the mapped memory for now.
 			 * You can't unmap it while it's on the send queue */
 			set_bit(RDS_MSG_MAPPED, &rm->m_flags);
 		}
 
+		/*
+		 * A number of cases require an RDS header to be sent
+		 * even if there is no data.
+		 * We permit 0-byte sends; rds-ping depends on this.
+		 * However, if there are exclusively attached silent ops,
+		 * we skip the hdr/data send, to enable silent operation.
+		 */
+		if (rm->data.op_nents == 0) {
+			int ops_present;
+			int all_ops_are_silent = 1;
+
+			ops_present = (rm->atomic.op_active || rm->rdma.op_active);
+			if (rm->atomic.op_active && !rm->atomic.op_silent)
+				all_ops_are_silent = 0;
+			if (rm->rdma.op_active && !rm->rdma.op_silent)
+				all_ops_are_silent = 0;
+
+			if (ops_present && all_ops_are_silent
+			    && !rm->m_rdma_cookie)
+				rm->data.op_active = 0;
+		}
+
 		if (rm->data.op_active && !conn->c_xmit_data_sent) {
 			ret = conn->c_trans->xmit(conn, rm,
 						  conn->c_xmit_hdr_off,
@@ -1009,8 +1017,7 @@ int rds_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
 	if (ret)
 		goto out;
 
-	if ((rm->m_rdma_cookie || rm->rdma.op_active) &&
-	    !conn->c_trans->xmit_rdma) {
+	if (rm->rdma.op_active && !conn->c_trans->xmit_rdma) {
 		if (printk_ratelimit())
 			printk(KERN_NOTICE "rdma_op %p conn xmit_rdma %p\n",
 			       &rm->rdma, conn->c_trans->xmit_rdma);
-- 
cgit v1.2.3


From 20c72bd5f5f902e5a8745d51573699605bf8d21c Mon Sep 17 00:00:00 2001
From: Andy Grover <andy.grover@oracle.com>
Date: Wed, 25 Aug 2010 05:51:28 -0700
Subject: RDS: Implement masked atomic operations

Add two CMSGs for masked versions of cswp and fadd. args
struct modified to use a union for different atomic op type's
arguments. Change IB to do masked atomic ops. Atomic op type
in rds_message similarly unionized.

Signed-off-by: Andy Grover <andy.grover@oracle.com>
---
 include/linux/rds.h | 12 ++++++++++++
 net/rds/ib_send.c   | 14 +++++++++-----
 net/rds/rdma.c      | 33 +++++++++++++++++++++++++++------
 net/rds/rds.h       | 14 ++++++++++++--
 net/rds/send.c      |  4 ++++
 5 files changed, 64 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rds.h b/include/linux/rds.h
index 109f1d343318..a2a5edb4a276 100644
--- a/include/linux/rds.h
+++ b/include/linux/rds.h
@@ -75,6 +75,8 @@
 #define RDS_CMSG_CONG_UPDATE		5
 #define RDS_CMSG_ATOMIC_FADD		6
 #define RDS_CMSG_ATOMIC_CSWP		7
+#define RDS_CMSG_MASKED_ATOMIC_FADD	8
+#define RDS_CMSG_MASKED_ATOMIC_CSWP	9
 
 #define RDS_INFO_FIRST			10000
 #define RDS_INFO_COUNTERS		10000
@@ -251,6 +253,16 @@ struct rds_atomic_args {
 		struct {
 			uint64_t	add;
 		} fadd;
+		struct {
+			uint64_t	compare;
+			uint64_t	swap;
+			uint64_t	compare_mask;
+			uint64_t	swap_mask;
+		} m_cswp;
+		struct {
+			uint64_t	add;
+			uint64_t	nocarry_mask;
+		} m_fadd;
 	};
 	uint64_t	flags;
 	uint64_t	user_token;
diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c
index 808544aebb70..71f373c421bc 100644
--- a/net/rds/ib_send.c
+++ b/net/rds/ib_send.c
@@ -807,13 +807,17 @@ int rds_ib_xmit_atomic(struct rds_connection *conn, struct rm_atomic_op *op)
 	send->s_queued = jiffies;
 
 	if (op->op_type == RDS_ATOMIC_TYPE_CSWP) {
-		send->s_wr.opcode = IB_WR_ATOMIC_CMP_AND_SWP;
-		send->s_wr.wr.atomic.compare_add = op->op_compare;
-		send->s_wr.wr.atomic.swap = op->op_swap_add;
+		send->s_wr.opcode = IB_WR_MASKED_ATOMIC_CMP_AND_SWP;
+		send->s_wr.wr.atomic.compare_add = op->op_m_cswp.compare;
+		send->s_wr.wr.atomic.swap = op->op_m_cswp.swap;
+		send->s_wr.wr.atomic.compare_add_mask = op->op_m_cswp.compare_mask;
+		send->s_wr.wr.atomic.swap_mask = op->op_m_cswp.swap_mask;
 	} else { /* FADD */
-		send->s_wr.opcode = IB_WR_ATOMIC_FETCH_AND_ADD;
-		send->s_wr.wr.atomic.compare_add = op->op_swap_add;
+		send->s_wr.opcode = IB_WR_MASKED_ATOMIC_FETCH_AND_ADD;
+		send->s_wr.wr.atomic.compare_add = op->op_m_fadd.add;
 		send->s_wr.wr.atomic.swap = 0;
+		send->s_wr.wr.atomic.compare_add_mask = op->op_m_fadd.nocarry_mask;
+		send->s_wr.wr.atomic.swap_mask = 0;
 	}
 	nr_sig = rds_ib_set_wr_signal_state(ic, send, op->op_notify);
 	send->s_wr.num_sge = 1;
diff --git a/net/rds/rdma.c b/net/rds/rdma.c
index 48781fe4431c..48064673fc76 100644
--- a/net/rds/rdma.c
+++ b/net/rds/rdma.c
@@ -738,13 +738,34 @@ int rds_cmsg_atomic(struct rds_sock *rs, struct rds_message *rm,
 
 	args = CMSG_DATA(cmsg);
 
-	if (cmsg->cmsg_type == RDS_CMSG_ATOMIC_CSWP) {
-		rm->atomic.op_type = RDS_ATOMIC_TYPE_CSWP;
-		rm->atomic.op_swap_add = args->cswp.swap;
-		rm->atomic.op_compare = args->cswp.compare;
-	} else {
+	/* Nonmasked & masked cmsg ops converted to masked hw ops */
+	switch (cmsg->cmsg_type) {
+	case RDS_CMSG_ATOMIC_FADD:
+		rm->atomic.op_type = RDS_ATOMIC_TYPE_FADD;
+		rm->atomic.op_m_fadd.add = args->fadd.add;
+		rm->atomic.op_m_fadd.nocarry_mask = 0;
+		break;
+	case RDS_CMSG_MASKED_ATOMIC_FADD:
 		rm->atomic.op_type = RDS_ATOMIC_TYPE_FADD;
-		rm->atomic.op_swap_add = args->fadd.add;
+		rm->atomic.op_m_fadd.add = args->m_fadd.add;
+		rm->atomic.op_m_fadd.nocarry_mask = args->m_fadd.nocarry_mask;
+		break;
+	case RDS_CMSG_ATOMIC_CSWP:
+		rm->atomic.op_type = RDS_ATOMIC_TYPE_CSWP;
+		rm->atomic.op_m_cswp.compare = args->cswp.compare;
+		rm->atomic.op_m_cswp.swap = args->cswp.swap;
+		rm->atomic.op_m_cswp.compare_mask = ~0;
+		rm->atomic.op_m_cswp.swap_mask = ~0;
+		break;
+	case RDS_CMSG_MASKED_ATOMIC_CSWP:
+		rm->atomic.op_type = RDS_ATOMIC_TYPE_CSWP;
+		rm->atomic.op_m_cswp.compare = args->m_cswp.compare;
+		rm->atomic.op_m_cswp.swap = args->m_cswp.swap;
+		rm->atomic.op_m_cswp.compare_mask = args->m_cswp.compare_mask;
+		rm->atomic.op_m_cswp.swap_mask = args->m_cswp.swap_mask;
+		break;
+	default:
+		BUG(); /* should never happen */
 	}
 
 	rm->atomic.op_notify = !!(args->flags & RDS_RDMA_NOTIFY_ME);
diff --git a/net/rds/rds.h b/net/rds/rds.h
index aadaddba88a7..8103dcf8b976 100644
--- a/net/rds/rds.h
+++ b/net/rds/rds.h
@@ -316,8 +316,18 @@ struct rds_message {
 	struct {
 		struct rm_atomic_op {
 			int			op_type;
-			uint64_t		op_swap_add;
-			uint64_t		op_compare;
+			union {
+				struct {
+					uint64_t	compare;
+					uint64_t	swap;
+					uint64_t	compare_mask;
+					uint64_t	swap_mask;
+				} op_m_cswp;
+				struct {
+					uint64_t	add;
+					uint64_t	nocarry_mask;
+				} op_m_fadd;
+			};
 
 			u32			op_rkey;
 			u64			op_remote_addr;
diff --git a/net/rds/send.c b/net/rds/send.c
index 81471b25373b..9b951a0ab6b7 100644
--- a/net/rds/send.c
+++ b/net/rds/send.c
@@ -843,6 +843,8 @@ static int rds_rm_size(struct msghdr *msg, int data_len)
 
 		case RDS_CMSG_ATOMIC_CSWP:
 		case RDS_CMSG_ATOMIC_FADD:
+		case RDS_CMSG_MASKED_ATOMIC_CSWP:
+		case RDS_CMSG_MASKED_ATOMIC_FADD:
 			cmsg_groups |= 1;
 			size += sizeof(struct scatterlist);
 			break;
@@ -894,6 +896,8 @@ static int rds_cmsg_send(struct rds_sock *rs, struct rds_message *rm,
 			break;
 		case RDS_CMSG_ATOMIC_CSWP:
 		case RDS_CMSG_ATOMIC_FADD:
+		case RDS_CMSG_MASKED_ATOMIC_CSWP:
+		case RDS_CMSG_MASKED_ATOMIC_FADD:
 			ret = rds_cmsg_atomic(rs, rm, cmsg);
 			break;
 
-- 
cgit v1.2.3


From fd128dfa50cfc4f2959dc4aa5d7468d33b988332 Mon Sep 17 00:00:00 2001
From: Andy Grover <andy.grover@oracle.com>
Date: Wed, 25 Aug 2010 09:32:17 -0700
Subject: RDS: Add rds.h to exported headers list

Also, a number of changes were made based on the assumption that
rds.h wasn't exported, so roll these back.

Signed-off-by: Andy Grover <andy.grover@oracle.com>
---
 include/linux/Kbuild |  1 +
 include/linux/rds.h  | 12 ++++++------
 2 files changed, 7 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 626b629429ff..c7fbf298ad68 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -302,6 +302,7 @@ header-y += quota.h
 header-y += radeonfb.h
 header-y += random.h
 header-y += raw.h
+header-y += rds.h
 header-y += reboot.h
 header-y += reiserfs_fs.h
 header-y += reiserfs_xattr.h
diff --git a/include/linux/rds.h b/include/linux/rds.h
index a2a5edb4a276..f371f885a352 100644
--- a/include/linux/rds.h
+++ b/include/linux/rds.h
@@ -95,7 +95,7 @@
 struct rds_info_counter {
 	u_int8_t	name[32];
 	u_int64_t	value;
-} __packed;
+} __attribute__((packed));
 
 #define RDS_INFO_CONNECTION_FLAG_SENDING	0x01
 #define RDS_INFO_CONNECTION_FLAG_CONNECTING	0x02
@@ -110,7 +110,7 @@ struct rds_info_connection {
 	__be32		faddr;
 	u_int8_t	transport[TRANSNAMSIZ];		/* null term ascii */
 	u_int8_t	flags;
-} __packed;
+} __attribute__((packed));
 
 struct rds_info_flow {
 	__be32		laddr;
@@ -118,7 +118,7 @@ struct rds_info_flow {
 	u_int32_t	bytes;
 	__be16		lport;
 	__be16		fport;
-} __packed;
+} __attribute__((packed));
 
 #define RDS_INFO_MESSAGE_FLAG_ACK               0x01
 #define RDS_INFO_MESSAGE_FLAG_FAST_ACK          0x02
@@ -131,7 +131,7 @@ struct rds_info_message {
 	__be16		lport;
 	__be16		fport;
 	u_int8_t	flags;
-} __packed;
+} __attribute__((packed));
 
 struct rds_info_socket {
 	u_int32_t	sndbuf;
@@ -141,7 +141,7 @@ struct rds_info_socket {
 	__be16		connected_port;
 	u_int32_t	rcvbuf;
 	u_int64_t	inum;
-} __packed;
+} __attribute__((packed));
 
 struct rds_info_tcp_socket {
 	__be32          local_addr;
@@ -153,7 +153,7 @@ struct rds_info_tcp_socket {
 	u_int32_t       last_sent_nxt;
 	u_int32_t       last_expected_una;
 	u_int32_t       last_seen_una;
-} __packed;
+} __attribute__((packed));
 
 #define RDS_IB_GID_LEN	16
 struct rds_info_rdma_connection {
-- 
cgit v1.2.3


From a46f561b774d90d8616473d56696e7d44fa1c9f1 Mon Sep 17 00:00:00 2001
From: Andy Grover <andy.grover@oracle.com>
Date: Wed, 25 Aug 2010 09:34:10 -0700
Subject: RDS: rds.h: Replace u_int[size]_t with uint[size]_t

Replace e.g. u_int32_t types with the more common uint32_t.

Reported-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Andy Grover <andy.grover@oracle.com>
---
 include/linux/rds.h | 58 ++++++++++++++++++++++++++---------------------------
 1 file changed, 29 insertions(+), 29 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rds.h b/include/linux/rds.h
index f371f885a352..3576b31b6b7b 100644
--- a/include/linux/rds.h
+++ b/include/linux/rds.h
@@ -93,8 +93,8 @@
 #define RDS_INFO_LAST			10010
 
 struct rds_info_counter {
-	u_int8_t	name[32];
-	u_int64_t	value;
+	uint8_t	name[32];
+	uint64_t	value;
 } __attribute__((packed));
 
 #define RDS_INFO_CONNECTION_FLAG_SENDING	0x01
@@ -104,18 +104,18 @@ struct rds_info_counter {
 #define TRANSNAMSIZ	16
 
 struct rds_info_connection {
-	u_int64_t	next_tx_seq;
-	u_int64_t	next_rx_seq;
+	uint64_t	next_tx_seq;
+	uint64_t	next_rx_seq;
 	__be32		laddr;
 	__be32		faddr;
-	u_int8_t	transport[TRANSNAMSIZ];		/* null term ascii */
-	u_int8_t	flags;
+	uint8_t	transport[TRANSNAMSIZ];		/* null term ascii */
+	uint8_t	flags;
 } __attribute__((packed));
 
 struct rds_info_flow {
 	__be32		laddr;
 	__be32		faddr;
-	u_int32_t	bytes;
+	uint32_t	bytes;
 	__be16		lport;
 	__be16		fport;
 } __attribute__((packed));
@@ -124,23 +124,23 @@ struct rds_info_flow {
 #define RDS_INFO_MESSAGE_FLAG_FAST_ACK          0x02
 
 struct rds_info_message {
-	u_int64_t	seq;
-	u_int32_t	len;
+	uint64_t	seq;
+	uint32_t	len;
 	__be32		laddr;
 	__be32		faddr;
 	__be16		lport;
 	__be16		fport;
-	u_int8_t	flags;
+	uint8_t	flags;
 } __attribute__((packed));
 
 struct rds_info_socket {
-	u_int32_t	sndbuf;
+	uint32_t	sndbuf;
 	__be32		bound_addr;
 	__be32		connected_addr;
 	__be16		bound_port;
 	__be16		connected_port;
-	u_int32_t	rcvbuf;
-	u_int64_t	inum;
+	uint32_t	rcvbuf;
+	uint64_t	inum;
 } __attribute__((packed));
 
 struct rds_info_tcp_socket {
@@ -148,11 +148,11 @@ struct rds_info_tcp_socket {
 	__be16          local_port;
 	__be32          peer_addr;
 	__be16          peer_port;
-	u_int64_t       hdr_rem;
-	u_int64_t       data_rem;
-	u_int32_t       last_sent_nxt;
-	u_int32_t       last_expected_una;
-	u_int32_t       last_seen_una;
+	uint64_t       hdr_rem;
+	uint64_t       data_rem;
+	uint32_t       last_sent_nxt;
+	uint32_t       last_expected_una;
+	uint32_t       last_seen_una;
 } __attribute__((packed));
 
 #define RDS_IB_GID_LEN	16
@@ -207,38 +207,38 @@ struct rds_info_rdma_connection {
  * (so that the application does not have to worry about
  * alignment).
  */
-typedef u_int64_t	rds_rdma_cookie_t;
+typedef uint64_t	rds_rdma_cookie_t;
 
 struct rds_iovec {
-	u_int64_t	addr;
-	u_int64_t	bytes;
+	uint64_t	addr;
+	uint64_t	bytes;
 };
 
 struct rds_get_mr_args {
 	struct rds_iovec vec;
-	u_int64_t	cookie_addr;
+	uint64_t	cookie_addr;
 	uint64_t	flags;
 };
 
 struct rds_get_mr_for_dest_args {
 	struct sockaddr_storage	dest_addr;
 	struct rds_iovec 	vec;
-	u_int64_t		cookie_addr;
+	uint64_t		cookie_addr;
 	uint64_t		flags;
 };
 
 struct rds_free_mr_args {
 	rds_rdma_cookie_t cookie;
-	u_int64_t	flags;
+	uint64_t	flags;
 };
 
 struct rds_rdma_args {
 	rds_rdma_cookie_t cookie;
 	struct rds_iovec remote_vec;
-	u_int64_t	local_vec_addr;
-	u_int64_t	nr_local;
-	u_int64_t	flags;
-	u_int64_t	user_token;
+	uint64_t	local_vec_addr;
+	uint64_t	nr_local;
+	uint64_t	flags;
+	uint64_t	user_token;
 };
 
 struct rds_atomic_args {
@@ -269,7 +269,7 @@ struct rds_atomic_args {
 };
 
 struct rds_rdma_notify {
-	u_int64_t	user_token;
+	uint64_t	user_token;
 	int32_t		status;
 };
 
-- 
cgit v1.2.3


From 905d64c89e2a9d71d0606904b7c3908633db6072 Mon Sep 17 00:00:00 2001
From: Andy Grover <andy.grover@oracle.com>
Date: Wed, 8 Sep 2010 18:03:54 -0700
Subject: RDS: Remove dead struct from rds.h

flows are an obsolete date type.

Signed-off-by: Andy Grover <andy.grover@oracle.com>
---
 include/linux/rds.h | 8 --------
 1 file changed, 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rds.h b/include/linux/rds.h
index 3576b31b6b7b..91950950aa59 100644
--- a/include/linux/rds.h
+++ b/include/linux/rds.h
@@ -112,14 +112,6 @@ struct rds_info_connection {
 	uint8_t	flags;
 } __attribute__((packed));
 
-struct rds_info_flow {
-	__be32		laddr;
-	__be32		faddr;
-	uint32_t	bytes;
-	__be16		lport;
-	__be16		fport;
-} __attribute__((packed));
-
 #define RDS_INFO_MESSAGE_FLAG_ACK               0x01
 #define RDS_INFO_MESSAGE_FLAG_FAST_ACK          0x02
 
-- 
cgit v1.2.3


From 01a08546af311c065f34727787dd0cc8dc0c216f Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 31 Aug 2010 10:28:16 +0200
Subject: sched: Add book scheduling domain

On top of the SMT and MC scheduling domains this adds the BOOK scheduling
domain. This is useful for NUMA like machines which do not have an interface
which tells which piece of memory is attached to which node or where the
hardware performs striping.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20100831082844.253053798@de.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h    |  1 +
 include/linux/topology.h |  6 ++++
 kernel/sched.c           | 77 ++++++++++++++++++++++++++++++++++++++++++++++--
 3 files changed, 82 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1e2a6db2d7dd..b51c53c285b8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -875,6 +875,7 @@ enum sched_domain_level {
 	SD_LV_NONE = 0,
 	SD_LV_SIBLING,
 	SD_LV_MC,
+	SD_LV_BOOK,
 	SD_LV_CPU,
 	SD_LV_NODE,
 	SD_LV_ALLNODES,
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 64e084ff5e5c..b91a40e847d2 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -201,6 +201,12 @@ int arch_update_cpu_topology(void);
 	.balance_interval	= 64,					\
 }
 
+#ifdef CONFIG_SCHED_BOOK
+#ifndef SD_BOOK_INIT
+#error Please define an appropriate SD_BOOK_INIT in include/asm/topology.h!!!
+#endif
+#endif /* CONFIG_SCHED_BOOK */
+
 #ifdef CONFIG_NUMA
 #ifndef SD_NODE_INIT
 #error Please define an appropriate SD_NODE_INIT in include/asm/topology.h!!!
diff --git a/kernel/sched.c b/kernel/sched.c
index 1a0c084b1cf9..26f83e2f1534 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6506,6 +6506,7 @@ struct s_data {
 	cpumask_var_t		nodemask;
 	cpumask_var_t		this_sibling_map;
 	cpumask_var_t		this_core_map;
+	cpumask_var_t		this_book_map;
 	cpumask_var_t		send_covered;
 	cpumask_var_t		tmpmask;
 	struct sched_group	**sched_group_nodes;
@@ -6517,6 +6518,7 @@ enum s_alloc {
 	sa_rootdomain,
 	sa_tmpmask,
 	sa_send_covered,
+	sa_this_book_map,
 	sa_this_core_map,
 	sa_this_sibling_map,
 	sa_nodemask,
@@ -6570,6 +6572,31 @@ cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
 }
 #endif /* CONFIG_SCHED_MC */
 
+/*
+ * book sched-domains:
+ */
+#ifdef CONFIG_SCHED_BOOK
+static DEFINE_PER_CPU(struct static_sched_domain, book_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_book);
+
+static int
+cpu_to_book_group(int cpu, const struct cpumask *cpu_map,
+		  struct sched_group **sg, struct cpumask *mask)
+{
+	int group = cpu;
+#ifdef CONFIG_SCHED_MC
+	cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map);
+	group = cpumask_first(mask);
+#elif defined(CONFIG_SCHED_SMT)
+	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
+	group = cpumask_first(mask);
+#endif
+	if (sg)
+		*sg = &per_cpu(sched_group_book, group).sg;
+	return group;
+}
+#endif /* CONFIG_SCHED_BOOK */
+
 static DEFINE_PER_CPU(struct static_sched_domain, phys_domains);
 static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys);
 
@@ -6578,7 +6605,10 @@ cpu_to_phys_group(int cpu, const struct cpumask *cpu_map,
 		  struct sched_group **sg, struct cpumask *mask)
 {
 	int group;
-#ifdef CONFIG_SCHED_MC
+#ifdef CONFIG_SCHED_BOOK
+	cpumask_and(mask, cpu_book_mask(cpu), cpu_map);
+	group = cpumask_first(mask);
+#elif defined(CONFIG_SCHED_MC)
 	cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map);
 	group = cpumask_first(mask);
 #elif defined(CONFIG_SCHED_SMT)
@@ -6839,6 +6869,9 @@ SD_INIT_FUNC(CPU)
 #ifdef CONFIG_SCHED_MC
  SD_INIT_FUNC(MC)
 #endif
+#ifdef CONFIG_SCHED_BOOK
+ SD_INIT_FUNC(BOOK)
+#endif
 
 static int default_relax_domain_level = -1;
 
@@ -6888,6 +6921,8 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 		free_cpumask_var(d->tmpmask); /* fall through */
 	case sa_send_covered:
 		free_cpumask_var(d->send_covered); /* fall through */
+	case sa_this_book_map:
+		free_cpumask_var(d->this_book_map); /* fall through */
 	case sa_this_core_map:
 		free_cpumask_var(d->this_core_map); /* fall through */
 	case sa_this_sibling_map:
@@ -6934,8 +6969,10 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 		return sa_nodemask;
 	if (!alloc_cpumask_var(&d->this_core_map, GFP_KERNEL))
 		return sa_this_sibling_map;
-	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
+	if (!alloc_cpumask_var(&d->this_book_map, GFP_KERNEL))
 		return sa_this_core_map;
+	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
+		return sa_this_book_map;
 	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
 		return sa_send_covered;
 	d->rd = alloc_rootdomain();
@@ -6993,6 +7030,23 @@ static struct sched_domain *__build_cpu_sched_domain(struct s_data *d,
 	return sd;
 }
 
+static struct sched_domain *__build_book_sched_domain(struct s_data *d,
+	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
+	struct sched_domain *parent, int i)
+{
+	struct sched_domain *sd = parent;
+#ifdef CONFIG_SCHED_BOOK
+	sd = &per_cpu(book_domains, i).sd;
+	SD_INIT(sd, BOOK);
+	set_domain_attribute(sd, attr);
+	cpumask_and(sched_domain_span(sd), cpu_map, cpu_book_mask(i));
+	sd->parent = parent;
+	parent->child = sd;
+	cpu_to_book_group(i, cpu_map, &sd->groups, d->tmpmask);
+#endif
+	return sd;
+}
+
 static struct sched_domain *__build_mc_sched_domain(struct s_data *d,
 	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
 	struct sched_domain *parent, int i)
@@ -7049,6 +7103,15 @@ static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 						&cpu_to_core_group,
 						d->send_covered, d->tmpmask);
 		break;
+#endif
+#ifdef CONFIG_SCHED_BOOK
+	case SD_LV_BOOK: /* set up book groups */
+		cpumask_and(d->this_book_map, cpu_map, cpu_book_mask(cpu));
+		if (cpu == cpumask_first(d->this_book_map))
+			init_sched_build_groups(d->this_book_map, cpu_map,
+						&cpu_to_book_group,
+						d->send_covered, d->tmpmask);
+		break;
 #endif
 	case SD_LV_CPU: /* set up physical groups */
 		cpumask_and(d->nodemask, cpumask_of_node(cpu), cpu_map);
@@ -7097,12 +7160,14 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 
 		sd = __build_numa_sched_domains(&d, cpu_map, attr, i);
 		sd = __build_cpu_sched_domain(&d, cpu_map, attr, sd, i);
+		sd = __build_book_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
 	}
 
 	for_each_cpu(i, cpu_map) {
 		build_sched_groups(&d, SD_LV_SIBLING, cpu_map, i);
+		build_sched_groups(&d, SD_LV_BOOK, cpu_map, i);
 		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
 	}
 
@@ -7133,6 +7198,12 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		init_sched_groups_power(i, sd);
 	}
 #endif
+#ifdef CONFIG_SCHED_BOOK
+	for_each_cpu(i, cpu_map) {
+		sd = &per_cpu(book_domains, i).sd;
+		init_sched_groups_power(i, sd);
+	}
+#endif
 
 	for_each_cpu(i, cpu_map) {
 		sd = &per_cpu(phys_domains, i).sd;
@@ -7158,6 +7229,8 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		sd = &per_cpu(cpu_domains, i).sd;
 #elif defined(CONFIG_SCHED_MC)
 		sd = &per_cpu(core_domains, i).sd;
+#elif defined(CONFIG_SCHED_BOOK)
+		sd = &per_cpu(book_domains, i).sd;
 #else
 		sd = &per_cpu(phys_domains, i).sd;
 #endif
-- 
cgit v1.2.3


From 51b0fe39549a04858001922919ab355dee9bdfcf Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 11 Jun 2010 13:35:57 +0200
Subject: perf: Deconstify struct pmu

sed -ie 's/const struct pmu\>/struct pmu/g' `git grep -l "const struct pmu\>"`

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/alpha/kernel/perf_event.c           |  4 ++--
 arch/arm/kernel/perf_event.c             |  2 +-
 arch/powerpc/kernel/perf_event.c         |  8 ++++----
 arch/powerpc/kernel/perf_event_fsl_emb.c |  2 +-
 arch/sh/kernel/perf_event.c              |  4 ++--
 arch/sparc/kernel/perf_event.c           | 10 +++++-----
 arch/x86/kernel/cpu/perf_event.c         | 14 +++++++-------
 include/linux/perf_event.h               | 10 +++++-----
 kernel/perf_event.c                      | 26 +++++++++++++-------------
 9 files changed, 40 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c
index 51c39fa41693..56fa41590381 100644
--- a/arch/alpha/kernel/perf_event.c
+++ b/arch/alpha/kernel/perf_event.c
@@ -642,7 +642,7 @@ static int __hw_perf_event_init(struct perf_event *event)
 	return 0;
 }
 
-static const struct pmu pmu = {
+static struct pmu pmu = {
 	.enable		= alpha_pmu_enable,
 	.disable	= alpha_pmu_disable,
 	.read		= alpha_pmu_read,
@@ -653,7 +653,7 @@ static const struct pmu pmu = {
 /*
  * Main entry point to initialise a HW performance event.
  */
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+struct pmu *hw_perf_event_init(struct perf_event *event)
 {
 	int err;
 
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 64ca8c3ab94b..0671e92c5111 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -491,7 +491,7 @@ __hw_perf_event_init(struct perf_event *event)
 	return err;
 }
 
-const struct pmu *
+struct pmu *
 hw_perf_event_init(struct perf_event *event)
 {
 	int err = 0;
diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c
index d301a30445e0..5f78681ad902 100644
--- a/arch/powerpc/kernel/perf_event.c
+++ b/arch/powerpc/kernel/perf_event.c
@@ -857,7 +857,7 @@ static void power_pmu_unthrottle(struct perf_event *event)
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-void power_pmu_start_txn(const struct pmu *pmu)
+void power_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
@@ -870,7 +870,7 @@ void power_pmu_start_txn(const struct pmu *pmu)
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-void power_pmu_cancel_txn(const struct pmu *pmu)
+void power_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
@@ -882,7 +882,7 @@ void power_pmu_cancel_txn(const struct pmu *pmu)
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-int power_pmu_commit_txn(const struct pmu *pmu)
+int power_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	long i, n;
@@ -1014,7 +1014,7 @@ static int hw_perf_cache_event(u64 config, u64 *eventp)
 	return 0;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+struct pmu *hw_perf_event_init(struct perf_event *event)
 {
 	u64 ev;
 	unsigned long flags;
diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c
index 1ba45471ae43..d7619b5e7a6e 100644
--- a/arch/powerpc/kernel/perf_event_fsl_emb.c
+++ b/arch/powerpc/kernel/perf_event_fsl_emb.c
@@ -428,7 +428,7 @@ static int hw_perf_cache_event(u64 config, u64 *eventp)
 	return 0;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+struct pmu *hw_perf_event_init(struct perf_event *event)
 {
 	u64 ev;
 	struct perf_event *events[MAX_HWEVENTS];
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 7a3dc3567258..395572c94c6a 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -257,13 +257,13 @@ static void sh_pmu_read(struct perf_event *event)
 	sh_perf_event_update(event, &event->hw, event->hw.idx);
 }
 
-static const struct pmu pmu = {
+static struct pmu pmu = {
 	.enable		= sh_pmu_enable,
 	.disable	= sh_pmu_disable,
 	.read		= sh_pmu_read,
 };
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+struct pmu *hw_perf_event_init(struct perf_event *event)
 {
 	int err = __hw_perf_event_init(event);
 	if (unlikely(err)) {
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 4bc402938575..481b894a5018 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -1099,7 +1099,7 @@ static int __hw_perf_event_init(struct perf_event *event)
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-static void sparc_pmu_start_txn(const struct pmu *pmu)
+static void sparc_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
@@ -1111,7 +1111,7 @@ static void sparc_pmu_start_txn(const struct pmu *pmu)
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-static void sparc_pmu_cancel_txn(const struct pmu *pmu)
+static void sparc_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
@@ -1123,7 +1123,7 @@ static void sparc_pmu_cancel_txn(const struct pmu *pmu)
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-static int sparc_pmu_commit_txn(const struct pmu *pmu)
+static int sparc_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int n;
@@ -1142,7 +1142,7 @@ static int sparc_pmu_commit_txn(const struct pmu *pmu)
 	return 0;
 }
 
-static const struct pmu pmu = {
+static struct pmu pmu = {
 	.enable		= sparc_pmu_enable,
 	.disable	= sparc_pmu_disable,
 	.read		= sparc_pmu_read,
@@ -1152,7 +1152,7 @@ static const struct pmu pmu = {
 	.commit_txn	= sparc_pmu_commit_txn,
 };
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+struct pmu *hw_perf_event_init(struct perf_event *event)
 {
 	int err = __hw_perf_event_init(event);
 
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index de6569c04cd0..fdd97f2e9961 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -618,7 +618,7 @@ static void x86_pmu_enable_all(int added)
 	}
 }
 
-static const struct pmu pmu;
+static struct pmu pmu;
 
 static inline int is_x86_event(struct perf_event *event)
 {
@@ -1427,7 +1427,7 @@ static inline void x86_pmu_read(struct perf_event *event)
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-static void x86_pmu_start_txn(const struct pmu *pmu)
+static void x86_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -1440,7 +1440,7 @@ static void x86_pmu_start_txn(const struct pmu *pmu)
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-static void x86_pmu_cancel_txn(const struct pmu *pmu)
+static void x86_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -1457,7 +1457,7 @@ static void x86_pmu_cancel_txn(const struct pmu *pmu)
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-static int x86_pmu_commit_txn(const struct pmu *pmu)
+static int x86_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int assign[X86_PMC_IDX_MAX];
@@ -1483,7 +1483,7 @@ static int x86_pmu_commit_txn(const struct pmu *pmu)
 	return 0;
 }
 
-static const struct pmu pmu = {
+static struct pmu pmu = {
 	.enable		= x86_pmu_enable,
 	.disable	= x86_pmu_disable,
 	.start		= x86_pmu_start,
@@ -1569,9 +1569,9 @@ out:
 	return ret;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+struct pmu *hw_perf_event_init(struct perf_event *event)
 {
-	const struct pmu *tmp;
+	struct pmu *tmp;
 	int err;
 
 	err = __hw_perf_event_init(event);
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 000610c4de71..09d048b52115 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -578,19 +578,19 @@ struct pmu {
 	 * Start the transaction, after this ->enable() doesn't need
 	 * to do schedulability tests.
 	 */
-	void (*start_txn)	(const struct pmu *pmu);
+	void (*start_txn)	(struct pmu *pmu);
 	/*
 	 * If ->start_txn() disabled the ->enable() schedulability test
 	 * then ->commit_txn() is required to perform one. On success
 	 * the transaction is closed. On error the transaction is kept
 	 * open until ->cancel_txn() is called.
 	 */
-	int  (*commit_txn)	(const struct pmu *pmu);
+	int  (*commit_txn)	(struct pmu *pmu);
 	/*
 	 * Will cancel the transaction, assumes ->disable() is called for
 	 * each successfull ->enable() during the transaction.
 	 */
-	void (*cancel_txn)	(const struct pmu *pmu);
+	void (*cancel_txn)	(struct pmu *pmu);
 };
 
 /**
@@ -669,7 +669,7 @@ struct perf_event {
 	int				nr_siblings;
 	int				group_flags;
 	struct perf_event		*group_leader;
-	const struct pmu		*pmu;
+	struct pmu		*pmu;
 
 	enum perf_event_active_state	state;
 	unsigned int			attach_state;
@@ -849,7 +849,7 @@ struct perf_output_handle {
  */
 extern int perf_max_events;
 
-extern const struct pmu *hw_perf_event_init(struct perf_event *event);
+extern struct pmu *hw_perf_event_init(struct perf_event *event);
 
 extern void perf_event_task_sched_in(struct task_struct *task);
 extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 2d74f31220ad..fb46fd13f31f 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -75,7 +75,7 @@ static DEFINE_SPINLOCK(perf_resource_lock);
 /*
  * Architecture provided APIs - weak aliases:
  */
-extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event)
+extern __weak struct pmu *hw_perf_event_init(struct perf_event *event)
 {
 	return NULL;
 }
@@ -691,7 +691,7 @@ group_sched_in(struct perf_event *group_event,
 	       struct perf_event_context *ctx)
 {
 	struct perf_event *event, *partial_group = NULL;
-	const struct pmu *pmu = group_event->pmu;
+	struct pmu *pmu = group_event->pmu;
 	bool txn = false;
 
 	if (group_event->state == PERF_EVENT_STATE_OFF)
@@ -4501,7 +4501,7 @@ static int perf_swevent_int(struct perf_event *event)
 	return 0;
 }
 
-static const struct pmu perf_ops_generic = {
+static struct pmu perf_ops_generic = {
 	.enable		= perf_swevent_enable,
 	.disable	= perf_swevent_disable,
 	.start		= perf_swevent_int,
@@ -4614,7 +4614,7 @@ static void cpu_clock_perf_event_read(struct perf_event *event)
 	cpu_clock_perf_event_update(event);
 }
 
-static const struct pmu perf_ops_cpu_clock = {
+static struct pmu perf_ops_cpu_clock = {
 	.enable		= cpu_clock_perf_event_enable,
 	.disable	= cpu_clock_perf_event_disable,
 	.read		= cpu_clock_perf_event_read,
@@ -4671,7 +4671,7 @@ static void task_clock_perf_event_read(struct perf_event *event)
 	task_clock_perf_event_update(event, time);
 }
 
-static const struct pmu perf_ops_task_clock = {
+static struct pmu perf_ops_task_clock = {
 	.enable		= task_clock_perf_event_enable,
 	.disable	= task_clock_perf_event_disable,
 	.read		= task_clock_perf_event_read,
@@ -4785,7 +4785,7 @@ static int swevent_hlist_get(struct perf_event *event)
 
 #ifdef CONFIG_EVENT_TRACING
 
-static const struct pmu perf_ops_tracepoint = {
+static struct pmu perf_ops_tracepoint = {
 	.enable		= perf_trace_enable,
 	.disable	= perf_trace_disable,
 	.start		= perf_swevent_int,
@@ -4849,7 +4849,7 @@ static void tp_perf_event_destroy(struct perf_event *event)
 	perf_trace_destroy(event);
 }
 
-static const struct pmu *tp_perf_event_init(struct perf_event *event)
+static struct pmu *tp_perf_event_init(struct perf_event *event)
 {
 	int err;
 
@@ -4896,7 +4896,7 @@ static void perf_event_free_filter(struct perf_event *event)
 
 #else
 
-static const struct pmu *tp_perf_event_init(struct perf_event *event)
+static struct pmu *tp_perf_event_init(struct perf_event *event)
 {
 	return NULL;
 }
@@ -4918,7 +4918,7 @@ static void bp_perf_event_destroy(struct perf_event *event)
 	release_bp_slot(event);
 }
 
-static const struct pmu *bp_perf_event_init(struct perf_event *bp)
+static struct pmu *bp_perf_event_init(struct perf_event *bp)
 {
 	int err;
 
@@ -4942,7 +4942,7 @@ void perf_bp_event(struct perf_event *bp, void *data)
 		perf_swevent_add(bp, 1, 1, &sample, regs);
 }
 #else
-static const struct pmu *bp_perf_event_init(struct perf_event *bp)
+static struct pmu *bp_perf_event_init(struct perf_event *bp)
 {
 	return NULL;
 }
@@ -4964,9 +4964,9 @@ static void sw_perf_event_destroy(struct perf_event *event)
 	swevent_hlist_put(event);
 }
 
-static const struct pmu *sw_perf_event_init(struct perf_event *event)
+static struct pmu *sw_perf_event_init(struct perf_event *event)
 {
-	const struct pmu *pmu = NULL;
+	struct pmu *pmu = NULL;
 	u64 event_id = event->attr.config;
 
 	/*
@@ -5028,7 +5028,7 @@ perf_event_alloc(struct perf_event_attr *attr,
 		   perf_overflow_handler_t overflow_handler,
 		   gfp_t gfpflags)
 {
-	const struct pmu *pmu;
+	struct pmu *pmu;
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
 	long err;
-- 
cgit v1.2.3


From b0a873ebbf87bf38bf70b5e39a7cadc96099fa13 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 11 Jun 2010 13:35:08 +0200
Subject: perf: Register PMU implementations

Simple registration interface for struct pmu, this provides the
infrastructure for removing all the weak functions.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/alpha/kernel/perf_event.c           |  37 +-
 arch/arm/kernel/perf_event.c             |  38 +-
 arch/powerpc/kernel/perf_event.c         |  46 +--
 arch/powerpc/kernel/perf_event_fsl_emb.c |  37 +-
 arch/sh/kernel/perf_event.c              |  35 +-
 arch/sparc/kernel/perf_event.c           |  29 +-
 arch/x86/kernel/cpu/perf_event.c         |  45 ++-
 include/linux/perf_event.h               |  10 +-
 kernel/hw_breakpoint.c                   |  35 +-
 kernel/perf_event.c                      | 588 +++++++++++++++----------------
 10 files changed, 488 insertions(+), 412 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c
index 56fa41590381..19660b5c298f 100644
--- a/arch/alpha/kernel/perf_event.c
+++ b/arch/alpha/kernel/perf_event.c
@@ -642,34 +642,39 @@ static int __hw_perf_event_init(struct perf_event *event)
 	return 0;
 }
 
-static struct pmu pmu = {
-	.enable		= alpha_pmu_enable,
-	.disable	= alpha_pmu_disable,
-	.read		= alpha_pmu_read,
-	.unthrottle	= alpha_pmu_unthrottle,
-};
-
-
 /*
  * Main entry point to initialise a HW performance event.
  */
-struct pmu *hw_perf_event_init(struct perf_event *event)
+static int alpha_pmu_event_init(struct perf_event *event)
 {
 	int err;
 
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (!alpha_pmu)
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 
 	/* Do the real initialisation work. */
 	err = __hw_perf_event_init(event);
 
-	if (err)
-		return ERR_PTR(err);
-
-	return &pmu;
+	return err;
 }
 
-
+static struct pmu pmu = {
+	.event_init	= alpha_pmu_event_init,
+	.enable		= alpha_pmu_enable,
+	.disable	= alpha_pmu_disable,
+	.read		= alpha_pmu_read,
+	.unthrottle	= alpha_pmu_unthrottle,
+};
 
 /*
  * Main entry point - enable HW performance counters.
@@ -838,5 +843,7 @@ void __init init_hw_perf_events(void)
 	/* And set up PMU specification */
 	alpha_pmu = &ev67_pmu;
 	perf_max_events = alpha_pmu->num_pmcs;
+
+	perf_pmu_register(&pmu);
 }
 
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 0671e92c5111..f62f9db35db3 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -306,12 +306,7 @@ out:
 	return err;
 }
 
-static struct pmu pmu = {
-	.enable	    = armpmu_enable,
-	.disable    = armpmu_disable,
-	.unthrottle = armpmu_unthrottle,
-	.read	    = armpmu_read,
-};
+static struct pmu pmu;
 
 static int
 validate_event(struct cpu_hw_events *cpuc,
@@ -491,20 +486,29 @@ __hw_perf_event_init(struct perf_event *event)
 	return err;
 }
 
-struct pmu *
-hw_perf_event_init(struct perf_event *event)
+static int armpmu_event_init(struct perf_event *event)
 {
 	int err = 0;
 
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (!armpmu)
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 
 	event->destroy = hw_perf_event_destroy;
 
 	if (!atomic_inc_not_zero(&active_events)) {
 		if (atomic_read(&active_events) > perf_max_events) {
 			atomic_dec(&active_events);
-			return ERR_PTR(-ENOSPC);
+			return -ENOSPC;
 		}
 
 		mutex_lock(&pmu_reserve_mutex);
@@ -518,15 +522,23 @@ hw_perf_event_init(struct perf_event *event)
 	}
 
 	if (err)
-		return ERR_PTR(err);
+		return err;
 
 	err = __hw_perf_event_init(event);
 	if (err)
 		hw_perf_event_destroy(event);
 
-	return err ? ERR_PTR(err) : &pmu;
+	return err;
 }
 
+static struct pmu pmu = {
+	.event_init = armpmu_event_init,
+	.enable	    = armpmu_enable,
+	.disable    = armpmu_disable,
+	.unthrottle = armpmu_unthrottle,
+	.read	    = armpmu_read,
+};
+
 void
 hw_perf_enable(void)
 {
@@ -2994,6 +3006,8 @@ init_hw_perf_events(void)
 		perf_max_events = -1;
 	}
 
+	perf_pmu_register(&pmu);
+
 	return 0;
 }
 arch_initcall(init_hw_perf_events);
diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c
index 5f78681ad902..19131b2614b9 100644
--- a/arch/powerpc/kernel/perf_event.c
+++ b/arch/powerpc/kernel/perf_event.c
@@ -904,16 +904,6 @@ int power_pmu_commit_txn(struct pmu *pmu)
 	return 0;
 }
 
-struct pmu power_pmu = {
-	.enable		= power_pmu_enable,
-	.disable	= power_pmu_disable,
-	.read		= power_pmu_read,
-	.unthrottle	= power_pmu_unthrottle,
-	.start_txn	= power_pmu_start_txn,
-	.cancel_txn	= power_pmu_cancel_txn,
-	.commit_txn	= power_pmu_commit_txn,
-};
-
 /*
  * Return 1 if we might be able to put event on a limited PMC,
  * or 0 if not.
@@ -1014,7 +1004,7 @@ static int hw_perf_cache_event(u64 config, u64 *eventp)
 	return 0;
 }
 
-struct pmu *hw_perf_event_init(struct perf_event *event)
+static int power_pmu_event_init(struct perf_event *event)
 {
 	u64 ev;
 	unsigned long flags;
@@ -1026,25 +1016,27 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 	struct cpu_hw_events *cpuhw;
 
 	if (!ppmu)
-		return ERR_PTR(-ENXIO);
+		return -ENOENT;
+
 	switch (event->attr.type) {
 	case PERF_TYPE_HARDWARE:
 		ev = event->attr.config;
 		if (ev >= ppmu->n_generic || ppmu->generic_events[ev] == 0)
-			return ERR_PTR(-EOPNOTSUPP);
+			return -EOPNOTSUPP;
 		ev = ppmu->generic_events[ev];
 		break;
 	case PERF_TYPE_HW_CACHE:
 		err = hw_perf_cache_event(event->attr.config, &ev);
 		if (err)
-			return ERR_PTR(err);
+			return err;
 		break;
 	case PERF_TYPE_RAW:
 		ev = event->attr.config;
 		break;
 	default:
-		return ERR_PTR(-EINVAL);
+		return -ENOENT;
 	}
+
 	event->hw.config_base = ev;
 	event->hw.idx = 0;
 
@@ -1081,7 +1073,7 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 			 */
 			ev = normal_pmc_alternative(ev, flags);
 			if (!ev)
-				return ERR_PTR(-EINVAL);
+				return -EINVAL;
 		}
 	}
 
@@ -1095,19 +1087,19 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 		n = collect_events(event->group_leader, ppmu->n_counter - 1,
 				   ctrs, events, cflags);
 		if (n < 0)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 	events[n] = ev;
 	ctrs[n] = event;
 	cflags[n] = flags;
 	if (check_excludes(ctrs, cflags, n, 1))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	cpuhw = &get_cpu_var(cpu_hw_events);
 	err = power_check_constraints(cpuhw, events, cflags, n + 1);
 	put_cpu_var(cpu_hw_events);
 	if (err)
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	event->hw.config = events[n];
 	event->hw.event_base = cflags[n];
@@ -1132,11 +1124,20 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 	}
 	event->destroy = hw_perf_event_destroy;
 
-	if (err)
-		return ERR_PTR(err);
-	return &power_pmu;
+	return err;
 }
 
+struct pmu power_pmu = {
+	.event_init	= power_pmu_event_init,
+	.enable		= power_pmu_enable,
+	.disable	= power_pmu_disable,
+	.read		= power_pmu_read,
+	.unthrottle	= power_pmu_unthrottle,
+	.start_txn	= power_pmu_start_txn,
+	.cancel_txn	= power_pmu_cancel_txn,
+	.commit_txn	= power_pmu_commit_txn,
+};
+
 /*
  * A counter has overflowed; update its count and record
  * things if requested.  Note that interrupts are hard-disabled
@@ -1342,6 +1343,7 @@ int register_power_pmu(struct power_pmu *pmu)
 		freeze_events_kernel = MMCR0_FCHV;
 #endif /* CONFIG_PPC64 */
 
+	perf_pmu_register(&power_pmu);
 	perf_cpu_notifier(power_pmu_notifier);
 
 	return 0;
diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c
index d7619b5e7a6e..ea6a804e43fd 100644
--- a/arch/powerpc/kernel/perf_event_fsl_emb.c
+++ b/arch/powerpc/kernel/perf_event_fsl_emb.c
@@ -378,13 +378,6 @@ static void fsl_emb_pmu_unthrottle(struct perf_event *event)
 	local_irq_restore(flags);
 }
 
-static struct pmu fsl_emb_pmu = {
-	.enable		= fsl_emb_pmu_enable,
-	.disable	= fsl_emb_pmu_disable,
-	.read		= fsl_emb_pmu_read,
-	.unthrottle	= fsl_emb_pmu_unthrottle,
-};
-
 /*
  * Release the PMU if this is the last perf_event.
  */
@@ -428,7 +421,7 @@ static int hw_perf_cache_event(u64 config, u64 *eventp)
 	return 0;
 }
 
-struct pmu *hw_perf_event_init(struct perf_event *event)
+static int fsl_emb_pmu_event_init(struct perf_event *event)
 {
 	u64 ev;
 	struct perf_event *events[MAX_HWEVENTS];
@@ -441,14 +434,14 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 	case PERF_TYPE_HARDWARE:
 		ev = event->attr.config;
 		if (ev >= ppmu->n_generic || ppmu->generic_events[ev] == 0)
-			return ERR_PTR(-EOPNOTSUPP);
+			return -EOPNOTSUPP;
 		ev = ppmu->generic_events[ev];
 		break;
 
 	case PERF_TYPE_HW_CACHE:
 		err = hw_perf_cache_event(event->attr.config, &ev);
 		if (err)
-			return ERR_PTR(err);
+			return err;
 		break;
 
 	case PERF_TYPE_RAW:
@@ -456,12 +449,12 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 		break;
 
 	default:
-		return ERR_PTR(-EINVAL);
+		return -ENOENT;
 	}
 
 	event->hw.config = ppmu->xlate_event(ev);
 	if (!(event->hw.config & FSL_EMB_EVENT_VALID))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	/*
 	 * If this is in a group, check if it can go on with all the
@@ -473,7 +466,7 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 		n = collect_events(event->group_leader,
 		                   ppmu->n_counter - 1, events);
 		if (n < 0)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 
 	if (event->hw.config & FSL_EMB_EVENT_RESTRICTED) {
@@ -484,7 +477,7 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 		}
 
 		if (num_restricted >= ppmu->n_restricted)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 
 	event->hw.idx = -1;
@@ -497,7 +490,7 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 	if (event->attr.exclude_kernel)
 		event->hw.config_base |= PMLCA_FCS;
 	if (event->attr.exclude_idle)
-		return ERR_PTR(-ENOTSUPP);
+		return -ENOTSUPP;
 
 	event->hw.last_period = event->hw.sample_period;
 	local64_set(&event->hw.period_left, event->hw.last_period);
@@ -523,11 +516,17 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 	}
 	event->destroy = hw_perf_event_destroy;
 
-	if (err)
-		return ERR_PTR(err);
-	return &fsl_emb_pmu;
+	return err;
 }
 
+static struct pmu fsl_emb_pmu = {
+	.event_init	= fsl_emb_pmu_event_init,
+	.enable		= fsl_emb_pmu_enable,
+	.disable	= fsl_emb_pmu_disable,
+	.read		= fsl_emb_pmu_read,
+	.unthrottle	= fsl_emb_pmu_unthrottle,
+};
+
 /*
  * A counter has overflowed; update its count and record
  * things if requested.  Note that interrupts are hard-disabled
@@ -651,5 +650,7 @@ int register_fsl_emb_pmu(struct fsl_emb_pmu *pmu)
 	pr_info("%s performance monitor hardware support registered\n",
 		pmu->name);
 
+	perf_pmu_register(&fsl_emb_pmu);
+
 	return 0;
 }
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 395572c94c6a..8cb206597e0c 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -257,26 +257,38 @@ static void sh_pmu_read(struct perf_event *event)
 	sh_perf_event_update(event, &event->hw, event->hw.idx);
 }
 
-static struct pmu pmu = {
-	.enable		= sh_pmu_enable,
-	.disable	= sh_pmu_disable,
-	.read		= sh_pmu_read,
-};
-
-struct pmu *hw_perf_event_init(struct perf_event *event)
+static int sh_pmu_event_init(struct perf_event *event)
 {
-	int err = __hw_perf_event_init(event);
+	int err;
+
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HW_CACHE:
+	case PERF_TYPE_HARDWARE:
+		err = __hw_perf_event_init(event);
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (unlikely(err)) {
 		if (event->destroy)
 			event->destroy(event);
-		return ERR_PTR(err);
 	}
 
-	return &pmu;
+	return err;
 }
 
+static struct pmu pmu = {
+	.event_init	= sh_pmu_event_init,
+	.enable		= sh_pmu_enable,
+	.disable	= sh_pmu_disable,
+	.read		= sh_pmu_read,
+};
+
 static void sh_pmu_setup(int cpu)
-{
+
 	struct cpu_hw_events *cpuhw = &per_cpu(cpu_hw_events, cpu);
 
 	memset(cpuhw, 0, sizeof(struct cpu_hw_events));
@@ -325,6 +337,7 @@ int __cpuinit register_sh_pmu(struct sh_pmu *pmu)
 
 	WARN_ON(pmu->num_events > MAX_HWEVENTS);
 
+	perf_pmu_register(&pmu);
 	perf_cpu_notifier(sh_pmu_notifier);
 	return 0;
 }
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 481b894a5018..bed4327f5a7a 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -1025,7 +1025,7 @@ out:
 	return ret;
 }
 
-static int __hw_perf_event_init(struct perf_event *event)
+static int sparc_pmu_event_init(struct perf_event *event)
 {
 	struct perf_event_attr *attr = &event->attr;
 	struct perf_event *evts[MAX_HWEVENTS];
@@ -1038,17 +1038,27 @@ static int __hw_perf_event_init(struct perf_event *event)
 	if (atomic_read(&nmi_active) < 0)
 		return -ENODEV;
 
-	if (attr->type == PERF_TYPE_HARDWARE) {
+	switch (attr->type) {
+	case PERF_TYPE_HARDWARE:
 		if (attr->config >= sparc_pmu->max_events)
 			return -EINVAL;
 		pmap = sparc_pmu->event_map(attr->config);
-	} else if (attr->type == PERF_TYPE_HW_CACHE) {
+		break;
+
+	case PERF_TYPE_HW_CACHE:
 		pmap = sparc_map_cache_event(attr->config);
 		if (IS_ERR(pmap))
 			return PTR_ERR(pmap);
-	} else
+		break;
+
+	case PERF_TYPE_RAW:
 		return -EOPNOTSUPP;
 
+	default:
+		return -ENOENT;
+
+	}
+
 	/* We save the enable bits in the config_base.  */
 	hwc->config_base = sparc_pmu->irq_bit;
 	if (!attr->exclude_user)
@@ -1143,6 +1153,7 @@ static int sparc_pmu_commit_txn(struct pmu *pmu)
 }
 
 static struct pmu pmu = {
+	.event_init	= sparc_pmu_event_init,
 	.enable		= sparc_pmu_enable,
 	.disable	= sparc_pmu_disable,
 	.read		= sparc_pmu_read,
@@ -1152,15 +1163,6 @@ static struct pmu pmu = {
 	.commit_txn	= sparc_pmu_commit_txn,
 };
 
-struct pmu *hw_perf_event_init(struct perf_event *event)
-{
-	int err = __hw_perf_event_init(event);
-
-	if (err)
-		return ERR_PTR(err);
-	return &pmu;
-}
-
 void perf_event_print_debug(void)
 {
 	unsigned long flags;
@@ -1280,6 +1282,7 @@ void __init init_hw_perf_events(void)
 	/* All sparc64 PMUs currently have 2 events.  */
 	perf_max_events = 2;
 
+	perf_pmu_register(&pmu);
 	register_die_notifier(&perf_event_nmi_notifier);
 }
 
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index fdd97f2e9961..2c89264ee791 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -530,7 +530,7 @@ static int x86_pmu_hw_config(struct perf_event *event)
 /*
  * Setup the hardware configuration for a given attr_type
  */
-static int __hw_perf_event_init(struct perf_event *event)
+static int __x86_pmu_event_init(struct perf_event *event)
 {
 	int err;
 
@@ -1414,6 +1414,7 @@ void __init init_hw_perf_events(void)
 	pr_info("... fixed-purpose events:   %d\n",     x86_pmu.num_counters_fixed);
 	pr_info("... event mask:             %016Lx\n", x86_pmu.intel_ctrl);
 
+	perf_pmu_register(&pmu);
 	perf_cpu_notifier(x86_pmu_notifier);
 }
 
@@ -1483,18 +1484,6 @@ static int x86_pmu_commit_txn(struct pmu *pmu)
 	return 0;
 }
 
-static struct pmu pmu = {
-	.enable		= x86_pmu_enable,
-	.disable	= x86_pmu_disable,
-	.start		= x86_pmu_start,
-	.stop		= x86_pmu_stop,
-	.read		= x86_pmu_read,
-	.unthrottle	= x86_pmu_unthrottle,
-	.start_txn	= x86_pmu_start_txn,
-	.cancel_txn	= x86_pmu_cancel_txn,
-	.commit_txn	= x86_pmu_commit_txn,
-};
-
 /*
  * validate that we can schedule this event
  */
@@ -1569,12 +1558,22 @@ out:
 	return ret;
 }
 
-struct pmu *hw_perf_event_init(struct perf_event *event)
+int x86_pmu_event_init(struct perf_event *event)
 {
 	struct pmu *tmp;
 	int err;
 
-	err = __hw_perf_event_init(event);
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
+	err = __x86_pmu_event_init(event);
 	if (!err) {
 		/*
 		 * we temporarily connect event to its pmu
@@ -1594,12 +1593,24 @@ struct pmu *hw_perf_event_init(struct perf_event *event)
 	if (err) {
 		if (event->destroy)
 			event->destroy(event);
-		return ERR_PTR(err);
 	}
 
-	return &pmu;
+	return err;
 }
 
+static struct pmu pmu = {
+	.event_init	= x86_pmu_event_init,
+	.enable		= x86_pmu_enable,
+	.disable	= x86_pmu_disable,
+	.start		= x86_pmu_start,
+	.stop		= x86_pmu_stop,
+	.read		= x86_pmu_read,
+	.unthrottle	= x86_pmu_unthrottle,
+	.start_txn	= x86_pmu_start_txn,
+	.cancel_txn	= x86_pmu_cancel_txn,
+	.commit_txn	= x86_pmu_commit_txn,
+};
+
 /*
  * callchain support
  */
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 09d048b52115..ab72f56eb372 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -561,6 +561,13 @@ struct perf_event;
  * struct pmu - generic performance monitoring unit
  */
 struct pmu {
+	struct list_head		entry;
+
+	/*
+	 * Should return -ENOENT when the @event doesn't match this pmu
+	 */
+	int (*event_init)		(struct perf_event *event);
+
 	int (*enable)			(struct perf_event *event);
 	void (*disable)			(struct perf_event *event);
 	int (*start)			(struct perf_event *event);
@@ -849,7 +856,8 @@ struct perf_output_handle {
  */
 extern int perf_max_events;
 
-extern struct pmu *hw_perf_event_init(struct perf_event *event);
+extern int perf_pmu_register(struct pmu *pmu);
+extern void perf_pmu_unregister(struct pmu *pmu);
 
 extern void perf_event_task_sched_in(struct task_struct *task);
 extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index d71a987fd2bf..e9c5cfa1fd20 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -565,6 +565,34 @@ static struct notifier_block hw_breakpoint_exceptions_nb = {
 	.priority = 0x7fffffff
 };
 
+static void bp_perf_event_destroy(struct perf_event *event)
+{
+	release_bp_slot(event);
+}
+
+static int hw_breakpoint_event_init(struct perf_event *bp)
+{
+	int err;
+
+	if (bp->attr.type != PERF_TYPE_BREAKPOINT)
+		return -ENOENT;
+
+	err = register_perf_hw_breakpoint(bp);
+	if (err)
+		return err;
+
+	bp->destroy = bp_perf_event_destroy;
+
+	return 0;
+}
+
+static struct pmu perf_breakpoint = {
+	.event_init	= hw_breakpoint_event_init,
+	.enable		= arch_install_hw_breakpoint,
+	.disable	= arch_uninstall_hw_breakpoint,
+	.read		= hw_breakpoint_pmu_read,
+};
+
 static int __init init_hw_breakpoint(void)
 {
 	unsigned int **task_bp_pinned;
@@ -586,6 +614,8 @@ static int __init init_hw_breakpoint(void)
 
 	constraints_initialized = 1;
 
+	perf_pmu_register(&perf_breakpoint);
+
 	return register_die_notifier(&hw_breakpoint_exceptions_nb);
 
  err_alloc:
@@ -601,8 +631,3 @@ static int __init init_hw_breakpoint(void)
 core_initcall(init_hw_breakpoint);
 
 
-struct pmu perf_ops_bp = {
-	.enable		= arch_install_hw_breakpoint,
-	.disable	= arch_uninstall_hw_breakpoint,
-	.read		= hw_breakpoint_pmu_read,
-};
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index fb46fd13f31f..288ce43de57c 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -31,7 +31,6 @@
 #include <linux/kernel_stat.h>
 #include <linux/perf_event.h>
 #include <linux/ftrace_event.h>
-#include <linux/hw_breakpoint.h>
 
 #include <asm/irq_regs.h>
 
@@ -72,14 +71,6 @@ static atomic64_t perf_event_id;
  */
 static DEFINE_SPINLOCK(perf_resource_lock);
 
-/*
- * Architecture provided APIs - weak aliases:
- */
-extern __weak struct pmu *hw_perf_event_init(struct perf_event *event)
-{
-	return NULL;
-}
-
 void __weak hw_perf_disable(void)		{ barrier(); }
 void __weak hw_perf_enable(void)		{ barrier(); }
 
@@ -4501,182 +4492,6 @@ static int perf_swevent_int(struct perf_event *event)
 	return 0;
 }
 
-static struct pmu perf_ops_generic = {
-	.enable		= perf_swevent_enable,
-	.disable	= perf_swevent_disable,
-	.start		= perf_swevent_int,
-	.stop		= perf_swevent_void,
-	.read		= perf_swevent_read,
-	.unthrottle	= perf_swevent_void, /* hwc->interrupts already reset */
-};
-
-/*
- * hrtimer based swevent callback
- */
-
-static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
-{
-	enum hrtimer_restart ret = HRTIMER_RESTART;
-	struct perf_sample_data data;
-	struct pt_regs *regs;
-	struct perf_event *event;
-	u64 period;
-
-	event = container_of(hrtimer, struct perf_event, hw.hrtimer);
-	event->pmu->read(event);
-
-	perf_sample_data_init(&data, 0);
-	data.period = event->hw.last_period;
-	regs = get_irq_regs();
-
-	if (regs && !perf_exclude_event(event, regs)) {
-		if (!(event->attr.exclude_idle && current->pid == 0))
-			if (perf_event_overflow(event, 0, &data, regs))
-				ret = HRTIMER_NORESTART;
-	}
-
-	period = max_t(u64, 10000, event->hw.sample_period);
-	hrtimer_forward_now(hrtimer, ns_to_ktime(period));
-
-	return ret;
-}
-
-static void perf_swevent_start_hrtimer(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-
-	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-	hwc->hrtimer.function = perf_swevent_hrtimer;
-	if (hwc->sample_period) {
-		u64 period;
-
-		if (hwc->remaining) {
-			if (hwc->remaining < 0)
-				period = 10000;
-			else
-				period = hwc->remaining;
-			hwc->remaining = 0;
-		} else {
-			period = max_t(u64, 10000, hwc->sample_period);
-		}
-		__hrtimer_start_range_ns(&hwc->hrtimer,
-				ns_to_ktime(period), 0,
-				HRTIMER_MODE_REL, 0);
-	}
-}
-
-static void perf_swevent_cancel_hrtimer(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-
-	if (hwc->sample_period) {
-		ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer);
-		hwc->remaining = ktime_to_ns(remaining);
-
-		hrtimer_cancel(&hwc->hrtimer);
-	}
-}
-
-/*
- * Software event: cpu wall time clock
- */
-
-static void cpu_clock_perf_event_update(struct perf_event *event)
-{
-	int cpu = raw_smp_processor_id();
-	s64 prev;
-	u64 now;
-
-	now = cpu_clock(cpu);
-	prev = local64_xchg(&event->hw.prev_count, now);
-	local64_add(now - prev, &event->count);
-}
-
-static int cpu_clock_perf_event_enable(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-	int cpu = raw_smp_processor_id();
-
-	local64_set(&hwc->prev_count, cpu_clock(cpu));
-	perf_swevent_start_hrtimer(event);
-
-	return 0;
-}
-
-static void cpu_clock_perf_event_disable(struct perf_event *event)
-{
-	perf_swevent_cancel_hrtimer(event);
-	cpu_clock_perf_event_update(event);
-}
-
-static void cpu_clock_perf_event_read(struct perf_event *event)
-{
-	cpu_clock_perf_event_update(event);
-}
-
-static struct pmu perf_ops_cpu_clock = {
-	.enable		= cpu_clock_perf_event_enable,
-	.disable	= cpu_clock_perf_event_disable,
-	.read		= cpu_clock_perf_event_read,
-};
-
-/*
- * Software event: task time clock
- */
-
-static void task_clock_perf_event_update(struct perf_event *event, u64 now)
-{
-	u64 prev;
-	s64 delta;
-
-	prev = local64_xchg(&event->hw.prev_count, now);
-	delta = now - prev;
-	local64_add(delta, &event->count);
-}
-
-static int task_clock_perf_event_enable(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-	u64 now;
-
-	now = event->ctx->time;
-
-	local64_set(&hwc->prev_count, now);
-
-	perf_swevent_start_hrtimer(event);
-
-	return 0;
-}
-
-static void task_clock_perf_event_disable(struct perf_event *event)
-{
-	perf_swevent_cancel_hrtimer(event);
-	task_clock_perf_event_update(event, event->ctx->time);
-
-}
-
-static void task_clock_perf_event_read(struct perf_event *event)
-{
-	u64 time;
-
-	if (!in_nmi()) {
-		update_context_time(event->ctx);
-		time = event->ctx->time;
-	} else {
-		u64 now = perf_clock();
-		u64 delta = now - event->ctx->timestamp;
-		time = event->ctx->time + delta;
-	}
-
-	task_clock_perf_event_update(event, time);
-}
-
-static struct pmu perf_ops_task_clock = {
-	.enable		= task_clock_perf_event_enable,
-	.disable	= task_clock_perf_event_disable,
-	.read		= task_clock_perf_event_read,
-};
-
 /* Deref the hlist from the update side */
 static inline struct swevent_hlist *
 swevent_hlist_deref(struct perf_cpu_context *cpuctx)
@@ -4783,17 +4598,63 @@ static int swevent_hlist_get(struct perf_event *event)
 	return err;
 }
 
-#ifdef CONFIG_EVENT_TRACING
+atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
 
-static struct pmu perf_ops_tracepoint = {
-	.enable		= perf_trace_enable,
-	.disable	= perf_trace_disable,
+static void sw_perf_event_destroy(struct perf_event *event)
+{
+	u64 event_id = event->attr.config;
+
+	WARN_ON(event->parent);
+
+	atomic_dec(&perf_swevent_enabled[event_id]);
+	swevent_hlist_put(event);
+}
+
+static int perf_swevent_init(struct perf_event *event)
+{
+	int event_id = event->attr.config;
+
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	switch (event_id) {
+	case PERF_COUNT_SW_CPU_CLOCK:
+	case PERF_COUNT_SW_TASK_CLOCK:
+		return -ENOENT;
+
+	default:
+		break;
+	}
+
+	if (event_id > PERF_COUNT_SW_MAX)
+		return -ENOENT;
+
+	if (!event->parent) {
+		int err;
+
+		err = swevent_hlist_get(event);
+		if (err)
+			return err;
+
+		atomic_inc(&perf_swevent_enabled[event_id]);
+		event->destroy = sw_perf_event_destroy;
+	}
+
+	return 0;
+}
+
+static struct pmu perf_swevent = {
+	.event_init	= perf_swevent_init,
+	.enable		= perf_swevent_enable,
+	.disable	= perf_swevent_disable,
 	.start		= perf_swevent_int,
 	.stop		= perf_swevent_void,
 	.read		= perf_swevent_read,
-	.unthrottle	= perf_swevent_void,
+	.unthrottle	= perf_swevent_void, /* hwc->interrupts already reset */
 };
 
+#ifdef CONFIG_EVENT_TRACING
+
 static int perf_tp_filter_match(struct perf_event *event,
 				struct perf_sample_data *data)
 {
@@ -4849,10 +4710,13 @@ static void tp_perf_event_destroy(struct perf_event *event)
 	perf_trace_destroy(event);
 }
 
-static struct pmu *tp_perf_event_init(struct perf_event *event)
+static int perf_tp_event_init(struct perf_event *event)
 {
 	int err;
 
+	if (event->attr.type != PERF_TYPE_TRACEPOINT)
+		return -ENOENT;
+
 	/*
 	 * Raw tracepoint data is a severe data leak, only allow root to
 	 * have these.
@@ -4860,15 +4724,30 @@ static struct pmu *tp_perf_event_init(struct perf_event *event)
 	if ((event->attr.sample_type & PERF_SAMPLE_RAW) &&
 			perf_paranoid_tracepoint_raw() &&
 			!capable(CAP_SYS_ADMIN))
-		return ERR_PTR(-EPERM);
+		return -EPERM;
 
 	err = perf_trace_init(event);
 	if (err)
-		return NULL;
+		return err;
 
 	event->destroy = tp_perf_event_destroy;
 
-	return &perf_ops_tracepoint;
+	return 0;
+}
+
+static struct pmu perf_tracepoint = {
+	.event_init	= perf_tp_event_init,
+	.enable		= perf_trace_enable,
+	.disable	= perf_trace_disable,
+	.start		= perf_swevent_int,
+	.stop		= perf_swevent_void,
+	.read		= perf_swevent_read,
+	.unthrottle	= perf_swevent_void,
+};
+
+static inline void perf_tp_register(void)
+{
+	perf_pmu_register(&perf_tracepoint);
 }
 
 static int perf_event_set_filter(struct perf_event *event, void __user *arg)
@@ -4896,9 +4775,8 @@ static void perf_event_free_filter(struct perf_event *event)
 
 #else
 
-static struct pmu *tp_perf_event_init(struct perf_event *event)
+static inline void perf_tp_register(void)
 {
-	return NULL;
 }
 
 static int perf_event_set_filter(struct perf_event *event, void __user *arg)
@@ -4913,105 +4791,247 @@ static void perf_event_free_filter(struct perf_event *event)
 #endif /* CONFIG_EVENT_TRACING */
 
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
-static void bp_perf_event_destroy(struct perf_event *event)
+void perf_bp_event(struct perf_event *bp, void *data)
 {
-	release_bp_slot(event);
+	struct perf_sample_data sample;
+	struct pt_regs *regs = data;
+
+	perf_sample_data_init(&sample, bp->attr.bp_addr);
+
+	if (!perf_exclude_event(bp, regs))
+		perf_swevent_add(bp, 1, 1, &sample, regs);
 }
+#endif
+
+/*
+ * hrtimer based swevent callback
+ */
 
-static struct pmu *bp_perf_event_init(struct perf_event *bp)
+static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
 {
-	int err;
+	enum hrtimer_restart ret = HRTIMER_RESTART;
+	struct perf_sample_data data;
+	struct pt_regs *regs;
+	struct perf_event *event;
+	u64 period;
 
-	err = register_perf_hw_breakpoint(bp);
-	if (err)
-		return ERR_PTR(err);
+	event = container_of(hrtimer, struct perf_event, hw.hrtimer);
+	event->pmu->read(event);
+
+	perf_sample_data_init(&data, 0);
+	data.period = event->hw.last_period;
+	regs = get_irq_regs();
+
+	if (regs && !perf_exclude_event(event, regs)) {
+		if (!(event->attr.exclude_idle && current->pid == 0))
+			if (perf_event_overflow(event, 0, &data, regs))
+				ret = HRTIMER_NORESTART;
+	}
 
-	bp->destroy = bp_perf_event_destroy;
+	period = max_t(u64, 10000, event->hw.sample_period);
+	hrtimer_forward_now(hrtimer, ns_to_ktime(period));
 
-	return &perf_ops_bp;
+	return ret;
 }
 
-void perf_bp_event(struct perf_event *bp, void *data)
+static void perf_swevent_start_hrtimer(struct perf_event *event)
 {
-	struct perf_sample_data sample;
-	struct pt_regs *regs = data;
+	struct hw_perf_event *hwc = &event->hw;
 
-	perf_sample_data_init(&sample, bp->attr.bp_addr);
+	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hwc->hrtimer.function = perf_swevent_hrtimer;
+	if (hwc->sample_period) {
+		u64 period;
 
-	if (!perf_exclude_event(bp, regs))
-		perf_swevent_add(bp, 1, 1, &sample, regs);
+		if (hwc->remaining) {
+			if (hwc->remaining < 0)
+				period = 10000;
+			else
+				period = hwc->remaining;
+			hwc->remaining = 0;
+		} else {
+			period = max_t(u64, 10000, hwc->sample_period);
+		}
+		__hrtimer_start_range_ns(&hwc->hrtimer,
+				ns_to_ktime(period), 0,
+				HRTIMER_MODE_REL, 0);
+	}
 }
-#else
-static struct pmu *bp_perf_event_init(struct perf_event *bp)
+
+static void perf_swevent_cancel_hrtimer(struct perf_event *event)
 {
-	return NULL;
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (hwc->sample_period) {
+		ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer);
+		hwc->remaining = ktime_to_ns(remaining);
+
+		hrtimer_cancel(&hwc->hrtimer);
+	}
 }
 
-void perf_bp_event(struct perf_event *bp, void *regs)
+/*
+ * Software event: cpu wall time clock
+ */
+
+static void cpu_clock_event_update(struct perf_event *event)
 {
+	int cpu = raw_smp_processor_id();
+	s64 prev;
+	u64 now;
+
+	now = cpu_clock(cpu);
+	prev = local64_xchg(&event->hw.prev_count, now);
+	local64_add(now - prev, &event->count);
 }
-#endif
 
-atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
+static int cpu_clock_event_enable(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int cpu = raw_smp_processor_id();
 
-static void sw_perf_event_destroy(struct perf_event *event)
+	local64_set(&hwc->prev_count, cpu_clock(cpu));
+	perf_swevent_start_hrtimer(event);
+
+	return 0;
+}
+
+static void cpu_clock_event_disable(struct perf_event *event)
 {
-	u64 event_id = event->attr.config;
+	perf_swevent_cancel_hrtimer(event);
+	cpu_clock_event_update(event);
+}
 
-	WARN_ON(event->parent);
+static void cpu_clock_event_read(struct perf_event *event)
+{
+	cpu_clock_event_update(event);
+}
 
-	atomic_dec(&perf_swevent_enabled[event_id]);
-	swevent_hlist_put(event);
+static int cpu_clock_event_init(struct perf_event *event)
+{
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	if (event->attr.config != PERF_COUNT_SW_CPU_CLOCK)
+		return -ENOENT;
+
+	return 0;
 }
 
-static struct pmu *sw_perf_event_init(struct perf_event *event)
+static struct pmu perf_cpu_clock = {
+	.event_init	= cpu_clock_event_init,
+	.enable		= cpu_clock_event_enable,
+	.disable	= cpu_clock_event_disable,
+	.read		= cpu_clock_event_read,
+};
+
+/*
+ * Software event: task time clock
+ */
+
+static void task_clock_event_update(struct perf_event *event, u64 now)
 {
-	struct pmu *pmu = NULL;
-	u64 event_id = event->attr.config;
+	u64 prev;
+	s64 delta;
 
-	/*
-	 * Software events (currently) can't in general distinguish
-	 * between user, kernel and hypervisor events.
-	 * However, context switches and cpu migrations are considered
-	 * to be kernel events, and page faults are never hypervisor
-	 * events.
-	 */
-	switch (event_id) {
-	case PERF_COUNT_SW_CPU_CLOCK:
-		pmu = &perf_ops_cpu_clock;
+	prev = local64_xchg(&event->hw.prev_count, now);
+	delta = now - prev;
+	local64_add(delta, &event->count);
+}
 
-		break;
-	case PERF_COUNT_SW_TASK_CLOCK:
-		/*
-		 * If the user instantiates this as a per-cpu event,
-		 * use the cpu_clock event instead.
-		 */
-		if (event->ctx->task)
-			pmu = &perf_ops_task_clock;
-		else
-			pmu = &perf_ops_cpu_clock;
+static int task_clock_event_enable(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 now;
 
-		break;
-	case PERF_COUNT_SW_PAGE_FAULTS:
-	case PERF_COUNT_SW_PAGE_FAULTS_MIN:
-	case PERF_COUNT_SW_PAGE_FAULTS_MAJ:
-	case PERF_COUNT_SW_CONTEXT_SWITCHES:
-	case PERF_COUNT_SW_CPU_MIGRATIONS:
-	case PERF_COUNT_SW_ALIGNMENT_FAULTS:
-	case PERF_COUNT_SW_EMULATION_FAULTS:
-		if (!event->parent) {
-			int err;
-
-			err = swevent_hlist_get(event);
-			if (err)
-				return ERR_PTR(err);
+	now = event->ctx->time;
 
-			atomic_inc(&perf_swevent_enabled[event_id]);
-			event->destroy = sw_perf_event_destroy;
+	local64_set(&hwc->prev_count, now);
+
+	perf_swevent_start_hrtimer(event);
+
+	return 0;
+}
+
+static void task_clock_event_disable(struct perf_event *event)
+{
+	perf_swevent_cancel_hrtimer(event);
+	task_clock_event_update(event, event->ctx->time);
+
+}
+
+static void task_clock_event_read(struct perf_event *event)
+{
+	u64 time;
+
+	if (!in_nmi()) {
+		update_context_time(event->ctx);
+		time = event->ctx->time;
+	} else {
+		u64 now = perf_clock();
+		u64 delta = now - event->ctx->timestamp;
+		time = event->ctx->time + delta;
+	}
+
+	task_clock_event_update(event, time);
+}
+
+static int task_clock_event_init(struct perf_event *event)
+{
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	if (event->attr.config != PERF_COUNT_SW_TASK_CLOCK)
+		return -ENOENT;
+
+	return 0;
+}
+
+static struct pmu perf_task_clock = {
+	.event_init	= task_clock_event_init,
+	.enable		= task_clock_event_enable,
+	.disable	= task_clock_event_disable,
+	.read		= task_clock_event_read,
+};
+
+static LIST_HEAD(pmus);
+static DEFINE_MUTEX(pmus_lock);
+static struct srcu_struct pmus_srcu;
+
+int perf_pmu_register(struct pmu *pmu)
+{
+	mutex_lock(&pmus_lock);
+	list_add_rcu(&pmu->entry, &pmus);
+	mutex_unlock(&pmus_lock);
+
+	return 0;
+}
+
+void perf_pmu_unregister(struct pmu *pmu)
+{
+	mutex_lock(&pmus_lock);
+	list_del_rcu(&pmu->entry);
+	mutex_unlock(&pmus_lock);
+
+	synchronize_srcu(&pmus_srcu);
+}
+
+struct pmu *perf_init_event(struct perf_event *event)
+{
+	struct pmu *pmu = NULL;
+	int idx;
+
+	idx = srcu_read_lock(&pmus_srcu);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		int ret = pmu->event_init(event);
+		if (!ret)
+			break;
+		if (ret != -ENOENT) {
+			pmu = ERR_PTR(ret);
+			break;
 		}
-		pmu = &perf_ops_generic;
-		break;
 	}
+	srcu_read_unlock(&pmus_srcu, idx);
 
 	return pmu;
 }
@@ -5092,29 +5112,8 @@ perf_event_alloc(struct perf_event_attr *attr,
 	if (attr->inherit && (attr->read_format & PERF_FORMAT_GROUP))
 		goto done;
 
-	switch (attr->type) {
-	case PERF_TYPE_RAW:
-	case PERF_TYPE_HARDWARE:
-	case PERF_TYPE_HW_CACHE:
-		pmu = hw_perf_event_init(event);
-		break;
-
-	case PERF_TYPE_SOFTWARE:
-		pmu = sw_perf_event_init(event);
-		break;
-
-	case PERF_TYPE_TRACEPOINT:
-		pmu = tp_perf_event_init(event);
-		break;
+	pmu = perf_init_event(event);
 
-	case PERF_TYPE_BREAKPOINT:
-		pmu = bp_perf_event_init(event);
-		break;
-
-
-	default:
-		break;
-	}
 done:
 	err = 0;
 	if (!pmu)
@@ -5979,22 +5978,15 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu)
 	return NOTIFY_OK;
 }
 
-/*
- * This has to have a higher priority than migration_notifier in sched.c.
- */
-static struct notifier_block __cpuinitdata perf_cpu_nb = {
-	.notifier_call		= perf_cpu_notify,
-	.priority		= 20,
-};
-
 void __init perf_event_init(void)
 {
 	perf_event_init_all_cpus();
-	perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_UP_PREPARE,
-			(void *)(long)smp_processor_id());
-	perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_ONLINE,
-			(void *)(long)smp_processor_id());
-	register_cpu_notifier(&perf_cpu_nb);
+	init_srcu_struct(&pmus_srcu);
+	perf_pmu_register(&perf_swevent);
+	perf_pmu_register(&perf_cpu_clock);
+	perf_pmu_register(&perf_task_clock);
+	perf_tp_register();
+	perf_cpu_notifier(perf_cpu_notify);
 }
 
 static ssize_t perf_show_reserve_percpu(struct sysdev_class *class,
-- 
cgit v1.2.3


From 24cd7f54a0d47e1d5b3de29e2456bfbd2d8447b7 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 11 Jun 2010 17:32:03 +0200
Subject: perf: Reduce perf_disable() usage

Since the current perf_disable() usage is only an optimization,
remove it for now. This eases the removal of the __weak
hw_perf_enable() interface.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/arm/kernel/perf_event.c             |  3 +++
 arch/powerpc/kernel/perf_event.c         |  3 +++
 arch/powerpc/kernel/perf_event_fsl_emb.c |  8 +++++--
 arch/sh/kernel/perf_event.c              | 11 +++++++---
 arch/sparc/kernel/perf_event.c           |  3 +++
 arch/x86/kernel/cpu/perf_event.c         | 22 ++++++++++++-------
 include/linux/perf_event.h               | 20 ++++++++---------
 kernel/perf_event.c                      | 37 +-------------------------------
 8 files changed, 48 insertions(+), 59 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index f62f9db35db3..afc92c580d18 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -277,6 +277,8 @@ armpmu_enable(struct perf_event *event)
 	int idx;
 	int err = 0;
 
+	perf_disable();
+
 	/* If we don't have a space for the counter then finish early. */
 	idx = armpmu->get_event_idx(cpuc, hwc);
 	if (idx < 0) {
@@ -303,6 +305,7 @@ armpmu_enable(struct perf_event *event)
 	perf_event_update_userpage(event);
 
 out:
+	perf_enable();
 	return err;
 }
 
diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c
index 19131b2614b9..c1408821dbc2 100644
--- a/arch/powerpc/kernel/perf_event.c
+++ b/arch/powerpc/kernel/perf_event.c
@@ -861,6 +861,7 @@ void power_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
+	perf_disable();
 	cpuhw->group_flag |= PERF_EVENT_TXN;
 	cpuhw->n_txn_start = cpuhw->n_events;
 }
@@ -875,6 +876,7 @@ void power_pmu_cancel_txn(struct pmu *pmu)
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_enable();
 }
 
 /*
@@ -901,6 +903,7 @@ int power_pmu_commit_txn(struct pmu *pmu)
 		cpuhw->event[i]->hw.config = cpuhw->events[i];
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_enable();
 	return 0;
 }
 
diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c
index ea6a804e43fd..9bc84a7fd901 100644
--- a/arch/powerpc/kernel/perf_event_fsl_emb.c
+++ b/arch/powerpc/kernel/perf_event_fsl_emb.c
@@ -262,7 +262,7 @@ static int collect_events(struct perf_event *group, int max_count,
 	return n;
 }
 
-/* perf must be disabled, context locked on entry */
+/* context locked on entry */
 static int fsl_emb_pmu_enable(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuhw;
@@ -271,6 +271,7 @@ static int fsl_emb_pmu_enable(struct perf_event *event)
 	u64 val;
 	int i;
 
+	perf_disable();
 	cpuhw = &get_cpu_var(cpu_hw_events);
 
 	if (event->hw.config & FSL_EMB_EVENT_RESTRICTED)
@@ -310,15 +311,17 @@ static int fsl_emb_pmu_enable(struct perf_event *event)
 	ret = 0;
  out:
 	put_cpu_var(cpu_hw_events);
+	perf_enable();
 	return ret;
 }
 
-/* perf must be disabled, context locked on entry */
+/* context locked on entry */
 static void fsl_emb_pmu_disable(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuhw;
 	int i = event->hw.idx;
 
+	perf_disable();
 	if (i < 0)
 		goto out;
 
@@ -346,6 +349,7 @@ static void fsl_emb_pmu_disable(struct perf_event *event)
 	cpuhw->n_events--;
 
  out:
+	perf_enable();
 	put_cpu_var(cpu_hw_events);
 }
 
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 8cb206597e0c..d042989ceb45 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -230,11 +230,14 @@ static int sh_pmu_enable(struct perf_event *event)
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
+	int ret = -EAGAIN;
+
+	perf_disable();
 
 	if (test_and_set_bit(idx, cpuc->used_mask)) {
 		idx = find_first_zero_bit(cpuc->used_mask, sh_pmu->num_events);
 		if (idx == sh_pmu->num_events)
-			return -EAGAIN;
+			goto out;
 
 		set_bit(idx, cpuc->used_mask);
 		hwc->idx = idx;
@@ -248,8 +251,10 @@ static int sh_pmu_enable(struct perf_event *event)
 	sh_pmu->enable(hwc, idx);
 
 	perf_event_update_userpage(event);
-
-	return 0;
+	ret = 0;
+out:
+	perf_enable();
+	return ret;
 }
 
 static void sh_pmu_read(struct perf_event *event)
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index bed4327f5a7a..d0131deeeaf6 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -1113,6 +1113,7 @@ static void sparc_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
+	perf_disable();
 	cpuhw->group_flag |= PERF_EVENT_TXN;
 }
 
@@ -1126,6 +1127,7 @@ static void sparc_pmu_cancel_txn(struct pmu *pmu)
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_enable();
 }
 
 /*
@@ -1149,6 +1151,7 @@ static int sparc_pmu_commit_txn(struct pmu *pmu)
 		return -EAGAIN;
 
 	cpuc->group_flag &= ~PERF_EVENT_TXN;
+	perf_enable();
 	return 0;
 }
 
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 2c89264ee791..846070ce49c3 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -969,10 +969,11 @@ static int x86_pmu_enable(struct perf_event *event)
 
 	hwc = &event->hw;
 
+	perf_disable();
 	n0 = cpuc->n_events;
-	n = collect_events(cpuc, event, false);
-	if (n < 0)
-		return n;
+	ret = n = collect_events(cpuc, event, false);
+	if (ret < 0)
+		goto out;
 
 	/*
 	 * If group events scheduling transaction was started,
@@ -980,23 +981,26 @@ static int x86_pmu_enable(struct perf_event *event)
 	 * at commit time(->commit_txn) as a whole
 	 */
 	if (cpuc->group_flag & PERF_EVENT_TXN)
-		goto out;
+		goto done_collect;
 
 	ret = x86_pmu.schedule_events(cpuc, n, assign);
 	if (ret)
-		return ret;
+		goto out;
 	/*
 	 * copy new assignment, now we know it is possible
 	 * will be used by hw_perf_enable()
 	 */
 	memcpy(cpuc->assign, assign, n*sizeof(int));
 
-out:
+done_collect:
 	cpuc->n_events = n;
 	cpuc->n_added += n - n0;
 	cpuc->n_txn += n - n0;
 
-	return 0;
+	ret = 0;
+out:
+	perf_enable();
+	return ret;
 }
 
 static int x86_pmu_start(struct perf_event *event)
@@ -1432,6 +1436,7 @@ static void x86_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
+	perf_disable();
 	cpuc->group_flag |= PERF_EVENT_TXN;
 	cpuc->n_txn = 0;
 }
@@ -1451,6 +1456,7 @@ static void x86_pmu_cancel_txn(struct pmu *pmu)
 	 */
 	cpuc->n_added -= cpuc->n_txn;
 	cpuc->n_events -= cpuc->n_txn;
+	perf_enable();
 }
 
 /*
@@ -1480,7 +1486,7 @@ static int x86_pmu_commit_txn(struct pmu *pmu)
 	memcpy(cpuc->assign, assign, n*sizeof(int));
 
 	cpuc->group_flag &= ~PERF_EVENT_TXN;
-
+	perf_enable();
 	return 0;
 }
 
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index ab72f56eb372..243286a8ded7 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -564,26 +564,26 @@ struct pmu {
 	struct list_head		entry;
 
 	/*
-	 * Should return -ENOENT when the @event doesn't match this pmu
+	 * Should return -ENOENT when the @event doesn't match this PMU.
 	 */
 	int (*event_init)		(struct perf_event *event);
 
-	int (*enable)			(struct perf_event *event);
+	int  (*enable)			(struct perf_event *event);
 	void (*disable)			(struct perf_event *event);
-	int (*start)			(struct perf_event *event);
+	int  (*start)			(struct perf_event *event);
 	void (*stop)			(struct perf_event *event);
 	void (*read)			(struct perf_event *event);
 	void (*unthrottle)		(struct perf_event *event);
 
 	/*
-	 * Group events scheduling is treated as a transaction, add group
-	 * events as a whole and perform one schedulability test. If the test
-	 * fails, roll back the whole group
+	 * Group events scheduling is treated as a transaction, add
+	 * group events as a whole and perform one schedulability test.
+	 * If the test fails, roll back the whole group
 	 */
 
 	/*
-	 * Start the transaction, after this ->enable() doesn't need
-	 * to do schedulability tests.
+	 * Start the transaction, after this ->enable() doesn't need to
+	 * do schedulability tests.
 	 */
 	void (*start_txn)	(struct pmu *pmu);
 	/*
@@ -594,8 +594,8 @@ struct pmu {
 	 */
 	int  (*commit_txn)	(struct pmu *pmu);
 	/*
-	 * Will cancel the transaction, assumes ->disable() is called for
-	 * each successfull ->enable() during the transaction.
+	 * Will cancel the transaction, assumes ->disable() is called
+	 * for each successfull ->enable() during the transaction.
 	 */
 	void (*cancel_txn)	(struct pmu *pmu);
 };
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 149ca18371b7..9a98ce953561 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -478,11 +478,6 @@ static void __perf_event_remove_from_context(void *info)
 		return;
 
 	raw_spin_lock(&ctx->lock);
-	/*
-	 * Protect the list operation against NMI by disabling the
-	 * events on a global level.
-	 */
-	perf_disable();
 
 	event_sched_out(event, cpuctx, ctx);
 
@@ -498,7 +493,6 @@ static void __perf_event_remove_from_context(void *info)
 			    perf_max_events - perf_reserved_percpu);
 	}
 
-	perf_enable();
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -803,12 +797,6 @@ static void __perf_install_in_context(void *info)
 	ctx->is_active = 1;
 	update_context_time(ctx);
 
-	/*
-	 * Protect the list operation against NMI by disabling the
-	 * events on a global level. NOP for non NMI based events.
-	 */
-	perf_disable();
-
 	add_event_to_ctx(event, ctx);
 
 	if (event->cpu != -1 && event->cpu != smp_processor_id())
@@ -850,8 +838,6 @@ static void __perf_install_in_context(void *info)
 		cpuctx->max_pertask--;
 
 unlock:
-	perf_enable();
-
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -972,12 +958,10 @@ static void __perf_event_enable(void *info)
 	if (!group_can_go_on(event, cpuctx, 1)) {
 		err = -EEXIST;
 	} else {
-		perf_disable();
 		if (event == leader)
 			err = group_sched_in(event, cpuctx, ctx);
 		else
 			err = event_sched_in(event, cpuctx, ctx);
-		perf_enable();
 	}
 
 	if (err) {
@@ -1090,9 +1074,8 @@ static void ctx_sched_out(struct perf_event_context *ctx,
 		goto out;
 	update_context_time(ctx);
 
-	perf_disable();
 	if (!ctx->nr_active)
-		goto out_enable;
+		goto out;
 
 	if (event_type & EVENT_PINNED) {
 		list_for_each_entry(event, &ctx->pinned_groups, group_entry)
@@ -1103,9 +1086,6 @@ static void ctx_sched_out(struct perf_event_context *ctx,
 		list_for_each_entry(event, &ctx->flexible_groups, group_entry)
 			group_sched_out(event, cpuctx, ctx);
 	}
-
- out_enable:
-	perf_enable();
 out:
 	raw_spin_unlock(&ctx->lock);
 }
@@ -1364,8 +1344,6 @@ ctx_sched_in(struct perf_event_context *ctx,
 
 	ctx->timestamp = perf_clock();
 
-	perf_disable();
-
 	/*
 	 * First go through the list and put on any pinned groups
 	 * in order to give them the best chance of going on.
@@ -1377,7 +1355,6 @@ ctx_sched_in(struct perf_event_context *ctx,
 	if (event_type & EVENT_FLEXIBLE)
 		ctx_flexible_sched_in(ctx, cpuctx);
 
-	perf_enable();
 out:
 	raw_spin_unlock(&ctx->lock);
 }
@@ -1425,8 +1402,6 @@ void perf_event_task_sched_in(struct task_struct *task)
 	if (cpuctx->task_ctx == ctx)
 		return;
 
-	perf_disable();
-
 	/*
 	 * We want to keep the following priority order:
 	 * cpu pinned (that don't need to move), task pinned,
@@ -1439,8 +1414,6 @@ void perf_event_task_sched_in(struct task_struct *task)
 	ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE);
 
 	cpuctx->task_ctx = ctx;
-
-	perf_enable();
 }
 
 #define MAX_INTERRUPTS (~0ULL)
@@ -1555,11 +1528,9 @@ static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count)
 	hwc->sample_period = sample_period;
 
 	if (local64_read(&hwc->period_left) > 8*sample_period) {
-		perf_disable();
 		perf_event_stop(event);
 		local64_set(&hwc->period_left, 0);
 		perf_event_start(event);
-		perf_enable();
 	}
 }
 
@@ -1588,15 +1559,12 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
 		 */
 		if (interrupts == MAX_INTERRUPTS) {
 			perf_log_throttle(event, 1);
-			perf_disable();
 			event->pmu->unthrottle(event);
-			perf_enable();
 		}
 
 		if (!event->attr.freq || !event->attr.sample_freq)
 			continue;
 
-		perf_disable();
 		event->pmu->read(event);
 		now = local64_read(&event->count);
 		delta = now - hwc->freq_count_stamp;
@@ -1604,7 +1572,6 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
 
 		if (delta > 0)
 			perf_adjust_period(event, TICK_NSEC, delta);
-		perf_enable();
 	}
 	raw_spin_unlock(&ctx->lock);
 }
@@ -1647,7 +1614,6 @@ void perf_event_task_tick(struct task_struct *curr)
 	if (!rotate)
 		return;
 
-	perf_disable();
 	cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
 		task_ctx_sched_out(ctx, EVENT_FLEXIBLE);
@@ -1659,7 +1625,6 @@ void perf_event_task_tick(struct task_struct *curr)
 	cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
 		task_ctx_sched_in(curr, EVENT_FLEXIBLE);
-	perf_enable();
 }
 
 static int event_enable_on_exec(struct perf_event *event,
-- 
cgit v1.2.3


From 33696fc0d141bbbcb12f75b69608ea83282e3117 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 14 Jun 2010 08:49:00 +0200
Subject: perf: Per PMU disable

Changes perf_disable() into perf_pmu_disable().

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/alpha/kernel/perf_event.c           | 30 +++++++++++++------------
 arch/arm/kernel/perf_event.c             | 28 +++++++++++------------
 arch/powerpc/kernel/perf_event.c         | 24 +++++++++++---------
 arch/powerpc/kernel/perf_event_fsl_emb.c | 18 ++++++++-------
 arch/sh/kernel/perf_event.c              | 38 +++++++++++++++++---------------
 arch/sparc/kernel/perf_event.c           | 20 +++++++++--------
 arch/x86/kernel/cpu/perf_event.c         | 16 ++++++++------
 include/linux/perf_event.h               | 13 ++++++-----
 kernel/perf_event.c                      | 31 ++++++++++++++++----------
 9 files changed, 119 insertions(+), 99 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c
index 19660b5c298f..3e260731f8e6 100644
--- a/arch/alpha/kernel/perf_event.c
+++ b/arch/alpha/kernel/perf_event.c
@@ -435,7 +435,7 @@ static int alpha_pmu_enable(struct perf_event *event)
 	 * nevertheless we disable the PMCs first to enable a potential
 	 * final PMI to occur before we disable interrupts.
 	 */
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 	local_irq_save(flags);
 
 	/* Default to error to be returned */
@@ -456,7 +456,7 @@ static int alpha_pmu_enable(struct perf_event *event)
 	}
 
 	local_irq_restore(flags);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 
 	return ret;
 }
@@ -474,7 +474,7 @@ static void alpha_pmu_disable(struct perf_event *event)
 	unsigned long flags;
 	int j;
 
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 	local_irq_save(flags);
 
 	for (j = 0; j < cpuc->n_events; j++) {
@@ -502,7 +502,7 @@ static void alpha_pmu_disable(struct perf_event *event)
 	}
 
 	local_irq_restore(flags);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 }
 
 
@@ -668,18 +668,10 @@ static int alpha_pmu_event_init(struct perf_event *event)
 	return err;
 }
 
-static struct pmu pmu = {
-	.event_init	= alpha_pmu_event_init,
-	.enable		= alpha_pmu_enable,
-	.disable	= alpha_pmu_disable,
-	.read		= alpha_pmu_read,
-	.unthrottle	= alpha_pmu_unthrottle,
-};
-
 /*
  * Main entry point - enable HW performance counters.
  */
-void hw_perf_enable(void)
+static void alpha_pmu_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -705,7 +697,7 @@ void hw_perf_enable(void)
  * Main entry point - disable HW performance counters.
  */
 
-void hw_perf_disable(void)
+static void alpha_pmu_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -718,6 +710,16 @@ void hw_perf_disable(void)
 	wrperfmon(PERFMON_CMD_DISABLE, cpuc->idx_mask);
 }
 
+static struct pmu pmu = {
+	.pmu_enable	= alpha_pmu_pmu_enable,
+	.pmu_disable	= alpha_pmu_pmu_disable,
+	.event_init	= alpha_pmu_event_init,
+	.enable		= alpha_pmu_enable,
+	.disable	= alpha_pmu_disable,
+	.read		= alpha_pmu_read,
+	.unthrottle	= alpha_pmu_unthrottle,
+};
+
 
 /*
  * Main entry point - don't know when this is called but it
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index afc92c580d18..3343f3f4b973 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -277,7 +277,7 @@ armpmu_enable(struct perf_event *event)
 	int idx;
 	int err = 0;
 
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	/* If we don't have a space for the counter then finish early. */
 	idx = armpmu->get_event_idx(cpuc, hwc);
@@ -305,7 +305,7 @@ armpmu_enable(struct perf_event *event)
 	perf_event_update_userpage(event);
 
 out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	return err;
 }
 
@@ -534,16 +534,7 @@ static int armpmu_event_init(struct perf_event *event)
 	return err;
 }
 
-static struct pmu pmu = {
-	.event_init = armpmu_event_init,
-	.enable	    = armpmu_enable,
-	.disable    = armpmu_disable,
-	.unthrottle = armpmu_unthrottle,
-	.read	    = armpmu_read,
-};
-
-void
-hw_perf_enable(void)
+static void armpmu_pmu_enable(struct pmu *pmu)
 {
 	/* Enable all of the perf events on hardware. */
 	int idx;
@@ -564,13 +555,22 @@ hw_perf_enable(void)
 	armpmu->start();
 }
 
-void
-hw_perf_disable(void)
+static void armpmu_pmu_disable(struct pmu *pmu)
 {
 	if (armpmu)
 		armpmu->stop();
 }
 
+static struct pmu pmu = {
+	.pmu_enable = armpmu_pmu_enable,
+	.pmu_disable= armpmu_pmu_disable,
+	.event_init = armpmu_event_init,
+	.enable	    = armpmu_enable,
+	.disable    = armpmu_disable,
+	.unthrottle = armpmu_unthrottle,
+	.read	    = armpmu_read,
+};
+
 /*
  * ARMv6 Performance counter handling code.
  *
diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c
index c1408821dbc2..deb84bbcb0e6 100644
--- a/arch/powerpc/kernel/perf_event.c
+++ b/arch/powerpc/kernel/perf_event.c
@@ -517,7 +517,7 @@ static void write_mmcr0(struct cpu_hw_events *cpuhw, unsigned long mmcr0)
  * Disable all events to prevent PMU interrupts and to allow
  * events to be added or removed.
  */
-void hw_perf_disable(void)
+static void power_pmu_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -565,7 +565,7 @@ void hw_perf_disable(void)
  * If we were previously disabled and events were added, then
  * put the new config on the PMU.
  */
-void hw_perf_enable(void)
+static void power_pmu_pmu_enable(struct pmu *pmu)
 {
 	struct perf_event *event;
 	struct cpu_hw_events *cpuhw;
@@ -735,7 +735,7 @@ static int power_pmu_enable(struct perf_event *event)
 	int ret = -EAGAIN;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	/*
 	 * Add the event to the list (if there is room)
@@ -769,7 +769,7 @@ nocheck:
 
 	ret = 0;
  out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 	return ret;
 }
@@ -784,7 +784,7 @@ static void power_pmu_disable(struct perf_event *event)
 	unsigned long flags;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	power_pmu_read(event);
 
@@ -821,7 +821,7 @@ static void power_pmu_disable(struct perf_event *event)
 		cpuhw->mmcr[0] &= ~(MMCR0_PMXE | MMCR0_FCECE);
 	}
 
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
@@ -837,7 +837,7 @@ static void power_pmu_unthrottle(struct perf_event *event)
 	if (!event->hw.idx || !event->hw.sample_period)
 		return;
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 	power_pmu_read(event);
 	left = event->hw.sample_period;
 	event->hw.last_period = left;
@@ -848,7 +848,7 @@ static void power_pmu_unthrottle(struct perf_event *event)
 	local64_set(&event->hw.prev_count, val);
 	local64_set(&event->hw.period_left, left);
 	perf_event_update_userpage(event);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
@@ -861,7 +861,7 @@ void power_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
-	perf_disable();
+	perf_pmu_disable(pmu);
 	cpuhw->group_flag |= PERF_EVENT_TXN;
 	cpuhw->n_txn_start = cpuhw->n_events;
 }
@@ -876,7 +876,7 @@ void power_pmu_cancel_txn(struct pmu *pmu)
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
-	perf_enable();
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -903,7 +903,7 @@ int power_pmu_commit_txn(struct pmu *pmu)
 		cpuhw->event[i]->hw.config = cpuhw->events[i];
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
-	perf_enable();
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
@@ -1131,6 +1131,8 @@ static int power_pmu_event_init(struct perf_event *event)
 }
 
 struct pmu power_pmu = {
+	.pmu_enable	= power_pmu_pmu_enable,
+	.pmu_disable	= power_pmu_pmu_disable,
 	.event_init	= power_pmu_event_init,
 	.enable		= power_pmu_enable,
 	.disable	= power_pmu_disable,
diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c
index 9bc84a7fd901..84b1974c628f 100644
--- a/arch/powerpc/kernel/perf_event_fsl_emb.c
+++ b/arch/powerpc/kernel/perf_event_fsl_emb.c
@@ -177,7 +177,7 @@ static void fsl_emb_pmu_read(struct perf_event *event)
  * Disable all events to prevent PMU interrupts and to allow
  * events to be added or removed.
  */
-void hw_perf_disable(void)
+static void fsl_emb_pmu_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -216,7 +216,7 @@ void hw_perf_disable(void)
  * If we were previously disabled and events were added, then
  * put the new config on the PMU.
  */
-void hw_perf_enable(void)
+static void fsl_emb_pmu_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -271,7 +271,7 @@ static int fsl_emb_pmu_enable(struct perf_event *event)
 	u64 val;
 	int i;
 
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 	cpuhw = &get_cpu_var(cpu_hw_events);
 
 	if (event->hw.config & FSL_EMB_EVENT_RESTRICTED)
@@ -311,7 +311,7 @@ static int fsl_emb_pmu_enable(struct perf_event *event)
 	ret = 0;
  out:
 	put_cpu_var(cpu_hw_events);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	return ret;
 }
 
@@ -321,7 +321,7 @@ static void fsl_emb_pmu_disable(struct perf_event *event)
 	struct cpu_hw_events *cpuhw;
 	int i = event->hw.idx;
 
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 	if (i < 0)
 		goto out;
 
@@ -349,7 +349,7 @@ static void fsl_emb_pmu_disable(struct perf_event *event)
 	cpuhw->n_events--;
 
  out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	put_cpu_var(cpu_hw_events);
 }
 
@@ -367,7 +367,7 @@ static void fsl_emb_pmu_unthrottle(struct perf_event *event)
 	if (event->hw.idx < 0 || !event->hw.sample_period)
 		return;
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 	fsl_emb_pmu_read(event);
 	left = event->hw.sample_period;
 	event->hw.last_period = left;
@@ -378,7 +378,7 @@ static void fsl_emb_pmu_unthrottle(struct perf_event *event)
 	local64_set(&event->hw.prev_count, val);
 	local64_set(&event->hw.period_left, left);
 	perf_event_update_userpage(event);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
@@ -524,6 +524,8 @@ static int fsl_emb_pmu_event_init(struct perf_event *event)
 }
 
 static struct pmu fsl_emb_pmu = {
+	.pmu_enable	= fsl_emb_pmu_pmu_enable,
+	.pmu_disable	= fsl_emb_pmu_pmu_disable,
 	.event_init	= fsl_emb_pmu_event_init,
 	.enable		= fsl_emb_pmu_enable,
 	.disable	= fsl_emb_pmu_disable,
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index d042989ceb45..4bbe19058a58 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -232,7 +232,7 @@ static int sh_pmu_enable(struct perf_event *event)
 	int idx = hwc->idx;
 	int ret = -EAGAIN;
 
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	if (test_and_set_bit(idx, cpuc->used_mask)) {
 		idx = find_first_zero_bit(cpuc->used_mask, sh_pmu->num_events);
@@ -253,7 +253,7 @@ static int sh_pmu_enable(struct perf_event *event)
 	perf_event_update_userpage(event);
 	ret = 0;
 out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	return ret;
 }
 
@@ -285,7 +285,25 @@ static int sh_pmu_event_init(struct perf_event *event)
 	return err;
 }
 
+static void sh_pmu_pmu_enable(struct pmu *pmu)
+{
+	if (!sh_pmu_initialized())
+		return;
+
+	sh_pmu->enable_all();
+}
+
+static void sh_pmu_pmu_disable(struct pmu *pmu)
+{
+	if (!sh_pmu_initialized())
+		return;
+
+	sh_pmu->disable_all();
+}
+
 static struct pmu pmu = {
+	.pmu_enable	= sh_pmu_pmu_enable,
+	.pmu_disable	= sh_pmu_pmu_disable,
 	.event_init	= sh_pmu_event_init,
 	.enable		= sh_pmu_enable,
 	.disable	= sh_pmu_disable,
@@ -316,22 +334,6 @@ sh_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
 	return NOTIFY_OK;
 }
 
-void hw_perf_enable(void)
-{
-	if (!sh_pmu_initialized())
-		return;
-
-	sh_pmu->enable_all();
-}
-
-void hw_perf_disable(void)
-{
-	if (!sh_pmu_initialized())
-		return;
-
-	sh_pmu->disable_all();
-}
-
 int __cpuinit register_sh_pmu(struct sh_pmu *pmu)
 {
 	if (sh_pmu)
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index d0131deeeaf6..37cae676536c 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -664,7 +664,7 @@ out:
 	return pcr;
 }
 
-void hw_perf_enable(void)
+static void sparc_pmu_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 pcr;
@@ -691,7 +691,7 @@ void hw_perf_enable(void)
 	pcr_ops->write(cpuc->pcr);
 }
 
-void hw_perf_disable(void)
+static void sparc_pmu_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 val;
@@ -718,7 +718,7 @@ static void sparc_pmu_disable(struct perf_event *event)
 	int i;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	for (i = 0; i < cpuc->n_events; i++) {
 		if (event == cpuc->event[i]) {
@@ -748,7 +748,7 @@ static void sparc_pmu_disable(struct perf_event *event)
 		}
 	}
 
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
@@ -991,7 +991,7 @@ static int sparc_pmu_enable(struct perf_event *event)
 	unsigned long flags;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	n0 = cpuc->n_events;
 	if (n0 >= perf_max_events)
@@ -1020,7 +1020,7 @@ nocheck:
 
 	ret = 0;
 out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 	return ret;
 }
@@ -1113,7 +1113,7 @@ static void sparc_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
-	perf_disable();
+	perf_pmu_disable(pmu);
 	cpuhw->group_flag |= PERF_EVENT_TXN;
 }
 
@@ -1127,7 +1127,7 @@ static void sparc_pmu_cancel_txn(struct pmu *pmu)
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
-	perf_enable();
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -1151,11 +1151,13 @@ static int sparc_pmu_commit_txn(struct pmu *pmu)
 		return -EAGAIN;
 
 	cpuc->group_flag &= ~PERF_EVENT_TXN;
-	perf_enable();
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
 static struct pmu pmu = {
+	.pmu_enable	= sparc_pmu_pmu_enable,
+	.pmu_disable	= sparc_pmu_pmu_disable,
 	.event_init	= sparc_pmu_event_init,
 	.enable		= sparc_pmu_enable,
 	.disable	= sparc_pmu_disable,
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 846070ce49c3..79705ac45019 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -583,7 +583,7 @@ static void x86_pmu_disable_all(void)
 	}
 }
 
-void hw_perf_disable(void)
+static void x86_pmu_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -803,7 +803,7 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc,
 static int x86_pmu_start(struct perf_event *event);
 static void x86_pmu_stop(struct perf_event *event);
 
-void hw_perf_enable(void)
+static void x86_pmu_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct perf_event *event;
@@ -969,7 +969,7 @@ static int x86_pmu_enable(struct perf_event *event)
 
 	hwc = &event->hw;
 
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 	n0 = cpuc->n_events;
 	ret = n = collect_events(cpuc, event, false);
 	if (ret < 0)
@@ -999,7 +999,7 @@ done_collect:
 
 	ret = 0;
 out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	return ret;
 }
 
@@ -1436,7 +1436,7 @@ static void x86_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
-	perf_disable();
+	perf_pmu_disable(pmu);
 	cpuc->group_flag |= PERF_EVENT_TXN;
 	cpuc->n_txn = 0;
 }
@@ -1456,7 +1456,7 @@ static void x86_pmu_cancel_txn(struct pmu *pmu)
 	 */
 	cpuc->n_added -= cpuc->n_txn;
 	cpuc->n_events -= cpuc->n_txn;
-	perf_enable();
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -1486,7 +1486,7 @@ static int x86_pmu_commit_txn(struct pmu *pmu)
 	memcpy(cpuc->assign, assign, n*sizeof(int));
 
 	cpuc->group_flag &= ~PERF_EVENT_TXN;
-	perf_enable();
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
@@ -1605,6 +1605,8 @@ int x86_pmu_event_init(struct perf_event *event)
 }
 
 static struct pmu pmu = {
+	.pmu_enable	= x86_pmu_pmu_enable,
+	.pmu_disable	= x86_pmu_pmu_disable,
 	.event_init	= x86_pmu_event_init,
 	.enable		= x86_pmu_enable,
 	.disable	= x86_pmu_disable,
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 243286a8ded7..6abf103fb7f8 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -563,6 +563,11 @@ struct perf_event;
 struct pmu {
 	struct list_head		entry;
 
+	int				*pmu_disable_count;
+
+	void (*pmu_enable)		(struct pmu *pmu);
+	void (*pmu_disable)		(struct pmu *pmu);
+
 	/*
 	 * Should return -ENOENT when the @event doesn't match this PMU.
 	 */
@@ -868,10 +873,8 @@ extern void perf_event_free_task(struct task_struct *task);
 extern void set_perf_event_pending(void);
 extern void perf_event_do_pending(void);
 extern void perf_event_print_debug(void);
-extern void __perf_disable(void);
-extern bool __perf_enable(void);
-extern void perf_disable(void);
-extern void perf_enable(void);
+extern void perf_pmu_disable(struct pmu *pmu);
+extern void perf_pmu_enable(struct pmu *pmu);
 extern int perf_event_task_disable(void);
 extern int perf_event_task_enable(void);
 extern void perf_event_update_userpage(struct perf_event *event);
@@ -1056,8 +1059,6 @@ static inline void perf_event_exit_task(struct task_struct *child)	{ }
 static inline void perf_event_free_task(struct task_struct *task)	{ }
 static inline void perf_event_do_pending(void)				{ }
 static inline void perf_event_print_debug(void)				{ }
-static inline void perf_disable(void)					{ }
-static inline void perf_enable(void)					{ }
 static inline int perf_event_task_disable(void)				{ return -EINVAL; }
 static inline int perf_event_task_enable(void)				{ return -EINVAL; }
 
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 9a98ce953561..5ed0c06765bb 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -71,23 +71,20 @@ static atomic64_t perf_event_id;
  */
 static DEFINE_SPINLOCK(perf_resource_lock);
 
-void __weak hw_perf_disable(void)		{ barrier(); }
-void __weak hw_perf_enable(void)		{ barrier(); }
-
 void __weak perf_event_print_debug(void)	{ }
 
-static DEFINE_PER_CPU(int, perf_disable_count);
-
-void perf_disable(void)
+void perf_pmu_disable(struct pmu *pmu)
 {
-	if (!__get_cpu_var(perf_disable_count)++)
-		hw_perf_disable();
+	int *count = this_cpu_ptr(pmu->pmu_disable_count);
+	if (!(*count)++)
+		pmu->pmu_disable(pmu);
 }
 
-void perf_enable(void)
+void perf_pmu_enable(struct pmu *pmu)
 {
-	if (!--__get_cpu_var(perf_disable_count))
-		hw_perf_enable();
+	int *count = this_cpu_ptr(pmu->pmu_disable_count);
+	if (!--(*count))
+		pmu->pmu_enable(pmu);
 }
 
 static void get_ctx(struct perf_event_context *ctx)
@@ -4970,11 +4967,19 @@ static struct srcu_struct pmus_srcu;
 
 int perf_pmu_register(struct pmu *pmu)
 {
+	int ret;
+
 	mutex_lock(&pmus_lock);
+	ret = -ENOMEM;
+	pmu->pmu_disable_count = alloc_percpu(int);
+	if (!pmu->pmu_disable_count)
+		goto unlock;
 	list_add_rcu(&pmu->entry, &pmus);
+	ret = 0;
+unlock:
 	mutex_unlock(&pmus_lock);
 
-	return 0;
+	return ret;
 }
 
 void perf_pmu_unregister(struct pmu *pmu)
@@ -4984,6 +4989,8 @@ void perf_pmu_unregister(struct pmu *pmu)
 	mutex_unlock(&pmus_lock);
 
 	synchronize_srcu(&pmus_srcu);
+
+	free_percpu(pmu->pmu_disable_count);
 }
 
 struct pmu *perf_init_event(struct perf_event *event)
-- 
cgit v1.2.3


From ad5133b7030d04ce7701aa7cbe98f561347c79c2 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Tue, 15 Jun 2010 12:22:39 +0200
Subject: perf: Default PMU ops

Provide default implementations for the pmu txn methods, this
allows us to remove some conditional code.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h | 10 ++++----
 kernel/perf_event.c        | 64 +++++++++++++++++++++++++++++++++++++---------
 2 files changed, 57 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 6abf103fb7f8..bf85733597ec 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -565,8 +565,8 @@ struct pmu {
 
 	int				*pmu_disable_count;
 
-	void (*pmu_enable)		(struct pmu *pmu);
-	void (*pmu_disable)		(struct pmu *pmu);
+	void (*pmu_enable)		(struct pmu *pmu); /* optional */
+	void (*pmu_disable)		(struct pmu *pmu); /* optional */
 
 	/*
 	 * Should return -ENOENT when the @event doesn't match this PMU.
@@ -590,19 +590,19 @@ struct pmu {
 	 * Start the transaction, after this ->enable() doesn't need to
 	 * do schedulability tests.
 	 */
-	void (*start_txn)	(struct pmu *pmu);
+	void (*start_txn)	(struct pmu *pmu); /* optional */
 	/*
 	 * If ->start_txn() disabled the ->enable() schedulability test
 	 * then ->commit_txn() is required to perform one. On success
 	 * the transaction is closed. On error the transaction is kept
 	 * open until ->cancel_txn() is called.
 	 */
-	int  (*commit_txn)	(struct pmu *pmu);
+	int  (*commit_txn)	(struct pmu *pmu); /* optional */
 	/*
 	 * Will cancel the transaction, assumes ->disable() is called
 	 * for each successfull ->enable() during the transaction.
 	 */
-	void (*cancel_txn)	(struct pmu *pmu);
+	void (*cancel_txn)	(struct pmu *pmu); /* optional */
 };
 
 /**
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 5ed0c06765bb..8ef4ba3bcb1f 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -674,21 +674,14 @@ group_sched_in(struct perf_event *group_event,
 {
 	struct perf_event *event, *partial_group = NULL;
 	struct pmu *pmu = group_event->pmu;
-	bool txn = false;
 
 	if (group_event->state == PERF_EVENT_STATE_OFF)
 		return 0;
 
-	/* Check if group transaction availabe */
-	if (pmu->start_txn)
-		txn = true;
-
-	if (txn)
-		pmu->start_txn(pmu);
+	pmu->start_txn(pmu);
 
 	if (event_sched_in(group_event, cpuctx, ctx)) {
-		if (txn)
-			pmu->cancel_txn(pmu);
+		pmu->cancel_txn(pmu);
 		return -EAGAIN;
 	}
 
@@ -702,7 +695,7 @@ group_sched_in(struct perf_event *group_event,
 		}
 	}
 
-	if (!txn || !pmu->commit_txn(pmu))
+	if (!pmu->commit_txn(pmu))
 		return 0;
 
 group_error:
@@ -717,8 +710,7 @@ group_error:
 	}
 	event_sched_out(group_event, cpuctx, ctx);
 
-	if (txn)
-		pmu->cancel_txn(pmu);
+	pmu->cancel_txn(pmu);
 
 	return -EAGAIN;
 }
@@ -4965,6 +4957,31 @@ static LIST_HEAD(pmus);
 static DEFINE_MUTEX(pmus_lock);
 static struct srcu_struct pmus_srcu;
 
+static void perf_pmu_nop_void(struct pmu *pmu)
+{
+}
+
+static int perf_pmu_nop_int(struct pmu *pmu)
+{
+	return 0;
+}
+
+static void perf_pmu_start_txn(struct pmu *pmu)
+{
+	perf_pmu_disable(pmu);
+}
+
+static int perf_pmu_commit_txn(struct pmu *pmu)
+{
+	perf_pmu_enable(pmu);
+	return 0;
+}
+
+static void perf_pmu_cancel_txn(struct pmu *pmu)
+{
+	perf_pmu_enable(pmu);
+}
+
 int perf_pmu_register(struct pmu *pmu)
 {
 	int ret;
@@ -4974,6 +4991,29 @@ int perf_pmu_register(struct pmu *pmu)
 	pmu->pmu_disable_count = alloc_percpu(int);
 	if (!pmu->pmu_disable_count)
 		goto unlock;
+
+	if (!pmu->start_txn) {
+		if (pmu->pmu_enable) {
+			/*
+			 * If we have pmu_enable/pmu_disable calls, install
+			 * transaction stubs that use that to try and batch
+			 * hardware accesses.
+			 */
+			pmu->start_txn  = perf_pmu_start_txn;
+			pmu->commit_txn = perf_pmu_commit_txn;
+			pmu->cancel_txn = perf_pmu_cancel_txn;
+		} else {
+			pmu->start_txn  = perf_pmu_nop_void;
+			pmu->commit_txn = perf_pmu_nop_int;
+			pmu->cancel_txn = perf_pmu_nop_void;
+		}
+	}
+
+	if (!pmu->pmu_enable) {
+		pmu->pmu_enable  = perf_pmu_nop_void;
+		pmu->pmu_disable = perf_pmu_nop_void;
+	}
+
 	list_add_rcu(&pmu->entry, &pmus);
 	ret = 0;
 unlock:
-- 
cgit v1.2.3


From fa407f35e0298d841e4088f95a7f9cf6e725c6d5 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 24 Jun 2010 12:35:12 +0200
Subject: perf: Shrink hw_perf_event

Use hw_perf_event::period_left instead of hw_perf_event::remaining
and win back 8 bytes.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |  1 -
 kernel/perf_event.c        | 13 ++++++-------
 2 files changed, 6 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index bf85733597ec..8cafa15af60d 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -529,7 +529,6 @@ struct hw_perf_event {
 			int		last_cpu;
 		};
 		struct { /* software */
-			s64		remaining;
 			struct hrtimer	hrtimer;
 		};
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 8ef4ba3bcb1f..1a6cdbf0d091 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -4800,14 +4800,13 @@ static void perf_swevent_start_hrtimer(struct perf_event *event)
 	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	hwc->hrtimer.function = perf_swevent_hrtimer;
 	if (hwc->sample_period) {
-		u64 period;
+		s64 period = local64_read(&hwc->period_left);
 
-		if (hwc->remaining) {
-			if (hwc->remaining < 0)
+		if (period) {
+			if (period < 0)
 				period = 10000;
-			else
-				period = hwc->remaining;
-			hwc->remaining = 0;
+
+			local64_set(&hwc->period_left, 0);
 		} else {
 			period = max_t(u64, 10000, hwc->sample_period);
 		}
@@ -4823,7 +4822,7 @@ static void perf_swevent_cancel_hrtimer(struct perf_event *event)
 
 	if (hwc->sample_period) {
 		ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer);
-		hwc->remaining = ktime_to_ns(remaining);
+		local64_set(&hwc->period_left, ktime_to_ns(remaining));
 
 		hrtimer_cancel(&hwc->hrtimer);
 	}
-- 
cgit v1.2.3


From a4eaf7f14675cb512d69f0c928055e73d0c6d252 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Wed, 16 Jun 2010 14:37:10 +0200
Subject: perf: Rework the PMU methods

Replace pmu::{enable,disable,start,stop,unthrottle} with
pmu::{add,del,start,stop}, all of which take a flags argument.

The new interface extends the capability to stop a counter while
keeping it scheduled on the PMU. We replace the throttled state with
the generic stopped state.

This also allows us to efficiently stop/start counters over certain
code paths (like IRQ handlers).

It also allows scheduling a counter without it starting, allowing for
a generic frozen state (useful for rotating stopped counters).

The stopped state is implemented in two different ways, depending on
how the architecture implemented the throttled state:

 1) We disable the counter:
    a) the pmu has per-counter enable bits, we flip that
    b) we program a NOP event, preserving the counter state

 2) We store the counter state and ignore all read/overflow events

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
Cc: Deng-Cheng Zhu <dengcheng.zhu@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Michael Cree <mcree@orcon.net.nz>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/alpha/kernel/perf_event.c            |  71 +++++++++++----
 arch/arm/kernel/perf_event.c              |  96 ++++++++++++--------
 arch/powerpc/kernel/perf_event.c          | 105 ++++++++++++++--------
 arch/powerpc/kernel/perf_event_fsl_emb.c  | 107 ++++++++++++++---------
 arch/sh/kernel/perf_event.c               |  75 +++++++++++-----
 arch/sparc/kernel/perf_event.c            | 109 ++++++++++++++---------
 arch/x86/kernel/cpu/perf_event.c          | 106 ++++++++++++----------
 arch/x86/kernel/cpu/perf_event_intel.c    |   2 +-
 arch/x86/kernel/cpu/perf_event_intel_ds.c |   2 +-
 include/linux/ftrace_event.h              |   4 +-
 include/linux/perf_event.h                |  54 +++++++++---
 kernel/hw_breakpoint.c                    |  29 ++++++-
 kernel/perf_event.c                       | 140 +++++++++++++++---------------
 kernel/trace/trace_event_perf.c           |   7 +-
 14 files changed, 576 insertions(+), 331 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c
index 3e260731f8e6..380ef02d557a 100644
--- a/arch/alpha/kernel/perf_event.c
+++ b/arch/alpha/kernel/perf_event.c
@@ -307,7 +307,7 @@ again:
 			     new_raw_count) != prev_raw_count)
 		goto again;
 
-	delta = (new_raw_count  - (prev_raw_count & alpha_pmu->pmc_count_mask[idx])) + ovf;
+	delta = (new_raw_count - (prev_raw_count & alpha_pmu->pmc_count_mask[idx])) + ovf;
 
 	/* It is possible on very rare occasions that the PMC has overflowed
 	 * but the interrupt is yet to come.  Detect and fix this situation.
@@ -402,14 +402,13 @@ static void maybe_change_configuration(struct cpu_hw_events *cpuc)
 		struct hw_perf_event *hwc = &pe->hw;
 		int idx = hwc->idx;
 
-		if (cpuc->current_idx[j] != PMC_NO_INDEX) {
-			cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
-			continue;
+		if (cpuc->current_idx[j] == PMC_NO_INDEX) {
+			alpha_perf_event_set_period(pe, hwc, idx);
+			cpuc->current_idx[j] = idx;
 		}
 
-		alpha_perf_event_set_period(pe, hwc, idx);
-		cpuc->current_idx[j] = idx;
-		cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
+		if (!(hwc->state & PERF_HES_STOPPED))
+			cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
 	}
 	cpuc->config = cpuc->event[0]->hw.config_base;
 }
@@ -420,7 +419,7 @@ static void maybe_change_configuration(struct cpu_hw_events *cpuc)
  *  - this function is called from outside this module via the pmu struct
  *    returned from perf event initialisation.
  */
-static int alpha_pmu_enable(struct perf_event *event)
+static int alpha_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int n0;
@@ -455,6 +454,10 @@ static int alpha_pmu_enable(struct perf_event *event)
 		}
 	}
 
+	hwc->state = PERF_HES_UPTODATE;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_STOPPED;
+
 	local_irq_restore(flags);
 	perf_pmu_enable(event->pmu);
 
@@ -467,7 +470,7 @@ static int alpha_pmu_enable(struct perf_event *event)
  *  - this function is called from outside this module via the pmu struct
  *    returned from perf event initialisation.
  */
-static void alpha_pmu_disable(struct perf_event *event)
+static void alpha_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
@@ -514,13 +517,44 @@ static void alpha_pmu_read(struct perf_event *event)
 }
 
 
-static void alpha_pmu_unthrottle(struct perf_event *event)
+static void alpha_pmu_stop(struct perf_event *event, int flags)
 {
 	struct hw_perf_event *hwc = &event->hw;
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		cpuc->idx_mask &= !(1UL<<hwc->idx);
+		hwc->state |= PERF_HES_STOPPED;
+	}
+
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		alpha_perf_event_update(event, hwc, hwc->idx, 0);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+
+	if (cpuc->enabled)
+		wrperfmon(PERFMON_CMD_ENABLE, (1UL<<hwc->idx));
+}
+
+
+static void alpha_pmu_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
+		return;
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+		alpha_perf_event_set_period(event, hwc, hwc->idx);
+	}
+
+	hwc->state = 0;
+
 	cpuc->idx_mask |= 1UL<<hwc->idx;
-	wrperfmon(PERFMON_CMD_ENABLE, (1UL<<hwc->idx));
+	if (cpuc->enabled)
+		wrperfmon(PERFMON_CMD_ENABLE, (1UL<<hwc->idx));
 }
 
 
@@ -671,7 +705,7 @@ static int alpha_pmu_event_init(struct perf_event *event)
 /*
  * Main entry point - enable HW performance counters.
  */
-static void alpha_pmu_pmu_enable(struct pmu *pmu)
+static void alpha_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -697,7 +731,7 @@ static void alpha_pmu_pmu_enable(struct pmu *pmu)
  * Main entry point - disable HW performance counters.
  */
 
-static void alpha_pmu_pmu_disable(struct pmu *pmu)
+static void alpha_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -711,13 +745,14 @@ static void alpha_pmu_pmu_disable(struct pmu *pmu)
 }
 
 static struct pmu pmu = {
-	.pmu_enable	= alpha_pmu_pmu_enable,
-	.pmu_disable	= alpha_pmu_pmu_disable,
+	.pmu_enable	= alpha_pmu_enable,
+	.pmu_disable	= alpha_pmu_disable,
 	.event_init	= alpha_pmu_event_init,
-	.enable		= alpha_pmu_enable,
-	.disable	= alpha_pmu_disable,
+	.add		= alpha_pmu_add,
+	.del		= alpha_pmu_del,
+	.start		= alpha_pmu_start,
+	.stop		= alpha_pmu_stop,
 	.read		= alpha_pmu_read,
-	.unthrottle	= alpha_pmu_unthrottle,
 };
 
 
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 3343f3f4b973..448cfa6b3ef0 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -221,46 +221,56 @@ again:
 }
 
 static void
-armpmu_disable(struct perf_event *event)
+armpmu_read(struct perf_event *event)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
-	int idx = hwc->idx;
-
-	WARN_ON(idx < 0);
-
-	clear_bit(idx, cpuc->active_mask);
-	armpmu->disable(hwc, idx);
-
-	barrier();
 
-	armpmu_event_update(event, hwc, idx);
-	cpuc->events[idx] = NULL;
-	clear_bit(idx, cpuc->used_mask);
+	/* Don't read disabled counters! */
+	if (hwc->idx < 0)
+		return;
 
-	perf_event_update_userpage(event);
+	armpmu_event_update(event, hwc, hwc->idx);
 }
 
 static void
-armpmu_read(struct perf_event *event)
+armpmu_stop(struct perf_event *event, int flags)
 {
 	struct hw_perf_event *hwc = &event->hw;
 
-	/* Don't read disabled counters! */
-	if (hwc->idx < 0)
+	if (!armpmu)
 		return;
 
-	armpmu_event_update(event, hwc, hwc->idx);
+	/*
+	 * ARM pmu always has to update the counter, so ignore
+	 * PERF_EF_UPDATE, see comments in armpmu_start().
+	 */
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		armpmu->disable(hwc, hwc->idx);
+		barrier(); /* why? */
+		armpmu_event_update(event, hwc, hwc->idx);
+		hwc->state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	}
 }
 
 static void
-armpmu_unthrottle(struct perf_event *event)
+armpmu_start(struct perf_event *event, int flags)
 {
 	struct hw_perf_event *hwc = &event->hw;
 
+	if (!armpmu)
+		return;
+
+	/*
+	 * ARM pmu always has to reprogram the period, so ignore
+	 * PERF_EF_RELOAD, see the comment below.
+	 */
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+
+	hwc->state = 0;
 	/*
 	 * Set the period again. Some counters can't be stopped, so when we
-	 * were throttled we simply disabled the IRQ source and the counter
+	 * were stopped we simply disabled the IRQ source and the counter
 	 * may have been left counting. If we don't do this step then we may
 	 * get an interrupt too soon or *way* too late if the overflow has
 	 * happened since disabling.
@@ -269,8 +279,25 @@ armpmu_unthrottle(struct perf_event *event)
 	armpmu->enable(hwc, hwc->idx);
 }
 
+static void
+armpmu_del(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	WARN_ON(idx < 0);
+
+	clear_bit(idx, cpuc->active_mask);
+	armpmu_stop(event, PERF_EF_UPDATE);
+	cpuc->events[idx] = NULL;
+	clear_bit(idx, cpuc->used_mask);
+
+	perf_event_update_userpage(event);
+}
+
 static int
-armpmu_enable(struct perf_event *event)
+armpmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
@@ -295,11 +322,9 @@ armpmu_enable(struct perf_event *event)
 	cpuc->events[idx] = event;
 	set_bit(idx, cpuc->active_mask);
 
-	/* Set the period for the event. */
-	armpmu_event_set_period(event, hwc, idx);
-
-	/* Enable the event. */
-	armpmu->enable(hwc, idx);
+	hwc->state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	if (flags & PERF_EF_START)
+		armpmu_start(event, PERF_EF_RELOAD);
 
 	/* Propagate our changes to the userspace mapping. */
 	perf_event_update_userpage(event);
@@ -534,7 +559,7 @@ static int armpmu_event_init(struct perf_event *event)
 	return err;
 }
 
-static void armpmu_pmu_enable(struct pmu *pmu)
+static void armpmu_enable(struct pmu *pmu)
 {
 	/* Enable all of the perf events on hardware. */
 	int idx;
@@ -555,20 +580,21 @@ static void armpmu_pmu_enable(struct pmu *pmu)
 	armpmu->start();
 }
 
-static void armpmu_pmu_disable(struct pmu *pmu)
+static void armpmu_disable(struct pmu *pmu)
 {
 	if (armpmu)
 		armpmu->stop();
 }
 
 static struct pmu pmu = {
-	.pmu_enable = armpmu_pmu_enable,
-	.pmu_disable= armpmu_pmu_disable,
-	.event_init = armpmu_event_init,
-	.enable	    = armpmu_enable,
-	.disable    = armpmu_disable,
-	.unthrottle = armpmu_unthrottle,
-	.read	    = armpmu_read,
+	.pmu_enable	= armpmu_enable,
+	.pmu_disable	= armpmu_disable,
+	.event_init	= armpmu_event_init,
+	.add		= armpmu_add,
+	.del		= armpmu_del,
+	.start		= armpmu_start,
+	.stop		= armpmu_stop,
+	.read		= armpmu_read,
 };
 
 /*
diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c
index deb84bbcb0e6..9cb4924b6c07 100644
--- a/arch/powerpc/kernel/perf_event.c
+++ b/arch/powerpc/kernel/perf_event.c
@@ -402,6 +402,9 @@ static void power_pmu_read(struct perf_event *event)
 {
 	s64 val, delta, prev;
 
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	if (!event->hw.idx)
 		return;
 	/*
@@ -517,7 +520,7 @@ static void write_mmcr0(struct cpu_hw_events *cpuhw, unsigned long mmcr0)
  * Disable all events to prevent PMU interrupts and to allow
  * events to be added or removed.
  */
-static void power_pmu_pmu_disable(struct pmu *pmu)
+static void power_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -565,7 +568,7 @@ static void power_pmu_pmu_disable(struct pmu *pmu)
  * If we were previously disabled and events were added, then
  * put the new config on the PMU.
  */
-static void power_pmu_pmu_enable(struct pmu *pmu)
+static void power_pmu_enable(struct pmu *pmu)
 {
 	struct perf_event *event;
 	struct cpu_hw_events *cpuhw;
@@ -672,6 +675,8 @@ static void power_pmu_pmu_enable(struct pmu *pmu)
 		}
 		local64_set(&event->hw.prev_count, val);
 		event->hw.idx = idx;
+		if (event->hw.state & PERF_HES_STOPPED)
+			val = 0;
 		write_pmc(idx, val);
 		perf_event_update_userpage(event);
 	}
@@ -727,7 +732,7 @@ static int collect_events(struct perf_event *group, int max_count,
  * re-enable the PMU in order to get hw_perf_enable to do the
  * actual work of reconfiguring the PMU.
  */
-static int power_pmu_enable(struct perf_event *event)
+static int power_pmu_add(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -749,6 +754,9 @@ static int power_pmu_enable(struct perf_event *event)
 	cpuhw->events[n0] = event->hw.config;
 	cpuhw->flags[n0] = event->hw.event_base;
 
+	if (!(ef_flags & PERF_EF_START))
+		event->hw.state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
@@ -777,7 +785,7 @@ nocheck:
 /*
  * Remove a event from the PMU.
  */
-static void power_pmu_disable(struct perf_event *event)
+static void power_pmu_del(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuhw;
 	long i;
@@ -826,27 +834,53 @@ static void power_pmu_disable(struct perf_event *event)
 }
 
 /*
- * Re-enable interrupts on a event after they were throttled
- * because they were coming too fast.
+ * POWER-PMU does not support disabling individual counters, hence
+ * program their cycle counter to their max value and ignore the interrupts.
  */
-static void power_pmu_unthrottle(struct perf_event *event)
+
+static void power_pmu_start(struct perf_event *event, int ef_flags)
 {
-	s64 val, left;
 	unsigned long flags;
+	s64 left;
 
 	if (!event->hw.idx || !event->hw.sample_period)
 		return;
+
+	if (!(event->hw.state & PERF_HES_STOPPED))
+		return;
+
+	if (ef_flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+
+	event->hw.state = 0;
+	left = local64_read(&event->hw.period_left);
+	write_pmc(event->hw.idx, left);
+
+	perf_event_update_userpage(event);
+	perf_pmu_enable(event->pmu);
+	local_irq_restore(flags);
+}
+
+static void power_pmu_stop(struct perf_event *event, int ef_flags)
+{
+	unsigned long flags;
+
+	if (!event->hw.idx || !event->hw.sample_period)
+		return;
+
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	local_irq_save(flags);
 	perf_pmu_disable(event->pmu);
+
 	power_pmu_read(event);
-	left = event->hw.sample_period;
-	event->hw.last_period = left;
-	val = 0;
-	if (left < 0x80000000L)
-		val = 0x80000000L - left;
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
+	event->hw.state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	write_pmc(event->hw.idx, 0);
+
 	perf_event_update_userpage(event);
 	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
@@ -1131,13 +1165,14 @@ static int power_pmu_event_init(struct perf_event *event)
 }
 
 struct pmu power_pmu = {
-	.pmu_enable	= power_pmu_pmu_enable,
-	.pmu_disable	= power_pmu_pmu_disable,
+	.pmu_enable	= power_pmu_enable,
+	.pmu_disable	= power_pmu_disable,
 	.event_init	= power_pmu_event_init,
-	.enable		= power_pmu_enable,
-	.disable	= power_pmu_disable,
+	.add		= power_pmu_add,
+	.del		= power_pmu_del,
+	.start		= power_pmu_start,
+	.stop		= power_pmu_stop,
 	.read		= power_pmu_read,
-	.unthrottle	= power_pmu_unthrottle,
 	.start_txn	= power_pmu_start_txn,
 	.cancel_txn	= power_pmu_cancel_txn,
 	.commit_txn	= power_pmu_commit_txn,
@@ -1155,6 +1190,11 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
 	s64 prev, delta, left;
 	int record = 0;
 
+	if (event->hw.state & PERF_HES_STOPPED) {
+		write_pmc(event->hw.idx, 0);
+		return;
+	}
+
 	/* we don't have to worry about interrupts here */
 	prev = local64_read(&event->hw.prev_count);
 	delta = (val - prev) & 0xfffffffful;
@@ -1177,6 +1217,11 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
 			val = 0x80000000LL - left;
 	}
 
+	write_pmc(event->hw.idx, val);
+	local64_set(&event->hw.prev_count, val);
+	local64_set(&event->hw.period_left, left);
+	perf_event_update_userpage(event);
+
 	/*
 	 * Finally record data if requested.
 	 */
@@ -1189,23 +1234,9 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
 		if (event->attr.sample_type & PERF_SAMPLE_ADDR)
 			perf_get_data_addr(regs, &data.addr);
 
-		if (perf_event_overflow(event, nmi, &data, regs)) {
-			/*
-			 * Interrupts are coming too fast - throttle them
-			 * by setting the event to 0, so it will be
-			 * at least 2^30 cycles until the next interrupt
-			 * (assuming each event counts at most 2 counts
-			 * per cycle).
-			 */
-			val = 0;
-			left = ~0ULL >> 1;
-		}
+		if (perf_event_overflow(event, nmi, &data, regs))
+			power_pmu_stop(event, 0);
 	}
-
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
-	perf_event_update_userpage(event);
 }
 
 /*
diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c
index 84b1974c628f..7ecca59ddf77 100644
--- a/arch/powerpc/kernel/perf_event_fsl_emb.c
+++ b/arch/powerpc/kernel/perf_event_fsl_emb.c
@@ -156,6 +156,9 @@ static void fsl_emb_pmu_read(struct perf_event *event)
 {
 	s64 val, delta, prev;
 
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	/*
 	 * Performance monitor interrupts come even when interrupts
 	 * are soft-disabled, as long as interrupts are hard-enabled.
@@ -177,7 +180,7 @@ static void fsl_emb_pmu_read(struct perf_event *event)
  * Disable all events to prevent PMU interrupts and to allow
  * events to be added or removed.
  */
-static void fsl_emb_pmu_pmu_disable(struct pmu *pmu)
+static void fsl_emb_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -216,7 +219,7 @@ static void fsl_emb_pmu_pmu_disable(struct pmu *pmu)
  * If we were previously disabled and events were added, then
  * put the new config on the PMU.
  */
-static void fsl_emb_pmu_pmu_enable(struct pmu *pmu)
+static void fsl_emb_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -263,7 +266,7 @@ static int collect_events(struct perf_event *group, int max_count,
 }
 
 /* context locked on entry */
-static int fsl_emb_pmu_enable(struct perf_event *event)
+static int fsl_emb_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuhw;
 	int ret = -EAGAIN;
@@ -302,6 +305,12 @@ static int fsl_emb_pmu_enable(struct perf_event *event)
 			val = 0x80000000L - left;
 	}
 	local64_set(&event->hw.prev_count, val);
+
+	if (!(flags & PERF_EF_START)) {
+		event->hw.state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+		val = 0;
+	}
+
 	write_pmc(i, val);
 	perf_event_update_userpage(event);
 
@@ -316,7 +325,7 @@ static int fsl_emb_pmu_enable(struct perf_event *event)
 }
 
 /* context locked on entry */
-static void fsl_emb_pmu_disable(struct perf_event *event)
+static void fsl_emb_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuhw;
 	int i = event->hw.idx;
@@ -353,30 +362,49 @@ static void fsl_emb_pmu_disable(struct perf_event *event)
 	put_cpu_var(cpu_hw_events);
 }
 
-/*
- * Re-enable interrupts on a event after they were throttled
- * because they were coming too fast.
- *
- * Context is locked on entry, but perf is not disabled.
- */
-static void fsl_emb_pmu_unthrottle(struct perf_event *event)
+static void fsl_emb_pmu_start(struct perf_event *event, int ef_flags)
+{
+	unsigned long flags;
+	s64 left;
+
+	if (event->hw.idx < 0 || !event->hw.sample_period)
+		return;
+
+	if (!(event->hw.state & PERF_HES_STOPPED))
+		return;
+
+	if (ef_flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+
+	event->hw.state = 0;
+	left = local64_read(&event->hw.period_left);
+	write_pmc(event->hw.idx, left);
+
+	perf_event_update_userpage(event);
+	perf_pmu_enable(event->pmu);
+	local_irq_restore(flags);
+}
+
+static void fsl_emb_pmu_stop(struct perf_event *event, int ef_flags)
 {
-	s64 val, left;
 	unsigned long flags;
 
 	if (event->hw.idx < 0 || !event->hw.sample_period)
 		return;
+
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	local_irq_save(flags);
 	perf_pmu_disable(event->pmu);
+
 	fsl_emb_pmu_read(event);
-	left = event->hw.sample_period;
-	event->hw.last_period = left;
-	val = 0;
-	if (left < 0x80000000L)
-		val = 0x80000000L - left;
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
+	event->hw.state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	write_pmc(event->hw.idx, 0);
+
 	perf_event_update_userpage(event);
 	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
@@ -524,13 +552,14 @@ static int fsl_emb_pmu_event_init(struct perf_event *event)
 }
 
 static struct pmu fsl_emb_pmu = {
-	.pmu_enable	= fsl_emb_pmu_pmu_enable,
-	.pmu_disable	= fsl_emb_pmu_pmu_disable,
+	.pmu_enable	= fsl_emb_pmu_enable,
+	.pmu_disable	= fsl_emb_pmu_disable,
 	.event_init	= fsl_emb_pmu_event_init,
-	.enable		= fsl_emb_pmu_enable,
-	.disable	= fsl_emb_pmu_disable,
+	.add		= fsl_emb_pmu_add,
+	.del		= fsl_emb_pmu_del,
+	.start		= fsl_emb_pmu_start,
+	.stop		= fsl_emb_pmu_stop,
 	.read		= fsl_emb_pmu_read,
-	.unthrottle	= fsl_emb_pmu_unthrottle,
 };
 
 /*
@@ -545,6 +574,11 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
 	s64 prev, delta, left;
 	int record = 0;
 
+	if (event->hw.state & PERF_HES_STOPPED) {
+		write_pmc(event->hw.idx, 0);
+		return;
+	}
+
 	/* we don't have to worry about interrupts here */
 	prev = local64_read(&event->hw.prev_count);
 	delta = (val - prev) & 0xfffffffful;
@@ -567,6 +601,11 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
 			val = 0x80000000LL - left;
 	}
 
+	write_pmc(event->hw.idx, val);
+	local64_set(&event->hw.prev_count, val);
+	local64_set(&event->hw.period_left, left);
+	perf_event_update_userpage(event);
+
 	/*
 	 * Finally record data if requested.
 	 */
@@ -576,23 +615,9 @@ static void record_and_restart(struct perf_event *event, unsigned long val,
 		perf_sample_data_init(&data, 0);
 		data.period = event->hw.last_period;
 
-		if (perf_event_overflow(event, nmi, &data, regs)) {
-			/*
-			 * Interrupts are coming too fast - throttle them
-			 * by setting the event to 0, so it will be
-			 * at least 2^30 cycles until the next interrupt
-			 * (assuming each event counts at most 2 counts
-			 * per cycle).
-			 */
-			val = 0;
-			left = ~0ULL >> 1;
-		}
+		if (perf_event_overflow(event, nmi, &data, regs))
+			fsl_emb_pmu_stop(event, 0);
 	}
-
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
-	perf_event_update_userpage(event);
 }
 
 static void perf_event_interrupt(struct pt_regs *regs)
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 4bbe19058a58..cf39c4873468 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -206,26 +206,52 @@ again:
 	local64_add(delta, &event->count);
 }
 
-static void sh_pmu_disable(struct perf_event *event)
+static void sh_pmu_stop(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
 
-	clear_bit(idx, cpuc->active_mask);
-	sh_pmu->disable(hwc, idx);
+	if (!(event->hw.state & PERF_HES_STOPPED)) {
+		sh_pmu->disable(hwc, idx);
+		cpuc->events[idx] = NULL;
+		event->hw.state |= PERF_HES_STOPPED;
+	}
 
-	barrier();
+	if ((flags & PERF_EF_UPDATE) && !(event->hw.state & PERF_HES_UPTODATE)) {
+		sh_perf_event_update(event, &event->hw, idx);
+		event->hw.state |= PERF_HES_UPTODATE;
+	}
+}
 
-	sh_perf_event_update(event, &event->hw, idx);
+static void sh_pmu_start(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
 
-	cpuc->events[idx] = NULL;
-	clear_bit(idx, cpuc->used_mask);
+	if (WARN_ON_ONCE(idx == -1))
+		return;
+
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	cpuc->events[idx] = event;
+	event->hw.state = 0;
+	sh_pmu->enable(hwc, idx);
+}
+
+static void sh_pmu_del(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	sh_pmu_stop(event, PERF_EF_UPDATE);
+	__clear_bit(event->hw.idx, cpuc->used_mask);
 
 	perf_event_update_userpage(event);
 }
 
-static int sh_pmu_enable(struct perf_event *event)
+static int sh_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
@@ -234,21 +260,20 @@ static int sh_pmu_enable(struct perf_event *event)
 
 	perf_pmu_disable(event->pmu);
 
-	if (test_and_set_bit(idx, cpuc->used_mask)) {
+	if (__test_and_set_bit(idx, cpuc->used_mask)) {
 		idx = find_first_zero_bit(cpuc->used_mask, sh_pmu->num_events);
 		if (idx == sh_pmu->num_events)
 			goto out;
 
-		set_bit(idx, cpuc->used_mask);
+		__set_bit(idx, cpuc->used_mask);
 		hwc->idx = idx;
 	}
 
 	sh_pmu->disable(hwc, idx);
 
-	cpuc->events[idx] = event;
-	set_bit(idx, cpuc->active_mask);
-
-	sh_pmu->enable(hwc, idx);
+	event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (flags & PERF_EF_START)
+		sh_pmu_start(event, PERF_EF_RELOAD);
 
 	perf_event_update_userpage(event);
 	ret = 0;
@@ -285,7 +310,7 @@ static int sh_pmu_event_init(struct perf_event *event)
 	return err;
 }
 
-static void sh_pmu_pmu_enable(struct pmu *pmu)
+static void sh_pmu_enable(struct pmu *pmu)
 {
 	if (!sh_pmu_initialized())
 		return;
@@ -293,7 +318,7 @@ static void sh_pmu_pmu_enable(struct pmu *pmu)
 	sh_pmu->enable_all();
 }
 
-static void sh_pmu_pmu_disable(struct pmu *pmu)
+static void sh_pmu_disable(struct pmu *pmu)
 {
 	if (!sh_pmu_initialized())
 		return;
@@ -302,11 +327,13 @@ static void sh_pmu_pmu_disable(struct pmu *pmu)
 }
 
 static struct pmu pmu = {
-	.pmu_enable	= sh_pmu_pmu_enable,
-	.pmu_disable	= sh_pmu_pmu_disable,
+	.pmu_enable	= sh_pmu_enable,
+	.pmu_disable	= sh_pmu_disable,
 	.event_init	= sh_pmu_event_init,
-	.enable		= sh_pmu_enable,
-	.disable	= sh_pmu_disable,
+	.add		= sh_pmu_add,
+	.del		= sh_pmu_del,
+	.start		= sh_pmu_start,
+	.stop		= sh_pmu_stop,
 	.read		= sh_pmu_read,
 };
 
@@ -334,15 +361,15 @@ sh_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
 	return NOTIFY_OK;
 }
 
-int __cpuinit register_sh_pmu(struct sh_pmu *pmu)
+int __cpuinit register_sh_pmu(struct sh_pmu *_pmu)
 {
 	if (sh_pmu)
 		return -EBUSY;
-	sh_pmu = pmu;
+	sh_pmu = _pmu;
 
-	pr_info("Performance Events: %s support registered\n", pmu->name);
+	pr_info("Performance Events: %s support registered\n", _pmu->name);
 
-	WARN_ON(pmu->num_events > MAX_HWEVENTS);
+	WARN_ON(_pmu->num_events > MAX_HWEVENTS);
 
 	perf_pmu_register(&pmu);
 	perf_cpu_notifier(sh_pmu_notifier);
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 37cae676536c..516be2314b54 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -658,13 +658,16 @@ static u64 maybe_change_configuration(struct cpu_hw_events *cpuc, u64 pcr)
 
 		enc = perf_event_get_enc(cpuc->events[i]);
 		pcr &= ~mask_for_index(idx);
-		pcr |= event_encoding(enc, idx);
+		if (hwc->state & PERF_HES_STOPPED)
+			pcr |= nop_for_index(idx);
+		else
+			pcr |= event_encoding(enc, idx);
 	}
 out:
 	return pcr;
 }
 
-static void sparc_pmu_pmu_enable(struct pmu *pmu)
+static void sparc_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 pcr;
@@ -691,7 +694,7 @@ static void sparc_pmu_pmu_enable(struct pmu *pmu)
 	pcr_ops->write(cpuc->pcr);
 }
 
-static void sparc_pmu_pmu_disable(struct pmu *pmu)
+static void sparc_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 val;
@@ -710,10 +713,53 @@ static void sparc_pmu_pmu_disable(struct pmu *pmu)
 	pcr_ops->write(cpuc->pcr);
 }
 
-static void sparc_pmu_disable(struct perf_event *event)
+static int active_event_index(struct cpu_hw_events *cpuc,
+			      struct perf_event *event)
+{
+	int i;
+
+	for (i = 0; i < cpuc->n_events; i++) {
+		if (cpuc->event[i] == event)
+			break;
+	}
+	BUG_ON(i == cpuc->n_events);
+	return cpuc->current_idx[i];
+}
+
+static void sparc_pmu_start(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx = active_event_index(cpuc, event);
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+		sparc_perf_event_set_period(event, &event->hw, idx);
+	}
+
+	event->hw.state = 0;
+
+	sparc_pmu_enable_event(cpuc, &event->hw, idx);
+}
+
+static void sparc_pmu_stop(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx = active_event_index(cpuc, event);
+
+	if (!(event->hw.state & PERF_HES_STOPPED)) {
+		sparc_pmu_disable_event(cpuc, &event->hw, idx);
+		event->hw.state |= PERF_HES_STOPPED;
+	}
+
+	if (!(event->hw.state & PERF_HES_UPTODATE) && (flags & PERF_EF_UPDATE)) {
+		sparc_perf_event_update(event, &event->hw, idx);
+		event->hw.state |= PERF_HES_UPTODATE;
+	}
+}
+
+static void sparc_pmu_del(struct perf_event *event, int _flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	struct hw_perf_event *hwc = &event->hw;
 	unsigned long flags;
 	int i;
 
@@ -722,7 +768,10 @@ static void sparc_pmu_disable(struct perf_event *event)
 
 	for (i = 0; i < cpuc->n_events; i++) {
 		if (event == cpuc->event[i]) {
-			int idx = cpuc->current_idx[i];
+			/* Absorb the final count and turn off the
+			 * event.
+			 */
+			sparc_pmu_stop(event, PERF_EF_UPDATE);
 
 			/* Shift remaining entries down into
 			 * the existing slot.
@@ -734,13 +783,6 @@ static void sparc_pmu_disable(struct perf_event *event)
 					cpuc->current_idx[i];
 			}
 
-			/* Absorb the final count and turn off the
-			 * event.
-			 */
-			sparc_pmu_disable_event(cpuc, hwc, idx);
-			barrier();
-			sparc_perf_event_update(event, hwc, idx);
-
 			perf_event_update_userpage(event);
 
 			cpuc->n_events--;
@@ -752,19 +794,6 @@ static void sparc_pmu_disable(struct perf_event *event)
 	local_irq_restore(flags);
 }
 
-static int active_event_index(struct cpu_hw_events *cpuc,
-			      struct perf_event *event)
-{
-	int i;
-
-	for (i = 0; i < cpuc->n_events; i++) {
-		if (cpuc->event[i] == event)
-			break;
-	}
-	BUG_ON(i == cpuc->n_events);
-	return cpuc->current_idx[i];
-}
-
 static void sparc_pmu_read(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -774,15 +803,6 @@ static void sparc_pmu_read(struct perf_event *event)
 	sparc_perf_event_update(event, hwc, idx);
 }
 
-static void sparc_pmu_unthrottle(struct perf_event *event)
-{
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	int idx = active_event_index(cpuc, event);
-	struct hw_perf_event *hwc = &event->hw;
-
-	sparc_pmu_enable_event(cpuc, hwc, idx);
-}
-
 static atomic_t active_events = ATOMIC_INIT(0);
 static DEFINE_MUTEX(pmc_grab_mutex);
 
@@ -984,7 +1004,7 @@ static int collect_events(struct perf_event *group, int max_count,
 	return n;
 }
 
-static int sparc_pmu_enable(struct perf_event *event)
+static int sparc_pmu_add(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int n0, ret = -EAGAIN;
@@ -1001,6 +1021,10 @@ static int sparc_pmu_enable(struct perf_event *event)
 	cpuc->events[n0] = event->hw.event_base;
 	cpuc->current_idx[n0] = PIC_NO_INDEX;
 
+	event->hw.state = PERF_HES_UPTODATE;
+	if (!(ef_flags & PERF_EF_START))
+		event->hw.state |= PERF_HES_STOPPED;
+
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
@@ -1156,13 +1180,14 @@ static int sparc_pmu_commit_txn(struct pmu *pmu)
 }
 
 static struct pmu pmu = {
-	.pmu_enable	= sparc_pmu_pmu_enable,
-	.pmu_disable	= sparc_pmu_pmu_disable,
+	.pmu_enable	= sparc_pmu_enable,
+	.pmu_disable	= sparc_pmu_disable,
 	.event_init	= sparc_pmu_event_init,
-	.enable		= sparc_pmu_enable,
-	.disable	= sparc_pmu_disable,
+	.add		= sparc_pmu_add,
+	.del		= sparc_pmu_del,
+	.start		= sparc_pmu_start,
+	.stop		= sparc_pmu_stop,
 	.read		= sparc_pmu_read,
-	.unthrottle	= sparc_pmu_unthrottle,
 	.start_txn	= sparc_pmu_start_txn,
 	.cancel_txn	= sparc_pmu_cancel_txn,
 	.commit_txn	= sparc_pmu_commit_txn,
@@ -1243,7 +1268,7 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self,
 			continue;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			sparc_pmu_disable_event(cpuc, hwc, idx);
+			sparc_pmu_stop(event, 0);
 	}
 
 	return NOTIFY_STOP;
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 79705ac45019..dd6fec710677 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -583,7 +583,7 @@ static void x86_pmu_disable_all(void)
 	}
 }
 
-static void x86_pmu_pmu_disable(struct pmu *pmu)
+static void x86_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -800,10 +800,10 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc,
 		hwc->last_tag == cpuc->tags[i];
 }
 
-static int x86_pmu_start(struct perf_event *event);
-static void x86_pmu_stop(struct perf_event *event);
+static void x86_pmu_start(struct perf_event *event, int flags);
+static void x86_pmu_stop(struct perf_event *event, int flags);
 
-static void x86_pmu_pmu_enable(struct pmu *pmu)
+static void x86_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct perf_event *event;
@@ -839,7 +839,14 @@ static void x86_pmu_pmu_enable(struct pmu *pmu)
 			    match_prev_assignment(hwc, cpuc, i))
 				continue;
 
-			x86_pmu_stop(event);
+			/*
+			 * Ensure we don't accidentally enable a stopped
+			 * counter simply because we rescheduled.
+			 */
+			if (hwc->state & PERF_HES_STOPPED)
+				hwc->state |= PERF_HES_ARCH;
+
+			x86_pmu_stop(event, PERF_EF_UPDATE);
 		}
 
 		for (i = 0; i < cpuc->n_events; i++) {
@@ -851,7 +858,10 @@ static void x86_pmu_pmu_enable(struct pmu *pmu)
 			else if (i < n_running)
 				continue;
 
-			x86_pmu_start(event);
+			if (hwc->state & PERF_HES_ARCH)
+				continue;
+
+			x86_pmu_start(event, PERF_EF_RELOAD);
 		}
 		cpuc->n_added = 0;
 		perf_events_lapic_init();
@@ -952,15 +962,12 @@ static void x86_pmu_enable_event(struct perf_event *event)
 }
 
 /*
- * activate a single event
+ * Add a single event to the PMU.
  *
  * The event is added to the group of enabled events
  * but only if it can be scehduled with existing events.
- *
- * Called with PMU disabled. If successful and return value 1,
- * then guaranteed to call perf_enable() and hw_perf_enable()
  */
-static int x86_pmu_enable(struct perf_event *event)
+static int x86_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc;
@@ -975,10 +982,14 @@ static int x86_pmu_enable(struct perf_event *event)
 	if (ret < 0)
 		goto out;
 
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_ARCH;
+
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
-	 * at commit time(->commit_txn) as a whole
+	 * at commit time (->commit_txn) as a whole
 	 */
 	if (cpuc->group_flag & PERF_EVENT_TXN)
 		goto done_collect;
@@ -1003,27 +1014,28 @@ out:
 	return ret;
 }
 
-static int x86_pmu_start(struct perf_event *event)
+static void x86_pmu_start(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx = event->hw.idx;
 
-	if (idx == -1)
-		return -EAGAIN;
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
+
+	if (WARN_ON_ONCE(idx == -1))
+		return;
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+		x86_perf_event_set_period(event);
+	}
+
+	event->hw.state = 0;
 
-	x86_perf_event_set_period(event);
 	cpuc->events[idx] = event;
 	__set_bit(idx, cpuc->active_mask);
 	x86_pmu.enable(event);
 	perf_event_update_userpage(event);
-
-	return 0;
-}
-
-static void x86_pmu_unthrottle(struct perf_event *event)
-{
-	int ret = x86_pmu_start(event);
-	WARN_ON_ONCE(ret);
 }
 
 void perf_event_print_debug(void)
@@ -1080,27 +1092,29 @@ void perf_event_print_debug(void)
 	local_irq_restore(flags);
 }
 
-static void x86_pmu_stop(struct perf_event *event)
+static void x86_pmu_stop(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
-	int idx = hwc->idx;
-
-	if (!__test_and_clear_bit(idx, cpuc->active_mask))
-		return;
-
-	x86_pmu.disable(event);
 
-	/*
-	 * Drain the remaining delta count out of a event
-	 * that we are disabling:
-	 */
-	x86_perf_event_update(event);
+	if (__test_and_clear_bit(hwc->idx, cpuc->active_mask)) {
+		x86_pmu.disable(event);
+		cpuc->events[hwc->idx] = NULL;
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+	}
 
-	cpuc->events[idx] = NULL;
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		/*
+		 * Drain the remaining delta count out of a event
+		 * that we are disabling:
+		 */
+		x86_perf_event_update(event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
 }
 
-static void x86_pmu_disable(struct perf_event *event)
+static void x86_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int i;
@@ -1113,7 +1127,7 @@ static void x86_pmu_disable(struct perf_event *event)
 	if (cpuc->group_flag & PERF_EVENT_TXN)
 		return;
 
-	x86_pmu_stop(event);
+	x86_pmu_stop(event, PERF_EF_UPDATE);
 
 	for (i = 0; i < cpuc->n_events; i++) {
 		if (event == cpuc->event_list[i]) {
@@ -1165,7 +1179,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs)
 			continue;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			x86_pmu_stop(event);
+			x86_pmu_stop(event, 0);
 	}
 
 	if (handled)
@@ -1605,15 +1619,17 @@ int x86_pmu_event_init(struct perf_event *event)
 }
 
 static struct pmu pmu = {
-	.pmu_enable	= x86_pmu_pmu_enable,
-	.pmu_disable	= x86_pmu_pmu_disable,
+	.pmu_enable	= x86_pmu_enable,
+	.pmu_disable	= x86_pmu_disable,
+
 	.event_init	= x86_pmu_event_init,
-	.enable		= x86_pmu_enable,
-	.disable	= x86_pmu_disable,
+
+	.add		= x86_pmu_add,
+	.del		= x86_pmu_del,
 	.start		= x86_pmu_start,
 	.stop		= x86_pmu_stop,
 	.read		= x86_pmu_read,
-	.unthrottle	= x86_pmu_unthrottle,
+
 	.start_txn	= x86_pmu_start_txn,
 	.cancel_txn	= x86_pmu_cancel_txn,
 	.commit_txn	= x86_pmu_commit_txn,
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index ee05c90012d2..82395f2378ec 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -763,7 +763,7 @@ again:
 		data.period = event->hw.last_period;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			x86_pmu_stop(event);
+			x86_pmu_stop(event, 0);
 	}
 
 	/*
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 18018d1311cd..9893a2f77b7a 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -491,7 +491,7 @@ static void __intel_pmu_pebs_event(struct perf_event *event,
 		regs.flags &= ~PERF_EFLAGS_EXACT;
 
 	if (perf_event_overflow(event, 1, &data, &regs))
-		x86_pmu_stop(event);
+		x86_pmu_stop(event, 0);
 }
 
 static void intel_pmu_drain_pebs_core(struct pt_regs *iregs)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 5f8ad7bec636..8beabb958f61 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -252,8 +252,8 @@ DECLARE_PER_CPU(struct pt_regs, perf_trace_regs);
 
 extern int  perf_trace_init(struct perf_event *event);
 extern void perf_trace_destroy(struct perf_event *event);
-extern int  perf_trace_enable(struct perf_event *event);
-extern void perf_trace_disable(struct perf_event *event);
+extern int  perf_trace_add(struct perf_event *event, int flags);
+extern void perf_trace_del(struct perf_event *event, int flags);
 extern int  ftrace_profile_set_filter(struct perf_event *event, int event_id,
 				     char *filter_str);
 extern void ftrace_profile_free_filter(struct perf_event *event);
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 8cafa15af60d..402073c61669 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -538,6 +538,7 @@ struct hw_perf_event {
 		};
 #endif
 	};
+	int				state;
 	local64_t			prev_count;
 	u64				sample_period;
 	u64				last_period;
@@ -549,6 +550,13 @@ struct hw_perf_event {
 #endif
 };
 
+/*
+ * hw_perf_event::state flags
+ */
+#define PERF_HES_STOPPED	0x01 /* the counter is stopped */
+#define PERF_HES_UPTODATE	0x02 /* event->count up-to-date */
+#define PERF_HES_ARCH		0x04
+
 struct perf_event;
 
 /*
@@ -564,42 +572,62 @@ struct pmu {
 
 	int				*pmu_disable_count;
 
+	/*
+	 * Fully disable/enable this PMU, can be used to protect from the PMI
+	 * as well as for lazy/batch writing of the MSRs.
+	 */
 	void (*pmu_enable)		(struct pmu *pmu); /* optional */
 	void (*pmu_disable)		(struct pmu *pmu); /* optional */
 
 	/*
+	 * Try and initialize the event for this PMU.
 	 * Should return -ENOENT when the @event doesn't match this PMU.
 	 */
 	int (*event_init)		(struct perf_event *event);
 
-	int  (*enable)			(struct perf_event *event);
-	void (*disable)			(struct perf_event *event);
-	int  (*start)			(struct perf_event *event);
-	void (*stop)			(struct perf_event *event);
+#define PERF_EF_START	0x01		/* start the counter when adding    */
+#define PERF_EF_RELOAD	0x02		/* reload the counter when starting */
+#define PERF_EF_UPDATE	0x04		/* update the counter when stopping */
+
+	/*
+	 * Adds/Removes a counter to/from the PMU, can be done inside
+	 * a transaction, see the ->*_txn() methods.
+	 */
+	int  (*add)			(struct perf_event *event, int flags);
+	void (*del)			(struct perf_event *event, int flags);
+
+	/*
+	 * Starts/Stops a counter present on the PMU. The PMI handler
+	 * should stop the counter when perf_event_overflow() returns
+	 * !0. ->start() will be used to continue.
+	 */
+	void (*start)			(struct perf_event *event, int flags);
+	void (*stop)			(struct perf_event *event, int flags);
+
+	/*
+	 * Updates the counter value of the event.
+	 */
 	void (*read)			(struct perf_event *event);
-	void (*unthrottle)		(struct perf_event *event);
 
 	/*
 	 * Group events scheduling is treated as a transaction, add
 	 * group events as a whole and perform one schedulability test.
 	 * If the test fails, roll back the whole group
-	 */
-
-	/*
-	 * Start the transaction, after this ->enable() doesn't need to
+	 *
+	 * Start the transaction, after this ->add() doesn't need to
 	 * do schedulability tests.
 	 */
 	void (*start_txn)	(struct pmu *pmu); /* optional */
 	/*
-	 * If ->start_txn() disabled the ->enable() schedulability test
+	 * If ->start_txn() disabled the ->add() schedulability test
 	 * then ->commit_txn() is required to perform one. On success
 	 * the transaction is closed. On error the transaction is kept
 	 * open until ->cancel_txn() is called.
 	 */
 	int  (*commit_txn)	(struct pmu *pmu); /* optional */
 	/*
-	 * Will cancel the transaction, assumes ->disable() is called
-	 * for each successfull ->enable() during the transaction.
+	 * Will cancel the transaction, assumes ->del() is called
+	 * for each successfull ->add() during the transaction.
 	 */
 	void (*cancel_txn)	(struct pmu *pmu); /* optional */
 };
@@ -680,7 +708,7 @@ struct perf_event {
 	int				nr_siblings;
 	int				group_flags;
 	struct perf_event		*group_leader;
-	struct pmu		*pmu;
+	struct pmu			*pmu;
 
 	enum perf_event_active_state	state;
 	unsigned int			attach_state;
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index e9c5cfa1fd20..6f150095cafe 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -586,10 +586,35 @@ static int hw_breakpoint_event_init(struct perf_event *bp)
 	return 0;
 }
 
+static int hw_breakpoint_add(struct perf_event *bp, int flags)
+{
+	if (!(flags & PERF_EF_START))
+		bp->hw.state = PERF_HES_STOPPED;
+
+	return arch_install_hw_breakpoint(bp);
+}
+
+static void hw_breakpoint_del(struct perf_event *bp, int flags)
+{
+	arch_uninstall_hw_breakpoint(bp);
+}
+
+static void hw_breakpoint_start(struct perf_event *bp, int flags)
+{
+	bp->hw.state = 0;
+}
+
+static void hw_breakpoint_stop(struct perf_event *bp, int flags)
+{
+	bp->hw.state = PERF_HES_STOPPED;
+}
+
 static struct pmu perf_breakpoint = {
 	.event_init	= hw_breakpoint_event_init,
-	.enable		= arch_install_hw_breakpoint,
-	.disable	= arch_uninstall_hw_breakpoint,
+	.add		= hw_breakpoint_add,
+	.del		= hw_breakpoint_del,
+	.start		= hw_breakpoint_start,
+	.stop		= hw_breakpoint_stop,
 	.read		= hw_breakpoint_pmu_read,
 };
 
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 1a6cdbf0d091..3bace4fd0355 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -424,7 +424,7 @@ event_sched_out(struct perf_event *event,
 		event->state = PERF_EVENT_STATE_OFF;
 	}
 	event->tstamp_stopped = ctx->time;
-	event->pmu->disable(event);
+	event->pmu->del(event, 0);
 	event->oncpu = -1;
 
 	if (!is_software_event(event))
@@ -649,7 +649,7 @@ event_sched_in(struct perf_event *event,
 	 */
 	smp_wmb();
 
-	if (event->pmu->enable(event)) {
+	if (event->pmu->add(event, PERF_EF_START)) {
 		event->state = PERF_EVENT_STATE_INACTIVE;
 		event->oncpu = -1;
 		return -EAGAIN;
@@ -1482,22 +1482,6 @@ do {					\
 	return div64_u64(dividend, divisor);
 }
 
-static void perf_event_stop(struct perf_event *event)
-{
-	if (!event->pmu->stop)
-		return event->pmu->disable(event);
-
-	return event->pmu->stop(event);
-}
-
-static int perf_event_start(struct perf_event *event)
-{
-	if (!event->pmu->start)
-		return event->pmu->enable(event);
-
-	return event->pmu->start(event);
-}
-
 static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count)
 {
 	struct hw_perf_event *hwc = &event->hw;
@@ -1517,9 +1501,9 @@ static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count)
 	hwc->sample_period = sample_period;
 
 	if (local64_read(&hwc->period_left) > 8*sample_period) {
-		perf_event_stop(event);
+		event->pmu->stop(event, PERF_EF_UPDATE);
 		local64_set(&hwc->period_left, 0);
-		perf_event_start(event);
+		event->pmu->start(event, PERF_EF_RELOAD);
 	}
 }
 
@@ -1548,7 +1532,7 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
 		 */
 		if (interrupts == MAX_INTERRUPTS) {
 			perf_log_throttle(event, 1);
-			event->pmu->unthrottle(event);
+			event->pmu->start(event, 0);
 		}
 
 		if (!event->attr.freq || !event->attr.sample_freq)
@@ -2506,6 +2490,9 @@ int perf_event_task_disable(void)
 
 static int perf_event_index(struct perf_event *event)
 {
+	if (event->hw.state & PERF_HES_STOPPED)
+		return 0;
+
 	if (event->state != PERF_EVENT_STATE_ACTIVE)
 		return 0;
 
@@ -4120,8 +4107,6 @@ static int __perf_event_overflow(struct perf_event *event, int nmi,
 	struct hw_perf_event *hwc = &event->hw;
 	int ret = 0;
 
-	throttle = (throttle && event->pmu->unthrottle != NULL);
-
 	if (!throttle) {
 		hwc->interrupts++;
 	} else {
@@ -4246,7 +4231,7 @@ static void perf_swevent_overflow(struct perf_event *event, u64 overflow,
 	}
 }
 
-static void perf_swevent_add(struct perf_event *event, u64 nr,
+static void perf_swevent_event(struct perf_event *event, u64 nr,
 			       int nmi, struct perf_sample_data *data,
 			       struct pt_regs *regs)
 {
@@ -4272,6 +4257,9 @@ static void perf_swevent_add(struct perf_event *event, u64 nr,
 static int perf_exclude_event(struct perf_event *event,
 			      struct pt_regs *regs)
 {
+	if (event->hw.state & PERF_HES_STOPPED)
+		return 0;
+
 	if (regs) {
 		if (event->attr.exclude_user && user_mode(regs))
 			return 1;
@@ -4371,7 +4359,7 @@ static void do_perf_sw_event(enum perf_type_id type, u32 event_id,
 
 	hlist_for_each_entry_rcu(event, node, head, hlist_entry) {
 		if (perf_swevent_match(event, type, event_id, data, regs))
-			perf_swevent_add(event, nr, nmi, data, regs);
+			perf_swevent_event(event, nr, nmi, data, regs);
 	}
 end:
 	rcu_read_unlock();
@@ -4415,7 +4403,7 @@ static void perf_swevent_read(struct perf_event *event)
 {
 }
 
-static int perf_swevent_enable(struct perf_event *event)
+static int perf_swevent_add(struct perf_event *event, int flags)
 {
 	struct hw_perf_event *hwc = &event->hw;
 	struct perf_cpu_context *cpuctx;
@@ -4428,6 +4416,8 @@ static int perf_swevent_enable(struct perf_event *event)
 		perf_swevent_set_period(event);
 	}
 
+	hwc->state = !(flags & PERF_EF_START);
+
 	head = find_swevent_head(cpuctx, event);
 	if (WARN_ON_ONCE(!head))
 		return -EINVAL;
@@ -4437,18 +4427,19 @@ static int perf_swevent_enable(struct perf_event *event)
 	return 0;
 }
 
-static void perf_swevent_disable(struct perf_event *event)
+static void perf_swevent_del(struct perf_event *event, int flags)
 {
 	hlist_del_rcu(&event->hlist_entry);
 }
 
-static void perf_swevent_void(struct perf_event *event)
+static void perf_swevent_start(struct perf_event *event, int flags)
 {
+	event->hw.state = 0;
 }
 
-static int perf_swevent_int(struct perf_event *event)
+static void perf_swevent_stop(struct perf_event *event, int flags)
 {
-	return 0;
+	event->hw.state = PERF_HES_STOPPED;
 }
 
 /* Deref the hlist from the update side */
@@ -4604,12 +4595,11 @@ static int perf_swevent_init(struct perf_event *event)
 
 static struct pmu perf_swevent = {
 	.event_init	= perf_swevent_init,
-	.enable		= perf_swevent_enable,
-	.disable	= perf_swevent_disable,
-	.start		= perf_swevent_int,
-	.stop		= perf_swevent_void,
+	.add		= perf_swevent_add,
+	.del		= perf_swevent_del,
+	.start		= perf_swevent_start,
+	.stop		= perf_swevent_stop,
 	.read		= perf_swevent_read,
-	.unthrottle	= perf_swevent_void, /* hwc->interrupts already reset */
 };
 
 #ifdef CONFIG_EVENT_TRACING
@@ -4657,7 +4647,7 @@ void perf_tp_event(u64 addr, u64 count, void *record, int entry_size,
 
 	hlist_for_each_entry_rcu(event, node, head, hlist_entry) {
 		if (perf_tp_event_match(event, &data, regs))
-			perf_swevent_add(event, count, 1, &data, regs);
+			perf_swevent_event(event, count, 1, &data, regs);
 	}
 
 	perf_swevent_put_recursion_context(rctx);
@@ -4696,12 +4686,11 @@ static int perf_tp_event_init(struct perf_event *event)
 
 static struct pmu perf_tracepoint = {
 	.event_init	= perf_tp_event_init,
-	.enable		= perf_trace_enable,
-	.disable	= perf_trace_disable,
-	.start		= perf_swevent_int,
-	.stop		= perf_swevent_void,
+	.add		= perf_trace_add,
+	.del		= perf_trace_del,
+	.start		= perf_swevent_start,
+	.stop		= perf_swevent_stop,
 	.read		= perf_swevent_read,
-	.unthrottle	= perf_swevent_void,
 };
 
 static inline void perf_tp_register(void)
@@ -4757,8 +4746,8 @@ void perf_bp_event(struct perf_event *bp, void *data)
 
 	perf_sample_data_init(&sample, bp->attr.bp_addr);
 
-	if (!perf_exclude_event(bp, regs))
-		perf_swevent_add(bp, 1, 1, &sample, regs);
+	if (!bp->hw.state && !perf_exclude_event(bp, regs))
+		perf_swevent_event(bp, 1, 1, &sample, regs);
 }
 #endif
 
@@ -4834,32 +4823,39 @@ static void perf_swevent_cancel_hrtimer(struct perf_event *event)
 
 static void cpu_clock_event_update(struct perf_event *event)
 {
-	int cpu = raw_smp_processor_id();
 	s64 prev;
 	u64 now;
 
-	now = cpu_clock(cpu);
+	now = local_clock();
 	prev = local64_xchg(&event->hw.prev_count, now);
 	local64_add(now - prev, &event->count);
 }
 
-static int cpu_clock_event_enable(struct perf_event *event)
+static void cpu_clock_event_start(struct perf_event *event, int flags)
 {
-	struct hw_perf_event *hwc = &event->hw;
-	int cpu = raw_smp_processor_id();
-
-	local64_set(&hwc->prev_count, cpu_clock(cpu));
+	local64_set(&event->hw.prev_count, local_clock());
 	perf_swevent_start_hrtimer(event);
-
-	return 0;
 }
 
-static void cpu_clock_event_disable(struct perf_event *event)
+static void cpu_clock_event_stop(struct perf_event *event, int flags)
 {
 	perf_swevent_cancel_hrtimer(event);
 	cpu_clock_event_update(event);
 }
 
+static int cpu_clock_event_add(struct perf_event *event, int flags)
+{
+	if (flags & PERF_EF_START)
+		cpu_clock_event_start(event, flags);
+
+	return 0;
+}
+
+static void cpu_clock_event_del(struct perf_event *event, int flags)
+{
+	cpu_clock_event_stop(event, flags);
+}
+
 static void cpu_clock_event_read(struct perf_event *event)
 {
 	cpu_clock_event_update(event);
@@ -4878,8 +4874,10 @@ static int cpu_clock_event_init(struct perf_event *event)
 
 static struct pmu perf_cpu_clock = {
 	.event_init	= cpu_clock_event_init,
-	.enable		= cpu_clock_event_enable,
-	.disable	= cpu_clock_event_disable,
+	.add		= cpu_clock_event_add,
+	.del		= cpu_clock_event_del,
+	.start		= cpu_clock_event_start,
+	.stop		= cpu_clock_event_stop,
 	.read		= cpu_clock_event_read,
 };
 
@@ -4897,25 +4895,29 @@ static void task_clock_event_update(struct perf_event *event, u64 now)
 	local64_add(delta, &event->count);
 }
 
-static int task_clock_event_enable(struct perf_event *event)
+static void task_clock_event_start(struct perf_event *event, int flags)
 {
-	struct hw_perf_event *hwc = &event->hw;
-	u64 now;
-
-	now = event->ctx->time;
-
-	local64_set(&hwc->prev_count, now);
-
+	local64_set(&event->hw.prev_count, event->ctx->time);
 	perf_swevent_start_hrtimer(event);
-
-	return 0;
 }
 
-static void task_clock_event_disable(struct perf_event *event)
+static void task_clock_event_stop(struct perf_event *event, int flags)
 {
 	perf_swevent_cancel_hrtimer(event);
 	task_clock_event_update(event, event->ctx->time);
+}
+
+static int task_clock_event_add(struct perf_event *event, int flags)
+{
+	if (flags & PERF_EF_START)
+		task_clock_event_start(event, flags);
 
+	return 0;
+}
+
+static void task_clock_event_del(struct perf_event *event, int flags)
+{
+	task_clock_event_stop(event, PERF_EF_UPDATE);
 }
 
 static void task_clock_event_read(struct perf_event *event)
@@ -4947,8 +4949,10 @@ static int task_clock_event_init(struct perf_event *event)
 
 static struct pmu perf_task_clock = {
 	.event_init	= task_clock_event_init,
-	.enable		= task_clock_event_enable,
-	.disable	= task_clock_event_disable,
+	.add		= task_clock_event_add,
+	.del		= task_clock_event_del,
+	.start		= task_clock_event_start,
+	.stop		= task_clock_event_stop,
 	.read		= task_clock_event_read,
 };
 
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index f3bbcd1c90c8..39c059ca670e 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -101,7 +101,7 @@ int perf_trace_init(struct perf_event *p_event)
 	return ret;
 }
 
-int perf_trace_enable(struct perf_event *p_event)
+int perf_trace_add(struct perf_event *p_event, int flags)
 {
 	struct ftrace_event_call *tp_event = p_event->tp_event;
 	struct hlist_head __percpu *pcpu_list;
@@ -111,13 +111,16 @@ int perf_trace_enable(struct perf_event *p_event)
 	if (WARN_ON_ONCE(!pcpu_list))
 		return -EINVAL;
 
+	if (!(flags & PERF_EF_START))
+		p_event->hw.state = PERF_HES_STOPPED;
+
 	list = this_cpu_ptr(pcpu_list);
 	hlist_add_head_rcu(&p_event->hlist_entry, list);
 
 	return 0;
 }
 
-void perf_trace_disable(struct perf_event *p_event)
+void perf_trace_del(struct perf_event *p_event, int flags)
 {
 	hlist_del_rcu(&p_event->hlist_entry);
 }
-- 
cgit v1.2.3


From 15ac9a395a753cb28c674e7ea80386ffdff21785 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 6 Sep 2010 15:51:45 +0200
Subject: perf: Remove the sysfs bits

Neither the overcommit nor the reservation sysfs parameter were
actually working, remove them as they'll only get in the way.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/alpha/kernel/perf_event.c   |   3 +-
 arch/arm/kernel/perf_event.c     |   9 +--
 arch/sparc/kernel/perf_event.c   |   9 +--
 arch/x86/kernel/cpu/perf_event.c |   1 -
 include/linux/perf_event.h       |   6 --
 kernel/perf_event.c              | 124 ---------------------------------------
 6 files changed, 5 insertions(+), 147 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c
index 380ef02d557a..9bb8c024080c 100644
--- a/arch/alpha/kernel/perf_event.c
+++ b/arch/alpha/kernel/perf_event.c
@@ -808,7 +808,7 @@ static void alpha_perf_event_irq_handler(unsigned long la_ptr,
 	wrperfmon(PERFMON_CMD_DISABLE, cpuc->idx_mask);
 
 	/* la_ptr is the counter that overflowed. */
-	if (unlikely(la_ptr >= perf_max_events)) {
+	if (unlikely(la_ptr >= alpha_pmu->num_pmcs)) {
 		/* This should never occur! */
 		irq_err_count++;
 		pr_warning("PMI: silly index %ld\n", la_ptr);
@@ -879,7 +879,6 @@ void __init init_hw_perf_events(void)
 
 	/* And set up PMU specification */
 	alpha_pmu = &ev67_pmu;
-	perf_max_events = alpha_pmu->num_pmcs;
 
 	perf_pmu_register(&pmu);
 }
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 448cfa6b3ef0..45d6a35217c1 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -534,7 +534,7 @@ static int armpmu_event_init(struct perf_event *event)
 	event->destroy = hw_perf_event_destroy;
 
 	if (!atomic_inc_not_zero(&active_events)) {
-		if (atomic_read(&active_events) > perf_max_events) {
+		if (atomic_read(&active_events) > armpmu.num_events) {
 			atomic_dec(&active_events);
 			return -ENOSPC;
 		}
@@ -2974,14 +2974,12 @@ init_hw_perf_events(void)
 			armpmu = &armv6pmu;
 			memcpy(armpmu_perf_cache_map, armv6_perf_cache_map,
 					sizeof(armv6_perf_cache_map));
-			perf_max_events	= armv6pmu.num_events;
 			break;
 		case 0xB020:	/* ARM11mpcore */
 			armpmu = &armv6mpcore_pmu;
 			memcpy(armpmu_perf_cache_map,
 			       armv6mpcore_perf_cache_map,
 			       sizeof(armv6mpcore_perf_cache_map));
-			perf_max_events = armv6mpcore_pmu.num_events;
 			break;
 		case 0xC080:	/* Cortex-A8 */
 			armv7pmu.id = ARM_PERF_PMU_ID_CA8;
@@ -2993,7 +2991,6 @@ init_hw_perf_events(void)
 			/* Reset PMNC and read the nb of CNTx counters
 			    supported */
 			armv7pmu.num_events = armv7_reset_read_pmnc();
-			perf_max_events = armv7pmu.num_events;
 			break;
 		case 0xC090:	/* Cortex-A9 */
 			armv7pmu.id = ARM_PERF_PMU_ID_CA9;
@@ -3005,7 +3002,6 @@ init_hw_perf_events(void)
 			/* Reset PMNC and read the nb of CNTx counters
 			    supported */
 			armv7pmu.num_events = armv7_reset_read_pmnc();
-			perf_max_events = armv7pmu.num_events;
 			break;
 		}
 	/* Intel CPUs [xscale]. */
@@ -3016,13 +3012,11 @@ init_hw_perf_events(void)
 			armpmu = &xscale1pmu;
 			memcpy(armpmu_perf_cache_map, xscale_perf_cache_map,
 					sizeof(xscale_perf_cache_map));
-			perf_max_events	= xscale1pmu.num_events;
 			break;
 		case 2:
 			armpmu = &xscale2pmu;
 			memcpy(armpmu_perf_cache_map, xscale_perf_cache_map,
 					sizeof(xscale_perf_cache_map));
-			perf_max_events	= xscale2pmu.num_events;
 			break;
 		}
 	}
@@ -3032,7 +3026,6 @@ init_hw_perf_events(void)
 				arm_pmu_names[armpmu->id], armpmu->num_events);
 	} else {
 		pr_info("no hardware support available\n");
-		perf_max_events = -1;
 	}
 
 	perf_pmu_register(&pmu);
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 516be2314b54..f9a706759364 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -897,7 +897,7 @@ static int sparc_check_constraints(struct perf_event **evts,
 	if (!n_ev)
 		return 0;
 
-	if (n_ev > perf_max_events)
+	if (n_ev > MAX_HWEVENTS)
 		return -1;
 
 	msk0 = perf_event_get_msk(events[0]);
@@ -1014,7 +1014,7 @@ static int sparc_pmu_add(struct perf_event *event, int ef_flags)
 	perf_pmu_disable(event->pmu);
 
 	n0 = cpuc->n_events;
-	if (n0 >= perf_max_events)
+	if (n0 >= MAX_HWEVENTS)
 		goto out;
 
 	cpuc->event[n0] = event;
@@ -1097,7 +1097,7 @@ static int sparc_pmu_event_init(struct perf_event *event)
 	n = 0;
 	if (event->group_leader != event) {
 		n = collect_events(event->group_leader,
-				   perf_max_events - 1,
+				   MAX_HWEVENTS - 1,
 				   evts, events, current_idx_dmy);
 		if (n < 0)
 			return -EINVAL;
@@ -1309,9 +1309,6 @@ void __init init_hw_perf_events(void)
 
 	pr_cont("Supported PMU type is '%s'\n", sparc_pmu_type);
 
-	/* All sparc64 PMUs currently have 2 events.  */
-	perf_max_events = 2;
-
 	perf_pmu_register(&pmu);
 	register_die_notifier(&perf_event_nmi_notifier);
 }
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index dd6fec710677..0fb17050360f 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1396,7 +1396,6 @@ void __init init_hw_perf_events(void)
 		x86_pmu.num_counters = X86_PMC_MAX_GENERIC;
 	}
 	x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1;
-	perf_max_events = x86_pmu.num_counters;
 
 	if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) {
 		WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!",
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 402073c61669..b22176d3ebdf 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -860,7 +860,6 @@ struct perf_cpu_context {
 	struct perf_event_context	ctx;
 	struct perf_event_context	*task_ctx;
 	int				active_oncpu;
-	int				max_pertask;
 	int				exclusive;
 	struct swevent_hlist		*swevent_hlist;
 	struct mutex			hlist_mutex;
@@ -883,11 +882,6 @@ struct perf_output_handle {
 
 #ifdef CONFIG_PERF_EVENTS
 
-/*
- * Set by architecture code:
- */
-extern int perf_max_events;
-
 extern int perf_pmu_register(struct pmu *pmu);
 extern void perf_pmu_unregister(struct pmu *pmu);
 
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 3bace4fd0355..8462e69409ae 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -39,10 +39,6 @@
  */
 static DEFINE_PER_CPU(struct perf_cpu_context, perf_cpu_context);
 
-int perf_max_events __read_mostly = 1;
-static int perf_reserved_percpu __read_mostly;
-static int perf_overcommit __read_mostly = 1;
-
 static atomic_t nr_events __read_mostly;
 static atomic_t nr_mmap_events __read_mostly;
 static atomic_t nr_comm_events __read_mostly;
@@ -66,11 +62,6 @@ int sysctl_perf_event_sample_rate __read_mostly = 100000;
 
 static atomic64_t perf_event_id;
 
-/*
- * Lock for (sysadmin-configurable) event reservations:
- */
-static DEFINE_SPINLOCK(perf_resource_lock);
-
 void __weak perf_event_print_debug(void)	{ }
 
 void perf_pmu_disable(struct pmu *pmu)
@@ -480,16 +471,6 @@ static void __perf_event_remove_from_context(void *info)
 
 	list_del_event(event, ctx);
 
-	if (!ctx->task) {
-		/*
-		 * Allow more per task events with respect to the
-		 * reservation:
-		 */
-		cpuctx->max_pertask =
-			min(perf_max_events - ctx->nr_events,
-			    perf_max_events - perf_reserved_percpu);
-	}
-
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -823,9 +804,6 @@ static void __perf_install_in_context(void *info)
 		}
 	}
 
-	if (!err && !ctx->task && cpuctx->max_pertask)
-		cpuctx->max_pertask--;
-
 unlock:
 	raw_spin_unlock(&ctx->lock);
 }
@@ -5930,10 +5908,6 @@ static void __cpuinit perf_event_init_cpu(int cpu)
 
 	cpuctx = &per_cpu(perf_cpu_context, cpu);
 
-	spin_lock(&perf_resource_lock);
-	cpuctx->max_pertask = perf_max_events - perf_reserved_percpu;
-	spin_unlock(&perf_resource_lock);
-
 	mutex_lock(&cpuctx->hlist_mutex);
 	if (cpuctx->hlist_refcount > 0) {
 		struct swevent_hlist *hlist;
@@ -6008,101 +5982,3 @@ void __init perf_event_init(void)
 	perf_tp_register();
 	perf_cpu_notifier(perf_cpu_notify);
 }
-
-static ssize_t perf_show_reserve_percpu(struct sysdev_class *class,
-					struct sysdev_class_attribute *attr,
-					char *buf)
-{
-	return sprintf(buf, "%d\n", perf_reserved_percpu);
-}
-
-static ssize_t
-perf_set_reserve_percpu(struct sysdev_class *class,
-			struct sysdev_class_attribute *attr,
-			const char *buf,
-			size_t count)
-{
-	struct perf_cpu_context *cpuctx;
-	unsigned long val;
-	int err, cpu, mpt;
-
-	err = strict_strtoul(buf, 10, &val);
-	if (err)
-		return err;
-	if (val > perf_max_events)
-		return -EINVAL;
-
-	spin_lock(&perf_resource_lock);
-	perf_reserved_percpu = val;
-	for_each_online_cpu(cpu) {
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		raw_spin_lock_irq(&cpuctx->ctx.lock);
-		mpt = min(perf_max_events - cpuctx->ctx.nr_events,
-			  perf_max_events - perf_reserved_percpu);
-		cpuctx->max_pertask = mpt;
-		raw_spin_unlock_irq(&cpuctx->ctx.lock);
-	}
-	spin_unlock(&perf_resource_lock);
-
-	return count;
-}
-
-static ssize_t perf_show_overcommit(struct sysdev_class *class,
-				    struct sysdev_class_attribute *attr,
-				    char *buf)
-{
-	return sprintf(buf, "%d\n", perf_overcommit);
-}
-
-static ssize_t
-perf_set_overcommit(struct sysdev_class *class,
-		    struct sysdev_class_attribute *attr,
-		    const char *buf, size_t count)
-{
-	unsigned long val;
-	int err;
-
-	err = strict_strtoul(buf, 10, &val);
-	if (err)
-		return err;
-	if (val > 1)
-		return -EINVAL;
-
-	spin_lock(&perf_resource_lock);
-	perf_overcommit = val;
-	spin_unlock(&perf_resource_lock);
-
-	return count;
-}
-
-static SYSDEV_CLASS_ATTR(
-				reserve_percpu,
-				0644,
-				perf_show_reserve_percpu,
-				perf_set_reserve_percpu
-			);
-
-static SYSDEV_CLASS_ATTR(
-				overcommit,
-				0644,
-				perf_show_overcommit,
-				perf_set_overcommit
-			);
-
-static struct attribute *perfclass_attrs[] = {
-	&attr_reserve_percpu.attr,
-	&attr_overcommit.attr,
-	NULL
-};
-
-static struct attribute_group perfclass_attr_group = {
-	.attrs			= perfclass_attrs,
-	.name			= "perf_events",
-};
-
-static int __init perf_event_sysfs_init(void)
-{
-	return sysfs_create_group(&cpu_sysdev_class.kset.kobj,
-				  &perfclass_attr_group);
-}
-device_initcall(perf_event_sysfs_init);
-- 
cgit v1.2.3


From b28ab83c595e767f2028276b7398d17f2253cec0 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 6 Sep 2010 14:48:15 +0200
Subject: perf: Remove the swevent hash-table from the cpu context

Separate the swevent hash-table from the cpu_context bits in
preparation for per pmu cpu contexts.

This keeps the swevent hash a global entity.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |   6 ---
 kernel/perf_event.c        | 104 +++++++++++++++++++++++++--------------------
 2 files changed, 58 insertions(+), 52 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index b22176d3ebdf..4ab4f0ca09a1 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -861,12 +861,6 @@ struct perf_cpu_context {
 	struct perf_event_context	*task_ctx;
 	int				active_oncpu;
 	int				exclusive;
-	struct swevent_hlist		*swevent_hlist;
-	struct mutex			hlist_mutex;
-	int				hlist_refcount;
-
-	/* Recursion avoidance in each contexts */
-	int				recursion[PERF_NR_CONTEXTS];
 };
 
 struct perf_output_handle {
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index a3c86a8335c4..2c47ed6c4f26 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -4154,6 +4154,17 @@ int perf_event_overflow(struct perf_event *event, int nmi,
  * Generic software event infrastructure
  */
 
+struct swevent_htable {
+	struct swevent_hlist		*swevent_hlist;
+	struct mutex			hlist_mutex;
+	int				hlist_refcount;
+
+	/* Recursion avoidance in each contexts */
+	int				recursion[PERF_NR_CONTEXTS];
+};
+
+static DEFINE_PER_CPU(struct swevent_htable, swevent_htable);
+
 /*
  * We directly increment event->count and keep a second value in
  * event->hw.period_left to count intervals. This period event
@@ -4286,11 +4297,11 @@ __find_swevent_head(struct swevent_hlist *hlist, u64 type, u32 event_id)
 
 /* For the read side: events when they trigger */
 static inline struct hlist_head *
-find_swevent_head_rcu(struct perf_cpu_context *ctx, u64 type, u32 event_id)
+find_swevent_head_rcu(struct swevent_htable *swhash, u64 type, u32 event_id)
 {
 	struct swevent_hlist *hlist;
 
-	hlist = rcu_dereference(ctx->swevent_hlist);
+	hlist = rcu_dereference(swhash->swevent_hlist);
 	if (!hlist)
 		return NULL;
 
@@ -4299,7 +4310,7 @@ find_swevent_head_rcu(struct perf_cpu_context *ctx, u64 type, u32 event_id)
 
 /* For the event head insertion and removal in the hlist */
 static inline struct hlist_head *
-find_swevent_head(struct perf_cpu_context *ctx, struct perf_event *event)
+find_swevent_head(struct swevent_htable *swhash, struct perf_event *event)
 {
 	struct swevent_hlist *hlist;
 	u32 event_id = event->attr.config;
@@ -4310,7 +4321,7 @@ find_swevent_head(struct perf_cpu_context *ctx, struct perf_event *event)
 	 * and release. Which makes the protected version suitable here.
 	 * The context lock guarantees that.
 	 */
-	hlist = rcu_dereference_protected(ctx->swevent_hlist,
+	hlist = rcu_dereference_protected(swhash->swevent_hlist,
 					  lockdep_is_held(&event->ctx->lock));
 	if (!hlist)
 		return NULL;
@@ -4323,17 +4334,13 @@ static void do_perf_sw_event(enum perf_type_id type, u32 event_id,
 				    struct perf_sample_data *data,
 				    struct pt_regs *regs)
 {
-	struct perf_cpu_context *cpuctx;
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 	struct perf_event *event;
 	struct hlist_node *node;
 	struct hlist_head *head;
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-
 	rcu_read_lock();
-
-	head = find_swevent_head_rcu(cpuctx, type, event_id);
-
+	head = find_swevent_head_rcu(swhash, type, event_id);
 	if (!head)
 		goto end;
 
@@ -4347,17 +4354,17 @@ end:
 
 int perf_swevent_get_recursion_context(void)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 
-	return get_recursion_context(cpuctx->recursion);
+	return get_recursion_context(swhash->recursion);
 }
 EXPORT_SYMBOL_GPL(perf_swevent_get_recursion_context);
 
 void inline perf_swevent_put_recursion_context(int rctx)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 
-	put_recursion_context(cpuctx->recursion, rctx);
+	put_recursion_context(swhash->recursion, rctx);
 }
 
 void __perf_sw_event(u32 event_id, u64 nr, int nmi,
@@ -4385,12 +4392,10 @@ static void perf_swevent_read(struct perf_event *event)
 
 static int perf_swevent_add(struct perf_event *event, int flags)
 {
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 	struct hw_perf_event *hwc = &event->hw;
-	struct perf_cpu_context *cpuctx;
 	struct hlist_head *head;
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-
 	if (hwc->sample_period) {
 		hwc->last_period = hwc->sample_period;
 		perf_swevent_set_period(event);
@@ -4398,7 +4403,7 @@ static int perf_swevent_add(struct perf_event *event, int flags)
 
 	hwc->state = !(flags & PERF_EF_START);
 
-	head = find_swevent_head(cpuctx, event);
+	head = find_swevent_head(swhash, event);
 	if (WARN_ON_ONCE(!head))
 		return -EINVAL;
 
@@ -4424,10 +4429,10 @@ static void perf_swevent_stop(struct perf_event *event, int flags)
 
 /* Deref the hlist from the update side */
 static inline struct swevent_hlist *
-swevent_hlist_deref(struct perf_cpu_context *cpuctx)
+swevent_hlist_deref(struct swevent_htable *swhash)
 {
-	return rcu_dereference_protected(cpuctx->swevent_hlist,
-					 lockdep_is_held(&cpuctx->hlist_mutex));
+	return rcu_dereference_protected(swhash->swevent_hlist,
+					 lockdep_is_held(&swhash->hlist_mutex));
 }
 
 static void swevent_hlist_release_rcu(struct rcu_head *rcu_head)
@@ -4438,27 +4443,27 @@ static void swevent_hlist_release_rcu(struct rcu_head *rcu_head)
 	kfree(hlist);
 }
 
-static void swevent_hlist_release(struct perf_cpu_context *cpuctx)
+static void swevent_hlist_release(struct swevent_htable *swhash)
 {
-	struct swevent_hlist *hlist = swevent_hlist_deref(cpuctx);
+	struct swevent_hlist *hlist = swevent_hlist_deref(swhash);
 
 	if (!hlist)
 		return;
 
-	rcu_assign_pointer(cpuctx->swevent_hlist, NULL);
+	rcu_assign_pointer(swhash->swevent_hlist, NULL);
 	call_rcu(&hlist->rcu_head, swevent_hlist_release_rcu);
 }
 
 static void swevent_hlist_put_cpu(struct perf_event *event, int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
-	mutex_lock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
 
-	if (!--cpuctx->hlist_refcount)
-		swevent_hlist_release(cpuctx);
+	if (!--swhash->hlist_refcount)
+		swevent_hlist_release(swhash);
 
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_unlock(&swhash->hlist_mutex);
 }
 
 static void swevent_hlist_put(struct perf_event *event)
@@ -4476,12 +4481,12 @@ static void swevent_hlist_put(struct perf_event *event)
 
 static int swevent_hlist_get_cpu(struct perf_event *event, int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 	int err = 0;
 
-	mutex_lock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
 
-	if (!swevent_hlist_deref(cpuctx) && cpu_online(cpu)) {
+	if (!swevent_hlist_deref(swhash) && cpu_online(cpu)) {
 		struct swevent_hlist *hlist;
 
 		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
@@ -4489,11 +4494,11 @@ static int swevent_hlist_get_cpu(struct perf_event *event, int cpu)
 			err = -ENOMEM;
 			goto exit;
 		}
-		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+		rcu_assign_pointer(swhash->swevent_hlist, hlist);
 	}
-	cpuctx->hlist_refcount++;
+	swhash->hlist_refcount++;
 exit:
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_unlock(&swhash->hlist_mutex);
 
 	return err;
 }
@@ -5889,12 +5894,15 @@ int perf_event_init_task(struct task_struct *child)
 
 static void __init perf_event_init_all_cpus(void)
 {
-	int cpu;
 	struct perf_cpu_context *cpuctx;
+	struct swevent_htable *swhash;
+	int cpu;
 
 	for_each_possible_cpu(cpu) {
+		swhash = &per_cpu(swevent_htable, cpu);
+		mutex_init(&swhash->hlist_mutex);
+
 		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		mutex_init(&cpuctx->hlist_mutex);
 		__perf_event_init_context(&cpuctx->ctx, NULL);
 	}
 }
@@ -5902,18 +5910,21 @@ static void __init perf_event_init_all_cpus(void)
 static void __cpuinit perf_event_init_cpu(int cpu)
 {
 	struct perf_cpu_context *cpuctx;
+	struct swevent_htable *swhash;
 
 	cpuctx = &per_cpu(perf_cpu_context, cpu);
 
-	mutex_lock(&cpuctx->hlist_mutex);
-	if (cpuctx->hlist_refcount > 0) {
+	swhash = &per_cpu(swevent_htable, cpu);
+
+	mutex_lock(&swhash->hlist_mutex);
+	if (swhash->hlist_refcount > 0) {
 		struct swevent_hlist *hlist;
 
-		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
-		WARN_ON_ONCE(!hlist);
-		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+		hlist = kzalloc_node(sizeof(*hlist), GFP_KERNEL, cpu_to_node(cpu));
+		WARN_ON(!hlist);
+		rcu_assign_pointer(swhash->swevent_hlist, hlist);
 	}
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_unlock(&swhash->hlist_mutex);
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -5931,11 +5942,12 @@ static void __perf_event_exit_cpu(void *info)
 static void perf_event_exit_cpu(int cpu)
 {
 	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 	struct perf_event_context *ctx = &cpuctx->ctx;
 
-	mutex_lock(&cpuctx->hlist_mutex);
-	swevent_hlist_release(cpuctx);
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
+	swevent_hlist_release(swhash);
+	mutex_unlock(&swhash->hlist_mutex);
 
 	mutex_lock(&ctx->mutex);
 	smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1);
-- 
cgit v1.2.3


From b5ab4cd563e7ab49b27957704112a8ecade54e1f Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 6 Sep 2010 16:32:21 +0200
Subject: perf: Per cpu-context rotation timer

Give each cpu-context its own timer so that it is a self contained
entity, this eases the way for per-pmu-per-cpu contexts as well as
provides the basic infrastructure to allow different rotation
times per pmu.

Things to look at:
 - folding the tick and these TICK_NSEC timers
 - separate task context rotation

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |  5 ++-
 kernel/perf_event.c        | 80 ++++++++++++++++++++++++++++++++++++----------
 kernel/sched.c             |  2 --
 3 files changed, 65 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 4ab4f0ca09a1..fa04537df55b 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -861,6 +861,8 @@ struct perf_cpu_context {
 	struct perf_event_context	*task_ctx;
 	int				active_oncpu;
 	int				exclusive;
+	u64				timer_interval;
+	struct hrtimer			timer;
 };
 
 struct perf_output_handle {
@@ -881,7 +883,6 @@ extern void perf_pmu_unregister(struct pmu *pmu);
 
 extern void perf_event_task_sched_in(struct task_struct *task);
 extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
-extern void perf_event_task_tick(struct task_struct *task);
 extern int perf_event_init_task(struct task_struct *child);
 extern void perf_event_exit_task(struct task_struct *child);
 extern void perf_event_free_task(struct task_struct *task);
@@ -1067,8 +1068,6 @@ perf_event_task_sched_in(struct task_struct *task)			{ }
 static inline void
 perf_event_task_sched_out(struct task_struct *task,
 			    struct task_struct *next)			{ }
-static inline void
-perf_event_task_tick(struct task_struct *task)				{ }
 static inline int perf_event_init_task(struct task_struct *child)	{ return 0; }
 static inline void perf_event_exit_task(struct task_struct *child)	{ }
 static inline void perf_event_free_task(struct task_struct *task)	{ }
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 2c47ed6c4f26..d75e4c8727f9 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -78,6 +78,25 @@ void perf_pmu_enable(struct pmu *pmu)
 		pmu->pmu_enable(pmu);
 }
 
+static void perf_pmu_rotate_start(void)
+{
+	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+
+	if (hrtimer_active(&cpuctx->timer))
+		return;
+
+	__hrtimer_start_range_ns(&cpuctx->timer,
+			ns_to_ktime(cpuctx->timer_interval), 0,
+			HRTIMER_MODE_REL_PINNED, 0);
+}
+
+static void perf_pmu_rotate_stop(void)
+{
+	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+
+	hrtimer_cancel(&cpuctx->timer);
+}
+
 static void get_ctx(struct perf_event_context *ctx)
 {
 	WARN_ON(!atomic_inc_not_zero(&ctx->refcount));
@@ -281,6 +300,8 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx)
 	}
 
 	list_add_rcu(&event->event_entry, &ctx->event_list);
+	if (!ctx->nr_events)
+		perf_pmu_rotate_start();
 	ctx->nr_events++;
 	if (event->attr.inherit_stat)
 		ctx->nr_stat++;
@@ -1383,6 +1404,12 @@ void perf_event_task_sched_in(struct task_struct *task)
 	ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE);
 
 	cpuctx->task_ctx = ctx;
+
+	/*
+	 * Since these rotations are per-cpu, we need to ensure the
+	 * cpu-context we got scheduled on is actually rotating.
+	 */
+	perf_pmu_rotate_start();
 }
 
 #define MAX_INTERRUPTS (~0ULL)
@@ -1487,7 +1514,7 @@ static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count)
 	}
 }
 
-static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
+static void perf_ctx_adjust_freq(struct perf_event_context *ctx, u64 period)
 {
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
@@ -1524,7 +1551,7 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
 		hwc->freq_count_stamp = now;
 
 		if (delta > 0)
-			perf_adjust_period(event, TICK_NSEC, delta);
+			perf_adjust_period(event, period, delta);
 	}
 	raw_spin_unlock(&ctx->lock);
 }
@@ -1542,30 +1569,39 @@ static void rotate_ctx(struct perf_event_context *ctx)
 	raw_spin_unlock(&ctx->lock);
 }
 
-void perf_event_task_tick(struct task_struct *curr)
+/*
+ * Cannot race with ->pmu_rotate_start() because this is ran from hardirq
+ * context, and ->pmu_rotate_start() is called with irqs disabled (both are
+ * cpu affine, so there are no SMP races).
+ */
+static enum hrtimer_restart perf_event_context_tick(struct hrtimer *timer)
 {
+	enum hrtimer_restart restart = HRTIMER_NORESTART;
 	struct perf_cpu_context *cpuctx;
 	struct perf_event_context *ctx;
 	int rotate = 0;
 
-	if (!atomic_read(&nr_events))
-		return;
+	cpuctx = container_of(timer, struct perf_cpu_context, timer);
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-	if (cpuctx->ctx.nr_events &&
-	    cpuctx->ctx.nr_events != cpuctx->ctx.nr_active)
-		rotate = 1;
+	if (cpuctx->ctx.nr_events) {
+		restart = HRTIMER_RESTART;
+		if (cpuctx->ctx.nr_events != cpuctx->ctx.nr_active)
+			rotate = 1;
+	}
 
-	ctx = curr->perf_event_ctxp;
-	if (ctx && ctx->nr_events && ctx->nr_events != ctx->nr_active)
-		rotate = 1;
+	ctx = current->perf_event_ctxp;
+	if (ctx && ctx->nr_events) {
+		restart = HRTIMER_RESTART;
+		if (ctx->nr_events != ctx->nr_active)
+			rotate = 1;
+	}
 
-	perf_ctx_adjust_freq(&cpuctx->ctx);
+	perf_ctx_adjust_freq(&cpuctx->ctx, cpuctx->timer_interval);
 	if (ctx)
-		perf_ctx_adjust_freq(ctx);
+		perf_ctx_adjust_freq(ctx, cpuctx->timer_interval);
 
 	if (!rotate)
-		return;
+		goto done;
 
 	cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
@@ -1577,7 +1613,12 @@ void perf_event_task_tick(struct task_struct *curr)
 
 	cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
-		task_ctx_sched_in(curr, EVENT_FLEXIBLE);
+		task_ctx_sched_in(current, EVENT_FLEXIBLE);
+
+done:
+	hrtimer_forward_now(timer, ns_to_ktime(cpuctx->timer_interval));
+
+	return restart;
 }
 
 static int event_enable_on_exec(struct perf_event *event,
@@ -4786,7 +4827,7 @@ static void perf_swevent_start_hrtimer(struct perf_event *event)
 		}
 		__hrtimer_start_range_ns(&hwc->hrtimer,
 				ns_to_ktime(period), 0,
-				HRTIMER_MODE_REL, 0);
+				HRTIMER_MODE_REL_PINNED, 0);
 	}
 }
 
@@ -5904,6 +5945,9 @@ static void __init perf_event_init_all_cpus(void)
 
 		cpuctx = &per_cpu(perf_cpu_context, cpu);
 		__perf_event_init_context(&cpuctx->ctx, NULL);
+		cpuctx->timer_interval = TICK_NSEC;
+		hrtimer_init(&cpuctx->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+		cpuctx->timer.function = perf_event_context_tick;
 	}
 }
 
@@ -5934,6 +5978,8 @@ static void __perf_event_exit_cpu(void *info)
 	struct perf_event_context *ctx = &cpuctx->ctx;
 	struct perf_event *event, *tmp;
 
+	perf_pmu_rotate_stop();
+
 	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
 		__perf_event_remove_from_context(event);
 	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, group_entry)
diff --git a/kernel/sched.c b/kernel/sched.c
index 09b574e7f4df..66a02ba83c01 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3578,8 +3578,6 @@ void scheduler_tick(void)
 	curr->sched_class->task_tick(rq, curr, 0);
 	raw_spin_unlock(&rq->lock);
 
-	perf_event_task_tick(curr);
-
 #ifdef CONFIG_SMP
 	rq->idle_at_tick = idle_cpu(cpu);
 	trigger_load_balance(rq, cpu);
-- 
cgit v1.2.3


From 108b02cfce04ee90b0a07ee0b104baffd39f5934 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 6 Sep 2010 14:32:03 +0200
Subject: perf: Per-pmu-per-cpu contexts

Allocate per-cpu contexts per pmu.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |   4 +-
 kernel/perf_event.c        | 178 ++++++++++++++++++++++++++++-----------------
 2 files changed, 113 insertions(+), 69 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index fa04537df55b..22155ef3b362 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -570,7 +570,8 @@ struct perf_event;
 struct pmu {
 	struct list_head		entry;
 
-	int				*pmu_disable_count;
+	int * __percpu			pmu_disable_count;
+	struct perf_cpu_context * __percpu pmu_cpu_context;
 
 	/*
 	 * Fully disable/enable this PMU, can be used to protect from the PMI
@@ -808,6 +809,7 @@ struct perf_event {
  * Used as a container for task events and CPU events as well:
  */
 struct perf_event_context {
+	struct pmu			*pmu;
 	/*
 	 * Protect the states of the events in the list,
 	 * nr_active, and the list:
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index d75e4c8727f9..8ca6e690ffe3 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -34,16 +34,15 @@
 
 #include <asm/irq_regs.h>
 
-/*
- * Each CPU has a list of per CPU events:
- */
-static DEFINE_PER_CPU(struct perf_cpu_context, perf_cpu_context);
-
 static atomic_t nr_events __read_mostly;
 static atomic_t nr_mmap_events __read_mostly;
 static atomic_t nr_comm_events __read_mostly;
 static atomic_t nr_task_events __read_mostly;
 
+static LIST_HEAD(pmus);
+static DEFINE_MUTEX(pmus_lock);
+static struct srcu_struct pmus_srcu;
+
 /*
  * perf event paranoia level:
  *  -1 - not paranoid at all
@@ -78,9 +77,9 @@ void perf_pmu_enable(struct pmu *pmu)
 		pmu->pmu_enable(pmu);
 }
 
-static void perf_pmu_rotate_start(void)
+static void perf_pmu_rotate_start(struct pmu *pmu)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
 
 	if (hrtimer_active(&cpuctx->timer))
 		return;
@@ -90,9 +89,9 @@ static void perf_pmu_rotate_start(void)
 			HRTIMER_MODE_REL_PINNED, 0);
 }
 
-static void perf_pmu_rotate_stop(void)
+static void perf_pmu_rotate_stop(struct pmu *pmu)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
 
 	hrtimer_cancel(&cpuctx->timer);
 }
@@ -301,7 +300,7 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx)
 
 	list_add_rcu(&event->event_entry, &ctx->event_list);
 	if (!ctx->nr_events)
-		perf_pmu_rotate_start();
+		perf_pmu_rotate_start(ctx->pmu);
 	ctx->nr_events++;
 	if (event->attr.inherit_stat)
 		ctx->nr_stat++;
@@ -466,6 +465,12 @@ group_sched_out(struct perf_event *group_event,
 		cpuctx->exclusive = 0;
 }
 
+static inline struct perf_cpu_context *
+__get_cpu_context(struct perf_event_context *ctx)
+{
+	return this_cpu_ptr(ctx->pmu->pmu_cpu_context);
+}
+
 /*
  * Cross CPU call to remove a performance event
  *
@@ -474,9 +479,9 @@ group_sched_out(struct perf_event *group_event,
  */
 static void __perf_event_remove_from_context(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a task context, we need to check whether it is
@@ -556,8 +561,8 @@ retry:
 static void __perf_event_disable(void *info)
 {
 	struct perf_event *event = info;
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a per-task event, need to check whether this
@@ -765,10 +770,10 @@ static void add_event_to_ctx(struct perf_event *event,
  */
 static void __perf_install_in_context(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
 	struct perf_event *leader = event->group_leader;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 	int err;
 
 	/*
@@ -912,9 +917,9 @@ static void __perf_event_mark_enabled(struct perf_event *event,
 static void __perf_event_enable(void *info)
 {
 	struct perf_event *event = info;
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = event->ctx;
 	struct perf_event *leader = event->group_leader;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 	int err;
 
 	/*
@@ -1188,15 +1193,19 @@ static void perf_event_sync_stat(struct perf_event_context *ctx,
 void perf_event_task_sched_out(struct task_struct *task,
 				 struct task_struct *next)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = task->perf_event_ctxp;
 	struct perf_event_context *next_ctx;
 	struct perf_event_context *parent;
+	struct perf_cpu_context *cpuctx;
 	int do_switch = 1;
 
 	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
 
-	if (likely(!ctx || !cpuctx->task_ctx))
+	if (likely(!ctx))
+		return;
+
+	cpuctx = __get_cpu_context(ctx);
+	if (!cpuctx->task_ctx)
 		return;
 
 	rcu_read_lock();
@@ -1242,7 +1251,7 @@ void perf_event_task_sched_out(struct task_struct *task,
 static void task_ctx_sched_out(struct perf_event_context *ctx,
 			       enum event_type_t event_type)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	if (!cpuctx->task_ctx)
 		return;
@@ -1360,8 +1369,8 @@ static void cpu_ctx_sched_in(struct perf_cpu_context *cpuctx,
 static void task_ctx_sched_in(struct task_struct *task,
 			      enum event_type_t event_type)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	if (likely(!ctx))
 		return;
@@ -1383,12 +1392,13 @@ static void task_ctx_sched_in(struct task_struct *task,
  */
 void perf_event_task_sched_in(struct task_struct *task)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_cpu_context *cpuctx;
 
 	if (likely(!ctx))
 		return;
 
+	cpuctx = __get_cpu_context(ctx);
 	if (cpuctx->task_ctx == ctx)
 		return;
 
@@ -1409,7 +1419,7 @@ void perf_event_task_sched_in(struct task_struct *task)
 	 * Since these rotations are per-cpu, we need to ensure the
 	 * cpu-context we got scheduled on is actually rotating.
 	 */
-	perf_pmu_rotate_start();
+	perf_pmu_rotate_start(ctx->pmu);
 }
 
 #define MAX_INTERRUPTS (~0ULL)
@@ -1687,9 +1697,9 @@ out:
  */
 static void __perf_event_read(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a task context, we need to check whether it is
@@ -1962,7 +1972,8 @@ __perf_event_init_context(struct perf_event_context *ctx,
 	ctx->task = task;
 }
 
-static struct perf_event_context *find_get_context(pid_t pid, int cpu)
+static struct perf_event_context *
+find_get_context(struct pmu *pmu, pid_t pid, int cpu)
 {
 	struct perf_event_context *ctx;
 	struct perf_cpu_context *cpuctx;
@@ -1986,7 +1997,7 @@ static struct perf_event_context *find_get_context(pid_t pid, int cpu)
 		if (!cpu_online(cpu))
 			return ERR_PTR(-ENODEV);
 
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
+		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
 		ctx = &cpuctx->ctx;
 		get_ctx(ctx);
 
@@ -2030,6 +2041,7 @@ retry:
 		if (!ctx)
 			goto errout;
 		__perf_event_init_context(ctx, task);
+		ctx->pmu = pmu;
 		get_ctx(ctx);
 		if (cmpxchg(&task->perf_event_ctxp, NULL, ctx)) {
 			/*
@@ -3745,18 +3757,20 @@ static void perf_event_task_ctx(struct perf_event_context *ctx,
 
 static void perf_event_task_event(struct perf_task_event *task_event)
 {
-	struct perf_cpu_context *cpuctx;
 	struct perf_event_context *ctx = task_event->task_ctx;
+	struct perf_cpu_context *cpuctx;
+	struct pmu *pmu;
 
-	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_task_ctx(&cpuctx->ctx, task_event);
+	rcu_read_lock_sched();
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_task_ctx(&cpuctx->ctx, task_event);
+	}
 	if (!ctx)
 		ctx = rcu_dereference(current->perf_event_ctxp);
 	if (ctx)
 		perf_event_task_ctx(ctx, task_event);
-	put_cpu_var(perf_cpu_context);
-	rcu_read_unlock();
+	rcu_read_unlock_sched();
 }
 
 static void perf_event_task(struct task_struct *task,
@@ -3861,6 +3875,7 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event)
 	struct perf_cpu_context *cpuctx;
 	struct perf_event_context *ctx;
 	unsigned int size;
+	struct pmu *pmu;
 	char comm[TASK_COMM_LEN];
 
 	memset(comm, 0, sizeof(comm));
@@ -3872,14 +3887,15 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event)
 
 	comm_event->event_id.header.size = sizeof(comm_event->event_id) + size;
 
-	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_comm_ctx(&cpuctx->ctx, comm_event);
+	rcu_read_lock_sched();
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_comm_ctx(&cpuctx->ctx, comm_event);
+	}
 	ctx = rcu_dereference(current->perf_event_ctxp);
 	if (ctx)
 		perf_event_comm_ctx(ctx, comm_event);
-	put_cpu_var(perf_cpu_context);
-	rcu_read_unlock();
+	rcu_read_unlock_sched();
 }
 
 void perf_event_comm(struct task_struct *task)
@@ -3989,6 +4005,7 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 	char tmp[16];
 	char *buf = NULL;
 	const char *name;
+	struct pmu *pmu;
 
 	memset(tmp, 0, sizeof(tmp));
 
@@ -4040,14 +4057,16 @@ got_name:
 
 	mmap_event->event_id.header.size = sizeof(mmap_event->event_id) + size;
 
-	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_mmap_ctx(&cpuctx->ctx, mmap_event, vma->vm_flags & VM_EXEC);
+	rcu_read_lock_sched();
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_mmap_ctx(&cpuctx->ctx, mmap_event,
+					vma->vm_flags & VM_EXEC);
+	}
 	ctx = rcu_dereference(current->perf_event_ctxp);
 	if (ctx)
 		perf_event_mmap_ctx(ctx, mmap_event, vma->vm_flags & VM_EXEC);
-	put_cpu_var(perf_cpu_context);
-	rcu_read_unlock();
+	rcu_read_unlock_sched();
 
 	kfree(buf);
 }
@@ -4982,10 +5001,6 @@ static struct pmu perf_task_clock = {
 	.read		= task_clock_event_read,
 };
 
-static LIST_HEAD(pmus);
-static DEFINE_MUTEX(pmus_lock);
-static struct srcu_struct pmus_srcu;
-
 static void perf_pmu_nop_void(struct pmu *pmu)
 {
 }
@@ -5013,7 +5028,7 @@ static void perf_pmu_cancel_txn(struct pmu *pmu)
 
 int perf_pmu_register(struct pmu *pmu)
 {
-	int ret;
+	int cpu, ret;
 
 	mutex_lock(&pmus_lock);
 	ret = -ENOMEM;
@@ -5021,6 +5036,21 @@ int perf_pmu_register(struct pmu *pmu)
 	if (!pmu->pmu_disable_count)
 		goto unlock;
 
+	pmu->pmu_cpu_context = alloc_percpu(struct perf_cpu_context);
+	if (!pmu->pmu_cpu_context)
+		goto free_pdc;
+
+	for_each_possible_cpu(cpu) {
+		struct perf_cpu_context *cpuctx;
+
+		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
+		__perf_event_init_context(&cpuctx->ctx, NULL);
+		cpuctx->ctx.pmu = pmu;
+		cpuctx->timer_interval = TICK_NSEC;
+		hrtimer_init(&cpuctx->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+		cpuctx->timer.function = perf_event_context_tick;
+	}
+
 	if (!pmu->start_txn) {
 		if (pmu->pmu_enable) {
 			/*
@@ -5049,6 +5079,10 @@ unlock:
 	mutex_unlock(&pmus_lock);
 
 	return ret;
+
+free_pdc:
+	free_percpu(pmu->pmu_disable_count);
+	goto unlock;
 }
 
 void perf_pmu_unregister(struct pmu *pmu)
@@ -5057,9 +5091,14 @@ void perf_pmu_unregister(struct pmu *pmu)
 	list_del_rcu(&pmu->entry);
 	mutex_unlock(&pmus_lock);
 
+	/*
+	 * We use the pmu list either under SRCU or preempt_disable,
+	 * synchronize_srcu() implies synchronize_sched() so we're good.
+	 */
 	synchronize_srcu(&pmus_srcu);
 
 	free_percpu(pmu->pmu_disable_count);
+	free_percpu(pmu->pmu_cpu_context);
 }
 
 struct pmu *perf_init_event(struct perf_event *event)
@@ -5374,7 +5413,7 @@ SYSCALL_DEFINE5(perf_event_open,
 	/*
 	 * Get the target context (task or percpu):
 	 */
-	ctx = find_get_context(pid, cpu);
+	ctx = find_get_context(event->pmu, pid, cpu);
 	if (IS_ERR(ctx)) {
 		err = PTR_ERR(ctx);
 		goto err_alloc;
@@ -5489,7 +5528,7 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
 		goto err;
 	}
 
-	ctx = find_get_context(pid, cpu);
+	ctx = find_get_context(event->pmu, pid, cpu);
 	if (IS_ERR(ctx)) {
 		err = PTR_ERR(ctx);
 		goto err_free;
@@ -5833,6 +5872,7 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent,
 			return -ENOMEM;
 
 		__perf_event_init_context(child_ctx, child);
+		child_ctx->pmu = event->pmu;
 		child->perf_event_ctxp = child_ctx;
 		get_task_struct(child);
 	}
@@ -5935,30 +5975,18 @@ int perf_event_init_task(struct task_struct *child)
 
 static void __init perf_event_init_all_cpus(void)
 {
-	struct perf_cpu_context *cpuctx;
 	struct swevent_htable *swhash;
 	int cpu;
 
 	for_each_possible_cpu(cpu) {
 		swhash = &per_cpu(swevent_htable, cpu);
 		mutex_init(&swhash->hlist_mutex);
-
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		__perf_event_init_context(&cpuctx->ctx, NULL);
-		cpuctx->timer_interval = TICK_NSEC;
-		hrtimer_init(&cpuctx->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-		cpuctx->timer.function = perf_event_context_tick;
 	}
 }
 
 static void __cpuinit perf_event_init_cpu(int cpu)
 {
-	struct perf_cpu_context *cpuctx;
-	struct swevent_htable *swhash;
-
-	cpuctx = &per_cpu(perf_cpu_context, cpu);
-
-	swhash = &per_cpu(swevent_htable, cpu);
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
 	mutex_lock(&swhash->hlist_mutex);
 	if (swhash->hlist_refcount > 0) {
@@ -5972,32 +6000,46 @@ static void __cpuinit perf_event_init_cpu(int cpu)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static void __perf_event_exit_cpu(void *info)
+static void __perf_event_exit_context(void *__info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = &cpuctx->ctx;
+	struct perf_event_context *ctx = __info;
 	struct perf_event *event, *tmp;
 
-	perf_pmu_rotate_stop();
+	perf_pmu_rotate_stop(ctx->pmu);
 
 	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
 		__perf_event_remove_from_context(event);
 	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, group_entry)
 		__perf_event_remove_from_context(event);
 }
+
+static void perf_event_exit_cpu_context(int cpu)
+{
+	struct perf_event_context *ctx;
+	struct pmu *pmu;
+	int idx;
+
+	idx = srcu_read_lock(&pmus_srcu);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		ctx = &this_cpu_ptr(pmu->pmu_cpu_context)->ctx;
+
+		mutex_lock(&ctx->mutex);
+		smp_call_function_single(cpu, __perf_event_exit_context, ctx, 1);
+		mutex_unlock(&ctx->mutex);
+	}
+	srcu_read_unlock(&pmus_srcu, idx);
+
+}
+
 static void perf_event_exit_cpu(int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
 	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
-	struct perf_event_context *ctx = &cpuctx->ctx;
 
 	mutex_lock(&swhash->hlist_mutex);
 	swevent_hlist_release(swhash);
 	mutex_unlock(&swhash->hlist_mutex);
 
-	mutex_lock(&ctx->mutex);
-	smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1);
-	mutex_unlock(&ctx->mutex);
+	perf_event_exit_cpu_context(cpu);
 }
 #else
 static inline void perf_event_exit_cpu(int cpu) { }
-- 
cgit v1.2.3


From 8dc85d547285668e509f86c177bcd4ea055bcaaf Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 2 Sep 2010 16:50:03 +0200
Subject: perf: Multiple task contexts

Provide the infrastructure for multiple task contexts.

A more flexible approach would have resulted in more pointer chases
in the scheduling hot-paths. This approach has the limitation of a
static number of task contexts.

Since I expect most external PMUs to be system wide, or at least node
wide (as per the intel uncore unit) they won't actually need a task
context.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |   1 +
 include/linux/sched.h      |   8 +-
 kernel/perf_event.c        | 336 +++++++++++++++++++++++++++++++--------------
 3 files changed, 239 insertions(+), 106 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 22155ef3b362..9ecfd856ce6e 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -572,6 +572,7 @@ struct pmu {
 
 	int * __percpu			pmu_disable_count;
 	struct perf_cpu_context * __percpu pmu_cpu_context;
+	int				task_ctx_nr;
 
 	/*
 	 * Fully disable/enable this PMU, can be used to protect from the PMI
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1e2a6db2d7dd..89d6023c6f82 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1160,6 +1160,12 @@ struct sched_rt_entity {
 
 struct rcu_node;
 
+enum perf_event_task_context {
+	perf_invalid_context = -1,
+	perf_hw_context = 0,
+	perf_nr_task_contexts,
+};
+
 struct task_struct {
 	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
 	void *stack;
@@ -1431,7 +1437,7 @@ struct task_struct {
 	struct futex_pi_state *pi_state_cache;
 #endif
 #ifdef CONFIG_PERF_EVENTS
-	struct perf_event_context *perf_event_ctxp;
+	struct perf_event_context *perf_event_ctxp[perf_nr_task_contexts];
 	struct mutex perf_event_mutex;
 	struct list_head perf_event_list;
 #endif
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 13d98d756347..7223ea875861 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -148,13 +148,13 @@ static u64 primary_event_id(struct perf_event *event)
  * the context could get moved to another task.
  */
 static struct perf_event_context *
-perf_lock_task_context(struct task_struct *task, unsigned long *flags)
+perf_lock_task_context(struct task_struct *task, int ctxn, unsigned long *flags)
 {
 	struct perf_event_context *ctx;
 
 	rcu_read_lock();
 retry:
-	ctx = rcu_dereference(task->perf_event_ctxp);
+	ctx = rcu_dereference(task->perf_event_ctxp[ctxn]);
 	if (ctx) {
 		/*
 		 * If this context is a clone of another, it might
@@ -167,7 +167,7 @@ retry:
 		 * can't get swapped on us any more.
 		 */
 		raw_spin_lock_irqsave(&ctx->lock, *flags);
-		if (ctx != rcu_dereference(task->perf_event_ctxp)) {
+		if (ctx != rcu_dereference(task->perf_event_ctxp[ctxn])) {
 			raw_spin_unlock_irqrestore(&ctx->lock, *flags);
 			goto retry;
 		}
@@ -186,12 +186,13 @@ retry:
  * can't get swapped to another task.  This also increments its
  * reference count so that the context can't get freed.
  */
-static struct perf_event_context *perf_pin_task_context(struct task_struct *task)
+static struct perf_event_context *
+perf_pin_task_context(struct task_struct *task, int ctxn)
 {
 	struct perf_event_context *ctx;
 	unsigned long flags;
 
-	ctx = perf_lock_task_context(task, &flags);
+	ctx = perf_lock_task_context(task, ctxn, &flags);
 	if (ctx) {
 		++ctx->pin_count;
 		raw_spin_unlock_irqrestore(&ctx->lock, flags);
@@ -1179,28 +1180,15 @@ static void perf_event_sync_stat(struct perf_event_context *ctx,
 	}
 }
 
-/*
- * Called from scheduler to remove the events of the current task,
- * with interrupts disabled.
- *
- * We stop each event and update the event value in event->count.
- *
- * This does not protect us against NMI, but disable()
- * sets the disabled bit in the control field of event _before_
- * accessing the event control register. If a NMI hits, then it will
- * not restart the event.
- */
-void perf_event_task_sched_out(struct task_struct *task,
-				 struct task_struct *next)
+void perf_event_context_sched_out(struct task_struct *task, int ctxn,
+				  struct task_struct *next)
 {
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_event_context *ctx = task->perf_event_ctxp[ctxn];
 	struct perf_event_context *next_ctx;
 	struct perf_event_context *parent;
 	struct perf_cpu_context *cpuctx;
 	int do_switch = 1;
 
-	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
-
 	if (likely(!ctx))
 		return;
 
@@ -1210,7 +1198,7 @@ void perf_event_task_sched_out(struct task_struct *task,
 
 	rcu_read_lock();
 	parent = rcu_dereference(ctx->parent_ctx);
-	next_ctx = next->perf_event_ctxp;
+	next_ctx = next->perf_event_ctxp[ctxn];
 	if (parent && next_ctx &&
 	    rcu_dereference(next_ctx->parent_ctx) == parent) {
 		/*
@@ -1229,8 +1217,8 @@ void perf_event_task_sched_out(struct task_struct *task,
 			 * XXX do we need a memory barrier of sorts
 			 * wrt to rcu_dereference() of perf_event_ctxp
 			 */
-			task->perf_event_ctxp = next_ctx;
-			next->perf_event_ctxp = ctx;
+			task->perf_event_ctxp[ctxn] = next_ctx;
+			next->perf_event_ctxp[ctxn] = ctx;
 			ctx->task = next;
 			next_ctx->task = task;
 			do_switch = 0;
@@ -1248,6 +1236,31 @@ void perf_event_task_sched_out(struct task_struct *task,
 	}
 }
 
+#define for_each_task_context_nr(ctxn)					\
+	for ((ctxn) = 0; (ctxn) < perf_nr_task_contexts; (ctxn)++)
+
+/*
+ * Called from scheduler to remove the events of the current task,
+ * with interrupts disabled.
+ *
+ * We stop each event and update the event value in event->count.
+ *
+ * This does not protect us against NMI, but disable()
+ * sets the disabled bit in the control field of event _before_
+ * accessing the event control register. If a NMI hits, then it will
+ * not restart the event.
+ */
+void perf_event_task_sched_out(struct task_struct *task,
+			       struct task_struct *next)
+{
+	int ctxn;
+
+	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
+
+	for_each_task_context_nr(ctxn)
+		perf_event_context_sched_out(task, ctxn, next);
+}
+
 static void task_ctx_sched_out(struct perf_event_context *ctx,
 			       enum event_type_t event_type)
 {
@@ -1366,38 +1379,23 @@ static void cpu_ctx_sched_in(struct perf_cpu_context *cpuctx,
 	ctx_sched_in(ctx, cpuctx, event_type);
 }
 
-static void task_ctx_sched_in(struct task_struct *task,
+static void task_ctx_sched_in(struct perf_event_context *ctx,
 			      enum event_type_t event_type)
 {
-	struct perf_event_context *ctx = task->perf_event_ctxp;
-	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
+	struct perf_cpu_context *cpuctx;
 
-	if (likely(!ctx))
-		return;
+       	cpuctx = __get_cpu_context(ctx);
 	if (cpuctx->task_ctx == ctx)
 		return;
+
 	ctx_sched_in(ctx, cpuctx, event_type);
 	cpuctx->task_ctx = ctx;
 }
-/*
- * Called from scheduler to add the events of the current task
- * with interrupts disabled.
- *
- * We restore the event value and then enable it.
- *
- * This does not protect us against NMI, but enable()
- * sets the enabled bit in the control field of event _before_
- * accessing the event control register. If a NMI hits, then it will
- * keep the event running.
- */
-void perf_event_task_sched_in(struct task_struct *task)
+
+void perf_event_context_sched_in(struct perf_event_context *ctx)
 {
-	struct perf_event_context *ctx = task->perf_event_ctxp;
 	struct perf_cpu_context *cpuctx;
 
-	if (likely(!ctx))
-		return;
-
 	cpuctx = __get_cpu_context(ctx);
 	if (cpuctx->task_ctx == ctx)
 		return;
@@ -1422,6 +1420,31 @@ void perf_event_task_sched_in(struct task_struct *task)
 	perf_pmu_rotate_start(ctx->pmu);
 }
 
+/*
+ * Called from scheduler to add the events of the current task
+ * with interrupts disabled.
+ *
+ * We restore the event value and then enable it.
+ *
+ * This does not protect us against NMI, but enable()
+ * sets the enabled bit in the control field of event _before_
+ * accessing the event control register. If a NMI hits, then it will
+ * keep the event running.
+ */
+void perf_event_task_sched_in(struct task_struct *task)
+{
+	struct perf_event_context *ctx;
+	int ctxn;
+
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (likely(!ctx))
+			continue;
+
+		perf_event_context_sched_in(ctx);
+	}
+}
+
 #define MAX_INTERRUPTS (~0ULL)
 
 static void perf_log_throttle(struct perf_event *event, int enable);
@@ -1588,7 +1611,7 @@ static enum hrtimer_restart perf_event_context_tick(struct hrtimer *timer)
 {
 	enum hrtimer_restart restart = HRTIMER_NORESTART;
 	struct perf_cpu_context *cpuctx;
-	struct perf_event_context *ctx;
+	struct perf_event_context *ctx = NULL;
 	int rotate = 0;
 
 	cpuctx = container_of(timer, struct perf_cpu_context, timer);
@@ -1599,7 +1622,7 @@ static enum hrtimer_restart perf_event_context_tick(struct hrtimer *timer)
 			rotate = 1;
 	}
 
-	ctx = current->perf_event_ctxp;
+	ctx = cpuctx->task_ctx;
 	if (ctx && ctx->nr_events) {
 		restart = HRTIMER_RESTART;
 		if (ctx->nr_events != ctx->nr_active)
@@ -1623,7 +1646,7 @@ static enum hrtimer_restart perf_event_context_tick(struct hrtimer *timer)
 
 	cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
-		task_ctx_sched_in(current, EVENT_FLEXIBLE);
+		task_ctx_sched_in(ctx, EVENT_FLEXIBLE);
 
 done:
 	hrtimer_forward_now(timer, ns_to_ktime(cpuctx->timer_interval));
@@ -1650,20 +1673,18 @@ static int event_enable_on_exec(struct perf_event *event,
  * Enable all of a task's events that have been marked enable-on-exec.
  * This expects task == current.
  */
-static void perf_event_enable_on_exec(struct task_struct *task)
+static void perf_event_enable_on_exec(struct perf_event_context *ctx)
 {
-	struct perf_event_context *ctx;
 	struct perf_event *event;
 	unsigned long flags;
 	int enabled = 0;
 	int ret;
 
 	local_irq_save(flags);
-	ctx = task->perf_event_ctxp;
 	if (!ctx || !ctx->nr_events)
 		goto out;
 
-	__perf_event_task_sched_out(ctx);
+	task_ctx_sched_out(ctx, EVENT_ALL);
 
 	raw_spin_lock(&ctx->lock);
 
@@ -1687,7 +1708,7 @@ static void perf_event_enable_on_exec(struct task_struct *task)
 
 	raw_spin_unlock(&ctx->lock);
 
-	perf_event_task_sched_in(task);
+	perf_event_context_sched_in(ctx);
 out:
 	local_irq_restore(flags);
 }
@@ -1995,7 +2016,7 @@ find_get_context(struct pmu *pmu, pid_t pid, int cpu)
 	struct perf_cpu_context *cpuctx;
 	struct task_struct *task;
 	unsigned long flags;
-	int err;
+	int ctxn, err;
 
 	if (pid == -1 && cpu != -1) {
 		/* Must be root to operate on a CPU event: */
@@ -2044,8 +2065,13 @@ find_get_context(struct pmu *pmu, pid_t pid, int cpu)
 	if (!ptrace_may_access(task, PTRACE_MODE_READ))
 		goto errout;
 
+	err = -EINVAL;
+	ctxn = pmu->task_ctx_nr;
+	if (ctxn < 0)
+		goto errout;
+
 retry:
-	ctx = perf_lock_task_context(task, &flags);
+	ctx = perf_lock_task_context(task, ctxn, &flags);
 	if (ctx) {
 		unclone_ctx(ctx);
 		raw_spin_unlock_irqrestore(&ctx->lock, flags);
@@ -2059,7 +2085,7 @@ retry:
 
 		get_ctx(ctx);
 
-		if (cmpxchg(&task->perf_event_ctxp, NULL, ctx)) {
+		if (cmpxchg(&task->perf_event_ctxp[ctxn], NULL, ctx)) {
 			/*
 			 * We raced with some other task; use
 			 * the context they set.
@@ -3773,19 +3799,26 @@ static void perf_event_task_ctx(struct perf_event_context *ctx,
 
 static void perf_event_task_event(struct perf_task_event *task_event)
 {
-	struct perf_event_context *ctx = task_event->task_ctx;
 	struct perf_cpu_context *cpuctx;
+	struct perf_event_context *ctx;
 	struct pmu *pmu;
+	int ctxn;
 
 	rcu_read_lock_sched();
 	list_for_each_entry_rcu(pmu, &pmus, entry) {
 		cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
 		perf_event_task_ctx(&cpuctx->ctx, task_event);
+
+		ctx = task_event->task_ctx;
+		if (!ctx) {
+			ctxn = pmu->task_ctx_nr;
+			if (ctxn < 0)
+				continue;
+			ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		}
+		if (ctx)
+			perf_event_task_ctx(ctx, task_event);
 	}
-	if (!ctx)
-		ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_task_ctx(ctx, task_event);
 	rcu_read_unlock_sched();
 }
 
@@ -3890,9 +3923,10 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event)
 {
 	struct perf_cpu_context *cpuctx;
 	struct perf_event_context *ctx;
+	char comm[TASK_COMM_LEN];
 	unsigned int size;
 	struct pmu *pmu;
-	char comm[TASK_COMM_LEN];
+	int ctxn;
 
 	memset(comm, 0, sizeof(comm));
 	strlcpy(comm, comm_event->task->comm, sizeof(comm));
@@ -3907,19 +3941,31 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event)
 	list_for_each_entry_rcu(pmu, &pmus, entry) {
 		cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
 		perf_event_comm_ctx(&cpuctx->ctx, comm_event);
+
+		ctxn = pmu->task_ctx_nr;
+		if (ctxn < 0)
+			continue;
+
+		ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		if (ctx)
+			perf_event_comm_ctx(ctx, comm_event);
 	}
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_comm_ctx(ctx, comm_event);
 	rcu_read_unlock_sched();
 }
 
 void perf_event_comm(struct task_struct *task)
 {
 	struct perf_comm_event comm_event;
+	struct perf_event_context *ctx;
+	int ctxn;
 
-	if (task->perf_event_ctxp)
-		perf_event_enable_on_exec(task);
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (!ctx)
+			continue;
+
+		perf_event_enable_on_exec(ctx);
+	}
 
 	if (!atomic_read(&nr_comm_events))
 		return;
@@ -4022,6 +4068,7 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 	char *buf = NULL;
 	const char *name;
 	struct pmu *pmu;
+	int ctxn;
 
 	memset(tmp, 0, sizeof(tmp));
 
@@ -4078,10 +4125,17 @@ got_name:
 		cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
 		perf_event_mmap_ctx(&cpuctx->ctx, mmap_event,
 					vma->vm_flags & VM_EXEC);
+
+		ctxn = pmu->task_ctx_nr;
+		if (ctxn < 0)
+			continue;
+
+		ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		if (ctx) {
+			perf_event_mmap_ctx(ctx, mmap_event,
+					vma->vm_flags & VM_EXEC);
+		}
 	}
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_mmap_ctx(ctx, mmap_event, vma->vm_flags & VM_EXEC);
 	rcu_read_unlock_sched();
 
 	kfree(buf);
@@ -5042,6 +5096,43 @@ static void perf_pmu_cancel_txn(struct pmu *pmu)
 	perf_pmu_enable(pmu);
 }
 
+/*
+ * Ensures all contexts with the same task_ctx_nr have the same
+ * pmu_cpu_context too.
+ */
+static void *find_pmu_context(int ctxn)
+{
+	struct pmu *pmu;
+
+	if (ctxn < 0)
+		return NULL;
+
+	list_for_each_entry(pmu, &pmus, entry) {
+		if (pmu->task_ctx_nr == ctxn)
+			return pmu->pmu_cpu_context;
+	}
+
+	return NULL;
+}
+
+static void free_pmu_context(void * __percpu cpu_context)
+{
+	struct pmu *pmu;
+
+	mutex_lock(&pmus_lock);
+	/*
+	 * Like a real lame refcount.
+	 */
+	list_for_each_entry(pmu, &pmus, entry) {
+		if (pmu->pmu_cpu_context == cpu_context)
+			goto out;
+	}
+
+	free_percpu(cpu_context);
+out:
+	mutex_unlock(&pmus_lock);
+}
+
 int perf_pmu_register(struct pmu *pmu)
 {
 	int cpu, ret;
@@ -5052,6 +5143,10 @@ int perf_pmu_register(struct pmu *pmu)
 	if (!pmu->pmu_disable_count)
 		goto unlock;
 
+	pmu->pmu_cpu_context = find_pmu_context(pmu->task_ctx_nr);
+	if (pmu->pmu_cpu_context)
+		goto got_cpu_context;
+
 	pmu->pmu_cpu_context = alloc_percpu(struct perf_cpu_context);
 	if (!pmu->pmu_cpu_context)
 		goto free_pdc;
@@ -5067,6 +5162,7 @@ int perf_pmu_register(struct pmu *pmu)
 		cpuctx->timer.function = perf_event_context_tick;
 	}
 
+got_cpu_context:
 	if (!pmu->start_txn) {
 		if (pmu->pmu_enable) {
 			/*
@@ -5114,7 +5210,7 @@ void perf_pmu_unregister(struct pmu *pmu)
 	synchronize_srcu(&pmus_srcu);
 
 	free_percpu(pmu->pmu_disable_count);
-	free_percpu(pmu->pmu_cpu_context);
+	free_pmu_context(pmu->pmu_cpu_context);
 }
 
 struct pmu *perf_init_event(struct perf_event *event)
@@ -5628,16 +5724,13 @@ __perf_event_exit_task(struct perf_event *child_event,
 	}
 }
 
-/*
- * When a child task exits, feed back event values to parent events.
- */
-void perf_event_exit_task(struct task_struct *child)
+static void perf_event_exit_task_context(struct task_struct *child, int ctxn)
 {
 	struct perf_event *child_event, *tmp;
 	struct perf_event_context *child_ctx;
 	unsigned long flags;
 
-	if (likely(!child->perf_event_ctxp)) {
+	if (likely(!child->perf_event_ctxp[ctxn])) {
 		perf_event_task(child, NULL, 0);
 		return;
 	}
@@ -5649,7 +5742,7 @@ void perf_event_exit_task(struct task_struct *child)
 	 * scheduled, so we are now safe from rescheduling changing
 	 * our context.
 	 */
-	child_ctx = child->perf_event_ctxp;
+	child_ctx = child->perf_event_ctxp[ctxn];
 	__perf_event_task_sched_out(child_ctx);
 
 	/*
@@ -5658,7 +5751,7 @@ void perf_event_exit_task(struct task_struct *child)
 	 * incremented the context's refcount before we do put_ctx below.
 	 */
 	raw_spin_lock(&child_ctx->lock);
-	child->perf_event_ctxp = NULL;
+	child->perf_event_ctxp[ctxn] = NULL;
 	/*
 	 * If this context is a clone; unclone it so it can't get
 	 * swapped to another process while we're removing all
@@ -5711,6 +5804,17 @@ again:
 	put_ctx(child_ctx);
 }
 
+/*
+ * When a child task exits, feed back event values to parent events.
+ */
+void perf_event_exit_task(struct task_struct *child)
+{
+	int ctxn;
+
+	for_each_task_context_nr(ctxn)
+		perf_event_exit_task_context(child, ctxn);
+}
+
 static void perf_free_event(struct perf_event *event,
 			    struct perf_event_context *ctx)
 {
@@ -5732,32 +5836,37 @@ static void perf_free_event(struct perf_event *event,
 
 /*
  * free an unexposed, unused context as created by inheritance by
- * init_task below, used by fork() in case of fail.
+ * perf_event_init_task below, used by fork() in case of fail.
  */
 void perf_event_free_task(struct task_struct *task)
 {
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_event_context *ctx;
 	struct perf_event *event, *tmp;
+	int ctxn;
 
-	if (!ctx)
-		return;
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (!ctx)
+			continue;
 
-	mutex_lock(&ctx->mutex);
+		mutex_lock(&ctx->mutex);
 again:
-	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
-		perf_free_event(event, ctx);
+		list_for_each_entry_safe(event, tmp, &ctx->pinned_groups,
+				group_entry)
+			perf_free_event(event, ctx);
 
-	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups,
-				 group_entry)
-		perf_free_event(event, ctx);
+		list_for_each_entry_safe(event, tmp, &ctx->flexible_groups,
+				group_entry)
+			perf_free_event(event, ctx);
 
-	if (!list_empty(&ctx->pinned_groups) ||
-	    !list_empty(&ctx->flexible_groups))
-		goto again;
+		if (!list_empty(&ctx->pinned_groups) ||
+				!list_empty(&ctx->flexible_groups))
+			goto again;
 
-	mutex_unlock(&ctx->mutex);
+		mutex_unlock(&ctx->mutex);
 
-	put_ctx(ctx);
+		put_ctx(ctx);
+	}
 }
 
 /*
@@ -5863,17 +5972,18 @@ static int inherit_group(struct perf_event *parent_event,
 static int
 inherit_task_group(struct perf_event *event, struct task_struct *parent,
 		   struct perf_event_context *parent_ctx,
-		   struct task_struct *child,
+		   struct task_struct *child, int ctxn,
 		   int *inherited_all)
 {
 	int ret;
-	struct perf_event_context *child_ctx = child->perf_event_ctxp;
+	struct perf_event_context *child_ctx;
 
 	if (!event->attr.inherit) {
 		*inherited_all = 0;
 		return 0;
 	}
 
+       	child_ctx = child->perf_event_ctxp[ctxn];
 	if (!child_ctx) {
 		/*
 		 * This is executed from the parent task context, so
@@ -5886,7 +5996,7 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent,
 		if (!child_ctx)
 			return -ENOMEM;
 
-		child->perf_event_ctxp = child_ctx;
+		child->perf_event_ctxp[ctxn] = child_ctx;
 	}
 
 	ret = inherit_group(event, parent, parent_ctx,
@@ -5901,7 +6011,7 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent,
 /*
  * Initialize the perf_event context in task_struct
  */
-int perf_event_init_task(struct task_struct *child)
+int perf_event_init_context(struct task_struct *child, int ctxn)
 {
 	struct perf_event_context *child_ctx, *parent_ctx;
 	struct perf_event_context *cloned_ctx;
@@ -5910,19 +6020,19 @@ int perf_event_init_task(struct task_struct *child)
 	int inherited_all = 1;
 	int ret = 0;
 
-	child->perf_event_ctxp = NULL;
+	child->perf_event_ctxp[ctxn] = NULL;
 
 	mutex_init(&child->perf_event_mutex);
 	INIT_LIST_HEAD(&child->perf_event_list);
 
-	if (likely(!parent->perf_event_ctxp))
+	if (likely(!parent->perf_event_ctxp[ctxn]))
 		return 0;
 
 	/*
 	 * If the parent's context is a clone, pin it so it won't get
 	 * swapped under us.
 	 */
-	parent_ctx = perf_pin_task_context(parent);
+	parent_ctx = perf_pin_task_context(parent, ctxn);
 
 	/*
 	 * No need to check if parent_ctx != NULL here; since we saw
@@ -5942,20 +6052,20 @@ int perf_event_init_task(struct task_struct *child)
 	 * the list, not manipulating it:
 	 */
 	list_for_each_entry(event, &parent_ctx->pinned_groups, group_entry) {
-		ret = inherit_task_group(event, parent, parent_ctx, child,
-					 &inherited_all);
+		ret = inherit_task_group(event, parent, parent_ctx,
+					 child, ctxn, &inherited_all);
 		if (ret)
 			break;
 	}
 
 	list_for_each_entry(event, &parent_ctx->flexible_groups, group_entry) {
-		ret = inherit_task_group(event, parent, parent_ctx, child,
-					 &inherited_all);
+		ret = inherit_task_group(event, parent, parent_ctx,
+					 child, ctxn, &inherited_all);
 		if (ret)
 			break;
 	}
 
-	child_ctx = child->perf_event_ctxp;
+	child_ctx = child->perf_event_ctxp[ctxn];
 
 	if (child_ctx && inherited_all) {
 		/*
@@ -5984,6 +6094,22 @@ int perf_event_init_task(struct task_struct *child)
 	return ret;
 }
 
+/*
+ * Initialize the perf_event context in task_struct
+ */
+int perf_event_init_task(struct task_struct *child)
+{
+	int ctxn, ret;
+
+	for_each_task_context_nr(ctxn) {
+		ret = perf_event_init_context(child, ctxn);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static void __init perf_event_init_all_cpus(void)
 {
 	struct swevent_htable *swhash;
-- 
cgit v1.2.3


From 89a1e18731959e9953fae15ddc1a983eb15a4f19 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Tue, 7 Sep 2010 17:34:50 +0200
Subject: perf: Provide a separate task context for swevents

Since software events are always schedulable, mixing them up with
hardware events (who are not) can lead to funny scheduling oddities.

Giving them their own context solves this.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus <paulus@samba.org>
Cc: stephane eranian <eranian@googlemail.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Lin Ming <ming.m.lin@intel.com>
Cc: Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |  9 +--------
 include/linux/sched.h      |  1 +
 kernel/hw_breakpoint.c     |  2 ++
 kernel/perf_event.c        | 40 +++++++++++++++++++++++++++++-----------
 4 files changed, 33 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 9ecfd856ce6e..c1173520f14d 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -952,14 +952,7 @@ extern int perf_event_overflow(struct perf_event *event, int nmi,
  */
 static inline int is_software_event(struct perf_event *event)
 {
-	switch (event->attr.type) {
-	case PERF_TYPE_SOFTWARE:
-	case PERF_TYPE_TRACEPOINT:
-	/* for now the breakpoint stuff also works as software event */
-	case PERF_TYPE_BREAKPOINT:
-		return 1;
-	}
-	return 0;
+	return event->pmu->task_ctx_nr == perf_sw_context;
 }
 
 extern atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 89d6023c6f82..eb3c1ceec06e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1163,6 +1163,7 @@ struct rcu_node;
 enum perf_event_task_context {
 	perf_invalid_context = -1,
 	perf_hw_context = 0,
+	perf_sw_context,
 	perf_nr_task_contexts,
 };
 
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index 6f150095cafe..3b2aaffb65f0 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -610,6 +610,8 @@ static void hw_breakpoint_stop(struct perf_event *bp, int flags)
 }
 
 static struct pmu perf_breakpoint = {
+	.task_ctx_nr	= perf_sw_context, /* could eventually get its own */
+
 	.event_init	= hw_breakpoint_event_init,
 	.add		= hw_breakpoint_add,
 	.del		= hw_breakpoint_del,
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 7223ea875861..357ee8d5e8ae 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -4709,6 +4709,8 @@ static int perf_swevent_init(struct perf_event *event)
 }
 
 static struct pmu perf_swevent = {
+	.task_ctx_nr	= perf_sw_context,
+
 	.event_init	= perf_swevent_init,
 	.add		= perf_swevent_add,
 	.del		= perf_swevent_del,
@@ -4800,6 +4802,8 @@ static int perf_tp_event_init(struct perf_event *event)
 }
 
 static struct pmu perf_tracepoint = {
+	.task_ctx_nr	= perf_sw_context,
+
 	.event_init	= perf_tp_event_init,
 	.add		= perf_trace_add,
 	.del		= perf_trace_del,
@@ -4988,6 +4992,8 @@ static int cpu_clock_event_init(struct perf_event *event)
 }
 
 static struct pmu perf_cpu_clock = {
+	.task_ctx_nr	= perf_sw_context,
+
 	.event_init	= cpu_clock_event_init,
 	.add		= cpu_clock_event_add,
 	.del		= cpu_clock_event_del,
@@ -5063,6 +5069,8 @@ static int task_clock_event_init(struct perf_event *event)
 }
 
 static struct pmu perf_task_clock = {
+	.task_ctx_nr	= perf_sw_context,
+
 	.event_init	= task_clock_event_init,
 	.add		= task_clock_event_add,
 	.del		= task_clock_event_del,
@@ -5490,6 +5498,7 @@ SYSCALL_DEFINE5(perf_event_open,
 	struct perf_event_context *ctx;
 	struct file *event_file = NULL;
 	struct file *group_file = NULL;
+	struct pmu *pmu;
 	int event_fd;
 	int fput_needed = 0;
 	int err;
@@ -5522,20 +5531,11 @@ SYSCALL_DEFINE5(perf_event_open,
 		goto err_fd;
 	}
 
-	/*
-	 * Get the target context (task or percpu):
-	 */
-	ctx = find_get_context(event->pmu, pid, cpu);
-	if (IS_ERR(ctx)) {
-		err = PTR_ERR(ctx);
-		goto err_alloc;
-	}
-
 	if (group_fd != -1) {
 		group_leader = perf_fget_light(group_fd, &fput_needed);
 		if (IS_ERR(group_leader)) {
 			err = PTR_ERR(group_leader);
-			goto err_context;
+			goto err_alloc;
 		}
 		group_file = group_leader->filp;
 		if (flags & PERF_FLAG_FD_OUTPUT)
@@ -5544,6 +5544,23 @@ SYSCALL_DEFINE5(perf_event_open,
 			group_leader = NULL;
 	}
 
+	/*
+	 * Special case software events and allow them to be part of
+	 * any hardware group.
+	 */
+	pmu = event->pmu;
+	if ((pmu->task_ctx_nr == perf_sw_context) && group_leader)
+		pmu = group_leader->pmu;
+
+	/*
+	 * Get the target context (task or percpu):
+	 */
+	ctx = find_get_context(pmu, pid, cpu);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto err_group_fd;
+	}
+
 	/*
 	 * Look up the group leader (we will attach this event to it):
 	 */
@@ -5605,8 +5622,9 @@ SYSCALL_DEFINE5(perf_event_open,
 	return event_fd;
 
 err_context:
-	fput_light(group_file, fput_needed);
 	put_ctx(ctx);
+err_group_fd:
+	fput_light(group_file, fput_needed);
 err_alloc:
 	free_event(event);
 err_fd:
-- 
cgit v1.2.3


From 4e231c7962ce711c7d8c2a4dc23ecd1e8fc28363 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 9 Sep 2010 21:01:59 +0200
Subject: perf: Fix up delayed_put_task_struct()

I missed a perf_event_ctxp user when converting it to an array. Pull this
last user into perf_event.c as well and fix it up.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h | 2 ++
 kernel/exit.c              | 4 +---
 kernel/perf_event.c        | 8 ++++++++
 3 files changed, 11 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index c1173520f14d..93bf53aa50e5 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -889,6 +889,7 @@ extern void perf_event_task_sched_out(struct task_struct *task, struct task_stru
 extern int perf_event_init_task(struct task_struct *child);
 extern void perf_event_exit_task(struct task_struct *child);
 extern void perf_event_free_task(struct task_struct *task);
+extern void perf_event_delayed_put(struct task_struct *task);
 extern void set_perf_event_pending(void);
 extern void perf_event_do_pending(void);
 extern void perf_event_print_debug(void);
@@ -1067,6 +1068,7 @@ perf_event_task_sched_out(struct task_struct *task,
 static inline int perf_event_init_task(struct task_struct *child)	{ return 0; }
 static inline void perf_event_exit_task(struct task_struct *child)	{ }
 static inline void perf_event_free_task(struct task_struct *task)	{ }
+static inline void perf_event_delayed_put(struct task_struct *task)	{ }
 static inline void perf_event_do_pending(void)				{ }
 static inline void perf_event_print_debug(void)				{ }
 static inline int perf_event_task_disable(void)				{ return -EINVAL; }
diff --git a/kernel/exit.c b/kernel/exit.c
index 03120229db28..e2bdf37f9fde 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -149,9 +149,7 @@ static void delayed_put_task_struct(struct rcu_head *rhp)
 {
 	struct task_struct *tsk = container_of(rhp, struct task_struct, rcu);
 
-#ifdef CONFIG_PERF_EVENTS
-	WARN_ON_ONCE(tsk->perf_event_ctxp);
-#endif
+	perf_event_delayed_put(tsk);
 	trace_sched_process_free(tsk);
 	put_task_struct(tsk);
 }
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 9819a69a61a1..eaf1c5de6dcc 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -5893,6 +5893,14 @@ again:
 	}
 }
 
+void perf_event_delayed_put(struct task_struct *task)
+{
+	int ctxn;
+
+	for_each_task_context_nr(ctxn)
+		WARN_ON_ONCE(task->perf_event_ctxp[ctxn]);
+}
+
 /*
  * inherit a event from parent task to child task:
  */
-- 
cgit v1.2.3


From 0da2f50944976e890ccc9436ab88c0da87788d02 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 11:56:16 +0200
Subject: ide: remove unnecessary blk_queue_flushing() test in do_ide_request()

Unplugging from a request function doesn't really help much (it's
already in the request_fn) and soon block layer will be updated to mix
barrier sequence with other commands, so there's no need to treat
queue flushing any differently.

ide was the only user of blk_queue_flushing().  Remove it.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 drivers/ide/ide-io.c   | 13 -------------
 include/linux/blkdev.h |  1 -
 2 files changed, 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index a381be814070..999dac054bcc 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -441,19 +441,6 @@ void do_ide_request(struct request_queue *q)
 	struct request	*rq = NULL;
 	ide_startstop_t	startstop;
 
-	/*
-	 * drive is doing pre-flush, ordered write, post-flush sequence. even
-	 * though that is 3 requests, it must be seen as a single transaction.
-	 * we must not preempt this drive until that is complete
-	 */
-	if (blk_queue_flushing(q))
-		/*
-		 * small race where queue could get replugged during
-		 * the 3-request flush cycle, just yank the plug since
-		 * we want it to finish asap
-		 */
-		blk_remove_plug(q);
-
 	spin_unlock_irq(q->queue_lock);
 
 	/* HLD do_request() callback might sleep, make sure it's okay */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 2c54906f678f..015375c7d031 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -521,7 +521,6 @@ enum {
 #define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
 #define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
 #define blk_queue_add_random(q)	test_bit(QUEUE_FLAG_ADD_RANDOM, &(q)->queue_flags)
-#define blk_queue_flushing(q)	((q)->ordseq)
 #define blk_queue_stackable(q)	\
 	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
 #define blk_queue_discard(q)	test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags)
-- 
cgit v1.2.3


From 6958f145459ca7ad9715024de97445addacb8510 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 11:56:16 +0200
Subject: block: kill QUEUE_ORDERED_BY_TAG

Nobody is making meaningful use of ORDERED_BY_TAG now and queue
draining for barrier requests will be removed soon which will render
the advantage of tag ordering moot.  Kill ORDERED_BY_TAG.  The
following users are affected.

* brd: converted to ORDERED_DRAIN.
* virtio_blk: ORDERED_TAG path was already marked deprecated.  Removed.
* xen-blkfront: ORDERED_TAG case dropped.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-barrier.c          | 35 +++++++----------------------------
 drivers/block/brd.c          |  2 +-
 drivers/block/virtio_blk.c   |  9 ---------
 drivers/block/xen-blkfront.c |  8 +++-----
 drivers/scsi/sd.c            |  4 +---
 include/linux/blkdev.h       | 17 +----------------
 6 files changed, 13 insertions(+), 62 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-barrier.c b/block/blk-barrier.c
index f0faefca032f..c807e9ca3a68 100644
--- a/block/blk-barrier.c
+++ b/block/blk-barrier.c
@@ -26,10 +26,7 @@ int blk_queue_ordered(struct request_queue *q, unsigned ordered)
 	if (ordered != QUEUE_ORDERED_NONE &&
 	    ordered != QUEUE_ORDERED_DRAIN &&
 	    ordered != QUEUE_ORDERED_DRAIN_FLUSH &&
-	    ordered != QUEUE_ORDERED_DRAIN_FUA &&
-	    ordered != QUEUE_ORDERED_TAG &&
-	    ordered != QUEUE_ORDERED_TAG_FLUSH &&
-	    ordered != QUEUE_ORDERED_TAG_FUA) {
+	    ordered != QUEUE_ORDERED_DRAIN_FUA) {
 		printk(KERN_ERR "blk_queue_ordered: bad value %d\n", ordered);
 		return -EINVAL;
 	}
@@ -155,21 +152,9 @@ static inline bool start_ordered(struct request_queue *q, struct request **rqp)
 	 * For an empty barrier, there's no actual BAR request, which
 	 * in turn makes POSTFLUSH unnecessary.  Mask them off.
 	 */
-	if (!blk_rq_sectors(rq)) {
+	if (!blk_rq_sectors(rq))
 		q->ordered &= ~(QUEUE_ORDERED_DO_BAR |
 				QUEUE_ORDERED_DO_POSTFLUSH);
-		/*
-		 * Empty barrier on a write-through device w/ ordered
-		 * tag has no command to issue and without any command
-		 * to issue, ordering by tag can't be used.  Drain
-		 * instead.
-		 */
-		if ((q->ordered & QUEUE_ORDERED_BY_TAG) &&
-		    !(q->ordered & QUEUE_ORDERED_DO_PREFLUSH)) {
-			q->ordered &= ~QUEUE_ORDERED_BY_TAG;
-			q->ordered |= QUEUE_ORDERED_BY_DRAIN;
-		}
-	}
 
 	/* stash away the original request */
 	blk_dequeue_request(rq);
@@ -210,7 +195,7 @@ static inline bool start_ordered(struct request_queue *q, struct request **rqp)
 	} else
 		skip |= QUEUE_ORDSEQ_PREFLUSH;
 
-	if ((q->ordered & QUEUE_ORDERED_BY_DRAIN) && queue_in_flight(q))
+	if (queue_in_flight(q))
 		rq = NULL;
 	else
 		skip |= QUEUE_ORDSEQ_DRAIN;
@@ -257,16 +242,10 @@ bool blk_do_ordered(struct request_queue *q, struct request **rqp)
 	    rq != &q->pre_flush_rq && rq != &q->post_flush_rq)
 		return true;
 
-	if (q->ordered & QUEUE_ORDERED_BY_TAG) {
-		/* Ordered by tag.  Blocking the next barrier is enough. */
-		if (is_barrier && rq != &q->bar_rq)
-			*rqp = NULL;
-	} else {
-		/* Ordered by draining.  Wait for turn. */
-		WARN_ON(blk_ordered_req_seq(rq) < blk_ordered_cur_seq(q));
-		if (blk_ordered_req_seq(rq) > blk_ordered_cur_seq(q))
-			*rqp = NULL;
-	}
+	/* Ordered by draining.  Wait for turn. */
+	WARN_ON(blk_ordered_req_seq(rq) < blk_ordered_cur_seq(q));
+	if (blk_ordered_req_seq(rq) > blk_ordered_cur_seq(q))
+		*rqp = NULL;
 
 	return true;
 }
diff --git a/drivers/block/brd.c b/drivers/block/brd.c
index 1c7f63792ff8..47a41272d26b 100644
--- a/drivers/block/brd.c
+++ b/drivers/block/brd.c
@@ -482,7 +482,7 @@ static struct brd_device *brd_alloc(int i)
 	if (!brd->brd_queue)
 		goto out_free_dev;
 	blk_queue_make_request(brd->brd_queue, brd_make_request);
-	blk_queue_ordered(brd->brd_queue, QUEUE_ORDERED_TAG);
+	blk_queue_ordered(brd->brd_queue, QUEUE_ORDERED_DRAIN);
 	blk_queue_max_hw_sectors(brd->brd_queue, 1024);
 	blk_queue_bounce_limit(brd->brd_queue, BLK_BOUNCE_ANY);
 
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 2aafafca2b13..79652809eee8 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -395,15 +395,6 @@ static int __devinit virtblk_probe(struct virtio_device *vdev)
 		 * to implement write barrier support.
 		 */
 		blk_queue_ordered(q, QUEUE_ORDERED_DRAIN_FLUSH);
-	} else if (virtio_has_feature(vdev, VIRTIO_BLK_F_BARRIER)) {
-		/*
-		 * If the BARRIER feature is supported the host expects us
-		 * to order request by tags.  This implies there is not
-		 * volatile write cache on the host, and that the host
-		 * never re-orders outstanding I/O.  This feature is not
-		 * useful for real life scenarious and deprecated.
-		 */
-		blk_queue_ordered(q, QUEUE_ORDERED_TAG);
 	} else {
 		/*
 		 * If the FLUSH feature is not supported we must assume that
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index ac1b682edecb..50ec6f834996 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -424,8 +424,7 @@ static int xlvbd_barrier(struct blkfront_info *info)
 	const char *barrier;
 
 	switch (info->feature_barrier) {
-	case QUEUE_ORDERED_DRAIN:	barrier = "enabled (drain)"; break;
-	case QUEUE_ORDERED_TAG:		barrier = "enabled (tag)"; break;
+	case QUEUE_ORDERED_DRAIN:	barrier = "enabled"; break;
 	case QUEUE_ORDERED_NONE:	barrier = "disabled"; break;
 	default:			return -EINVAL;
 	}
@@ -1078,8 +1077,7 @@ static void blkfront_connect(struct blkfront_info *info)
 	 * we're dealing with a very old backend which writes
 	 * synchronously; draining will do what needs to get done.
 	 *
-	 * If there are barriers, then we can do full queued writes
-	 * with tagged barriers.
+	 * If there are barriers, then we use flush.
 	 *
 	 * If barriers are not supported, then there's no much we can
 	 * do, so just set ordering to NONE.
@@ -1087,7 +1085,7 @@ static void blkfront_connect(struct blkfront_info *info)
 	if (err)
 		info->feature_barrier = QUEUE_ORDERED_DRAIN;
 	else if (barrier)
-		info->feature_barrier = QUEUE_ORDERED_TAG;
+		info->feature_barrier = QUEUE_ORDERED_DRAIN_FLUSH;
 	else
 		info->feature_barrier = QUEUE_ORDERED_NONE;
 
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 2714becc2eaf..cdfc51ab9cf2 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -2151,9 +2151,7 @@ static int sd_revalidate_disk(struct gendisk *disk)
 
 	/*
 	 * We now have all cache related info, determine how we deal
-	 * with ordered requests.  Note that as the current SCSI
-	 * dispatch function can alter request order, we cannot use
-	 * QUEUE_ORDERED_TAG_* even when ordered tag is supported.
+	 * with ordered requests.
 	 */
 	if (sdkp->WCE)
 		ordered = sdkp->DPOFUA
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 015375c7d031..7077bc0d6138 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -470,12 +470,7 @@ enum {
 	 * DRAIN	: ordering by draining is enough
 	 * DRAIN_FLUSH	: ordering by draining w/ pre and post flushes
 	 * DRAIN_FUA	: ordering by draining w/ pre flush and FUA write
-	 * TAG		: ordering by tag is enough
-	 * TAG_FLUSH	: ordering by tag w/ pre and post flushes
-	 * TAG_FUA	: ordering by tag w/ pre flush and FUA write
 	 */
-	QUEUE_ORDERED_BY_DRAIN		= 0x01,
-	QUEUE_ORDERED_BY_TAG		= 0x02,
 	QUEUE_ORDERED_DO_PREFLUSH	= 0x10,
 	QUEUE_ORDERED_DO_BAR		= 0x20,
 	QUEUE_ORDERED_DO_POSTFLUSH	= 0x40,
@@ -483,8 +478,7 @@ enum {
 
 	QUEUE_ORDERED_NONE		= 0x00,
 
-	QUEUE_ORDERED_DRAIN		= QUEUE_ORDERED_BY_DRAIN |
-					  QUEUE_ORDERED_DO_BAR,
+	QUEUE_ORDERED_DRAIN		= QUEUE_ORDERED_DO_BAR,
 	QUEUE_ORDERED_DRAIN_FLUSH	= QUEUE_ORDERED_DRAIN |
 					  QUEUE_ORDERED_DO_PREFLUSH |
 					  QUEUE_ORDERED_DO_POSTFLUSH,
@@ -492,15 +486,6 @@ enum {
 					  QUEUE_ORDERED_DO_PREFLUSH |
 					  QUEUE_ORDERED_DO_FUA,
 
-	QUEUE_ORDERED_TAG		= QUEUE_ORDERED_BY_TAG |
-					  QUEUE_ORDERED_DO_BAR,
-	QUEUE_ORDERED_TAG_FLUSH		= QUEUE_ORDERED_TAG |
-					  QUEUE_ORDERED_DO_PREFLUSH |
-					  QUEUE_ORDERED_DO_POSTFLUSH,
-	QUEUE_ORDERED_TAG_FUA		= QUEUE_ORDERED_TAG |
-					  QUEUE_ORDERED_DO_PREFLUSH |
-					  QUEUE_ORDERED_DO_FUA,
-
 	/*
 	 * Ordered operation sequence
 	 */
-- 
cgit v1.2.3


From 4913efe456c987057e5d36a3f0a55422a9072cae Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 11:56:16 +0200
Subject: block: deprecate barrier and replace blk_queue_ordered() with
 blk_queue_flush()

Barrier is deemed too heavy and will soon be replaced by FLUSH/FUA
requests.  Deprecate barrier.  All REQ_HARDBARRIERs are failed with
-EOPNOTSUPP and blk_queue_ordered() is replaced with simpler
blk_queue_flush().

blk_queue_flush() takes combinations of REQ_FLUSH and FUA.  If a
device has write cache and can flush it, it should set REQ_FLUSH.  If
the device can handle FUA writes, it should also set REQ_FUA.

All blk_queue_ordered() users are converted.

* ORDERED_DRAIN is mapped to 0 which is the default value.
* ORDERED_DRAIN_FLUSH is mapped to REQ_FLUSH.
* ORDERED_DRAIN_FLUSH_FUA is mapped to REQ_FLUSH | REQ_FUA.

Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Boaz Harrosh <bharrosh@panasas.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Chris Wright <chrisw@sous-sol.org>
Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Alasdair G Kergon <agk@redhat.com>
Cc: Pierre Ossman <drzeus@drzeus.cx>
Cc: Stefan Weinhuber <wein@de.ibm.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-barrier.c          | 29 -----------------------------
 block/blk-core.c             |  6 ++++--
 block/blk-settings.c         | 20 ++++++++++++++++++++
 drivers/block/brd.c          |  1 -
 drivers/block/loop.c         |  2 +-
 drivers/block/osdblk.c       |  2 +-
 drivers/block/ps3disk.c      |  2 +-
 drivers/block/virtio_blk.c   | 25 +++++++++----------------
 drivers/block/xen-blkfront.c | 43 ++++++++++++-------------------------------
 drivers/ide/ide-disk.c       | 13 ++++++-------
 drivers/md/dm.c              |  2 +-
 drivers/mmc/card/queue.c     |  1 -
 drivers/s390/block/dasd.c    |  1 -
 drivers/scsi/sd.c            | 16 ++++++++--------
 include/linux/blkdev.h       |  6 ++++--
 15 files changed, 67 insertions(+), 102 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-barrier.c b/block/blk-barrier.c
index c807e9ca3a68..ed0aba5463ab 100644
--- a/block/blk-barrier.c
+++ b/block/blk-barrier.c
@@ -9,35 +9,6 @@
 
 #include "blk.h"
 
-/**
- * blk_queue_ordered - does this queue support ordered writes
- * @q:        the request queue
- * @ordered:  one of QUEUE_ORDERED_*
- *
- * Description:
- *   For journalled file systems, doing ordered writes on a commit
- *   block instead of explicitly doing wait_on_buffer (which is bad
- *   for performance) can be a big win. Block drivers supporting this
- *   feature should call this function and indicate so.
- *
- **/
-int blk_queue_ordered(struct request_queue *q, unsigned ordered)
-{
-	if (ordered != QUEUE_ORDERED_NONE &&
-	    ordered != QUEUE_ORDERED_DRAIN &&
-	    ordered != QUEUE_ORDERED_DRAIN_FLUSH &&
-	    ordered != QUEUE_ORDERED_DRAIN_FUA) {
-		printk(KERN_ERR "blk_queue_ordered: bad value %d\n", ordered);
-		return -EINVAL;
-	}
-
-	q->ordered = ordered;
-	q->next_ordered = ordered;
-
-	return 0;
-}
-EXPORT_SYMBOL(blk_queue_ordered);
-
 /*
  * Cache flushing for ordered writes handling
  */
diff --git a/block/blk-core.c b/block/blk-core.c
index ee1a1e7e63cc..f06354183b29 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1203,11 +1203,13 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 	const unsigned int ff = bio->bi_rw & REQ_FAILFAST_MASK;
 	int rw_flags;
 
-	if ((bio->bi_rw & REQ_HARDBARRIER) &&
-	    (q->next_ordered == QUEUE_ORDERED_NONE)) {
+	/* REQ_HARDBARRIER is no more */
+	if (WARN_ONCE(bio->bi_rw & REQ_HARDBARRIER,
+		"block: HARDBARRIER is deprecated, use FLUSH/FUA instead\n")) {
 		bio_endio(bio, -EOPNOTSUPP);
 		return 0;
 	}
+
 	/*
 	 * low level driver can indicate that it wants pages above a
 	 * certain limit bounced to low memory (ie for highmem, or even
diff --git a/block/blk-settings.c b/block/blk-settings.c
index a234f4bf1d6f..9b18afcfe925 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -794,6 +794,26 @@ void blk_queue_update_dma_alignment(struct request_queue *q, int mask)
 }
 EXPORT_SYMBOL(blk_queue_update_dma_alignment);
 
+/**
+ * blk_queue_flush - configure queue's cache flush capability
+ * @q:		the request queue for the device
+ * @flush:	0, REQ_FLUSH or REQ_FLUSH | REQ_FUA
+ *
+ * Tell block layer cache flush capability of @q.  If it supports
+ * flushing, REQ_FLUSH should be set.  If it supports bypassing
+ * write cache for individual writes, REQ_FUA should be set.
+ */
+void blk_queue_flush(struct request_queue *q, unsigned int flush)
+{
+	WARN_ON_ONCE(flush & ~(REQ_FLUSH | REQ_FUA));
+
+	if (WARN_ON_ONCE(!(flush & REQ_FLUSH) && (flush & REQ_FUA)))
+		flush &= ~REQ_FUA;
+
+	q->flush_flags = flush & (REQ_FLUSH | REQ_FUA);
+}
+EXPORT_SYMBOL_GPL(blk_queue_flush);
+
 static int __init blk_settings_init(void)
 {
 	blk_max_low_pfn = max_low_pfn - 1;
diff --git a/drivers/block/brd.c b/drivers/block/brd.c
index 47a41272d26b..fa33f97722ba 100644
--- a/drivers/block/brd.c
+++ b/drivers/block/brd.c
@@ -482,7 +482,6 @@ static struct brd_device *brd_alloc(int i)
 	if (!brd->brd_queue)
 		goto out_free_dev;
 	blk_queue_make_request(brd->brd_queue, brd_make_request);
-	blk_queue_ordered(brd->brd_queue, QUEUE_ORDERED_DRAIN);
 	blk_queue_max_hw_sectors(brd->brd_queue, 1024);
 	blk_queue_bounce_limit(brd->brd_queue, BLK_BOUNCE_ANY);
 
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index c3a4a2e176da..953d1e12f4d4 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -832,7 +832,7 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
 	lo->lo_queue->unplug_fn = loop_unplug;
 
 	if (!(lo_flags & LO_FLAGS_READ_ONLY) && file->f_op->fsync)
-		blk_queue_ordered(lo->lo_queue, QUEUE_ORDERED_DRAIN_FLUSH);
+		blk_queue_flush(lo->lo_queue, REQ_FLUSH);
 
 	set_capacity(lo->lo_disk, size);
 	bd_set_size(bdev, size << 9);
diff --git a/drivers/block/osdblk.c b/drivers/block/osdblk.c
index 2284b4f05c62..72d62462433d 100644
--- a/drivers/block/osdblk.c
+++ b/drivers/block/osdblk.c
@@ -439,7 +439,7 @@ static int osdblk_init_disk(struct osdblk_device *osdev)
 	blk_queue_stack_limits(q, osd_request_queue(osdev->osd));
 
 	blk_queue_prep_rq(q, blk_queue_start_tag);
-	blk_queue_ordered(q, QUEUE_ORDERED_DRAIN_FLUSH);
+	blk_queue_flush(q, REQ_FLUSH);
 
 	disk->queue = q;
 
diff --git a/drivers/block/ps3disk.c b/drivers/block/ps3disk.c
index e9da874d0419..4911f9e57bc7 100644
--- a/drivers/block/ps3disk.c
+++ b/drivers/block/ps3disk.c
@@ -468,7 +468,7 @@ static int __devinit ps3disk_probe(struct ps3_system_bus_device *_dev)
 	blk_queue_dma_alignment(queue, dev->blk_size-1);
 	blk_queue_logical_block_size(queue, dev->blk_size);
 
-	blk_queue_ordered(queue, QUEUE_ORDERED_DRAIN_FLUSH);
+	blk_queue_flush(queue, REQ_FLUSH);
 
 	blk_queue_max_segments(queue, -1);
 	blk_queue_max_segment_size(queue, dev->bounce_size);
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 79652809eee8..d10b635b3946 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -388,22 +388,15 @@ static int __devinit virtblk_probe(struct virtio_device *vdev)
 	vblk->disk->driverfs_dev = &vdev->dev;
 	index++;
 
-	if (virtio_has_feature(vdev, VIRTIO_BLK_F_FLUSH)) {
-		/*
-		 * If the FLUSH feature is supported we do have support for
-		 * flushing a volatile write cache on the host.  Use that
-		 * to implement write barrier support.
-		 */
-		blk_queue_ordered(q, QUEUE_ORDERED_DRAIN_FLUSH);
-	} else {
-		/*
-		 * If the FLUSH feature is not supported we must assume that
-		 * the host does not perform any kind of volatile write
-		 * caching. We still need to drain the queue to provider
-		 * proper barrier semantics.
-		 */
-		blk_queue_ordered(q, QUEUE_ORDERED_DRAIN);
-	}
+	/*
+	 * If the FLUSH feature is supported we do have support for
+	 * flushing a volatile write cache on the host.  Use that to
+	 * implement write barrier support; otherwise, we must assume
+	 * that the host does not perform any kind of volatile write
+	 * caching.
+	 */
+	if (virtio_has_feature(vdev, VIRTIO_BLK_F_FLUSH))
+		blk_queue_flush(q, REQ_FLUSH);
 
 	/* If disk is read-only in the host, the guest should obey */
 	if (virtio_has_feature(vdev, VIRTIO_BLK_F_RO))
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index 50ec6f834996..0b1eea643262 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -95,7 +95,7 @@ struct blkfront_info
 	struct gnttab_free_callback callback;
 	struct blk_shadow shadow[BLK_RING_SIZE];
 	unsigned long shadow_free;
-	int feature_barrier;
+	unsigned int feature_flush;
 	int is_ready;
 };
 
@@ -418,25 +418,12 @@ static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size)
 }
 
 
-static int xlvbd_barrier(struct blkfront_info *info)
+static void xlvbd_flush(struct blkfront_info *info)
 {
-	int err;
-	const char *barrier;
-
-	switch (info->feature_barrier) {
-	case QUEUE_ORDERED_DRAIN:	barrier = "enabled"; break;
-	case QUEUE_ORDERED_NONE:	barrier = "disabled"; break;
-	default:			return -EINVAL;
-	}
-
-	err = blk_queue_ordered(info->rq, info->feature_barrier);
-
-	if (err)
-		return err;
-
+	blk_queue_flush(info->rq, info->feature_flush);
 	printk(KERN_INFO "blkfront: %s: barriers %s\n",
-	       info->gd->disk_name, barrier);
-	return 0;
+	       info->gd->disk_name,
+	       info->feature_flush ? "enabled" : "disabled");
 }
 
 
@@ -515,7 +502,7 @@ static int xlvbd_alloc_gendisk(blkif_sector_t capacity,
 	info->rq = gd->queue;
 	info->gd = gd;
 
-	xlvbd_barrier(info);
+	xlvbd_flush(info);
 
 	if (vdisk_info & VDISK_READONLY)
 		set_disk_ro(gd, 1);
@@ -661,8 +648,8 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
 				printk(KERN_WARNING "blkfront: %s: write barrier op failed\n",
 				       info->gd->disk_name);
 				error = -EOPNOTSUPP;
-				info->feature_barrier = QUEUE_ORDERED_NONE;
-				xlvbd_barrier(info);
+				info->feature_flush = 0;
+				xlvbd_flush(info);
 			}
 			/* fall through */
 		case BLKIF_OP_READ:
@@ -1075,19 +1062,13 @@ static void blkfront_connect(struct blkfront_info *info)
 	/*
 	 * If there's no "feature-barrier" defined, then it means
 	 * we're dealing with a very old backend which writes
-	 * synchronously; draining will do what needs to get done.
+	 * synchronously; nothing to do.
 	 *
 	 * If there are barriers, then we use flush.
-	 *
-	 * If barriers are not supported, then there's no much we can
-	 * do, so just set ordering to NONE.
 	 */
-	if (err)
-		info->feature_barrier = QUEUE_ORDERED_DRAIN;
-	else if (barrier)
-		info->feature_barrier = QUEUE_ORDERED_DRAIN_FLUSH;
-	else
-		info->feature_barrier = QUEUE_ORDERED_NONE;
+	info->feature_flush = 0;
+	if (!err && barrier)
+		info->feature_flush = REQ_FLUSH;
 
 	err = xlvbd_alloc_gendisk(sectors, info, binfo, sector_size);
 	if (err) {
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 7433e07de30e..7c5b01ce51d2 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -516,10 +516,10 @@ static int ide_do_setfeature(ide_drive_t *drive, u8 feature, u8 nsect)
 	return ide_no_data_taskfile(drive, &cmd);
 }
 
-static void update_ordered(ide_drive_t *drive)
+static void update_flush(ide_drive_t *drive)
 {
 	u16 *id = drive->id;
-	unsigned ordered = QUEUE_ORDERED_NONE;
+	unsigned flush = 0;
 
 	if (drive->dev_flags & IDE_DFLAG_WCACHE) {
 		unsigned long long capacity;
@@ -543,13 +543,12 @@ static void update_ordered(ide_drive_t *drive)
 		       drive->name, barrier ? "" : "not ");
 
 		if (barrier) {
-			ordered = QUEUE_ORDERED_DRAIN_FLUSH;
+			flush = REQ_FLUSH;
 			blk_queue_prep_rq(drive->queue, idedisk_prep_fn);
 		}
-	} else
-		ordered = QUEUE_ORDERED_DRAIN;
+	}
 
-	blk_queue_ordered(drive->queue, ordered);
+	blk_queue_flush(drive->queue, flush);
 }
 
 ide_devset_get_flag(wcache, IDE_DFLAG_WCACHE);
@@ -572,7 +571,7 @@ static int set_wcache(ide_drive_t *drive, int arg)
 		}
 	}
 
-	update_ordered(drive);
+	update_flush(drive);
 
 	return err;
 }
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index ac384b2a6a33..b1d92be8f990 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -2245,7 +2245,7 @@ static int dm_init_request_based_queue(struct mapped_device *md)
 	blk_queue_softirq_done(md->queue, dm_softirq_done);
 	blk_queue_prep_rq(md->queue, dm_prep_fn);
 	blk_queue_lld_busy(md->queue, dm_lld_busy);
-	blk_queue_ordered(md->queue, QUEUE_ORDERED_DRAIN_FLUSH);
+	blk_queue_flush(md->queue, REQ_FLUSH);
 
 	elv_register_queue(md->queue);
 
diff --git a/drivers/mmc/card/queue.c b/drivers/mmc/card/queue.c
index e876678176be..9c0b42bfe089 100644
--- a/drivers/mmc/card/queue.c
+++ b/drivers/mmc/card/queue.c
@@ -128,7 +128,6 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, spinlock_t *lock
 	mq->req = NULL;
 
 	blk_queue_prep_rq(mq->queue, mmc_prep_request);
-	blk_queue_ordered(mq->queue, QUEUE_ORDERED_DRAIN);
 	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, mq->queue);
 	if (mmc_can_erase(card)) {
 		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mq->queue);
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index 8373ca0de8e0..9b106d83b0cd 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -2197,7 +2197,6 @@ static void dasd_setup_queue(struct dasd_block *block)
 	 */
 	blk_queue_max_segment_size(block->request_queue, PAGE_SIZE);
 	blk_queue_segment_boundary(block->request_queue, PAGE_SIZE - 1);
-	blk_queue_ordered(block->request_queue, QUEUE_ORDERED_DRAIN);
 }
 
 /*
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index cdfc51ab9cf2..63bd01ae534f 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -2109,7 +2109,7 @@ static int sd_revalidate_disk(struct gendisk *disk)
 	struct scsi_disk *sdkp = scsi_disk(disk);
 	struct scsi_device *sdp = sdkp->device;
 	unsigned char *buffer;
-	unsigned ordered;
+	unsigned flush = 0;
 
 	SCSI_LOG_HLQUEUE(3, sd_printk(KERN_INFO, sdkp,
 				      "sd_revalidate_disk\n"));
@@ -2151,15 +2151,15 @@ static int sd_revalidate_disk(struct gendisk *disk)
 
 	/*
 	 * We now have all cache related info, determine how we deal
-	 * with ordered requests.
+	 * with flush requests.
 	 */
-	if (sdkp->WCE)
-		ordered = sdkp->DPOFUA
-			? QUEUE_ORDERED_DRAIN_FUA : QUEUE_ORDERED_DRAIN_FLUSH;
-	else
-		ordered = QUEUE_ORDERED_DRAIN;
+	if (sdkp->WCE) {
+		flush |= REQ_FLUSH;
+		if (sdkp->DPOFUA)
+			flush |= REQ_FUA;
+	}
 
-	blk_queue_ordered(sdkp->disk->queue, ordered);
+	blk_queue_flush(sdkp->disk->queue, flush);
 
 	set_capacity(disk, sdkp->capacity);
 	kfree(buffer);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 7077bc0d6138..e97911d4dec3 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -355,8 +355,10 @@ struct request_queue
 	struct blk_trace	*blk_trace;
 #endif
 	/*
-	 * reserved for flush operations
+	 * for flush operations
 	 */
+	unsigned int		flush_flags;
+
 	unsigned int		ordered, next_ordered, ordseq;
 	int			orderr, ordcolor;
 	struct request		pre_flush_rq, bar_rq, post_flush_rq;
@@ -865,8 +867,8 @@ extern void blk_queue_update_dma_alignment(struct request_queue *, int);
 extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
 extern void blk_queue_rq_timed_out(struct request_queue *, rq_timed_out_fn *);
 extern void blk_queue_rq_timeout(struct request_queue *, unsigned int);
+extern void blk_queue_flush(struct request_queue *q, unsigned int flush);
 extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);
-extern int blk_queue_ordered(struct request_queue *, unsigned);
 extern bool blk_do_ordered(struct request_queue *, struct request **);
 extern unsigned blk_ordered_cur_seq(struct request_queue *);
 extern unsigned blk_ordered_req_seq(struct request *);
-- 
cgit v1.2.3


From dd831006d5be7f74c3fe7aef82380c51c3637960 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 11:56:16 +0200
Subject: block: misc cleanups in barrier code

Make the following cleanups in preparation of barrier/flush update.

* blk_do_ordered() declaration is moved from include/linux/blkdev.h to
  block/blk.h.

* blk_do_ordered() now returns pointer to struct request, with %NULL
  meaning "try the next request" and ERR_PTR(-EAGAIN) "try again
  later".  The third case will be dropped with further changes.

* In the initialization of proxy barrier request, data direction is
  already set by init_request_from_bio().  Drop unnecessary explicit
  REQ_WRITE setting and move init_request_from_bio() above REQ_FUA
  flag setting.

* add_request() is collapsed into __make_request().

These changes don't make any functional difference.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-barrier.c    | 32 ++++++++++++++------------------
 block/blk-core.c       | 21 ++++-----------------
 block/blk.h            |  7 +++++--
 include/linux/blkdev.h |  1 -
 4 files changed, 23 insertions(+), 38 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-barrier.c b/block/blk-barrier.c
index ed0aba5463ab..f1be85ba2bb5 100644
--- a/block/blk-barrier.c
+++ b/block/blk-barrier.c
@@ -110,9 +110,9 @@ static void queue_flush(struct request_queue *q, unsigned which)
 	elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
 }
 
-static inline bool start_ordered(struct request_queue *q, struct request **rqp)
+static inline struct request *start_ordered(struct request_queue *q,
+					    struct request *rq)
 {
-	struct request *rq = *rqp;
 	unsigned skip = 0;
 
 	q->orderr = 0;
@@ -149,11 +149,9 @@ static inline bool start_ordered(struct request_queue *q, struct request **rqp)
 
 		/* initialize proxy request and queue it */
 		blk_rq_init(q, rq);
-		if (bio_data_dir(q->orig_bar_rq->bio) == WRITE)
-			rq->cmd_flags |= REQ_WRITE;
+		init_request_from_bio(rq, q->orig_bar_rq->bio);
 		if (q->ordered & QUEUE_ORDERED_DO_FUA)
 			rq->cmd_flags |= REQ_FUA;
-		init_request_from_bio(rq, q->orig_bar_rq->bio);
 		rq->end_io = bar_end_io;
 
 		elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
@@ -171,27 +169,26 @@ static inline bool start_ordered(struct request_queue *q, struct request **rqp)
 	else
 		skip |= QUEUE_ORDSEQ_DRAIN;
 
-	*rqp = rq;
-
 	/*
 	 * Complete skipped sequences.  If whole sequence is complete,
-	 * return false to tell elevator that this request is gone.
+	 * return %NULL to tell elevator that this request is gone.
 	 */
-	return !blk_ordered_complete_seq(q, skip, 0);
+	if (blk_ordered_complete_seq(q, skip, 0))
+		rq = NULL;
+	return rq;
 }
 
-bool blk_do_ordered(struct request_queue *q, struct request **rqp)
+struct request *blk_do_ordered(struct request_queue *q, struct request *rq)
 {
-	struct request *rq = *rqp;
 	const int is_barrier = rq->cmd_type == REQ_TYPE_FS &&
 				(rq->cmd_flags & REQ_HARDBARRIER);
 
 	if (!q->ordseq) {
 		if (!is_barrier)
-			return true;
+			return rq;
 
 		if (q->next_ordered != QUEUE_ORDERED_NONE)
-			return start_ordered(q, rqp);
+			return start_ordered(q, rq);
 		else {
 			/*
 			 * Queue ordering not supported.  Terminate
@@ -199,8 +196,7 @@ bool blk_do_ordered(struct request_queue *q, struct request **rqp)
 			 */
 			blk_dequeue_request(rq);
 			__blk_end_request_all(rq, -EOPNOTSUPP);
-			*rqp = NULL;
-			return false;
+			return NULL;
 		}
 	}
 
@@ -211,14 +207,14 @@ bool blk_do_ordered(struct request_queue *q, struct request **rqp)
 	/* Special requests are not subject to ordering rules. */
 	if (rq->cmd_type != REQ_TYPE_FS &&
 	    rq != &q->pre_flush_rq && rq != &q->post_flush_rq)
-		return true;
+		return rq;
 
 	/* Ordered by draining.  Wait for turn. */
 	WARN_ON(blk_ordered_req_seq(rq) < blk_ordered_cur_seq(q));
 	if (blk_ordered_req_seq(rq) > blk_ordered_cur_seq(q))
-		*rqp = NULL;
+		rq = ERR_PTR(-EAGAIN);
 
-	return true;
+	return rq;
 }
 
 static void bio_end_empty_barrier(struct bio *bio, int err)
diff --git a/block/blk-core.c b/block/blk-core.c
index f06354183b29..f8d37a8e2c55 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1037,22 +1037,6 @@ void blk_insert_request(struct request_queue *q, struct request *rq,
 }
 EXPORT_SYMBOL(blk_insert_request);
 
-/*
- * add-request adds a request to the linked list.
- * queue lock is held and interrupts disabled, as we muck with the
- * request queue list.
- */
-static inline void add_request(struct request_queue *q, struct request *req)
-{
-	drive_stat_acct(req, 1);
-
-	/*
-	 * elevator indicated where it wants this request to be
-	 * inserted at elevator_merge time
-	 */
-	__elv_add_request(q, req, ELEVATOR_INSERT_SORT, 0);
-}
-
 static void part_round_stats_single(int cpu, struct hd_struct *part,
 				    unsigned long now)
 {
@@ -1316,7 +1300,10 @@ get_rq:
 		req->cpu = blk_cpu_to_group(smp_processor_id());
 	if (queue_should_plug(q) && elv_queue_empty(q))
 		blk_plug_device(q);
-	add_request(q, req);
+
+	/* insert the request into the elevator */
+	drive_stat_acct(req, 1);
+	__elv_add_request(q, req, ELEVATOR_INSERT_SORT, 0);
 out:
 	if (unplug || !queue_should_plug(q))
 		__generic_unplug_device(q);
diff --git a/block/blk.h b/block/blk.h
index 6e7dc87141e4..874eb4ea8093 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -51,6 +51,8 @@ static inline void blk_clear_rq_complete(struct request *rq)
  */
 #define ELV_ON_HASH(rq)		(!hlist_unhashed(&(rq)->hash))
 
+struct request *blk_do_ordered(struct request_queue *q, struct request *rq);
+
 static inline struct request *__elv_next_request(struct request_queue *q)
 {
 	struct request *rq;
@@ -58,8 +60,9 @@ static inline struct request *__elv_next_request(struct request_queue *q)
 	while (1) {
 		while (!list_empty(&q->queue_head)) {
 			rq = list_entry_rq(q->queue_head.next);
-			if (blk_do_ordered(q, &rq))
-				return rq;
+			rq = blk_do_ordered(q, rq);
+			if (rq)
+				return !IS_ERR(rq) ? rq : NULL;
 		}
 
 		if (!q->elevator->ops->elevator_dispatch_fn(q, 0))
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index e97911d4dec3..996549d71923 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -869,7 +869,6 @@ extern void blk_queue_rq_timed_out(struct request_queue *, rq_timed_out_fn *);
 extern void blk_queue_rq_timeout(struct request_queue *, unsigned int);
 extern void blk_queue_flush(struct request_queue *q, unsigned int flush);
 extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);
-extern bool blk_do_ordered(struct request_queue *, struct request **);
 extern unsigned blk_ordered_cur_seq(struct request_queue *);
 extern unsigned blk_ordered_req_seq(struct request *);
 extern bool blk_ordered_complete_seq(struct request_queue *, unsigned, int);
-- 
cgit v1.2.3


From 28e7d1845216538303bb95d679d8fd4de50e2f1a Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 11:56:16 +0200
Subject: block: drop barrier ordering by queue draining

Filesystems will take all the responsibilities for ordering requests
around commit writes and will only indicate how the commit writes
themselves should be handled by block layers.  This patch drops
barrier ordering by queue draining from block layer.  Ordering by
draining implementation was somewhat invasive to request handling.
List of notable changes follow.

* Each queue has 1 bit color which is flipped on each barrier issue.
  This is used to track whether a given request is issued before the
  current barrier or not.  REQ_ORDERED_COLOR flag and coloring
  implementation in __elv_add_request() are removed.

* Requests which shouldn't be processed yet for draining were stalled
  by returning -EAGAIN from blk_do_ordered() according to the test
  result between blk_ordered_req_seq() and blk_blk_ordered_cur_seq().
  This logic is removed.

* Draining completion logic in elv_completed_request() removed.

* All barrier sequence requests were queued to request queue and then
  trckled to lower layer according to progress and thus maintaining
  request orders during requeue was necessary.  This is replaced by
  queueing the next request in the barrier sequence only after the
  current one is complete from blk_ordered_complete_seq(), which
  removes the need for multiple proxy requests in struct request_queue
  and the request sorting logic in the ELEVATOR_INSERT_REQUEUE path of
  elv_insert().

* As barriers no longer have ordering constraints, there's no need to
  dump the whole elevator onto the dispatch queue on each barrier.
  Insert barriers at the front instead.

* If other barrier requests come to the front of the dispatch queue
  while one is already in progress, they are stored in
  q->pending_barriers and restored to dispatch queue one-by-one after
  each barrier completion from blk_ordered_complete_seq().

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-barrier.c       | 220 ++++++++++++++++++----------------------------
 block/blk-core.c          |  11 ++-
 block/blk.h               |   2 +-
 block/elevator.c          |  79 ++---------------
 include/linux/blk_types.h |   2 -
 include/linux/blkdev.h    |  19 ++--
 6 files changed, 113 insertions(+), 220 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-barrier.c b/block/blk-barrier.c
index f1be85ba2bb5..e8b2e5c091b1 100644
--- a/block/blk-barrier.c
+++ b/block/blk-barrier.c
@@ -9,6 +9,8 @@
 
 #include "blk.h"
 
+static struct request *queue_next_ordseq(struct request_queue *q);
+
 /*
  * Cache flushing for ordered writes handling
  */
@@ -19,38 +21,10 @@ unsigned blk_ordered_cur_seq(struct request_queue *q)
 	return 1 << ffz(q->ordseq);
 }
 
-unsigned blk_ordered_req_seq(struct request *rq)
-{
-	struct request_queue *q = rq->q;
-
-	BUG_ON(q->ordseq == 0);
-
-	if (rq == &q->pre_flush_rq)
-		return QUEUE_ORDSEQ_PREFLUSH;
-	if (rq == &q->bar_rq)
-		return QUEUE_ORDSEQ_BAR;
-	if (rq == &q->post_flush_rq)
-		return QUEUE_ORDSEQ_POSTFLUSH;
-
-	/*
-	 * !fs requests don't need to follow barrier ordering.  Always
-	 * put them at the front.  This fixes the following deadlock.
-	 *
-	 * http://thread.gmane.org/gmane.linux.kernel/537473
-	 */
-	if (rq->cmd_type != REQ_TYPE_FS)
-		return QUEUE_ORDSEQ_DRAIN;
-
-	if ((rq->cmd_flags & REQ_ORDERED_COLOR) ==
-	    (q->orig_bar_rq->cmd_flags & REQ_ORDERED_COLOR))
-		return QUEUE_ORDSEQ_DRAIN;
-	else
-		return QUEUE_ORDSEQ_DONE;
-}
-
-bool blk_ordered_complete_seq(struct request_queue *q, unsigned seq, int error)
+static struct request *blk_ordered_complete_seq(struct request_queue *q,
+						unsigned seq, int error)
 {
-	struct request *rq;
+	struct request *next_rq = NULL;
 
 	if (error && !q->orderr)
 		q->orderr = error;
@@ -58,16 +32,22 @@ bool blk_ordered_complete_seq(struct request_queue *q, unsigned seq, int error)
 	BUG_ON(q->ordseq & seq);
 	q->ordseq |= seq;
 
-	if (blk_ordered_cur_seq(q) != QUEUE_ORDSEQ_DONE)
-		return false;
-
-	/*
-	 * Okay, sequence complete.
-	 */
-	q->ordseq = 0;
-	rq = q->orig_bar_rq;
-	__blk_end_request_all(rq, q->orderr);
-	return true;
+	if (blk_ordered_cur_seq(q) != QUEUE_ORDSEQ_DONE) {
+		/* not complete yet, queue the next ordered sequence */
+		next_rq = queue_next_ordseq(q);
+	} else {
+		/* complete this barrier request */
+		__blk_end_request_all(q->orig_bar_rq, q->orderr);
+		q->orig_bar_rq = NULL;
+		q->ordseq = 0;
+
+		/* dispatch the next barrier if there's one */
+		if (!list_empty(&q->pending_barriers)) {
+			next_rq = list_entry_rq(q->pending_barriers.next);
+			list_move(&next_rq->queuelist, &q->queue_head);
+		}
+	}
+	return next_rq;
 }
 
 static void pre_flush_end_io(struct request *rq, int error)
@@ -88,133 +68,105 @@ static void post_flush_end_io(struct request *rq, int error)
 	blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_POSTFLUSH, error);
 }
 
-static void queue_flush(struct request_queue *q, unsigned which)
+static void queue_flush(struct request_queue *q, struct request *rq,
+			rq_end_io_fn *end_io)
 {
-	struct request *rq;
-	rq_end_io_fn *end_io;
-
-	if (which == QUEUE_ORDERED_DO_PREFLUSH) {
-		rq = &q->pre_flush_rq;
-		end_io = pre_flush_end_io;
-	} else {
-		rq = &q->post_flush_rq;
-		end_io = post_flush_end_io;
-	}
-
 	blk_rq_init(q, rq);
 	rq->cmd_type = REQ_TYPE_FS;
-	rq->cmd_flags = REQ_HARDBARRIER | REQ_FLUSH;
+	rq->cmd_flags = REQ_FLUSH;
 	rq->rq_disk = q->orig_bar_rq->rq_disk;
 	rq->end_io = end_io;
 
 	elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
 }
 
-static inline struct request *start_ordered(struct request_queue *q,
-					    struct request *rq)
+static struct request *queue_next_ordseq(struct request_queue *q)
 {
-	unsigned skip = 0;
-
-	q->orderr = 0;
-	q->ordered = q->next_ordered;
-	q->ordseq |= QUEUE_ORDSEQ_STARTED;
-
-	/*
-	 * For an empty barrier, there's no actual BAR request, which
-	 * in turn makes POSTFLUSH unnecessary.  Mask them off.
-	 */
-	if (!blk_rq_sectors(rq))
-		q->ordered &= ~(QUEUE_ORDERED_DO_BAR |
-				QUEUE_ORDERED_DO_POSTFLUSH);
-
-	/* stash away the original request */
-	blk_dequeue_request(rq);
-	q->orig_bar_rq = rq;
-	rq = NULL;
-
-	/*
-	 * Queue ordered sequence.  As we stack them at the head, we
-	 * need to queue in reverse order.  Note that we rely on that
-	 * no fs request uses ELEVATOR_INSERT_FRONT and thus no fs
-	 * request gets inbetween ordered sequence.
-	 */
-	if (q->ordered & QUEUE_ORDERED_DO_POSTFLUSH) {
-		queue_flush(q, QUEUE_ORDERED_DO_POSTFLUSH);
-		rq = &q->post_flush_rq;
-	} else
-		skip |= QUEUE_ORDSEQ_POSTFLUSH;
+	struct request *rq = &q->bar_rq;
 
-	if (q->ordered & QUEUE_ORDERED_DO_BAR) {
-		rq = &q->bar_rq;
+	switch (blk_ordered_cur_seq(q)) {
+	case QUEUE_ORDSEQ_PREFLUSH:
+		queue_flush(q, rq, pre_flush_end_io);
+		break;
 
+	case QUEUE_ORDSEQ_BAR:
 		/* initialize proxy request and queue it */
 		blk_rq_init(q, rq);
 		init_request_from_bio(rq, q->orig_bar_rq->bio);
+		rq->cmd_flags &= ~REQ_HARDBARRIER;
 		if (q->ordered & QUEUE_ORDERED_DO_FUA)
 			rq->cmd_flags |= REQ_FUA;
 		rq->end_io = bar_end_io;
 
 		elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
-	} else
-		skip |= QUEUE_ORDSEQ_BAR;
+		break;
 
-	if (q->ordered & QUEUE_ORDERED_DO_PREFLUSH) {
-		queue_flush(q, QUEUE_ORDERED_DO_PREFLUSH);
-		rq = &q->pre_flush_rq;
-	} else
-		skip |= QUEUE_ORDSEQ_PREFLUSH;
+	case QUEUE_ORDSEQ_POSTFLUSH:
+		queue_flush(q, rq, post_flush_end_io);
+		break;
 
-	if (queue_in_flight(q))
-		rq = NULL;
-	else
-		skip |= QUEUE_ORDSEQ_DRAIN;
-
-	/*
-	 * Complete skipped sequences.  If whole sequence is complete,
-	 * return %NULL to tell elevator that this request is gone.
-	 */
-	if (blk_ordered_complete_seq(q, skip, 0))
-		rq = NULL;
+	default:
+		BUG();
+	}
 	return rq;
 }
 
 struct request *blk_do_ordered(struct request_queue *q, struct request *rq)
 {
-	const int is_barrier = rq->cmd_type == REQ_TYPE_FS &&
-				(rq->cmd_flags & REQ_HARDBARRIER);
-
-	if (!q->ordseq) {
-		if (!is_barrier)
-			return rq;
-
-		if (q->next_ordered != QUEUE_ORDERED_NONE)
-			return start_ordered(q, rq);
-		else {
-			/*
-			 * Queue ordering not supported.  Terminate
-			 * with prejudice.
-			 */
-			blk_dequeue_request(rq);
-			__blk_end_request_all(rq, -EOPNOTSUPP);
-			return NULL;
-		}
+	unsigned skip = 0;
+
+	if (!(rq->cmd_flags & REQ_HARDBARRIER))
+		return rq;
+
+	if (q->ordseq) {
+		/*
+		 * Barrier is already in progress and they can't be
+		 * processed in parallel.  Queue for later processing.
+		 */
+		list_move_tail(&rq->queuelist, &q->pending_barriers);
+		return NULL;
+	}
+
+	if (unlikely(q->next_ordered == QUEUE_ORDERED_NONE)) {
+		/*
+		 * Queue ordering not supported.  Terminate
+		 * with prejudice.
+		 */
+		blk_dequeue_request(rq);
+		__blk_end_request_all(rq, -EOPNOTSUPP);
+		return NULL;
 	}
 
 	/*
-	 * Ordered sequence in progress
+	 * Start a new ordered sequence
 	 */
+	q->orderr = 0;
+	q->ordered = q->next_ordered;
+	q->ordseq |= QUEUE_ORDSEQ_STARTED;
 
-	/* Special requests are not subject to ordering rules. */
-	if (rq->cmd_type != REQ_TYPE_FS &&
-	    rq != &q->pre_flush_rq && rq != &q->post_flush_rq)
-		return rq;
+	/*
+	 * For an empty barrier, there's no actual BAR request, which
+	 * in turn makes POSTFLUSH unnecessary.  Mask them off.
+	 */
+	if (!blk_rq_sectors(rq))
+		q->ordered &= ~(QUEUE_ORDERED_DO_BAR |
+				QUEUE_ORDERED_DO_POSTFLUSH);
 
-	/* Ordered by draining.  Wait for turn. */
-	WARN_ON(blk_ordered_req_seq(rq) < blk_ordered_cur_seq(q));
-	if (blk_ordered_req_seq(rq) > blk_ordered_cur_seq(q))
-		rq = ERR_PTR(-EAGAIN);
+	/* stash away the original request */
+	blk_dequeue_request(rq);
+	q->orig_bar_rq = rq;
 
-	return rq;
+	if (!(q->ordered & QUEUE_ORDERED_DO_PREFLUSH))
+		skip |= QUEUE_ORDSEQ_PREFLUSH;
+
+	if (!(q->ordered & QUEUE_ORDERED_DO_BAR))
+		skip |= QUEUE_ORDSEQ_BAR;
+
+	if (!(q->ordered & QUEUE_ORDERED_DO_POSTFLUSH))
+		skip |= QUEUE_ORDSEQ_POSTFLUSH;
+
+	/* complete skipped sequences and return the first sequence */
+	return blk_ordered_complete_seq(q, skip, 0);
 }
 
 static void bio_end_empty_barrier(struct bio *bio, int err)
diff --git a/block/blk-core.c b/block/blk-core.c
index f8d37a8e2c55..d316662682c8 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -520,6 +520,7 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 	init_timer(&q->unplug_timer);
 	setup_timer(&q->timeout, blk_rq_timed_out_timer, (unsigned long) q);
 	INIT_LIST_HEAD(&q->timeout_list);
+	INIT_LIST_HEAD(&q->pending_barriers);
 	INIT_WORK(&q->unplug_work, blk_unplug_work);
 
 	kobject_init(&q->kobj, &blk_queue_ktype);
@@ -1185,6 +1186,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 	const bool sync = (bio->bi_rw & REQ_SYNC);
 	const bool unplug = (bio->bi_rw & REQ_UNPLUG);
 	const unsigned int ff = bio->bi_rw & REQ_FAILFAST_MASK;
+	int where = ELEVATOR_INSERT_SORT;
 	int rw_flags;
 
 	/* REQ_HARDBARRIER is no more */
@@ -1203,7 +1205,12 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 
 	spin_lock_irq(q->queue_lock);
 
-	if (unlikely((bio->bi_rw & REQ_HARDBARRIER)) || elv_queue_empty(q))
+	if (bio->bi_rw & REQ_HARDBARRIER) {
+		where = ELEVATOR_INSERT_FRONT;
+		goto get_rq;
+	}
+
+	if (elv_queue_empty(q))
 		goto get_rq;
 
 	el_ret = elv_merge(q, &req, bio);
@@ -1303,7 +1310,7 @@ get_rq:
 
 	/* insert the request into the elevator */
 	drive_stat_acct(req, 1);
-	__elv_add_request(q, req, ELEVATOR_INSERT_SORT, 0);
+	__elv_add_request(q, req, where, 0);
 out:
 	if (unplug || !queue_should_plug(q))
 		__generic_unplug_device(q);
diff --git a/block/blk.h b/block/blk.h
index 874eb4ea8093..08081e4b294e 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -62,7 +62,7 @@ static inline struct request *__elv_next_request(struct request_queue *q)
 			rq = list_entry_rq(q->queue_head.next);
 			rq = blk_do_ordered(q, rq);
 			if (rq)
-				return !IS_ERR(rq) ? rq : NULL;
+				return rq;
 		}
 
 		if (!q->elevator->ops->elevator_dispatch_fn(q, 0))
diff --git a/block/elevator.c b/block/elevator.c
index ec585c9554d3..241c69c45c5f 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -617,8 +617,6 @@ void elv_quiesce_end(struct request_queue *q)
 
 void elv_insert(struct request_queue *q, struct request *rq, int where)
 {
-	struct list_head *pos;
-	unsigned ordseq;
 	int unplug_it = 1;
 
 	trace_block_rq_insert(q, rq);
@@ -626,9 +624,16 @@ void elv_insert(struct request_queue *q, struct request *rq, int where)
 	rq->q = q;
 
 	switch (where) {
+	case ELEVATOR_INSERT_REQUEUE:
+		/*
+		 * Most requeues happen because of a busy condition,
+		 * don't force unplug of the queue for that case.
+		 * Clear unplug_it and fall through.
+		 */
+		unplug_it = 0;
+
 	case ELEVATOR_INSERT_FRONT:
 		rq->cmd_flags |= REQ_SOFTBARRIER;
-
 		list_add(&rq->queuelist, &q->queue_head);
 		break;
 
@@ -668,36 +673,6 @@ void elv_insert(struct request_queue *q, struct request *rq, int where)
 		q->elevator->ops->elevator_add_req_fn(q, rq);
 		break;
 
-	case ELEVATOR_INSERT_REQUEUE:
-		/*
-		 * If ordered flush isn't in progress, we do front
-		 * insertion; otherwise, requests should be requeued
-		 * in ordseq order.
-		 */
-		rq->cmd_flags |= REQ_SOFTBARRIER;
-
-		/*
-		 * Most requeues happen because of a busy condition,
-		 * don't force unplug of the queue for that case.
-		 */
-		unplug_it = 0;
-
-		if (q->ordseq == 0) {
-			list_add(&rq->queuelist, &q->queue_head);
-			break;
-		}
-
-		ordseq = blk_ordered_req_seq(rq);
-
-		list_for_each(pos, &q->queue_head) {
-			struct request *pos_rq = list_entry_rq(pos);
-			if (ordseq <= blk_ordered_req_seq(pos_rq))
-				break;
-		}
-
-		list_add_tail(&rq->queuelist, pos);
-		break;
-
 	default:
 		printk(KERN_ERR "%s: bad insertion point %d\n",
 		       __func__, where);
@@ -716,26 +691,8 @@ void elv_insert(struct request_queue *q, struct request *rq, int where)
 void __elv_add_request(struct request_queue *q, struct request *rq, int where,
 		       int plug)
 {
-	if (q->ordcolor)
-		rq->cmd_flags |= REQ_ORDERED_COLOR;
-
 	if (rq->cmd_flags & (REQ_SOFTBARRIER | REQ_HARDBARRIER)) {
-		/*
-		 * toggle ordered color
-		 */
-		if (rq->cmd_flags & REQ_HARDBARRIER)
-			q->ordcolor ^= 1;
-
-		/*
-		 * barriers implicitly indicate back insertion
-		 */
-		if (where == ELEVATOR_INSERT_SORT)
-			where = ELEVATOR_INSERT_BACK;
-
-		/*
-		 * this request is scheduling boundary, update
-		 * end_sector
-		 */
+		/* barriers are scheduling boundary, update end_sector */
 		if (rq->cmd_type == REQ_TYPE_FS ||
 		    (rq->cmd_flags & REQ_DISCARD)) {
 			q->end_sector = rq_end_sector(rq);
@@ -855,24 +812,6 @@ void elv_completed_request(struct request_queue *q, struct request *rq)
 		    e->ops->elevator_completed_req_fn)
 			e->ops->elevator_completed_req_fn(q, rq);
 	}
-
-	/*
-	 * Check if the queue is waiting for fs requests to be
-	 * drained for flush sequence.
-	 */
-	if (unlikely(q->ordseq)) {
-		struct request *next = NULL;
-
-		if (!list_empty(&q->queue_head))
-			next = list_entry_rq(q->queue_head.next);
-
-		if (!queue_in_flight(q) &&
-		    blk_ordered_cur_seq(q) == QUEUE_ORDSEQ_DRAIN &&
-		    (!next || blk_ordered_req_seq(next) > QUEUE_ORDSEQ_DRAIN)) {
-			blk_ordered_complete_seq(q, QUEUE_ORDSEQ_DRAIN, 0);
-			__blk_run_queue(q);
-		}
-	}
 }
 
 #define to_elv(atr) container_of((atr), struct elv_fs_entry, attr)
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index ca83a97c9715..9192282b4259 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -143,7 +143,6 @@ enum rq_flag_bits {
 	__REQ_FAILED,		/* set if the request failed */
 	__REQ_QUIET,		/* don't worry about errors */
 	__REQ_PREEMPT,		/* set for "ide_preempt" requests */
-	__REQ_ORDERED_COLOR,	/* is before or after barrier */
 	__REQ_ALLOCED,		/* request came from our alloc pool */
 	__REQ_COPY_USER,	/* contains copies of user pages */
 	__REQ_INTEGRITY,	/* integrity metadata has been remapped */
@@ -184,7 +183,6 @@ enum rq_flag_bits {
 #define REQ_FAILED		(1 << __REQ_FAILED)
 #define REQ_QUIET		(1 << __REQ_QUIET)
 #define REQ_PREEMPT		(1 << __REQ_PREEMPT)
-#define REQ_ORDERED_COLOR	(1 << __REQ_ORDERED_COLOR)
 #define REQ_ALLOCED		(1 << __REQ_ALLOCED)
 #define REQ_COPY_USER		(1 << __REQ_COPY_USER)
 #define REQ_INTEGRITY		(1 << __REQ_INTEGRITY)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 996549d71923..20a3710a481b 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -360,9 +360,10 @@ struct request_queue
 	unsigned int		flush_flags;
 
 	unsigned int		ordered, next_ordered, ordseq;
-	int			orderr, ordcolor;
-	struct request		pre_flush_rq, bar_rq, post_flush_rq;
+	int			orderr;
+	struct request		bar_rq;
 	struct request		*orig_bar_rq;
+	struct list_head	pending_barriers;
 
 	struct mutex		sysfs_lock;
 
@@ -491,12 +492,11 @@ enum {
 	/*
 	 * Ordered operation sequence
 	 */
-	QUEUE_ORDSEQ_STARTED	= 0x01,	/* flushing in progress */
-	QUEUE_ORDSEQ_DRAIN	= 0x02,	/* waiting for the queue to be drained */
-	QUEUE_ORDSEQ_PREFLUSH	= 0x04,	/* pre-flushing in progress */
-	QUEUE_ORDSEQ_BAR	= 0x08,	/* original barrier req in progress */
-	QUEUE_ORDSEQ_POSTFLUSH	= 0x10,	/* post-flushing in progress */
-	QUEUE_ORDSEQ_DONE	= 0x20,
+	QUEUE_ORDSEQ_STARTED	= (1 << 0), /* flushing in progress */
+	QUEUE_ORDSEQ_PREFLUSH	= (1 << 1), /* pre-flushing in progress */
+	QUEUE_ORDSEQ_BAR	= (1 << 2), /* barrier write in progress */
+	QUEUE_ORDSEQ_POSTFLUSH	= (1 << 3), /* post-flushing in progress */
+	QUEUE_ORDSEQ_DONE	= (1 << 4),
 };
 
 #define blk_queue_plugged(q)	test_bit(QUEUE_FLAG_PLUGGED, &(q)->queue_flags)
@@ -869,9 +869,6 @@ extern void blk_queue_rq_timed_out(struct request_queue *, rq_timed_out_fn *);
 extern void blk_queue_rq_timeout(struct request_queue *, unsigned int);
 extern void blk_queue_flush(struct request_queue *q, unsigned int flush);
 extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);
-extern unsigned blk_ordered_cur_seq(struct request_queue *);
-extern unsigned blk_ordered_req_seq(struct request *);
-extern bool blk_ordered_complete_seq(struct request_queue *, unsigned, int);
 
 extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *);
 extern void blk_dump_rq_flags(struct request *, char *);
-- 
cgit v1.2.3


From dd4c133f387c48f526022860ad70354637a80f4c Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 11:56:16 +0200
Subject: block: rename barrier/ordered to flush

With ordering requirements dropped, barrier and ordered are misnomers.
Now all block layer does is sequencing FLUSH and FUA.  Rename them to
flush.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-core.c       | 21 ++++++-----
 block/blk-flush.c      | 98 +++++++++++++++++++++++++-------------------------
 block/blk.h            |  4 +--
 include/linux/blkdev.h | 24 ++++++-------
 4 files changed, 72 insertions(+), 75 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-core.c b/block/blk-core.c
index d316662682c8..8870ae40179d 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -136,7 +136,7 @@ static void req_bio_endio(struct request *rq, struct bio *bio,
 {
 	struct request_queue *q = rq->q;
 
-	if (&q->bar_rq != rq) {
+	if (&q->flush_rq != rq) {
 		if (error)
 			clear_bit(BIO_UPTODATE, &bio->bi_flags);
 		else if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
@@ -160,13 +160,12 @@ static void req_bio_endio(struct request *rq, struct bio *bio,
 		if (bio->bi_size == 0)
 			bio_endio(bio, error);
 	} else {
-
 		/*
-		 * Okay, this is the barrier request in progress, just
-		 * record the error;
+		 * Okay, this is the sequenced flush request in
+		 * progress, just record the error;
 		 */
-		if (error && !q->orderr)
-			q->orderr = error;
+		if (error && !q->flush_err)
+			q->flush_err = error;
 	}
 }
 
@@ -520,7 +519,7 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 	init_timer(&q->unplug_timer);
 	setup_timer(&q->timeout, blk_rq_timed_out_timer, (unsigned long) q);
 	INIT_LIST_HEAD(&q->timeout_list);
-	INIT_LIST_HEAD(&q->pending_barriers);
+	INIT_LIST_HEAD(&q->pending_flushes);
 	INIT_WORK(&q->unplug_work, blk_unplug_work);
 
 	kobject_init(&q->kobj, &blk_queue_ktype);
@@ -1764,11 +1763,11 @@ static void blk_account_io_completion(struct request *req, unsigned int bytes)
 static void blk_account_io_done(struct request *req)
 {
 	/*
-	 * Account IO completion.  bar_rq isn't accounted as a normal
-	 * IO on queueing nor completion.  Accounting the containing
-	 * request is enough.
+	 * Account IO completion.  flush_rq isn't accounted as a
+	 * normal IO on queueing nor completion.  Accounting the
+	 * containing request is enough.
 	 */
-	if (blk_do_io_stat(req) && req != &req->q->bar_rq) {
+	if (blk_do_io_stat(req) && req != &req->q->flush_rq) {
 		unsigned long duration = jiffies - req->start_time;
 		const int rw = rq_data_dir(req);
 		struct hd_struct *part;
diff --git a/block/blk-flush.c b/block/blk-flush.c
index e8b2e5c091b1..dd873225da97 100644
--- a/block/blk-flush.c
+++ b/block/blk-flush.c
@@ -9,41 +9,38 @@
 
 #include "blk.h"
 
-static struct request *queue_next_ordseq(struct request_queue *q);
+static struct request *queue_next_fseq(struct request_queue *q);
 
-/*
- * Cache flushing for ordered writes handling
- */
-unsigned blk_ordered_cur_seq(struct request_queue *q)
+unsigned blk_flush_cur_seq(struct request_queue *q)
 {
-	if (!q->ordseq)
+	if (!q->flush_seq)
 		return 0;
-	return 1 << ffz(q->ordseq);
+	return 1 << ffz(q->flush_seq);
 }
 
-static struct request *blk_ordered_complete_seq(struct request_queue *q,
-						unsigned seq, int error)
+static struct request *blk_flush_complete_seq(struct request_queue *q,
+					      unsigned seq, int error)
 {
 	struct request *next_rq = NULL;
 
-	if (error && !q->orderr)
-		q->orderr = error;
+	if (error && !q->flush_err)
+		q->flush_err = error;
 
-	BUG_ON(q->ordseq & seq);
-	q->ordseq |= seq;
+	BUG_ON(q->flush_seq & seq);
+	q->flush_seq |= seq;
 
-	if (blk_ordered_cur_seq(q) != QUEUE_ORDSEQ_DONE) {
-		/* not complete yet, queue the next ordered sequence */
-		next_rq = queue_next_ordseq(q);
+	if (blk_flush_cur_seq(q) != QUEUE_FSEQ_DONE) {
+		/* not complete yet, queue the next flush sequence */
+		next_rq = queue_next_fseq(q);
 	} else {
-		/* complete this barrier request */
-		__blk_end_request_all(q->orig_bar_rq, q->orderr);
-		q->orig_bar_rq = NULL;
-		q->ordseq = 0;
-
-		/* dispatch the next barrier if there's one */
-		if (!list_empty(&q->pending_barriers)) {
-			next_rq = list_entry_rq(q->pending_barriers.next);
+		/* complete this flush request */
+		__blk_end_request_all(q->orig_flush_rq, q->flush_err);
+		q->orig_flush_rq = NULL;
+		q->flush_seq = 0;
+
+		/* dispatch the next flush if there's one */
+		if (!list_empty(&q->pending_flushes)) {
+			next_rq = list_entry_rq(q->pending_flushes.next);
 			list_move(&next_rq->queuelist, &q->queue_head);
 		}
 	}
@@ -53,19 +50,19 @@ static struct request *blk_ordered_complete_seq(struct request_queue *q,
 static void pre_flush_end_io(struct request *rq, int error)
 {
 	elv_completed_request(rq->q, rq);
-	blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_PREFLUSH, error);
+	blk_flush_complete_seq(rq->q, QUEUE_FSEQ_PREFLUSH, error);
 }
 
-static void bar_end_io(struct request *rq, int error)
+static void flush_data_end_io(struct request *rq, int error)
 {
 	elv_completed_request(rq->q, rq);
-	blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_BAR, error);
+	blk_flush_complete_seq(rq->q, QUEUE_FSEQ_DATA, error);
 }
 
 static void post_flush_end_io(struct request *rq, int error)
 {
 	elv_completed_request(rq->q, rq);
-	blk_ordered_complete_seq(rq->q, QUEUE_ORDSEQ_POSTFLUSH, error);
+	blk_flush_complete_seq(rq->q, QUEUE_FSEQ_POSTFLUSH, error);
 }
 
 static void queue_flush(struct request_queue *q, struct request *rq,
@@ -74,34 +71,34 @@ static void queue_flush(struct request_queue *q, struct request *rq,
 	blk_rq_init(q, rq);
 	rq->cmd_type = REQ_TYPE_FS;
 	rq->cmd_flags = REQ_FLUSH;
-	rq->rq_disk = q->orig_bar_rq->rq_disk;
+	rq->rq_disk = q->orig_flush_rq->rq_disk;
 	rq->end_io = end_io;
 
 	elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
 }
 
-static struct request *queue_next_ordseq(struct request_queue *q)
+static struct request *queue_next_fseq(struct request_queue *q)
 {
-	struct request *rq = &q->bar_rq;
+	struct request *rq = &q->flush_rq;
 
-	switch (blk_ordered_cur_seq(q)) {
-	case QUEUE_ORDSEQ_PREFLUSH:
+	switch (blk_flush_cur_seq(q)) {
+	case QUEUE_FSEQ_PREFLUSH:
 		queue_flush(q, rq, pre_flush_end_io);
 		break;
 
-	case QUEUE_ORDSEQ_BAR:
+	case QUEUE_FSEQ_DATA:
 		/* initialize proxy request and queue it */
 		blk_rq_init(q, rq);
-		init_request_from_bio(rq, q->orig_bar_rq->bio);
+		init_request_from_bio(rq, q->orig_flush_rq->bio);
 		rq->cmd_flags &= ~REQ_HARDBARRIER;
 		if (q->ordered & QUEUE_ORDERED_DO_FUA)
 			rq->cmd_flags |= REQ_FUA;
-		rq->end_io = bar_end_io;
+		rq->end_io = flush_data_end_io;
 
 		elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
 		break;
 
-	case QUEUE_ORDSEQ_POSTFLUSH:
+	case QUEUE_FSEQ_POSTFLUSH:
 		queue_flush(q, rq, post_flush_end_io);
 		break;
 
@@ -111,19 +108,20 @@ static struct request *queue_next_ordseq(struct request_queue *q)
 	return rq;
 }
 
-struct request *blk_do_ordered(struct request_queue *q, struct request *rq)
+struct request *blk_do_flush(struct request_queue *q, struct request *rq)
 {
 	unsigned skip = 0;
 
 	if (!(rq->cmd_flags & REQ_HARDBARRIER))
 		return rq;
 
-	if (q->ordseq) {
+	if (q->flush_seq) {
 		/*
-		 * Barrier is already in progress and they can't be
-		 * processed in parallel.  Queue for later processing.
+		 * Sequenced flush is already in progress and they
+		 * can't be processed in parallel.  Queue for later
+		 * processing.
 		 */
-		list_move_tail(&rq->queuelist, &q->pending_barriers);
+		list_move_tail(&rq->queuelist, &q->pending_flushes);
 		return NULL;
 	}
 
@@ -138,11 +136,11 @@ struct request *blk_do_ordered(struct request_queue *q, struct request *rq)
 	}
 
 	/*
-	 * Start a new ordered sequence
+	 * Start a new flush sequence
 	 */
-	q->orderr = 0;
+	q->flush_err = 0;
 	q->ordered = q->next_ordered;
-	q->ordseq |= QUEUE_ORDSEQ_STARTED;
+	q->flush_seq |= QUEUE_FSEQ_STARTED;
 
 	/*
 	 * For an empty barrier, there's no actual BAR request, which
@@ -154,19 +152,19 @@ struct request *blk_do_ordered(struct request_queue *q, struct request *rq)
 
 	/* stash away the original request */
 	blk_dequeue_request(rq);
-	q->orig_bar_rq = rq;
+	q->orig_flush_rq = rq;
 
 	if (!(q->ordered & QUEUE_ORDERED_DO_PREFLUSH))
-		skip |= QUEUE_ORDSEQ_PREFLUSH;
+		skip |= QUEUE_FSEQ_PREFLUSH;
 
 	if (!(q->ordered & QUEUE_ORDERED_DO_BAR))
-		skip |= QUEUE_ORDSEQ_BAR;
+		skip |= QUEUE_FSEQ_DATA;
 
 	if (!(q->ordered & QUEUE_ORDERED_DO_POSTFLUSH))
-		skip |= QUEUE_ORDSEQ_POSTFLUSH;
+		skip |= QUEUE_FSEQ_POSTFLUSH;
 
 	/* complete skipped sequences and return the first sequence */
-	return blk_ordered_complete_seq(q, skip, 0);
+	return blk_flush_complete_seq(q, skip, 0);
 }
 
 static void bio_end_empty_barrier(struct bio *bio, int err)
diff --git a/block/blk.h b/block/blk.h
index 08081e4b294e..24b92bd78f37 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -51,7 +51,7 @@ static inline void blk_clear_rq_complete(struct request *rq)
  */
 #define ELV_ON_HASH(rq)		(!hlist_unhashed(&(rq)->hash))
 
-struct request *blk_do_ordered(struct request_queue *q, struct request *rq);
+struct request *blk_do_flush(struct request_queue *q, struct request *rq);
 
 static inline struct request *__elv_next_request(struct request_queue *q)
 {
@@ -60,7 +60,7 @@ static inline struct request *__elv_next_request(struct request_queue *q)
 	while (1) {
 		while (!list_empty(&q->queue_head)) {
 			rq = list_entry_rq(q->queue_head.next);
-			rq = blk_do_ordered(q, rq);
+			rq = blk_do_flush(q, rq);
 			if (rq)
 				return rq;
 		}
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 20a3710a481b..1cd83ec077db 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -357,13 +357,13 @@ struct request_queue
 	/*
 	 * for flush operations
 	 */
+	unsigned int		ordered, next_ordered;
 	unsigned int		flush_flags;
-
-	unsigned int		ordered, next_ordered, ordseq;
-	int			orderr;
-	struct request		bar_rq;
-	struct request		*orig_bar_rq;
-	struct list_head	pending_barriers;
+	unsigned int		flush_seq;
+	int			flush_err;
+	struct request		flush_rq;
+	struct request		*orig_flush_rq;
+	struct list_head	pending_flushes;
 
 	struct mutex		sysfs_lock;
 
@@ -490,13 +490,13 @@ enum {
 					  QUEUE_ORDERED_DO_FUA,
 
 	/*
-	 * Ordered operation sequence
+	 * FLUSH/FUA sequences.
 	 */
-	QUEUE_ORDSEQ_STARTED	= (1 << 0), /* flushing in progress */
-	QUEUE_ORDSEQ_PREFLUSH	= (1 << 1), /* pre-flushing in progress */
-	QUEUE_ORDSEQ_BAR	= (1 << 2), /* barrier write in progress */
-	QUEUE_ORDSEQ_POSTFLUSH	= (1 << 3), /* post-flushing in progress */
-	QUEUE_ORDSEQ_DONE	= (1 << 4),
+	QUEUE_FSEQ_STARTED	= (1 << 0), /* flushing in progress */
+	QUEUE_FSEQ_PREFLUSH	= (1 << 1), /* pre-flushing in progress */
+	QUEUE_FSEQ_DATA		= (1 << 2), /* data write in progress */
+	QUEUE_FSEQ_POSTFLUSH	= (1 << 3), /* post-flushing in progress */
+	QUEUE_FSEQ_DONE		= (1 << 4),
 };
 
 #define blk_queue_plugged(q)	test_bit(QUEUE_FLAG_PLUGGED, &(q)->queue_flags)
-- 
cgit v1.2.3


From 4fed947cb311e5aa51781d316cefca836352f6ce Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 11:56:17 +0200
Subject: block: implement REQ_FLUSH/FUA based interface for FLUSH/FUA requests

Now that the backend conversion is complete, export sequenced
FLUSH/FUA capability through REQ_FLUSH/FUA flags.  REQ_FLUSH means the
device cache should be flushed before executing the request.  REQ_FUA
means that the data in the request should be on non-volatile media on
completion.

Block layer will choose the correct way of implementing the semantics
and execute it.  The request may be passed to the device directly if
the device can handle it; otherwise, it will be sequenced using one or
more proxy requests.  Devices will never see REQ_FLUSH and/or FUA
which it doesn't support.

Also, unlike the original REQ_HARDBARRIER, REQ_FLUSH/FUA requests are
never failed with -EOPNOTSUPP.  If the underlying device doesn't
support FLUSH/FUA, the block layer simply make those noop.  IOW, it no
longer distinguishes between writeback cache which doesn't support
cache flush and writethrough/no cache.  Devices which have WB cache
w/o flush are very difficult to come by these days and there's nothing
much we can do anyway, so it doesn't make sense to require everyone to
implement -EOPNOTSUPP handling.  This will simplify filesystems and
block drivers as they can drop -EOPNOTSUPP retry logic for barriers.

* QUEUE_ORDERED_* are removed and QUEUE_FSEQ_* are moved into
  blk-flush.c.

* REQ_FLUSH w/o data can also be directly passed to drivers without
  sequencing but some drivers assume that zero length requests don't
  have rq->bio which isn't true for these requests requiring the use
  of proxy requests.

* REQ_COMMON_MASK now includes REQ_FLUSH | REQ_FUA so that they are
  copied from bio to request.

* WRITE_BARRIER is marked deprecated and WRITE_FLUSH, WRITE_FUA and
  WRITE_FLUSH_FUA are added.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-core.c            |  2 +-
 block/blk-flush.c           | 85 ++++++++++++++++++++++++---------------------
 block/blk.h                 |  3 ++
 include/linux/blk_types.h   |  2 +-
 include/linux/blkdev.h      | 38 ++------------------
 include/linux/buffer_head.h |  2 +-
 include/linux/fs.h          | 19 ++++++----
 7 files changed, 67 insertions(+), 84 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-core.c b/block/blk-core.c
index 8870ae40179d..18455c4f618a 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1204,7 +1204,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 
 	spin_lock_irq(q->queue_lock);
 
-	if (bio->bi_rw & REQ_HARDBARRIER) {
+	if (bio->bi_rw & (REQ_FLUSH | REQ_FUA)) {
 		where = ELEVATOR_INSERT_FRONT;
 		goto get_rq;
 	}
diff --git a/block/blk-flush.c b/block/blk-flush.c
index dd873225da97..452c552e9ead 100644
--- a/block/blk-flush.c
+++ b/block/blk-flush.c
@@ -1,5 +1,5 @@
 /*
- * Functions related to barrier IO handling
+ * Functions to sequence FLUSH and FUA writes.
  */
 #include <linux/kernel.h>
 #include <linux/module.h>
@@ -9,6 +9,15 @@
 
 #include "blk.h"
 
+/* FLUSH/FUA sequences */
+enum {
+	QUEUE_FSEQ_STARTED	= (1 << 0), /* flushing in progress */
+	QUEUE_FSEQ_PREFLUSH	= (1 << 1), /* pre-flushing in progress */
+	QUEUE_FSEQ_DATA		= (1 << 2), /* data write in progress */
+	QUEUE_FSEQ_POSTFLUSH	= (1 << 3), /* post-flushing in progress */
+	QUEUE_FSEQ_DONE		= (1 << 4),
+};
+
 static struct request *queue_next_fseq(struct request_queue *q);
 
 unsigned blk_flush_cur_seq(struct request_queue *q)
@@ -79,6 +88,7 @@ static void queue_flush(struct request_queue *q, struct request *rq,
 
 static struct request *queue_next_fseq(struct request_queue *q)
 {
+	struct request *orig_rq = q->orig_flush_rq;
 	struct request *rq = &q->flush_rq;
 
 	switch (blk_flush_cur_seq(q)) {
@@ -87,12 +97,11 @@ static struct request *queue_next_fseq(struct request_queue *q)
 		break;
 
 	case QUEUE_FSEQ_DATA:
-		/* initialize proxy request and queue it */
+		/* initialize proxy request, inherit FLUSH/FUA and queue it */
 		blk_rq_init(q, rq);
-		init_request_from_bio(rq, q->orig_flush_rq->bio);
-		rq->cmd_flags &= ~REQ_HARDBARRIER;
-		if (q->ordered & QUEUE_ORDERED_DO_FUA)
-			rq->cmd_flags |= REQ_FUA;
+		init_request_from_bio(rq, orig_rq->bio);
+		rq->cmd_flags &= ~(REQ_FLUSH | REQ_FUA);
+		rq->cmd_flags |= orig_rq->cmd_flags & (REQ_FLUSH | REQ_FUA);
 		rq->end_io = flush_data_end_io;
 
 		elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
@@ -110,60 +119,58 @@ static struct request *queue_next_fseq(struct request_queue *q)
 
 struct request *blk_do_flush(struct request_queue *q, struct request *rq)
 {
+	unsigned int fflags = q->flush_flags; /* may change, cache it */
+	bool has_flush = fflags & REQ_FLUSH, has_fua = fflags & REQ_FUA;
+	bool do_preflush = has_flush && (rq->cmd_flags & REQ_FLUSH);
+	bool do_postflush = has_flush && !has_fua && (rq->cmd_flags & REQ_FUA);
 	unsigned skip = 0;
 
-	if (!(rq->cmd_flags & REQ_HARDBARRIER))
+	/*
+	 * Special case.  If there's data but flush is not necessary,
+	 * the request can be issued directly.
+	 *
+	 * Flush w/o data should be able to be issued directly too but
+	 * currently some drivers assume that rq->bio contains
+	 * non-zero data if it isn't NULL and empty FLUSH requests
+	 * getting here usually have bio's without data.
+	 */
+	if (blk_rq_sectors(rq) && !do_preflush && !do_postflush) {
+		rq->cmd_flags &= ~REQ_FLUSH;
+		if (!has_fua)
+			rq->cmd_flags &= ~REQ_FUA;
 		return rq;
+	}
 
+	/*
+	 * Sequenced flushes can't be processed in parallel.  If
+	 * another one is already in progress, queue for later
+	 * processing.
+	 */
 	if (q->flush_seq) {
-		/*
-		 * Sequenced flush is already in progress and they
-		 * can't be processed in parallel.  Queue for later
-		 * processing.
-		 */
 		list_move_tail(&rq->queuelist, &q->pending_flushes);
 		return NULL;
 	}
 
-	if (unlikely(q->next_ordered == QUEUE_ORDERED_NONE)) {
-		/*
-		 * Queue ordering not supported.  Terminate
-		 * with prejudice.
-		 */
-		blk_dequeue_request(rq);
-		__blk_end_request_all(rq, -EOPNOTSUPP);
-		return NULL;
-	}
-
 	/*
 	 * Start a new flush sequence
 	 */
 	q->flush_err = 0;
-	q->ordered = q->next_ordered;
 	q->flush_seq |= QUEUE_FSEQ_STARTED;
 
-	/*
-	 * For an empty barrier, there's no actual BAR request, which
-	 * in turn makes POSTFLUSH unnecessary.  Mask them off.
-	 */
-	if (!blk_rq_sectors(rq))
-		q->ordered &= ~(QUEUE_ORDERED_DO_BAR |
-				QUEUE_ORDERED_DO_POSTFLUSH);
-
-	/* stash away the original request */
+	/* adjust FLUSH/FUA of the original request and stash it away */
+	rq->cmd_flags &= ~REQ_FLUSH;
+	if (!has_fua)
+		rq->cmd_flags &= ~REQ_FUA;
 	blk_dequeue_request(rq);
 	q->orig_flush_rq = rq;
 
-	if (!(q->ordered & QUEUE_ORDERED_DO_PREFLUSH))
+	/* skip unneded sequences and return the first one */
+	if (!do_preflush)
 		skip |= QUEUE_FSEQ_PREFLUSH;
-
-	if (!(q->ordered & QUEUE_ORDERED_DO_BAR))
+	if (!blk_rq_sectors(rq))
 		skip |= QUEUE_FSEQ_DATA;
-
-	if (!(q->ordered & QUEUE_ORDERED_DO_POSTFLUSH))
+	if (!do_postflush)
 		skip |= QUEUE_FSEQ_POSTFLUSH;
-
-	/* complete skipped sequences and return the first sequence */
 	return blk_flush_complete_seq(q, skip, 0);
 }
 
diff --git a/block/blk.h b/block/blk.h
index 24b92bd78f37..a09c18b19116 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -60,6 +60,9 @@ static inline struct request *__elv_next_request(struct request_queue *q)
 	while (1) {
 		while (!list_empty(&q->queue_head)) {
 			rq = list_entry_rq(q->queue_head.next);
+			if (!(rq->cmd_flags & (REQ_FLUSH | REQ_FUA)) ||
+			    rq == &q->flush_rq)
+				return rq;
 			rq = blk_do_flush(q, rq);
 			if (rq)
 				return rq;
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index 9192282b4259..179799479e6f 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -167,7 +167,7 @@ enum rq_flag_bits {
 	(REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | REQ_FAILFAST_DRIVER)
 #define REQ_COMMON_MASK \
 	(REQ_WRITE | REQ_FAILFAST_MASK | REQ_HARDBARRIER | REQ_SYNC | \
-	 REQ_META| REQ_DISCARD | REQ_NOIDLE)
+	 REQ_META | REQ_DISCARD | REQ_NOIDLE | REQ_FLUSH | REQ_FUA)
 
 #define REQ_UNPLUG		(1 << __REQ_UNPLUG)
 #define REQ_RAHEAD		(1 << __REQ_RAHEAD)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 1cd83ec077db..8ef705f800ab 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -357,7 +357,6 @@ struct request_queue
 	/*
 	 * for flush operations
 	 */
-	unsigned int		ordered, next_ordered;
 	unsigned int		flush_flags;
 	unsigned int		flush_seq;
 	int			flush_err;
@@ -465,40 +464,6 @@ static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
 	__clear_bit(flag, &q->queue_flags);
 }
 
-enum {
-	/*
-	 * Hardbarrier is supported with one of the following methods.
-	 *
-	 * NONE		: hardbarrier unsupported
-	 * DRAIN	: ordering by draining is enough
-	 * DRAIN_FLUSH	: ordering by draining w/ pre and post flushes
-	 * DRAIN_FUA	: ordering by draining w/ pre flush and FUA write
-	 */
-	QUEUE_ORDERED_DO_PREFLUSH	= 0x10,
-	QUEUE_ORDERED_DO_BAR		= 0x20,
-	QUEUE_ORDERED_DO_POSTFLUSH	= 0x40,
-	QUEUE_ORDERED_DO_FUA		= 0x80,
-
-	QUEUE_ORDERED_NONE		= 0x00,
-
-	QUEUE_ORDERED_DRAIN		= QUEUE_ORDERED_DO_BAR,
-	QUEUE_ORDERED_DRAIN_FLUSH	= QUEUE_ORDERED_DRAIN |
-					  QUEUE_ORDERED_DO_PREFLUSH |
-					  QUEUE_ORDERED_DO_POSTFLUSH,
-	QUEUE_ORDERED_DRAIN_FUA		= QUEUE_ORDERED_DRAIN |
-					  QUEUE_ORDERED_DO_PREFLUSH |
-					  QUEUE_ORDERED_DO_FUA,
-
-	/*
-	 * FLUSH/FUA sequences.
-	 */
-	QUEUE_FSEQ_STARTED	= (1 << 0), /* flushing in progress */
-	QUEUE_FSEQ_PREFLUSH	= (1 << 1), /* pre-flushing in progress */
-	QUEUE_FSEQ_DATA		= (1 << 2), /* data write in progress */
-	QUEUE_FSEQ_POSTFLUSH	= (1 << 3), /* post-flushing in progress */
-	QUEUE_FSEQ_DONE		= (1 << 4),
-};
-
 #define blk_queue_plugged(q)	test_bit(QUEUE_FLAG_PLUGGED, &(q)->queue_flags)
 #define blk_queue_tagged(q)	test_bit(QUEUE_FLAG_QUEUED, &(q)->queue_flags)
 #define blk_queue_stopped(q)	test_bit(QUEUE_FLAG_STOPPED, &(q)->queue_flags)
@@ -578,7 +543,8 @@ static inline void blk_clear_queue_full(struct request_queue *q, int sync)
  * it already be started by driver.
  */
 #define RQ_NOMERGE_FLAGS	\
-	(REQ_NOMERGE | REQ_STARTED | REQ_HARDBARRIER | REQ_SOFTBARRIER)
+	(REQ_NOMERGE | REQ_STARTED | REQ_HARDBARRIER | REQ_SOFTBARRIER | \
+	 REQ_FLUSH | REQ_FUA)
 #define rq_mergeable(rq)	\
 	(!((rq)->cmd_flags & RQ_NOMERGE_FLAGS) && \
 	 (((rq)->cmd_flags & REQ_DISCARD) || \
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index ec94c12f21da..fc999f583fda 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
@@ -32,7 +32,7 @@ enum bh_state_bits {
 	BH_Delay,	/* Buffer is not yet allocated on disk */
 	BH_Boundary,	/* Block is followed by a discontiguity */
 	BH_Write_EIO,	/* I/O error on write */
-	BH_Eopnotsupp,	/* operation not supported (barrier) */
+	BH_Eopnotsupp,	/* DEPRECATED: operation not supported (barrier) */
 	BH_Unwritten,	/* Buffer is allocated on disk but not written */
 	BH_Quiet,	/* Buffer Error Prinks to be quiet */
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 76041b614758..352c48627381 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -135,12 +135,13 @@ struct inodes_stat_t {
  *			immediately after submission. The write equivalent
  *			of READ_SYNC.
  * WRITE_ODIRECT_PLUG	Special case write for O_DIRECT only.
- * WRITE_BARRIER	Like WRITE_SYNC, but tells the block layer that all
- *			previously submitted writes must be safely on storage
- *			before this one is started. Also guarantees that when
- *			this write is complete, it itself is also safely on
- *			storage. Prevents reordering of writes on both sides
- *			of this IO.
+ * WRITE_BARRIER	DEPRECATED. Always fails. Use FLUSH/FUA instead.
+ * WRITE_FLUSH		Like WRITE_SYNC but with preceding cache flush.
+ * WRITE_FUA		Like WRITE_SYNC but data is guaranteed to be on
+ *			non-volatile media on completion.
+ * WRITE_FLUSH_FUA	Combination of WRITE_FLUSH and FUA. The IO is preceded
+ *			by a cache flush and data is guaranteed to be on
+ *			non-volatile media on completion.
  *
  */
 #define RW_MASK			REQ_WRITE
@@ -158,6 +159,12 @@ struct inodes_stat_t {
 #define WRITE_META		(WRITE | REQ_META)
 #define WRITE_BARRIER		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG | \
 				 REQ_HARDBARRIER)
+#define WRITE_FLUSH		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG | \
+				 REQ_FLUSH)
+#define WRITE_FUA		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG | \
+				 REQ_FUA)
+#define WRITE_FLUSH_FUA		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG | \
+				 REQ_FLUSH | REQ_FUA)
 
 /*
  * These aren't really reads or writes, they pass down information about
-- 
cgit v1.2.3


From 3a2edd0d6ddbd5fa3b389ea6db811285415ce6c8 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Fri, 3 Sep 2010 11:56:18 +0200
Subject: block: make __blk_rq_prep_clone() copy most command flags

Currently __blk_rq_prep_clone() copies only REQ_WRITE and REQ_DISCARD.
There's no reason to omit other command flags and REQ_FUA needs to be
copied to implement FUA support in request-based dm.

REQ_COMMON_MASK which specifies flags to be copied from bio to request
already identifies all the command flags.  Define REQ_CLONE_MASK to be
the same as REQ_COMMON_MASK for clarity and make __blk_rq_prep_clone()
copy all flags in the mask.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-core.c          | 4 +---
 include/linux/blk_types.h | 1 +
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-core.c b/block/blk-core.c
index 495bdc4a23da..2a5b19204546 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -2505,9 +2505,7 @@ EXPORT_SYMBOL_GPL(blk_rq_unprep_clone);
 static void __blk_rq_prep_clone(struct request *dst, struct request *src)
 {
 	dst->cpu = src->cpu;
-	dst->cmd_flags = (rq_data_dir(src) | REQ_NOMERGE);
-	if (src->cmd_flags & REQ_DISCARD)
-		dst->cmd_flags |= REQ_DISCARD;
+	dst->cmd_flags = (src->cmd_flags & REQ_CLONE_MASK) | REQ_NOMERGE;
 	dst->cmd_type = src->cmd_type;
 	dst->__sector = blk_rq_pos(src);
 	dst->__data_len = blk_rq_bytes(src);
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index 179799479e6f..36edadf5b41a 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -168,6 +168,7 @@ enum rq_flag_bits {
 #define REQ_COMMON_MASK \
 	(REQ_WRITE | REQ_FAILFAST_MASK | REQ_HARDBARRIER | REQ_SYNC | \
 	 REQ_META | REQ_DISCARD | REQ_NOIDLE | REQ_FLUSH | REQ_FUA)
+#define REQ_CLONE_MASK		REQ_COMMON_MASK
 
 #define REQ_UNPLUG		(1 << __REQ_UNPLUG)
 #define REQ_RAHEAD		(1 << __REQ_RAHEAD)
-- 
cgit v1.2.3


From 2cf6d26a354ab6362e301b5a323832b02867df47 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@infradead.org>
Date: Wed, 18 Aug 2010 05:29:10 -0400
Subject: block: pass gfp_mask and flags to sb_issue_discard

We'll need to get rid of the BLKDEV_IFL_BARRIER flag, and to facilitate
that and to make the interface less confusing pass all flags explicitly.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 fs/ext4/mballoc.c      |  3 ++-
 fs/fat/fatent.c        |  4 +++-
 include/linux/blkdev.h | 11 +++++------
 3 files changed, 10 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 4b4ad4b7ce57..df44b345f662 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2566,7 +2566,8 @@ static inline void ext4_issue_discard(struct super_block *sb,
 	discard_block = block + ext4_group_first_block_no(sb, block_group);
 	trace_ext4_discard_blocks(sb,
 			(unsigned long long) discard_block, count);
-	ret = sb_issue_discard(sb, discard_block, count);
+	ret = sb_issue_discard(sb, discard_block, count, GFP_NOFS,
+			       BLKDEV_IFL_WAIT | BLKDEV_IFL_BARRIER);
 	if (ret == EOPNOTSUPP) {
 		ext4_warning(sb, "discard not supported, disabling");
 		clear_opt(EXT4_SB(sb)->s_mount_opt, DISCARD);
diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c
index 81184d3b75a3..3a56a82f5658 100644
--- a/fs/fat/fatent.c
+++ b/fs/fat/fatent.c
@@ -577,7 +577,9 @@ int fat_free_clusters(struct inode *inode, int cluster)
 
 				sb_issue_discard(sb,
 					fat_clus_to_blknr(sbi, first_cl),
-					nr_clus * sbi->sec_per_clus);
+					nr_clus * sbi->sec_per_clus,
+					GFP_NOFS,
+					BLKDEV_IFL_WAIT | BLKDEV_IFL_BARRIER);
 
 				first_cl = cluster;
 			}
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 8ef705f800ab..6b305eb4a343 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -881,13 +881,12 @@ extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
 extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
 			sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
-static inline int sb_issue_discard(struct super_block *sb,
-				   sector_t block, sector_t nr_blocks)
+static inline int sb_issue_discard(struct super_block *sb, sector_t block,
+		sector_t nr_blocks, gfp_t gfp_mask, unsigned long flags)
 {
-	block <<= (sb->s_blocksize_bits - 9);
-	nr_blocks <<= (sb->s_blocksize_bits - 9);
-	return blkdev_issue_discard(sb->s_bdev, block, nr_blocks, GFP_NOFS,
-				   BLKDEV_IFL_WAIT | BLKDEV_IFL_BARRIER);
+	return blkdev_issue_discard(sb->s_bdev, block << (sb->s_blocksize_bits - 9),
+				    nr_blocks << (sb->s_blocksize_bits - 9),
+				    gfp_mask, flags);
 }
 
 extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
-- 
cgit v1.2.3


From 31725e65c7214b52b607eba705fc4f306be4d5a5 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@infradead.org>
Date: Wed, 18 Aug 2010 05:29:21 -0400
Subject: block: remove the WRITE_BARRIER flag

It's unused now.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 include/linux/fs.h | 3 ---
 1 file changed, 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 352c48627381..d6add69bc170 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -135,7 +135,6 @@ struct inodes_stat_t {
  *			immediately after submission. The write equivalent
  *			of READ_SYNC.
  * WRITE_ODIRECT_PLUG	Special case write for O_DIRECT only.
- * WRITE_BARRIER	DEPRECATED. Always fails. Use FLUSH/FUA instead.
  * WRITE_FLUSH		Like WRITE_SYNC but with preceding cache flush.
  * WRITE_FUA		Like WRITE_SYNC but data is guaranteed to be on
  *			non-volatile media on completion.
@@ -157,8 +156,6 @@ struct inodes_stat_t {
 #define WRITE_SYNC		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG)
 #define WRITE_ODIRECT_PLUG	(WRITE | REQ_SYNC)
 #define WRITE_META		(WRITE | REQ_META)
-#define WRITE_BARRIER		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG | \
-				 REQ_HARDBARRIER)
 #define WRITE_FLUSH		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG | \
 				 REQ_FLUSH)
 #define WRITE_FUA		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG | \
-- 
cgit v1.2.3


From 8c5553678237b7121355108e03c36086037d8975 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@infradead.org>
Date: Wed, 18 Aug 2010 05:29:22 -0400
Subject: block: remove the BLKDEV_IFL_BARRIER flag

Remove support for barriers on discards, which is unused now.  Also
remove the DISCARD_NOBARRIER I/O type in favour of just setting the
rw flags up locally in blkdev_issue_discard.

tj: Also remove DISCARD_SECURE and use REQ_SECURE directly.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-lib.c        | 18 ++----------------
 include/linux/blkdev.h |  2 --
 include/linux/fs.h     |  8 --------
 3 files changed, 2 insertions(+), 26 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-lib.c b/block/blk-lib.c
index c392029a104e..fe2e6ed0f510 100644
--- a/block/blk-lib.c
+++ b/block/blk-lib.c
@@ -39,8 +39,7 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 {
 	DECLARE_COMPLETION_ONSTACK(wait);
 	struct request_queue *q = bdev_get_queue(bdev);
-	int type = flags & BLKDEV_IFL_BARRIER ?
-		DISCARD_BARRIER : DISCARD_NOBARRIER;
+	int type = REQ_WRITE | REQ_DISCARD;
 	unsigned int max_discard_sectors;
 	struct bio *bio;
 	int ret = 0;
@@ -65,7 +64,7 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 	if (flags & BLKDEV_IFL_SECURE) {
 		if (!blk_queue_secdiscard(q))
 			return -EOPNOTSUPP;
-		type |= DISCARD_SECURE;
+		type |= REQ_SECURE;
 	}
 
 	while (nr_sects && !ret) {
@@ -162,12 +161,6 @@ int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
 	bb.wait = &wait;
 	bb.end_io = NULL;
 
-	if (flags & BLKDEV_IFL_BARRIER) {
-		/* issue async barrier before the data */
-		ret = blkdev_issue_flush(bdev, gfp_mask, NULL, 0);
-		if (ret)
-			return ret;
-	}
 submit:
 	ret = 0;
 	while (nr_sects != 0) {
@@ -199,13 +192,6 @@ submit:
 		issued++;
 		submit_bio(WRITE, bio);
 	}
-	/*
-	 * When all data bios are in flight. Send final barrier if requeted.
-	 */
-	if (nr_sects == 0 && flags & BLKDEV_IFL_BARRIER)
-		ret = blkdev_issue_flush(bdev, gfp_mask, NULL,
-					flags & BLKDEV_IFL_WAIT);
-
 
 	if (flags & BLKDEV_IFL_WAIT)
 		/* Wait for bios in-flight */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 6b305eb4a343..cfcb3a610605 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -869,11 +869,9 @@ static inline struct request *blk_map_queue_find_tag(struct blk_queue_tag *bqt,
 }
 enum{
 	BLKDEV_WAIT,	/* wait for completion */
-	BLKDEV_BARRIER,	/* issue request with barrier */
 	BLKDEV_SECURE,	/* secure discard */
 };
 #define BLKDEV_IFL_WAIT		(1 << BLKDEV_WAIT)
-#define BLKDEV_IFL_BARRIER	(1 << BLKDEV_BARRIER)
 #define BLKDEV_IFL_SECURE	(1 << BLKDEV_SECURE)
 extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *,
 			unsigned long);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index d6add69bc170..6b0f6e9993a3 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -163,14 +163,6 @@ struct inodes_stat_t {
 #define WRITE_FLUSH_FUA		(WRITE | REQ_SYNC | REQ_NOIDLE | REQ_UNPLUG | \
 				 REQ_FLUSH | REQ_FUA)
 
-/*
- * These aren't really reads or writes, they pass down information about
- * parts of device that are now unused by the file system.
- */
-#define DISCARD_NOBARRIER	(WRITE | REQ_DISCARD)
-#define DISCARD_BARRIER		(WRITE | REQ_DISCARD | REQ_HARDBARRIER)
-#define DISCARD_SECURE		(DISCARD_NOBARRIER | REQ_SECURE)
-
 #define SEL_IN		1
 #define SEL_OUT		2
 #define SEL_EX		4
-- 
cgit v1.2.3


From 0edd55faea7c8081bc826234b917501738a6218f Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@infradead.org>
Date: Wed, 18 Aug 2010 05:29:23 -0400
Subject: block: remove the BH_Eopnotsupp flag

This flag was only set for barrier buffers, which we don't submit
anymore.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 fs/buffer.c                 | 7 +------
 fs/fat/misc.c               | 5 +----
 include/linux/buffer_head.h | 2 --
 3 files changed, 2 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/fs/buffer.c b/fs/buffer.c
index 3e7dca279d1c..7f0b9b083f77 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -156,7 +156,7 @@ void end_buffer_write_sync(struct buffer_head *bh, int uptodate)
 	if (uptodate) {
 		set_buffer_uptodate(bh);
 	} else {
-		if (!buffer_eopnotsupp(bh) && !quiet_error(bh)) {
+		if (!quiet_error(bh)) {
 			buffer_io_error(bh);
 			printk(KERN_WARNING "lost page write due to "
 					"I/O error on %s\n",
@@ -2891,7 +2891,6 @@ static void end_bio_bh_io_sync(struct bio *bio, int err)
 
 	if (err == -EOPNOTSUPP) {
 		set_bit(BIO_EOPNOTSUPP, &bio->bi_flags);
-		set_bit(BH_Eopnotsupp, &bh->b_state);
 	}
 
 	if (unlikely (test_bit(BIO_QUIET,&bio->bi_flags)))
@@ -3031,10 +3030,6 @@ int __sync_dirty_buffer(struct buffer_head *bh, int rw)
 		bh->b_end_io = end_buffer_write_sync;
 		ret = submit_bh(rw, bh);
 		wait_on_buffer(bh);
-		if (buffer_eopnotsupp(bh)) {
-			clear_buffer_eopnotsupp(bh);
-			ret = -EOPNOTSUPP;
-		}
 		if (!ret && !buffer_uptodate(bh))
 			ret = -EIO;
 	} else {
diff --git a/fs/fat/misc.c b/fs/fat/misc.c
index 1736f2356388..970e682ea754 100644
--- a/fs/fat/misc.c
+++ b/fs/fat/misc.c
@@ -255,10 +255,7 @@ int fat_sync_bhs(struct buffer_head **bhs, int nr_bhs)
 
 	for (i = 0; i < nr_bhs; i++) {
 		wait_on_buffer(bhs[i]);
-		if (buffer_eopnotsupp(bhs[i])) {
-			clear_buffer_eopnotsupp(bhs[i]);
-			err = -EOPNOTSUPP;
-		} else if (!err && !buffer_uptodate(bhs[i]))
+		if (!err && !buffer_uptodate(bhs[i]))
 			err = -EIO;
 	}
 	return err;
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index fc999f583fda..dd1b25b2641c 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
@@ -32,7 +32,6 @@ enum bh_state_bits {
 	BH_Delay,	/* Buffer is not yet allocated on disk */
 	BH_Boundary,	/* Block is followed by a discontiguity */
 	BH_Write_EIO,	/* I/O error on write */
-	BH_Eopnotsupp,	/* DEPRECATED: operation not supported (barrier) */
 	BH_Unwritten,	/* Buffer is allocated on disk but not written */
 	BH_Quiet,	/* Buffer Error Prinks to be quiet */
 
@@ -124,7 +123,6 @@ BUFFER_FNS(Async_Write, async_write)
 BUFFER_FNS(Delay, delay)
 BUFFER_FNS(Boundary, boundary)
 BUFFER_FNS(Write_EIO, write_io_error)
-BUFFER_FNS(Eopnotsupp, eopnotsupp)
 BUFFER_FNS(Unwritten, unwritten)
 
 #define bh_offset(bh)		((unsigned long)(bh)->b_data & ~PAGE_MASK)
-- 
cgit v1.2.3


From 30ca22c70e3ef0a96ff84de69cd7e8561b416cb2 Mon Sep 17 00:00:00 2001
From: "Patrick J. LoPresti" <lopresti@gmail.com>
Date: Thu, 22 Jul 2010 15:03:41 -0700
Subject: ext3/ext4: Factor out disk addressability check

As part of adding support for OCFS2 to mount huge volumes, we need to
check that the sector_t and page cache of the system are capable of
addressing the entire volume.

An identical check already appears in ext3 and ext4.  This patch moves
the addressability check into its own function in fs/libfs.c and
modifies ext3 and ext4 to invoke it.

[Edited to -EINVAL instead of BUG_ON() for bad blocksize_bits -- Joel]

Signed-off-by: Patrick LoPresti <lopresti@gmail.com>
Cc: linux-ext4@vger.kernel.org
Acked-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ext3/super.c    |  4 ++--
 fs/ext4/super.c    |  8 +++-----
 fs/libfs.c         | 29 +++++++++++++++++++++++++++++
 include/linux/fs.h |  2 ++
 4 files changed, 36 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 5dbf4dba03c4..a367dd044280 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -1849,8 +1849,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 		goto failed_mount;
 	}
 
-	if (le32_to_cpu(es->s_blocks_count) >
-		    (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) {
+	if (generic_check_addressable(sb->s_blocksize_bits,
+				      le32_to_cpu(es->s_blocks_count))) {
 		ext3_msg(sb, KERN_ERR,
 			"error: filesystem is too large to mount safely");
 		if (sizeof(sector_t) < 8)
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 26147746c272..7f47c366bf15 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -2831,15 +2831,13 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	 * Test whether we have more sectors than will fit in sector_t,
 	 * and whether the max offset is addressable by the page cache.
 	 */
-	if ((ext4_blocks_count(es) >
-	     (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) ||
-	    (ext4_blocks_count(es) >
-	     (pgoff_t)(~0ULL) >> (PAGE_CACHE_SHIFT - sb->s_blocksize_bits))) {
+	ret = generic_check_addressable(sb->s_blocksize_bits,
+					ext4_blocks_count(es));
+	if (ret) {
 		ext4_msg(sb, KERN_ERR, "filesystem"
 			 " too large to mount safely on this system");
 		if (sizeof(sector_t) < 8)
 			ext4_msg(sb, KERN_WARNING, "CONFIG_LBDAF not enabled");
-		ret = -EFBIG;
 		goto failed_mount;
 	}
 
diff --git a/fs/libfs.c b/fs/libfs.c
index 0a9da95317f7..8debe7b33769 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -913,6 +913,35 @@ int generic_file_fsync(struct file *file, int datasync)
 }
 EXPORT_SYMBOL(generic_file_fsync);
 
+/**
+ * generic_check_addressable - Check addressability of file system
+ * @blocksize_bits:	log of file system block size
+ * @num_blocks:		number of blocks in file system
+ *
+ * Determine whether a file system with @num_blocks blocks (and a
+ * block size of 2**@blocksize_bits) is addressable by the sector_t
+ * and page cache of the system.  Return 0 if so and -EFBIG otherwise.
+ */
+int generic_check_addressable(unsigned blocksize_bits, u64 num_blocks)
+{
+	u64 last_fs_block = num_blocks - 1;
+
+	if (unlikely(num_blocks == 0))
+		return 0;
+
+	if ((blocksize_bits < 9) || (blocksize_bits > PAGE_CACHE_SHIFT))
+		return -EINVAL;
+
+	if ((last_fs_block >
+	     (sector_t)(~0ULL) >> (blocksize_bits - 9)) ||
+	    (last_fs_block >
+	     (pgoff_t)(~0ULL) >> (PAGE_CACHE_SHIFT - blocksize_bits))) {
+		return -EFBIG;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(generic_check_addressable);
+
 /*
  * No-op implementation of ->fsync for in-memory filesystems.
  */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 76041b614758..1a759f40ab9e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2374,6 +2374,8 @@ extern ssize_t simple_write_to_buffer(void *to, size_t available, loff_t *ppos,
 
 extern int generic_file_fsync(struct file *, int);
 
+extern int generic_check_addressable(unsigned, u64);
+
 #ifdef CONFIG_MIGRATION
 extern int buffer_migrate_page(struct address_space *,
 				struct page *, struct page *);
-- 
cgit v1.2.3


From c8bf1336824ebd698d37b71763e1c43190f2229a Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 10 Sep 2010 20:07:38 +0200
Subject: Consolidate min_not_zero

We have several users of min_not_zero, each of them using their own
definition.  Move the define to kernel.h.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@carl.home.kernel.dk>
---
 block/blk-settings.c               |  5 -----
 drivers/block/drbd/drbd_receiver.c |  1 -
 drivers/md/dm-snap.c               |  2 --
 drivers/md/dm-table.c              |  5 -----
 include/linux/kernel.h             | 10 ++++++++++
 5 files changed, 10 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-settings.c b/block/blk-settings.c
index a234f4bf1d6f..8d592b559bd3 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -455,11 +455,6 @@ void blk_queue_io_opt(struct request_queue *q, unsigned int opt)
 }
 EXPORT_SYMBOL(blk_queue_io_opt);
 
-/*
- * Returns the minimum that is _not_ zero, unless both are zero.
- */
-#define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r))
-
 /**
  * blk_queue_stack_limits - inherit underlying queue limits for stacked drivers
  * @t:	the stacking driver (top)
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 081522d3c742..484ecbb6b772 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -2972,7 +2972,6 @@ static int receive_sizes(struct drbd_conf *mdev, struct p_header *h)
 	 * we still need to figure out whether we accept that. */
 	mdev->p_size = p_size;
 
-#define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r))
 	if (get_ldev(mdev)) {
 		warn_if_differ_considerably(mdev, "lower level device sizes",
 			   p_size, drbd_get_max_capacity(mdev->ldev));
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 5974d3094d97..f30f6e8d594e 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -706,8 +706,6 @@ static int dm_add_exception(void *context, chunk_t old, chunk_t new)
 	return 0;
 }
 
-#define min_not_zero(l, r) (((l) == 0) ? (r) : (((r) == 0) ? (l) : min(l, r)))
-
 /*
  * Return a minimum chunk size of all snapshots that have the specified origin.
  * Return zero if the origin has no snapshots.
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index f9fc07d7a4b9..90267f8d64ee 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -486,11 +486,6 @@ static int __table_get_device(struct dm_table *t, struct dm_target *ti,
 	return 0;
 }
 
-/*
- * Returns the minimum that is _not_ zero, unless both are zero.
- */
-#define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r))
-
 int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
 			 sector_t start, sector_t len, void *data)
 {
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 2b0a35e6bc69..f5df2f4acb0d 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -640,6 +640,16 @@ static inline void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) { }
 	(void) (&_max1 == &_max2);		\
 	_max1 > _max2 ? _max1 : _max2; })
 
+/**
+ * min_not_zero - return the minimum that is _not_ zero, unless both are zero
+ * @x: value1
+ * @y: value2
+ */
+#define min_not_zero(x, y) ({			\
+	typeof(x) __x = (x);			\
+	typeof(y) __y = (y);			\
+	__x == 0 ? __y : ((__y == 0) ? __x : min(__x, __y)); })
+
 /**
  * clamp - return a value clamped to a given range with strict typechecking
  * @val: current value
-- 
cgit v1.2.3


From 13f05c8d8e98bbdce89158bfdb2e380940695a88 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 10 Sep 2010 20:50:10 +0200
Subject: block/scsi: Provide a limit on the number of integrity segments

Some controllers have a hardware limit on the number of protection
information scatter-gather list segments they can handle.

Introduce a max_integrity_segments limit in the block layer and provide
a new scsi_host_template setting that allows HBA drivers to provide a
value suitable for the hardware.

Add support for honoring the integrity segment limit when merging both
bios and requests.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@carl.home.kernel.dk>
---
 block/blk-integrity.c     | 93 ++++++++++++++++++++++++++++++++++++-----------
 block/blk-merge.c         | 23 +++++++-----
 block/blk-settings.c      |  3 ++
 block/blk-sysfs.c         | 11 ++++++
 block/blk.h               |  8 ----
 drivers/scsi/hosts.c      |  1 +
 drivers/scsi/scsi_lib.c   | 26 +++++++++----
 drivers/scsi/scsi_sysfs.c |  2 +
 include/linux/bio.h       |  4 ++
 include/linux/blkdev.h    | 33 +++++++++++++++--
 include/scsi/scsi.h       |  6 +++
 include/scsi/scsi_host.h  |  7 ++++
 12 files changed, 167 insertions(+), 50 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-integrity.c b/block/blk-integrity.c
index edce1ef7933d..885cbb59967e 100644
--- a/block/blk-integrity.c
+++ b/block/blk-integrity.c
@@ -32,24 +32,37 @@ static struct kmem_cache *integrity_cachep;
 
 /**
  * blk_rq_count_integrity_sg - Count number of integrity scatterlist elements
- * @rq:		request with integrity metadata attached
+ * @q:		request queue
+ * @bio:	bio with integrity metadata attached
  *
  * Description: Returns the number of elements required in a
- * scatterlist corresponding to the integrity metadata in a request.
+ * scatterlist corresponding to the integrity metadata in a bio.
  */
-int blk_rq_count_integrity_sg(struct request *rq)
+int blk_rq_count_integrity_sg(struct request_queue *q, struct bio *bio)
 {
-	struct bio_vec *iv, *ivprv;
-	struct req_iterator iter;
-	unsigned int segments;
+	struct bio_vec *iv, *ivprv = NULL;
+	unsigned int segments = 0;
+	unsigned int seg_size = 0;
+	unsigned int i = 0;
 
-	ivprv = NULL;
-	segments = 0;
+	bio_for_each_integrity_vec(iv, bio, i) {
 
-	rq_for_each_integrity_segment(iv, rq, iter) {
+		if (ivprv) {
+			if (!BIOVEC_PHYS_MERGEABLE(ivprv, iv))
+				goto new_segment;
+
+			if (!BIOVEC_SEG_BOUNDARY(q, ivprv, iv))
+				goto new_segment;
+
+			if (seg_size + iv->bv_len > queue_max_segment_size(q))
+				goto new_segment;
 
-		if (!ivprv || !BIOVEC_PHYS_MERGEABLE(ivprv, iv))
+			seg_size += iv->bv_len;
+		} else {
+new_segment:
 			segments++;
+			seg_size = iv->bv_len;
+		}
 
 		ivprv = iv;
 	}
@@ -60,30 +73,34 @@ EXPORT_SYMBOL(blk_rq_count_integrity_sg);
 
 /**
  * blk_rq_map_integrity_sg - Map integrity metadata into a scatterlist
- * @rq:		request with integrity metadata attached
+ * @q:		request queue
+ * @bio:	bio with integrity metadata attached
  * @sglist:	target scatterlist
  *
  * Description: Map the integrity vectors in request into a
  * scatterlist.  The scatterlist must be big enough to hold all
  * elements.  I.e. sized using blk_rq_count_integrity_sg().
  */
-int blk_rq_map_integrity_sg(struct request *rq, struct scatterlist *sglist)
+int blk_rq_map_integrity_sg(struct request_queue *q, struct bio *bio,
+			    struct scatterlist *sglist)
 {
-	struct bio_vec *iv, *ivprv;
-	struct req_iterator iter;
-	struct scatterlist *sg;
-	unsigned int segments;
+	struct bio_vec *iv, *ivprv = NULL;
+	struct scatterlist *sg = NULL;
+	unsigned int segments = 0;
+	unsigned int i = 0;
 
-	ivprv = NULL;
-	sg = NULL;
-	segments = 0;
-
-	rq_for_each_integrity_segment(iv, rq, iter) {
+	bio_for_each_integrity_vec(iv, bio, i) {
 
 		if (ivprv) {
 			if (!BIOVEC_PHYS_MERGEABLE(ivprv, iv))
 				goto new_segment;
 
+			if (!BIOVEC_SEG_BOUNDARY(q, ivprv, iv))
+				goto new_segment;
+
+			if (sg->length + iv->bv_len > queue_max_segment_size(q))
+				goto new_segment;
+
 			sg->length += iv->bv_len;
 		} else {
 new_segment:
@@ -162,6 +179,40 @@ int blk_integrity_compare(struct gendisk *gd1, struct gendisk *gd2)
 }
 EXPORT_SYMBOL(blk_integrity_compare);
 
+int blk_integrity_merge_rq(struct request_queue *q, struct request *req,
+			   struct request *next)
+{
+	if (blk_integrity_rq(req) != blk_integrity_rq(next))
+		return -1;
+
+	if (req->nr_integrity_segments + next->nr_integrity_segments >
+	    q->limits.max_integrity_segments)
+		return -1;
+
+	return 0;
+}
+EXPORT_SYMBOL(blk_integrity_merge_rq);
+
+int blk_integrity_merge_bio(struct request_queue *q, struct request *req,
+			    struct bio *bio)
+{
+	int nr_integrity_segs;
+	struct bio *next = bio->bi_next;
+
+	bio->bi_next = NULL;
+	nr_integrity_segs = blk_rq_count_integrity_sg(q, bio);
+	bio->bi_next = next;
+
+	if (req->nr_integrity_segments + nr_integrity_segs >
+	    q->limits.max_integrity_segments)
+		return -1;
+
+	req->nr_integrity_segments += nr_integrity_segs;
+
+	return 0;
+}
+EXPORT_SYMBOL(blk_integrity_merge_bio);
+
 struct integrity_sysfs_entry {
 	struct attribute attr;
 	ssize_t (*show)(struct blk_integrity *, char *);
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 3b0cd4249671..6a725461654d 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -205,12 +205,11 @@ static inline int ll_new_hw_segment(struct request_queue *q,
 {
 	int nr_phys_segs = bio_phys_segments(q, bio);
 
-	if (req->nr_phys_segments + nr_phys_segs > queue_max_segments(q)) {
-		req->cmd_flags |= REQ_NOMERGE;
-		if (req == q->last_merge)
-			q->last_merge = NULL;
-		return 0;
-	}
+	if (req->nr_phys_segments + nr_phys_segs > queue_max_segments(q))
+		goto no_merge;
+
+	if (bio_integrity(bio) && blk_integrity_merge_bio(q, req, bio))
+		goto no_merge;
 
 	/*
 	 * This will form the start of a new hw segment.  Bump both
@@ -218,6 +217,12 @@ static inline int ll_new_hw_segment(struct request_queue *q,
 	 */
 	req->nr_phys_segments += nr_phys_segs;
 	return 1;
+
+no_merge:
+	req->cmd_flags |= REQ_NOMERGE;
+	if (req == q->last_merge)
+		q->last_merge = NULL;
+	return 0;
 }
 
 int ll_back_merge_fn(struct request_queue *q, struct request *req,
@@ -301,6 +306,9 @@ static int ll_merge_requests_fn(struct request_queue *q, struct request *req,
 	if (total_phys_segments > queue_max_segments(q))
 		return 0;
 
+	if (blk_integrity_rq(req) && blk_integrity_merge_rq(q, req, next))
+		return 0;
+
 	/* Merge is OK... */
 	req->nr_phys_segments = total_phys_segments;
 	return 1;
@@ -372,9 +380,6 @@ static int attempt_merge(struct request_queue *q, struct request *req,
 	    || next->special)
 		return 0;
 
-	if (blk_integrity_rq(req) != blk_integrity_rq(next))
-		return 0;
-
 	/*
 	 * If we are allowed to merge, then append bio list
 	 * from next to rq and release next. merge_requests_fn
diff --git a/block/blk-settings.c b/block/blk-settings.c
index 8d592b559bd3..f8f2ddf20613 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -111,6 +111,7 @@ EXPORT_SYMBOL_GPL(blk_queue_lld_busy);
 void blk_set_default_limits(struct queue_limits *lim)
 {
 	lim->max_segments = BLK_MAX_SEGMENTS;
+	lim->max_integrity_segments = 0;
 	lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
 	lim->max_segment_size = BLK_MAX_SEGMENT_SIZE;
 	lim->max_sectors = BLK_DEF_MAX_SECTORS;
@@ -509,6 +510,8 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
 					    b->seg_boundary_mask);
 
 	t->max_segments = min_not_zero(t->max_segments, b->max_segments);
+	t->max_integrity_segments = min_not_zero(t->max_integrity_segments,
+						 b->max_integrity_segments);
 
 	t->max_segment_size = min_not_zero(t->max_segment_size,
 					   b->max_segment_size);
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 001ab18078f5..b014f7739e98 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -112,6 +112,11 @@ static ssize_t queue_max_segments_show(struct request_queue *q, char *page)
 	return queue_var_show(queue_max_segments(q), (page));
 }
 
+static ssize_t queue_max_integrity_segments_show(struct request_queue *q, char *page)
+{
+	return queue_var_show(q->limits.max_integrity_segments, (page));
+}
+
 static ssize_t queue_max_segment_size_show(struct request_queue *q, char *page)
 {
 	if (test_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags))
@@ -288,6 +293,11 @@ static struct queue_sysfs_entry queue_max_segments_entry = {
 	.show = queue_max_segments_show,
 };
 
+static struct queue_sysfs_entry queue_max_integrity_segments_entry = {
+	.attr = {.name = "max_integrity_segments", .mode = S_IRUGO },
+	.show = queue_max_integrity_segments_show,
+};
+
 static struct queue_sysfs_entry queue_max_segment_size_entry = {
 	.attr = {.name = "max_segment_size", .mode = S_IRUGO },
 	.show = queue_max_segment_size_show,
@@ -375,6 +385,7 @@ static struct attribute *default_attrs[] = {
 	&queue_max_hw_sectors_entry.attr,
 	&queue_max_sectors_entry.attr,
 	&queue_max_segments_entry.attr,
+	&queue_max_integrity_segments_entry.attr,
 	&queue_max_segment_size_entry.attr,
 	&queue_iosched_entry.attr,
 	&queue_hw_sector_size_entry.attr,
diff --git a/block/blk.h b/block/blk.h
index 6e7dc87141e4..6738831ba447 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -132,14 +132,6 @@ static inline int queue_congestion_off_threshold(struct request_queue *q)
 	return q->nr_congestion_off;
 }
 
-#if defined(CONFIG_BLK_DEV_INTEGRITY)
-
-#define rq_for_each_integrity_segment(bvl, _rq, _iter)		\
-	__rq_for_each_bio(_iter.bio, _rq)			\
-		bip_for_each_vec(bvl, _iter.bio->bi_integrity, _iter.i)
-
-#endif /* BLK_DEV_INTEGRITY */
-
 static inline int blk_cpu_to_group(int cpu)
 {
 #ifdef CONFIG_SCHED_MC
diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 8a8f803439e1..10478153641b 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -376,6 +376,7 @@ struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize)
 	shost->this_id = sht->this_id;
 	shost->can_queue = sht->can_queue;
 	shost->sg_tablesize = sht->sg_tablesize;
+	shost->sg_prot_tablesize = sht->sg_prot_tablesize;
 	shost->cmd_per_lun = sht->cmd_per_lun;
 	shost->unchecked_isa_dma = sht->unchecked_isa_dma;
 	shost->use_clustering = sht->use_clustering;
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 9ade720422c6..861c0b937ac9 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -968,11 +968,13 @@ static int scsi_init_sgtable(struct request *req, struct scsi_data_buffer *sdb,
  */
 int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 {
-	int error = scsi_init_sgtable(cmd->request, &cmd->sdb, gfp_mask);
+	struct request *rq = cmd->request;
+
+	int error = scsi_init_sgtable(rq, &cmd->sdb, gfp_mask);
 	if (error)
 		goto err_exit;
 
-	if (blk_bidi_rq(cmd->request)) {
+	if (blk_bidi_rq(rq)) {
 		struct scsi_data_buffer *bidi_sdb = kmem_cache_zalloc(
 			scsi_sdb_cache, GFP_ATOMIC);
 		if (!bidi_sdb) {
@@ -980,28 +982,28 @@ int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 			goto err_exit;
 		}
 
-		cmd->request->next_rq->special = bidi_sdb;
-		error = scsi_init_sgtable(cmd->request->next_rq, bidi_sdb,
-								    GFP_ATOMIC);
+		rq->next_rq->special = bidi_sdb;
+		error = scsi_init_sgtable(rq->next_rq, bidi_sdb, GFP_ATOMIC);
 		if (error)
 			goto err_exit;
 	}
 
-	if (blk_integrity_rq(cmd->request)) {
+	if (blk_integrity_rq(rq)) {
 		struct scsi_data_buffer *prot_sdb = cmd->prot_sdb;
 		int ivecs, count;
 
 		BUG_ON(prot_sdb == NULL);
-		ivecs = blk_rq_count_integrity_sg(cmd->request);
+		ivecs = blk_rq_count_integrity_sg(rq->q, rq->bio);
 
 		if (scsi_alloc_sgtable(prot_sdb, ivecs, gfp_mask)) {
 			error = BLKPREP_DEFER;
 			goto err_exit;
 		}
 
-		count = blk_rq_map_integrity_sg(cmd->request,
+		count = blk_rq_map_integrity_sg(rq->q, rq->bio,
 						prot_sdb->table.sgl);
 		BUG_ON(unlikely(count > ivecs));
+		BUG_ON(unlikely(count > queue_max_integrity_segments(rq->q)));
 
 		cmd->prot_sdb = prot_sdb;
 		cmd->prot_sdb->table.nents = count;
@@ -1625,6 +1627,14 @@ struct request_queue *__scsi_alloc_queue(struct Scsi_Host *shost,
 	blk_queue_max_segments(q, min_t(unsigned short, shost->sg_tablesize,
 					SCSI_MAX_SG_CHAIN_SEGMENTS));
 
+	if (scsi_host_prot_dma(shost)) {
+		shost->sg_prot_tablesize =
+			min_not_zero(shost->sg_prot_tablesize,
+				     (unsigned short)SCSI_MAX_PROT_SG_SEGMENTS);
+		BUG_ON(shost->sg_prot_tablesize < shost->sg_tablesize);
+		blk_queue_max_integrity_segments(q, shost->sg_prot_tablesize);
+	}
+
 	blk_queue_max_hw_sectors(q, shost->max_sectors);
 	blk_queue_bounce_limit(q, scsi_calculate_bounce_limit(shost));
 	blk_queue_segment_boundary(q, shost->dma_boundary);
diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c
index c3f67373a4f8..20ad59dff730 100644
--- a/drivers/scsi/scsi_sysfs.c
+++ b/drivers/scsi/scsi_sysfs.c
@@ -251,6 +251,7 @@ shost_rd_attr(host_busy, "%hu\n");
 shost_rd_attr(cmd_per_lun, "%hd\n");
 shost_rd_attr(can_queue, "%hd\n");
 shost_rd_attr(sg_tablesize, "%hu\n");
+shost_rd_attr(sg_prot_tablesize, "%hu\n");
 shost_rd_attr(unchecked_isa_dma, "%d\n");
 shost_rd_attr(prot_capabilities, "%u\n");
 shost_rd_attr(prot_guard_type, "%hd\n");
@@ -262,6 +263,7 @@ static struct attribute *scsi_sysfs_shost_attrs[] = {
 	&dev_attr_cmd_per_lun.attr,
 	&dev_attr_can_queue.attr,
 	&dev_attr_sg_tablesize.attr,
+	&dev_attr_sg_prot_tablesize.attr,
 	&dev_attr_unchecked_isa_dma.attr,
 	&dev_attr_proc_name.attr,
 	&dev_attr_scan.attr,
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 5274103434ad..2c3fd7421607 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -496,6 +496,10 @@ static inline struct bio *bio_list_get(struct bio_list *bl)
 #define bip_for_each_vec(bvl, bip, i)					\
 	__bip_for_each_vec(bvl, bip, i, (bip)->bip_idx)
 
+#define bio_for_each_integrity_vec(_bvl, _bio, _iter)			\
+	for_each_bio(_bio)						\
+		bip_for_each_vec(_bvl, _bio->bi_integrity, _iter)
+
 #define bio_integrity(bio) (bio->bi_integrity != NULL)
 
 extern struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *, gfp_t, unsigned int, struct bio_set *);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 2c54906f678f..7e661106270a 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -124,6 +124,9 @@ struct request {
 	 * physical address coalescing is performed.
 	 */
 	unsigned short nr_phys_segments;
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+	unsigned short nr_integrity_segments;
+#endif
 
 	unsigned short ioprio;
 
@@ -243,6 +246,7 @@ struct queue_limits {
 
 	unsigned short		logical_block_size;
 	unsigned short		max_segments;
+	unsigned short		max_integrity_segments;
 
 	unsigned char		misaligned;
 	unsigned char		discard_misaligned;
@@ -1213,8 +1217,13 @@ struct blk_integrity {
 extern int blk_integrity_register(struct gendisk *, struct blk_integrity *);
 extern void blk_integrity_unregister(struct gendisk *);
 extern int blk_integrity_compare(struct gendisk *, struct gendisk *);
-extern int blk_rq_map_integrity_sg(struct request *, struct scatterlist *);
-extern int blk_rq_count_integrity_sg(struct request *);
+extern int blk_rq_map_integrity_sg(struct request_queue *, struct bio *,
+				   struct scatterlist *);
+extern int blk_rq_count_integrity_sg(struct request_queue *, struct bio *);
+extern int blk_integrity_merge_rq(struct request_queue *, struct request *,
+				  struct request *);
+extern int blk_integrity_merge_bio(struct request_queue *, struct request *,
+				   struct bio *);
 
 static inline
 struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
@@ -1235,16 +1244,32 @@ static inline int blk_integrity_rq(struct request *rq)
 	return bio_integrity(rq->bio);
 }
 
+static inline void blk_queue_max_integrity_segments(struct request_queue *q,
+						    unsigned int segs)
+{
+	q->limits.max_integrity_segments = segs;
+}
+
+static inline unsigned short
+queue_max_integrity_segments(struct request_queue *q)
+{
+	return q->limits.max_integrity_segments;
+}
+
 #else /* CONFIG_BLK_DEV_INTEGRITY */
 
 #define blk_integrity_rq(rq)			(0)
-#define blk_rq_count_integrity_sg(a)		(0)
-#define blk_rq_map_integrity_sg(a, b)		(0)
+#define blk_rq_count_integrity_sg(a, b)		(0)
+#define blk_rq_map_integrity_sg(a, b, c)	(0)
 #define bdev_get_integrity(a)			(0)
 #define blk_get_integrity(a)			(0)
 #define blk_integrity_compare(a, b)		(0)
 #define blk_integrity_register(a, b)		(0)
 #define blk_integrity_unregister(a)		do { } while (0);
+#define blk_queue_max_integrity_segments(a, b)	do { } while (0);
+#define queue_max_integrity_segments(a)		(0)
+#define blk_integrity_merge_rq(a, b, c)		(0)
+#define blk_integrity_merge_bio(a, b, c)	(0)
 
 #endif /* CONFIG_BLK_DEV_INTEGRITY */
 
diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h
index 8fcb6e0e9e72..d63533a4a59e 100644
--- a/include/scsi/scsi.h
+++ b/include/scsi/scsi.h
@@ -31,6 +31,12 @@ struct scsi_cmnd;
 #define SCSI_MAX_SG_CHAIN_SEGMENTS	SCSI_MAX_SG_SEGMENTS
 #endif
 
+/*
+ * DIX-capable adapters effectively support infinite chaining for the
+ * protection information scatterlist
+ */
+#define SCSI_MAX_PROT_SG_SEGMENTS	0xFFFF
+
 /*
  * Special value for scanning to specify scanning or rescanning of all
  * possible channels, (target) ids, or luns on a given shost.
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index b7bdecb7b76e..d0a6a845f204 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -388,6 +388,7 @@ struct scsi_host_template {
 	 * of scatter-gather.
 	 */
 	unsigned short sg_tablesize;
+	unsigned short sg_prot_tablesize;
 
 	/*
 	 * Set this if the host adapter has limitations beside segment count.
@@ -599,6 +600,7 @@ struct Scsi_Host {
 	int can_queue;
 	short cmd_per_lun;
 	short unsigned int sg_tablesize;
+	short unsigned int sg_prot_tablesize;
 	short unsigned int max_sectors;
 	unsigned long dma_boundary;
 	/* 
@@ -823,6 +825,11 @@ static inline unsigned int scsi_host_get_prot(struct Scsi_Host *shost)
 	return shost->prot_capabilities;
 }
 
+static inline int scsi_host_prot_dma(struct Scsi_Host *shost)
+{
+	return shost->prot_capabilities >= SHOST_DIX_TYPE0_PROTECTION;
+}
+
 static inline unsigned int scsi_host_dif_capable(struct Scsi_Host *shost, unsigned int target_type)
 {
 	static unsigned char cap[] = { 0,
-- 
cgit v1.2.3


From 637bbdc5b83615ef9f45f50399d1c7f27473c713 Mon Sep 17 00:00:00 2001
From: Dave Young <hidave.darkstar@gmail.com>
Date: Mon, 13 Sep 2010 20:19:03 +0800
Subject: sched: Remove unused PF_ALIGNWARN flag

PF_ALIGNWARN is not implemented and it is for 486 as the
comment.

It is not likely someone will implement this flag feature.
So here remove this flag and leave the valuable 0x00000001 for
future use.

Signed-off-by: Dave Young <hidave.darkstar@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <20100913121903.GB22238@darkstar>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index b51c53c285b8..cdf56693ecbf 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1682,8 +1682,6 @@ extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *
 /*
  * Per process flags
  */
-#define PF_ALIGNWARN	0x00000001	/* Print alignment warning msgs */
-					/* Not implemented yet, only for 486*/
 #define PF_STARTING	0x00000002	/* being created */
 #define PF_EXITING	0x00000004	/* getting shut down */
 #define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
-- 
cgit v1.2.3


From 38a81da2205f94e8a2a834b51a6b99c91fc7c2e8 Mon Sep 17 00:00:00 2001
From: Matt Helsley <matthltc@us.ibm.com>
Date: Mon, 13 Sep 2010 13:01:20 -0700
Subject: perf events: Clean up pid passing

The kernel perf event creation path shouldn't use find_task_by_vpid()
because a vpid exists in a specific namespace. find_task_by_vpid() uses
current's pid namespace which isn't always the correct namespace to use
for the vpid in all the places perf_event_create_kernel_counter() (and
thus find_get_context()) is called.

The goal is to clean up pid namespace handling and prevent bugs like:

	https://bugzilla.kernel.org/show_bug.cgi?id=17281

Instead of using pids switch find_get_context() to use task struct
pointers directly. The syscall is responsible for resolving the pid to
a task struct. This moves the pid namespace resolution into the syscall
much like every other syscall that takes pid parameters.

Signed-off-by: Matt Helsley <matthltc@us.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Robin Green <greenrd@greenrd.org>
Cc: Prasad <prasad@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
LKML-Reference: <a134e5e392ab0204961fd1a62c84a222bf5874a9.1284407763.git.matthltc@us.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/arm/oprofile/common.c |  2 +-
 include/linux/perf_event.h |  2 +-
 kernel/hw_breakpoint.c     |  5 ++---
 kernel/perf_event.c        | 21 ++++++++++-----------
 kernel/watchdog.c          |  2 +-
 5 files changed, 15 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c
index 0691176899ff..aad63e611b36 100644
--- a/arch/arm/oprofile/common.c
+++ b/arch/arm/oprofile/common.c
@@ -96,7 +96,7 @@ static int op_create_counter(int cpu, int event)
 		return ret;
 
 	pevent = perf_event_create_kernel_counter(&counter_config[event].attr,
-						  cpu, -1,
+						  cpu, NULL,
 						  op_overflow_handler);
 
 	if (IS_ERR(pevent)) {
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 93bf53aa50e5..39d8860b2684 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -902,7 +902,7 @@ extern int perf_event_release_kernel(struct perf_event *event);
 extern struct perf_event *
 perf_event_create_kernel_counter(struct perf_event_attr *attr,
 				int cpu,
-				pid_t pid,
+				struct task_struct *task,
 				perf_overflow_handler_t callback);
 extern u64 perf_event_read_value(struct perf_event *event,
 				 u64 *enabled, u64 *running);
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index 6122f02cfedf..3b714e839c10 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -433,8 +433,7 @@ register_user_hw_breakpoint(struct perf_event_attr *attr,
 			    perf_overflow_handler_t triggered,
 			    struct task_struct *tsk)
 {
-	return perf_event_create_kernel_counter(attr, -1, task_pid_vnr(tsk),
-						triggered);
+	return perf_event_create_kernel_counter(attr, -1, tsk, triggered);
 }
 EXPORT_SYMBOL_GPL(register_user_hw_breakpoint);
 
@@ -516,7 +515,7 @@ register_wide_hw_breakpoint(struct perf_event_attr *attr,
 	get_online_cpus();
 	for_each_online_cpu(cpu) {
 		pevent = per_cpu_ptr(cpu_events, cpu);
-		bp = perf_event_create_kernel_counter(attr, cpu, -1, triggered);
+		bp = perf_event_create_kernel_counter(attr, cpu, NULL, triggered);
 
 		*pevent = bp;
 
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 3f5309db72f1..86f394e15d53 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -2053,15 +2053,14 @@ errout:
 }
 
 static struct perf_event_context *
-find_get_context(struct pmu *pmu, pid_t pid, int cpu)
+find_get_context(struct pmu *pmu, struct task_struct *task, int cpu)
 {
 	struct perf_event_context *ctx;
 	struct perf_cpu_context *cpuctx;
-	struct task_struct *task;
 	unsigned long flags;
 	int ctxn, err;
 
-	if (pid == -1 && cpu != -1) {
+	if (!task && cpu != -1) {
 		/* Must be root to operate on a CPU event: */
 		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
 			return ERR_PTR(-EACCES);
@@ -2084,10 +2083,6 @@ find_get_context(struct pmu *pmu, pid_t pid, int cpu)
 		return ctx;
 	}
 
-	task = find_lively_task_by_vpid(pid);
-	if (IS_ERR(task))
-		return (void*)task;
-
 	err = -EINVAL;
 	ctxn = pmu->task_ctx_nr;
 	if (ctxn < 0)
@@ -5527,6 +5522,7 @@ SYSCALL_DEFINE5(perf_event_open,
 	struct perf_event_context *ctx;
 	struct file *event_file = NULL;
 	struct file *group_file = NULL;
+	struct task_struct *task = NULL;
 	struct pmu *pmu;
 	int event_fd;
 	int fput_needed = 0;
@@ -5581,10 +5577,13 @@ SYSCALL_DEFINE5(perf_event_open,
 	if ((pmu->task_ctx_nr == perf_sw_context) && group_leader)
 		pmu = group_leader->pmu;
 
+	if (pid != -1)
+		task = find_lively_task_by_vpid(pid);
+
 	/*
 	 * Get the target context (task or percpu):
 	 */
-	ctx = find_get_context(pmu, pid, cpu);
+	ctx = find_get_context(pmu, task, cpu);
 	if (IS_ERR(ctx)) {
 		err = PTR_ERR(ctx);
 		goto err_group_fd;
@@ -5666,11 +5665,11 @@ err_fd:
  *
  * @attr: attributes of the counter to create
  * @cpu: cpu in which the counter is bound
- * @pid: task to profile
+ * @task: task to profile (NULL for percpu)
  */
 struct perf_event *
 perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
-				 pid_t pid,
+				 struct task_struct *task,
 				 perf_overflow_handler_t overflow_handler)
 {
 	struct perf_event_context *ctx;
@@ -5687,7 +5686,7 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
 		goto err;
 	}
 
-	ctx = find_get_context(event->pmu, pid, cpu);
+	ctx = find_get_context(event->pmu, task, cpu);
 	if (IS_ERR(ctx)) {
 		err = PTR_ERR(ctx);
 		goto err_free;
diff --git a/kernel/watchdog.c b/kernel/watchdog.c
index 89eadbb9cefe..dc8e16824b51 100644
--- a/kernel/watchdog.c
+++ b/kernel/watchdog.c
@@ -358,7 +358,7 @@ static int watchdog_nmi_enable(int cpu)
 	/* Try to register using hardware perf events */
 	wd_attr = &wd_hw_attr;
 	wd_attr->sample_period = hw_nmi_get_sample_period();
-	event = perf_event_create_kernel_counter(wd_attr, cpu, -1, watchdog_overflow_callback);
+	event = perf_event_create_kernel_counter(wd_attr, cpu, NULL, watchdog_overflow_callback);
 	if (!IS_ERR(event)) {
 		printk(KERN_INFO "NMI watchdog enabled, takes one hw-pmu counter.\n");
 		goto out_save;
-- 
cgit v1.2.3


From 144177991ca624841ddbd1e7edff958fc0f6d1fe Mon Sep 17 00:00:00 2001
From: Namhyung Kim <namhyung@gmail.com>
Date: Wed, 15 Sep 2010 13:08:27 +0200
Subject: block: fix an address space warning in blk-map.c

Change type of 2nd parameter of blk_rq_aligned() into unsigned long
and remove unnecessary casting. Now we can call it with 'uaddr'
instead of 'ubuf' in __blk_rq_map_user() so that it can remove
following warnings from sparse:

 block/blk-map.c:57:31: warning: incorrect type in argument 2 (different address spaces)
 block/blk-map.c:57:31:    expected void *addr
 block/blk-map.c:57:31:    got void [noderef] <asn:1>*ubuf

However blk_rq_map_kern() needs one more local variable to handle it.

Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-map.c        | 5 +++--
 include/linux/blkdev.h | 4 ++--
 2 files changed, 5 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-map.c b/block/blk-map.c
index c65d7593f7f1..ac0f7d46db63 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -54,7 +54,7 @@ static int __blk_rq_map_user(struct request_queue *q, struct request *rq,
 	 * direct dma. else, set up kernel bounce buffers
 	 */
 	uaddr = (unsigned long) ubuf;
-	if (blk_rq_aligned(q, ubuf, len) && !map_data)
+	if (blk_rq_aligned(q, uaddr, len) && !map_data)
 		bio = bio_map_user(q, NULL, uaddr, len, reading, gfp_mask);
 	else
 		bio = bio_copy_user(q, map_data, uaddr, len, reading, gfp_mask);
@@ -288,6 +288,7 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf,
 		    unsigned int len, gfp_t gfp_mask)
 {
 	int reading = rq_data_dir(rq) == READ;
+	unsigned long addr = (unsigned long) kbuf;
 	int do_copy = 0;
 	struct bio *bio;
 	int ret;
@@ -297,7 +298,7 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf,
 	if (!len || !kbuf)
 		return -EINVAL;
 
-	do_copy = !blk_rq_aligned(q, kbuf, len) || object_is_on_stack(kbuf);
+	do_copy = !blk_rq_aligned(q, addr, len) || object_is_on_stack(kbuf);
 	if (do_copy)
 		bio = bio_copy_kern(q, kbuf, len, gfp_mask, reading);
 	else
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 7e661106270a..780824edac16 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1097,11 +1097,11 @@ static inline int queue_dma_alignment(struct request_queue *q)
 	return q ? q->dma_alignment : 511;
 }
 
-static inline int blk_rq_aligned(struct request_queue *q, void *addr,
+static inline int blk_rq_aligned(struct request_queue *q, unsigned long addr,
 				 unsigned int len)
 {
 	unsigned int alignment = queue_dma_alignment(q) | q->dma_pad_mask;
-	return !((unsigned long)addr & alignment) && !(len & alignment);
+	return !(addr & alignment) && !(len & alignment);
 }
 
 /* assumes size > 256 */
-- 
cgit v1.2.3


From 6d1d8050b4bc89d0165d29b58e894aeba2564a97 Mon Sep 17 00:00:00 2001
From: Will Drewry <wad@chromium.org>
Date: Tue, 31 Aug 2010 15:47:05 -0500
Subject: block, partition: add partition_meta_info to hd_struct

I'm reposting this patch series as v4 since there have been no additional
comments, and I cleaned up one extra bit of unneeded code (in 3/3). The patches
are against Linus's tree: 2bfc96a127bc1cc94d26bfaa40159966064f9c8c
(2.6.36-rc3).

Would this patchset be suitable for inclusion in an mm branch?

This changes adds a partition_meta_info struct which itself contains a
union of structures that provide partition table specific metadata.

This change leaves the union empty. The subsequent patch includes an
implementation for CONFIG_EFI_PARTITION-based metadata.

Signed-off-by: Will Drewry <wad@chromium.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/genhd.c         |  1 +
 block/ioctl.c         |  2 +-
 fs/partitions/check.c | 23 +++++++++++++++++++---
 fs/partitions/check.h |  3 +++
 include/linux/genhd.h | 53 +++++++++++++++++++++++++++++++++++++++++++++++++--
 5 files changed, 76 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/block/genhd.c b/block/genhd.c
index 59a2db6fecef..c8da12055264 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1004,6 +1004,7 @@ static void disk_release(struct device *dev)
 	kfree(disk->random);
 	disk_replace_part_tbl(disk, NULL);
 	free_part_stats(&disk->part0);
+	free_part_info(&disk->part0);
 	kfree(disk);
 }
 struct class block_class = {
diff --git a/block/ioctl.c b/block/ioctl.c
index d8052f0dabd3..2c15fe0912c4 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -62,7 +62,7 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 
 			/* all seems OK */
 			part = add_partition(disk, partno, start, length,
-					     ADDPART_FLAG_NONE);
+					     ADDPART_FLAG_NONE, NULL);
 			mutex_unlock(&bdev->bd_mutex);
 			return IS_ERR(part) ? PTR_ERR(part) : 0;
 		case BLKPG_DEL_PARTITION:
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 79fbf3f390f0..6dfbee03ccc6 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -352,6 +352,7 @@ static void part_release(struct device *dev)
 {
 	struct hd_struct *p = dev_to_part(dev);
 	free_part_stats(p);
+	free_part_info(p);
 	kfree(p);
 }
 
@@ -401,7 +402,8 @@ static DEVICE_ATTR(whole_disk, S_IRUSR | S_IRGRP | S_IROTH,
 		   whole_disk_show, NULL);
 
 struct hd_struct *add_partition(struct gendisk *disk, int partno,
-				sector_t start, sector_t len, int flags)
+				sector_t start, sector_t len, int flags,
+				struct partition_meta_info *info)
 {
 	struct hd_struct *p;
 	dev_t devt = MKDEV(0, 0);
@@ -438,6 +440,14 @@ struct hd_struct *add_partition(struct gendisk *disk, int partno,
 	p->partno = partno;
 	p->policy = get_disk_ro(disk);
 
+	if (info) {
+		struct partition_meta_info *pinfo = alloc_part_info(disk);
+		if (!pinfo)
+			goto out_free_stats;
+		memcpy(pinfo, info, sizeof(*info));
+		p->info = pinfo;
+	}
+
 	dname = dev_name(ddev);
 	if (isdigit(dname[strlen(dname) - 1]))
 		dev_set_name(pdev, "%sp%d", dname, partno);
@@ -451,7 +461,7 @@ struct hd_struct *add_partition(struct gendisk *disk, int partno,
 
 	err = blk_alloc_devt(p, &devt);
 	if (err)
-		goto out_free_stats;
+		goto out_free_info;
 	pdev->devt = devt;
 
 	/* delay uevent until 'holders' subdir is created */
@@ -481,6 +491,8 @@ struct hd_struct *add_partition(struct gendisk *disk, int partno,
 
 	return p;
 
+out_free_info:
+	free_part_info(p);
 out_free_stats:
 	free_part_stats(p);
 out_free:
@@ -642,6 +654,7 @@ rescan:
 	/* add partitions */
 	for (p = 1; p < state->limit; p++) {
 		sector_t size, from;
+		struct partition_meta_info *info = NULL;
 
 		size = state->parts[p].size;
 		if (!size)
@@ -675,8 +688,12 @@ rescan:
 				size = get_capacity(disk) - from;
 			}
 		}
+
+		if (state->parts[p].has_info)
+			info = &state->parts[p].info;
 		part = add_partition(disk, p, from, size,
-				     state->parts[p].flags);
+				     state->parts[p].flags,
+				     &state->parts[p].info);
 		if (IS_ERR(part)) {
 			printk(KERN_ERR " %s: p%d could not be added: %ld\n",
 			       disk->disk_name, p, -PTR_ERR(part));
diff --git a/fs/partitions/check.h b/fs/partitions/check.h
index 8e4e103ba216..d68bf4dc3bc2 100644
--- a/fs/partitions/check.h
+++ b/fs/partitions/check.h
@@ -1,5 +1,6 @@
 #include <linux/pagemap.h>
 #include <linux/blkdev.h>
+#include <linux/genhd.h>
 
 /*
  * add_gd_partition adds a partitions details to the devices partition
@@ -12,6 +13,8 @@ struct parsed_partitions {
 		sector_t from;
 		sector_t size;
 		int flags;
+		bool has_info;
+		struct partition_meta_info info;
 	} parts[DISK_MAX_PARTS];
 	int next;
 	int limit;
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 5f2f4c4d8fb0..66e26b5a1537 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -12,6 +12,7 @@
 #include <linux/types.h>
 #include <linux/kdev_t.h>
 #include <linux/rcupdate.h>
+#include <linux/slab.h>
 
 #ifdef CONFIG_BLOCK
 
@@ -86,7 +87,15 @@ struct disk_stats {
 	unsigned long io_ticks;
 	unsigned long time_in_queue;
 };
-	
+
+#define PARTITION_META_INFO_VOLNAMELTH	64
+#define PARTITION_META_INFO_UUIDLTH	16
+
+struct partition_meta_info {
+	u8 uuid[PARTITION_META_INFO_UUIDLTH];	/* always big endian */
+	u8 volname[PARTITION_META_INFO_VOLNAMELTH];
+};
+
 struct hd_struct {
 	sector_t start_sect;
 	sector_t nr_sects;
@@ -95,6 +104,7 @@ struct hd_struct {
 	struct device __dev;
 	struct kobject *holder_dir;
 	int policy, partno;
+	struct partition_meta_info *info;
 #ifdef CONFIG_FAIL_MAKE_REQUEST
 	int make_it_fail;
 #endif
@@ -181,6 +191,30 @@ static inline struct gendisk *part_to_disk(struct hd_struct *part)
 	return NULL;
 }
 
+static inline void part_pack_uuid(const u8 *uuid_str, u8 *to)
+{
+	int i;
+	for (i = 0; i < 16; ++i) {
+		*to++ = (hex_to_bin(*uuid_str) << 4) |
+			(hex_to_bin(*(uuid_str + 1)));
+		uuid_str += 2;
+		switch (i) {
+		case 3:
+		case 5:
+		case 7:
+		case 9:
+			uuid_str++;
+			continue;
+		}
+	}
+}
+
+static inline char *part_unpack_uuid(const u8 *uuid, char *out)
+{
+	sprintf(out, "%pU", uuid);
+	return out;
+}
+
 static inline int disk_max_parts(struct gendisk *disk)
 {
 	if (disk->flags & GENHD_FL_EXT_DEVT)
@@ -342,6 +376,19 @@ static inline int part_in_flight(struct hd_struct *part)
 	return part->in_flight[0] + part->in_flight[1];
 }
 
+static inline struct partition_meta_info *alloc_part_info(struct gendisk *disk)
+{
+	if (disk)
+		return kzalloc_node(sizeof(struct partition_meta_info),
+				    GFP_KERNEL, disk->node_id);
+	return kzalloc(sizeof(struct partition_meta_info), GFP_KERNEL);
+}
+
+static inline void free_part_info(struct hd_struct *part)
+{
+	kfree(part->info);
+}
+
 /* block/blk-core.c */
 extern void part_round_stats(int cpu, struct hd_struct *part);
 
@@ -533,7 +580,9 @@ extern int disk_expand_part_tbl(struct gendisk *disk, int target);
 extern int rescan_partitions(struct gendisk *disk, struct block_device *bdev);
 extern struct hd_struct * __must_check add_partition(struct gendisk *disk,
 						     int partno, sector_t start,
-						     sector_t len, int flags);
+						     sector_t len, int flags,
+						     struct partition_meta_info
+						       *info);
 extern void delete_partition(struct gendisk *, int);
 extern void printk_all_partitions(void);
 
-- 
cgit v1.2.3


From 3661ca66a42e306aaf53246fb75aec1ea01be0f0 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 15 Sep 2010 13:05:29 -0700
Subject: memblock: Fix section mismatch warnings

Stephen found a bunch of section mismatch warnings with the
new memblock changes.

Use __init_memblock to replace __init in memblock.c and remove
__init in memblock.h. We should not use __init in header files.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Tested-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Yinghai Lu <Yinghai@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
LKML-Reference: <4C912709.2090201@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/memblock.h | 24 ++++++++++++------------
 mm/memblock.c            | 14 +++++++-------
 2 files changed, 19 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 7d285271130d..5096458c7535 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -51,39 +51,39 @@ u64 memblock_find_in_range(u64 start, u64 end, u64 size, u64 align);
 int memblock_free_reserved_regions(void);
 int memblock_reserve_reserved_regions(void);
 
-extern void __init memblock_init(void);
-extern void __init memblock_analyze(void);
+extern void memblock_init(void);
+extern void memblock_analyze(void);
 extern long memblock_add(phys_addr_t base, phys_addr_t size);
 extern long memblock_remove(phys_addr_t base, phys_addr_t size);
-extern long __init memblock_free(phys_addr_t base, phys_addr_t size);
-extern long __init memblock_reserve(phys_addr_t base, phys_addr_t size);
+extern long memblock_free(phys_addr_t base, phys_addr_t size);
+extern long memblock_reserve(phys_addr_t base, phys_addr_t size);
 
 /* The numa aware allocator is only available if
  * CONFIG_ARCH_POPULATES_NODE_MAP is set
  */
-extern phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align,
+extern phys_addr_t memblock_alloc_nid(phys_addr_t size, phys_addr_t align,
 					int nid);
-extern phys_addr_t __init memblock_alloc_try_nid(phys_addr_t size, phys_addr_t align,
+extern phys_addr_t memblock_alloc_try_nid(phys_addr_t size, phys_addr_t align,
 					    int nid);
 
-extern phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align);
+extern phys_addr_t memblock_alloc(phys_addr_t size, phys_addr_t align);
 
 /* Flags for memblock_alloc_base() amd __memblock_alloc_base() */
 #define MEMBLOCK_ALLOC_ANYWHERE	(~(phys_addr_t)0)
 #define MEMBLOCK_ALLOC_ACCESSIBLE	0
 
-extern phys_addr_t __init memblock_alloc_base(phys_addr_t size,
+extern phys_addr_t memblock_alloc_base(phys_addr_t size,
 					 phys_addr_t align,
 					 phys_addr_t max_addr);
-extern phys_addr_t __init __memblock_alloc_base(phys_addr_t size,
+extern phys_addr_t __memblock_alloc_base(phys_addr_t size,
 					   phys_addr_t align,
 					   phys_addr_t max_addr);
-extern phys_addr_t __init memblock_phys_mem_size(void);
+extern phys_addr_t memblock_phys_mem_size(void);
 extern phys_addr_t memblock_end_of_DRAM(void);
-extern void __init memblock_enforce_memory_limit(phys_addr_t memory_limit);
+extern void memblock_enforce_memory_limit(phys_addr_t memory_limit);
 extern int memblock_is_memory(phys_addr_t addr);
 extern int memblock_is_region_memory(phys_addr_t base, phys_addr_t size);
-extern int __init memblock_is_reserved(phys_addr_t addr);
+extern int memblock_is_reserved(phys_addr_t addr);
 extern int memblock_is_region_reserved(phys_addr_t base, phys_addr_t size);
 
 extern void memblock_dump_all(void);
diff --git a/mm/memblock.c b/mm/memblock.c
index 65e3ba8d09fb..d5d63ac1fd83 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -125,8 +125,8 @@ static phys_addr_t __init memblock_find_region(phys_addr_t start, phys_addr_t en
 	return MEMBLOCK_ERROR;
 }
 
-static phys_addr_t __init memblock_find_base(phys_addr_t size, phys_addr_t align,
-					phys_addr_t start, phys_addr_t end)
+static phys_addr_t __init_memblock memblock_find_base(phys_addr_t size,
+			phys_addr_t align, phys_addr_t start, phys_addr_t end)
 {
 	long i;
 
@@ -439,12 +439,12 @@ long __init_memblock memblock_remove(phys_addr_t base, phys_addr_t size)
 	return __memblock_remove(&memblock.memory, base, size);
 }
 
-long __init memblock_free(phys_addr_t base, phys_addr_t size)
+long __init_memblock memblock_free(phys_addr_t base, phys_addr_t size)
 {
 	return __memblock_remove(&memblock.reserved, base, size);
 }
 
-long __init memblock_reserve(phys_addr_t base, phys_addr_t size)
+long __init_memblock memblock_reserve(phys_addr_t base, phys_addr_t size)
 {
 	struct memblock_type *_rgn = &memblock.reserved;
 
@@ -671,12 +671,12 @@ int __init memblock_is_reserved(phys_addr_t addr)
 	return memblock_search(&memblock.reserved, addr) != -1;
 }
 
-int memblock_is_memory(phys_addr_t addr)
+int __init_memblock memblock_is_memory(phys_addr_t addr)
 {
 	return memblock_search(&memblock.memory, addr) != -1;
 }
 
-int memblock_is_region_memory(phys_addr_t base, phys_addr_t size)
+int __init_memblock memblock_is_region_memory(phys_addr_t base, phys_addr_t size)
 {
 	int idx = memblock_search(&memblock.reserved, base);
 
@@ -693,7 +693,7 @@ int __init_memblock memblock_is_region_reserved(phys_addr_t base, phys_addr_t si
 }
 
 
-void __init memblock_set_current_limit(phys_addr_t limit)
+void __init_memblock memblock_set_current_limit(phys_addr_t limit)
 {
 	memblock.current_limit = limit;
 }
-- 
cgit v1.2.3


From e035587305011432ee07f69f9738b3c7ef7f3684 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Tue, 14 Sep 2010 09:10:03 +0000
Subject: ethtool: Complete kernel-doc comments for RX flow filter and hash
 control

There are now several interfaces within the ethtool API for getting
and setting RX flow filtering and hashing behaviour, most of which are
poorly documented.  This adds kernel-doc comments for all these
interfaces, based on the existing incomplete comments and on the
initial implementations.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool.h | 129 ++++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 113 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 991269e5b152..4b3ba05b11a8 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -314,9 +314,20 @@ enum ethtool_flags {
 };
 
 /* The following structures are for supporting RX network flow
- * classification configuration. Note, all multibyte fields, e.g.,
- * ip4src, ip4dst, psrc, pdst, spi, etc. are expected to be in network
- * byte order.
+ * classification and RX n-tuple configuration. Note, all multibyte
+ * fields, e.g., ip4src, ip4dst, psrc, pdst, spi, etc. are expected to
+ * be in network byte order.
+ */
+
+/**
+ * struct ethtool_tcpip4_spec - flow specification for TCP/IPv4 etc.
+ * @ip4src: Source host
+ * @ip4dst: Destination host
+ * @psrc: Source port
+ * @pdst: Destination port
+ * @tos: Type-of-service
+ *
+ * This can be used to specify a TCP/IPv4, UDP/IPv4 or SCTP/IPv4 flow.
  */
 struct ethtool_tcpip4_spec {
 	__be32	ip4src;
@@ -326,6 +337,15 @@ struct ethtool_tcpip4_spec {
 	__u8    tos;
 };
 
+/**
+ * struct ethtool_ah_espip4_spec - flow specification for IPsec/IPv4
+ * @ip4src: Source host
+ * @ip4dst: Destination host
+ * @spi: Security parameters index
+ * @tos: Type-of-service
+ *
+ * This can be used to specify an IPsec transport or tunnel over IPv4.
+ */
 struct ethtool_ah_espip4_spec {
 	__be32	ip4src;
 	__be32	ip4dst;
@@ -348,6 +368,15 @@ struct ethtool_ether_spec {
 #define	ETH_RX_NFC_IP4	1
 #define	ETH_RX_NFC_IP6	2
 
+/**
+ * struct ethtool_usrip4_spec - general flow specification for IPv4
+ * @ip4src: Source host
+ * @ip4dst: Destination host
+ * @l4_4_bytes: First 4 bytes of transport (layer 4) header
+ * @tos: Type-of-service
+ * @ip_ver: Value must be %ETH_RX_NFC_IP4; mask must be 0
+ * @proto: Transport protocol number; mask must be 0
+ */
 struct ethtool_usrip4_spec {
 	__be32	ip4src;
 	__be32	ip4dst;
@@ -357,6 +386,15 @@ struct ethtool_usrip4_spec {
 	__u8    proto;
 };
 
+/**
+ * struct ethtool_rx_flow_spec - specification for RX flow filter
+ * @flow_type: Type of match to perform, e.g. %TCP_V4_FLOW
+ * @h_u: Flow fields to match (dependent on @flow_type)
+ * @m_u: Masks for flow field bits to be ignored
+ * @ring_cookie: RX ring/queue index to deliver to, or %RX_CLS_FLOW_DISC
+ *	if packets should be discarded
+ * @location: Index of filter in hardware table
+ */
 struct ethtool_rx_flow_spec {
 	__u32		flow_type;
 	union {
@@ -369,32 +407,87 @@ struct ethtool_rx_flow_spec {
 		struct ethtool_ether_spec		ether_spec;
 		struct ethtool_usrip4_spec		usr_ip4_spec;
 		__u8					hdata[64];
-	} h_u, m_u; /* entry, mask */
+	} h_u, m_u;
 	__u64		ring_cookie;
 	__u32		location;
 };
 
+/**
+ * struct ethtool_rxnfc - command to get or set RX flow classification rules
+ * @cmd: Specific command number - %ETHTOOL_GRXFH, %ETHTOOL_SRXFH,
+ *	%ETHTOOL_GRXRINGS, %ETHTOOL_GRXCLSRLCNT, %ETHTOOL_GRXCLSRULE,
+ *	%ETHTOOL_GRXCLSRLALL, %ETHTOOL_SRXCLSRLDEL or %ETHTOOL_SRXCLSRLINS
+ * @flow_type: Type of flow to be affected, e.g. %TCP_V4_FLOW
+ * @data: Command-dependent value
+ * @fs: Flow filter specification
+ * @rule_cnt: Number of rules to be affected
+ * @rule_locs: Array of valid rule indices
+ *
+ * For %ETHTOOL_GRXFH and %ETHTOOL_SRXFH, @data is a bitmask indicating
+ * the fields included in the flow hash, e.g. %RXH_IP_SRC.  The following
+ * structure fields must not be used.
+ *
+ * For %ETHTOOL_GRXRINGS, @data is set to the number of RX rings/queues
+ * on return.
+ *
+ * For %ETHTOOL_GRXCLSRLCNT, @rule_cnt is set to the number of defined
+ * rules on return.
+ *
+ * For %ETHTOOL_GRXCLSRULE, @fs.@location specifies the index of an
+ * existing filter rule on entry and @fs contains the rule on return.
+ *
+ * For %ETHTOOL_GRXCLSRLALL, @rule_cnt specifies the array size of the
+ * user buffer for @rule_locs on entry.  On return, @data is the size
+ * of the filter table and @rule_locs contains the indices of the
+ * defined rules.
+ *
+ * For %ETHTOOL_SRXCLSRLINS, @fs specifies the filter rule to add or
+ * update.  @fs.@location specifies the index to use and must not be
+ * ignored.
+ *
+ * For %ETHTOOL_SRXCLSRLDEL, @fs.@location specifies the index of an
+ * existing filter rule on entry.
+ *
+ * Implementation of indexed classification rules generally requires a
+ * TCAM.
+ */
 struct ethtool_rxnfc {
 	__u32				cmd;
 	__u32				flow_type;
-	/* The rx flow hash value or the rule DB size */
 	__u64				data;
-	/* The following fields are not valid and must not be used for
-	 * the ETHTOOL_{G,X}RXFH commands. */
 	struct ethtool_rx_flow_spec	fs;
 	__u32				rule_cnt;
 	__u32				rule_locs[0];
 };
 
+/**
+ * struct ethtool_rxfh_indir - command to get or set RX flow hash indirection
+ * @cmd: Specific command number - %ETHTOOL_GRXFHINDIR or %ETHTOOL_SRXFHINDIR
+ * @size: On entry, the array size of the user buffer.  On return from
+ *	%ETHTOOL_GRXFHINDIR, the array size of the hardware indirection table.
+ * @ring_index: RX ring/queue index for each hash value
+ */
 struct ethtool_rxfh_indir {
 	__u32	cmd;
-	/* On entry, this is the array size of the user buffer.  On
-	 * return from ETHTOOL_GRXFHINDIR, this is the array size of
-	 * the hardware indirection table. */
 	__u32	size;
-	__u32	ring_index[0];	/* ring/queue index for each hash value */
+	__u32	ring_index[0];
 };
 
+/**
+ * struct ethtool_rx_ntuple_flow_spec - specification for RX flow filter
+ * @flow_type: Type of match to perform, e.g. %TCP_V4_FLOW
+ * @h_u: Flow field values to match (dependent on @flow_type)
+ * @m_u: Masks for flow field value bits to be ignored
+ * @vlan_tag: VLAN tag to match
+ * @vlan_tag_mask: Mask for VLAN tag bits to be ignored
+ * @data: Driver-dependent data to match
+ * @data_mask: Mask for driver-dependent data bits to be ignored
+ * @action: RX ring/queue index to deliver to (non-negative) or other action
+ *	(negative, e.g. %ETHTOOL_RXNTUPLE_ACTION_DROP)
+ *
+ * Zero values in @h_u may be ignored, as if all the corresponding
+ * mask bits were set.
+ */
 struct ethtool_rx_ntuple_flow_spec {
 	__u32		 flow_type;
 	union {
@@ -407,18 +500,22 @@ struct ethtool_rx_ntuple_flow_spec {
 		struct ethtool_ether_spec		ether_spec;
 		struct ethtool_usrip4_spec		usr_ip4_spec;
 		__u8					hdata[64];
-	} h_u, m_u; /* entry, mask */
+	} h_u, m_u;
 
 	__u16	        vlan_tag;
 	__u16	        vlan_tag_mask;
-	__u64		data;      /* user-defined flow spec data */
-	__u64		data_mask; /* user-defined flow spec mask */
+	__u64		data;
+	__u64		data_mask;
 
-	/* signed to distinguish between queue and actions (DROP) */
 	__s32		action;
-#define ETHTOOL_RXNTUPLE_ACTION_DROP -1
+#define ETHTOOL_RXNTUPLE_ACTION_DROP -1		/* drop packet */
 };
 
+/**
+ * struct ethtool_rx_ntuple - command to set RX flow filter
+ * @cmd: Command number - %ETHTOOL_SRXNTUPLE
+ * @fs: Flow filter specification
+ */
 struct ethtool_rx_ntuple {
 	__u32					cmd;
 	struct ethtool_rx_ntuple_flow_spec	fs;
-- 
cgit v1.2.3


From e0de7c93b950b9e784894efc4b529c6958cb747a Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Tue, 14 Sep 2010 09:13:08 +0000
Subject: ethtool: Remove unimplemented flow specification types

struct ethtool_rawip4_spec and struct ethtool_ether_spec are neither
commented nor used by any driver, so remove them.  Adjust padding in
the user-visible unions that included these structures.

Fix references to struct ethtool_rawip4_spec in
ethtool_get_rx_ntuple(), which should use struct ethtool_usrip4_spec.

struct ethtool_usrip4_spec cannot hold IPv6 host addresses and there
is no separate structure that can, so remove ETH_RX_NFC_IP6 and the
reference to it in niu.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/niu.c       | 19 +++++--------------
 include/linux/ethtool.h | 21 ++-------------------
 net/core/ethtool.c      |  8 ++++----
 3 files changed, 11 insertions(+), 37 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/niu.c b/drivers/net/niu.c
index 8e1859c801a4..e36a83845a1c 100644
--- a/drivers/net/niu.c
+++ b/drivers/net/niu.c
@@ -7462,10 +7462,12 @@ static int niu_add_ethtool_tcam_entry(struct niu *np,
 	if (fsp->flow_type == IP_USER_FLOW) {
 		int i;
 		int add_usr_cls = 0;
-		int ipv6 = 0;
 		struct ethtool_usrip4_spec *uspec = &fsp->h_u.usr_ip4_spec;
 		struct ethtool_usrip4_spec *umask = &fsp->m_u.usr_ip4_spec;
 
+		if (uspec->ip_ver != ETH_RX_NFC_IP4)
+			return -EINVAL;
+
 		niu_lock_parent(np, flags);
 
 		for (i = 0; i < NIU_L3_PROG_CLS; i++) {
@@ -7494,9 +7496,7 @@ static int niu_add_ethtool_tcam_entry(struct niu *np,
 				default:
 					break;
 				}
-				if (uspec->ip_ver == ETH_RX_NFC_IP6)
-					ipv6 = 1;
-				ret = tcam_user_ip_class_set(np, class, ipv6,
+				ret = tcam_user_ip_class_set(np, class, 0,
 							     uspec->proto,
 							     uspec->tos,
 							     umask->tos);
@@ -7553,16 +7553,7 @@ static int niu_add_ethtool_tcam_entry(struct niu *np,
 		ret = -EINVAL;
 		goto out;
 	case IP_USER_FLOW:
-		if (fsp->h_u.usr_ip4_spec.ip_ver == ETH_RX_NFC_IP4) {
-			niu_get_tcamkey_from_ip4fs(fsp, tp, l2_rdc_table,
-						   class);
-		} else {
-			/* Not yet implemented */
-			netdev_info(np->dev, "niu%d: In %s(): usr flow for IPv6 not implemented\n",
-				    parent->index, __func__);
-			ret = -EINVAL;
-			goto out;
-		}
+		niu_get_tcamkey_from_ip4fs(fsp, tp, l2_rdc_table, class);
 		break;
 	default:
 		netdev_info(np->dev, "niu%d: In %s(): Unknown flow type %d\n",
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 4b3ba05b11a8..d64e246a39e7 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -353,20 +353,7 @@ struct ethtool_ah_espip4_spec {
 	__u8    tos;
 };
 
-struct ethtool_rawip4_spec {
-	__be32	ip4src;
-	__be32	ip4dst;
-	__u8	hdata[64];
-};
-
-struct ethtool_ether_spec {
-	__be16	ether_type;
-	__u8	frame_size;
-	__u8	eframe[16];
-};
-
 #define	ETH_RX_NFC_IP4	1
-#define	ETH_RX_NFC_IP6	2
 
 /**
  * struct ethtool_usrip4_spec - general flow specification for IPv4
@@ -403,10 +390,8 @@ struct ethtool_rx_flow_spec {
 		struct ethtool_tcpip4_spec		sctp_ip4_spec;
 		struct ethtool_ah_espip4_spec		ah_ip4_spec;
 		struct ethtool_ah_espip4_spec		esp_ip4_spec;
-		struct ethtool_rawip4_spec		raw_ip4_spec;
-		struct ethtool_ether_spec		ether_spec;
 		struct ethtool_usrip4_spec		usr_ip4_spec;
-		__u8					hdata[64];
+		__u8					hdata[72];
 	} h_u, m_u;
 	__u64		ring_cookie;
 	__u32		location;
@@ -496,10 +481,8 @@ struct ethtool_rx_ntuple_flow_spec {
 		struct ethtool_tcpip4_spec		sctp_ip4_spec;
 		struct ethtool_ah_espip4_spec		ah_ip4_spec;
 		struct ethtool_ah_espip4_spec		esp_ip4_spec;
-		struct ethtool_rawip4_spec		raw_ip4_spec;
-		struct ethtool_ether_spec		ether_spec;
 		struct ethtool_usrip4_spec		usr_ip4_spec;
-		__u8					hdata[64];
+		__u8					hdata[72];
 	} h_u, m_u;
 
 	__u16	        vlan_tag;
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 970eb9817bbc..fcd62757704d 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -673,19 +673,19 @@ static int ethtool_get_rx_ntuple(struct net_device *dev, void __user *useraddr)
 			break;
 		case IP_USER_FLOW:
 			sprintf(p, "\tSrc IP addr: 0x%x\n",
-				fsc->fs.h_u.raw_ip4_spec.ip4src);
+				fsc->fs.h_u.usr_ip4_spec.ip4src);
 			p += ETH_GSTRING_LEN;
 			num_strings++;
 			sprintf(p, "\tSrc IP mask: 0x%x\n",
-				fsc->fs.m_u.raw_ip4_spec.ip4src);
+				fsc->fs.m_u.usr_ip4_spec.ip4src);
 			p += ETH_GSTRING_LEN;
 			num_strings++;
 			sprintf(p, "\tDest IP addr: 0x%x\n",
-				fsc->fs.h_u.raw_ip4_spec.ip4dst);
+				fsc->fs.h_u.usr_ip4_spec.ip4dst);
 			p += ETH_GSTRING_LEN;
 			num_strings++;
 			sprintf(p, "\tDest IP mask: 0x%x\n",
-				fsc->fs.m_u.raw_ip4_spec.ip4dst);
+				fsc->fs.m_u.usr_ip4_spec.ip4dst);
 			p += ETH_GSTRING_LEN;
 			num_strings++;
 			break;
-- 
cgit v1.2.3


From 7dff59efbb0e8b0f81c95fd40379c0d0c757c808 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Wed, 15 Sep 2010 11:07:15 +0000
Subject: net: add rtnl_dereference()

We sometime want to dereference an rcu protected pointer while
holding RTNL. Use a macro to hide all lockdep details.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h | 9 +++++++++
 1 file changed, 9 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 263690d991a8..68c436bddc88 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -760,6 +760,15 @@ extern int lockdep_rtnl_is_held(void);
 	rcu_dereference_check(p, rcu_read_lock_held() ||	\
 				 lockdep_rtnl_is_held())
 
+/**
+ * rtnl_dereference - rcu_dereference with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ *
+ * Do an rcu_dereference(p), but check caller holds RTNL
+ */
+#define rtnl_dereference(p)					\
+	rcu_dereference_check(p, lockdep_rtnl_is_held())
+
 extern void rtnetlink_init(void);
 extern void __rtnl_unlock(void);
 
-- 
cgit v1.2.3


From 7417fa83c1a8b75a03bd9b9b358999f38e771eab Mon Sep 17 00:00:00 2001
From: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Date: Wed, 15 Sep 2010 12:30:12 +0000
Subject: Phonet: hook resource routing to userspace via ioctl()'s
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I wish we could use something cleaner, such as bind(). But that would
not work since resource subscription is orthogonal/in addition to the
normal object ID allocated via bind(). This is similar to multicasting
which also uses ioctl()'s.

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/phonet.h |  2 ++
 net/phonet/datagram.c  | 13 +++++++++++++
 net/phonet/socket.c    |  1 +
 3 files changed, 16 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/phonet.h b/include/linux/phonet.h
index 76edadf046d3..85e14a83283b 100644
--- a/include/linux/phonet.h
+++ b/include/linux/phonet.h
@@ -47,6 +47,8 @@
 
 /* ioctls */
 #define SIOCPNGETOBJECT		(SIOCPROTOPRIVATE + 0)
+#define SIOCPNADDRESOURCE	(SIOCPROTOPRIVATE + 14)
+#define SIOCPNDELRESOURCE	(SIOCPROTOPRIVATE + 15)
 
 /* Phonet protocol header */
 struct phonethdr {
diff --git a/net/phonet/datagram.c b/net/phonet/datagram.c
index 1bd38db4fe1e..2f032381bd45 100644
--- a/net/phonet/datagram.c
+++ b/net/phonet/datagram.c
@@ -52,6 +52,19 @@ static int pn_ioctl(struct sock *sk, int cmd, unsigned long arg)
 		answ = skb ? skb->len : 0;
 		release_sock(sk);
 		return put_user(answ, (int __user *)arg);
+
+	case SIOCPNADDRESOURCE:
+	case SIOCPNDELRESOURCE: {
+			u32 res;
+			if (get_user(res, (u32 __user *)arg))
+				return -EFAULT;
+			if (res >= 256)
+				return -EINVAL;
+			if (cmd == SIOCPNADDRESOURCE)
+				return pn_sock_bind_res(sk, res);
+			else
+				return pn_sock_unbind_res(sk, res);
+		}
 	}
 
 	return -ENOIOCTLCMD;
diff --git a/net/phonet/socket.c b/net/phonet/socket.c
index 4c29a23e9007..d4f41afc0583 100644
--- a/net/phonet/socket.c
+++ b/net/phonet/socket.c
@@ -158,6 +158,7 @@ void pn_sock_unhash(struct sock *sk)
 	spin_lock_bh(&pnsocks.lock);
 	sk_del_node_init(sk);
 	spin_unlock_bh(&pnsocks.lock);
+	pn_sock_unbind_all_res(sk);
 }
 EXPORT_SYMBOL(pn_sock_unhash);
 
-- 
cgit v1.2.3


From 95ae6b228f814fc0528d0506ee9f18ac333d6851 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Wed, 15 Sep 2010 04:04:31 +0000
Subject: ipv4: ip_ptr cleanups

dev->ip_ptr is protected by rtnl and rcu.

Yet some places dont use appropriate primitives and/or locking rules.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/plip.c           |  8 ++++++--
 drivers/net/via-velocity.h   | 11 +++++++----
 drivers/net/wan/hdlc_cisco.c |  4 +++-
 include/linux/inetdevice.h   | 14 +++++---------
 include/linux/netdevice.h    |  2 +-
 net/core/dev.c               |  2 +-
 net/ipv4/devinet.c           |  4 ++--
 net/ipv4/ipmr.c              |  2 +-
 net/mac80211/main.c          |  2 +-
 9 files changed, 27 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/plip.c b/drivers/net/plip.c
index 7e82a82422cf..ca4df7f4cf21 100644
--- a/drivers/net/plip.c
+++ b/drivers/net/plip.c
@@ -995,8 +995,10 @@ plip_tx_packet(struct sk_buff *skb, struct net_device *dev)
 static void
 plip_rewrite_address(const struct net_device *dev, struct ethhdr *eth)
 {
-	const struct in_device *in_dev = dev->ip_ptr;
+	const struct in_device *in_dev;
 
+	rcu_read_lock();
+	in_dev = __in_dev_get_rcu(dev);
 	if (in_dev) {
 		/* Any address will do - we take the first */
 		const struct in_ifaddr *ifa = in_dev->ifa_list;
@@ -1006,6 +1008,7 @@ plip_rewrite_address(const struct net_device *dev, struct ethhdr *eth)
 			memcpy(eth->h_dest+2, &ifa->ifa_address, 4);
 		}
 	}
+	rcu_read_unlock();
 }
 
 static int
@@ -1088,7 +1091,8 @@ plip_open(struct net_device *dev)
 	   when the device address isn't identical to the address of a
 	   received frame, the kernel incorrectly drops it).             */
 
-	if ((in_dev=dev->ip_ptr) != NULL) {
+	in_dev=__in_dev_get_rtnl(dev);
+	if (in_dev) {
 		/* Any address will do - we take the first. We already
 		   have the first two bytes filled with 0xfc, from
 		   plip_init_dev(). */
diff --git a/drivers/net/via-velocity.h b/drivers/net/via-velocity.h
index f7b33ae7a703..b5e120b0074b 100644
--- a/drivers/net/via-velocity.h
+++ b/drivers/net/via-velocity.h
@@ -1504,22 +1504,25 @@ struct velocity_info {
  *	addresses on this chain then we use the first - multi-IP WOL is not
  *	supported.
  *
- *	CHECK ME: locking
  */
 
 static inline int velocity_get_ip(struct velocity_info *vptr)
 {
-	struct in_device *in_dev = (struct in_device *) vptr->dev->ip_ptr;
+	struct in_device *in_dev;
 	struct in_ifaddr *ifa;
+	int res = -ENOENT;
 
+	rcu_read_lock();
+	in_dev = __in_dev_get_rcu(vptr->dev);
 	if (in_dev != NULL) {
 		ifa = (struct in_ifaddr *) in_dev->ifa_list;
 		if (ifa != NULL) {
 			memcpy(vptr->ip_addr, &ifa->ifa_address, 4);
-			return 0;
+			res = 0;
 		}
 	}
-	return -ENOENT;
+	rcu_read_unlock();
+	return res;
 }
 
 /**
diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c
index b38ffa149aba..b1e5e5b69c2a 100644
--- a/drivers/net/wan/hdlc_cisco.c
+++ b/drivers/net/wan/hdlc_cisco.c
@@ -191,7 +191,8 @@ static int cisco_rx(struct sk_buff *skb)
 
 		switch (ntohl (cisco_data->type)) {
 		case CISCO_ADDR_REQ: /* Stolen from syncppp.c :-) */
-			in_dev = dev->ip_ptr;
+			rcu_read_lock();
+			in_dev = __in_dev_get_rcu(dev);
 			addr = 0;
 			mask = ~cpu_to_be32(0); /* is the mask correct? */
 
@@ -211,6 +212,7 @@ static int cisco_rx(struct sk_buff *skb)
 				cisco_keepalive_send(dev, CISCO_ADDR_REPLY,
 						     addr, mask);
 			}
+			rcu_read_unlock();
 			dev_kfree_skb_any(skb);
 			return NET_RX_SUCCESS;
 
diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index 2be1a1a2beb9..1ec09bb4a3ab 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -9,6 +9,7 @@
 #include <linux/rcupdate.h>
 #include <linux/timer.h>
 #include <linux/sysctl.h>
+#include <linux/rtnetlink.h>
 
 enum
 {
@@ -198,14 +199,10 @@ static __inline__ int bad_mask(__be32 mask, __be32 addr)
 
 static inline struct in_device *__in_dev_get_rcu(const struct net_device *dev)
 {
-	struct in_device *in_dev = dev->ip_ptr;
-	if (in_dev)
-		in_dev = rcu_dereference(in_dev);
-	return in_dev;
+	return rcu_dereference(dev->ip_ptr);
 }
 
-static __inline__ struct in_device *
-in_dev_get(const struct net_device *dev)
+static inline struct in_device *in_dev_get(const struct net_device *dev)
 {
 	struct in_device *in_dev;
 
@@ -217,10 +214,9 @@ in_dev_get(const struct net_device *dev)
 	return in_dev;
 }
 
-static __inline__ struct in_device *
-__in_dev_get_rtnl(const struct net_device *dev)
+static inline struct in_device *__in_dev_get_rtnl(const struct net_device *dev)
 {
-	return (struct in_device*)dev->ip_ptr;
+	return rcu_dereference_check(dev->ip_ptr, lockdep_rtnl_is_held());
 }
 
 extern void in_dev_finish_destroy(struct in_device *idev);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index af05186d5b36..8992fffb8104 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -942,7 +942,7 @@ struct net_device {
 	void			*dsa_ptr;	/* dsa specific data */
 #endif
 	void 			*atalk_ptr;	/* AppleTalk link 	*/
-	void			*ip_ptr;	/* IPv4 specific data	*/
+	struct in_device __rcu	*ip_ptr;	/* IPv4 specific data	*/
 	void                    *dn_ptr;        /* DECnet specific data */
 	void                    *ip6_ptr;       /* IPv6 specific data */
 	void			*ec_ptr;	/* Econet specific data	*/
diff --git a/net/core/dev.c b/net/core/dev.c
index fc2dc933bee5..5bdce97b8175 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5286,7 +5286,7 @@ void netdev_run_todo(void)
 
 		/* paranoia */
 		BUG_ON(atomic_read(&dev->refcnt));
-		WARN_ON(dev->ip_ptr);
+		WARN_ON(rcu_dereference_raw(dev->ip_ptr));
 		WARN_ON(dev->ip6_ptr);
 		WARN_ON(dev->dn_ptr);
 
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index da14c49284f4..c2ff48fa18c7 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -209,7 +209,7 @@ static void inetdev_destroy(struct in_device *in_dev)
 		inet_free_ifa(ifa);
 	}
 
-	dev->ip_ptr = NULL;
+	rcu_assign_pointer(dev->ip_ptr, NULL);
 
 	devinet_sysctl_unregister(in_dev);
 	neigh_parms_release(&arp_tbl, in_dev->arp_parms);
@@ -1059,7 +1059,7 @@ static int inetdev_event(struct notifier_block *this, unsigned long event,
 	switch (event) {
 	case NETDEV_REGISTER:
 		printk(KERN_DEBUG "inetdev_event: bug\n");
-		dev->ip_ptr = NULL;
+		rcu_assign_pointer(dev->ip_ptr, NULL);
 		break;
 	case NETDEV_UP:
 		if (!inetdev_valid_mtu(dev->mtu))
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 179fcab866fc..10b24c02deb0 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -724,7 +724,7 @@ static int vif_add(struct net *net, struct mr_table *mrt,
 	case 0:
 		if (vifc->vifc_flags == VIFF_USE_IFINDEX) {
 			dev = dev_get_by_index(net, vifc->vifc_lcl_ifindex);
-			if (dev && dev->ip_ptr == NULL) {
+			if (dev && __in_dev_get_rtnl(dev) == NULL) {
 				dev_put(dev);
 				return -EADDRNOTAVAIL;
 			}
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 4935b843bcca..b8cf2821f00d 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -362,7 +362,7 @@ static int ieee80211_ifa_changed(struct notifier_block *nb,
 	if (sdata->vif.type != NL80211_IFTYPE_STATION)
 		return NOTIFY_DONE;
 
-	idev = sdata->dev->ip_ptr;
+	idev = __in_dev_get_rtnl(sdata->dev);
 	if (!idev)
 		return NOTIFY_DONE;
 
-- 
cgit v1.2.3


From e43473b7f223ec866f7db273697e76c337c390f9 Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@redhat.com>
Date: Wed, 15 Sep 2010 17:06:35 -0400
Subject: blkio: Core implementation of throttle policy

o Actual implementation of throttling policy in block layer. Currently it
  implements READ and WRITE bytes per second throttling logic. IOPS throttling
  comes in later patches.

Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/Kconfig             |  12 +
 block/Makefile            |   1 +
 block/blk-core.c          |  24 ++
 block/blk-throttle.c      | 909 ++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/blk_types.h |   3 +
 include/linux/blkdev.h    |  24 ++
 init/Kconfig              |   9 +-
 7 files changed, 979 insertions(+), 3 deletions(-)
 create mode 100644 block/blk-throttle.c

(limited to 'include/linux')

diff --git a/block/Kconfig b/block/Kconfig
index 9be0b56eaee1..6c9213ef15a1 100644
--- a/block/Kconfig
+++ b/block/Kconfig
@@ -77,6 +77,18 @@ config BLK_DEV_INTEGRITY
 	T10/SCSI Data Integrity Field or the T13/ATA External Path
 	Protection.  If in doubt, say N.
 
+config BLK_DEV_THROTTLING
+	bool "Block layer bio throttling support"
+	depends on BLK_CGROUP=y && EXPERIMENTAL
+	default n
+	---help---
+	Block layer bio throttling support. It can be used to limit
+	the IO rate to a device. IO rate policies are per cgroup and
+	one needs to mount and use blkio cgroup controller for creating
+	cgroups and specifying per device IO rate policies.
+
+	See Documentation/cgroups/blkio-controller.txt for more information.
+
 endif # BLOCK
 
 config BLOCK_COMPAT
diff --git a/block/Makefile b/block/Makefile
index 0bb499a739cd..c850d5ef80a2 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_BLOCK) := elevator.o blk-core.o blk-tag.o blk-sysfs.o \
 
 obj-$(CONFIG_BLK_DEV_BSG)	+= bsg.o
 obj-$(CONFIG_BLK_CGROUP)	+= blk-cgroup.o
+obj-$(CONFIG_BLK_DEV_THROTTLING)	+= blk-throttle.o
 obj-$(CONFIG_IOSCHED_NOOP)	+= noop-iosched.o
 obj-$(CONFIG_IOSCHED_DEADLINE)	+= deadline-iosched.o
 obj-$(CONFIG_IOSCHED_CFQ)	+= cfq-iosched.o
diff --git a/block/blk-core.c b/block/blk-core.c
index 8d07c1b7e701..797d5095eb83 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -382,6 +382,7 @@ void blk_sync_queue(struct request_queue *q)
 	del_timer_sync(&q->unplug_timer);
 	del_timer_sync(&q->timeout);
 	cancel_work_sync(&q->unplug_work);
+	throtl_shutdown_timer_wq(q);
 }
 EXPORT_SYMBOL(blk_sync_queue);
 
@@ -459,6 +460,8 @@ void blk_cleanup_queue(struct request_queue *q)
 	if (q->elevator)
 		elevator_exit(q->elevator);
 
+	blk_throtl_exit(q);
+
 	blk_put_queue(q);
 }
 EXPORT_SYMBOL(blk_cleanup_queue);
@@ -515,6 +518,11 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 		return NULL;
 	}
 
+	if (blk_throtl_init(q)) {
+		kmem_cache_free(blk_requestq_cachep, q);
+		return NULL;
+	}
+
 	setup_timer(&q->backing_dev_info.laptop_mode_wb_timer,
 		    laptop_mode_timer_fn, (unsigned long) q);
 	init_timer(&q->unplug_timer);
@@ -1522,6 +1530,15 @@ static inline void __generic_make_request(struct bio *bio)
 			goto end_io;
 		}
 
+		blk_throtl_bio(q, &bio);
+
+		/*
+		 * If bio = NULL, bio has been throttled and will be submitted
+		 * later.
+		 */
+		if (!bio)
+			break;
+
 		trace_block_bio_queue(q, bio);
 
 		ret = q->make_request_fn(q, bio);
@@ -2580,6 +2597,13 @@ int kblockd_schedule_work(struct request_queue *q, struct work_struct *work)
 }
 EXPORT_SYMBOL(kblockd_schedule_work);
 
+int kblockd_schedule_delayed_work(struct request_queue *q,
+			struct delayed_work *dwork, unsigned long delay)
+{
+	return queue_delayed_work(kblockd_workqueue, dwork, delay);
+}
+EXPORT_SYMBOL(kblockd_schedule_delayed_work);
+
 int __init blk_dev_init(void)
 {
 	BUILD_BUG_ON(__REQ_NR_BITS > 8 *
diff --git a/block/blk-throttle.c b/block/blk-throttle.c
new file mode 100644
index 000000000000..4b492011e0de
--- /dev/null
+++ b/block/blk-throttle.c
@@ -0,0 +1,909 @@
+/*
+ * Interface for controlling IO bandwidth on a request queue
+ *
+ * Copyright (C) 2010 Vivek Goyal <vgoyal@redhat.com>
+ */
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/blkdev.h>
+#include <linux/bio.h>
+#include <linux/blktrace_api.h>
+#include "blk-cgroup.h"
+
+/* Max dispatch from a group in 1 round */
+static int throtl_grp_quantum = 8;
+
+/* Total max dispatch from all groups in one round */
+static int throtl_quantum = 32;
+
+/* Throttling is performed over 100ms slice and after that slice is renewed */
+static unsigned long throtl_slice = HZ/10;	/* 100 ms */
+
+struct throtl_rb_root {
+	struct rb_root rb;
+	struct rb_node *left;
+	unsigned int count;
+	unsigned long min_disptime;
+};
+
+#define THROTL_RB_ROOT	(struct throtl_rb_root) { .rb = RB_ROOT, .left = NULL, \
+			.count = 0, .min_disptime = 0}
+
+#define rb_entry_tg(node)	rb_entry((node), struct throtl_grp, rb_node)
+
+struct throtl_grp {
+	/* List of throtl groups on the request queue*/
+	struct hlist_node tg_node;
+
+	/* active throtl group service_tree member */
+	struct rb_node rb_node;
+
+	/*
+	 * Dispatch time in jiffies. This is the estimated time when group
+	 * will unthrottle and is ready to dispatch more bio. It is used as
+	 * key to sort active groups in service tree.
+	 */
+	unsigned long disptime;
+
+	struct blkio_group blkg;
+	atomic_t ref;
+	unsigned int flags;
+
+	/* Two lists for READ and WRITE */
+	struct bio_list bio_lists[2];
+
+	/* Number of queued bios on READ and WRITE lists */
+	unsigned int nr_queued[2];
+
+	/* bytes per second rate limits */
+	uint64_t bps[2];
+
+	/* Number of bytes disptached in current slice */
+	uint64_t bytes_disp[2];
+
+	/* When did we start a new slice */
+	unsigned long slice_start[2];
+	unsigned long slice_end[2];
+};
+
+struct throtl_data
+{
+	/* List of throtl groups */
+	struct hlist_head tg_list;
+
+	/* service tree for active throtl groups */
+	struct throtl_rb_root tg_service_tree;
+
+	struct throtl_grp root_tg;
+	struct request_queue *queue;
+
+	/* Total Number of queued bios on READ and WRITE lists */
+	unsigned int nr_queued[2];
+
+	/*
+	 * number of total undestroyed groups (excluding root group)
+	 */
+	unsigned int nr_undestroyed_grps;
+
+	/* Work for dispatching throttled bios */
+	struct delayed_work throtl_work;
+};
+
+enum tg_state_flags {
+	THROTL_TG_FLAG_on_rr = 0,	/* on round-robin busy list */
+};
+
+#define THROTL_TG_FNS(name)						\
+static inline void throtl_mark_tg_##name(struct throtl_grp *tg)		\
+{									\
+	(tg)->flags |= (1 << THROTL_TG_FLAG_##name);			\
+}									\
+static inline void throtl_clear_tg_##name(struct throtl_grp *tg)	\
+{									\
+	(tg)->flags &= ~(1 << THROTL_TG_FLAG_##name);			\
+}									\
+static inline int throtl_tg_##name(const struct throtl_grp *tg)		\
+{									\
+	return ((tg)->flags & (1 << THROTL_TG_FLAG_##name)) != 0;	\
+}
+
+THROTL_TG_FNS(on_rr);
+
+#define throtl_log_tg(td, tg, fmt, args...)				\
+	blk_add_trace_msg((td)->queue, "throtl %s " fmt,		\
+				blkg_path(&(tg)->blkg), ##args);      	\
+
+#define throtl_log(td, fmt, args...)	\
+	blk_add_trace_msg((td)->queue, "throtl " fmt, ##args)
+
+static inline struct throtl_grp *tg_of_blkg(struct blkio_group *blkg)
+{
+	if (blkg)
+		return container_of(blkg, struct throtl_grp, blkg);
+
+	return NULL;
+}
+
+static inline int total_nr_queued(struct throtl_data *td)
+{
+	return (td->nr_queued[0] + td->nr_queued[1]);
+}
+
+static inline struct throtl_grp *throtl_ref_get_tg(struct throtl_grp *tg)
+{
+	atomic_inc(&tg->ref);
+	return tg;
+}
+
+static void throtl_put_tg(struct throtl_grp *tg)
+{
+	BUG_ON(atomic_read(&tg->ref) <= 0);
+	if (!atomic_dec_and_test(&tg->ref))
+		return;
+	kfree(tg);
+}
+
+static struct throtl_grp * throtl_find_alloc_tg(struct throtl_data *td,
+			struct cgroup *cgroup)
+{
+	struct blkio_cgroup *blkcg = cgroup_to_blkio_cgroup(cgroup);
+	struct throtl_grp *tg = NULL;
+	void *key = td;
+	struct backing_dev_info *bdi = &td->queue->backing_dev_info;
+	unsigned int major, minor;
+
+	/*
+	 * TODO: Speed up blkiocg_lookup_group() by maintaining a radix
+	 * tree of blkg (instead of traversing through hash list all
+	 * the time.
+	 */
+	tg = tg_of_blkg(blkiocg_lookup_group(blkcg, key));
+
+	/* Fill in device details for root group */
+	if (tg && !tg->blkg.dev && bdi->dev && dev_name(bdi->dev)) {
+		sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor);
+		tg->blkg.dev = MKDEV(major, minor);
+		goto done;
+	}
+
+	if (tg)
+		goto done;
+
+	tg = kzalloc_node(sizeof(*tg), GFP_ATOMIC, td->queue->node);
+	if (!tg)
+		goto done;
+
+	INIT_HLIST_NODE(&tg->tg_node);
+	RB_CLEAR_NODE(&tg->rb_node);
+	bio_list_init(&tg->bio_lists[0]);
+	bio_list_init(&tg->bio_lists[1]);
+
+	/*
+	 * Take the initial reference that will be released on destroy
+	 * This can be thought of a joint reference by cgroup and
+	 * request queue which will be dropped by either request queue
+	 * exit or cgroup deletion path depending on who is exiting first.
+	 */
+	atomic_set(&tg->ref, 1);
+
+	/* Add group onto cgroup list */
+	sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor);
+	blkiocg_add_blkio_group(blkcg, &tg->blkg, (void *)td,
+				MKDEV(major, minor), BLKIO_POLICY_THROTL);
+
+	tg->bps[READ] = blkcg_get_read_bps(blkcg, tg->blkg.dev);
+	tg->bps[WRITE] = blkcg_get_write_bps(blkcg, tg->blkg.dev);
+
+	hlist_add_head(&tg->tg_node, &td->tg_list);
+	td->nr_undestroyed_grps++;
+done:
+	return tg;
+}
+
+static struct throtl_grp * throtl_get_tg(struct throtl_data *td)
+{
+	struct cgroup *cgroup;
+	struct throtl_grp *tg = NULL;
+
+	rcu_read_lock();
+	cgroup = task_cgroup(current, blkio_subsys_id);
+	tg = throtl_find_alloc_tg(td, cgroup);
+	if (!tg)
+		tg = &td->root_tg;
+	rcu_read_unlock();
+	return tg;
+}
+
+static struct throtl_grp *throtl_rb_first(struct throtl_rb_root *root)
+{
+	/* Service tree is empty */
+	if (!root->count)
+		return NULL;
+
+	if (!root->left)
+		root->left = rb_first(&root->rb);
+
+	if (root->left)
+		return rb_entry_tg(root->left);
+
+	return NULL;
+}
+
+static void rb_erase_init(struct rb_node *n, struct rb_root *root)
+{
+	rb_erase(n, root);
+	RB_CLEAR_NODE(n);
+}
+
+static void throtl_rb_erase(struct rb_node *n, struct throtl_rb_root *root)
+{
+	if (root->left == n)
+		root->left = NULL;
+	rb_erase_init(n, &root->rb);
+	--root->count;
+}
+
+static void update_min_dispatch_time(struct throtl_rb_root *st)
+{
+	struct throtl_grp *tg;
+
+	tg = throtl_rb_first(st);
+	if (!tg)
+		return;
+
+	st->min_disptime = tg->disptime;
+}
+
+static void
+tg_service_tree_add(struct throtl_rb_root *st, struct throtl_grp *tg)
+{
+	struct rb_node **node = &st->rb.rb_node;
+	struct rb_node *parent = NULL;
+	struct throtl_grp *__tg;
+	unsigned long key = tg->disptime;
+	int left = 1;
+
+	while (*node != NULL) {
+		parent = *node;
+		__tg = rb_entry_tg(parent);
+
+		if (time_before(key, __tg->disptime))
+			node = &parent->rb_left;
+		else {
+			node = &parent->rb_right;
+			left = 0;
+		}
+	}
+
+	if (left)
+		st->left = &tg->rb_node;
+
+	rb_link_node(&tg->rb_node, parent, node);
+	rb_insert_color(&tg->rb_node, &st->rb);
+}
+
+static void __throtl_enqueue_tg(struct throtl_data *td, struct throtl_grp *tg)
+{
+	struct throtl_rb_root *st = &td->tg_service_tree;
+
+	tg_service_tree_add(st, tg);
+	throtl_mark_tg_on_rr(tg);
+	st->count++;
+}
+
+static void throtl_enqueue_tg(struct throtl_data *td, struct throtl_grp *tg)
+{
+	if (!throtl_tg_on_rr(tg))
+		__throtl_enqueue_tg(td, tg);
+}
+
+static void __throtl_dequeue_tg(struct throtl_data *td, struct throtl_grp *tg)
+{
+	throtl_rb_erase(&tg->rb_node, &td->tg_service_tree);
+	throtl_clear_tg_on_rr(tg);
+}
+
+static void throtl_dequeue_tg(struct throtl_data *td, struct throtl_grp *tg)
+{
+	if (throtl_tg_on_rr(tg))
+		__throtl_dequeue_tg(td, tg);
+}
+
+static void throtl_schedule_next_dispatch(struct throtl_data *td)
+{
+	struct throtl_rb_root *st = &td->tg_service_tree;
+
+	/*
+	 * If there are more bios pending, schedule more work.
+	 */
+	if (!total_nr_queued(td))
+		return;
+
+	BUG_ON(!st->count);
+
+	update_min_dispatch_time(st);
+
+	if (time_before_eq(st->min_disptime, jiffies))
+		throtl_schedule_delayed_work(td->queue, 0);
+	else
+		throtl_schedule_delayed_work(td->queue,
+				(st->min_disptime - jiffies));
+}
+
+static inline void
+throtl_start_new_slice(struct throtl_data *td, struct throtl_grp *tg, bool rw)
+{
+	tg->bytes_disp[rw] = 0;
+	tg->slice_start[rw] = jiffies;
+	tg->slice_end[rw] = jiffies + throtl_slice;
+	throtl_log_tg(td, tg, "[%c] new slice start=%lu end=%lu jiffies=%lu",
+			rw == READ ? 'R' : 'W', tg->slice_start[rw],
+			tg->slice_end[rw], jiffies);
+}
+
+static inline void throtl_extend_slice(struct throtl_data *td,
+		struct throtl_grp *tg, bool rw, unsigned long jiffy_end)
+{
+	tg->slice_end[rw] = roundup(jiffy_end, throtl_slice);
+	throtl_log_tg(td, tg, "[%c] extend slice start=%lu end=%lu jiffies=%lu",
+			rw == READ ? 'R' : 'W', tg->slice_start[rw],
+			tg->slice_end[rw], jiffies);
+}
+
+/* Determine if previously allocated or extended slice is complete or not */
+static bool
+throtl_slice_used(struct throtl_data *td, struct throtl_grp *tg, bool rw)
+{
+	if (time_in_range(jiffies, tg->slice_start[rw], tg->slice_end[rw]))
+		return 0;
+
+	return 1;
+}
+
+/* Trim the used slices and adjust slice start accordingly */
+static inline void
+throtl_trim_slice(struct throtl_data *td, struct throtl_grp *tg, bool rw)
+{
+	unsigned long nr_slices, bytes_trim, time_elapsed;
+
+	BUG_ON(time_before(tg->slice_end[rw], tg->slice_start[rw]));
+
+	/*
+	 * If bps are unlimited (-1), then time slice don't get
+	 * renewed. Don't try to trim the slice if slice is used. A new
+	 * slice will start when appropriate.
+	 */
+	if (throtl_slice_used(td, tg, rw))
+		return;
+
+	time_elapsed = jiffies - tg->slice_start[rw];
+
+	nr_slices = time_elapsed / throtl_slice;
+
+	if (!nr_slices)
+		return;
+
+	bytes_trim = (tg->bps[rw] * throtl_slice * nr_slices)/HZ;
+
+	if (!bytes_trim)
+		return;
+
+	if (tg->bytes_disp[rw] >= bytes_trim)
+		tg->bytes_disp[rw] -= bytes_trim;
+	else
+		tg->bytes_disp[rw] = 0;
+
+	tg->slice_start[rw] += nr_slices * throtl_slice;
+
+	throtl_log_tg(td, tg, "[%c] trim slice nr=%lu bytes=%lu"
+			" start=%lu end=%lu jiffies=%lu",
+			rw == READ ? 'R' : 'W', nr_slices, bytes_trim,
+			tg->slice_start[rw], tg->slice_end[rw], jiffies);
+}
+
+/*
+ * Returns whether one can dispatch a bio or not. Also returns approx number
+ * of jiffies to wait before this bio is with-in IO rate and can be dispatched
+ */
+static bool tg_may_dispatch(struct throtl_data *td, struct throtl_grp *tg,
+				struct bio *bio, unsigned long *wait)
+{
+	bool rw = bio_data_dir(bio);
+	u64 bytes_allowed, extra_bytes;
+	unsigned long jiffy_elapsed, jiffy_wait, jiffy_elapsed_rnd;
+
+	/*
+	 * Currently whole state machine of group depends on first bio
+	 * queued in the group bio list. So one should not be calling
+	 * this function with a different bio if there are other bios
+	 * queued.
+	 */
+	BUG_ON(tg->nr_queued[rw] && bio != bio_list_peek(&tg->bio_lists[rw]));
+
+	/* If tg->bps = -1, then BW is unlimited */
+	if (tg->bps[rw] == -1) {
+		if (wait)
+			*wait = 0;
+		return 1;
+	}
+
+	/*
+	 * If previous slice expired, start a new one otherwise renew/extend
+	 * existing slice to make sure it is at least throtl_slice interval
+	 * long since now.
+	 */
+	if (throtl_slice_used(td, tg, rw))
+		throtl_start_new_slice(td, tg, rw);
+	else {
+		if (time_before(tg->slice_end[rw], jiffies + throtl_slice))
+			throtl_extend_slice(td, tg, rw, jiffies + throtl_slice);
+	}
+
+	jiffy_elapsed = jiffy_elapsed_rnd = jiffies - tg->slice_start[rw];
+
+	/* Slice has just started. Consider one slice interval */
+	if (!jiffy_elapsed)
+		jiffy_elapsed_rnd = throtl_slice;
+
+	jiffy_elapsed_rnd = roundup(jiffy_elapsed_rnd, throtl_slice);
+
+	bytes_allowed = (tg->bps[rw] * jiffies_to_msecs(jiffy_elapsed_rnd))
+				/ MSEC_PER_SEC;
+
+	if (tg->bytes_disp[rw] + bio->bi_size <= bytes_allowed) {
+		if (wait)
+			*wait = 0;
+		return 1;
+	}
+
+	/* Calc approx time to dispatch */
+	extra_bytes = tg->bytes_disp[rw] + bio->bi_size - bytes_allowed;
+	jiffy_wait = div64_u64(extra_bytes * HZ, tg->bps[rw]);
+
+	if (!jiffy_wait)
+		jiffy_wait = 1;
+
+	/*
+	 * This wait time is without taking into consideration the rounding
+	 * up we did. Add that time also.
+	 */
+	jiffy_wait = jiffy_wait + (jiffy_elapsed_rnd - jiffy_elapsed);
+
+	if (wait)
+		*wait = jiffy_wait;
+
+	if (time_before(tg->slice_end[rw], jiffies + jiffy_wait))
+		throtl_extend_slice(td, tg, rw, jiffies + jiffy_wait);
+
+	return 0;
+}
+
+static void throtl_charge_bio(struct throtl_grp *tg, struct bio *bio)
+{
+	bool rw = bio_data_dir(bio);
+	bool sync = bio->bi_rw & REQ_SYNC;
+
+	/* Charge the bio to the group */
+	tg->bytes_disp[rw] += bio->bi_size;
+
+	/*
+	 * TODO: This will take blkg->stats_lock. Figure out a way
+	 * to avoid this cost.
+	 */
+	blkiocg_update_dispatch_stats(&tg->blkg, bio->bi_size, rw, sync);
+
+}
+
+static void throtl_add_bio_tg(struct throtl_data *td, struct throtl_grp *tg,
+			struct bio *bio)
+{
+	bool rw = bio_data_dir(bio);
+
+	bio_list_add(&tg->bio_lists[rw], bio);
+	/* Take a bio reference on tg */
+	throtl_ref_get_tg(tg);
+	tg->nr_queued[rw]++;
+	td->nr_queued[rw]++;
+	throtl_enqueue_tg(td, tg);
+}
+
+static void tg_update_disptime(struct throtl_data *td, struct throtl_grp *tg)
+{
+	unsigned long read_wait = -1, write_wait = -1, min_wait = -1, disptime;
+	struct bio *bio;
+
+	if ((bio = bio_list_peek(&tg->bio_lists[READ])))
+		tg_may_dispatch(td, tg, bio, &read_wait);
+
+	if ((bio = bio_list_peek(&tg->bio_lists[WRITE])))
+		tg_may_dispatch(td, tg, bio, &write_wait);
+
+	min_wait = min(read_wait, write_wait);
+	disptime = jiffies + min_wait;
+
+	/*
+	 * If group is already on active tree, then update dispatch time
+	 * only if it is lesser than existing dispatch time. Otherwise
+	 * always update the dispatch time
+	 */
+
+	if (throtl_tg_on_rr(tg) && time_before(disptime, tg->disptime))
+		return;
+
+	/* Update dispatch time */
+	throtl_dequeue_tg(td, tg);
+	tg->disptime = disptime;
+	throtl_enqueue_tg(td, tg);
+}
+
+static void tg_dispatch_one_bio(struct throtl_data *td, struct throtl_grp *tg,
+				bool rw, struct bio_list *bl)
+{
+	struct bio *bio;
+
+	bio = bio_list_pop(&tg->bio_lists[rw]);
+	tg->nr_queued[rw]--;
+	/* Drop bio reference on tg */
+	throtl_put_tg(tg);
+
+	BUG_ON(td->nr_queued[rw] <= 0);
+	td->nr_queued[rw]--;
+
+	throtl_charge_bio(tg, bio);
+	bio_list_add(bl, bio);
+	bio->bi_rw |= REQ_THROTTLED;
+
+	throtl_trim_slice(td, tg, rw);
+}
+
+static int throtl_dispatch_tg(struct throtl_data *td, struct throtl_grp *tg,
+				struct bio_list *bl)
+{
+	unsigned int nr_reads = 0, nr_writes = 0;
+	unsigned int max_nr_reads = throtl_grp_quantum*3/4;
+	unsigned int max_nr_writes = throtl_grp_quantum - nr_reads;
+	struct bio *bio;
+
+	/* Try to dispatch 75% READS and 25% WRITES */
+
+	while ((bio = bio_list_peek(&tg->bio_lists[READ]))
+		&& tg_may_dispatch(td, tg, bio, NULL)) {
+
+		tg_dispatch_one_bio(td, tg, bio_data_dir(bio), bl);
+		nr_reads++;
+
+		if (nr_reads >= max_nr_reads)
+			break;
+	}
+
+	while ((bio = bio_list_peek(&tg->bio_lists[WRITE]))
+		&& tg_may_dispatch(td, tg, bio, NULL)) {
+
+		tg_dispatch_one_bio(td, tg, bio_data_dir(bio), bl);
+		nr_writes++;
+
+		if (nr_writes >= max_nr_writes)
+			break;
+	}
+
+	return nr_reads + nr_writes;
+}
+
+static int throtl_select_dispatch(struct throtl_data *td, struct bio_list *bl)
+{
+	unsigned int nr_disp = 0;
+	struct throtl_grp *tg;
+	struct throtl_rb_root *st = &td->tg_service_tree;
+
+	while (1) {
+		tg = throtl_rb_first(st);
+
+		if (!tg)
+			break;
+
+		if (time_before(jiffies, tg->disptime))
+			break;
+
+		throtl_dequeue_tg(td, tg);
+
+		nr_disp += throtl_dispatch_tg(td, tg, bl);
+
+		if (tg->nr_queued[0] || tg->nr_queued[1]) {
+			tg_update_disptime(td, tg);
+			throtl_enqueue_tg(td, tg);
+		}
+
+		if (nr_disp >= throtl_quantum)
+			break;
+	}
+
+	return nr_disp;
+}
+
+/* Dispatch throttled bios. Should be called without queue lock held. */
+static int throtl_dispatch(struct request_queue *q)
+{
+	struct throtl_data *td = q->td;
+	unsigned int nr_disp = 0;
+	struct bio_list bio_list_on_stack;
+	struct bio *bio;
+
+	spin_lock_irq(q->queue_lock);
+
+	if (!total_nr_queued(td))
+		goto out;
+
+	bio_list_init(&bio_list_on_stack);
+
+	throtl_log(td, "dispatch nr_queued=%lu read=%u write=%u",
+			total_nr_queued(td), td->nr_queued[READ],
+			td->nr_queued[WRITE]);
+
+	nr_disp = throtl_select_dispatch(td, &bio_list_on_stack);
+
+	if (nr_disp)
+		throtl_log(td, "bios disp=%u", nr_disp);
+
+	throtl_schedule_next_dispatch(td);
+out:
+	spin_unlock_irq(q->queue_lock);
+
+	/*
+	 * If we dispatched some requests, unplug the queue to make sure
+	 * immediate dispatch
+	 */
+	if (nr_disp) {
+		while((bio = bio_list_pop(&bio_list_on_stack)))
+			generic_make_request(bio);
+		blk_unplug(q);
+	}
+	return nr_disp;
+}
+
+void blk_throtl_work(struct work_struct *work)
+{
+	struct throtl_data *td = container_of(work, struct throtl_data,
+					throtl_work.work);
+	struct request_queue *q = td->queue;
+
+	throtl_dispatch(q);
+}
+
+/* Call with queue lock held */
+void throtl_schedule_delayed_work(struct request_queue *q, unsigned long delay)
+{
+
+	struct throtl_data *td = q->td;
+	struct delayed_work *dwork = &td->throtl_work;
+
+	if (total_nr_queued(td) > 0) {
+		/*
+		 * We might have a work scheduled to be executed in future.
+		 * Cancel that and schedule a new one.
+		 */
+		__cancel_delayed_work(dwork);
+		kblockd_schedule_delayed_work(q, dwork, delay);
+		throtl_log(td, "schedule work. delay=%lu jiffies=%lu",
+				delay, jiffies);
+	}
+}
+EXPORT_SYMBOL(throtl_schedule_delayed_work);
+
+static void
+throtl_destroy_tg(struct throtl_data *td, struct throtl_grp *tg)
+{
+	/* Something wrong if we are trying to remove same group twice */
+	BUG_ON(hlist_unhashed(&tg->tg_node));
+
+	hlist_del_init(&tg->tg_node);
+
+	/*
+	 * Put the reference taken at the time of creation so that when all
+	 * queues are gone, group can be destroyed.
+	 */
+	throtl_put_tg(tg);
+	td->nr_undestroyed_grps--;
+}
+
+static void throtl_release_tgs(struct throtl_data *td)
+{
+	struct hlist_node *pos, *n;
+	struct throtl_grp *tg;
+
+	hlist_for_each_entry_safe(tg, pos, n, &td->tg_list, tg_node) {
+		/*
+		 * If cgroup removal path got to blk_group first and removed
+		 * it from cgroup list, then it will take care of destroying
+		 * cfqg also.
+		 */
+		if (!blkiocg_del_blkio_group(&tg->blkg))
+			throtl_destroy_tg(td, tg);
+	}
+}
+
+static void throtl_td_free(struct throtl_data *td)
+{
+	kfree(td);
+}
+
+/*
+ * Blk cgroup controller notification saying that blkio_group object is being
+ * delinked as associated cgroup object is going away. That also means that
+ * no new IO will come in this group. So get rid of this group as soon as
+ * any pending IO in the group is finished.
+ *
+ * This function is called under rcu_read_lock(). key is the rcu protected
+ * pointer. That means "key" is a valid throtl_data pointer as long as we are
+ * rcu read lock.
+ *
+ * "key" was fetched from blkio_group under blkio_cgroup->lock. That means
+ * it should not be NULL as even if queue was going away, cgroup deltion
+ * path got to it first.
+ */
+void throtl_unlink_blkio_group(void *key, struct blkio_group *blkg)
+{
+	unsigned long flags;
+	struct throtl_data *td = key;
+
+	spin_lock_irqsave(td->queue->queue_lock, flags);
+	throtl_destroy_tg(td, tg_of_blkg(blkg));
+	spin_unlock_irqrestore(td->queue->queue_lock, flags);
+}
+
+static void throtl_update_blkio_group_read_bps (struct blkio_group *blkg,
+			u64 read_bps)
+{
+	tg_of_blkg(blkg)->bps[READ] = read_bps;
+}
+
+static void throtl_update_blkio_group_write_bps (struct blkio_group *blkg,
+			u64 write_bps)
+{
+	tg_of_blkg(blkg)->bps[WRITE] = write_bps;
+}
+
+void throtl_shutdown_timer_wq(struct request_queue *q)
+{
+	struct throtl_data *td = q->td;
+
+	cancel_delayed_work_sync(&td->throtl_work);
+}
+
+static struct blkio_policy_type blkio_policy_throtl = {
+	.ops = {
+		.blkio_unlink_group_fn = throtl_unlink_blkio_group,
+		.blkio_update_group_read_bps_fn =
+					throtl_update_blkio_group_read_bps,
+		.blkio_update_group_write_bps_fn =
+					throtl_update_blkio_group_write_bps,
+	},
+};
+
+int blk_throtl_bio(struct request_queue *q, struct bio **biop)
+{
+	struct throtl_data *td = q->td;
+	struct throtl_grp *tg;
+	struct bio *bio = *biop;
+	bool rw = bio_data_dir(bio), update_disptime = true;
+
+	if (bio->bi_rw & REQ_THROTTLED) {
+		bio->bi_rw &= ~REQ_THROTTLED;
+		return 0;
+	}
+
+	spin_lock_irq(q->queue_lock);
+	tg = throtl_get_tg(td);
+
+	if (tg->nr_queued[rw]) {
+		/*
+		 * There is already another bio queued in same dir. No
+		 * need to update dispatch time.
+		 */
+		update_disptime = false;
+		goto queue_bio;
+	}
+
+	/* Bio is with-in rate limit of group */
+	if (tg_may_dispatch(td, tg, bio, NULL)) {
+		throtl_charge_bio(tg, bio);
+		goto out;
+	}
+
+queue_bio:
+	throtl_log_tg(td, tg, "[%c] bio. disp=%u sz=%u bps=%llu"
+			" queued=%d/%d", rw == READ ? 'R' : 'W',
+			tg->bytes_disp[rw], bio->bi_size, tg->bps[rw],
+			tg->nr_queued[READ], tg->nr_queued[WRITE]);
+
+	throtl_add_bio_tg(q->td, tg, bio);
+	*biop = NULL;
+
+	if (update_disptime) {
+		tg_update_disptime(td, tg);
+		throtl_schedule_next_dispatch(td);
+	}
+
+out:
+	spin_unlock_irq(q->queue_lock);
+	return 0;
+}
+
+int blk_throtl_init(struct request_queue *q)
+{
+	struct throtl_data *td;
+	struct throtl_grp *tg;
+
+	td = kzalloc_node(sizeof(*td), GFP_KERNEL, q->node);
+	if (!td)
+		return -ENOMEM;
+
+	INIT_HLIST_HEAD(&td->tg_list);
+	td->tg_service_tree = THROTL_RB_ROOT;
+
+	/* Init root group */
+	tg = &td->root_tg;
+	INIT_HLIST_NODE(&tg->tg_node);
+	RB_CLEAR_NODE(&tg->rb_node);
+	bio_list_init(&tg->bio_lists[0]);
+	bio_list_init(&tg->bio_lists[1]);
+
+	/* Practically unlimited BW */
+	tg->bps[0] = tg->bps[1] = -1;
+	atomic_set(&tg->ref, 1);
+
+	INIT_DELAYED_WORK(&td->throtl_work, blk_throtl_work);
+
+	rcu_read_lock();
+	blkiocg_add_blkio_group(&blkio_root_cgroup, &tg->blkg, (void *)td,
+					0, BLKIO_POLICY_THROTL);
+	rcu_read_unlock();
+
+	/* Attach throtl data to request queue */
+	td->queue = q;
+	q->td = td;
+	return 0;
+}
+
+void blk_throtl_exit(struct request_queue *q)
+{
+	struct throtl_data *td = q->td;
+	bool wait = false;
+
+	BUG_ON(!td);
+
+	throtl_shutdown_timer_wq(q);
+
+	spin_lock_irq(q->queue_lock);
+	throtl_release_tgs(td);
+	blkiocg_del_blkio_group(&td->root_tg.blkg);
+
+	/* If there are other groups */
+	if (td->nr_undestroyed_grps >= 1)
+		wait = true;
+
+	spin_unlock_irq(q->queue_lock);
+
+	/*
+	 * Wait for tg->blkg->key accessors to exit their grace periods.
+	 * Do this wait only if there are other undestroyed groups out
+	 * there (other than root group). This can happen if cgroup deletion
+	 * path claimed the responsibility of cleaning up a group before
+	 * queue cleanup code get to the group.
+	 *
+	 * Do not call synchronize_rcu() unconditionally as there are drivers
+	 * which create/delete request queue hundreds of times during scan/boot
+	 * and synchronize_rcu() can take significant time and slow down boot.
+	 */
+	if (wait)
+		synchronize_rcu();
+	throtl_td_free(td);
+}
+
+static int __init throtl_init(void)
+{
+	blkio_policy_register(&blkio_policy_throtl);
+	return 0;
+}
+
+module_init(throtl_init);
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index ca83a97c9715..10a0c291b55a 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -130,6 +130,8 @@ enum rq_flag_bits {
 	/* bio only flags */
 	__REQ_UNPLUG,		/* unplug the immediately after submission */
 	__REQ_RAHEAD,		/* read ahead, can fail anytime */
+	__REQ_THROTTLED,	/* This bio has already been subjected to
+				 * throttling rules. Don't do it again. */
 
 	/* request only flags */
 	__REQ_SORTED,		/* elevator knows about this request */
@@ -172,6 +174,7 @@ enum rq_flag_bits {
 
 #define REQ_UNPLUG		(1 << __REQ_UNPLUG)
 #define REQ_RAHEAD		(1 << __REQ_RAHEAD)
+#define REQ_THROTTLED		(1 << __REQ_THROTTLED)
 
 #define REQ_SORTED		(1 << __REQ_SORTED)
 #define REQ_SOFTBARRIER		(1 << __REQ_SOFTBARRIER)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 780824edac16..1341df5806df 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -371,6 +371,11 @@ struct request_queue
 #if defined(CONFIG_BLK_DEV_BSG)
 	struct bsg_class_device bsg_dev;
 #endif
+
+#ifdef CONFIG_BLK_DEV_THROTTLING
+	/* Throttle data */
+	struct throtl_data *td;
+#endif
 };
 
 #define QUEUE_FLAG_CLUSTER	0	/* cluster several segments into 1 */
@@ -1131,6 +1136,7 @@ static inline void put_dev_sector(Sector p)
 
 struct work_struct;
 int kblockd_schedule_work(struct request_queue *q, struct work_struct *work);
+int kblockd_schedule_delayed_work(struct request_queue *q, struct delayed_work *dwork, unsigned long delay);
 
 #ifdef CONFIG_BLK_CGROUP
 /*
@@ -1174,6 +1180,24 @@ static inline uint64_t rq_io_start_time_ns(struct request *req)
 }
 #endif
 
+#ifdef CONFIG_BLK_DEV_THROTTLING
+extern int blk_throtl_init(struct request_queue *q);
+extern void blk_throtl_exit(struct request_queue *q);
+extern int blk_throtl_bio(struct request_queue *q, struct bio **bio);
+extern void throtl_schedule_delayed_work(struct request_queue *q, unsigned long delay);
+extern void throtl_shutdown_timer_wq(struct request_queue *q);
+#else /* CONFIG_BLK_DEV_THROTTLING */
+static inline int blk_throtl_bio(struct request_queue *q, struct bio **bio)
+{
+	return 0;
+}
+
+static inline int blk_throtl_init(struct request_queue *q) { return 0; }
+static inline int blk_throtl_exit(struct request_queue *q) { return 0; }
+static inline void throtl_schedule_delayed_work(struct request_queue *q, unsigned long delay) {}
+static inline void throtl_shutdown_timer_wq(struct request_queue *q) {}
+#endif /* CONFIG_BLK_DEV_THROTTLING */
+
 #define MODULE_ALIAS_BLOCKDEV(major,minor) \
 	MODULE_ALIAS("block-major-" __stringify(major) "-" __stringify(minor))
 #define MODULE_ALIAS_BLOCKDEV_MAJOR(major) \
diff --git a/init/Kconfig b/init/Kconfig
index 2de5b1cbadd9..950ba26f7233 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -634,11 +634,14 @@ config BLK_CGROUP
 
 	Currently, CFQ IO scheduler uses it to recognize task groups and
 	control disk bandwidth allocation (proportional time slice allocation)
-	to such task groups.
+	to such task groups. It is also used by bio throttling logic in
+	block layer to implement upper limit in IO rates on a device.
 
 	This option only enables generic Block IO controller infrastructure.
-	One needs to also enable actual IO controlling logic in CFQ for it
-	to take effect. (CONFIG_CFQ_GROUP_IOSCHED=y).
+	One needs to also enable actual IO controlling logic/policy. For
+	enabling proportional weight division of disk bandwidth in CFQ seti
+	CONFIG_CFQ_GROUP_IOSCHED=y and for enabling throttling policy set
+	CONFIG_BLK_THROTTLE=y.
 
 	See Documentation/cgroups/blkio-controller.txt for more information.
 
-- 
cgit v1.2.3


From 1ec5584e3edf9c4bf2c88c846534d19cf986ba11 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sun, 15 Aug 2010 21:50:52 +0200
Subject: libfs: use generic_file_llseek for simple_attr

Simple attribute files need to be seekable to
allow resetting the file for another read.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 include/linux/fs.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 76041b614758..c8effc81f0c9 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2450,6 +2450,7 @@ static const struct file_operations __fops = {				\
 	.release = simple_attr_release,					\
 	.read	 = simple_attr_read,					\
 	.write	 = simple_attr_write,					\
+	.llseek	 = generic_file_llseek,					\
 };
 
 static inline void __attribute__((format(printf, 1, 2)))
-- 
cgit v1.2.3


From dd3932eddf428571762596e17b65f5dc92ca361b Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 16 Sep 2010 20:51:46 +0200
Subject: block: remove BLKDEV_IFL_WAIT

All the blkdev_issue_* helpers can only sanely be used for synchronous
caller.  To issue cache flushes or barriers asynchronously the caller needs
to set up a bio by itself with a completion callback to move the asynchronous
state machine ahead.  So drop the BLKDEV_IFL_WAIT flag that is always
specified when calling blkdev_issue_* and also remove the now unused flags
argument to blkdev_issue_flush and blkdev_issue_zeroout.  For
blkdev_issue_discard we need to keep it for the secure discard flag, which
gains a more descriptive name and loses the bitops vs flag confusion.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-flush.c                  | 25 +++++++++++--------------
 block/blk-lib.c                    | 21 ++++++++-------------
 block/ioctl.c                      |  4 ++--
 drivers/block/drbd/drbd_int.h      |  3 +--
 drivers/block/drbd/drbd_receiver.c |  2 +-
 fs/block_dev.c                     |  2 +-
 fs/btrfs/extent-tree.c             |  3 +--
 fs/ext3/fsync.c                    |  3 +--
 fs/ext4/fsync.c                    |  5 ++---
 fs/ext4/mballoc.c                  |  3 +--
 fs/fat/fatent.c                    |  3 +--
 fs/gfs2/rgrp.c                     |  5 ++---
 fs/jbd2/checkpoint.c               |  3 +--
 fs/jbd2/commit.c                   |  6 ++----
 fs/nilfs2/the_nilfs.c              |  4 ++--
 fs/reiserfs/file.c                 |  3 +--
 fs/xfs/linux-2.6/xfs_super.c       |  3 +--
 include/linux/blkdev.h             | 14 +++++---------
 mm/swapfile.c                      |  6 +++---
 19 files changed, 47 insertions(+), 71 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-flush.c b/block/blk-flush.c
index 62b7df9bca9d..54b123d6563e 100644
--- a/block/blk-flush.c
+++ b/block/blk-flush.c
@@ -205,7 +205,6 @@ static void bio_end_flush(struct bio *bio, int err)
  * @bdev:	blockdev to issue flush for
  * @gfp_mask:	memory allocation flags (for bio_alloc)
  * @error_sector:	error sector
- * @flags:	BLKDEV_IFL_* flags to control behaviour
  *
  * Description:
  *    Issue a flush for the block device in question. Caller can supply
@@ -214,7 +213,7 @@ static void bio_end_flush(struct bio *bio, int err)
  *    request was pushed in some internal queue for later handling.
  */
 int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask,
-		sector_t *error_sector, unsigned long flags)
+		sector_t *error_sector)
 {
 	DECLARE_COMPLETION_ONSTACK(wait);
 	struct request_queue *q;
@@ -240,21 +239,19 @@ int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask,
 	bio = bio_alloc(gfp_mask, 0);
 	bio->bi_end_io = bio_end_flush;
 	bio->bi_bdev = bdev;
-	if (test_bit(BLKDEV_WAIT, &flags))
-		bio->bi_private = &wait;
+	bio->bi_private = &wait;
 
 	bio_get(bio);
 	submit_bio(WRITE_FLUSH, bio);
-	if (test_bit(BLKDEV_WAIT, &flags)) {
-		wait_for_completion(&wait);
-		/*
-		 * The driver must store the error location in ->bi_sector, if
-		 * it supports it. For non-stacked drivers, this should be
-		 * copied from blk_rq_pos(rq).
-		 */
-		if (error_sector)
-			*error_sector = bio->bi_sector;
-	}
+	wait_for_completion(&wait);
+
+	/*
+	 * The driver must store the error location in ->bi_sector, if
+	 * it supports it. For non-stacked drivers, this should be
+	 * copied from blk_rq_pos(rq).
+	 */
+	if (error_sector)
+               *error_sector = bio->bi_sector;
 
 	if (!bio_flagged(bio, BIO_UPTODATE))
 		ret = -EIO;
diff --git a/block/blk-lib.c b/block/blk-lib.c
index fe2e6ed0f510..1a320d2406b0 100644
--- a/block/blk-lib.c
+++ b/block/blk-lib.c
@@ -61,7 +61,7 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 		max_discard_sectors &= ~(disc_sects - 1);
 	}
 
-	if (flags & BLKDEV_IFL_SECURE) {
+	if (flags & BLKDEV_DISCARD_SECURE) {
 		if (!blk_queue_secdiscard(q))
 			return -EOPNOTSUPP;
 		type |= REQ_SECURE;
@@ -77,8 +77,7 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 		bio->bi_sector = sector;
 		bio->bi_end_io = blkdev_discard_end_io;
 		bio->bi_bdev = bdev;
-		if (flags & BLKDEV_IFL_WAIT)
-			bio->bi_private = &wait;
+		bio->bi_private = &wait;
 
 		if (nr_sects > max_discard_sectors) {
 			bio->bi_size = max_discard_sectors << 9;
@@ -92,8 +91,7 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 		bio_get(bio);
 		submit_bio(type, bio);
 
-		if (flags & BLKDEV_IFL_WAIT)
-			wait_for_completion(&wait);
+		wait_for_completion(&wait);
 
 		if (bio_flagged(bio, BIO_EOPNOTSUPP))
 			ret = -EOPNOTSUPP;
@@ -139,7 +137,6 @@ static void bio_batch_end_io(struct bio *bio, int err)
  * @sector:	start sector
  * @nr_sects:	number of sectors to write
  * @gfp_mask:	memory allocation flags (for bio_alloc)
- * @flags:	BLKDEV_IFL_* flags to control behaviour
  *
  * Description:
  *  Generate and issue number of bios with zerofiled pages.
@@ -148,7 +145,7 @@ static void bio_batch_end_io(struct bio *bio, int err)
  */
 
 int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
-			sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
+			sector_t nr_sects, gfp_t gfp_mask)
 {
 	int ret;
 	struct bio *bio;
@@ -174,8 +171,7 @@ submit:
 		bio->bi_sector = sector;
 		bio->bi_bdev   = bdev;
 		bio->bi_end_io = bio_batch_end_io;
-		if (flags & BLKDEV_IFL_WAIT)
-			bio->bi_private = &bb;
+		bio->bi_private = &bb;
 
 		while (nr_sects != 0) {
 			sz = min((sector_t) PAGE_SIZE >> 9 , nr_sects);
@@ -193,10 +189,9 @@ submit:
 		submit_bio(WRITE, bio);
 	}
 
-	if (flags & BLKDEV_IFL_WAIT)
-		/* Wait for bios in-flight */
-		while ( issued != atomic_read(&bb.done))
-			wait_for_completion(&wait);
+	/* Wait for bios in-flight */
+	while (issued != atomic_read(&bb.done))
+		wait_for_completion(&wait);
 
 	if (!test_bit(BIO_UPTODATE, &bb.flags))
 		/* One of bios in the batch was completed with error.*/
diff --git a/block/ioctl.c b/block/ioctl.c
index d8052f0dabd3..cb2b9099862b 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -116,7 +116,7 @@ static int blkdev_reread_part(struct block_device *bdev)
 static int blk_ioctl_discard(struct block_device *bdev, uint64_t start,
 			     uint64_t len, int secure)
 {
-	unsigned long flags = BLKDEV_IFL_WAIT;
+	unsigned long flags = 0;
 
 	if (start & 511)
 		return -EINVAL;
@@ -128,7 +128,7 @@ static int blk_ioctl_discard(struct block_device *bdev, uint64_t start,
 	if (start + len > (bdev->bd_inode->i_size >> 9))
 		return -EINVAL;
 	if (secure)
-		flags |= BLKDEV_IFL_SECURE;
+		flags |= BLKDEV_DISCARD_SECURE;
 	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL, flags);
 }
 
diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index 352441b0f92f..c2ef476f5711 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -2321,8 +2321,7 @@ static inline void drbd_md_flush(struct drbd_conf *mdev)
 	if (test_bit(MD_NO_BARRIER, &mdev->flags))
 		return;
 
-	r = blkdev_issue_flush(mdev->ldev->md_bdev, GFP_KERNEL, NULL,
-			BLKDEV_IFL_WAIT);
+	r = blkdev_issue_flush(mdev->ldev->md_bdev, GFP_KERNEL, NULL);
 	if (r) {
 		set_bit(MD_NO_BARRIER, &mdev->flags);
 		dev_err(DEV, "meta data flush failed with status %d, disabling md-flushes\n", r);
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 081522d3c742..df15e7f0e7b7 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -975,7 +975,7 @@ static enum finish_epoch drbd_flush_after_epoch(struct drbd_conf *mdev, struct d
 
 	if (mdev->write_ordering >= WO_bdev_flush && get_ldev(mdev)) {
 		rv = blkdev_issue_flush(mdev->ldev->backing_bdev, GFP_KERNEL,
-					NULL, BLKDEV_IFL_WAIT);
+					NULL);
 		if (rv) {
 			dev_err(DEV, "local disk flush failed with status %d\n", rv);
 			/* would rather check on EOPNOTSUPP, but that is not reliable.
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 50e8c8582faa..b737451e2e9d 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -370,7 +370,7 @@ int blkdev_fsync(struct file *filp, int datasync)
 	 */
 	mutex_unlock(&bd_inode->i_mutex);
 
-	error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL, BLKDEV_IFL_WAIT);
+	error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL);
 	if (error == -EOPNOTSUPP)
 		error = 0;
 
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 43dc9ea9aef6..0b81ecdb101c 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -1695,8 +1695,7 @@ static int remove_extent_backref(struct btrfs_trans_handle *trans,
 static void btrfs_issue_discard(struct block_device *bdev,
 				u64 start, u64 len)
 {
-	blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_KERNEL,
-			BLKDEV_IFL_WAIT);
+	blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_KERNEL, 0);
 }
 
 static int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
diff --git a/fs/ext3/fsync.c b/fs/ext3/fsync.c
index d7e9f74dc3a6..09b13bb34c94 100644
--- a/fs/ext3/fsync.c
+++ b/fs/ext3/fsync.c
@@ -90,7 +90,6 @@ int ext3_sync_file(struct file *file, int datasync)
 	 * storage
 	 */
 	if (needs_barrier)
-		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL,
-				BLKDEV_IFL_WAIT);
+		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
 	return ret;
 }
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c
index 592adf2e546e..3f3ff5ee8f9d 100644
--- a/fs/ext4/fsync.c
+++ b/fs/ext4/fsync.c
@@ -128,10 +128,9 @@ int ext4_sync_file(struct file *file, int datasync)
 		    (journal->j_fs_dev != journal->j_dev) &&
 		    (journal->j_flags & JBD2_BARRIER))
 			blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL,
-					NULL, BLKDEV_IFL_WAIT);
+					NULL);
 		ret = jbd2_log_wait_commit(journal, commit_tid);
 	} else if (journal->j_flags & JBD2_BARRIER)
-		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL,
-			BLKDEV_IFL_WAIT);
+		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
 	return ret;
 }
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index a22bfef3da95..19aa0d44d822 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2566,8 +2566,7 @@ static inline void ext4_issue_discard(struct super_block *sb,
 	discard_block = block + ext4_group_first_block_no(sb, block_group);
 	trace_ext4_discard_blocks(sb,
 			(unsigned long long) discard_block, count);
-	ret = sb_issue_discard(sb, discard_block, count, GFP_NOFS,
-			       BLKDEV_IFL_WAIT);
+	ret = sb_issue_discard(sb, discard_block, count, GFP_NOFS, 0);
 	if (ret == EOPNOTSUPP) {
 		ext4_warning(sb, "discard not supported, disabling");
 		clear_opt(EXT4_SB(sb)->s_mount_opt, DISCARD);
diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c
index f9a0b7ae8648..b47d2c9f4fa1 100644
--- a/fs/fat/fatent.c
+++ b/fs/fat/fatent.c
@@ -578,8 +578,7 @@ int fat_free_clusters(struct inode *inode, int cluster)
 				sb_issue_discard(sb,
 					fat_clus_to_blknr(sbi, first_cl),
 					nr_clus * sbi->sec_per_clus,
-					GFP_NOFS,
-					BLKDEV_IFL_WAIT);
+					GFP_NOFS, 0);
 
 				first_cl = cluster;
 			}
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 379316472918..38b3ea1abacc 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -854,7 +854,7 @@ static void gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset,
 				if ((start + nr_sects) != blk) {
 					rv = blkdev_issue_discard(bdev, start,
 							    nr_sects, GFP_NOFS,
-							    BLKDEV_IFL_WAIT);
+							    0);
 					if (rv)
 						goto fail;
 					nr_sects = 0;
@@ -868,8 +868,7 @@ start_new_extent:
 		}
 	}
 	if (nr_sects) {
-		rv = blkdev_issue_discard(bdev, start, nr_sects, GFP_NOFS,
-					 BLKDEV_IFL_WAIT);
+		rv = blkdev_issue_discard(bdev, start, nr_sects, GFP_NOFS, 0);
 		if (rv)
 			goto fail;
 	}
diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 5247e7ffdcb4..6571a056e55d 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -532,8 +532,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	 */
 	if ((journal->j_fs_dev != journal->j_dev) &&
 	    (journal->j_flags & JBD2_BARRIER))
-		blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL,
-			BLKDEV_IFL_WAIT);
+		blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
 	if (!(journal->j_flags & JBD2_ABORT))
 		jbd2_journal_update_superblock(journal, 1);
 	return 0;
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index f204e27f44d1..cb43c605cfaa 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -684,8 +684,7 @@ start_journal_io:
 	if (commit_transaction->t_flushed_data_blocks &&
 	    (journal->j_fs_dev != journal->j_dev) &&
 	    (journal->j_flags & JBD2_BARRIER))
-		blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL,
-			BLKDEV_IFL_WAIT);
+		blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
 
 	/* Done it all: now write the commit record asynchronously. */
 	if (JBD2_HAS_INCOMPAT_FEATURE(journal,
@@ -810,8 +809,7 @@ wait_for_iobuf:
 	if (JBD2_HAS_INCOMPAT_FEATURE(journal,
 				      JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT) &&
 	    journal->j_flags & JBD2_BARRIER) {
-		blkdev_issue_flush(journal->j_dev, GFP_KERNEL, NULL,
-				   BLKDEV_IFL_WAIT);
+		blkdev_issue_flush(journal->j_dev, GFP_KERNEL, NULL);
 	}
 
 	if (err)
diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c
index 400b2caef4d8..d97310f07bef 100644
--- a/fs/nilfs2/the_nilfs.c
+++ b/fs/nilfs2/the_nilfs.c
@@ -774,7 +774,7 @@ int nilfs_discard_segments(struct the_nilfs *nilfs, __u64 *segnump,
 			ret = blkdev_issue_discard(nilfs->ns_bdev,
 						   start * sects_per_block,
 						   nblocks * sects_per_block,
-						   GFP_NOFS, BLKDEV_IFL_WAIT);
+						   GFP_NOFS, 0);
 			if (ret < 0)
 				return ret;
 			nblocks = 0;
@@ -784,7 +784,7 @@ int nilfs_discard_segments(struct the_nilfs *nilfs, __u64 *segnump,
 		ret = blkdev_issue_discard(nilfs->ns_bdev,
 					   start * sects_per_block,
 					   nblocks * sects_per_block,
-					   GFP_NOFS, BLKDEV_IFL_WAIT);
+					   GFP_NOFS, 0);
 	return ret;
 }
 
diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index 6846371498b6..91f080cc76c8 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -152,8 +152,7 @@ static int reiserfs_sync_file(struct file *filp, int datasync)
 	barrier_done = reiserfs_commit_for_inode(inode);
 	reiserfs_write_unlock(inode->i_sb);
 	if (barrier_done != 1 && reiserfs_barrier_flush(inode->i_sb))
-		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL, 
-			BLKDEV_IFL_WAIT);
+		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
 	if (barrier_done < 0)
 		return barrier_done;
 	return (err < 0) ? -EIO : 0;
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 15c35b62ff14..5fa7a30cc3f0 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -693,8 +693,7 @@ void
 xfs_blkdev_issue_flush(
 	xfs_buftarg_t		*buftarg)
 {
-	blkdev_issue_flush(buftarg->bt_bdev, GFP_KERNEL, NULL,
-			BLKDEV_IFL_WAIT);
+	blkdev_issue_flush(buftarg->bt_bdev, GFP_KERNEL, NULL);
 }
 
 STATIC void
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index cfcb3a610605..accbd0e5c893 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -867,18 +867,14 @@ static inline struct request *blk_map_queue_find_tag(struct blk_queue_tag *bqt,
 		return NULL;
 	return bqt->tag_index[tag];
 }
-enum{
-	BLKDEV_WAIT,	/* wait for completion */
-	BLKDEV_SECURE,	/* secure discard */
-};
-#define BLKDEV_IFL_WAIT		(1 << BLKDEV_WAIT)
-#define BLKDEV_IFL_SECURE	(1 << BLKDEV_SECURE)
-extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *,
-			unsigned long);
+
+#define BLKDEV_DISCARD_SECURE  0x01    /* secure discard */
+
+extern int blkdev_issue_flush(struct block_device *, gfp_t, sector_t *);
 extern int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
 extern int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
-			sector_t nr_sects, gfp_t gfp_mask, unsigned long flags);
+			sector_t nr_sects, gfp_t gfp_mask);
 static inline int sb_issue_discard(struct super_block *sb, sector_t block,
 		sector_t nr_blocks, gfp_t gfp_mask, unsigned long flags)
 {
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 68cda164dff6..e132e1708acc 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -141,7 +141,7 @@ static int discard_swap(struct swap_info_struct *si)
 	nr_blocks = ((sector_t)se->nr_pages - 1) << (PAGE_SHIFT - 9);
 	if (nr_blocks) {
 		err = blkdev_issue_discard(si->bdev, start_block,
-				nr_blocks, GFP_KERNEL, BLKDEV_IFL_WAIT);
+				nr_blocks, GFP_KERNEL, 0);
 		if (err)
 			return err;
 		cond_resched();
@@ -152,7 +152,7 @@ static int discard_swap(struct swap_info_struct *si)
 		nr_blocks = (sector_t)se->nr_pages << (PAGE_SHIFT - 9);
 
 		err = blkdev_issue_discard(si->bdev, start_block,
-				nr_blocks, GFP_KERNEL, BLKDEV_IFL_WAIT);
+				nr_blocks, GFP_KERNEL, 0);
 		if (err)
 			break;
 
@@ -191,7 +191,7 @@ static void discard_swap_cluster(struct swap_info_struct *si,
 			start_block <<= PAGE_SHIFT - 9;
 			nr_blocks <<= PAGE_SHIFT - 9;
 			if (blkdev_issue_discard(si->bdev, start_block,
-				    nr_blocks, GFP_NOIO, BLKDEV_IFL_WAIT))
+				    nr_blocks, GFP_NOIO, 0))
 				break;
 		}
 
-- 
cgit v1.2.3


From 074ac8df9f93f2a35a356d92fd7f16cd846f0a03 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 16 Sep 2010 14:58:22 +0200
Subject: cfg80211/nl80211: introduce p2p device types

This adds P2P-STA and P2P-GO as device types so
we can distinguish between those and normal STA
or AP (respectively) type interfaces.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h |  4 ++++
 net/wireless/core.c     |  2 ++
 net/wireless/mlme.c     |  3 ++-
 net/wireless/nl80211.c  | 56 ++++++++++++++++++++++++++++++++++---------------
 net/wireless/sme.c      |  9 +++++---
 net/wireless/util.c     | 15 ++++++++++---
 6 files changed, 65 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 31603e8b5581..f0518b0278a9 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -1020,6 +1020,8 @@ enum nl80211_attrs {
  * @NL80211_IFTYPE_WDS: wireless distribution interface
  * @NL80211_IFTYPE_MONITOR: monitor interface receiving all frames
  * @NL80211_IFTYPE_MESH_POINT: mesh point
+ * @NL80211_IFTYPE_P2P_CLIENT: P2P client
+ * @NL80211_IFTYPE_P2P_GO: P2P group owner
  * @NL80211_IFTYPE_MAX: highest interface type number currently defined
  * @NUM_NL80211_IFTYPES: number of defined interface types
  *
@@ -1036,6 +1038,8 @@ enum nl80211_iftype {
 	NL80211_IFTYPE_WDS,
 	NL80211_IFTYPE_MONITOR,
 	NL80211_IFTYPE_MESH_POINT,
+	NL80211_IFTYPE_P2P_CLIENT,
+	NL80211_IFTYPE_P2P_GO,
 
 	/* keep last */
 	NUM_NL80211_IFTYPES,
diff --git a/net/wireless/core.c b/net/wireless/core.c
index b8191cf86226..ff9615a7ee7a 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -726,6 +726,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb,
 			dev->ethtool_ops = &cfg80211_ethtool_ops;
 
 		if ((wdev->iftype == NL80211_IFTYPE_STATION ||
+		     wdev->iftype == NL80211_IFTYPE_P2P_CLIENT ||
 		     wdev->iftype == NL80211_IFTYPE_ADHOC) && !wdev->use_4addr)
 			dev->priv_flags |= IFF_DONT_BRIDGE;
 		break;
@@ -734,6 +735,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb,
 		case NL80211_IFTYPE_ADHOC:
 			cfg80211_leave_ibss(rdev, dev, true);
 			break;
+		case NL80211_IFTYPE_P2P_CLIENT:
 		case NL80211_IFTYPE_STATION:
 			wdev_lock(wdev);
 #ifdef CONFIG_CFG80211_WEXT
diff --git a/net/wireless/mlme.c b/net/wireless/mlme.c
index 8515b1e5c578..46f371160896 100644
--- a/net/wireless/mlme.c
+++ b/net/wireless/mlme.c
@@ -882,7 +882,8 @@ int cfg80211_mlme_mgmt_tx(struct cfg80211_registered_device *rdev,
 		if (!wdev->current_bss ||
 		    memcmp(wdev->current_bss->pub.bssid, mgmt->bssid,
 			   ETH_ALEN) != 0 ||
-		    (wdev->iftype == NL80211_IFTYPE_STATION &&
+		    ((wdev->iftype == NL80211_IFTYPE_STATION ||
+		      wdev->iftype == NL80211_IFTYPE_P2P_CLIENT) &&
 		     memcmp(wdev->current_bss->pub.bssid, mgmt->da,
 			    ETH_ALEN) != 0)) {
 			wdev_unlock(wdev);
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 1d6ef24254fe..f15b1af2c768 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -410,12 +410,14 @@ static int nl80211_key_allowed(struct wireless_dev *wdev)
 	switch (wdev->iftype) {
 	case NL80211_IFTYPE_AP:
 	case NL80211_IFTYPE_AP_VLAN:
+	case NL80211_IFTYPE_P2P_GO:
 		break;
 	case NL80211_IFTYPE_ADHOC:
 		if (!wdev->current_bss)
 			return -ENOLINK;
 		break;
 	case NL80211_IFTYPE_STATION:
+	case NL80211_IFTYPE_P2P_CLIENT:
 		if (wdev->sme_state != CFG80211_SME_CONNECTED)
 			return -ENOLINK;
 		break;
@@ -766,7 +768,8 @@ static bool nl80211_can_set_dev_channel(struct wireless_dev *wdev)
 		wdev->iftype == NL80211_IFTYPE_AP ||
 		wdev->iftype == NL80211_IFTYPE_WDS ||
 		wdev->iftype == NL80211_IFTYPE_MESH_POINT ||
-		wdev->iftype == NL80211_IFTYPE_MONITOR;
+		wdev->iftype == NL80211_IFTYPE_MONITOR ||
+		wdev->iftype == NL80211_IFTYPE_P2P_GO;
 }
 
 static int __nl80211_set_channel(struct cfg80211_registered_device *rdev,
@@ -1693,7 +1696,8 @@ static int nl80211_addset_beacon(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto unlock_rtnl;
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -1785,7 +1789,8 @@ static int nl80211_del_beacon(struct sk_buff *skb, struct genl_info *info)
 		goto out;
 	}
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -2128,10 +2133,12 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info)
 	switch (dev->ieee80211_ptr->iftype) {
 	case NL80211_IFTYPE_AP:
 	case NL80211_IFTYPE_AP_VLAN:
+	case NL80211_IFTYPE_P2P_GO:
 		/* disallow mesh-specific things */
 		if (params.plink_action)
 			err = -EINVAL;
 		break;
+	case NL80211_IFTYPE_P2P_CLIENT:
 	case NL80211_IFTYPE_STATION:
 		/* disallow everything but AUTHORIZED flag */
 		if (params.plink_action)
@@ -2233,7 +2240,8 @@ static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info)
 		goto out_rtnl;
 
 	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN) {
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) {
 		err = -EINVAL;
 		goto out;
 	}
@@ -2286,7 +2294,8 @@ static int nl80211_del_station(struct sk_buff *skb, struct genl_info *info)
 
 	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
 	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) {
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) {
 		err = -EINVAL;
 		goto out;
 	}
@@ -2660,7 +2669,8 @@ static int nl80211_set_bss(struct sk_buff *skb, struct genl_info *info)
 		goto out;
 	}
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -3363,6 +3373,7 @@ static int nl80211_send_bss(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 	}
 
 	switch (wdev->iftype) {
+	case NL80211_IFTYPE_P2P_CLIENT:
 	case NL80211_IFTYPE_STATION:
 		if (intbss == wdev->current_bss)
 			NLA_PUT_U32(msg, NL80211_BSS_STATUS,
@@ -3649,7 +3660,8 @@ static int nl80211_authenticate(struct sk_buff *skb, struct genl_info *info)
 		goto out;
 	}
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -3804,7 +3816,8 @@ static int nl80211_associate(struct sk_buff *skb, struct genl_info *info)
 		goto out;
 	}
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -3888,7 +3901,8 @@ static int nl80211_deauthenticate(struct sk_buff *skb, struct genl_info *info)
 		goto out;
 	}
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -3954,7 +3968,8 @@ static int nl80211_disassociate(struct sk_buff *skb, struct genl_info *info)
 		goto out;
 	}
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -4332,7 +4347,8 @@ static int nl80211_connect(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto unlock_rtnl;
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -4408,7 +4424,8 @@ static int nl80211_disconnect(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto unlock_rtnl;
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -4496,7 +4513,8 @@ static int nl80211_setdel_pmksa(struct sk_buff *skb, struct genl_info *info)
 	pmksa.pmkid = nla_data(info->attrs[NL80211_ATTR_PMKID]);
 	pmksa.bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -4541,7 +4559,8 @@ static int nl80211_flush_pmksa(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto out_rtnl;
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION) {
+	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -4823,7 +4842,8 @@ static int nl80211_register_mgmt(struct sk_buff *skb, struct genl_info *info)
 		goto unlock_rtnl;
 
 	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC) {
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -4875,7 +4895,8 @@ static int nl80211_tx_mgmt(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC) {
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto out;
 	}
@@ -5093,7 +5114,8 @@ static int nl80211_set_cqm_rssi(struct genl_info *info,
 		goto unlock_rdev;
 	}
 
-	if (wdev->iftype != NL80211_IFTYPE_STATION) {
+	if (wdev->iftype != NL80211_IFTYPE_STATION &&
+	    wdev->iftype != NL80211_IFTYPE_P2P_CLIENT) {
 		err = -EOPNOTSUPP;
 		goto unlock_rdev;
 	}
diff --git a/net/wireless/sme.c b/net/wireless/sme.c
index a8c2d6b877ae..f161b9844542 100644
--- a/net/wireless/sme.c
+++ b/net/wireless/sme.c
@@ -411,7 +411,8 @@ void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid,
 
 	ASSERT_WDEV_LOCK(wdev);
 
-	if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION))
+	if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION &&
+		    wdev->iftype != NL80211_IFTYPE_P2P_CLIENT))
 		return;
 
 	if (wdev->sme_state != CFG80211_SME_CONNECTING)
@@ -548,7 +549,8 @@ void __cfg80211_roamed(struct wireless_dev *wdev, const u8 *bssid,
 
 	ASSERT_WDEV_LOCK(wdev);
 
-	if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION))
+	if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION &&
+		    wdev->iftype != NL80211_IFTYPE_P2P_CLIENT))
 		return;
 
 	if (wdev->sme_state != CFG80211_SME_CONNECTED)
@@ -644,7 +646,8 @@ void __cfg80211_disconnected(struct net_device *dev, const u8 *ie,
 
 	ASSERT_WDEV_LOCK(wdev);
 
-	if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION))
+	if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION &&
+		    wdev->iftype != NL80211_IFTYPE_P2P_CLIENT))
 		return;
 
 	if (wdev->sme_state != CFG80211_SME_CONNECTED)
diff --git a/net/wireless/util.c b/net/wireless/util.c
index bca32eb8f446..fb5448f7d55a 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -326,7 +326,8 @@ int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
 		cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS)) {
 	case cpu_to_le16(IEEE80211_FCTL_TODS):
 		if (unlikely(iftype != NL80211_IFTYPE_AP &&
-			     iftype != NL80211_IFTYPE_AP_VLAN))
+			     iftype != NL80211_IFTYPE_AP_VLAN &&
+			     iftype != NL80211_IFTYPE_P2P_GO))
 			return -1;
 		break;
 	case cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS):
@@ -354,7 +355,8 @@ int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr,
 		break;
 	case cpu_to_le16(IEEE80211_FCTL_FROMDS):
 		if ((iftype != NL80211_IFTYPE_STATION &&
-		    iftype != NL80211_IFTYPE_MESH_POINT) ||
+		     iftype != NL80211_IFTYPE_P2P_CLIENT &&
+		     iftype != NL80211_IFTYPE_MESH_POINT) ||
 		    (is_multicast_ether_addr(dst) &&
 		     !compare_ether_addr(src, addr)))
 			return -1;
@@ -431,6 +433,7 @@ int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr,
 	switch (iftype) {
 	case NL80211_IFTYPE_AP:
 	case NL80211_IFTYPE_AP_VLAN:
+	case NL80211_IFTYPE_P2P_GO:
 		fc |= cpu_to_le16(IEEE80211_FCTL_FROMDS);
 		/* DA BSSID SA */
 		memcpy(hdr.addr1, skb->data, ETH_ALEN);
@@ -439,6 +442,7 @@ int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr,
 		hdrlen = 24;
 		break;
 	case NL80211_IFTYPE_STATION:
+	case NL80211_IFTYPE_P2P_CLIENT:
 		fc |= cpu_to_le16(IEEE80211_FCTL_TODS);
 		/* BSSID SA DA */
 		memcpy(hdr.addr1, bssid, ETH_ALEN);
@@ -778,7 +782,9 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev,
 
 	/* if it's part of a bridge, reject changing type to station/ibss */
 	if ((dev->priv_flags & IFF_BRIDGE_PORT) &&
-	    (ntype == NL80211_IFTYPE_ADHOC || ntype == NL80211_IFTYPE_STATION))
+	    (ntype == NL80211_IFTYPE_ADHOC ||
+	     ntype == NL80211_IFTYPE_STATION ||
+	     ntype == NL80211_IFTYPE_P2P_CLIENT))
 		return -EBUSY;
 
 	if (ntype != otype) {
@@ -789,6 +795,7 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev,
 			cfg80211_leave_ibss(rdev, dev, false);
 			break;
 		case NL80211_IFTYPE_STATION:
+		case NL80211_IFTYPE_P2P_CLIENT:
 			cfg80211_disconnect(rdev, dev,
 					    WLAN_REASON_DEAUTH_LEAVING, true);
 			break;
@@ -817,9 +824,11 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev,
 			if (dev->ieee80211_ptr->use_4addr)
 				break;
 			/* fall through */
+		case NL80211_IFTYPE_P2P_CLIENT:
 		case NL80211_IFTYPE_ADHOC:
 			dev->priv_flags |= IFF_DONT_BRIDGE;
 			break;
+		case NL80211_IFTYPE_P2P_GO:
 		case NL80211_IFTYPE_AP:
 		case NL80211_IFTYPE_AP_VLAN:
 		case NL80211_IFTYPE_WDS:
-- 
cgit v1.2.3


From cd13539b8bc9ae884e6d8d9374c594adff4304e4 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 16 Sep 2010 02:58:13 +0000
Subject: net: shrinks struct net_device

commit ab95bfe01 (net: replace hooks in __netif_receive_skb) added
rx_handler at wrong place, between two cache line aligned objects,
creating a big hole (a full cache line)

Move rx_handler and rx_handler_data before rx_queue, filling existing
hole.

Move master field in the cache line(s) used in receive path.

This saves 64 bytes (or L1_CACHE_BYTES), and avoids two possible
cache misses in receive path.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 8992fffb8104..ec17887a5bca 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -918,10 +918,6 @@ struct net_device {
 	unsigned short		needed_headroom;
 	unsigned short		needed_tailroom;
 
-	struct net_device	*master; /* Pointer to master device of a group,
-					  * which this device is member of.
-					  */
-
 	/* Interface address info. */
 	unsigned char		perm_addr[MAX_ADDR_LEN]; /* permanent hw address */
 	unsigned char		addr_assign_type; /* hw address assignment type */
@@ -951,7 +947,7 @@ struct net_device {
 						   assign before registering */
 
 /*
- * Cache line mostly used on receive path (including eth_type_trans())
+ * Cache lines mostly used on receive path (including eth_type_trans())
  */
 	unsigned long		last_rx;	/* Time of last Rx
 						 * This should not be set in
@@ -961,6 +957,10 @@ struct net_device {
 						 * avoid dirtying this cache line.
 						 */
 
+	struct net_device	*master; /* Pointer to master device of a group,
+					  * which this device is member of.
+					  */
+
 	/* Interface address info used in eth_type_trans() */
 	unsigned char		*dev_addr;	/* hw address, (before bcast
 						   because most packets are
@@ -980,10 +980,14 @@ struct net_device {
 	unsigned int		num_rx_queues;
 #endif
 
-	struct netdev_queue	rx_queue;
 	rx_handler_func_t	*rx_handler;
 	void			*rx_handler_data;
 
+	struct netdev_queue	rx_queue; /* use two cache lines */
+
+/*
+ * Cache lines mostly used on transmit path
+ */
 	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
 
 	/* Number of TX queues allocated at alloc_netdev_mq() time  */
@@ -997,9 +1001,7 @@ struct net_device {
 
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 	spinlock_t		tx_global_lock;
-/*
- * One part is mostly used on xmit path (device)
- */
+
 	/* These may be needed for future network-power-down code. */
 
 	/*
-- 
cgit v1.2.3


From b04243ef7006cda301819f54ee7ce0a3632489e3 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 17 Sep 2010 11:28:48 +0200
Subject: perf: Complete software pmu grouping

Aside from allowing software events into a !software group,
allow adding !software events to pure software groups.

Once we've moved the software group and attached the first
!software event, the group will no longer be a pure software
group and hence no longer be eligible for movement, at which
point the straight ctx comparison is correct again.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <20100917093009.410784731@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |  6 +++++
 kernel/perf_event.c        | 65 ++++++++++++++++++++++++++++++++++++++++++----
 2 files changed, 66 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 39d8860b2684..165287fd2cc4 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -804,12 +804,18 @@ struct perf_event {
 #endif /* CONFIG_PERF_EVENTS */
 };
 
+enum perf_event_context_type {
+	task_context,
+	cpu_context,
+};
+
 /**
  * struct perf_event_context - event context structure
  *
  * Used as a container for task events and CPU events as well:
  */
 struct perf_event_context {
+	enum perf_event_context_type	type;
 	struct pmu			*pmu;
 	/*
 	 * Protect the states of the events in the list,
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index ce95617f5d2c..6d7eef5f3c41 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -5184,6 +5184,7 @@ int perf_pmu_register(struct pmu *pmu)
 
 		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
 		__perf_event_init_context(&cpuctx->ctx);
+		cpuctx->ctx.type = cpu_context;
 		cpuctx->ctx.pmu = pmu;
 		cpuctx->timer_interval = TICK_NSEC;
 		hrtimer_init(&cpuctx->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
@@ -5517,7 +5518,8 @@ SYSCALL_DEFINE5(perf_event_open,
 		struct perf_event_attr __user *, attr_uptr,
 		pid_t, pid, int, cpu, int, group_fd, unsigned long, flags)
 {
-	struct perf_event *event, *group_leader = NULL, *output_event = NULL;
+	struct perf_event *group_leader = NULL, *output_event = NULL;
+	struct perf_event *event, *sibling;
 	struct perf_event_attr attr;
 	struct perf_event_context *ctx;
 	struct file *event_file = NULL;
@@ -5525,6 +5527,7 @@ SYSCALL_DEFINE5(perf_event_open,
 	struct task_struct *task = NULL;
 	struct pmu *pmu;
 	int event_fd;
+	int move_group = 0;
 	int fput_needed = 0;
 	int err;
 
@@ -5574,8 +5577,29 @@ SYSCALL_DEFINE5(perf_event_open,
 	 * any hardware group.
 	 */
 	pmu = event->pmu;
-	if ((pmu->task_ctx_nr == perf_sw_context) && group_leader)
-		pmu = group_leader->pmu;
+
+	if (group_leader &&
+	    (is_software_event(event) != is_software_event(group_leader))) {
+		if (is_software_event(event)) {
+			/*
+			 * If event and group_leader are not both a software
+			 * event, and event is, then group leader is not.
+			 *
+			 * Allow the addition of software events to !software
+			 * groups, this is safe because software events never
+			 * fail to schedule.
+			 */
+			pmu = group_leader->pmu;
+		} else if (is_software_event(group_leader) &&
+			   (group_leader->group_flags & PERF_GROUP_SOFTWARE)) {
+			/*
+			 * In case the group is a pure software group, and we
+			 * try to add a hardware event, move the whole group to
+			 * the hardware context.
+			 */
+			move_group = 1;
+		}
+	}
 
 	if (pid != -1)
 		task = find_lively_task_by_vpid(pid);
@@ -5605,8 +5629,14 @@ SYSCALL_DEFINE5(perf_event_open,
 		 * Do not allow to attach to a group in a different
 		 * task or CPU context:
 		 */
-		if (group_leader->ctx != ctx)
-			goto err_context;
+		if (move_group) {
+			if (group_leader->ctx->type != ctx->type)
+				goto err_context;
+		} else {
+			if (group_leader->ctx != ctx)
+				goto err_context;
+		}
+
 		/*
 		 * Only a group leader can be exclusive or pinned
 		 */
@@ -5626,9 +5656,34 @@ SYSCALL_DEFINE5(perf_event_open,
 		goto err_context;
 	}
 
+	if (move_group) {
+		struct perf_event_context *gctx = group_leader->ctx;
+
+		mutex_lock(&gctx->mutex);
+		perf_event_remove_from_context(group_leader);
+		list_for_each_entry(sibling, &group_leader->sibling_list,
+				    group_entry) {
+			perf_event_remove_from_context(sibling);
+			put_ctx(gctx);
+		}
+		mutex_unlock(&gctx->mutex);
+		put_ctx(gctx);
+	}
+
 	event->filp = event_file;
 	WARN_ON_ONCE(ctx->parent_ctx);
 	mutex_lock(&ctx->mutex);
+
+	if (move_group) {
+		perf_install_in_context(ctx, group_leader, cpu);
+		get_ctx(ctx);
+		list_for_each_entry(sibling, &group_leader->sibling_list,
+				    group_entry) {
+			perf_install_in_context(ctx, sibling, cpu);
+			get_ctx(ctx);
+		}
+	}
+
 	perf_install_in_context(ctx, event, cpu);
 	++ctx->generation;
 	mutex_unlock(&ctx->mutex);
-- 
cgit v1.2.3


From e9d2b064149ff7ef4acbc65a1b9374ac8b218d3e Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 17 Sep 2010 11:28:50 +0200
Subject: perf: Undo the per cpu-context timer stuff

Revert the timer per cpu-context timers because of unfortunate
nohz interaction. Fixing that would have been somewhat ugly, so
go back to driving things from the regular tick. Provide a
jiffies interval feature for people who want slower rotations.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20100917093009.519845633@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |  6 ++--
 kernel/perf_event.c        | 79 ++++++++++++++++++++++++++++------------------
 kernel/sched.c             |  2 ++
 3 files changed, 55 insertions(+), 32 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 165287fd2cc4..61b1e2d760fd 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -870,8 +870,8 @@ struct perf_cpu_context {
 	struct perf_event_context	*task_ctx;
 	int				active_oncpu;
 	int				exclusive;
-	u64				timer_interval;
-	struct hrtimer			timer;
+	struct list_head		rotation_list;
+	int				jiffies_interval;
 };
 
 struct perf_output_handle {
@@ -1065,6 +1065,7 @@ extern int perf_swevent_get_recursion_context(void);
 extern void perf_swevent_put_recursion_context(int rctx);
 extern void perf_event_enable(struct perf_event *event);
 extern void perf_event_disable(struct perf_event *event);
+extern void perf_event_task_tick(void);
 #else
 static inline void
 perf_event_task_sched_in(struct task_struct *task)			{ }
@@ -1099,6 +1100,7 @@ static inline int  perf_swevent_get_recursion_context(void)		{ return -1; }
 static inline void perf_swevent_put_recursion_context(int rctx)		{ }
 static inline void perf_event_enable(struct perf_event *event)		{ }
 static inline void perf_event_disable(struct perf_event *event)		{ }
+static inline void perf_event_task_tick(void)				{ }
 #endif
 
 #define perf_output_put(handle, x) \
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 27332e5f51a7..baae1367e945 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -77,23 +77,22 @@ void perf_pmu_enable(struct pmu *pmu)
 		pmu->pmu_enable(pmu);
 }
 
+static DEFINE_PER_CPU(struct list_head, rotation_list);
+
+/*
+ * perf_pmu_rotate_start() and perf_rotate_context() are fully serialized
+ * because they're strictly cpu affine and rotate_start is called with IRQs
+ * disabled, while rotate_context is called from IRQ context.
+ */
 static void perf_pmu_rotate_start(struct pmu *pmu)
 {
 	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+	struct list_head *head = &__get_cpu_var(rotation_list);
 
-	if (hrtimer_active(&cpuctx->timer))
-		return;
+	WARN_ON(!irqs_disabled());
 
-	__hrtimer_start_range_ns(&cpuctx->timer,
-			ns_to_ktime(cpuctx->timer_interval), 0,
-			HRTIMER_MODE_REL_PINNED, 0);
-}
-
-static void perf_pmu_rotate_stop(struct pmu *pmu)
-{
-	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
-
-	hrtimer_cancel(&cpuctx->timer);
+	if (list_empty(&cpuctx->rotation_list))
+		list_add(&cpuctx->rotation_list, head);
 }
 
 static void get_ctx(struct perf_event_context *ctx)
@@ -1607,36 +1606,33 @@ static void rotate_ctx(struct perf_event_context *ctx)
 }
 
 /*
- * Cannot race with ->pmu_rotate_start() because this is ran from hardirq
- * context, and ->pmu_rotate_start() is called with irqs disabled (both are
- * cpu affine, so there are no SMP races).
+ * perf_pmu_rotate_start() and perf_rotate_context() are fully serialized
+ * because they're strictly cpu affine and rotate_start is called with IRQs
+ * disabled, while rotate_context is called from IRQ context.
  */
-static enum hrtimer_restart perf_event_context_tick(struct hrtimer *timer)
+static void perf_rotate_context(struct perf_cpu_context *cpuctx)
 {
-	enum hrtimer_restart restart = HRTIMER_NORESTART;
-	struct perf_cpu_context *cpuctx;
+	u64 interval = (u64)cpuctx->jiffies_interval * TICK_NSEC;
 	struct perf_event_context *ctx = NULL;
-	int rotate = 0;
-
-	cpuctx = container_of(timer, struct perf_cpu_context, timer);
+	int rotate = 0, remove = 1;
 
 	if (cpuctx->ctx.nr_events) {
-		restart = HRTIMER_RESTART;
+		remove = 0;
 		if (cpuctx->ctx.nr_events != cpuctx->ctx.nr_active)
 			rotate = 1;
 	}
 
 	ctx = cpuctx->task_ctx;
 	if (ctx && ctx->nr_events) {
-		restart = HRTIMER_RESTART;
+		remove = 0;
 		if (ctx->nr_events != ctx->nr_active)
 			rotate = 1;
 	}
 
 	perf_pmu_disable(cpuctx->ctx.pmu);
-	perf_ctx_adjust_freq(&cpuctx->ctx, cpuctx->timer_interval);
+	perf_ctx_adjust_freq(&cpuctx->ctx, interval);
 	if (ctx)
-		perf_ctx_adjust_freq(ctx, cpuctx->timer_interval);
+		perf_ctx_adjust_freq(ctx, interval);
 
 	if (!rotate)
 		goto done;
@@ -1654,10 +1650,24 @@ static enum hrtimer_restart perf_event_context_tick(struct hrtimer *timer)
 		task_ctx_sched_in(ctx, EVENT_FLEXIBLE);
 
 done:
+	if (remove)
+		list_del_init(&cpuctx->rotation_list);
+
 	perf_pmu_enable(cpuctx->ctx.pmu);
-	hrtimer_forward_now(timer, ns_to_ktime(cpuctx->timer_interval));
+}
+
+void perf_event_task_tick(void)
+{
+	struct list_head *head = &__get_cpu_var(rotation_list);
+	struct perf_cpu_context *cpuctx, *tmp;
 
-	return restart;
+	WARN_ON(!irqs_disabled());
+
+	list_for_each_entry_safe(cpuctx, tmp, head, rotation_list) {
+		if (cpuctx->jiffies_interval == 1 ||
+				!(jiffies % cpuctx->jiffies_interval))
+			perf_rotate_context(cpuctx);
+	}
 }
 
 static int event_enable_on_exec(struct perf_event *event,
@@ -5186,9 +5196,8 @@ int perf_pmu_register(struct pmu *pmu)
 		__perf_event_init_context(&cpuctx->ctx);
 		cpuctx->ctx.type = cpu_context;
 		cpuctx->ctx.pmu = pmu;
-		cpuctx->timer_interval = TICK_NSEC;
-		hrtimer_init(&cpuctx->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-		cpuctx->timer.function = perf_event_context_tick;
+		cpuctx->jiffies_interval = 1;
+		INIT_LIST_HEAD(&cpuctx->rotation_list);
 	}
 
 got_cpu_context:
@@ -6229,6 +6238,7 @@ static void __init perf_event_init_all_cpus(void)
 	for_each_possible_cpu(cpu) {
 		swhash = &per_cpu(swevent_htable, cpu);
 		mutex_init(&swhash->hlist_mutex);
+		INIT_LIST_HEAD(&per_cpu(rotation_list, cpu));
 	}
 }
 
@@ -6248,6 +6258,15 @@ static void __cpuinit perf_event_init_cpu(int cpu)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
+static void perf_pmu_rotate_stop(struct pmu *pmu)
+{
+	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+
+	WARN_ON(!irqs_disabled());
+
+	list_del_init(&cpuctx->rotation_list);
+}
+
 static void __perf_event_exit_context(void *__info)
 {
 	struct perf_event_context *ctx = __info;
diff --git a/kernel/sched.c b/kernel/sched.c
index 1c3ea7a55b7b..794819eab9ca 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3584,6 +3584,8 @@ void scheduler_tick(void)
 	curr->sched_class->task_tick(rq, curr, 0);
 	raw_spin_unlock(&rq->lock);
 
+	perf_event_task_tick();
+
 #ifdef CONFIG_SMP
 	rq->idle_at_tick = idle_cpu(cpu);
 	trigger_load_balance(rq, cpu);
-- 
cgit v1.2.3


From 3575792e005dc9994f15ae72c1c6f401d134177d Mon Sep 17 00:00:00 2001
From: Julian Anastasov <ja@ssi.bg>
Date: Fri, 17 Sep 2010 14:18:16 +0200
Subject: ipvs: extend connection flags to 32 bits

- the sync protocol supports 16 bits only, so bits 0..15 should be
used only for flags that should go to backup server, bits 16 and
above should be allocated for flags not sent to backup.

- use IP_VS_CONN_F_DEST_MASK as mask of connection flags in
destination that can be changed by user space

- allow IP_VS_CONN_F_ONE_PACKET to be set in destination

Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/ip_vs.h           |  8 ++++++++
 include/net/ip_vs.h             |  2 +-
 net/netfilter/ipvs/ip_vs_conn.c | 16 ++++++++++------
 net/netfilter/ipvs/ip_vs_core.c | 11 ++++++-----
 net/netfilter/ipvs/ip_vs_ctl.c  |  5 +++--
 5 files changed, 28 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ip_vs.h b/include/linux/ip_vs.h
index 9708de265bb1..003d75f6ffe1 100644
--- a/include/linux/ip_vs.h
+++ b/include/linux/ip_vs.h
@@ -70,6 +70,7 @@
 
 /*
  *      IPVS Connection Flags
+ *      Only flags 0..15 are sent to backup server
  */
 #define IP_VS_CONN_F_FWD_MASK	0x0007		/* mask for the fwd methods */
 #define IP_VS_CONN_F_MASQ	0x0000		/* masquerading/NAT */
@@ -88,6 +89,13 @@
 #define IP_VS_CONN_F_TEMPLATE	0x1000		/* template, not connection */
 #define IP_VS_CONN_F_ONE_PACKET	0x2000		/* forward only one packet */
 
+/* Flags that are not sent to backup server start from bit 16 */
+
+/* Connection flags from destination that can be changed by user space */
+#define IP_VS_CONN_F_DEST_MASK (IP_VS_CONN_F_FWD_MASK | \
+				IP_VS_CONN_F_ONE_PACKET | \
+				0)
+
 #define IP_VS_SCHEDNAME_MAXLEN	16
 #define IP_VS_IFNAME_MAXLEN	16
 
diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index f976885f686f..62698a9c1631 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -366,6 +366,7 @@ struct ip_vs_conn {
 	union nf_inet_addr       caddr;          /* client address */
 	union nf_inet_addr       vaddr;          /* virtual address */
 	union nf_inet_addr       daddr;          /* destination address */
+	volatile __u32           flags;          /* status flags */
 	__be16                   cport;
 	__be16                   vport;
 	__be16                   dport;
@@ -378,7 +379,6 @@ struct ip_vs_conn {
 
 	/* Flags and state transition */
 	spinlock_t              lock;           /* lock for state transition */
-	volatile __u16          flags;          /* status flags */
 	volatile __u16          state;          /* state info */
 	volatile __u16          old_state;      /* old state, to be used for
 						 * state transition triggerd
diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c
index b71c69a2db13..9fe1da7bcf16 100644
--- a/net/netfilter/ipvs/ip_vs_conn.c
+++ b/net/netfilter/ipvs/ip_vs_conn.c
@@ -505,6 +505,8 @@ static inline int ip_vs_dest_totalconns(struct ip_vs_dest *dest)
 static inline void
 ip_vs_bind_dest(struct ip_vs_conn *cp, struct ip_vs_dest *dest)
 {
+	unsigned int conn_flags;
+
 	/* if dest is NULL, then return directly */
 	if (!dest)
 		return;
@@ -512,16 +514,18 @@ ip_vs_bind_dest(struct ip_vs_conn *cp, struct ip_vs_dest *dest)
 	/* Increase the refcnt counter of the dest */
 	atomic_inc(&dest->refcnt);
 
+	conn_flags = atomic_read(&dest->conn_flags);
+	if (cp->protocol != IPPROTO_UDP)
+		conn_flags &= ~IP_VS_CONN_F_ONE_PACKET;
 	/* Bind with the destination and its corresponding transmitter */
-	if ((cp->flags & IP_VS_CONN_F_SYNC) &&
-	    (!(cp->flags & IP_VS_CONN_F_TEMPLATE)))
+	if (cp->flags & IP_VS_CONN_F_SYNC) {
 		/* if the connection is not template and is created
 		 * by sync, preserve the activity flag.
 		 */
-		cp->flags |= atomic_read(&dest->conn_flags) &
-			     (~IP_VS_CONN_F_INACTIVE);
-	else
-		cp->flags |= atomic_read(&dest->conn_flags);
+		if (!(cp->flags & IP_VS_CONN_F_TEMPLATE))
+			conn_flags &= ~IP_VS_CONN_F_INACTIVE;
+	}
+	cp->flags |= conn_flags;
 	cp->dest = dest;
 
 	IP_VS_DBG_BUF(7, "Bind-dest %s c:%s:%d v:%s:%d "
diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index 0c043b6ce65e..319991d4d251 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -194,7 +194,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 	struct ip_vs_dest *dest;
 	struct ip_vs_conn *ct;
 	__be16  dport;			/* destination port to forward */
-	__be16  flags;
+	unsigned int flags;
 	union nf_inet_addr snet;	/* source network of the client,
 					   after masking */
 
@@ -382,7 +382,8 @@ ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
 	struct ip_vs_conn *cp = NULL;
 	struct ip_vs_iphdr iph;
 	struct ip_vs_dest *dest;
-	__be16 _ports[2], *pptr, flags;
+	__be16 _ports[2], *pptr;
+	unsigned int flags;
 
 	ip_vs_fill_iphdr(svc->af, skb_network_header(skb), &iph);
 	pptr = skb_header_pointer(skb, iph.len, sizeof(_ports), _ports);
@@ -473,9 +474,9 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 	if (sysctl_ip_vs_cache_bypass && svc->fwmark && unicast) {
 		int ret, cs;
 		struct ip_vs_conn *cp;
-		__u16 flags = (svc->flags & IP_VS_SVC_F_ONEPACKET &&
-				iph.protocol == IPPROTO_UDP)?
-				IP_VS_CONN_F_ONE_PACKET : 0;
+		unsigned int flags = (svc->flags & IP_VS_SVC_F_ONEPACKET &&
+				      iph.protocol == IPPROTO_UDP)?
+				      IP_VS_CONN_F_ONE_PACKET : 0;
 		union nf_inet_addr daddr =  { .all = { 0, 0, 0, 0 } };
 
 		ip_vs_service_put(svc);
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index ca8ec8c4f311..7bd41d28080c 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -765,7 +765,8 @@ __ip_vs_update_dest(struct ip_vs_service *svc,
 
 	/* set the weight and the flags */
 	atomic_set(&dest->weight, udest->weight);
-	conn_flags = udest->conn_flags | IP_VS_CONN_F_INACTIVE;
+	conn_flags = udest->conn_flags & IP_VS_CONN_F_DEST_MASK;
+	conn_flags |= IP_VS_CONN_F_INACTIVE;
 
 	/* check if local node and update the flags */
 #ifdef CONFIG_IP_VS_IPV6
@@ -782,7 +783,7 @@ __ip_vs_update_dest(struct ip_vs_service *svc,
 		}
 
 	/* set the IP_VS_CONN_F_NOOUTPUT flag if not masquerading/NAT */
-	if ((conn_flags & IP_VS_CONN_F_FWD_MASK) != 0) {
+	if ((conn_flags & IP_VS_CONN_F_FWD_MASK) != IP_VS_CONN_F_MASQ) {
 		conn_flags |= IP_VS_CONN_F_NOOUTPUT;
 	} else {
 		/*
-- 
cgit v1.2.3


From dfb8fb96ae2b5126cd0c08c0ccd7c42e1f46568a Mon Sep 17 00:00:00 2001
From: Giuseppe CAVALLARO <peppe.cavallaro@st.com>
Date: Fri, 17 Sep 2010 03:23:39 +0000
Subject: stmmac: add CSR Clock range selection

This patch adds the CSR Clock range selection.

Original patch from Johannes Stezenbach fixed the CSR
in the stmmac_mdio. We agreed to provide this through
the platform instead of.
Also thanks to Johannes for having tested it on ARM.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Johannes Stezenbach <js@sig21.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/stmmac/stmmac.h      | 1 +
 drivers/net/stmmac/stmmac_main.c | 1 +
 drivers/net/stmmac/stmmac_mdio.c | 5 +++--
 include/linux/stmmac.h           | 1 +
 4 files changed, 6 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/stmmac/stmmac.h b/drivers/net/stmmac/stmmac.h
index d0ddab0d21c2..12d1cb00c0d7 100644
--- a/drivers/net/stmmac/stmmac.h
+++ b/drivers/net/stmmac/stmmac.h
@@ -78,6 +78,7 @@ struct stmmac_priv {
 	unsigned int flow_ctrl;
 	unsigned int pause;
 	struct mii_bus *mii;
+	int mii_clk_csr;
 
 	u32 msg_enable;
 	spinlock_t lock;
diff --git a/drivers/net/stmmac/stmmac_main.c b/drivers/net/stmmac/stmmac_main.c
index 03c160c6d75c..a169b1441d50 100644
--- a/drivers/net/stmmac/stmmac_main.c
+++ b/drivers/net/stmmac/stmmac_main.c
@@ -1704,6 +1704,7 @@ static int stmmac_dvr_probe(struct platform_device *pdev)
 	plat_dat = pdev->dev.platform_data;
 	priv->bus_id = plat_dat->bus_id;
 	priv->pbl = plat_dat->pbl;	/* TLI */
+	priv->mii_clk_csr = plat_dat->clk_csr;
 	priv->is_gmac = plat_dat->has_gmac;	/* GMAC is on board */
 	priv->enh_desc = plat_dat->enh_desc;
 	priv->ioaddr = addr;
diff --git a/drivers/net/stmmac/stmmac_mdio.c b/drivers/net/stmmac/stmmac_mdio.c
index 03dea1401571..d7441616357d 100644
--- a/drivers/net/stmmac/stmmac_mdio.c
+++ b/drivers/net/stmmac/stmmac_mdio.c
@@ -53,7 +53,7 @@ static int stmmac_mdio_read(struct mii_bus *bus, int phyaddr, int phyreg)
 	int data;
 	u16 regValue = (((phyaddr << 11) & (0x0000F800)) |
 			((phyreg << 6) & (0x000007C0)));
-	regValue |= MII_BUSY;	/* in case of GMAC */
+	regValue |= MII_BUSY | ((priv->mii_clk_csr & 7) << 2);
 
 	do {} while (((readl(priv->ioaddr + mii_address)) & MII_BUSY) == 1);
 	writel(regValue, priv->ioaddr + mii_address);
@@ -85,7 +85,8 @@ static int stmmac_mdio_write(struct mii_bus *bus, int phyaddr, int phyreg,
 	    (((phyaddr << 11) & (0x0000F800)) | ((phyreg << 6) & (0x000007C0)))
 	    | MII_WRITE;
 
-	value |= MII_BUSY;
+	value |= MII_BUSY | ((priv->mii_clk_csr & 7) << 2);
+
 
 	/* Wait until any existing MII operation is complete */
 	do {} while (((readl(priv->ioaddr + mii_address)) & MII_BUSY) == 1);
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index a4adf0de6ed6..c87c88ccffc0 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -32,6 +32,7 @@
 struct plat_stmmacenet_data {
 	int bus_id;
 	int pbl;
+	int clk_csr;
 	int has_gmac;
 	int enh_desc;
 	void (*fix_mac_speed)(void *priv, unsigned int speed);
-- 
cgit v1.2.3


From ebbb293f8b3021ae2009fcb7cb3b8a52fb5fd06a Mon Sep 17 00:00:00 2001
From: Giuseppe CAVALLARO <peppe.cavallaro@st.com>
Date: Fri, 17 Sep 2010 03:23:40 +0000
Subject: stmmac: consolidate and tidy-up the COE support

The first version of the driver had hard-coded the logic
for handling the checksum offloading.
This was designed according to the chips included in
the STM platforms where:
o MAC10/100 supports no COE at all.
o GMAC fully supports RX/TX COE.

This is not good for other chip configurations where,
for example, the mac10/100 supports the tx csum in HW
or when the GMAC has no IPC.

Thanks to Johannes Stezenbach; he provided me a first
draft of this patch that only reviewed the IPC for the
GMAC devices.

This patch also helps on SPEAr platforms where the
MAC10/100 can perform the TX csum in HW.
Thanks to Deepak SIKRI for his support on this.

In the end, GMAC devices for STM platforms have
a bugged Jumbo frame support that needs to have
the Tx COE disabled for oversized frames (due to
limited buffer sizes). This information is also
passed through the driver's platform structure.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: Johannes Stezenbach <js@sig21.net>
Signed-off-by: Deepak SIKRI <deepak.sikri@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/stmmac/common.h         |  4 +--
 drivers/net/stmmac/dwmac1000.h      |  2 +-
 drivers/net/stmmac/dwmac1000_core.c | 13 ++++++++
 drivers/net/stmmac/dwmac100_core.c  |  6 ++++
 drivers/net/stmmac/stmmac.h         |  4 ++-
 drivers/net/stmmac/stmmac_ethtool.c |  2 +-
 drivers/net/stmmac/stmmac_main.c    | 66 ++++++++++++++++++-------------------
 include/linux/stmmac.h              |  2 ++
 8 files changed, 60 insertions(+), 39 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/stmmac/common.h b/drivers/net/stmmac/common.h
index e8cbcb5c206e..673ef86a063f 100644
--- a/drivers/net/stmmac/common.h
+++ b/drivers/net/stmmac/common.h
@@ -102,8 +102,6 @@ struct stmmac_extra_stats {
 
 #define SF_DMA_MODE 1 /* DMA STORE-AND-FORWARD Operation Mode */
 
-#define HW_CSUM 1
-#define NO_HW_CSUM 0
 enum rx_frame_status { /* IPC status */
 	good_frame = 0,
 	discard_frame = 1,
@@ -205,6 +203,8 @@ struct stmmac_dma_ops {
 struct stmmac_ops {
 	/* MAC core initialization */
 	void (*core_init) (void __iomem *ioaddr) ____cacheline_aligned;
+	/* Support checksum offload engine */
+	int  (*rx_coe) (void __iomem *ioaddr);
 	/* Dump MAC registers */
 	void (*dump_regs) (void __iomem *ioaddr);
 	/* Handle extra events on specific interrupts hw dependent */
diff --git a/drivers/net/stmmac/dwmac1000.h b/drivers/net/stmmac/dwmac1000.h
index 8b20b19971cb..81ee4fd04386 100644
--- a/drivers/net/stmmac/dwmac1000.h
+++ b/drivers/net/stmmac/dwmac1000.h
@@ -99,7 +99,7 @@ enum inter_frame_gap {
 #define GMAC_CONTROL_RE		0x00000004 /* Receiver Enable */
 
 #define GMAC_CORE_INIT (GMAC_CONTROL_JD | GMAC_CONTROL_PS | GMAC_CONTROL_ACS | \
-			GMAC_CONTROL_IPC | GMAC_CONTROL_JE | GMAC_CONTROL_BE)
+			GMAC_CONTROL_JE | GMAC_CONTROL_BE)
 
 /* GMAC Frame Filter defines */
 #define GMAC_FRAME_FILTER_PR	0x00000001	/* Promiscuous Mode */
diff --git a/drivers/net/stmmac/dwmac1000_core.c b/drivers/net/stmmac/dwmac1000_core.c
index f1f426146f40..c18c85993179 100644
--- a/drivers/net/stmmac/dwmac1000_core.c
+++ b/drivers/net/stmmac/dwmac1000_core.c
@@ -50,6 +50,18 @@ static void dwmac1000_core_init(void __iomem *ioaddr)
 #endif
 }
 
+static int dwmac1000_rx_coe_supported(void __iomem *ioaddr)
+{
+	u32 value = readl(ioaddr + GMAC_CONTROL);
+
+	value |= GMAC_CONTROL_IPC;
+	writel(value, ioaddr + GMAC_CONTROL);
+
+	value = readl(ioaddr + GMAC_CONTROL);
+
+	return !!(value & GMAC_CONTROL_IPC);
+}
+
 static void dwmac1000_dump_regs(void __iomem *ioaddr)
 {
 	int i;
@@ -202,6 +214,7 @@ static void dwmac1000_irq_status(void __iomem *ioaddr)
 
 struct stmmac_ops dwmac1000_ops = {
 	.core_init = dwmac1000_core_init,
+	.rx_coe = dwmac1000_rx_coe_supported,
 	.dump_regs = dwmac1000_dump_regs,
 	.host_irq_status = dwmac1000_irq_status,
 	.set_filter = dwmac1000_set_filter,
diff --git a/drivers/net/stmmac/dwmac100_core.c b/drivers/net/stmmac/dwmac100_core.c
index db06c04ce480..58a914b27003 100644
--- a/drivers/net/stmmac/dwmac100_core.c
+++ b/drivers/net/stmmac/dwmac100_core.c
@@ -42,6 +42,11 @@ static void dwmac100_core_init(void __iomem *ioaddr)
 #endif
 }
 
+static int dwmac100_rx_coe_supported(void __iomem *ioaddr)
+{
+	return 0;
+}
+
 static void dwmac100_dump_mac_regs(void __iomem *ioaddr)
 {
 	pr_info("\t----------------------------------------------\n"
@@ -165,6 +170,7 @@ static void dwmac100_pmt(void __iomem *ioaddr, unsigned long mode)
 
 struct stmmac_ops dwmac100_ops = {
 	.core_init = dwmac100_core_init,
+	.rx_coe = dwmac100_rx_coe_supported,
 	.dump_regs = dwmac100_dump_mac_regs,
 	.host_irq_status = dwmac100_irq_status,
 	.set_filter = dwmac100_set_filter,
diff --git a/drivers/net/stmmac/stmmac.h b/drivers/net/stmmac/stmmac.h
index 12d1cb00c0d7..92154ff7d702 100644
--- a/drivers/net/stmmac/stmmac.h
+++ b/drivers/net/stmmac/stmmac.h
@@ -51,7 +51,6 @@ struct stmmac_priv {
 	int is_gmac;
 	dma_addr_t dma_rx_phy;
 	unsigned int dma_rx_size;
-	int rx_csum;
 	unsigned int dma_buf_sz;
 	struct device *device;
 	struct mac_device_info *hw;
@@ -92,6 +91,9 @@ struct stmmac_priv {
 	struct vlan_group *vlgrp;
 #endif
 	int enh_desc;
+	int rx_coe;
+	int bugged_jumbo;
+	int no_csum_insertion;
 };
 
 #ifdef CONFIG_STM_DRIVERS
diff --git a/drivers/net/stmmac/stmmac_ethtool.c b/drivers/net/stmmac/stmmac_ethtool.c
index 63b68e61afce..b32c16ae55c6 100644
--- a/drivers/net/stmmac/stmmac_ethtool.c
+++ b/drivers/net/stmmac/stmmac_ethtool.c
@@ -209,7 +209,7 @@ u32 stmmac_ethtool_get_rx_csum(struct net_device *dev)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
 
-	return priv->rx_csum;
+	return priv->rx_coe;
 }
 
 static void
diff --git a/drivers/net/stmmac/stmmac_main.c b/drivers/net/stmmac/stmmac_main.c
index a169b1441d50..a908f7201aae 100644
--- a/drivers/net/stmmac/stmmac_main.c
+++ b/drivers/net/stmmac/stmmac_main.c
@@ -134,13 +134,6 @@ static int buf_sz = DMA_BUFFER_SIZE;
 module_param(buf_sz, int, S_IRUGO | S_IWUSR);
 MODULE_PARM_DESC(buf_sz, "DMA buffer size");
 
-/* In case of Giga ETH, we can enable/disable the COE for the
- * transmit HW checksum computation.
- * Note that, if tx csum is off in HW, SG will be still supported. */
-static int tx_coe = HW_CSUM;
-module_param(tx_coe, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(tx_coe, "GMAC COE type 2 [on/off]");
-
 static const u32 default_msg_level = (NETIF_MSG_DRV | NETIF_MSG_PROBE |
 				      NETIF_MSG_LINK | NETIF_MSG_IFUP |
 				      NETIF_MSG_IFDOWN | NETIF_MSG_TIMER);
@@ -569,29 +562,22 @@ static void free_dma_desc_resources(struct stmmac_priv *priv)
  *  stmmac_dma_operation_mode - HW DMA operation mode
  *  @priv : pointer to the private device structure.
  *  Description: it sets the DMA operation mode: tx/rx DMA thresholds
- *  or Store-And-Forward capability. It also verifies the COE for the
- *  transmission in case of Giga ETH.
+ *  or Store-And-Forward capability.
  */
 static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
 {
-	if (!priv->is_gmac) {
-		/* MAC 10/100 */
-		priv->hw->dma->dma_mode(priv->ioaddr, tc, 0);
-		priv->tx_coe = NO_HW_CSUM;
-	} else {
-		if ((priv->dev->mtu <= ETH_DATA_LEN) && (tx_coe)) {
-			priv->hw->dma->dma_mode(priv->ioaddr,
-						SF_DMA_MODE, SF_DMA_MODE);
-			tc = SF_DMA_MODE;
-			priv->tx_coe = HW_CSUM;
-		} else {
-			/* Checksum computation is performed in software. */
-			priv->hw->dma->dma_mode(priv->ioaddr, tc,
-						SF_DMA_MODE);
-			priv->tx_coe = NO_HW_CSUM;
-		}
-	}
-	tx_coe = priv->tx_coe;
+	if (likely((priv->tx_coe) && (!priv->no_csum_insertion))) {
+		/* In case of GMAC, SF mode has to be enabled
+		 * to perform the TX COE. This depends on:
+		 * 1) TX COE if actually supported
+		 * 2) There is no bugged Jumbo frame support
+		 *    that needs to not insert csum in the TDES.
+		 */
+		priv->hw->dma->dma_mode(priv->ioaddr,
+					SF_DMA_MODE, SF_DMA_MODE);
+		tc = SF_DMA_MODE;
+	} else
+		priv->hw->dma->dma_mode(priv->ioaddr, tc, SF_DMA_MODE);
 }
 
 /**
@@ -858,6 +844,12 @@ static int stmmac_open(struct net_device *dev)
 	/* Initialize the MAC Core */
 	priv->hw->mac->core_init(priv->ioaddr);
 
+	priv->rx_coe = priv->hw->mac->rx_coe(priv->ioaddr);
+	if (priv->rx_coe)
+		pr_info("stmmac: Rx Checksum Offload Engine supported\n");
+	if (priv->tx_coe)
+		pr_info("\tTX Checksum insertion supported\n");
+
 	priv->shutdown = 0;
 
 	/* Initialise the MMC (if present) to disable all interrupts. */
@@ -1066,7 +1058,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
 		return stmmac_sw_tso(priv, skb);
 
 	if (likely((skb->ip_summed == CHECKSUM_PARTIAL))) {
-		if (likely(priv->tx_coe == NO_HW_CSUM))
+		if (unlikely((!priv->tx_coe) || (priv->no_csum_insertion)))
 			skb_checksum_help(skb);
 		else
 			csum_insertion = 1;
@@ -1390,6 +1382,15 @@ static int stmmac_change_mtu(struct net_device *dev, int new_mtu)
 		return -EINVAL;
 	}
 
+	/* Some GMAC devices have a bugged Jumbo frame support that
+	 * needs to have the Tx COE disabled for oversized frames
+	 * (due to limited buffer sizes). In this case we disable
+	 * the TX csum insertionin the TDES and not use SF. */
+	if ((priv->bugged_jumbo) && (priv->dev->mtu > ETH_DATA_LEN))
+		priv->no_csum_insertion = 1;
+	else
+		priv->no_csum_insertion = 0;
+
 	dev->mtu = new_mtu;
 
 	return 0;
@@ -1510,9 +1511,6 @@ static int stmmac_probe(struct net_device *dev)
 #endif
 	priv->msg_enable = netif_msg_init(debug, default_msg_level);
 
-	if (priv->is_gmac)
-		priv->rx_csum = 1;
-
 	if (flow_ctrl)
 		priv->flow_ctrl = FLOW_AUTO;	/* RX/TX pause on */
 
@@ -1662,7 +1660,7 @@ static int stmmac_dvr_probe(struct platform_device *pdev)
 		ret = -ENODEV;
 		goto out;
 	}
-	pr_info("done!\n");
+	pr_info("\tdone!\n");
 
 	if (!request_mem_region(res->start, resource_size(res),
 				pdev->name)) {
@@ -1705,6 +1703,8 @@ static int stmmac_dvr_probe(struct platform_device *pdev)
 	priv->bus_id = plat_dat->bus_id;
 	priv->pbl = plat_dat->pbl;	/* TLI */
 	priv->mii_clk_csr = plat_dat->clk_csr;
+	priv->tx_coe = plat_dat->tx_coe;
+	priv->bugged_jumbo = plat_dat->bugged_jumbo;
 	priv->is_gmac = plat_dat->has_gmac;	/* GMAC is on board */
 	priv->enh_desc = plat_dat->enh_desc;
 	priv->ioaddr = addr;
@@ -1966,8 +1966,6 @@ static int __init stmmac_cmdline_opt(char *str)
 			strict_strtoul(opt + 7, 0, (unsigned long *)&buf_sz);
 		else if (!strncmp(opt, "tc:", 3))
 			strict_strtoul(opt + 3, 0, (unsigned long *)&tc);
-		else if (!strncmp(opt, "tx_coe:", 7))
-			strict_strtoul(opt + 7, 0, (unsigned long *)&tx_coe);
 		else if (!strncmp(opt, "watchdog:", 9))
 			strict_strtoul(opt + 9, 0, (unsigned long *)&watchdog);
 		else if (!strncmp(opt, "flow_ctrl:", 10))
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index c87c88ccffc0..1d8baf719211 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -35,6 +35,8 @@ struct plat_stmmacenet_data {
 	int clk_csr;
 	int has_gmac;
 	int enh_desc;
+	int tx_coe;
+	int bugged_jumbo;
 	void (*fix_mac_speed)(void *priv, unsigned int speed);
 	void (*bus_setup)(void __iomem *ioaddr);
 #ifdef CONFIG_STM_DRIVERS
-- 
cgit v1.2.3


From be2902daee80b655cebd482b5ee91ffc29408121 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu, 16 Sep 2010 11:28:07 +0000
Subject: ethtool, ixgbe: Move RX n-tuple mask fixup to ethtool

The ethtool utility does not set masks for flow parameters that are
not specified, so if both value and mask are 0 then this must be
treated as equivalent to a mask with all bits set.  Currently that is
done in the only driver that implements RX n-tuple filtering, ixgbe.
Move it to the ethtool core.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ixgbe/ixgbe_82599.c | 57 ++++++++++-------------------------------
 include/linux/ethtool.h         |  5 ++--
 net/core/ethtool.c              | 34 ++++++++++++++++++++++++
 3 files changed, 51 insertions(+), 45 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/ixgbe/ixgbe_82599.c b/drivers/net/ixgbe/ixgbe_82599.c
index 3e06a61da921..e80657c75506 100644
--- a/drivers/net/ixgbe/ixgbe_82599.c
+++ b/drivers/net/ixgbe/ixgbe_82599.c
@@ -1910,56 +1910,27 @@ s32 ixgbe_fdir_add_perfect_filter_82599(struct ixgbe_hw *hw,
 	              (dst_port << IXGBE_FDIRPORT_DESTINATION_SHIFT)));
 
 	/*
-	 * Program the relevant mask registers.  If src/dst_port or src/dst_addr
-	 * are zero, then assume a full mask for that field.  Also assume that
-	 * a VLAN of 0 is unspecified, so mask that out as well.  L4type
-	 * cannot be masked out in this implementation.
+	 * Program the relevant mask registers.  L4type cannot be
+	 * masked out in this implementation.
 	 *
 	 * This also assumes IPv4 only.  IPv6 masking isn't supported at this
 	 * point in time.
 	 */
-	if (src_ipv4 == 0)
-		IXGBE_WRITE_REG(hw, IXGBE_FDIRSIP4M, 0xffffffff);
-	else
-		IXGBE_WRITE_REG(hw, IXGBE_FDIRSIP4M, input_masks->src_ip_mask);
-
-	if (dst_ipv4 == 0)
-		IXGBE_WRITE_REG(hw, IXGBE_FDIRDIP4M, 0xffffffff);
-	else
-		IXGBE_WRITE_REG(hw, IXGBE_FDIRDIP4M, input_masks->dst_ip_mask);
+	IXGBE_WRITE_REG(hw, IXGBE_FDIRSIP4M, input_masks->src_ip_mask);
+	IXGBE_WRITE_REG(hw, IXGBE_FDIRDIP4M, input_masks->dst_ip_mask);
 
 	switch (l4type & IXGBE_ATR_L4TYPE_MASK) {
 	case IXGBE_ATR_L4TYPE_TCP:
-		if (src_port == 0)
-			IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM, 0xffff);
-		else
-			IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM,
-			                input_masks->src_port_mask);
-
-		if (dst_port == 0)
-			IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM,
-			               (IXGBE_READ_REG(hw, IXGBE_FDIRTCPM) |
-			                (0xffff << 16)));
-		else
-			IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM,
-			               (IXGBE_READ_REG(hw, IXGBE_FDIRTCPM) |
-			                (input_masks->dst_port_mask << 16)));
+		IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM, input_masks->src_port_mask);
+		IXGBE_WRITE_REG(hw, IXGBE_FDIRTCPM,
+				(IXGBE_READ_REG(hw, IXGBE_FDIRTCPM) |
+				 (input_masks->dst_port_mask << 16)));
 		break;
 	case IXGBE_ATR_L4TYPE_UDP:
-		if (src_port == 0)
-			IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM, 0xffff);
-		else
-			IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM,
-			                input_masks->src_port_mask);
-
-		if (dst_port == 0)
-			IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM,
-			               (IXGBE_READ_REG(hw, IXGBE_FDIRUDPM) |
-			                (0xffff << 16)));
-		else
-			IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM,
-			               (IXGBE_READ_REG(hw, IXGBE_FDIRUDPM) |
-			                (input_masks->src_port_mask << 16)));
+		IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM, input_masks->src_port_mask);
+		IXGBE_WRITE_REG(hw, IXGBE_FDIRUDPM,
+				(IXGBE_READ_REG(hw, IXGBE_FDIRUDPM) |
+				 (input_masks->src_port_mask << 16)));
 		break;
 	default:
 		/* this already would have failed above */
@@ -1967,11 +1938,11 @@ s32 ixgbe_fdir_add_perfect_filter_82599(struct ixgbe_hw *hw,
 	}
 
 	/* Program the last mask register, FDIRM */
-	if (input_masks->vlan_id_mask || !vlan_id)
+	if (input_masks->vlan_id_mask)
 		/* Mask both VLAN and VLANP - bits 0 and 1 */
 		fdirm |= 0x3;
 
-	if (input_masks->data_mask || !flex_bytes)
+	if (input_masks->data_mask)
 		/* Flex bytes need masking, so mask the whole thing - bit 4 */
 		fdirm |= 0x10;
 
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index d64e246a39e7..00334eebbe26 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -470,8 +470,9 @@ struct ethtool_rxfh_indir {
  * @action: RX ring/queue index to deliver to (non-negative) or other action
  *	(negative, e.g. %ETHTOOL_RXNTUPLE_ACTION_DROP)
  *
- * Zero values in @h_u may be ignored, as if all the corresponding
- * mask bits were set.
+ * For flow types %TCP_V4_FLOW, %UDP_V4_FLOW and %SCTP_V4_FLOW, where
+ * a field value and mask are both zero this is treated as if all mask
+ * bits are set i.e. the field is ignored.
  */
 struct ethtool_rx_ntuple_flow_spec {
 	__u32		 flow_type;
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 248c25c3e820..91ffce20c36b 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -485,6 +485,38 @@ static void __rx_ntuple_filter_add(struct ethtool_rx_ntuple_list *list,
 	list->count++;
 }
 
+/*
+ * ethtool does not (or did not) set masks for flow parameters that are
+ * not specified, so if both value and mask are 0 then this must be
+ * treated as equivalent to a mask with all bits set.  Implement that
+ * here rather than in drivers.
+ */
+static void rx_ntuple_fix_masks(struct ethtool_rx_ntuple_flow_spec *fs)
+{
+	struct ethtool_tcpip4_spec *entry = &fs->h_u.tcp_ip4_spec;
+	struct ethtool_tcpip4_spec *mask = &fs->m_u.tcp_ip4_spec;
+
+	if (fs->flow_type != TCP_V4_FLOW &&
+	    fs->flow_type != UDP_V4_FLOW &&
+	    fs->flow_type != SCTP_V4_FLOW)
+		return;
+
+	if (!(entry->ip4src | mask->ip4src))
+		mask->ip4src = htonl(0xffffffff);
+	if (!(entry->ip4dst | mask->ip4dst))
+		mask->ip4dst = htonl(0xffffffff);
+	if (!(entry->psrc | mask->psrc))
+		mask->psrc = htons(0xffff);
+	if (!(entry->pdst | mask->pdst))
+		mask->pdst = htons(0xffff);
+	if (!(entry->tos | mask->tos))
+		mask->tos = 0xff;
+	if (!(fs->vlan_tag | fs->vlan_tag_mask))
+		fs->vlan_tag_mask = 0xffff;
+	if (!(fs->data | fs->data_mask))
+		fs->data_mask = 0xffffffffffffffffULL;
+}
+
 static noinline_for_stack int ethtool_set_rx_ntuple(struct net_device *dev,
 						    void __user *useraddr)
 {
@@ -499,6 +531,8 @@ static noinline_for_stack int ethtool_set_rx_ntuple(struct net_device *dev,
 	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
 		return -EFAULT;
 
+	rx_ntuple_fix_masks(&cmd.fs);
+
 	/*
 	 * Cache filter in dev struct for GET operation only if
 	 * the underlying driver doesn't have its own GET operation, and
-- 
cgit v1.2.3


From 07af7a2bfa853db3957a22f9a41f437bf0f10e63 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu, 16 Sep 2010 11:34:26 +0000
Subject: ethtool: Add comments for valid use of flow types

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool.h | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 00334eebbe26..b67af60a8890 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -841,21 +841,21 @@ struct ethtool_ops {
 #define WAKE_MAGICSECURE	(1 << 6) /* only meaningful if WAKE_MAGIC */
 
 /* L3-L4 network traffic flow types */
-#define	TCP_V4_FLOW	0x01
-#define	UDP_V4_FLOW	0x02
-#define	SCTP_V4_FLOW	0x03
-#define	AH_ESP_V4_FLOW	0x04
-#define	TCP_V6_FLOW	0x05
-#define	UDP_V6_FLOW	0x06
-#define	SCTP_V6_FLOW	0x07
-#define	AH_ESP_V6_FLOW	0x08
-#define	AH_V4_FLOW	0x09
-#define	ESP_V4_FLOW	0x0a
-#define	AH_V6_FLOW	0x0b
-#define	ESP_V6_FLOW	0x0c
-#define	IP_USER_FLOW	0x0d
-#define	IPV4_FLOW	0x10
-#define	IPV6_FLOW	0x11
+#define	TCP_V4_FLOW	0x01	/* hash or spec (tcp_ip4_spec) */
+#define	UDP_V4_FLOW	0x02	/* hash or spec (udp_ip4_spec) */
+#define	SCTP_V4_FLOW	0x03	/* hash or spec (sctp_ip4_spec) */
+#define	AH_ESP_V4_FLOW	0x04	/* hash only */
+#define	TCP_V6_FLOW	0x05	/* hash only */
+#define	UDP_V6_FLOW	0x06	/* hash only */
+#define	SCTP_V6_FLOW	0x07	/* hash only */
+#define	AH_ESP_V6_FLOW	0x08	/* hash only */
+#define	AH_V4_FLOW	0x09	/* hash or spec (ah_ip4_spec) */
+#define	ESP_V4_FLOW	0x0a	/* hash or spec (esp_ip4_spec) */
+#define	AH_V6_FLOW	0x0b	/* hash only */
+#define	ESP_V6_FLOW	0x0c	/* hash only */
+#define	IP_USER_FLOW	0x0d	/* spec only (usr_ip4_spec) */
+#define	IPV4_FLOW	0x10	/* hash only */
+#define	IPV6_FLOW	0x11	/* hash only */
 
 /* L3-L4 network traffic flow hash options */
 #define	RXH_L2DA	(1 << 1)
-- 
cgit v1.2.3


From a3c74c52570c0c4ac90c9a0216de800c39089ba7 Mon Sep 17 00:00:00 2001
From: Namhyung Kim <namhyung@gmail.com>
Date: Tue, 14 Sep 2010 21:43:47 +0900
Subject: futex: Mark restart_block.futex.uaddr[2] __user

@uaddr and @uaddr2 fields in restart_block.futex are user
pointers. Add __user and remove unnecessary casts.

Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Darren Hart <dvhltc@us.ibm.com>
LKML-Reference: <1284468228-8723-2-git-send-email-namhyung@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/thread_info.h | 4 ++--
 kernel/futex.c              | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h
index a8cc4e13434c..c90696544176 100644
--- a/include/linux/thread_info.h
+++ b/include/linux/thread_info.h
@@ -23,12 +23,12 @@ struct restart_block {
 		};
 		/* For futex_wait and futex_wait_requeue_pi */
 		struct {
-			u32 *uaddr;
+			u32 __user *uaddr;
 			u32 val;
 			u32 flags;
 			u32 bitset;
 			u64 time;
-			u32 *uaddr2;
+			u32 __user *uaddr2;
 		} futex;
 		/* For nanosleep */
 		struct {
diff --git a/kernel/futex.c b/kernel/futex.c
index 464de2751ff9..45e448a5e440 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -1843,7 +1843,7 @@ retry:
 
 	restart = &current_thread_info()->restart_block;
 	restart->fn = futex_wait_restart;
-	restart->futex.uaddr = (u32 *)uaddr;
+	restart->futex.uaddr = uaddr;
 	restart->futex.val = val;
 	restart->futex.time = abs_time->tv64;
 	restart->futex.bitset = bitset;
@@ -1869,7 +1869,7 @@ out:
 
 static long futex_wait_restart(struct restart_block *restart)
 {
-	u32 __user *uaddr = (u32 __user *)restart->futex.uaddr;
+	u32 __user *uaddr = restart->futex.uaddr;
 	int fshared = 0;
 	ktime_t t, *tp = NULL;
 
-- 
cgit v1.2.3


From f0a7a98d1d400e2a5fd9a63ed56d30d30f2864cb Mon Sep 17 00:00:00 2001
From: Rabin Vincent <rabin.vincent@stericsson.com>
Date: Mon, 13 Sep 2010 13:04:02 +0100
Subject: ARM: 6373/1: tc35892-gpio: add setup/remove callbacks

For board-specific initialization.

Cc: Samuel Ortiz <sameo@linux.intel.com>
Cc: linux-kernel@vger.kernel.org
Acked-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/gpio/tc35892-gpio.c | 8 ++++++++
 include/linux/mfd/tc35892.h | 4 ++++
 2 files changed, 12 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/gpio/tc35892-gpio.c b/drivers/gpio/tc35892-gpio.c
index 1be6288780de..7e10c935a047 100644
--- a/drivers/gpio/tc35892-gpio.c
+++ b/drivers/gpio/tc35892-gpio.c
@@ -322,6 +322,9 @@ static int __devinit tc35892_gpio_probe(struct platform_device *pdev)
 		goto out_freeirq;
 	}
 
+	if (pdata->setup)
+		pdata->setup(tc35892, tc35892_gpio->chip.base);
+
 	platform_set_drvdata(pdev, tc35892_gpio);
 
 	return 0;
@@ -338,9 +341,14 @@ out_free:
 static int __devexit tc35892_gpio_remove(struct platform_device *pdev)
 {
 	struct tc35892_gpio *tc35892_gpio = platform_get_drvdata(pdev);
+	struct tc35892 *tc35892 = tc35892_gpio->tc35892;
+	struct tc35892_gpio_platform_data *pdata = tc35892->pdata->gpio;
 	int irq = platform_get_irq(pdev, 0);
 	int ret;
 
+	if (pdata->remove)
+		pdata->remove(tc35892, tc35892_gpio->chip.base);
+
 	ret = gpiochip_remove(&tc35892_gpio->chip);
 	if (ret < 0) {
 		dev_err(tc35892_gpio->dev,
diff --git a/include/linux/mfd/tc35892.h b/include/linux/mfd/tc35892.h
index e47f770d3068..eff3094ca84e 100644
--- a/include/linux/mfd/tc35892.h
+++ b/include/linux/mfd/tc35892.h
@@ -111,9 +111,13 @@ extern int tc35892_set_bits(struct tc35892 *tc35892, u8 reg, u8 mask, u8 val);
  * struct tc35892_gpio_platform_data - TC35892 GPIO platform data
  * @gpio_base: first gpio number assigned to TC35892.  A maximum of
  *	       %TC35892_NR_GPIOS GPIOs will be allocated.
+ * @setup: callback for board-specific initialization
+ * @remove: callback for board-specific teardown
  */
 struct tc35892_gpio_platform_data {
 	int gpio_base;
+	void (*setup)(struct tc35892 *tc35892, unsigned gpio_base);
+	void (*remove)(struct tc35892 *tc35892, unsigned gpio_base);
 };
 
 /**
-- 
cgit v1.2.3


From 81dcaf6516d8bbd75b894862c8ae7bba04380cfe Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 16 Sep 2010 10:17:35 +0200
Subject: workqueue: implement alloc_ordered_workqueue()

alloc_ordered_workqueue() creates a workqueue which processes each
work itemp one by one in the queued order.  This will be used to
replace create_freezeable_workqueue() and
create_singlethread_workqueue().

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 include/linux/workqueue.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 25e02c941bac..07c48925a8fc 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -306,6 +306,24 @@ __alloc_workqueue_key(const char *name, unsigned int flags, int max_active,
 	__alloc_workqueue_key((name), (flags), (max_active), NULL, NULL)
 #endif
 
+/**
+ * alloc_ordered_workqueue - allocate an ordered workqueue
+ * @name: name of the workqueue
+ * @flags: WQ_* flags (only WQ_FREEZEABLE and WQ_RESCUER are meaningful)
+ *
+ * Allocate an ordered workqueue.  An ordered workqueue executes at
+ * most one work item at any given time in the queued order.  They are
+ * implemented as unbound workqueues with @max_active of one.
+ *
+ * RETURNS:
+ * Pointer to the allocated workqueue on success, %NULL on failure.
+ */
+static inline struct workqueue_struct *
+alloc_ordered_workqueue(const char *name, unsigned int flags)
+{
+	return alloc_workqueue(name, WQ_UNBOUND | flags, 1);
+}
+
 #define create_workqueue(name)					\
 	alloc_workqueue((name), WQ_RESCUER, 1)
 #define create_freezeable_workqueue(name)			\
-- 
cgit v1.2.3


From 401a8d048eadfbe1b1c1bf53d3b614fcc894c61a Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 16 Sep 2010 10:36:00 +0200
Subject: workqueue: cleanup flush/cancel functions

Make the following cleanup changes.

* Relocate flush/cancel function prototypes and definitions.

* Relocate wait_on_cpu_work() and wait_on_work() before
  try_to_grab_pending().  These will be used to implement
  flush_work_sync().

* Make all flush/cancel functions return bool instead of int.

* Update wait_on_cpu_work() and wait_on_work() to return %true if they
  actually waited.

* Add / update comments.

This patch doesn't cause any functional changes.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 include/linux/workqueue.h |  18 ++---
 kernel/workqueue.c        | 175 +++++++++++++++++++++++++---------------------
 2 files changed, 103 insertions(+), 90 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 07c48925a8fc..bb9b683ea6fa 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -343,7 +343,6 @@ extern int queue_delayed_work_on(int cpu, struct workqueue_struct *wq,
 
 extern void flush_workqueue(struct workqueue_struct *wq);
 extern void flush_scheduled_work(void);
-extern void flush_delayed_work(struct delayed_work *work);
 
 extern int schedule_work(struct work_struct *work);
 extern int schedule_work_on(int cpu, struct work_struct *work);
@@ -355,8 +354,11 @@ extern int keventd_up(void);
 
 int execute_in_process_context(work_func_t fn, struct execute_work *);
 
-extern int flush_work(struct work_struct *work);
-extern int cancel_work_sync(struct work_struct *work);
+extern bool flush_work(struct work_struct *work);
+extern bool cancel_work_sync(struct work_struct *work);
+
+extern bool flush_delayed_work(struct delayed_work *dwork);
+extern bool cancel_delayed_work_sync(struct delayed_work *dwork);
 
 extern void workqueue_set_max_active(struct workqueue_struct *wq,
 				     int max_active);
@@ -370,9 +372,9 @@ extern unsigned int work_busy(struct work_struct *work);
  * it returns 1 and the work doesn't re-arm itself. Run flush_workqueue() or
  * cancel_work_sync() to wait on it.
  */
-static inline int cancel_delayed_work(struct delayed_work *work)
+static inline bool cancel_delayed_work(struct delayed_work *work)
 {
-	int ret;
+	bool ret;
 
 	ret = del_timer_sync(&work->timer);
 	if (ret)
@@ -385,9 +387,9 @@ static inline int cancel_delayed_work(struct delayed_work *work)
  * if it returns 0 the timer function may be running and the queueing is in
  * progress.
  */
-static inline int __cancel_delayed_work(struct delayed_work *work)
+static inline bool __cancel_delayed_work(struct delayed_work *work)
 {
-	int ret;
+	bool ret;
 
 	ret = del_timer(&work->timer);
 	if (ret)
@@ -395,8 +397,6 @@ static inline int __cancel_delayed_work(struct delayed_work *work)
 	return ret;
 }
 
-extern int cancel_delayed_work_sync(struct delayed_work *work);
-
 /* Obsolete. use cancel_delayed_work_sync() */
 static inline
 void cancel_rearming_delayed_workqueue(struct workqueue_struct *wq,
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index f77afd939229..1240b9d94b03 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -2327,16 +2327,24 @@ out_unlock:
 EXPORT_SYMBOL_GPL(flush_workqueue);
 
 /**
- * flush_work - block until a work_struct's callback has terminated
- * @work: the work which is to be flushed
+ * flush_work - wait for a work to finish executing the last queueing instance
+ * @work: the work to flush
  *
- * Returns false if @work has already terminated.
+ * Wait until @work has finished execution.  This function considers
+ * only the last queueing instance of @work.  If @work has been
+ * enqueued across different CPUs on a non-reentrant workqueue or on
+ * multiple workqueues, @work might still be executing on return on
+ * some of the CPUs from earlier queueing.
  *
- * It is expected that, prior to calling flush_work(), the caller has
- * arranged for the work to not be requeued, otherwise it doesn't make
- * sense to use this function.
+ * If @work was queued only on a non-reentrant, ordered or unbound
+ * workqueue, @work is guaranteed to be idle on return if it hasn't
+ * been requeued since flush started.
+ *
+ * RETURNS:
+ * %true if flush_work() waited for the work to finish execution,
+ * %false if it was already idle.
  */
-int flush_work(struct work_struct *work)
+bool flush_work(struct work_struct *work)
 {
 	struct worker *worker = NULL;
 	struct global_cwq *gcwq;
@@ -2374,13 +2382,49 @@ int flush_work(struct work_struct *work)
 
 	wait_for_completion(&barr.done);
 	destroy_work_on_stack(&barr.work);
-	return 1;
+	return true;
 already_gone:
 	spin_unlock_irq(&gcwq->lock);
-	return 0;
+	return false;
 }
 EXPORT_SYMBOL_GPL(flush_work);
 
+static bool wait_on_cpu_work(struct global_cwq *gcwq, struct work_struct *work)
+{
+	struct wq_barrier barr;
+	struct worker *worker;
+
+	spin_lock_irq(&gcwq->lock);
+
+	worker = find_worker_executing_work(gcwq, work);
+	if (unlikely(worker))
+		insert_wq_barrier(worker->current_cwq, &barr, work, worker);
+
+	spin_unlock_irq(&gcwq->lock);
+
+	if (unlikely(worker)) {
+		wait_for_completion(&barr.done);
+		destroy_work_on_stack(&barr.work);
+		return true;
+	} else
+		return false;
+}
+
+static bool wait_on_work(struct work_struct *work)
+{
+	bool ret = false;
+	int cpu;
+
+	might_sleep();
+
+	lock_map_acquire(&work->lockdep_map);
+	lock_map_release(&work->lockdep_map);
+
+	for_each_gcwq_cpu(cpu)
+		ret |= wait_on_cpu_work(get_gcwq(cpu), work);
+	return ret;
+}
+
 /*
  * Upon a successful return (>= 0), the caller "owns" WORK_STRUCT_PENDING bit,
  * so this work can't be re-armed in any way.
@@ -2423,39 +2467,7 @@ static int try_to_grab_pending(struct work_struct *work)
 	return ret;
 }
 
-static void wait_on_cpu_work(struct global_cwq *gcwq, struct work_struct *work)
-{
-	struct wq_barrier barr;
-	struct worker *worker;
-
-	spin_lock_irq(&gcwq->lock);
-
-	worker = find_worker_executing_work(gcwq, work);
-	if (unlikely(worker))
-		insert_wq_barrier(worker->current_cwq, &barr, work, worker);
-
-	spin_unlock_irq(&gcwq->lock);
-
-	if (unlikely(worker)) {
-		wait_for_completion(&barr.done);
-		destroy_work_on_stack(&barr.work);
-	}
-}
-
-static void wait_on_work(struct work_struct *work)
-{
-	int cpu;
-
-	might_sleep();
-
-	lock_map_acquire(&work->lockdep_map);
-	lock_map_release(&work->lockdep_map);
-
-	for_each_gcwq_cpu(cpu)
-		wait_on_cpu_work(get_gcwq(cpu), work);
-}
-
-static int __cancel_work_timer(struct work_struct *work,
+static bool __cancel_work_timer(struct work_struct *work,
 				struct timer_list* timer)
 {
 	int ret;
@@ -2472,42 +2484,60 @@ static int __cancel_work_timer(struct work_struct *work,
 }
 
 /**
- * cancel_work_sync - block until a work_struct's callback has terminated
- * @work: the work which is to be flushed
- *
- * Returns true if @work was pending.
+ * cancel_work_sync - cancel a work and wait for it to finish
+ * @work: the work to cancel
  *
- * cancel_work_sync() will cancel the work if it is queued. If the work's
- * callback appears to be running, cancel_work_sync() will block until it
- * has completed.
+ * Cancel @work and wait for its execution to finish.  This function
+ * can be used even if the work re-queues itself or migrates to
+ * another workqueue.  On return from this function, @work is
+ * guaranteed to be not pending or executing on any CPU.
  *
- * It is possible to use this function if the work re-queues itself. It can
- * cancel the work even if it migrates to another workqueue, however in that
- * case it only guarantees that work->func() has completed on the last queued
- * workqueue.
- *
- * cancel_work_sync(&delayed_work->work) should be used only if ->timer is not
- * pending, otherwise it goes into a busy-wait loop until the timer expires.
+ * cancel_work_sync(&delayed_work->work) must not be used for
+ * delayed_work's.  Use cancel_delayed_work_sync() instead.
  *
- * The caller must ensure that workqueue_struct on which this work was last
+ * The caller must ensure that the workqueue on which @work was last
  * queued can't be destroyed before this function returns.
+ *
+ * RETURNS:
+ * %true if @work was pending, %false otherwise.
  */
-int cancel_work_sync(struct work_struct *work)
+bool cancel_work_sync(struct work_struct *work)
 {
 	return __cancel_work_timer(work, NULL);
 }
 EXPORT_SYMBOL_GPL(cancel_work_sync);
 
 /**
- * cancel_delayed_work_sync - reliably kill off a delayed work.
- * @dwork: the delayed work struct
+ * flush_delayed_work - wait for a dwork to finish executing the last queueing
+ * @dwork: the delayed work to flush
+ *
+ * Delayed timer is cancelled and the pending work is queued for
+ * immediate execution.  Like flush_work(), this function only
+ * considers the last queueing instance of @dwork.
+ *
+ * RETURNS:
+ * %true if flush_work() waited for the work to finish execution,
+ * %false if it was already idle.
+ */
+bool flush_delayed_work(struct delayed_work *dwork)
+{
+	if (del_timer_sync(&dwork->timer))
+		__queue_work(raw_smp_processor_id(),
+			     get_work_cwq(&dwork->work)->wq, &dwork->work);
+	return flush_work(&dwork->work);
+}
+EXPORT_SYMBOL(flush_delayed_work);
+
+/**
+ * cancel_delayed_work_sync - cancel a delayed work and wait for it to finish
+ * @dwork: the delayed work cancel
  *
- * Returns true if @dwork was pending.
+ * This is cancel_work_sync() for delayed works.
  *
- * It is possible to use this function if @dwork rearms itself via queue_work()
- * or queue_delayed_work(). See also the comment for cancel_work_sync().
+ * RETURNS:
+ * %true if @dwork was pending, %false otherwise.
  */
-int cancel_delayed_work_sync(struct delayed_work *dwork)
+bool cancel_delayed_work_sync(struct delayed_work *dwork)
 {
 	return __cancel_work_timer(&dwork->work, &dwork->timer);
 }
@@ -2558,23 +2588,6 @@ int schedule_delayed_work(struct delayed_work *dwork,
 }
 EXPORT_SYMBOL(schedule_delayed_work);
 
-/**
- * flush_delayed_work - block until a dwork_struct's callback has terminated
- * @dwork: the delayed work which is to be flushed
- *
- * Any timeout is cancelled, and any pending work is run immediately.
- */
-void flush_delayed_work(struct delayed_work *dwork)
-{
-	if (del_timer_sync(&dwork->timer)) {
-		__queue_work(get_cpu(), get_work_cwq(&dwork->work)->wq,
-			     &dwork->work);
-		put_cpu();
-	}
-	flush_work(&dwork->work);
-}
-EXPORT_SYMBOL(flush_delayed_work);
-
 /**
  * schedule_delayed_work_on - queue work in global workqueue on CPU after delay
  * @cpu: cpu to use
-- 
cgit v1.2.3


From 09383498c5d35262e643bfdbae84826177a3c624 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 16 Sep 2010 10:48:29 +0200
Subject: workqueue: implement flush[_delayed]_work_sync()

Implement flush[_delayed]_work_sync().  These are flush functions
which also make sure no CPU is still executing the target work from
earlier queueing instances.  These are similar to
cancel[_delayed]_work_sync() except that the target work item is
flushed instead of cancelled.

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 include/linux/workqueue.h |  2 ++
 kernel/workqueue.c        | 56 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 58 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index bb9b683ea6fa..e33ff4a91703 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -355,9 +355,11 @@ extern int keventd_up(void);
 int execute_in_process_context(work_func_t fn, struct execute_work *);
 
 extern bool flush_work(struct work_struct *work);
+extern bool flush_work_sync(struct work_struct *work);
 extern bool cancel_work_sync(struct work_struct *work);
 
 extern bool flush_delayed_work(struct delayed_work *dwork);
+extern bool flush_delayed_work_sync(struct delayed_work *work);
 extern bool cancel_delayed_work_sync(struct delayed_work *dwork);
 
 extern void workqueue_set_max_active(struct workqueue_struct *wq,
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 33d31d768706..19e4bc15ee99 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -2435,6 +2435,41 @@ static bool wait_on_work(struct work_struct *work)
 	return ret;
 }
 
+/**
+ * flush_work_sync - wait until a work has finished execution
+ * @work: the work to flush
+ *
+ * Wait until @work has finished execution.  On return, it's
+ * guaranteed that all queueing instances of @work which happened
+ * before this function is called are finished.  In other words, if
+ * @work hasn't been requeued since this function was called, @work is
+ * guaranteed to be idle on return.
+ *
+ * RETURNS:
+ * %true if flush_work_sync() waited for the work to finish execution,
+ * %false if it was already idle.
+ */
+bool flush_work_sync(struct work_struct *work)
+{
+	struct wq_barrier barr;
+	bool pending, waited;
+
+	/* we'll wait for executions separately, queue barr only if pending */
+	pending = start_flush_work(work, &barr, false);
+
+	/* wait for executions to finish */
+	waited = wait_on_work(work);
+
+	/* wait for the pending one */
+	if (pending) {
+		wait_for_completion(&barr.done);
+		destroy_work_on_stack(&barr.work);
+	}
+
+	return pending || waited;
+}
+EXPORT_SYMBOL_GPL(flush_work_sync);
+
 /*
  * Upon a successful return (>= 0), the caller "owns" WORK_STRUCT_PENDING bit,
  * so this work can't be re-armed in any way.
@@ -2538,6 +2573,27 @@ bool flush_delayed_work(struct delayed_work *dwork)
 }
 EXPORT_SYMBOL(flush_delayed_work);
 
+/**
+ * flush_delayed_work_sync - wait for a dwork to finish
+ * @dwork: the delayed work to flush
+ *
+ * Delayed timer is cancelled and the pending work is queued for
+ * execution immediately.  Other than timer handling, its behavior
+ * is identical to flush_work_sync().
+ *
+ * RETURNS:
+ * %true if flush_work_sync() waited for the work to finish execution,
+ * %false if it was already idle.
+ */
+bool flush_delayed_work_sync(struct delayed_work *dwork)
+{
+	if (del_timer_sync(&dwork->timer))
+		__queue_work(raw_smp_processor_id(),
+			     get_work_cwq(&dwork->work)->wq, &dwork->work);
+	return flush_work_sync(&dwork->work);
+}
+EXPORT_SYMBOL(flush_delayed_work_sync);
+
 /**
  * cancel_delayed_work_sync - cancel a delayed work and wait for it to finish
  * @dwork: the delayed work cancel
-- 
cgit v1.2.3


From 8f8f103d8466e627ecef7894248eb79407d9047c Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Sun, 19 Sep 2010 11:24:02 -0700
Subject: net: reorder struct netdev_hw_addr

Move 'synced' and 'global_use' fields before 'refcount', to shrinks
struct netdev_hw_addr by 8 bytes (on 64bit arches).

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ec17887a5bca..f7f1302138af 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -228,9 +228,9 @@ struct netdev_hw_addr {
 #define NETDEV_HW_ADDR_T_SLAVE		3
 #define NETDEV_HW_ADDR_T_UNICAST	4
 #define NETDEV_HW_ADDR_T_MULTICAST	5
-	int			refcount;
 	bool			synced;
 	bool			global_use;
+	int			refcount;
 	struct rcu_head		rcu_head;
 };
 
-- 
cgit v1.2.3


From a18213d1d2a469956845b437f5d1d0401ab22e8b Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Sun, 19 Sep 2010 20:08:00 +0200
Subject: dccp: Replace magic CCID-specific numbers by symbolic constants

The constants DCCPO_{MIN,MAX}_CCID_SPECIFIC are nowhere used in the code, but
instead for the CCID-specific options numbers are used.

This patch unifies the use of CCID-specific option numbers, by adding symbolic
names reflecting the definitions in RFC 4340, 10.3.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
---
 include/linux/dccp.h |  6 ++++--
 net/dccp/options.c   | 13 +++----------
 2 files changed, 7 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 7434a8353e23..7187bd8a75f6 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -165,8 +165,10 @@ enum {
 	DCCPO_TIMESTAMP_ECHO = 42,
 	DCCPO_ELAPSED_TIME = 43,
 	DCCPO_MAX = 45,
-	DCCPO_MIN_CCID_SPECIFIC = 128,
-	DCCPO_MAX_CCID_SPECIFIC = 255,
+	DCCPO_MIN_RX_CCID_SPECIFIC = 128,	/* from sender to receiver */
+	DCCPO_MAX_RX_CCID_SPECIFIC = 191,
+	DCCPO_MIN_TX_CCID_SPECIFIC = 192,	/* from receiver to sender */
+	DCCPO_MAX_TX_CCID_SPECIFIC = 255,
 };
 /* maximum size of a single TLV-encoded DCCP option (sans type/len bytes) */
 #define DCCP_SINGLE_OPT_MAXLEN	253
diff --git a/net/dccp/options.c b/net/dccp/options.c
index e4983e3d2616..92718511eac5 100644
--- a/net/dccp/options.c
+++ b/net/dccp/options.c
@@ -96,18 +96,11 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
 		}
 
 		/*
-		 * CCID-Specific Options (from RFC 4340, sec. 10.3):
-		 *
-		 * Option numbers 128 through 191 are for options sent from the
-		 * HC-Sender to the HC-Receiver; option numbers 192 through 255
-		 * are for options sent from the HC-Receiver to	the HC-Sender.
-		 *
 		 * CCID-specific options are ignored during connection setup, as
 		 * negotiation may still be in progress (see RFC 4340, 10.3).
 		 * The same applies to Ack Vectors, as these depend on the CCID.
-		 *
 		 */
-		if (dreq != NULL && (opt >= 128 ||
+		if (dreq != NULL && (opt >= DCCPO_MIN_RX_CCID_SPECIFIC ||
 		    opt == DCCPO_ACK_VECTOR_0 || opt == DCCPO_ACK_VECTOR_1))
 			goto ignore_option;
 
@@ -226,12 +219,12 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
 			dccp_pr_debug("%s rx opt: ELAPSED_TIME=%d\n",
 				      dccp_role(sk), elapsed_time);
 			break;
-		case 128 ... 191:
+		case DCCPO_MIN_RX_CCID_SPECIFIC ... DCCPO_MAX_RX_CCID_SPECIFIC:
 			if (ccid_hc_rx_parse_options(dp->dccps_hc_rx_ccid, sk,
 						     pkt_type, opt, value, len))
 				goto out_invalid_option;
 			break;
-		case 192 ... 255:
+		case DCCPO_MIN_TX_CCID_SPECIFIC ... DCCPO_MAX_TX_CCID_SPECIFIC:
 			if (ccid_hc_tx_parse_options(dp->dccps_hc_tx_ccid, sk,
 						     pkt_type, opt, value, len))
 				goto out_invalid_option;
-- 
cgit v1.2.3


From 8b8e2ec1eeca7f6941bc81cefc9663018d6ceb57 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 16 Sep 2010 19:21:28 +0200
Subject: percpu: Add {get,put}_cpu_ptr

These are similar to {get,put}_cpu_var() except for dynamically
allocated per-cpu memory.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Tejun Heo <tj@kernel.org>
LKML-Reference: <20100917093009.252867712@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/percpu.h | 9 +++++++++
 1 file changed, 9 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 49466b13c5c6..0eb50832aa00 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -39,6 +39,15 @@
 	preempt_enable();				\
 } while (0)
 
+#define get_cpu_ptr(var) ({				\
+	preempt_disable();				\
+	this_cpu_ptr(var); })
+
+#define put_cpu_ptr(var) do {				\
+	(void)(var);					\
+	preempt_enable();				\
+} while (0)
+
 #ifdef CONFIG_SMP
 
 /* minimum unit size, also is the maximum supported allocation size */
-- 
cgit v1.2.3


From c1f9a095600e07fefe64eb94eb711f410100824a Mon Sep 17 00:00:00 2001
From: Ohad Ben-Cohen <ohad@wizery.com>
Date: Thu, 16 Sep 2010 13:16:02 +0200
Subject: wl12xx: make wl12xx.h common to both spi and sdio

Move wl12xx.h outside of the spi-specific location,
so it can be shared with both spi and sdio solutions.

Update all users of spi/wl12xx.h accordingly

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: Luciano Coelho <luciano.coelho@nokia.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 MAINTAINERS                                  |  2 +-
 arch/arm/mach-omap2/board-omap3pandora.c     |  1 +
 arch/arm/mach-omap2/board-rx51-peripherals.c |  2 +-
 drivers/net/wireless/wl12xx/wl1251_sdio.c    |  2 +-
 drivers/net/wireless/wl12xx/wl1251_spi.c     |  2 +-
 drivers/net/wireless/wl12xx/wl1271_spi.c     |  2 +-
 include/linux/spi/wl12xx.h                   | 34 ----------------------------
 include/linux/wl12xx.h                       | 34 ++++++++++++++++++++++++++++
 8 files changed, 40 insertions(+), 39 deletions(-)
 delete mode 100644 include/linux/spi/wl12xx.h
 create mode 100644 include/linux/wl12xx.h

(limited to 'include/linux')

diff --git a/MAINTAINERS b/MAINTAINERS
index 80cea03a737e..e9aec08e575a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6324,7 +6324,7 @@ W:	http://wireless.kernel.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-testing.git
 S:	Maintained
 F:	drivers/net/wireless/wl12xx/wl1271*
-F:	include/linux/spi/wl12xx.h
+F:	include/linux/wl12xx.h
 
 WL3501 WIRELESS PCMCIA CARD DRIVER
 M:	Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
diff --git a/arch/arm/mach-omap2/board-omap3pandora.c b/arch/arm/mach-omap2/board-omap3pandora.c
index c0f4f12eba54..9b62b6283c6e 100644
--- a/arch/arm/mach-omap2/board-omap3pandora.c
+++ b/arch/arm/mach-omap2/board-omap3pandora.c
@@ -25,6 +25,7 @@
 #include <linux/spi/ads7846.h>
 #include <linux/regulator/machine.h>
 #include <linux/i2c/twl.h>
+#include <linux/wl12xx.h>
 #include <linux/leds.h>
 #include <linux/input.h>
 #include <linux/input/matrix_keypad.h>
diff --git a/arch/arm/mach-omap2/board-rx51-peripherals.c b/arch/arm/mach-omap2/board-rx51-peripherals.c
index 03483920ed6e..1cf994299cbe 100644
--- a/arch/arm/mach-omap2/board-rx51-peripherals.c
+++ b/arch/arm/mach-omap2/board-rx51-peripherals.c
@@ -14,7 +14,7 @@
 #include <linux/input.h>
 #include <linux/input/matrix_keypad.h>
 #include <linux/spi/spi.h>
-#include <linux/spi/wl12xx.h>
+#include <linux/wl12xx.h>
 #include <linux/i2c.h>
 #include <linux/i2c/twl.h>
 #include <linux/clk.h>
diff --git a/drivers/net/wireless/wl12xx/wl1251_sdio.c b/drivers/net/wireless/wl12xx/wl1251_sdio.c
index c0b68b0a9aa8..74ba9ced5393 100644
--- a/drivers/net/wireless/wl12xx/wl1251_sdio.c
+++ b/drivers/net/wireless/wl12xx/wl1251_sdio.c
@@ -24,7 +24,7 @@
 #include <linux/mmc/sdio_func.h>
 #include <linux/mmc/sdio_ids.h>
 #include <linux/platform_device.h>
-#include <linux/spi/wl12xx.h>
+#include <linux/wl12xx.h>
 #include <linux/irq.h>
 
 #include "wl1251.h"
diff --git a/drivers/net/wireless/wl12xx/wl1251_spi.c b/drivers/net/wireless/wl12xx/wl1251_spi.c
index 334ded9881c0..320de79667a6 100644
--- a/drivers/net/wireless/wl12xx/wl1251_spi.c
+++ b/drivers/net/wireless/wl12xx/wl1251_spi.c
@@ -24,7 +24,7 @@
 #include <linux/slab.h>
 #include <linux/crc7.h>
 #include <linux/spi/spi.h>
-#include <linux/spi/wl12xx.h>
+#include <linux/wl12xx.h>
 
 #include "wl1251.h"
 #include "wl1251_reg.h"
diff --git a/drivers/net/wireless/wl12xx/wl1271_spi.c b/drivers/net/wireless/wl12xx/wl1271_spi.c
index 4cb99c541e2a..c3fdab75ad2a 100644
--- a/drivers/net/wireless/wl12xx/wl1271_spi.c
+++ b/drivers/net/wireless/wl12xx/wl1271_spi.c
@@ -25,7 +25,7 @@
 #include <linux/module.h>
 #include <linux/crc7.h>
 #include <linux/spi/spi.h>
-#include <linux/spi/wl12xx.h>
+#include <linux/wl12xx.h>
 #include <linux/slab.h>
 
 #include "wl1271.h"
diff --git a/include/linux/spi/wl12xx.h b/include/linux/spi/wl12xx.h
deleted file mode 100644
index a20bccf0b5c2..000000000000
--- a/include/linux/spi/wl12xx.h
+++ /dev/null
@@ -1,34 +0,0 @@
-/*
- * This file is part of wl12xx
- *
- * Copyright (C) 2009 Nokia Corporation
- *
- * Contact: Luciano Coelho <luciano.coelho@nokia.com>
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * version 2 as published by the Free Software Foundation.
- *
- * 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., 51 Franklin St, Fifth Floor, Boston, MA
- * 02110-1301 USA
- *
- */
-
-#ifndef _LINUX_SPI_WL12XX_H
-#define _LINUX_SPI_WL12XX_H
-
-struct wl12xx_platform_data {
-	void (*set_power)(bool enable);
-	/* SDIO only: IRQ number if WLAN_IRQ line is used, 0 for SDIO IRQs */
-	int irq;
-	bool use_eeprom;
-};
-
-#endif
diff --git a/include/linux/wl12xx.h b/include/linux/wl12xx.h
new file mode 100644
index 000000000000..015687a1776d
--- /dev/null
+++ b/include/linux/wl12xx.h
@@ -0,0 +1,34 @@
+/*
+ * This file is part of wl12xx
+ *
+ * Copyright (C) 2009 Nokia Corporation
+ *
+ * Contact: Luciano Coelho <luciano.coelho@nokia.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation.
+ *
+ * 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., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ *
+ */
+
+#ifndef _LINUX_WL12XX_H
+#define _LINUX_WL12XX_H
+
+struct wl12xx_platform_data {
+	void (*set_power)(bool enable);
+	/* SDIO only: IRQ number if WLAN_IRQ line is used, 0 for SDIO IRQs */
+	int irq;
+	bool use_eeprom;
+};
+
+#endif
-- 
cgit v1.2.3


From 61ee7007a5d61aa066076da578e8e8084e122d7d Mon Sep 17 00:00:00 2001
From: Ohad Ben-Cohen <ohad@wizery.com>
Date: Thu, 16 Sep 2010 01:31:12 +0200
Subject: wl12xx: add platform data passing support

Add a simple mechanism to pass platform data to the
SDIO instances of wl12xx.

This way there is no confusion over who owns the 'embedded data',
typechecking is preserved, and no possibility for the wrong driver to
pick up the data.

Originally proposed by Russell King.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/Makefile                      |  2 ++
 drivers/net/wireless/wl12xx/Kconfig                |  5 +++-
 drivers/net/wireless/wl12xx/wl12xx_platform_data.c | 28 ++++++++++++++++++++++
 include/linux/wl12xx.h                             |  3 +++
 4 files changed, 37 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/wireless/wl12xx/wl12xx_platform_data.c

(limited to 'include/linux')

diff --git a/drivers/net/wireless/Makefile b/drivers/net/wireless/Makefile
index 5d4ce4d2b32b..85af697574a6 100644
--- a/drivers/net/wireless/Makefile
+++ b/drivers/net/wireless/Makefile
@@ -50,5 +50,7 @@ obj-$(CONFIG_ATH_COMMON)	+= ath/
 obj-$(CONFIG_MAC80211_HWSIM)	+= mac80211_hwsim.o
 
 obj-$(CONFIG_WL12XX)	+= wl12xx/
+# small builtin driver bit
+obj-$(CONFIG_WL12XX_PLATFORM_DATA)	+= wl12xx/wl12xx_platform_data.o
 
 obj-$(CONFIG_IWM)	+= iwmc3200wifi/
diff --git a/drivers/net/wireless/wl12xx/Kconfig b/drivers/net/wireless/wl12xx/Kconfig
index 2f98058be451..4a8bb25c1739 100644
--- a/drivers/net/wireless/wl12xx/Kconfig
+++ b/drivers/net/wireless/wl12xx/Kconfig
@@ -74,4 +74,7 @@ config WL1271_SDIO
 	  If you choose to build a module, it'll be called
 	  wl1271_sdio. Say N if unsure.
 
-
+config WL12XX_PLATFORM_DATA
+	bool
+	depends on WL1271_SDIO != n
+	default y
diff --git a/drivers/net/wireless/wl12xx/wl12xx_platform_data.c b/drivers/net/wireless/wl12xx/wl12xx_platform_data.c
new file mode 100644
index 000000000000..973b11060a8f
--- /dev/null
+++ b/drivers/net/wireless/wl12xx/wl12xx_platform_data.c
@@ -0,0 +1,28 @@
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/wl12xx.h>
+
+static const struct wl12xx_platform_data *platform_data;
+
+int __init wl12xx_set_platform_data(const struct wl12xx_platform_data *data)
+{
+	if (platform_data)
+		return -EBUSY;
+	if (!data)
+		return -EINVAL;
+
+	platform_data = kmemdup(data, sizeof(*data), GFP_KERNEL);
+	if (!platform_data)
+		return -ENOMEM;
+
+	return 0;
+}
+
+const struct wl12xx_platform_data *wl12xx_get_platform_data(void)
+{
+	if (!platform_data)
+		return ERR_PTR(-ENODEV);
+
+	return platform_data;
+}
+EXPORT_SYMBOL(wl12xx_get_platform_data);
diff --git a/include/linux/wl12xx.h b/include/linux/wl12xx.h
index 015687a1776d..bd70563107fa 100644
--- a/include/linux/wl12xx.h
+++ b/include/linux/wl12xx.h
@@ -31,4 +31,7 @@ struct wl12xx_platform_data {
 	bool use_eeprom;
 };
 
+int wl12xx_set_platform_data(const struct wl12xx_platform_data *data);
+const struct wl12xx_platform_data *wl12xx_get_platform_data(void);
+
 #endif
-- 
cgit v1.2.3


From 15cea99306ae14ce5f7c3d3989bcc17202e2b0be Mon Sep 17 00:00:00 2001
From: Ohad Ben-Cohen <ohad@wizery.com>
Date: Thu, 16 Sep 2010 01:31:51 +0200
Subject: wl1271: make ref_clock configurable by board

The wl1271 device is using a reference clock that may change
between board to board.

Make the ref_clock parameter configurable by board settings
instead of having a hard coded value in the sources.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/wl12xx/wl1271.h      |  1 +
 drivers/net/wireless/wl12xx/wl1271_boot.c | 11 +++++++----
 drivers/net/wireless/wl12xx/wl1271_boot.h |  1 -
 drivers/net/wireless/wl12xx/wl1271_sdio.c |  1 +
 drivers/net/wireless/wl12xx/wl1271_spi.c  |  2 ++
 include/linux/wl12xx.h                    |  1 +
 6 files changed, 12 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/wireless/wl12xx/wl1271.h b/drivers/net/wireless/wl12xx/wl1271.h
index faa5925efe30..4134f4495b95 100644
--- a/drivers/net/wireless/wl12xx/wl1271.h
+++ b/drivers/net/wireless/wl12xx/wl1271.h
@@ -330,6 +330,7 @@ struct wl1271 {
 
 	void (*set_power)(bool enable);
 	int irq;
+	int ref_clock;
 
 	spinlock_t wl_lock;
 
diff --git a/drivers/net/wireless/wl12xx/wl1271_boot.c b/drivers/net/wireless/wl12xx/wl1271_boot.c
index f36430b0336d..fc21db810812 100644
--- a/drivers/net/wireless/wl12xx/wl1271_boot.c
+++ b/drivers/net/wireless/wl12xx/wl1271_boot.c
@@ -457,17 +457,20 @@ int wl1271_boot(struct wl1271 *wl)
 {
 	int ret = 0;
 	u32 tmp, clk, pause;
+	int ref_clock = wl->ref_clock;
 
 	wl1271_boot_hw_version(wl);
 
-	if (REF_CLOCK == 0 || REF_CLOCK == 2 || REF_CLOCK == 4)
+	if (ref_clock == 0 || ref_clock == 2 || ref_clock == 4)
 		/* ref clk: 19.2/38.4/38.4-XTAL */
 		clk = 0x3;
-	else if (REF_CLOCK == 1 || REF_CLOCK == 3)
+	else if (ref_clock == 1 || ref_clock == 3)
 		/* ref clk: 26/52 */
 		clk = 0x5;
+	else
+		return -EINVAL;
 
-	if (REF_CLOCK != 0) {
+	if (ref_clock != 0) {
 		u16 val;
 		/* Set clock type (open drain) */
 		val = wl1271_top_reg_read(wl, OCP_REG_CLK_TYPE);
@@ -516,7 +519,7 @@ int wl1271_boot(struct wl1271 *wl)
 	wl1271_debug(DEBUG_BOOT, "clk2 0x%x", clk);
 
 	/* 2 */
-	clk |= (REF_CLOCK << 1) << 4;
+	clk |= (ref_clock << 1) << 4;
 	wl1271_write32(wl, DRPW_SCRATCH_START, clk);
 
 	wl1271_set_partition(wl, &part_table[PART_WORK]);
diff --git a/drivers/net/wireless/wl12xx/wl1271_boot.h b/drivers/net/wireless/wl12xx/wl1271_boot.h
index f829699d597e..f73b0b15a280 100644
--- a/drivers/net/wireless/wl12xx/wl1271_boot.h
+++ b/drivers/net/wireless/wl12xx/wl1271_boot.h
@@ -46,7 +46,6 @@ struct wl1271_static_data {
 /* delay between retries */
 #define INIT_LOOP_DELAY 50
 
-#define REF_CLOCK            2
 #define WU_COUNTER_PAUSE_VAL 0x3FF
 #define WELP_ARM_COMMAND_VAL 0x4
 
diff --git a/drivers/net/wireless/wl12xx/wl1271_sdio.c b/drivers/net/wireless/wl12xx/wl1271_sdio.c
index 987ecdc9406d..f2f04663627c 100644
--- a/drivers/net/wireless/wl12xx/wl1271_sdio.c
+++ b/drivers/net/wireless/wl12xx/wl1271_sdio.c
@@ -234,6 +234,7 @@ static int __devinit wl1271_probe(struct sdio_func *func,
 	}
 
 	wl->irq = wlan_data->irq;
+	wl->ref_clock = wlan_data->board_ref_clock;
 
 	ret = request_irq(wl->irq, wl1271_irq, 0, DRIVER_NAME, wl);
 	if (ret < 0) {
diff --git a/drivers/net/wireless/wl12xx/wl1271_spi.c b/drivers/net/wireless/wl12xx/wl1271_spi.c
index de56d8d9ee66..ced0a9e2c7e1 100644
--- a/drivers/net/wireless/wl12xx/wl1271_spi.c
+++ b/drivers/net/wireless/wl12xx/wl1271_spi.c
@@ -372,6 +372,8 @@ static int __devinit wl1271_probe(struct spi_device *spi)
 		goto out_free;
 	}
 
+	wl->ref_clock = pdata->board_ref_clock;
+
 	wl->irq = spi->irq;
 	if (wl->irq < 0) {
 		wl1271_error("irq missing in platform data");
diff --git a/include/linux/wl12xx.h b/include/linux/wl12xx.h
index bd70563107fa..95deae3968f4 100644
--- a/include/linux/wl12xx.h
+++ b/include/linux/wl12xx.h
@@ -29,6 +29,7 @@ struct wl12xx_platform_data {
 	/* SDIO only: IRQ number if WLAN_IRQ line is used, 0 for SDIO IRQs */
 	int irq;
 	bool use_eeprom;
+	int board_ref_clock;
 };
 
 int wl12xx_set_platform_data(const struct wl12xx_platform_data *data);
-- 
cgit v1.2.3


From f4bc17cdd205ebaa3807c2aa973719bb5ce6a5b2 Mon Sep 17 00:00:00 2001
From: Julian Anastasov <ja@ssi.bg>
Date: Tue, 21 Sep 2010 17:35:41 +0200
Subject: ipvs: netfilter connection tracking changes

	Add more code to IPVS to work with Netfilter connection
tracking and fix some problems.

- Allow IPVS to be compiled without connection tracking as in
2.6.35 and before. This can avoid keeping conntracks for all
IPVS connections because this costs memory. ip_vs_ftp still
depends on connection tracking and NAT as implemented for 2.6.36.

- Add sysctl var "conntrack" to enable connection tracking for
all IPVS connections. For loaded IPVS directors it needs
tuning of nf_conntrack_max limit.

- Add IP_VS_CONN_F_NFCT connection flag to request the connection
to use connection tracking. This allows user space to provide this
flag, for example, in dest->conn_flags. This can be useful to
request connection tracking per real server instead of forcing it
for all connections with the "conntrack" sysctl. This flag is
set currently only by ip_vs_ftp and of course by "conntrack" sysctl.

- Add ip_vs_nfct.c file to hold all connection tracking code,
by this way main code should not depend of netfilter conntrack
support.

- Return back the ip_vs_post_routing handler as in 2.6.35 and use
skb->ipvs_property=1 to allow IPVS to work without connection
tracking

Connection tracking:

- most of the code is already in 2.6.36-rc

- alter conntrack reply tuple for LVS-NAT connections when first packet
from client is forwarded and conntrack state is NEW or RELATED.
Additionally, alter reply for RELATED connections from real server,
again for packet in original direction.

- add IP_VS_XMIT_TUNNEL to confirm conntrack (without altering
reply) for LVS-TUN early because we want to call nf_reset. It is
needed because we add IPIP header and the original conntrack
should be preserved, not destroyed. The transmitted IPIP packets
can reuse same conntrack, so we do not set skb->ipvs_property.

- try to destroy conntrack when the IPVS connection is destroyed.
It is not fatal if conntrack disappears before that, it depends
on the used timers.

Fix problems from long time:

- add skb->ip_summed = CHECKSUM_NONE for the LVS-TUN transmitters

Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/ip_vs.h           |   2 +
 include/net/ip_vs.h             |  44 +++++-
 net/netfilter/ipvs/Kconfig      |  13 +-
 net/netfilter/ipvs/Makefile     |   5 +-
 net/netfilter/ipvs/ip_vs_conn.c |  13 ++
 net/netfilter/ipvs/ip_vs_core.c |  46 ++++++-
 net/netfilter/ipvs/ip_vs_ctl.c  |  12 ++
 net/netfilter/ipvs/ip_vs_ftp.c  | 146 +-------------------
 net/netfilter/ipvs/ip_vs_nfct.c | 292 ++++++++++++++++++++++++++++++++++++++++
 net/netfilter/ipvs/ip_vs_xmit.c |  98 +++++++-------
 10 files changed, 475 insertions(+), 196 deletions(-)
 create mode 100644 net/netfilter/ipvs/ip_vs_nfct.c

(limited to 'include/linux')

diff --git a/include/linux/ip_vs.h b/include/linux/ip_vs.h
index 003d75f6ffe1..df7728613720 100644
--- a/include/linux/ip_vs.h
+++ b/include/linux/ip_vs.h
@@ -90,10 +90,12 @@
 #define IP_VS_CONN_F_ONE_PACKET	0x2000		/* forward only one packet */
 
 /* Flags that are not sent to backup server start from bit 16 */
+#define IP_VS_CONN_F_NFCT	(1 << 16)	/* use netfilter conntrack */
 
 /* Connection flags from destination that can be changed by user space */
 #define IP_VS_CONN_F_DEST_MASK (IP_VS_CONN_F_FWD_MASK | \
 				IP_VS_CONN_F_ONE_PACKET | \
+				IP_VS_CONN_F_NFCT | \
 				0)
 
 #define IP_VS_SCHEDNAME_MAXLEN	16
diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index 62698a9c1631..e8ec5231eae9 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -25,7 +25,9 @@
 #include <linux/ip.h>
 #include <linux/ipv6.h>			/* for struct ipv6hdr */
 #include <net/ipv6.h>			/* for ipv6_addr_copy */
-
+#ifdef CONFIG_IP_VS_NFCT
+#include <net/netfilter/nf_conntrack.h>
+#endif
 
 /* Connections' size value needed by ip_vs_ctl.c */
 extern int ip_vs_conn_tab_size;
@@ -798,6 +800,7 @@ extern int sysctl_ip_vs_expire_nodest_conn;
 extern int sysctl_ip_vs_expire_quiescent_template;
 extern int sysctl_ip_vs_sync_threshold[2];
 extern int sysctl_ip_vs_nat_icmp_send;
+extern int sysctl_ip_vs_conntrack;
 extern struct ip_vs_stats ip_vs_stats;
 extern const struct ctl_path net_vs_ctl_path[];
 
@@ -955,8 +958,47 @@ static inline __wsum ip_vs_check_diff2(__be16 old, __be16 new, __wsum oldsum)
 	return csum_partial(diff, sizeof(diff), oldsum);
 }
 
+#ifdef CONFIG_IP_VS_NFCT
+/*
+ *      Netfilter connection tracking
+ *      (from ip_vs_nfct.c)
+ */
+static inline int ip_vs_conntrack_enabled(void)
+{
+	return sysctl_ip_vs_conntrack;
+}
+
 extern void ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp,
 				   int outin);
+extern int ip_vs_confirm_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp);
+extern void ip_vs_nfct_expect_related(struct sk_buff *skb, struct nf_conn *ct,
+				      struct ip_vs_conn *cp, u_int8_t proto,
+				      const __be16 port, int from_rs);
+extern void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp);
+
+#else
+
+static inline int ip_vs_conntrack_enabled(void)
+{
+	return 0;
+}
+
+static inline void ip_vs_update_conntrack(struct sk_buff *skb,
+					  struct ip_vs_conn *cp, int outin)
+{
+}
+
+static inline int ip_vs_confirm_conntrack(struct sk_buff *skb,
+					  struct ip_vs_conn *cp)
+{
+	return NF_ACCEPT;
+}
+
+static inline void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp)
+{
+}
+/* CONFIG_IP_VS_NFCT */
+#endif
 
 #endif /* __KERNEL__ */
 
diff --git a/net/netfilter/ipvs/Kconfig b/net/netfilter/ipvs/Kconfig
index 46a77d5c3887..af3c9f48f2d7 100644
--- a/net/netfilter/ipvs/Kconfig
+++ b/net/netfilter/ipvs/Kconfig
@@ -3,7 +3,7 @@
 #
 menuconfig IP_VS
 	tristate "IP virtual server support"
-	depends on NET && INET && NETFILTER && NF_CONNTRACK
+	depends on NET && INET && NETFILTER
 	---help---
 	  IP Virtual Server support will let you build a high-performance
 	  virtual server based on cluster of two or more real servers. This
@@ -235,7 +235,8 @@ comment 'IPVS application helper'
 
 config	IP_VS_FTP
   	tristate "FTP protocol helper"
-        depends on IP_VS_PROTO_TCP && NF_NAT
+        depends on IP_VS_PROTO_TCP && NF_CONNTRACK && NF_NAT
+	select IP_VS_NFCT
 	---help---
 	  FTP is a protocol that transfers IP address and/or port number in
 	  the payload. In the virtual server via Network Address Translation,
@@ -247,4 +248,12 @@ config	IP_VS_FTP
 	  If you want to compile it in kernel, say Y. To compile it as a
 	  module, choose M here. If unsure, say N.
 
+config	IP_VS_NFCT
+	bool "Netfilter connection tracking"
+	depends on NF_CONNTRACK
+	---help---
+	  The Netfilter connection tracking support allows the IPVS
+	  connection state to be exported to the Netfilter framework
+	  for filtering purposes.
+
 endif # IP_VS
diff --git a/net/netfilter/ipvs/Makefile b/net/netfilter/ipvs/Makefile
index e3baefd7066e..349fe8819b89 100644
--- a/net/netfilter/ipvs/Makefile
+++ b/net/netfilter/ipvs/Makefile
@@ -9,10 +9,13 @@ ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_UDP) += ip_vs_proto_udp.o
 ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_AH_ESP) += ip_vs_proto_ah_esp.o
 ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_SCTP) += ip_vs_proto_sctp.o
 
+ip_vs-extra_objs-y :=
+ip_vs-extra_objs-$(CONFIG_IP_VS_NFCT) += ip_vs_nfct.o
+
 ip_vs-objs :=	ip_vs_conn.o ip_vs_core.o ip_vs_ctl.o ip_vs_sched.o	   \
 		ip_vs_xmit.o ip_vs_app.o ip_vs_sync.o	   		   \
 		ip_vs_est.o ip_vs_proto.o 				   \
-		$(ip_vs_proto-objs-y)
+		$(ip_vs_proto-objs-y) $(ip_vs-extra_objs-y)
 
 
 # IPVS core
diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c
index 9fe1da7bcf16..a970d9691496 100644
--- a/net/netfilter/ipvs/ip_vs_conn.c
+++ b/net/netfilter/ipvs/ip_vs_conn.c
@@ -721,6 +721,9 @@ static void ip_vs_conn_expire(unsigned long data)
 		if (cp->control)
 			ip_vs_control_del(cp);
 
+		if (cp->flags & IP_VS_CONN_F_NFCT)
+			ip_vs_conn_drop_conntrack(cp);
+
 		if (unlikely(cp->app != NULL))
 			ip_vs_unbind_app(cp);
 		ip_vs_unbind_dest(cp);
@@ -816,6 +819,16 @@ ip_vs_conn_new(int af, int proto, const union nf_inet_addr *caddr, __be16 cport,
 	if (unlikely(pp && atomic_read(&pp->appcnt)))
 		ip_vs_bind_app(cp, pp);
 
+	/*
+	 * Allow conntrack to be preserved. By default, conntrack
+	 * is created and destroyed for every packet.
+	 * Sometimes keeping conntrack can be useful for
+	 * IP_VS_CONN_F_ONE_PACKET too.
+	 */
+
+	if (ip_vs_conntrack_enabled())
+		cp->flags |= IP_VS_CONN_F_NFCT;
+
 	/* Hash it in the ip_vs_conn_tab finally */
 	ip_vs_conn_hash(cp);
 
diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index 319991d4d251..7fbc80d81fe8 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -537,6 +537,23 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 	return NF_DROP;
 }
 
+/*
+ * It is hooked before NF_IP_PRI_NAT_SRC at the NF_INET_POST_ROUTING
+ * chain and is used to avoid double NAT and confirmation when we do
+ * not want to keep the conntrack structure
+ */
+static unsigned int ip_vs_post_routing(unsigned int hooknum,
+				       struct sk_buff *skb,
+				       const struct net_device *in,
+				       const struct net_device *out,
+				       int (*okfn)(struct sk_buff *))
+{
+	if (!skb->ipvs_property)
+		return NF_ACCEPT;
+	/* The packet was sent from IPVS, exit this chain */
+	return NF_STOP;
+}
+
 __sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset)
 {
 	return csum_fold(skb_checksum(skb, offset, skb->len - offset, 0));
@@ -695,7 +712,10 @@ static int handle_response_icmp(int af, struct sk_buff *skb,
 	/* do the statistics and put it back */
 	ip_vs_out_stats(cp, skb);
 
-	skb->ipvs_property = 1;
+	if (!(cp->flags & IP_VS_CONN_F_NFCT))
+		skb->ipvs_property = 1;
+	else
+		ip_vs_update_conntrack(skb, cp, 0);
 	verdict = NF_ACCEPT;
 
 out:
@@ -928,17 +948,19 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp,
 
 	ip_vs_out_stats(cp, skb);
 	ip_vs_set_state(cp, IP_VS_DIR_OUTPUT, skb, pp);
-	ip_vs_update_conntrack(skb, cp, 0);
+	if (!(cp->flags & IP_VS_CONN_F_NFCT))
+		skb->ipvs_property = 1;
+	else
+		ip_vs_update_conntrack(skb, cp, 0);
 	ip_vs_conn_put(cp);
 
-	skb->ipvs_property = 1;
-
 	LeaveFunction(11);
 	return NF_ACCEPT;
 
 drop:
 	ip_vs_conn_put(cp);
 	kfree_skb(skb);
+	LeaveFunction(11);
 	return NF_STOLEN;
 }
 
@@ -1483,6 +1505,14 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 		.hooknum        = NF_INET_FORWARD,
 		.priority       = 99,
 	},
+	/* Before the netfilter connection tracking, exit from POST_ROUTING */
+	{
+		.hook		= ip_vs_post_routing,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET,
+		.hooknum        = NF_INET_POST_ROUTING,
+		.priority       = NF_IP_PRI_NAT_SRC-1,
+	},
 #ifdef CONFIG_IP_VS_IPV6
 	/* After packet filtering, forward packet through VS/DR, VS/TUN,
 	 * or VS/NAT(change destination), so that filtering rules can be
@@ -1511,6 +1541,14 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = {
 		.hooknum        = NF_INET_FORWARD,
 		.priority       = 99,
 	},
+	/* Before the netfilter connection tracking, exit from POST_ROUTING */
+	{
+		.hook		= ip_vs_post_routing,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET6,
+		.hooknum        = NF_INET_POST_ROUTING,
+		.priority       = NF_IP6_PRI_NAT_SRC-1,
+	},
 #endif
 };
 
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index 7bd41d28080c..d2d842f292c6 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -88,6 +88,9 @@ int sysctl_ip_vs_expire_nodest_conn = 0;
 int sysctl_ip_vs_expire_quiescent_template = 0;
 int sysctl_ip_vs_sync_threshold[2] = { 3, 50 };
 int sysctl_ip_vs_nat_icmp_send = 0;
+#ifdef CONFIG_IP_VS_NFCT
+int sysctl_ip_vs_conntrack;
+#endif
 
 
 #ifdef CONFIG_IP_VS_DEBUG
@@ -1580,6 +1583,15 @@ static struct ctl_table vs_vars[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_do_defense_mode,
 	},
+#ifdef CONFIG_IP_VS_NFCT
+	{
+		.procname	= "conntrack",
+		.data		= &sysctl_ip_vs_conntrack,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+#endif
 	{
 		.procname	= "secure_tcp",
 		.data		= &sysctl_ip_vs_secure_tcp,
diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c
index 7e9af5b76d9e..9cd375f94d61 100644
--- a/net/netfilter/ipvs/ip_vs_ftp.c
+++ b/net/netfilter/ipvs/ip_vs_ftp.c
@@ -20,17 +20,6 @@
  *
  * Author:	Wouter Gadeyne
  *
- *
- * Code for ip_vs_expect_related and ip_vs_expect_callback is taken from
- * http://www.ssi.bg/~ja/nfct/:
- *
- * ip_vs_nfct.c:	Netfilter connection tracking support for IPVS
- *
- * Portions Copyright (C) 2001-2002
- * Antefacto Ltd, 181 Parnell St, Dublin 1, Ireland.
- *
- * Portions Copyright (C) 2003-2008
- * Julian Anastasov
  */
 
 #define KMSG_COMPONENT "IPVS"
@@ -58,16 +47,6 @@
 #define SERVER_STRING "227 Entering Passive Mode ("
 #define CLIENT_STRING "PORT "
 
-#define FMT_TUPLE	"%pI4:%u->%pI4:%u/%u"
-#define ARG_TUPLE(T)	&(T)->src.u3.ip, ntohs((T)->src.u.all), \
-			&(T)->dst.u3.ip, ntohs((T)->dst.u.all), \
-			(T)->dst.protonum
-
-#define FMT_CONN	"%pI4:%u->%pI4:%u->%pI4:%u/%u:%u"
-#define ARG_CONN(C)	&((C)->caddr.ip), ntohs((C)->cport), \
-			&((C)->vaddr.ip), ntohs((C)->vport), \
-			&((C)->daddr.ip), ntohs((C)->dport), \
-			(C)->protocol, (C)->state
 
 /*
  * List of ports (up to IP_VS_APP_MAX_PORTS) to be handled by helper
@@ -85,6 +64,8 @@ static int ip_vs_ftp_pasv;
 static int
 ip_vs_ftp_init_conn(struct ip_vs_app *app, struct ip_vs_conn *cp)
 {
+	/* We use connection tracking for the command connection */
+	cp->flags |= IP_VS_CONN_F_NFCT;
 	return 0;
 }
 
@@ -148,120 +129,6 @@ static int ip_vs_ftp_get_addrport(char *data, char *data_limit,
 	return 1;
 }
 
-/*
- * Called from init_conntrack() as expectfn handler.
- */
-static void
-ip_vs_expect_callback(struct nf_conn *ct,
-		      struct nf_conntrack_expect *exp)
-{
-	struct nf_conntrack_tuple *orig, new_reply;
-	struct ip_vs_conn *cp;
-
-	if (exp->tuple.src.l3num != PF_INET)
-		return;
-
-	/*
-	 * We assume that no NF locks are held before this callback.
-	 * ip_vs_conn_out_get and ip_vs_conn_in_get should match their
-	 * expectations even if they use wildcard values, now we provide the
-	 * actual values from the newly created original conntrack direction.
-	 * The conntrack is confirmed when packet reaches IPVS hooks.
-	 */
-
-	/* RS->CLIENT */
-	orig = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
-	cp = ip_vs_conn_out_get(exp->tuple.src.l3num, orig->dst.protonum,
-				&orig->src.u3, orig->src.u.tcp.port,
-				&orig->dst.u3, orig->dst.u.tcp.port);
-	if (cp) {
-		/* Change reply CLIENT->RS to CLIENT->VS */
-		new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-		IP_VS_DBG(7, "%s(): ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", "
-			  FMT_TUPLE ", found inout cp=" FMT_CONN "\n",
-			  __func__, ct, ct->status,
-			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
-			  ARG_CONN(cp));
-		new_reply.dst.u3 = cp->vaddr;
-		new_reply.dst.u.tcp.port = cp->vport;
-		IP_VS_DBG(7, "%s(): ct=%p, new tuples=" FMT_TUPLE ", " FMT_TUPLE
-			  ", inout cp=" FMT_CONN "\n",
-			  __func__, ct,
-			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
-			  ARG_CONN(cp));
-		goto alter;
-	}
-
-	/* CLIENT->VS */
-	cp = ip_vs_conn_in_get(exp->tuple.src.l3num, orig->dst.protonum,
-			       &orig->src.u3, orig->src.u.tcp.port,
-			       &orig->dst.u3, orig->dst.u.tcp.port);
-	if (cp) {
-		/* Change reply VS->CLIENT to RS->CLIENT */
-		new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-		IP_VS_DBG(7, "%s(): ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", "
-			  FMT_TUPLE ", found outin cp=" FMT_CONN "\n",
-			  __func__, ct, ct->status,
-			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
-			  ARG_CONN(cp));
-		new_reply.src.u3 = cp->daddr;
-		new_reply.src.u.tcp.port = cp->dport;
-		IP_VS_DBG(7, "%s(): ct=%p, new tuples=" FMT_TUPLE ", "
-			  FMT_TUPLE ", outin cp=" FMT_CONN "\n",
-			  __func__, ct,
-			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
-			  ARG_CONN(cp));
-		goto alter;
-	}
-
-	IP_VS_DBG(7, "%s(): ct=%p, status=0x%lX, tuple=" FMT_TUPLE
-		  " - unknown expect\n",
-		  __func__, ct, ct->status, ARG_TUPLE(orig));
-	return;
-
-alter:
-	/* Never alter conntrack for non-NAT conns */
-	if (IP_VS_FWD_METHOD(cp) == IP_VS_CONN_F_MASQ)
-		nf_conntrack_alter_reply(ct, &new_reply);
-	ip_vs_conn_put(cp);
-	return;
-}
-
-/*
- * Create NF conntrack expectation with wildcard (optional) source port.
- * Then the default callback function will alter the reply and will confirm
- * the conntrack entry when the first packet comes.
- */
-static void
-ip_vs_expect_related(struct sk_buff *skb, struct nf_conn *ct,
-		     struct ip_vs_conn *cp, u_int8_t proto,
-		     const __be16 *port, int from_rs)
-{
-	struct nf_conntrack_expect *exp;
-
-	BUG_ON(!ct || ct == &nf_conntrack_untracked);
-
-	exp = nf_ct_expect_alloc(ct);
-	if (!exp)
-		return;
-
-	if (from_rs)
-		nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-				  nf_ct_l3num(ct), &cp->daddr, &cp->caddr,
-				  proto, port, &cp->cport);
-	else
-		nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT,
-				  nf_ct_l3num(ct), &cp->caddr, &cp->vaddr,
-				  proto, port, &cp->vport);
-
-	exp->expectfn = ip_vs_expect_callback;
-
-	IP_VS_DBG(7, "%s(): ct=%p, expect tuple=" FMT_TUPLE "\n",
-		  __func__, ct, ARG_TUPLE(&exp->tuple));
-	nf_ct_expect_related(exp);
-	nf_ct_expect_put(exp);
-}
-
 /*
  * Look at outgoing ftp packets to catch the response to a PASV command
  * from the server (inside-to-outside).
@@ -335,7 +202,8 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp,
 					      &cp->caddr, 0,
 					      &cp->vaddr, port,
 					      &from, port,
-					      IP_VS_CONN_F_NO_CPORT,
+					      IP_VS_CONN_F_NO_CPORT |
+					      IP_VS_CONN_F_NFCT,
 					      cp->dest);
 			if (!n_cp)
 				return 0;
@@ -371,8 +239,8 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp,
 						       start-data, end-start,
 						       buf, buf_len);
 			if (ret)
-				ip_vs_expect_related(skb, ct, n_cp,
-						     IPPROTO_TCP, NULL, 0);
+				ip_vs_nfct_expect_related(skb, ct, n_cp,
+							  IPPROTO_TCP, 0, 0);
 		}
 
 		/*
@@ -487,7 +355,7 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp,
 				      &to, port,
 				      &cp->vaddr, htons(ntohs(cp->vport)-1),
 				      &cp->daddr, htons(ntohs(cp->dport)-1),
-				      0,
+				      IP_VS_CONN_F_NFCT,
 				      cp->dest);
 		if (!n_cp)
 			return 0;
diff --git a/net/netfilter/ipvs/ip_vs_nfct.c b/net/netfilter/ipvs/ip_vs_nfct.c
new file mode 100644
index 000000000000..c038458d0290
--- /dev/null
+++ b/net/netfilter/ipvs/ip_vs_nfct.c
@@ -0,0 +1,292 @@
+/*
+ * ip_vs_nfct.c:	Netfilter connection tracking support for IPVS
+ *
+ * Portions Copyright (C) 2001-2002
+ * Antefacto Ltd, 181 Parnell St, Dublin 1, Ireland.
+ *
+ * Portions Copyright (C) 2003-2010
+ * Julian Anastasov
+ *
+ *
+ * This code 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, or
+ * (at your option) any later version.
+ *
+ * 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
+ *
+ *
+ * Authors:
+ * Ben North <ben@redfrontdoor.org>
+ * Julian Anastasov <ja@ssi.bg>		Reorganize and sync with latest kernels
+ * Hannes Eder <heder@google.com>	Extend NFCT support for FTP, ipvs match
+ *
+ *
+ * Current status:
+ *
+ * - provide conntrack confirmation for new and related connections, by
+ * this way we can see their proper conntrack state in all hooks
+ * - support for all forwarding methods, not only NAT
+ * - FTP support (NAT), ability to support other NAT apps with expectations
+ * - to correctly create expectations for related NAT connections the proper
+ * NF conntrack support must be already installed, eg. ip_vs_ftp requires
+ * nf_conntrack_ftp ... iptables_nat for the same ports (but no iptables
+ * NAT rules are needed)
+ * - alter reply for NAT when forwarding packet in original direction:
+ * conntrack from client in NEW or RELATED (Passive FTP DATA) state or
+ * when RELATED conntrack is created from real server (Active FTP DATA)
+ * - if iptables_nat is not loaded the Passive FTP will not work (the
+ * PASV response can not be NAT-ed) but Active FTP should work
+ *
+ */
+
+#define KMSG_COMPONENT "IPVS"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/compiler.h>
+#include <linux/vmalloc.h>
+#include <linux/skbuff.h>
+#include <net/ip.h>
+#include <linux/netfilter.h>
+#include <linux/netfilter_ipv4.h>
+#include <net/ip_vs.h>
+#include <net/netfilter/nf_conntrack_core.h>
+#include <net/netfilter/nf_conntrack_expect.h>
+#include <net/netfilter/nf_conntrack_helper.h>
+#include <net/netfilter/nf_conntrack_zones.h>
+
+
+#define FMT_TUPLE	"%pI4:%u->%pI4:%u/%u"
+#define ARG_TUPLE(T)	&(T)->src.u3.ip, ntohs((T)->src.u.all), \
+			&(T)->dst.u3.ip, ntohs((T)->dst.u.all), \
+			(T)->dst.protonum
+
+#define FMT_CONN	"%pI4:%u->%pI4:%u->%pI4:%u/%u:%u"
+#define ARG_CONN(C)	&((C)->caddr.ip), ntohs((C)->cport), \
+			&((C)->vaddr.ip), ntohs((C)->vport), \
+			&((C)->daddr.ip), ntohs((C)->dport), \
+			(C)->protocol, (C)->state
+
+void
+ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp, int outin)
+{
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn *ct = ct = nf_ct_get(skb, &ctinfo);
+	struct nf_conntrack_tuple new_tuple;
+
+	if (ct == NULL || nf_ct_is_confirmed(ct) || nf_ct_is_untracked(ct) ||
+	    nf_ct_is_dying(ct))
+		return;
+
+	/* Never alter conntrack for non-NAT conns */
+	if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ)
+		return;
+
+	/* Alter reply only in original direction */
+	if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL)
+		return;
+
+	/*
+	 * The connection is not yet in the hashtable, so we update it.
+	 * CIP->VIP will remain the same, so leave the tuple in
+	 * IP_CT_DIR_ORIGINAL untouched.  When the reply comes back from the
+	 * real-server we will see RIP->DIP.
+	 */
+	new_tuple = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
+	/*
+	 * This will also take care of UDP and other protocols.
+	 */
+	if (outin) {
+		new_tuple.src.u3 = cp->daddr;
+		if (new_tuple.dst.protonum != IPPROTO_ICMP &&
+		    new_tuple.dst.protonum != IPPROTO_ICMPV6)
+			new_tuple.src.u.tcp.port = cp->dport;
+	} else {
+		new_tuple.dst.u3 = cp->vaddr;
+		if (new_tuple.dst.protonum != IPPROTO_ICMP &&
+		    new_tuple.dst.protonum != IPPROTO_ICMPV6)
+			new_tuple.dst.u.tcp.port = cp->vport;
+	}
+	IP_VS_DBG(7, "%s: Updating conntrack ct=%p, status=0x%lX, "
+		  "ctinfo=%d, old reply=" FMT_TUPLE
+		  ", new reply=" FMT_TUPLE ", cp=" FMT_CONN "\n",
+		  __func__, ct, ct->status, ctinfo,
+		  ARG_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple),
+		  ARG_TUPLE(&new_tuple), ARG_CONN(cp));
+	nf_conntrack_alter_reply(ct, &new_tuple);
+}
+
+int ip_vs_confirm_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp)
+{
+	return nf_conntrack_confirm(skb);
+}
+
+/*
+ * Called from init_conntrack() as expectfn handler.
+ */
+static void ip_vs_nfct_expect_callback(struct nf_conn *ct,
+	struct nf_conntrack_expect *exp)
+{
+	struct nf_conntrack_tuple *orig, new_reply;
+	struct ip_vs_conn *cp;
+
+	if (exp->tuple.src.l3num != PF_INET)
+		return;
+
+	/*
+	 * We assume that no NF locks are held before this callback.
+	 * ip_vs_conn_out_get and ip_vs_conn_in_get should match their
+	 * expectations even if they use wildcard values, now we provide the
+	 * actual values from the newly created original conntrack direction.
+	 * The conntrack is confirmed when packet reaches IPVS hooks.
+	 */
+
+	/* RS->CLIENT */
+	orig = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple;
+	cp = ip_vs_conn_out_get(exp->tuple.src.l3num, orig->dst.protonum,
+				&orig->src.u3, orig->src.u.tcp.port,
+				&orig->dst.u3, orig->dst.u.tcp.port);
+	if (cp) {
+		/* Change reply CLIENT->RS to CLIENT->VS */
+		new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
+		IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", "
+			  FMT_TUPLE ", found inout cp=" FMT_CONN "\n",
+			  __func__, ct, ct->status,
+			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
+			  ARG_CONN(cp));
+		new_reply.dst.u3 = cp->vaddr;
+		new_reply.dst.u.tcp.port = cp->vport;
+		IP_VS_DBG(7, "%s: ct=%p, new tuples=" FMT_TUPLE ", " FMT_TUPLE
+			  ", inout cp=" FMT_CONN "\n",
+			  __func__, ct,
+			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
+			  ARG_CONN(cp));
+		goto alter;
+	}
+
+	/* CLIENT->VS */
+	cp = ip_vs_conn_in_get(exp->tuple.src.l3num, orig->dst.protonum,
+			       &orig->src.u3, orig->src.u.tcp.port,
+			       &orig->dst.u3, orig->dst.u.tcp.port);
+	if (cp) {
+		/* Change reply VS->CLIENT to RS->CLIENT */
+		new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
+		IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", "
+			  FMT_TUPLE ", found outin cp=" FMT_CONN "\n",
+			  __func__, ct, ct->status,
+			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
+			  ARG_CONN(cp));
+		new_reply.src.u3 = cp->daddr;
+		new_reply.src.u.tcp.port = cp->dport;
+		IP_VS_DBG(7, "%s: ct=%p, new tuples=" FMT_TUPLE ", "
+			  FMT_TUPLE ", outin cp=" FMT_CONN "\n",
+			  __func__, ct,
+			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
+			  ARG_CONN(cp));
+		goto alter;
+	}
+
+	IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuple=" FMT_TUPLE
+		  " - unknown expect\n",
+		  __func__, ct, ct->status, ARG_TUPLE(orig));
+	return;
+
+alter:
+	/* Never alter conntrack for non-NAT conns */
+	if (IP_VS_FWD_METHOD(cp) == IP_VS_CONN_F_MASQ)
+		nf_conntrack_alter_reply(ct, &new_reply);
+	ip_vs_conn_put(cp);
+	return;
+}
+
+/*
+ * Create NF conntrack expectation with wildcard (optional) source port.
+ * Then the default callback function will alter the reply and will confirm
+ * the conntrack entry when the first packet comes.
+ * Use port 0 to expect connection from any port.
+ */
+void ip_vs_nfct_expect_related(struct sk_buff *skb, struct nf_conn *ct,
+			       struct ip_vs_conn *cp, u_int8_t proto,
+			       const __be16 port, int from_rs)
+{
+	struct nf_conntrack_expect *exp;
+
+	if (ct == NULL || nf_ct_is_untracked(ct))
+		return;
+
+	exp = nf_ct_expect_alloc(ct);
+	if (!exp)
+		return;
+
+	nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct),
+			from_rs ? &cp->daddr : &cp->caddr,
+			from_rs ? &cp->caddr : &cp->vaddr,
+			proto, port ? &port : NULL,
+			from_rs ? &cp->cport : &cp->vport);
+
+	exp->expectfn = ip_vs_nfct_expect_callback;
+
+	IP_VS_DBG(7, "%s: ct=%p, expect tuple=" FMT_TUPLE "\n",
+		__func__, ct, ARG_TUPLE(&exp->tuple));
+	nf_ct_expect_related(exp);
+	nf_ct_expect_put(exp);
+}
+EXPORT_SYMBOL(ip_vs_nfct_expect_related);
+
+/*
+ * Our connection was terminated, try to drop the conntrack immediately
+ */
+void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp)
+{
+	struct nf_conntrack_tuple_hash *h;
+	struct nf_conn *ct;
+	struct nf_conntrack_tuple tuple;
+
+	if (!cp->cport)
+		return;
+
+	tuple = (struct nf_conntrack_tuple) {
+		.dst = { .protonum = cp->protocol, .dir = IP_CT_DIR_ORIGINAL } };
+	tuple.src.u3 = cp->caddr;
+	tuple.src.u.all = cp->cport;
+	tuple.src.l3num = cp->af;
+	tuple.dst.u3 = cp->vaddr;
+	tuple.dst.u.all = cp->vport;
+
+	IP_VS_DBG(7, "%s: dropping conntrack with tuple=" FMT_TUPLE
+		" for conn " FMT_CONN "\n",
+		__func__, ARG_TUPLE(&tuple), ARG_CONN(cp));
+
+	h = nf_conntrack_find_get(&init_net, NF_CT_DEFAULT_ZONE, &tuple);
+	if (h) {
+		ct = nf_ct_tuplehash_to_ctrack(h);
+		/* Show what happens instead of calling nf_ct_kill() */
+		if (del_timer(&ct->timeout)) {
+			IP_VS_DBG(7, "%s: ct=%p, deleted conntrack timer for tuple="
+				FMT_TUPLE "\n",
+				__func__, ct, ARG_TUPLE(&tuple));
+			if (ct->timeout.function)
+				ct->timeout.function(ct->timeout.data);
+		} else {
+			IP_VS_DBG(7, "%s: ct=%p, no conntrack timer for tuple="
+				FMT_TUPLE "\n",
+				__func__, ct, ARG_TUPLE(&tuple));
+		}
+		nf_ct_put(ct);
+	} else {
+		IP_VS_DBG(7, "%s: no conntrack for tuple=" FMT_TUPLE "\n",
+			__func__, ARG_TUPLE(&tuple));
+	}
+}
+
diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c
index 49df6bea6a2d..8817afa34e6a 100644
--- a/net/netfilter/ipvs/ip_vs_xmit.c
+++ b/net/netfilter/ipvs/ip_vs_xmit.c
@@ -28,7 +28,6 @@
 #include <net/ip6_route.h>
 #include <linux/icmpv6.h>
 #include <linux/netfilter.h>
-#include <net/netfilter/nf_conntrack.h>
 #include <linux/netfilter_ipv4.h>
 
 #include <net/ip_vs.h>
@@ -194,12 +193,37 @@ ip_vs_dst_reset(struct ip_vs_dest *dest)
 	dst_release(old_dst);
 }
 
-#define IP_VS_XMIT(pf, skb, rt)				\
+#define IP_VS_XMIT_TUNNEL(skb, cp)				\
+({								\
+	int __ret = NF_ACCEPT;					\
+								\
+	if (unlikely((cp)->flags & IP_VS_CONN_F_NFCT))		\
+		__ret = ip_vs_confirm_conntrack(skb, cp);	\
+	if (__ret == NF_ACCEPT) {				\
+		nf_reset(skb);					\
+		(skb)->ip_summed = CHECKSUM_NONE;		\
+	}							\
+	__ret;							\
+})
+
+#define IP_VS_XMIT_NAT(pf, skb, cp)				\
 do {							\
-	(skb)->ipvs_property = 1;			\
+	if (likely(!((cp)->flags & IP_VS_CONN_F_NFCT)))	\
+		(skb)->ipvs_property = 1;		\
+	else						\
+		ip_vs_update_conntrack(skb, cp, 1);	\
 	skb_forward_csum(skb);				\
 	NF_HOOK(pf, NF_INET_LOCAL_OUT, (skb), NULL,	\
-		(rt)->dst.dev, dst_output);		\
+		skb_dst(skb)->dev, dst_output);		\
+} while (0)
+
+#define IP_VS_XMIT(pf, skb, cp)				\
+do {							\
+	if (likely(!((cp)->flags & IP_VS_CONN_F_NFCT)))	\
+		(skb)->ipvs_property = 1;		\
+	skb_forward_csum(skb);				\
+	NF_HOOK(pf, NF_INET_LOCAL_OUT, (skb), NULL,	\
+		skb_dst(skb)->dev, dst_output);		\
 } while (0)
 
 
@@ -271,7 +295,7 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(NFPROTO_IPV4, skb, rt);
+	IP_VS_XMIT(NFPROTO_IPV4, skb, cp);
 
 	LeaveFunction(10);
 	return NF_STOLEN;
@@ -335,7 +359,7 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(NFPROTO_IPV6, skb, rt);
+	IP_VS_XMIT(NFPROTO_IPV6, skb, cp);
 
 	LeaveFunction(10);
 	return NF_STOLEN;
@@ -349,36 +373,6 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 }
 #endif
 
-void
-ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp, int outin)
-{
-	struct nf_conn *ct = (struct nf_conn *)skb->nfct;
-	struct nf_conntrack_tuple new_tuple;
-
-	if (ct == NULL || nf_ct_is_untracked(ct) || nf_ct_is_confirmed(ct))
-		return;
-
-	/*
-	 * The connection is not yet in the hashtable, so we update it.
-	 * CIP->VIP will remain the same, so leave the tuple in
-	 * IP_CT_DIR_ORIGINAL untouched.  When the reply comes back from the
-	 * real-server we will see RIP->DIP.
-	 */
-	new_tuple = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-	if (outin)
-		new_tuple.src.u3 = cp->daddr;
-	else
-		new_tuple.dst.u3 = cp->vaddr;
-	/*
-	 * This will also take care of UDP and other protocols.
-	 */
-	if (outin)
-		new_tuple.src.u.tcp.port = cp->dport;
-	else
-		new_tuple.dst.u.tcp.port = cp->vport;
-	nf_conntrack_alter_reply(ct, &new_tuple);
-}
-
 /*
  *      NAT transmitter (only for outside-to-inside nat forwarding)
  *      Not used for related ICMP
@@ -434,8 +428,6 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	IP_VS_DBG_PKT(10, pp, skb, 0, "After DNAT");
 
-	ip_vs_update_conntrack(skb, cp, 1);
-
 	/* FIXME: when application helper enlarges the packet and the length
 	   is larger than the MTU of outgoing device, there will be still
 	   MTU problem. */
@@ -443,7 +435,7 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(NFPROTO_IPV4, skb, rt);
+	IP_VS_XMIT_NAT(NFPROTO_IPV4, skb, cp);
 
 	LeaveFunction(10);
 	return NF_STOLEN;
@@ -451,8 +443,8 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
   tx_error_icmp:
 	dst_link_failure(skb);
   tx_error:
-	LeaveFunction(10);
 	kfree_skb(skb);
+	LeaveFunction(10);
 	return NF_STOLEN;
   tx_error_put:
 	ip_rt_put(rt);
@@ -512,8 +504,6 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 
 	IP_VS_DBG_PKT(10, pp, skb, 0, "After DNAT");
 
-	ip_vs_update_conntrack(skb, cp, 1);
-
 	/* FIXME: when application helper enlarges the packet and the length
 	   is larger than the MTU of outgoing device, there will be still
 	   MTU problem. */
@@ -521,7 +511,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(NFPROTO_IPV6, skb, rt);
+	IP_VS_XMIT_NAT(NFPROTO_IPV6, skb, cp);
 
 	LeaveFunction(10);
 	return NF_STOLEN;
@@ -571,6 +561,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	struct iphdr  *iph;			/* Our new IP header */
 	unsigned int max_headroom;		/* The extra header space needed */
 	int    mtu;
+	int ret;
 
 	EnterFunction(10);
 
@@ -655,7 +646,11 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	ip_local_out(skb);
+	ret = IP_VS_XMIT_TUNNEL(skb, cp);
+	if (ret == NF_ACCEPT)
+		ip_local_out(skb);
+	else if (ret == NF_DROP)
+		kfree_skb(skb);
 
 	LeaveFunction(10);
 
@@ -681,6 +676,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	struct ipv6hdr  *iph;		/* Our new IP header */
 	unsigned int max_headroom;	/* The extra header space needed */
 	int    mtu;
+	int ret;
 
 	EnterFunction(10);
 
@@ -761,7 +757,11 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	ip6_local_out(skb);
+	ret = IP_VS_XMIT_TUNNEL(skb, cp);
+	if (ret == NF_ACCEPT)
+		ip6_local_out(skb);
+	else if (ret == NF_DROP)
+		kfree_skb(skb);
 
 	LeaveFunction(10);
 
@@ -820,7 +820,7 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(NFPROTO_IPV4, skb, rt);
+	IP_VS_XMIT(NFPROTO_IPV4, skb, cp);
 
 	LeaveFunction(10);
 	return NF_STOLEN;
@@ -873,7 +873,7 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(NFPROTO_IPV6, skb, rt);
+	IP_VS_XMIT(NFPROTO_IPV6, skb, cp);
 
 	LeaveFunction(10);
 	return NF_STOLEN;
@@ -947,7 +947,7 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(NFPROTO_IPV4, skb, rt);
+	IP_VS_XMIT(NFPROTO_IPV4, skb, cp);
 
 	rc = NF_STOLEN;
 	goto out;
@@ -1022,7 +1022,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
 	/* Another hack: avoid icmp_send in ip_fragment */
 	skb->local_df = 1;
 
-	IP_VS_XMIT(NFPROTO_IPV6, skb, rt);
+	IP_VS_XMIT(NFPROTO_IPV6, skb, cp);
 
 	rc = NF_STOLEN;
 	goto out;
-- 
cgit v1.2.3


From c22ab7816fd81efceefa96b00c4ad62cf657964b Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Mon, 20 Sep 2010 08:41:47 +0000
Subject: ethtool: Define RX n-tuple action to clear a rule

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index b67af60a8890..3350870001fe 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -492,11 +492,12 @@ struct ethtool_rx_ntuple_flow_spec {
 	__u64		data_mask;
 
 	__s32		action;
-#define ETHTOOL_RXNTUPLE_ACTION_DROP -1		/* drop packet */
+#define ETHTOOL_RXNTUPLE_ACTION_DROP	(-1)	/* drop packet */
+#define ETHTOOL_RXNTUPLE_ACTION_CLEAR	(-2)	/* clear filter */
 };
 
 /**
- * struct ethtool_rx_ntuple - command to set RX flow filter
+ * struct ethtool_rx_ntuple - command to set or clear RX flow filter
  * @cmd: Command number - %ETHTOOL_SRXNTUPLE
  * @fs: Flow filter specification
  */
-- 
cgit v1.2.3


From 6099e3dea9aaa6127cea0610533221c9e956f009 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Mon, 20 Sep 2010 08:42:08 +0000
Subject: ethtool: Add Ethernet MAC-level filtering/steering

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 3350870001fe..8a3338ceb438 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -14,6 +14,7 @@
 #define _LINUX_ETHTOOL_H
 
 #include <linux/types.h>
+#include <linux/if_ether.h>
 
 /* This should work for both 32 and 64 bit userland. */
 struct ethtool_cmd {
@@ -391,6 +392,7 @@ struct ethtool_rx_flow_spec {
 		struct ethtool_ah_espip4_spec		ah_ip4_spec;
 		struct ethtool_ah_espip4_spec		esp_ip4_spec;
 		struct ethtool_usrip4_spec		usr_ip4_spec;
+		struct ethhdr				ether_spec;
 		__u8					hdata[72];
 	} h_u, m_u;
 	__u64		ring_cookie;
@@ -483,6 +485,7 @@ struct ethtool_rx_ntuple_flow_spec {
 		struct ethtool_ah_espip4_spec		ah_ip4_spec;
 		struct ethtool_ah_espip4_spec		esp_ip4_spec;
 		struct ethtool_usrip4_spec		usr_ip4_spec;
+		struct ethhdr				ether_spec;
 		__u8					hdata[72];
 	} h_u, m_u;
 
@@ -841,7 +844,7 @@ struct ethtool_ops {
 #define WAKE_MAGIC		(1 << 5)
 #define WAKE_MAGICSECURE	(1 << 6) /* only meaningful if WAKE_MAGIC */
 
-/* L3-L4 network traffic flow types */
+/* L2-L4 network traffic flow types */
 #define	TCP_V4_FLOW	0x01	/* hash or spec (tcp_ip4_spec) */
 #define	UDP_V4_FLOW	0x02	/* hash or spec (udp_ip4_spec) */
 #define	SCTP_V4_FLOW	0x03	/* hash or spec (sctp_ip4_spec) */
@@ -857,6 +860,7 @@ struct ethtool_ops {
 #define	IP_USER_FLOW	0x0d	/* spec only (usr_ip4_spec) */
 #define	IPV4_FLOW	0x10	/* hash only */
 #define	IPV6_FLOW	0x11	/* hash only */
+#define	ETHER_FLOW	0x12	/* spec only (ether_spec) */
 
 /* L3-L4 network traffic flow hash options */
 #define	RXH_L2DA	(1 << 1)
-- 
cgit v1.2.3


From 756e64a0b106f1a2ca96889c39ea0d48131105c0 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 21 Sep 2010 06:43:54 +0000
Subject: net: constify some ppp/pptp structs

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pppoe.c      | 2 +-
 drivers/net/pppox.c      | 4 ++--
 drivers/net/pptp.c       | 8 ++++----
 include/linux/if_pppox.h | 2 +-
 net/l2tp/l2tp_ppp.c      | 2 +-
 5 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/pppoe.c b/drivers/net/pppoe.c
index c07de359dc07..d72fb0519a2a 100644
--- a/drivers/net/pppoe.c
+++ b/drivers/net/pppoe.c
@@ -1124,7 +1124,7 @@ static const struct proto_ops pppoe_ops = {
 	.ioctl		= pppox_ioctl,
 };
 
-static struct pppox_proto pppoe_proto = {
+static const struct pppox_proto pppoe_proto = {
 	.create	= pppoe_create,
 	.ioctl	= pppoe_ioctl,
 	.owner	= THIS_MODULE,
diff --git a/drivers/net/pppox.c b/drivers/net/pppox.c
index d4191ef9cad1..8c0d170dabcd 100644
--- a/drivers/net/pppox.c
+++ b/drivers/net/pppox.c
@@ -36,9 +36,9 @@
 
 #include <asm/uaccess.h>
 
-static struct pppox_proto *pppox_protos[PX_MAX_PROTO + 1];
+static const struct pppox_proto *pppox_protos[PX_MAX_PROTO + 1];
 
-int register_pppox_proto(int proto_num, struct pppox_proto *pp)
+int register_pppox_proto(int proto_num, const struct pppox_proto *pp)
 {
 	if (proto_num < 0 || proto_num > PX_MAX_PROTO)
 		return -EINVAL;
diff --git a/drivers/net/pptp.c b/drivers/net/pptp.c
index 761f0eced724..ccbc91326bfa 100644
--- a/drivers/net/pptp.c
+++ b/drivers/net/pptp.c
@@ -53,7 +53,7 @@ static struct pppox_sock **callid_sock;
 static DEFINE_SPINLOCK(chan_lock);
 
 static struct proto pptp_sk_proto __read_mostly;
-static struct ppp_channel_ops pptp_chan_ops;
+static const struct ppp_channel_ops pptp_chan_ops;
 static const struct proto_ops pptp_ops;
 
 #define PPP_LCP_ECHOREQ 0x09
@@ -628,7 +628,7 @@ static int pptp_ppp_ioctl(struct ppp_channel *chan, unsigned int cmd,
 	return err;
 }
 
-static struct ppp_channel_ops pptp_chan_ops = {
+static const struct ppp_channel_ops pptp_chan_ops = {
 	.start_xmit = pptp_xmit,
 	.ioctl      = pptp_ppp_ioctl,
 };
@@ -659,12 +659,12 @@ static const struct proto_ops pptp_ops = {
 	.ioctl      = pppox_ioctl,
 };
 
-static struct pppox_proto pppox_pptp_proto = {
+static const struct pppox_proto pppox_pptp_proto = {
 	.create = pptp_create,
 	.owner  = THIS_MODULE,
 };
 
-static struct gre_protocol gre_pptp_protocol = {
+static const struct gre_protocol gre_pptp_protocol = {
 	.handler = pptp_rcv,
 };
 
diff --git a/include/linux/if_pppox.h b/include/linux/if_pppox.h
index 29bcd55851eb..397921b09ef9 100644
--- a/include/linux/if_pppox.h
+++ b/include/linux/if_pppox.h
@@ -204,7 +204,7 @@ struct pppox_proto {
 	struct module	*owner;
 };
 
-extern int register_pppox_proto(int proto_num, struct pppox_proto *pp);
+extern int register_pppox_proto(int proto_num, const struct pppox_proto *pp);
 extern void unregister_pppox_proto(int proto_num);
 extern void pppox_unbind_sock(struct sock *sk);/* delete ppp-channel binding */
 extern int pppox_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg);
diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c
index ff954b3e94b6..39a21d0c61c4 100644
--- a/net/l2tp/l2tp_ppp.c
+++ b/net/l2tp/l2tp_ppp.c
@@ -1768,7 +1768,7 @@ static const struct proto_ops pppol2tp_ops = {
 	.ioctl		= pppox_ioctl,
 };
 
-static struct pppox_proto pppol2tp_proto = {
+static const struct pppox_proto pppol2tp_proto = {
 	.create		= pppol2tp_create,
 	.ioctl		= pppol2tp_ioctl
 };
-- 
cgit v1.2.3


From 8b008faf92ac8f7eeb65e8cd36077601af7c46db Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Wed, 22 Sep 2010 08:36:59 +0200
Subject: netfilter: ctnetlink: allow to specify the expectation flags

With this patch, you can specify the expectation flags for user-space
created expectations.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/nf_conntrack_common.h | 4 ++++
 include/linux/netfilter/nfnetlink_conntrack.h | 1 +
 include/net/netfilter/nf_conntrack_expect.h   | 3 ---
 net/netfilter/nf_conntrack_netlink.c          | 8 +++++++-
 4 files changed, 12 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h
index 1afd18c855ec..fdc50cae861f 100644
--- a/include/linux/netfilter/nf_conntrack_common.h
+++ b/include/linux/netfilter/nf_conntrack_common.h
@@ -100,6 +100,10 @@ enum ip_conntrack_expect_events {
 	IPEXP_NEW,		/* new expectation */
 };
 
+/* expectation flags */
+#define NF_CT_EXPECT_PERMANENT		0x1
+#define NF_CT_EXPECT_INACTIVE		0x2
+
 #ifdef __KERNEL__
 struct ip_conntrack_stat {
 	unsigned int searched;
diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index 9ed534c991b9..455f0ce4f430 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -161,6 +161,7 @@ enum ctattr_expect {
 	CTA_EXPECT_ID,
 	CTA_EXPECT_HELP_NAME,
 	CTA_EXPECT_ZONE,
+	CTA_EXPECT_FLAGS,
 	__CTA_EXPECT_MAX
 };
 #define CTA_EXPECT_MAX (__CTA_EXPECT_MAX - 1)
diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index 11e815084fcf..96bb42af5fae 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -67,9 +67,6 @@ struct nf_conntrack_expect_policy {
 
 #define NF_CT_EXPECT_CLASS_DEFAULT	0
 
-#define NF_CT_EXPECT_PERMANENT	0x1
-#define NF_CT_EXPECT_INACTIVE	0x2
-
 int nf_conntrack_expect_init(struct net *net);
 void nf_conntrack_expect_fini(struct net *net);
 
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 37533a30413b..0804e0ef6500 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -1577,6 +1577,7 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb,
 
 	NLA_PUT_BE32(skb, CTA_EXPECT_TIMEOUT, htonl(timeout));
 	NLA_PUT_BE32(skb, CTA_EXPECT_ID, htonl((unsigned long)exp));
+	NLA_PUT_BE32(skb, CTA_EXPECT_FLAGS, htonl(exp->flags));
 	helper = rcu_dereference(nfct_help(master)->helper);
 	if (helper)
 		NLA_PUT_STRING(skb, CTA_EXPECT_HELP_NAME, helper->name);
@@ -1734,6 +1735,7 @@ static const struct nla_policy exp_nla_policy[CTA_EXPECT_MAX+1] = {
 	[CTA_EXPECT_ID]		= { .type = NLA_U32 },
 	[CTA_EXPECT_HELP_NAME]	= { .type = NLA_NUL_STRING },
 	[CTA_EXPECT_ZONE]	= { .type = NLA_U16 },
+	[CTA_EXPECT_FLAGS]	= { .type = NLA_U32 },
 };
 
 static int
@@ -1933,9 +1935,13 @@ ctnetlink_create_expect(struct net *net, u16 zone,
 		goto out;
 	}
 
+	if (cda[CTA_EXPECT_FLAGS])
+		exp->flags = ntohl(nla_get_be32(cda[CTA_EXPECT_FLAGS]));
+	else
+		exp->flags = 0;
+
 	exp->class = 0;
 	exp->expectfn = NULL;
-	exp->flags = 0;
 	exp->master = ct;
 	exp->helper = NULL;
 	memcpy(&exp->tuple, &tuple, sizeof(struct nf_conntrack_tuple));
-- 
cgit v1.2.3


From 676cb02dc32adef13d9efb5ea52079e4ede1e3ec Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 20 Jul 2009 23:33:49 +0200
Subject: softirqs: Make wakeup_softirqd static

No users outside of kernel/softirq.c

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/interrupt.h | 1 -
 kernel/softirq.c          | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index a0384a4d1e6f..0a9141e69241 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -410,7 +410,6 @@ extern void softirq_init(void);
 #define __raise_softirq_irqoff(nr) do { or_softirq_pending(1UL << (nr)); } while (0)
 extern void raise_softirq_irqoff(unsigned int nr);
 extern void raise_softirq(unsigned int nr);
-extern void wakeup_softirqd(void);
 
 /* This is the worklist that queues up per-cpu softirq work.
  *
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 07b4f1b1a73a..80f6e3bd1d2a 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -67,7 +67,7 @@ char *softirq_to_name[NR_SOFTIRQS] = {
  * to the pending events, so lets the scheduler to balance
  * the softirq load for us.
  */
-void wakeup_softirqd(void)
+static void wakeup_softirqd(void)
 {
 	/* Interrupts are disabled: no need to stop preemption */
 	struct task_struct *tsk = __get_cpu_var(ksoftirqd);
-- 
cgit v1.2.3


From bf5438fca2950b03c21ad868090cc1a8fcd49536 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Fri, 17 Sep 2010 11:09:00 -0400
Subject: jump label: Base patch for jump label

base patch to implement 'jump labeling'. Based on a new 'asm goto' inline
assembly gcc mechanism, we can now branch to labels from an 'asm goto'
statment. This allows us to create a 'no-op' fastpath, which can subsequently
be patched with a jump to the slowpath code. This is useful for code which
might be rarely used, but which we'd like to be able to call, if needed.
Tracepoints are the current usecase that these are being implemented for.

Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jason Baron <jbaron@redhat.com>
LKML-Reference: <ee8b3595967989fdaf84e698dc7447d315ce972a.1284733808.git.jbaron@redhat.com>

[ cleaned up some formating ]

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 Makefile                           |   5 +
 arch/Kconfig                       |   3 +
 arch/x86/include/asm/alternative.h |   3 +-
 arch/x86/kernel/alternative.c      |   2 +-
 include/asm-generic/vmlinux.lds.h  |  10 ++
 include/linux/jump_label.h         |  58 +++++++
 include/linux/module.h             |   5 +-
 kernel/Makefile                    |   2 +-
 kernel/jump_label.c                | 346 +++++++++++++++++++++++++++++++++++++
 kernel/kprobes.c                   |   1 +
 kernel/module.c                    |   6 +
 scripts/gcc-goto.sh                |   5 +
 12 files changed, 442 insertions(+), 4 deletions(-)
 create mode 100644 include/linux/jump_label.h
 create mode 100644 kernel/jump_label.c
 create mode 100644 scripts/gcc-goto.sh

(limited to 'include/linux')

diff --git a/Makefile b/Makefile
index 92ab33f16cf0..a906378d505d 100644
--- a/Makefile
+++ b/Makefile
@@ -591,6 +591,11 @@ KBUILD_CFLAGS	+= $(call cc-option,-fno-strict-overflow)
 # conserve stack if available
 KBUILD_CFLAGS   += $(call cc-option,-fconserve-stack)
 
+# check for 'asm goto'
+ifeq ($(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-goto.sh $(CC)), y)
+	KBUILD_CFLAGS += -DCC_HAVE_ASM_GOTO
+endif
+
 # Add user supplied CPPFLAGS, AFLAGS and CFLAGS as the last assignments
 # But warn user when we do so
 warn-assign = \
diff --git a/arch/Kconfig b/arch/Kconfig
index 4877a8c8ee16..1462d8492d89 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -158,4 +158,7 @@ config HAVE_PERF_EVENTS_NMI
 	  subsystem.  Also has support for calculating CPU cycle events
 	  to determine how many clock cycles in a given period.
 
+config HAVE_ARCH_JUMP_LABEL
+	bool
+
 source "kernel/gcov/Kconfig"
diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h
index 634bf782dca5..76561d20ea2f 100644
--- a/arch/x86/include/asm/alternative.h
+++ b/arch/x86/include/asm/alternative.h
@@ -4,6 +4,7 @@
 #include <linux/types.h>
 #include <linux/stddef.h>
 #include <linux/stringify.h>
+#include <linux/jump_label.h>
 #include <asm/asm.h>
 
 /*
@@ -182,7 +183,7 @@ extern void *text_poke_early(void *addr, const void *opcode, size_t len);
 extern void *text_poke(void *addr, const void *opcode, size_t len);
 extern void *text_poke_smp(void *addr, const void *opcode, size_t len);
 
-#if defined(CONFIG_DYNAMIC_FTRACE)
+#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL)
 #define IDEAL_NOP_SIZE_5 5
 extern unsigned char ideal_nop5[IDEAL_NOP_SIZE_5];
 extern void arch_init_ideal_nop5(void);
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index 083bd010d92c..cb0e6d385f6d 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -641,7 +641,7 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len)
 	return addr;
 }
 
-#if defined(CONFIG_DYNAMIC_FTRACE)
+#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL)
 
 unsigned char ideal_nop5[IDEAL_NOP_SIZE_5];
 
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 8a92a170fb7d..ef2af9948eac 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -220,6 +220,8 @@
 									\
 	BUG_TABLE							\
 									\
+	JUMP_TABLE							\
+									\
 	/* PCI quirks */						\
 	.pci_fixup        : AT(ADDR(.pci_fixup) - LOAD_OFFSET) {	\
 		VMLINUX_SYMBOL(__start_pci_fixups_early) = .;		\
@@ -563,6 +565,14 @@
 #define BUG_TABLE
 #endif
 
+#define JUMP_TABLE							\
+	. = ALIGN(8);							\
+	__jump_table : AT(ADDR(__jump_table) - LOAD_OFFSET) {		\
+		VMLINUX_SYMBOL(__start___jump_table) = .;		\
+		*(__jump_table)						\
+		VMLINUX_SYMBOL(__stop___jump_table) = .;		\
+	}
+
 #ifdef CONFIG_PM_TRACE
 #define TRACEDATA							\
 	. = ALIGN(4);							\
diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h
new file mode 100644
index 000000000000..de58656d28e0
--- /dev/null
+++ b/include/linux/jump_label.h
@@ -0,0 +1,58 @@
+#ifndef _LINUX_JUMP_LABEL_H
+#define _LINUX_JUMP_LABEL_H
+
+#if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_HAVE_ARCH_JUMP_LABEL)
+# include <asm/jump_label.h>
+# define HAVE_JUMP_LABEL
+#endif
+
+enum jump_label_type {
+	JUMP_LABEL_ENABLE,
+	JUMP_LABEL_DISABLE
+};
+
+struct module;
+
+#ifdef HAVE_JUMP_LABEL
+
+extern struct jump_entry __start___jump_table[];
+extern struct jump_entry __stop___jump_table[];
+
+extern void arch_jump_label_transform(struct jump_entry *entry,
+				 enum jump_label_type type);
+extern void jump_label_update(unsigned long key, enum jump_label_type type);
+extern void jump_label_apply_nops(struct module *mod);
+extern void arch_jump_label_text_poke_early(jump_label_t addr);
+
+#define enable_jump_label(key) \
+	jump_label_update((unsigned long)key, JUMP_LABEL_ENABLE);
+
+#define disable_jump_label(key) \
+	jump_label_update((unsigned long)key, JUMP_LABEL_DISABLE);
+
+#else
+
+#define JUMP_LABEL(key, label)			\
+do {						\
+	if (unlikely(*key))			\
+		goto label;			\
+} while (0)
+
+#define enable_jump_label(cond_var)	\
+do {					\
+       *(cond_var) = 1;			\
+} while (0)
+
+#define disable_jump_label(cond_var)	\
+do {					\
+       *(cond_var) = 0;			\
+} while (0)
+
+static inline int jump_label_apply_nops(struct module *mod)
+{
+	return 0;
+}
+
+#endif
+
+#endif
diff --git a/include/linux/module.h b/include/linux/module.h
index 8a6b9fdc7ffa..403ac26023ce 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -350,7 +350,10 @@ struct module
 	struct tracepoint *tracepoints;
 	unsigned int num_tracepoints;
 #endif
-
+#ifdef HAVE_JUMP_LABEL
+	struct jump_entry *jump_entries;
+	unsigned int num_jump_entries;
+#endif
 #ifdef CONFIG_TRACING
 	const char **trace_bprintk_fmt_start;
 	unsigned int num_trace_bprintk_fmt;
diff --git a/kernel/Makefile b/kernel/Makefile
index 0b72d1a74be0..d52b473c99a1 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -10,7 +10,7 @@ obj-y     = sched.o fork.o exec_domain.o panic.o printk.o \
 	    kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
 	    hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \
 	    notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \
-	    async.o range.o
+	    async.o range.o jump_label.o
 obj-$(CONFIG_HAVE_EARLY_RES) += early_res.o
 obj-y += groups.o
 
diff --git a/kernel/jump_label.c b/kernel/jump_label.c
new file mode 100644
index 000000000000..460fd40112b3
--- /dev/null
+++ b/kernel/jump_label.c
@@ -0,0 +1,346 @@
+/*
+ * jump label support
+ *
+ * Copyright (C) 2009 Jason Baron <jbaron@redhat.com>
+ *
+ */
+#include <linux/jump_label.h>
+#include <linux/memory.h>
+#include <linux/uaccess.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/jhash.h>
+#include <linux/slab.h>
+#include <linux/sort.h>
+#include <linux/err.h>
+
+#ifdef HAVE_JUMP_LABEL
+
+#define JUMP_LABEL_HASH_BITS 6
+#define JUMP_LABEL_TABLE_SIZE (1 << JUMP_LABEL_HASH_BITS)
+static struct hlist_head jump_label_table[JUMP_LABEL_TABLE_SIZE];
+
+/* mutex to protect coming/going of the the jump_label table */
+static DEFINE_MUTEX(jump_label_mutex);
+
+struct jump_label_entry {
+	struct hlist_node hlist;
+	struct jump_entry *table;
+	int nr_entries;
+	/* hang modules off here */
+	struct hlist_head modules;
+	unsigned long key;
+};
+
+struct jump_label_module_entry {
+	struct hlist_node hlist;
+	struct jump_entry *table;
+	int nr_entries;
+	struct module *mod;
+};
+
+static int jump_label_cmp(const void *a, const void *b)
+{
+	const struct jump_entry *jea = a;
+	const struct jump_entry *jeb = b;
+
+	if (jea->key < jeb->key)
+		return -1;
+
+	if (jea->key > jeb->key)
+		return 1;
+
+	return 0;
+}
+
+static void
+sort_jump_label_entries(struct jump_entry *start, struct jump_entry *stop)
+{
+	unsigned long size;
+
+	size = (((unsigned long)stop - (unsigned long)start)
+					/ sizeof(struct jump_entry));
+	sort(start, size, sizeof(struct jump_entry), jump_label_cmp, NULL);
+}
+
+static struct jump_label_entry *get_jump_label_entry(jump_label_t key)
+{
+	struct hlist_head *head;
+	struct hlist_node *node;
+	struct jump_label_entry *e;
+	u32 hash = jhash((void *)&key, sizeof(jump_label_t), 0);
+
+	head = &jump_label_table[hash & (JUMP_LABEL_TABLE_SIZE - 1)];
+	hlist_for_each_entry(e, node, head, hlist) {
+		if (key == e->key)
+			return e;
+	}
+	return NULL;
+}
+
+static struct jump_label_entry *
+add_jump_label_entry(jump_label_t key, int nr_entries, struct jump_entry *table)
+{
+	struct hlist_head *head;
+	struct jump_label_entry *e;
+	u32 hash;
+
+	e = get_jump_label_entry(key);
+	if (e)
+		return ERR_PTR(-EEXIST);
+
+	e = kmalloc(sizeof(struct jump_label_entry), GFP_KERNEL);
+	if (!e)
+		return ERR_PTR(-ENOMEM);
+
+	hash = jhash((void *)&key, sizeof(jump_label_t), 0);
+	head = &jump_label_table[hash & (JUMP_LABEL_TABLE_SIZE - 1)];
+	e->key = key;
+	e->table = table;
+	e->nr_entries = nr_entries;
+	INIT_HLIST_HEAD(&(e->modules));
+	hlist_add_head(&e->hlist, head);
+	return e;
+}
+
+static int
+build_jump_label_hashtable(struct jump_entry *start, struct jump_entry *stop)
+{
+	struct jump_entry *iter, *iter_begin;
+	struct jump_label_entry *entry;
+	int count;
+
+	sort_jump_label_entries(start, stop);
+	iter = start;
+	while (iter < stop) {
+		entry = get_jump_label_entry(iter->key);
+		if (!entry) {
+			iter_begin = iter;
+			count = 0;
+			while ((iter < stop) &&
+				(iter->key == iter_begin->key)) {
+				iter++;
+				count++;
+			}
+			entry = add_jump_label_entry(iter_begin->key,
+							count, iter_begin);
+			if (IS_ERR(entry))
+				return PTR_ERR(entry);
+		 } else {
+			WARN_ONCE(1, KERN_ERR "build_jump_hashtable: unexpected entry!\n");
+			return -1;
+		}
+	}
+	return 0;
+}
+
+/***
+ * jump_label_update - update jump label text
+ * @key -  key value associated with a a jump label
+ * @type - enum set to JUMP_LABEL_ENABLE or JUMP_LABEL_DISABLE
+ *
+ * Will enable/disable the jump for jump label @key, depending on the
+ * value of @type.
+ *
+ */
+
+void jump_label_update(unsigned long key, enum jump_label_type type)
+{
+	struct jump_entry *iter;
+	struct jump_label_entry *entry;
+	struct hlist_node *module_node;
+	struct jump_label_module_entry *e_module;
+	int count;
+
+	mutex_lock(&jump_label_mutex);
+	entry = get_jump_label_entry((jump_label_t)key);
+	if (entry) {
+		count = entry->nr_entries;
+		iter = entry->table;
+		while (count--) {
+			if (kernel_text_address(iter->code))
+				arch_jump_label_transform(iter, type);
+			iter++;
+		}
+		/* eanble/disable jump labels in modules */
+		hlist_for_each_entry(e_module, module_node, &(entry->modules),
+							hlist) {
+			count = e_module->nr_entries;
+			iter = e_module->table;
+			while (count--) {
+				if (kernel_text_address(iter->code))
+					arch_jump_label_transform(iter, type);
+				iter++;
+			}
+		}
+	}
+	mutex_unlock(&jump_label_mutex);
+}
+
+static __init int init_jump_label(void)
+{
+	int ret;
+	struct jump_entry *iter_start = __start___jump_table;
+	struct jump_entry *iter_stop = __stop___jump_table;
+	struct jump_entry *iter;
+
+	mutex_lock(&jump_label_mutex);
+	ret = build_jump_label_hashtable(__start___jump_table,
+					 __stop___jump_table);
+	iter = iter_start;
+	while (iter < iter_stop) {
+		arch_jump_label_text_poke_early(iter->code);
+		iter++;
+	}
+	mutex_unlock(&jump_label_mutex);
+	return ret;
+}
+early_initcall(init_jump_label);
+
+#ifdef CONFIG_MODULES
+
+static struct jump_label_module_entry *
+add_jump_label_module_entry(struct jump_label_entry *entry,
+			    struct jump_entry *iter_begin,
+			    int count, struct module *mod)
+{
+	struct jump_label_module_entry *e;
+
+	e = kmalloc(sizeof(struct jump_label_module_entry), GFP_KERNEL);
+	if (!e)
+		return ERR_PTR(-ENOMEM);
+	e->mod = mod;
+	e->nr_entries = count;
+	e->table = iter_begin;
+	hlist_add_head(&e->hlist, &entry->modules);
+	return e;
+}
+
+static int add_jump_label_module(struct module *mod)
+{
+	struct jump_entry *iter, *iter_begin;
+	struct jump_label_entry *entry;
+	struct jump_label_module_entry *module_entry;
+	int count;
+
+	/* if the module doesn't have jump label entries, just return */
+	if (!mod->num_jump_entries)
+		return 0;
+
+	sort_jump_label_entries(mod->jump_entries,
+				mod->jump_entries + mod->num_jump_entries);
+	iter = mod->jump_entries;
+	while (iter < mod->jump_entries + mod->num_jump_entries) {
+		entry = get_jump_label_entry(iter->key);
+		iter_begin = iter;
+		count = 0;
+		while ((iter < mod->jump_entries + mod->num_jump_entries) &&
+			(iter->key == iter_begin->key)) {
+				iter++;
+				count++;
+		}
+		if (!entry) {
+			entry = add_jump_label_entry(iter_begin->key, 0, NULL);
+			if (IS_ERR(entry))
+				return PTR_ERR(entry);
+		}
+		module_entry = add_jump_label_module_entry(entry, iter_begin,
+							   count, mod);
+		if (IS_ERR(module_entry))
+			return PTR_ERR(module_entry);
+	}
+	return 0;
+}
+
+static void remove_jump_label_module(struct module *mod)
+{
+	struct hlist_head *head;
+	struct hlist_node *node, *node_next, *module_node, *module_node_next;
+	struct jump_label_entry *e;
+	struct jump_label_module_entry *e_module;
+	int i;
+
+	/* if the module doesn't have jump label entries, just return */
+	if (!mod->num_jump_entries)
+		return;
+
+	for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) {
+		head = &jump_label_table[i];
+		hlist_for_each_entry_safe(e, node, node_next, head, hlist) {
+			hlist_for_each_entry_safe(e_module, module_node,
+						  module_node_next,
+						  &(e->modules), hlist) {
+				if (e_module->mod == mod) {
+					hlist_del(&e_module->hlist);
+					kfree(e_module);
+				}
+			}
+			if (hlist_empty(&e->modules) && (e->nr_entries == 0)) {
+				hlist_del(&e->hlist);
+				kfree(e);
+			}
+		}
+	}
+}
+
+static int
+jump_label_module_notify(struct notifier_block *self, unsigned long val,
+			 void *data)
+{
+	struct module *mod = data;
+	int ret = 0;
+
+	switch (val) {
+	case MODULE_STATE_COMING:
+		mutex_lock(&jump_label_mutex);
+		ret = add_jump_label_module(mod);
+		if (ret)
+			remove_jump_label_module(mod);
+		mutex_unlock(&jump_label_mutex);
+		break;
+	case MODULE_STATE_GOING:
+		mutex_lock(&jump_label_mutex);
+		remove_jump_label_module(mod);
+		mutex_unlock(&jump_label_mutex);
+		break;
+	}
+	return ret;
+}
+
+/***
+ * apply_jump_label_nops - patch module jump labels with arch_get_jump_label_nop()
+ * @mod: module to patch
+ *
+ * Allow for run-time selection of the optimal nops. Before the module
+ * loads patch these with arch_get_jump_label_nop(), which is specified by
+ * the arch specific jump label code.
+ */
+void jump_label_apply_nops(struct module *mod)
+{
+	struct jump_entry *iter;
+
+	/* if the module doesn't have jump label entries, just return */
+	if (!mod->num_jump_entries)
+		return;
+
+	iter = mod->jump_entries;
+	while (iter < mod->jump_entries + mod->num_jump_entries) {
+		arch_jump_label_text_poke_early(iter->code);
+		iter++;
+	}
+}
+
+struct notifier_block jump_label_module_nb = {
+	.notifier_call = jump_label_module_notify,
+	.priority = 0,
+};
+
+static __init int init_jump_label_module(void)
+{
+	return register_module_notifier(&jump_label_module_nb);
+}
+early_initcall(init_jump_label_module);
+
+#endif /* CONFIG_MODULES */
+
+#endif
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 6dd5359e1f0e..18904e42a918 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -47,6 +47,7 @@
 #include <linux/memory.h>
 #include <linux/ftrace.h>
 #include <linux/cpu.h>
+#include <linux/jump_label.h>
 
 #include <asm-generic/sections.h>
 #include <asm/cacheflush.h>
diff --git a/kernel/module.c b/kernel/module.c
index d0b5f8db11b4..eba134157ef6 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -55,6 +55,7 @@
 #include <linux/async.h>
 #include <linux/percpu.h>
 #include <linux/kmemleak.h>
+#include <linux/jump_label.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/module.h>
@@ -2308,6 +2309,11 @@ static void find_module_sections(struct module *mod, struct load_info *info)
 					sizeof(*mod->tracepoints),
 					&mod->num_tracepoints);
 #endif
+#ifdef HAVE_JUMP_LABEL
+	mod->jump_entries = section_objs(info, "__jump_table",
+					sizeof(*mod->jump_entries),
+					&mod->num_jump_entries);
+#endif
 #ifdef CONFIG_EVENT_TRACING
 	mod->trace_events = section_objs(info, "_ftrace_events",
 					 sizeof(*mod->trace_events),
diff --git a/scripts/gcc-goto.sh b/scripts/gcc-goto.sh
new file mode 100644
index 000000000000..8e82424be7aa
--- /dev/null
+++ b/scripts/gcc-goto.sh
@@ -0,0 +1,5 @@
+#!/bin/sh
+# Test for gcc 'asm goto' suport
+# Copyright (C) 2010, Jason Baron <jbaron@redhat.com>
+
+echo "int main(void) { entry: asm goto (\"\"::::entry); return 0; }" | $1 -x c - -c -o /dev/null >/dev/null 2>&1 && echo "y"
-- 
cgit v1.2.3


From 4c3ef6d79328c0e23ade60cbfc8d496123a6855c Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Fri, 17 Sep 2010 11:09:08 -0400
Subject: jump label: Add jump_label_text_reserved() to reserve jump points

Add a jump_label_text_reserved(void *start, void *end), so that other
pieces of code that want to modify kernel text, can first verify that
jump label has not reserved the instruction.

Acked-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Jason Baron <jbaron@redhat.com>
LKML-Reference: <06236663a3a7b1c1f13576bb9eccb6d9c17b7bfe.1284733808.git.jbaron@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 arch/x86/kernel/kprobes.c  |  3 +-
 include/linux/jump_label.h |  8 ++++-
 kernel/jump_label.c        | 83 ++++++++++++++++++++++++++++++++++++++++++++++
 kernel/kprobes.c           |  3 +-
 4 files changed, 94 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index e05952af5d26..1cbd54c0df99 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -1218,7 +1218,8 @@ static int __kprobes copy_optimized_instructions(u8 *dest, u8 *src)
 	}
 	/* Check whether the address range is reserved */
 	if (ftrace_text_reserved(src, src + len - 1) ||
-	    alternatives_text_reserved(src, src + len - 1))
+	    alternatives_text_reserved(src, src + len - 1) ||
+	    jump_label_text_reserved(src, src + len - 1))
 		return -EBUSY;
 
 	return len;
diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h
index de58656d28e0..b72cd9f92c2e 100644
--- a/include/linux/jump_label.h
+++ b/include/linux/jump_label.h
@@ -20,9 +20,10 @@ extern struct jump_entry __stop___jump_table[];
 
 extern void arch_jump_label_transform(struct jump_entry *entry,
 				 enum jump_label_type type);
+extern void arch_jump_label_text_poke_early(jump_label_t addr);
 extern void jump_label_update(unsigned long key, enum jump_label_type type);
 extern void jump_label_apply_nops(struct module *mod);
-extern void arch_jump_label_text_poke_early(jump_label_t addr);
+extern int jump_label_text_reserved(void *start, void *end);
 
 #define enable_jump_label(key) \
 	jump_label_update((unsigned long)key, JUMP_LABEL_ENABLE);
@@ -53,6 +54,11 @@ static inline int jump_label_apply_nops(struct module *mod)
 	return 0;
 }
 
+static inline int jump_label_text_reserved(void *start, void *end)
+{
+	return 0;
+}
+
 #endif
 
 #endif
diff --git a/kernel/jump_label.c b/kernel/jump_label.c
index 460fd40112b3..7be868bf25c6 100644
--- a/kernel/jump_label.c
+++ b/kernel/jump_label.c
@@ -177,6 +177,89 @@ void jump_label_update(unsigned long key, enum jump_label_type type)
 	mutex_unlock(&jump_label_mutex);
 }
 
+static int addr_conflict(struct jump_entry *entry, void *start, void *end)
+{
+	if (entry->code <= (unsigned long)end &&
+		entry->code + JUMP_LABEL_NOP_SIZE > (unsigned long)start)
+		return 1;
+
+	return 0;
+}
+
+#ifdef CONFIG_MODULES
+
+static int module_conflict(void *start, void *end)
+{
+	struct hlist_head *head;
+	struct hlist_node *node, *node_next, *module_node, *module_node_next;
+	struct jump_label_entry *e;
+	struct jump_label_module_entry *e_module;
+	struct jump_entry *iter;
+	int i, count;
+	int conflict = 0;
+
+	for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) {
+		head = &jump_label_table[i];
+		hlist_for_each_entry_safe(e, node, node_next, head, hlist) {
+			hlist_for_each_entry_safe(e_module, module_node,
+							module_node_next,
+							&(e->modules), hlist) {
+				count = e_module->nr_entries;
+				iter = e_module->table;
+				while (count--) {
+					if (addr_conflict(iter, start, end)) {
+						conflict = 1;
+						goto out;
+					}
+					iter++;
+				}
+			}
+		}
+	}
+out:
+	return conflict;
+}
+
+#endif
+
+/***
+ * jump_label_text_reserved - check if addr range is reserved
+ * @start: start text addr
+ * @end: end text addr
+ *
+ * checks if the text addr located between @start and @end
+ * overlaps with any of the jump label patch addresses. Code
+ * that wants to modify kernel text should first verify that
+ * it does not overlap with any of the jump label addresses.
+ *
+ * returns 1 if there is an overlap, 0 otherwise
+ */
+int jump_label_text_reserved(void *start, void *end)
+{
+	struct jump_entry *iter;
+	struct jump_entry *iter_start = __start___jump_table;
+	struct jump_entry *iter_stop = __start___jump_table;
+	int conflict = 0;
+
+	mutex_lock(&jump_label_mutex);
+	iter = iter_start;
+	while (iter < iter_stop) {
+		if (addr_conflict(iter, start, end)) {
+			conflict = 1;
+			goto out;
+		}
+		iter++;
+	}
+
+	/* now check modules */
+#ifdef CONFIG_MODULES
+	conflict = module_conflict(start, end);
+#endif
+out:
+	mutex_unlock(&jump_label_mutex);
+	return conflict;
+}
+
 static __init int init_jump_label(void)
 {
 	int ret;
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 18904e42a918..ec4210c6501e 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -1147,7 +1147,8 @@ int __kprobes register_kprobe(struct kprobe *p)
 	preempt_disable();
 	if (!kernel_text_address((unsigned long) p->addr) ||
 	    in_kprobes_functions((unsigned long) p->addr) ||
-	    ftrace_text_reserved(p->addr, p->addr)) {
+	    ftrace_text_reserved(p->addr, p->addr) ||
+	    jump_label_text_reserved(p->addr, p->addr)) {
 		preempt_enable();
 		return -EINVAL;
 	}
-- 
cgit v1.2.3


From 8f7b50c514206211cc282a4247f7b12f18dee674 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Fri, 17 Sep 2010 11:09:13 -0400
Subject: jump label: Tracepoint support for jump labels

Make use of the jump label infrastructure for tracepoints.

Signed-off-by: Jason Baron <jbaron@redhat.com>
LKML-Reference: <a9ba2056e2c9cf332c3c300b577463ce66ff23a8.1284733808.git.jbaron@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 include/linux/tracepoint.h |  5 ++++-
 kernel/tracepoint.c        | 14 ++++++++++++--
 2 files changed, 16 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h
index 103d1b61aacb..a4a90b6726ce 100644
--- a/include/linux/tracepoint.h
+++ b/include/linux/tracepoint.h
@@ -17,6 +17,7 @@
 #include <linux/errno.h>
 #include <linux/types.h>
 #include <linux/rcupdate.h>
+#include <linux/jump_label.h>
 
 struct module;
 struct tracepoint;
@@ -145,7 +146,9 @@ static inline void tracepoint_update_probe_range(struct tracepoint *begin,
 	extern struct tracepoint __tracepoint_##name;			\
 	static inline void trace_##name(proto)				\
 	{								\
-		if (unlikely(__tracepoint_##name.state))		\
+		JUMP_LABEL(&__tracepoint_##name.state, do_trace);	\
+		return;							\
+do_trace:								\
 			__DO_TRACE(&__tracepoint_##name,		\
 				TP_PROTO(data_proto),			\
 				TP_ARGS(data_args));			\
diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c
index c77f3eceea25..d6073a50a6ca 100644
--- a/kernel/tracepoint.c
+++ b/kernel/tracepoint.c
@@ -25,6 +25,7 @@
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/sched.h>
+#include <linux/jump_label.h>
 
 extern struct tracepoint __start___tracepoints[];
 extern struct tracepoint __stop___tracepoints[];
@@ -263,7 +264,13 @@ static void set_tracepoint(struct tracepoint_entry **entry,
 	 * is used.
 	 */
 	rcu_assign_pointer(elem->funcs, (*entry)->funcs);
-	elem->state = active;
+	if (!elem->state && active) {
+		enable_jump_label(&elem->state);
+		elem->state = active;
+	} else if (elem->state && !active) {
+		disable_jump_label(&elem->state);
+		elem->state = active;
+	}
 }
 
 /*
@@ -277,7 +284,10 @@ static void disable_tracepoint(struct tracepoint *elem)
 	if (elem->unregfunc && elem->state)
 		elem->unregfunc();
 
-	elem->state = 0;
+	if (elem->state) {
+		disable_jump_label(&elem->state);
+		elem->state = 0;
+	}
 	rcu_assign_pointer(elem->funcs, NULL);
 }
 
-- 
cgit v1.2.3


From 52159d98be6f26c48f5e02c7ab3c9848a85979b5 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Fri, 17 Sep 2010 11:09:17 -0400
Subject: jump label: Convert dynamic debug to use jump labels

Convert the 'dynamic debug' infrastructure to use jump labels.

Signed-off-by: Jason Baron <jbaron@redhat.com>
LKML-Reference: <b77627358cea3e27d7be4386f45f66219afb8452.1284733808.git.jbaron@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 include/linux/dynamic_debug.h | 39 ++++++++++++++------------
 lib/dynamic_debug.c           | 42 ++--------------------------
 scripts/Makefile.lib          | 11 +-------
 scripts/basic/Makefile        |  2 +-
 scripts/basic/hash.c          | 64 -------------------------------------------
 5 files changed, 26 insertions(+), 132 deletions(-)
 delete mode 100644 scripts/basic/hash.c

(limited to 'include/linux')

diff --git a/include/linux/dynamic_debug.h b/include/linux/dynamic_debug.h
index 52c0da4bdd18..bef3cda44c4c 100644
--- a/include/linux/dynamic_debug.h
+++ b/include/linux/dynamic_debug.h
@@ -1,6 +1,8 @@
 #ifndef _DYNAMIC_DEBUG_H
 #define _DYNAMIC_DEBUG_H
 
+#include <linux/jump_label.h>
+
 /* dynamic_printk_enabled, and dynamic_printk_enabled2 are bitmasks in which
  * bit n is set to 1 if any modname hashes into the bucket n, 0 otherwise. They
  * use independent hash functions, to reduce the chance of false positives.
@@ -22,8 +24,6 @@ struct _ddebug {
 	const char *function;
 	const char *filename;
 	const char *format;
-	char primary_hash;
-	char secondary_hash;
 	unsigned int lineno:24;
 	/*
  	 * The flags field controls the behaviour at the callsite.
@@ -33,6 +33,7 @@ struct _ddebug {
 #define _DPRINTK_FLAGS_PRINT   (1<<0)  /* printk() a message using the format */
 #define _DPRINTK_FLAGS_DEFAULT 0
 	unsigned int flags:8;
+	char enabled;
 } __attribute__((aligned(8)));
 
 
@@ -42,33 +43,35 @@ int ddebug_add_module(struct _ddebug *tab, unsigned int n,
 #if defined(CONFIG_DYNAMIC_DEBUG)
 extern int ddebug_remove_module(const char *mod_name);
 
-#define __dynamic_dbg_enabled(dd)  ({	     \
-	int __ret = 0;							     \
-	if (unlikely((dynamic_debug_enabled & (1LL << DEBUG_HASH)) &&	     \
-			(dynamic_debug_enabled2 & (1LL << DEBUG_HASH2))))   \
-				if (unlikely(dd.flags))			     \
-					__ret = 1;			     \
-	__ret; })
-
 #define dynamic_pr_debug(fmt, ...) do {					\
+	__label__ do_printk;						\
+	__label__ out;							\
 	static struct _ddebug descriptor				\
 	__used								\
 	__attribute__((section("__verbose"), aligned(8))) =		\
-	{ KBUILD_MODNAME, __func__, __FILE__, fmt, DEBUG_HASH,	\
-		DEBUG_HASH2, __LINE__, _DPRINTK_FLAGS_DEFAULT };	\
-	if (__dynamic_dbg_enabled(descriptor))				\
-		printk(KERN_DEBUG pr_fmt(fmt),	##__VA_ARGS__);		\
+	{ KBUILD_MODNAME, __func__, __FILE__, fmt, __LINE__,		\
+		_DPRINTK_FLAGS_DEFAULT };				\
+	JUMP_LABEL(&descriptor.enabled, do_printk);			\
+	goto out;							\
+do_printk:								\
+	printk(KERN_DEBUG pr_fmt(fmt),	##__VA_ARGS__);			\
+out:	;								\
 	} while (0)
 
 
 #define dynamic_dev_dbg(dev, fmt, ...) do {				\
+	__label__ do_printk;						\
+	__label__ out;							\
 	static struct _ddebug descriptor				\
 	__used								\
 	__attribute__((section("__verbose"), aligned(8))) =		\
-	{ KBUILD_MODNAME, __func__, __FILE__, fmt, DEBUG_HASH,	\
-		DEBUG_HASH2, __LINE__, _DPRINTK_FLAGS_DEFAULT };	\
-	if (__dynamic_dbg_enabled(descriptor))				\
-		dev_printk(KERN_DEBUG, dev, fmt, ##__VA_ARGS__);	\
+	{ KBUILD_MODNAME, __func__, __FILE__, fmt, __LINE__,		\
+		_DPRINTK_FLAGS_DEFAULT };				\
+	JUMP_LABEL(&descriptor.enabled, do_printk);			\
+	goto out;							\
+do_printk:								\
+	dev_printk(KERN_DEBUG, dev, fmt, ##__VA_ARGS__);		\
+out:	;								\
 	} while (0)
 
 #else
diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c
index 02afc2533728..e925c7b960f1 100644
--- a/lib/dynamic_debug.c
+++ b/lib/dynamic_debug.c
@@ -26,19 +26,11 @@
 #include <linux/dynamic_debug.h>
 #include <linux/debugfs.h>
 #include <linux/slab.h>
+#include <linux/jump_label.h>
 
 extern struct _ddebug __start___verbose[];
 extern struct _ddebug __stop___verbose[];
 
-/* dynamic_debug_enabled, and dynamic_debug_enabled2 are bitmasks in which
- * bit n is set to 1 if any modname hashes into the bucket n, 0 otherwise. They
- * use independent hash functions, to reduce the chance of false positives.
- */
-long long dynamic_debug_enabled;
-EXPORT_SYMBOL_GPL(dynamic_debug_enabled);
-long long dynamic_debug_enabled2;
-EXPORT_SYMBOL_GPL(dynamic_debug_enabled2);
-
 struct ddebug_table {
 	struct list_head link;
 	char *mod_name;
@@ -87,26 +79,6 @@ static char *ddebug_describe_flags(struct _ddebug *dp, char *buf,
 	return buf;
 }
 
-/*
- * must be called with ddebug_lock held
- */
-
-static int disabled_hash(char hash, bool first_table)
-{
-	struct ddebug_table *dt;
-	char table_hash_value;
-
-	list_for_each_entry(dt, &ddebug_tables, link) {
-		if (first_table)
-			table_hash_value = dt->ddebugs->primary_hash;
-		else
-			table_hash_value = dt->ddebugs->secondary_hash;
-		if (dt->num_enabled && (hash == table_hash_value))
-			return 0;
-	}
-	return 1;
-}
-
 /*
  * Search the tables for _ddebug's which match the given
  * `query' and apply the `flags' and `mask' to them.  Tells
@@ -170,17 +142,9 @@ static void ddebug_change(const struct ddebug_query *query,
 				dt->num_enabled++;
 			dp->flags = newflags;
 			if (newflags) {
-				dynamic_debug_enabled |=
-						(1LL << dp->primary_hash);
-				dynamic_debug_enabled2 |=
-						(1LL << dp->secondary_hash);
+				enable_jump_label(&dp->enabled);
 			} else {
-				if (disabled_hash(dp->primary_hash, true))
-					dynamic_debug_enabled &=
-						~(1LL << dp->primary_hash);
-				if (disabled_hash(dp->secondary_hash, false))
-					dynamic_debug_enabled2 &=
-						~(1LL << dp->secondary_hash);
+				disable_jump_label(&dp->enabled);
 			}
 			if (verbose)
 				printk(KERN_INFO
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index 54fd1b700131..7bfcf1a09ac5 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -101,14 +101,6 @@ basename_flags = -D"KBUILD_BASENAME=KBUILD_STR($(call name-fix,$(basetarget)))"
 modname_flags  = $(if $(filter 1,$(words $(modname))),\
                  -D"KBUILD_MODNAME=KBUILD_STR($(call name-fix,$(modname)))")
 
-#hash values
-ifdef CONFIG_DYNAMIC_DEBUG
-debug_flags = -D"DEBUG_HASH=$(shell ./scripts/basic/hash djb2 $(@D)$(modname))"\
-              -D"DEBUG_HASH2=$(shell ./scripts/basic/hash r5 $(@D)$(modname))"
-else
-debug_flags =
-endif
-
 orig_c_flags   = $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) $(KBUILD_SUBDIR_CCFLAGS) \
                  $(ccflags-y) $(CFLAGS_$(basetarget).o)
 _c_flags       = $(filter-out $(CFLAGS_REMOVE_$(basetarget).o), $(orig_c_flags))
@@ -152,8 +144,7 @@ endif
 
 c_flags        = -Wp,-MD,$(depfile) $(NOSTDINC_FLAGS) $(LINUXINCLUDE)     \
 		 $(__c_flags) $(modkern_cflags)                           \
-		 -D"KBUILD_STR(s)=\#s" $(basename_flags) $(modname_flags) \
-		  $(debug_flags)
+		 -D"KBUILD_STR(s)=\#s" $(basename_flags) $(modname_flags)
 
 a_flags        = -Wp,-MD,$(depfile) $(NOSTDINC_FLAGS) $(LINUXINCLUDE)     \
 		 $(__a_flags) $(modkern_aflags)
diff --git a/scripts/basic/Makefile b/scripts/basic/Makefile
index 09559951df12..4c324a1f1e0e 100644
--- a/scripts/basic/Makefile
+++ b/scripts/basic/Makefile
@@ -9,7 +9,7 @@
 # fixdep: 	 Used to generate dependency information during build process
 # docproc:	 Used in Documentation/DocBook
 
-hostprogs-y	:= fixdep docproc hash
+hostprogs-y	:= fixdep docproc
 always		:= $(hostprogs-y)
 
 # fixdep is needed to compile other host programs
diff --git a/scripts/basic/hash.c b/scripts/basic/hash.c
deleted file mode 100644
index 2ef5d3f666b8..000000000000
--- a/scripts/basic/hash.c
+++ /dev/null
@@ -1,64 +0,0 @@
-/*
- * Copyright (C) 2008 Red Hat, Inc., Jason Baron <jbaron@redhat.com>
- *
- */
-
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-
-#define DYNAMIC_DEBUG_HASH_BITS 6
-
-static const char *program;
-
-static void usage(void)
-{
-	printf("Usage: %s <djb2|r5> <modname>\n", program);
-	exit(1);
-}
-
-/* djb2 hashing algorithm by Dan Bernstein. From:
- * http://www.cse.yorku.ca/~oz/hash.html
- */
-
-static unsigned int djb2_hash(char *str)
-{
-	unsigned long hash = 5381;
-	int c;
-
-	c = *str;
-	while (c) {
-		hash = ((hash << 5) + hash) + c;
-		c = *++str;
-	}
-	return (unsigned int)(hash & ((1 << DYNAMIC_DEBUG_HASH_BITS) - 1));
-}
-
-static unsigned int r5_hash(char *str)
-{
-	unsigned long hash = 0;
-	int c;
-
-	c = *str;
-	while (c) {
-		hash = (hash + (c << 4) + (c >> 4)) * 11;
-		c = *++str;
-	}
-	return (unsigned int)(hash & ((1 << DYNAMIC_DEBUG_HASH_BITS) - 1));
-}
-
-int main(int argc, char *argv[])
-{
-	program = argv[0];
-
-	if (argc != 3)
-		usage();
-	if (!strcmp(argv[1], "djb2"))
-		printf("%d\n", djb2_hash(argv[2]));
-	else if (!strcmp(argv[1], "r5"))
-		printf("%d\n", r5_hash(argv[2]));
-	else
-		usage();
-	exit(0);
-}
-
-- 
cgit v1.2.3


From d3f3cf859db17cc5f8156c5bfcd032413e44483b Mon Sep 17 00:00:00 2001
From: Mathieu Lacage <mathieu.lacage@sophia.inria.fr>
Date: Sat, 14 Aug 2010 15:02:44 +0200
Subject: missing inline keyword for static function in linux/dmaengine.h

Add a missing inline keyword for static function in linux/dmaengine.h to
avoid duplicate symbol definitions.

Signed-off-by: Mathieu Lacage <mathieu.lacage@sophia.inria.fr>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 include/linux/dmaengine.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index c61d4ca27bcc..e2106495cc11 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -548,7 +548,7 @@ static inline bool dma_dev_has_pq_continue(struct dma_device *dma)
 	return (dma->max_pq & DMA_HAS_PQ_CONTINUE) == DMA_HAS_PQ_CONTINUE;
 }
 
-static unsigned short dma_dev_to_maxpq(struct dma_device *dma)
+static inline unsigned short dma_dev_to_maxpq(struct dma_device *dma)
 {
 	return dma->max_pq & ~DMA_HAS_PQ_CONTINUE;
 }
-- 
cgit v1.2.3


From b3a084b9b684622b149e8dcf03855bf0d5fb588b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Wed, 22 Sep 2010 08:38:44 +0200
Subject: rcu: rcu_read_lock_bh_held(): disabling irqs also disables bh

rcu_dereference_bh() doesnt know yet about hard irq being disabled, so
lockdep can trigger in netpoll_rx() after commit f0f9deae9e7c4 (netpoll:
Disable IRQ around RCU dereference in netpoll_rx)

Reported-by: Miles Lane <miles.lane@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Tested-by: Miles Lane <miles.lane@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/rcupdate.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 9fbc54a2585d..83af1f8d8b74 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -454,7 +454,7 @@ static inline notrace void rcu_read_unlock_sched_notrace(void)
  * Makes rcu_dereference_check() do the dirty work.
  */
 #define rcu_dereference_bh(p) \
-		rcu_dereference_check(p, rcu_read_lock_bh_held())
+		rcu_dereference_check(p, rcu_read_lock_bh_held() || irqs_disabled())
 
 /**
  * rcu_dereference_sched - fetch RCU-protected pointer, checking for RCU-sched
-- 
cgit v1.2.3


From 53ecfba259f54b6967a35d19f4a564e3bc07997f Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Mon, 13 Sep 2010 17:24:21 -0700
Subject: rcu: only one evaluation of arg in rcu_dereference_check() unless
 sparse

The current version of the __rcu_access_pointer(), __rcu_dereference_check(),
and __rcu_dereference_protected() macros evaluate their "p" argument
three times, not counting typeof()s.  This is bad news if that argument
contains a side effect.  This commit therefore evaluates this argument
only once in normal kernel builds.  However, the straightforward approach
defeats sparse's RCU-pointer checking, so when __CHECKER__ is defined,
the additional pair of evaluations of the "p" argument are performed in
order to permit sparse to detect misuse of RCU-protected pointers.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
---
 include/linux/rcupdate.h | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 89414d67d961..03cda7bed985 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -310,24 +310,32 @@ extern int rcu_my_thread_group_empty(void);
  * (e.g., __rcu_bh, * __rcu_sched, and __srcu), should this make sense in
  * the future.
  */
+
+#ifdef __CHECKER__
+#define rcu_dereference_sparse(p, space) \
+	((void)(((typeof(*p) space *)p) == p))
+#else /* #ifdef __CHECKER__ */
+#define rcu_dereference_sparse(p, space)
+#endif /* #else #ifdef __CHECKER__ */
+
 #define __rcu_access_pointer(p, space) \
 	({ \
 		typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \
-		(void) (((typeof (*p) space *)p) == p); \
+		rcu_dereference_sparse(p, space); \
 		((typeof(*p) __force __kernel *)(_________p1)); \
 	})
 #define __rcu_dereference_check(p, c, space) \
 	({ \
 		typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \
 		rcu_lockdep_assert(c); \
-		(void) (((typeof (*p) space *)p) == p); \
+		rcu_dereference_sparse(p, space); \
 		smp_read_barrier_depends(); \
 		((typeof(*p) __force __kernel *)(_________p1)); \
 	})
 #define __rcu_dereference_protected(p, c, space) \
 	({ \
 		rcu_lockdep_assert(c); \
-		(void) (((typeof (*p) space *)p) == p); \
+		rcu_dereference_sparse(p, space); \
 		((typeof(*p) __force __kernel *)(p)); \
 	})
 
-- 
cgit v1.2.3


From d1ea13c6e2cce0106531852daaa93dd97aec9580 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 23 Sep 2010 18:40:07 +0200
Subject: genirq: Cleanup irq_chip->typename leftovers

3 years transition phase is enough. Cleanup the last users and remove
the cruft.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Leo Chen <leochen@broadcom.com>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Chris Zankel <chris@zankel.net>
---
 arch/arm/mach-bcmring/irq.c          | 6 +++---
 arch/m32r/kernel/irq.c               | 2 +-
 arch/m32r/platforms/m32104ut/setup.c | 2 +-
 arch/m32r/platforms/m32700ut/setup.c | 8 ++++----
 arch/m32r/platforms/mappi/setup.c    | 2 +-
 arch/m32r/platforms/mappi2/setup.c   | 2 +-
 arch/m32r/platforms/mappi3/setup.c   | 2 +-
 arch/m32r/platforms/oaks32r/setup.c  | 2 +-
 arch/m32r/platforms/opsput/setup.c   | 6 +++---
 arch/m32r/platforms/usrv/setup.c     | 4 ++--
 arch/tile/kernel/irq.c               | 4 ++--
 arch/um/kernel/irq.c                 | 6 +++---
 arch/xtensa/kernel/irq.c             | 2 +-
 include/linux/irq.h                  | 6 ------
 kernel/irq/chip.c                    | 2 --
 15 files changed, 24 insertions(+), 32 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/mach-bcmring/irq.c b/arch/arm/mach-bcmring/irq.c
index dc1c4939b0ce..e3152631eb37 100644
--- a/arch/arm/mach-bcmring/irq.c
+++ b/arch/arm/mach-bcmring/irq.c
@@ -67,21 +67,21 @@ static void bcmring_unmask_irq2(unsigned int irq)
 }
 
 static struct irq_chip bcmring_irq0_chip = {
-	.typename = "ARM-INTC0",
+	.name = "ARM-INTC0",
 	.ack = bcmring_mask_irq0,
 	.mask = bcmring_mask_irq0,	/* mask a specific interrupt, blocking its delivery. */
 	.unmask = bcmring_unmask_irq0,	/* unmaks an interrupt */
 };
 
 static struct irq_chip bcmring_irq1_chip = {
-	.typename = "ARM-INTC1",
+	.name = "ARM-INTC1",
 	.ack = bcmring_mask_irq1,
 	.mask = bcmring_mask_irq1,
 	.unmask = bcmring_unmask_irq1,
 };
 
 static struct irq_chip bcmring_irq2_chip = {
-	.typename = "ARM-SINTC",
+	.name = "ARM-SINTC",
 	.ack = bcmring_mask_irq2,
 	.mask = bcmring_mask_irq2,
 	.unmask = bcmring_unmask_irq2,
diff --git a/arch/m32r/kernel/irq.c b/arch/m32r/kernel/irq.c
index 3c71f776872c..7db26f1f082d 100644
--- a/arch/m32r/kernel/irq.c
+++ b/arch/m32r/kernel/irq.c
@@ -51,7 +51,7 @@ int show_interrupts(struct seq_file *p, void *v)
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 #endif
-		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, " %14s", irq_desc[i].chip->name);
 		seq_printf(p, "  %s", action->name);
 
 		for (action=action->next; action; action = action->next)
diff --git a/arch/m32r/platforms/m32104ut/setup.c b/arch/m32r/platforms/m32104ut/setup.c
index 922fdfdadeaa..402a59d7219b 100644
--- a/arch/m32r/platforms/m32104ut/setup.c
+++ b/arch/m32r/platforms/m32104ut/setup.c
@@ -65,7 +65,7 @@ static void shutdown_m32104ut_irq(unsigned int irq)
 
 static struct irq_chip m32104ut_irq_type =
 {
-	.typename = "M32104UT-IRQ",
+	.name = "M32104UT-IRQ",
 	.startup = startup_m32104ut_irq,
 	.shutdown = shutdown_m32104ut_irq,
 	.enable = enable_m32104ut_irq,
diff --git a/arch/m32r/platforms/m32700ut/setup.c b/arch/m32r/platforms/m32700ut/setup.c
index 9c1bc7487c1e..80b1a026795a 100644
--- a/arch/m32r/platforms/m32700ut/setup.c
+++ b/arch/m32r/platforms/m32700ut/setup.c
@@ -71,7 +71,7 @@ static void shutdown_m32700ut_irq(unsigned int irq)
 
 static struct irq_chip m32700ut_irq_type =
 {
-	.typename = "M32700UT-IRQ",
+	.name = "M32700UT-IRQ",
 	.startup = startup_m32700ut_irq,
 	.shutdown = shutdown_m32700ut_irq,
 	.enable = enable_m32700ut_irq,
@@ -148,7 +148,7 @@ static void shutdown_m32700ut_pld_irq(unsigned int irq)
 
 static struct irq_chip m32700ut_pld_irq_type =
 {
-	.typename = "M32700UT-PLD-IRQ",
+	.name = "M32700UT-PLD-IRQ",
 	.startup = startup_m32700ut_pld_irq,
 	.shutdown = shutdown_m32700ut_pld_irq,
 	.enable = enable_m32700ut_pld_irq,
@@ -217,7 +217,7 @@ static void shutdown_m32700ut_lanpld_irq(unsigned int irq)
 
 static struct irq_chip m32700ut_lanpld_irq_type =
 {
-	.typename = "M32700UT-PLD-LAN-IRQ",
+	.name = "M32700UT-PLD-LAN-IRQ",
 	.startup = startup_m32700ut_lanpld_irq,
 	.shutdown = shutdown_m32700ut_lanpld_irq,
 	.enable = enable_m32700ut_lanpld_irq,
@@ -286,7 +286,7 @@ static void shutdown_m32700ut_lcdpld_irq(unsigned int irq)
 
 static struct irq_chip m32700ut_lcdpld_irq_type =
 {
-	.typename = "M32700UT-PLD-LCD-IRQ",
+	.name = "M32700UT-PLD-LCD-IRQ",
 	.startup = startup_m32700ut_lcdpld_irq,
 	.shutdown = shutdown_m32700ut_lcdpld_irq,
 	.enable = enable_m32700ut_lcdpld_irq,
diff --git a/arch/m32r/platforms/mappi/setup.c b/arch/m32r/platforms/mappi/setup.c
index fb4b17799b66..ea00c84d6b1b 100644
--- a/arch/m32r/platforms/mappi/setup.c
+++ b/arch/m32r/platforms/mappi/setup.c
@@ -65,7 +65,7 @@ static void shutdown_mappi_irq(unsigned int irq)
 
 static struct irq_chip mappi_irq_type =
 {
-	.typename = "MAPPI-IRQ",
+	.name = "MAPPI-IRQ",
 	.startup = startup_mappi_irq,
 	.shutdown = shutdown_mappi_irq,
 	.enable = enable_mappi_irq,
diff --git a/arch/m32r/platforms/mappi2/setup.c b/arch/m32r/platforms/mappi2/setup.c
index 6a65eda0a056..c049376d0270 100644
--- a/arch/m32r/platforms/mappi2/setup.c
+++ b/arch/m32r/platforms/mappi2/setup.c
@@ -72,7 +72,7 @@ static void shutdown_mappi2_irq(unsigned int irq)
 
 static struct irq_chip mappi2_irq_type =
 {
-	.typename = "MAPPI2-IRQ",
+	.name = "MAPPI2-IRQ",
 	.startup = startup_mappi2_irq,
 	.shutdown = shutdown_mappi2_irq,
 	.enable = enable_mappi2_irq,
diff --git a/arch/m32r/platforms/mappi3/setup.c b/arch/m32r/platforms/mappi3/setup.c
index 9c337aeac94b..882de25c6e8c 100644
--- a/arch/m32r/platforms/mappi3/setup.c
+++ b/arch/m32r/platforms/mappi3/setup.c
@@ -72,7 +72,7 @@ static void shutdown_mappi3_irq(unsigned int irq)
 
 static struct irq_chip mappi3_irq_type =
 {
-	.typename = "MAPPI3-IRQ",
+	.name = "MAPPI3-IRQ",
 	.startup = startup_mappi3_irq,
 	.shutdown = shutdown_mappi3_irq,
 	.enable = enable_mappi3_irq,
diff --git a/arch/m32r/platforms/oaks32r/setup.c b/arch/m32r/platforms/oaks32r/setup.c
index ed865741c38d..d11d93bf74f5 100644
--- a/arch/m32r/platforms/oaks32r/setup.c
+++ b/arch/m32r/platforms/oaks32r/setup.c
@@ -63,7 +63,7 @@ static void shutdown_oaks32r_irq(unsigned int irq)
 
 static struct irq_chip oaks32r_irq_type =
 {
-	.typename = "OAKS32R-IRQ",
+	.name = "OAKS32R-IRQ",
 	.startup = startup_oaks32r_irq,
 	.shutdown = shutdown_oaks32r_irq,
 	.enable = enable_oaks32r_irq,
diff --git a/arch/m32r/platforms/opsput/setup.c b/arch/m32r/platforms/opsput/setup.c
index 80d680657019..5f3402a2fbaf 100644
--- a/arch/m32r/platforms/opsput/setup.c
+++ b/arch/m32r/platforms/opsput/setup.c
@@ -72,7 +72,7 @@ static void shutdown_opsput_irq(unsigned int irq)
 
 static struct irq_chip opsput_irq_type =
 {
-	.typename = "OPSPUT-IRQ",
+	.name = "OPSPUT-IRQ",
 	.startup = startup_opsput_irq,
 	.shutdown = shutdown_opsput_irq,
 	.enable = enable_opsput_irq,
@@ -149,7 +149,7 @@ static void shutdown_opsput_pld_irq(unsigned int irq)
 
 static struct irq_chip opsput_pld_irq_type =
 {
-	.typename = "OPSPUT-PLD-IRQ",
+	.name = "OPSPUT-PLD-IRQ",
 	.startup = startup_opsput_pld_irq,
 	.shutdown = shutdown_opsput_pld_irq,
 	.enable = enable_opsput_pld_irq,
@@ -218,7 +218,7 @@ static void shutdown_opsput_lanpld_irq(unsigned int irq)
 
 static struct irq_chip opsput_lanpld_irq_type =
 {
-	.typename = "OPSPUT-PLD-LAN-IRQ",
+	.name = "OPSPUT-PLD-LAN-IRQ",
 	.startup = startup_opsput_lanpld_irq,
 	.shutdown = shutdown_opsput_lanpld_irq,
 	.enable = enable_opsput_lanpld_irq,
diff --git a/arch/m32r/platforms/usrv/setup.c b/arch/m32r/platforms/usrv/setup.c
index 757302660af8..1beac7a51ed4 100644
--- a/arch/m32r/platforms/usrv/setup.c
+++ b/arch/m32r/platforms/usrv/setup.c
@@ -63,7 +63,7 @@ static void shutdown_mappi_irq(unsigned int irq)
 
 static struct irq_chip mappi_irq_type =
 {
-	.typename = "M32700-IRQ",
+	.name = "M32700-IRQ",
 	.startup = startup_mappi_irq,
 	.shutdown = shutdown_mappi_irq,
 	.enable = enable_mappi_irq,
@@ -136,7 +136,7 @@ static void shutdown_m32700ut_pld_irq(unsigned int irq)
 
 static struct irq_chip m32700ut_pld_irq_type =
 {
-	.typename = "USRV-PLD-IRQ",
+	.name = "USRV-PLD-IRQ",
 	.startup = startup_m32700ut_pld_irq,
 	.shutdown = shutdown_m32700ut_pld_irq,
 	.enable = enable_m32700ut_pld_irq,
diff --git a/arch/tile/kernel/irq.c b/arch/tile/kernel/irq.c
index 596c60086930..9a27d563fc30 100644
--- a/arch/tile/kernel/irq.c
+++ b/arch/tile/kernel/irq.c
@@ -208,7 +208,7 @@ static void tile_irq_chip_eoi(unsigned int irq)
 }
 
 static struct irq_chip tile_irq_chip = {
-	.typename = "tile_irq_chip",
+	.name = "tile_irq_chip",
 	.ack = tile_irq_chip_ack,
 	.eoi = tile_irq_chip_eoi,
 	.mask = tile_irq_chip_mask,
@@ -288,7 +288,7 @@ int show_interrupts(struct seq_file *p, void *v)
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 #endif
-		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, " %14s", irq_desc[i].chip->name);
 		seq_printf(p, "  %s", action->name);
 
 		for (action = action->next; action; action = action->next)
diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c
index a3f0b04d7101..a746e3037a5b 100644
--- a/arch/um/kernel/irq.c
+++ b/arch/um/kernel/irq.c
@@ -46,7 +46,7 @@ int show_interrupts(struct seq_file *p, void *v)
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 #endif
-		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, " %14s", irq_desc[i].chip->name);
 		seq_printf(p, "  %s", action->name);
 
 		for (action=action->next; action; action = action->next)
@@ -369,7 +369,7 @@ static void dummy(unsigned int irq)
 
 /* This is used for everything else than the timer. */
 static struct irq_chip normal_irq_type = {
-	.typename = "SIGIO",
+	.name = "SIGIO",
 	.release = free_irq_by_irq_and_dev,
 	.disable = dummy,
 	.enable = dummy,
@@ -378,7 +378,7 @@ static struct irq_chip normal_irq_type = {
 };
 
 static struct irq_chip SIGVTALRM_irq_type = {
-	.typename = "SIGVTALRM",
+	.name = "SIGVTALRM",
 	.release = free_irq_by_irq_and_dev,
 	.shutdown = dummy, /* never called */
 	.disable = dummy,
diff --git a/arch/xtensa/kernel/irq.c b/arch/xtensa/kernel/irq.c
index c64a5d387de5..87508886cbbd 100644
--- a/arch/xtensa/kernel/irq.c
+++ b/arch/xtensa/kernel/irq.c
@@ -92,7 +92,7 @@ int show_interrupts(struct seq_file *p, void *v)
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 #endif
-		seq_printf(p, " %14s", irq_desc[i].chip->typename);
+		seq_printf(p, " %14s", irq_desc[i].chip->name);
 		seq_printf(p, "  %s", action->name);
 
 		for (action=action->next; action; action = action->next)
diff --git a/include/linux/irq.h b/include/linux/irq.h
index c03243ad84b4..06273a2a17e7 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -106,7 +106,6 @@ struct msi_desc;
  * @bus_sync_unlock:	function to sync and unlock slow bus (i2c) chips
  *
  * @release:		release function solely used by UML
- * @typename:		obsoleted by name, kept as migration helper
  */
 struct irq_chip {
 	const char	*name;
@@ -135,11 +134,6 @@ struct irq_chip {
 #ifdef CONFIG_IRQ_RELEASE_METHOD
 	void		(*release)(unsigned int irq, void *dev_id);
 #endif
-	/*
-	 * For compatibility, ->typename is copied into ->name.
-	 * Will disappear.
-	 */
-	const char	*typename;
 };
 
 struct timer_rand_state;
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index b7091d5ca2f8..4ea775cc60f0 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -344,8 +344,6 @@ void irq_chip_set_defaults(struct irq_chip *chip)
 	if (!chip->shutdown)
 		chip->shutdown = chip->disable != default_disable ?
 			chip->disable : default_shutdown;
-	if (!chip->name)
-		chip->name = chip->typename;
 	if (!chip->end)
 		chip->end = dummy_irq_chip.end;
 }
-- 
cgit v1.2.3


From a02cec2155fbea457eca8881870fd2de1a4c4c76 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Wed, 22 Sep 2010 20:43:57 +0000
Subject: net: return operator cleanup

Change "return (EXPR);" to "return EXPR;"

return is not a function, parentheses are not required.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/atmdev.h                  |  2 +-
 include/linux/etherdevice.h             |  4 +--
 include/linux/netdevice.h               |  2 +-
 include/linux/skbuff.h                  |  6 ++---
 include/net/bluetooth/hci_core.h        |  2 +-
 include/net/bluetooth/l2cap.h           |  2 +-
 include/net/inet_ecn.h                  |  2 +-
 include/net/ip.h                        |  4 +--
 include/net/ipv6.h                      | 35 +++++++++++++-------------
 include/net/irda/irlap.h                |  2 +-
 include/net/irda/irlmp.h                |  2 +-
 include/net/irda/irttp.h                |  2 +-
 include/net/sch_generic.h               |  2 +-
 include/net/sctp/sctp.h                 | 12 ++++-----
 include/net/sctp/sm.h                   | 10 ++++----
 include/net/sctp/structs.h              |  2 +-
 include/net/sctp/tsnmap.h               |  2 +-
 include/net/tipc/tipc_msg.h             | 10 ++++----
 net/802/fc.c                            |  2 +-
 net/802/fddi.c                          | 12 ++++-----
 net/802/hippi.c                         |  2 +-
 net/802/tr.c                            |  2 +-
 net/8021q/vlan_core.c                   |  2 +-
 net/9p/client.c                         |  4 +--
 net/bluetooth/rfcomm/core.c             |  4 +--
 net/core/flow.c                         |  4 +--
 net/core/neighbour.c                    |  6 ++---
 net/core/utils.c                        |  2 +-
 net/dccp/ccids/lib/loss_interval.c      |  2 +-
 net/econet/af_econet.c                  |  4 +--
 net/ethernet/eth.c                      |  2 +-
 net/ipv4/arp.c                          |  2 +-
 net/ipv4/datagram.c                     |  2 +-
 net/ipv4/inet_diag.c                    |  2 +-
 net/ipv4/ip_fragment.c                  |  4 +--
 net/ipv4/ip_gre.c                       |  2 +-
 net/ipv4/netfilter/arp_tables.c         |  2 +-
 net/ipv4/route.c                        |  2 +-
 net/ipv4/tcp_input.c                    | 10 ++++----
 net/ipv4/tcp_minisocks.c                |  2 +-
 net/ipv4/tcp_output.c                   |  8 +++---
 net/ipv4/tcp_westwood.c                 |  2 +-
 net/ipv6/addrconf.c                     |  2 +-
 net/ipv6/addrlabel.c                    |  5 ++--
 net/ipv6/af_inet6.c                     |  6 ++---
 net/ipv6/exthdrs_core.c                 |  4 +--
 net/ipv6/ip6_output.c                   |  4 +--
 net/ipv6/ndisc.c                        |  8 +++---
 net/ipv6/netfilter/ip6_tables.c         | 14 +++++------
 net/ipv6/raw.c                          | 12 ++++-----
 net/ipv6/route.c                        | 14 +++++------
 net/ipv6/tcp_ipv6.c                     |  2 +-
 net/ipv6/xfrm6_policy.c                 |  2 +-
 net/irda/af_irda.c                      | 14 +++++------
 net/irda/discovery.c                    |  2 +-
 net/irda/ircomm/ircomm_tty.c            |  4 +--
 net/irda/irlmp.c                        |  2 +-
 net/irda/irlmp_frame.c                  |  2 +-
 net/irda/irnet/irnet_irda.c             | 22 ++++++++---------
 net/irda/irnet/irnet_ppp.c              |  8 +++---
 net/key/af_key.c                        |  4 +--
 net/mac80211/rate.c                     |  2 +-
 net/rfkill/input.c                      |  2 +-
 net/rose/rose_link.c                    |  4 +--
 net/sctp/protocol.c                     |  2 +-
 net/sctp/socket.c                       |  6 ++---
 net/sunrpc/auth_gss/auth_gss.c          |  2 +-
 net/sunrpc/auth_gss/gss_generic_token.c | 44 ++++++++++++++++-----------------
 net/sunrpc/auth_gss/gss_krb5_seqnum.c   |  2 +-
 net/sunrpc/auth_gss/gss_mech_switch.c   |  2 +-
 net/sunrpc/sched.c                      |  2 +-
 net/tipc/addr.c                         |  2 +-
 net/tipc/bcast.c                        |  2 +-
 net/tipc/bearer.c                       |  2 +-
 net/tipc/dbg.c                          |  4 +--
 net/tipc/link.c                         |  6 ++---
 net/tipc/link.h                         | 16 ++++++------
 net/tipc/msg.h                          |  6 ++---
 net/tipc/name_table.c                   |  2 +-
 net/tipc/node.c                         |  6 ++---
 net/tipc/port.h                         |  2 +-
 net/tipc/socket.c                       |  2 +-
 net/tipc/subscr.c                       |  2 +-
 net/wireless/core.h                     |  2 +-
 84 files changed, 220 insertions(+), 222 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/atmdev.h b/include/linux/atmdev.h
index f6481daf6e52..a8e4e832cdbb 100644
--- a/include/linux/atmdev.h
+++ b/include/linux/atmdev.h
@@ -449,7 +449,7 @@ void vcc_insert_socket(struct sock *sk);
 
 static inline int atm_guess_pdu2truesize(int size)
 {
-	return (SKB_DATA_ALIGN(size) + sizeof(struct skb_shared_info));
+	return SKB_DATA_ALIGN(size) + sizeof(struct skb_shared_info);
 }
 
 
diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h
index fb6aa6070921..f16a01081e15 100644
--- a/include/linux/etherdevice.h
+++ b/include/linux/etherdevice.h
@@ -71,7 +71,7 @@ static inline int is_zero_ether_addr(const u8 *addr)
  */
 static inline int is_multicast_ether_addr(const u8 *addr)
 {
-	return (0x01 & addr[0]);
+	return 0x01 & addr[0];
 }
 
 /**
@@ -82,7 +82,7 @@ static inline int is_multicast_ether_addr(const u8 *addr)
  */
 static inline int is_local_ether_addr(const u8 *addr)
 {
-	return (0x02 & addr[0]);
+	return 0x02 & addr[0];
 }
 
 /**
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f7f1302138af..45dcda5bfda9 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1676,7 +1676,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
  */
 static inline int netif_is_multiqueue(const struct net_device *dev)
 {
-	return (dev->num_tx_queues > 1);
+	return dev->num_tx_queues > 1;
 }
 
 extern void netif_set_real_num_tx_queues(struct net_device *dev,
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 9e8085a89589..b2c41d19735c 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -601,7 +601,7 @@ static inline int skb_queue_empty(const struct sk_buff_head *list)
 static inline bool skb_queue_is_last(const struct sk_buff_head *list,
 				     const struct sk_buff *skb)
 {
-	return (skb->next == (struct sk_buff *) list);
+	return skb->next == (struct sk_buff *)list;
 }
 
 /**
@@ -614,7 +614,7 @@ static inline bool skb_queue_is_last(const struct sk_buff_head *list,
 static inline bool skb_queue_is_first(const struct sk_buff_head *list,
 				      const struct sk_buff *skb)
 {
-	return (skb->prev == (struct sk_buff *) list);
+	return skb->prev == (struct sk_buff *)list;
 }
 
 /**
@@ -2156,7 +2156,7 @@ static inline u16 skb_get_rx_queue(const struct sk_buff *skb)
 
 static inline bool skb_rx_queue_recorded(const struct sk_buff *skb)
 {
-	return (skb->queue_mapping != 0);
+	return skb->queue_mapping != 0;
 }
 
 extern u16 skb_tx_hash(const struct net_device *dev,
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index 4568b938ca35..ebec8c9a929d 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -233,7 +233,7 @@ static inline void inquiry_cache_init(struct hci_dev *hdev)
 static inline int inquiry_cache_empty(struct hci_dev *hdev)
 {
 	struct inquiry_cache *c = &hdev->inq_cache;
-	return (c->list == NULL);
+	return c->list == NULL;
 }
 
 static inline long inquiry_cache_age(struct hci_dev *hdev)
diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h
index 6c241444f902..c819c8bf9b68 100644
--- a/include/net/bluetooth/l2cap.h
+++ b/include/net/bluetooth/l2cap.h
@@ -414,7 +414,7 @@ static inline int l2cap_tx_window_full(struct sock *sk)
 	if (sub < 0)
 		sub += 64;
 
-	return (sub == pi->remote_tx_win);
+	return sub == pi->remote_tx_win;
 }
 
 #define __get_txseq(ctrl) ((ctrl) & L2CAP_CTRL_TXSEQ) >> 1
diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h
index 9b5d08f4f6e8..88bdd010d65d 100644
--- a/include/net/inet_ecn.h
+++ b/include/net/inet_ecn.h
@@ -27,7 +27,7 @@ static inline int INET_ECN_is_not_ect(__u8 dsfield)
 
 static inline int INET_ECN_is_capable(__u8 dsfield)
 {
-	return (dsfield & INET_ECN_ECT_0);
+	return dsfield & INET_ECN_ECT_0;
 }
 
 static inline __u8 INET_ECN_encapsulate(__u8 outer, __u8 inner)
diff --git a/include/net/ip.h b/include/net/ip.h
index 7691aca133db..dbee3fe260e1 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -238,9 +238,9 @@ int ip_decrease_ttl(struct iphdr *iph)
 static inline
 int ip_dont_fragment(struct sock *sk, struct dst_entry *dst)
 {
-	return (inet_sk(sk)->pmtudisc == IP_PMTUDISC_DO ||
+	return  inet_sk(sk)->pmtudisc == IP_PMTUDISC_DO ||
 		(inet_sk(sk)->pmtudisc == IP_PMTUDISC_WANT &&
-		 !(dst_metric_locked(dst, RTAX_MTU))));
+		 !(dst_metric_locked(dst, RTAX_MTU)));
 }
 
 extern void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more);
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 1f8412410998..4a3cd2cd2f5e 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -262,7 +262,7 @@ static inline int ipv6_addr_scope(const struct in6_addr *addr)
 
 static inline int __ipv6_addr_src_scope(int type)
 {
-	return (type == IPV6_ADDR_ANY ? __IPV6_ADDR_SCOPE_INVALID : (type >> 16));
+	return (type == IPV6_ADDR_ANY) ? __IPV6_ADDR_SCOPE_INVALID : (type >> 16);
 }
 
 static inline int ipv6_addr_src_scope(const struct in6_addr *addr)
@@ -279,10 +279,10 @@ static inline int
 ipv6_masked_addr_cmp(const struct in6_addr *a1, const struct in6_addr *m,
 		     const struct in6_addr *a2)
 {
-	return (!!(((a1->s6_addr32[0] ^ a2->s6_addr32[0]) & m->s6_addr32[0]) |
-		   ((a1->s6_addr32[1] ^ a2->s6_addr32[1]) & m->s6_addr32[1]) |
-		   ((a1->s6_addr32[2] ^ a2->s6_addr32[2]) & m->s6_addr32[2]) |
-		   ((a1->s6_addr32[3] ^ a2->s6_addr32[3]) & m->s6_addr32[3])));
+	return !!(((a1->s6_addr32[0] ^ a2->s6_addr32[0]) & m->s6_addr32[0]) |
+		  ((a1->s6_addr32[1] ^ a2->s6_addr32[1]) & m->s6_addr32[1]) |
+		  ((a1->s6_addr32[2] ^ a2->s6_addr32[2]) & m->s6_addr32[2]) |
+		  ((a1->s6_addr32[3] ^ a2->s6_addr32[3]) & m->s6_addr32[3]));
 }
 
 static inline void ipv6_addr_copy(struct in6_addr *a1, const struct in6_addr *a2)
@@ -317,10 +317,10 @@ static inline void ipv6_addr_set(struct in6_addr *addr,
 static inline int ipv6_addr_equal(const struct in6_addr *a1,
 				  const struct in6_addr *a2)
 {
-	return (((a1->s6_addr32[0] ^ a2->s6_addr32[0]) |
-		 (a1->s6_addr32[1] ^ a2->s6_addr32[1]) |
-		 (a1->s6_addr32[2] ^ a2->s6_addr32[2]) |
-		 (a1->s6_addr32[3] ^ a2->s6_addr32[3])) == 0);
+	return ((a1->s6_addr32[0] ^ a2->s6_addr32[0]) |
+		(a1->s6_addr32[1] ^ a2->s6_addr32[1]) |
+		(a1->s6_addr32[2] ^ a2->s6_addr32[2]) |
+		(a1->s6_addr32[3] ^ a2->s6_addr32[3])) == 0;
 }
 
 static inline int __ipv6_prefix_equal(const __be32 *a1, const __be32 *a2,
@@ -373,20 +373,20 @@ int ip6_frag_match(struct inet_frag_queue *q, void *a);
 
 static inline int ipv6_addr_any(const struct in6_addr *a)
 {
-	return ((a->s6_addr32[0] | a->s6_addr32[1] | 
-		 a->s6_addr32[2] | a->s6_addr32[3] ) == 0); 
+	return (a->s6_addr32[0] | a->s6_addr32[1] |
+		a->s6_addr32[2] | a->s6_addr32[3]) == 0;
 }
 
 static inline int ipv6_addr_loopback(const struct in6_addr *a)
 {
-	return ((a->s6_addr32[0] | a->s6_addr32[1] |
-		 a->s6_addr32[2] | (a->s6_addr32[3] ^ htonl(1))) == 0);
+	return (a->s6_addr32[0] | a->s6_addr32[1] |
+		a->s6_addr32[2] | (a->s6_addr32[3] ^ htonl(1))) == 0;
 }
 
 static inline int ipv6_addr_v4mapped(const struct in6_addr *a)
 {
-	return ((a->s6_addr32[0] | a->s6_addr32[1] |
-		 (a->s6_addr32[2] ^ htonl(0x0000ffff))) == 0);
+	return (a->s6_addr32[0] | a->s6_addr32[1] |
+		 (a->s6_addr32[2] ^ htonl(0x0000ffff))) == 0;
 }
 
 /*
@@ -395,8 +395,7 @@ static inline int ipv6_addr_v4mapped(const struct in6_addr *a)
  */
 static inline int ipv6_addr_orchid(const struct in6_addr *a)
 {
-	return ((a->s6_addr32[0] & htonl(0xfffffff0))
-		== htonl(0x20010010));
+	return (a->s6_addr32[0] & htonl(0xfffffff0)) == htonl(0x20010010);
 }
 
 static inline void ipv6_addr_set_v4mapped(const __be32 addr,
@@ -441,7 +440,7 @@ static inline int __ipv6_addr_diff(const void *token1, const void *token2, int a
 	 *	if returned value is greater than prefix length.
 	 *					--ANK (980803)
 	 */
-	return (addrlen << 5);
+	return addrlen << 5;
 }
 
 static inline int ipv6_addr_diff(const struct in6_addr *a1, const struct in6_addr *a2)
diff --git a/include/net/irda/irlap.h b/include/net/irda/irlap.h
index 9d0c78ea92f5..17fcd964f9d9 100644
--- a/include/net/irda/irlap.h
+++ b/include/net/irda/irlap.h
@@ -282,7 +282,7 @@ static inline int irlap_is_primary(struct irlap_cb *self)
 	default:
 		ret = -1;
 	}
-	return(ret);
+	return ret;
 }
 
 /* Clear a pending IrLAP disconnect. - Jean II */
diff --git a/include/net/irda/irlmp.h b/include/net/irda/irlmp.h
index 3ffc1d0f93d6..fff11b7fe8a4 100644
--- a/include/net/irda/irlmp.h
+++ b/include/net/irda/irlmp.h
@@ -274,7 +274,7 @@ static inline int irlmp_lap_tx_queue_full(struct lsap_cb *self)
 	if (self->lap->irlap == NULL)
 		return 0;
 
-	return(IRLAP_GET_TX_QUEUE_LEN(self->lap->irlap) >= LAP_HIGH_THRESHOLD);
+	return IRLAP_GET_TX_QUEUE_LEN(self->lap->irlap) >= LAP_HIGH_THRESHOLD;
 }
 
 /* After doing a irlmp_dup(), this get one of the two socket back into
diff --git a/include/net/irda/irttp.h b/include/net/irda/irttp.h
index 11aee7a2972a..af4b87721d13 100644
--- a/include/net/irda/irttp.h
+++ b/include/net/irda/irttp.h
@@ -204,7 +204,7 @@ static inline int irttp_is_primary(struct tsap_cb *self)
 	    (self->lsap->lap == NULL) ||
 	    (self->lsap->lap->irlap == NULL))
 		return -2;
-	return(irlap_is_primary(self->lsap->lap->irlap));
+	return irlap_is_primary(self->lsap->lap->irlap);
 }
 
 #endif /* IRTTP_H */
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 3c8728aaab4e..eda8808fdacd 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -601,7 +601,7 @@ static inline u32 qdisc_l2t(struct qdisc_rate_table* rtab, unsigned int pktlen)
 		slot = 0;
 	slot >>= rtab->rate.cell_log;
 	if (slot > 255)
-		return (rtab->data[255]*(slot >> 8) + rtab->data[slot & 0xFF]);
+		return rtab->data[255]*(slot >> 8) + rtab->data[slot & 0xFF];
 	return rtab->data[slot];
 }
 
diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 2cb3980b1616..505845ddb0be 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -405,7 +405,7 @@ static inline void sctp_v6_del_protocol(void) { return; }
 /* Map an association to an assoc_id. */
 static inline sctp_assoc_t sctp_assoc2id(const struct sctp_association *asoc)
 {
-	return (asoc?asoc->assoc_id:0);
+	return asoc ? asoc->assoc_id : 0;
 }
 
 /* Look up the association by its id.  */
@@ -473,7 +473,7 @@ static inline void sctp_skb_set_owner_r(struct sk_buff *skb, struct sock *sk)
 /* Tests if the list has one and only one entry. */
 static inline int sctp_list_single_entry(struct list_head *head)
 {
-	return ((head->next != head) && (head->next == head->prev));
+	return (head->next != head) && (head->next == head->prev);
 }
 
 /* Generate a random jitter in the range of -50% ~ +50% of input RTO. */
@@ -631,13 +631,13 @@ static inline int sctp_sanity_check(void)
 /* This is the hash function for the SCTP port hash table. */
 static inline int sctp_phashfn(__u16 lport)
 {
-	return (lport & (sctp_port_hashsize - 1));
+	return lport & (sctp_port_hashsize - 1);
 }
 
 /* This is the hash function for the endpoint hash table. */
 static inline int sctp_ep_hashfn(__u16 lport)
 {
-	return (lport & (sctp_ep_hashsize - 1));
+	return lport & (sctp_ep_hashsize - 1);
 }
 
 /* This is the hash function for the association hash table. */
@@ -645,7 +645,7 @@ static inline int sctp_assoc_hashfn(__u16 lport, __u16 rport)
 {
 	int h = (lport << 16) + rport;
 	h ^= h>>8;
-	return (h & (sctp_assoc_hashsize - 1));
+	return h & (sctp_assoc_hashsize - 1);
 }
 
 /* This is the hash function for the association hash table.  This is
@@ -656,7 +656,7 @@ static inline int sctp_vtag_hashfn(__u16 lport, __u16 rport, __u32 vtag)
 {
 	int h = (lport << 16) + rport;
 	h ^= vtag;
-	return (h & (sctp_assoc_hashsize-1));
+	return h & (sctp_assoc_hashsize - 1);
 }
 
 #define sctp_for_each_hentry(epb, node, head) \
diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h
index 4088c89a9055..9352d12f02de 100644
--- a/include/net/sctp/sm.h
+++ b/include/net/sctp/sm.h
@@ -345,12 +345,12 @@ enum {
 
 static inline int TSN_lt(__u32 s, __u32 t)
 {
-	return (((s) - (t)) & TSN_SIGN_BIT);
+	return ((s) - (t)) & TSN_SIGN_BIT;
 }
 
 static inline int TSN_lte(__u32 s, __u32 t)
 {
-	return (((s) == (t)) || (((s) - (t)) & TSN_SIGN_BIT));
+	return ((s) == (t)) || (((s) - (t)) & TSN_SIGN_BIT);
 }
 
 /* Compare two SSNs */
@@ -369,12 +369,12 @@ enum {
 
 static inline int SSN_lt(__u16 s, __u16 t)
 {
-	return (((s) - (t)) & SSN_SIGN_BIT);
+	return ((s) - (t)) & SSN_SIGN_BIT;
 }
 
 static inline int SSN_lte(__u16 s, __u16 t)
 {
-	return (((s) == (t)) || (((s) - (t)) & SSN_SIGN_BIT));
+	return ((s) == (t)) || (((s) - (t)) & SSN_SIGN_BIT);
 }
 
 /*
@@ -388,7 +388,7 @@ enum {
 
 static inline int ADDIP_SERIAL_gte(__u16 s, __u16 t)
 {
-	return (((s) == (t)) || (((t) - (s)) & ADDIP_SERIAL_SIGN_BIT));
+	return ((s) == (t)) || (((t) - (s)) & ADDIP_SERIAL_SIGN_BIT);
 }
 
 /* Check VTAG of the packet matches the sender's own tag. */
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index f9e7473613bd..69fef4fb79c0 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -847,7 +847,7 @@ void sctp_packet_free(struct sctp_packet *);
 
 static inline int sctp_packet_empty(struct sctp_packet *packet)
 {
-	return (packet->size == packet->overhead);
+	return packet->size == packet->overhead;
 }
 
 /* This represents a remote transport address.
diff --git a/include/net/sctp/tsnmap.h b/include/net/sctp/tsnmap.h
index 4aabc5a96cf6..e7728bc14ccf 100644
--- a/include/net/sctp/tsnmap.h
+++ b/include/net/sctp/tsnmap.h
@@ -157,7 +157,7 @@ __u16 sctp_tsnmap_pending(struct sctp_tsnmap *map);
 /* Is there a gap in the TSN map?  */
 static inline int sctp_tsnmap_has_gap(const struct sctp_tsnmap *map)
 {
-	return (map->cumulative_tsn_ack_point != map->max_tsn_seen);
+	return map->cumulative_tsn_ack_point != map->max_tsn_seen;
 }
 
 /* Mark a duplicate TSN.  Note:  limit the storage of duplicate TSN
diff --git a/include/net/tipc/tipc_msg.h b/include/net/tipc/tipc_msg.h
index 2e159a812f83..ffe50b4e7b93 100644
--- a/include/net/tipc/tipc_msg.h
+++ b/include/net/tipc/tipc_msg.h
@@ -107,7 +107,7 @@ static inline u32 msg_hdr_sz(struct tipc_msg *m)
 
 static inline int msg_short(struct tipc_msg *m)
 {
-	return (msg_hdr_sz(m) == 24);
+	return msg_hdr_sz(m) == 24;
 }
 
 static inline u32 msg_size(struct tipc_msg *m)
@@ -117,7 +117,7 @@ static inline u32 msg_size(struct tipc_msg *m)
 
 static inline u32 msg_data_sz(struct tipc_msg *m)
 {
-	return (msg_size(m) - msg_hdr_sz(m));
+	return msg_size(m) - msg_hdr_sz(m);
 }
 
 static inline unchar *msg_data(struct tipc_msg *m)
@@ -132,17 +132,17 @@ static inline u32 msg_type(struct tipc_msg *m)
 
 static inline u32 msg_named(struct tipc_msg *m)
 {
-	return (msg_type(m) == TIPC_NAMED_MSG);
+	return msg_type(m) == TIPC_NAMED_MSG;
 }
 
 static inline u32 msg_mcast(struct tipc_msg *m)
 {
-	return (msg_type(m) == TIPC_MCAST_MSG);
+	return msg_type(m) == TIPC_MCAST_MSG;
 }
 
 static inline u32 msg_connected(struct tipc_msg *m)
 {
-	return (msg_type(m) == TIPC_CONN_MSG);
+	return msg_type(m) == TIPC_CONN_MSG;
 }
 
 static inline u32 msg_errcode(struct tipc_msg *m)
diff --git a/net/802/fc.c b/net/802/fc.c
index 34cf1ee014b8..1e49f2d4ea96 100644
--- a/net/802/fc.c
+++ b/net/802/fc.c
@@ -70,7 +70,7 @@ static int fc_header(struct sk_buff *skb, struct net_device *dev,
 	if(daddr)
 	{
 		memcpy(fch->daddr,daddr,dev->addr_len);
-		return(hdr_len);
+		return hdr_len;
 	}
 	return -hdr_len;
 }
diff --git a/net/802/fddi.c b/net/802/fddi.c
index 3ef0ab0a543a..94b3ad08f39a 100644
--- a/net/802/fddi.c
+++ b/net/802/fddi.c
@@ -82,10 +82,10 @@ static int fddi_header(struct sk_buff *skb, struct net_device *dev,
 	if (daddr != NULL)
 	{
 		memcpy(fddi->daddr, daddr, dev->addr_len);
-		return(hl);
+		return hl;
 	}
 
-	return(-hl);
+	return -hl;
 }
 
 
@@ -108,7 +108,7 @@ static int fddi_rebuild_header(struct sk_buff	*skb)
 	{
 		printk("%s: Don't know how to resolve type %04X addresses.\n",
 		       skb->dev->name, ntohs(fddi->hdr.llc_snap.ethertype));
-		return(0);
+		return 0;
 	}
 }
 
@@ -162,7 +162,7 @@ __be16 fddi_type_trans(struct sk_buff *skb, struct net_device *dev)
 
 	/* Assume 802.2 SNAP frames, for now */
 
-	return(type);
+	return type;
 }
 
 EXPORT_SYMBOL(fddi_type_trans);
@@ -170,9 +170,9 @@ EXPORT_SYMBOL(fddi_type_trans);
 int fddi_change_mtu(struct net_device *dev, int new_mtu)
 {
 	if ((new_mtu < FDDI_K_SNAP_HLEN) || (new_mtu > FDDI_K_SNAP_DLEN))
-		return(-EINVAL);
+		return -EINVAL;
 	dev->mtu = new_mtu;
-	return(0);
+	return 0;
 }
 EXPORT_SYMBOL(fddi_change_mtu);
 
diff --git a/net/802/hippi.c b/net/802/hippi.c
index cd3e8e929529..91aca8780fd0 100644
--- a/net/802/hippi.c
+++ b/net/802/hippi.c
@@ -152,7 +152,7 @@ int hippi_change_mtu(struct net_device *dev, int new_mtu)
 	if ((new_mtu < 68) || (new_mtu > 65280))
 		return -EINVAL;
 	dev->mtu = new_mtu;
-	return(0);
+	return 0;
 }
 EXPORT_SYMBOL(hippi_change_mtu);
 
diff --git a/net/802/tr.c b/net/802/tr.c
index 1c6e596074df..5e20cf8a074b 100644
--- a/net/802/tr.c
+++ b/net/802/tr.c
@@ -145,7 +145,7 @@ static int tr_header(struct sk_buff *skb, struct net_device *dev,
 	{
 		memcpy(trh->daddr,daddr,dev->addr_len);
 		tr_source_route(skb, trh, dev);
-		return(hdr_len);
+		return hdr_len;
 	}
 
 	return -hdr_len;
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 889f4ac4459a..0eb486d342dc 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -27,7 +27,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 	else if (vlan_id)
 		goto drop;
 
-	return (polling ? netif_receive_skb(skb) : netif_rx(skb));
+	return polling ? netif_receive_skb(skb) : netif_rx(skb);
 
 drop:
 	dev_kfree_skb_any(skb);
diff --git a/net/9p/client.c b/net/9p/client.c
index dc6f2f26d023..f34b9f510818 100644
--- a/net/9p/client.c
+++ b/net/9p/client.c
@@ -61,13 +61,13 @@ static const match_table_t tokens = {
 
 inline int p9_is_proto_dotl(struct p9_client *clnt)
 {
-	return (clnt->proto_version == p9_proto_2000L);
+	return clnt->proto_version == p9_proto_2000L;
 }
 EXPORT_SYMBOL(p9_is_proto_dotl);
 
 inline int p9_is_proto_dotu(struct p9_client *clnt)
 {
-	return (clnt->proto_version == p9_proto_2000u);
+	return clnt->proto_version == p9_proto_2000u;
 }
 EXPORT_SYMBOL(p9_is_proto_dotu);
 
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index 7dca91bb8c57..15ea84ba344e 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -179,13 +179,13 @@ static unsigned char rfcomm_crc_table[256] = {
 /* FCS on 2 bytes */
 static inline u8 __fcs(u8 *data)
 {
-	return (0xff - __crc(data));
+	return 0xff - __crc(data);
 }
 
 /* FCS on 3 bytes */
 static inline u8 __fcs2(u8 *data)
 {
-	return (0xff - rfcomm_crc_table[__crc(data) ^ data[2]]);
+	return 0xff - rfcomm_crc_table[__crc(data) ^ data[2]];
 }
 
 /* Check FCS */
diff --git a/net/core/flow.c b/net/core/flow.c
index b143b86b1f2a..127c8a7ffd61 100644
--- a/net/core/flow.c
+++ b/net/core/flow.c
@@ -176,8 +176,8 @@ static u32 flow_hash_code(struct flow_cache *fc,
 {
 	u32 *k = (u32 *) key;
 
-	return (jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd)
-		& (flow_cache_hash_size(fc) - 1));
+	return jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd)
+		& (flow_cache_hash_size(fc) - 1);
 }
 
 typedef unsigned long flow_compare_t;
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index a4e0a7482c2b..96b1a749abb4 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -122,7 +122,7 @@ static void neigh_cleanup_and_release(struct neighbour *neigh)
 
 unsigned long neigh_rand_reach_time(unsigned long base)
 {
-	return (base ? (net_random() % base) + (base >> 1) : 0);
+	return base ? (net_random() % base) + (base >> 1) : 0;
 }
 EXPORT_SYMBOL(neigh_rand_reach_time);
 
@@ -766,9 +766,9 @@ next_elt:
 static __inline__ int neigh_max_probes(struct neighbour *n)
 {
 	struct neigh_parms *p = n->parms;
-	return (n->nud_state & NUD_PROBE ?
+	return (n->nud_state & NUD_PROBE) ?
 		p->ucast_probes :
-		p->ucast_probes + p->app_probes + p->mcast_probes);
+		p->ucast_probes + p->app_probes + p->mcast_probes;
 }
 
 static void neigh_invalidate(struct neighbour *neigh)
diff --git a/net/core/utils.c b/net/core/utils.c
index ec6bb322f372..5fea0ab21902 100644
--- a/net/core/utils.c
+++ b/net/core/utils.c
@@ -75,7 +75,7 @@ __be32 in_aton(const char *str)
 				str++;
 		}
 	}
-	return(htonl(l));
+	return htonl(l);
 }
 EXPORT_SYMBOL(in_aton);
 
diff --git a/net/dccp/ccids/lib/loss_interval.c b/net/dccp/ccids/lib/loss_interval.c
index 8fc3cbf79071..497723c4d4bb 100644
--- a/net/dccp/ccids/lib/loss_interval.c
+++ b/net/dccp/ccids/lib/loss_interval.c
@@ -116,7 +116,7 @@ u8 tfrc_lh_update_i_mean(struct tfrc_loss_hist *lh, struct sk_buff *skb)
 	cur->li_length = len;
 	tfrc_lh_calc_i_mean(lh);
 
-	return (lh->i_mean < old_i_mean);
+	return lh->i_mean < old_i_mean;
 }
 
 /* Determine if `new_loss' does begin a new loss interval [RFC 4342, 10.2] */
diff --git a/net/econet/af_econet.c b/net/econet/af_econet.c
index baa98fb83552..f8c1ae4b41f0 100644
--- a/net/econet/af_econet.c
+++ b/net/econet/af_econet.c
@@ -392,7 +392,7 @@ static int econet_sendmsg(struct kiocb *iocb, struct socket *sock,
 		dev_queue_xmit(skb);
 		dev_put(dev);
 		mutex_unlock(&econet_mutex);
-		return(len);
+		return len;
 
 	out_free:
 		kfree_skb(skb);
@@ -637,7 +637,7 @@ static int econet_create(struct net *net, struct socket *sock, int protocol,
 	eo->num = protocol;
 
 	econet_insert_socket(&econet_sklist, sk);
-	return(0);
+	return 0;
 out:
 	return err;
 }
diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c
index 85e7b4551326..f00ef2f1d814 100644
--- a/net/ethernet/eth.c
+++ b/net/ethernet/eth.c
@@ -387,6 +387,6 @@ ssize_t sysfs_format_mac(char *buf, const unsigned char *addr, int len)
 
 	l = _format_mac_addr(buf, PAGE_SIZE, addr, len);
 	l += scnprintf(buf + l, PAGE_SIZE - l, "\n");
-	return ((ssize_t) l);
+	return (ssize_t)l;
 }
 EXPORT_SYMBOL(sysfs_format_mac);
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index dcfe7e961c10..4083c186fd30 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -567,7 +567,7 @@ static inline int arp_fwd_proxy(struct in_device *in_dev,
 	if (out_dev)
 		omi = IN_DEV_MEDIUM_ID(out_dev);
 
-	return (omi != imi && omi != -1);
+	return omi != imi && omi != -1;
 }
 
 /*
diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c
index 721a8a37b45c..174be6caa5c8 100644
--- a/net/ipv4/datagram.c
+++ b/net/ipv4/datagram.c
@@ -73,6 +73,6 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	inet->inet_id = jiffies;
 
 	sk_dst_set(sk, &rt->dst);
-	return(0);
+	return 0;
 }
 EXPORT_SYMBOL(ip4_datagram_connect);
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index e5fa2ddce320..ba8042665849 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -425,7 +425,7 @@ static int inet_diag_bc_run(const void *bc, int len,
 			bc += op->no;
 		}
 	}
-	return (len == 0);
+	return len == 0;
 }
 
 static int valid_cc(const void *bc, int len, int cc)
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index f4dc879e258e..168440834ade 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -116,11 +116,11 @@ static int ip4_frag_match(struct inet_frag_queue *q, void *a)
 	struct ip4_create_arg *arg = a;
 
 	qp = container_of(q, struct ipq, q);
-	return (qp->id == arg->iph->id &&
+	return	qp->id == arg->iph->id &&
 			qp->saddr == arg->iph->saddr &&
 			qp->daddr == arg->iph->daddr &&
 			qp->protocol == arg->iph->protocol &&
-			qp->user == arg->user);
+			qp->user == arg->user;
 }
 
 /* Memory Tracking Functions. */
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 714b6a80361d..0967d02fefd8 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -659,7 +659,7 @@ drop:
 	rcu_read_unlock();
 drop_nolock:
 	kfree_skb(skb);
-	return(0);
+	return 0;
 }
 
 static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index e8f4f9a57f12..8b642f152468 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -72,7 +72,7 @@ static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap,
 	for (i = 0; i < len; i++)
 		ret |= (hdr_addr[i] ^ ap->addr[i]) & ap->mask[i];
 
-	return (ret != 0);
+	return ret != 0;
 }
 
 /*
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index e24d48dd99d3..ae1d4a41f1c6 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2791,7 +2791,7 @@ static int ipv4_dst_blackhole(struct net *net, struct rtable **rp, struct flowi
 
 	dst_release(&(*rp)->dst);
 	*rp = rt;
-	return (rt ? 0 : -ENOMEM);
+	return rt ? 0 : -ENOMEM;
 }
 
 int ip_route_output_flow(struct net *net, struct rtable **rp, struct flowi *flp,
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 1bc87a05c734..51966b3f9719 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -2301,7 +2301,7 @@ static inline int tcp_dupack_heuristics(struct tcp_sock *tp)
 
 static inline int tcp_skb_timedout(struct sock *sk, struct sk_buff *skb)
 {
-	return (tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto);
+	return tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto;
 }
 
 static inline int tcp_head_timedout(struct sock *sk)
@@ -3398,8 +3398,8 @@ static void tcp_ack_probe(struct sock *sk)
 
 static inline int tcp_ack_is_dubious(const struct sock *sk, const int flag)
 {
-	return (!(flag & FLAG_NOT_DUP) || (flag & FLAG_CA_ALERT) ||
-		inet_csk(sk)->icsk_ca_state != TCP_CA_Open);
+	return !(flag & FLAG_NOT_DUP) || (flag & FLAG_CA_ALERT) ||
+		inet_csk(sk)->icsk_ca_state != TCP_CA_Open;
 }
 
 static inline int tcp_may_raise_cwnd(const struct sock *sk, const int flag)
@@ -3416,9 +3416,9 @@ static inline int tcp_may_update_window(const struct tcp_sock *tp,
 					const u32 ack, const u32 ack_seq,
 					const u32 nwin)
 {
-	return (after(ack, tp->snd_una) ||
+	return	after(ack, tp->snd_una) ||
 		after(ack_seq, tp->snd_wl1) ||
-		(ack_seq == tp->snd_wl1 && nwin > tp->snd_wnd));
+		(ack_seq == tp->snd_wl1 && nwin > tp->snd_wnd);
 }
 
 /* Update our send window.
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index f25b56cb85cb..43cf901d7659 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -55,7 +55,7 @@ static __inline__ int tcp_in_window(u32 seq, u32 end_seq, u32 s_win, u32 e_win)
 		return 1;
 	if (after(end_seq, s_win) && before(seq, e_win))
 		return 1;
-	return (seq == e_win && seq == end_seq);
+	return seq == e_win && seq == end_seq;
 }
 
 /*
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index ea09d2fd50c7..05b1ecf36763 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1370,9 +1370,9 @@ static inline int tcp_nagle_check(const struct tcp_sock *tp,
 				  const struct sk_buff *skb,
 				  unsigned mss_now, int nonagle)
 {
-	return (skb->len < mss_now &&
+	return skb->len < mss_now &&
 		((nonagle & TCP_NAGLE_CORK) ||
-		 (!nonagle && tp->packets_out && tcp_minshall_check(tp))));
+		 (!nonagle && tp->packets_out && tcp_minshall_check(tp)));
 }
 
 /* Return non-zero if the Nagle test allows this packet to be
@@ -1443,10 +1443,10 @@ int tcp_may_send_now(struct sock *sk)
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb = tcp_send_head(sk);
 
-	return (skb &&
+	return skb &&
 		tcp_snd_test(sk, skb, tcp_current_mss(sk),
 			     (tcp_skb_is_last(sk, skb) ?
-			      tp->nonagle : TCP_NAGLE_PUSH)));
+			      tp->nonagle : TCP_NAGLE_PUSH));
 }
 
 /* Trim TSO SKB to LEN bytes, put the remaining data into a new packet
diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c
index 20151d6a6241..a534dda5456e 100644
--- a/net/ipv4/tcp_westwood.c
+++ b/net/ipv4/tcp_westwood.c
@@ -80,7 +80,7 @@ static void tcp_westwood_init(struct sock *sk)
  */
 static inline u32 westwood_do_filter(u32 a, u32 b)
 {
-	return (((7 * a) + b) >> 3);
+	return ((7 * a) + b) >> 3;
 }
 
 static void westwood_filter(struct westwood *w, u32 delta)
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 5bc893e28008..89aa54394a08 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -243,7 +243,7 @@ static inline bool addrconf_qdisc_ok(const struct net_device *dev)
 /* Check if a route is valid prefix route */
 static inline int addrconf_is_prefix_route(const struct rt6_info *rt)
 {
-	return ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0);
+	return (rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0;
 }
 
 static void addrconf_del_timer(struct inet6_ifaddr *ifp)
diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c
index f0e774cea386..921dcf6c271a 100644
--- a/net/ipv6/addrlabel.c
+++ b/net/ipv6/addrlabel.c
@@ -513,10 +513,9 @@ static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb)
 
 static inline int ip6addrlbl_msgsize(void)
 {
-	return (NLMSG_ALIGN(sizeof(struct ifaddrlblmsg))
+	return NLMSG_ALIGN(sizeof(struct ifaddrlblmsg))
 		+ nla_total_size(16)	/* IFAL_ADDRESS */
-		+ nla_total_size(4)	/* IFAL_LABEL */
-	);
+		+ nla_total_size(4);	/* IFAL_LABEL */
 }
 
 static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh,
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 56b9bf2516f4..60220985bb80 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -467,7 +467,7 @@ int inet6_getname(struct socket *sock, struct sockaddr *uaddr,
 	if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL)
 		sin->sin6_scope_id = sk->sk_bound_dev_if;
 	*uaddr_len = sizeof(*sin);
-	return(0);
+	return 0;
 }
 
 EXPORT_SYMBOL(inet6_getname);
@@ -488,7 +488,7 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 	case SIOCADDRT:
 	case SIOCDELRT:
 
-		return(ipv6_route_ioctl(net, cmd, (void __user *)arg));
+		return ipv6_route_ioctl(net, cmd, (void __user *)arg);
 
 	case SIOCSIFADDR:
 		return addrconf_add_ifaddr(net, (void __user *) arg);
@@ -502,7 +502,7 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		return sk->sk_prot->ioctl(sk, cmd, arg);
 	}
 	/*NOTREACHED*/
-	return(0);
+	return 0;
 }
 
 EXPORT_SYMBOL(inet6_ioctl);
diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c
index e1caa5d526c2..14ed0a955b56 100644
--- a/net/ipv6/exthdrs_core.c
+++ b/net/ipv6/exthdrs_core.c
@@ -13,12 +13,12 @@ int ipv6_ext_hdr(u8 nexthdr)
 	/*
 	 * find out if nexthdr is an extension header or a protocol
 	 */
-	return ( (nexthdr == NEXTHDR_HOP)	||
+	return   (nexthdr == NEXTHDR_HOP)	||
 		 (nexthdr == NEXTHDR_ROUTING)	||
 		 (nexthdr == NEXTHDR_FRAGMENT)	||
 		 (nexthdr == NEXTHDR_AUTH)	||
 		 (nexthdr == NEXTHDR_NONE)	||
-		 (nexthdr == NEXTHDR_DEST) );
+		 (nexthdr == NEXTHDR_DEST);
 }
 
 /*
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 1838927a2243..efbbbce68f9e 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -870,8 +870,8 @@ static inline int ip6_rt_check(struct rt6key *rt_key,
 			       struct in6_addr *fl_addr,
 			       struct in6_addr *addr_cache)
 {
-	return ((rt_key->plen != 128 || !ipv6_addr_equal(fl_addr, &rt_key->addr)) &&
-		(addr_cache == NULL || !ipv6_addr_equal(fl_addr, addr_cache)));
+	return (rt_key->plen != 128 || !ipv6_addr_equal(fl_addr, &rt_key->addr)) &&
+		(addr_cache == NULL || !ipv6_addr_equal(fl_addr, addr_cache));
 }
 
 static struct dst_entry *ip6_sk_dst_check(struct sock *sk,
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 69a0051cea67..b3dd844cd34f 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -228,12 +228,12 @@ static struct nd_opt_hdr *ndisc_next_option(struct nd_opt_hdr *cur,
 	do {
 		cur = ((void *)cur) + (cur->nd_opt_len << 3);
 	} while(cur < end && cur->nd_opt_type != type);
-	return (cur <= end && cur->nd_opt_type == type ? cur : NULL);
+	return cur <= end && cur->nd_opt_type == type ? cur : NULL;
 }
 
 static inline int ndisc_is_useropt(struct nd_opt_hdr *opt)
 {
-	return (opt->nd_opt_type == ND_OPT_RDNSS);
+	return opt->nd_opt_type == ND_OPT_RDNSS;
 }
 
 static struct nd_opt_hdr *ndisc_next_useropt(struct nd_opt_hdr *cur,
@@ -244,7 +244,7 @@ static struct nd_opt_hdr *ndisc_next_useropt(struct nd_opt_hdr *cur,
 	do {
 		cur = ((void *)cur) + (cur->nd_opt_len << 3);
 	} while(cur < end && !ndisc_is_useropt(cur));
-	return (cur <= end && ndisc_is_useropt(cur) ? cur : NULL);
+	return cur <= end && ndisc_is_useropt(cur) ? cur : NULL;
 }
 
 static struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len,
@@ -319,7 +319,7 @@ static inline u8 *ndisc_opt_addr_data(struct nd_opt_hdr *p,
 	int prepad = ndisc_addr_option_pad(dev->type);
 	if (lladdrlen != NDISC_OPT_SPACE(dev->addr_len + prepad))
 		return NULL;
-	return (lladdr + prepad);
+	return lladdr + prepad;
 }
 
 int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int dir)
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 8e754be92c24..6b331e9b5706 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -82,13 +82,13 @@ EXPORT_SYMBOL_GPL(ip6t_alloc_initial_table);
 int
 ip6t_ext_hdr(u8 nexthdr)
 {
-	return ( (nexthdr == IPPROTO_HOPOPTS)   ||
-		 (nexthdr == IPPROTO_ROUTING)   ||
-		 (nexthdr == IPPROTO_FRAGMENT)  ||
-		 (nexthdr == IPPROTO_ESP)       ||
-		 (nexthdr == IPPROTO_AH)        ||
-		 (nexthdr == IPPROTO_NONE)      ||
-		 (nexthdr == IPPROTO_DSTOPTS) );
+	return  (nexthdr == IPPROTO_HOPOPTS)   ||
+		(nexthdr == IPPROTO_ROUTING)   ||
+		(nexthdr == IPPROTO_FRAGMENT)  ||
+		(nexthdr == IPPROTO_ESP)       ||
+		(nexthdr == IPPROTO_AH)        ||
+		(nexthdr == IPPROTO_NONE)      ||
+		(nexthdr == IPPROTO_DSTOPTS);
 }
 
 /* Returns whether matches rule or not. */
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index e677937a07fc..45e6efb7f171 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -764,7 +764,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk,
 			return -EINVAL;
 
 		if (sin6->sin6_family && sin6->sin6_family != AF_INET6)
-			return(-EAFNOSUPPORT);
+			return -EAFNOSUPPORT;
 
 		/* port is the proto value [0..255] carried in nexthdr */
 		proto = ntohs(sin6->sin6_port);
@@ -772,10 +772,10 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk,
 		if (!proto)
 			proto = inet->inet_num;
 		else if (proto != inet->inet_num)
-			return(-EINVAL);
+			return -EINVAL;
 
 		if (proto > 255)
-			return(-EINVAL);
+			return -EINVAL;
 
 		daddr = &sin6->sin6_addr;
 		if (np->sndflow) {
@@ -985,7 +985,7 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname,
 			/* You may get strange result with a positive odd offset;
 			   RFC2292bis agrees with me. */
 			if (val > 0 && (val&1))
-				return(-EINVAL);
+				return -EINVAL;
 			if (val < 0) {
 				rp->checksum = 0;
 			} else {
@@ -997,7 +997,7 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname,
 			break;
 
 		default:
-			return(-ENOPROTOOPT);
+			return -ENOPROTOOPT;
 	}
 }
 
@@ -1190,7 +1190,7 @@ static int rawv6_init_sk(struct sock *sk)
 	default:
 		break;
 	}
-	return(0);
+	return 0;
 }
 
 struct proto rawv6_prot = {
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index d126365ac046..25b0beda4331 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -217,14 +217,14 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
 
 static __inline__ int rt6_check_expired(const struct rt6_info *rt)
 {
-	return (rt->rt6i_flags & RTF_EXPIRES &&
-		time_after(jiffies, rt->rt6i_expires));
+	return (rt->rt6i_flags & RTF_EXPIRES) &&
+		time_after(jiffies, rt->rt6i_expires);
 }
 
 static inline int rt6_need_strict(struct in6_addr *daddr)
 {
-	return (ipv6_addr_type(daddr) &
-		(IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK));
+	return ipv6_addr_type(daddr) &
+		(IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK);
 }
 
 /*
@@ -440,7 +440,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
 		  __func__, match);
 
 	net = dev_net(rt0->rt6i_dev);
-	return (match ? match : net->ipv6.ip6_null_entry);
+	return match ? match : net->ipv6.ip6_null_entry;
 }
 
 #ifdef CONFIG_IPV6_ROUTE_INFO
@@ -859,7 +859,7 @@ int ip6_dst_blackhole(struct sock *sk, struct dst_entry **dstp, struct flowi *fl
 
 	dst_release(*dstp);
 	*dstp = new;
-	return (new ? 0 : -ENOMEM);
+	return new ? 0 : -ENOMEM;
 }
 EXPORT_SYMBOL_GPL(ip6_dst_blackhole);
 
@@ -1070,7 +1070,7 @@ static int ip6_dst_gc(struct dst_ops *ops)
 		net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1;
 out:
 	net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>rt_elasticity;
-	return (atomic_read(&ops->entries) > rt_max_size);
+	return atomic_read(&ops->entries) > rt_max_size;
 }
 
 /* Clean host part of a prefix. Not necessary in radix tree,
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index fe6d40418c0b..8d93f6d81979 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -139,7 +139,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 		return -EINVAL;
 
 	if (usin->sin6_family != AF_INET6)
-		return(-EAFNOSUPPORT);
+		return -EAFNOSUPPORT;
 
 	memset(&fl, 0, sizeof(fl));
 
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index 6baeabbbca82..39676eac3a37 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -199,7 +199,7 @@ static inline int xfrm6_garbage_collect(struct dst_ops *ops)
 	struct net *net = container_of(ops, struct net, xfrm.xfrm6_dst_ops);
 
 	xfrm6_policy_afinfo.garbage_collect(net);
-	return (atomic_read(&ops->entries) > ops->gc_thresh * 2);
+	return atomic_read(&ops->entries) > ops->gc_thresh * 2;
 }
 
 static void xfrm6_update_pmtu(struct dst_entry *dst, u32 mtu)
diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c
index fd55b5135de5..bf3635129b17 100644
--- a/net/irda/af_irda.c
+++ b/net/irda/af_irda.c
@@ -573,9 +573,9 @@ static int irda_find_lsap_sel(struct irda_sock *self, char *name)
 		/* Requested object/attribute doesn't exist */
 		if((self->errno == IAS_CLASS_UNKNOWN) ||
 		   (self->errno == IAS_ATTRIB_UNKNOWN))
-			return (-EADDRNOTAVAIL);
+			return -EADDRNOTAVAIL;
 		else
-			return (-EHOSTUNREACH);
+			return -EHOSTUNREACH;
 	}
 
 	/* Get the remote TSAP selector */
@@ -663,7 +663,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name)
 					   __func__, name);
 				self->daddr = DEV_ADDR_ANY;
 				kfree(discoveries);
-				return(-ENOTUNIQ);
+				return -ENOTUNIQ;
 			}
 			/* First time we found that one, save it ! */
 			daddr = self->daddr;
@@ -677,7 +677,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name)
 			IRDA_DEBUG(0, "%s(), unexpected IAS query failure\n", __func__);
 			self->daddr = DEV_ADDR_ANY;
 			kfree(discoveries);
-			return(-EHOSTUNREACH);
+			return -EHOSTUNREACH;
 			break;
 		}
 	}
@@ -689,7 +689,7 @@ static int irda_discover_daddr_and_lsap_sel(struct irda_sock *self, char *name)
 		IRDA_DEBUG(1, "%s(), cannot discover service ''%s'' in any device !!!\n",
 			   __func__, name);
 		self->daddr = DEV_ADDR_ANY;
-		return(-EADDRNOTAVAIL);
+		return -EADDRNOTAVAIL;
 	}
 
 	/* Revert back to discovered device & service */
@@ -2465,9 +2465,9 @@ bed:
 			/* Requested object/attribute doesn't exist */
 			if((self->errno == IAS_CLASS_UNKNOWN) ||
 			   (self->errno == IAS_ATTRIB_UNKNOWN))
-				return (-EADDRNOTAVAIL);
+				return -EADDRNOTAVAIL;
 			else
-				return (-EHOSTUNREACH);
+				return -EHOSTUNREACH;
 		}
 
 		/* Translate from internal to user structure */
diff --git a/net/irda/discovery.c b/net/irda/discovery.c
index c1c8ae939126..36c3f037f172 100644
--- a/net/irda/discovery.c
+++ b/net/irda/discovery.c
@@ -315,7 +315,7 @@ struct irda_device_info *irlmp_copy_discoveries(hashbin_t *log, int *pn,
 
 	/* Get the actual number of device in the buffer and return */
 	*pn = i;
-	return(buffer);
+	return buffer;
 }
 
 #ifdef CONFIG_PROC_FS
diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c
index faa82ca2dfdc..a39cca8331df 100644
--- a/net/irda/ircomm/ircomm_tty.c
+++ b/net/irda/ircomm/ircomm_tty.c
@@ -449,8 +449,8 @@ static int ircomm_tty_open(struct tty_struct *tty, struct file *filp)
 		}
 
 #ifdef SERIAL_DO_RESTART
-		return ((self->flags & ASYNC_HUP_NOTIFY) ?
-			-EAGAIN : -ERESTARTSYS);
+		return (self->flags & ASYNC_HUP_NOTIFY) ?
+			-EAGAIN : -ERESTARTSYS;
 #else
 		return -EAGAIN;
 #endif
diff --git a/net/irda/irlmp.c b/net/irda/irlmp.c
index 0e7d8bde145d..6115a44c0a24 100644
--- a/net/irda/irlmp.c
+++ b/net/irda/irlmp.c
@@ -939,7 +939,7 @@ struct irda_device_info *irlmp_get_discoveries(int *pn, __u16 mask, int nslots)
 	}
 
 	/* Return current cached discovery log */
-	return(irlmp_copy_discoveries(irlmp->cachelog, pn, mask, TRUE));
+	return irlmp_copy_discoveries(irlmp->cachelog, pn, mask, TRUE);
 }
 EXPORT_SYMBOL(irlmp_get_discoveries);
 
diff --git a/net/irda/irlmp_frame.c b/net/irda/irlmp_frame.c
index 3750884094da..062e63b1c5c4 100644
--- a/net/irda/irlmp_frame.c
+++ b/net/irda/irlmp_frame.c
@@ -448,7 +448,7 @@ static struct lsap_cb *irlmp_find_lsap(struct lap_cb *self, __u8 dlsap_sel,
 	    (self->cache.slsap_sel == slsap_sel) &&
 	    (self->cache.dlsap_sel == dlsap_sel))
 	{
-		return (self->cache.lsap);
+		return self->cache.lsap;
 	}
 #endif
 
diff --git a/net/irda/irnet/irnet_irda.c b/net/irda/irnet/irnet_irda.c
index e98e40d76f4f..7f17a8020e8a 100644
--- a/net/irda/irnet/irnet_irda.c
+++ b/net/irda/irnet/irnet_irda.c
@@ -238,7 +238,7 @@ irnet_ias_to_tsap(irnet_socket *	self,
   DEXIT(IRDA_SR_TRACE, "\n");
 
   /* Return the TSAP */
-  return(dtsap_sel);
+  return dtsap_sel;
 }
 
 /*------------------------------------------------------------------*/
@@ -301,7 +301,7 @@ irnet_connect_tsap(irnet_socket *	self)
     {
       clear_bit(0, &self->ttp_connect);
       DERROR(IRDA_SR_ERROR, "connect aborted!\n");
-      return(err);
+      return err;
     }
 
   /* Connect to remote device */
@@ -312,7 +312,7 @@ irnet_connect_tsap(irnet_socket *	self)
     {
       clear_bit(0, &self->ttp_connect);
       DERROR(IRDA_SR_ERROR, "connect aborted!\n");
-      return(err);
+      return err;
     }
 
   /* The above call is non-blocking.
@@ -321,7 +321,7 @@ irnet_connect_tsap(irnet_socket *	self)
    * See you there ;-) */
 
   DEXIT(IRDA_SR_TRACE, "\n");
-  return(err);
+  return err;
 }
 
 /*------------------------------------------------------------------*/
@@ -362,10 +362,10 @@ irnet_discover_next_daddr(irnet_socket *	self)
       /* The above request is non-blocking.
        * After a while, IrDA will call us back in irnet_discovervalue_confirm()
        * We will then call irnet_ias_to_tsap() and come back here again... */
-      return(0);
+      return 0;
     }
   else
-    return(1);
+    return 1;
 }
 
 /*------------------------------------------------------------------*/
@@ -436,7 +436,7 @@ irnet_discover_daddr_and_lsap_sel(irnet_socket *	self)
   /* Follow me in irnet_discovervalue_confirm() */
 
   DEXIT(IRDA_SR_TRACE, "\n");
-  return(0);
+  return 0;
 }
 
 /*------------------------------------------------------------------*/
@@ -485,7 +485,7 @@ irnet_dname_to_daddr(irnet_socket *	self)
   /* No luck ! */
   DEBUG(IRDA_SR_INFO, "cannot discover device ``%s'' !!!\n", self->rname);
   kfree(discoveries);
-  return(-EADDRNOTAVAIL);
+  return -EADDRNOTAVAIL;
 }
 
 
@@ -527,7 +527,7 @@ irda_irnet_create(irnet_socket *	self)
   INIT_WORK(&self->disconnect_work, irnet_ppp_disconnect);
 
   DEXIT(IRDA_SOCK_TRACE, "\n");
-  return(0);
+  return 0;
 }
 
 /*------------------------------------------------------------------*/
@@ -601,7 +601,7 @@ irda_irnet_connect(irnet_socket *	self)
    * We will finish the connection procedure in irnet_connect_tsap().
    */
   DEXIT(IRDA_SOCK_TRACE, "\n");
-  return(0);
+  return 0;
 }
 
 /*------------------------------------------------------------------*/
@@ -733,7 +733,7 @@ irnet_daddr_to_dname(irnet_socket *	self)
   /* No luck ! */
   DEXIT(IRDA_SERV_INFO, ": cannot discover device 0x%08x !!!\n", self->daddr);
   kfree(discoveries);
-  return(-EADDRNOTAVAIL);
+  return -EADDRNOTAVAIL;
 }
 
 /*------------------------------------------------------------------*/
diff --git a/net/irda/irnet/irnet_ppp.c b/net/irda/irnet/irnet_ppp.c
index dfe7b38dd4af..69f1fa64994e 100644
--- a/net/irda/irnet/irnet_ppp.c
+++ b/net/irda/irnet/irnet_ppp.c
@@ -166,7 +166,7 @@ irnet_ctrl_write(irnet_socket *	ap,
     }
 
   /* Success : we have parsed all commands successfully */
-  return(count);
+  return count;
 }
 
 #ifdef INITIAL_DISCOVERY
@@ -300,7 +300,7 @@ irnet_ctrl_read(irnet_socket *	ap,
 	}
 
       DEXIT(CTRL_TRACE, "\n");
-      return(strlen(event));
+      return strlen(event);
     }
 #endif /* INITIAL_DISCOVERY */
 
@@ -409,7 +409,7 @@ irnet_ctrl_read(irnet_socket *	ap,
     }
 
   DEXIT(CTRL_TRACE, "\n");
-  return(strlen(event));
+  return strlen(event);
 }
 
 /*------------------------------------------------------------------*/
@@ -623,7 +623,7 @@ dev_irnet_poll(struct file *	file,
     mask |= irnet_ctrl_poll(ap, file, wait);
 
   DEXIT(FS_TRACE, " - mask=0x%X\n", mask);
-  return(mask);
+  return mask;
 }
 
 /*------------------------------------------------------------------*/
diff --git a/net/key/af_key.c b/net/key/af_key.c
index 43040e97c474..d87c22df6f1e 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -565,12 +565,12 @@ pfkey_proto2satype(uint16_t proto)
 
 static uint8_t pfkey_proto_to_xfrm(uint8_t proto)
 {
-	return (proto == IPSEC_PROTO_ANY ? 0 : proto);
+	return proto == IPSEC_PROTO_ANY ? 0 : proto;
 }
 
 static uint8_t pfkey_proto_from_xfrm(uint8_t proto)
 {
-	return (proto ? proto : IPSEC_PROTO_ANY);
+	return proto ? proto : IPSEC_PROTO_ANY;
 }
 
 static inline int pfkey_sockaddr_len(sa_family_t family)
diff --git a/net/mac80211/rate.c b/net/mac80211/rate.c
index 4f772de2f213..b0cc385bf989 100644
--- a/net/mac80211/rate.c
+++ b/net/mac80211/rate.c
@@ -207,7 +207,7 @@ static bool rc_no_data_or_no_ack(struct ieee80211_tx_rate_control *txrc)
 
 	fc = hdr->frame_control;
 
-	return ((info->flags & IEEE80211_TX_CTL_NO_ACK) || !ieee80211_is_data(fc));
+	return (info->flags & IEEE80211_TX_CTL_NO_ACK) || !ieee80211_is_data(fc);
 }
 
 static void rc_send_low_broadcast(s8 *idx, u32 basic_rates, u8 max_rate_idx)
diff --git a/net/rfkill/input.c b/net/rfkill/input.c
index 3713d7ecab96..1bca6d49ec96 100644
--- a/net/rfkill/input.c
+++ b/net/rfkill/input.c
@@ -142,7 +142,7 @@ static unsigned long rfkill_last_scheduled;
 static unsigned long rfkill_ratelimit(const unsigned long last)
 {
 	const unsigned long delay = msecs_to_jiffies(RFKILL_OPS_DELAY);
-	return (time_after(jiffies, last + delay)) ? 0 : delay;
+	return time_after(jiffies, last + delay) ? 0 : delay;
 }
 
 static void rfkill_schedule_ratelimited(void)
diff --git a/net/rose/rose_link.c b/net/rose/rose_link.c
index a750a28e0221..fa5f5641a2c2 100644
--- a/net/rose/rose_link.c
+++ b/net/rose/rose_link.c
@@ -114,7 +114,7 @@ static int rose_send_frame(struct sk_buff *skb, struct rose_neigh *neigh)
 	if (ax25s)
 		ax25_cb_put(ax25s);
 
-	return (neigh->ax25 != NULL);
+	return neigh->ax25 != NULL;
 }
 
 /*
@@ -137,7 +137,7 @@ static int rose_link_up(struct rose_neigh *neigh)
 	if (ax25s)
 		ax25_cb_put(ax25s);
 
-	return (neigh->ax25 != NULL);
+	return neigh->ax25 != NULL;
 }
 
 /*
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index f774e657641a..1ef29c74d85e 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -799,7 +799,7 @@ static void sctp_inet_skb_msgname(struct sk_buff *skb, char *msgname, int *len)
 static int sctp_inet_af_supported(sa_family_t family, struct sctp_sock *sp)
 {
 	/* PF_INET only supports AF_INET addresses. */
-	return (AF_INET == family);
+	return AF_INET == family;
 }
 
 /* Address matching with wildcards allowed. */
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 6a691d84aef4..535659fdbaa1 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -3884,7 +3884,7 @@ static int sctp_getsockopt_sctp_status(struct sock *sk, int len,
 	}
 
 out:
-	return (retval);
+	return retval;
 }
 
 
@@ -3940,7 +3940,7 @@ static int sctp_getsockopt_peer_addr_info(struct sock *sk, int len,
 	}
 
 out:
-	return (retval);
+	return retval;
 }
 
 /* 7.1.12 Enable/Disable message fragmentation (SCTP_DISABLE_FRAGMENTS)
@@ -5594,7 +5594,7 @@ static int sctp_get_port(struct sock *sk, unsigned short snum)
 	/* Note: sk->sk_num gets filled in if ephemeral port request. */
 	ret = sctp_get_port_local(sk, &addr);
 
-	return (ret ? 1 : 0);
+	return ret ? 1 : 0;
 }
 
 /*
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index dcfc66bab2bb..597c493392ad 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -1049,7 +1049,7 @@ gss_match(struct auth_cred *acred, struct rpc_cred *rc, int flags)
 out:
 	if (acred->machine_cred != gss_cred->gc_machine_cred)
 		return 0;
-	return (rc->cr_uid == acred->uid);
+	return rc->cr_uid == acred->uid;
 }
 
 /*
diff --git a/net/sunrpc/auth_gss/gss_generic_token.c b/net/sunrpc/auth_gss/gss_generic_token.c
index 310b78e99456..c586e92bcf76 100644
--- a/net/sunrpc/auth_gss/gss_generic_token.c
+++ b/net/sunrpc/auth_gss/gss_generic_token.c
@@ -76,19 +76,19 @@ static int
 der_length_size( int length)
 {
 	if (length < (1<<7))
-		return(1);
+		return 1;
 	else if (length < (1<<8))
-		return(2);
+		return 2;
 #if (SIZEOF_INT == 2)
 	else
-		return(3);
+		return 3;
 #else
 	else if (length < (1<<16))
-		return(3);
+		return 3;
 	else if (length < (1<<24))
-		return(4);
+		return 4;
 	else
-		return(5);
+		return 5;
 #endif
 }
 
@@ -121,14 +121,14 @@ der_read_length(unsigned char **buf, int *bufsize)
 	int ret;
 
 	if (*bufsize < 1)
-		return(-1);
+		return -1;
 	sf = *(*buf)++;
 	(*bufsize)--;
 	if (sf & 0x80) {
 		if ((sf &= 0x7f) > ((*bufsize)-1))
-			return(-1);
+			return -1;
 		if (sf > SIZEOF_INT)
-			return (-1);
+			return -1;
 		ret = 0;
 		for (; sf; sf--) {
 			ret = (ret<<8) + (*(*buf)++);
@@ -138,7 +138,7 @@ der_read_length(unsigned char **buf, int *bufsize)
 		ret = sf;
 	}
 
-	return(ret);
+	return ret;
 }
 
 /* returns the length of a token, given the mech oid and the body size */
@@ -148,7 +148,7 @@ g_token_size(struct xdr_netobj *mech, unsigned int body_size)
 {
 	/* set body_size to sequence contents size */
 	body_size += 2 + (int) mech->len;         /* NEED overflow check */
-	return(1 + der_length_size(body_size) + body_size);
+	return 1 + der_length_size(body_size) + body_size;
 }
 
 EXPORT_SYMBOL_GPL(g_token_size);
@@ -186,27 +186,27 @@ g_verify_token_header(struct xdr_netobj *mech, int *body_size,
 	int ret = 0;
 
 	if ((toksize-=1) < 0)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 	if (*buf++ != 0x60)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 
 	if ((seqsize = der_read_length(&buf, &toksize)) < 0)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 
 	if (seqsize != toksize)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 
 	if ((toksize-=1) < 0)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 	if (*buf++ != 0x06)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 
 	if ((toksize-=1) < 0)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 	toid.len = *buf++;
 
 	if ((toksize-=toid.len) < 0)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 	toid.data = buf;
 	buf+=toid.len;
 
@@ -217,17 +217,17 @@ g_verify_token_header(struct xdr_netobj *mech, int *body_size,
       to return G_BAD_TOK_HEADER if the token header is in fact bad */
 
 	if ((toksize-=2) < 0)
-		return(G_BAD_TOK_HEADER);
+		return G_BAD_TOK_HEADER;
 
 	if (ret)
-		return(ret);
+		return ret;
 
 	if (!ret) {
 		*buf_in = buf;
 		*body_size = toksize;
 	}
 
-	return(ret);
+	return ret;
 }
 
 EXPORT_SYMBOL_GPL(g_verify_token_header);
diff --git a/net/sunrpc/auth_gss/gss_krb5_seqnum.c b/net/sunrpc/auth_gss/gss_krb5_seqnum.c
index 415c013ba382..62ac90c62cb1 100644
--- a/net/sunrpc/auth_gss/gss_krb5_seqnum.c
+++ b/net/sunrpc/auth_gss/gss_krb5_seqnum.c
@@ -162,5 +162,5 @@ krb5_get_seq_num(struct krb5_ctx *kctx,
 	*seqnum = ((plain[0]) |
 		   (plain[1] << 8) | (plain[2] << 16) | (plain[3] << 24));
 
-	return (0);
+	return 0;
 }
diff --git a/net/sunrpc/auth_gss/gss_mech_switch.c b/net/sunrpc/auth_gss/gss_mech_switch.c
index 2689de39dc78..8b4061049d76 100644
--- a/net/sunrpc/auth_gss/gss_mech_switch.c
+++ b/net/sunrpc/auth_gss/gss_mech_switch.c
@@ -331,7 +331,7 @@ gss_delete_sec_context(struct gss_ctx	**context_handle)
 			*context_handle);
 
 	if (!*context_handle)
-		return(GSS_S_NO_CONTEXT);
+		return GSS_S_NO_CONTEXT;
 	if ((*context_handle)->internal_ctx_id)
 		(*context_handle)->mech_type->gm_ops
 			->gss_delete_sec_context((*context_handle)
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index cace6049e4a5..aa5dbda6608c 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -376,7 +376,7 @@ int rpc_queue_empty(struct rpc_wait_queue *queue)
 	spin_lock_bh(&queue->lock);
 	res = queue->qlen;
 	spin_unlock_bh(&queue->lock);
-	return (res == 0);
+	return res == 0;
 }
 EXPORT_SYMBOL_GPL(rpc_queue_empty);
 
diff --git a/net/tipc/addr.c b/net/tipc/addr.c
index c048543ffbeb..2ddc351b3be9 100644
--- a/net/tipc/addr.c
+++ b/net/tipc/addr.c
@@ -89,7 +89,7 @@ int tipc_addr_domain_valid(u32 addr)
 
 int tipc_addr_node_valid(u32 addr)
 {
-	return (tipc_addr_domain_valid(addr) && tipc_node(addr));
+	return tipc_addr_domain_valid(addr) && tipc_node(addr);
 }
 
 int tipc_in_scope(u32 domain, u32 addr)
diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c
index b11248c2d788..ecfaac10d0b4 100644
--- a/net/tipc/bcast.c
+++ b/net/tipc/bcast.c
@@ -184,7 +184,7 @@ static void bclink_set_gap(struct tipc_node *n_ptr)
 
 static int bclink_ack_allowed(u32 n)
 {
-	return((n % TIPC_MIN_LINK_WIN) == tipc_own_tag);
+	return (n % TIPC_MIN_LINK_WIN) == tipc_own_tag;
 }
 
 
diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
index 52ae17b2583e..9c10c6b7c12b 100644
--- a/net/tipc/bearer.c
+++ b/net/tipc/bearer.c
@@ -63,7 +63,7 @@ static int media_name_valid(const char *name)
 	len = strlen(name);
 	if ((len + 1) > TIPC_MAX_MEDIA_NAME)
 		return 0;
-	return (strspn(name, tipc_alphabet) == len);
+	return strspn(name, tipc_alphabet) == len;
 }
 
 /**
diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c
index 1885a7edb0c8..6569d45bfb9a 100644
--- a/net/tipc/dbg.c
+++ b/net/tipc/dbg.c
@@ -134,7 +134,7 @@ void tipc_printbuf_reset(struct print_buf *pb)
 
 int tipc_printbuf_empty(struct print_buf *pb)
 {
-	return (!pb->buf || (pb->crs == pb->buf));
+	return !pb->buf || (pb->crs == pb->buf);
 }
 
 /**
@@ -169,7 +169,7 @@ int tipc_printbuf_validate(struct print_buf *pb)
 			tipc_printf(pb, err);
 		}
 	}
-	return (pb->crs - pb->buf + 1);
+	return pb->crs - pb->buf + 1;
 }
 
 /**
diff --git a/net/tipc/link.c b/net/tipc/link.c
index a6a3102bb4d6..b8cf1e9d0b86 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -239,13 +239,13 @@ int tipc_link_is_up(struct link *l_ptr)
 {
 	if (!l_ptr)
 		return 0;
-	return (link_working_working(l_ptr) || link_working_unknown(l_ptr));
+	return link_working_working(l_ptr) || link_working_unknown(l_ptr);
 }
 
 int tipc_link_is_active(struct link *l_ptr)
 {
-	return ((l_ptr->owner->active_links[0] == l_ptr) ||
-		(l_ptr->owner->active_links[1] == l_ptr));
+	return	(l_ptr->owner->active_links[0] == l_ptr) ||
+		(l_ptr->owner->active_links[1] == l_ptr);
 }
 
 /**
diff --git a/net/tipc/link.h b/net/tipc/link.h
index 2e5385c47d30..26151d30589d 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -279,12 +279,12 @@ static inline int between(u32 lower, u32 upper, u32 n)
 
 static inline int less_eq(u32 left, u32 right)
 {
-	return (mod(right - left) < 32768u);
+	return mod(right - left) < 32768u;
 }
 
 static inline int less(u32 left, u32 right)
 {
-	return (less_eq(left, right) && (mod(right) != mod(left)));
+	return less_eq(left, right) && (mod(right) != mod(left));
 }
 
 static inline u32 lesser(u32 left, u32 right)
@@ -299,32 +299,32 @@ static inline u32 lesser(u32 left, u32 right)
 
 static inline int link_working_working(struct link *l_ptr)
 {
-	return (l_ptr->state == WORKING_WORKING);
+	return l_ptr->state == WORKING_WORKING;
 }
 
 static inline int link_working_unknown(struct link *l_ptr)
 {
-	return (l_ptr->state == WORKING_UNKNOWN);
+	return l_ptr->state == WORKING_UNKNOWN;
 }
 
 static inline int link_reset_unknown(struct link *l_ptr)
 {
-	return (l_ptr->state == RESET_UNKNOWN);
+	return l_ptr->state == RESET_UNKNOWN;
 }
 
 static inline int link_reset_reset(struct link *l_ptr)
 {
-	return (l_ptr->state == RESET_RESET);
+	return l_ptr->state == RESET_RESET;
 }
 
 static inline int link_blocked(struct link *l_ptr)
 {
-	return (l_ptr->exp_msg_count || l_ptr->blocked);
+	return l_ptr->exp_msg_count || l_ptr->blocked;
 }
 
 static inline int link_congested(struct link *l_ptr)
 {
-	return (l_ptr->out_queue_size >= l_ptr->queue_limit[0]);
+	return l_ptr->out_queue_size >= l_ptr->queue_limit[0];
 }
 
 #endif
diff --git a/net/tipc/msg.h b/net/tipc/msg.h
index 995d2da35b01..031aad18efce 100644
--- a/net/tipc/msg.h
+++ b/net/tipc/msg.h
@@ -104,7 +104,7 @@ static inline u32 msg_user(struct tipc_msg *m)
 
 static inline u32 msg_isdata(struct tipc_msg *m)
 {
-	return (msg_user(m) <= TIPC_CRITICAL_IMPORTANCE);
+	return msg_user(m) <= TIPC_CRITICAL_IMPORTANCE;
 }
 
 static inline void msg_set_user(struct tipc_msg *m, u32 n)
@@ -289,7 +289,7 @@ static inline void msg_set_destnode(struct tipc_msg *m, u32 a)
 
 static inline int msg_is_dest(struct tipc_msg *m, u32 d)
 {
-	return(msg_short(m) || (msg_destnode(m) == d));
+	return msg_short(m) || (msg_destnode(m) == d);
 }
 
 static inline u32 msg_routed(struct tipc_msg *m)
@@ -632,7 +632,7 @@ static inline void msg_set_bcast_tag(struct tipc_msg *m, u32 n)
 
 static inline u32 msg_max_pkt(struct tipc_msg *m)
 {
-	return (msg_bits(m, 9, 16, 0xffff) * 4);
+	return msg_bits(m, 9, 16, 0xffff) * 4;
 }
 
 static inline void msg_set_max_pkt(struct tipc_msg *m, u32 n)
diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c
index c13c2c7c4b57..9ca4b0689237 100644
--- a/net/tipc/name_table.c
+++ b/net/tipc/name_table.c
@@ -116,7 +116,7 @@ DEFINE_RWLOCK(tipc_nametbl_lock);
 
 static int hash(int x)
 {
-	return(x & (tipc_nametbl_size - 1));
+	return x & (tipc_nametbl_size - 1);
 }
 
 /**
diff --git a/net/tipc/node.c b/net/tipc/node.c
index b702c7bf580f..7c49cd056df7 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -242,17 +242,17 @@ int tipc_node_has_active_links(struct tipc_node *n_ptr)
 
 int tipc_node_has_redundant_links(struct tipc_node *n_ptr)
 {
-	return (n_ptr->working_links > 1);
+	return n_ptr->working_links > 1;
 }
 
 static int tipc_node_has_active_routes(struct tipc_node *n_ptr)
 {
-	return (n_ptr && (n_ptr->last_router >= 0));
+	return n_ptr && (n_ptr->last_router >= 0);
 }
 
 int tipc_node_is_up(struct tipc_node *n_ptr)
 {
-	return (tipc_node_has_active_links(n_ptr) || tipc_node_has_active_routes(n_ptr));
+	return tipc_node_has_active_links(n_ptr) || tipc_node_has_active_routes(n_ptr);
 }
 
 struct tipc_node *tipc_node_attach_link(struct link *l_ptr)
diff --git a/net/tipc/port.h b/net/tipc/port.h
index 8d1652aab298..e74bd9563739 100644
--- a/net/tipc/port.h
+++ b/net/tipc/port.h
@@ -157,7 +157,7 @@ static inline u32 tipc_peer_node(struct port *p_ptr)
 
 static inline int tipc_port_congested(struct port *p_ptr)
 {
-	return((p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2));
+	return (p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2);
 }
 
 /**
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index f7ac94de24fe..33217fc3d697 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -1195,7 +1195,7 @@ static int rx_queue_full(struct tipc_msg *msg, u32 queue_size, u32 base)
 	if (msg_connected(msg))
 		threshold *= 4;
 
-	return (queue_size >= threshold);
+	return queue_size >= threshold;
 }
 
 /**
diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index ab6eab4c45e2..1a5b9a6bd128 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -604,6 +604,6 @@ int tipc_ispublished(struct tipc_name const *name)
 {
 	u32 domain = 0;
 
-	return(tipc_nametbl_translate(name->type, name->instance,&domain) != 0);
+	return tipc_nametbl_translate(name->type, name->instance, &domain) != 0;
 }
 
diff --git a/net/wireless/core.h b/net/wireless/core.h
index 37580e090a3d..5d89310b3587 100644
--- a/net/wireless/core.h
+++ b/net/wireless/core.h
@@ -86,7 +86,7 @@ struct cfg80211_registered_device *wiphy_to_dev(struct wiphy *wiphy)
 static inline
 bool wiphy_idx_valid(int wiphy_idx)
 {
-	return (wiphy_idx >= 0);
+	return wiphy_idx >= 0;
 }
 
 
-- 
cgit v1.2.3


From 543876c92837a8b208b5c99ec225c1f5a581900e Mon Sep 17 00:00:00 2001
From: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Date: Fri, 24 Sep 2010 21:27:41 -0700
Subject: stmmac: review the wake-up support

If the PM support is available this is passed
through the platform instead to be hard-coded
in the core files.
WoL on Magic Frame can be enabled by using
the ethtool support.

Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/stmmac/common.h         |  1 -
 drivers/net/stmmac/dwmac1000_core.c |  1 -
 drivers/net/stmmac/dwmac100_core.c  |  1 -
 drivers/net/stmmac/stmmac_ethtool.c | 14 +++++++++-----
 drivers/net/stmmac/stmmac_main.c    | 26 ++++++++++++++------------
 include/linux/stmmac.h              |  1 +
 6 files changed, 24 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/stmmac/common.h b/drivers/net/stmmac/common.h
index 673ef86a063f..dec7ce40c27a 100644
--- a/drivers/net/stmmac/common.h
+++ b/drivers/net/stmmac/common.h
@@ -238,7 +238,6 @@ struct mac_device_info {
 	struct stmmac_ops	*mac;
 	struct stmmac_desc_ops	*desc;
 	struct stmmac_dma_ops	*dma;
-	unsigned int pmt;	/* support Power-Down */
 	struct mii_regs mii;	/* MII register Addresses */
 	struct mac_link link;
 };
diff --git a/drivers/net/stmmac/dwmac1000_core.c b/drivers/net/stmmac/dwmac1000_core.c
index c18c85993179..65667b692024 100644
--- a/drivers/net/stmmac/dwmac1000_core.c
+++ b/drivers/net/stmmac/dwmac1000_core.c
@@ -239,7 +239,6 @@ struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr)
 	mac->mac = &dwmac1000_ops;
 	mac->dma = &dwmac1000_dma_ops;
 
-	mac->pmt = PMT_SUPPORTED;
 	mac->link.port = GMAC_CONTROL_PS;
 	mac->link.duplex = GMAC_CONTROL_DM;
 	mac->link.speed = GMAC_CONTROL_FES;
diff --git a/drivers/net/stmmac/dwmac100_core.c b/drivers/net/stmmac/dwmac100_core.c
index 58a914b27003..94eeccf3a8a0 100644
--- a/drivers/net/stmmac/dwmac100_core.c
+++ b/drivers/net/stmmac/dwmac100_core.c
@@ -193,7 +193,6 @@ struct mac_device_info *dwmac100_setup(void __iomem *ioaddr)
 	mac->mac = &dwmac100_ops;
 	mac->dma = &dwmac100_dma_ops;
 
-	mac->pmt = PMT_NOT_SUPPORTED;
 	mac->link.port = MAC_CONTROL_PS;
 	mac->link.duplex = MAC_CONTROL_F;
 	mac->link.speed = 0;
diff --git a/drivers/net/stmmac/stmmac_ethtool.c b/drivers/net/stmmac/stmmac_ethtool.c
index b32c16ae55c6..25a7e385f8ec 100644
--- a/drivers/net/stmmac/stmmac_ethtool.c
+++ b/drivers/net/stmmac/stmmac_ethtool.c
@@ -322,7 +322,7 @@ static void stmmac_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 	struct stmmac_priv *priv = netdev_priv(dev);
 
 	spin_lock_irq(&priv->lock);
-	if (priv->wolenabled == PMT_SUPPORTED) {
+	if (device_can_wakeup(priv->device)) {
 		wol->supported = WAKE_MAGIC;
 		wol->wolopts = priv->wolopts;
 	}
@@ -334,16 +334,20 @@ static int stmmac_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 	struct stmmac_priv *priv = netdev_priv(dev);
 	u32 support = WAKE_MAGIC;
 
-	if (priv->wolenabled == PMT_NOT_SUPPORTED)
+	if (!device_can_wakeup(priv->device))
 		return -EINVAL;
 
 	if (wol->wolopts & ~support)
 		return -EINVAL;
 
-	if (wol->wolopts == 0)
-		device_set_wakeup_enable(priv->device, 0);
-	else
+	if (wol->wolopts) {
+		pr_info("stmmac: wakeup enable\n");
 		device_set_wakeup_enable(priv->device, 1);
+		enable_irq_wake(dev->irq);
+	} else {
+		device_set_wakeup_enable(priv->device, 0);
+		disable_irq_wake(dev->irq);
+	}
 
 	spin_lock_irq(&priv->lock);
 	priv->wolopts = wol->wolopts;
diff --git a/drivers/net/stmmac/stmmac_main.c b/drivers/net/stmmac/stmmac_main.c
index a908f7201aae..823b9e6431d5 100644
--- a/drivers/net/stmmac/stmmac_main.c
+++ b/drivers/net/stmmac/stmmac_main.c
@@ -1568,9 +1568,8 @@ static int stmmac_mac_device_setup(struct net_device *dev)
 
 	priv->hw = device;
 
-	priv->wolenabled = priv->hw->pmt;	/* PMT supported */
-	if (priv->wolenabled == PMT_SUPPORTED)
-		priv->wolopts = WAKE_MAGIC;		/* Magic Frame */
+	if (device_can_wakeup(priv->device))
+		priv->wolopts = WAKE_MAGIC; /* Magic Frame as default */
 
 	return 0;
 }
@@ -1709,6 +1708,12 @@ static int stmmac_dvr_probe(struct platform_device *pdev)
 	priv->enh_desc = plat_dat->enh_desc;
 	priv->ioaddr = addr;
 
+	/* PMT module is not integrated in all the MAC devices. */
+	if (plat_dat->pmt) {
+		pr_info("\tPMT module supported\n");
+		device_set_wakeup_capable(&pdev->dev, 1);
+	}
+
 	platform_set_drvdata(pdev, ndev);
 
 	/* Set the I/O base addr */
@@ -1836,13 +1841,11 @@ static int stmmac_suspend(struct platform_device *pdev, pm_message_t state)
 
 		stmmac_mac_disable_tx(priv->ioaddr);
 
-		if (device_may_wakeup(&(pdev->dev))) {
-			/* Enable Power down mode by programming the PMT regs */
-			if (priv->wolenabled == PMT_SUPPORTED)
-				priv->hw->mac->pmt(priv->ioaddr, priv->wolopts);
-		} else {
+		/* Enable Power down mode by programming the PMT regs */
+		if (device_can_wakeup(priv->device))
+			priv->hw->mac->pmt(priv->ioaddr, priv->wolopts);
+		else
 			stmmac_mac_disable_rx(priv->ioaddr);
-		}
 	} else {
 		priv->shutdown = 1;
 		/* Although this can appear slightly redundant it actually
@@ -1877,9 +1880,8 @@ static int stmmac_resume(struct platform_device *pdev)
 	 * is received. Anyway, it's better to manually clear
 	 * this bit because it can generate problems while resuming
 	 * from another devices (e.g. serial console). */
-	if (device_may_wakeup(&(pdev->dev)))
-		if (priv->wolenabled == PMT_SUPPORTED)
-			priv->hw->mac->pmt(priv->ioaddr, 0);
+	if (device_can_wakeup(priv->device))
+		priv->hw->mac->pmt(priv->ioaddr, 0);
 
 	netif_device_attach(dev);
 
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index 1d8baf719211..d66c61774d95 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -37,6 +37,7 @@ struct plat_stmmacenet_data {
 	int enh_desc;
 	int tx_coe;
 	int bugged_jumbo;
+	int pmt;
 	void (*fix_mac_speed)(void *priv, unsigned int speed);
 	void (*bus_setup)(void __iomem *ioaddr);
 #ifdef CONFIG_STM_DRIVERS
-- 
cgit v1.2.3


From e4ecda1b60bfd2333c12bbe71b153d3b6bdc831a Mon Sep 17 00:00:00 2001
From: Mark Lord <mlord@pobox.com>
Date: Sat, 25 Sep 2010 11:17:22 +0200
Subject: Fix compile error in blk-exec.c for !CONFIG_DETECT_HUNG_TASK

Ensure that 'sysctl_hung_task_timeout_secs' is defined
even when CONFIG_DETECT_HUNG_TASK is not set.
This way we can safely reference it without need for
ifdefs in the code elsewhere.  eg. in block/blk-exec.c

Signed-off-by: Mark Lord <mlord@pobox.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 include/linux/sched.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1e2a6db2d7dd..dbafa9e34a2d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -336,6 +336,9 @@ extern unsigned long sysctl_hung_task_warnings;
 extern int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
 					 void __user *buffer,
 					 size_t *lenp, loff_t *ppos);
+#else
+/* Avoid need for ifdefs elsewhere in the code */
+enum { sysctl_hung_task_timeout_secs = 0 };
 #endif
 
 /* Attach to any functions which should be ignored in wchan output. */
-- 
cgit v1.2.3


From cb4dfe562cac6fcb544df752e40c1d78000d0712 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 23 Sep 2010 05:06:54 +0000
Subject: net: skb_frag_t can be smaller on small arches

On 32bit arches, if PAGE_SIZE is smaller than 65536, we can use 16bit
offset and size fields. This patch saves 72 bytes per skb on i386, or
128 bytes after rounding.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index b2c41d19735c..0b53c43ac92e 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -129,8 +129,13 @@ typedef struct skb_frag_struct skb_frag_t;
 
 struct skb_frag_struct {
 	struct page *page;
+#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536)
 	__u32 page_offset;
 	__u32 size;
+#else
+	__u16 page_offset;
+	__u16 size;
+#endif
 };
 
 #define HAVE_HW_TIME_STAMP
-- 
cgit v1.2.3


From a7855c78a24d6348e989bec616318e68c662e78b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 23 Sep 2010 23:51:51 +0000
Subject: net: loopback driver cleanup

loopback driver uses dev->ml_priv to store its percpu stats pointer.
It uses ugly casts "(void __percpu __force *)" to shut up sparse
complains.

Define an union to better document we use ml_priv in loopback driver and
define a lstats field with appropriate types.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/loopback.c    | 20 +++++---------------
 include/linux/netdevice.h |  6 ++++--
 2 files changed, 9 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index 9a0996795321..4b0e30b564e5 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -74,7 +74,6 @@ struct pcpu_lstats {
 static netdev_tx_t loopback_xmit(struct sk_buff *skb,
 				 struct net_device *dev)
 {
-	struct pcpu_lstats __percpu *pcpu_lstats;
 	struct pcpu_lstats *lb_stats;
 	int len;
 
@@ -83,8 +82,7 @@ static netdev_tx_t loopback_xmit(struct sk_buff *skb,
 	skb->protocol = eth_type_trans(skb, dev);
 
 	/* it's OK to use per_cpu_ptr() because BHs are off */
-	pcpu_lstats = (void __percpu __force *)dev->ml_priv;
-	lb_stats = this_cpu_ptr(pcpu_lstats);
+	lb_stats = this_cpu_ptr(dev->lstats);
 
 	len = skb->len;
 	if (likely(netif_rx(skb) == NET_RX_SUCCESS)) {
@@ -101,19 +99,17 @@ static netdev_tx_t loopback_xmit(struct sk_buff *skb,
 static struct rtnl_link_stats64 *loopback_get_stats64(struct net_device *dev,
 						      struct rtnl_link_stats64 *stats)
 {
-	const struct pcpu_lstats __percpu *pcpu_lstats;
 	u64 bytes = 0;
 	u64 packets = 0;
 	u64 drops = 0;
 	int i;
 
-	pcpu_lstats = (void __percpu __force *)dev->ml_priv;
 	for_each_possible_cpu(i) {
 		const struct pcpu_lstats *lb_stats;
 		u64 tbytes, tpackets;
 		unsigned int start;
 
-		lb_stats = per_cpu_ptr(pcpu_lstats, i);
+		lb_stats = per_cpu_ptr(dev->lstats, i);
 		do {
 			start = u64_stats_fetch_begin(&lb_stats->syncp);
 			tbytes = lb_stats->bytes;
@@ -147,22 +143,16 @@ static const struct ethtool_ops loopback_ethtool_ops = {
 
 static int loopback_dev_init(struct net_device *dev)
 {
-	struct pcpu_lstats __percpu *lstats;
-
-	lstats = alloc_percpu(struct pcpu_lstats);
-	if (!lstats)
+	dev->lstats = alloc_percpu(struct pcpu_lstats);
+	if (!dev->lstats)
 		return -ENOMEM;
 
-	dev->ml_priv = (void __force *)lstats;
 	return 0;
 }
 
 static void loopback_dev_free(struct net_device *dev)
 {
-	struct pcpu_lstats __percpu *lstats =
-		(void __percpu __force *)dev->ml_priv;
-
-	free_percpu(lstats);
+	free_percpu(dev->lstats);
 	free_netdev(dev);
 }
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 45dcda5bfda9..01bd4c82d982 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1050,8 +1050,10 @@ struct net_device {
 #endif
 
 	/* mid-layer private */
-	void			*ml_priv;
-
+	union {
+		void				*ml_priv;
+		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
+	};
 	/* GARP */
 	struct garp_port	*garp_port;
 
-- 
cgit v1.2.3


From 8d98efa84b790bdd62248eb0dfff17e9baf5c844 Mon Sep 17 00:00:00 2001
From: Kumar Sanghvi <kumar.sanghvi@stericsson.com>
Date: Sun, 26 Sep 2010 19:07:59 +0000
Subject: Phonet: Implement Pipe Controller to support Nokia Slim Modems

Phonet stack assumes the presence of Pipe Controller, either in Modem or
on Application Processing Engine user-space for the Pipe data.
Nokia Slim Modems like WG2.5 used in ST-Ericsson U8500 platform do not
implement Pipe controller in them.
This patch adds Pipe Controller implemenation to Phonet stack to support
Pipe data over Phonet stack for Nokia Slim Modems.

Signed-off-by: Kumar Sanghvi <kumar.sanghvi@stericsson.com>
Acked-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/phonet.h   |   5 +
 include/net/phonet/pep.h |  21 +++
 net/phonet/Kconfig       |  11 ++
 net/phonet/pep.c         | 448 ++++++++++++++++++++++++++++++++++++++++++++++-
 4 files changed, 479 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/phonet.h b/include/linux/phonet.h
index 85e14a83283b..96f5625d62fa 100644
--- a/include/linux/phonet.h
+++ b/include/linux/phonet.h
@@ -36,6 +36,11 @@
 /* Socket options for SOL_PNPIPE level */
 #define PNPIPE_ENCAP		1
 #define PNPIPE_IFINDEX		2
+#define PNPIPE_CREATE           3
+#define PNPIPE_ENABLE           4
+#define PNPIPE_DISABLE          5
+#define PNPIPE_DESTROY          6
+#define PNPIPE_INQ              7
 
 #define PNADDR_ANY		0
 #define PNADDR_BROADCAST	0xFC
diff --git a/include/net/phonet/pep.h b/include/net/phonet/pep.h
index 37f23dc05de8..def6cfa3f451 100644
--- a/include/net/phonet/pep.h
+++ b/include/net/phonet/pep.h
@@ -45,6 +45,10 @@ struct pep_sock {
 	u8			tx_fc;	/* TX flow control */
 	u8			init_enable;	/* auto-enable at creation */
 	u8			aligned;
+#ifdef CONFIG_PHONET_PIPECTRLR
+	u16                     remote_pep;
+	u8                      pipe_state;
+#endif
 };
 
 static inline struct pep_sock *pep_sk(struct sock *sk)
@@ -165,4 +169,21 @@ enum {
 	PEP_IND_READY,
 };
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+#define PNS_PEP_CONNECT_UTID           0x02
+#define PNS_PIPE_CREATED_IND_UTID      0x04
+#define PNS_PIPE_ENABLE_UTID           0x0A
+#define PNS_PIPE_ENABLED_IND_UTID      0x0C
+#define PNS_PIPE_DISABLE_UTID          0x0F
+#define PNS_PIPE_DISABLED_IND_UTID     0x11
+#define PNS_PEP_DISCONNECT_UTID        0x06
+
+/* Used for tracking state of a pipe */
+enum {
+	PIPE_IDLE,
+	PIPE_DISABLED,
+	PIPE_ENABLED,
+};
+#endif /* CONFIG_PHONET_PIPECTRLR */
+
 #endif
diff --git a/net/phonet/Kconfig b/net/phonet/Kconfig
index 6ec7d55b1769..901956ada9c8 100644
--- a/net/phonet/Kconfig
+++ b/net/phonet/Kconfig
@@ -14,3 +14,14 @@ config PHONET
 
 	  To compile this driver as a module, choose M here: the module
 	  will be called phonet. If unsure, say N.
+
+config PHONET_PIPECTRLR
+	bool "Phonet Pipe Controller"
+	depends on PHONET
+	default N
+	help
+	  The Pipe Controller implementation in Phonet stack to support Pipe
+	  data with Nokia Slim modems like WG2.5 used on ST-Ericsson U8500
+	  platform.
+
+	  If unsure, say N.
diff --git a/net/phonet/pep.c b/net/phonet/pep.c
index d0e7eb24c8b9..7bf23cf36b02 100644
--- a/net/phonet/pep.c
+++ b/net/phonet/pep.c
@@ -88,6 +88,15 @@ static int pep_reply(struct sock *sk, struct sk_buff *oskb,
 	const struct pnpipehdr *oph = pnp_hdr(oskb);
 	struct pnpipehdr *ph;
 	struct sk_buff *skb;
+#ifdef CONFIG_PHONET_PIPECTRLR
+	const struct phonethdr *hdr = pn_hdr(oskb);
+	struct sockaddr_pn spn = {
+		.spn_family = AF_PHONET,
+		.spn_resource = 0xD9,
+		.spn_dev = hdr->pn_sdev,
+		.spn_obj = hdr->pn_sobj,
+	};
+#endif
 
 	skb = alloc_skb(MAX_PNPIPE_HEADER + len, priority);
 	if (!skb)
@@ -105,10 +114,271 @@ static int pep_reply(struct sock *sk, struct sk_buff *oskb,
 	ph->pipe_handle = oph->pipe_handle;
 	ph->error_code = code;
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+	return pn_skb_send(sk, skb, &spn);
+#else
 	return pn_skb_send(sk, skb, &pipe_srv);
+#endif
 }
 
 #define PAD 0x00
+
+#ifdef CONFIG_PHONET_PIPECTRLR
+static u8 pipe_negotiate_fc(u8 *host_fc, u8 *remote_fc, int len)
+{
+	int i, j;
+	u8 base_fc, final_fc;
+
+	for (i = 0; i < len; i++) {
+		base_fc = host_fc[i];
+		for (j = 0; j < len; j++) {
+			if (remote_fc[j] == base_fc) {
+				final_fc = base_fc;
+				goto done;
+			}
+		}
+	}
+	return -EINVAL;
+
+done:
+	return final_fc;
+
+}
+
+static int pipe_get_flow_info(struct sock *sk, struct sk_buff *skb,
+		u8 *pref_rx_fc, u8 *req_tx_fc)
+{
+	struct pnpipehdr *hdr;
+	u8 n_sb;
+
+	if (!pskb_may_pull(skb, sizeof(*hdr) + 4))
+		return -EINVAL;
+
+	hdr = pnp_hdr(skb);
+	n_sb = hdr->data[4];
+
+	__skb_pull(skb, sizeof(*hdr) + 4);
+	while (n_sb > 0) {
+		u8 type, buf[3], len = sizeof(buf);
+		u8 *data = pep_get_sb(skb, &type, &len, buf);
+
+		if (data == NULL)
+			return -EINVAL;
+
+		switch (type) {
+		case PN_PIPE_SB_REQUIRED_FC_TX:
+			if (len < 3 || (data[2] | data[3] | data[4]) > 3)
+				break;
+			req_tx_fc[0] = data[2];
+			req_tx_fc[1] = data[3];
+			req_tx_fc[2] = data[4];
+			break;
+
+		case PN_PIPE_SB_PREFERRED_FC_RX:
+			if (len < 3 || (data[2] | data[3] | data[4]) > 3)
+				break;
+			pref_rx_fc[0] = data[2];
+			pref_rx_fc[1] = data[3];
+			pref_rx_fc[2] = data[4];
+			break;
+
+		}
+		n_sb--;
+	}
+	return 0;
+}
+
+static int pipe_handler_send_req(struct sock *sk, u16 dobj, u8 utid,
+		u8 msg_id, u8 p_handle, gfp_t priority)
+{
+	int len;
+	struct pnpipehdr *ph;
+	struct sk_buff *skb;
+	struct sockaddr_pn spn = {
+		.spn_family = AF_PHONET,
+		.spn_resource = 0xD9,
+		.spn_dev = pn_dev(dobj),
+		.spn_obj = pn_obj(dobj),
+	};
+
+	static const u8 data[4] = {
+		PAD, PAD, PAD, PAD,
+	};
+
+	switch (msg_id) {
+	case PNS_PEP_CONNECT_REQ:
+		len = sizeof(data);
+		break;
+
+	case PNS_PEP_DISCONNECT_REQ:
+	case PNS_PEP_ENABLE_REQ:
+	case PNS_PEP_DISABLE_REQ:
+		len = 0;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	skb = alloc_skb(MAX_PNPIPE_HEADER + len, priority);
+	if (!skb)
+		return -ENOMEM;
+	skb_set_owner_w(skb, sk);
+
+	skb_reserve(skb, MAX_PNPIPE_HEADER);
+	if (len) {
+		__skb_put(skb, len);
+		skb_copy_to_linear_data(skb, data, len);
+	}
+	__skb_push(skb, sizeof(*ph));
+	skb_reset_transport_header(skb);
+	ph = pnp_hdr(skb);
+	ph->utid = utid;
+	ph->message_id = msg_id;
+	ph->pipe_handle = p_handle;
+	ph->error_code = PN_PIPE_NO_ERROR;
+
+	return pn_skb_send(sk, skb, &spn);
+}
+
+static int pipe_handler_send_created_ind(struct sock *sk, u16 dobj,
+		u8 utid, u8 p_handle, u8 msg_id, u8 tx_fc, u8 rx_fc)
+{
+	int err_code;
+	struct pnpipehdr *ph;
+	struct sk_buff *skb;
+	struct sockaddr_pn spn = {
+		.spn_family = AF_PHONET,
+		.spn_resource = 0xD9,
+		.spn_dev = pn_dev(dobj),
+		.spn_obj = pn_obj(dobj),
+	};
+
+	static u8 data[4] = {
+		0x03, 0x04,
+	};
+	data[2] = tx_fc;
+	data[3] = rx_fc;
+
+	/*
+	 * actually, below is number of sub-blocks and not error code.
+	 * Pipe_created_ind message format does not have any
+	 * error code field. However, the Phonet stack will always send
+	 * an error code as part of pnpipehdr. So, use that err_code to
+	 * specify the number of sub-blocks.
+	 */
+	err_code = 0x01;
+
+	skb = alloc_skb(MAX_PNPIPE_HEADER + sizeof(data), GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+	skb_set_owner_w(skb, sk);
+
+	skb_reserve(skb, MAX_PNPIPE_HEADER);
+	__skb_put(skb, sizeof(data));
+	skb_copy_to_linear_data(skb, data, sizeof(data));
+	__skb_push(skb, sizeof(*ph));
+	skb_reset_transport_header(skb);
+	ph = pnp_hdr(skb);
+	ph->utid = utid;
+	ph->message_id = msg_id;
+	ph->pipe_handle = p_handle;
+	ph->error_code = err_code;
+
+	return pn_skb_send(sk, skb, &spn);
+}
+
+static int pipe_handler_send_ind(struct sock *sk, u16 dobj, u8 utid,
+		u8 p_handle, u8 msg_id)
+{
+	int err_code;
+	struct pnpipehdr *ph;
+	struct sk_buff *skb;
+	struct sockaddr_pn spn = {
+		.spn_family = AF_PHONET,
+		.spn_resource = 0xD9,
+		.spn_dev = pn_dev(dobj),
+		.spn_obj = pn_obj(dobj),
+	};
+
+	/*
+	 * actually, below is a filler.
+	 * Pipe_enabled/disabled_ind message format does not have any
+	 * error code field. However, the Phonet stack will always send
+	 * an error code as part of pnpipehdr. So, use that err_code to
+	 * specify the filler value.
+	 */
+	err_code = 0x0;
+
+	skb = alloc_skb(MAX_PNPIPE_HEADER, GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+	skb_set_owner_w(skb, sk);
+
+	skb_reserve(skb, MAX_PNPIPE_HEADER);
+	__skb_push(skb, sizeof(*ph));
+	skb_reset_transport_header(skb);
+	ph = pnp_hdr(skb);
+	ph->utid = utid;
+	ph->message_id = msg_id;
+	ph->pipe_handle = p_handle;
+	ph->error_code = err_code;
+
+	return pn_skb_send(sk, skb, &spn);
+}
+
+static int pipe_handler_enable_pipe(struct sock *sk, int cmd)
+{
+	int ret;
+	struct pep_sock *pn = pep_sk(sk);
+
+	switch (cmd) {
+	case PNPIPE_ENABLE:
+		ret = pipe_handler_send_req(sk, pn->pn_sk.sobject,
+				PNS_PIPE_ENABLE_UTID, PNS_PEP_ENABLE_REQ,
+				pn->pipe_handle, GFP_ATOMIC);
+		break;
+
+	case PNPIPE_DISABLE:
+		ret = pipe_handler_send_req(sk, pn->pn_sk.sobject,
+				PNS_PIPE_DISABLE_UTID, PNS_PEP_DISABLE_REQ,
+				pn->pipe_handle, GFP_ATOMIC);
+		break;
+
+	default:
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+static int pipe_handler_create_pipe(struct sock *sk, int pipe_handle, int cmd)
+{
+	int ret;
+	struct pep_sock *pn = pep_sk(sk);
+
+	switch (cmd) {
+	case PNPIPE_CREATE:
+		ret = pipe_handler_send_req(sk, pn->pn_sk.sobject,
+				PNS_PEP_CONNECT_UTID, PNS_PEP_CONNECT_REQ,
+				pipe_handle, GFP_ATOMIC);
+		break;
+
+	case PNPIPE_DESTROY:
+		ret = pipe_handler_send_req(sk, pn->remote_pep,
+				PNS_PEP_DISCONNECT_UTID,
+				PNS_PEP_DISCONNECT_REQ,
+				pn->pipe_handle, GFP_ATOMIC);
+		break;
+
+	default:
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+#endif
+
 static int pep_accept_conn(struct sock *sk, struct sk_buff *skb)
 {
 	static const u8 data[20] = {
@@ -173,6 +443,14 @@ static int pipe_snd_status(struct sock *sk, u8 type, u8 status, gfp_t priority)
 	struct pep_sock *pn = pep_sk(sk);
 	struct pnpipehdr *ph;
 	struct sk_buff *skb;
+#ifdef CONFIG_PHONET_PIPECTRLR
+	struct sockaddr_pn spn = {
+		.spn_family = AF_PHONET,
+		.spn_resource = 0xD9,
+		.spn_dev = pn_dev(pn->remote_pep),
+		.spn_obj = pn_obj(pn->remote_pep),
+	};
+#endif
 
 	skb = alloc_skb(MAX_PNPIPE_HEADER + 4, priority);
 	if (!skb)
@@ -192,7 +470,11 @@ static int pipe_snd_status(struct sock *sk, u8 type, u8 status, gfp_t priority)
 	ph->data[3] = PAD;
 	ph->data[4] = status;
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+	return pn_skb_send(sk, skb, &spn);
+#else
 	return pn_skb_send(sk, skb, &pipe_srv);
+#endif
 }
 
 /* Send our RX flow control information to the sender.
@@ -308,6 +590,12 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 	struct pnpipehdr *hdr = pnp_hdr(skb);
 	struct sk_buff_head *queue;
 	int err = 0;
+#ifdef CONFIG_PHONET_PIPECTRLR
+	struct phonethdr *ph = pn_hdr(skb);
+	static u8 host_pref_rx_fc[3], host_req_tx_fc[3];
+	u8 remote_pref_rx_fc[3], remote_req_tx_fc[3];
+	u8 negotiated_rx_fc, negotiated_tx_fc;
+#endif
 
 	BUG_ON(sk->sk_state == TCP_CLOSE_WAIT);
 
@@ -316,6 +604,40 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 		pep_reject_conn(sk, skb, PN_PIPE_ERR_PEP_IN_USE);
 		break;
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+	case PNS_PEP_CONNECT_RESP:
+		if ((ph->pn_sdev == pn_dev(pn->remote_pep)) &&
+				(ph->pn_sobj == pn_obj(pn->remote_pep))) {
+			pipe_get_flow_info(sk, skb, remote_pref_rx_fc,
+					remote_req_tx_fc);
+
+			 negotiated_tx_fc = pipe_negotiate_fc(remote_req_tx_fc,
+					 host_pref_rx_fc,
+					 sizeof(host_pref_rx_fc));
+			 negotiated_rx_fc = pipe_negotiate_fc(host_req_tx_fc,
+					 remote_pref_rx_fc,
+					 sizeof(host_pref_rx_fc));
+
+			pn->pipe_state = PIPE_DISABLED;
+			pipe_handler_send_created_ind(sk, pn->remote_pep,
+					PNS_PIPE_CREATED_IND_UTID,
+					pn->pipe_handle, PNS_PIPE_CREATED_IND,
+					negotiated_tx_fc, negotiated_rx_fc);
+			pipe_handler_send_created_ind(sk, pn->pn_sk.sobject,
+					PNS_PIPE_CREATED_IND_UTID,
+					pn->pipe_handle, PNS_PIPE_CREATED_IND,
+					negotiated_tx_fc, negotiated_rx_fc);
+		} else {
+			pipe_handler_send_req(sk, pn->remote_pep,
+					PNS_PEP_CONNECT_UTID,
+					PNS_PEP_CONNECT_REQ, pn->pipe_handle,
+					GFP_ATOMIC);
+			pipe_get_flow_info(sk, skb, host_pref_rx_fc,
+					host_req_tx_fc);
+		}
+		break;
+#endif
+
 	case PNS_PEP_DISCONNECT_REQ:
 		pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC);
 		sk->sk_state = TCP_CLOSE_WAIT;
@@ -323,11 +645,41 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 			sk->sk_state_change(sk);
 		break;
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+	case PNS_PEP_DISCONNECT_RESP:
+		pn->pipe_state = PIPE_IDLE;
+		pipe_handler_send_req(sk, pn->pn_sk.sobject,
+				PNS_PEP_DISCONNECT_UTID,
+				PNS_PEP_DISCONNECT_REQ, pn->pipe_handle,
+				GFP_KERNEL);
+		break;
+#endif
+
 	case PNS_PEP_ENABLE_REQ:
 		/* Wait for PNS_PIPE_(ENABLED|REDIRECTED)_IND */
 		pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC);
 		break;
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+	case PNS_PEP_ENABLE_RESP:
+		if ((ph->pn_sdev == pn_dev(pn->remote_pep)) &&
+				(ph->pn_sobj == pn_obj(pn->remote_pep))) {
+			pn->pipe_state = PIPE_ENABLED;
+			pipe_handler_send_ind(sk, pn->remote_pep,
+					PNS_PIPE_ENABLED_IND_UTID,
+					pn->pipe_handle, PNS_PIPE_ENABLED_IND);
+			pipe_handler_send_ind(sk, pn->pn_sk.sobject,
+					PNS_PIPE_ENABLED_IND_UTID,
+					pn->pipe_handle, PNS_PIPE_ENABLED_IND);
+		} else
+			pipe_handler_send_req(sk, pn->remote_pep,
+					PNS_PIPE_ENABLE_UTID,
+					PNS_PEP_ENABLE_REQ, pn->pipe_handle,
+					GFP_KERNEL);
+
+		break;
+#endif
+
 	case PNS_PEP_RESET_REQ:
 		switch (hdr->state_after_reset) {
 		case PN_PIPE_DISABLE:
@@ -346,6 +698,27 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 		pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC);
 		break;
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+	case PNS_PEP_DISABLE_RESP:
+		if ((ph->pn_sdev == pn_dev(pn->remote_pep)) &&
+				(ph->pn_sobj == pn_obj(pn->remote_pep))) {
+			pn->pipe_state = PIPE_DISABLED;
+			pipe_handler_send_ind(sk, pn->remote_pep,
+					PNS_PIPE_DISABLED_IND_UTID,
+					pn->pipe_handle,
+					PNS_PIPE_DISABLED_IND);
+			pipe_handler_send_ind(sk, pn->pn_sk.sobject,
+					PNS_PIPE_DISABLED_IND_UTID,
+					pn->pipe_handle,
+					PNS_PIPE_DISABLED_IND);
+		} else
+			pipe_handler_send_req(sk, pn->remote_pep,
+					PNS_PIPE_DISABLE_UTID,
+					PNS_PEP_DISABLE_REQ, pn->pipe_handle,
+					GFP_KERNEL);
+		break;
+#endif
+
 	case PNS_PEP_CTRL_REQ:
 		if (skb_queue_len(&pn->ctrlreq_queue) >= PNPIPE_CTRLREQ_MAX) {
 			atomic_inc(&sk->sk_drops);
@@ -519,6 +892,9 @@ static int pep_connreq_rcv(struct sock *sk, struct sk_buff *skb)
 	newpn->rx_fc = newpn->tx_fc = PN_LEGACY_FLOW_CONTROL;
 	newpn->init_enable = enabled;
 	newpn->aligned = aligned;
+#ifdef CONFIG_PHONET_PIPECTRLR
+	newpn->remote_pep = pn->remote_pep;
+#endif
 
 	BUG_ON(!skb_queue_empty(&newsk->sk_receive_queue));
 	skb_queue_head(&newsk->sk_receive_queue, skb);
@@ -781,6 +1157,10 @@ static int pep_setsockopt(struct sock *sk, int level, int optname,
 {
 	struct pep_sock *pn = pep_sk(sk);
 	int val = 0, err = 0;
+#ifdef CONFIG_PHONET_PIPECTRLR
+	int remote_pep;
+	int pipe_handle;
+#endif
 
 	if (level != SOL_PNPIPE)
 		return -ENOPROTOOPT;
@@ -791,6 +1171,48 @@ static int pep_setsockopt(struct sock *sk, int level, int optname,
 
 	lock_sock(sk);
 	switch (optname) {
+#ifdef CONFIG_PHONET_PIPECTRLR
+	case PNPIPE_CREATE:
+		if (val) {
+			if (pn->pipe_state > PIPE_IDLE) {
+				err = -EFAULT;
+				break;
+			}
+			remote_pep = val & 0xFFFF;
+			pipe_handle =  (val >> 16) & 0xFF;
+			pn->remote_pep = remote_pep;
+			err = pipe_handler_create_pipe(sk, pipe_handle,
+					PNPIPE_CREATE);
+			break;
+		}
+
+	case PNPIPE_ENABLE:
+		if (pn->pipe_state != PIPE_DISABLED) {
+			err = -EFAULT;
+			break;
+		}
+		err = pipe_handler_enable_pipe(sk, PNPIPE_ENABLE);
+		break;
+
+	case PNPIPE_DISABLE:
+		if (pn->pipe_state != PIPE_ENABLED) {
+			err = -EFAULT;
+			break;
+		}
+
+		err = pipe_handler_enable_pipe(sk, PNPIPE_DISABLE);
+		break;
+
+	case PNPIPE_DESTROY:
+		if (pn->pipe_state < PIPE_DISABLED) {
+			err = -EFAULT;
+			break;
+		}
+
+		err = pipe_handler_create_pipe(sk, 0x0, PNPIPE_DESTROY);
+		break;
+#endif
+
 	case PNPIPE_ENCAP:
 		if (val && val != PNPIPE_ENCAP_IP) {
 			err = -EINVAL;
@@ -840,6 +1262,13 @@ static int pep_getsockopt(struct sock *sk, int level, int optname,
 	case PNPIPE_ENCAP:
 		val = pn->ifindex ? PNPIPE_ENCAP_IP : PNPIPE_ENCAP_NONE;
 		break;
+
+#ifdef CONFIG_PHONET_PIPECTRLR
+	case PNPIPE_INQ:
+		val = pn->pipe_state;
+		break;
+#endif
+
 	case PNPIPE_IFINDEX:
 		val = pn->ifindex;
 		break;
@@ -859,7 +1288,14 @@ static int pipe_skb_send(struct sock *sk, struct sk_buff *skb)
 {
 	struct pep_sock *pn = pep_sk(sk);
 	struct pnpipehdr *ph;
-	int err;
+#ifdef CONFIG_PHONET_PIPECTRLR
+	struct sockaddr_pn spn = {
+		.spn_family = AF_PHONET,
+		.spn_resource = 0xD9,
+		.spn_dev = pn_dev(pn->remote_pep),
+		.spn_obj = pn_obj(pn->remote_pep),
+	};
+#endif
 
 	if (pn_flow_safe(pn->tx_fc) &&
 	    !atomic_add_unless(&pn->tx_credits, -1, 0)) {
@@ -877,11 +1313,11 @@ static int pipe_skb_send(struct sock *sk, struct sk_buff *skb)
 	} else
 		ph->message_id = PNS_PIPE_DATA;
 	ph->pipe_handle = pn->pipe_handle;
-
-	err = pn_skb_send(sk, skb, &pipe_srv);
-	if (err && pn_flow_safe(pn->tx_fc))
-		atomic_inc(&pn->tx_credits);
-	return err;
+#ifdef CONFIG_PHONET_PIPECTRLR
+	return pn_skb_send(sk, skb, &spn);
+#else
+	return pn_skb_send(sk, skb, &pipe_srv);
+#endif
 }
 
 static int pep_sendmsg(struct kiocb *iocb, struct sock *sk,
-- 
cgit v1.2.3


From 290b895e0ba4552dfcfc4bd35759c192345b934a Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 27 Sep 2010 00:33:35 +0000
Subject: tunnels: prepare percpu accounting

Tunnels are going to use percpu for their accounting.

They are going to use a new tstats field in net_device.

skb_tunnel_rx() is changed to be a wrapper around __skb_tunnel_rx()

IPTUNNEL_XMIT() is changed to be a wrapper around __IPTUNNEL_XMIT()

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 +
 include/net/dst.h         | 24 +++++++++++++++++++-----
 include/net/ipip.h        | 12 +++++++-----
 3 files changed, 27 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 01bd4c82d982..83de0eb7a071 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1053,6 +1053,7 @@ struct net_device {
 	union {
 		void				*ml_priv;
 		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
+		struct pcpu_tstats __percpu	*tstats; /* tunnel stats */
 	};
 	/* GARP */
 	struct garp_port	*garp_port;
diff --git a/include/net/dst.h b/include/net/dst.h
index 02386505033d..aa53fbc34b2b 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -227,6 +227,23 @@ static inline void skb_dst_force(struct sk_buff *skb)
 }
 
 
+/**
+ *	__skb_tunnel_rx - prepare skb for rx reinsert
+ *	@skb: buffer
+ *	@dev: tunnel device
+ *
+ *	After decapsulation, packet is going to re-enter (netif_rx()) our stack,
+ *	so make some cleanups. (no accounting done)
+ */
+static inline void __skb_tunnel_rx(struct sk_buff *skb, struct net_device *dev)
+{
+	skb->dev = dev;
+	skb->rxhash = 0;
+	skb_set_queue_mapping(skb, 0);
+	skb_dst_drop(skb);
+	nf_reset(skb);
+}
+
 /**
  *	skb_tunnel_rx - prepare skb for rx reinsert
  *	@skb: buffer
@@ -234,17 +251,14 @@ static inline void skb_dst_force(struct sk_buff *skb)
  *
  *	After decapsulation, packet is going to re-enter (netif_rx()) our stack,
  *	so make some cleanups, and perform accounting.
+ *	Note: this accounting is not SMP safe.
  */
 static inline void skb_tunnel_rx(struct sk_buff *skb, struct net_device *dev)
 {
-	skb->dev = dev;
 	/* TODO : stats should be SMP safe */
 	dev->stats.rx_packets++;
 	dev->stats.rx_bytes += skb->len;
-	skb->rxhash = 0;
-	skb_set_queue_mapping(skb, 0);
-	skb_dst_drop(skb);
-	nf_reset(skb);
+	__skb_tunnel_rx(skb, dev);
 }
 
 /* Children define the path of the packet through the
diff --git a/include/net/ipip.h b/include/net/ipip.h
index 65caea8b414f..58abbf966b0c 100644
--- a/include/net/ipip.h
+++ b/include/net/ipip.h
@@ -45,7 +45,7 @@ struct ip_tunnel_prl_entry {
 	struct rcu_head			rcu_head;
 };
 
-#define IPTUNNEL_XMIT() do {						\
+#define __IPTUNNEL_XMIT(stats1, stats2) do {				\
 	int err;							\
 	int pkt_len = skb->len - skb_transport_offset(skb);		\
 									\
@@ -54,12 +54,14 @@ struct ip_tunnel_prl_entry {
 									\
 	err = ip_local_out(skb);					\
 	if (likely(net_xmit_eval(err) == 0)) {				\
-		txq->tx_bytes += pkt_len;				\
-		txq->tx_packets++;					\
+		(stats1)->tx_bytes += pkt_len;				\
+		(stats1)->tx_packets++;					\
 	} else {							\
-		stats->tx_errors++;					\
-		stats->tx_aborted_errors++;				\
+		(stats2)->tx_errors++;					\
+		(stats2)->tx_aborted_errors++;				\
 	}								\
 } while (0)
 
+#define IPTUNNEL_XMIT() __IPTUNNEL_XMIT(txq, stats)
+
 #endif
-- 
cgit v1.2.3


From 62fe0b40abb3484413800edaef9b087a20059acf Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Mon, 27 Sep 2010 08:24:33 +0000
Subject: net: Allow changing number of RX queues after device allocation

For RPS, we create a kobject for each RX queue based on the number of
queues passed to alloc_netdev_mq().  However, drivers generally do not
determine the numbers of hardware queues to use until much later, so
this usually represents the maximum number the driver may use and not
the actual number in use.

For TX queues, drivers can update the actual number using
netif_set_real_num_tx_queues().  Add a corresponding function for RX
queues, netif_set_real_num_rx_queues().

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 16 +++++++++++++++-
 net/core/dev.c            | 45 +++++++++++++++++++++++++++++++++++++++++----
 net/core/net-sysfs.c      | 32 ++++++++++++++++++--------------
 net/core/net-sysfs.h      |  4 ++++
 4 files changed, 78 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 83de0eb7a071..b15732e22eee 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -976,8 +976,11 @@ struct net_device {
 
 	struct netdev_rx_queue	*_rx;
 
-	/* Number of RX queues allocated at alloc_netdev_mq() time  */
+	/* Number of RX queues allocated at register_netdev() time */
 	unsigned int		num_rx_queues;
+
+	/* Number of RX queues currently active in device */
+	unsigned int		real_num_rx_queues;
 #endif
 
 	rx_handler_func_t	*rx_handler;
@@ -1685,6 +1688,17 @@ static inline int netif_is_multiqueue(const struct net_device *dev)
 extern void netif_set_real_num_tx_queues(struct net_device *dev,
 					 unsigned int txq);
 
+#ifdef CONFIG_RPS
+extern int netif_set_real_num_rx_queues(struct net_device *dev,
+					unsigned int rxq);
+#else
+static inline int netif_set_real_num_rx_queues(struct net_device *dev,
+						unsigned int rxq)
+{
+	return 0;
+}
+#endif
+
 /* Use this variant when it is known for sure that it
  * is executing from hardware interrupt context or with hardware interrupts
  * disabled.
diff --git a/net/core/dev.c b/net/core/dev.c
index 42b200fdf12e..48ad47f402ad 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1567,6 +1567,41 @@ void netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq)
 }
 EXPORT_SYMBOL(netif_set_real_num_tx_queues);
 
+#ifdef CONFIG_RPS
+/**
+ *	netif_set_real_num_rx_queues - set actual number of RX queues used
+ *	@dev: Network device
+ *	@rxq: Actual number of RX queues
+ *
+ *	This must be called either with the rtnl_lock held or before
+ *	registration of the net device.  Returns 0 on success, or a
+ *	negative error code.  If called before registration, it also
+ *	sets the maximum number of queues, and always succeeds.
+ */
+int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq)
+{
+	int rc;
+
+	if (dev->reg_state == NETREG_REGISTERED) {
+		ASSERT_RTNL();
+
+		if (rxq > dev->num_rx_queues)
+			return -EINVAL;
+
+		rc = net_rx_queue_update_kobjects(dev, dev->real_num_rx_queues,
+						  rxq);
+		if (rc)
+			return rc;
+	} else {
+		dev->num_rx_queues = rxq;
+	}
+
+	dev->real_num_rx_queues = rxq;
+	return 0;
+}
+EXPORT_SYMBOL(netif_set_real_num_rx_queues);
+#endif
+
 static inline void __netif_reschedule(struct Qdisc *q)
 {
 	struct softnet_data *sd;
@@ -2352,10 +2387,11 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb,
 
 	if (skb_rx_queue_recorded(skb)) {
 		u16 index = skb_get_rx_queue(skb);
-		if (unlikely(index >= dev->num_rx_queues)) {
-			WARN_ONCE(dev->num_rx_queues > 1, "%s received packet "
-				"on queue %u, but number of RX queues is %u\n",
-				dev->name, index, dev->num_rx_queues);
+		if (unlikely(index >= dev->real_num_rx_queues)) {
+			WARN_ONCE(dev->real_num_rx_queues > 1,
+				  "%s received packet on queue %u, but number "
+				  "of RX queues is %u\n",
+				  dev->name, index, dev->real_num_rx_queues);
 			goto done;
 		}
 		rxqueue = dev->_rx + index;
@@ -5472,6 +5508,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 
 #ifdef CONFIG_RPS
 	dev->num_rx_queues = queue_count;
+	dev->real_num_rx_queues = queue_count;
 #endif
 
 	dev->gso_max_size = GSO_MAX_SIZE;
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 76485a3f910b..fa81fd0a488f 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -742,34 +742,38 @@ static int rx_queue_add_kobject(struct net_device *net, int index)
 	return error;
 }
 
-static int rx_queue_register_kobjects(struct net_device *net)
+int
+net_rx_queue_update_kobjects(struct net_device *net, int old_num, int new_num)
 {
 	int i;
 	int error = 0;
 
-	net->queues_kset = kset_create_and_add("queues",
-	    NULL, &net->dev.kobj);
-	if (!net->queues_kset)
-		return -ENOMEM;
-	for (i = 0; i < net->num_rx_queues; i++) {
+	for (i = old_num; i < new_num; i++) {
 		error = rx_queue_add_kobject(net, i);
-		if (error)
+		if (error) {
+			new_num = old_num;
 			break;
+		}
 	}
 
-	if (error)
-		while (--i >= 0)
-			kobject_put(&net->_rx[i].kobj);
+	while (--i >= new_num)
+		kobject_put(&net->_rx[i].kobj);
 
 	return error;
 }
 
-static void rx_queue_remove_kobjects(struct net_device *net)
+static int rx_queue_register_kobjects(struct net_device *net)
 {
-	int i;
+	net->queues_kset = kset_create_and_add("queues",
+	    NULL, &net->dev.kobj);
+	if (!net->queues_kset)
+		return -ENOMEM;
+	return net_rx_queue_update_kobjects(net, 0, net->real_num_rx_queues);
+}
 
-	for (i = 0; i < net->num_rx_queues; i++)
-		kobject_put(&net->_rx[i].kobj);
+static void rx_queue_remove_kobjects(struct net_device *net)
+{
+	net_rx_queue_update_kobjects(net, net->real_num_rx_queues, 0);
 	kset_unregister(net->queues_kset);
 }
 #endif /* CONFIG_RPS */
diff --git a/net/core/net-sysfs.h b/net/core/net-sysfs.h
index 805555e8b187..778e1571548d 100644
--- a/net/core/net-sysfs.h
+++ b/net/core/net-sysfs.h
@@ -4,4 +4,8 @@
 int netdev_kobject_init(void);
 int netdev_register_kobject(struct net_device *);
 void netdev_unregister_kobject(struct net_device *);
+#ifdef CONFIG_RPS
+int net_rx_queue_update_kobjects(struct net_device *, int old_num, int new_num);
+#endif
+
 #endif
-- 
cgit v1.2.3


From 3171d026291d08c2a4cfe06302ce308b09605c4b Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Mon, 27 Sep 2010 08:24:49 +0000
Subject: net: Add netif_copy_real_num_queues() for use by virtual net drivers

This sets the active numbers of queues on a net device to match another.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b15732e22eee..c2bec990bd17 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1699,6 +1699,18 @@ static inline int netif_set_real_num_rx_queues(struct net_device *dev,
 }
 #endif
 
+static inline int netif_copy_real_num_queues(struct net_device *to_dev,
+					     const struct net_device *from_dev)
+{
+	netif_set_real_num_tx_queues(to_dev, from_dev->real_num_tx_queues);
+#ifdef CONFIG_RPS
+	return netif_set_real_num_rx_queues(to_dev,
+					    from_dev->real_num_rx_queues);
+#else
+	return 0;
+#endif
+}
+
 /* Use this variant when it is known for sure that it
  * is executing from hardware interrupt context or with hardware interrupts
  * disabled.
-- 
cgit v1.2.3


From bc01befdcf3e40979eb518085a075cbf0aacede0 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Tue, 28 Sep 2010 21:06:34 +0200
Subject: netfilter: ctnetlink: add support for user-space expectation helpers

This patch adds the basic infrastructure to support user-space
expectation helpers via ctnetlink and the netfilter queuing
infrastructure NFQUEUE. Basically, this patch:

* adds NF_CT_EXPECT_USERSPACE flag to identify user-space
  created expectations. I have also added a sanity check in
  __nf_ct_expect_check() to avoid that kernel-space helpers
  may create an expectation if the master conntrack has no
  helper assigned.
* adds some branches to check if the master conntrack helper
  exists, otherwise we skip the code that refers to kernel-space
  helper such as the local expectation list and the expectation
  policy.
* allows to set the timeout for user-space expectations with
  no helper assigned.
* a list of expectations created from user-space that depends
  on ctnetlink (if this module is removed, they are deleted).
* includes USERSPACE in the /proc output for expectations
  that have been created by a user-space helper.

This patch also modifies ctnetlink to skip including the helper
name in the Netlink messages if no kernel-space helper is set
(since no user-space expectation has not kernel-space kernel
assigned).

You can access an example user-space FTP conntrack helper at:
http://people.netfilter.org/pablo/userspace-conntrack-helpers/nf-ftp-helper-userspace-POC.tar.bz

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/nf_conntrack_common.h |  1 +
 include/net/netfilter/nf_conntrack_expect.h   |  1 +
 net/netfilter/nf_conntrack_expect.c           | 62 ++++++++++++++++++++-------
 net/netfilter/nf_conntrack_netlink.c          | 46 +++++++++++++-------
 4 files changed, 79 insertions(+), 31 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h
index fdc50cae861f..23a1a08578a8 100644
--- a/include/linux/netfilter/nf_conntrack_common.h
+++ b/include/linux/netfilter/nf_conntrack_common.h
@@ -103,6 +103,7 @@ enum ip_conntrack_expect_events {
 /* expectation flags */
 #define NF_CT_EXPECT_PERMANENT		0x1
 #define NF_CT_EXPECT_INACTIVE		0x2
+#define NF_CT_EXPECT_USERSPACE		0x4
 
 #ifdef __KERNEL__
 struct ip_conntrack_stat {
diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index 96bb42af5fae..416b83844485 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -85,6 +85,7 @@ nf_ct_find_expectation(struct net *net, u16 zone,
 void nf_ct_unlink_expect(struct nf_conntrack_expect *exp);
 void nf_ct_remove_expectations(struct nf_conn *ct);
 void nf_ct_unexpect_related(struct nf_conntrack_expect *exp);
+void nf_ct_remove_userspace_expectations(void);
 
 /* Allocate space for an expectation: this is mandatory before calling
    nf_ct_expect_related.  You will have to call put afterwards. */
diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index acb29ccaa41f..b30a1f2aac00 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -38,20 +38,23 @@ static int nf_ct_expect_hash_rnd_initted __read_mostly;
 
 static struct kmem_cache *nf_ct_expect_cachep __read_mostly;
 
+static HLIST_HEAD(nf_ct_userspace_expect_list);
+
 /* nf_conntrack_expect helper functions */
 void nf_ct_unlink_expect(struct nf_conntrack_expect *exp)
 {
 	struct nf_conn_help *master_help = nfct_help(exp->master);
 	struct net *net = nf_ct_exp_net(exp);
 
-	NF_CT_ASSERT(master_help);
 	NF_CT_ASSERT(!timer_pending(&exp->timeout));
 
 	hlist_del_rcu(&exp->hnode);
 	net->ct.expect_count--;
 
 	hlist_del(&exp->lnode);
-	master_help->expecting[exp->class]--;
+	if (!(exp->flags & NF_CT_EXPECT_USERSPACE))
+		master_help->expecting[exp->class]--;
+
 	nf_ct_expect_put(exp);
 
 	NF_CT_STAT_INC(net, expect_delete);
@@ -320,16 +323,21 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp)
 
 	atomic_inc(&exp->use);
 
-	hlist_add_head(&exp->lnode, &master_help->expectations);
-	master_help->expecting[exp->class]++;
+	if (master_help) {
+		hlist_add_head(&exp->lnode, &master_help->expectations);
+		master_help->expecting[exp->class]++;
+	} else if (exp->flags & NF_CT_EXPECT_USERSPACE)
+		hlist_add_head(&exp->lnode, &nf_ct_userspace_expect_list);
 
 	hlist_add_head_rcu(&exp->hnode, &net->ct.expect_hash[h]);
 	net->ct.expect_count++;
 
 	setup_timer(&exp->timeout, nf_ct_expectation_timed_out,
 		    (unsigned long)exp);
-	p = &master_help->helper->expect_policy[exp->class];
-	exp->timeout.expires = jiffies + p->timeout * HZ;
+	if (master_help) {
+		p = &master_help->helper->expect_policy[exp->class];
+		exp->timeout.expires = jiffies + p->timeout * HZ;
+	}
 	add_timer(&exp->timeout);
 
 	atomic_inc(&exp->use);
@@ -380,7 +388,9 @@ static inline int __nf_ct_expect_check(struct nf_conntrack_expect *expect)
 	unsigned int h;
 	int ret = 1;
 
-	if (!master_help->helper) {
+	/* Don't allow expectations created from kernel-space with no helper */
+	if (!(expect->flags & NF_CT_EXPECT_USERSPACE) &&
+	    (!master_help || (master_help && !master_help->helper))) {
 		ret = -ESHUTDOWN;
 		goto out;
 	}
@@ -398,13 +408,16 @@ static inline int __nf_ct_expect_check(struct nf_conntrack_expect *expect)
 		}
 	}
 	/* Will be over limit? */
-	p = &master_help->helper->expect_policy[expect->class];
-	if (p->max_expected &&
-	    master_help->expecting[expect->class] >= p->max_expected) {
-		evict_oldest_expect(master, expect);
-		if (master_help->expecting[expect->class] >= p->max_expected) {
-			ret = -EMFILE;
-			goto out;
+	if (master_help) {
+		p = &master_help->helper->expect_policy[expect->class];
+		if (p->max_expected &&
+		    master_help->expecting[expect->class] >= p->max_expected) {
+			evict_oldest_expect(master, expect);
+			if (master_help->expecting[expect->class]
+						>= p->max_expected) {
+				ret = -EMFILE;
+				goto out;
+			}
 		}
 	}
 
@@ -439,6 +452,21 @@ out:
 }
 EXPORT_SYMBOL_GPL(nf_ct_expect_related_report);
 
+void nf_ct_remove_userspace_expectations(void)
+{
+	struct nf_conntrack_expect *exp;
+	struct hlist_node *n, *next;
+
+	hlist_for_each_entry_safe(exp, n, next,
+				  &nf_ct_userspace_expect_list, lnode) {
+		if (del_timer(&exp->timeout)) {
+			nf_ct_unlink_expect(exp);
+			nf_ct_expect_put(exp);
+		}
+	}
+}
+EXPORT_SYMBOL_GPL(nf_ct_remove_userspace_expectations);
+
 #ifdef CONFIG_PROC_FS
 struct ct_expect_iter_state {
 	struct seq_net_private p;
@@ -529,8 +557,12 @@ static int exp_seq_show(struct seq_file *s, void *v)
 		seq_printf(s, "PERMANENT");
 		delim = ",";
 	}
-	if (expect->flags & NF_CT_EXPECT_INACTIVE)
+	if (expect->flags & NF_CT_EXPECT_INACTIVE) {
 		seq_printf(s, "%sINACTIVE", delim);
+		delim = ",";
+	}
+	if (expect->flags & NF_CT_EXPECT_USERSPACE)
+		seq_printf(s, "%sUSERSPACE", delim);
 
 	helper = rcu_dereference(nfct_help(expect->master)->helper);
 	if (helper) {
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 0804e0ef6500..b4077be5f663 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -1560,8 +1560,8 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb,
 			  const struct nf_conntrack_expect *exp)
 {
 	struct nf_conn *master = exp->master;
-	struct nf_conntrack_helper *helper;
 	long timeout = (exp->timeout.expires - jiffies) / HZ;
+	struct nf_conn_help *help;
 
 	if (timeout < 0)
 		timeout = 0;
@@ -1578,9 +1578,14 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb,
 	NLA_PUT_BE32(skb, CTA_EXPECT_TIMEOUT, htonl(timeout));
 	NLA_PUT_BE32(skb, CTA_EXPECT_ID, htonl((unsigned long)exp));
 	NLA_PUT_BE32(skb, CTA_EXPECT_FLAGS, htonl(exp->flags));
-	helper = rcu_dereference(nfct_help(master)->helper);
-	if (helper)
-		NLA_PUT_STRING(skb, CTA_EXPECT_HELP_NAME, helper->name);
+	help = nfct_help(master);
+	if (help) {
+		struct nf_conntrack_helper *helper;
+
+		helper = rcu_dereference(help->helper);
+		if (helper)
+			NLA_PUT_STRING(skb, CTA_EXPECT_HELP_NAME, helper->name);
+	}
 
 	return 0;
 
@@ -1921,24 +1926,32 @@ ctnetlink_create_expect(struct net *net, u16 zone,
 	if (!h)
 		return -ENOENT;
 	ct = nf_ct_tuplehash_to_ctrack(h);
-	help = nfct_help(ct);
-
-	if (!help || !help->helper) {
-		/* such conntrack hasn't got any helper, abort */
-		err = -EOPNOTSUPP;
-		goto out;
-	}
-
 	exp = nf_ct_expect_alloc(ct);
 	if (!exp) {
 		err = -ENOMEM;
 		goto out;
 	}
+	help = nfct_help(ct);
+	if (!help) {
+		if (!cda[CTA_EXPECT_TIMEOUT]) {
+			err = -EINVAL;
+			goto out;
+		}
+		exp->timeout.expires =
+		  jiffies + ntohl(nla_get_be32(cda[CTA_EXPECT_TIMEOUT])) * HZ;
 
-	if (cda[CTA_EXPECT_FLAGS])
-		exp->flags = ntohl(nla_get_be32(cda[CTA_EXPECT_FLAGS]));
-	else
-		exp->flags = 0;
+		exp->flags = NF_CT_EXPECT_USERSPACE;
+		if (cda[CTA_EXPECT_FLAGS]) {
+			exp->flags |=
+				ntohl(nla_get_be32(cda[CTA_EXPECT_FLAGS]));
+		}
+	} else {
+		if (cda[CTA_EXPECT_FLAGS]) {
+			exp->flags = ntohl(nla_get_be32(cda[CTA_EXPECT_FLAGS]));
+			exp->flags &= ~NF_CT_EXPECT_USERSPACE;
+		} else
+			exp->flags = 0;
+	}
 
 	exp->class = 0;
 	exp->expectfn = NULL;
@@ -2109,6 +2122,7 @@ static void __exit ctnetlink_exit(void)
 {
 	pr_info("ctnetlink: unregistering from nfnetlink.\n");
 
+	nf_ct_remove_userspace_expectations();
 #ifdef CONFIG_NF_CONNTRACK_EVENTS
 	nf_ct_expect_unregister_notifier(&ctnl_notifier_exp);
 	nf_conntrack_unregister_notifier(&ctnl_notifier);
-- 
cgit v1.2.3


From 6d81f41c58c69ddde497e9e640ba5805aa26e78c Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 27 Sep 2010 20:50:33 +0000
Subject: dummy: percpu stats and lockless xmit

Converts dummy network device driver to :

- percpu stats

- 64bit stats

- lockless xmit (NETIF_F_LLTX)

- performance features added (NETIF_F_SG | NETIF_F_FRAGLIST |
NETIF_F_TSO | NETIF_F_NO_CSUM | NETIF_F_HIGHDMA)

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dummy.c       | 58 ++++++++++++++++++++++++++++++++++++++++++++---
 include/linux/netdevice.h |  1 +
 2 files changed, 56 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/dummy.c b/drivers/net/dummy.c
index 37dcfdc63456..ff2d29b17858 100644
--- a/drivers/net/dummy.c
+++ b/drivers/net/dummy.c
@@ -36,6 +36,7 @@
 #include <linux/moduleparam.h>
 #include <linux/rtnetlink.h>
 #include <net/rtnetlink.h>
+#include <linux/u64_stats_sync.h>
 
 static int numdummies = 1;
 
@@ -55,21 +56,69 @@ static void set_multicast_list(struct net_device *dev)
 {
 }
 
+struct pcpu_dstats {
+	u64			tx_packets;
+	u64			tx_bytes;
+	struct u64_stats_sync	syncp;
+};
+
+static struct rtnl_link_stats64 *dummy_get_stats64(struct net_device *dev,
+						   struct rtnl_link_stats64 *stats)
+{
+	int i;
+
+	for_each_possible_cpu(i) {
+		const struct pcpu_dstats *dstats;
+		u64 tbytes, tpackets;
+		unsigned int start;
+
+		dstats = per_cpu_ptr(dev->dstats, i);
+		do {
+			start = u64_stats_fetch_begin(&dstats->syncp);
+			tbytes = dstats->tx_bytes;
+			tpackets = dstats->tx_packets;
+		} while (u64_stats_fetch_retry(&dstats->syncp, start));
+		stats->tx_bytes += tbytes;
+		stats->tx_packets += tpackets;
+	}
+	return stats;
+}
 
 static netdev_tx_t dummy_xmit(struct sk_buff *skb, struct net_device *dev)
 {
-	dev->stats.tx_packets++;
-	dev->stats.tx_bytes += skb->len;
+	struct pcpu_dstats *dstats = this_cpu_ptr(dev->dstats);
+
+	u64_stats_update_begin(&dstats->syncp);
+	dstats->tx_packets++;
+	dstats->tx_bytes += skb->len;
+	u64_stats_update_end(&dstats->syncp);
 
 	dev_kfree_skb(skb);
 	return NETDEV_TX_OK;
 }
 
+static int dummy_dev_init(struct net_device *dev)
+{
+	dev->dstats = alloc_percpu(struct pcpu_dstats);
+	if (!dev->dstats)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void dummy_dev_free(struct net_device *dev)
+{
+	free_percpu(dev->dstats);
+	free_netdev(dev);
+}
+
 static const struct net_device_ops dummy_netdev_ops = {
+	.ndo_init		= dummy_dev_init,
 	.ndo_start_xmit		= dummy_xmit,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_multicast_list = set_multicast_list,
 	.ndo_set_mac_address	= dummy_set_address,
+	.ndo_get_stats64	= dummy_get_stats64,
 };
 
 static void dummy_setup(struct net_device *dev)
@@ -78,14 +127,17 @@ static void dummy_setup(struct net_device *dev)
 
 	/* Initialize the device structure. */
 	dev->netdev_ops = &dummy_netdev_ops;
-	dev->destructor = free_netdev;
+	dev->destructor = dummy_dev_free;
 
 	/* Fill in device structure with ethernet-generic values. */
 	dev->tx_queue_len = 0;
 	dev->flags |= IFF_NOARP;
 	dev->flags &= ~IFF_MULTICAST;
+	dev->features	|= NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_TSO;
+	dev->features	|= NETIF_F_NO_CSUM | NETIF_F_HIGHDMA | NETIF_F_LLTX;
 	random_ether_addr(dev->dev_addr);
 }
+
 static int dummy_validate(struct nlattr *tb[], struct nlattr *data[])
 {
 	if (tb[IFLA_ADDRESS]) {
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c2bec990bd17..6f0845e0b888 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1057,6 +1057,7 @@ struct net_device {
 		void				*ml_priv;
 		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
 		struct pcpu_tstats __percpu	*tstats; /* tunnel stats */
+		struct pcpu_dstats __percpu	*dstats; /* dummy stats */
 	};
 	/* GARP */
 	struct garp_port	*garp_port;
-- 
cgit v1.2.3


From bfa5ae63b823f4ffd3483a05f60a93a4a7b7d680 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 28 Sep 2010 05:58:37 +0000
Subject: net: rename netdev rx_queue to ingress_queue

There is some confusion with rx_queue name after RPS, and net drivers
private rx_queue fields.

I suggest to rename "struct net_device"->rx_queue to ingress_queue.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  2 +-
 net/core/dev.c            |  8 ++++----
 net/sched/sch_api.c       | 14 +++++++-------
 net/sched/sch_generic.c   |  8 ++++----
 4 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 6f0845e0b888..ceed3474014a 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -986,7 +986,7 @@ struct net_device {
 	rx_handler_func_t	*rx_handler;
 	void			*rx_handler_data;
 
-	struct netdev_queue	rx_queue; /* use two cache lines */
+	struct netdev_queue	ingress_queue; /* use two cache lines */
 
 /*
  * Cache lines mostly used on transmit path
diff --git a/net/core/dev.c b/net/core/dev.c
index 50daccad6a53..a313bab1b754 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2720,7 +2720,7 @@ static int ing_filter(struct sk_buff *skb)
 	skb->tc_verd = SET_TC_RTTL(skb->tc_verd, ttl);
 	skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS);
 
-	rxq = &dev->rx_queue;
+	rxq = &dev->ingress_queue;
 
 	q = rxq->qdisc;
 	if (q != &noop_qdisc) {
@@ -2737,7 +2737,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb,
 					 struct packet_type **pt_prev,
 					 int *ret, struct net_device *orig_dev)
 {
-	if (skb->dev->rx_queue.qdisc == &noop_qdisc)
+	if (skb->dev->ingress_queue.qdisc == &noop_qdisc)
 		goto out;
 
 	if (*pt_prev) {
@@ -4940,7 +4940,7 @@ static void __netdev_init_queue_locks_one(struct net_device *dev,
 static void netdev_init_queue_locks(struct net_device *dev)
 {
 	netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL);
-	__netdev_init_queue_locks_one(dev, &dev->rx_queue, NULL);
+	__netdev_init_queue_locks_one(dev, &dev->ingress_queue, NULL);
 }
 
 unsigned long netdev_fix_features(unsigned long features, const char *name)
@@ -5452,7 +5452,7 @@ static void netdev_init_one_queue(struct net_device *dev,
 
 static void netdev_init_queues(struct net_device *dev)
 {
-	netdev_init_one_queue(dev, &dev->rx_queue, NULL);
+	netdev_init_one_queue(dev, &dev->ingress_queue, NULL);
 	netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL);
 	spin_lock_init(&dev->tx_global_lock);
 }
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 6fb3d41c0e41..b8020784d0e9 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -240,7 +240,7 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 	if (q)
 		goto out;
 
-	q = qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle);
+	q = qdisc_match_from_root(dev->ingress_queue.qdisc_sleeping, handle);
 out:
 	return q;
 }
@@ -701,7 +701,7 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent,
 		}
 
 		for (i = 0; i < num_q; i++) {
-			struct netdev_queue *dev_queue = &dev->rx_queue;
+			struct netdev_queue *dev_queue = &dev->ingress_queue;
 
 			if (!ingress)
 				dev_queue = netdev_get_tx_queue(dev, i);
@@ -979,7 +979,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 					return -ENOENT;
 				q = qdisc_leaf(p, clid);
 			} else { /* ingress */
-				q = dev->rx_queue.qdisc_sleeping;
+				q = dev->ingress_queue.qdisc_sleeping;
 			}
 		} else {
 			q = dev->qdisc;
@@ -1044,7 +1044,7 @@ replay:
 					return -ENOENT;
 				q = qdisc_leaf(p, clid);
 			} else { /*ingress */
-				q = dev->rx_queue.qdisc_sleeping;
+				q = dev->ingress_queue.qdisc_sleeping;
 			}
 		} else {
 			q = dev->qdisc;
@@ -1124,7 +1124,7 @@ create_n_graft:
 	if (!(n->nlmsg_flags&NLM_F_CREATE))
 		return -ENOENT;
 	if (clid == TC_H_INGRESS)
-		q = qdisc_create(dev, &dev->rx_queue, p,
+		q = qdisc_create(dev, &dev->ingress_queue, p,
 				 tcm->tcm_parent, tcm->tcm_parent,
 				 tca, &err);
 	else {
@@ -1304,7 +1304,7 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 		if (tc_dump_qdisc_root(dev->qdisc, skb, cb, &q_idx, s_q_idx) < 0)
 			goto done;
 
-		dev_queue = &dev->rx_queue;
+		dev_queue = &dev->ingress_queue;
 		if (tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, &q_idx, s_q_idx) < 0)
 			goto done;
 
@@ -1595,7 +1595,7 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	if (tc_dump_tclass_root(dev->qdisc, skb, tcm, cb, &t, s_t) < 0)
 		goto done;
 
-	dev_queue = &dev->rx_queue;
+	dev_queue = &dev->ingress_queue;
 	if (tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, &t, s_t) < 0)
 		goto done;
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 2aeb3a4386a1..545278a1c478 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -753,7 +753,7 @@ void dev_activate(struct net_device *dev)
 
 	need_watchdog = 0;
 	netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog);
-	transition_one_qdisc(dev, &dev->rx_queue, NULL);
+	transition_one_qdisc(dev, &dev->ingress_queue, NULL);
 
 	if (need_watchdog) {
 		dev->trans_start = jiffies;
@@ -812,7 +812,7 @@ static bool some_qdisc_is_busy(struct net_device *dev)
 void dev_deactivate(struct net_device *dev)
 {
 	netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc);
-	dev_deactivate_queue(dev, &dev->rx_queue, &noop_qdisc);
+	dev_deactivate_queue(dev, &dev->ingress_queue, &noop_qdisc);
 
 	dev_watchdog_down(dev);
 
@@ -838,7 +838,7 @@ void dev_init_scheduler(struct net_device *dev)
 {
 	dev->qdisc = &noop_qdisc;
 	netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc);
-	dev_init_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc);
+	dev_init_scheduler_queue(dev, &dev->ingress_queue, &noop_qdisc);
 
 	setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev);
 }
@@ -861,7 +861,7 @@ static void shutdown_scheduler_queue(struct net_device *dev,
 void dev_shutdown(struct net_device *dev)
 {
 	netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
-	shutdown_scheduler_queue(dev, &dev->rx_queue, &noop_qdisc);
+	shutdown_scheduler_queue(dev, &dev->ingress_queue, &noop_qdisc);
 	qdisc_destroy(dev->qdisc);
 	dev->qdisc = &noop_qdisc;
 
-- 
cgit v1.2.3


From 65836112fc24bdf009554481b36b6ba0a690b855 Mon Sep 17 00:00:00 2001
From: Ohad Ben-Cohen <ohad@wizery.com>
Date: Tue, 28 Sep 2010 20:20:28 +0200
Subject: wl12xx: fix non-wl12xx build scenarios

Support building wl1271-equipped boards without building the
wl1271 driver itself, e.g.:

CONFIG_MACH_OMAP_ZOOM3=y
CONFIG_WL12XX is not set

Reported-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Luciano Coelho <luciano.coelho@nokia.com>
---
 include/linux/wl12xx.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/wl12xx.h b/include/linux/wl12xx.h
index 95deae3968f4..4f902e1908aa 100644
--- a/include/linux/wl12xx.h
+++ b/include/linux/wl12xx.h
@@ -32,7 +32,20 @@ struct wl12xx_platform_data {
 	int board_ref_clock;
 };
 
+#ifdef CONFIG_WL12XX_PLATFORM_DATA
+
 int wl12xx_set_platform_data(const struct wl12xx_platform_data *data);
+
+#else
+
+static inline
+int wl12xx_set_platform_data(const struct wl12xx_platform_data *data)
+{
+	return -ENOSYS;
+}
+
+#endif
+
 const struct wl12xx_platform_data *wl12xx_get_platform_data(void);
 
 #endif
-- 
cgit v1.2.3


From 6110a1f43c27b516e16d5ce8860fca50748c2a87 Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Thu, 30 Sep 2010 21:19:07 -0400
Subject: intel_idle: Voluntary leave_mm before entering deeper

Avoid TLB flush IPIs for the cores in deeper c-states by voluntary leave_mm()
before entering into that state. CPUs tend to flush TLB in those c-states
anyways.

acpi_idle does this with C3-type states, but it was not caried over
when intel_idle was introduced.  intel_idle can apply it
to C-states in addition to those that ACPI might export as C3...

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/idle/intel_idle.c | 18 ++++++++++++++----
 include/linux/cpuidle.h   |  1 +
 2 files changed, 15 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 96bf38097996..0906fc5b69b9 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -108,7 +108,7 @@ static struct cpuidle_state nehalem_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.name = "NHM-C3",
 		.desc = "MWAIT 0x10",
 		.driver_data = (void *) 0x10,
-		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
 		.exit_latency = 20,
 		.power_usage = 500,
 		.target_residency = 80,
@@ -117,7 +117,7 @@ static struct cpuidle_state nehalem_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.name = "NHM-C6",
 		.desc = "MWAIT 0x20",
 		.driver_data = (void *) 0x20,
-		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
 		.exit_latency = 200,
 		.power_usage = 350,
 		.target_residency = 800,
@@ -149,7 +149,7 @@ static struct cpuidle_state atom_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.name = "ATM-C4",
 		.desc = "MWAIT 0x30",
 		.driver_data = (void *) 0x30,
-		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
 		.exit_latency = 100,
 		.power_usage = 250,
 		.target_residency = 400,
@@ -159,7 +159,7 @@ static struct cpuidle_state atom_cstates[MWAIT_MAX_NUM_CSTATES] = {
 		.name = "ATM-C6",
 		.desc = "MWAIT 0x40",
 		.driver_data = (void *) 0x40,
-		.flags = CPUIDLE_FLAG_TIME_VALID,
+		.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
 		.exit_latency = 200,
 		.power_usage = 150,
 		.target_residency = 800,
@@ -185,6 +185,16 @@ static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state)
 
 	local_irq_disable();
 
+	/*
+	 * If the state flag indicates that the TLB will be flushed or if this
+	 * is the deepest c-state supported, do a voluntary leave mm to avoid
+	 * costly and mostly unnecessary wakeups for flushing the user TLB's
+	 * associated with the active mm.
+	 */
+	if (state->flags & CPUIDLE_FLAG_TLB_FLUSHED ||
+	    (&dev->states[dev->state_count - 1] == state))
+		leave_mm(cpu);
+
 	if (!(lapic_timer_reliable_states & (1 << (cstate))))
 		clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &cpu);
 
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index 36ca9721a0c2..1be416bbbb82 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -53,6 +53,7 @@ struct cpuidle_state {
 #define CPUIDLE_FLAG_BALANCED	(0x40) /* medium latency, moderate savings */
 #define CPUIDLE_FLAG_DEEP	(0x80) /* high latency, large savings */
 #define CPUIDLE_FLAG_IGNORE	(0x100) /* ignore during this idle period */
+#define CPUIDLE_FLAG_TLB_FLUSHED (0x200) /* tlb will be flushed */
 
 #define CPUIDLE_DRIVER_FLAGS_MASK (0xFFFF0000)
 
-- 
cgit v1.2.3


From 82efee1499a27c06f5afb11b07db384fdb3f7004 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 30 Sep 2010 03:31:56 +0000
Subject: ipv4: introduce __ip_dev_find()

ip_dev_find(net, addr) finds a device given an IPv4 source address and
takes a reference on it.

Introduce __ip_dev_find(), taking a third argument, to optionally take
the device reference. Callers not asking the reference to be taken
should be in an rcu_read_lock() protected section.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/inetdevice.h |  7 ++++++-
 net/ipv4/fib_frontend.c    | 32 +++++++++++++++++++-------------
 2 files changed, 25 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index 1ec09bb4a3ab..ccd5b07d678d 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -159,7 +159,12 @@ struct in_ifaddr {
 extern int register_inetaddr_notifier(struct notifier_block *nb);
 extern int unregister_inetaddr_notifier(struct notifier_block *nb);
 
-extern struct net_device *ip_dev_find(struct net *net, __be32 addr);
+extern struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref);
+static inline struct net_device *ip_dev_find(struct net *net, __be32 addr)
+{
+	return __ip_dev_find(net, addr, true);
+}
+
 extern int		inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b);
 extern int		devinet_ioctl(struct net *net, unsigned int cmd, void __user *);
 extern void		devinet_init(void);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 981f3c59b334..4a69a957872b 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -147,34 +147,40 @@ static void fib_flush(struct net *net)
 		rt_cache_flush(net, -1);
 }
 
-/*
- *	Find the first device with a given source address.
+/**
+ * __ip_dev_find - find the first device with a given source address.
+ * @net: the net namespace
+ * @addr: the source address
+ * @devref: if true, take a reference on the found device
+ *
+ * If a caller uses devref=false, it should be protected by RCU
  */
-
-struct net_device * ip_dev_find(struct net *net, __be32 addr)
+struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref)
 {
-	struct flowi fl = { .nl_u = { .ip4_u = { .daddr = addr } },
-			    .flags = FLOWI_FLAG_MATCH_ANY_IIF };
-	struct fib_result res;
+	struct flowi fl = {
+		.nl_u = {
+			.ip4_u = {
+				.daddr = addr
+			}
+		},
+		.flags = FLOWI_FLAG_MATCH_ANY_IIF
+	};
+	struct fib_result res = { 0 };
 	struct net_device *dev = NULL;
 
-#ifdef CONFIG_IP_MULTIPLE_TABLES
-	res.r = NULL;
-#endif
-
 	if (fib_lookup(net, &fl, &res))
 		return NULL;
 	if (res.type != RTN_LOCAL)
 		goto out;
 	dev = FIB_RES_DEV(res);
 
-	if (dev)
+	if (dev && devref)
 		dev_hold(dev);
 out:
 	fib_res_put(&res);
 	return dev;
 }
-EXPORT_SYMBOL(ip_dev_find);
+EXPORT_SYMBOL(__ip_dev_find);
 
 /*
  * Find address type as if only "dev" was present in the system. If
-- 
cgit v1.2.3


From 5c80cc78de46aef6cd5e714208da05c3f7f548f8 Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <andreas.herrmann3@amd.com>
Date: Thu, 30 Sep 2010 14:43:16 +0200
Subject: x86, amd_nb: Enable GART support for AMD family 0x15 CPUs

AMD CPU family 0x15 still supports GART for compatibility reasons.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
LKML-Reference: <20100930124316.GG20545@loge.amd.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 arch/x86/kernel/amd_nb.c | 4 +++-
 include/linux/pci_ids.h  | 1 +
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/amd_nb.c b/arch/x86/kernel/amd_nb.c
index 4ffc38df7ac5..8f6463d8ed0d 100644
--- a/arch/x86/kernel/amd_nb.c
+++ b/arch/x86/kernel/amd_nb.c
@@ -15,6 +15,7 @@ static u32 *flush_words;
 struct pci_device_id k8_nb_ids[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_MISC) },
 	{}
 };
 EXPORT_SYMBOL(k8_nb_ids);
@@ -45,7 +46,8 @@ int cache_k8_northbridges(void)
 		k8_northbridges.num++;
 
 	/* some CPU families (e.g. family 0x11) do not support GART */
-	if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10)
+	if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10 ||
+	    boot_cpu_data.x86 == 0x15)
 		k8_northbridges.gart_supported = 1;
 
 	k8_northbridges.nb_misc = kmalloc((k8_northbridges.num + 1) *
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 10d33309e9a6..edc0279be72f 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -514,6 +514,7 @@
 #define PCI_DEVICE_ID_AMD_11H_NB_DRAM	0x1302
 #define PCI_DEVICE_ID_AMD_11H_NB_MISC	0x1303
 #define PCI_DEVICE_ID_AMD_11H_NB_LINK	0x1304
+#define PCI_DEVICE_ID_AMD_15H_NB_MISC	0x1603
 #define PCI_DEVICE_ID_AMD_LANCE		0x2000
 #define PCI_DEVICE_ID_AMD_LANCE_HOME	0x2001
 #define PCI_DEVICE_ID_AMD_SCSI		0x2020
-- 
cgit v1.2.3


From a8c9486b816f74d4645144db9e8fa2f711c1fc4b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Fri, 1 Oct 2010 16:15:08 +0000
Subject: ipmr: RCU protection for mfc_cache_array

Use RCU & RTNL protection for mfc_cache_array[]

ipmr_cache_find() is called under rcu_read_lock();

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/mroute.h |  1 +
 net/ipv4/ipmr.c        | 87 +++++++++++++++++++++++++++-----------------------
 2 files changed, 48 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index fa04b246c9ae..0fa7a3a874c8 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -213,6 +213,7 @@ struct mfc_cache {
 			unsigned char ttls[MAXVIFS];	/* TTL thresholds		*/
 		} res;
 	} mfc_un;
+	struct rcu_head	rcu;
 };
 
 #define MFC_STATIC		1
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index e2db2ea616ff..cbb6dabe024f 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -577,11 +577,18 @@ static int vif_delete(struct mr_table *mrt, int vifi, int notify,
 	return 0;
 }
 
-static inline void ipmr_cache_free(struct mfc_cache *c)
+static void ipmr_cache_free_rcu(struct rcu_head *head)
 {
+	struct mfc_cache *c = container_of(head, struct mfc_cache, rcu);
+
 	kmem_cache_free(mrt_cachep, c);
 }
 
+static inline void ipmr_cache_free(struct mfc_cache *c)
+{
+	call_rcu(&c->rcu, ipmr_cache_free_rcu);
+}
+
 /* Destroy an unresolved cache entry, killing queued skbs
    and reporting error to netlink readers.
  */
@@ -781,6 +788,7 @@ static int vif_add(struct net *net, struct mr_table *mrt,
 	return 0;
 }
 
+/* called with rcu_read_lock() */
 static struct mfc_cache *ipmr_cache_find(struct mr_table *mrt,
 					 __be32 origin,
 					 __be32 mcastgrp)
@@ -788,7 +796,7 @@ static struct mfc_cache *ipmr_cache_find(struct mr_table *mrt,
 	int line = MFC_HASH(mcastgrp, origin);
 	struct mfc_cache *c;
 
-	list_for_each_entry(c, &mrt->mfc_cache_array[line], list) {
+	list_for_each_entry_rcu(c, &mrt->mfc_cache_array[line], list) {
 		if (c->mfc_origin == origin && c->mfc_mcastgrp == mcastgrp)
 			return c;
 	}
@@ -801,19 +809,20 @@ static struct mfc_cache *ipmr_cache_find(struct mr_table *mrt,
 static struct mfc_cache *ipmr_cache_alloc(void)
 {
 	struct mfc_cache *c = kmem_cache_zalloc(mrt_cachep, GFP_KERNEL);
-	if (c == NULL)
-		return NULL;
-	c->mfc_un.res.minvif = MAXVIFS;
+
+	if (c)
+		c->mfc_un.res.minvif = MAXVIFS;
 	return c;
 }
 
 static struct mfc_cache *ipmr_cache_alloc_unres(void)
 {
 	struct mfc_cache *c = kmem_cache_zalloc(mrt_cachep, GFP_ATOMIC);
-	if (c == NULL)
-		return NULL;
-	skb_queue_head_init(&c->mfc_un.unres.unresolved);
-	c->mfc_un.unres.expires = jiffies + 10*HZ;
+
+	if (c) {
+		skb_queue_head_init(&c->mfc_un.unres.unresolved);
+		c->mfc_un.unres.expires = jiffies + 10*HZ;
+	}
 	return c;
 }
 
@@ -1040,9 +1049,7 @@ static int ipmr_mfc_delete(struct mr_table *mrt, struct mfcctl *mfc)
 	list_for_each_entry_safe(c, next, &mrt->mfc_cache_array[line], list) {
 		if (c->mfc_origin == mfc->mfcc_origin.s_addr &&
 		    c->mfc_mcastgrp == mfc->mfcc_mcastgrp.s_addr) {
-			write_lock_bh(&mrt_lock);
-			list_del(&c->list);
-			write_unlock_bh(&mrt_lock);
+			list_del_rcu(&c->list);
 
 			ipmr_cache_free(c);
 			return 0;
@@ -1095,9 +1102,7 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt,
 	if (!mrtsock)
 		c->mfc_flags |= MFC_STATIC;
 
-	write_lock_bh(&mrt_lock);
-	list_add(&c->list, &mrt->mfc_cache_array[line]);
-	write_unlock_bh(&mrt_lock);
+	list_add_rcu(&c->list, &mrt->mfc_cache_array[line]);
 
 	/*
 	 *	Check to see if we resolved a queued list. If so we
@@ -1149,12 +1154,9 @@ static void mroute_clean_tables(struct mr_table *mrt)
 	 */
 	for (i = 0; i < MFC_LINES; i++) {
 		list_for_each_entry_safe(c, next, &mrt->mfc_cache_array[i], list) {
-			if (c->mfc_flags&MFC_STATIC)
+			if (c->mfc_flags & MFC_STATIC)
 				continue;
-			write_lock_bh(&mrt_lock);
-			list_del(&c->list);
-			write_unlock_bh(&mrt_lock);
-
+			list_del_rcu(&c->list);
 			ipmr_cache_free(c);
 		}
 	}
@@ -1422,19 +1424,19 @@ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg)
 		if (copy_from_user(&sr, arg, sizeof(sr)))
 			return -EFAULT;
 
-		read_lock(&mrt_lock);
+		rcu_read_lock();
 		c = ipmr_cache_find(mrt, sr.src.s_addr, sr.grp.s_addr);
 		if (c) {
 			sr.pktcnt = c->mfc_un.res.pkt;
 			sr.bytecnt = c->mfc_un.res.bytes;
 			sr.wrong_if = c->mfc_un.res.wrong_if;
-			read_unlock(&mrt_lock);
+			rcu_read_unlock();
 
 			if (copy_to_user(arg, &sr, sizeof(sr)))
 				return -EFAULT;
 			return 0;
 		}
-		read_unlock(&mrt_lock);
+		rcu_read_unlock();
 		return -EADDRNOTAVAIL;
 	default:
 		return -ENOIOCTLCMD;
@@ -1764,7 +1766,7 @@ int ip_mr_input(struct sk_buff *skb)
 		    }
 	}
 
-	read_lock(&mrt_lock);
+	/* already under rcu_read_lock() */
 	cache = ipmr_cache_find(mrt, ip_hdr(skb)->saddr, ip_hdr(skb)->daddr);
 
 	/*
@@ -1776,13 +1778,12 @@ int ip_mr_input(struct sk_buff *skb)
 		if (local) {
 			struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
 			ip_local_deliver(skb);
-			if (skb2 == NULL) {
-				read_unlock(&mrt_lock);
+			if (skb2 == NULL)
 				return -ENOBUFS;
-			}
 			skb = skb2;
 		}
 
+		read_lock(&mrt_lock);
 		vif = ipmr_find_vif(mrt, skb->dev);
 		if (vif >= 0) {
 			int err2 = ipmr_cache_unresolved(mrt, vif, skb);
@@ -1795,8 +1796,8 @@ int ip_mr_input(struct sk_buff *skb)
 		return -ENODEV;
 	}
 
+	read_lock(&mrt_lock);
 	ip_mr_forward(net, mrt, skb, cache, local);
-
 	read_unlock(&mrt_lock);
 
 	if (local)
@@ -1963,7 +1964,7 @@ int ipmr_get_route(struct net *net,
 	if (mrt == NULL)
 		return -ENOENT;
 
-	read_lock(&mrt_lock);
+	rcu_read_lock();
 	cache = ipmr_cache_find(mrt, rt->rt_src, rt->rt_dst);
 
 	if (cache == NULL) {
@@ -1973,18 +1974,21 @@ int ipmr_get_route(struct net *net,
 		int vif;
 
 		if (nowait) {
-			read_unlock(&mrt_lock);
+			rcu_read_unlock();
 			return -EAGAIN;
 		}
 
 		dev = skb->dev;
+		read_lock(&mrt_lock);
 		if (dev == NULL || (vif = ipmr_find_vif(mrt, dev)) < 0) {
 			read_unlock(&mrt_lock);
+			rcu_read_unlock();
 			return -ENODEV;
 		}
 		skb2 = skb_clone(skb, GFP_ATOMIC);
 		if (!skb2) {
 			read_unlock(&mrt_lock);
+			rcu_read_unlock();
 			return -ENOMEM;
 		}
 
@@ -1997,13 +2001,16 @@ int ipmr_get_route(struct net *net,
 		iph->version = 0;
 		err = ipmr_cache_unresolved(mrt, vif, skb2);
 		read_unlock(&mrt_lock);
+		rcu_read_unlock();
 		return err;
 	}
 
-	if (!nowait && (rtm->rtm_flags&RTM_F_NOTIFY))
+	read_lock(&mrt_lock);
+	if (!nowait && (rtm->rtm_flags & RTM_F_NOTIFY))
 		cache->mfc_flags |= MFC_NOTIFY;
 	err = __ipmr_fill_mroute(mrt, skb, cache, rtm);
 	read_unlock(&mrt_lock);
+	rcu_read_unlock();
 	return err;
 }
 
@@ -2055,14 +2062,14 @@ static int ipmr_rtm_dumproute(struct sk_buff *skb, struct netlink_callback *cb)
 	s_h = cb->args[1];
 	s_e = cb->args[2];
 
-	read_lock(&mrt_lock);
+	rcu_read_lock();
 	ipmr_for_each_table(mrt, net) {
 		if (t < s_t)
 			goto next_table;
 		if (t > s_t)
 			s_h = 0;
 		for (h = s_h; h < MFC_LINES; h++) {
-			list_for_each_entry(mfc, &mrt->mfc_cache_array[h], list) {
+			list_for_each_entry_rcu(mfc, &mrt->mfc_cache_array[h], list) {
 				if (e < s_e)
 					goto next_entry;
 				if (ipmr_fill_mroute(mrt, skb,
@@ -2080,7 +2087,7 @@ next_table:
 		t++;
 	}
 done:
-	read_unlock(&mrt_lock);
+	rcu_read_unlock();
 
 	cb->args[2] = e;
 	cb->args[1] = h;
@@ -2213,14 +2220,14 @@ static struct mfc_cache *ipmr_mfc_seq_idx(struct net *net,
 	struct mr_table *mrt = it->mrt;
 	struct mfc_cache *mfc;
 
-	read_lock(&mrt_lock);
+	rcu_read_lock();
 	for (it->ct = 0; it->ct < MFC_LINES; it->ct++) {
 		it->cache = &mrt->mfc_cache_array[it->ct];
-		list_for_each_entry(mfc, it->cache, list)
+		list_for_each_entry_rcu(mfc, it->cache, list)
 			if (pos-- == 0)
 				return mfc;
 	}
-	read_unlock(&mrt_lock);
+	rcu_read_unlock();
 
 	spin_lock_bh(&mfc_unres_lock);
 	it->cache = &mrt->mfc_unres_queue;
@@ -2279,7 +2286,7 @@ static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 	}
 
 	/* exhausted cache_array, show unresolved */
-	read_unlock(&mrt_lock);
+	rcu_read_unlock();
 	it->cache = &mrt->mfc_unres_queue;
 	it->ct = 0;
 
@@ -2302,7 +2309,7 @@ static void ipmr_mfc_seq_stop(struct seq_file *seq, void *v)
 	if (it->cache == &mrt->mfc_unres_queue)
 		spin_unlock_bh(&mfc_unres_lock);
 	else if (it->cache == &mrt->mfc_cache_array[it->ct])
-		read_unlock(&mrt_lock);
+		rcu_read_unlock();
 }
 
 static int ipmr_mfc_seq_show(struct seq_file *seq, void *v)
@@ -2426,7 +2433,7 @@ int __init ip_mr_init(void)
 
 	mrt_cachep = kmem_cache_create("ip_mrt_cache",
 				       sizeof(struct mfc_cache),
-				       0, SLAB_HWCACHE_ALIGN|SLAB_PANIC,
+				       0, SLAB_HWCACHE_ALIGN | SLAB_PANIC,
 				       NULL);
 	if (!mrt_cachep)
 		return -ENOMEM;
-- 
cgit v1.2.3


From ff7dcd44dd446db2c3e13bdedf2d52b8e0127f16 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 27 Sep 2010 12:44:25 +0000
Subject: genirq: Create irq_data

Low level chip functions need access to irq_desc->handler_data,
irq_desc->chip_data and irq_desc->msi_desc. We hand down the irq
number to the low level functions, so they need to lookup irq_desc.
With sparse irq this means a radix tree lookup.

We could hand down irq_desc itself, but low level chip functions have
no need to fiddle with it directly and we want to restrict access to
irq_desc further.

Preparatory patch for new chip functions.

Note, that the ugly anon union/struct is there to avoid a full tree
wide clean up for now. This is not going to last 3 years like __do_IRQ()

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20100927121841.645542300@linutronix.de>
Reviewed-by: H. Peter Anvin <hpa@zytor.com>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h | 90 +++++++++++++++++++++++++++++++++++++----------------
 kernel/irq/handle.c | 39 +++++++++++------------
 2 files changed, 82 insertions(+), 47 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 06273a2a17e7..363c76ff82c8 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -83,6 +83,37 @@ typedef	void (*irq_flow_handler_t)(unsigned int irq,
 struct proc_dir_entry;
 struct msi_desc;
 
+/**
+ * struct irq_data - per irq and irq chip data passed down to chip functions
+ * @irq:		interrupt number
+ * @node:		node index useful for balancing
+ * @chip:		low level interrupt hardware access
+ * @handler_data:	per-IRQ data for the irq_chip methods
+ * @chip_data:		platform-specific per-chip private data for the chip
+ *			methods, to allow shared chip implementations
+ * @msi_desc:		MSI descriptor
+ * @affinity:		IRQ affinity on SMP
+ * @irq_2_iommu:	iommu with this irq
+ *
+ * The fields here need to overlay the ones in irq_desc until we
+ * cleaned up the direct references and switched everything over to
+ * irq_data.
+ */
+struct irq_data {
+	unsigned int		irq;
+	unsigned int		node;
+	struct irq_chip		*chip;
+	void			*handler_data;
+	void			*chip_data;
+	struct msi_desc		*msi_desc;
+#ifdef CONFIG_SMP
+	cpumask_var_t		affinity;
+#endif
+#ifdef CONFIG_INTR_REMAP
+	struct irq_2_iommu      *irq_2_iommu;
+#endif
+};
+
 /**
  * struct irq_chip - hardware interrupt chip descriptor
  *
@@ -140,16 +171,10 @@ struct timer_rand_state;
 struct irq_2_iommu;
 /**
  * struct irq_desc - interrupt descriptor
- * @irq:		interrupt number for this descriptor
+ * @irq_data:		per irq and chip data passed down to chip functions
  * @timer_rand_state:	pointer to timer rand state struct
  * @kstat_irqs:		irq stats per cpu
- * @irq_2_iommu:	iommu with this irq
  * @handle_irq:		highlevel irq-events handler [if NULL, __do_IRQ()]
- * @chip:		low level interrupt hardware access
- * @msi_desc:		MSI descriptor
- * @handler_data:	per-IRQ data for the irq_chip methods
- * @chip_data:		platform-specific per-chip private data for the chip
- *			methods, to allow shared chip implementations
  * @action:		the irq action chain
  * @status:		status information
  * @depth:		disable-depth, for nested irq_disable() calls
@@ -158,8 +183,6 @@ struct irq_2_iommu;
  * @last_unhandled:	aging timer for unhandled count
  * @irqs_unhandled:	stats field for spurious unhandled interrupts
  * @lock:		locking for SMP
- * @affinity:		IRQ affinity on SMP
- * @node:		node index useful for balancing
  * @pending_mask:	pending rebalanced interrupts
  * @threads_active:	number of irqaction threads currently running
  * @wait_for_threads:	wait queue for sync_irq to wait for threaded handlers
@@ -167,17 +190,32 @@ struct irq_2_iommu;
  * @name:		flow handler name for /proc/interrupts output
  */
 struct irq_desc {
-	unsigned int		irq;
-	struct timer_rand_state *timer_rand_state;
-	unsigned int            *kstat_irqs;
+
+	/*
+	 * This union will go away, once we fixed the direct access to
+	 * irq_desc all over the place. The direct fields are a 1:1
+	 * overlay of irq_data.
+	 */
+	union {
+		struct irq_data		irq_data;
+		struct {
+			unsigned int		irq;
+			unsigned int		node;
+			struct irq_chip		*chip;
+			void			*handler_data;
+			void			*chip_data;
+			struct msi_desc		*msi_desc;
+#ifdef CONFIG_SMP
+			cpumask_var_t		affinity;
+#endif
 #ifdef CONFIG_INTR_REMAP
-	struct irq_2_iommu      *irq_2_iommu;
+			struct irq_2_iommu      *irq_2_iommu;
 #endif
+		};
+	};
+	struct timer_rand_state *timer_rand_state;
+	unsigned int            *kstat_irqs;
 	irq_flow_handler_t	handle_irq;
-	struct irq_chip		*chip;
-	struct msi_desc		*msi_desc;
-	void			*handler_data;
-	void			*chip_data;
 	struct irqaction	*action;	/* IRQ action list */
 	unsigned int		status;		/* IRQ status */
 
@@ -188,9 +226,7 @@ struct irq_desc {
 	unsigned int		irqs_unhandled;
 	raw_spinlock_t		lock;
 #ifdef CONFIG_SMP
-	cpumask_var_t		affinity;
 	const struct cpumask	*affinity_hint;
-	unsigned int		node;
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	cpumask_var_t		pending_mask;
 #endif
@@ -406,15 +442,15 @@ extern int set_irq_chip_data(unsigned int irq, void *data);
 extern int set_irq_type(unsigned int irq, unsigned int type);
 extern int set_irq_msi(unsigned int irq, struct msi_desc *entry);
 
-#define get_irq_chip(irq)	(irq_to_desc(irq)->chip)
-#define get_irq_chip_data(irq)	(irq_to_desc(irq)->chip_data)
-#define get_irq_data(irq)	(irq_to_desc(irq)->handler_data)
-#define get_irq_msi(irq)	(irq_to_desc(irq)->msi_desc)
+#define get_irq_chip(irq)	(irq_to_desc(irq)->irq_data.chip)
+#define get_irq_chip_data(irq)	(irq_to_desc(irq)->irq_data.chip_data)
+#define get_irq_data(irq)	(irq_to_desc(irq)->irq_data.handler_data)
+#define get_irq_msi(irq)	(irq_to_desc(irq)->irq_data.msi_desc)
 
-#define get_irq_desc_chip(desc)		((desc)->chip)
-#define get_irq_desc_chip_data(desc)	((desc)->chip_data)
-#define get_irq_desc_data(desc)		((desc)->handler_data)
-#define get_irq_desc_msi(desc)		((desc)->msi_desc)
+#define get_irq_desc_chip(desc)		((desc)->irq_data.chip)
+#define get_irq_desc_chip_data(desc)	((desc)->irq_data.chip_data)
+#define get_irq_desc_data(desc)		((desc)->irq_data.handler_data)
+#define get_irq_desc_msi(desc)		((desc)->irq_data.msi_desc)
 
 #endif /* CONFIG_GENERIC_HARDIRQS */
 
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 27e5c6911223..099d4fc368c3 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -75,12 +75,10 @@ EXPORT_SYMBOL_GPL(nr_irqs);
 #ifdef CONFIG_SPARSE_IRQ
 
 static struct irq_desc irq_desc_init = {
-	.irq	    = -1,
-	.status	    = IRQ_DISABLED,
-	.chip	    = &no_irq_chip,
-	.handle_irq = handle_bad_irq,
-	.depth      = 1,
-	.lock       = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
+	.status		= IRQ_DISABLED,
+	.handle_irq	= handle_bad_irq,
+	.depth		= 1,
+	.lock		= __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
 };
 
 void __ref init_kstat_irqs(struct irq_desc *desc, int node, int nr)
@@ -105,7 +103,7 @@ static void init_one_irq_desc(int irq, struct irq_desc *desc, int node)
 	memcpy(desc, &irq_desc_init, sizeof(struct irq_desc));
 
 	raw_spin_lock_init(&desc->lock);
-	desc->irq = irq;
+	desc->irq_data.irq = irq;
 #ifdef CONFIG_SMP
 	desc->node = node;
 #endif
@@ -151,12 +149,10 @@ void replace_irq_desc(unsigned int irq, struct irq_desc *desc)
 
 static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_smp = {
 	[0 ... NR_IRQS_LEGACY-1] = {
-		.irq	    = -1,
-		.status	    = IRQ_DISABLED,
-		.chip	    = &no_irq_chip,
-		.handle_irq = handle_bad_irq,
-		.depth	    = 1,
-		.lock	    = __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
+		.status		= IRQ_DISABLED,
+		.handle_irq	= handle_bad_irq,
+		.depth		= 1,
+		.lock		= __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
 	}
 };
 
@@ -183,8 +179,11 @@ int __init early_irq_init(void)
 	kstat_irqs_legacy = kzalloc_node(NR_IRQS_LEGACY * nr_cpu_ids *
 					  sizeof(int), GFP_NOWAIT, node);
 
+	irq_desc_init.irq_data.chip = &no_irq_chip;
+
 	for (i = 0; i < legacy_count; i++) {
-		desc[i].irq = i;
+		desc[i].irq_data.irq = i;
+		desc[i].irq_data.chip = &no_irq_chip;
 #ifdef CONFIG_SMP
 		desc[i].node = node;
 #endif
@@ -241,11 +240,10 @@ out_unlock:
 
 struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
 	[0 ... NR_IRQS-1] = {
-		.status = IRQ_DISABLED,
-		.chip = &no_irq_chip,
-		.handle_irq = handle_bad_irq,
-		.depth = 1,
-		.lock = __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock),
+		.status		= IRQ_DISABLED,
+		.handle_irq	= handle_bad_irq,
+		.depth		= 1,
+		.lock		= __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock),
 	}
 };
 
@@ -264,7 +262,8 @@ int __init early_irq_init(void)
 	count = ARRAY_SIZE(irq_desc);
 
 	for (i = 0; i < count; i++) {
-		desc[i].irq = i;
+		desc[i].irq_data.irq = i;
+		desc[i].irq_data.chip = &no_irq_chip;
 		alloc_desc_masks(&desc[i], 0, true);
 		init_desc_masks(&desc[i]);
 		desc[i].kstat_irqs = kstat_irqs_all[i];
-- 
cgit v1.2.3


From 6b8ff3120c758340505dddf08ad685ebb841d5d5 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 1 Oct 2010 12:58:38 +0200
Subject: genirq: Convert core code to irq_data

Convert all references in the core code to orq, chip, handler_data,
chip_data, msi_desc, affinity to irq_data.*

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h       | 10 +++---
 kernel/irq/autoprobe.c    | 14 ++++-----
 kernel/irq/chip.c         | 78 +++++++++++++++++++++++------------------------
 kernel/irq/handle.c       | 16 +++++-----
 kernel/irq/internals.h    | 12 ++++----
 kernel/irq/manage.c       | 54 ++++++++++++++++----------------
 kernel/irq/migration.c    | 10 +++---
 kernel/irq/numa_migrate.c |  8 ++---
 kernel/irq/proc.c         |  8 ++---
 kernel/irq/resend.c       |  5 +--
 kernel/irq/spurious.c     |  6 ++--
 11 files changed, 111 insertions(+), 110 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 363c76ff82c8..002351d83c3f 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -475,12 +475,12 @@ static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
 		gfp = GFP_NOWAIT;
 
 #ifdef CONFIG_CPUMASK_OFFSTACK
-	if (!alloc_cpumask_var_node(&desc->affinity, gfp, node))
+	if (!alloc_cpumask_var_node(&desc->irq_data.affinity, gfp, node))
 		return false;
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	if (!alloc_cpumask_var_node(&desc->pending_mask, gfp, node)) {
-		free_cpumask_var(desc->affinity);
+		free_cpumask_var(desc->irq_data.affinity);
 		return false;
 	}
 #endif
@@ -490,7 +490,7 @@ static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
 
 static inline void init_desc_masks(struct irq_desc *desc)
 {
-	cpumask_setall(desc->affinity);
+	cpumask_setall(desc->irq_data.affinity);
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	cpumask_clear(desc->pending_mask);
 #endif
@@ -510,7 +510,7 @@ static inline void init_copy_desc_masks(struct irq_desc *old_desc,
 					struct irq_desc *new_desc)
 {
 #ifdef CONFIG_CPUMASK_OFFSTACK
-	cpumask_copy(new_desc->affinity, old_desc->affinity);
+	cpumask_copy(new_desc->irq_data.affinity, old_desc->irq_data.affinity);
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	cpumask_copy(new_desc->pending_mask, old_desc->pending_mask);
@@ -521,7 +521,7 @@ static inline void init_copy_desc_masks(struct irq_desc *old_desc,
 static inline void free_desc_masks(struct irq_desc *old_desc,
 				   struct irq_desc *new_desc)
 {
-	free_cpumask_var(old_desc->affinity);
+	free_cpumask_var(old_desc->irq_data.affinity);
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	free_cpumask_var(old_desc->pending_mask);
diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c
index 2295a31ef110..f9bf9b228033 100644
--- a/kernel/irq/autoprobe.c
+++ b/kernel/irq/autoprobe.c
@@ -57,9 +57,9 @@ unsigned long probe_irq_on(void)
 			 * Some chips need to know about probing in
 			 * progress:
 			 */
-			if (desc->chip->set_type)
-				desc->chip->set_type(i, IRQ_TYPE_PROBE);
-			desc->chip->startup(i);
+			if (desc->irq_data.chip->set_type)
+				desc->irq_data.chip->set_type(i, IRQ_TYPE_PROBE);
+			desc->irq_data.chip->startup(i);
 		}
 		raw_spin_unlock_irq(&desc->lock);
 	}
@@ -76,7 +76,7 @@ unsigned long probe_irq_on(void)
 		raw_spin_lock_irq(&desc->lock);
 		if (!desc->action && !(desc->status & IRQ_NOPROBE)) {
 			desc->status |= IRQ_AUTODETECT | IRQ_WAITING;
-			if (desc->chip->startup(i))
+			if (desc->irq_data.chip->startup(i))
 				desc->status |= IRQ_PENDING;
 		}
 		raw_spin_unlock_irq(&desc->lock);
@@ -98,7 +98,7 @@ unsigned long probe_irq_on(void)
 			/* It triggered already - consider it spurious. */
 			if (!(status & IRQ_WAITING)) {
 				desc->status = status & ~IRQ_AUTODETECT;
-				desc->chip->shutdown(i);
+				desc->irq_data.chip->shutdown(i);
 			} else
 				if (i < 32)
 					mask |= 1 << i;
@@ -137,7 +137,7 @@ unsigned int probe_irq_mask(unsigned long val)
 				mask |= 1 << i;
 
 			desc->status = status & ~IRQ_AUTODETECT;
-			desc->chip->shutdown(i);
+			desc->irq_data.chip->shutdown(i);
 		}
 		raw_spin_unlock_irq(&desc->lock);
 	}
@@ -181,7 +181,7 @@ int probe_irq_off(unsigned long val)
 				nr_of_irqs++;
 			}
 			desc->status = status & ~IRQ_AUTODETECT;
-			desc->chip->shutdown(i);
+			desc->irq_data.chip->shutdown(i);
 		}
 		raw_spin_unlock_irq(&desc->lock);
 	}
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 4ea775cc60f0..e0e93ff10afd 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -32,18 +32,18 @@ static void dynamic_irq_init_x(unsigned int irq, bool keep_chip_data)
 	/* Ensure we don't have left over values from a previous use of this irq */
 	raw_spin_lock_irqsave(&desc->lock, flags);
 	desc->status = IRQ_DISABLED;
-	desc->chip = &no_irq_chip;
+	desc->irq_data.chip = &no_irq_chip;
 	desc->handle_irq = handle_bad_irq;
 	desc->depth = 1;
-	desc->msi_desc = NULL;
-	desc->handler_data = NULL;
+	desc->irq_data.msi_desc = NULL;
+	desc->irq_data.handler_data = NULL;
 	if (!keep_chip_data)
-		desc->chip_data = NULL;
+		desc->irq_data.chip_data = NULL;
 	desc->action = NULL;
 	desc->irq_count = 0;
 	desc->irqs_unhandled = 0;
 #ifdef CONFIG_SMP
-	cpumask_setall(desc->affinity);
+	cpumask_setall(desc->irq_data.affinity);
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	cpumask_clear(desc->pending_mask);
 #endif
@@ -64,7 +64,7 @@ void dynamic_irq_init(unsigned int irq)
  *	dynamic_irq_init_keep_chip_data - initialize a dynamically allocated irq
  *	@irq:	irq number to initialize
  *
- *	does not set irq_to_desc(irq)->chip_data to NULL
+ *	does not set irq_to_desc(irq)->irq_data.chip_data to NULL
  */
 void dynamic_irq_init_keep_chip_data(unsigned int irq)
 {
@@ -88,12 +88,12 @@ static void dynamic_irq_cleanup_x(unsigned int irq, bool keep_chip_data)
 			irq);
 		return;
 	}
-	desc->msi_desc = NULL;
-	desc->handler_data = NULL;
+	desc->irq_data.msi_desc = NULL;
+	desc->irq_data.handler_data = NULL;
 	if (!keep_chip_data)
-		desc->chip_data = NULL;
+		desc->irq_data.chip_data = NULL;
 	desc->handle_irq = handle_bad_irq;
-	desc->chip = &no_irq_chip;
+	desc->irq_data.chip = &no_irq_chip;
 	desc->name = NULL;
 	clear_kstat_irqs(desc);
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
@@ -112,7 +112,7 @@ void dynamic_irq_cleanup(unsigned int irq)
  *	dynamic_irq_cleanup_keep_chip_data - cleanup a dynamically allocated irq
  *	@irq:	irq number to initialize
  *
- *	does not set irq_to_desc(irq)->chip_data to NULL
+ *	does not set irq_to_desc(irq)->irq_data.chip_data to NULL
  */
 void dynamic_irq_cleanup_keep_chip_data(unsigned int irq)
 {
@@ -140,7 +140,7 @@ int set_irq_chip(unsigned int irq, struct irq_chip *chip)
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
 	irq_chip_set_defaults(chip);
-	desc->chip = chip;
+	desc->irq_data.chip = chip;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 
 	return 0;
@@ -193,7 +193,7 @@ int set_irq_data(unsigned int irq, void *data)
 	}
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->handler_data = data;
+	desc->irq_data.handler_data = data;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 	return 0;
 }
@@ -218,7 +218,7 @@ int set_irq_msi(unsigned int irq, struct msi_desc *entry)
 	}
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->msi_desc = entry;
+	desc->irq_data.msi_desc = entry;
 	if (entry)
 		entry->irq = irq;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
@@ -243,13 +243,13 @@ int set_irq_chip_data(unsigned int irq, void *data)
 		return -EINVAL;
 	}
 
-	if (!desc->chip) {
+	if (!desc->irq_data.chip) {
 		printk(KERN_ERR "BUG: bad set_irq_chip_data(IRQ#%d)\n", irq);
 		return -EINVAL;
 	}
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->chip_data = data;
+	desc->irq_data.chip_data = data;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 
 	return 0;
@@ -291,7 +291,7 @@ static void default_enable(unsigned int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 
-	desc->chip->unmask(irq);
+	desc->irq_data.chip->unmask(irq);
 	desc->status &= ~IRQ_MASKED;
 }
 
@@ -309,7 +309,7 @@ static unsigned int default_startup(unsigned int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 
-	desc->chip->enable(irq);
+	desc->irq_data.chip->enable(irq);
 	return 0;
 }
 
@@ -320,7 +320,7 @@ static void default_shutdown(unsigned int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 
-	desc->chip->mask(irq);
+	desc->irq_data.chip->mask(irq);
 	desc->status |= IRQ_MASKED;
 }
 
@@ -350,28 +350,28 @@ void irq_chip_set_defaults(struct irq_chip *chip)
 
 static inline void mask_ack_irq(struct irq_desc *desc, int irq)
 {
-	if (desc->chip->mask_ack)
-		desc->chip->mask_ack(irq);
+	if (desc->irq_data.chip->mask_ack)
+		desc->irq_data.chip->mask_ack(irq);
 	else {
-		desc->chip->mask(irq);
-		if (desc->chip->ack)
-			desc->chip->ack(irq);
+		desc->irq_data.chip->mask(irq);
+		if (desc->irq_data.chip->ack)
+			desc->irq_data.chip->ack(irq);
 	}
 	desc->status |= IRQ_MASKED;
 }
 
 static inline void mask_irq(struct irq_desc *desc, int irq)
 {
-	if (desc->chip->mask) {
-		desc->chip->mask(irq);
+	if (desc->irq_data.chip->mask) {
+		desc->irq_data.chip->mask(irq);
 		desc->status |= IRQ_MASKED;
 	}
 }
 
 static inline void unmask_irq(struct irq_desc *desc, int irq)
 {
-	if (desc->chip->unmask) {
-		desc->chip->unmask(irq);
+	if (desc->irq_data.chip->unmask) {
+		desc->irq_data.chip->unmask(irq);
 		desc->status &= ~IRQ_MASKED;
 	}
 }
@@ -552,7 +552,7 @@ handle_fasteoi_irq(unsigned int irq, struct irq_desc *desc)
 	raw_spin_lock(&desc->lock);
 	desc->status &= ~IRQ_INPROGRESS;
 out:
-	desc->chip->eoi(irq);
+	desc->irq_data.chip->eoi(irq);
 
 	raw_spin_unlock(&desc->lock);
 }
@@ -594,8 +594,8 @@ handle_edge_irq(unsigned int irq, struct irq_desc *desc)
 	kstat_incr_irqs_this_cpu(irq, desc);
 
 	/* Start handling the irq */
-	if (desc->chip->ack)
-		desc->chip->ack(irq);
+	if (desc->irq_data.chip->ack)
+		desc->irq_data.chip->ack(irq);
 
 	/* Mark the IRQ currently in progress.*/
 	desc->status |= IRQ_INPROGRESS;
@@ -648,15 +648,15 @@ handle_percpu_irq(unsigned int irq, struct irq_desc *desc)
 
 	kstat_incr_irqs_this_cpu(irq, desc);
 
-	if (desc->chip->ack)
-		desc->chip->ack(irq);
+	if (desc->irq_data.chip->ack)
+		desc->irq_data.chip->ack(irq);
 
 	action_ret = handle_IRQ_event(irq, desc->action);
 	if (!noirqdebug)
 		note_interrupt(irq, desc, action_ret);
 
-	if (desc->chip->eoi)
-		desc->chip->eoi(irq);
+	if (desc->irq_data.chip->eoi)
+		desc->irq_data.chip->eoi(irq);
 }
 
 void
@@ -674,7 +674,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained,
 
 	if (!handle)
 		handle = handle_bad_irq;
-	else if (desc->chip == &no_irq_chip) {
+	else if (desc->irq_data.chip == &no_irq_chip) {
 		printk(KERN_WARNING "Trying to install %sinterrupt handler "
 		       "for IRQ%d\n", is_chained ? "chained " : "", irq);
 		/*
@@ -684,7 +684,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained,
 		 * prevent us to setup the interrupt at all. Switch it to
 		 * dummy_irq_chip for easy transition.
 		 */
-		desc->chip = &dummy_irq_chip;
+		desc->irq_data.chip = &dummy_irq_chip;
 	}
 
 	chip_bus_lock(irq, desc);
@@ -692,7 +692,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained,
 
 	/* Uninstall? */
 	if (handle == handle_bad_irq) {
-		if (desc->chip != &no_irq_chip)
+		if (desc->irq_data.chip != &no_irq_chip)
 			mask_ack_irq(desc, irq);
 		desc->status |= IRQ_DISABLED;
 		desc->depth = 1;
@@ -704,7 +704,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained,
 		desc->status &= ~IRQ_DISABLED;
 		desc->status |= IRQ_NOREQUEST | IRQ_NOPROBE;
 		desc->depth = 0;
-		desc->chip->startup(irq);
+		desc->irq_data.chip->startup(irq);
 	}
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 	chip_bus_sync_unlock(irq, desc);
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 099d4fc368c3..fc27d76e83ef 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -105,7 +105,7 @@ static void init_one_irq_desc(int irq, struct irq_desc *desc, int node)
 	raw_spin_lock_init(&desc->lock);
 	desc->irq_data.irq = irq;
 #ifdef CONFIG_SMP
-	desc->node = node;
+	desc->irq_data.node = node;
 #endif
 	lockdep_set_class(&desc->lock, &irq_desc_lock_class);
 	init_kstat_irqs(desc, node, nr_cpu_ids);
@@ -185,7 +185,7 @@ int __init early_irq_init(void)
 		desc[i].irq_data.irq = i;
 		desc[i].irq_data.chip = &no_irq_chip;
 #ifdef CONFIG_SMP
-		desc[i].node = node;
+		desc[i].irq_data.node = node;
 #endif
 		desc[i].kstat_irqs = kstat_irqs_legacy + i * nr_cpu_ids;
 		lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
@@ -456,20 +456,20 @@ unsigned int __do_IRQ(unsigned int irq)
 		/*
 		 * No locking required for CPU-local interrupts:
 		 */
-		if (desc->chip->ack)
-			desc->chip->ack(irq);
+		if (desc->irq_data.chip->ack)
+			desc->irq_data.chip->ack(irq);
 		if (likely(!(desc->status & IRQ_DISABLED))) {
 			action_ret = handle_IRQ_event(irq, desc->action);
 			if (!noirqdebug)
 				note_interrupt(irq, desc, action_ret);
 		}
-		desc->chip->end(irq);
+		desc->irq_data.chip->end(irq);
 		return 1;
 	}
 
 	raw_spin_lock(&desc->lock);
-	if (desc->chip->ack)
-		desc->chip->ack(irq);
+	if (desc->irq_data.chip->ack)
+		desc->irq_data.chip->ack(irq);
 	/*
 	 * REPLAY is when Linux resends an IRQ that was dropped earlier
 	 * WAITING is used by probe to mark irqs that are being tested
@@ -529,7 +529,7 @@ out:
 	 * The ->end() handler has to deal with interrupts which got
 	 * disabled while the handler was running.
 	 */
-	desc->chip->end(irq);
+	desc->irq_data.chip->end(irq);
 	raw_spin_unlock(&desc->lock);
 
 	return 1;
diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h
index c63f3bc88f0b..a805a00cfd28 100644
--- a/kernel/irq/internals.h
+++ b/kernel/irq/internals.h
@@ -43,14 +43,14 @@ extern void irq_set_thread_affinity(struct irq_desc *desc);
 /* Inline functions for support of irq chips on slow busses */
 static inline void chip_bus_lock(unsigned int irq, struct irq_desc *desc)
 {
-	if (unlikely(desc->chip->bus_lock))
-		desc->chip->bus_lock(irq);
+	if (unlikely(desc->irq_data.chip->bus_lock))
+		desc->irq_data.chip->bus_lock(irq);
 }
 
 static inline void chip_bus_sync_unlock(unsigned int irq, struct irq_desc *desc)
 {
-	if (unlikely(desc->chip->bus_sync_unlock))
-		desc->chip->bus_sync_unlock(irq);
+	if (unlikely(desc->irq_data.chip->bus_sync_unlock))
+		desc->irq_data.chip->bus_sync_unlock(irq);
 }
 
 /*
@@ -67,8 +67,8 @@ static inline void print_irq_desc(unsigned int irq, struct irq_desc *desc)
 		irq, desc, desc->depth, desc->irq_count, desc->irqs_unhandled);
 	printk("->handle_irq():  %p, ", desc->handle_irq);
 	print_symbol("%s\n", (unsigned long)desc->handle_irq);
-	printk("->chip(): %p, ", desc->chip);
-	print_symbol("%s\n", (unsigned long)desc->chip);
+	printk("->irq_data.chip(): %p, ", desc->irq_data.chip);
+	print_symbol("%s\n", (unsigned long)desc->irq_data.chip);
 	printk("->action(): %p\n", desc->action);
 	if (desc->action) {
 		printk("->action->handler(): %p, ", desc->action->handler);
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index c3003e9d91a3..4dfb19521d9f 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -73,8 +73,8 @@ int irq_can_set_affinity(unsigned int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 
-	if (CHECK_IRQ_PER_CPU(desc->status) || !desc->chip ||
-	    !desc->chip->set_affinity)
+	if (CHECK_IRQ_PER_CPU(desc->status) || !desc->irq_data.chip ||
+	    !desc->irq_data.chip->set_affinity)
 		return 0;
 
 	return 1;
@@ -111,15 +111,15 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 	struct irq_desc *desc = irq_to_desc(irq);
 	unsigned long flags;
 
-	if (!desc->chip->set_affinity)
+	if (!desc->irq_data.chip->set_affinity)
 		return -EINVAL;
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	if (desc->status & IRQ_MOVE_PCNTXT) {
-		if (!desc->chip->set_affinity(irq, cpumask)) {
-			cpumask_copy(desc->affinity, cpumask);
+		if (!desc->irq_data.chip->set_affinity(irq, cpumask)) {
+			cpumask_copy(desc->irq_data.affinity, cpumask);
 			irq_set_thread_affinity(desc);
 		}
 	}
@@ -128,8 +128,8 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 		cpumask_copy(desc->pending_mask, cpumask);
 	}
 #else
-	if (!desc->chip->set_affinity(irq, cpumask)) {
-		cpumask_copy(desc->affinity, cpumask);
+	if (!desc->irq_data.chip->set_affinity(irq, cpumask)) {
+		cpumask_copy(desc->irq_data.affinity, cpumask);
 		irq_set_thread_affinity(desc);
 	}
 #endif
@@ -168,16 +168,16 @@ static int setup_affinity(unsigned int irq, struct irq_desc *desc)
 	 * one of the targets is online.
 	 */
 	if (desc->status & (IRQ_AFFINITY_SET | IRQ_NO_BALANCING)) {
-		if (cpumask_any_and(desc->affinity, cpu_online_mask)
+		if (cpumask_any_and(desc->irq_data.affinity, cpu_online_mask)
 		    < nr_cpu_ids)
 			goto set_affinity;
 		else
 			desc->status &= ~IRQ_AFFINITY_SET;
 	}
 
-	cpumask_and(desc->affinity, cpu_online_mask, irq_default_affinity);
+	cpumask_and(desc->irq_data.affinity, cpu_online_mask, irq_default_affinity);
 set_affinity:
-	desc->chip->set_affinity(irq, desc->affinity);
+	desc->irq_data.chip->set_affinity(irq, desc->irq_data.affinity);
 
 	return 0;
 }
@@ -223,7 +223,7 @@ void __disable_irq(struct irq_desc *desc, unsigned int irq, bool suspend)
 
 	if (!desc->depth++) {
 		desc->status |= IRQ_DISABLED;
-		desc->chip->disable(irq);
+		desc->irq_data.chip->disable(irq);
 	}
 }
 
@@ -313,7 +313,7 @@ void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume)
  *	IRQ line is re-enabled.
  *
  *	This function may be called from IRQ context only when
- *	desc->chip->bus_lock and desc->chip->bus_sync_unlock are NULL !
+ *	desc->irq_data.chip->bus_lock and desc->chip->bus_sync_unlock are NULL !
  */
 void enable_irq(unsigned int irq)
 {
@@ -336,8 +336,8 @@ static int set_irq_wake_real(unsigned int irq, unsigned int on)
 	struct irq_desc *desc = irq_to_desc(irq);
 	int ret = -ENXIO;
 
-	if (desc->chip->set_wake)
-		ret = desc->chip->set_wake(irq, on);
+	if (desc->irq_data.chip->set_wake)
+		ret = desc->irq_data.chip->set_wake(irq, on);
 
 	return ret;
 }
@@ -432,7 +432,7 @@ int __irq_set_trigger(struct irq_desc *desc, unsigned int irq,
 		unsigned long flags)
 {
 	int ret;
-	struct irq_chip *chip = desc->chip;
+	struct irq_chip *chip = desc->irq_data.chip;
 
 	if (!chip || !chip->set_type) {
 		/*
@@ -457,8 +457,8 @@ int __irq_set_trigger(struct irq_desc *desc, unsigned int irq,
 		desc->status &= ~(IRQ_LEVEL | IRQ_TYPE_SENSE_MASK);
 		desc->status |= flags;
 
-		if (chip != desc->chip)
-			irq_chip_set_defaults(desc->chip);
+		if (chip != desc->irq_data.chip)
+			irq_chip_set_defaults(desc->irq_data.chip);
 	}
 
 	return ret;
@@ -528,7 +528,7 @@ again:
 
 	if (!(desc->status & IRQ_DISABLED) && (desc->status & IRQ_MASKED)) {
 		desc->status &= ~IRQ_MASKED;
-		desc->chip->unmask(irq);
+		desc->irq_data.chip->unmask(irq);
 	}
 	raw_spin_unlock_irq(&desc->lock);
 	chip_bus_sync_unlock(irq, desc);
@@ -556,7 +556,7 @@ irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action)
 	}
 
 	raw_spin_lock_irq(&desc->lock);
-	cpumask_copy(mask, desc->affinity);
+	cpumask_copy(mask, desc->irq_data.affinity);
 	raw_spin_unlock_irq(&desc->lock);
 
 	set_cpus_allowed_ptr(current, mask);
@@ -657,7 +657,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new)
 	if (!desc)
 		return -EINVAL;
 
-	if (desc->chip == &no_irq_chip)
+	if (desc->irq_data.chip == &no_irq_chip)
 		return -ENOSYS;
 	/*
 	 * Some drivers like serial.c use request_irq() heavily,
@@ -752,7 +752,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new)
 	}
 
 	if (!shared) {
-		irq_chip_set_defaults(desc->chip);
+		irq_chip_set_defaults(desc->irq_data.chip);
 
 		init_waitqueue_head(&desc->wait_for_threads);
 
@@ -779,7 +779,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new)
 		if (!(desc->status & IRQ_NOAUTOEN)) {
 			desc->depth = 0;
 			desc->status &= ~IRQ_DISABLED;
-			desc->chip->startup(irq);
+			desc->irq_data.chip->startup(irq);
 		} else
 			/* Undo nested disables: */
 			desc->depth = 1;
@@ -912,17 +912,17 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id)
 
 	/* Currently used only by UML, might disappear one day: */
 #ifdef CONFIG_IRQ_RELEASE_METHOD
-	if (desc->chip->release)
-		desc->chip->release(irq, dev_id);
+	if (desc->irq_data.chip->release)
+		desc->irq_data.chip->release(irq, dev_id);
 #endif
 
 	/* If this was the last handler, shut down the IRQ line: */
 	if (!desc->action) {
 		desc->status |= IRQ_DISABLED;
-		if (desc->chip->shutdown)
-			desc->chip->shutdown(irq);
+		if (desc->irq_data.chip->shutdown)
+			desc->irq_data.chip->shutdown(irq);
 		else
-			desc->chip->disable(irq);
+			desc->irq_data.chip->disable(irq);
 	}
 
 #ifdef CONFIG_SMP
diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c
index 241962280836..f923c37e651a 100644
--- a/kernel/irq/migration.c
+++ b/kernel/irq/migration.c
@@ -24,7 +24,7 @@ void move_masked_irq(int irq)
 	if (unlikely(cpumask_empty(desc->pending_mask)))
 		return;
 
-	if (!desc->chip->set_affinity)
+	if (!desc->irq_data.chip->set_affinity)
 		return;
 
 	assert_raw_spin_locked(&desc->lock);
@@ -43,8 +43,8 @@ void move_masked_irq(int irq)
 	 */
 	if (likely(cpumask_any_and(desc->pending_mask, cpu_online_mask)
 		   < nr_cpu_ids))
-		if (!desc->chip->set_affinity(irq, desc->pending_mask)) {
-			cpumask_copy(desc->affinity, desc->pending_mask);
+		if (!desc->irq_data.chip->set_affinity(irq, desc->pending_mask)) {
+			cpumask_copy(desc->irq_data.affinity, desc->pending_mask);
 			irq_set_thread_affinity(desc);
 		}
 
@@ -61,8 +61,8 @@ void move_native_irq(int irq)
 	if (unlikely(desc->status & IRQ_DISABLED))
 		return;
 
-	desc->chip->mask(irq);
+	desc->irq_data.chip->mask(irq);
 	move_masked_irq(irq);
-	desc->chip->unmask(irq);
+	desc->irq_data.chip->unmask(irq);
 }
 
diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c
index 65d3845665ac..e7f1f16402c1 100644
--- a/kernel/irq/numa_migrate.c
+++ b/kernel/irq/numa_migrate.c
@@ -44,7 +44,7 @@ static bool init_copy_one_irq_desc(int irq, struct irq_desc *old_desc,
 		return false;
 	}
 	raw_spin_lock_init(&desc->lock);
-	desc->node = node;
+	desc->irq_data.node = node;
 	lockdep_set_class(&desc->lock, &irq_desc_lock_class);
 	init_copy_kstat_irqs(old_desc, desc, node, nr_cpu_ids);
 	init_copy_desc_masks(old_desc, desc);
@@ -66,7 +66,7 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc,
 	unsigned int irq;
 	unsigned long flags;
 
-	irq = old_desc->irq;
+	irq = old_desc->irq_data.irq;
 
 	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
 
@@ -109,10 +109,10 @@ out_unlock:
 struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
 {
 	/* those static or target node is -1, do not move them */
-	if (desc->irq < NR_IRQS_LEGACY || node == -1)
+	if (desc->irq_data.irq < NR_IRQS_LEGACY || node == -1)
 		return desc;
 
-	if (desc->node != node)
+	if (desc->irq_data.node != node)
 		desc = __real_move_irq_desc(desc, node);
 
 	return desc;
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index 09a2ee540bd2..9b0da94b5b2b 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -21,7 +21,7 @@ static struct proc_dir_entry *root_irq_dir;
 static int irq_affinity_proc_show(struct seq_file *m, void *v)
 {
 	struct irq_desc *desc = irq_to_desc((long)m->private);
-	const struct cpumask *mask = desc->affinity;
+	const struct cpumask *mask = desc->irq_data.affinity;
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	if (desc->status & IRQ_MOVE_PENDING)
@@ -65,7 +65,7 @@ static ssize_t irq_affinity_proc_write(struct file *file,
 	cpumask_var_t new_value;
 	int err;
 
-	if (!irq_to_desc(irq)->chip->set_affinity || no_irq_affinity ||
+	if (!irq_to_desc(irq)->irq_data.chip->set_affinity || no_irq_affinity ||
 	    irq_balancing_disabled(irq))
 		return -EIO;
 
@@ -185,7 +185,7 @@ static int irq_node_proc_show(struct seq_file *m, void *v)
 {
 	struct irq_desc *desc = irq_to_desc((long) m->private);
 
-	seq_printf(m, "%d\n", desc->node);
+	seq_printf(m, "%d\n", desc->irq_data.node);
 	return 0;
 }
 
@@ -269,7 +269,7 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc)
 {
 	char name [MAX_NAMELEN];
 
-	if (!root_irq_dir || (desc->chip == &no_irq_chip) || desc->dir)
+	if (!root_irq_dir || (desc->irq_data.chip == &no_irq_chip) || desc->dir)
 		return;
 
 	memset(name, 0, MAX_NAMELEN);
diff --git a/kernel/irq/resend.c b/kernel/irq/resend.c
index 090c3763f3a2..47c56a097928 100644
--- a/kernel/irq/resend.c
+++ b/kernel/irq/resend.c
@@ -60,7 +60,7 @@ void check_irq_resend(struct irq_desc *desc, unsigned int irq)
 	/*
 	 * Make sure the interrupt is enabled, before resending it:
 	 */
-	desc->chip->enable(irq);
+	desc->irq_data.chip->enable(irq);
 
 	/*
 	 * We do not resend level type interrupts. Level type
@@ -70,7 +70,8 @@ void check_irq_resend(struct irq_desc *desc, unsigned int irq)
 	if ((status & (IRQ_LEVEL | IRQ_PENDING | IRQ_REPLAY)) == IRQ_PENDING) {
 		desc->status = (status & ~IRQ_PENDING) | IRQ_REPLAY;
 
-		if (!desc->chip->retrigger || !desc->chip->retrigger(irq)) {
+		if (!desc->irq_data.chip->retrigger ||
+		    !desc->irq_data.chip->retrigger(irq)) {
 #ifdef CONFIG_HARDIRQS_SW_RESEND
 			/* Set it pending and activate the softirq: */
 			set_bit(irq, irqs_resend);
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index 89fb90ae534f..36c2c9289e2b 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -78,8 +78,8 @@ static int try_one_irq(int irq, struct irq_desc *desc)
 	 * If we did actual work for the real IRQ line we must let the
 	 * IRQ controller clean up too
 	 */
-	if (work && desc->chip && desc->chip->end)
-		desc->chip->end(irq);
+	if (work && desc->irq_data.chip && desc->irq_data.chip->end)
+		desc->irq_data.chip->end(irq);
 	raw_spin_unlock(&desc->lock);
 
 	return ok;
@@ -254,7 +254,7 @@ void note_interrupt(unsigned int irq, struct irq_desc *desc,
 		printk(KERN_EMERG "Disabling IRQ #%d\n", irq);
 		desc->status |= IRQ_DISABLED | IRQ_SPURIOUS_DISABLED;
 		desc->depth++;
-		desc->chip->disable(irq);
+		desc->irq_data.chip->disable(irq);
 
 		mod_timer(&poll_spurious_irq_timer,
 			  jiffies + POLL_SPURIOUS_IRQ_INTERVAL);
-- 
cgit v1.2.3


From f8822657e799b02c55556c99a601261e207a299d Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 27 Sep 2010 12:44:32 +0000
Subject: genirq: Provide advanced irq chip functions

The low level irq chip functions want access to irq_desc->irq_data.
Provide new functions which hand down irq_data instead of the irq
number so these functions avoid to call irq_to_desc() which is a radix
tree lookup in case of sparse irq.

This provides all the old functions except one: end(). end() is a
relict of __do_IRQ() and will just go away with the __do_IRQ() code.

The replacement for set_affinity() has an extra argument "bool
force". The reason for this is to notify the low level code, that the
move has to be done right away and cannot be delayed until the next
interrupt happens. That's necessary to handle the irq fixup on cpu
unplug in the generic code.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20100927121841.742126604@linutronix.de>
Reviewed-by: H. Peter Anvin <hpa@zytor.com>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h | 66 ++++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 50 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 002351d83c3f..0c83cbd2df4e 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -118,23 +118,38 @@ struct irq_data {
  * struct irq_chip - hardware interrupt chip descriptor
  *
  * @name:		name for /proc/interrupts
- * @startup:		start up the interrupt (defaults to ->enable if NULL)
- * @shutdown:		shut down the interrupt (defaults to ->disable if NULL)
- * @enable:		enable the interrupt (defaults to chip->unmask if NULL)
- * @disable:		disable the interrupt
- * @ack:		start of a new interrupt
- * @mask:		mask an interrupt source
- * @mask_ack:		ack and mask an interrupt source
- * @unmask:		unmask an interrupt source
- * @eoi:		end of interrupt - chip level
- * @end:		end of interrupt - flow level
- * @set_affinity:	set the CPU affinity on SMP machines
- * @retrigger:		resend an IRQ to the CPU
- * @set_type:		set the flow type (IRQ_TYPE_LEVEL/etc.) of an IRQ
- * @set_wake:		enable/disable power-management wake-on of an IRQ
+ * @startup:		deprecated, replaced by irq_startup
+ * @shutdown:		deprecated, replaced by irq_shutdown
+ * @enable:		deprecated, replaced by irq_enable
+ * @disable:		deprecated, replaced by irq_disable
+ * @ack:		deprecated, replaced by irq_ack
+ * @mask:		deprecated, replaced by irq_mask
+ * @mask_ack:		deprecated, replaced by irq_mask_ack
+ * @unmask:		deprecated, replaced by irq_unmask
+ * @eoi:		deprecated, replaced by irq_eoi
+ * @end:		deprecated, will go away with __do_IRQ()
+ * @set_affinity:	deprecated, replaced by irq_set_affinity
+ * @retrigger:		deprecated, replaced by irq_retrigger
+ * @set_type:		deprecated, replaced by irq_set_type
+ * @set_wake:		deprecated, replaced by irq_wake
+ * @bus_lock:		deprecated, replaced by irq_bus_lock
+ * @bus_sync_unlock:	deprecated, replaced by irq_bus_sync_unlock
  *
- * @bus_lock:		function to lock access to slow bus (i2c) chips
- * @bus_sync_unlock:	function to sync and unlock slow bus (i2c) chips
+ * @irq_startup:	start up the interrupt (defaults to ->enable if NULL)
+ * @irq_shutdown:	shut down the interrupt (defaults to ->disable if NULL)
+ * @irq_enable:		enable the interrupt (defaults to chip->unmask if NULL)
+ * @irq_disable:	disable the interrupt
+ * @irq_ack:		start of a new interrupt
+ * @irq_mask:		mask an interrupt source
+ * @irq_mask_ack:	ack and mask an interrupt source
+ * @irq_unmask:		unmask an interrupt source
+ * @irq_eoi:		end of interrupt
+ * @irq_set_affinity:	set the CPU affinity on SMP machines
+ * @irq_retrigger:	resend an IRQ to the CPU
+ * @irq_set_type:	set the flow type (IRQ_TYPE_LEVEL/etc.) of an IRQ
+ * @irq_set_wake:	enable/disable power-management wake-on of an IRQ
+ * @irq_bus_lock:	function to lock access to slow bus (i2c) chips
+ * @irq_bus_sync_unlock:function to sync and unlock slow bus (i2c) chips
  *
  * @release:		release function solely used by UML
  */
@@ -161,6 +176,25 @@ struct irq_chip {
 	void		(*bus_lock)(unsigned int irq);
 	void		(*bus_sync_unlock)(unsigned int irq);
 
+	unsigned int	(*irq_startup)(struct irq_data *data);
+	void		(*irq_shutdown)(struct irq_data *data);
+	void		(*irq_enable)(struct irq_data *data);
+	void		(*irq_disable)(struct irq_data *data);
+
+	void		(*irq_ack)(struct irq_data *data);
+	void		(*irq_mask)(struct irq_data *data);
+	void		(*irq_mask_ack)(struct irq_data *data);
+	void		(*irq_unmask)(struct irq_data *data);
+	void		(*irq_eoi)(struct irq_data *data);
+
+	int		(*irq_set_affinity)(struct irq_data *data, const struct cpumask *dest, bool force);
+	int		(*irq_retrigger)(struct irq_data *data);
+	int		(*irq_set_type)(struct irq_data *data, unsigned int flow_type);
+	int		(*irq_set_wake)(struct irq_data *data, unsigned int on);
+
+	void		(*irq_bus_lock)(struct irq_data *data);
+	void		(*irq_bus_sync_unlock)(struct irq_data *data);
+
 	/* Currently used only by UML, might disappear one day.*/
 #ifdef CONFIG_IRQ_RELEASE_METHOD
 	void		(*release)(unsigned int irq, void *dev_id);
-- 
cgit v1.2.3


From bd151412263a67b5321e9dd1d5b4bf6d96fdebf3 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 1 Oct 2010 15:17:14 +0200
Subject: genirq: Provide config option to disable deprecated code

This option covers now the old chip functions and the irq_desc data
fields which are moving to struct irq_data. More stuff will follow.

Pretty handy for testing a conversion, whether something broke or not.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h    |  8 +++++++-
 kernel/irq/Kconfig     |  4 ++++
 kernel/irq/chip.c      |  8 +++++++-
 kernel/irq/handle.c    |  9 +++++++--
 kernel/irq/internals.h | 10 ++++++++++
 kernel/irq/spurious.c  |  6 ++++--
 6 files changed, 39 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 0c83cbd2df4e..82ed8231394a 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -155,6 +155,7 @@ struct irq_data {
  */
 struct irq_chip {
 	const char	*name;
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
 	unsigned int	(*startup)(unsigned int irq);
 	void		(*shutdown)(unsigned int irq);
 	void		(*enable)(unsigned int irq);
@@ -175,7 +176,7 @@ struct irq_chip {
 
 	void		(*bus_lock)(unsigned int irq);
 	void		(*bus_sync_unlock)(unsigned int irq);
-
+#endif
 	unsigned int	(*irq_startup)(struct irq_data *data);
 	void		(*irq_shutdown)(struct irq_data *data);
 	void		(*irq_enable)(struct irq_data *data);
@@ -225,6 +226,9 @@ struct irq_2_iommu;
  */
 struct irq_desc {
 
+#ifdef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
+	struct irq_data		irq_data;
+#else
 	/*
 	 * This union will go away, once we fixed the direct access to
 	 * irq_desc all over the place. The direct fields are a 1:1
@@ -247,6 +251,8 @@ struct irq_desc {
 #endif
 		};
 	};
+#endif
+
 	struct timer_rand_state *timer_rand_state;
 	unsigned int            *kstat_irqs;
 	irq_flow_handler_t	handle_irq;
diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig
index e0fc6cd78aa0..a42c0191d71a 100644
--- a/kernel/irq/Kconfig
+++ b/kernel/irq/Kconfig
@@ -12,6 +12,10 @@ config GENERIC_HARDIRQS
 config GENERIC_HARDIRQS_NO__DO_IRQ
        def_bool y
 
+# Select this to disable the deprecated stuff
+config GENERIC_HARDIRQS_NO_DEPRECATED
+       def_bool n
+
 # Options selectable by the architecture code
 config HAVE_SPARSE_IRQ
        def_bool n
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index f2c4d28c508a..323547983f15 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -324,6 +324,7 @@ static void default_shutdown(struct irq_data *data)
 	desc->status |= IRQ_MASKED;
 }
 
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
 /* Temporary migration helpers */
 static void compat_irq_mask(struct irq_data *data)
 {
@@ -400,12 +401,14 @@ static void compat_bus_sync_unlock(struct irq_data *data)
 {
 	data->chip->bus_sync_unlock(data->irq);
 }
+#endif
 
 /*
  * Fixup enable/disable function pointers
  */
 void irq_chip_set_defaults(struct irq_chip *chip)
 {
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
 	/*
 	 * Compat fixup functions need to be before we set the
 	 * defaults for enable/disable/startup/shutdown
@@ -418,7 +421,7 @@ void irq_chip_set_defaults(struct irq_chip *chip)
 		chip->irq_shutdown = compat_irq_shutdown;
 	if (chip->startup)
 		chip->irq_startup = compat_irq_startup;
-
+#endif
 	/*
 	 * The real defaults
 	 */
@@ -437,6 +440,8 @@ void irq_chip_set_defaults(struct irq_chip *chip)
 	if (!chip->irq_shutdown)
 		chip->irq_shutdown = chip->irq_disable != default_disable ?
 			chip->irq_disable : default_shutdown;
+
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
 	if (!chip->end)
 		chip->end = dummy_irq_chip.end;
 
@@ -465,6 +470,7 @@ void irq_chip_set_defaults(struct irq_chip *chip)
 		chip->irq_set_wake = compat_irq_set_wake;
 	if (chip->retrigger)
 		chip->irq_retrigger = compat_irq_retrigger;
+#endif
 }
 
 static inline void mask_ack_irq(struct irq_desc *desc)
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 8d0697f892a2..3fcef37154a1 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -309,7 +309,12 @@ static unsigned int noop_ret(struct irq_data *data)
 	return 0;
 }
 
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
 static void compat_noop(unsigned int irq) { }
+#define END_INIT .end = compat_noop
+#else
+#define END_INIT
+#endif
 
 /*
  * Generic no controller implementation
@@ -321,7 +326,7 @@ struct irq_chip no_irq_chip = {
 	.irq_enable	= noop,
 	.irq_disable	= noop,
 	.irq_ack	= ack_bad,
-	.end		= compat_noop,
+	END_INIT
 };
 
 /*
@@ -337,7 +342,7 @@ struct irq_chip dummy_irq_chip = {
 	.irq_ack	= noop,
 	.irq_mask	= noop,
 	.irq_unmask	= noop,
-	.end		= compat_noop,
+	END_INIT
 };
 
 /*
diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h
index ecafbfee5b12..b905f0ab1bb2 100644
--- a/kernel/irq/internals.h
+++ b/kernel/irq/internals.h
@@ -42,6 +42,16 @@ extern int irq_select_affinity_usr(unsigned int irq);
 
 extern void irq_set_thread_affinity(struct irq_desc *desc);
 
+#ifndef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
+static inline void irq_end(unsigned int irq, struct irq_desc *desc)
+{
+	if (desc->irq_data.chip && desc->irq_data.chip->end)
+		desc->irq_data.chip->end(irq);
+}
+#else
+static inline void irq_end(unsigned int irq, struct irq_desc *desc) { }
+#endif
+
 /* Inline functions for support of irq chips on slow busses */
 static inline void chip_bus_lock(struct irq_desc *desc)
 {
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index 9ee704d3a23c..3089d3b9d5f3 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -14,6 +14,8 @@
 #include <linux/moduleparam.h>
 #include <linux/timer.h>
 
+#include "internals.h"
+
 static int irqfixup __read_mostly;
 
 #define POLL_SPURIOUS_IRQ_INTERVAL (HZ/10)
@@ -78,8 +80,8 @@ static int try_one_irq(int irq, struct irq_desc *desc)
 	 * If we did actual work for the real IRQ line we must let the
 	 * IRQ controller clean up too
 	 */
-	if (work && desc->irq_data.chip && desc->irq_data.chip->end)
-		desc->irq_data.chip->end(irq);
+	if (work)
+		irq_end(irq, desc);
 	raw_spin_unlock(&desc->lock);
 
 	return ok;
-- 
cgit v1.2.3


From 001985b2c0cfad48e1dec8e30f4d432eac240dd2 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@verge.net.au>
Date: Sun, 22 Aug 2010 21:37:51 +0900
Subject: netfilter: nf_conntrack_sip: Add callid parser

Signed-off-by: Simon Horman <horms@verge.net.au>
Acked-by: Julian Anastasov <ja@ssi.bg>
---
 include/linux/netfilter/nf_conntrack_sip.h |  1 +
 net/netfilter/nf_conntrack_sip.c           | 39 ++++++++++++++++++++++++++++++
 2 files changed, 40 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h
index ff8cfbcf3b81..0ce91d56a5f2 100644
--- a/include/linux/netfilter/nf_conntrack_sip.h
+++ b/include/linux/netfilter/nf_conntrack_sip.h
@@ -89,6 +89,7 @@ enum sip_header_types {
 	SIP_HDR_VIA_TCP,
 	SIP_HDR_EXPIRES,
 	SIP_HDR_CONTENT_LENGTH,
+	SIP_HDR_CALL_ID,
 };
 
 enum sdp_header_types {
diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c
index 2fd1ea2c1bb3..715ce54d2fc4 100644
--- a/net/netfilter/nf_conntrack_sip.c
+++ b/net/netfilter/nf_conntrack_sip.c
@@ -130,6 +130,44 @@ static int digits_len(const struct nf_conn *ct, const char *dptr,
 	return len;
 }
 
+static int iswordc(const char c)
+{
+	if (isalnum(c) || c == '!' || c == '"' || c == '%' ||
+	    (c >= '(' && c <= '/') || c == ':' || c == '<' || c == '>' ||
+	    c == '?' || (c >= '[' && c <= ']') || c == '_' || c == '`' ||
+	    c == '{' || c == '}' || c == '~')
+		return 1;
+	return 0;
+}
+
+static int word_len(const char *dptr, const char *limit)
+{
+	int len = 0;
+	while (dptr < limit && iswordc(*dptr)) {
+		dptr++;
+		len++;
+	}
+	return len;
+}
+
+static int callid_len(const struct nf_conn *ct, const char *dptr,
+		      const char *limit, int *shift)
+{
+	int len, domain_len;
+
+	len = word_len(dptr, limit);
+	dptr += len;
+	if (!len || dptr == limit || *dptr != '@')
+		return len;
+	dptr++;
+	len++;
+
+	domain_len = word_len(dptr, limit);
+	if (!domain_len)
+		return 0;
+	return len + domain_len;
+}
+
 /* get media type + port length */
 static int media_len(const struct nf_conn *ct, const char *dptr,
 		     const char *limit, int *shift)
@@ -299,6 +337,7 @@ static const struct sip_header ct_sip_hdrs[] = {
 	[SIP_HDR_VIA_TCP]		= SIP_HDR("Via", "v", "TCP ", epaddr_len),
 	[SIP_HDR_EXPIRES]		= SIP_HDR("Expires", NULL, NULL, digits_len),
 	[SIP_HDR_CONTENT_LENGTH]	= SIP_HDR("Content-Length", "l", NULL, digits_len),
+	[SIP_HDR_CALL_ID]		= SIP_HDR("Call-Id", "i", NULL, callid_len),
 };
 
 static const char *sip_follow_continuation(const char *dptr, const char *limit)
-- 
cgit v1.2.3


From 85999283a21ab2dd37427fdd8c8e8af57223977c Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@verge.net.au>
Date: Sun, 22 Aug 2010 21:37:53 +0900
Subject: IPVS: Add struct ip_vs_pe

Signed-off-by: Simon Horman <horms@verge.net.au>
Acked-by: Julian Anastasov <ja@ssi.bg>
---
 include/linux/ip_vs.h           |  2 ++
 include/net/ip_vs.h             | 28 ++++++++++++++++-
 net/netfilter/ipvs/ip_vs_conn.c | 67 ++++++++++++++++++++++++++++++++++-------
 net/netfilter/ipvs/ip_vs_core.c | 36 +++++++++++++++++-----
 net/netfilter/ipvs/ip_vs_sync.c | 17 +++++++++--
 5 files changed, 129 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ip_vs.h b/include/linux/ip_vs.h
index df7728613720..0a9c44d64292 100644
--- a/include/linux/ip_vs.h
+++ b/include/linux/ip_vs.h
@@ -99,8 +99,10 @@
 				0)
 
 #define IP_VS_SCHEDNAME_MAXLEN	16
+#define IP_VS_PENAME_MAXLEN	16
 #define IP_VS_IFNAME_MAXLEN	16
 
+#define IP_VS_PEDATA_MAXLEN     255
 
 /*
  *	The struct ip_vs_service_user and struct ip_vs_dest_user are
diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index d4da774eca75..b6b309d05e4e 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -364,6 +364,10 @@ struct ip_vs_conn_param {
 	__be16				vport;
 	__u16				protocol;
 	u16				af;
+
+	const struct ip_vs_pe		*pe;
+	char				*pe_data;
+	__u8				pe_data_len;
 };
 
 /*
@@ -416,6 +420,9 @@ struct ip_vs_conn {
 	void                    *app_data;      /* Application private data */
 	struct ip_vs_seq        in_seq;         /* incoming seq. struct */
 	struct ip_vs_seq        out_seq;        /* outgoing seq. struct */
+
+	char			*pe_data;
+	__u8			pe_data_len;
 };
 
 
@@ -486,6 +493,9 @@ struct ip_vs_service {
 	struct ip_vs_scheduler	*scheduler;    /* bound scheduler object */
 	rwlock_t		sched_lock;    /* lock sched_data */
 	void			*sched_data;   /* scheduler application data */
+
+	/* alternate persistence engine */
+	struct ip_vs_pe		*pe;
 };
 
 
@@ -549,6 +559,20 @@ struct ip_vs_scheduler {
 				       const struct sk_buff *skb);
 };
 
+/* The persistence engine object */
+struct ip_vs_pe {
+	struct list_head	n_list;		/* d-linked list head */
+	char			*name;		/* scheduler name */
+	atomic_t		refcnt;		/* reference counter */
+	struct module		*module;	/* THIS_MODULE/NULL */
+
+	/* get the connection template, if any */
+	int (*fill_param)(struct ip_vs_conn_param *p, struct sk_buff *skb);
+	bool (*ct_match)(const struct ip_vs_conn_param *p,
+			 struct ip_vs_conn *ct);
+	u32 (*hashkey_raw)(const struct ip_vs_conn_param *p, u32 initval,
+			   bool inverse);
+};
 
 /*
  *	The application module object (a.k.a. app incarnation)
@@ -648,6 +672,8 @@ static inline void ip_vs_conn_fill_param(int af, int protocol,
 	p->cport = cport;
 	p->vaddr = vaddr;
 	p->vport = vport;
+	p->pe = NULL;
+	p->pe_data = NULL;
 }
 
 struct ip_vs_conn *ip_vs_conn_in_get(const struct ip_vs_conn_param *p);
@@ -803,7 +829,7 @@ extern int ip_vs_unbind_scheduler(struct ip_vs_service *svc);
 extern struct ip_vs_scheduler *ip_vs_scheduler_get(const char *sched_name);
 extern void ip_vs_scheduler_put(struct ip_vs_scheduler *scheduler);
 extern struct ip_vs_conn *
-ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb);
+ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb);
 extern int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
 			struct ip_vs_protocol *pp);
 
diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c
index deeb906a797b..06da21e97405 100644
--- a/net/netfilter/ipvs/ip_vs_conn.c
+++ b/net/netfilter/ipvs/ip_vs_conn.c
@@ -148,6 +148,42 @@ static unsigned int ip_vs_conn_hashkey(int af, unsigned proto,
 		& ip_vs_conn_tab_mask;
 }
 
+static unsigned int ip_vs_conn_hashkey_param(const struct ip_vs_conn_param *p,
+					     bool inverse)
+{
+	const union nf_inet_addr *addr;
+	__be16 port;
+
+	if (p->pe && p->pe->hashkey_raw)
+		return p->pe->hashkey_raw(p, ip_vs_conn_rnd, inverse) &
+			ip_vs_conn_tab_mask;
+
+	if (likely(!inverse)) {
+		addr = p->caddr;
+		port = p->cport;
+	} else {
+		addr = p->vaddr;
+		port = p->vport;
+	}
+
+	return ip_vs_conn_hashkey(p->af, p->protocol, addr, port);
+}
+
+static unsigned int ip_vs_conn_hashkey_conn(const struct ip_vs_conn *cp)
+{
+	struct ip_vs_conn_param p;
+
+	ip_vs_conn_fill_param(cp->af, cp->protocol, &cp->caddr, cp->cport,
+			      NULL, 0, &p);
+
+	if (cp->dest && cp->dest->svc->pe) {
+		p.pe = cp->dest->svc->pe;
+		p.pe_data = cp->pe_data;
+		p.pe_data_len = cp->pe_data_len;
+	}
+
+	return ip_vs_conn_hashkey_param(&p, false);
+}
 
 /*
  *	Hashes ip_vs_conn in ip_vs_conn_tab by proto,addr,port.
@@ -162,7 +198,7 @@ static inline int ip_vs_conn_hash(struct ip_vs_conn *cp)
 		return 0;
 
 	/* Hash by protocol, client address and port */
-	hash = ip_vs_conn_hashkey(cp->af, cp->protocol, &cp->caddr, cp->cport);
+	hash = ip_vs_conn_hashkey_conn(cp);
 
 	ct_write_lock(hash);
 	spin_lock(&cp->lock);
@@ -195,7 +231,7 @@ static inline int ip_vs_conn_unhash(struct ip_vs_conn *cp)
 	int ret;
 
 	/* unhash it and decrease its reference counter */
-	hash = ip_vs_conn_hashkey(cp->af, cp->protocol, &cp->caddr, cp->cport);
+	hash = ip_vs_conn_hashkey_conn(cp);
 
 	ct_write_lock(hash);
 	spin_lock(&cp->lock);
@@ -227,7 +263,7 @@ __ip_vs_conn_in_get(const struct ip_vs_conn_param *p)
 	unsigned hash;
 	struct ip_vs_conn *cp;
 
-	hash = ip_vs_conn_hashkey(p->af, p->protocol, p->caddr, p->cport);
+	hash = ip_vs_conn_hashkey_param(p, false);
 
 	ct_read_lock(hash);
 
@@ -312,11 +348,17 @@ struct ip_vs_conn *ip_vs_ct_in_get(const struct ip_vs_conn_param *p)
 	unsigned hash;
 	struct ip_vs_conn *cp;
 
-	hash = ip_vs_conn_hashkey(p->af, p->protocol, p->caddr, p->cport);
+	hash = ip_vs_conn_hashkey_param(p, false);
 
 	ct_read_lock(hash);
 
 	list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) {
+		if (p->pe && p->pe->ct_match) {
+			if (p->pe->ct_match(p, cp))
+				goto out;
+			continue;
+		}
+
 		if (cp->af == p->af &&
 		    ip_vs_addr_equal(p->af, p->caddr, &cp->caddr) &&
 		    /* protocol should only be IPPROTO_IP if
@@ -325,15 +367,14 @@ struct ip_vs_conn *ip_vs_ct_in_get(const struct ip_vs_conn_param *p)
 				     p->af, p->vaddr, &cp->vaddr) &&
 		    p->cport == cp->cport && p->vport == cp->vport &&
 		    cp->flags & IP_VS_CONN_F_TEMPLATE &&
-		    p->protocol == cp->protocol) {
-			/* HIT */
-			atomic_inc(&cp->refcnt);
+		    p->protocol == cp->protocol)
 			goto out;
-		}
 	}
 	cp = NULL;
 
   out:
+	if (cp)
+		atomic_inc(&cp->refcnt);
 	ct_read_unlock(hash);
 
 	IP_VS_DBG_BUF(9, "template lookup/in %s %s:%d->%s:%d %s\n",
@@ -357,7 +398,7 @@ struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p)
 	/*
 	 *	Check for "full" addressed entries
 	 */
-	hash = ip_vs_conn_hashkey(p->af, p->protocol, p->vaddr, p->vport);
+	hash = ip_vs_conn_hashkey_param(p, true);
 
 	ct_read_lock(hash);
 
@@ -722,6 +763,7 @@ static void ip_vs_conn_expire(unsigned long data)
 		if (cp->flags & IP_VS_CONN_F_NFCT)
 			ip_vs_conn_drop_conntrack(cp);
 
+		kfree(cp->pe_data);
 		if (unlikely(cp->app != NULL))
 			ip_vs_unbind_app(cp);
 		ip_vs_unbind_dest(cp);
@@ -782,6 +824,10 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p,
 			&cp->daddr, daddr);
 	cp->dport          = dport;
 	cp->flags	   = flags;
+	if (flags & IP_VS_CONN_F_TEMPLATE && p->pe_data) {
+		cp->pe_data = p->pe_data;
+		cp->pe_data_len = p->pe_data_len;
+	}
 	spin_lock_init(&cp->lock);
 
 	/*
@@ -832,7 +878,6 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p,
 	return cp;
 }
 
-
 /*
  *	/proc/net/ip_vs_conn entries
  */
@@ -848,7 +893,7 @@ static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos)
 		list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) {
 			if (pos-- == 0) {
 				seq->private = &ip_vs_conn_tab[idx];
-				return cp;
+			return cp;
 			}
 		}
 		ct_read_unlock_bh(idx);
diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
index 87602a62458e..ab9889380496 100644
--- a/net/netfilter/ipvs/ip_vs_core.c
+++ b/net/netfilter/ipvs/ip_vs_core.c
@@ -176,6 +176,19 @@ ip_vs_set_state(struct ip_vs_conn *cp, int direction,
 	return pp->state_transition(cp, direction, skb, pp);
 }
 
+static inline int
+ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc,
+			      struct sk_buff *skb, int protocol,
+			      const union nf_inet_addr *caddr, __be16 cport,
+			      const union nf_inet_addr *vaddr, __be16 vport,
+			      struct ip_vs_conn_param *p)
+{
+	ip_vs_conn_fill_param(svc->af, protocol, caddr, cport, vaddr, vport, p);
+	p->pe = svc->pe;
+	if (p->pe && p->pe->fill_param)
+		return p->pe->fill_param(p, skb);
+	return 0;
+}
 
 /*
  *  IPVS persistent scheduling function
@@ -186,7 +199,7 @@ ip_vs_set_state(struct ip_vs_conn *cp, int direction,
  */
 static struct ip_vs_conn *
 ip_vs_sched_persist(struct ip_vs_service *svc,
-		    const struct sk_buff *skb,
+		    struct sk_buff *skb,
 		    __be16 ports[2])
 {
 	struct ip_vs_conn *cp = NULL;
@@ -255,8 +268,9 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 				vaddr = &fwmark;
 			}
 		}
-		ip_vs_conn_fill_param(svc->af, protocol, &snet, 0,
-				      vaddr, vport, &param);
+		if (ip_vs_conn_fill_param_persist(svc, skb, protocol, &snet, 0,
+						  vaddr, vport, &param))
+			return NULL;
 	}
 
 	/* Check if a template already exists */
@@ -268,22 +282,30 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
 		dest = svc->scheduler->schedule(svc, skb);
 		if (!dest) {
 			IP_VS_DBG(1, "p-schedule: no dest found.\n");
+			kfree(param.pe_data);
 			return NULL;
 		}
 
 		if (ports[1] == svc->port && svc->port != FTPPORT)
 			dport = dest->port;
 
-		/* Create a template */
+		/* Create a template
+		 * This adds param.pe_data to the template,
+		 * and thus param.pe_data will be destroyed
+		 * when the template expires */
 		ct = ip_vs_conn_new(&param, &dest->addr, dport,
 				    IP_VS_CONN_F_TEMPLATE, dest);
-		if (ct == NULL)
+		if (ct == NULL) {
+			kfree(param.pe_data);
 			return NULL;
+		}
 
 		ct->timeout = svc->timeout;
-	} else
+	} else {
 		/* set destination with the found template */
 		dest = ct->dest;
+		kfree(param.pe_data);
+	}
 
 	dport = ports[1];
 	if (dport == svc->port && dest->port)
@@ -322,7 +344,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc,
  *  Protocols supported: TCP, UDP
  */
 struct ip_vs_conn *
-ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb)
+ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb)
 {
 	struct ip_vs_conn *cp = NULL;
 	struct ip_vs_iphdr iph;
diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c
index f68631f75f09..ab85aedea17e 100644
--- a/net/netfilter/ipvs/ip_vs_sync.c
+++ b/net/netfilter/ipvs/ip_vs_sync.c
@@ -288,6 +288,16 @@ void ip_vs_sync_conn(struct ip_vs_conn *cp)
 		ip_vs_sync_conn(cp->control);
 }
 
+static inline int
+ip_vs_conn_fill_param_sync(int af, int protocol,
+			   const union nf_inet_addr *caddr, __be16 cport,
+			   const union nf_inet_addr *vaddr, __be16 vport,
+			   struct ip_vs_conn_param *p)
+{
+	/* XXX: Need to take into account persistence engine */
+	ip_vs_conn_fill_param(af, protocol, caddr, cport, vaddr, vport, p);
+	return 0;
+}
 
 /*
  *      Process received multicast message and create the corresponding
@@ -372,11 +382,14 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen)
 		}
 
 		{
-			ip_vs_conn_fill_param(AF_INET, s->protocol,
+			if (ip_vs_conn_fill_param_sync(AF_INET, s->protocol,
 					      (union nf_inet_addr *)&s->caddr,
 					      s->cport,
 					      (union nf_inet_addr *)&s->vaddr,
-					      s->vport, &param);
+					      s->vport, &param)) {
+				pr_err("ip_vs_conn_fill_param_sync failed");
+				return;
+			}
 			if (!(flags & IP_VS_CONN_F_TEMPLATE))
 				cp = ip_vs_conn_in_get(&param);
 			else
-- 
cgit v1.2.3


From 0d1e71b04a04b6912e50926b9987c1e72facb1f3 Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@verge.net.au>
Date: Sun, 22 Aug 2010 21:37:54 +0900
Subject: IPVS: Allow configuration of persistence engines

Allow the persistence engine of a virtual service to be set, edited
and unset.

This feature only works with the netlink user-space interface.

Signed-off-by: Simon Horman <horms@verge.net.au>
Acked-by: Julian Anastasov <ja@ssi.bg>
---
 include/linux/ip_vs.h          |  3 +++
 include/net/ip_vs.h            |  1 +
 net/netfilter/ipvs/ip_vs_ctl.c | 57 +++++++++++++++++++++++++++++++++++++++---
 3 files changed, 58 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ip_vs.h b/include/linux/ip_vs.h
index 0a9c44d64292..5f43a3b2e3ad 100644
--- a/include/linux/ip_vs.h
+++ b/include/linux/ip_vs.h
@@ -336,6 +336,9 @@ enum {
 	IPVS_SVC_ATTR_NETMASK,		/* persistent netmask */
 
 	IPVS_SVC_ATTR_STATS,		/* nested attribute for service stats */
+
+	IPVS_SVC_ATTR_PE_NAME,		/* name of ct retriever */
+
 	__IPVS_SVC_ATTR_MAX,
 };
 
diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index a6b93a26d4e2..52fbe2308c38 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -444,6 +444,7 @@ struct ip_vs_service_user_kern {
 
 	/* virtual service options */
 	char			*sched_name;
+	char			*pe_name;
 	unsigned		flags;		/* virtual service flags */
 	unsigned		timeout;	/* persistent timeout in sec */
 	u32			netmask;	/* persistent netmask */
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index 5f20caf47a1d..a697591d0e23 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -1134,6 +1134,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u,
 {
 	int ret = 0;
 	struct ip_vs_scheduler *sched = NULL;
+	struct ip_vs_pe *pe = NULL;
 	struct ip_vs_service *svc = NULL;
 
 	/* increase the module use count */
@@ -1147,6 +1148,16 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u,
 		goto out_err;
 	}
 
+	if (u->pe_name && *u->pe_name) {
+		pe = ip_vs_pe_get(u->pe_name);
+		if (pe == NULL) {
+			pr_info("persistence engine module ip_vs_pe_%s "
+				"not found\n", u->pe_name);
+			ret = -ENOENT;
+			goto out_err;
+		}
+	}
+
 #ifdef CONFIG_IP_VS_IPV6
 	if (u->af == AF_INET6 && (u->netmask < 1 || u->netmask > 128)) {
 		ret = -EINVAL;
@@ -1184,6 +1195,10 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u,
 		goto out_err;
 	sched = NULL;
 
+	/* Bind the ct retriever */
+	ip_vs_bind_pe(svc, pe);
+	pe = NULL;
+
 	/* Update the virtual service counters */
 	if (svc->port == FTPPORT)
 		atomic_inc(&ip_vs_ftpsvc_counter);
@@ -1215,6 +1230,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u,
 		kfree(svc);
 	}
 	ip_vs_scheduler_put(sched);
+	ip_vs_pe_put(pe);
 
 	/* decrease the module use count */
 	ip_vs_use_count_dec();
@@ -1230,6 +1246,7 @@ static int
 ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u)
 {
 	struct ip_vs_scheduler *sched, *old_sched;
+	struct ip_vs_pe *pe = NULL, *old_pe = NULL;
 	int ret = 0;
 
 	/*
@@ -1242,6 +1259,17 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u)
 	}
 	old_sched = sched;
 
+	if (u->pe_name && *u->pe_name) {
+		pe = ip_vs_pe_get(u->pe_name);
+		if (pe == NULL) {
+			pr_info("persistence engine module ip_vs_pe_%s "
+				"not found\n", u->pe_name);
+			ret = -ENOENT;
+			goto out;
+		}
+		old_pe = pe;
+	}
+
 #ifdef CONFIG_IP_VS_IPV6
 	if (u->af == AF_INET6 && (u->netmask < 1 || u->netmask > 128)) {
 		ret = -EINVAL;
@@ -1293,12 +1321,17 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u)
 		}
 	}
 
+	old_pe = svc->pe;
+	if (pe != old_pe) {
+		ip_vs_unbind_pe(svc);
+		ip_vs_bind_pe(svc, pe);
+	}
+
   out_unlock:
 	write_unlock_bh(&__ip_vs_svc_lock);
-#ifdef CONFIG_IP_VS_IPV6
   out:
-#endif
 	ip_vs_scheduler_put(old_sched);
+	ip_vs_pe_put(old_pe);
 	return ret;
 }
 
@@ -1312,6 +1345,9 @@ static void __ip_vs_del_service(struct ip_vs_service *svc)
 {
 	struct ip_vs_dest *dest, *nxt;
 	struct ip_vs_scheduler *old_sched;
+	struct ip_vs_pe *old_pe;
+
+	pr_info("%s: enter\n", __func__);
 
 	/* Count only IPv4 services for old get/setsockopt interface */
 	if (svc->af == AF_INET)
@@ -1324,6 +1360,11 @@ static void __ip_vs_del_service(struct ip_vs_service *svc)
 	ip_vs_unbind_scheduler(svc);
 	ip_vs_scheduler_put(old_sched);
 
+	/* Unbind persistence engine */
+	old_pe = svc->pe;
+	ip_vs_unbind_pe(svc);
+	ip_vs_pe_put(old_pe);
+
 	/* Unbind app inc */
 	if (svc->inc) {
 		ip_vs_app_inc_put(svc->inc);
@@ -2026,6 +2067,8 @@ static const unsigned char set_arglen[SET_CMDID(IP_VS_SO_SET_MAX)+1] = {
 static void ip_vs_copy_usvc_compat(struct ip_vs_service_user_kern *usvc,
 				  struct ip_vs_service_user *usvc_compat)
 {
+	memset(usvc, 0, sizeof(*usvc));
+
 	usvc->af		= AF_INET;
 	usvc->protocol		= usvc_compat->protocol;
 	usvc->addr.ip		= usvc_compat->addr;
@@ -2043,6 +2086,8 @@ static void ip_vs_copy_usvc_compat(struct ip_vs_service_user_kern *usvc,
 static void ip_vs_copy_udest_compat(struct ip_vs_dest_user_kern *udest,
 				   struct ip_vs_dest_user *udest_compat)
 {
+	memset(udest, 0, sizeof(*udest));
+
 	udest->addr.ip		= udest_compat->addr;
 	udest->port		= udest_compat->port;
 	udest->conn_flags	= udest_compat->conn_flags;
@@ -2539,6 +2584,8 @@ static const struct nla_policy ip_vs_svc_policy[IPVS_SVC_ATTR_MAX + 1] = {
 	[IPVS_SVC_ATTR_FWMARK]		= { .type = NLA_U32 },
 	[IPVS_SVC_ATTR_SCHED_NAME]	= { .type = NLA_NUL_STRING,
 					    .len = IP_VS_SCHEDNAME_MAXLEN },
+	[IPVS_SVC_ATTR_PE_NAME]		= { .type = NLA_NUL_STRING,
+					    .len = IP_VS_PENAME_MAXLEN },
 	[IPVS_SVC_ATTR_FLAGS]		= { .type = NLA_BINARY,
 					    .len = sizeof(struct ip_vs_flags) },
 	[IPVS_SVC_ATTR_TIMEOUT]		= { .type = NLA_U32 },
@@ -2615,6 +2662,8 @@ static int ip_vs_genl_fill_service(struct sk_buff *skb,
 	}
 
 	NLA_PUT_STRING(skb, IPVS_SVC_ATTR_SCHED_NAME, svc->scheduler->name);
+	if (svc->pe)
+		NLA_PUT_STRING(skb, IPVS_SVC_ATTR_PE_NAME, svc->pe->name);
 	NLA_PUT(skb, IPVS_SVC_ATTR_FLAGS, sizeof(flags), &flags);
 	NLA_PUT_U32(skb, IPVS_SVC_ATTR_TIMEOUT, svc->timeout / HZ);
 	NLA_PUT_U32(skb, IPVS_SVC_ATTR_NETMASK, svc->netmask);
@@ -2741,11 +2790,12 @@ static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc,
 
 	/* If a full entry was requested, check for the additional fields */
 	if (full_entry) {
-		struct nlattr *nla_sched, *nla_flags, *nla_timeout,
+		struct nlattr *nla_sched, *nla_flags, *nla_pe, *nla_timeout,
 			      *nla_netmask;
 		struct ip_vs_flags flags;
 
 		nla_sched = attrs[IPVS_SVC_ATTR_SCHED_NAME];
+		nla_pe = attrs[IPVS_SVC_ATTR_PE_NAME];
 		nla_flags = attrs[IPVS_SVC_ATTR_FLAGS];
 		nla_timeout = attrs[IPVS_SVC_ATTR_TIMEOUT];
 		nla_netmask = attrs[IPVS_SVC_ATTR_NETMASK];
@@ -2763,6 +2813,7 @@ static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc,
 		usvc->flags = (usvc->flags & ~flags.mask) |
 			      (flags.flags & flags.mask);
 		usvc->sched_name = nla_data(nla_sched);
+		usvc->pe_name = nla_pe ? nla_data(nla_pe) : NULL;
 		usvc->timeout = nla_get_u32(nla_timeout);
 		usvc->netmask = nla_get_u32(nla_netmask);
 	}
-- 
cgit v1.2.3


From 2e54eb96e2c801f33d95b5dade15212ac4d6c4a5 Mon Sep 17 00:00:00 2001
From: Petr Vandrovec <petr@vandrovec.name>
Date: Mon, 27 Sep 2010 01:47:33 +0200
Subject: BKL: Remove BKL from ncpfs

Dozen of changes in ncpfs to provide some locking other than BKL.

In readdir cache unlock and mark complete first page as last operation,
so it can be used for synchronization, as code intended.

When updating dentry name on case insensitive filesystems do at least
some basic locking...

Hold i_mutex when updating inode fields.

Push some ncp_conn_is_valid down to ncp_request.  Connection can become
invalid at any moment, and fewer error code paths to test the better.

Use i_size_{read,write} to modify file size.

Set inode's backing_dev_info as ncpfs has its own special bdi.

In ioctl unbreak ioctls invoked on filesystem mounted 'ro' - tests are
for inode writeable or owner match, but were turned to filesystem
writeable and inode writeable or owner match.  Also collect all permission
checks in single place.

Add some locking, and remove comments saying that it would be cool to
add some locks to the code.

Constify some pointers.

Signed-off-by: Petr Vandrovec <petr@vandrovec.name>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 fs/ncpfs/dir.c            | 221 +++++++++++++---------
 fs/ncpfs/file.c           |  25 +--
 fs/ncpfs/inode.c          |  41 ++--
 fs/ncpfs/ioctl.c          | 470 +++++++++++++++++++++++++---------------------
 fs/ncpfs/ncplib_kernel.c  | 101 +++++-----
 fs/ncpfs/ncplib_kernel.h  |  15 +-
 fs/ncpfs/ncpsign_kernel.c |  10 +-
 fs/ncpfs/sock.c           |   1 -
 include/linux/ncp_fs.h    |  28 ---
 include/linux/ncp_fs_sb.h |   4 +-
 10 files changed, 487 insertions(+), 429 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c
index 9578cbe0cd58..aac8832e919e 100644
--- a/fs/ncpfs/dir.c
+++ b/fs/ncpfs/dir.c
@@ -95,6 +95,34 @@ const struct dentry_operations ncp_root_dentry_operations =
 };
 
 
+#define ncp_namespace(i)	(NCP_SERVER(i)->name_space[NCP_FINFO(i)->volNumber])
+
+static inline int ncp_preserve_entry_case(struct inode *i, __u32 nscreator)
+{
+#ifdef CONFIG_NCPFS_SMALLDOS
+	int ns = ncp_namespace(i);
+
+	if ((ns == NW_NS_DOS)
+#ifdef CONFIG_NCPFS_OS2_NS
+		|| ((ns == NW_NS_OS2) && (nscreator == NW_NS_DOS))
+#endif /* CONFIG_NCPFS_OS2_NS */
+	   )
+		return 0;
+#endif /* CONFIG_NCPFS_SMALLDOS */
+	return 1;
+}
+
+#define ncp_preserve_case(i)	(ncp_namespace(i) != NW_NS_DOS)
+
+static inline int ncp_case_sensitive(struct dentry *dentry)
+{
+#ifdef CONFIG_NCPFS_NFS_NS
+	return ncp_namespace(dentry->d_inode) == NW_NS_NFS;
+#else
+	return 0;
+#endif /* CONFIG_NCPFS_NFS_NS */
+}
+
 /*
  * Note: leave the hash unchanged if the directory
  * is case-sensitive.
@@ -102,13 +130,12 @@ const struct dentry_operations ncp_root_dentry_operations =
 static int 
 ncp_hash_dentry(struct dentry *dentry, struct qstr *this)
 {
-	struct nls_table *t;
-	unsigned long hash;
-	int i;
-
-	t = NCP_IO_TABLE(dentry);
+	if (!ncp_case_sensitive(dentry)) {
+		struct nls_table *t;
+		unsigned long hash;
+		int i;
 
-	if (!ncp_case_sensitive(dentry->d_inode)) {
+		t = NCP_IO_TABLE(dentry);
 		hash = init_name_hash();
 		for (i=0; i<this->len ; i++)
 			hash = partial_name_hash(ncp_tolower(t, this->name[i]),
@@ -124,7 +151,7 @@ ncp_compare_dentry(struct dentry *dentry, struct qstr *a, struct qstr *b)
 	if (a->len != b->len)
 		return 1;
 
-	if (ncp_case_sensitive(dentry->d_inode))
+	if (ncp_case_sensitive(dentry))
 		return strncmp(a->name, b->name, a->len);
 
 	return ncp_strnicmp(NCP_IO_TABLE(dentry), a->name, b->name, a->len);
@@ -266,7 +293,7 @@ leave_me:;
 
 
 static int
-__ncp_lookup_validate(struct dentry *dentry)
+ncp_lookup_validate(struct dentry *dentry, struct nameidata *nd)
 {
 	struct ncp_server *server;
 	struct dentry *parent;
@@ -283,9 +310,6 @@ __ncp_lookup_validate(struct dentry *dentry)
 
 	server = NCP_SERVER(dir);
 
-	if (!ncp_conn_valid(server))
-		goto finished;
-
 	/*
 	 * Inspired by smbfs:
 	 * The default validation is based on dentry age:
@@ -304,8 +328,11 @@ __ncp_lookup_validate(struct dentry *dentry)
 	if (ncp_is_server_root(dir)) {
 		res = ncp_io2vol(server, __name, &len, dentry->d_name.name,
 				 dentry->d_name.len, 1);
-		if (!res)
+		if (!res) {
 			res = ncp_lookup_volume(server, __name, &(finfo.i));
+			if (!res)
+				ncp_update_known_namespace(server, finfo.i.volNumber, NULL);
+		}
 	} else {
 		res = ncp_io2vol(server, __name, &len, dentry->d_name.name,
 				 dentry->d_name.len, !ncp_preserve_case(dir));
@@ -320,13 +347,17 @@ __ncp_lookup_validate(struct dentry *dentry)
 	 * what we remember, it's not valid any more.
 	 */
 	if (!res) {
-		if (finfo.i.dirEntNum == NCP_FINFO(dentry->d_inode)->dirEntNum) {
+		struct inode *inode = dentry->d_inode;
+
+		mutex_lock(&inode->i_mutex);
+		if (finfo.i.dirEntNum == NCP_FINFO(inode)->dirEntNum) {
 			ncp_new_dentry(dentry);
 			val=1;
 		} else
 			DDPRINTK("ncp_lookup_validate: found, but dirEntNum changed\n");
 
-		ncp_update_inode2(dentry->d_inode, &finfo);
+		ncp_update_inode2(inode, &finfo);
+		mutex_unlock(&inode->i_mutex);
 	}
 
 finished:
@@ -335,16 +366,6 @@ finished:
 	return val;
 }
 
-static int
-ncp_lookup_validate(struct dentry * dentry, struct nameidata *nd)
-{
-	int res;
-	lock_kernel();
-	res = __ncp_lookup_validate(dentry);
-	unlock_kernel();
-	return res;
-}
-
 static struct dentry *
 ncp_dget_fpos(struct dentry *dentry, struct dentry *parent, unsigned long fpos)
 {
@@ -411,8 +432,6 @@ static int ncp_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	int result, mtime_valid = 0;
 	time_t mtime = 0;
 
-	lock_kernel();
-
 	ctl.page  = NULL;
 	ctl.cache = NULL;
 
@@ -421,6 +440,7 @@ static int ncp_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		(int) filp->f_pos);
 
 	result = -EIO;
+	/* Do not generate '.' and '..' when server is dead. */
 	if (!ncp_conn_valid(server))
 		goto out;
 
@@ -532,6 +552,12 @@ read_really:
 	ctl.head.end = ctl.fpos - 1;
 	ctl.head.eof = ctl.valid;
 finished:
+	if (ctl.page) {
+		kunmap(ctl.page);
+		SetPageUptodate(ctl.page);
+		unlock_page(ctl.page);
+		page_cache_release(ctl.page);
+	}
 	if (page) {
 		cache->head = ctl.head;
 		kunmap(page);
@@ -539,23 +565,17 @@ finished:
 		unlock_page(page);
 		page_cache_release(page);
 	}
-	if (ctl.page) {
-		kunmap(ctl.page);
-		SetPageUptodate(ctl.page);
-		unlock_page(ctl.page);
-		page_cache_release(ctl.page);
-	}
 out:
-	unlock_kernel();
 	return result;
 }
 
 static int
 ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
-		struct ncp_cache_control *ctrl, struct ncp_entry_info *entry)
+		struct ncp_cache_control *ctrl, struct ncp_entry_info *entry,
+		int inval_childs)
 {
 	struct dentry *newdent, *dentry = filp->f_path.dentry;
-	struct inode *newino, *inode = dentry->d_inode;
+	struct inode *dir = dentry->d_inode;
 	struct ncp_cache_control ctl = *ctrl;
 	struct qstr qname;
 	int valid = 0;
@@ -564,9 +584,9 @@ ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
 	__u8 __name[NCP_MAXPATHLEN + 1];
 
 	qname.len = sizeof(__name);
-	if (ncp_vol2io(NCP_SERVER(inode), __name, &qname.len,
+	if (ncp_vol2io(NCP_SERVER(dir), __name, &qname.len,
 			entry->i.entryName, entry->i.nameLen,
-			!ncp_preserve_entry_case(inode, entry->i.NSCreator)))
+			!ncp_preserve_entry_case(dir, entry->i.NSCreator)))
 		return 1; /* I'm not sure */
 
 	qname.name = __name;
@@ -584,22 +604,64 @@ ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
 			goto end_advance;
 	} else {
 		hashed = 1;
-		memcpy((char *) newdent->d_name.name, qname.name,
-							newdent->d_name.len);
+
+		/* If case sensitivity changed for this volume, all entries below this one
+		   should be thrown away.  This entry itself is not affected, as its case
+		   sensitivity is controlled by its own parent. */
+		if (inval_childs)
+			shrink_dcache_parent(newdent);
+
+		/*
+		 * It is not as dangerous as it looks.  NetWare's OS2 namespace is
+		 * case preserving yet case insensitive.  So we update dentry's name
+		 * as received from server.  We found dentry via d_lookup with our
+		 * hash, so we know that hash does not change, and so replacing name
+		 * should be reasonably safe.
+		 */
+		if (qname.len == newdent->d_name.len &&
+		    memcmp(newdent->d_name.name, qname.name, newdent->d_name.len)) {
+			struct inode *inode = newdent->d_inode;
+
+			/*
+			 * Inside ncpfs all uses of d_name are either for debugging,
+			 * or on functions which acquire inode mutex (mknod, creat,
+			 * lookup).  So grab i_mutex here, to be sure.  d_path
+			 * uses dcache_lock when generating path, so we should too.
+			 * And finally d_compare is protected by dentry's d_lock, so
+			 * here we go.
+			 */
+			if (inode)
+				mutex_lock(&inode->i_mutex);
+			spin_lock(&dcache_lock);
+			spin_lock(&newdent->d_lock);
+			memcpy((char *) newdent->d_name.name, qname.name,
+								newdent->d_name.len);
+			spin_unlock(&newdent->d_lock);
+			spin_unlock(&dcache_lock);
+			if (inode)
+				mutex_unlock(&inode->i_mutex);
+		}
 	}
 
 	if (!newdent->d_inode) {
+		struct inode *inode;
+
 		entry->opened = 0;
-		entry->ino = iunique(inode->i_sb, 2);
-		newino = ncp_iget(inode->i_sb, entry);
-		if (newino) {
+		entry->ino = iunique(dir->i_sb, 2);
+		inode = ncp_iget(dir->i_sb, entry);
+		if (inode) {
 			newdent->d_op = &ncp_dentry_operations;
-			d_instantiate(newdent, newino);
+			d_instantiate(newdent, inode);
 			if (!hashed)
 				d_rehash(newdent);
 		}
-	} else
-		ncp_update_inode2(newdent->d_inode, entry);
+	} else {
+		struct inode *inode = newdent->d_inode;
+
+		mutex_lock(&inode->i_mutex);
+		ncp_update_inode2(inode, entry);
+		mutex_unlock(&inode->i_mutex);
+	}
 
 	if (newdent->d_inode) {
 		ino = newdent->d_inode->i_ino;
@@ -617,7 +679,7 @@ ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir,
 		ctl.cache = NULL;
 		ctl.idx  -= NCP_DIRCACHE_SIZE;
 		ctl.ofs  += 1;
-		ctl.page  = grab_cache_page(&inode->i_data, ctl.ofs);
+		ctl.page  = grab_cache_page(&dir->i_data, ctl.ofs);
 		if (ctl.page)
 			ctl.cache = kmap(ctl.page);
 	}
@@ -633,7 +695,7 @@ end_advance:
 		if (!ino)
 			ino = find_inode_number(dentry, &qname);
 		if (!ino)
-			ino = iunique(inode->i_sb, 2);
+			ino = iunique(dir->i_sb, 2);
 		ctl.filled = filldir(dirent, qname.name, qname.len,
 				     filp->f_pos, ino, DT_UNKNOWN);
 		if (!ctl.filled)
@@ -660,6 +722,7 @@ ncp_read_volume_list(struct file *filp, void *dirent, filldir_t filldir,
 			(unsigned long) filp->f_pos);
 
 	for (i = 0; i < NCP_NUMBER_OF_VOLUMES; i++) {
+		int inval_dentry;
 
 		if (ncp_get_volume_info_with_number(server, i, &info) != 0)
 			return;
@@ -675,8 +738,9 @@ ncp_read_volume_list(struct file *filp, void *dirent, filldir_t filldir,
 				info.volume_name);
 			continue;
 		}
+		inval_dentry = ncp_update_known_namespace(server, entry.i.volNumber, NULL);
 		entry.volume = entry.i.volNumber;
-		if (!ncp_fill_cache(filp, dirent, filldir, ctl, &entry))
+		if (!ncp_fill_cache(filp, dirent, filldir, ctl, &entry, inval_dentry))
 			return;
 	}
 }
@@ -739,7 +803,7 @@ ncp_do_readdir(struct file *filp, void *dirent, filldir_t filldir,
 			rpl += onerpl;
 			rpls -= onerpl;
 			entry.volume = entry.i.volNumber;
-			if (!ncp_fill_cache(filp, dirent, filldir, ctl, &entry))
+			if (!ncp_fill_cache(filp, dirent, filldir, ctl, &entry, 0))
 				break;
 		}
 	} while (more);
@@ -775,17 +839,19 @@ int ncp_conn_logged_in(struct super_block *sb)
 		if (dent) {
 			struct inode* ino = dent->d_inode;
 			if (ino) {
+				ncp_update_known_namespace(server, volNumber, NULL);
 				NCP_FINFO(ino)->volNumber = volNumber;
 				NCP_FINFO(ino)->dirEntNum = dirEntNum;
 				NCP_FINFO(ino)->DosDirNum = DosDirNum;
+				result = 0;
 			} else {
 				DPRINTK("ncpfs: sb->s_root->d_inode == NULL!\n");
 			}
 		} else {
 			DPRINTK("ncpfs: sb->s_root == NULL!\n");
 		}
-	}
-	result = 0;
+	} else
+		result = 0;
 
 out:
 	return result;
@@ -799,7 +865,6 @@ static struct dentry *ncp_lookup(struct inode *dir, struct dentry *dentry, struc
 	int error, res, len;
 	__u8 __name[NCP_MAXPATHLEN + 1];
 
-	lock_kernel();
 	error = -EIO;
 	if (!ncp_conn_valid(server))
 		goto finished;
@@ -813,6 +878,8 @@ static struct dentry *ncp_lookup(struct inode *dir, struct dentry *dentry, struc
 				 dentry->d_name.len, 1);
 		if (!res)
 			res = ncp_lookup_volume(server, __name, &(finfo.i));
+			if (!res)
+				ncp_update_known_namespace(server, finfo.i.volNumber, NULL);
 	} else {
 		res = ncp_io2vol(server, __name, &len, dentry->d_name.name,
 				 dentry->d_name.len, !ncp_preserve_case(dir));
@@ -846,7 +913,6 @@ add_entry:
 
 finished:
 	PPRINTK("ncp_lookup: result=%d\n", error);
-	unlock_kernel();
 	return ERR_PTR(error);
 }
 
@@ -887,11 +953,6 @@ int ncp_create_new(struct inode *dir, struct dentry *dentry, int mode,
 	PPRINTK("ncp_create_new: creating %s/%s, mode=%x\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name, mode);
 
-	error = -EIO;
-	lock_kernel();
-	if (!ncp_conn_valid(server))
-		goto out;
-
 	ncp_age_dentry(server, dentry);
 	len = sizeof(__name);
 	error = ncp_io2vol(server, __name, &len, dentry->d_name.name,
@@ -917,6 +978,8 @@ int ncp_create_new(struct inode *dir, struct dentry *dentry, int mode,
 		if (result) {
 			if (result == 0x87)
 				error = -ENAMETOOLONG;
+			else if (result < 0)
+				error = result;
 			DPRINTK("ncp_create: %s/%s failed\n",
 				dentry->d_parent->d_name.name, dentry->d_name.name);
 			goto out;
@@ -935,7 +998,6 @@ int ncp_create_new(struct inode *dir, struct dentry *dentry, int mode,
 
 	error = ncp_instantiate(dir, dentry, &finfo);
 out:
-	unlock_kernel();
 	return error;
 }
 
@@ -955,11 +1017,6 @@ static int ncp_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	DPRINTK("ncp_mkdir: making %s/%s\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name);
 
-	error = -EIO;
-	lock_kernel();
-	if (!ncp_conn_valid(server))
-		goto out;
-
 	ncp_age_dentry(server, dentry);
 	len = sizeof(__name);
 	error = ncp_io2vol(server, __name, &len, dentry->d_name.name,
@@ -967,12 +1024,11 @@ static int ncp_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (error)
 		goto out;
 
-	error = -EACCES;
-	if (ncp_open_create_file_or_subdir(server, dir, __name,
+	error = ncp_open_create_file_or_subdir(server, dir, __name,
 					   OC_MODE_CREATE, aDIR,
 					   cpu_to_le16(0xffff),
-					   &finfo) == 0)
-	{
+					   &finfo);
+	if (error == 0) {
 		if (ncp_is_nfs_extras(server, finfo.volume)) {
 			mode |= S_IFDIR;
 			finfo.i.nfs.mode = mode;
@@ -983,9 +1039,10 @@ static int ncp_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 				goto out;
 		}
 		error = ncp_instantiate(dir, dentry, &finfo);
+	} else if (error > 0) {
+		error = -EACCES;
 	}
 out:
-	unlock_kernel();
 	return error;
 }
 
@@ -998,11 +1055,6 @@ static int ncp_rmdir(struct inode *dir, struct dentry *dentry)
 	DPRINTK("ncp_rmdir: removing %s/%s\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name);
 
-	error = -EIO;
-	lock_kernel();
-	if (!ncp_conn_valid(server))
-		goto out;
-
 	error = -EBUSY;
 	if (!d_unhashed(dentry))
 		goto out;
@@ -1036,11 +1088,10 @@ static int ncp_rmdir(struct inode *dir, struct dentry *dentry)
 			error = -ENOENT;
 			break;
 		default:
-			error = -EACCES;
+			error = result < 0 ? result : -EACCES;
 			break;
        	}
 out:
-	unlock_kernel();
 	return error;
 }
 
@@ -1050,15 +1101,10 @@ static int ncp_unlink(struct inode *dir, struct dentry *dentry)
 	struct ncp_server *server;
 	int error;
 
-	lock_kernel();
 	server = NCP_SERVER(dir);
 	DPRINTK("ncp_unlink: unlinking %s/%s\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name);
 	
-	error = -EIO;
-	if (!ncp_conn_valid(server))
-		goto out;
-
 	/*
 	 * Check whether to close the file ...
 	 */
@@ -1097,12 +1143,9 @@ static int ncp_unlink(struct inode *dir, struct dentry *dentry)
 			error = -ENOENT;
 			break;
 		default:
-			error = -EACCES;
+			error = error < 0 ? error : -EACCES;
 			break;
 	}
-		
-out:
-	unlock_kernel();
 	return error;
 }
 
@@ -1118,11 +1161,6 @@ static int ncp_rename(struct inode *old_dir, struct dentry *old_dentry,
 		old_dentry->d_parent->d_name.name, old_dentry->d_name.name,
 		new_dentry->d_parent->d_name.name, new_dentry->d_name.name);
 
-	error = -EIO;
-	lock_kernel();
-	if (!ncp_conn_valid(server))
-		goto out;
-
 	ncp_age_dentry(server, old_dentry);
 	ncp_age_dentry(server, new_dentry);
 
@@ -1161,11 +1199,10 @@ static int ncp_rename(struct inode *old_dir, struct dentry *old_dentry,
 			error = -ENOENT;
 			break;
 		default:
-			error = -EACCES;
+			error = error < 0 ? error : -EACCES;
 			break;
 	}
 out:
-	unlock_kernel();
 	return error;
 }
 
diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c
index 3639cc5cbdae..6c754f70c529 100644
--- a/fs/ncpfs/file.c
+++ b/fs/ncpfs/file.c
@@ -113,9 +113,6 @@ ncp_file_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 	DPRINTK("ncp_file_read: enter %s/%s\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name);
 
-	if (!ncp_conn_valid(NCP_SERVER(inode)))
-		return -EIO;
-
 	pos = *ppos;
 
 	if ((ssize_t) count < 0) {
@@ -192,13 +189,11 @@ ncp_file_write(struct file *file, const char __user *buf, size_t count, loff_t *
 
 	DPRINTK("ncp_file_write: enter %s/%s\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name);
-	if (!ncp_conn_valid(NCP_SERVER(inode)))
-		return -EIO;
 	if ((ssize_t) count < 0)
 		return -EINVAL;
 	pos = *ppos;
 	if (file->f_flags & O_APPEND) {
-		pos = inode->i_size;
+		pos = i_size_read(inode);
 	}
 
 	if (pos + count > MAX_NON_LFS && !(file->f_flags&O_LARGEFILE)) {
@@ -264,8 +259,11 @@ ncp_file_write(struct file *file, const char __user *buf, size_t count, loff_t *
 
 	*ppos = pos;
 
-	if (pos > inode->i_size) {
-		inode->i_size = pos;
+	if (pos > i_size_read(inode)) {
+		mutex_lock(&inode->i_mutex);
+		if (pos > i_size_read(inode))
+			i_size_write(inode, pos);
+		mutex_unlock(&inode->i_mutex);
 	}
 	DPRINTK("ncp_file_write: exit %s/%s\n",
 		dentry->d_parent->d_name.name, dentry->d_name.name);
@@ -281,18 +279,9 @@ static int ncp_release(struct inode *inode, struct file *file) {
 	return 0;
 }
 
-static loff_t ncp_remote_llseek(struct file *file, loff_t offset, int origin)
-{
-	loff_t ret;
-	lock_kernel();
-	ret = generic_file_llseek_unlocked(file, offset, origin);
-	unlock_kernel();
-	return ret;
-}
-
 const struct file_operations ncp_file_operations =
 {
-	.llseek 	= ncp_remote_llseek,
+	.llseek		= generic_file_llseek,
 	.read		= ncp_file_read,
 	.write		= ncp_file_write,
 	.unlocked_ioctl	= ncp_ioctl,
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index b4de38cf49f5..5f4e58d93fdd 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -139,7 +139,7 @@ static void ncp_update_dates(struct inode *inode, struct nw_info_struct *nwi)
 		inode->i_mode = nwi->nfs.mode;
 	}
 
-	inode->i_blocks = (inode->i_size + NCP_BLOCK_SIZE - 1) >> NCP_BLOCK_SHIFT;
+	inode->i_blocks = (i_size_read(inode) + NCP_BLOCK_SIZE - 1) >> NCP_BLOCK_SHIFT;
 
 	inode->i_mtime.tv_sec = ncp_date_dos2unix(nwi->modifyTime, nwi->modifyDate);
 	inode->i_ctime.tv_sec = ncp_date_dos2unix(nwi->creationTime, nwi->creationDate);
@@ -158,18 +158,21 @@ static void ncp_update_attrs(struct inode *inode, struct ncp_entry_info *nwinfo)
 		inode->i_mode = server->m.dir_mode;
 		/* for directories dataStreamSize seems to be some
 		   Object ID ??? */
-		inode->i_size = NCP_BLOCK_SIZE;
+		i_size_write(inode, NCP_BLOCK_SIZE);
 	} else {
+		u32 size;
+
 		inode->i_mode = server->m.file_mode;
-		inode->i_size = le32_to_cpu(nwi->dataStreamSize);
+		size = le32_to_cpu(nwi->dataStreamSize);
+		i_size_write(inode, size);
 #ifdef CONFIG_NCPFS_EXTRAS
 		if ((server->m.flags & (NCP_MOUNT_EXTRAS|NCP_MOUNT_SYMLINKS)) 
 		 && (nwi->attributes & aSHARED)) {
 			switch (nwi->attributes & (aHIDDEN|aSYSTEM)) {
 				case aHIDDEN:
 					if (server->m.flags & NCP_MOUNT_SYMLINKS) {
-						if (/* (inode->i_size >= NCP_MIN_SYMLINK_SIZE)
-						 && */ (inode->i_size <= NCP_MAX_SYMLINK_SIZE)) {
+						if (/* (size >= NCP_MIN_SYMLINK_SIZE)
+						 && */ (size <= NCP_MAX_SYMLINK_SIZE)) {
 							inode->i_mode = (inode->i_mode & ~S_IFMT) | S_IFLNK;
 							NCP_FINFO(inode)->flags |= NCPI_KLUDGE_SYMLINK;
 							break;
@@ -208,7 +211,7 @@ void ncp_update_inode2(struct inode* inode, struct ncp_entry_info *nwinfo)
 }
 
 /*
- * Fill in the inode based on the ncp_entry_info structure.
+ * Fill in the inode based on the ncp_entry_info structure.  Used only for brand new inodes.
  */
 static void ncp_set_attr(struct inode *inode, struct ncp_entry_info *nwinfo)
 {
@@ -254,6 +257,7 @@ ncp_iget(struct super_block *sb, struct ncp_entry_info *info)
 	if (inode) {
 		atomic_set(&NCP_FINFO(inode)->opened, info->opened);
 
+		inode->i_mapping->backing_dev_info = sb->s_bdi;
 		inode->i_ino = info->ino;
 		ncp_set_attr(inode, info);
 		if (S_ISREG(inode->i_mode)) {
@@ -565,10 +569,12 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
 /*	server->conn_status = 0;	*/
 /*	server->root_dentry = NULL;	*/
 /*	server->root_setuped = 0;	*/
+	mutex_init(&server->root_setup_lock);
 #ifdef CONFIG_NCPFS_PACKET_SIGNING
 /*	server->sign_wanted = 0;	*/
 /*	server->sign_active = 0;	*/
 #endif
+	init_rwsem(&server->auth_rwsem);
 	server->auth.auth_type = NCP_AUTH_NONE;
 /*	server->auth.object_name_len = 0;	*/
 /*	server->auth.object_name = NULL;	*/
@@ -593,7 +599,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
 	server->nls_io = load_nls_default();
 #endif /* CONFIG_NCPFS_NLS */
 
-	server->dentry_ttl = 0;	/* no caching */
+	atomic_set(&server->dentry_ttl, 0);	/* no caching */
 
 	INIT_LIST_HEAD(&server->tx.requests);
 	mutex_init(&server->rcv.creq_mutex);
@@ -658,8 +664,10 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
 				goto out_disconnect;
 			}
 		}
+		ncp_lock_server(server);
 		if (options & 2)
 			server->sign_wanted = 1;
+		ncp_unlock_server(server);
 	}
 	else 
 #endif	/* CONFIG_NCPFS_PACKET_SIGNING */
@@ -720,6 +728,9 @@ out_nls:
 	unload_nls(server->nls_io);
 	unload_nls(server->nls_vol);
 #endif
+	mutex_destroy(&server->rcv.creq_mutex);
+	mutex_destroy(&server->root_setup_lock);
+	mutex_destroy(&server->mutex);
 out_fput2:
 	if (server->info_filp)
 		fput(server->info_filp);
@@ -743,8 +754,6 @@ static void ncp_put_super(struct super_block *sb)
 {
 	struct ncp_server *server = NCP_SBP(sb);
 
-	lock_kernel();
-
 	ncp_lock_server(server);
 	ncp_disconnect(server);
 	ncp_unlock_server(server);
@@ -756,6 +765,9 @@ static void ncp_put_super(struct super_block *sb)
 	unload_nls(server->nls_vol);
 	unload_nls(server->nls_io);
 #endif /* CONFIG_NCPFS_NLS */
+	mutex_destroy(&server->rcv.creq_mutex);
+	mutex_destroy(&server->root_setup_lock);
+	mutex_destroy(&server->mutex);
 
 	if (server->info_filp)
 		fput(server->info_filp);
@@ -771,8 +783,6 @@ static void ncp_put_super(struct super_block *sb)
 	vfree(server->packet);
 	sb->s_fs_info = NULL;
 	kfree(server);
-
-	unlock_kernel();
 }
 
 static int ncp_statfs(struct dentry *dentry, struct kstatfs *buf)
@@ -851,10 +861,8 @@ int ncp_notify_change(struct dentry *dentry, struct iattr *attr)
 
 	result = -EIO;
 
-	lock_kernel();	
-
 	server = NCP_SERVER(inode);
-	if ((!server) || !ncp_conn_valid(server))
+	if (!server)	/* How this could happen? */
 		goto out;
 
 	/* ageing the dentry to force validation */
@@ -981,8 +989,6 @@ int ncp_notify_change(struct dentry *dentry, struct iattr *attr)
 		result = ncp_modify_file_or_subdir_dos_info(NCP_SERVER(inode),
 				      inode, info_mask, &info);
 		if (result != 0) {
-			result = -EACCES;
-
 			if (info_mask == (DM_CREATE_TIME | DM_CREATE_DATE)) {
 				/* NetWare seems not to allow this. I
 				   do not know why. So, just tell the
@@ -1005,7 +1011,8 @@ int ncp_notify_change(struct dentry *dentry, struct iattr *attr)
 	mark_inode_dirty(inode);
 
 out:
-	unlock_kernel();
+	if (result > 0)
+		result = -EACCES;
 	return result;
 }
 
diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c
index 84a8cfc4e38e..c2a1f9a155c3 100644
--- a/fs/ncpfs/ioctl.c
+++ b/fs/ncpfs/ioctl.c
@@ -35,16 +35,11 @@
 #define NCP_PACKET_SIZE_INTERNAL 65536
 
 static int
-ncp_get_fs_info(struct ncp_server * server, struct file *file,
+ncp_get_fs_info(struct ncp_server * server, struct inode *inode,
 		struct ncp_fs_info __user *arg)
 {
-	struct inode *inode = file->f_path.dentry->d_inode;
 	struct ncp_fs_info info;
 
-	if (file_permission(file, MAY_WRITE) != 0
-	    && current_uid() != server->m.mounted_uid)
-		return -EACCES;
-
 	if (copy_from_user(&info, arg, sizeof(info)))
 		return -EFAULT;
 
@@ -65,16 +60,11 @@ ncp_get_fs_info(struct ncp_server * server, struct file *file,
 }
 
 static int
-ncp_get_fs_info_v2(struct ncp_server * server, struct file *file,
+ncp_get_fs_info_v2(struct ncp_server * server, struct inode *inode,
 		   struct ncp_fs_info_v2 __user * arg)
 {
-	struct inode *inode = file->f_path.dentry->d_inode;
 	struct ncp_fs_info_v2 info2;
 
-	if (file_permission(file, MAY_WRITE) != 0
-	    && current_uid() != server->m.mounted_uid)
-		return -EACCES;
-
 	if (copy_from_user(&info2, arg, sizeof(info2)))
 		return -EFAULT;
 
@@ -136,16 +126,11 @@ struct compat_ncp_privatedata_ioctl
 #define NCP_IOC_SETPRIVATEDATA_32	_IOR('n', 10, struct compat_ncp_privatedata_ioctl)
 
 static int
-ncp_get_compat_fs_info_v2(struct ncp_server * server, struct file *file,
+ncp_get_compat_fs_info_v2(struct ncp_server * server, struct inode *inode,
 		   struct compat_ncp_fs_info_v2 __user * arg)
 {
-	struct inode *inode = file->f_path.dentry->d_inode;
 	struct compat_ncp_fs_info_v2 info2;
 
-	if (file_permission(file, MAY_WRITE) != 0
-	    && current_uid() != server->m.mounted_uid)
-		return -EACCES;
-
 	if (copy_from_user(&info2, arg, sizeof(info2)))
 		return -EFAULT;
 
@@ -182,11 +167,8 @@ ncp_set_charsets(struct ncp_server* server, struct ncp_nls_ioctl __user *arg)
 	struct nls_table *iocharset;
 	struct nls_table *oldset_io;
 	struct nls_table *oldset_cp;
-
-	if (!capable(CAP_SYS_ADMIN))
-		return -EACCES;
-	if (server->root_setuped)
-		return -EBUSY;
+	int utf8;
+	int err;
 
 	if (copy_from_user(&user, arg, sizeof(user)))
 		return -EFAULT;
@@ -206,28 +188,40 @@ ncp_set_charsets(struct ncp_server* server, struct ncp_nls_ioctl __user *arg)
 	user.iocharset[NCP_IOCSNAME_LEN] = 0;
 	if (!user.iocharset[0] || !strcmp(user.iocharset, "default")) {
 		iocharset = load_nls_default();
-		NCP_CLR_FLAG(server, NCP_FLAG_UTF8);
+		utf8 = 0;
 	} else if (!strcmp(user.iocharset, "utf8")) {
 		iocharset = load_nls_default();
-		NCP_SET_FLAG(server, NCP_FLAG_UTF8);
+		utf8 = 1;
 	} else {
 		iocharset = load_nls(user.iocharset);
 		if (!iocharset) {
 			unload_nls(codepage);
 			return -EBADRQC;
 		}
-		NCP_CLR_FLAG(server, NCP_FLAG_UTF8);
+		utf8 = 0;
 	}
 
-	oldset_cp = server->nls_vol;
-	server->nls_vol = codepage;
-	oldset_io = server->nls_io;
-	server->nls_io = iocharset;
-
+	mutex_lock(&server->root_setup_lock);
+	if (server->root_setuped) {
+		oldset_cp = codepage;
+		oldset_io = iocharset;
+		err = -EBUSY;
+	} else {
+		if (utf8)
+			NCP_SET_FLAG(server, NCP_FLAG_UTF8);
+		else
+			NCP_CLR_FLAG(server, NCP_FLAG_UTF8);
+		oldset_cp = server->nls_vol;
+		server->nls_vol = codepage;
+		oldset_io = server->nls_io;
+		server->nls_io = iocharset;
+		err = 0;
+	}
+	mutex_unlock(&server->root_setup_lock);
 	unload_nls(oldset_cp);
 	unload_nls(oldset_io);
 
-	return 0;
+	return err;
 }
 
 static int
@@ -237,6 +231,7 @@ ncp_get_charsets(struct ncp_server* server, struct ncp_nls_ioctl __user *arg)
 	int len;
 
 	memset(&user, 0, sizeof(user));
+	mutex_lock(&server->root_setup_lock);
 	if (server->nls_vol && server->nls_vol->charset) {
 		len = strlen(server->nls_vol->charset);
 		if (len > NCP_IOCSNAME_LEN)
@@ -254,6 +249,7 @@ ncp_get_charsets(struct ncp_server* server, struct ncp_nls_ioctl __user *arg)
 		strncpy(user.iocharset,	server->nls_io->charset, len);
 		user.iocharset[len] = 0;
 	}
+	mutex_unlock(&server->root_setup_lock);
 
 	if (copy_to_user(arg, &user, sizeof(user)))
 		return -EFAULT;
@@ -261,25 +257,19 @@ ncp_get_charsets(struct ncp_server* server, struct ncp_nls_ioctl __user *arg)
 }
 #endif /* CONFIG_NCPFS_NLS */
 
-static long __ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
+static long __ncp_ioctl(struct inode *inode, unsigned int cmd, unsigned long arg)
 {
-	struct inode *inode = filp->f_dentry->d_inode;
 	struct ncp_server *server = NCP_SERVER(inode);
 	int result;
 	struct ncp_ioctl_request request;
 	char* bouncebuffer;
 	void __user *argp = (void __user *)arg;
-	uid_t uid = current_uid();
 
 	switch (cmd) {
 #ifdef CONFIG_COMPAT
 	case NCP_IOC_NCPREQUEST_32:
 #endif
 	case NCP_IOC_NCPREQUEST:
-		if (file_permission(filp, MAY_WRITE) != 0
-		    && uid != server->m.mounted_uid)
-			return -EACCES;
-
 #ifdef CONFIG_COMPAT
 		if (cmd == NCP_IOC_NCPREQUEST_32) {
 			struct compat_ncp_ioctl_request request32;
@@ -314,7 +304,7 @@ static long __ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		server->current_size = request.size;
 		memcpy(server->packet, bouncebuffer, request.size);
 
-		result = ncp_request2(server, request.function, 
+		result = ncp_request2(server, request.function,
 			bouncebuffer, NCP_PACKET_SIZE_INTERNAL);
 		if (result < 0)
 			result = -EIO;
@@ -331,69 +321,69 @@ static long __ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 
 	case NCP_IOC_CONN_LOGGED_IN:
 
-		if (!capable(CAP_SYS_ADMIN))
-			return -EACCES;
 		if (!(server->m.int_flags & NCP_IMOUNT_LOGGEDIN_POSSIBLE))
 			return -EINVAL;
+		mutex_lock(&server->root_setup_lock);
 		if (server->root_setuped)
-			return -EBUSY;
-		server->root_setuped = 1;
-		return ncp_conn_logged_in(inode->i_sb);
+			result = -EBUSY;
+		else {
+			result = ncp_conn_logged_in(inode->i_sb);
+			if (result == 0)
+				server->root_setuped = 1;
+		}
+		mutex_unlock(&server->root_setup_lock);
+		return result;
 
 	case NCP_IOC_GET_FS_INFO:
-		return ncp_get_fs_info(server, filp, argp);
+		return ncp_get_fs_info(server, inode, argp);
 
 	case NCP_IOC_GET_FS_INFO_V2:
-		return ncp_get_fs_info_v2(server, filp, argp);
+		return ncp_get_fs_info_v2(server, inode, argp);
 
 #ifdef CONFIG_COMPAT
 	case NCP_IOC_GET_FS_INFO_V2_32:
-		return ncp_get_compat_fs_info_v2(server, filp, argp);
+		return ncp_get_compat_fs_info_v2(server, inode, argp);
 #endif
 	/* we have too many combinations of CONFIG_COMPAT,
 	 * CONFIG_64BIT and CONFIG_UID16, so just handle
 	 * any of the possible ioctls */
 	case NCP_IOC_GETMOUNTUID16:
-	case NCP_IOC_GETMOUNTUID32:
-	case NCP_IOC_GETMOUNTUID64:
-		if (file_permission(filp, MAY_READ) != 0
-			&& uid != server->m.mounted_uid)
-			return -EACCES;
-
-		if (cmd == NCP_IOC_GETMOUNTUID16) {
+		{
 			u16 uid;
+
 			SET_UID(uid, server->m.mounted_uid);
 			if (put_user(uid, (u16 __user *)argp))
 				return -EFAULT;
-		} else if (cmd == NCP_IOC_GETMOUNTUID32) {
-			if (put_user(server->m.mounted_uid,
-						(u32 __user *)argp))
-				return -EFAULT;
-		} else {
-			if (put_user(server->m.mounted_uid,
-						(u64 __user *)argp))
-				return -EFAULT;
+			return 0;
 		}
+	case NCP_IOC_GETMOUNTUID32:
+		if (put_user(server->m.mounted_uid,
+			     (u32 __user *)argp))
+			return -EFAULT;
+		return 0;
+	case NCP_IOC_GETMOUNTUID64:
+		if (put_user(server->m.mounted_uid,
+			     (u64 __user *)argp))
+			return -EFAULT;
 		return 0;
 
 	case NCP_IOC_GETROOT:
 		{
 			struct ncp_setroot_ioctl sr;
 
-			if (file_permission(filp, MAY_READ) != 0
-			    && uid != server->m.mounted_uid)
-				return -EACCES;
-
+			result = -EACCES;
+			mutex_lock(&server->root_setup_lock);
 			if (server->m.mounted_vol[0]) {
 				struct dentry* dentry = inode->i_sb->s_root;
 
 				if (dentry) {
 					struct inode* s_inode = dentry->d_inode;
-				
+
 					if (s_inode) {
 						sr.volNumber = NCP_FINFO(s_inode)->volNumber;
 						sr.dirEntNum = NCP_FINFO(s_inode)->dirEntNum;
 						sr.namespace = server->name_space[sr.volNumber];
+						result = 0;
 					} else
 						DPRINTK("ncpfs: s_root->d_inode==NULL\n");
 				} else
@@ -402,10 +392,12 @@ static long __ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 				sr.volNumber = -1;
 				sr.namespace = 0;
 				sr.dirEntNum = 0;
+				result = 0;
 			}
-			if (copy_to_user(argp, &sr, sizeof(sr)))
-				return -EFAULT;
-			return 0;
+			mutex_unlock(&server->root_setup_lock);
+			if (!result && copy_to_user(argp, &sr, sizeof(sr)))
+				result = -EFAULT;
+			return result;
 		}
 
 	case NCP_IOC_SETROOT:
@@ -416,103 +408,114 @@ static long __ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 			__le32 dosde;
 			struct dentry* dentry;
 
-			if (!capable(CAP_SYS_ADMIN))
-			{
-				return -EACCES;
-			}
-			if (server->root_setuped) return -EBUSY;
 			if (copy_from_user(&sr, argp, sizeof(sr)))
 				return -EFAULT;
-			if (sr.volNumber < 0) {
-				server->m.mounted_vol[0] = 0;
-				vnum = NCP_NUMBER_OF_VOLUMES;
-				de = 0;
-				dosde = 0;
-			} else if (sr.volNumber >= NCP_NUMBER_OF_VOLUMES) {
-				return -EINVAL;
-			} else if (ncp_mount_subdir(server, sr.volNumber,
-						sr.namespace, sr.dirEntNum,
-						&vnum, &de, &dosde)) {
-				return -ENOENT;
-			}
-			
-			dentry = inode->i_sb->s_root;
-			server->root_setuped = 1;
-			if (dentry) {
-				struct inode* s_inode = dentry->d_inode;
-				
-				if (s_inode) {
-					NCP_FINFO(s_inode)->volNumber = vnum;
-					NCP_FINFO(s_inode)->dirEntNum = de;
-					NCP_FINFO(s_inode)->DosDirNum = dosde;
+			mutex_lock(&server->root_setup_lock);
+			if (server->root_setuped)
+				result = -EBUSY;
+			else {
+				if (sr.volNumber < 0) {
+					server->m.mounted_vol[0] = 0;
+					vnum = NCP_NUMBER_OF_VOLUMES;
+					de = 0;
+					dosde = 0;
+					result = 0;
+				} else if (sr.volNumber >= NCP_NUMBER_OF_VOLUMES) {
+					result = -EINVAL;
+				} else if (ncp_mount_subdir(server, sr.volNumber,
+							sr.namespace, sr.dirEntNum,
+							&vnum, &de, &dosde)) {
+					result = -ENOENT;
 				} else
-					DPRINTK("ncpfs: s_root->d_inode==NULL\n");
-			} else
-				DPRINTK("ncpfs: s_root==NULL\n");
+					result = 0;
+
+				if (result == 0) {
+					dentry = inode->i_sb->s_root;
+					if (dentry) {
+						struct inode* s_inode = dentry->d_inode;
+
+						if (s_inode) {
+							NCP_FINFO(s_inode)->volNumber = vnum;
+							NCP_FINFO(s_inode)->dirEntNum = de;
+							NCP_FINFO(s_inode)->DosDirNum = dosde;
+							server->root_setuped = 1;
+						} else {
+							DPRINTK("ncpfs: s_root->d_inode==NULL\n");
+							result = -EIO;
+						}
+					} else {
+						DPRINTK("ncpfs: s_root==NULL\n");
+						result = -EIO;
+					}
+				}
+				result = 0;
+			}
+			mutex_unlock(&server->root_setup_lock);
 
-			return 0;
+			return result;
 		}
 
-#ifdef CONFIG_NCPFS_PACKET_SIGNING	
+#ifdef CONFIG_NCPFS_PACKET_SIGNING
 	case NCP_IOC_SIGN_INIT:
-		if (file_permission(filp, MAY_WRITE) != 0
-		    && uid != server->m.mounted_uid)
-			return -EACCES;
-
-		if (argp) {
-			if (server->sign_wanted)
-			{
-				struct ncp_sign_init sign;
+		{
+			struct ncp_sign_init sign;
 
+			if (argp)
 				if (copy_from_user(&sign, argp, sizeof(sign)))
 					return -EFAULT;
-				memcpy(server->sign_root,sign.sign_root,8);
-				memcpy(server->sign_last,sign.sign_last,16);
-				server->sign_active = 1;
+			ncp_lock_server(server);
+			mutex_lock(&server->rcv.creq_mutex);
+			if (argp) {
+				if (server->sign_wanted) {
+					memcpy(server->sign_root,sign.sign_root,8);
+					memcpy(server->sign_last,sign.sign_last,16);
+					server->sign_active = 1;
+				}
+				/* ignore when signatures not wanted */
+			} else {
+				server->sign_active = 0;
 			}
-			/* ignore when signatures not wanted */
-		} else {
-			server->sign_active = 0;
+			mutex_unlock(&server->rcv.creq_mutex);
+			ncp_unlock_server(server);
+			return 0;
 		}
-		return 0;		
-		
+
         case NCP_IOC_SIGN_WANTED:
-		if (file_permission(filp, MAY_READ) != 0
-		    && uid != server->m.mounted_uid)
-			return -EACCES;
-		
-                if (put_user(server->sign_wanted, (int __user *)argp))
-			return -EFAULT;
-                return 0;
+		{
+			int state;
+
+			ncp_lock_server(server);
+			state = server->sign_wanted;
+			ncp_unlock_server(server);
+			if (put_user(state, (int __user *)argp))
+				return -EFAULT;
+			return 0;
+		}
 
 	case NCP_IOC_SET_SIGN_WANTED:
 		{
 			int newstate;
 
-			if (file_permission(filp, MAY_WRITE) != 0
-			    && uid != server->m.mounted_uid)
-				return -EACCES;
-
 			/* get only low 8 bits... */
 			if (get_user(newstate, (unsigned char __user *)argp))
 				return -EFAULT;
+			result = 0;
+			ncp_lock_server(server);
 			if (server->sign_active) {
 				/* cannot turn signatures OFF when active */
-				if (!newstate) return -EINVAL;
+				if (!newstate)
+					result = -EINVAL;
 			} else {
 				server->sign_wanted = newstate != 0;
 			}
-			return 0;
+			ncp_unlock_server(server);
+			return result;
 		}
 
 #endif /* CONFIG_NCPFS_PACKET_SIGNING */
 
 #ifdef CONFIG_NCPFS_IOCTL_LOCKING
 	case NCP_IOC_LOCKUNLOCK:
-		if (file_permission(filp, MAY_WRITE) != 0
-		    && uid != server->m.mounted_uid)
-			return -EACCES;
-
 		{
 			struct ncp_lock_ioctl	 rqdata;
 
@@ -541,16 +544,13 @@ static long __ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 			{
 				return result;
 			}
-			result = -EIO;
-			if (!ncp_conn_valid(server))
-				goto outrel;
 			result = -EISDIR;
 			if (!S_ISREG(inode->i_mode))
 				goto outrel;
 			if (rqdata.cmd == NCP_LOCK_CLEAR)
 			{
 				result = ncp_ClearPhysicalRecord(NCP_SERVER(inode),
-							NCP_FINFO(inode)->file_handle, 
+							NCP_FINFO(inode)->file_handle,
 							rqdata.offset,
 							rqdata.length);
 				if (result > 0) result = 0;	/* no such lock */
@@ -573,7 +573,7 @@ static long __ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 							rqdata.timeout);
 				if (result > 0) result = -EAGAIN;
 			}
-outrel:			
+outrel:
 			ncp_inode_close(inode);
 			return result;
 		}
@@ -581,60 +581,62 @@ outrel:
 
 #ifdef CONFIG_COMPAT
 	case NCP_IOC_GETOBJECTNAME_32:
-		if (uid != server->m.mounted_uid)
-			return -EACCES;
 		{
 			struct compat_ncp_objectname_ioctl user;
 			size_t outl;
 
 			if (copy_from_user(&user, argp, sizeof(user)))
 				return -EFAULT;
+			down_read(&server->auth_rwsem);
 			user.auth_type = server->auth.auth_type;
 			outl = user.object_name_len;
 			user.object_name_len = server->auth.object_name_len;
 			if (outl > user.object_name_len)
 				outl = user.object_name_len;
+			result = 0;
 			if (outl) {
 				if (copy_to_user(compat_ptr(user.object_name),
 						 server->auth.object_name,
-						 outl)) return -EFAULT;
+						 outl))
+					result = -EFAULT;
 			}
-			if (copy_to_user(argp, &user, sizeof(user)))
-				return -EFAULT;
-			return 0;
+			up_read(&server->auth_rwsem);
+			if (!result && copy_to_user(argp, &user, sizeof(user)))
+				result = -EFAULT;
+			return result;
 		}
 #endif
 
 	case NCP_IOC_GETOBJECTNAME:
-		if (uid != server->m.mounted_uid)
-			return -EACCES;
 		{
 			struct ncp_objectname_ioctl user;
 			size_t outl;
 
 			if (copy_from_user(&user, argp, sizeof(user)))
 				return -EFAULT;
+			down_read(&server->auth_rwsem);
 			user.auth_type = server->auth.auth_type;
 			outl = user.object_name_len;
 			user.object_name_len = server->auth.object_name_len;
 			if (outl > user.object_name_len)
 				outl = user.object_name_len;
+			result = 0;
 			if (outl) {
 				if (copy_to_user(user.object_name,
 						 server->auth.object_name,
-						 outl)) return -EFAULT;
+						 outl))
+					result = -EFAULT;
 			}
-			if (copy_to_user(argp, &user, sizeof(user)))
-				return -EFAULT;
-			return 0;
+			up_read(&server->auth_rwsem);
+			if (!result && copy_to_user(argp, &user, sizeof(user)))
+				result = -EFAULT;
+			return result;
 		}
 
 #ifdef CONFIG_COMPAT
 	case NCP_IOC_SETOBJECTNAME_32:
 #endif
 	case NCP_IOC_SETOBJECTNAME:
-		if (uid != server->m.mounted_uid)
-			return -EACCES;
 		{
 			struct ncp_objectname_ioctl user;
 			void* newname;
@@ -666,9 +668,7 @@ outrel:
 			} else {
 				newname = NULL;
 			}
-			/* enter critical section */
-			/* maybe that kfree can sleep so do that this way */
-			/* it is at least more SMP friendly (in future...) */
+			down_write(&server->auth_rwsem);
 			oldname = server->auth.object_name;
 			oldnamelen = server->auth.object_name_len;
 			oldprivate = server->priv.data;
@@ -678,7 +678,7 @@ outrel:
 			server->auth.object_name = newname;
 			server->priv.len = 0;
 			server->priv.data = NULL;
-			/* leave critical section */
+			up_write(&server->auth_rwsem);
 			kfree(oldprivate);
 			kfree(oldname);
 			return 0;
@@ -688,8 +688,6 @@ outrel:
 	case NCP_IOC_GETPRIVATEDATA_32:
 #endif
 	case NCP_IOC_GETPRIVATEDATA:
-		if (uid != server->m.mounted_uid)
-			return -EACCES;
 		{
 			struct ncp_privatedata_ioctl user;
 			size_t outl;
@@ -706,14 +704,20 @@ outrel:
 			if (copy_from_user(&user, argp, sizeof(user)))
 				return -EFAULT;
 
+			down_read(&server->auth_rwsem);
 			outl = user.len;
 			user.len = server->priv.len;
 			if (outl > user.len) outl = user.len;
+			result = 0;
 			if (outl) {
 				if (copy_to_user(user.data,
 						 server->priv.data,
-						 outl)) return -EFAULT;
+						 outl))
+					result = -EFAULT;
 			}
+			up_read(&server->auth_rwsem);
+			if (result)
+				return result;
 #ifdef CONFIG_COMPAT
 			if (cmd == NCP_IOC_GETPRIVATEDATA_32) {
 				struct compat_ncp_privatedata_ioctl user32;
@@ -733,8 +737,6 @@ outrel:
 	case NCP_IOC_SETPRIVATEDATA_32:
 #endif
 	case NCP_IOC_SETPRIVATEDATA:
-		if (uid != server->m.mounted_uid)
-			return -EACCES;
 		{
 			struct ncp_privatedata_ioctl user;
 			void* new;
@@ -762,12 +764,12 @@ outrel:
 			} else {
 				new = NULL;
 			}
-			/* enter critical section */
+			down_write(&server->auth_rwsem);
 			old = server->priv.data;
 			oldlen = server->priv.len;
 			server->priv.len = user.len;
 			server->priv.data = new;
-			/* leave critical section */
+			up_write(&server->auth_rwsem);
 			kfree(old);
 			return 0;
 		}
@@ -775,17 +777,13 @@ outrel:
 #ifdef CONFIG_NCPFS_NLS
 	case NCP_IOC_SETCHARSETS:
 		return ncp_set_charsets(server, argp);
-		
+
 	case NCP_IOC_GETCHARSETS:
 		return ncp_get_charsets(server, argp);
 
 #endif /* CONFIG_NCPFS_NLS */
 
 	case NCP_IOC_SETDENTRYTTL:
-		if (file_permission(filp, MAY_WRITE) != 0 &&
-		    uid != server->m.mounted_uid)
-			return -EACCES;
-
 		{
 			u_int32_t user;
 
@@ -795,13 +793,13 @@ outrel:
 			if (user > 20000)
 				return -EINVAL;
 			user = (user * HZ) / 1000;
-			server->dentry_ttl = user;
+			atomic_set(&server->dentry_ttl, user);
 			return 0;
 		}
-		
+
 	case NCP_IOC_GETDENTRYTTL:
 		{
-			u_int32_t user = (server->dentry_ttl * 1000) / HZ;
+			u_int32_t user = (atomic_read(&server->dentry_ttl) * 1000) / HZ;
 			if (copy_to_user(argp, &user, sizeof(user)))
 				return -EFAULT;
 			return 0;
@@ -811,59 +809,103 @@ outrel:
 	return -EINVAL;
 }
 
-static int ncp_ioctl_need_write(unsigned int cmd)
+long ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
+	struct inode *inode = filp->f_dentry->d_inode;
+	struct ncp_server *server = NCP_SERVER(inode);
+	uid_t uid = current_uid();
+	int need_drop_write = 0;
+	long ret;
+
 	switch (cmd) {
-	case NCP_IOC_GET_FS_INFO:
-	case NCP_IOC_GET_FS_INFO_V2:
-	case NCP_IOC_NCPREQUEST:
-	case NCP_IOC_SETDENTRYTTL:
-	case NCP_IOC_SIGN_INIT:
-	case NCP_IOC_LOCKUNLOCK:
-	case NCP_IOC_SET_SIGN_WANTED:
-		return 1;
-	case NCP_IOC_GETOBJECTNAME:
-	case NCP_IOC_SETOBJECTNAME:
-	case NCP_IOC_GETPRIVATEDATA:
-	case NCP_IOC_SETPRIVATEDATA:
 	case NCP_IOC_SETCHARSETS:
-	case NCP_IOC_GETCHARSETS:
 	case NCP_IOC_CONN_LOGGED_IN:
-	case NCP_IOC_GETDENTRYTTL:
-	case NCP_IOC_GETMOUNTUID2:
-	case NCP_IOC_SIGN_WANTED:
-	case NCP_IOC_GETROOT:
 	case NCP_IOC_SETROOT:
-		return 0;
-	default:
-		/* unknown IOCTL command, assume write */
-		return 1;
+		if (!capable(CAP_SYS_ADMIN)) {
+			ret = -EACCES;
+			goto out;
+		}
+		break;
 	}
-}
-
-long ncp_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
-{
-	long ret;
-
-	lock_kernel();
-	if (ncp_ioctl_need_write(cmd)) {
+	if (server->m.mounted_uid != uid) {
+		switch (cmd) {
 		/*
-		 * inside the ioctl(), any failures which
-		 * are because of file_permission() are
-		 * -EACCESS, so it seems consistent to keep
-		 *  that here.
+		 * Only mount owner can issue these ioctls.  Information
+		 * necessary to authenticate to other NDS servers are
+		 * stored here.
 		 */
-		if (mnt_want_write(filp->f_path.mnt)) {
+		case NCP_IOC_GETOBJECTNAME:
+		case NCP_IOC_SETOBJECTNAME:
+		case NCP_IOC_GETPRIVATEDATA:
+		case NCP_IOC_SETPRIVATEDATA:
+#ifdef CONFIG_COMPAT
+		case NCP_IOC_GETOBJECTNAME_32:
+		case NCP_IOC_SETOBJECTNAME_32:
+		case NCP_IOC_GETPRIVATEDATA_32:
+		case NCP_IOC_SETPRIVATEDATA_32:
+#endif
 			ret = -EACCES;
 			goto out;
+		/*
+		 * These require write access on the inode if user id
+		 * does not match.  Note that they do not write to the
+		 * file...  But old code did mnt_want_write, so I keep
+		 * it as is.  Of course not for mountpoint owner, as
+		 * that breaks read-only mounts altogether as ncpmount
+		 * needs working NCP_IOC_NCPREQUEST and
+		 * NCP_IOC_GET_FS_INFO.  Some of these codes (setdentryttl,
+		 * signinit, setsignwanted) should be probably restricted
+		 * to owner only, or even more to CAP_SYS_ADMIN).
+		 */
+		case NCP_IOC_GET_FS_INFO:
+		case NCP_IOC_GET_FS_INFO_V2:
+		case NCP_IOC_NCPREQUEST:
+		case NCP_IOC_SETDENTRYTTL:
+		case NCP_IOC_SIGN_INIT:
+		case NCP_IOC_LOCKUNLOCK:
+		case NCP_IOC_SET_SIGN_WANTED:
+#ifdef CONFIG_COMPAT
+		case NCP_IOC_GET_FS_INFO_V2_32:
+		case NCP_IOC_NCPREQUEST_32:
+#endif
+			ret = mnt_want_write_file(filp);
+			if (ret)
+				goto out;
+			need_drop_write = 1;
+			ret = inode_permission(inode, MAY_WRITE);
+			if (ret)
+				goto outDropWrite;
+			break;
+		/*
+		 * Read access required.
+		 */
+		case NCP_IOC_GETMOUNTUID16:
+		case NCP_IOC_GETMOUNTUID32:
+		case NCP_IOC_GETMOUNTUID64:
+		case NCP_IOC_GETROOT:
+		case NCP_IOC_SIGN_WANTED:
+			ret = inode_permission(inode, MAY_READ);
+			if (ret)
+				goto out;
+			break;
+		/*
+		 * Anybody can read these.
+		 */
+		case NCP_IOC_GETCHARSETS:
+		case NCP_IOC_GETDENTRYTTL:
+		default:
+		/* Three codes below are protected by CAP_SYS_ADMIN above. */
+		case NCP_IOC_SETCHARSETS:
+		case NCP_IOC_CONN_LOGGED_IN:
+		case NCP_IOC_SETROOT:
+			break;
 		}
 	}
-	ret = __ncp_ioctl(filp, cmd, arg);
-	if (ncp_ioctl_need_write(cmd))
+	ret = __ncp_ioctl(inode, cmd, arg);
+outDropWrite:
+	if (need_drop_write)
 		mnt_drop_write(filp->f_path.mnt);
-
 out:
-	unlock_kernel();
 	return ret;
 }
 
@@ -872,10 +914,8 @@ long ncp_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	long ret;
 
-	lock_kernel();
 	arg = (unsigned long) compat_ptr(arg);
 	ret = ncp_ioctl(file, cmd, arg);
-	unlock_kernel();
 	return ret;
 }
 #endif
diff --git a/fs/ncpfs/ncplib_kernel.c b/fs/ncpfs/ncplib_kernel.c
index 0ec6237a5970..a95615a0b6ac 100644
--- a/fs/ncpfs/ncplib_kernel.c
+++ b/fs/ncpfs/ncplib_kernel.c
@@ -107,17 +107,17 @@ ncp_reply_data(struct ncp_server *server, int offset)
 	return &(server->packet[sizeof(struct ncp_reply_header) + offset]);
 }
 
-static inline u8 BVAL(void *data)
+static inline u8 BVAL(const void *data)
 {
-	return *(u8 *)data;
+	return *(const u8 *)data;
 }
 
 static u8 ncp_reply_byte(struct ncp_server *server, int offset)
 {
-	return *(u8 *)ncp_reply_data(server, offset);
+	return *(const u8 *)ncp_reply_data(server, offset);
 }
 
-static inline u16 WVAL_LH(void *data)
+static inline u16 WVAL_LH(const void *data)
 {
 	return get_unaligned_le16(data);
 }
@@ -134,7 +134,7 @@ ncp_reply_be16(struct ncp_server *server, int offset)
 	return get_unaligned_be16(ncp_reply_data(server, offset));
 }
 
-static inline u32 DVAL_LH(void *data)
+static inline u32 DVAL_LH(const void *data)
 {
 	return get_unaligned_le32(data);
 }
@@ -349,9 +349,9 @@ int ncp_dirhandle_free(struct ncp_server* server, __u8 dirhandle) {
 	return result;
 }
 
-void ncp_extract_file_info(void *structure, struct nw_info_struct *target)
+void ncp_extract_file_info(const void *structure, struct nw_info_struct *target)
 {
-	__u8 *name_len;
+	const __u8 *name_len;
 	const int info_struct_size = offsetof(struct nw_info_struct, nameLen);
 
 	memcpy(target, structure, info_struct_size);
@@ -364,7 +364,7 @@ void ncp_extract_file_info(void *structure, struct nw_info_struct *target)
 }
 
 #ifdef CONFIG_NCPFS_NFS_NS
-static inline void ncp_extract_nfs_info(unsigned char *structure,
+static inline void ncp_extract_nfs_info(const unsigned char *structure,
 				 struct nw_nfs_info *target)
 {
 	target->mode = DVAL_LH(structure);
@@ -417,7 +417,7 @@ int ncp_obtain_nfs_info(struct ncp_server *server,
  * Returns information for a (one-component) name relative to
  * the specified directory.
  */
-int ncp_obtain_info(struct ncp_server *server, struct inode *dir, char *path,
+int ncp_obtain_info(struct ncp_server *server, struct inode *dir, const char *path,
 			struct nw_info_struct *target)
 {
 	__u8  volnum = NCP_FINFO(dir)->volNumber;
@@ -452,16 +452,16 @@ out:
 #ifdef CONFIG_NCPFS_NFS_NS
 static int
 ncp_obtain_DOS_dir_base(struct ncp_server *server,
-		__u8 volnum, __le32 dirent,
-		char *path, /* At most 1 component */
+		__u8 ns, __u8 volnum, __le32 dirent,
+		const char *path, /* At most 1 component */
 		__le32 *DOS_dir_base)
 {
 	int result;
 
 	ncp_init_request(server);
 	ncp_add_byte(server, 6); /* subfunction */
-	ncp_add_byte(server, server->name_space[volnum]);
-	ncp_add_byte(server, server->name_space[volnum]);
+	ncp_add_byte(server, ns);
+	ncp_add_byte(server, ns);
 	ncp_add_word(server, cpu_to_le16(0x8006)); /* get all */
 	ncp_add_dword(server, RIM_DIRECTORY);
 	ncp_add_handle_path(server, volnum, dirent, 1, path);
@@ -523,10 +523,27 @@ ncp_get_known_namespace(struct ncp_server *server, __u8 volume)
 #endif	/* defined(CONFIG_NCPFS_OS2_NS) || defined(CONFIG_NCPFS_NFS_NS) */
 }
 
+int
+ncp_update_known_namespace(struct ncp_server *server, __u8 volume, int *ret_ns)
+{
+	int ns = ncp_get_known_namespace(server, volume);
+
+	if (ret_ns)
+		*ret_ns = ns;
+
+	DPRINTK("lookup_vol: namespace[%d] = %d\n",
+		volume, server->name_space[volume]);
+
+	if (server->name_space[volume] == ns)
+		return 0;
+	server->name_space[volume] = ns;
+	return 1;
+}
+
 static int
 ncp_ObtainSpecificDirBase(struct ncp_server *server,
 		__u8 nsSrc, __u8 nsDst, __u8 vol_num, __le32 dir_base,
-		char *path, /* At most 1 component */
+		const char *path, /* At most 1 component */
 		__le32 *dirEntNum, __le32 *DosDirNum)
 {
 	int result;
@@ -560,14 +577,13 @@ ncp_mount_subdir(struct ncp_server *server,
 {
 	int dstNS;
 	int result;
-	
-	dstNS = ncp_get_known_namespace(server, volNumber);
+
+	ncp_update_known_namespace(server, volNumber, &dstNS);
 	if ((result = ncp_ObtainSpecificDirBase(server, srcNS, dstNS, volNumber, 
 				      dirEntNum, NULL, newDirEnt, newDosEnt)) != 0)
 	{
 		return result;
 	}
-	server->name_space[volNumber] = dstNS;
 	*volume = volNumber;
 	server->m.mounted_vol[1] = 0;
 	server->m.mounted_vol[0] = 'X';
@@ -575,11 +591,10 @@ ncp_mount_subdir(struct ncp_server *server,
 }
 
 int 
-ncp_get_volume_root(struct ncp_server *server, const char *volname,
-		    __u32* volume, __le32* dirent, __le32* dosdirent)
+ncp_get_volume_root(struct ncp_server *server,
+		    const char *volname, __u32* volume, __le32* dirent, __le32* dosdirent)
 {
 	int result;
-	__u8 volnum;
 
 	DPRINTK("ncp_get_volume_root: looking up vol %s\n", volname);
 
@@ -601,21 +616,14 @@ ncp_get_volume_root(struct ncp_server *server, const char *volname,
 		return result;
 	}
 	*dirent = *dosdirent = ncp_reply_dword(server, 4);
-	volnum = ncp_reply_byte(server, 8);
+	*volume = ncp_reply_byte(server, 8);
 	ncp_unlock_server(server);
-	*volume = volnum;
-
-	server->name_space[volnum] = ncp_get_known_namespace(server, volnum);
-
-	DPRINTK("lookup_vol: namespace[%d] = %d\n",
-		volnum, server->name_space[volnum]);
-
 	return 0;
 }
 
 int
-ncp_lookup_volume(struct ncp_server *server, const char *volname,
-		  struct nw_info_struct *target)
+ncp_lookup_volume(struct ncp_server *server,
+		  const char *volname, struct nw_info_struct *target)
 {
 	int result;
 
@@ -625,6 +633,7 @@ ncp_lookup_volume(struct ncp_server *server, const char *volname,
 	if (result) {
 		return result;
 	}
+	ncp_update_known_namespace(server, target->volNumber, NULL);
 	target->nameLen = strlen(volname);
 	memcpy(target->entryName, volname, target->nameLen+1);
 	target->attributes = aDIR;
@@ -676,8 +685,8 @@ int ncp_modify_nfs_info(struct ncp_server *server, __u8 volnum, __le32 dirent,
 {
 	int result = 0;
 
+	ncp_init_request(server);
 	if (server->name_space[volnum] == NW_NS_NFS) {
-		ncp_init_request(server);
 		ncp_add_byte(server, 25);	/* subfunction */
 		ncp_add_byte(server, server->name_space[volnum]);
 		ncp_add_byte(server, NW_NS_NFS);
@@ -690,8 +699,8 @@ int ncp_modify_nfs_info(struct ncp_server *server, __u8 volnum, __le32 dirent,
 		ncp_add_dword_lh(server, 1);	/* nlinks */
 		ncp_add_dword_lh(server, rdev);
 		result = ncp_request(server, 87);
-		ncp_unlock_server(server);
 	}
+	ncp_unlock_server(server);
 	return result;
 }
 #endif
@@ -700,7 +709,7 @@ int ncp_modify_nfs_info(struct ncp_server *server, __u8 volnum, __le32 dirent,
 static int
 ncp_DeleteNSEntry(struct ncp_server *server,
 		  __u8 have_dir_base, __u8 volnum, __le32 dirent,
-		  char* name, __u8 ns, __le16 attr)
+		  const char* name, __u8 ns, __le16 attr)
 {
 	int result;
 
@@ -734,23 +743,25 @@ ncp_del_file_or_subdir2(struct ncp_server *server,
 
 int
 ncp_del_file_or_subdir(struct ncp_server *server,
-		       struct inode *dir, char *name)
+		       struct inode *dir, const char *name)
 {
 	__u8  volnum = NCP_FINFO(dir)->volNumber;
 	__le32 dirent = NCP_FINFO(dir)->dirEntNum;
+	int name_space;
 
+	name_space = server->name_space[volnum];
 #ifdef CONFIG_NCPFS_NFS_NS
-	if (server->name_space[volnum]==NW_NS_NFS)
+	if (name_space == NW_NS_NFS)
  	{
  		int result;
  
- 		result=ncp_obtain_DOS_dir_base(server, volnum, dirent, name, &dirent);
+		result=ncp_obtain_DOS_dir_base(server, name_space, volnum, dirent, name, &dirent);
  		if (result) return result;
- 		return ncp_DeleteNSEntry(server, 1, volnum, dirent, NULL, NW_NS_DOS, cpu_to_le16(0x8006));
+		name = NULL;
+		name_space = NW_NS_DOS;
  	}
- 	else
 #endif	/* CONFIG_NCPFS_NFS_NS */
- 		return ncp_DeleteNSEntry(server, 1, volnum, dirent, name, server->name_space[volnum], cpu_to_le16(0x8006));
+	return ncp_DeleteNSEntry(server, 1, volnum, dirent, name, name_space, cpu_to_le16(0x8006));
 }
 
 static inline void ConvertToNWfromDWORD(__u16 v0, __u16 v1, __u8 ret[6])
@@ -765,7 +776,7 @@ static inline void ConvertToNWfromDWORD(__u16 v0, __u16 v1, __u8 ret[6])
 /* If both dir and name are NULL, then in target there's already a
    looked-up entry that wants to be opened. */
 int ncp_open_create_file_or_subdir(struct ncp_server *server,
-				   struct inode *dir, char *name,
+				   struct inode *dir, const char *name,
 				   int open_create_mode,
 				   __le32 create_attributes,
 				   __le16 desired_acc_rights,
@@ -890,8 +901,8 @@ int ncp_search_for_fileset(struct ncp_server *server,
 
 static int
 ncp_RenameNSEntry(struct ncp_server *server,
-		  struct inode *old_dir, char *old_name, __le16 old_type,
-		  struct inode *new_dir, char *new_name)
+		  struct inode *old_dir, const char *old_name, __le16 old_type,
+		  struct inode *new_dir, const char *new_name)
 {
 	int result = -EINVAL;
 
@@ -929,8 +940,8 @@ out:
 }
 
 int ncp_ren_or_mov_file_or_subdir(struct ncp_server *server,
-				struct inode *old_dir, char *old_name,
-				struct inode *new_dir, char *new_name)
+				struct inode *old_dir, const char *old_name,
+				struct inode *new_dir, const char *new_name)
 {
         int result;
         __le16 old_type = cpu_to_le16(0x06);
@@ -958,7 +969,7 @@ int
 ncp_read_kernel(struct ncp_server *server, const char *file_id,
 	     __u32 offset, __u16 to_read, char *target, int *bytes_read)
 {
-	char *source;
+	const char *source;
 	int result;
 
 	ncp_init_request(server);
diff --git a/fs/ncpfs/ncplib_kernel.h b/fs/ncpfs/ncplib_kernel.h
index 2441d1ab57dc..3c57eca634ce 100644
--- a/fs/ncpfs/ncplib_kernel.h
+++ b/fs/ncpfs/ncplib_kernel.h
@@ -65,10 +65,11 @@ static inline void ncp_inode_close(struct inode *inode) {
 	atomic_dec(&NCP_FINFO(inode)->opened);
 }
 
-void ncp_extract_file_info(void* src, struct nw_info_struct* target);
-int ncp_obtain_info(struct ncp_server *server, struct inode *, char *,
+void ncp_extract_file_info(const void* src, struct nw_info_struct* target);
+int ncp_obtain_info(struct ncp_server *server, struct inode *, const char *,
 		struct nw_info_struct *target);
 int ncp_obtain_nfs_info(struct ncp_server *server, struct nw_info_struct *target);
+int ncp_update_known_namespace(struct ncp_server *server, __u8 volume, int *ret_ns);
 int ncp_get_volume_root(struct ncp_server *server, const char *volname,
 			__u32 *volume, __le32 *dirent, __le32 *dosdirent);
 int ncp_lookup_volume(struct ncp_server *, const char *, struct nw_info_struct *);
@@ -80,8 +81,8 @@ int ncp_modify_nfs_info(struct ncp_server *, __u8 volnum, __le32 dirent,
 			__u32 mode, __u32 rdev);
 
 int ncp_del_file_or_subdir2(struct ncp_server *, struct dentry*);
-int ncp_del_file_or_subdir(struct ncp_server *, struct inode *, char *);
-int ncp_open_create_file_or_subdir(struct ncp_server *, struct inode *, char *,
+int ncp_del_file_or_subdir(struct ncp_server *, struct inode *, const char *);
+int ncp_open_create_file_or_subdir(struct ncp_server *, struct inode *, const char *,
 				int, __le32, __le16, struct ncp_entry_info *);
 
 int ncp_initialize_search(struct ncp_server *, struct inode *,
@@ -93,7 +94,7 @@ int ncp_search_for_fileset(struct ncp_server *server,
 			   char** rbuf, size_t* rsize);
 
 int ncp_ren_or_mov_file_or_subdir(struct ncp_server *server,
-			      struct inode *, char *, struct inode *, char *);
+			      struct inode *, const char *, struct inode *, const char *);
 
 
 int
@@ -170,13 +171,13 @@ static inline int ncp_strnicmp(struct nls_table *t, const unsigned char *s1,
 #endif /* CONFIG_NCPFS_NLS */
 
 #define NCP_GET_AGE(dentry)	(jiffies - (dentry)->d_time)
-#define NCP_MAX_AGE(server)	((server)->dentry_ttl)
+#define NCP_MAX_AGE(server)	atomic_read(&(server)->dentry_ttl)
 #define NCP_TEST_AGE(server,dentry)	(NCP_GET_AGE(dentry) < NCP_MAX_AGE(server))
 
 static inline void
 ncp_age_dentry(struct ncp_server* server, struct dentry* dentry)
 {
-	dentry->d_time = jiffies - server->dentry_ttl;
+	dentry->d_time = jiffies - NCP_MAX_AGE(server);
 }
 
 static inline void
diff --git a/fs/ncpfs/ncpsign_kernel.c b/fs/ncpfs/ncpsign_kernel.c
index 7c0b5c21e6cf..d8b2d7e6910b 100644
--- a/fs/ncpfs/ncpsign_kernel.c
+++ b/fs/ncpfs/ncpsign_kernel.c
@@ -15,21 +15,21 @@
 
 /* i386: 32-bit, little endian, handles mis-alignment */
 #ifdef __i386__
-#define GET_LE32(p) (*(int *)(p))
+#define GET_LE32(p) (*(const int *)(p))
 #define PUT_LE32(p,v) { *(int *)(p)=v; }
 #else
 /* from include/ncplib.h */
-#define BVAL(buf,pos) (((__u8 *)(buf))[pos])
+#define BVAL(buf,pos) (((const __u8 *)(buf))[pos])
 #define PVAL(buf,pos) ((unsigned)BVAL(buf,pos))
-#define BSET(buf,pos,val) (BVAL(buf,pos) = (val))
+#define BSET(buf,pos,val) (((__u8 *)(buf))[pos] = (val))
 
 static inline __u16
-WVAL_LH(__u8 * buf, int pos)
+WVAL_LH(const __u8 * buf, int pos)
 {
 	return PVAL(buf, pos) | PVAL(buf, pos + 1) << 8;
 }
 static inline __u32
-DVAL_LH(__u8 * buf, int pos)
+DVAL_LH(const __u8 * buf, int pos)
 {
 	return WVAL_LH(buf, pos) | WVAL_LH(buf, pos + 2) << 16;
 }
diff --git a/fs/ncpfs/sock.c b/fs/ncpfs/sock.c
index c7ff6c700a6e..668bd267346e 100644
--- a/fs/ncpfs/sock.c
+++ b/fs/ncpfs/sock.c
@@ -746,7 +746,6 @@ static int ncp_do_request(struct ncp_server *server, int size,
 		return -EIO;
 	}
 	if (!ncp_conn_valid(server)) {
-		printk(KERN_ERR "ncpfs: Connection invalid!\n");
 		return -EIO;
 	}
 	{
diff --git a/include/linux/ncp_fs.h b/include/linux/ncp_fs.h
index 4522aed00906..ef663061d5ac 100644
--- a/include/linux/ncp_fs.h
+++ b/include/linux/ncp_fs.h
@@ -241,34 +241,6 @@ int ncp_mmap(struct file *, struct vm_area_struct *);
 /* linux/fs/ncpfs/ncplib_kernel.c */
 int ncp_make_closed(struct inode *);
 
-#define ncp_namespace(i)	(NCP_SERVER(i)->name_space[NCP_FINFO(i)->volNumber])
-
-static inline int ncp_preserve_entry_case(struct inode *i, __u32 nscreator)
-{
-#ifdef CONFIG_NCPFS_SMALLDOS
-	int ns = ncp_namespace(i);
-
-	if ((ns == NW_NS_DOS)
-#ifdef CONFIG_NCPFS_OS2_NS
-		|| ((ns == NW_NS_OS2) && (nscreator == NW_NS_DOS))
-#endif /* CONFIG_NCPFS_OS2_NS */
-				)
-		return 0;
-#endif /* CONFIG_NCPFS_SMALLDOS */
-	return 1;
-}
-
-#define ncp_preserve_case(i)	(ncp_namespace(i) != NW_NS_DOS)
-
-static inline int ncp_case_sensitive(struct inode *i)
-{
-#ifdef CONFIG_NCPFS_NFS_NS
-	return ncp_namespace(i) == NW_NS_NFS;
-#else
-	return 0;
-#endif	/* CONFIG_NCPFS_NFS_NS */
-} 
-
 #endif				/* __KERNEL__ */
 
 #endif				/* _LINUX_NCP_FS_H */
diff --git a/include/linux/ncp_fs_sb.h b/include/linux/ncp_fs_sb.h
index 8da05bc098ca..d64b0e894336 100644
--- a/include/linux/ncp_fs_sb.h
+++ b/include/linux/ncp_fs_sb.h
@@ -62,6 +62,7 @@ struct ncp_server {
 	int ncp_reply_size;
 
 	int root_setuped;
+	struct mutex root_setup_lock;
 
 	/* info for packet signing */
 	int sign_wanted;	/* 1=Server needs signed packets */
@@ -81,13 +82,14 @@ struct ncp_server {
 		size_t	len;
 		void*	data;
 	} priv;
+	struct rw_semaphore auth_rwsem;
 
 	/* nls info: codepage for volume and charset for I/O */
 	struct nls_table *nls_vol;
 	struct nls_table *nls_io;
 
 	/* maximum age in jiffies */
-	int dentry_ttl;
+	atomic_t dentry_ttl;
 
 	/* miscellaneous */
 	unsigned int flags;
-- 
cgit v1.2.3


From 24824a09e35402b8d58dcc5be803a5ad3937bdba Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Sat, 2 Oct 2010 06:11:55 +0000
Subject: net: dynamic ingress_queue allocation

ingress being not used very much, and net_device->ingress_queue being
quite a big object (128 or 256 bytes), use a dynamic allocation if
needed (tc qdisc add dev eth0 ingress ...)

dev_ingress_queue(dev) helper should be used only with RTNL taken.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  2 +-
 include/linux/rtnetlink.h |  8 ++++++++
 net/core/dev.c            | 34 ++++++++++++++++++++++++++--------
 net/sched/sch_api.c       | 42 ++++++++++++++++++++++++++++--------------
 net/sched/sch_generic.c   | 12 ++++++++----
 5 files changed, 71 insertions(+), 27 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ceed3474014a..92d81edd5808 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -986,7 +986,7 @@ struct net_device {
 	rx_handler_func_t	*rx_handler;
 	void			*rx_handler_data;
 
-	struct netdev_queue	ingress_queue; /* use two cache lines */
+	struct netdev_queue __rcu *ingress_queue;
 
 /*
  * Cache lines mostly used on transmit path
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 68c436bddc88..0bb7b48632bd 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -6,6 +6,7 @@
 #include <linux/if_link.h>
 #include <linux/if_addr.h>
 #include <linux/neighbour.h>
+#include <linux/netdevice.h>
 
 /* rtnetlink families. Values up to 127 are reserved for real address
  * families, values above 128 may be used arbitrarily.
@@ -769,6 +770,13 @@ extern int lockdep_rtnl_is_held(void);
 #define rtnl_dereference(p)					\
 	rcu_dereference_check(p, lockdep_rtnl_is_held())
 
+static inline struct netdev_queue *dev_ingress_queue(struct net_device *dev)
+{
+	return rtnl_dereference(dev->ingress_queue);
+}
+
+extern struct netdev_queue *dev_ingress_queue_create(struct net_device *dev);
+
 extern void rtnetlink_init(void);
 extern void __rtnl_unlock(void);
 
diff --git a/net/core/dev.c b/net/core/dev.c
index a313bab1b754..ce6ad88c980b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2702,11 +2702,10 @@ EXPORT_SYMBOL_GPL(br_fdb_test_addr_hook);
  * the ingress scheduler, you just cant add policies on ingress.
  *
  */
-static int ing_filter(struct sk_buff *skb)
+static int ing_filter(struct sk_buff *skb, struct netdev_queue *rxq)
 {
 	struct net_device *dev = skb->dev;
 	u32 ttl = G_TC_RTTL(skb->tc_verd);
-	struct netdev_queue *rxq;
 	int result = TC_ACT_OK;
 	struct Qdisc *q;
 
@@ -2720,8 +2719,6 @@ static int ing_filter(struct sk_buff *skb)
 	skb->tc_verd = SET_TC_RTTL(skb->tc_verd, ttl);
 	skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS);
 
-	rxq = &dev->ingress_queue;
-
 	q = rxq->qdisc;
 	if (q != &noop_qdisc) {
 		spin_lock(qdisc_lock(q));
@@ -2737,7 +2734,9 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb,
 					 struct packet_type **pt_prev,
 					 int *ret, struct net_device *orig_dev)
 {
-	if (skb->dev->ingress_queue.qdisc == &noop_qdisc)
+	struct netdev_queue *rxq = rcu_dereference(skb->dev->ingress_queue);
+
+	if (!rxq || rxq->qdisc == &noop_qdisc)
 		goto out;
 
 	if (*pt_prev) {
@@ -2745,7 +2744,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb,
 		*pt_prev = NULL;
 	}
 
-	switch (ing_filter(skb)) {
+	switch (ing_filter(skb, rxq)) {
 	case TC_ACT_SHOT:
 	case TC_ACT_STOLEN:
 		kfree_skb(skb);
@@ -4940,7 +4939,6 @@ static void __netdev_init_queue_locks_one(struct net_device *dev,
 static void netdev_init_queue_locks(struct net_device *dev)
 {
 	netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL);
-	__netdev_init_queue_locks_one(dev, &dev->ingress_queue, NULL);
 }
 
 unsigned long netdev_fix_features(unsigned long features, const char *name)
@@ -5452,11 +5450,29 @@ static void netdev_init_one_queue(struct net_device *dev,
 
 static void netdev_init_queues(struct net_device *dev)
 {
-	netdev_init_one_queue(dev, &dev->ingress_queue, NULL);
 	netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL);
 	spin_lock_init(&dev->tx_global_lock);
 }
 
+struct netdev_queue *dev_ingress_queue_create(struct net_device *dev)
+{
+	struct netdev_queue *queue = dev_ingress_queue(dev);
+
+#ifdef CONFIG_NET_CLS_ACT
+	if (queue)
+		return queue;
+	queue = kzalloc(sizeof(*queue), GFP_KERNEL);
+	if (!queue)
+		return NULL;
+	netdev_init_one_queue(dev, queue, NULL);
+	__netdev_init_queue_locks_one(dev, queue, NULL);
+	queue->qdisc = &noop_qdisc;
+	queue->qdisc_sleeping = &noop_qdisc;
+	rcu_assign_pointer(dev->ingress_queue, queue);
+#endif
+	return queue;
+}
+
 /**
  *	alloc_netdev_mq - allocate network device
  *	@sizeof_priv:	size of private data to allocate space for
@@ -5559,6 +5575,8 @@ void free_netdev(struct net_device *dev)
 
 	kfree(dev->_tx);
 
+	kfree(rcu_dereference_raw(dev->ingress_queue));
+
 	/* Flush device addresses */
 	dev_addr_flush(dev);
 
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index b8020784d0e9..b22ca2d1cebc 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -240,7 +240,10 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 	if (q)
 		goto out;
 
-	q = qdisc_match_from_root(dev->ingress_queue.qdisc_sleeping, handle);
+	if (dev_ingress_queue(dev))
+		q = qdisc_match_from_root(
+			dev_ingress_queue(dev)->qdisc_sleeping,
+			handle);
 out:
 	return q;
 }
@@ -690,6 +693,8 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent,
 		    (new && new->flags & TCQ_F_INGRESS)) {
 			num_q = 1;
 			ingress = 1;
+			if (!dev_ingress_queue(dev))
+				return -ENOENT;
 		}
 
 		if (dev->flags & IFF_UP)
@@ -701,7 +706,7 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent,
 		}
 
 		for (i = 0; i < num_q; i++) {
-			struct netdev_queue *dev_queue = &dev->ingress_queue;
+			struct netdev_queue *dev_queue = dev_ingress_queue(dev);
 
 			if (!ingress)
 				dev_queue = netdev_get_tx_queue(dev, i);
@@ -979,7 +984,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 					return -ENOENT;
 				q = qdisc_leaf(p, clid);
 			} else { /* ingress */
-				q = dev->ingress_queue.qdisc_sleeping;
+				if (dev_ingress_queue(dev))
+					q = dev_ingress_queue(dev)->qdisc_sleeping;
 			}
 		} else {
 			q = dev->qdisc;
@@ -1043,8 +1049,9 @@ replay:
 				if ((p = qdisc_lookup(dev, TC_H_MAJ(clid))) == NULL)
 					return -ENOENT;
 				q = qdisc_leaf(p, clid);
-			} else { /*ingress */
-				q = dev->ingress_queue.qdisc_sleeping;
+			} else { /* ingress */
+				if (dev_ingress_queue_create(dev))
+					q = dev_ingress_queue(dev)->qdisc_sleeping;
 			}
 		} else {
 			q = dev->qdisc;
@@ -1123,11 +1130,14 @@ replay:
 create_n_graft:
 	if (!(n->nlmsg_flags&NLM_F_CREATE))
 		return -ENOENT;
-	if (clid == TC_H_INGRESS)
-		q = qdisc_create(dev, &dev->ingress_queue, p,
-				 tcm->tcm_parent, tcm->tcm_parent,
-				 tca, &err);
-	else {
+	if (clid == TC_H_INGRESS) {
+		if (dev_ingress_queue(dev))
+			q = qdisc_create(dev, dev_ingress_queue(dev), p,
+					 tcm->tcm_parent, tcm->tcm_parent,
+					 tca, &err);
+		else
+			err = -ENOENT;
+	} else {
 		struct netdev_queue *dev_queue;
 
 		if (p && p->ops->cl_ops && p->ops->cl_ops->select_queue)
@@ -1304,8 +1314,10 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 		if (tc_dump_qdisc_root(dev->qdisc, skb, cb, &q_idx, s_q_idx) < 0)
 			goto done;
 
-		dev_queue = &dev->ingress_queue;
-		if (tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, &q_idx, s_q_idx) < 0)
+		dev_queue = dev_ingress_queue(dev);
+		if (dev_queue &&
+		    tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb,
+				       &q_idx, s_q_idx) < 0)
 			goto done;
 
 cont:
@@ -1595,8 +1607,10 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	if (tc_dump_tclass_root(dev->qdisc, skb, tcm, cb, &t, s_t) < 0)
 		goto done;
 
-	dev_queue = &dev->ingress_queue;
-	if (tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, &t, s_t) < 0)
+	dev_queue = dev_ingress_queue(dev);
+	if (dev_queue &&
+	    tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb,
+				&t, s_t) < 0)
 		goto done;
 
 done:
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 545278a1c478..3d57681bdb76 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -753,7 +753,8 @@ void dev_activate(struct net_device *dev)
 
 	need_watchdog = 0;
 	netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog);
-	transition_one_qdisc(dev, &dev->ingress_queue, NULL);
+	if (dev_ingress_queue(dev))
+		transition_one_qdisc(dev, dev_ingress_queue(dev), NULL);
 
 	if (need_watchdog) {
 		dev->trans_start = jiffies;
@@ -812,7 +813,8 @@ static bool some_qdisc_is_busy(struct net_device *dev)
 void dev_deactivate(struct net_device *dev)
 {
 	netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc);
-	dev_deactivate_queue(dev, &dev->ingress_queue, &noop_qdisc);
+	if (dev_ingress_queue(dev))
+		dev_deactivate_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
 
 	dev_watchdog_down(dev);
 
@@ -838,7 +840,8 @@ void dev_init_scheduler(struct net_device *dev)
 {
 	dev->qdisc = &noop_qdisc;
 	netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc);
-	dev_init_scheduler_queue(dev, &dev->ingress_queue, &noop_qdisc);
+	if (dev_ingress_queue(dev))
+		dev_init_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
 
 	setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev);
 }
@@ -861,7 +864,8 @@ static void shutdown_scheduler_queue(struct net_device *dev,
 void dev_shutdown(struct net_device *dev)
 {
 	netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
-	shutdown_scheduler_queue(dev, &dev->ingress_queue, &noop_qdisc);
+	if (dev_ingress_queue(dev))
+		shutdown_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc);
 	qdisc_destroy(dev->qdisc);
 	dev->qdisc = &noop_qdisc;
 
-- 
cgit v1.2.3


From 29fa060eab3f524d338566d34c1d9e704579ae5e Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 5 Oct 2010 00:29:48 -0700
Subject: net: relax rtnl_dereference()

rtnl_dereference() is used in contexts where RTNL is held, to fetch an
RCU protected pointer.

Updates to this pointer are prevented by RTNL, so we dont need
smp_read_barrier_depends() and the ACCESS_ONCE() provided in
rcu_dereference_check().

rtnl_dereference() is mainly a macro to document the locking invariant.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 0bb7b48632bd..d42f274418b8 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -755,20 +755,22 @@ extern int lockdep_rtnl_is_held(void);
  * @p: The pointer to read, prior to dereferencing
  *
  * Do an rcu_dereference(p), but check caller either holds rcu_read_lock()
- * or RTNL
+ * or RTNL. Note : Please prefer rtnl_dereference() or rcu_dereference()
  */
 #define rcu_dereference_rtnl(p)					\
 	rcu_dereference_check(p, rcu_read_lock_held() ||	\
 				 lockdep_rtnl_is_held())
 
 /**
- * rtnl_dereference - rcu_dereference with debug checking
+ * rtnl_dereference - fetch RCU pointer when updates are prevented by RTNL
  * @p: The pointer to read, prior to dereferencing
  *
- * Do an rcu_dereference(p), but check caller holds RTNL
+ * Return the value of the specified RCU-protected pointer, but omit
+ * both the smp_read_barrier_depends() and the ACCESS_ONCE(), because
+ * caller holds RTNL.
  */
 #define rtnl_dereference(p)					\
-	rcu_dereference_check(p, lockdep_rtnl_is_held())
+	rcu_dereference_protected(p, lockdep_rtnl_is_held())
 
 static inline struct netdev_queue *dev_ingress_queue(struct net_device *dev)
 {
-- 
cgit v1.2.3


From b89f432133851a01c0d28822f11cbdcc15781a75 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sat, 18 Sep 2010 15:09:31 +0200
Subject: fs/locks.c: prepare for BKL removal

This prepares the removal of the big kernel lock from the
file locking code. We still use the BKL as long as fs/lockd
uses it and ceph might sleep, but we can flip the definition
to a private spinlock as soon as that's done.
All users outside of fs/lockd get converted to use
lock_flocks() instead of lock_kernel() where appropriate.

Based on an earlier patch to use a spinlock from Matthew
Wilcox, who has attempted this a few times before, the
earliest patch from over 10 years ago turned it into
a semaphore, which ended up being slower than the BKL
and was subsequently reverted.

Someone should do some serious performance testing when
this becomes a spinlock, since this has caused problems
before. Using a spinlock should be at least as good
as the BKL in theory, but who knows...

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Miklos Szeredi <mszeredi@suse.cz>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Sage Weil <sage@newdream.net>
Cc: linux-kernel@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org
---
 fs/afs/flock.c      |   5 +--
 fs/cifs/cifsfs.c    |   4 +-
 fs/gfs2/file.c      |   2 +
 fs/locks.c          | 112 +++++++++++++++++++++++++++++++---------------------
 fs/nfs/delegation.c |  10 ++---
 fs/nfs/nfs4state.c  |  10 ++---
 fs/nfsd/nfs4state.c |   6 +--
 include/linux/fs.h  |  14 +++++--
 8 files changed, 97 insertions(+), 66 deletions(-)

(limited to 'include/linux')

diff --git a/fs/afs/flock.c b/fs/afs/flock.c
index 0931bc1325eb..757d664575dd 100644
--- a/fs/afs/flock.c
+++ b/fs/afs/flock.c
@@ -9,7 +9,6 @@
  * 2 of the License, or (at your option) any later version.
  */
 
-#include <linux/smp_lock.h>
 #include "internal.h"
 
 #define AFS_LOCK_GRANTED	0
@@ -274,7 +273,7 @@ static int afs_do_setlk(struct file *file, struct file_lock *fl)
 
 	type = (fl->fl_type == F_RDLCK) ? AFS_LOCK_READ : AFS_LOCK_WRITE;
 
-	lock_kernel();
+	lock_flocks();
 
 	/* make sure we've got a callback on this file and that our view of the
 	 * data version is up to date */
@@ -421,7 +420,7 @@ given_lock:
 	afs_vnode_fetch_status(vnode, NULL, key);
 
 error:
-	unlock_kernel();
+	unlock_flocks();
 	_leave(" = %d", ret);
 	return ret;
 
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 4e273f7793f6..50208c15309a 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -562,8 +562,8 @@ static loff_t cifs_llseek(struct file *file, loff_t offset, int origin)
 
 static int cifs_setlease(struct file *file, long arg, struct file_lock **lease)
 {
-	/* note that this is called by vfs setlease with the BKL held
-	   although I doubt that BKL is needed here in cifs */
+	/* note that this is called by vfs setlease with lock_flocks held
+	   to protect *lease from going away */
 	struct inode *inode = file->f_path.dentry->d_inode;
 
 	if (!(S_ISREG(inode->i_mode)))
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 4edd662c8232..8fcfefb96077 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -620,6 +620,8 @@ static ssize_t gfs2_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
  * cluster; until we do, disable leases (by just returning -EINVAL),
  * unless the administrator has requested purely local locking.
  *
+ * Locking: called under lock_flocks
+ *
  * Returns: errno
  */
 
diff --git a/fs/locks.c b/fs/locks.c
index ab24d49fc048..8b2b6ad56a09 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -143,6 +143,22 @@ int lease_break_time = 45;
 static LIST_HEAD(file_lock_list);
 static LIST_HEAD(blocked_list);
 
+/*
+ * Protects the two list heads above, plus the inode->i_flock list
+ * FIXME: should use a spinlock, once lockd and ceph are ready.
+ */
+void lock_flocks(void)
+{
+	lock_kernel();
+}
+EXPORT_SYMBOL_GPL(lock_flocks);
+
+void unlock_flocks(void)
+{
+	unlock_kernel();
+}
+EXPORT_SYMBOL_GPL(unlock_flocks);
+
 static struct kmem_cache *filelock_cache __read_mostly;
 
 /* Allocate an empty lock structure. */
@@ -511,9 +527,9 @@ static void __locks_delete_block(struct file_lock *waiter)
  */
 static void locks_delete_block(struct file_lock *waiter)
 {
-	lock_kernel();
+	lock_flocks();
 	__locks_delete_block(waiter);
-	unlock_kernel();
+	unlock_flocks();
 }
 
 /* Insert waiter into blocker's block list.
@@ -644,7 +660,7 @@ posix_test_lock(struct file *filp, struct file_lock *fl)
 {
 	struct file_lock *cfl;
 
-	lock_kernel();
+	lock_flocks();
 	for (cfl = filp->f_path.dentry->d_inode->i_flock; cfl; cfl = cfl->fl_next) {
 		if (!IS_POSIX(cfl))
 			continue;
@@ -657,7 +673,7 @@ posix_test_lock(struct file *filp, struct file_lock *fl)
 			fl->fl_pid = pid_vnr(cfl->fl_nspid);
 	} else
 		fl->fl_type = F_UNLCK;
-	unlock_kernel();
+	unlock_flocks();
 	return;
 }
 EXPORT_SYMBOL(posix_test_lock);
@@ -730,18 +746,16 @@ static int flock_lock_file(struct file *filp, struct file_lock *request)
 	int error = 0;
 	int found = 0;
 
-	lock_kernel();
-	if (request->fl_flags & FL_ACCESS)
-		goto find_conflict;
-
-	if (request->fl_type != F_UNLCK) {
-		error = -ENOMEM;
+	if (!(request->fl_flags & FL_ACCESS) && (request->fl_type != F_UNLCK)) {
 		new_fl = locks_alloc_lock();
-		if (new_fl == NULL)
-			goto out;
-		error = 0;
+		if (!new_fl)
+			return -ENOMEM;
 	}
 
+	lock_flocks();
+	if (request->fl_flags & FL_ACCESS)
+		goto find_conflict;
+
 	for_each_lock(inode, before) {
 		struct file_lock *fl = *before;
 		if (IS_POSIX(fl))
@@ -767,8 +781,11 @@ static int flock_lock_file(struct file *filp, struct file_lock *request)
 	 * If a higher-priority process was blocked on the old file lock,
 	 * give it the opportunity to lock the file.
 	 */
-	if (found)
+	if (found) {
+		unlock_flocks();
 		cond_resched();
+		lock_flocks();
+	}
 
 find_conflict:
 	for_each_lock(inode, before) {
@@ -794,7 +811,7 @@ find_conflict:
 	error = 0;
 
 out:
-	unlock_kernel();
+	unlock_flocks();
 	if (new_fl)
 		locks_free_lock(new_fl);
 	return error;
@@ -823,7 +840,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str
 		new_fl2 = locks_alloc_lock();
 	}
 
-	lock_kernel();
+	lock_flocks();
 	if (request->fl_type != F_UNLCK) {
 		for_each_lock(inode, before) {
 			fl = *before;
@@ -991,7 +1008,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str
 		locks_wake_up_blocks(left);
 	}
  out:
-	unlock_kernel();
+	unlock_flocks();
 	/*
 	 * Free any unused locks.
 	 */
@@ -1066,14 +1083,14 @@ int locks_mandatory_locked(struct inode *inode)
 	/*
 	 * Search the lock list for this inode for any POSIX locks.
 	 */
-	lock_kernel();
+	lock_flocks();
 	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
 		if (!IS_POSIX(fl))
 			continue;
 		if (fl->fl_owner != owner)
 			break;
 	}
-	unlock_kernel();
+	unlock_flocks();
 	return fl ? -EAGAIN : 0;
 }
 
@@ -1186,7 +1203,7 @@ int __break_lease(struct inode *inode, unsigned int mode)
 
 	new_fl = lease_alloc(NULL, want_write ? F_WRLCK : F_RDLCK);
 
-	lock_kernel();
+	lock_flocks();
 
 	time_out_leases(inode);
 
@@ -1247,8 +1264,10 @@ restart:
 			break_time++;
 	}
 	locks_insert_block(flock, new_fl);
+	unlock_flocks();
 	error = wait_event_interruptible_timeout(new_fl->fl_wait,
 						!new_fl->fl_next, break_time);
+	lock_flocks();
 	__locks_delete_block(new_fl);
 	if (error >= 0) {
 		if (error == 0)
@@ -1263,7 +1282,7 @@ restart:
 	}
 
 out:
-	unlock_kernel();
+	unlock_flocks();
 	if (!IS_ERR(new_fl))
 		locks_free_lock(new_fl);
 	return error;
@@ -1319,7 +1338,7 @@ int fcntl_getlease(struct file *filp)
 	struct file_lock *fl;
 	int type = F_UNLCK;
 
-	lock_kernel();
+	lock_flocks();
 	time_out_leases(filp->f_path.dentry->d_inode);
 	for (fl = filp->f_path.dentry->d_inode->i_flock; fl && IS_LEASE(fl);
 			fl = fl->fl_next) {
@@ -1328,7 +1347,7 @@ int fcntl_getlease(struct file *filp)
 			break;
 		}
 	}
-	unlock_kernel();
+	unlock_flocks();
 	return type;
 }
 
@@ -1341,7 +1360,7 @@ int fcntl_getlease(struct file *filp)
  *	The (input) flp->fl_lmops->fl_break function is required
  *	by break_lease().
  *
- *	Called with kernel lock held.
+ *	Called with file_lock_lock held.
  */
 int generic_setlease(struct file *filp, long arg, struct file_lock **flp)
 {
@@ -1436,7 +1455,15 @@ out:
 }
 EXPORT_SYMBOL(generic_setlease);
 
- /**
+static int __vfs_setlease(struct file *filp, long arg, struct file_lock **lease)
+{
+	if (filp->f_op && filp->f_op->setlease)
+		return filp->f_op->setlease(filp, arg, lease);
+	else
+		return generic_setlease(filp, arg, lease);
+}
+
+/**
  *	vfs_setlease        -       sets a lease on an open file
  *	@filp: file pointer
  *	@arg: type of lease to obtain
@@ -1467,12 +1494,9 @@ int vfs_setlease(struct file *filp, long arg, struct file_lock **lease)
 {
 	int error;
 
-	lock_kernel();
-	if (filp->f_op && filp->f_op->setlease)
-		error = filp->f_op->setlease(filp, arg, lease);
-	else
-		error = generic_setlease(filp, arg, lease);
-	unlock_kernel();
+	lock_flocks();
+	error = __vfs_setlease(filp, arg, lease);
+	unlock_flocks();
 
 	return error;
 }
@@ -1499,9 +1523,9 @@ int fcntl_setlease(unsigned int fd, struct file *filp, long arg)
 	if (error)
 		return error;
 
-	lock_kernel();
+	lock_flocks();
 
-	error = vfs_setlease(filp, arg, &flp);
+	error = __vfs_setlease(filp, arg, &flp);
 	if (error || arg == F_UNLCK)
 		goto out_unlock;
 
@@ -1516,7 +1540,7 @@ int fcntl_setlease(unsigned int fd, struct file *filp, long arg)
 
 	error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0);
 out_unlock:
-	unlock_kernel();
+	unlock_flocks();
 	return error;
 }
 
@@ -2020,7 +2044,7 @@ void locks_remove_flock(struct file *filp)
 			fl.fl_ops->fl_release_private(&fl);
 	}
 
-	lock_kernel();
+	lock_flocks();
 	before = &inode->i_flock;
 
 	while ((fl = *before) != NULL) {
@@ -2038,7 +2062,7 @@ void locks_remove_flock(struct file *filp)
  		}
 		before = &fl->fl_next;
 	}
-	unlock_kernel();
+	unlock_flocks();
 }
 
 /**
@@ -2053,12 +2077,12 @@ posix_unblock_lock(struct file *filp, struct file_lock *waiter)
 {
 	int status = 0;
 
-	lock_kernel();
+	lock_flocks();
 	if (waiter->fl_next)
 		__locks_delete_block(waiter);
 	else
 		status = -ENOENT;
-	unlock_kernel();
+	unlock_flocks();
 	return status;
 }
 
@@ -2172,7 +2196,7 @@ static int locks_show(struct seq_file *f, void *v)
 
 static void *locks_start(struct seq_file *f, loff_t *pos)
 {
-	lock_kernel();
+	lock_flocks();
 	f->private = (void *)1;
 	return seq_list_start(&file_lock_list, *pos);
 }
@@ -2184,7 +2208,7 @@ static void *locks_next(struct seq_file *f, void *v, loff_t *pos)
 
 static void locks_stop(struct seq_file *f, void *v)
 {
-	unlock_kernel();
+	unlock_flocks();
 }
 
 static const struct seq_operations locks_seq_operations = {
@@ -2231,7 +2255,7 @@ int lock_may_read(struct inode *inode, loff_t start, unsigned long len)
 {
 	struct file_lock *fl;
 	int result = 1;
-	lock_kernel();
+	lock_flocks();
 	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
 		if (IS_POSIX(fl)) {
 			if (fl->fl_type == F_RDLCK)
@@ -2248,7 +2272,7 @@ int lock_may_read(struct inode *inode, loff_t start, unsigned long len)
 		result = 0;
 		break;
 	}
-	unlock_kernel();
+	unlock_flocks();
 	return result;
 }
 
@@ -2271,7 +2295,7 @@ int lock_may_write(struct inode *inode, loff_t start, unsigned long len)
 {
 	struct file_lock *fl;
 	int result = 1;
-	lock_kernel();
+	lock_flocks();
 	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
 		if (IS_POSIX(fl)) {
 			if ((fl->fl_end < start) || (fl->fl_start > (start + len)))
@@ -2286,7 +2310,7 @@ int lock_may_write(struct inode *inode, loff_t start, unsigned long len)
 		result = 0;
 		break;
 	}
-	unlock_kernel();
+	unlock_flocks();
 	return result;
 }
 
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index b9c3c43cea1d..232a7eead33a 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -71,20 +71,20 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_
 	if (inode->i_flock == NULL)
 		goto out;
 
-	/* Protect inode->i_flock using the BKL */
-	lock_kernel();
+	/* Protect inode->i_flock using the file locks lock */
+	lock_flocks();
 	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
 		if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK)))
 			continue;
 		if (nfs_file_open_context(fl->fl_file) != ctx)
 			continue;
-		unlock_kernel();
+		unlock_flocks();
 		status = nfs4_lock_delegation_recall(state, fl);
 		if (status < 0)
 			goto out;
-		lock_kernel();
+		lock_flocks();
 	}
-	unlock_kernel();
+	unlock_flocks();
 out:
 	return status;
 }
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 3e2f19b04c06..96524c5dca6b 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -40,7 +40,7 @@
 
 #include <linux/kernel.h>
 #include <linux/slab.h>
-#include <linux/smp_lock.h>
+#include <linux/fs.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_idmap.h>
 #include <linux/kthread.h>
@@ -970,13 +970,13 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_
 	/* Guard against delegation returns and new lock/unlock calls */
 	down_write(&nfsi->rwsem);
 	/* Protect inode->i_flock using the BKL */
-	lock_kernel();
+	lock_flocks();
 	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
 		if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK)))
 			continue;
 		if (nfs_file_open_context(fl->fl_file)->state != state)
 			continue;
-		unlock_kernel();
+		unlock_flocks();
 		status = ops->recover_lock(state, fl);
 		switch (status) {
 			case 0:
@@ -1003,9 +1003,9 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_
 				/* kill_proc(fl->fl_pid, SIGLOST, 1); */
 				status = 0;
 		}
-		lock_kernel();
+		lock_flocks();
 	}
-	unlock_kernel();
+	unlock_flocks();
 out:
 	up_write(&nfsi->rwsem);
 	return status;
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index cf0d2ffb3c84..a7292fcf7718 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -33,7 +33,7 @@
 */
 
 #include <linux/file.h>
-#include <linux/smp_lock.h>
+#include <linux/fs.h>
 #include <linux/slab.h>
 #include <linux/namei.h>
 #include <linux/swap.h>
@@ -3895,7 +3895,7 @@ check_for_locks(struct nfs4_file *filp, struct nfs4_stateowner *lowner)
 	struct inode *inode = filp->fi_inode;
 	int status = 0;
 
-	lock_kernel();
+	lock_flocks();
 	for (flpp = &inode->i_flock; *flpp != NULL; flpp = &(*flpp)->fl_next) {
 		if ((*flpp)->fl_owner == (fl_owner_t)lowner) {
 			status = 1;
@@ -3903,7 +3903,7 @@ check_for_locks(struct nfs4_file *filp, struct nfs4_stateowner *lowner)
 		}
 	}
 out:
-	unlock_kernel();
+	unlock_flocks();
 	return status;
 }
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 63d069bd80b7..180325268237 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1093,10 +1093,6 @@ struct file_lock {
 
 #include <linux/fcntl.h>
 
-/* temporary stubs for BKL removal */
-#define lock_flocks() lock_kernel()
-#define unlock_flocks() unlock_kernel()
-
 extern void send_sigio(struct fown_struct *fown, int fd, int band);
 
 #ifdef CONFIG_FILE_LOCKING
@@ -1135,6 +1131,8 @@ extern int vfs_setlease(struct file *, long, struct file_lock **);
 extern int lease_modify(struct file_lock **, int);
 extern int lock_may_read(struct inode *, loff_t start, unsigned long count);
 extern int lock_may_write(struct inode *, loff_t start, unsigned long count);
+extern void lock_flocks(void);
+extern void unlock_flocks(void);
 #else /* !CONFIG_FILE_LOCKING */
 static inline int fcntl_getlk(struct file *file, struct flock __user *user)
 {
@@ -1277,6 +1275,14 @@ static inline int lock_may_write(struct inode *inode, loff_t start,
 	return 1;
 }
 
+static inline void lock_flocks(void)
+{
+}
+
+static inline void unlock_flocks(void)
+{
+}
+
 #endif /* !CONFIG_FILE_LOCKING */
 
 
-- 
cgit v1.2.3


From 17e5a8082894a4b66cb69e7ec16074f0f01281e1 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Wed, 29 Sep 2010 17:15:30 +0200
Subject: nl80211: allow drivers to indicate whether the survey data channel is
 in use

Some user space applications only want to display survey data for
the operating channel, however there is no API to get that yet.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 2 ++
 include/net/cfg80211.h  | 2 ++
 net/wireless/nl80211.c  | 2 ++
 3 files changed, 6 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index f0518b0278a9..edd21ae6acf7 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -1400,6 +1400,7 @@ enum nl80211_reg_rule_flags {
  * @__NL80211_SURVEY_INFO_INVALID: attribute number 0 is reserved
  * @NL80211_SURVEY_INFO_FREQUENCY: center frequency of channel
  * @NL80211_SURVEY_INFO_NOISE: noise level of channel (u8, dBm)
+ * @NL80211_SURVEY_INFO_IN_USE: channel is currently being used
  * @NL80211_SURVEY_INFO_MAX: highest survey info attribute number
  *	currently defined
  * @__NL80211_SURVEY_INFO_AFTER_LAST: internal use
@@ -1408,6 +1409,7 @@ enum nl80211_survey_info {
 	__NL80211_SURVEY_INFO_INVALID,
 	NL80211_SURVEY_INFO_FREQUENCY,
 	NL80211_SURVEY_INFO_NOISE,
+	NL80211_SURVEY_INFO_IN_USE,
 
 	/* keep last */
 	__NL80211_SURVEY_INFO_AFTER_LAST,
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index a0613ff62c97..ecc0403b918a 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -293,12 +293,14 @@ struct key_params {
  * enum survey_info_flags - survey information flags
  *
  * @SURVEY_INFO_NOISE_DBM: noise (in dBm) was filled in
+ * @SURVEY_INFO_IN_USE: channel is currently being used
  *
  * Used by the driver to indicate which info in &struct survey_info
  * it has filled in during the get_survey().
  */
 enum survey_info_flags {
 	SURVEY_INFO_NOISE_DBM = 1<<0,
+	SURVEY_INFO_IN_USE = 1<<1,
 };
 
 /**
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 9c84825803ce..0087c4323c53 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -3489,6 +3489,8 @@ static int nl80211_send_survey(struct sk_buff *msg, u32 pid, u32 seq,
 	if (survey->filled & SURVEY_INFO_NOISE_DBM)
 		NLA_PUT_U8(msg, NL80211_SURVEY_INFO_NOISE,
 			    survey->noise);
+	if (survey->filled & SURVEY_INFO_IN_USE)
+		NLA_PUT_FLAG(msg, NL80211_SURVEY_INFO_IN_USE);
 
 	nla_nest_end(msg, infoattr);
 
-- 
cgit v1.2.3


From e8347ebad2f1b15bddb6ed3ed5f767531eb52dc3 Mon Sep 17 00:00:00 2001
From: Bill Jordan <bjordan@ig88.(none)>
Date: Fri, 1 Oct 2010 13:54:28 -0400
Subject: cfg80211: patches to allow setting the WDS peer

Added a nl interface to set the peer bssid of a WDS interface.

Signed-off-by: Bill Jordan <bjordan@rajant.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h |  3 +++
 net/wireless/nl80211.c  | 54 +++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 57 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index edd21ae6acf7..73d9390d4ddb 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -387,6 +387,8 @@
  *	of any other interfaces, and other interfaces will again take
  *	precedence when they are used.
  *
+ * @NL80211_CMD_SET_WDS_PEER: Set the MAC address of the peer on a WDS interface.
+ *
  * @NL80211_CMD_MAX: highest used command number
  * @__NL80211_CMD_AFTER_LAST: internal use
  */
@@ -489,6 +491,7 @@ enum nl80211_commands {
 	NL80211_CMD_NOTIFY_CQM,
 
 	NL80211_CMD_SET_CHANNEL,
+	NL80211_CMD_SET_WDS_PEER,
 
 	/* add new commands above here */
 
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 21061ccee557..fd92b6b7ff04 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -603,6 +603,7 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 		NLA_PUT_U32(msg, i, NL80211_CMD_SET_WIPHY_NETNS);
 	}
 	CMD(set_channel, SET_CHANNEL);
+	CMD(set_wds_peer, SET_WDS_PEER);
 
 #undef CMD
 
@@ -833,6 +834,53 @@ static int nl80211_set_channel(struct sk_buff *skb, struct genl_info *info)
 	return result;
 }
 
+static int nl80211_set_wds_peer(struct sk_buff *skb, struct genl_info *info)
+{
+	struct cfg80211_registered_device *rdev;
+	struct wireless_dev *wdev;
+	struct net_device *dev;
+	u8 *bssid;
+	int err;
+
+	if (!info->attrs[NL80211_ATTR_MAC])
+		return -EINVAL;
+
+	rtnl_lock();
+
+	err = get_rdev_dev_by_info_ifindex(info, &rdev, &dev);
+	if (err)
+		goto unlock_rtnl;
+
+	wdev = dev->ieee80211_ptr;
+
+	if (netif_running(dev)) {
+		err = -EBUSY;
+		goto out;
+	}
+
+	if (!rdev->ops->set_wds_peer) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	if (wdev->iftype != NL80211_IFTYPE_WDS) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
+	bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
+	err = rdev->ops->set_wds_peer(wdev->wiphy, dev, bssid);
+
+out:
+	cfg80211_unlock_rdev(rdev);
+	dev_put(dev);
+unlock_rtnl:
+	rtnl_unlock();
+
+	return err;
+}
+
+
 static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *rdev;
@@ -5473,6 +5521,12 @@ static struct genl_ops nl80211_ops[] = {
 		.policy = nl80211_policy,
 		.flags = GENL_ADMIN_PERM,
 	},
+	{
+		.cmd = NL80211_CMD_SET_WDS_PEER,
+		.doit = nl80211_set_wds_peer,
+		.policy = nl80211_policy,
+		.flags = GENL_ADMIN_PERM,
+	},
 };
 
 static struct genl_multicast_group nl80211_mlme_mcgrp = {
-- 
cgit v1.2.3


From 691895e7e2204be9a717809fb78d6ff7c10b470a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 5 Oct 2010 16:19:42 +0200
Subject: nl80211: fix remain-on-channel documentation

The documentation for NL80211_CMD_REMAIN_ON_CHANNEL
isn't accurate, an interface index is required by
the command. Update it accordingly.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 73d9390d4ddb..c4efdfa24ed8 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -315,8 +315,8 @@
  *	channel for the specified amount of time. This can be used to do
  *	off-channel operations like transmit a Public Action frame and wait for
  *	a response while being associated to an AP on another channel.
- *	%NL80211_ATTR_WIPHY or %NL80211_ATTR_IFINDEX is used to specify which
- *	radio is used. %NL80211_ATTR_WIPHY_FREQ is used to specify the
+ *	%NL80211_ATTR_IFINDEX is used to specify which interface (and thus
+ *	radio) is used. %NL80211_ATTR_WIPHY_FREQ is used to specify the
  *	frequency for the operation and %NL80211_ATTR_WIPHY_CHANNEL_TYPE may be
  *	optionally used to specify additional channel parameters.
  *	%NL80211_ATTR_DURATION is used to specify the duration in milliseconds
-- 
cgit v1.2.3


From 5336377d6225959624146629ce3fc88ee8ecda3d Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Tue, 5 Oct 2010 11:29:27 -0700
Subject: modules: Fix module_bug_list list corruption race

With all the recent module loading cleanups, we've minimized the code
that sits under module_mutex, fixing various deadlocks and making it
possible to do most of the module loading in parallel.

However, that whole conversion totally missed the rather obscure code
that adds a new module to the list for BUG() handling.  That code was
doubly obscure because (a) the code itself lives in lib/bugs.c (for
dubious reasons) and (b) it gets called from the architecture-specific
"module_finalize()" rather than from generic code.

Calling it from arch-specific code makes no sense what-so-ever to begin
with, and is now actively wrong since that code isn't protected by the
module loading lock any more.

So this commit moves the "module_bug_{finalize,cleanup}()" calls away
from the arch-specific code, and into the generic code - and in the
process protects it with the module_mutex so that the list operations
are now safe.

Future fixups:
 - move the module list handling code into kernel/module.c where it
   belongs.
 - get rid of 'module_bug_list' and just use the regular list of modules
   (called 'modules' - imagine that) that we already create and maintain
   for other reasons.

Reported-and-tested-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Adrian Bunk <bunk@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/avr32/kernel/module.c   | 3 +--
 arch/h8300/kernel/module.c   | 3 +--
 arch/mn10300/kernel/module.c | 3 +--
 arch/parisc/kernel/module.c  | 3 +--
 arch/powerpc/kernel/module.c | 5 -----
 arch/s390/kernel/module.c    | 3 +--
 arch/sh/kernel/module.c      | 2 --
 arch/x86/kernel/module.c     | 3 +--
 include/linux/module.h       | 5 ++---
 kernel/module.c              | 4 ++++
 lib/bug.c                    | 6 ++----
 11 files changed, 14 insertions(+), 26 deletions(-)

(limited to 'include/linux')

diff --git a/arch/avr32/kernel/module.c b/arch/avr32/kernel/module.c
index 98f94d041d9c..a727f54d64d6 100644
--- a/arch/avr32/kernel/module.c
+++ b/arch/avr32/kernel/module.c
@@ -314,10 +314,9 @@ int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs,
 	vfree(module->arch.syminfo);
 	module->arch.syminfo = NULL;
 
-	return module_bug_finalize(hdr, sechdrs, module);
+	return 0;
 }
 
 void module_arch_cleanup(struct module *module)
 {
-	module_bug_cleanup(module);
 }
diff --git a/arch/h8300/kernel/module.c b/arch/h8300/kernel/module.c
index 0865e291c20d..db4953dc4e1b 100644
--- a/arch/h8300/kernel/module.c
+++ b/arch/h8300/kernel/module.c
@@ -112,10 +112,9 @@ int module_finalize(const Elf_Ehdr *hdr,
 		    const Elf_Shdr *sechdrs,
 		    struct module *me)
 {
-	return module_bug_finalize(hdr, sechdrs, me);
+	return 0;
 }
 
 void module_arch_cleanup(struct module *mod)
 {
-	module_bug_cleanup(mod);
 }
diff --git a/arch/mn10300/kernel/module.c b/arch/mn10300/kernel/module.c
index 6aea7fd76993..196a111e2e29 100644
--- a/arch/mn10300/kernel/module.c
+++ b/arch/mn10300/kernel/module.c
@@ -206,7 +206,7 @@ int module_finalize(const Elf_Ehdr *hdr,
 		    const Elf_Shdr *sechdrs,
 		    struct module *me)
 {
-	return module_bug_finalize(hdr, sechdrs, me);
+	return 0;
 }
 
 /*
@@ -214,5 +214,4 @@ int module_finalize(const Elf_Ehdr *hdr,
  */
 void module_arch_cleanup(struct module *mod)
 {
-	module_bug_cleanup(mod);
 }
diff --git a/arch/parisc/kernel/module.c b/arch/parisc/kernel/module.c
index 159a2b81e90c..6e81bb596e5b 100644
--- a/arch/parisc/kernel/module.c
+++ b/arch/parisc/kernel/module.c
@@ -941,11 +941,10 @@ int module_finalize(const Elf_Ehdr *hdr,
 	nsyms = newptr - (Elf_Sym *)symhdr->sh_addr;
 	DEBUGP("NEW num_symtab %lu\n", nsyms);
 	symhdr->sh_size = nsyms * sizeof(Elf_Sym);
-	return module_bug_finalize(hdr, sechdrs, me);
+	return 0;
 }
 
 void module_arch_cleanup(struct module *mod)
 {
 	deregister_unwind_table(mod);
-	module_bug_cleanup(mod);
 }
diff --git a/arch/powerpc/kernel/module.c b/arch/powerpc/kernel/module.c
index 477c663e0140..4ef93ae2235f 100644
--- a/arch/powerpc/kernel/module.c
+++ b/arch/powerpc/kernel/module.c
@@ -65,10 +65,6 @@ int module_finalize(const Elf_Ehdr *hdr,
 	const Elf_Shdr *sect;
 	int err;
 
-	err = module_bug_finalize(hdr, sechdrs, me);
-	if (err)
-		return err;
-
 	/* Apply feature fixups */
 	sect = find_section(hdr, sechdrs, "__ftr_fixup");
 	if (sect != NULL)
@@ -101,5 +97,4 @@ int module_finalize(const Elf_Ehdr *hdr,
 
 void module_arch_cleanup(struct module *mod)
 {
-	module_bug_cleanup(mod);
 }
diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c
index 22cfd634c355..f7167ee4604c 100644
--- a/arch/s390/kernel/module.c
+++ b/arch/s390/kernel/module.c
@@ -407,10 +407,9 @@ int module_finalize(const Elf_Ehdr *hdr,
 {
 	vfree(me->arch.syminfo);
 	me->arch.syminfo = NULL;
-	return module_bug_finalize(hdr, sechdrs, me);
+	return 0;
 }
 
 void module_arch_cleanup(struct module *mod)
 {
-	module_bug_cleanup(mod);
 }
diff --git a/arch/sh/kernel/module.c b/arch/sh/kernel/module.c
index 43adddfe4c04..ae0be697a89e 100644
--- a/arch/sh/kernel/module.c
+++ b/arch/sh/kernel/module.c
@@ -149,13 +149,11 @@ int module_finalize(const Elf_Ehdr *hdr,
 	int ret = 0;
 
 	ret |= module_dwarf_finalize(hdr, sechdrs, me);
-	ret |= module_bug_finalize(hdr, sechdrs, me);
 
 	return ret;
 }
 
 void module_arch_cleanup(struct module *mod)
 {
-	module_bug_cleanup(mod);
 	module_dwarf_cleanup(mod);
 }
diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c
index e0bc186d7501..1c355c550960 100644
--- a/arch/x86/kernel/module.c
+++ b/arch/x86/kernel/module.c
@@ -239,11 +239,10 @@ int module_finalize(const Elf_Ehdr *hdr,
 		apply_paravirt(pseg, pseg + para->sh_size);
 	}
 
-	return module_bug_finalize(hdr, sechdrs, me);
+	return 0;
 }
 
 void module_arch_cleanup(struct module *mod)
 {
 	alternatives_smp_module_del(mod);
-	module_bug_cleanup(mod);
 }
diff --git a/include/linux/module.h b/include/linux/module.h
index 8a6b9fdc7ffa..aace066bad8f 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -686,17 +686,16 @@ extern int module_sysfs_initialized;
 
 
 #ifdef CONFIG_GENERIC_BUG
-int  module_bug_finalize(const Elf_Ehdr *, const Elf_Shdr *,
+void module_bug_finalize(const Elf_Ehdr *, const Elf_Shdr *,
 			 struct module *);
 void module_bug_cleanup(struct module *);
 
 #else	/* !CONFIG_GENERIC_BUG */
 
-static inline int  module_bug_finalize(const Elf_Ehdr *hdr,
+static inline void module_bug_finalize(const Elf_Ehdr *hdr,
 					const Elf_Shdr *sechdrs,
 					struct module *mod)
 {
-	return 0;
 }
 static inline void module_bug_cleanup(struct module *mod) {}
 #endif	/* CONFIG_GENERIC_BUG */
diff --git a/kernel/module.c b/kernel/module.c
index d0b5f8db11b4..ccd641991842 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -1537,6 +1537,7 @@ static int __unlink_module(void *_mod)
 {
 	struct module *mod = _mod;
 	list_del(&mod->list);
+	module_bug_cleanup(mod);
 	return 0;
 }
 
@@ -2625,6 +2626,7 @@ static struct module *load_module(void __user *umod,
 	if (err < 0)
 		goto ddebug;
 
+	module_bug_finalize(info.hdr, info.sechdrs, mod);
 	list_add_rcu(&mod->list, &modules);
 	mutex_unlock(&module_mutex);
 
@@ -2650,6 +2652,8 @@ static struct module *load_module(void __user *umod,
 	mutex_lock(&module_mutex);
 	/* Unlink carefully: kallsyms could be walking list. */
 	list_del_rcu(&mod->list);
+	module_bug_cleanup(mod);
+
  ddebug:
 	if (!mod->taints)
 		dynamic_debug_remove(info.debug);
diff --git a/lib/bug.c b/lib/bug.c
index 7cdfad88128f..19552096d16b 100644
--- a/lib/bug.c
+++ b/lib/bug.c
@@ -72,8 +72,8 @@ static const struct bug_entry *module_find_bug(unsigned long bugaddr)
 	return NULL;
 }
 
-int module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs,
-			struct module *mod)
+void module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs,
+			 struct module *mod)
 {
 	char *secstrings;
 	unsigned int i;
@@ -97,8 +97,6 @@ int module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs,
 	 * could potentially lead to deadlock and thus be counter-productive.
 	 */
 	list_add(&mod->bug_list, &module_bug_list);
-
-	return 0;
 }
 
 void module_bug_cleanup(struct module *mod)
-- 
cgit v1.2.3


From 231d0aefd88e94129cb8fb84794f9bb788c6366e Mon Sep 17 00:00:00 2001
From: Evgeny Kuznetsov <ext-eugeny.kuznetsov@nokia.com>
Date: Tue, 5 Oct 2010 12:47:57 +0400
Subject: wait: using uninitialized member of wait queue

The "flags" member of "struct wait_queue_t" is used in several places in
the kernel code without beeing initialized by init_wait().  "flags" is
used in bitwise operations.

If "flags" not initialized then unexpected behaviour may take place.
Incorrect flags might used later in code.

Added initialization of "wait_queue_t.flags" with zero value into
"init_wait".

Signed-off-by: Evgeny Kuznetsov <EXT-Eugeny.Kuznetsov@nokia.com>
[ The bit we care about does end up being initialized by both
   prepare_to_wait() and add_to_wait_queue(), so this doesn't seem to
   cause actual bugs, but is definitely the right thing to do -Linus ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/wait.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/wait.h b/include/linux/wait.h
index 0836ccc57121..3efc9f3f43a0 100644
--- a/include/linux/wait.h
+++ b/include/linux/wait.h
@@ -614,6 +614,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key);
 		(wait)->private = current;				\
 		(wait)->func = autoremove_wake_function;		\
 		INIT_LIST_HEAD(&(wait)->task_list);			\
+		(wait)->flags = 0;					\
 	} while (0)
 
 /**
-- 
cgit v1.2.3


From 773e3f93577ffb493fb7c39b1a6ecf39b5748e87 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Tue, 5 Oct 2010 14:03:02 -0700
Subject: rcu: move check from rcu_dereference_bh to rcu_read_lock_bh_held

As suggested by Linus, push the irqs_disabled() down to the
rcu_read_lock_bh_held() level so that all callers get the benefit
of the correct check.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/rcupdate.h | 2 +-
 kernel/rcupdate.c        | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 83af1f8d8b74..9fbc54a2585d 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -454,7 +454,7 @@ static inline notrace void rcu_read_unlock_sched_notrace(void)
  * Makes rcu_dereference_check() do the dirty work.
  */
 #define rcu_dereference_bh(p) \
-		rcu_dereference_check(p, rcu_read_lock_bh_held() || irqs_disabled())
+		rcu_dereference_check(p, rcu_read_lock_bh_held())
 
 /**
  * rcu_dereference_sched - fetch RCU-protected pointer, checking for RCU-sched
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index 4d169835fb36..0af1dc70fece 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -86,7 +86,7 @@ int rcu_read_lock_bh_held(void)
 {
 	if (!debug_lockdep_rcu_enabled())
 		return 1;
-	return in_softirq();
+	return in_softirq() || irqs_disabled();
 }
 EXPORT_SYMBOL_GPL(rcu_read_lock_bh_held);
 
-- 
cgit v1.2.3


From caf586e5f23cebb2a68cbaf288d59dbbf2d74052 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 30 Sep 2010 21:06:55 +0000
Subject: net: add a core netdev->rx_dropped counter

In various situations, a device provides a packet to our stack and we
drop it before it enters protocol stack :
- softnet backlog full (accounted in /proc/net/softnet_stat)
- bad vlan tag (not accounted)
- unknown/unregistered protocol (not accounted)

We can handle a per-device counter of such dropped frames at core level,
and automatically adds it to the device provided stats (rx_dropped), so
that standard tools can be used (ifconfig, ip link, cat /proc/net/dev)

This is a generalization of commit 8990f468a (net: rx_dropped
accounting), thus reverting it.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/loopback.c    |  8 +-------
 include/linux/netdevice.h |  3 +++
 net/8021q/vlan.h          |  2 --
 net/8021q/vlan_core.c     |  2 ++
 net/8021q/vlan_dev.c      | 11 ++++-------
 net/core/dev.c            | 19 +++++++++++--------
 net/ipv4/ip_gre.c         |  3 +--
 net/ipv4/ipip.c           |  3 +--
 net/ipv6/ip6_tunnel.c     |  3 +--
 net/ipv6/ip6mr.c          |  3 +--
 net/ipv6/sit.c            |  3 +--
 11 files changed, 26 insertions(+), 34 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index 4b0e30b564e5..2d9663a1c54d 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -64,7 +64,6 @@ struct pcpu_lstats {
 	u64			packets;
 	u64			bytes;
 	struct u64_stats_sync	syncp;
-	unsigned long		drops;
 };
 
 /*
@@ -90,8 +89,7 @@ static netdev_tx_t loopback_xmit(struct sk_buff *skb,
 		lb_stats->bytes += len;
 		lb_stats->packets++;
 		u64_stats_update_end(&lb_stats->syncp);
-	} else
-		lb_stats->drops++;
+	}
 
 	return NETDEV_TX_OK;
 }
@@ -101,7 +99,6 @@ static struct rtnl_link_stats64 *loopback_get_stats64(struct net_device *dev,
 {
 	u64 bytes = 0;
 	u64 packets = 0;
-	u64 drops = 0;
 	int i;
 
 	for_each_possible_cpu(i) {
@@ -115,14 +112,11 @@ static struct rtnl_link_stats64 *loopback_get_stats64(struct net_device *dev,
 			tbytes = lb_stats->bytes;
 			tpackets = lb_stats->packets;
 		} while (u64_stats_fetch_retry(&lb_stats->syncp, start));
-		drops   += lb_stats->drops;
 		bytes   += tbytes;
 		packets += tpackets;
 	}
 	stats->rx_packets = packets;
 	stats->tx_packets = packets;
-	stats->rx_dropped = drops;
-	stats->rx_errors  = drops;
 	stats->rx_bytes   = bytes;
 	stats->tx_bytes   = bytes;
 	return stats;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 92d81edd5808..6abcef67b178 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -884,6 +884,9 @@ struct net_device {
 	int			iflink;
 
 	struct net_device_stats	stats;
+	atomic_long_t		rx_dropped; /* dropped packets by core network
+					     * Do not use this in drivers.
+					     */
 
 #ifdef CONFIG_WIRELESS_EXT
 	/* List of functions to handle Wireless Extensions (instead of ioctl).
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index b26ce343072c..8d9503ad01da 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -25,7 +25,6 @@ struct vlan_priority_tci_mapping {
  *	@rx_multicast: number of received multicast packets
  *	@syncp: synchronization point for 64bit counters
  *	@rx_errors: number of errors
- *	@rx_dropped: number of dropped packets
  */
 struct vlan_rx_stats {
 	u64			rx_packets;
@@ -33,7 +32,6 @@ struct vlan_rx_stats {
 	u64			rx_multicast;
 	struct u64_stats_sync	syncp;
 	unsigned long		rx_errors;
-	unsigned long		rx_dropped;
 };
 
 /**
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index b6d55a9304f2..dee727ce0291 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -33,6 +33,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 	return polling ? netif_receive_skb(skb) : netif_rx(skb);
 
 drop:
+	atomic_long_inc(&skb->dev->rx_dropped);
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
@@ -123,6 +124,7 @@ vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp,
 	return dev_gro_receive(napi, skb);
 
 drop:
+	atomic_long_inc(&skb->dev->rx_dropped);
 	return GRO_DROP;
 }
 
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index f6fbcc0f1af9..f54251edd40d 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -225,16 +225,15 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 		}
 	}
 
-	if (unlikely(netif_rx(skb) == NET_RX_DROP)) {
-		if (rx_stats)
-			rx_stats->rx_dropped++;
-	}
+	netif_rx(skb);
+
 	rcu_read_unlock();
 	return NET_RX_SUCCESS;
 
 err_unlock:
 	rcu_read_unlock();
 err_free:
+	atomic_long_inc(&dev->rx_dropped);
 	kfree_skb(skb);
 	return NET_RX_DROP;
 }
@@ -846,15 +845,13 @@ static struct rtnl_link_stats64 *vlan_dev_get_stats64(struct net_device *dev, st
 			accum.rx_packets += rxpackets;
 			accum.rx_bytes   += rxbytes;
 			accum.rx_multicast += rxmulticast;
-			/* rx_errors, rx_dropped are ulong, not protected by syncp */
+			/* rx_errors is ulong, not protected by syncp */
 			accum.rx_errors  += p->rx_errors;
-			accum.rx_dropped += p->rx_dropped;
 		}
 		stats->rx_packets = accum.rx_packets;
 		stats->rx_bytes   = accum.rx_bytes;
 		stats->rx_errors  = accum.rx_errors;
 		stats->multicast  = accum.rx_multicast;
-		stats->rx_dropped = accum.rx_dropped;
 	}
 	return stats;
 }
diff --git a/net/core/dev.c b/net/core/dev.c
index ce6ad88c980b..7d149550e8d6 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1483,8 +1483,9 @@ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb)
 	skb_orphan(skb);
 	nf_reset(skb);
 
-	if (!(dev->flags & IFF_UP) ||
-	    (skb->len > (dev->mtu + dev->hard_header_len))) {
+	if (unlikely(!(dev->flags & IFF_UP) ||
+		     (skb->len > (dev->mtu + dev->hard_header_len)))) {
+		atomic_long_inc(&dev->rx_dropped);
 		kfree_skb(skb);
 		return NET_RX_DROP;
 	}
@@ -2548,6 +2549,7 @@ enqueue:
 
 	local_irq_restore(flags);
 
+	atomic_long_inc(&skb->dev->rx_dropped);
 	kfree_skb(skb);
 	return NET_RX_DROP;
 }
@@ -2995,6 +2997,7 @@ ncls:
 	if (pt_prev) {
 		ret = pt_prev->func(skb, skb->dev, pt_prev, orig_dev);
 	} else {
+		atomic_long_inc(&skb->dev->rx_dropped);
 		kfree_skb(skb);
 		/* Jamal, now you will not able to escape explaining
 		 * me how you were going to use this. :-)
@@ -5429,14 +5432,14 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
 
 	if (ops->ndo_get_stats64) {
 		memset(storage, 0, sizeof(*storage));
-		return ops->ndo_get_stats64(dev, storage);
-	}
-	if (ops->ndo_get_stats) {
+		ops->ndo_get_stats64(dev, storage);
+	} else if (ops->ndo_get_stats) {
 		netdev_stats_to_stats64(storage, ops->ndo_get_stats(dev));
-		return storage;
+	} else {
+		netdev_stats_to_stats64(storage, &dev->stats);
+		dev_txq_stats_fold(dev, storage);
 	}
-	netdev_stats_to_stats64(storage, &dev->stats);
-	dev_txq_stats_fold(dev, storage);
+	storage->rx_dropped += atomic_long_read(&dev->rx_dropped);
 	return storage;
 }
 EXPORT_SYMBOL(dev_get_stats);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index fbe2c473a06a..9d421f4cf3ef 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -679,8 +679,7 @@ static int ipgre_rcv(struct sk_buff *skb)
 		skb_reset_network_header(skb);
 		ipgre_ecn_decapsulate(iph, skb);
 
-		if (netif_rx(skb) == NET_RX_DROP)
-			tunnel->dev->stats.rx_dropped++;
+		netif_rx(skb);
 
 		rcu_read_unlock();
 		return 0;
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 6ad46c28ede2..e9b816e6cd73 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -414,8 +414,7 @@ static int ipip_rcv(struct sk_buff *skb)
 
 		ipip_ecn_decapsulate(iph, skb);
 
-		if (netif_rx(skb) == NET_RX_DROP)
-			tunnel->dev->stats.rx_dropped++;
+		netif_rx(skb);
 
 		rcu_read_unlock();
 		return 0;
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 8be3c452af90..c2c0f89397b1 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -768,8 +768,7 @@ static int ip6_tnl_rcv(struct sk_buff *skb, __u16 protocol,
 
 		dscp_ecn_decapsulate(t, ipv6h, skb);
 
-		if (netif_rx(skb) == NET_RX_DROP)
-			t->dev->stats.rx_dropped++;
+		netif_rx(skb);
 
 		rcu_read_unlock();
 		return 0;
diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 2640c9be589d..6f32ffce7022 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -666,8 +666,7 @@ static int pim6_rcv(struct sk_buff *skb)
 
 	skb_tunnel_rx(skb, reg_dev);
 
-	if (netif_rx(skb) == NET_RX_DROP)
-		reg_dev->stats.rx_dropped++;
+	netif_rx(skb);
 
 	dev_put(reg_dev);
 	return 0;
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index d7701782b639..367a6cc584cc 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -600,8 +600,7 @@ static int ipip6_rcv(struct sk_buff *skb)
 
 		ipip6_ecn_decapsulate(iph, skb);
 
-		if (netif_rx(skb) == NET_RX_DROP)
-			tunnel->dev->stats.rx_dropped++;
+		netif_rx(skb);
 
 		rcu_read_unlock();
 		return 0;
-- 
cgit v1.2.3


From c2952c314b4fe61820ba8fd6c949eed636140d52 Mon Sep 17 00:00:00 2001
From: Flavio Leitner <fleitner@redhat.com>
Date: Tue, 5 Oct 2010 14:23:59 +0000
Subject: bonding: add retransmit membership reports tunable

Allow sysadmins to configure the number of multicast
membership report sent on a link failure event.

Signed-off-by: Flavio Leitner <fleitner@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/bonding.txt |  8 +++++++
 drivers/net/bonding/bond_main.c      | 15 ++++++++++++
 drivers/net/bonding/bond_sysfs.c     | 44 ++++++++++++++++++++++++++++++++++++
 drivers/net/bonding/bonding.h        |  2 ++
 include/linux/if_bonding.h           |  3 +++
 5 files changed, 72 insertions(+)

(limited to 'include/linux')

diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt
index d2b62b71b617..5dc638791d97 100644
--- a/Documentation/networking/bonding.txt
+++ b/Documentation/networking/bonding.txt
@@ -765,6 +765,14 @@ xmit_hash_policy
 	does not exist, and the layer2 policy is the only policy.  The
 	layer2+3 value was added for bonding version 3.2.2.
 
+resend_igmp
+
+	Specifies the number of IGMP membership reports to be issued after
+	a failover event. One membership report is issued immediately after
+	the failover, subsequent packets are sent in each 200ms interval.
+
+	The valid range is 0 - 255; the default value is 1. This option
+	was added for bonding version 3.7.0.
 
 3. Configuring Bonding Devices
 ==============================
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index ad6386671f28..6f5e6b453da6 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -109,6 +109,7 @@ static char *arp_validate;
 static char *fail_over_mac;
 static int all_slaves_active = 0;
 static struct bond_params bonding_defaults;
+static int resend_igmp = BOND_DEFAULT_RESEND_IGMP;
 
 module_param(max_bonds, int, 0);
 MODULE_PARM_DESC(max_bonds, "Max number of bonded devices");
@@ -163,6 +164,8 @@ module_param(all_slaves_active, int, 0);
 MODULE_PARM_DESC(all_slaves_active, "Keep all frames received on an interface"
 				     "by setting active flag for all slaves.  "
 				     "0 for never (default), 1 for always.");
+module_param(resend_igmp, int, 0);
+MODULE_PARM_DESC(resend_igmp, "Number of IGMP membership reports to send on link failure");
 
 /*----------------------------- Global variables ----------------------------*/
 
@@ -905,6 +908,9 @@ static void bond_resend_igmp_join_requests(struct bonding *bond)
 		}
 	}
 
+	if (--bond->igmp_retrans > 0)
+		queue_delayed_work(bond->wq, &bond->mcast_work, HZ/5);
+
 	read_unlock(&bond->lock);
 }
 
@@ -1213,6 +1219,7 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
 	 * all were sent on curr_active_slave */
 	if ((USES_PRIMARY(bond->params.mode) && new_active) ||
 	    bond->params.mode == BOND_MODE_ROUNDROBIN) {
+		bond->igmp_retrans = bond->params.resend_igmp;
 		queue_delayed_work(bond->wq, &bond->mcast_work, 0);
 	}
 }
@@ -4933,6 +4940,13 @@ static int bond_check_params(struct bond_params *params)
 		all_slaves_active = 0;
 	}
 
+	if (resend_igmp < 0 || resend_igmp > 255) {
+		pr_warning("Warning: resend_igmp (%d) should be between "
+			   "0 and 255, resetting to %d\n",
+			   resend_igmp, BOND_DEFAULT_RESEND_IGMP);
+		resend_igmp = BOND_DEFAULT_RESEND_IGMP;
+	}
+
 	/* reset values for TLB/ALB */
 	if ((bond_mode == BOND_MODE_TLB) ||
 	    (bond_mode == BOND_MODE_ALB)) {
@@ -5105,6 +5119,7 @@ static int bond_check_params(struct bond_params *params)
 	params->fail_over_mac = fail_over_mac_value;
 	params->tx_queues = tx_queues;
 	params->all_slaves_active = all_slaves_active;
+	params->resend_igmp = resend_igmp;
 
 	if (primary) {
 		strncpy(params->primary, primary, IFNAMSIZ);
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index c311aed9bd02..01b4c3f5d9e7 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -1592,6 +1592,49 @@ out:
 static DEVICE_ATTR(all_slaves_active, S_IRUGO | S_IWUSR,
 		   bonding_show_slaves_active, bonding_store_slaves_active);
 
+/*
+ * Show and set the number of IGMP membership reports to send on link failure
+ */
+static ssize_t bonding_show_resend_igmp(struct device *d,
+					 struct device_attribute *attr,
+					 char *buf)
+{
+	struct bonding *bond = to_bond(d);
+
+	return sprintf(buf, "%d\n", bond->params.resend_igmp);
+}
+
+static ssize_t bonding_store_resend_igmp(struct device *d,
+					  struct device_attribute *attr,
+					  const char *buf, size_t count)
+{
+	int new_value, ret = count;
+	struct bonding *bond = to_bond(d);
+
+	if (sscanf(buf, "%d", &new_value) != 1) {
+		pr_err("%s: no resend_igmp value specified.\n",
+		       bond->dev->name);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (new_value < 0) {
+		pr_err("%s: Invalid resend_igmp value %d not in range 0-255; rejected.\n",
+		       bond->dev->name, new_value);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	pr_info("%s: Setting resend_igmp to %d.\n",
+		bond->dev->name, new_value);
+	bond->params.resend_igmp = new_value;
+out:
+	return ret;
+}
+
+static DEVICE_ATTR(resend_igmp, S_IRUGO | S_IWUSR,
+		   bonding_show_resend_igmp, bonding_store_resend_igmp);
+
 static struct attribute *per_bond_attrs[] = {
 	&dev_attr_slaves.attr,
 	&dev_attr_mode.attr,
@@ -1619,6 +1662,7 @@ static struct attribute *per_bond_attrs[] = {
 	&dev_attr_ad_partner_mac.attr,
 	&dev_attr_queue_id.attr,
 	&dev_attr_all_slaves_active.attr,
+	&dev_attr_resend_igmp.attr,
 	NULL,
 };
 
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 308ed10dca90..c15f21347486 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -136,6 +136,7 @@ struct bond_params {
 	__be32 arp_targets[BOND_MAX_ARP_TARGETS];
 	int tx_queues;
 	int all_slaves_active;
+	int resend_igmp;
 };
 
 struct bond_parm_tbl {
@@ -202,6 +203,7 @@ struct bonding {
 	s8	 send_grat_arp;
 	s8	 send_unsol_na;
 	s8	 setup_by_slave;
+	s8       igmp_retrans;
 #ifdef CONFIG_PROC_FS
 	struct   proc_dir_entry *proc_entry;
 	char     proc_file_name[IFNAMSIZ];
diff --git a/include/linux/if_bonding.h b/include/linux/if_bonding.h
index 2c7994372bde..a17edda8a781 100644
--- a/include/linux/if_bonding.h
+++ b/include/linux/if_bonding.h
@@ -84,6 +84,9 @@
 #define BOND_DEFAULT_MAX_BONDS  1   /* Default maximum number of devices to support */
 
 #define BOND_DEFAULT_TX_QUEUES 16   /* Default number of tx queues per device */
+
+#define BOND_DEFAULT_RESEND_IGMP	1 /* Default number of IGMP membership reports */
+
 /* hashing types */
 #define BOND_XMIT_POLICY_LAYER2		0 /* layer 2 (MAC only), default */
 #define BOND_XMIT_POLICY_LAYER34	1 /* layer 3+4 (IP ^ (TCP || UDP)) */
-- 
cgit v1.2.3


From e31b82136d1adc7a599b6e99d3321e5831841f5a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 5 Oct 2010 19:39:30 +0200
Subject: cfg80211/mac80211: allow per-station GTKs

This adds API to allow adding per-station GTKs,
updates mac80211 to support it, and also allows
drivers to remove a key from hwaccel again when
this may be necessary due to multiple GTKs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwmc3200wifi/cfg80211.c |  7 +-
 drivers/net/wireless/libertas/cfg.c          |  4 +-
 drivers/net/wireless/rndis_wlan.c            | 12 ++--
 include/linux/nl80211.h                      | 12 ++++
 include/net/cfg80211.h                       |  9 ++-
 include/net/mac80211.h                       | 24 +++++++
 net/mac80211/cfg.c                           | 32 +++++++---
 net/mac80211/ieee80211_i.h                   |  2 -
 net/mac80211/key.c                           | 95 ++++++++++++++++++----------
 net/mac80211/key.h                           |  3 +
 net/mac80211/rx.c                            | 41 +++++++-----
 net/mac80211/sta_info.c                      | 10 +--
 net/mac80211/sta_info.h                      |  6 +-
 net/mac80211/tx.c                            |  2 +-
 net/wireless/core.h                          |  2 +-
 net/wireless/ibss.c                          |  2 +-
 net/wireless/nl80211.c                       | 87 ++++++++++++++++++++++---
 net/wireless/sme.c                           |  2 +-
 net/wireless/util.c                          | 12 +++-
 net/wireless/wext-compat.c                   | 38 +++++++----
 20 files changed, 293 insertions(+), 109 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/wireless/iwmc3200wifi/cfg80211.c b/drivers/net/wireless/iwmc3200wifi/cfg80211.c
index 60619678f4ec..c6c0eff9b5ed 100644
--- a/drivers/net/wireless/iwmc3200wifi/cfg80211.c
+++ b/drivers/net/wireless/iwmc3200wifi/cfg80211.c
@@ -161,7 +161,7 @@ static int iwm_key_init(struct iwm_key *key, u8 key_index,
 }
 
 static int iwm_cfg80211_add_key(struct wiphy *wiphy, struct net_device *ndev,
-				u8 key_index, const u8 *mac_addr,
+				u8 key_index, bool pairwise, const u8 *mac_addr,
 				struct key_params *params)
 {
 	struct iwm_priv *iwm = ndev_to_iwm(ndev);
@@ -181,7 +181,8 @@ static int iwm_cfg80211_add_key(struct wiphy *wiphy, struct net_device *ndev,
 }
 
 static int iwm_cfg80211_get_key(struct wiphy *wiphy, struct net_device *ndev,
-				u8 key_index, const u8 *mac_addr, void *cookie,
+				u8 key_index, bool pairwise, const u8 *mac_addr,
+				void *cookie,
 				void (*callback)(void *cookie,
 						 struct key_params*))
 {
@@ -206,7 +207,7 @@ static int iwm_cfg80211_get_key(struct wiphy *wiphy, struct net_device *ndev,
 
 
 static int iwm_cfg80211_del_key(struct wiphy *wiphy, struct net_device *ndev,
-				u8 key_index, const u8 *mac_addr)
+				u8 key_index, bool pairwise, const u8 *mac_addr)
 {
 	struct iwm_priv *iwm = ndev_to_iwm(ndev);
 	struct iwm_key *key = &iwm->keys[key_index];
diff --git a/drivers/net/wireless/libertas/cfg.c b/drivers/net/wireless/libertas/cfg.c
index cb3b855d949c..1dc44bc6511f 100644
--- a/drivers/net/wireless/libertas/cfg.c
+++ b/drivers/net/wireless/libertas/cfg.c
@@ -1438,7 +1438,7 @@ static int lbs_cfg_set_default_key(struct wiphy *wiphy,
 
 
 static int lbs_cfg_add_key(struct wiphy *wiphy, struct net_device *netdev,
-			   u8 idx, const u8 *mac_addr,
+			   u8 idx, bool pairwise, const u8 *mac_addr,
 			   struct key_params *params)
 {
 	struct lbs_private *priv = wiphy_priv(wiphy);
@@ -1498,7 +1498,7 @@ static int lbs_cfg_add_key(struct wiphy *wiphy, struct net_device *netdev,
 
 
 static int lbs_cfg_del_key(struct wiphy *wiphy, struct net_device *netdev,
-			   u8 key_index, const u8 *mac_addr)
+			   u8 key_index, bool pairwise, const u8 *mac_addr)
 {
 
 	lbs_deb_enter(LBS_DEB_CFG80211);
diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c
index 719573bbbf81..71b5971da597 100644
--- a/drivers/net/wireless/rndis_wlan.c
+++ b/drivers/net/wireless/rndis_wlan.c
@@ -540,11 +540,11 @@ static int rndis_set_channel(struct wiphy *wiphy, struct net_device *dev,
 	struct ieee80211_channel *chan, enum nl80211_channel_type channel_type);
 
 static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev,
-					u8 key_index, const u8 *mac_addr,
-					struct key_params *params);
+			 u8 key_index, bool pairwise, const u8 *mac_addr,
+			 struct key_params *params);
 
 static int rndis_del_key(struct wiphy *wiphy, struct net_device *netdev,
-					u8 key_index, const u8 *mac_addr);
+			 u8 key_index, bool pairwise, const u8 *mac_addr);
 
 static int rndis_set_default_key(struct wiphy *wiphy, struct net_device *netdev,
 								u8 key_index);
@@ -2308,8 +2308,8 @@ static int rndis_set_channel(struct wiphy *wiphy, struct net_device *netdev,
 }
 
 static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev,
-					u8 key_index, const u8 *mac_addr,
-					struct key_params *params)
+			 u8 key_index, bool pairwise, const u8 *mac_addr,
+			 struct key_params *params)
 {
 	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
 	struct usbnet *usbdev = priv->usbdev;
@@ -2344,7 +2344,7 @@ static int rndis_add_key(struct wiphy *wiphy, struct net_device *netdev,
 }
 
 static int rndis_del_key(struct wiphy *wiphy, struct net_device *netdev,
-					u8 key_index, const u8 *mac_addr)
+			 u8 key_index, bool pairwise, const u8 *mac_addr)
 {
 	struct rndis_wlan_private *priv = wiphy_priv(wiphy);
 	struct usbnet *usbdev = priv->usbdev;
diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index c4efdfa24ed8..e451f176e662 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -801,6 +801,9 @@ enum nl80211_commands {
  *      This is used in association with @NL80211_ATTR_WIPHY_TX_POWER_SETTING
  *      for non-automatic settings.
  *
+ * @NL80211_ATTR_SUPPORT_IBSS_RSN: The device supports IBSS RSN, which mostly
+ *	means support for per-station GTKs.
+ *
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
  */
@@ -968,6 +971,8 @@ enum nl80211_attrs {
 	NL80211_ATTR_CONTROL_PORT_ETHERTYPE,
 	NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT,
 
+	NL80211_ATTR_SUPPORT_IBSS_RSN,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
@@ -1659,11 +1664,14 @@ enum nl80211_auth_type {
  * @NL80211_KEYTYPE_GROUP: Group (broadcast/multicast) key
  * @NL80211_KEYTYPE_PAIRWISE: Pairwise (unicast/individual) key
  * @NL80211_KEYTYPE_PEERKEY: PeerKey (DLS)
+ * @NUM_NL80211_KEYTYPES: number of defined key types
  */
 enum nl80211_key_type {
 	NL80211_KEYTYPE_GROUP,
 	NL80211_KEYTYPE_PAIRWISE,
 	NL80211_KEYTYPE_PEERKEY,
+
+	NUM_NL80211_KEYTYPES
 };
 
 /**
@@ -1694,6 +1702,9 @@ enum nl80211_wpa_versions {
  *	CCMP keys, each six bytes in little endian
  * @NL80211_KEY_DEFAULT: flag indicating default key
  * @NL80211_KEY_DEFAULT_MGMT: flag indicating default management key
+ * @NL80211_KEY_TYPE: the key type from enum nl80211_key_type, if not
+ *	specified the default depends on whether a MAC address was
+ *	given with the command using the key or not (u32)
  * @__NL80211_KEY_AFTER_LAST: internal
  * @NL80211_KEY_MAX: highest key attribute
  */
@@ -1705,6 +1716,7 @@ enum nl80211_key_attributes {
 	NL80211_KEY_SEQ,
 	NL80211_KEY_DEFAULT,
 	NL80211_KEY_DEFAULT_MGMT,
+	NL80211_KEY_TYPE,
 
 	/* keep last */
 	__NL80211_KEY_AFTER_LAST,
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 5f4d8acf7abb..0f77515266b8 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -1130,13 +1130,14 @@ struct cfg80211_ops {
 				       struct vif_params *params);
 
 	int	(*add_key)(struct wiphy *wiphy, struct net_device *netdev,
-			   u8 key_index, const u8 *mac_addr,
+			   u8 key_index, bool pairwise, const u8 *mac_addr,
 			   struct key_params *params);
 	int	(*get_key)(struct wiphy *wiphy, struct net_device *netdev,
-			   u8 key_index, const u8 *mac_addr, void *cookie,
+			   u8 key_index, bool pairwise, const u8 *mac_addr,
+			   void *cookie,
 			   void (*callback)(void *cookie, struct key_params*));
 	int	(*del_key)(struct wiphy *wiphy, struct net_device *netdev,
-			   u8 key_index, const u8 *mac_addr);
+			   u8 key_index, bool pairwise, const u8 *mac_addr);
 	int	(*set_default_key)(struct wiphy *wiphy,
 				   struct net_device *netdev,
 				   u8 key_index);
@@ -1304,6 +1305,7 @@ struct cfg80211_ops {
  * @WIPHY_FLAG_CONTROL_PORT_PROTOCOL: This device supports setting the
  *	control port protocol ethertype. The device also honours the
  *	control_port_no_encrypt flag.
+ * @WIPHY_FLAG_IBSS_RSN: The device supports IBSS RSN.
  */
 enum wiphy_flags {
 	WIPHY_FLAG_CUSTOM_REGULATORY		= BIT(0),
@@ -1314,6 +1316,7 @@ enum wiphy_flags {
 	WIPHY_FLAG_4ADDR_AP			= BIT(5),
 	WIPHY_FLAG_4ADDR_STATION		= BIT(6),
 	WIPHY_FLAG_CONTROL_PORT_PROTOCOL	= BIT(7),
+	WIPHY_FLAG_IBSS_RSN			= BIT(7),
 };
 
 struct mac_address {
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 47316a653ae1..33aa2e39147b 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -1041,6 +1041,13 @@ enum ieee80211_tkip_key_type {
  * @IEEE80211_HW_NEED_DTIM_PERIOD:
  *	This device needs to know the DTIM period for the BSS before
  *	associating.
+ *
+ * @IEEE80211_HW_SUPPORTS_PER_STA_GTK: The device's crypto engine supports
+ *	per-station GTKs as used by IBSS RSN or during fast transition. If
+ *	the device doesn't support per-station GTKs, but can be asked not
+ *	to decrypt group addressed frames, then IBSS RSN support is still
+ *	possible but software crypto will be used. Advertise the wiphy flag
+ *	only in that case.
  */
 enum ieee80211_hw_flags {
 	IEEE80211_HW_HAS_RATE_CONTROL			= 1<<0,
@@ -1064,6 +1071,7 @@ enum ieee80211_hw_flags {
 	IEEE80211_HW_REPORTS_TX_ACK_STATUS		= 1<<18,
 	IEEE80211_HW_CONNECTION_MONITOR			= 1<<19,
 	IEEE80211_HW_SUPPORTS_CQM_RSSI			= 1<<20,
+	IEEE80211_HW_SUPPORTS_PER_STA_GTK		= 1<<21,
 };
 
 /**
@@ -2582,6 +2590,22 @@ void ieee80211_chswitch_done(struct ieee80211_vif *vif, bool success);
 void ieee80211_request_smps(struct ieee80211_vif *vif,
 			    enum ieee80211_smps_mode smps_mode);
 
+/**
+ * ieee80211_key_removed - disable hw acceleration for key
+ * @key_conf: The key hw acceleration should be disabled for
+ *
+ * This allows drivers to indicate that the given key has been
+ * removed from hardware acceleration, due to a new key that
+ * was added. Don't use this if the key can continue to be used
+ * for TX, if the key restriction is on RX only it is permitted
+ * to keep the key for TX only and not call this function.
+ *
+ * Due to locking constraints, it may only be called during
+ * @set_key. This function must be allowed to sleep, and the
+ * key it tries to disable may still be used until it returns.
+ */
+void ieee80211_key_removed(struct ieee80211_key_conf *key_conf);
+
 /* Rate control API */
 
 /**
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 94bf550bd4c9..8b0e874a3d65 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -103,7 +103,7 @@ static int ieee80211_change_iface(struct wiphy *wiphy,
 }
 
 static int ieee80211_add_key(struct wiphy *wiphy, struct net_device *dev,
-			     u8 key_idx, const u8 *mac_addr,
+			     u8 key_idx, bool pairwise, const u8 *mac_addr,
 			     struct key_params *params)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
@@ -131,6 +131,9 @@ static int ieee80211_add_key(struct wiphy *wiphy, struct net_device *dev,
 	if (IS_ERR(key))
 		return PTR_ERR(key);
 
+	if (pairwise)
+		key->conf.flags |= IEEE80211_KEY_FLAG_PAIRWISE;
+
 	mutex_lock(&sdata->local->sta_mtx);
 
 	if (mac_addr) {
@@ -153,7 +156,7 @@ static int ieee80211_add_key(struct wiphy *wiphy, struct net_device *dev,
 }
 
 static int ieee80211_del_key(struct wiphy *wiphy, struct net_device *dev,
-			     u8 key_idx, const u8 *mac_addr)
+			     u8 key_idx, bool pairwise, const u8 *mac_addr)
 {
 	struct ieee80211_sub_if_data *sdata;
 	struct sta_info *sta;
@@ -170,10 +173,17 @@ static int ieee80211_del_key(struct wiphy *wiphy, struct net_device *dev,
 		if (!sta)
 			goto out_unlock;
 
-		if (sta->key) {
-			ieee80211_key_free(sdata->local, sta->key);
-			WARN_ON(sta->key);
-			ret = 0;
+		if (pairwise) {
+			if (sta->ptk) {
+				ieee80211_key_free(sdata->local, sta->ptk);
+				ret = 0;
+			}
+		} else {
+			if (sta->gtk[key_idx]) {
+				ieee80211_key_free(sdata->local,
+						   sta->gtk[key_idx]);
+				ret = 0;
+			}
 		}
 
 		goto out_unlock;
@@ -195,7 +205,8 @@ static int ieee80211_del_key(struct wiphy *wiphy, struct net_device *dev,
 }
 
 static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev,
-			     u8 key_idx, const u8 *mac_addr, void *cookie,
+			     u8 key_idx, bool pairwise, const u8 *mac_addr,
+			     void *cookie,
 			     void (*callback)(void *cookie,
 					      struct key_params *params))
 {
@@ -203,7 +214,7 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev,
 	struct sta_info *sta = NULL;
 	u8 seq[6] = {0};
 	struct key_params params;
-	struct ieee80211_key *key;
+	struct ieee80211_key *key = NULL;
 	u32 iv32;
 	u16 iv16;
 	int err = -ENOENT;
@@ -217,7 +228,10 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev,
 		if (!sta)
 			goto out;
 
-		key = sta->key;
+		if (pairwise)
+			key = sta->ptk;
+		else if (key_idx < NUM_DEFAULT_KEYS)
+			key = sta->gtk[key_idx];
 	} else
 		key = sdata->keys[key_idx];
 
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 76c2b50ec6f8..f0610fa4fbe0 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -549,8 +549,6 @@ struct ieee80211_sub_if_data {
 	struct ieee80211_fragment_entry	fragments[IEEE80211_FRAGMENT_MAX];
 	unsigned int fragment_next;
 
-#define NUM_DEFAULT_KEYS 4
-#define NUM_DEFAULT_MGMT_KEYS 2
 	struct ieee80211_key *keys[NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS];
 	struct ieee80211_key *default_key;
 	struct ieee80211_key *default_mgmt_key;
diff --git a/net/mac80211/key.c b/net/mac80211/key.c
index 6a63d1abd14d..ccd676b2f599 100644
--- a/net/mac80211/key.c
+++ b/net/mac80211/key.c
@@ -68,15 +68,21 @@ static int ieee80211_key_enable_hw_accel(struct ieee80211_key *key)
 
 	might_sleep();
 
-	if (!key->local->ops->set_key) {
-		ret = -EOPNOTSUPP;
+	if (!key->local->ops->set_key)
 		goto out_unsupported;
-	}
 
 	assert_key_lock(key->local);
 
 	sta = get_sta_for_key(key);
 
+	/*
+	 * If this is a per-STA GTK, check if it
+	 * is supported; if not, return.
+	 */
+	if (sta && !(key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE) &&
+	    !(key->local->hw.flags & IEEE80211_HW_SUPPORTS_PER_STA_GTK))
+		goto out_unsupported;
+
 	sdata = key->sdata;
 	if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN)
 		sdata = container_of(sdata->bss,
@@ -85,31 +91,28 @@ static int ieee80211_key_enable_hw_accel(struct ieee80211_key *key)
 
 	ret = drv_set_key(key->local, SET_KEY, sdata, sta, &key->conf);
 
-	if (!ret)
+	if (!ret) {
 		key->flags |= KEY_FLAG_UPLOADED_TO_HARDWARE;
+		return 0;
+	}
 
-	if (ret && ret != -ENOSPC && ret != -EOPNOTSUPP)
+	if (ret != -ENOSPC && ret != -EOPNOTSUPP)
 		wiphy_err(key->local->hw.wiphy,
 			  "failed to set key (%d, %pM) to hardware (%d)\n",
 			  key->conf.keyidx, sta ? sta->addr : bcast_addr, ret);
 
-out_unsupported:
-	if (ret) {
-		switch (key->conf.cipher) {
-		case WLAN_CIPHER_SUITE_WEP40:
-		case WLAN_CIPHER_SUITE_WEP104:
-		case WLAN_CIPHER_SUITE_TKIP:
-		case WLAN_CIPHER_SUITE_CCMP:
-		case WLAN_CIPHER_SUITE_AES_CMAC:
-			/* all of these we can do in software */
-			ret = 0;
-			break;
-		default:
-			ret = -EINVAL;
-		}
+ out_unsupported:
+	switch (key->conf.cipher) {
+	case WLAN_CIPHER_SUITE_WEP40:
+	case WLAN_CIPHER_SUITE_WEP104:
+	case WLAN_CIPHER_SUITE_TKIP:
+	case WLAN_CIPHER_SUITE_CCMP:
+	case WLAN_CIPHER_SUITE_AES_CMAC:
+		/* all of these we can do in software */
+		return 0;
+	default:
+		return -EINVAL;
 	}
-
-	return ret;
 }
 
 static void ieee80211_key_disable_hw_accel(struct ieee80211_key *key)
@@ -147,6 +150,26 @@ static void ieee80211_key_disable_hw_accel(struct ieee80211_key *key)
 	key->flags &= ~KEY_FLAG_UPLOADED_TO_HARDWARE;
 }
 
+void ieee80211_key_removed(struct ieee80211_key_conf *key_conf)
+{
+	struct ieee80211_key *key;
+
+	key = container_of(key_conf, struct ieee80211_key, conf);
+
+	might_sleep();
+	assert_key_lock(key->local);
+
+	key->flags &= ~KEY_FLAG_UPLOADED_TO_HARDWARE;
+
+	/*
+	 * Flush TX path to avoid attempts to use this key
+	 * after this function returns. Until then, drivers
+	 * must be prepared to handle the key.
+	 */
+	synchronize_rcu();
+}
+EXPORT_SYMBOL_GPL(ieee80211_key_removed);
+
 static void __ieee80211_set_default_key(struct ieee80211_sub_if_data *sdata,
 					int idx)
 {
@@ -202,6 +225,7 @@ void ieee80211_set_default_mgmt_key(struct ieee80211_sub_if_data *sdata,
 
 static void __ieee80211_key_replace(struct ieee80211_sub_if_data *sdata,
 				    struct sta_info *sta,
+				    bool pairwise,
 				    struct ieee80211_key *old,
 				    struct ieee80211_key *new)
 {
@@ -210,8 +234,14 @@ static void __ieee80211_key_replace(struct ieee80211_sub_if_data *sdata,
 	if (new)
 		list_add(&new->list, &sdata->key_list);
 
-	if (sta) {
-		rcu_assign_pointer(sta->key, new);
+	if (sta && pairwise) {
+		rcu_assign_pointer(sta->ptk, new);
+	} else if (sta) {
+		if (old)
+			idx = old->conf.keyidx;
+		else
+			idx = new->conf.keyidx;
+		rcu_assign_pointer(sta->gtk[idx], new);
 	} else {
 		WARN_ON(new && old && new->conf.keyidx != old->conf.keyidx);
 
@@ -355,6 +385,7 @@ int ieee80211_key_link(struct ieee80211_key *key,
 {
 	struct ieee80211_key *old_key;
 	int idx, ret;
+	bool pairwise = key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE;
 
 	BUG_ON(!sdata);
 	BUG_ON(!key);
@@ -371,13 +402,6 @@ int ieee80211_key_link(struct ieee80211_key *key,
 		 */
 		if (test_sta_flags(sta, WLAN_STA_WME))
 			key->conf.flags |= IEEE80211_KEY_FLAG_WMM_STA;
-
-		/*
-		 * This key is for a specific sta interface,
-		 * inform the driver that it should try to store
-		 * this key as pairwise key.
-		 */
-		key->conf.flags |= IEEE80211_KEY_FLAG_PAIRWISE;
 	} else {
 		if (sdata->vif.type == NL80211_IFTYPE_STATION) {
 			struct sta_info *ap;
@@ -399,12 +423,14 @@ int ieee80211_key_link(struct ieee80211_key *key,
 
 	mutex_lock(&sdata->local->key_mtx);
 
-	if (sta)
-		old_key = sta->key;
+	if (sta && pairwise)
+		old_key = sta->ptk;
+	else if (sta)
+		old_key = sta->gtk[idx];
 	else
 		old_key = sdata->keys[idx];
 
-	__ieee80211_key_replace(sdata, sta, old_key, key);
+	__ieee80211_key_replace(sdata, sta, pairwise, old_key, key);
 	__ieee80211_key_destroy(old_key);
 
 	ieee80211_debugfs_key_add(key);
@@ -423,7 +449,8 @@ static void __ieee80211_key_free(struct ieee80211_key *key)
 	 */
 	if (key->sdata)
 		__ieee80211_key_replace(key->sdata, key->sta,
-					key, NULL);
+				key->conf.flags & IEEE80211_KEY_FLAG_PAIRWISE,
+				key, NULL);
 	__ieee80211_key_destroy(key);
 }
 
diff --git a/net/mac80211/key.h b/net/mac80211/key.h
index cb9a4a65cc68..0db1c0f5f697 100644
--- a/net/mac80211/key.h
+++ b/net/mac80211/key.h
@@ -16,6 +16,9 @@
 #include <linux/rcupdate.h>
 #include <net/mac80211.h>
 
+#define NUM_DEFAULT_KEYS 4
+#define NUM_DEFAULT_MGMT_KEYS 2
+
 #define WEP_IV_LEN		4
 #define WEP_ICV_LEN		4
 #define ALG_TKIP_KEY_LEN	32
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index b3e161ffa4b3..b67221def584 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -846,7 +846,7 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx)
 	int keyidx;
 	int hdrlen;
 	ieee80211_rx_result result = RX_DROP_UNUSABLE;
-	struct ieee80211_key *stakey = NULL;
+	struct ieee80211_key *sta_ptk = NULL;
 	int mmie_keyidx = -1;
 	__le16 fc;
 
@@ -888,15 +888,15 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx)
 	rx->key = NULL;
 
 	if (rx->sta)
-		stakey = rcu_dereference(rx->sta->key);
+		sta_ptk = rcu_dereference(rx->sta->ptk);
 
 	fc = hdr->frame_control;
 
 	if (!ieee80211_has_protected(fc))
 		mmie_keyidx = ieee80211_get_mmie_keyidx(rx->skb);
 
-	if (!is_multicast_ether_addr(hdr->addr1) && stakey) {
-		rx->key = stakey;
+	if (!is_multicast_ether_addr(hdr->addr1) && sta_ptk) {
+		rx->key = sta_ptk;
 		if ((status->flag & RX_FLAG_DECRYPTED) &&
 		    (status->flag & RX_FLAG_IV_STRIPPED))
 			return RX_CONTINUE;
@@ -912,7 +912,10 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx)
 		if (mmie_keyidx < NUM_DEFAULT_KEYS ||
 		    mmie_keyidx >= NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS)
 			return RX_DROP_MONITOR; /* unexpected BIP keyidx */
-		rx->key = rcu_dereference(rx->sdata->keys[mmie_keyidx]);
+		if (rx->sta)
+			rx->key = rcu_dereference(rx->sta->gtk[mmie_keyidx]);
+		if (!rx->key)
+			rx->key = rcu_dereference(rx->sdata->keys[mmie_keyidx]);
 	} else if (!ieee80211_has_protected(fc)) {
 		/*
 		 * The frame was not protected, so skip decryption. However, we
@@ -955,17 +958,25 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx)
 		skb_copy_bits(rx->skb, hdrlen + 3, &keyid, 1);
 		keyidx = keyid >> 6;
 
-		rx->key = rcu_dereference(rx->sdata->keys[keyidx]);
+		/* check per-station GTK first, if multicast packet */
+		if (is_multicast_ether_addr(hdr->addr1) && rx->sta)
+			rx->key = rcu_dereference(rx->sta->gtk[keyidx]);
 
-		/*
-		 * RSNA-protected unicast frames should always be sent with
-		 * pairwise or station-to-station keys, but for WEP we allow
-		 * using a key index as well.
-		 */
-		if (rx->key && rx->key->conf.cipher != WLAN_CIPHER_SUITE_WEP40 &&
-		    rx->key->conf.cipher != WLAN_CIPHER_SUITE_WEP104 &&
-		    !is_multicast_ether_addr(hdr->addr1))
-			rx->key = NULL;
+		/* if not found, try default key */
+		if (!rx->key) {
+			rx->key = rcu_dereference(rx->sdata->keys[keyidx]);
+
+			/*
+			 * RSNA-protected unicast frames should always be
+			 * sent with pairwise or station-to-station keys,
+			 * but for WEP we allow using a key index as well.
+			 */
+			if (rx->key &&
+			    rx->key->conf.cipher != WLAN_CIPHER_SUITE_WEP40 &&
+			    rx->key->conf.cipher != WLAN_CIPHER_SUITE_WEP104 &&
+			    !is_multicast_ether_addr(hdr->addr1))
+				rx->key = NULL;
+		}
 	}
 
 	if (rx->key) {
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index aeaf2d6fccc8..6d8f897d8763 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -616,7 +616,7 @@ static int __must_check __sta_info_destroy(struct sta_info *sta)
 	struct ieee80211_sub_if_data *sdata;
 	struct sk_buff *skb;
 	unsigned long flags;
-	int ret;
+	int ret, i;
 
 	might_sleep();
 
@@ -644,10 +644,10 @@ static int __must_check __sta_info_destroy(struct sta_info *sta)
 	if (ret)
 		return ret;
 
-	if (sta->key) {
-		ieee80211_key_free(local, sta->key);
-		WARN_ON(sta->key);
-	}
+	for (i = 0; i < NUM_DEFAULT_KEYS; i++)
+		ieee80211_key_free(local, sta->gtk[i]);
+	if (sta->ptk)
+		ieee80211_key_free(local, sta->ptk);
 
 	sta->dead = true;
 
diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index cf21a2e8134f..9265acadef32 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -199,7 +199,8 @@ enum plink_state {
  * @hnext: hash table linked list pointer
  * @local: pointer to the global information
  * @sdata: virtual interface this station belongs to
- * @key: peer key negotiated with this station, if any
+ * @ptk: peer key negotiated with this station, if any
+ * @gtk: group keys negotiated with this station, if any
  * @rate_ctrl: rate control algorithm reference
  * @rate_ctrl_priv: rate control private per-STA pointer
  * @last_tx_rate: rate used for last transmit, to report to userspace as
@@ -254,7 +255,8 @@ struct sta_info {
 	struct sta_info *hnext;
 	struct ieee80211_local *local;
 	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_key *key;
+	struct ieee80211_key *gtk[NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS];
+	struct ieee80211_key *ptk;
 	struct rate_control_ref *rate_ctrl;
 	void *rate_ctrl_priv;
 	spinlock_t lock;
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index 258fbdbedbdf..96c594309506 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -532,7 +532,7 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx)
 
 	if (unlikely(info->flags & IEEE80211_TX_INTFL_DONT_ENCRYPT))
 		tx->key = NULL;
-	else if (tx->sta && (key = rcu_dereference(tx->sta->key)))
+	else if (tx->sta && (key = rcu_dereference(tx->sta->ptk)))
 		tx->key = key;
 	else if (ieee80211_is_mgmt(hdr->frame_control) &&
 		 is_multicast_ether_addr(hdr->addr1) &&
diff --git a/net/wireless/core.h b/net/wireless/core.h
index 37580e090a3d..2d1d4c70113c 100644
--- a/net/wireless/core.h
+++ b/net/wireless/core.h
@@ -375,7 +375,7 @@ bool cfg80211_sme_failed_reassoc(struct wireless_dev *wdev);
 /* internal helpers */
 int cfg80211_validate_key_settings(struct cfg80211_registered_device *rdev,
 				   struct key_params *params, int key_idx,
-				   const u8 *mac_addr);
+				   bool pairwise, const u8 *mac_addr);
 void __cfg80211_disconnected(struct net_device *dev, const u8 *ie,
 			     size_t ie_len, u16 reason, bool from_ap);
 void cfg80211_sme_scan_done(struct net_device *dev);
diff --git a/net/wireless/ibss.c b/net/wireless/ibss.c
index 8cb6e08373b9..f33fbb79437c 100644
--- a/net/wireless/ibss.c
+++ b/net/wireless/ibss.c
@@ -160,7 +160,7 @@ static void __cfg80211_clear_ibss(struct net_device *dev, bool nowext)
 	 */
 	if (rdev->ops->del_key)
 		for (i = 0; i < 6; i++)
-			rdev->ops->del_key(wdev->wiphy, dev, i, NULL);
+			rdev->ops->del_key(wdev->wiphy, dev, i, false, NULL);
 
 	if (wdev->current_bss) {
 		cfg80211_unhold_bss(wdev->current_bss);
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 0c9497170f1f..8826888cc14e 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -93,6 +93,7 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = {
 	[NL80211_ATTR_KEY_CIPHER] = { .type = NLA_U32 },
 	[NL80211_ATTR_KEY_DEFAULT] = { .type = NLA_FLAG },
 	[NL80211_ATTR_KEY_SEQ] = { .type = NLA_BINARY, .len = 8 },
+	[NL80211_ATTR_KEY_TYPE] = { .type = NLA_U32 },
 
 	[NL80211_ATTR_BEACON_INTERVAL] = { .type = NLA_U32 },
 	[NL80211_ATTR_DTIM_PERIOD] = { .type = NLA_U32 },
@@ -168,7 +169,7 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = {
 	[NL80211_ATTR_FRAME_TYPE] = { .type = NLA_U16 },
 };
 
-/* policy for the attributes */
+/* policy for the key attributes */
 static const struct nla_policy nl80211_key_policy[NL80211_KEY_MAX + 1] = {
 	[NL80211_KEY_DATA] = { .type = NLA_BINARY, .len = WLAN_MAX_KEY_LEN },
 	[NL80211_KEY_IDX] = { .type = NLA_U8 },
@@ -176,6 +177,7 @@ static const struct nla_policy nl80211_key_policy[NL80211_KEY_MAX + 1] = {
 	[NL80211_KEY_SEQ] = { .type = NLA_BINARY, .len = 8 },
 	[NL80211_KEY_DEFAULT] = { .type = NLA_FLAG },
 	[NL80211_KEY_DEFAULT_MGMT] = { .type = NLA_FLAG },
+	[NL80211_KEY_TYPE] = { .type = NLA_U32 },
 };
 
 /* ifidx get helper */
@@ -306,6 +308,7 @@ static int nl80211_msg_put_channel(struct sk_buff *msg,
 struct key_parse {
 	struct key_params p;
 	int idx;
+	int type;
 	bool def, defmgmt;
 };
 
@@ -336,6 +339,12 @@ static int nl80211_parse_key_new(struct nlattr *key, struct key_parse *k)
 	if (tb[NL80211_KEY_CIPHER])
 		k->p.cipher = nla_get_u32(tb[NL80211_KEY_CIPHER]);
 
+	if (tb[NL80211_KEY_TYPE]) {
+		k->type = nla_get_u32(tb[NL80211_KEY_TYPE]);
+		if (k->type < 0 || k->type >= NUM_NL80211_KEYTYPES)
+			return -EINVAL;
+	}
+
 	return 0;
 }
 
@@ -360,6 +369,12 @@ static int nl80211_parse_key_old(struct genl_info *info, struct key_parse *k)
 	k->def = !!info->attrs[NL80211_ATTR_KEY_DEFAULT];
 	k->defmgmt = !!info->attrs[NL80211_ATTR_KEY_DEFAULT_MGMT];
 
+	if (info->attrs[NL80211_ATTR_KEY_TYPE]) {
+		k->type = nla_get_u32(info->attrs[NL80211_ATTR_KEY_TYPE]);
+		if (k->type < 0 || k->type >= NUM_NL80211_KEYTYPES)
+			return -EINVAL;
+	}
+
 	return 0;
 }
 
@@ -369,6 +384,7 @@ static int nl80211_parse_key(struct genl_info *info, struct key_parse *k)
 
 	memset(k, 0, sizeof(*k));
 	k->idx = -1;
+	k->type = -1;
 
 	if (info->attrs[NL80211_ATTR_KEY])
 		err = nl80211_parse_key_new(info->attrs[NL80211_ATTR_KEY], k);
@@ -433,7 +449,7 @@ nl80211_parse_connkeys(struct cfg80211_registered_device *rdev,
 		} else if (parse.defmgmt)
 			goto error;
 		err = cfg80211_validate_key_settings(rdev, &parse.p,
-						     parse.idx, NULL);
+						     parse.idx, false, NULL);
 		if (err)
 			goto error;
 		result->params[parse.idx].cipher = parse.p.cipher;
@@ -516,6 +532,9 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags,
 	NLA_PUT_U16(msg, NL80211_ATTR_MAX_SCAN_IE_LEN,
 		    dev->wiphy.max_scan_ie_len);
 
+	if (dev->wiphy.flags & WIPHY_FLAG_IBSS_RSN)
+		NLA_PUT_FLAG(msg, NL80211_ATTR_SUPPORT_IBSS_RSN);
+
 	NLA_PUT(msg, NL80211_ATTR_CIPHER_SUITES,
 		sizeof(u32) * dev->wiphy.n_cipher_suites,
 		dev->wiphy.cipher_suites);
@@ -1446,7 +1465,8 @@ static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info)
 	int err;
 	struct net_device *dev = info->user_ptr[1];
 	u8 key_idx = 0;
-	u8 *mac_addr = NULL;
+	const u8 *mac_addr = NULL;
+	bool pairwise;
 	struct get_key_cookie cookie = {
 		.error = 0,
 	};
@@ -1462,6 +1482,17 @@ static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info)
 	if (info->attrs[NL80211_ATTR_MAC])
 		mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
 
+	pairwise = !!mac_addr;
+	if (info->attrs[NL80211_ATTR_KEY_TYPE]) {
+		u32 kt = nla_get_u32(info->attrs[NL80211_ATTR_KEY_TYPE]);
+		if (kt >= NUM_NL80211_KEYTYPES)
+			return -EINVAL;
+		if (kt != NL80211_KEYTYPE_GROUP &&
+		    kt != NL80211_KEYTYPE_PAIRWISE)
+			return -EINVAL;
+		pairwise = kt == NL80211_KEYTYPE_PAIRWISE;
+	}
+
 	if (!rdev->ops->get_key)
 		return -EOPNOTSUPP;
 
@@ -1482,8 +1513,12 @@ static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info)
 	if (mac_addr)
 		NLA_PUT(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr);
 
-	err = rdev->ops->get_key(&rdev->wiphy, dev, key_idx, mac_addr,
-				&cookie, get_key_callback);
+	if (pairwise && mac_addr &&
+	    !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN))
+		return -ENOENT;
+
+	err = rdev->ops->get_key(&rdev->wiphy, dev, key_idx, pairwise,
+				 mac_addr, &cookie, get_key_callback);
 
 	if (err)
 		goto free_msg;
@@ -1553,7 +1588,7 @@ static int nl80211_new_key(struct sk_buff *skb, struct genl_info *info)
 	int err;
 	struct net_device *dev = info->user_ptr[1];
 	struct key_parse key;
-	u8 *mac_addr = NULL;
+	const u8 *mac_addr = NULL;
 
 	err = nl80211_parse_key(info, &key);
 	if (err)
@@ -1565,16 +1600,31 @@ static int nl80211_new_key(struct sk_buff *skb, struct genl_info *info)
 	if (info->attrs[NL80211_ATTR_MAC])
 		mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
 
+	if (key.type == -1) {
+		if (mac_addr)
+			key.type = NL80211_KEYTYPE_PAIRWISE;
+		else
+			key.type = NL80211_KEYTYPE_GROUP;
+	}
+
+	/* for now */
+	if (key.type != NL80211_KEYTYPE_PAIRWISE &&
+	    key.type != NL80211_KEYTYPE_GROUP)
+		return -EINVAL;
+
 	if (!rdev->ops->add_key)
 		return -EOPNOTSUPP;
 
-	if (cfg80211_validate_key_settings(rdev, &key.p, key.idx, mac_addr))
+	if (cfg80211_validate_key_settings(rdev, &key.p, key.idx,
+					   key.type == NL80211_KEYTYPE_PAIRWISE,
+					   mac_addr))
 		return -EINVAL;
 
 	wdev_lock(dev->ieee80211_ptr);
 	err = nl80211_key_allowed(dev->ieee80211_ptr);
 	if (!err)
 		err = rdev->ops->add_key(&rdev->wiphy, dev, key.idx,
+					 key.type == NL80211_KEYTYPE_PAIRWISE,
 					 mac_addr, &key.p);
 	wdev_unlock(dev->ieee80211_ptr);
 
@@ -1596,13 +1646,32 @@ static int nl80211_del_key(struct sk_buff *skb, struct genl_info *info)
 	if (info->attrs[NL80211_ATTR_MAC])
 		mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
 
+	if (key.type == -1) {
+		if (mac_addr)
+			key.type = NL80211_KEYTYPE_PAIRWISE;
+		else
+			key.type = NL80211_KEYTYPE_GROUP;
+	}
+
+	/* for now */
+	if (key.type != NL80211_KEYTYPE_PAIRWISE &&
+	    key.type != NL80211_KEYTYPE_GROUP)
+		return -EINVAL;
+
 	if (!rdev->ops->del_key)
 		return -EOPNOTSUPP;
 
 	wdev_lock(dev->ieee80211_ptr);
 	err = nl80211_key_allowed(dev->ieee80211_ptr);
+
+	if (key.type == NL80211_KEYTYPE_PAIRWISE && mac_addr &&
+	    !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN))
+		err = -ENOENT;
+
 	if (!err)
-		err = rdev->ops->del_key(&rdev->wiphy, dev, key.idx, mac_addr);
+		err = rdev->ops->del_key(&rdev->wiphy, dev, key.idx,
+					 key.type == NL80211_KEYTYPE_PAIRWISE,
+					 mac_addr);
 
 #ifdef CONFIG_CFG80211_WEXT
 	if (!err) {
@@ -3212,6 +3281,8 @@ static int nl80211_authenticate(struct sk_buff *skb, struct genl_info *info)
 		return err;
 
 	if (key.idx >= 0) {
+		if (key.type != -1 && key.type != NL80211_KEYTYPE_GROUP)
+			return -EINVAL;
 		if (!key.p.key || !key.p.key_len)
 			return -EINVAL;
 		if ((key.p.cipher != WLAN_CIPHER_SUITE_WEP40 ||
diff --git a/net/wireless/sme.c b/net/wireless/sme.c
index f161b9844542..e17b0bee6bdc 100644
--- a/net/wireless/sme.c
+++ b/net/wireless/sme.c
@@ -698,7 +698,7 @@ void __cfg80211_disconnected(struct net_device *dev, const u8 *ie,
 	 */
 	if (rdev->ops->del_key)
 		for (i = 0; i < 6; i++)
-			rdev->ops->del_key(wdev->wiphy, dev, i, NULL);
+			rdev->ops->del_key(wdev->wiphy, dev, i, false, NULL);
 
 #ifdef CONFIG_CFG80211_WEXT
 	memset(&wrqu, 0, sizeof(wrqu));
diff --git a/net/wireless/util.c b/net/wireless/util.c
index fb5448f7d55a..76120aeda57d 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -144,19 +144,25 @@ void ieee80211_set_bitrate_flags(struct wiphy *wiphy)
 
 int cfg80211_validate_key_settings(struct cfg80211_registered_device *rdev,
 				   struct key_params *params, int key_idx,
-				   const u8 *mac_addr)
+				   bool pairwise, const u8 *mac_addr)
 {
 	int i;
 
 	if (key_idx > 5)
 		return -EINVAL;
 
+	if (!pairwise && mac_addr && !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN))
+		return -EINVAL;
+
+	if (pairwise && !mac_addr)
+		return -EINVAL;
+
 	/*
 	 * Disallow pairwise keys with non-zero index unless it's WEP
 	 * (because current deployments use pairwise WEP keys with
 	 * non-zero indizes but 802.11i clearly specifies to use zero)
 	 */
-	if (mac_addr && key_idx &&
+	if (pairwise && key_idx &&
 	    params->cipher != WLAN_CIPHER_SUITE_WEP40 &&
 	    params->cipher != WLAN_CIPHER_SUITE_WEP104)
 		return -EINVAL;
@@ -677,7 +683,7 @@ void cfg80211_upload_connect_keys(struct wireless_dev *wdev)
 	for (i = 0; i < 6; i++) {
 		if (!wdev->connect_keys->params[i].cipher)
 			continue;
-		if (rdev->ops->add_key(wdev->wiphy, dev, i, NULL,
+		if (rdev->ops->add_key(wdev->wiphy, dev, i, false, NULL,
 					&wdev->connect_keys->params[i])) {
 			printk(KERN_ERR "%s: failed to set key %d\n",
 				dev->name, i);
diff --git a/net/wireless/wext-compat.c b/net/wireless/wext-compat.c
index 7e5c3a45f811..6002265289c6 100644
--- a/net/wireless/wext-compat.c
+++ b/net/wireless/wext-compat.c
@@ -432,14 +432,17 @@ int cfg80211_wext_giwretry(struct net_device *dev,
 EXPORT_SYMBOL_GPL(cfg80211_wext_giwretry);
 
 static int __cfg80211_set_encryption(struct cfg80211_registered_device *rdev,
-				     struct net_device *dev, const u8 *addr,
-				     bool remove, bool tx_key, int idx,
-				     struct key_params *params)
+				     struct net_device *dev, bool pairwise,
+				     const u8 *addr, bool remove, bool tx_key,
+				     int idx, struct key_params *params)
 {
 	struct wireless_dev *wdev = dev->ieee80211_ptr;
 	int err, i;
 	bool rejoin = false;
 
+	if (pairwise && !addr)
+		return -EINVAL;
+
 	if (!wdev->wext.keys) {
 		wdev->wext.keys = kzalloc(sizeof(*wdev->wext.keys),
 					      GFP_KERNEL);
@@ -478,7 +481,13 @@ static int __cfg80211_set_encryption(struct cfg80211_registered_device *rdev,
 				__cfg80211_leave_ibss(rdev, wdev->netdev, true);
 				rejoin = true;
 			}
-			err = rdev->ops->del_key(&rdev->wiphy, dev, idx, addr);
+
+			if (!pairwise && addr &&
+			    !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN))
+				err = -ENOENT;
+			else
+				err = rdev->ops->del_key(&rdev->wiphy, dev, idx,
+							 pairwise, addr);
 		}
 		wdev->wext.connect.privacy = false;
 		/*
@@ -507,12 +516,13 @@ static int __cfg80211_set_encryption(struct cfg80211_registered_device *rdev,
 	if (addr)
 		tx_key = false;
 
-	if (cfg80211_validate_key_settings(rdev, params, idx, addr))
+	if (cfg80211_validate_key_settings(rdev, params, idx, pairwise, addr))
 		return -EINVAL;
 
 	err = 0;
 	if (wdev->current_bss)
-		err = rdev->ops->add_key(&rdev->wiphy, dev, idx, addr, params);
+		err = rdev->ops->add_key(&rdev->wiphy, dev, idx,
+					 pairwise, addr, params);
 	if (err)
 		return err;
 
@@ -563,17 +573,17 @@ static int __cfg80211_set_encryption(struct cfg80211_registered_device *rdev,
 }
 
 static int cfg80211_set_encryption(struct cfg80211_registered_device *rdev,
-				   struct net_device *dev, const u8 *addr,
-				   bool remove, bool tx_key, int idx,
-				   struct key_params *params)
+				   struct net_device *dev, bool pairwise,
+				   const u8 *addr, bool remove, bool tx_key,
+				   int idx, struct key_params *params)
 {
 	int err;
 
 	/* devlist mutex needed for possible IBSS re-join */
 	mutex_lock(&rdev->devlist_mtx);
 	wdev_lock(dev->ieee80211_ptr);
-	err = __cfg80211_set_encryption(rdev, dev, addr, remove,
-					tx_key, idx, params);
+	err = __cfg80211_set_encryption(rdev, dev, pairwise, addr,
+					remove, tx_key, idx, params);
 	wdev_unlock(dev->ieee80211_ptr);
 	mutex_unlock(&rdev->devlist_mtx);
 
@@ -635,7 +645,7 @@ int cfg80211_wext_siwencode(struct net_device *dev,
 	else if (!remove)
 		return -EINVAL;
 
-	return cfg80211_set_encryption(rdev, dev, NULL, remove,
+	return cfg80211_set_encryption(rdev, dev, false, NULL, remove,
 				       wdev->wext.default_key == -1,
 				       idx, &params);
 }
@@ -725,7 +735,9 @@ int cfg80211_wext_siwencodeext(struct net_device *dev,
 	}
 
 	return cfg80211_set_encryption(
-			rdev, dev, addr, remove,
+			rdev, dev,
+			!(ext->ext_flags & IW_ENCODE_EXT_GROUP_KEY),
+			addr, remove,
 			ext->ext_flags & IW_ENCODE_EXT_SET_TX_KEY,
 			idx, &params);
 }
-- 
cgit v1.2.3


From b206b4ef062d83c0875a085672ed50e8c8b01521 Mon Sep 17 00:00:00 2001
From: Bruno Randolf <br1@einfach.org>
Date: Wed, 6 Oct 2010 18:34:12 +0900
Subject: nl80211/mac80211: Add retry and failed transmission count to station
 info

This information is already available in mac80211, we just need to export it
via cfg80211 and nl80211.

Signed-off-by: Bruno Randolf <br1@einfach.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 4 ++++
 include/net/cfg80211.h  | 8 ++++++++
 net/mac80211/cfg.c      | 4 ++++
 net/wireless/nl80211.c  | 6 ++++++
 4 files changed, 22 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index e451f176e662..c08709fe36fc 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -1137,6 +1137,8 @@ enum nl80211_rate_info {
  * @NL80211_STA_INFO_RX_PACKETS: total received packet (u32, from this station)
  * @NL80211_STA_INFO_TX_PACKETS: total transmitted packets (u32, to this
  *	station)
+ * @NL80211_STA_INFO_TX_RETRIES: total retries (u32, to this station)
+ * @NL80211_STA_INFO_TX_FAILED: total failed packets (u32, to this station)
  */
 enum nl80211_sta_info {
 	__NL80211_STA_INFO_INVALID,
@@ -1150,6 +1152,8 @@ enum nl80211_sta_info {
 	NL80211_STA_INFO_TX_BITRATE,
 	NL80211_STA_INFO_RX_PACKETS,
 	NL80211_STA_INFO_TX_PACKETS,
+	NL80211_STA_INFO_TX_RETRIES,
+	NL80211_STA_INFO_TX_FAILED,
 
 	/* keep last */
 	__NL80211_STA_INFO_AFTER_LAST,
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 0f77515266b8..e76daaa7dc25 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -401,6 +401,8 @@ struct station_parameters {
  *  (tx_bitrate, tx_bitrate_flags and tx_bitrate_mcs)
  * @STATION_INFO_RX_PACKETS: @rx_packets filled
  * @STATION_INFO_TX_PACKETS: @tx_packets filled
+ * @STATION_INFO_TX_RETRIES: @tx_retries filled
+ * @STATION_INFO_TX_FAILED: @tx_failed filled
  */
 enum station_info_flags {
 	STATION_INFO_INACTIVE_TIME	= 1<<0,
@@ -413,6 +415,8 @@ enum station_info_flags {
 	STATION_INFO_TX_BITRATE		= 1<<7,
 	STATION_INFO_RX_PACKETS		= 1<<8,
 	STATION_INFO_TX_PACKETS		= 1<<9,
+	STATION_INFO_TX_RETRIES		= 1<<10,
+	STATION_INFO_TX_FAILED		= 1<<11,
 };
 
 /**
@@ -462,6 +466,8 @@ struct rate_info {
  * @txrate: current unicast bitrate to this station
  * @rx_packets: packets received from this station
  * @tx_packets: packets transmitted to this station
+ * @tx_retries: cumulative retry counts
+ * @tx_failed: number of failed transmissions (retries exceeded, no ACK)
  * @generation: generation number for nl80211 dumps.
  *	This number should increase every time the list of stations
  *	changes, i.e. when a station is added or removed, so that
@@ -479,6 +485,8 @@ struct station_info {
 	struct rate_info txrate;
 	u32 rx_packets;
 	u32 tx_packets;
+	u32 tx_retries;
+	u32 tx_failed;
 
 	int generation;
 };
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 8b0e874a3d65..2e5a3fb38efe 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -327,6 +327,8 @@ static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo)
 			STATION_INFO_TX_BYTES |
 			STATION_INFO_RX_PACKETS |
 			STATION_INFO_TX_PACKETS |
+			STATION_INFO_TX_RETRIES |
+			STATION_INFO_TX_FAILED |
 			STATION_INFO_TX_BITRATE;
 
 	sinfo->inactive_time = jiffies_to_msecs(jiffies - sta->last_rx);
@@ -334,6 +336,8 @@ static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo)
 	sinfo->tx_bytes = sta->tx_bytes;
 	sinfo->rx_packets = sta->rx_packets;
 	sinfo->tx_packets = sta->tx_packets;
+	sinfo->tx_retries = sta->tx_retry_count;
+	sinfo->tx_failed = sta->tx_retry_failed;
 
 	if ((sta->local->hw.flags & IEEE80211_HW_SIGNAL_DBM) ||
 	    (sta->local->hw.flags & IEEE80211_HW_SIGNAL_UNSPEC)) {
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 9942f0b061ff..524f55402838 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1890,6 +1890,12 @@ static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq,
 	if (sinfo->filled & STATION_INFO_TX_PACKETS)
 		NLA_PUT_U32(msg, NL80211_STA_INFO_TX_PACKETS,
 			    sinfo->tx_packets);
+	if (sinfo->filled & STATION_INFO_TX_RETRIES)
+		NLA_PUT_U32(msg, NL80211_STA_INFO_TX_RETRIES,
+			    sinfo->tx_retries);
+	if (sinfo->filled & STATION_INFO_TX_FAILED)
+		NLA_PUT_U32(msg, NL80211_STA_INFO_TX_FAILED,
+			    sinfo->tx_failed);
 	nla_nest_end(msg, sinfoattr);
 
 	return genlmsg_end(msg, hdr);
-- 
cgit v1.2.3


From 430c62fb2948d964cf8dc7f3e2f69623c04ef62f Mon Sep 17 00:00:00 2001
From: Jens Axboe <jaxboe@fusionio.com>
Date: Thu, 7 Oct 2010 09:35:16 +0200
Subject: elevator: fix oops on early call to elevator_change()

2.6.36 introduces an API for drivers to switch the IO scheduler
instead of manually calling the elevator exit and init functions.
This API was added since q->elevator must be cleared in between
those two calls. And since we already have this functionality
directly from use by the sysfs interface to switch schedulers
online, it was prudent to reuse it internally too.

But this API needs the queue to be in a fully initialized state
before it is called, or it will attempt to unregister elevator
kobjects before they have been added. This results in an oops
like this:

BUG: unable to handle kernel NULL pointer dereference at 0000000000000051
IP: [<ffffffff8116f15e>] sysfs_create_dir+0x2e/0xc0
PGD 47ddfc067 PUD 47c6a1067 PMD 0
Oops: 0000 [#1] PREEMPT SMP
last sysfs file: /sys/devices/pci0000:00/0000:00:02.0/0000:04:00.1/irq
CPU 2
Modules linked in: t(+) loop hid_apple usbhid ahci ehci_hcd uhci_hcd libahci usbcore nls_base igb

Pid: 7319, comm: modprobe Not tainted 2.6.36-rc6+ #132 QSSC-S4R/QSSC-S4R
RIP: 0010:[<ffffffff8116f15e>]  [<ffffffff8116f15e>] sysfs_create_dir+0x2e/0xc0
RSP: 0018:ffff88027da25d08  EFLAGS: 00010246
RAX: ffff88047c68c528 RBX: 00000000fffffffe RCX: 0000000000000000
RDX: 000000000000002f RSI: 000000000000002f RDI: ffff88047e196c88
RBP: ffff88027da25d38 R08: 0000000000000000 R09: d84156c5635688c0
R10: d84156c5635688c0 R11: 0000000000000000 R12: ffff88047e196c88
R13: 0000000000000000 R14: 0000000000000000 R15: ffff88047c68c528
FS:  00007fcb0b26f6e0(0000) GS:ffff880287400000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000051 CR3: 000000047e76e000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process modprobe (pid: 7319, threadinfo ffff88027da24000, task ffff88027d377090)
Stack:
 ffff88027da25d58 ffff88047c68c528 00000000fffffffe ffff88047e196c88
<0> ffff88047c68c528 ffff88047e05bd90 ffff88027da25d78 ffffffff8123fb77
<0> ffff88047e05bd90 0000000000000000 ffff88047e196c88 ffff88047c68c528
Call Trace:
 [<ffffffff8123fb77>] kobject_add_internal+0xe7/0x1f0
 [<ffffffff8123fd98>] kobject_add_varg+0x38/0x60
 [<ffffffff8123feb9>] kobject_add+0x69/0x90
 [<ffffffff8116efe0>] ? sysfs_remove_dir+0x20/0xa0
 [<ffffffff8103d48d>] ? sub_preempt_count+0x9d/0xe0
 [<ffffffff8143de20>] ? _raw_spin_unlock+0x30/0x50
 [<ffffffff8116efe0>] ? sysfs_remove_dir+0x20/0xa0
 [<ffffffff8116eff4>] ? sysfs_remove_dir+0x34/0xa0
 [<ffffffff81224204>] elv_register_queue+0x34/0xa0
 [<ffffffff81224aad>] elevator_change+0xfd/0x250
 [<ffffffffa007e000>] ? t_init+0x0/0x361 [t]
 [<ffffffffa007e000>] ? t_init+0x0/0x361 [t]
 [<ffffffffa007e0a8>] t_init+0xa8/0x361 [t]
 [<ffffffff810001de>] do_one_initcall+0x3e/0x170
 [<ffffffff8108c3fd>] sys_init_module+0xbd/0x220
 [<ffffffff81002f2b>] system_call_fastpath+0x16/0x1b
Code: e5 41 56 41 55 41 54 49 89 fc 53 48 83 ec 10 48 85 ff 74 52 48 8b 47 18 49 c7 c5 00 46 61 81 48 85 c0 74 04 4c 8b 68 30 45 31 f6 <41> 80 7d 51 00 74 0e 49 8b 44 24 28 4c 89 e7 ff 50 20 49 89 c6
RIP  [<ffffffff8116f15e>] sysfs_create_dir+0x2e/0xc0
 RSP <ffff88027da25d08>
CR2: 0000000000000051
---[ end trace a6541d3bf07945df ]---

Fix this by adding a registered bit to the elevator queue, which is
set when the sysfs kobjects have been registered.

Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/elevator.c         | 12 ++++++++----
 include/linux/elevator.h |  1 +
 2 files changed, 9 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/block/elevator.c b/block/elevator.c
index 205b09a5bd9e..4e11559aa2b0 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -938,6 +938,7 @@ int elv_register_queue(struct request_queue *q)
 			}
 		}
 		kobject_uevent(&e->kobj, KOBJ_ADD);
+		e->registered = 1;
 	}
 	return error;
 }
@@ -947,6 +948,7 @@ static void __elv_unregister_queue(struct elevator_queue *e)
 {
 	kobject_uevent(&e->kobj, KOBJ_REMOVE);
 	kobject_del(&e->kobj);
+	e->registered = 0;
 }
 
 void elv_unregister_queue(struct request_queue *q)
@@ -1042,11 +1044,13 @@ static int elevator_switch(struct request_queue *q, struct elevator_type *new_e)
 
 	spin_unlock_irq(q->queue_lock);
 
-	__elv_unregister_queue(old_elevator);
+	if (old_elevator->registered) {
+		__elv_unregister_queue(old_elevator);
 
-	err = elv_register_queue(q);
-	if (err)
-		goto fail_register;
+		err = elv_register_queue(q);
+		if (err)
+			goto fail_register;
+	}
 
 	/*
 	 * finally exit old elevator and turn off BYPASS.
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 926b50322a46..4fd978e7eb83 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -93,6 +93,7 @@ struct elevator_queue
 	struct elevator_type *elevator_type;
 	struct mutex sysfs_lock;
 	struct hlist_head *hash;
+	unsigned int registered:1;
 };
 
 /*
-- 
cgit v1.2.3


From bcdb714c8856c76383ca455294f0074168705eab Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Thu, 7 Oct 2010 14:08:53 +0100
Subject: Drop a couple of unnecessary asm/system.h inclusions

Drop inclusions of asm/system.h from linux/hardirq.h and linux/list.h as
they're no longer required and prevent the M68K arch's IRQ flag handling macros
from being made into inlined functions due to circular dependencies.

Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Greg Ungerer <gerg@uclinux.org>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
---
 include/linux/hardirq.h | 1 -
 include/linux/list.h    | 1 -
 2 files changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index d5b387669dab..7dfdc06c7e18 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -8,7 +8,6 @@
 #include <linux/lockdep.h>
 #include <linux/ftrace_irq.h>
 #include <asm/hardirq.h>
-#include <asm/system.h>
 
 /*
  * We put the hardirq and softirq counter into the preemption
diff --git a/include/linux/list.h b/include/linux/list.h
index d167b5d7c0ac..88a000617d77 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -5,7 +5,6 @@
 #include <linux/stddef.h>
 #include <linux/poison.h>
 #include <linux/prefetch.h>
-#include <asm/system.h>
 
 /*
  * Simple doubly linked list implementation.
-- 
cgit v1.2.3


From df9ee29270c11dba7d0fe0b83ce47a4d8e8d2101 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Thu, 7 Oct 2010 14:08:55 +0100
Subject: Fix IRQ flag handling naming

Fix the IRQ flag handling naming.  In linux/irqflags.h under one configuration,
it maps:

	local_irq_enable() -> raw_local_irq_enable()
	local_irq_disable() -> raw_local_irq_disable()
	local_irq_save() -> raw_local_irq_save()
	...

and under the other configuration, it maps:

	raw_local_irq_enable() -> local_irq_enable()
	raw_local_irq_disable() -> local_irq_disable()
	raw_local_irq_save() -> local_irq_save()
	...

This is quite confusing.  There should be one set of names expected of the
arch, and this should be wrapped to give another set of names that are expected
by users of this facility.

Change this to have the arch provide:

	flags = arch_local_save_flags()
	flags = arch_local_irq_save()
	arch_local_irq_restore(flags)
	arch_local_irq_disable()
	arch_local_irq_enable()
	arch_irqs_disabled_flags(flags)
	arch_irqs_disabled()
	arch_safe_halt()

Then linux/irqflags.h wraps these to provide:

	raw_local_save_flags(flags)
	raw_local_irq_save(flags)
	raw_local_irq_restore(flags)
	raw_local_irq_disable()
	raw_local_irq_enable()
	raw_irqs_disabled_flags(flags)
	raw_irqs_disabled()
	raw_safe_halt()

with type checking on the flags 'arguments', and then wraps those to provide:

	local_save_flags(flags)
	local_irq_save(flags)
	local_irq_restore(flags)
	local_irq_disable()
	local_irq_enable()
	irqs_disabled_flags(flags)
	irqs_disabled()
	safe_halt()

with tracing included if enabled.

The arch functions can now all be inline functions rather than some of them
having to be macros.

Signed-off-by: David Howells <dhowells@redhat.com> [X86, FRV, MN10300]
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com> [Tile]
Signed-off-by: Michal Simek <monstr@monstr.eu> [Microblaze]
Tested-by: Catalin Marinas <catalin.marinas@arm.com> [ARM]
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com> [AVR]
Acked-by: Tony Luck <tony.luck@intel.com> [IA-64]
Acked-by: Hirokazu Takata <takata@linux-m32r.org> [M32R]
Acked-by: Greg Ungerer <gerg@uclinux.org> [M68K/M68KNOMMU]
Acked-by: Ralf Baechle <ralf@linux-mips.org> [MIPS]
Acked-by: Kyle McMartin <kyle@mcmartin.ca> [PA-RISC]
Acked-by: Paul Mackerras <paulus@samba.org> [PowerPC]
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> [S390]
Acked-by: Chen Liqin <liqin.chen@sunplusct.com> [Score]
Acked-by: Matt Fleming <matt@console-pimps.org> [SH]
Acked-by: David S. Miller <davem@davemloft.net> [Sparc]
Acked-by: Chris Zankel <chris@zankel.net> [Xtensa]
Reviewed-by: Richard Henderson <rth@twiddle.net> [Alpha]
Reviewed-by: Yoshinori Sato <ysato@users.sourceforge.jp> [H8300]
Cc: starvik@axis.com [CRIS]
Cc: jesper.nilsson@axis.com [CRIS]
Cc: linux-cris-kernel@axis.com
---
 arch/alpha/include/asm/irqflags.h          |  67 ++++++++++
 arch/alpha/include/asm/system.h            |  28 -----
 arch/arm/include/asm/irqflags.h            | 145 +++++++++++++---------
 arch/avr32/include/asm/irqflags.h          |  29 ++---
 arch/blackfin/include/asm/irqflags.h       |  12 --
 arch/blackfin/kernel/trace.c               |   1 +
 arch/cris/include/arch-v10/arch/irqflags.h |  45 +++++++
 arch/cris/include/arch-v10/arch/system.h   |  16 ---
 arch/cris/include/arch-v32/arch/irqflags.h |  46 +++++++
 arch/cris/include/arch-v32/arch/system.h   |  22 ----
 arch/cris/include/asm/irqflags.h           |   1 +
 arch/cris/include/asm/system.h             |   1 +
 arch/frv/include/asm/irqflags.h            | 158 +++++++++++++++++++++++
 arch/frv/include/asm/system.h              | 136 --------------------
 arch/h8300/include/asm/irqflags.h          |  43 +++++++
 arch/h8300/include/asm/system.h            |  24 +---
 arch/ia64/include/asm/irqflags.h           |  94 ++++++++++++++
 arch/ia64/include/asm/system.h             |  76 ------------
 arch/m32r/include/asm/irqflags.h           | 104 ++++++++++++++++
 arch/m32r/include/asm/system.h             |  66 +---------
 arch/m68k/include/asm/entry_no.h           |   2 +-
 arch/m68k/include/asm/irqflags.h           |  76 ++++++++++++
 arch/m68k/include/asm/system_mm.h          |  25 +---
 arch/m68k/include/asm/system_no.h          |  57 +--------
 arch/m68knommu/kernel/asm-offsets.c        |   2 -
 arch/m68knommu/platform/coldfire/head.S    |   1 +
 arch/microblaze/include/asm/irqflags.h     | 193 +++++++++++++++--------------
 arch/mips/include/asm/irqflags.h           |  53 ++++----
 arch/mips/kernel/smtc.c                    |   4 +-
 arch/mn10300/include/asm/irqflags.h        | 123 ++++++++++++++++++
 arch/mn10300/include/asm/system.h          | 109 +---------------
 arch/mn10300/kernel/entry.S                |   1 +
 arch/parisc/include/asm/irqflags.h         |  46 +++++++
 arch/parisc/include/asm/system.h           |  19 +--
 arch/powerpc/include/asm/hw_irq.h          | 113 ++++++++++-------
 arch/powerpc/include/asm/irqflags.h        |   2 +-
 arch/powerpc/kernel/exceptions-64s.S       |   4 +-
 arch/powerpc/kernel/irq.c                  |   4 +-
 arch/s390/include/asm/irqflags.h           |  51 ++++----
 arch/s390/include/asm/system.h             |   2 +-
 arch/s390/kernel/mem_detect.c              |   4 +-
 arch/s390/mm/init.c                        |   3 +-
 arch/s390/mm/maccess.c                     |   4 +-
 arch/score/include/asm/irqflags.h          | 187 +++++++++++++++-------------
 arch/sh/include/asm/irqflags.h             |   4 +-
 arch/sh/kernel/irq_32.c                    |  12 +-
 arch/sparc/include/asm/irqflags_32.h       |  35 +++---
 arch/sparc/include/asm/irqflags_64.h       |  29 ++---
 arch/sparc/kernel/irq_32.c                 |  13 +-
 arch/sparc/prom/p1275.c                    |   2 +-
 arch/tile/include/asm/irqflags.h           |  36 +++---
 arch/x86/include/asm/irqflags.h            |  32 ++---
 arch/x86/include/asm/paravirt.h            |  16 +--
 arch/x86/xen/spinlock.c                    |   2 +-
 arch/xtensa/include/asm/irqflags.h         |  58 +++++++++
 arch/xtensa/include/asm/system.h           |  33 +----
 drivers/s390/char/sclp.c                   |   2 +-
 include/asm-generic/atomic.h               |   5 +-
 include/asm-generic/cmpxchg-local.h        |   1 +
 include/asm-generic/hardirq.h              |   1 -
 include/asm-generic/irqflags.h             |  52 ++++----
 include/linux/irqflags.h                   | 107 +++++++++-------
 include/linux/spinlock.h                   |   1 +
 63 files changed, 1482 insertions(+), 1158 deletions(-)
 create mode 100644 arch/alpha/include/asm/irqflags.h
 create mode 100644 arch/cris/include/arch-v10/arch/irqflags.h
 create mode 100644 arch/cris/include/arch-v32/arch/irqflags.h
 create mode 100644 arch/cris/include/asm/irqflags.h
 create mode 100644 arch/frv/include/asm/irqflags.h
 create mode 100644 arch/h8300/include/asm/irqflags.h
 create mode 100644 arch/ia64/include/asm/irqflags.h
 create mode 100644 arch/m32r/include/asm/irqflags.h
 create mode 100644 arch/m68k/include/asm/irqflags.h
 create mode 100644 arch/mn10300/include/asm/irqflags.h
 create mode 100644 arch/parisc/include/asm/irqflags.h
 create mode 100644 arch/xtensa/include/asm/irqflags.h

(limited to 'include/linux')

diff --git a/arch/alpha/include/asm/irqflags.h b/arch/alpha/include/asm/irqflags.h
new file mode 100644
index 000000000000..299bbc7e9d71
--- /dev/null
+++ b/arch/alpha/include/asm/irqflags.h
@@ -0,0 +1,67 @@
+#ifndef __ALPHA_IRQFLAGS_H
+#define __ALPHA_IRQFLAGS_H
+
+#include <asm/system.h>
+
+#define IPL_MIN		0
+#define IPL_SW0		1
+#define IPL_SW1		2
+#define IPL_DEV0	3
+#define IPL_DEV1	4
+#define IPL_TIMER	5
+#define IPL_PERF	6
+#define IPL_POWERFAIL	6
+#define IPL_MCHECK	7
+#define IPL_MAX		7
+
+#ifdef CONFIG_ALPHA_BROKEN_IRQ_MASK
+#undef IPL_MIN
+#define IPL_MIN		__min_ipl
+extern int __min_ipl;
+#endif
+
+#define getipl()		(rdps() & 7)
+#define setipl(ipl)		((void) swpipl(ipl))
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	return rdps();
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	setipl(IPL_MAX);
+	barrier();
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags = swpipl(IPL_MAX);
+	barrier();
+	return flags;
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	barrier();
+	setipl(IPL_MIN);
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	barrier();
+	setipl(flags);
+	barrier();
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return flags == IPL_MAX;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(getipl());
+}
+
+#endif /* __ALPHA_IRQFLAGS_H */
diff --git a/arch/alpha/include/asm/system.h b/arch/alpha/include/asm/system.h
index 5aa40cca4f23..9f78e6934637 100644
--- a/arch/alpha/include/asm/system.h
+++ b/arch/alpha/include/asm/system.h
@@ -259,34 +259,6 @@ __CALL_PAL_RW2(wrperfmon, unsigned long, unsigned long, unsigned long);
 __CALL_PAL_W1(wrusp, unsigned long);
 __CALL_PAL_W1(wrvptptr, unsigned long);
 
-#define IPL_MIN		0
-#define IPL_SW0		1
-#define IPL_SW1		2
-#define IPL_DEV0	3
-#define IPL_DEV1	4
-#define IPL_TIMER	5
-#define IPL_PERF	6
-#define IPL_POWERFAIL	6
-#define IPL_MCHECK	7
-#define IPL_MAX		7
-
-#ifdef CONFIG_ALPHA_BROKEN_IRQ_MASK
-#undef IPL_MIN
-#define IPL_MIN		__min_ipl
-extern int __min_ipl;
-#endif
-
-#define getipl()		(rdps() & 7)
-#define setipl(ipl)		((void) swpipl(ipl))
-
-#define local_irq_disable()			do { setipl(IPL_MAX); barrier(); } while(0)
-#define local_irq_enable()			do { barrier(); setipl(IPL_MIN); } while(0)
-#define local_save_flags(flags)	((flags) = rdps())
-#define local_irq_save(flags)	do { (flags) = swpipl(IPL_MAX); barrier(); } while(0)
-#define local_irq_restore(flags)	do { barrier(); setipl(flags); barrier(); } while(0)
-
-#define irqs_disabled()	(getipl() == IPL_MAX)
-
 /*
  * TB routines..
  */
diff --git a/arch/arm/include/asm/irqflags.h b/arch/arm/include/asm/irqflags.h
index 6d09974e6646..1e6cca55c750 100644
--- a/arch/arm/include/asm/irqflags.h
+++ b/arch/arm/include/asm/irqflags.h
@@ -10,66 +10,85 @@
  */
 #if __LINUX_ARM_ARCH__ >= 6
 
-#define raw_local_irq_save(x)					\
-	({							\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_irq_save\n"	\
-	"cpsid	i"						\
-	: "=r" (x) : : "memory", "cc");				\
-	})
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags;
+
+	asm volatile(
+		"	mrs	%0, cpsr	@ arch_local_irq_save\n"
+		"	cpsid	i"
+		: "=r" (flags) : : "memory", "cc");
+	return flags;
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	asm volatile(
+		"	cpsie i			@ arch_local_irq_enable"
+		:
+		:
+		: "memory", "cc");
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	asm volatile(
+		"	cpsid i			@ arch_local_irq_disable"
+		:
+		:
+		: "memory", "cc");
+}
 
-#define raw_local_irq_enable()  __asm__("cpsie i	@ __sti" : : : "memory", "cc")
-#define raw_local_irq_disable() __asm__("cpsid i	@ __cli" : : : "memory", "cc")
 #define local_fiq_enable()  __asm__("cpsie f	@ __stf" : : : "memory", "cc")
 #define local_fiq_disable() __asm__("cpsid f	@ __clf" : : : "memory", "cc")
-
 #else
 
 /*
  * Save the current interrupt enable state & disable IRQs
  */
-#define raw_local_irq_save(x)					\
-	({							\
-		unsigned long temp;				\
-		(void) (&temp == &x);				\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_irq_save\n"	\
-"	orr	%1, %0, #128\n"					\
-"	msr	cpsr_c, %1"					\
-	: "=r" (x), "=r" (temp)					\
-	:							\
-	: "memory", "cc");					\
-	})
-	
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags, temp;
+
+	asm volatile(
+		"	mrs	%0, cpsr	@ arch_local_irq_save\n"
+		"	orr	%1, %0, #128\n"
+		"	msr	cpsr_c, %1"
+		: "=r" (flags), "=r" (temp)
+		:
+		: "memory", "cc");
+	return flags;
+}
+
 /*
  * Enable IRQs
  */
-#define raw_local_irq_enable()					\
-	({							\
-		unsigned long temp;				\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_irq_enable\n"	\
-"	bic	%0, %0, #128\n"					\
-"	msr	cpsr_c, %0"					\
-	: "=r" (temp)						\
-	:							\
-	: "memory", "cc");					\
-	})
+static inline void arch_local_irq_enable(void)
+{
+	unsigned long temp;
+	asm volatile(
+		"	mrs	%0, cpsr	@ arch_local_irq_enable\n"
+		"	bic	%0, %0, #128\n"
+		"	msr	cpsr_c, %0"
+		: "=r" (temp)
+		:
+		: "memory", "cc");
+}
 
 /*
  * Disable IRQs
  */
-#define raw_local_irq_disable()					\
-	({							\
-		unsigned long temp;				\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_irq_disable\n"	\
-"	orr	%0, %0, #128\n"					\
-"	msr	cpsr_c, %0"					\
-	: "=r" (temp)						\
-	:							\
-	: "memory", "cc");					\
-	})
+static inline void arch_local_irq_disable(void)
+{
+	unsigned long temp;
+	asm volatile(
+		"	mrs	%0, cpsr	@ arch_local_irq_disable\n"
+		"	orr	%0, %0, #128\n"
+		"	msr	cpsr_c, %0"
+		: "=r" (temp)
+		:
+		: "memory", "cc");
+}
 
 /*
  * Enable FIQs
@@ -106,27 +125,31 @@
 /*
  * Save the current interrupt enable state.
  */
-#define raw_local_save_flags(x)					\
-	({							\
-	__asm__ __volatile__(					\
-	"mrs	%0, cpsr		@ local_save_flags"	\
-	: "=r" (x) : : "memory", "cc");				\
-	})
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile(
+		"	mrs	%0, cpsr	@ local_save_flags"
+		: "=r" (flags) : : "memory", "cc");
+	return flags;
+}
 
 /*
  * restore saved IRQ & FIQ state
  */
-#define raw_local_irq_restore(x)				\
-	__asm__ __volatile__(					\
-	"msr	cpsr_c, %0		@ local_irq_restore\n"	\
-	:							\
-	: "r" (x)						\
-	: "memory", "cc")
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile(
+		"	msr	cpsr_c, %0	@ local_irq_restore"
+		:
+		: "r" (flags)
+		: "memory", "cc");
+}
 
-#define raw_irqs_disabled_flags(flags)	\
-({					\
-	(int)((flags) & PSR_I_BIT);	\
-})
+static inline int arch_irqs_disabled_flags(unsigned long flags)
+{
+	return flags & PSR_I_BIT;
+}
 
 #endif
 #endif
diff --git a/arch/avr32/include/asm/irqflags.h b/arch/avr32/include/asm/irqflags.h
index 93570daac38a..006e9487372d 100644
--- a/arch/avr32/include/asm/irqflags.h
+++ b/arch/avr32/include/asm/irqflags.h
@@ -8,16 +8,14 @@
 #ifndef __ASM_AVR32_IRQFLAGS_H
 #define __ASM_AVR32_IRQFLAGS_H
 
+#include <linux/types.h>
 #include <asm/sysreg.h>
 
-static inline unsigned long __raw_local_save_flags(void)
+static inline unsigned long arch_local_save_flags(void)
 {
 	return sysreg_read(SR);
 }
 
-#define raw_local_save_flags(x)					\
-	do { (x) = __raw_local_save_flags(); } while (0)
-
 /*
  * This will restore ALL status register flags, not only the interrupt
  * mask flag.
@@ -25,44 +23,39 @@ static inline unsigned long __raw_local_save_flags(void)
  * The empty asm statement informs the compiler of this fact while
  * also serving as a barrier.
  */
-static inline void raw_local_irq_restore(unsigned long flags)
+static inline void arch_local_irq_restore(unsigned long flags)
 {
 	sysreg_write(SR, flags);
 	asm volatile("" : : : "memory", "cc");
 }
 
-static inline void raw_local_irq_disable(void)
+static inline void arch_local_irq_disable(void)
 {
 	asm volatile("ssrf %0" : : "n"(SYSREG_GM_OFFSET) : "memory");
 }
 
-static inline void raw_local_irq_enable(void)
+static inline void arch_local_irq_enable(void)
 {
 	asm volatile("csrf %0" : : "n"(SYSREG_GM_OFFSET) : "memory");
 }
 
-static inline int raw_irqs_disabled_flags(unsigned long flags)
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
 {
 	return (flags & SYSREG_BIT(GM)) != 0;
 }
 
-static inline int raw_irqs_disabled(void)
+static inline bool arch_irqs_disabled(void)
 {
-	unsigned long flags = __raw_local_save_flags();
-
-	return raw_irqs_disabled_flags(flags);
+	return arch_irqs_disabled_flags(arch_local_save_flags());
 }
 
-static inline unsigned long __raw_local_irq_save(void)
+static inline unsigned long arch_local_irq_save(void)
 {
-	unsigned long flags = __raw_local_save_flags();
+	unsigned long flags = arch_local_save_flags();
 
-	raw_local_irq_disable();
+	arch_local_irq_disable();
 
 	return flags;
 }
 
-#define raw_local_irq_save(flags)				\
-	do { (flags) = __raw_local_irq_save(); } while (0)
-
 #endif /* __ASM_AVR32_IRQFLAGS_H */
diff --git a/arch/blackfin/include/asm/irqflags.h b/arch/blackfin/include/asm/irqflags.h
index 994d76791016..41c4d70544ef 100644
--- a/arch/blackfin/include/asm/irqflags.h
+++ b/arch/blackfin/include/asm/irqflags.h
@@ -218,16 +218,4 @@ static inline void hard_local_irq_restore(unsigned long flags)
 
 
 #endif /* !CONFIG_IPIPE */
-
-/*
- * Raw interface to linux/irqflags.h.
- */
-#define raw_local_save_flags(flags)	do { (flags) = arch_local_save_flags(); } while (0)
-#define raw_local_irq_save(flags)	do { (flags) = arch_local_irq_save(); } while (0)
-#define raw_local_irq_restore(flags)	arch_local_irq_restore(flags)
-#define raw_local_irq_enable()		arch_local_irq_enable()
-#define raw_local_irq_disable()		arch_local_irq_disable()
-#define raw_irqs_disabled_flags(flags)	arch_irqs_disabled_flags(flags)
-#define raw_irqs_disabled()		arch_irqs_disabled()
-
 #endif
diff --git a/arch/blackfin/kernel/trace.c b/arch/blackfin/kernel/trace.c
index 59fcdf6b0138..05b550891ce5 100644
--- a/arch/blackfin/kernel/trace.c
+++ b/arch/blackfin/kernel/trace.c
@@ -15,6 +15,7 @@
 #include <linux/kallsyms.h>
 #include <linux/err.h>
 #include <linux/fs.h>
+#include <linux/irq.h>
 #include <asm/dma.h>
 #include <asm/trace.h>
 #include <asm/fixed_code.h>
diff --git a/arch/cris/include/arch-v10/arch/irqflags.h b/arch/cris/include/arch-v10/arch/irqflags.h
new file mode 100644
index 000000000000..75ef18991240
--- /dev/null
+++ b/arch/cris/include/arch-v10/arch/irqflags.h
@@ -0,0 +1,45 @@
+#ifndef __ASM_CRIS_ARCH_IRQFLAGS_H
+#define __ASM_CRIS_ARCH_IRQFLAGS_H
+
+#include <linux/types.h>
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile("move $ccr,%0" : "=rm" (flags) : : "memory");
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	asm volatile("di" : : : "memory");
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	asm volatile("ei" : : : "memory");
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags = arch_local_save_flags();
+	arch_local_irq_disable();
+	return flags;
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile("move %0,$ccr" : : "rm" (flags) : "memory");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return !(flags & (1 << 5));
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* __ASM_CRIS_ARCH_IRQFLAGS_H */
diff --git a/arch/cris/include/arch-v10/arch/system.h b/arch/cris/include/arch-v10/arch/system.h
index 4a9cd36c9e16..935fde34aa15 100644
--- a/arch/cris/include/arch-v10/arch/system.h
+++ b/arch/cris/include/arch-v10/arch/system.h
@@ -44,20 +44,4 @@ static inline unsigned long _get_base(char * addr)
 struct __xchg_dummy { unsigned long a[100]; };
 #define __xg(x) ((struct __xchg_dummy *)(x))
 
-/* interrupt control.. */
-#define local_save_flags(x)	__asm__ __volatile__ ("move $ccr,%0" : "=rm" (x) : : "memory");
-#define local_irq_restore(x) 	__asm__ __volatile__ ("move %0,$ccr" : : "rm" (x) : "memory");
-#define local_irq_disable() 	__asm__ __volatile__ ( "di" : : :"memory");
-#define local_irq_enable()	__asm__ __volatile__ ( "ei" : : :"memory");
-
-#define irqs_disabled()			\
-({					\
-	unsigned long flags;		\
-	local_save_flags(flags);	\
-	!(flags & (1<<5));		\
-})
-
-/* For spinlocks etc */
-#define local_irq_save(x) __asm__ __volatile__ ("move $ccr,%0\n\tdi" : "=rm" (x) : : "memory");
-
 #endif
diff --git a/arch/cris/include/arch-v32/arch/irqflags.h b/arch/cris/include/arch-v32/arch/irqflags.h
new file mode 100644
index 000000000000..041851f8ec6f
--- /dev/null
+++ b/arch/cris/include/arch-v32/arch/irqflags.h
@@ -0,0 +1,46 @@
+#ifndef __ASM_CRIS_ARCH_IRQFLAGS_H
+#define __ASM_CRIS_ARCH_IRQFLAGS_H
+
+#include <linux/types.h>
+#include <arch/ptrace.h>
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile("move $ccs,%0" : "=rm" (flags) : : "memory");
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	asm volatile("di" : : : "memory");
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	asm volatile("ei" : : : "memory");
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags = arch_local_save_flags();
+	arch_local_irq_disable();
+	return flags;
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile("move %0,$ccs" : : "rm" (flags) : "memory");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return !(flags & (1 << I_CCS_BITNR));
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* __ASM_CRIS_ARCH_IRQFLAGS_H */
diff --git a/arch/cris/include/arch-v32/arch/system.h b/arch/cris/include/arch-v32/arch/system.h
index 6ca90f1f110a..76cea99eaa60 100644
--- a/arch/cris/include/arch-v32/arch/system.h
+++ b/arch/cris/include/arch-v32/arch/system.h
@@ -44,26 +44,4 @@ static inline unsigned long rdsp(void)
 struct __xchg_dummy { unsigned long a[100]; };
 #define __xg(x) ((struct __xchg_dummy *)(x))
 
-/* Used for interrupt control. */
-#define local_save_flags(x) \
-	__asm__ __volatile__ ("move $ccs, %0" : "=rm" (x) : : "memory");
-
-#define local_irq_restore(x) \
-	__asm__ __volatile__ ("move %0, $ccs" : : "rm" (x) : "memory");
-
-#define local_irq_disable()  __asm__ __volatile__ ("di" : : : "memory");
-#define local_irq_enable()   __asm__ __volatile__ ("ei" : : : "memory");
-
-#define irqs_disabled()		\
-({				\
-	unsigned long flags;	\
-				\
-	local_save_flags(flags);\
-	!(flags & (1 << I_CCS_BITNR));	\
-})
-
-/* Used for spinlocks, etc. */
-#define local_irq_save(x) \
-	__asm__ __volatile__ ("move $ccs, %0\n\tdi" : "=rm" (x) : : "memory");
-
 #endif /* _ASM_CRIS_ARCH_SYSTEM_H */
diff --git a/arch/cris/include/asm/irqflags.h b/arch/cris/include/asm/irqflags.h
new file mode 100644
index 000000000000..943ba5ca6d2c
--- /dev/null
+++ b/arch/cris/include/asm/irqflags.h
@@ -0,0 +1 @@
+#include <arch/irqflags.h>
diff --git a/arch/cris/include/asm/system.h b/arch/cris/include/asm/system.h
index 8657b084a922..ea10592f7d75 100644
--- a/arch/cris/include/asm/system.h
+++ b/arch/cris/include/asm/system.h
@@ -1,6 +1,7 @@
 #ifndef __ASM_CRIS_SYSTEM_H
 #define __ASM_CRIS_SYSTEM_H
 
+#include <linux/irqflags.h>
 #include <arch/system.h>
 
 /* the switch_to macro calls resume, an asm function in entry.S which does the actual
diff --git a/arch/frv/include/asm/irqflags.h b/arch/frv/include/asm/irqflags.h
new file mode 100644
index 000000000000..82f0b5363f42
--- /dev/null
+++ b/arch/frv/include/asm/irqflags.h
@@ -0,0 +1,158 @@
+/* FR-V interrupt handling
+ *
+ * Copyright (C) 2010 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public Licence
+ * as published by the Free Software Foundation; either version
+ * 2 of the Licence, or (at your option) any later version.
+ */
+
+#ifndef _ASM_IRQFLAGS_H
+#define _ASM_IRQFLAGS_H
+
+/*
+ * interrupt flag manipulation
+ * - use virtual interrupt management since touching the PSR is slow
+ *   - ICC2.Z: T if interrupts virtually disabled
+ *   - ICC2.C: F if interrupts really disabled
+ * - if Z==1 upon interrupt:
+ *   - C is set to 0
+ *   - interrupts are really disabled
+ *   - entry.S returns immediately
+ * - uses TIHI (TRAP if Z==0 && C==0) #2 to really reenable interrupts
+ *   - if taken, the trap:
+ *     - sets ICC2.C
+ *     - enables interrupts
+ */
+static inline void arch_local_irq_disable(void)
+{
+	/* set Z flag, but don't change the C flag */
+	asm volatile("	andcc	gr0,gr0,gr0,icc2	\n"
+		     :
+		     :
+		     : "memory", "icc2"
+		     );
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	/* clear Z flag and then test the C flag */
+	asm volatile("  oricc	gr0,#1,gr0,icc2		\n"
+		     "	tihi	icc2,gr0,#2		\n"
+		     :
+		     :
+		     : "memory", "icc2"
+		     );
+}
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+
+	asm volatile("movsg ccr,%0"
+		     : "=r"(flags)
+		     :
+		     : "memory");
+
+	/* shift ICC2.Z to bit 0 */
+	flags >>= 26;
+
+	/* make flags 1 if interrupts disabled, 0 otherwise */
+	return flags & 1UL;
+
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags = arch_local_save_flags();
+	arch_local_irq_disable();
+	return flags;
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	/* load the Z flag by turning 1 if disabled into 0 if disabled
+	 * and thus setting the Z flag but not the C flag */
+	asm volatile("  xoricc	%0,#1,gr0,icc2		\n"
+		     /* then trap if Z=0 and C=0 */
+		     "	tihi	icc2,gr0,#2		\n"
+		     :
+		     : "r"(flags)
+		     : "memory", "icc2"
+		     );
+
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return flags;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+/*
+ * real interrupt flag manipulation
+ */
+#define __arch_local_irq_disable()			\
+do {							\
+	unsigned long psr;				\
+	asm volatile("	movsg	psr,%0		\n"	\
+		     "	andi	%0,%2,%0	\n"	\
+		     "	ori	%0,%1,%0	\n"	\
+		     "	movgs	%0,psr		\n"	\
+		     : "=r"(psr)			\
+		     : "i" (PSR_PIL_14), "i" (~PSR_PIL)	\
+		     : "memory");			\
+} while (0)
+
+#define __arch_local_irq_enable()			\
+do {							\
+	unsigned long psr;				\
+	asm volatile("	movsg	psr,%0		\n"	\
+		     "	andi	%0,%1,%0	\n"	\
+		     "	movgs	%0,psr		\n"	\
+		     : "=r"(psr)			\
+		     : "i" (~PSR_PIL)			\
+		     : "memory");			\
+} while (0)
+
+#define __arch_local_save_flags(flags)		\
+do {						\
+	typecheck(unsigned long, flags);	\
+	asm("movsg psr,%0"			\
+	    : "=r"(flags)			\
+	    :					\
+	    : "memory");			\
+} while (0)
+
+#define	__arch_local_irq_save(flags)			\
+do {							\
+	unsigned long npsr;				\
+	typecheck(unsigned long, flags);		\
+	asm volatile("	movsg	psr,%0		\n"	\
+		     "	andi	%0,%3,%1	\n"	\
+		     "	ori	%1,%2,%1	\n"	\
+		     "	movgs	%1,psr		\n"	\
+		     : "=r"(flags), "=r"(npsr)		\
+		     : "i" (PSR_PIL_14), "i" (~PSR_PIL)	\
+		     : "memory");			\
+} while (0)
+
+#define	__arch_local_irq_restore(flags)			\
+do {							\
+	typecheck(unsigned long, flags);		\
+	asm volatile("	movgs	%0,psr		\n"	\
+		     :					\
+		     : "r" (flags)			\
+		     : "memory");			\
+} while (0)
+
+#define __arch_irqs_disabled()			\
+	((__get_PSR() & PSR_PIL) >= PSR_PIL_14)
+
+#endif /* _ASM_IRQFLAGS_H */
diff --git a/arch/frv/include/asm/system.h b/arch/frv/include/asm/system.h
index efd22d9077ac..0a6d8d9ca45b 100644
--- a/arch/frv/include/asm/system.h
+++ b/arch/frv/include/asm/system.h
@@ -36,142 +36,6 @@ do {									\
 	mb();								\
 } while(0)
 
-/*
- * interrupt flag manipulation
- * - use virtual interrupt management since touching the PSR is slow
- *   - ICC2.Z: T if interrupts virtually disabled
- *   - ICC2.C: F if interrupts really disabled
- * - if Z==1 upon interrupt:
- *   - C is set to 0
- *   - interrupts are really disabled
- *   - entry.S returns immediately
- * - uses TIHI (TRAP if Z==0 && C==0) #2 to really reenable interrupts
- *   - if taken, the trap:
- *     - sets ICC2.C
- *     - enables interrupts
- */
-#define local_irq_disable()					\
-do {								\
-	/* set Z flag, but don't change the C flag */		\
-	asm volatile("	andcc	gr0,gr0,gr0,icc2	\n"	\
-		     :						\
-		     :						\
-		     : "memory", "icc2"				\
-		     );						\
-} while(0)
-
-#define local_irq_enable()					\
-do {								\
-	/* clear Z flag and then test the C flag */		\
-	asm volatile("  oricc	gr0,#1,gr0,icc2		\n"	\
-		     "	tihi	icc2,gr0,#2		\n"	\
-		     :						\
-		     :						\
-		     : "memory", "icc2"				\
-		     );						\
-} while(0)
-
-#define local_save_flags(flags)					\
-do {								\
-	typecheck(unsigned long, flags);			\
-	asm volatile("movsg ccr,%0"				\
-		     : "=r"(flags)				\
-		     :						\
-		     : "memory");				\
-								\
-	/* shift ICC2.Z to bit 0 */				\
-	flags >>= 26;						\
-								\
-	/* make flags 1 if interrupts disabled, 0 otherwise */	\
-	flags &= 1UL;						\
-} while(0)
-
-#define irqs_disabled() \
-	({unsigned long flags; local_save_flags(flags); !!flags; })
-
-#define	local_irq_save(flags)			\
-do {						\
-	typecheck(unsigned long, flags);	\
-	local_save_flags(flags);		\
-	local_irq_disable();			\
-} while(0)
-
-#define	local_irq_restore(flags)					\
-do {									\
-	typecheck(unsigned long, flags);				\
-									\
-	/* load the Z flag by turning 1 if disabled into 0 if disabled	\
-	 * and thus setting the Z flag but not the C flag */		\
-	asm volatile("  xoricc	%0,#1,gr0,icc2		\n"		\
-		     /* then test Z=0 and C=0 */			\
-		     "	tihi	icc2,gr0,#2		\n"		\
-		     :							\
-		     : "r"(flags)					\
-		     : "memory", "icc2"					\
-		     );							\
-									\
-} while(0)
-
-/*
- * real interrupt flag manipulation
- */
-#define __local_irq_disable()				\
-do {							\
-	unsigned long psr;				\
-	asm volatile("	movsg	psr,%0		\n"	\
-		     "	andi	%0,%2,%0	\n"	\
-		     "	ori	%0,%1,%0	\n"	\
-		     "	movgs	%0,psr		\n"	\
-		     : "=r"(psr)			\
-		     : "i" (PSR_PIL_14), "i" (~PSR_PIL)	\
-		     : "memory");			\
-} while(0)
-
-#define __local_irq_enable()				\
-do {							\
-	unsigned long psr;				\
-	asm volatile("	movsg	psr,%0		\n"	\
-		     "	andi	%0,%1,%0	\n"	\
-		     "	movgs	%0,psr		\n"	\
-		     : "=r"(psr)			\
-		     : "i" (~PSR_PIL)			\
-		     : "memory");			\
-} while(0)
-
-#define __local_save_flags(flags)		\
-do {						\
-	typecheck(unsigned long, flags);	\
-	asm("movsg psr,%0"			\
-	    : "=r"(flags)			\
-	    :					\
-	    : "memory");			\
-} while(0)
-
-#define	__local_irq_save(flags)				\
-do {							\
-	unsigned long npsr;				\
-	typecheck(unsigned long, flags);		\
-	asm volatile("	movsg	psr,%0		\n"	\
-		     "	andi	%0,%3,%1	\n"	\
-		     "	ori	%1,%2,%1	\n"	\
-		     "	movgs	%1,psr		\n"	\
-		     : "=r"(flags), "=r"(npsr)		\
-		     : "i" (PSR_PIL_14), "i" (~PSR_PIL)	\
-		     : "memory");			\
-} while(0)
-
-#define	__local_irq_restore(flags)			\
-do {							\
-	typecheck(unsigned long, flags);		\
-	asm volatile("	movgs	%0,psr		\n"	\
-		     :					\
-		     : "r" (flags)			\
-		     : "memory");			\
-} while(0)
-
-#define __irqs_disabled() \
-	((__get_PSR() & PSR_PIL) >= PSR_PIL_14)
-
 /*
  * Force strict CPU ordering.
  */
diff --git a/arch/h8300/include/asm/irqflags.h b/arch/h8300/include/asm/irqflags.h
new file mode 100644
index 000000000000..9617cd57aebd
--- /dev/null
+++ b/arch/h8300/include/asm/irqflags.h
@@ -0,0 +1,43 @@
+#ifndef _H8300_IRQFLAGS_H
+#define _H8300_IRQFLAGS_H
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile ("stc ccr,%w0" : "=r" (flags));
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	asm volatile ("orc  #0x80,ccr" : : : "memory");
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	asm volatile ("andc #0x7f,ccr" : : : "memory");
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags = arch_local_save_flags();
+	arch_local_irq_disable();
+	return flags;
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile ("ldc %w0,ccr" : : "r" (flags) : "memory");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & 0x80) == 0x80;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* _H8300_IRQFLAGS_H */
diff --git a/arch/h8300/include/asm/system.h b/arch/h8300/include/asm/system.h
index 16bf1560ff68..2c2382e50d93 100644
--- a/arch/h8300/include/asm/system.h
+++ b/arch/h8300/include/asm/system.h
@@ -2,6 +2,7 @@
 #define _H8300_SYSTEM_H
 
 #include <linux/linkage.h>
+#include <linux/irqflags.h>
 
 struct pt_regs;
 
@@ -51,31 +52,8 @@ asmlinkage void resume(void);
   (last) = _last; 					    \
 }
 
-#define __sti() asm volatile ("andc #0x7f,ccr")
-#define __cli() asm volatile ("orc  #0x80,ccr")
-
-#define __save_flags(x) \
-       asm volatile ("stc ccr,%w0":"=r" (x))
-
-#define __restore_flags(x) \
-       asm volatile ("ldc %w0,ccr": :"r" (x))
-
-#define	irqs_disabled()			\
-({					\
-	unsigned char flags;		\
-	__save_flags(flags);	        \
-	((flags & 0x80) == 0x80);	\
-})
-
 #define iret() __asm__ __volatile__ ("rte": : :"memory", "sp", "cc")
 
-/* For spinlocks etc */
-#define local_irq_disable()	__cli()
-#define local_irq_enable()      __sti()
-#define local_irq_save(x)	({ __save_flags(x); local_irq_disable(); })
-#define local_irq_restore(x)	__restore_flags(x)
-#define local_save_flags(x)     __save_flags(x)
-
 /*
  * Force strict CPU ordering.
  * Not really required on H8...
diff --git a/arch/ia64/include/asm/irqflags.h b/arch/ia64/include/asm/irqflags.h
new file mode 100644
index 000000000000..f82d6be2ecd2
--- /dev/null
+++ b/arch/ia64/include/asm/irqflags.h
@@ -0,0 +1,94 @@
+/*
+ * IRQ flags defines.
+ *
+ * Copyright (C) 1998-2003 Hewlett-Packard Co
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
+ * Copyright (C) 1999 Asit Mallick <asit.k.mallick@intel.com>
+ * Copyright (C) 1999 Don Dugger <don.dugger@intel.com>
+ */
+
+#ifndef _ASM_IA64_IRQFLAGS_H
+#define _ASM_IA64_IRQFLAGS_H
+
+#ifdef CONFIG_IA64_DEBUG_IRQ
+extern unsigned long last_cli_ip;
+static inline void arch_maybe_save_ip(unsigned long flags)
+{
+	if (flags & IA64_PSR_I)
+		last_cli_ip = ia64_getreg(_IA64_REG_IP);
+}
+#else
+#define arch_maybe_save_ip(flags) do {} while (0)
+#endif
+
+/*
+ * - clearing psr.i is implicitly serialized (visible by next insn)
+ * - setting psr.i requires data serialization
+ * - we need a stop-bit before reading PSR because we sometimes
+ *   write a floating-point register right before reading the PSR
+ *   and that writes to PSR.mfl
+ */
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	ia64_stop();
+#ifdef CONFIG_PARAVIRT
+	return ia64_get_psr_i();
+#else
+	return ia64_getreg(_IA64_REG_PSR);
+#endif
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags = arch_local_save_flags();
+
+	ia64_stop();
+	ia64_rsm(IA64_PSR_I);
+	arch_maybe_save_ip(flags);
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+#ifdef CONFIG_IA64_DEBUG_IRQ
+	arch_local_irq_save();
+#else
+	ia64_stop();
+	ia64_rsm(IA64_PSR_I);
+#endif
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	ia64_stop();
+	ia64_ssm(IA64_PSR_I);
+	ia64_srlz_d();
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+#ifdef CONFIG_IA64_DEBUG_IRQ
+	unsigned long old_psr = arch_local_save_flags();
+#endif
+	ia64_intrin_local_irq_restore(flags & IA64_PSR_I);
+	arch_maybe_save_ip(old_psr & ~flags);
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & IA64_PSR_I) == 0;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+static inline void arch_safe_halt(void)
+{
+	ia64_pal_halt_light();	/* PAL_HALT_LIGHT */
+}
+
+
+#endif /* _ASM_IA64_IRQFLAGS_H */
diff --git a/arch/ia64/include/asm/system.h b/arch/ia64/include/asm/system.h
index 9f342a574ce8..2feb7f64c035 100644
--- a/arch/ia64/include/asm/system.h
+++ b/arch/ia64/include/asm/system.h
@@ -107,87 +107,11 @@ extern struct ia64_boot_param {
  */
 #define set_mb(var, value)	do { (var) = (value); mb(); } while (0)
 
-#define safe_halt()         ia64_pal_halt_light()    /* PAL_HALT_LIGHT */
-
 /*
  * The group barrier in front of the rsm & ssm are necessary to ensure
  * that none of the previous instructions in the same group are
  * affected by the rsm/ssm.
  */
-/* For spinlocks etc */
-
-/*
- * - clearing psr.i is implicitly serialized (visible by next insn)
- * - setting psr.i requires data serialization
- * - we need a stop-bit before reading PSR because we sometimes
- *   write a floating-point register right before reading the PSR
- *   and that writes to PSR.mfl
- */
-#ifdef CONFIG_PARAVIRT
-#define __local_save_flags()	ia64_get_psr_i()
-#else
-#define __local_save_flags()	ia64_getreg(_IA64_REG_PSR)
-#endif
-
-#define __local_irq_save(x)			\
-do {						\
-	ia64_stop();				\
-	(x) = __local_save_flags();		\
-	ia64_stop();				\
-	ia64_rsm(IA64_PSR_I);			\
-} while (0)
-
-#define __local_irq_disable()			\
-do {						\
-	ia64_stop();				\
-	ia64_rsm(IA64_PSR_I);			\
-} while (0)
-
-#define __local_irq_restore(x)	ia64_intrin_local_irq_restore((x) & IA64_PSR_I)
-
-#ifdef CONFIG_IA64_DEBUG_IRQ
-
-  extern unsigned long last_cli_ip;
-
-# define __save_ip()		last_cli_ip = ia64_getreg(_IA64_REG_IP)
-
-# define local_irq_save(x)					\
-do {								\
-	unsigned long __psr;					\
-								\
-	__local_irq_save(__psr);				\
-	if (__psr & IA64_PSR_I)					\
-		__save_ip();					\
-	(x) = __psr;						\
-} while (0)
-
-# define local_irq_disable()	do { unsigned long __x; local_irq_save(__x); } while (0)
-
-# define local_irq_restore(x)					\
-do {								\
-	unsigned long __old_psr, __psr = (x);			\
-								\
-	local_save_flags(__old_psr);				\
-	__local_irq_restore(__psr);				\
-	if ((__old_psr & IA64_PSR_I) && !(__psr & IA64_PSR_I))	\
-		__save_ip();					\
-} while (0)
-
-#else /* !CONFIG_IA64_DEBUG_IRQ */
-# define local_irq_save(x)	__local_irq_save(x)
-# define local_irq_disable()	__local_irq_disable()
-# define local_irq_restore(x)	__local_irq_restore(x)
-#endif /* !CONFIG_IA64_DEBUG_IRQ */
-
-#define local_irq_enable()	({ ia64_stop(); ia64_ssm(IA64_PSR_I); ia64_srlz_d(); })
-#define local_save_flags(flags)	({ ia64_stop(); (flags) = __local_save_flags(); })
-
-#define irqs_disabled()				\
-({						\
-	unsigned long __ia64_id_flags;		\
-	local_save_flags(__ia64_id_flags);	\
-	(__ia64_id_flags & IA64_PSR_I) == 0;	\
-})
 
 #ifdef __KERNEL__
 
diff --git a/arch/m32r/include/asm/irqflags.h b/arch/m32r/include/asm/irqflags.h
new file mode 100644
index 000000000000..1f92d29982ae
--- /dev/null
+++ b/arch/m32r/include/asm/irqflags.h
@@ -0,0 +1,104 @@
+/*
+ * 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.
+ *
+ * Copyright (C) 2001  Hiroyuki Kondo, Hirokazu Takata, and Hitoshi Yamamoto
+ * Copyright (C) 2004, 2006  Hirokazu Takata <takata at linux-m32r.org>
+ */
+
+#ifndef _ASM_M32R_IRQFLAGS_H
+#define _ASM_M32R_IRQFLAGS_H
+
+#include <linux/types.h>
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile("mvfc %0,psw" : "=r"(flags));
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+#if !defined(CONFIG_CHIP_M32102) && !defined(CONFIG_CHIP_M32104)
+	asm volatile (
+		"clrpsw #0x40 -> nop"
+		: : : "memory");
+#else
+	unsigned long tmpreg0, tmpreg1;
+	asm volatile (
+		"ld24	%0, #0	; Use 32-bit insn.			\n\t"
+		"mvfc	%1, psw	; No interrupt can be accepted here.	\n\t"
+		"mvtc	%0, psw						\n\t"
+		"and3	%0, %1, #0xffbf					\n\t"
+		"mvtc	%0, psw						\n\t"
+		: "=&r" (tmpreg0), "=&r" (tmpreg1)
+		:
+		: "cbit", "memory");
+#endif
+}
+
+static inline void arch_local_irq_enable(void)
+{
+#if !defined(CONFIG_CHIP_M32102) && !defined(CONFIG_CHIP_M32104)
+	asm volatile (
+		"setpsw #0x40 -> nop"
+		: : : "memory");
+#else
+	unsigned long tmpreg;
+	asm volatile (
+		"mvfc	%0, psw;		\n\t"
+		"or3	%0, %0, #0x0040;	\n\t"
+		"mvtc	%0, psw;		\n\t"
+		: "=&r" (tmpreg)
+		:
+		: "cbit", "memory");
+#endif
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags;
+
+#if !(defined(CONFIG_CHIP_M32102) || defined(CONFIG_CHIP_M32104))
+	asm volatile (
+		"mvfc	%0, psw;	\n\t"
+		"clrpsw	#0x40 -> nop;	\n\t"
+		: "=r" (flags)
+		:
+		: "memory");
+#else
+	unsigned long tmpreg;
+	asm volatile (
+		"ld24	%1, #0		\n\t"
+		"mvfc	%0, psw		\n\t"
+		"mvtc	%1, psw		\n\t"
+		"and3	%1, %0, #0xffbf	\n\t"
+		"mvtc	%1, psw		\n\t"
+		: "=r" (flags), "=&r" (tmpreg)
+		:
+		: "cbit", "memory");
+#endif
+	return flags;
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile("mvtc %0,psw"
+		     :
+		     : "r" (flags)
+		     : "cbit", "memory");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return !(flags & 0x40);
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* _ASM_M32R_IRQFLAGS_H */
diff --git a/arch/m32r/include/asm/system.h b/arch/m32r/include/asm/system.h
index c980f5ba8de7..13c46794ccb1 100644
--- a/arch/m32r/include/asm/system.h
+++ b/arch/m32r/include/asm/system.h
@@ -11,6 +11,7 @@
  */
 
 #include <linux/compiler.h>
+#include <linux/irqflags.h>
 #include <asm/assembler.h>
 
 #ifdef __KERNEL__
@@ -54,71 +55,6 @@
 	); \
 } while(0)
 
-/* Interrupt Control */
-#if !defined(CONFIG_CHIP_M32102) && !defined(CONFIG_CHIP_M32104)
-#define local_irq_enable() \
-	__asm__ __volatile__ ("setpsw #0x40 -> nop": : :"memory")
-#define local_irq_disable() \
-	__asm__ __volatile__ ("clrpsw #0x40 -> nop": : :"memory")
-#else	/* CONFIG_CHIP_M32102 || CONFIG_CHIP_M32104 */
-static inline void local_irq_enable(void)
-{
-	unsigned long tmpreg;
-	__asm__ __volatile__(
-		"mvfc	%0, psw;		\n\t"
-		"or3	%0, %0, #0x0040;	\n\t"
-		"mvtc	%0, psw;		\n\t"
-	: "=&r" (tmpreg) : : "cbit", "memory");
-}
-
-static inline void local_irq_disable(void)
-{
-	unsigned long tmpreg0, tmpreg1;
-	__asm__ __volatile__(
-		"ld24	%0, #0	; Use 32-bit insn. \n\t"
-		"mvfc	%1, psw	; No interrupt can be accepted here. \n\t"
-		"mvtc	%0, psw	\n\t"
-		"and3	%0, %1, #0xffbf	\n\t"
-		"mvtc	%0, psw	\n\t"
-	: "=&r" (tmpreg0), "=&r" (tmpreg1) : : "cbit", "memory");
-}
-#endif	/* CONFIG_CHIP_M32102 || CONFIG_CHIP_M32104 */
-
-#define local_save_flags(x) \
-	__asm__ __volatile__("mvfc %0,psw" : "=r"(x) : /* no input */)
-
-#define local_irq_restore(x) \
-	__asm__ __volatile__("mvtc %0,psw" : /* no outputs */ \
-		: "r" (x) : "cbit", "memory")
-
-#if !(defined(CONFIG_CHIP_M32102) || defined(CONFIG_CHIP_M32104))
-#define local_irq_save(x)				\
-	__asm__ __volatile__(				\
-  		"mvfc	%0, psw;		\n\t"	\
-	  	"clrpsw	#0x40 -> nop;		\n\t"	\
-  		: "=r" (x) : /* no input */ : "memory")
-#else	/* CONFIG_CHIP_M32102 || CONFIG_CHIP_M32104 */
-#define local_irq_save(x) 				\
-	({						\
-		unsigned long tmpreg;			\
-		__asm__ __volatile__( 			\
-			"ld24	%1, #0 \n\t" 		\
-			"mvfc	%0, psw \n\t"		\
-			"mvtc	%1, psw \n\t"		\
-			"and3	%1, %0, #0xffbf \n\t"	\
-			"mvtc	%1, psw \n\t" 		\
-			: "=r" (x), "=&r" (tmpreg)	\
-			: : "cbit", "memory");		\
-	})
-#endif	/* CONFIG_CHIP_M32102 || CONFIG_CHIP_M32104 */
-
-#define irqs_disabled()					\
-	({						\
-		unsigned long flags;			\
-		local_save_flags(flags);		\
-		!(flags & 0x40);			\
-	})
-
 #define nop()	__asm__ __volatile__ ("nop" : : )
 
 #define xchg(ptr, x)							\
diff --git a/arch/m68k/include/asm/entry_no.h b/arch/m68k/include/asm/entry_no.h
index 907ed03d792f..80e41492aa2a 100644
--- a/arch/m68k/include/asm/entry_no.h
+++ b/arch/m68k/include/asm/entry_no.h
@@ -28,7 +28,7 @@
  *			M68K		  COLDFIRE
  */
 
-#define ALLOWINT 0xf8ff
+#define ALLOWINT (~0x700)
 
 #ifdef __ASSEMBLY__
 
diff --git a/arch/m68k/include/asm/irqflags.h b/arch/m68k/include/asm/irqflags.h
new file mode 100644
index 000000000000..4a5b284a1550
--- /dev/null
+++ b/arch/m68k/include/asm/irqflags.h
@@ -0,0 +1,76 @@
+#ifndef _M68K_IRQFLAGS_H
+#define _M68K_IRQFLAGS_H
+
+#include <linux/types.h>
+#include <linux/hardirq.h>
+#include <linux/preempt.h>
+#include <asm/thread_info.h>
+#include <asm/entry.h>
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile ("movew %%sr,%0" : "=d" (flags) : : "memory");
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+#ifdef CONFIG_COLDFIRE
+	asm volatile (
+		"move	%/sr,%%d0	\n\t"
+		"ori.l	#0x0700,%%d0	\n\t"
+		"move	%%d0,%/sr	\n"
+		: /* no outputs */
+		:
+		: "cc", "%d0", "memory");
+#else
+	asm volatile ("oriw  #0x0700,%%sr" : : : "memory");
+#endif
+}
+
+static inline void arch_local_irq_enable(void)
+{
+#if defined(CONFIG_COLDFIRE)
+	asm volatile (
+		"move	%/sr,%%d0	\n\t"
+		"andi.l	#0xf8ff,%%d0	\n\t"
+		"move	%%d0,%/sr	\n"
+		: /* no outputs */
+		:
+		: "cc", "%d0", "memory");
+#else
+# if defined(CONFIG_MMU)
+	if (MACH_IS_Q40 || !hardirq_count())
+# endif
+		asm volatile (
+			"andiw %0,%%sr"
+			:
+			: "i" (ALLOWINT)
+			: "memory");
+#endif
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags = arch_local_save_flags();
+	arch_local_irq_disable();
+	return flags;
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile ("movew %0,%%sr" : : "d" (flags) : "memory");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & ~ALLOWINT) != 0;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* _M68K_IRQFLAGS_H */
diff --git a/arch/m68k/include/asm/system_mm.h b/arch/m68k/include/asm/system_mm.h
index dbb6515ffd5b..12053c44cccf 100644
--- a/arch/m68k/include/asm/system_mm.h
+++ b/arch/m68k/include/asm/system_mm.h
@@ -3,6 +3,7 @@
 
 #include <linux/linkage.h>
 #include <linux/kernel.h>
+#include <linux/irqflags.h>
 #include <asm/segment.h>
 #include <asm/entry.h>
 
@@ -62,30 +63,6 @@ asmlinkage void resume(void);
 #define smp_wmb()	barrier()
 #define smp_read_barrier_depends()	((void)0)
 
-/* interrupt control.. */
-#if 0
-#define local_irq_enable() asm volatile ("andiw %0,%%sr": : "i" (ALLOWINT) : "memory")
-#else
-#include <linux/hardirq.h>
-#define local_irq_enable() ({							\
-	if (MACH_IS_Q40 || !hardirq_count())					\
-		asm volatile ("andiw %0,%%sr": : "i" (ALLOWINT) : "memory");	\
-})
-#endif
-#define local_irq_disable() asm volatile ("oriw  #0x0700,%%sr": : : "memory")
-#define local_save_flags(x) asm volatile ("movew %%sr,%0":"=d" (x) : : "memory")
-#define local_irq_restore(x) asm volatile ("movew %0,%%sr": :"d" (x) : "memory")
-
-static inline int irqs_disabled(void)
-{
-	unsigned long flags;
-	local_save_flags(flags);
-	return flags & ~ALLOWINT;
-}
-
-/* For spinlocks etc */
-#define local_irq_save(x)	({ local_save_flags(x); local_irq_disable(); })
-
 #define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
 
 struct __xchg_dummy { unsigned long a[100]; };
diff --git a/arch/m68k/include/asm/system_no.h b/arch/m68k/include/asm/system_no.h
index 3c0718d74398..20126c09794e 100644
--- a/arch/m68k/include/asm/system_no.h
+++ b/arch/m68k/include/asm/system_no.h
@@ -2,6 +2,7 @@
 #define _M68KNOMMU_SYSTEM_H
 
 #include <linux/linkage.h>
+#include <linux/irqflags.h>
 #include <asm/segment.h>
 #include <asm/entry.h>
 
@@ -46,54 +47,6 @@ asmlinkage void resume(void);
   (last) = _last;						\
 }
 
-#ifdef CONFIG_COLDFIRE
-#define local_irq_enable() __asm__ __volatile__ (		\
-	"move %/sr,%%d0\n\t"					\
-	"andi.l #0xf8ff,%%d0\n\t"				\
-	"move %%d0,%/sr\n"					\
-	: /* no outputs */					\
-	:							\
-        : "cc", "%d0", "memory")
-#define local_irq_disable() __asm__ __volatile__ (		\
-	"move %/sr,%%d0\n\t"					\
-	"ori.l #0x0700,%%d0\n\t"				\
-	"move %%d0,%/sr\n"					\
-	: /* no outputs */					\
-	:							\
-	: "cc", "%d0", "memory")
-/* For spinlocks etc */
-#define local_irq_save(x) __asm__ __volatile__ (		\
-	"movew %%sr,%0\n\t"					\
-	"movew #0x0700,%%d0\n\t"				\
-	"or.l  %0,%%d0\n\t"					\
-	"movew %%d0,%/sr"					\
-	: "=d" (x)						\
-	:							\
-	: "cc", "%d0", "memory")
-#else
-
-/* portable version */ /* FIXME - see entry.h*/
-#define ALLOWINT 0xf8ff
-
-#define local_irq_enable() asm volatile ("andiw %0,%%sr": : "i" (ALLOWINT) : "memory")
-#define local_irq_disable() asm volatile ("oriw  #0x0700,%%sr": : : "memory")
-#endif
-
-#define local_save_flags(x) asm volatile ("movew %%sr,%0":"=d" (x) : : "memory")
-#define local_irq_restore(x) asm volatile ("movew %0,%%sr": :"d" (x) : "memory")
-
-/* For spinlocks etc */
-#ifndef local_irq_save
-#define local_irq_save(x) do { local_save_flags(x); local_irq_disable(); } while (0)
-#endif
-
-#define	irqs_disabled()			\
-({					\
-	unsigned long flags;		\
-	local_save_flags(flags);	\
-	((flags & 0x0700) == 0x0700);	\
-})
-
 #define iret() __asm__ __volatile__ ("rte": : :"memory", "sp", "cc")
 
 /*
@@ -206,12 +159,4 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz
 #define arch_align_stack(x) (x)
 
 
-static inline int irqs_disabled_flags(unsigned long flags)
-{
-	if (flags & 0x0700)
-		return 0;
-	else
-		return 1;
-}
-
 #endif /* _M68KNOMMU_SYSTEM_H */
diff --git a/arch/m68knommu/kernel/asm-offsets.c b/arch/m68knommu/kernel/asm-offsets.c
index 9a8876f715d8..24335022fa2c 100644
--- a/arch/m68knommu/kernel/asm-offsets.c
+++ b/arch/m68knommu/kernel/asm-offsets.c
@@ -74,8 +74,6 @@ int main(void)
 
 	DEFINE(PT_PTRACED, PT_PTRACED);
 
-	DEFINE(THREAD_SIZE, THREAD_SIZE);
-
 	/* Offsets in thread_info structure */
 	DEFINE(TI_TASK, offsetof(struct thread_info, task));
 	DEFINE(TI_EXECDOMAIN, offsetof(struct thread_info, exec_domain));
diff --git a/arch/m68knommu/platform/coldfire/head.S b/arch/m68knommu/platform/coldfire/head.S
index 4b91aa24eb00..0b2d7c7adf79 100644
--- a/arch/m68knommu/platform/coldfire/head.S
+++ b/arch/m68knommu/platform/coldfire/head.S
@@ -15,6 +15,7 @@
 #include <asm/coldfire.h>
 #include <asm/mcfcache.h>
 #include <asm/mcfsim.h>
+#include <asm/thread_info.h>
 
 /*****************************************************************************/
 
diff --git a/arch/microblaze/include/asm/irqflags.h b/arch/microblaze/include/asm/irqflags.h
index 2c38c6d80176..5fd31905775d 100644
--- a/arch/microblaze/include/asm/irqflags.h
+++ b/arch/microblaze/include/asm/irqflags.h
@@ -9,103 +9,114 @@
 #ifndef _ASM_MICROBLAZE_IRQFLAGS_H
 #define _ASM_MICROBLAZE_IRQFLAGS_H
 
-#include <linux/irqflags.h>
+#include <linux/types.h>
 #include <asm/registers.h>
 
-# if CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR
-
-# define raw_local_irq_save(flags)			\
-	do {						\
-		asm volatile ("	msrclr %0, %1;		\
-				nop;"			\
-				: "=r"(flags)		\
-				: "i"(MSR_IE)		\
-				: "memory");		\
-	} while (0)
-
-# define raw_local_irq_disable()			\
-	do {						\
-		asm volatile ("	msrclr r0, %0;		\
-				nop;"			\
-				:			\
-				: "i"(MSR_IE)		\
-				: "memory");		\
-	} while (0)
-
-# define raw_local_irq_enable()				\
-	do {						\
-		asm volatile ("	msrset	r0, %0;		\
-				nop;"			\
-				:			\
-				: "i"(MSR_IE)		\
-				: "memory");		\
-	} while (0)
-
-# else /* CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR == 0 */
-
-# define raw_local_irq_save(flags)				\
-	do {							\
-		register unsigned tmp;				\
-		asm volatile ("	mfs	%0, rmsr;		\
-				nop;				\
-				andi	%1, %0, %2;		\
-				mts	rmsr, %1;		\
-				nop;"				\
-				: "=r"(flags), "=r" (tmp)	\
-				: "i"(~MSR_IE)			\
-				: "memory");			\
-	} while (0)
-
-# define raw_local_irq_disable()				\
-	do {							\
-		register unsigned tmp;				\
-		asm volatile ("	mfs	%0, rmsr;		\
-				nop;				\
-				andi	%0, %0, %1;		\
-				mts	rmsr, %0;		\
-				nop;"			\
-				: "=r"(tmp)			\
-				: "i"(~MSR_IE)			\
-				: "memory");			\
-	} while (0)
-
-# define raw_local_irq_enable()					\
-	do {							\
-		register unsigned tmp;				\
-		asm volatile ("	mfs	%0, rmsr;		\
-				nop;				\
-				ori	%0, %0, %1;		\
-				mts	rmsr, %0;		\
-				nop;"				\
-				: "=r"(tmp)			\
-				: "i"(MSR_IE)			\
-				: "memory");			\
-	} while (0)
-
-# endif /* CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */
-
-#define raw_local_irq_restore(flags)				\
-	do {							\
-		asm volatile ("	mts	rmsr, %0;		\
-				nop;"				\
-				:				\
-				: "r"(flags)			\
-				: "memory");			\
-	} while (0)
-
-static inline unsigned long get_msr(void)
+#ifdef CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags;
+	asm volatile("	msrclr %0, %1	\n"
+		     "	nop		\n"
+		     : "=r"(flags)
+		     : "i"(MSR_IE)
+		     : "memory");
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	/* this uses r0 without declaring it - is that correct? */
+	asm volatile("	msrclr r0, %0	\n"
+		     "	nop		\n"
+		     :
+		     : "i"(MSR_IE)
+		     : "memory");
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	/* this uses r0 without declaring it - is that correct? */
+	asm volatile("	msrset	r0, %0	\n"
+		     "	nop		\n"
+		     :
+		     : "i"(MSR_IE)
+		     : "memory");
+}
+
+#else /* !CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags, tmp;
+	asm volatile ("	mfs	%0, rmsr	\n"
+		      "	nop			\n"
+		      "	andi	%1, %0, %2	\n"
+		      "	mts	rmsr, %1	\n"
+		      "	nop			\n"
+		      : "=r"(flags), "=r"(tmp)
+		      : "i"(~MSR_IE)
+		      : "memory");
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	unsigned long tmp;
+	asm volatile("	mfs	%0, rmsr	\n"
+		     "	nop			\n"
+		     "	andi	%0, %0, %1	\n"
+		     "	mts	rmsr, %0	\n"
+		     "	nop			\n"
+		     : "=r"(tmp)
+		     : "i"(~MSR_IE)
+		     : "memory");
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	unsigned long tmp;
+	asm volatile("	mfs	%0, rmsr	\n"
+		     "	nop			\n"
+		     "	ori	%0, %0, %1	\n"
+		     "	mts	rmsr, %0	\n"
+		     "	nop			\n"
+		     : "=r"(tmp)
+		     : "i"(MSR_IE)
+		     : "memory");
+}
+
+#endif /* CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */
+
+static inline unsigned long arch_local_save_flags(void)
 {
 	unsigned long flags;
-	asm volatile ("	mfs	%0, rmsr;	\
-			nop;"			\
-			: "=r"(flags)		\
-			:			\
-			: "memory");		\
+	asm volatile("	mfs	%0, rmsr	\n"
+		     "	nop			\n"
+		     : "=r"(flags)
+		     :
+		     : "memory");
 	return flags;
 }
 
-#define raw_local_save_flags(flags)	((flags) = get_msr())
-#define raw_irqs_disabled()		((get_msr() & MSR_IE) == 0)
-#define raw_irqs_disabled_flags(flags)	((flags & MSR_IE) == 0)
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile("	mts	rmsr, %0	\n"
+		     "	nop			\n"
+		     :
+		     : "r"(flags)
+		     : "memory");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & MSR_IE) == 0;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
 
 #endif /* _ASM_MICROBLAZE_IRQFLAGS_H */
diff --git a/arch/mips/include/asm/irqflags.h b/arch/mips/include/asm/irqflags.h
index 701ec0ba8fa9..9ef3b0d17896 100644
--- a/arch/mips/include/asm/irqflags.h
+++ b/arch/mips/include/asm/irqflags.h
@@ -17,7 +17,7 @@
 #include <asm/hazards.h>
 
 __asm__(
-	"	.macro	raw_local_irq_enable				\n"
+	"	.macro	arch_local_irq_enable				\n"
 	"	.set	push						\n"
 	"	.set	reorder						\n"
 	"	.set	noat						\n"
@@ -40,7 +40,7 @@ __asm__(
 
 extern void smtc_ipi_replay(void);
 
-static inline void raw_local_irq_enable(void)
+static inline void arch_local_irq_enable(void)
 {
 #ifdef CONFIG_MIPS_MT_SMTC
 	/*
@@ -50,7 +50,7 @@ static inline void raw_local_irq_enable(void)
 	smtc_ipi_replay();
 #endif
 	__asm__ __volatile__(
-		"raw_local_irq_enable"
+		"arch_local_irq_enable"
 		: /* no outputs */
 		: /* no inputs */
 		: "memory");
@@ -76,7 +76,7 @@ static inline void raw_local_irq_enable(void)
  * Workaround: mask EXL bit of the result or place a nop before mfc0.
  */
 __asm__(
-	"	.macro	raw_local_irq_disable\n"
+	"	.macro	arch_local_irq_disable\n"
 	"	.set	push						\n"
 	"	.set	noat						\n"
 #ifdef CONFIG_MIPS_MT_SMTC
@@ -97,17 +97,17 @@ __asm__(
 	"	.set	pop						\n"
 	"	.endm							\n");
 
-static inline void raw_local_irq_disable(void)
+static inline void arch_local_irq_disable(void)
 {
 	__asm__ __volatile__(
-		"raw_local_irq_disable"
+		"arch_local_irq_disable"
 		: /* no outputs */
 		: /* no inputs */
 		: "memory");
 }
 
 __asm__(
-	"	.macro	raw_local_save_flags flags			\n"
+	"	.macro	arch_local_save_flags flags			\n"
 	"	.set	push						\n"
 	"	.set	reorder						\n"
 #ifdef CONFIG_MIPS_MT_SMTC
@@ -118,13 +118,15 @@ __asm__(
 	"	.set	pop						\n"
 	"	.endm							\n");
 
-#define raw_local_save_flags(x)						\
-__asm__ __volatile__(							\
-	"raw_local_save_flags %0"					\
-	: "=r" (x))
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile("arch_local_save_flags %0" : "=r" (flags));
+	return flags;
+}
 
 __asm__(
-	"	.macro	raw_local_irq_save result			\n"
+	"	.macro	arch_local_irq_save result			\n"
 	"	.set	push						\n"
 	"	.set	reorder						\n"
 	"	.set	noat						\n"
@@ -148,15 +150,18 @@ __asm__(
 	"	.set	pop						\n"
 	"	.endm							\n");
 
-#define raw_local_irq_save(x)						\
-__asm__ __volatile__(							\
-	"raw_local_irq_save\t%0"					\
-	: "=r" (x)							\
-	: /* no inputs */						\
-	: "memory")
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags;
+	asm volatile("arch_local_irq_save\t%0"
+		     : "=r" (flags)
+		     : /* no inputs */
+		     : "memory");
+	return flags;
+}
 
 __asm__(
-	"	.macro	raw_local_irq_restore flags			\n"
+	"	.macro	arch_local_irq_restore flags			\n"
 	"	.set	push						\n"
 	"	.set	noreorder					\n"
 	"	.set	noat						\n"
@@ -196,7 +201,7 @@ __asm__(
 	"	.endm							\n");
 
 
-static inline void raw_local_irq_restore(unsigned long flags)
+static inline void arch_local_irq_restore(unsigned long flags)
 {
 	unsigned long __tmp1;
 
@@ -211,24 +216,24 @@ static inline void raw_local_irq_restore(unsigned long flags)
 #endif
 
 	__asm__ __volatile__(
-		"raw_local_irq_restore\t%0"
+		"arch_local_irq_restore\t%0"
 		: "=r" (__tmp1)
 		: "0" (flags)
 		: "memory");
 }
 
-static inline void __raw_local_irq_restore(unsigned long flags)
+static inline void __arch_local_irq_restore(unsigned long flags)
 {
 	unsigned long __tmp1;
 
 	__asm__ __volatile__(
-		"raw_local_irq_restore\t%0"
+		"arch_local_irq_restore\t%0"
 		: "=r" (__tmp1)
 		: "0" (flags)
 		: "memory");
 }
 
-static inline int raw_irqs_disabled_flags(unsigned long flags)
+static inline int arch_irqs_disabled_flags(unsigned long flags)
 {
 #ifdef CONFIG_MIPS_MT_SMTC
 	/*
diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c
index cfeb2c155896..39c08254b0f1 100644
--- a/arch/mips/kernel/smtc.c
+++ b/arch/mips/kernel/smtc.c
@@ -1038,7 +1038,7 @@ void deferred_smtc_ipi(void)
 		 * but it's more efficient, given that we're already
 		 * running down the IPI queue.
 		 */
-		__raw_local_irq_restore(flags);
+		__arch_local_irq_restore(flags);
 	}
 }
 
@@ -1190,7 +1190,7 @@ void smtc_ipi_replay(void)
 		/*
 		 ** But use a raw restore here to avoid recursion.
 		 */
-		__raw_local_irq_restore(flags);
+		__arch_local_irq_restore(flags);
 
 		if (pipi) {
 			self_ipi(pipi);
diff --git a/arch/mn10300/include/asm/irqflags.h b/arch/mn10300/include/asm/irqflags.h
new file mode 100644
index 000000000000..5e529a117cb2
--- /dev/null
+++ b/arch/mn10300/include/asm/irqflags.h
@@ -0,0 +1,123 @@
+/* MN10300 IRQ flag handling
+ *
+ * Copyright (C) 2010 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public Licence
+ * as published by the Free Software Foundation; either version
+ * 2 of the Licence, or (at your option) any later version.
+ */
+
+#ifndef _ASM_IRQFLAGS_H
+#define _ASM_IRQFLAGS_H
+
+#include <asm/cpu-regs.h>
+
+/*
+ * interrupt control
+ * - "disabled": run in IM1/2
+ *   - level 0 - GDB stub
+ *   - level 1 - virtual serial DMA (if present)
+ *   - level 5 - normal interrupt priority
+ *   - level 6 - timer interrupt
+ * - "enabled":  run in IM7
+ */
+#ifdef CONFIG_MN10300_TTYSM
+#define MN10300_CLI_LEVEL	EPSW_IM_2
+#else
+#define MN10300_CLI_LEVEL	EPSW_IM_1
+#endif
+
+#ifndef __ASSEMBLY__
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+
+	asm volatile("mov epsw,%0" : "=d"(flags));
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	asm volatile(
+		"	and %0,epsw	\n"
+		"	or %1,epsw	\n"
+		"	nop		\n"
+		"	nop		\n"
+		"	nop		\n"
+		:
+		: "i"(~EPSW_IM), "i"(EPSW_IE | MN10300_CLI_LEVEL)
+		: "memory");
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags;
+
+	flags = arch_local_save_flags();
+	arch_local_irq_disable();
+	return flags;
+}
+
+/*
+ * we make sure arch_irq_enable() doesn't cause priority inversion
+ */
+extern unsigned long __mn10300_irq_enabled_epsw;
+
+static inline void arch_local_irq_enable(void)
+{
+	unsigned long tmp;
+
+	asm volatile(
+		"	mov	epsw,%0		\n"
+		"	and	%1,%0		\n"
+		"	or	%2,%0		\n"
+		"	mov	%0,epsw		\n"
+		: "=&d"(tmp)
+		: "i"(~EPSW_IM), "r"(__mn10300_irq_enabled_epsw)
+		: "memory");
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile(
+		"	mov %0,epsw	\n"
+		"	nop		\n"
+		"	nop		\n"
+		"	nop		\n"
+		:
+		: "d"(flags)
+		: "memory", "cc");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & EPSW_IM) <= MN10300_CLI_LEVEL;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+/*
+ * Hook to save power by halting the CPU
+ * - called from the idle loop
+ * - must reenable interrupts (which takes three instruction cycles to complete)
+ */
+static inline void arch_safe_halt(void)
+{
+	asm volatile(
+		"	or	%0,epsw	\n"
+		"	nop		\n"
+		"	nop		\n"
+		"	bset	%2,(%1)	\n"
+		:
+		: "i"(EPSW_IE|EPSW_IM), "n"(&CPUM), "i"(CPUM_SLEEP)
+		: "cc");
+}
+
+#endif /* __ASSEMBLY__ */
+#endif /* _ASM_IRQFLAGS_H */
diff --git a/arch/mn10300/include/asm/system.h b/arch/mn10300/include/asm/system.h
index 3636c054dcd5..9f7c7e17c01e 100644
--- a/arch/mn10300/include/asm/system.h
+++ b/arch/mn10300/include/asm/system.h
@@ -17,6 +17,7 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/kernel.h>
+#include <linux/irqflags.h>
 
 struct task_struct;
 struct thread_struct;
@@ -79,114 +80,6 @@ do {									\
 #define read_barrier_depends()		do {} while (0)
 #define smp_read_barrier_depends()	do {} while (0)
 
-/*****************************************************************************/
-/*
- * interrupt control
- * - "disabled": run in IM1/2
- *   - level 0 - GDB stub
- *   - level 1 - virtual serial DMA (if present)
- *   - level 5 - normal interrupt priority
- *   - level 6 - timer interrupt
- * - "enabled":  run in IM7
- */
-#ifdef CONFIG_MN10300_TTYSM
-#define MN10300_CLI_LEVEL	EPSW_IM_2
-#else
-#define MN10300_CLI_LEVEL	EPSW_IM_1
-#endif
-
-#define local_save_flags(x)			\
-do {						\
-	typecheck(unsigned long, x);		\
-	asm volatile(				\
-		"	mov epsw,%0	\n"	\
-		: "=d"(x)			\
-		);				\
-} while (0)
-
-#define local_irq_disable()						\
-do {									\
-	asm volatile(							\
-		"	and %0,epsw	\n"				\
-		"	or %1,epsw	\n"				\
-		"	nop		\n"				\
-		"	nop		\n"				\
-		"	nop		\n"				\
-		:							\
-		: "i"(~EPSW_IM), "i"(EPSW_IE | MN10300_CLI_LEVEL)	\
-		);							\
-} while (0)
-
-#define local_irq_save(x)			\
-do {						\
-	local_save_flags(x);			\
-	local_irq_disable();			\
-} while (0)
-
-/*
- * we make sure local_irq_enable() doesn't cause priority inversion
- */
-#ifndef __ASSEMBLY__
-
-extern unsigned long __mn10300_irq_enabled_epsw;
-
-#endif
-
-#define local_irq_enable()						\
-do {									\
-	unsigned long tmp;						\
-									\
-	asm volatile(							\
-		"	mov	epsw,%0		\n"			\
-		"	and	%1,%0		\n"			\
-		"	or	%2,%0		\n"			\
-		"	mov	%0,epsw		\n"			\
-		: "=&d"(tmp)						\
-		: "i"(~EPSW_IM), "r"(__mn10300_irq_enabled_epsw)	\
-		: "cc"							\
-		);							\
-} while (0)
-
-#define local_irq_restore(x)			\
-do {						\
-	typecheck(unsigned long, x);		\
-	asm volatile(				\
-		"	mov %0,epsw	\n"	\
-		"	nop		\n"	\
-		"	nop		\n"	\
-		"	nop		\n"	\
-		:				\
-		: "d"(x)			\
-		: "memory", "cc"		\
-		);				\
-} while (0)
-
-#define irqs_disabled()				\
-({						\
-	unsigned long flags;			\
-	local_save_flags(flags);		\
-	(flags & EPSW_IM) <= MN10300_CLI_LEVEL;	\
-})
-
-/* hook to save power by halting the CPU
- * - called from the idle loop
- * - must reenable interrupts (which takes three instruction cycles to complete)
- */
-#define safe_halt()							\
-do {									\
-	asm volatile("	or	%0,epsw	\n"				\
-		     "	nop		\n"				\
-		     "	nop		\n"				\
-		     "	bset	%2,(%1)	\n"				\
-		     :							\
-		     : "i"(EPSW_IE|EPSW_IM), "n"(&CPUM), "i"(CPUM_SLEEP)\
-		     : "cc"						\
-		     );							\
-} while (0)
-
-#define STI	or EPSW_IE|EPSW_IM,epsw
-#define CLI	and ~EPSW_IM,epsw; or EPSW_IE|MN10300_CLI_LEVEL,epsw; nop; nop; nop
-
 /*****************************************************************************/
 /*
  * MN10300 doesn't actually have an exchange instruction
diff --git a/arch/mn10300/kernel/entry.S b/arch/mn10300/kernel/entry.S
index d9ed5a15c547..3d394b4eefba 100644
--- a/arch/mn10300/kernel/entry.S
+++ b/arch/mn10300/kernel/entry.S
@@ -16,6 +16,7 @@
 #include <linux/linkage.h>
 #include <asm/smp.h>
 #include <asm/system.h>
+#include <asm/irqflags.h>
 #include <asm/thread_info.h>
 #include <asm/intctl-regs.h>
 #include <asm/busctl-regs.h>
diff --git a/arch/parisc/include/asm/irqflags.h b/arch/parisc/include/asm/irqflags.h
new file mode 100644
index 000000000000..34f9cb9b4754
--- /dev/null
+++ b/arch/parisc/include/asm/irqflags.h
@@ -0,0 +1,46 @@
+#ifndef __PARISC_IRQFLAGS_H
+#define __PARISC_IRQFLAGS_H
+
+#include <linux/types.h>
+#include <asm/psw.h>
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile("ssm 0, %0" : "=r" (flags) : : "memory");
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	asm volatile("rsm %0,%%r0\n" : : "i" (PSW_I) : "memory");
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	asm volatile("ssm %0,%%r0\n" : : "i" (PSW_I) : "memory");
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags;
+	asm volatile("rsm %1,%0" : "=r" (flags) : "i" (PSW_I) : "memory");
+	return flags;
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile("mtsm %0" : : "r" (flags) : "memory");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & PSW_I) == 0;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* __PARISC_IRQFLAGS_H */
diff --git a/arch/parisc/include/asm/system.h b/arch/parisc/include/asm/system.h
index 2ab4af58ecb9..b19e63a8e848 100644
--- a/arch/parisc/include/asm/system.h
+++ b/arch/parisc/include/asm/system.h
@@ -1,7 +1,7 @@
 #ifndef __PARISC_SYSTEM_H
 #define __PARISC_SYSTEM_H
 
-#include <asm/psw.h>
+#include <linux/irqflags.h>
 
 /* The program status word as bitfields.  */
 struct pa_psw {
@@ -48,23 +48,6 @@ extern struct task_struct *_switch_to(struct task_struct *, struct task_struct *
 	(last) = _switch_to(prev, next);			\
 } while(0)
 
-/* interrupt control */
-#define local_save_flags(x)	__asm__ __volatile__("ssm 0, %0" : "=r" (x) : : "memory")
-#define local_irq_disable()	__asm__ __volatile__("rsm %0,%%r0\n" : : "i" (PSW_I) : "memory" )
-#define local_irq_enable()	__asm__ __volatile__("ssm %0,%%r0\n" : : "i" (PSW_I) : "memory" )
-
-#define local_irq_save(x) \
-	__asm__ __volatile__("rsm %1,%0" : "=r" (x) :"i" (PSW_I) : "memory" )
-#define local_irq_restore(x) \
-	__asm__ __volatile__("mtsm %0" : : "r" (x) : "memory" )
-
-#define irqs_disabled()			\
-({					\
-	unsigned long flags;		\
-	local_save_flags(flags);	\
-	(flags & PSW_I) == 0;		\
-})
-
 #define mfctl(reg)	({		\
 	unsigned long cr;		\
 	__asm__ __volatile__(		\
diff --git a/arch/powerpc/include/asm/hw_irq.h b/arch/powerpc/include/asm/hw_irq.h
index bd100fcf40d0..ff08b70b36d4 100644
--- a/arch/powerpc/include/asm/hw_irq.h
+++ b/arch/powerpc/include/asm/hw_irq.h
@@ -16,42 +16,57 @@ extern void timer_interrupt(struct pt_regs *);
 #ifdef CONFIG_PPC64
 #include <asm/paca.h>
 
-static inline unsigned long local_get_flags(void)
+static inline unsigned long arch_local_save_flags(void)
 {
 	unsigned long flags;
 
-	__asm__ __volatile__("lbz %0,%1(13)"
-	: "=r" (flags)
-	: "i" (offsetof(struct paca_struct, soft_enabled)));
+	asm volatile(
+		"lbz %0,%1(13)"
+		: "=r" (flags)
+		: "i" (offsetof(struct paca_struct, soft_enabled)));
 
 	return flags;
 }
 
-static inline unsigned long raw_local_irq_disable(void)
+static inline unsigned long arch_local_irq_disable(void)
 {
 	unsigned long flags, zero;
 
-	__asm__ __volatile__("li %1,0; lbz %0,%2(13); stb %1,%2(13)"
-	: "=r" (flags), "=&r" (zero)
-	: "i" (offsetof(struct paca_struct, soft_enabled))
-	: "memory");
+	asm volatile(
+		"li %1,0; lbz %0,%2(13); stb %1,%2(13)"
+		: "=r" (flags), "=&r" (zero)
+		: "i" (offsetof(struct paca_struct, soft_enabled))
+		: "memory");
 
 	return flags;
 }
 
-extern void raw_local_irq_restore(unsigned long);
+extern void arch_local_irq_restore(unsigned long);
 extern void iseries_handle_interrupts(void);
 
-#define raw_local_irq_enable()		raw_local_irq_restore(1)
-#define raw_local_save_flags(flags)	((flags) = local_get_flags())
-#define raw_local_irq_save(flags)	((flags) = raw_local_irq_disable())
+static inline void arch_local_irq_enable(void)
+{
+	arch_local_irq_restore(1);
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	return arch_local_irq_disable();
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return flags == 0;
+}
 
-#define raw_irqs_disabled()		(local_get_flags() == 0)
-#define raw_irqs_disabled_flags(flags)	((flags) == 0)
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
 
 #ifdef CONFIG_PPC_BOOK3E
-#define __hard_irq_enable()	__asm__ __volatile__("wrteei 1": : :"memory");
-#define __hard_irq_disable()	__asm__ __volatile__("wrteei 0": : :"memory");
+#define __hard_irq_enable()	asm volatile("wrteei 1" : : : "memory");
+#define __hard_irq_disable()	asm volatile("wrteei 0" : : : "memory");
 #else
 #define __hard_irq_enable()	__mtmsrd(mfmsr() | MSR_EE, 1)
 #define __hard_irq_disable()	__mtmsrd(mfmsr() & ~MSR_EE, 1)
@@ -64,64 +79,66 @@ extern void iseries_handle_interrupts(void);
 		get_paca()->hard_enabled = 0;	\
 	} while(0)
 
-#else
+#else /* CONFIG_PPC64 */
 
-#if defined(CONFIG_BOOKE)
 #define SET_MSR_EE(x)	mtmsr(x)
-#define raw_local_irq_restore(flags)	__asm__ __volatile__("wrtee %0" : : "r" (flags) : "memory")
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	return mfmsr();
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+#if defined(CONFIG_BOOKE)
+	asm volatile("wrtee %0" : : "r" (flags) : "memory");
 #else
-#define SET_MSR_EE(x)	mtmsr(x)
-#define raw_local_irq_restore(flags)	mtmsr(flags)
+	mtmsr(flags);
 #endif
+}
 
-static inline void raw_local_irq_disable(void)
+static inline unsigned long arch_local_irq_save(void)
 {
+	unsigned long flags = arch_local_save_flags();
 #ifdef CONFIG_BOOKE
-	__asm__ __volatile__("wrteei 0": : :"memory");
+	asm volatile("wrteei 0" : : : "memory");
 #else
-	unsigned long msr;
-
-	msr = mfmsr();
-	SET_MSR_EE(msr & ~MSR_EE);
+	SET_MSR_EE(flags & ~MSR_EE);
 #endif
+	return flags;
 }
 
-static inline void raw_local_irq_enable(void)
+static inline void arch_local_irq_disable(void)
 {
 #ifdef CONFIG_BOOKE
-	__asm__ __volatile__("wrteei 1": : :"memory");
+	asm volatile("wrteei 0" : : : "memory");
 #else
-	unsigned long msr;
-
-	msr = mfmsr();
-	SET_MSR_EE(msr | MSR_EE);
+	arch_local_irq_save();
 #endif
 }
 
-static inline void raw_local_irq_save_ptr(unsigned long *flags)
+static inline void arch_local_irq_enable(void)
 {
-	unsigned long msr;
-	msr = mfmsr();
-	*flags = msr;
 #ifdef CONFIG_BOOKE
-	__asm__ __volatile__("wrteei 0": : :"memory");
+	asm volatile("wrteei 1" : : : "memory");
 #else
-	SET_MSR_EE(msr & ~MSR_EE);
+	unsigned long msr = mfmsr();
+	SET_MSR_EE(msr | MSR_EE);
 #endif
 }
 
-#define raw_local_save_flags(flags)	((flags) = mfmsr())
-#define raw_local_irq_save(flags)	raw_local_irq_save_ptr(&flags)
-#define raw_irqs_disabled()		((mfmsr() & MSR_EE) == 0)
-#define raw_irqs_disabled_flags(flags)	(((flags) & MSR_EE) == 0)
-
-#define hard_irq_disable()		raw_local_irq_disable()
-
-static inline int irqs_disabled_flags(unsigned long flags)
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
 {
 	return (flags & MSR_EE) == 0;
 }
 
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#define hard_irq_disable()		arch_local_irq_disable()
+
 #endif /* CONFIG_PPC64 */
 
 /*
diff --git a/arch/powerpc/include/asm/irqflags.h b/arch/powerpc/include/asm/irqflags.h
index 5f68ecfdf516..b85d8ddbb666 100644
--- a/arch/powerpc/include/asm/irqflags.h
+++ b/arch/powerpc/include/asm/irqflags.h
@@ -6,7 +6,7 @@
 
 #ifndef __ASSEMBLY__
 /*
- * Get definitions for raw_local_save_flags(x), etc.
+ * Get definitions for arch_local_save_flags(x), etc.
  */
 #include <asm/hw_irq.h>
 
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index f53029a01554..39b0c48872d2 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -818,12 +818,12 @@ END_FW_FTR_SECTION_IFCLR(FW_FEATURE_ISERIES)
 
 	/*
 	 * hash_page couldn't handle it, set soft interrupt enable back
-	 * to what it was before the trap.  Note that .raw_local_irq_restore
+	 * to what it was before the trap.  Note that .arch_local_irq_restore
 	 * handles any interrupts pending at this point.
 	 */
 	ld	r3,SOFTE(r1)
 	TRACE_AND_RESTORE_IRQ_PARTIAL(r3, 11f)
-	bl	.raw_local_irq_restore
+	bl	.arch_local_irq_restore
 	b	11f
 
 /* We have a data breakpoint exception - handle it */
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index 4a65386995d7..1903290f5469 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -116,7 +116,7 @@ static inline notrace void set_soft_enabled(unsigned long enable)
 	: : "r" (enable), "i" (offsetof(struct paca_struct, soft_enabled)));
 }
 
-notrace void raw_local_irq_restore(unsigned long en)
+notrace void arch_local_irq_restore(unsigned long en)
 {
 	/*
 	 * get_paca()->soft_enabled = en;
@@ -192,7 +192,7 @@ notrace void raw_local_irq_restore(unsigned long en)
 
 	__hard_irq_enable();
 }
-EXPORT_SYMBOL(raw_local_irq_restore);
+EXPORT_SYMBOL(arch_local_irq_restore);
 #endif /* CONFIG_PPC64 */
 
 static int show_other_interrupts(struct seq_file *p, int prec)
diff --git a/arch/s390/include/asm/irqflags.h b/arch/s390/include/asm/irqflags.h
index 15b3ac253898..865d6d891ace 100644
--- a/arch/s390/include/asm/irqflags.h
+++ b/arch/s390/include/asm/irqflags.h
@@ -8,8 +8,8 @@
 
 #include <linux/types.h>
 
-/* store then or system mask. */
-#define __raw_local_irq_stosm(__or)					\
+/* store then OR system mask. */
+#define __arch_local_irq_stosm(__or)					\
 ({									\
 	unsigned long __mask;						\
 	asm volatile(							\
@@ -18,8 +18,8 @@
 	__mask;								\
 })
 
-/* store then and system mask. */
-#define __raw_local_irq_stnsm(__and)					\
+/* store then AND system mask. */
+#define __arch_local_irq_stnsm(__and)					\
 ({									\
 	unsigned long __mask;						\
 	asm volatile(							\
@@ -29,39 +29,44 @@
 })
 
 /* set system mask. */
-#define __raw_local_irq_ssm(__mask)					\
-({									\
-	asm volatile("ssm   %0" : : "Q" (__mask) : "memory");		\
-})
+static inline void __arch_local_irq_ssm(unsigned long flags)
+{
+	asm volatile("ssm   %0" : : "Q" (flags) : "memory");
+}
 
-/* interrupt control.. */
-static inline unsigned long raw_local_irq_enable(void)
+static inline unsigned long arch_local_save_flags(void)
 {
-	return __raw_local_irq_stosm(0x03);
+	return __arch_local_irq_stosm(0x00);
 }
 
-static inline unsigned long raw_local_irq_disable(void)
+static inline unsigned long arch_local_irq_save(void)
 {
-	return __raw_local_irq_stnsm(0xfc);
+	return __arch_local_irq_stnsm(0xfc);
 }
 
-#define raw_local_save_flags(x)						\
-do {									\
-	typecheck(unsigned long, x);					\
-	(x) = __raw_local_irq_stosm(0x00);				\
-} while (0)
+static inline void arch_local_irq_disable(void)
+{
+	arch_local_irq_save();
+}
 
-static inline void raw_local_irq_restore(unsigned long flags)
+static inline void arch_local_irq_enable(void)
 {
-	__raw_local_irq_ssm(flags);
+	__arch_local_irq_stosm(0x03);
 }
 
-static inline int raw_irqs_disabled_flags(unsigned long flags)
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	__arch_local_irq_ssm(flags);
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
 {
 	return !(flags & (3UL << (BITS_PER_LONG - 8)));
 }
 
-/* For spinlocks etc */
-#define raw_local_irq_save(x)	((x) = raw_local_irq_disable())
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
 
 #endif /* __ASM_IRQFLAGS_H */
diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h
index cef66210c846..8e8a50eeed92 100644
--- a/arch/s390/include/asm/system.h
+++ b/arch/s390/include/asm/system.h
@@ -399,7 +399,7 @@ static inline unsigned long __cmpxchg_local(volatile void *ptr,
 static inline void
 __set_psw_mask(unsigned long mask)
 {
-	__load_psw_mask(mask | (__raw_local_irq_stosm(0x00) & ~(-1UL >> 8)));
+	__load_psw_mask(mask | (arch_local_save_flags() & ~(-1UL >> 8)));
 }
 
 #define local_mcck_enable()  __set_psw_mask(psw_kernel_bits)
diff --git a/arch/s390/kernel/mem_detect.c b/arch/s390/kernel/mem_detect.c
index 559af0d07878..0fbe4e32f7ba 100644
--- a/arch/s390/kernel/mem_detect.c
+++ b/arch/s390/kernel/mem_detect.c
@@ -54,11 +54,11 @@ void detect_memory_layout(struct mem_chunk chunk[])
 	 * right thing and we don't get scheduled away with low address
 	 * protection disabled.
 	 */
-	flags = __raw_local_irq_stnsm(0xf8);
+	flags = __arch_local_irq_stnsm(0xf8);
 	__ctl_store(cr0, 0, 0);
 	__ctl_clear_bit(0, 28);
 	find_memory_chunks(chunk);
 	__ctl_load(cr0, 0, 0);
-	__raw_local_irq_ssm(flags);
+	arch_local_irq_restore(flags);
 }
 EXPORT_SYMBOL(detect_memory_layout);
diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index 30eb6d02ddb8..94b8ba2ec857 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -50,7 +50,6 @@ EXPORT_SYMBOL(empty_zero_page);
  */
 void __init paging_init(void)
 {
-	static const int ssm_mask = 0x04000000L;
 	unsigned long max_zone_pfns[MAX_NR_ZONES];
 	unsigned long pgd_type;
 
@@ -72,7 +71,7 @@ void __init paging_init(void)
 	__ctl_load(S390_lowcore.kernel_asce, 1, 1);
 	__ctl_load(S390_lowcore.kernel_asce, 7, 7);
 	__ctl_load(S390_lowcore.kernel_asce, 13, 13);
-	__raw_local_irq_ssm(ssm_mask);
+	arch_local_irq_restore(4UL << (BITS_PER_LONG - 8));
 
 	atomic_set(&init_mm.context.attach_count, 1);
 
diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c
index a8c2af8c650f..71a4b0d34be0 100644
--- a/arch/s390/mm/maccess.c
+++ b/arch/s390/mm/maccess.c
@@ -71,7 +71,7 @@ int memcpy_real(void *dest, void *src, size_t count)
 
 	if (!count)
 		return 0;
-	flags = __raw_local_irq_stnsm(0xf8UL);
+	flags = __arch_local_irq_stnsm(0xf8UL);
 	asm volatile (
 		"0:	mvcle	%1,%2,0x0\n"
 		"1:	jo	0b\n"
@@ -82,6 +82,6 @@ int memcpy_real(void *dest, void *src, size_t count)
 		  "+d" (_len2), "=m" (*((long *) dest))
 		: "m" (*((long *) src))
 		: "cc", "memory");
-	__raw_local_irq_ssm(flags);
+	arch_local_irq_restore(flags);
 	return rc;
 }
diff --git a/arch/score/include/asm/irqflags.h b/arch/score/include/asm/irqflags.h
index 690a6cae7294..5c7563891e28 100644
--- a/arch/score/include/asm/irqflags.h
+++ b/arch/score/include/asm/irqflags.h
@@ -3,107 +3,118 @@
 
 #ifndef __ASSEMBLY__
 
-#define raw_local_irq_save(x)			\
-{						\
-	__asm__ __volatile__(			\
-		"mfcr	r8, cr0;"		\
-		"li	r9, 0xfffffffe;"	\
-		"nop;"				\
-		"mv	%0, r8;"		\
-		"and	r8, r8, r9;"		\
-		"mtcr	r8, cr0;"		\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		: "=r" (x)			\
-		:				\
-		: "r8", "r9"			\
-		);				\
+#include <linux/types.h>
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+
+	asm volatile(
+		"	mfcr	r8, cr0		\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	mv	%0, r8		\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	ldi	r9, 0x1		\n"
+		"	and	%0, %0, r9	\n"
+		: "=r" (flags)
+		:
+		: "r8", "r9");
+	return flags;
 }
 
-#define raw_local_irq_restore(x)		\
-{						\
-	__asm__ __volatile__(			\
-		"mfcr	r8, cr0;"		\
-		"ldi	r9, 0x1;"		\
-		"and	%0, %0, r9;"		\
-		"or	r8, r8, %0;"		\
-		"mtcr	r8, cr0;"		\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		:				\
-		: "r"(x)			\
-		: "r8", "r9"			\
-		);				\
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags
+
+	asm volatile(
+		"	mfcr	r8, cr0		\n"
+		"	li	r9, 0xfffffffe	\n"
+		"	nop			\n"
+		"	mv	%0, r8		\n"
+		"	and	r8, r8, r9	\n"
+		"	mtcr	r8, cr0		\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		: "=r" (flags)
+		:
+		: "r8", "r9", "memory");
+
+	return flags;
 }
 
-#define raw_local_irq_enable(void)		\
-{						\
-	__asm__ __volatile__(			\
-		"mfcr\tr8,cr0;"			\
-		"nop;"				\
-		"nop;"				\
-		"ori\tr8,0x1;"			\
-		"mtcr\tr8,cr0;"			\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		:				\
-		:				\
-		: "r8");			\
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile(
+		"	mfcr	r8, cr0		\n"
+		"	ldi	r9, 0x1		\n"
+		"	and	%0, %0, r9	\n"
+		"	or	r8, r8, %0	\n"
+		"	mtcr	r8, cr0		\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		:
+		: "r"(flags)
+		: "r8", "r9", "memory");
 }
 
-#define raw_local_irq_disable(void)		\
-{						\
-	__asm__ __volatile__(			\
-		"mfcr\tr8,cr0;"			\
-		"nop;"				\
-		"nop;"				\
-		"srli\tr8,r8,1;"		\
-		"slli\tr8,r8,1;"		\
-		"mtcr\tr8,cr0;"			\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		:				\
-		:				\
-		: "r8");			\
+static inline void arch_local_irq_enable(void)
+{
+	asm volatile(
+		"	mfcr	r8,cr0		\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	ori	r8,0x1		\n"
+		"	mtcr	r8,cr0		\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		:
+		:
+		: "r8", "memory");
 }
 
-#define raw_local_save_flags(x)			\
-{						\
-	__asm__ __volatile__(			\
-		"mfcr	r8, cr0;"		\
-		"nop;"				\
-		"nop;"				\
-		"mv	%0, r8;"		\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"nop;"				\
-		"ldi	r9, 0x1;"		\
-		"and	%0, %0, r9;"		\
-		: "=r" (x)			\
-		:				\
-		: "r8", "r9"			\
-		);				\
+static inline void arch_local_irq_disable(void)
+{
+	asm volatile(
+		"	mfcr	r8,cr0		\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	srli	r8,r8,1		\n"
+		"	slli	r8,r8,1		\n"
+		"	mtcr	r8,cr0		\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		"	nop			\n"
+		:
+		:
+		: "r8", "memory");
 }
 
-static inline int raw_irqs_disabled_flags(unsigned long flags)
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
 {
 	return !(flags & 1);
 }
 
-#endif
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* __ASSEMBLY__ */
 
 #endif /* _ASM_SCORE_IRQFLAGS_H */
diff --git a/arch/sh/include/asm/irqflags.h b/arch/sh/include/asm/irqflags.h
index a741153b41c2..43b7608606c3 100644
--- a/arch/sh/include/asm/irqflags.h
+++ b/arch/sh/include/asm/irqflags.h
@@ -1,8 +1,8 @@
 #ifndef __ASM_SH_IRQFLAGS_H
 #define __ASM_SH_IRQFLAGS_H
 
-#define RAW_IRQ_DISABLED	0xf0
-#define RAW_IRQ_ENABLED		0x00
+#define ARCH_IRQ_DISABLED	0xf0
+#define ARCH_IRQ_ENABLED	0x00
 
 #include <asm-generic/irqflags.h>
 
diff --git a/arch/sh/kernel/irq_32.c b/arch/sh/kernel/irq_32.c
index e33ab15831f9..e5a755be9129 100644
--- a/arch/sh/kernel/irq_32.c
+++ b/arch/sh/kernel/irq_32.c
@@ -10,11 +10,11 @@
 #include <linux/irqflags.h>
 #include <linux/module.h>
 
-void notrace raw_local_irq_restore(unsigned long flags)
+void notrace arch_local_irq_restore(unsigned long flags)
 {
 	unsigned long __dummy0, __dummy1;
 
-	if (flags == RAW_IRQ_DISABLED) {
+	if (flags == ARCH_IRQ_DISABLED) {
 		__asm__ __volatile__ (
 			"stc	sr, %0\n\t"
 			"or	#0xf0, %0\n\t"
@@ -33,14 +33,14 @@ void notrace raw_local_irq_restore(unsigned long flags)
 #endif
 			"ldc	%0, sr\n\t"
 			: "=&r" (__dummy0), "=r" (__dummy1)
-			: "1" (~RAW_IRQ_DISABLED)
+			: "1" (~ARCH_IRQ_DISABLED)
 			: "memory"
 		);
 	}
 }
-EXPORT_SYMBOL(raw_local_irq_restore);
+EXPORT_SYMBOL(arch_local_irq_restore);
 
-unsigned long notrace __raw_local_save_flags(void)
+unsigned long notrace arch_local_save_flags(void)
 {
 	unsigned long flags;
 
@@ -54,4 +54,4 @@ unsigned long notrace __raw_local_save_flags(void)
 
 	return flags;
 }
-EXPORT_SYMBOL(__raw_local_save_flags);
+EXPORT_SYMBOL(arch_local_save_flags);
diff --git a/arch/sparc/include/asm/irqflags_32.h b/arch/sparc/include/asm/irqflags_32.h
index 0fca9d97d44f..d4d0711de0f9 100644
--- a/arch/sparc/include/asm/irqflags_32.h
+++ b/arch/sparc/include/asm/irqflags_32.h
@@ -5,33 +5,40 @@
  *
  * This file gets included from lowlevel asm headers too, to provide
  * wrapped versions of the local_irq_*() APIs, based on the
- * raw_local_irq_*() functions from the lowlevel headers.
+ * arch_local_irq_*() functions from the lowlevel headers.
  */
 #ifndef _ASM_IRQFLAGS_H
 #define _ASM_IRQFLAGS_H
 
 #ifndef __ASSEMBLY__
 
-extern void raw_local_irq_restore(unsigned long);
-extern unsigned long __raw_local_irq_save(void);
-extern void raw_local_irq_enable(void);
+#include <linux/types.h>
 
-static inline unsigned long getipl(void)
+extern void arch_local_irq_restore(unsigned long);
+extern unsigned long arch_local_irq_save(void);
+extern void arch_local_irq_enable(void);
+
+static inline unsigned long arch_local_save_flags(void)
 {
-        unsigned long retval;
+	unsigned long flags;
+
+	asm volatile("rd        %%psr, %0" : "=r" (flags));
+	return flags;
+}
 
-        __asm__ __volatile__("rd        %%psr, %0" : "=r" (retval));
-        return retval;
+static inline void arch_local_irq_disable(void)
+{
+	arch_local_irq_save();
 }
 
-#define raw_local_save_flags(flags) ((flags) = getipl())
-#define raw_local_irq_save(flags)   ((flags) = __raw_local_irq_save())
-#define raw_local_irq_disable()     ((void) __raw_local_irq_save())
-#define raw_irqs_disabled()         ((getipl() & PSR_PIL) != 0)
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & PSR_PIL) != 0;
+}
 
-static inline int raw_irqs_disabled_flags(unsigned long flags)
+static inline bool arch_irqs_disabled(void)
 {
-        return ((flags & PSR_PIL) != 0);
+	return arch_irqs_disabled_flags(arch_local_save_flags());
 }
 
 #endif /* (__ASSEMBLY__) */
diff --git a/arch/sparc/include/asm/irqflags_64.h b/arch/sparc/include/asm/irqflags_64.h
index bfa1ea45b4cd..aab969c82c2b 100644
--- a/arch/sparc/include/asm/irqflags_64.h
+++ b/arch/sparc/include/asm/irqflags_64.h
@@ -5,7 +5,7 @@
  *
  * This file gets included from lowlevel asm headers too, to provide
  * wrapped versions of the local_irq_*() APIs, based on the
- * raw_local_irq_*() functions from the lowlevel headers.
+ * arch_local_irq_*() functions from the lowlevel headers.
  */
 #ifndef _ASM_IRQFLAGS_H
 #define _ASM_IRQFLAGS_H
@@ -14,7 +14,7 @@
 
 #ifndef __ASSEMBLY__
 
-static inline unsigned long __raw_local_save_flags(void)
+static inline unsigned long arch_local_save_flags(void)
 {
 	unsigned long flags;
 
@@ -26,10 +26,7 @@ static inline unsigned long __raw_local_save_flags(void)
 	return flags;
 }
 
-#define raw_local_save_flags(flags) \
-		do { (flags) = __raw_local_save_flags(); } while (0)
-
-static inline void raw_local_irq_restore(unsigned long flags)
+static inline void arch_local_irq_restore(unsigned long flags)
 {
 	__asm__ __volatile__(
 		"wrpr	%0, %%pil"
@@ -39,7 +36,7 @@ static inline void raw_local_irq_restore(unsigned long flags)
 	);
 }
 
-static inline void raw_local_irq_disable(void)
+static inline void arch_local_irq_disable(void)
 {
 	__asm__ __volatile__(
 		"wrpr	%0, %%pil"
@@ -49,7 +46,7 @@ static inline void raw_local_irq_disable(void)
 	);
 }
 
-static inline void raw_local_irq_enable(void)
+static inline void arch_local_irq_enable(void)
 {
 	__asm__ __volatile__(
 		"wrpr	0, %%pil"
@@ -59,22 +56,17 @@ static inline void raw_local_irq_enable(void)
 	);
 }
 
-static inline int raw_irqs_disabled_flags(unsigned long flags)
+static inline int arch_irqs_disabled_flags(unsigned long flags)
 {
 	return (flags > 0);
 }
 
-static inline int raw_irqs_disabled(void)
+static inline int arch_irqs_disabled(void)
 {
-	unsigned long flags = __raw_local_save_flags();
-
-	return raw_irqs_disabled_flags(flags);
+	return arch_irqs_disabled_flags(arch_local_save_flags());
 }
 
-/*
- * For spinlocks, etc:
- */
-static inline unsigned long __raw_local_irq_save(void)
+static inline unsigned long arch_local_irq_save(void)
 {
 	unsigned long flags, tmp;
 
@@ -100,9 +92,6 @@ static inline unsigned long __raw_local_irq_save(void)
 	return flags;
 }
 
-#define raw_local_irq_save(flags) \
-		do { (flags) = __raw_local_irq_save(); } while (0)
-
 #endif /* (__ASSEMBLY__) */
 
 #endif /* !(_ASM_IRQFLAGS_H) */
diff --git a/arch/sparc/kernel/irq_32.c b/arch/sparc/kernel/irq_32.c
index e1af43728329..0116d8d10def 100644
--- a/arch/sparc/kernel/irq_32.c
+++ b/arch/sparc/kernel/irq_32.c
@@ -57,7 +57,7 @@
 #define SMP_NOP2
 #define SMP_NOP3
 #endif /* SMP */
-unsigned long __raw_local_irq_save(void)
+unsigned long arch_local_irq_save(void)
 {
 	unsigned long retval;
 	unsigned long tmp;
@@ -74,8 +74,9 @@ unsigned long __raw_local_irq_save(void)
 
 	return retval;
 }
+EXPORT_SYMBOL(arch_local_irq_save);
 
-void raw_local_irq_enable(void)
+void arch_local_irq_enable(void)
 {
 	unsigned long tmp;
 
@@ -89,8 +90,9 @@ void raw_local_irq_enable(void)
 		: "i" (PSR_PIL)
 		: "memory");
 }
+EXPORT_SYMBOL(arch_local_irq_enable);
 
-void raw_local_irq_restore(unsigned long old_psr)
+void arch_local_irq_restore(unsigned long old_psr)
 {
 	unsigned long tmp;
 
@@ -105,10 +107,7 @@ void raw_local_irq_restore(unsigned long old_psr)
 		: "i" (PSR_PIL), "r" (old_psr)
 		: "memory");
 }
-
-EXPORT_SYMBOL(__raw_local_irq_save);
-EXPORT_SYMBOL(raw_local_irq_enable);
-EXPORT_SYMBOL(raw_local_irq_restore);
+EXPORT_SYMBOL(arch_local_irq_restore);
 
 /*
  * Dave Redman (djhr@tadpole.co.uk)
diff --git a/arch/sparc/prom/p1275.c b/arch/sparc/prom/p1275.c
index fa6e4e219b9c..d9850c2b9bf2 100644
--- a/arch/sparc/prom/p1275.c
+++ b/arch/sparc/prom/p1275.c
@@ -39,7 +39,7 @@ void p1275_cmd_direct(unsigned long *args)
 	unsigned long flags;
 
 	raw_local_save_flags(flags);
-	raw_local_irq_restore(PIL_NMI);
+	raw_local_irq_restore((unsigned long)PIL_NMI);
 	raw_spin_lock(&prom_entry_lock);
 
 	prom_world(1);
diff --git a/arch/tile/include/asm/irqflags.h b/arch/tile/include/asm/irqflags.h
index 45cf67c2f286..a11d4837ee4d 100644
--- a/arch/tile/include/asm/irqflags.h
+++ b/arch/tile/include/asm/irqflags.h
@@ -103,55 +103,57 @@ DECLARE_PER_CPU(unsigned long long, interrupts_enabled_mask);
 #define INITIAL_INTERRUPTS_ENABLED INT_MASK(INT_MEM_ERROR)
 
 /* Disable interrupts. */
-#define raw_local_irq_disable() \
+#define arch_local_irq_disable() \
 	interrupt_mask_set_mask(LINUX_MASKABLE_INTERRUPTS)
 
 /* Disable all interrupts, including NMIs. */
-#define raw_local_irq_disable_all() \
+#define arch_local_irq_disable_all() \
 	interrupt_mask_set_mask(-1UL)
 
 /* Re-enable all maskable interrupts. */
-#define raw_local_irq_enable() \
+#define arch_local_irq_enable() \
 	interrupt_mask_reset_mask(__get_cpu_var(interrupts_enabled_mask))
 
 /* Disable or enable interrupts based on flag argument. */
-#define raw_local_irq_restore(disabled) do { \
+#define arch_local_irq_restore(disabled) do { \
 	if (disabled) \
-		raw_local_irq_disable(); \
+		arch_local_irq_disable(); \
 	else \
-		raw_local_irq_enable(); \
+		arch_local_irq_enable(); \
 } while (0)
 
 /* Return true if "flags" argument means interrupts are disabled. */
-#define raw_irqs_disabled_flags(flags) ((flags) != 0)
+#define arch_irqs_disabled_flags(flags) ((flags) != 0)
 
 /* Return true if interrupts are currently disabled. */
-#define raw_irqs_disabled() interrupt_mask_check(INT_MEM_ERROR)
+#define arch_irqs_disabled() interrupt_mask_check(INT_MEM_ERROR)
 
 /* Save whether interrupts are currently disabled. */
-#define raw_local_save_flags(flags) ((flags) = raw_irqs_disabled())
+#define arch_local_save_flags() arch_irqs_disabled()
 
 /* Save whether interrupts are currently disabled, then disable them. */
-#define raw_local_irq_save(flags) \
-	do { raw_local_save_flags(flags); raw_local_irq_disable(); } while (0)
+#define arch_local_irq_save() ({ \
+	unsigned long __flags = arch_local_save_flags(); \
+	arch_local_irq_disable(); \
+	__flags; })
 
 /* Prevent the given interrupt from being enabled next time we enable irqs. */
-#define raw_local_irq_mask(interrupt) \
+#define arch_local_irq_mask(interrupt) \
 	(__get_cpu_var(interrupts_enabled_mask) &= ~INT_MASK(interrupt))
 
 /* Prevent the given interrupt from being enabled immediately. */
-#define raw_local_irq_mask_now(interrupt) do { \
-	raw_local_irq_mask(interrupt); \
+#define arch_local_irq_mask_now(interrupt) do { \
+	arch_local_irq_mask(interrupt); \
 	interrupt_mask_set(interrupt); \
 } while (0)
 
 /* Allow the given interrupt to be enabled next time we enable irqs. */
-#define raw_local_irq_unmask(interrupt) \
+#define arch_local_irq_unmask(interrupt) \
 	(__get_cpu_var(interrupts_enabled_mask) |= INT_MASK(interrupt))
 
 /* Allow the given interrupt to be enabled immediately, if !irqs_disabled. */
-#define raw_local_irq_unmask_now(interrupt) do { \
-	raw_local_irq_unmask(interrupt); \
+#define arch_local_irq_unmask_now(interrupt) do { \
+	arch_local_irq_unmask(interrupt); \
 	if (!irqs_disabled()) \
 		interrupt_mask_reset(interrupt); \
 } while (0)
diff --git a/arch/x86/include/asm/irqflags.h b/arch/x86/include/asm/irqflags.h
index 9e2b952f810a..5745ce8bf108 100644
--- a/arch/x86/include/asm/irqflags.h
+++ b/arch/x86/include/asm/irqflags.h
@@ -61,22 +61,22 @@ static inline void native_halt(void)
 #else
 #ifndef __ASSEMBLY__
 
-static inline unsigned long __raw_local_save_flags(void)
+static inline unsigned long arch_local_save_flags(void)
 {
 	return native_save_fl();
 }
 
-static inline void raw_local_irq_restore(unsigned long flags)
+static inline void arch_local_irq_restore(unsigned long flags)
 {
 	native_restore_fl(flags);
 }
 
-static inline void raw_local_irq_disable(void)
+static inline void arch_local_irq_disable(void)
 {
 	native_irq_disable();
 }
 
-static inline void raw_local_irq_enable(void)
+static inline void arch_local_irq_enable(void)
 {
 	native_irq_enable();
 }
@@ -85,7 +85,7 @@ static inline void raw_local_irq_enable(void)
  * Used in the idle loop; sti takes one instruction cycle
  * to complete:
  */
-static inline void raw_safe_halt(void)
+static inline void arch_safe_halt(void)
 {
 	native_safe_halt();
 }
@@ -102,12 +102,10 @@ static inline void halt(void)
 /*
  * For spinlocks, etc:
  */
-static inline unsigned long __raw_local_irq_save(void)
+static inline unsigned long arch_local_irq_save(void)
 {
-	unsigned long flags = __raw_local_save_flags();
-
-	raw_local_irq_disable();
-
+	unsigned long flags = arch_local_save_flags();
+	arch_local_irq_disable();
 	return flags;
 }
 #else
@@ -153,22 +151,16 @@ static inline unsigned long __raw_local_irq_save(void)
 #endif /* CONFIG_PARAVIRT */
 
 #ifndef __ASSEMBLY__
-#define raw_local_save_flags(flags)				\
-	do { (flags) = __raw_local_save_flags(); } while (0)
-
-#define raw_local_irq_save(flags)				\
-	do { (flags) = __raw_local_irq_save(); } while (0)
-
-static inline int raw_irqs_disabled_flags(unsigned long flags)
+static inline int arch_irqs_disabled_flags(unsigned long flags)
 {
 	return !(flags & X86_EFLAGS_IF);
 }
 
-static inline int raw_irqs_disabled(void)
+static inline int arch_irqs_disabled(void)
 {
-	unsigned long flags = __raw_local_save_flags();
+	unsigned long flags = arch_local_save_flags();
 
-	return raw_irqs_disabled_flags(flags);
+	return arch_irqs_disabled_flags(flags);
 }
 
 #else
diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index 5653f43d90e5..499954c530da 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -105,7 +105,7 @@ static inline void write_cr8(unsigned long x)
 }
 #endif
 
-static inline void raw_safe_halt(void)
+static inline void arch_safe_halt(void)
 {
 	PVOP_VCALL0(pv_irq_ops.safe_halt);
 }
@@ -829,32 +829,32 @@ static __always_inline void arch_spin_unlock(struct arch_spinlock *lock)
 #define __PV_IS_CALLEE_SAVE(func)			\
 	((struct paravirt_callee_save) { func })
 
-static inline unsigned long __raw_local_save_flags(void)
+static inline unsigned long arch_local_save_flags(void)
 {
 	return PVOP_CALLEE0(unsigned long, pv_irq_ops.save_fl);
 }
 
-static inline void raw_local_irq_restore(unsigned long f)
+static inline void arch_local_irq_restore(unsigned long f)
 {
 	PVOP_VCALLEE1(pv_irq_ops.restore_fl, f);
 }
 
-static inline void raw_local_irq_disable(void)
+static inline void arch_local_irq_disable(void)
 {
 	PVOP_VCALLEE0(pv_irq_ops.irq_disable);
 }
 
-static inline void raw_local_irq_enable(void)
+static inline void arch_local_irq_enable(void)
 {
 	PVOP_VCALLEE0(pv_irq_ops.irq_enable);
 }
 
-static inline unsigned long __raw_local_irq_save(void)
+static inline unsigned long arch_local_irq_save(void)
 {
 	unsigned long f;
 
-	f = __raw_local_save_flags();
-	raw_local_irq_disable();
+	f = arch_local_save_flags();
+	arch_local_irq_disable();
 	return f;
 }
 
diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index e0500646585d..23e061b9327b 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -224,7 +224,7 @@ static noinline int xen_spin_lock_slow(struct arch_spinlock *lock, bool irq_enab
 			goto out;
 		}
 
-		flags = __raw_local_save_flags();
+		flags = arch_local_save_flags();
 		if (irq_enable) {
 			ADD_STATS(taken_slow_irqenable, 1);
 			raw_local_irq_enable();
diff --git a/arch/xtensa/include/asm/irqflags.h b/arch/xtensa/include/asm/irqflags.h
new file mode 100644
index 000000000000..dae9a8bdcb17
--- /dev/null
+++ b/arch/xtensa/include/asm/irqflags.h
@@ -0,0 +1,58 @@
+/*
+ * Xtensa IRQ flags handling functions
+ *
+ * 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.
+ *
+ * Copyright (C) 2001 - 2005 Tensilica Inc.
+ */
+
+#ifndef _XTENSA_IRQFLAGS_H
+#define _XTENSA_IRQFLAGS_H
+
+#include <linux/types.h>
+
+static inline unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile("rsr %0,"__stringify(PS) : "=a" (flags));
+	return flags;
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags;
+	asm volatile("rsil %0, "__stringify(LOCKLEVEL)
+		     : "=a" (flags) :: "memory");
+	return flags;
+}
+
+static inline void arch_local_irq_disable(void)
+{
+	arch_local_irq_save();
+}
+
+static inline void arch_local_irq_enable(void)
+{
+	unsigned long flags;
+	asm volatile("rsil %0, 0" : "=a" (flags) :: "memory");
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile("wsr %0, "__stringify(PS)" ; rsync"
+		     :: "a" (flags) : "memory");
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & 0xf) != 0;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* _XTENSA_IRQFLAGS_H */
diff --git a/arch/xtensa/include/asm/system.h b/arch/xtensa/include/asm/system.h
index 62b1e8f3c13c..1e7e09ab6cd7 100644
--- a/arch/xtensa/include/asm/system.h
+++ b/arch/xtensa/include/asm/system.h
@@ -12,41 +12,10 @@
 #define _XTENSA_SYSTEM_H
 
 #include <linux/stringify.h>
+#include <linux/irqflags.h>
 
 #include <asm/processor.h>
 
-/* interrupt control */
-
-#define local_save_flags(x)						\
-	__asm__ __volatile__ ("rsr %0,"__stringify(PS) : "=a" (x));
-#define local_irq_restore(x)	do {					\
-	__asm__ __volatile__ ("wsr %0, "__stringify(PS)" ; rsync" 	\
-	    		      :: "a" (x) : "memory"); } while(0);
-#define local_irq_save(x)	do {					\
-	__asm__ __volatile__ ("rsil %0, "__stringify(LOCKLEVEL) 	\
-	    		      : "=a" (x) :: "memory");} while(0);
-
-static inline void local_irq_disable(void)
-{
-	unsigned long flags;
-	__asm__ __volatile__ ("rsil %0, "__stringify(LOCKLEVEL)
-	    		      : "=a" (flags) :: "memory");
-}
-static inline void local_irq_enable(void)
-{
-	unsigned long flags;
-	__asm__ __volatile__ ("rsil %0, 0" : "=a" (flags) :: "memory");
-
-}
-
-static inline int irqs_disabled(void)
-{
-	unsigned long flags;
-	local_save_flags(flags);
-	return flags & 0xf;
-}
-
-
 #define smp_read_barrier_depends() do { } while(0)
 #define read_barrier_depends() do { } while(0)
 
diff --git a/drivers/s390/char/sclp.c b/drivers/s390/char/sclp.c
index f6d72e1f2a38..5707a80b96b6 100644
--- a/drivers/s390/char/sclp.c
+++ b/drivers/s390/char/sclp.c
@@ -468,7 +468,7 @@ sclp_sync_wait(void)
 	cr0_sync &= 0xffff00a0;
 	cr0_sync |= 0x00000200;
 	__ctl_load(cr0_sync, 0, 0);
-	__raw_local_irq_stosm(0x01);
+	__arch_local_irq_stosm(0x01);
 	/* Loop until driver state indicates finished request */
 	while (sclp_running_state != sclp_running_state_idle) {
 		/* Check for expired request timer */
diff --git a/include/asm-generic/atomic.h b/include/asm-generic/atomic.h
index e53347fbf1da..fd57b8477fab 100644
--- a/include/asm-generic/atomic.h
+++ b/include/asm-generic/atomic.h
@@ -43,6 +43,7 @@
  */
 #define atomic_set(v, i) (((v)->counter) = (i))
 
+#include <linux/irqflags.h>
 #include <asm/system.h>
 
 /**
@@ -57,7 +58,7 @@ static inline int atomic_add_return(int i, atomic_t *v)
 	unsigned long flags;
 	int temp;
 
-	raw_local_irq_save(flags); /* Don't trace it in a irqsoff handler */
+	raw_local_irq_save(flags); /* Don't trace it in an irqsoff handler */
 	temp = v->counter;
 	temp += i;
 	v->counter = temp;
@@ -78,7 +79,7 @@ static inline int atomic_sub_return(int i, atomic_t *v)
 	unsigned long flags;
 	int temp;
 
-	raw_local_irq_save(flags); /* Don't trace it in a irqsoff handler */
+	raw_local_irq_save(flags); /* Don't trace it in an irqsoff handler */
 	temp = v->counter;
 	temp -= i;
 	v->counter = temp;
diff --git a/include/asm-generic/cmpxchg-local.h b/include/asm-generic/cmpxchg-local.h
index b2ba2fc8829a..2533fddd34a6 100644
--- a/include/asm-generic/cmpxchg-local.h
+++ b/include/asm-generic/cmpxchg-local.h
@@ -2,6 +2,7 @@
 #define __ASM_GENERIC_CMPXCHG_LOCAL_H
 
 #include <linux/types.h>
+#include <linux/irqflags.h>
 
 extern unsigned long wrong_size_cmpxchg(volatile void *ptr);
 
diff --git a/include/asm-generic/hardirq.h b/include/asm-generic/hardirq.h
index 62f59080e5cc..c0771aa248cf 100644
--- a/include/asm-generic/hardirq.h
+++ b/include/asm-generic/hardirq.h
@@ -3,7 +3,6 @@
 
 #include <linux/cache.h>
 #include <linux/threads.h>
-#include <linux/irq.h>
 
 typedef struct {
 	unsigned int __softirq_pending;
diff --git a/include/asm-generic/irqflags.h b/include/asm-generic/irqflags.h
index 9aebf618275a..1f40d0024cf3 100644
--- a/include/asm-generic/irqflags.h
+++ b/include/asm-generic/irqflags.h
@@ -5,68 +5,62 @@
  * All architectures should implement at least the first two functions,
  * usually inline assembly will be the best way.
  */
-#ifndef RAW_IRQ_DISABLED
-#define RAW_IRQ_DISABLED 0
-#define RAW_IRQ_ENABLED 1
+#ifndef ARCH_IRQ_DISABLED
+#define ARCH_IRQ_DISABLED 0
+#define ARCH_IRQ_ENABLED 1
 #endif
 
 /* read interrupt enabled status */
-#ifndef __raw_local_save_flags
-unsigned long __raw_local_save_flags(void);
+#ifndef arch_local_save_flags
+unsigned long arch_local_save_flags(void);
 #endif
 
 /* set interrupt enabled status */
-#ifndef raw_local_irq_restore
-void raw_local_irq_restore(unsigned long flags);
+#ifndef arch_local_irq_restore
+void arch_local_irq_restore(unsigned long flags);
 #endif
 
 /* get status and disable interrupts */
-#ifndef __raw_local_irq_save
-static inline unsigned long __raw_local_irq_save(void)
+#ifndef arch_local_irq_save
+static inline unsigned long arch_local_irq_save(void)
 {
 	unsigned long flags;
-	flags = __raw_local_save_flags();
-	raw_local_irq_restore(RAW_IRQ_DISABLED);
+	flags = arch_local_save_flags();
+	arch_local_irq_restore(ARCH_IRQ_DISABLED);
 	return flags;
 }
 #endif
 
 /* test flags */
-#ifndef raw_irqs_disabled_flags
-static inline int raw_irqs_disabled_flags(unsigned long flags)
+#ifndef arch_irqs_disabled_flags
+static inline int arch_irqs_disabled_flags(unsigned long flags)
 {
-	return flags == RAW_IRQ_DISABLED;
+	return flags == ARCH_IRQ_DISABLED;
 }
 #endif
 
 /* unconditionally enable interrupts */
-#ifndef raw_local_irq_enable
-static inline void raw_local_irq_enable(void)
+#ifndef arch_local_irq_enable
+static inline void arch_local_irq_enable(void)
 {
-	raw_local_irq_restore(RAW_IRQ_ENABLED);
+	arch_local_irq_restore(ARCH_IRQ_ENABLED);
 }
 #endif
 
 /* unconditionally disable interrupts */
-#ifndef raw_local_irq_disable
-static inline void raw_local_irq_disable(void)
+#ifndef arch_local_irq_disable
+static inline void arch_local_irq_disable(void)
 {
-	raw_local_irq_restore(RAW_IRQ_DISABLED);
+	arch_local_irq_restore(ARCH_IRQ_DISABLED);
 }
 #endif
 
 /* test hardware interrupt enable bit */
-#ifndef raw_irqs_disabled
-static inline int raw_irqs_disabled(void)
+#ifndef arch_irqs_disabled
+static inline int arch_irqs_disabled(void)
 {
-	return raw_irqs_disabled_flags(__raw_local_save_flags());
+	return arch_irqs_disabled_flags(arch_local_save_flags());
 }
 #endif
 
-#define raw_local_save_flags(flags) \
-	do { (flags) = __raw_local_save_flags(); } while (0)
-
-#define raw_local_irq_save(flags) \
-	do { (flags) = __raw_local_irq_save(); } while (0)
-
 #endif /* __ASM_GENERIC_IRQFLAGS_H */
diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
index 006bf45eae30..d176d658fe25 100644
--- a/include/linux/irqflags.h
+++ b/include/linux/irqflags.h
@@ -12,6 +12,7 @@
 #define _LINUX_TRACE_IRQFLAGS_H
 
 #include <linux/typecheck.h>
+#include <asm/irqflags.h>
 
 #ifdef CONFIG_TRACE_IRQFLAGS
   extern void trace_softirqs_on(unsigned long ip);
@@ -52,17 +53,45 @@
 # define start_critical_timings() do { } while (0)
 #endif
 
-#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
-
-#include <asm/irqflags.h>
+/*
+ * Wrap the arch provided IRQ routines to provide appropriate checks.
+ */
+#define raw_local_irq_disable()		arch_local_irq_disable()
+#define raw_local_irq_enable()		arch_local_irq_enable()
+#define raw_local_irq_save(flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		flags = arch_local_irq_save();		\
+	} while (0)
+#define raw_local_irq_restore(flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		arch_local_irq_restore(flags);		\
+	} while (0)
+#define raw_local_save_flags(flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		flags = arch_local_save_flags();	\
+	} while (0)
+#define raw_irqs_disabled_flags(flags)			\
+	({						\
+		typecheck(unsigned long, flags);	\
+		arch_irqs_disabled_flags(flags);	\
+	})
+#define raw_irqs_disabled()		(arch_irqs_disabled())
+#define raw_safe_halt()			arch_safe_halt()
 
+/*
+ * The local_irq_*() APIs are equal to the raw_local_irq*()
+ * if !TRACE_IRQFLAGS.
+ */
+#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
 #define local_irq_enable() \
 	do { trace_hardirqs_on(); raw_local_irq_enable(); } while (0)
 #define local_irq_disable() \
 	do { raw_local_irq_disable(); trace_hardirqs_off(); } while (0)
 #define local_irq_save(flags)				\
 	do {						\
-		typecheck(unsigned long, flags);	\
 		raw_local_irq_save(flags);		\
 		trace_hardirqs_off();			\
 	} while (0)
@@ -70,7 +99,6 @@
 
 #define local_irq_restore(flags)			\
 	do {						\
-		typecheck(unsigned long, flags);	\
 		if (raw_irqs_disabled_flags(flags)) {	\
 			raw_local_irq_restore(flags);	\
 			trace_hardirqs_off();		\
@@ -79,51 +107,44 @@
 			raw_local_irq_restore(flags);	\
 		}					\
 	} while (0)
-#else /* !CONFIG_TRACE_IRQFLAGS_SUPPORT */
-/*
- * The local_irq_*() APIs are equal to the raw_local_irq*()
- * if !TRACE_IRQFLAGS.
- */
-# define raw_local_irq_disable()	local_irq_disable()
-# define raw_local_irq_enable()		local_irq_enable()
-# define raw_local_irq_save(flags)			\
-	do {						\
-		typecheck(unsigned long, flags);	\
-		local_irq_save(flags);			\
-	} while (0)
-# define raw_local_irq_restore(flags)			\
+#define local_save_flags(flags)				\
 	do {						\
-		typecheck(unsigned long, flags);	\
-		local_irq_restore(flags);		\
+		raw_local_save_flags(flags);		\
 	} while (0)
-#endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */
 
-#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
-#define safe_halt()						\
-	do {							\
-		trace_hardirqs_on();				\
-		raw_safe_halt();				\
-	} while (0)
+#define irqs_disabled_flags(flags)			\
+	({						\
+		raw_irqs_disabled_flags(flags);		\
+	})
 
-#define local_save_flags(flags)				\
-	do {						\
-		typecheck(unsigned long, flags);	\
-		raw_local_save_flags(flags);		\
+#define irqs_disabled()					\
+	({						\
+		unsigned long _flags;			\
+		raw_local_save_flags(_flags);		\
+		raw_irqs_disabled_flags(_flags);	\
+	})
+
+#define safe_halt()				\
+	do {					\
+		trace_hardirqs_on();		\
+		raw_safe_halt();		\
 	} while (0)
 
-#define irqs_disabled()						\
-({								\
-	unsigned long _flags;					\
-								\
-	raw_local_save_flags(_flags);				\
-	raw_irqs_disabled_flags(_flags);			\
-})
 
-#define irqs_disabled_flags(flags)		\
-({						\
-	typecheck(unsigned long, flags);	\
-	raw_irqs_disabled_flags(flags);		\
-})
+#else /* !CONFIG_TRACE_IRQFLAGS_SUPPORT */
+
+#define local_irq_enable()	do { raw_local_irq_enable(); } while (0)
+#define local_irq_disable()	do { raw_local_irq_disable(); } while (0)
+#define local_irq_save(flags)					\
+	do {							\
+		raw_local_irq_save(flags);			\
+	} while (0)
+#define local_irq_restore(flags) do { raw_local_irq_restore(flags); } while (0)
+#define local_save_flags(flags)	do { raw_local_save_flags(flags); } while (0)
+#define irqs_disabled()		(raw_irqs_disabled())
+#define irqs_disabled_flags(flags) (raw_irqs_disabled_flags(flags))
+#define safe_halt()		do { raw_safe_halt(); } while (0)
+
 #endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */
 
 #endif
diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index f8854655860e..80e535897de6 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -50,6 +50,7 @@
 #include <linux/preempt.h>
 #include <linux/linkage.h>
 #include <linux/compiler.h>
+#include <linux/irqflags.h>
 #include <linux/thread_info.h>
 #include <linux/kernel.h>
 #include <linux/stringify.h>
-- 
cgit v1.2.3


From 01723a9566f9e9ce4c75e5c4c9f6dc20600871a7 Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@stericsson.com>
Date: Tue, 7 Sep 2010 22:43:19 +0100
Subject: ARM: 6368/1: move the PrimeCell IDs to use macros

This make four macros for the PrimeCell ID register available to
drivers that use them witout using the PrimeCell/AMBA bus
abstraction and struct amba_device. It also moves the magic
PrimeCell CID "B105F00D" to the bus.h header file.

Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/amba/bus.c       |  2 +-
 include/linux/amba/bus.h | 15 +++++++++++----
 2 files changed, 12 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/amba/bus.c b/drivers/amba/bus.c
index d31590e7011b..2737b9752205 100644
--- a/drivers/amba/bus.c
+++ b/drivers/amba/bus.c
@@ -298,7 +298,7 @@ int amba_device_register(struct amba_device *dev, struct resource *parent)
 
 		amba_put_disable_pclk(dev);
 
-		if (cid == 0xb105f00d)
+		if (cid == AMBA_CID)
 			dev->periphid = pid;
 
 		if (!dev->periphid)
diff --git a/include/linux/amba/bus.h b/include/linux/amba/bus.h
index b0c174012436..c6454cca0447 100644
--- a/include/linux/amba/bus.h
+++ b/include/linux/amba/bus.h
@@ -20,6 +20,7 @@
 #include <linux/resource.h>
 
 #define AMBA_NR_IRQS	2
+#define AMBA_CID	0xb105f00d
 
 struct clk;
 
@@ -70,9 +71,15 @@ void amba_release_regions(struct amba_device *);
 #define amba_pclk_disable(d)	\
 	do { if (!IS_ERR((d)->pclk)) clk_disable((d)->pclk); } while (0)
 
-#define amba_config(d)	(((d)->periphid >> 24) & 0xff)
-#define amba_rev(d)	(((d)->periphid >> 20) & 0x0f)
-#define amba_manf(d)	(((d)->periphid >> 12) & 0xff)
-#define amba_part(d)	((d)->periphid & 0xfff)
+/* Some drivers don't use the struct amba_device */
+#define AMBA_CONFIG_BITS(a) (((a) >> 24) & 0xff)
+#define AMBA_REV_BITS(a) (((a) >> 20) & 0x0f)
+#define AMBA_MANF_BITS(a) (((a) >> 12) & 0xff)
+#define AMBA_PART_BITS(a) ((a) & 0xfff)
+
+#define amba_config(d)	AMBA_CONFIG_BITS((d)->periphid)
+#define amba_rev(d)	AMBA_REV_BITS((d)->periphid)
+#define amba_manf(d)	AMBA_MANF_BITS((d)->periphid)
+#define amba_part(d)	AMBA_PART_BITS((d)->periphid)
 
 #endif
-- 
cgit v1.2.3


From 03789f26722a15ccfe6f191e9fb3d356f2f18a1e Mon Sep 17 00:00:00 2001
From: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Date: Fri, 8 Oct 2010 04:02:02 +0000
Subject: Phonet: cleanup pipe enable socket option
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The current code works like this:

  int garbage, status;
  socklen_t len = sizeof(status);

  /* enable pipe */
  setsockopt(fd, SOL_PNPIPE, PNPIPE_ENABLE, &garbage, sizeof(garbage));
  /* disable pipe */
  setsockopt(fd, SOL_PNPIPE, PNPIPE_DISABLE, &garbage, sizeof(garbage));
  /* get status */
  getsockopt(fd, SOL_PNPIPE, PNPIPE_INQ, &status, &len);

...which does not follow the usual socket option pattern. This patch
merges all three "options" into a single gettable&settable option,
before Linux 2.6.37 gets out:

  int status;
  socklen_t len = sizeof(status);

  /* enable pipe */
  status = 1;
  setsockopt(fd, SOL_PNPIPE, PNPIPE_ENABLE, &status, sizeof(status));
  /* disable pipe */
  status = 0;
  setsockopt(fd, SOL_PNPIPE, PNPIPE_ENABLE, &status, sizeof(status));
  /* get status */
  getsockopt(fd, SOL_PNPIPE, PNPIPE_ENABLE, &status, &len);

This also fixes the error code from EFAULT to ENOTCONN.

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Cc: Kumar Sanghvi <kumar.sanghvi@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/phonet.txt | 15 ++------
 include/linux/phonet.h              |  3 +-
 net/phonet/pep.c                    | 72 ++++++++++++++++---------------------
 3 files changed, 34 insertions(+), 56 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/networking/phonet.txt b/Documentation/networking/phonet.txt
index cccf5ff07ec2..2d9bc2b711fc 100644
--- a/Documentation/networking/phonet.txt
+++ b/Documentation/networking/phonet.txt
@@ -213,12 +213,9 @@ The implementation adds socket options at SOL_PNPIPE level:
 	It then updates the pipe state associated with the sequenced socket to
 	be PIPE_DISABLED.
 
-  PNPIPE_ENABLE
-	It follows the same sequence as above for enabling a pipe by sending
-	PNS_PEP_ENABLE_REQ initially and then sending PNS_PEP_ENABLED_IND after
-	getting responses from sequenced socket and remote-pep.
-	It will also update the pipe state associated with the sequenced socket
-	to PIPE_ENABLED.
+  PNPIPE_ENABLE accepts one integer value (int). If set to zero, the pipe
+    is disabled. If the value is non-zero, the pipe is enabled. If the pipe
+    is not (yet) connected, ENOTCONN is error is returned.
 
    PNPIPE_DESTROY
 	This will send out PNS_PEP_DISCONNECT_REQ on the sequenced socket and
@@ -226,12 +223,6 @@ The implementation adds socket options at SOL_PNPIPE level:
 	It will also update the pipe state associated with the sequenced socket
 	to PIPE_IDLE
 
-   PNPIPE_INQ
-	This getsocktopt allows the user-space running on the sequenced socket
-	to examine the pipe state associated with that socket ie. whether the
-	pipe is created (PIPE_DISABLED) or enabled (PIPE_ENABLED) or disabled
-	(PIPE_DISABLED) or no pipe exists (PIPE_IDLE).
-
 After a pipe has been created and enabled successfully, the Pipe data can be
 exchanged between the host-pep and remote-pep (modem).
 
diff --git a/include/linux/phonet.h b/include/linux/phonet.h
index 96f5625d62fa..e27cbf931740 100644
--- a/include/linux/phonet.h
+++ b/include/linux/phonet.h
@@ -38,9 +38,8 @@
 #define PNPIPE_IFINDEX		2
 #define PNPIPE_CREATE           3
 #define PNPIPE_ENABLE           4
-#define PNPIPE_DISABLE          5
+/* unused slot */
 #define PNPIPE_DESTROY          6
-#define PNPIPE_INQ              7
 
 #define PNADDR_ANY		0
 #define PNADDR_BROADCAST	0xFC
diff --git a/net/phonet/pep.c b/net/phonet/pep.c
index aa3d8700d213..f818f76d297d 100644
--- a/net/phonet/pep.c
+++ b/net/phonet/pep.c
@@ -327,29 +327,20 @@ static int pipe_handler_send_ind(struct sock *sk, u16 dobj, u8 utid,
 	return pn_skb_send(sk, skb, &spn);
 }
 
-static int pipe_handler_enable_pipe(struct sock *sk, int cmd)
+static int pipe_handler_enable_pipe(struct sock *sk, int enable)
 {
-	int ret;
 	struct pep_sock *pn = pep_sk(sk);
-
-	switch (cmd) {
-	case PNPIPE_ENABLE:
-		ret = pipe_handler_send_req(sk, pn->pn_sk.sobject,
-				PNS_PIPE_ENABLE_UTID, PNS_PEP_ENABLE_REQ,
-				pn->pipe_handle, GFP_ATOMIC);
-		break;
-
-	case PNPIPE_DISABLE:
-		ret = pipe_handler_send_req(sk, pn->pn_sk.sobject,
-				PNS_PIPE_DISABLE_UTID, PNS_PEP_DISABLE_REQ,
-				pn->pipe_handle, GFP_ATOMIC);
-		break;
-
-	default:
-		ret = -EINVAL;
+	int utid, req;
+
+	if (enable) {
+		utid = PNS_PIPE_ENABLE_UTID;
+		req = PNS_PEP_ENABLE_REQ;
+	} else {
+		utid = PNS_PIPE_DISABLE_UTID;
+		req = PNS_PEP_DISABLE_REQ;
 	}
-
-	return ret;
+	return pipe_handler_send_req(sk, pn->pn_sk.sobject, utid, req,
+			pn->pipe_handle, GFP_ATOMIC);
 }
 
 static int pipe_handler_create_pipe(struct sock *sk, int pipe_handle, int cmd)
@@ -1187,23 +1178,6 @@ static int pep_setsockopt(struct sock *sk, int level, int optname,
 			break;
 		}
 
-	case PNPIPE_ENABLE:
-		if (pn->pipe_state != PIPE_DISABLED) {
-			err = -EFAULT;
-			break;
-		}
-		err = pipe_handler_enable_pipe(sk, PNPIPE_ENABLE);
-		break;
-
-	case PNPIPE_DISABLE:
-		if (pn->pipe_state != PIPE_ENABLED) {
-			err = -EFAULT;
-			break;
-		}
-
-		err = pipe_handler_enable_pipe(sk, PNPIPE_DISABLE);
-		break;
-
 	case PNPIPE_DESTROY:
 		if (pn->pipe_state < PIPE_DISABLED) {
 			err = -EFAULT;
@@ -1239,6 +1213,17 @@ static int pep_setsockopt(struct sock *sk, int level, int optname,
 			err = 0;
 		}
 		goto out_norel;
+
+#ifdef CONFIG_PHONET_PIPECTRLR
+	case PNPIPE_ENABLE:
+		if (pn->pipe_state <= PIPE_IDLE) {
+			err = -ENOTCONN;
+			break;
+		}
+		err = pipe_handler_enable_pipe(sk, val);
+		break;
+#endif
+
 	default:
 		err = -ENOPROTOOPT;
 	}
@@ -1264,15 +1249,18 @@ static int pep_getsockopt(struct sock *sk, int level, int optname,
 		val = pn->ifindex ? PNPIPE_ENCAP_IP : PNPIPE_ENCAP_NONE;
 		break;
 
+	case PNPIPE_IFINDEX:
+		val = pn->ifindex;
+		break;
+
 #ifdef CONFIG_PHONET_PIPECTRLR
-	case PNPIPE_INQ:
-		val = pn->pipe_state;
+	case PNPIPE_ENABLE:
+		if (pn->pipe_state <= PIPE_IDLE)
+			return -ENOTCONN;
+		val = pn->pipe_state != PIPE_DISABLED;
 		break;
 #endif
 
-	case PNPIPE_IFINDEX:
-		val = pn->ifindex;
-		break;
 	default:
 		return -ENOPROTOOPT;
 	}
-- 
cgit v1.2.3


From 3e51d3c924aea8a1f1372e6c615b0a37b528121d Mon Sep 17 00:00:00 2001
From: Kai Makisara <Kai.Makisara@kolumbus.fi>
Date: Sat, 9 Oct 2010 00:17:56 +0300
Subject: [SCSI] st: add MTWEOFI to write filemarks without flushing drive
 buffer

This patch adds a new MTIOCTOP operation MTWEOFI that writes filemarks with
immediate bit set. This means that the drive does not flush its buffer and the
next file can be started immediately. This speeds up writing in applications
that have to write multiple small files.

Signed-off-by: Kai Makisara <kai.makisara@kolumbus.fi>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
---
 Documentation/scsi/st.txt | 15 ++++++++++++++-
 drivers/scsi/st.c         | 15 +++++++++------
 include/linux/mtio.h      |  1 +
 3 files changed, 24 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/scsi/st.txt b/Documentation/scsi/st.txt
index 40752602c050..691ca292c24d 100644
--- a/Documentation/scsi/st.txt
+++ b/Documentation/scsi/st.txt
@@ -2,7 +2,7 @@ This file contains brief information about the SCSI tape driver.
 The driver is currently maintained by Kai Mäkisara (email
 Kai.Makisara@kolumbus.fi)
 
-Last modified: Sun Feb 24 21:59:07 2008 by kai.makisara
+Last modified: Sun Aug 29 18:25:47 2010 by kai.makisara
 
 
 BASICS
@@ -85,6 +85,17 @@ writing and the last operation has been a write. Two filemarks can be
 optionally written. In both cases end of data is signified by
 returning zero bytes for two consecutive reads.
 
+Writing filemarks without the immediate bit set in the SCSI command block acts
+as a synchronization point, i.e., all remaining data form the drive buffers is
+written to tape before the command returns. This makes sure that write errors
+are caught at that point, but this takes time. In some applications, several
+consecutive files must be written fast. The MTWEOFI operation can be used to
+write the filemarks without flushing the drive buffer. Writing filemark at
+close() is always flushing the drive buffers. However, if the previous
+operation is MTWEOFI, close() does not write a filemark. This can be used if
+the program wants to close/open the tape device between files and wants to
+skip waiting.
+
 If rewind, offline, bsf, or seek is done and previous tape operation was
 write, a filemark is written before moving tape.
 
@@ -301,6 +312,8 @@ MTBSR   Space backward over count records.
 MTFSS   Space forward over count setmarks.
 MTBSS   Space backward over count setmarks.
 MTWEOF  Write count filemarks.
+MTWEOFI	Write count filemarks with immediate bit set (i.e., does not
+	wait until data is on tape)
 MTWSM   Write count setmarks.
 MTREW   Rewind tape.
 MTOFFL  Set device off line (often rewind plus eject).
diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c
index 24211d0efa6d..9e2c3a72ff4d 100644
--- a/drivers/scsi/st.c
+++ b/drivers/scsi/st.c
@@ -9,7 +9,7 @@
    Steve Hirsch, Andreas Koppenh"ofer, Michael Leodolter, Eyal Lebedinsky,
    Michael Schaefer, J"org Weule, and Eric Youngdale.
 
-   Copyright 1992 - 2008 Kai Makisara
+   Copyright 1992 - 2010 Kai Makisara
    email Kai.Makisara@kolumbus.fi
 
    Some small formal changes - aeb, 950809
@@ -17,7 +17,7 @@
    Last modified: 18-JAN-1998 Richard Gooch <rgooch@atnf.csiro.au> Devfs support
  */
 
-static const char *verstr = "20081215";
+static const char *verstr = "20100829";
 
 #include <linux/module.h>
 
@@ -2696,18 +2696,21 @@ static int st_int_ioctl(struct scsi_tape *STp, unsigned int cmd_in, unsigned lon
 		}
 		break;
 	case MTWEOF:
+	case MTWEOFI:
 	case MTWSM:
 		if (STp->write_prot)
 			return (-EACCES);
 		cmd[0] = WRITE_FILEMARKS;
 		if (cmd_in == MTWSM)
 			cmd[1] = 2;
+		if (cmd_in == MTWEOFI)
+			cmd[1] |= 1;
 		cmd[2] = (arg >> 16);
 		cmd[3] = (arg >> 8);
 		cmd[4] = arg;
 		timeout = STp->device->request_queue->rq_timeout;
                 DEBC(
-                     if (cmd_in == MTWEOF)
+		     if (cmd_in != MTWSM)
                                printk(ST_DEB_MSG "%s: Writing %d filemarks.\n", name,
 				 cmd[2] * 65536 + cmd[3] * 256 + cmd[4]);
                      else
@@ -2883,8 +2886,8 @@ static int st_int_ioctl(struct scsi_tape *STp, unsigned int cmd_in, unsigned lon
 		else if (chg_eof)
 			STps->eof = ST_NOEOF;
 
-		if (cmd_in == MTWEOF)
-			STps->rw = ST_IDLE;
+		if (cmd_in == MTWEOF || cmd_in == MTWEOFI)
+			STps->rw = ST_IDLE;  /* prevent automatic WEOF at close */
 	} else { /* SCSI command was not completely successful. Don't return
                     from this block without releasing the SCSI command block! */
 		struct st_cmdstatus *cmdstatp = &STp->buffer->cmdstat;
@@ -2901,7 +2904,7 @@ static int st_int_ioctl(struct scsi_tape *STp, unsigned int cmd_in, unsigned lon
 		else
 			undone = 0;
 
-		if (cmd_in == MTWEOF &&
+		if ((cmd_in == MTWEOF || cmd_in == MTWEOFI) &&
 		    cmdstatp->have_sense &&
 		    (cmdstatp->flags & SENSE_EOM)) {
 			if (cmdstatp->sense_hdr.sense_key == NO_SENSE ||
diff --git a/include/linux/mtio.h b/include/linux/mtio.h
index ef01d6aa5934..8f825756c459 100644
--- a/include/linux/mtio.h
+++ b/include/linux/mtio.h
@@ -63,6 +63,7 @@ struct	mtop {
 #define MTCOMPRESSION 32/* control compression with SCSI mode page 15 */
 #define MTSETPART 33	/* Change the active tape partition */
 #define MTMKPART  34	/* Format the tape with one or two partitions */
+#define MTWEOFI	35	/* write an end-of-file record (mark) in immediate mode */
 
 /* structure for MTIOCGET - mag tape get status command */
 
-- 
cgit v1.2.3


From 708ff2a0097b02d32d375b66996661f36cd4d6d1 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Wed, 29 Sep 2010 18:08:50 +0900
Subject: bitops: make asm-generic/bitops/find.h more generic

asm-generic/bitops/find.h has the extern declarations of find_next_bit()
and find_next_zero_bit() and the macro definitions of find_first_bit()
and find_first_zero_bit(). It is only usable by the architectures which
enables CONFIG_GENERIC_FIND_NEXT_BIT and disables
CONFIG_GENERIC_FIND_FIRST_BIT.

x86 and tile enable both CONFIG_GENERIC_FIND_NEXT_BIT and
CONFIG_GENERIC_FIND_FIRST_BIT. These architectures cannot include
asm-generic/bitops/find.h in their asm/bitops.h. So ifdefed extern
declarations of find_first_bit and find_first_zero_bit() are put in
linux/bitops.h.

This makes asm-generic/bitops/find.h usable by these architectures
and use it. Also this change is needed for the forthcoming duplicated
extern declarations cleanup.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: Chris Metcalf <cmetcalf@tilera.com>
---
 arch/tile/include/asm/bitops.h    |  1 +
 arch/x86/include/asm/bitops.h     |  2 ++
 include/asm-generic/bitops/find.h | 25 +++++++++++++++++++++++++
 include/linux/bitops.h            | 22 ----------------------
 4 files changed, 28 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/arch/tile/include/asm/bitops.h b/arch/tile/include/asm/bitops.h
index 6832b4be8990..6d4f0ff2c68c 100644
--- a/arch/tile/include/asm/bitops.h
+++ b/arch/tile/include/asm/bitops.h
@@ -120,6 +120,7 @@ static inline unsigned long __arch_hweight64(__u64 w)
 
 #include <asm-generic/bitops/const_hweight.h>
 #include <asm-generic/bitops/lock.h>
+#include <asm-generic/bitops/find.h>
 #include <asm-generic/bitops/sched.h>
 #include <asm-generic/bitops/ext2-non-atomic.h>
 #include <asm-generic/bitops/minix.h>
diff --git a/arch/x86/include/asm/bitops.h b/arch/x86/include/asm/bitops.h
index bafd80defa43..903683b07e42 100644
--- a/arch/x86/include/asm/bitops.h
+++ b/arch/x86/include/asm/bitops.h
@@ -440,6 +440,8 @@ static inline int fls(int x)
 
 #ifdef __KERNEL__
 
+#include <asm-generic/bitops/find.h>
+
 #include <asm-generic/bitops/sched.h>
 
 #define ARCH_HAS_FAST_MULTIPLIER 1
diff --git a/include/asm-generic/bitops/find.h b/include/asm-generic/bitops/find.h
index 1914e9742512..30afec0db7d7 100644
--- a/include/asm-generic/bitops/find.h
+++ b/include/asm-generic/bitops/find.h
@@ -9,7 +9,32 @@ extern unsigned long find_next_zero_bit(const unsigned long *addr, unsigned
 		long size, unsigned long offset);
 #endif
 
+#ifdef CONFIG_GENERIC_FIND_FIRST_BIT
+
+/**
+ * find_first_bit - find the first set bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum size to search
+ *
+ * Returns the bit number of the first set bit.
+ */
+extern unsigned long find_first_bit(const unsigned long *addr,
+				    unsigned long size);
+
+/**
+ * find_first_zero_bit - find the first cleared bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum size to search
+ *
+ * Returns the bit number of the first cleared bit.
+ */
+extern unsigned long find_first_zero_bit(const unsigned long *addr,
+					 unsigned long size);
+#else /* CONFIG_GENERIC_FIND_FIRST_BIT */
+
 #define find_first_bit(addr, size) find_next_bit((addr), (size), 0)
 #define find_first_zero_bit(addr, size) find_next_zero_bit((addr), (size), 0)
 
+#endif /* CONFIG_GENERIC_FIND_FIRST_BIT */
+
 #endif /*_ASM_GENERIC_BITOPS_FIND_H_ */
diff --git a/include/linux/bitops.h b/include/linux/bitops.h
index fc68053378ce..adb0f113f571 100644
--- a/include/linux/bitops.h
+++ b/include/linux/bitops.h
@@ -136,28 +136,6 @@ static inline unsigned long __ffs64(u64 word)
 }
 
 #ifdef __KERNEL__
-#ifdef CONFIG_GENERIC_FIND_FIRST_BIT
-
-/**
- * find_first_bit - find the first set bit in a memory region
- * @addr: The address to start the search at
- * @size: The maximum size to search
- *
- * Returns the bit number of the first set bit.
- */
-extern unsigned long find_first_bit(const unsigned long *addr,
-				    unsigned long size);
-
-/**
- * find_first_zero_bit - find the first cleared bit in a memory region
- * @addr: The address to start the search at
- * @size: The maximum size to search
- *
- * Returns the bit number of the first cleared bit.
- */
-extern unsigned long find_first_zero_bit(const unsigned long *addr,
-					 unsigned long size);
-#endif /* CONFIG_GENERIC_FIND_FIRST_BIT */
 
 #ifdef CONFIG_GENERIC_FIND_LAST_BIT
 /**
-- 
cgit v1.2.3


From d852a6afd91fc928128f59ebff381838c365e358 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Wed, 29 Sep 2010 18:08:51 +0900
Subject: bitops: remove duplicated extern declarations

If CONFIG_GENERIC_FIND_NEXT_BIT is enabled, find_next_bit() and
find_next_zero_bit() are doubly declared in asm-generic/bitops/find.h
and linux/bitops.h.

asm/bitops.h includes asm-generic/bitops/find.h if and only if the
architecture enables CONFIG_GENERIC_FIND_NEXT_BIT. And asm/bitops.h
is included by linux/bitops.h

So we can just remove the extern declarations of find_next_bit() and
find_next_zero_bit() in linux/bitops.h.

Also we can remove unneeded #ifndef CONFIG_GENERIC_FIND_NEXT_BIT in
asm-generic/bitops/find.h.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 include/asm-generic/bitops/find.h | 14 ++++++++++++--
 include/linux/bitops.h            | 23 -----------------------
 2 files changed, 12 insertions(+), 25 deletions(-)

(limited to 'include/linux')

diff --git a/include/asm-generic/bitops/find.h b/include/asm-generic/bitops/find.h
index 30afec0db7d7..110fa700f853 100644
--- a/include/asm-generic/bitops/find.h
+++ b/include/asm-generic/bitops/find.h
@@ -1,13 +1,23 @@
 #ifndef _ASM_GENERIC_BITOPS_FIND_H_
 #define _ASM_GENERIC_BITOPS_FIND_H_
 
-#ifndef CONFIG_GENERIC_FIND_NEXT_BIT
+/**
+ * find_next_bit - find the next set bit in a memory region
+ * @addr: The address to base the search on
+ * @offset: The bitnumber to start searching at
+ * @size: The bitmap size in bits
+ */
 extern unsigned long find_next_bit(const unsigned long *addr, unsigned long
 		size, unsigned long offset);
 
+/**
+ * find_next_zero_bit - find the next cleared bit in a memory region
+ * @addr: The address to base the search on
+ * @offset: The bitnumber to start searching at
+ * @size: The bitmap size in bits
+ */
 extern unsigned long find_next_zero_bit(const unsigned long *addr, unsigned
 		long size, unsigned long offset);
-#endif
 
 #ifdef CONFIG_GENERIC_FIND_FIRST_BIT
 
diff --git a/include/linux/bitops.h b/include/linux/bitops.h
index adb0f113f571..827cc95711ef 100644
--- a/include/linux/bitops.h
+++ b/include/linux/bitops.h
@@ -149,28 +149,5 @@ extern unsigned long find_last_bit(const unsigned long *addr,
 				   unsigned long size);
 #endif /* CONFIG_GENERIC_FIND_LAST_BIT */
 
-#ifdef CONFIG_GENERIC_FIND_NEXT_BIT
-
-/**
- * find_next_bit - find the next set bit in a memory region
- * @addr: The address to base the search on
- * @offset: The bitnumber to start searching at
- * @size: The bitmap size in bits
- */
-extern unsigned long find_next_bit(const unsigned long *addr,
-				   unsigned long size, unsigned long offset);
-
-/**
- * find_next_zero_bit - find the next cleared bit in a memory region
- * @addr: The address to base the search on
- * @offset: The bitnumber to start searching at
- * @size: The bitmap size in bits
- */
-
-extern unsigned long find_next_zero_bit(const unsigned long *addr,
-					unsigned long size,
-					unsigned long offset);
-
-#endif /* CONFIG_GENERIC_FIND_NEXT_BIT */
 #endif /* __KERNEL__ */
 #endif
-- 
cgit v1.2.3


From 3bf101ba42a1c89b5afbc7492e7647dae5e18735 Mon Sep 17 00:00:00 2001
From: Matt Fleming <matt@console-pimps.org>
Date: Mon, 27 Sep 2010 20:22:24 +0100
Subject: perf: Add helper function to return number of counters

The number of counters for the registered pmu is needed in a few places
so provide a helper function that returns this number.

Signed-off-by: Matt Fleming <matt@console-pimps.org>
Tested-by: Will Deacon <will.deacon@arm.com>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/arm/kernel/perf_event.c |  6 ++++++
 arch/arm/oprofile/common.c   | 31 ++++++++++++++++++-------------
 arch/sh/kernel/perf_event.c  |  9 +++++++++
 include/linux/perf_event.h   |  1 +
 4 files changed, 34 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 417c392ddf1c..3b0aedfb96e7 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -123,6 +123,12 @@ armpmu_get_max_events(void)
 }
 EXPORT_SYMBOL_GPL(armpmu_get_max_events);
 
+int perf_num_counters(void)
+{
+	return armpmu_get_max_events();
+}
+EXPORT_SYMBOL_GPL(perf_num_counters);
+
 #define HW_OP_UNSUPPORTED		0xFFFF
 
 #define C(_x) \
diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c
index d660cb8dab36..1e971a7fcf82 100644
--- a/arch/arm/oprofile/common.c
+++ b/arch/arm/oprofile/common.c
@@ -43,7 +43,7 @@ static DEFINE_MUTEX(op_arm_mutex);
 
 static struct op_counter_config *counter_config;
 static struct perf_event **perf_events[nr_cpumask_bits];
-static int perf_num_counters;
+static int num_counters;
 
 /*
  * Overflow callback for oprofile.
@@ -54,11 +54,11 @@ static void op_overflow_handler(struct perf_event *event, int unused,
 	int id;
 	u32 cpu = smp_processor_id();
 
-	for (id = 0; id < perf_num_counters; ++id)
+	for (id = 0; id < num_counters; ++id)
 		if (perf_events[cpu][id] == event)
 			break;
 
-	if (id != perf_num_counters)
+	if (id != num_counters)
 		oprofile_add_sample(regs, id);
 	else
 		pr_warning("oprofile: ignoring spurious overflow "
@@ -76,7 +76,7 @@ static void op_perf_setup(void)
 	u32 size = sizeof(struct perf_event_attr);
 	struct perf_event_attr *attr;
 
-	for (i = 0; i < perf_num_counters; ++i) {
+	for (i = 0; i < num_counters; ++i) {
 		attr = &counter_config[i].attr;
 		memset(attr, 0, size);
 		attr->type		= PERF_TYPE_RAW;
@@ -131,7 +131,7 @@ static int op_perf_start(void)
 	int cpu, event, ret = 0;
 
 	for_each_online_cpu(cpu) {
-		for (event = 0; event < perf_num_counters; ++event) {
+		for (event = 0; event < num_counters; ++event) {
 			ret = op_create_counter(cpu, event);
 			if (ret)
 				goto out;
@@ -150,7 +150,7 @@ static void op_perf_stop(void)
 	int cpu, event;
 
 	for_each_online_cpu(cpu)
-		for (event = 0; event < perf_num_counters; ++event)
+		for (event = 0; event < num_counters; ++event)
 			op_destroy_counter(cpu, event);
 }
 
@@ -179,7 +179,7 @@ static int op_arm_create_files(struct super_block *sb, struct dentry *root)
 {
 	unsigned int i;
 
-	for (i = 0; i < perf_num_counters; i++) {
+	for (i = 0; i < num_counters; i++) {
 		struct dentry *dir;
 		char buf[4];
 
@@ -353,14 +353,19 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
 
 	memset(&perf_events, 0, sizeof(perf_events));
 
-	perf_num_counters = armpmu_get_max_events();
+	num_counters = perf_num_counters();
+	if (num_counters <= 0) {
+		pr_info("oprofile: no performance counters\n");
+		ret = -ENODEV;
+		goto out;
+	}
 
-	counter_config = kcalloc(perf_num_counters,
+	counter_config = kcalloc(num_counters,
 			sizeof(struct op_counter_config), GFP_KERNEL);
 
 	if (!counter_config) {
 		pr_info("oprofile: failed to allocate %d "
-				"counters\n", perf_num_counters);
+				"counters\n", num_counters);
 		ret = -ENOMEM;
 		goto out;
 	}
@@ -370,11 +375,11 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
 		goto out;
 
 	for_each_possible_cpu(cpu) {
-		perf_events[cpu] = kcalloc(perf_num_counters,
+		perf_events[cpu] = kcalloc(num_counters,
 				sizeof(struct perf_event *), GFP_KERNEL);
 		if (!perf_events[cpu]) {
 			pr_info("oprofile: failed to allocate %d perf events "
-					"for cpu %d\n", perf_num_counters, cpu);
+					"for cpu %d\n", num_counters, cpu);
 			ret = -ENOMEM;
 			goto out;
 		}
@@ -409,7 +414,7 @@ void __exit oprofile_arch_exit(void)
 	struct perf_event *event;
 
 	for_each_possible_cpu(cpu) {
-		for (id = 0; id < perf_num_counters; ++id) {
+		for (id = 0; id < num_counters; ++id) {
 			event = perf_events[cpu][id];
 			if (event)
 				perf_event_release_kernel(event);
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 7a3dc3567258..2cb9ad59d4b1 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -59,6 +59,15 @@ static inline int sh_pmu_initialized(void)
 	return !!sh_pmu;
 }
 
+int perf_num_counters(void)
+{
+	if (!sh_pmu)
+		return 0;
+
+	return sh_pmu->num_events;
+}
+EXPORT_SYMBOL_GPL(perf_num_counters);
+
 /*
  * Release the PMU if this is the last perf_event.
  */
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 716f99b682c1..1a0219247183 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -849,6 +849,7 @@ extern int perf_max_events;
 
 extern const struct pmu *hw_perf_event_init(struct perf_event *event);
 
+extern int perf_num_counters(void);
 extern void perf_event_task_sched_in(struct task_struct *task);
 extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
 extern void perf_event_task_tick(struct task_struct *task);
-- 
cgit v1.2.3


From 6370a6ad3b53df90b4700977f7718118a2cd524a Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Mon, 11 Oct 2010 15:12:27 +0200
Subject: workqueue: add and use WQ_MEM_RECLAIM flag

Add WQ_MEM_RECLAIM flag which currently maps to WQ_RESCUER, mark
WQ_RESCUER as internal and replace all external WQ_RESCUER usages to
WQ_MEM_RECLAIM.

This makes the API users express the intent of the workqueue instead
of indicating the internal mechanism used to guarantee forward
progress.  This is also to make it cleaner to add more semantics to
WQ_MEM_RECLAIM.  For example, if deemed necessary, memory reclaim
workqueues can be made highpri.

This patch doesn't introduce any functional change.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jeff Garzik <jgarzik@pobox.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Steven Whitehouse <swhiteho@redhat.com>
---
 Documentation/workqueue.txt | 29 +++++++++++++++--------------
 drivers/ata/libata-sff.c    |  2 +-
 fs/gfs2/main.c              |  2 +-
 fs/xfs/linux-2.6/xfs_buf.c  |  2 +-
 include/linux/workqueue.h   | 11 ++++++-----
 kernel/workqueue.c          |  7 +++++++
 6 files changed, 31 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/workqueue.txt b/Documentation/workqueue.txt
index e4498a2872c3..996a27d9b8db 100644
--- a/Documentation/workqueue.txt
+++ b/Documentation/workqueue.txt
@@ -196,11 +196,11 @@ resources, scheduled and executed.
 	suspend operations.  Work items on the wq are drained and no
 	new work item starts execution until thawed.
 
-  WQ_RESCUER
+  WQ_MEM_RECLAIM
 
 	All wq which might be used in the memory reclaim paths _MUST_
-	have this flag set.  This reserves one worker exclusively for
-	the execution of this wq under memory pressure.
+	have this flag set.  The wq is guaranteed to have at least one
+	execution context regardless of memory pressure.
 
   WQ_HIGHPRI
 
@@ -356,11 +356,11 @@ If q1 has WQ_CPU_INTENSIVE set,
 
 6. Guidelines
 
-* Do not forget to use WQ_RESCUER if a wq may process work items which
-  are used during memory reclaim.  Each wq with WQ_RESCUER set has one
-  rescuer thread reserved for it.  If there is dependency among
-  multiple work items used during memory reclaim, they should be
-  queued to separate wq each with WQ_RESCUER.
+* Do not forget to use WQ_MEM_RECLAIM if a wq may process work items
+  which are used during memory reclaim.  Each wq with WQ_MEM_RECLAIM
+  set has an execution context reserved for it.  If there is
+  dependency among multiple work items used during memory reclaim,
+  they should be queued to separate wq each with WQ_MEM_RECLAIM.
 
 * Unless strict ordering is required, there is no need to use ST wq.
 
@@ -368,12 +368,13 @@ If q1 has WQ_CPU_INTENSIVE set,
   recommended.  In most use cases, concurrency level usually stays
   well under the default limit.
 
-* A wq serves as a domain for forward progress guarantee (WQ_RESCUER),
-  flush and work item attributes.  Work items which are not involved
-  in memory reclaim and don't need to be flushed as a part of a group
-  of work items, and don't require any special attribute, can use one
-  of the system wq.  There is no difference in execution
-  characteristics between using a dedicated wq and a system wq.
+* A wq serves as a domain for forward progress guarantee
+  (WQ_MEM_RECLAIM, flush and work item attributes.  Work items which
+  are not involved in memory reclaim and don't need to be flushed as a
+  part of a group of work items, and don't require any special
+  attribute, can use one of the system wq.  There is no difference in
+  execution characteristics between using a dedicated wq and a system
+  wq.
 
 * Unless work items are expected to consume a huge amount of CPU
   cycles, using a bound wq is usually beneficial due to the increased
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index e30c537cce32..f5296bb19ec0 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -3335,7 +3335,7 @@ void ata_sff_port_init(struct ata_port *ap)
 
 int __init ata_sff_init(void)
 {
-	ata_sff_wq = alloc_workqueue("ata_sff", WQ_RESCUER, WQ_MAX_ACTIVE);
+	ata_sff_wq = alloc_workqueue("ata_sff", WQ_MEM_RECLAIM, WQ_MAX_ACTIVE);
 	if (!ata_sff_wq)
 		return -ENOMEM;
 
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index b1e9630eb46a..1c5f46075d52 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -140,7 +140,7 @@ static int __init init_gfs2_fs(void)
 
 	error = -ENOMEM;
 	gfs_recovery_wq = alloc_workqueue("gfs_recovery",
-					  WQ_NON_REENTRANT | WQ_RESCUER, 0);
+					  WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0);
 	if (!gfs_recovery_wq)
 		goto fail_wq;
 
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 286e36e21dae..6838aefca71f 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -1933,7 +1933,7 @@ xfs_buf_init(void)
 		goto out;
 
 	xfslogd_workqueue = alloc_workqueue("xfslogd",
-					WQ_RESCUER | WQ_HIGHPRI, 1);
+					WQ_MEM_RECLAIM | WQ_HIGHPRI, 1);
 	if (!xfslogd_workqueue)
 		goto out_free_buf_zone;
 
diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index e33ff4a91703..03bbe903e5ce 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -243,11 +243,12 @@ enum {
 	WQ_NON_REENTRANT	= 1 << 0, /* guarantee non-reentrance */
 	WQ_UNBOUND		= 1 << 1, /* not bound to any cpu */
 	WQ_FREEZEABLE		= 1 << 2, /* freeze during suspend */
-	WQ_RESCUER		= 1 << 3, /* has an rescue worker */
+	WQ_MEM_RECLAIM		= 1 << 3, /* may be used for memory reclaim */
 	WQ_HIGHPRI		= 1 << 4, /* high priority */
 	WQ_CPU_INTENSIVE	= 1 << 5, /* cpu instensive workqueue */
 
 	WQ_DYING		= 1 << 6, /* internal: workqueue is dying */
+	WQ_RESCUER		= 1 << 7, /* internal: workqueue has rescuer */
 
 	WQ_MAX_ACTIVE		= 512,	  /* I like 512, better ideas? */
 	WQ_MAX_UNBOUND_PER_CPU	= 4,	  /* 4 * #cpus for unbound wq */
@@ -309,7 +310,7 @@ __alloc_workqueue_key(const char *name, unsigned int flags, int max_active,
 /**
  * alloc_ordered_workqueue - allocate an ordered workqueue
  * @name: name of the workqueue
- * @flags: WQ_* flags (only WQ_FREEZEABLE and WQ_RESCUER are meaningful)
+ * @flags: WQ_* flags (only WQ_FREEZEABLE and WQ_MEM_RECLAIM are meaningful)
  *
  * Allocate an ordered workqueue.  An ordered workqueue executes at
  * most one work item at any given time in the queued order.  They are
@@ -325,11 +326,11 @@ alloc_ordered_workqueue(const char *name, unsigned int flags)
 }
 
 #define create_workqueue(name)					\
-	alloc_workqueue((name), WQ_RESCUER, 1)
+	alloc_workqueue((name), WQ_MEM_RECLAIM, 1)
 #define create_freezeable_workqueue(name)			\
-	alloc_workqueue((name), WQ_FREEZEABLE | WQ_UNBOUND | WQ_RESCUER, 1)
+	alloc_workqueue((name), WQ_FREEZEABLE | WQ_UNBOUND | WQ_MEM_RECLAIM, 1)
 #define create_singlethread_workqueue(name)			\
-	alloc_workqueue((name), WQ_UNBOUND | WQ_RESCUER, 1)
+	alloc_workqueue((name), WQ_UNBOUND | WQ_MEM_RECLAIM, 1)
 
 extern void destroy_workqueue(struct workqueue_struct *wq);
 
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index b57a8babdec3..2c6871cbcbee 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -2847,6 +2847,13 @@ struct workqueue_struct *__alloc_workqueue_key(const char *name,
 	struct workqueue_struct *wq;
 	unsigned int cpu;
 
+	/*
+	 * Workqueues which may be used during memory reclaim should
+	 * have a rescuer to guarantee forward progress.
+	 */
+	if (flags & WQ_MEM_RECLAIM)
+		flags |= WQ_RESCUER;
+
 	/*
 	 * Unbound workqueues aren't concurrency managed and should be
 	 * dispatched to workers immediately.
-- 
cgit v1.2.3


From 84c7991059c9c4530cc911137c5bf508a41ed129 Mon Sep 17 00:00:00 2001
From: Matt Fleming <matt@console-pimps.org>
Date: Sun, 3 Oct 2010 21:41:13 +0100
Subject: perf: New helper function for pmu name

Introduce perf_pmu_name() helper function that returns the name of the
pmu. This gives us a generic way to get the name of a pmu regardless of
how an architecture identifies it internally.

Signed-off-by: Matt Fleming <matt@console-pimps.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/sh/kernel/perf_event.c | 9 +++++++++
 include/linux/perf_event.h  | 1 +
 kernel/perf_event.c         | 5 +++++
 3 files changed, 15 insertions(+)

(limited to 'include/linux')

diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 2cb9ad59d4b1..55fe89bbdfe0 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -59,6 +59,15 @@ static inline int sh_pmu_initialized(void)
 	return !!sh_pmu;
 }
 
+const char *perf_pmu_name(void)
+{
+	if (!sh_pmu)
+		return NULL;
+
+	return sh_pmu->name;
+}
+EXPORT_SYMBOL_GPL(perf_pmu_name);
+
 int perf_num_counters(void)
 {
 	if (!sh_pmu)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 1a0219247183..33f08dafda2f 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -850,6 +850,7 @@ extern int perf_max_events;
 extern const struct pmu *hw_perf_event_init(struct perf_event *event);
 
 extern int perf_num_counters(void);
+extern const char *perf_pmu_name(void);
 extern void perf_event_task_sched_in(struct task_struct *task);
 extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
 extern void perf_event_task_tick(struct task_struct *task);
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 403d1804b198..e2534691db0d 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -85,6 +85,11 @@ void __weak hw_perf_enable(void)		{ barrier(); }
 
 void __weak perf_event_print_debug(void)	{ }
 
+extern __weak const char *perf_pmu_name(void)
+{
+	return "pmu";
+}
+
 static DEFINE_PER_CPU(int, perf_disable_count);
 
 void perf_disable(void)
-- 
cgit v1.2.3


From 56946331b28d53232115a155ba662ab3dc598952 Mon Sep 17 00:00:00 2001
From: Matt Fleming <matt@console-pimps.org>
Date: Fri, 8 Oct 2010 21:42:17 +0100
Subject: oprofile: Make op_name_from_perf_id() global

Make op_name_from_perf_id() global so that we have a way for each
architecture to construct an oprofile name for op->cpu_type. We need to
remove the argument from the function prototype so that we can hide all
implementation details inside the function.

Signed-off-by: Matt Fleming <matt@console-pimps.org>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/arm/oprofile/common.c | 6 ++++--
 include/linux/oprofile.h   | 4 ++++
 2 files changed, 8 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c
index 1e971a7fcf82..4f67cfab2c59 100644
--- a/arch/arm/oprofile/common.c
+++ b/arch/arm/oprofile/common.c
@@ -155,8 +155,10 @@ static void op_perf_stop(void)
 }
 
 
-static char *op_name_from_perf_id(enum arm_perf_pmu_ids id)
+char *op_name_from_perf_id(void)
 {
+	enum arm_perf_pmu_ids id = armpmu_get_pmu_id();
+
 	switch (id) {
 	case ARM_PERF_PMU_ID_XSCALE1:
 		return "arm/xscale1";
@@ -391,7 +393,7 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
 	ops->start		= op_arm_start;
 	ops->stop		= op_arm_stop;
 	ops->shutdown		= op_arm_stop;
-	ops->cpu_type		= op_name_from_perf_id(armpmu_get_pmu_id());
+	ops->cpu_type		= op_name_from_perf_id();
 
 	if (!ops->cpu_type)
 		ret = -ENODEV;
diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h
index 5171639ecf0f..1574d4aca721 100644
--- a/include/linux/oprofile.h
+++ b/include/linux/oprofile.h
@@ -185,4 +185,8 @@ int oprofile_add_data(struct op_entry *entry, unsigned long val);
 int oprofile_add_data64(struct op_entry *entry, u64 val);
 int oprofile_write_commit(struct op_entry *entry);
 
+#ifdef CONFIG_PERF_EVENTS
+char *op_name_from_perf_id(void);
+#endif /* CONFIG_PERF_EVENTS */
+
 #endif /* OPROFILE_H */
-- 
cgit v1.2.3


From 3d90a00763b51e1db344a7430c966be723b67a29 Mon Sep 17 00:00:00 2001
From: Matt Fleming <matt@console-pimps.org>
Date: Mon, 27 Sep 2010 20:45:08 +0100
Subject: oprofile: Abstract the perf-events backend

Move the perf-events backend from arch/arm/oprofile into
drivers/oprofile so that the code can be shared between architectures.

This allows each architecture to maintain only a single copy of the PMU
accessor functions instead of one for both perf and OProfile. It also
becomes possible for other architectures to delete much of their
OProfile code in favour of the common code now available in
drivers/oprofile/oprofile_perf.c.

Signed-off-by: Matt Fleming <matt@console-pimps.org>
Tested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/arm/oprofile/Makefile       |   4 +
 arch/arm/oprofile/common.c       | 319 --------------------------------------
 drivers/oprofile/oprofile_perf.c | 326 +++++++++++++++++++++++++++++++++++++++
 include/linux/oprofile.h         |   3 +
 4 files changed, 333 insertions(+), 319 deletions(-)
 create mode 100644 drivers/oprofile/oprofile_perf.c

(limited to 'include/linux')

diff --git a/arch/arm/oprofile/Makefile b/arch/arm/oprofile/Makefile
index e666eafed152..b2215c61cdf0 100644
--- a/arch/arm/oprofile/Makefile
+++ b/arch/arm/oprofile/Makefile
@@ -6,4 +6,8 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \
 		oprofilefs.o oprofile_stats.o \
 		timer_int.o )
 
+ifeq ($(CONFIG_HW_PERF_EVENTS),y)
+DRIVER_OBJS += $(addprefix ../../../drivers/oprofile/, oprofile_perf.o)
+endif
+
 oprofile-y				:= $(DRIVER_OBJS) common.o
diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c
index 8718311cb530..8aa974491dfc 100644
--- a/arch/arm/oprofile/common.c
+++ b/arch/arm/oprofile/common.c
@@ -25,136 +25,6 @@
 #include <asm/ptrace.h>
 
 #ifdef CONFIG_HW_PERF_EVENTS
-/*
- * Per performance monitor configuration as set via oprofilefs.
- */
-struct op_counter_config {
-	unsigned long count;
-	unsigned long enabled;
-	unsigned long event;
-	unsigned long unit_mask;
-	unsigned long kernel;
-	unsigned long user;
-	struct perf_event_attr attr;
-};
-
-static int oprofile_perf_enabled;
-static DEFINE_MUTEX(oprofile_perf_mutex);
-
-static struct op_counter_config *counter_config;
-static struct perf_event **perf_events[nr_cpumask_bits];
-static int num_counters;
-
-/*
- * Overflow callback for oprofile.
- */
-static void op_overflow_handler(struct perf_event *event, int unused,
-			struct perf_sample_data *data, struct pt_regs *regs)
-{
-	int id;
-	u32 cpu = smp_processor_id();
-
-	for (id = 0; id < num_counters; ++id)
-		if (perf_events[cpu][id] == event)
-			break;
-
-	if (id != num_counters)
-		oprofile_add_sample(regs, id);
-	else
-		pr_warning("oprofile: ignoring spurious overflow "
-				"on cpu %u\n", cpu);
-}
-
-/*
- * Called by oprofile_perf_setup to create perf attributes to mirror the oprofile
- * settings in counter_config. Attributes are created as `pinned' events and
- * so are permanently scheduled on the PMU.
- */
-static void op_perf_setup(void)
-{
-	int i;
-	u32 size = sizeof(struct perf_event_attr);
-	struct perf_event_attr *attr;
-
-	for (i = 0; i < num_counters; ++i) {
-		attr = &counter_config[i].attr;
-		memset(attr, 0, size);
-		attr->type		= PERF_TYPE_RAW;
-		attr->size		= size;
-		attr->config		= counter_config[i].event;
-		attr->sample_period	= counter_config[i].count;
-		attr->pinned		= 1;
-	}
-}
-
-static int op_create_counter(int cpu, int event)
-{
-	int ret = 0;
-	struct perf_event *pevent;
-
-	if (!counter_config[event].enabled || (perf_events[cpu][event] != NULL))
-		return ret;
-
-	pevent = perf_event_create_kernel_counter(&counter_config[event].attr,
-						  cpu, -1,
-						  op_overflow_handler);
-
-	if (IS_ERR(pevent)) {
-		ret = PTR_ERR(pevent);
-	} else if (pevent->state != PERF_EVENT_STATE_ACTIVE) {
-		pr_warning("oprofile: failed to enable event %d "
-				"on CPU %d\n", event, cpu);
-		ret = -EBUSY;
-	} else {
-		perf_events[cpu][event] = pevent;
-	}
-
-	return ret;
-}
-
-static void op_destroy_counter(int cpu, int event)
-{
-	struct perf_event *pevent = perf_events[cpu][event];
-
-	if (pevent) {
-		perf_event_release_kernel(pevent);
-		perf_events[cpu][event] = NULL;
-	}
-}
-
-/*
- * Called by oprofile_perf_start to create active perf events based on the
- * perviously configured attributes.
- */
-static int op_perf_start(void)
-{
-	int cpu, event, ret = 0;
-
-	for_each_online_cpu(cpu) {
-		for (event = 0; event < num_counters; ++event) {
-			ret = op_create_counter(cpu, event);
-			if (ret)
-				goto out;
-		}
-	}
-
-out:
-	return ret;
-}
-
-/*
- * Called by oprofile_perf_stop at the end of a profiling run.
- */
-static void op_perf_stop(void)
-{
-	int cpu, event;
-
-	for_each_online_cpu(cpu)
-		for (event = 0; event < num_counters; ++event)
-			op_destroy_counter(cpu, event);
-}
-
-
 char *op_name_from_perf_id(void)
 {
 	enum arm_perf_pmu_ids id = armpmu_get_pmu_id();
@@ -177,116 +47,6 @@ char *op_name_from_perf_id(void)
 	}
 }
 
-static int oprofile_perf_create_files(struct super_block *sb, struct dentry *root)
-{
-	unsigned int i;
-
-	for (i = 0; i < num_counters; i++) {
-		struct dentry *dir;
-		char buf[4];
-
-		snprintf(buf, sizeof buf, "%d", i);
-		dir = oprofilefs_mkdir(sb, root, buf);
-		oprofilefs_create_ulong(sb, dir, "enabled", &counter_config[i].enabled);
-		oprofilefs_create_ulong(sb, dir, "event", &counter_config[i].event);
-		oprofilefs_create_ulong(sb, dir, "count", &counter_config[i].count);
-		oprofilefs_create_ulong(sb, dir, "unit_mask", &counter_config[i].unit_mask);
-		oprofilefs_create_ulong(sb, dir, "kernel", &counter_config[i].kernel);
-		oprofilefs_create_ulong(sb, dir, "user", &counter_config[i].user);
-	}
-
-	return 0;
-}
-
-static int oprofile_perf_setup(void)
-{
-	spin_lock(&oprofilefs_lock);
-	op_perf_setup();
-	spin_unlock(&oprofilefs_lock);
-	return 0;
-}
-
-static int oprofile_perf_start(void)
-{
-	int ret = -EBUSY;
-
-	mutex_lock(&oprofile_perf_mutex);
-	if (!oprofile_perf_enabled) {
-		ret = 0;
-		op_perf_start();
-		oprofile_perf_enabled = 1;
-	}
-	mutex_unlock(&oprofile_perf_mutex);
-	return ret;
-}
-
-static void oprofile_perf_stop(void)
-{
-	mutex_lock(&oprofile_perf_mutex);
-	if (oprofile_perf_enabled)
-		op_perf_stop();
-	oprofile_perf_enabled = 0;
-	mutex_unlock(&oprofile_perf_mutex);
-}
-
-#ifdef CONFIG_PM
-static int oprofile_perf_suspend(struct platform_device *dev, pm_message_t state)
-{
-	mutex_lock(&oprofile_perf_mutex);
-	if (oprofile_perf_enabled)
-		op_perf_stop();
-	mutex_unlock(&oprofile_perf_mutex);
-	return 0;
-}
-
-static int oprofile_perf_resume(struct platform_device *dev)
-{
-	mutex_lock(&oprofile_perf_mutex);
-	if (oprofile_perf_enabled && op_perf_start())
-		oprofile_perf_enabled = 0;
-	mutex_unlock(&oprofile_perf_mutex);
-	return 0;
-}
-
-static struct platform_driver oprofile_driver = {
-	.driver		= {
-		.name		= "oprofile-perf",
-	},
-	.resume		= oprofile_perf_resume,
-	.suspend	= oprofile_perf_suspend,
-};
-
-static struct platform_device *oprofile_pdev;
-
-static int __init init_driverfs(void)
-{
-	int ret;
-
-	ret = platform_driver_register(&oprofile_driver);
-	if (ret)
-		goto out;
-
-	oprofile_pdev =	platform_device_register_simple(
-				oprofile_driver.driver.name, 0, NULL, 0);
-	if (IS_ERR(oprofile_pdev)) {
-		ret = PTR_ERR(oprofile_pdev);
-		platform_driver_unregister(&oprofile_driver);
-	}
-
-out:
-	return ret;
-}
-
-static void __exit exit_driverfs(void)
-{
-	platform_device_unregister(oprofile_pdev);
-	platform_driver_unregister(&oprofile_driver);
-}
-#else
-static int __init init_driverfs(void) { return 0; }
-#define exit_driverfs() do { } while (0)
-#endif /* CONFIG_PM */
-
 static int report_trace(struct stackframe *frame, void *d)
 {
 	unsigned int *depth = d;
@@ -349,66 +109,6 @@ static void arm_backtrace(struct pt_regs * const regs, unsigned int depth)
 		tail = user_backtrace(tail);
 }
 
-int __init oprofile_perf_init(struct oprofile_operations *ops)
-{
-	int cpu, ret = 0;
-
-	memset(&perf_events, 0, sizeof(perf_events));
-
-	num_counters = perf_num_counters();
-	if (num_counters <= 0) {
-		pr_info("oprofile: no performance counters\n");
-		ret = -ENODEV;
-		goto out;
-	}
-
-	counter_config = kcalloc(num_counters,
-			sizeof(struct op_counter_config), GFP_KERNEL);
-
-	if (!counter_config) {
-		pr_info("oprofile: failed to allocate %d "
-				"counters\n", num_counters);
-		ret = -ENOMEM;
-		goto out;
-	}
-
-	ret = init_driverfs();
-	if (ret)
-		goto out;
-
-	for_each_possible_cpu(cpu) {
-		perf_events[cpu] = kcalloc(num_counters,
-				sizeof(struct perf_event *), GFP_KERNEL);
-		if (!perf_events[cpu]) {
-			pr_info("oprofile: failed to allocate %d perf events "
-					"for cpu %d\n", num_counters, cpu);
-			ret = -ENOMEM;
-			goto out;
-		}
-	}
-
-	ops->create_files	= oprofile_perf_create_files;
-	ops->setup		= oprofile_perf_setup;
-	ops->start		= oprofile_perf_start;
-	ops->stop		= oprofile_perf_stop;
-	ops->shutdown		= oprofile_perf_stop;
-	ops->cpu_type		= op_name_from_perf_id();
-
-	if (!ops->cpu_type)
-		ret = -ENODEV;
-	else
-		pr_info("oprofile: using %s\n", ops->cpu_type);
-
-out:
-	if (ret) {
-		for_each_possible_cpu(cpu)
-			kfree(perf_events[cpu]);
-		kfree(counter_config);
-	}
-
-	return ret;
-}
-
 int __init oprofile_arch_init(struct oprofile_operations *ops)
 {
 	ops->backtrace		= arm_backtrace;
@@ -416,25 +116,6 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
 	return oprofile_perf_init(ops);
 }
 
-void __exit oprofile_perf_exit(void)
-{
-	int cpu, id;
-	struct perf_event *event;
-
-	for_each_possible_cpu(cpu) {
-		for (id = 0; id < num_counters; ++id) {
-			event = perf_events[cpu][id];
-			if (event)
-				perf_event_release_kernel(event);
-		}
-
-		kfree(perf_events[cpu]);
-	}
-
-	kfree(counter_config);
-	exit_driverfs();
-}
-
 void __exit oprofile_arch_exit(void)
 {
 	oprofile_perf_exit();
diff --git a/drivers/oprofile/oprofile_perf.c b/drivers/oprofile/oprofile_perf.c
new file mode 100644
index 000000000000..ebb40cb87474
--- /dev/null
+++ b/drivers/oprofile/oprofile_perf.c
@@ -0,0 +1,326 @@
+/*
+ * Copyright 2010 ARM Ltd.
+ *
+ * Perf-events backend for OProfile.
+ */
+#include <linux/perf_event.h>
+#include <linux/oprofile.h>
+#include <linux/slab.h>
+
+/*
+ * Per performance monitor configuration as set via oprofilefs.
+ */
+struct op_counter_config {
+	unsigned long count;
+	unsigned long enabled;
+	unsigned long event;
+	unsigned long unit_mask;
+	unsigned long kernel;
+	unsigned long user;
+	struct perf_event_attr attr;
+};
+
+static int oprofile_perf_enabled;
+static DEFINE_MUTEX(oprofile_perf_mutex);
+
+static struct op_counter_config *counter_config;
+static struct perf_event **perf_events[nr_cpumask_bits];
+static int num_counters;
+
+/*
+ * Overflow callback for oprofile.
+ */
+static void op_overflow_handler(struct perf_event *event, int unused,
+			struct perf_sample_data *data, struct pt_regs *regs)
+{
+	int id;
+	u32 cpu = smp_processor_id();
+
+	for (id = 0; id < num_counters; ++id)
+		if (perf_events[cpu][id] == event)
+			break;
+
+	if (id != num_counters)
+		oprofile_add_sample(regs, id);
+	else
+		pr_warning("oprofile: ignoring spurious overflow "
+				"on cpu %u\n", cpu);
+}
+
+/*
+ * Called by oprofile_perf_setup to create perf attributes to mirror the oprofile
+ * settings in counter_config. Attributes are created as `pinned' events and
+ * so are permanently scheduled on the PMU.
+ */
+static void op_perf_setup(void)
+{
+	int i;
+	u32 size = sizeof(struct perf_event_attr);
+	struct perf_event_attr *attr;
+
+	for (i = 0; i < num_counters; ++i) {
+		attr = &counter_config[i].attr;
+		memset(attr, 0, size);
+		attr->type		= PERF_TYPE_RAW;
+		attr->size		= size;
+		attr->config		= counter_config[i].event;
+		attr->sample_period	= counter_config[i].count;
+		attr->pinned		= 1;
+	}
+}
+
+static int op_create_counter(int cpu, int event)
+{
+	int ret = 0;
+	struct perf_event *pevent;
+
+	if (!counter_config[event].enabled || (perf_events[cpu][event] != NULL))
+		return ret;
+
+	pevent = perf_event_create_kernel_counter(&counter_config[event].attr,
+						  cpu, -1,
+						  op_overflow_handler);
+
+	if (IS_ERR(pevent)) {
+		ret = PTR_ERR(pevent);
+	} else if (pevent->state != PERF_EVENT_STATE_ACTIVE) {
+		pr_warning("oprofile: failed to enable event %d "
+				"on CPU %d\n", event, cpu);
+		ret = -EBUSY;
+	} else {
+		perf_events[cpu][event] = pevent;
+	}
+
+	return ret;
+}
+
+static void op_destroy_counter(int cpu, int event)
+{
+	struct perf_event *pevent = perf_events[cpu][event];
+
+	if (pevent) {
+		perf_event_release_kernel(pevent);
+		perf_events[cpu][event] = NULL;
+	}
+}
+
+/*
+ * Called by oprofile_perf_start to create active perf events based on the
+ * perviously configured attributes.
+ */
+static int op_perf_start(void)
+{
+	int cpu, event, ret = 0;
+
+	for_each_online_cpu(cpu) {
+		for (event = 0; event < num_counters; ++event) {
+			ret = op_create_counter(cpu, event);
+			if (ret)
+				goto out;
+		}
+	}
+
+out:
+	return ret;
+}
+
+/*
+ * Called by oprofile_perf_stop at the end of a profiling run.
+ */
+static void op_perf_stop(void)
+{
+	int cpu, event;
+
+	for_each_online_cpu(cpu)
+		for (event = 0; event < num_counters; ++event)
+			op_destroy_counter(cpu, event);
+}
+
+static int oprofile_perf_create_files(struct super_block *sb, struct dentry *root)
+{
+	unsigned int i;
+
+	for (i = 0; i < num_counters; i++) {
+		struct dentry *dir;
+		char buf[4];
+
+		snprintf(buf, sizeof buf, "%d", i);
+		dir = oprofilefs_mkdir(sb, root, buf);
+		oprofilefs_create_ulong(sb, dir, "enabled", &counter_config[i].enabled);
+		oprofilefs_create_ulong(sb, dir, "event", &counter_config[i].event);
+		oprofilefs_create_ulong(sb, dir, "count", &counter_config[i].count);
+		oprofilefs_create_ulong(sb, dir, "unit_mask", &counter_config[i].unit_mask);
+		oprofilefs_create_ulong(sb, dir, "kernel", &counter_config[i].kernel);
+		oprofilefs_create_ulong(sb, dir, "user", &counter_config[i].user);
+	}
+
+	return 0;
+}
+
+static int oprofile_perf_setup(void)
+{
+	spin_lock(&oprofilefs_lock);
+	op_perf_setup();
+	spin_unlock(&oprofilefs_lock);
+	return 0;
+}
+
+static int oprofile_perf_start(void)
+{
+	int ret = -EBUSY;
+
+	mutex_lock(&oprofile_perf_mutex);
+	if (!oprofile_perf_enabled) {
+		ret = 0;
+		op_perf_start();
+		oprofile_perf_enabled = 1;
+	}
+	mutex_unlock(&oprofile_perf_mutex);
+	return ret;
+}
+
+static void oprofile_perf_stop(void)
+{
+	mutex_lock(&oprofile_perf_mutex);
+	if (oprofile_perf_enabled)
+		op_perf_stop();
+	oprofile_perf_enabled = 0;
+	mutex_unlock(&oprofile_perf_mutex);
+}
+
+#ifdef CONFIG_PM
+static int oprofile_perf_suspend(struct platform_device *dev, pm_message_t state)
+{
+	mutex_lock(&oprofile_perf_mutex);
+	if (oprofile_perf_enabled)
+		op_perf_stop();
+	mutex_unlock(&oprofile_perf_mutex);
+	return 0;
+}
+
+static int oprofile_perf_resume(struct platform_device *dev)
+{
+	mutex_lock(&oprofile_perf_mutex);
+	if (oprofile_perf_enabled && op_perf_start())
+		oprofile_perf_enabled = 0;
+	mutex_unlock(&oprofile_perf_mutex);
+	return 0;
+}
+
+static struct platform_driver oprofile_driver = {
+	.driver		= {
+		.name		= "oprofile-perf",
+	},
+	.resume		= oprofile_perf_resume,
+	.suspend	= oprofile_perf_suspend,
+};
+
+static struct platform_device *oprofile_pdev;
+
+static int __init init_driverfs(void)
+{
+	int ret;
+
+	ret = platform_driver_register(&oprofile_driver);
+	if (ret)
+		goto out;
+
+	oprofile_pdev =	platform_device_register_simple(
+				oprofile_driver.driver.name, 0, NULL, 0);
+	if (IS_ERR(oprofile_pdev)) {
+		ret = PTR_ERR(oprofile_pdev);
+		platform_driver_unregister(&oprofile_driver);
+	}
+
+out:
+	return ret;
+}
+
+static void __exit exit_driverfs(void)
+{
+	platform_device_unregister(oprofile_pdev);
+	platform_driver_unregister(&oprofile_driver);
+}
+#else
+static int __init init_driverfs(void) { return 0; }
+#define exit_driverfs() do { } while (0)
+#endif /* CONFIG_PM */
+
+int __init oprofile_perf_init(struct oprofile_operations *ops)
+{
+	int cpu, ret = 0;
+
+	memset(&perf_events, 0, sizeof(perf_events));
+
+	num_counters = perf_num_counters();
+	if (num_counters <= 0) {
+		pr_info("oprofile: no performance counters\n");
+		ret = -ENODEV;
+		goto out;
+	}
+
+	counter_config = kcalloc(num_counters,
+			sizeof(struct op_counter_config), GFP_KERNEL);
+
+	if (!counter_config) {
+		pr_info("oprofile: failed to allocate %d "
+				"counters\n", num_counters);
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = init_driverfs();
+	if (ret)
+		goto out;
+
+	for_each_possible_cpu(cpu) {
+		perf_events[cpu] = kcalloc(num_counters,
+				sizeof(struct perf_event *), GFP_KERNEL);
+		if (!perf_events[cpu]) {
+			pr_info("oprofile: failed to allocate %d perf events "
+					"for cpu %d\n", num_counters, cpu);
+			ret = -ENOMEM;
+			goto out;
+		}
+	}
+
+	ops->create_files	= oprofile_perf_create_files;
+	ops->setup		= oprofile_perf_setup;
+	ops->start		= oprofile_perf_start;
+	ops->stop		= oprofile_perf_stop;
+	ops->shutdown		= oprofile_perf_stop;
+	ops->cpu_type		= op_name_from_perf_id();
+
+	if (!ops->cpu_type)
+		ret = -ENODEV;
+	else
+		pr_info("oprofile: using %s\n", ops->cpu_type);
+
+out:
+	if (ret) {
+		for_each_possible_cpu(cpu)
+			kfree(perf_events[cpu]);
+		kfree(counter_config);
+	}
+
+	return ret;
+}
+
+void __exit oprofile_perf_exit(void)
+{
+	int cpu, id;
+	struct perf_event *event;
+
+	for_each_possible_cpu(cpu) {
+		for (id = 0; id < num_counters; ++id) {
+			event = perf_events[cpu][id];
+			if (event)
+				perf_event_release_kernel(event);
+		}
+
+		kfree(perf_events[cpu]);
+	}
+
+	kfree(counter_config);
+	exit_driverfs();
+}
diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h
index 1574d4aca721..d67a8330b41e 100644
--- a/include/linux/oprofile.h
+++ b/include/linux/oprofile.h
@@ -15,6 +15,7 @@
 
 #include <linux/types.h>
 #include <linux/spinlock.h>
+#include <linux/init.h>
 #include <asm/atomic.h>
  
 /* Each escaped entry is prefixed by ESCAPE_CODE
@@ -186,6 +187,8 @@ int oprofile_add_data64(struct op_entry *entry, u64 val);
 int oprofile_write_commit(struct op_entry *entry);
 
 #ifdef CONFIG_PERF_EVENTS
+int __init oprofile_perf_init(struct oprofile_operations *ops);
+void __exit oprofile_perf_exit(void);
 char *op_name_from_perf_id(void);
 #endif /* CONFIG_PERF_EVENTS */
 
-- 
cgit v1.2.3


From 34d101dd6204bd100fc2e6f7b5f9a10f959ce2c9 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 11 Oct 2010 09:16:57 -0700
Subject: neigh: speedup neigh_hh_init()

When a new dst is used to send a frame, neigh_resolve_output() tries to
associate an struct hh_cache to this dst, calling neigh_hh_init() with
the neigh rwlock write locked.

Most of the time, hh_cache is already known and linked into neighbour,
so we find it and increment its refcount.

This patch changes the logic so that we call neigh_hh_init() with
neighbour lock read locked only, so that fast path can be run in
parallel by concurrent cpus.

This brings part of the speedup we got with commit c7d4426a98a5f
(introduce DST_NOCACHE flag) for non cached dsts, even for cached ones,
removing one of the contention point that routers hit on multiqueue
enabled machines.

Further improvements would need to use a seqlock instead of an rwlock to
protect neigh->ha[], to not dirty neigh too often and remove two atomic
ops.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  6 +++
 net/core/dst.c            |  4 +-
 net/core/neighbour.c      | 99 +++++++++++++++++++++++++++++------------------
 3 files changed, 69 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 6abcef67b178..4160db3721ba 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -281,6 +281,12 @@ struct hh_cache {
 	unsigned long	hh_data[HH_DATA_ALIGN(LL_MAX_HEADER) / sizeof(long)];
 };
 
+static inline void hh_cache_put(struct hh_cache *hh)
+{
+	if (atomic_dec_and_test(&hh->hh_refcnt))
+		kfree(hh);
+}
+
 /* Reserve HH_DATA_MOD byte aligned hard_header_len, but at least that much.
  * Alternative is:
  *   dev->hard_header_len ? (dev->hard_header_len +
diff --git a/net/core/dst.c b/net/core/dst.c
index 6c41b1fac3db..978a1ee1f7d0 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -228,8 +228,8 @@ again:
 	child = dst->child;
 
 	dst->hh = NULL;
-	if (hh && atomic_dec_and_test(&hh->hh_refcnt))
-		kfree(hh);
+	if (hh)
+		hh_cache_put(hh);
 
 	if (neigh) {
 		dst->neighbour = NULL;
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 3ffafaa0414c..2044906ecd1a 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -709,8 +709,7 @@ void neigh_destroy(struct neighbour *neigh)
 		write_seqlock_bh(&hh->hh_lock);
 		hh->hh_output = neigh_blackhole;
 		write_sequnlock_bh(&hh->hh_lock);
-		if (atomic_dec_and_test(&hh->hh_refcnt))
-			kfree(hh);
+		hh_cache_put(hh);
 	}
 
 	skb_queue_purge(&neigh->arp_queue);
@@ -1210,39 +1209,67 @@ struct neighbour *neigh_event_ns(struct neigh_table *tbl,
 }
 EXPORT_SYMBOL(neigh_event_ns);
 
+static inline bool neigh_hh_lookup(struct neighbour *n, struct dst_entry *dst,
+				   __be16 protocol)
+{
+	struct hh_cache *hh;
+
+	for (hh = n->hh; hh; hh = hh->hh_next) {
+		if (hh->hh_type == protocol) {
+			atomic_inc(&hh->hh_refcnt);
+			if (unlikely(cmpxchg(&dst->hh, NULL, hh) != NULL))
+				hh_cache_put(hh);
+			return true;
+		}
+	}
+	return false;
+}
+
+/* called with read_lock_bh(&n->lock); */
 static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst,
 			  __be16 protocol)
 {
 	struct hh_cache	*hh;
 	struct net_device *dev = dst->dev;
 
-	for (hh = n->hh; hh; hh = hh->hh_next)
-		if (hh->hh_type == protocol)
-			break;
+	if (likely(neigh_hh_lookup(n, dst, protocol)))
+		return;
 
-	if (!hh && (hh = kzalloc(sizeof(*hh), GFP_ATOMIC)) != NULL) {
-		seqlock_init(&hh->hh_lock);
-		hh->hh_type = protocol;
-		atomic_set(&hh->hh_refcnt, 0);
-		hh->hh_next = NULL;
+	/* slow path */
+	hh = kzalloc(sizeof(*hh), GFP_ATOMIC);
+	if (!hh)
+		return;
 
-		if (dev->header_ops->cache(n, hh)) {
-			kfree(hh);
-			hh = NULL;
-		} else {
-			atomic_inc(&hh->hh_refcnt);
-			hh->hh_next = n->hh;
-			n->hh	    = hh;
-			if (n->nud_state & NUD_CONNECTED)
-				hh->hh_output = n->ops->hh_output;
-			else
-				hh->hh_output = n->ops->output;
-		}
+	seqlock_init(&hh->hh_lock);
+	hh->hh_type = protocol;
+	atomic_set(&hh->hh_refcnt, 2);
+
+	if (dev->header_ops->cache(n, hh)) {
+		kfree(hh);
+		return;
 	}
-	if (hh)	{
-		atomic_inc(&hh->hh_refcnt);
-		dst->hh = hh;
+	read_unlock(&n->lock);
+	write_lock(&n->lock);
+
+	/* must check if another thread already did the insert */
+	if (neigh_hh_lookup(n, dst, protocol)) {
+		kfree(hh);
+		goto end;
 	}
+
+	if (n->nud_state & NUD_CONNECTED)
+		hh->hh_output = n->ops->hh_output;
+	else
+		hh->hh_output = n->ops->output;
+
+	hh->hh_next = n->hh;
+	n->hh	    = hh;
+
+	if (unlikely(cmpxchg(&dst->hh, NULL, hh) != NULL))
+		hh_cache_put(hh);
+end:
+	write_unlock(&n->lock);
+	read_lock(&n->lock);
 }
 
 /* This function can be used in contexts, where only old dev_queue_xmit
@@ -1281,21 +1308,17 @@ int neigh_resolve_output(struct sk_buff *skb)
 	if (!neigh_event_send(neigh, skb)) {
 		int err;
 		struct net_device *dev = neigh->dev;
+
+		read_lock_bh(&neigh->lock);
 		if (dev->header_ops->cache &&
 		    !dst->hh &&
-		    !(dst->flags & DST_NOCACHE)) {
-			write_lock_bh(&neigh->lock);
-			if (!dst->hh)
-				neigh_hh_init(neigh, dst, dst->ops->protocol);
-			err = dev_hard_header(skb, dev, ntohs(skb->protocol),
-					      neigh->ha, NULL, skb->len);
-			write_unlock_bh(&neigh->lock);
-		} else {
-			read_lock_bh(&neigh->lock);
-			err = dev_hard_header(skb, dev, ntohs(skb->protocol),
-					      neigh->ha, NULL, skb->len);
-			read_unlock_bh(&neigh->lock);
-		}
+		    !(dst->flags & DST_NOCACHE))
+			neigh_hh_init(neigh, dst, dst->ops->protocol);
+
+		err = dev_hard_header(skb, dev, ntohs(skb->protocol),
+				      neigh->ha, NULL, skb->len);
+		read_unlock_bh(&neigh->lock);
+
 		if (err >= 0)
 			rc = neigh->ops->queue_xmit(skb);
 		else
-- 
cgit v1.2.3


From 8610c29a2c9f273886b1c31ae4d92c69d4326262 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@openwrt.org>
Date: Sat, 9 Oct 2010 02:39:29 +0200
Subject: cfg80211: add channel utilization stats to the survey command

Using these, user space can calculate a relative channel utilization
with arbitrary intervals by regularly taking snapshots of the survey
results.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 15 +++++++++++++++
 include/net/cfg80211.h  | 20 ++++++++++++++++++++
 net/wireless/nl80211.c  | 15 +++++++++++++++
 3 files changed, 50 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index c08709fe36fc..0edb2566c14c 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -1413,6 +1413,16 @@ enum nl80211_reg_rule_flags {
  * @NL80211_SURVEY_INFO_FREQUENCY: center frequency of channel
  * @NL80211_SURVEY_INFO_NOISE: noise level of channel (u8, dBm)
  * @NL80211_SURVEY_INFO_IN_USE: channel is currently being used
+ * @NL80211_SURVEY_INFO_CHANNEL_TIME: amount of time (in ms) that the radio
+ *	spent on this channel
+ * @NL80211_SURVEY_INFO_CHANNEL_TIME_BUSY: amount of the time the primary
+ *	channel was sensed busy (either due to activity or energy detect)
+ * @NL80211_SURVEY_INFO_CHANNEL_TIME_EXT_BUSY: amount of time the extension
+ *	channel was sensed busy
+ * @NL80211_SURVEY_INFO_CHANNEL_TIME_RX: amount of time the radio spent
+ *	receiving data
+ * @NL80211_SURVEY_INFO_CHANNEL_TIME_TX: amount of time the radio spent
+ *	transmitting data
  * @NL80211_SURVEY_INFO_MAX: highest survey info attribute number
  *	currently defined
  * @__NL80211_SURVEY_INFO_AFTER_LAST: internal use
@@ -1422,6 +1432,11 @@ enum nl80211_survey_info {
 	NL80211_SURVEY_INFO_FREQUENCY,
 	NL80211_SURVEY_INFO_NOISE,
 	NL80211_SURVEY_INFO_IN_USE,
+	NL80211_SURVEY_INFO_CHANNEL_TIME,
+	NL80211_SURVEY_INFO_CHANNEL_TIME_BUSY,
+	NL80211_SURVEY_INFO_CHANNEL_TIME_EXT_BUSY,
+	NL80211_SURVEY_INFO_CHANNEL_TIME_RX,
+	NL80211_SURVEY_INFO_CHANNEL_TIME_TX,
 
 	/* keep last */
 	__NL80211_SURVEY_INFO_AFTER_LAST,
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index f920a06f363e..24d5b5869272 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -294,6 +294,11 @@ struct key_params {
  *
  * @SURVEY_INFO_NOISE_DBM: noise (in dBm) was filled in
  * @SURVEY_INFO_IN_USE: channel is currently being used
+ * @SURVEY_INFO_CHANNEL_TIME: channel active time (in ms) was filled in
+ * @SURVEY_INFO_CHANNEL_TIME_BUSY: channel busy time was filled in
+ * @SURVEY_INFO_CHANNEL_TIME_EXT_BUSY: extension channel busy time was filled in
+ * @SURVEY_INFO_CHANNEL_TIME_RX: channel receive time was filled in
+ * @SURVEY_INFO_CHANNEL_TIME_TX: channel transmit time was filled in
  *
  * Used by the driver to indicate which info in &struct survey_info
  * it has filled in during the get_survey().
@@ -301,6 +306,11 @@ struct key_params {
 enum survey_info_flags {
 	SURVEY_INFO_NOISE_DBM = 1<<0,
 	SURVEY_INFO_IN_USE = 1<<1,
+	SURVEY_INFO_CHANNEL_TIME = 1<<2,
+	SURVEY_INFO_CHANNEL_TIME_BUSY = 1<<3,
+	SURVEY_INFO_CHANNEL_TIME_EXT_BUSY = 1<<4,
+	SURVEY_INFO_CHANNEL_TIME_RX = 1<<5,
+	SURVEY_INFO_CHANNEL_TIME_TX = 1<<6,
 };
 
 /**
@@ -310,6 +320,11 @@ enum survey_info_flags {
  * @filled: bitflag of flags from &enum survey_info_flags
  * @noise: channel noise in dBm. This and all following fields are
  *     optional
+ * @channel_time: amount of time in ms the radio spent on the channel
+ * @channel_time_busy: amount of time the primary channel was sensed busy
+ * @channel_time_ext_busy: amount of time the extension channel was sensed busy
+ * @channel_time_rx: amount of time the radio spent receiving data
+ * @channel_time_tx: amount of time the radio spent transmitting data
  *
  * Used by dump_survey() to report back per-channel survey information.
  *
@@ -318,6 +333,11 @@ enum survey_info_flags {
  */
 struct survey_info {
 	struct ieee80211_channel *channel;
+	u64 channel_time;
+	u64 channel_time_busy;
+	u64 channel_time_ext_busy;
+	u64 channel_time_rx;
+	u64 channel_time_tx;
 	u32 filled;
 	s8 noise;
 };
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 882dc921103b..c506241f8637 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -3153,6 +3153,21 @@ static int nl80211_send_survey(struct sk_buff *msg, u32 pid, u32 seq,
 			    survey->noise);
 	if (survey->filled & SURVEY_INFO_IN_USE)
 		NLA_PUT_FLAG(msg, NL80211_SURVEY_INFO_IN_USE);
+	if (survey->filled & SURVEY_INFO_CHANNEL_TIME)
+		NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME,
+			    survey->channel_time);
+	if (survey->filled & SURVEY_INFO_CHANNEL_TIME_BUSY)
+		NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME_BUSY,
+			    survey->channel_time_busy);
+	if (survey->filled & SURVEY_INFO_CHANNEL_TIME_EXT_BUSY)
+		NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME_EXT_BUSY,
+			    survey->channel_time_ext_busy);
+	if (survey->filled & SURVEY_INFO_CHANNEL_TIME_RX)
+		NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME_RX,
+			    survey->channel_time_rx);
+	if (survey->filled & SURVEY_INFO_CHANNEL_TIME_TX)
+		NLA_PUT_U64(msg, NL80211_SURVEY_INFO_CHANNEL_TIME_TX,
+			    survey->channel_time_tx);
 
 	nla_nest_end(msg, infoattr);
 
-- 
cgit v1.2.3


From cfdfa4d3a0c7aa1287c61326a7714f262466157a Mon Sep 17 00:00:00 2001
From: Steve deRosier <steve@cozybit.com>
Date: Sat, 9 Oct 2010 17:23:28 -0700
Subject: mac80211: Update mesh constants to approved IEEE ANA values

This patch updates IEEE802.11 mesh constants to be consistent with newly
approved values. It modifies some values, as well as adds many new constants
in preparation for updating mesh code to the current 802.11s drafts. ANA
numbers were taken from:
https://mentor.ieee.org/802.11/dcn/09/11-09-0031-12-0000-ana-database-assigned-number-authority.xls

A few notes are in order:
1. This will break backwards compatibility with existing Linux kernels as
over-the-air constants have changed.
2. Some old and obsolete constants have been retained for now as the mesh code
itself hasn't been updated yet to the new 802.11s draft. This was desired to
keep the existing mesh scheme working until it can be updated. Adding the
approved values is the first step in updating the mesh code.
3. Obsolete constants have been clearly marked.
4. All ANA approved 802.11s constants have been added.

Signed-off-by: Steve deRosier <steve@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h | 71 +++++++++++++++++++++++++++++++++++++----------
 1 file changed, 56 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 97b2eae6a22c..ed5a03cbe184 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -986,6 +986,7 @@ struct ieee80211_ht_info {
 #define WLAN_AUTH_OPEN 0
 #define WLAN_AUTH_SHARED_KEY 1
 #define WLAN_AUTH_FT 2
+#define WLAN_AUTH_SAE 3
 #define WLAN_AUTH_LEAP 128
 
 #define WLAN_AUTH_CHALLENGE_LEN 128
@@ -1072,6 +1073,10 @@ enum ieee80211_statuscode {
 	WLAN_STATUS_NO_DIRECT_LINK = 48,
 	WLAN_STATUS_STA_NOT_PRESENT = 49,
 	WLAN_STATUS_STA_NOT_QSTA = 50,
+	/* 802.11s */
+	WLAN_STATUS_ANTI_CLOG_REQUIRED = 76,
+	WLAN_STATUS_FCG_NOT_SUPP = 78,
+	WLAN_STATUS_STA_NO_TBTT = 78,
 };
 
 
@@ -1112,6 +1117,22 @@ enum ieee80211_reasoncode {
 	WLAN_REASON_QSTA_REQUIRE_SETUP = 38,
 	WLAN_REASON_QSTA_TIMEOUT = 39,
 	WLAN_REASON_QSTA_CIPHER_NOT_SUPP = 45,
+	/* 802.11s */
+	WLAN_REASON_MESH_PEER_CANCELED = 52,
+	WLAN_REASON_MESH_MAX_PEERS = 53,
+	WLAN_REASON_MESH_CONFIG = 54,
+	WLAN_REASON_MESH_CLOSE = 55,
+	WLAN_REASON_MESH_MAX_RETRIES = 56,
+	WLAN_REASON_MESH_CONFIRM_TIMEOUT = 57,
+	WLAN_REASON_MESH_INVALID_GTK = 58,
+	WLAN_REASON_MESH_INCONSISTENT_PARAM = 59,
+	WLAN_REASON_MESH_INVALID_SECURITY = 60,
+	WLAN_REASON_MESH_PATH_ERROR = 61,
+	WLAN_REASON_MESH_PATH_NOFORWARD = 62,
+	WLAN_REASON_MESH_PATH_DEST_UNREACHABLE = 63,
+	WLAN_REASON_MAC_EXISTS_IN_MBSS = 64,
+	WLAN_REASON_MESH_CHAN_REGULATORY = 65,
+	WLAN_REASON_MESH_CHAN = 66,
 };
 
 
@@ -1139,20 +1160,33 @@ enum ieee80211_eid {
 	WLAN_EID_TS_DELAY = 43,
 	WLAN_EID_TCLAS_PROCESSING = 44,
 	WLAN_EID_QOS_CAPA = 46,
-	/* 802.11s
-	 *
-	 * All mesh EID numbers are pending IEEE 802.11 ANA approval.
-	 * The numbers have been incremented from those suggested in
-	 * 802.11s/D2.0 so that MESH_CONFIG does not conflict with
-	 * EXT_SUPP_RATES.
+	/* 802.11s */
+	WLAN_EID_MESH_CONFIG = 113,
+	WLAN_EID_MESH_ID = 114,
+	WLAN_EID_LINK_METRIC_REPORT = 115,
+	WLAN_EID_CONGESTION_NOTIFICATION = 116,
+	/* Note that the Peer Link IE has been replaced with the similar
+	 * Peer Management IE.  We will keep the former definition until mesh
+	 * code is changed to comply with latest 802.11s drafts.
 	 */
-	WLAN_EID_MESH_CONFIG = 51,
-	WLAN_EID_MESH_ID = 52,
-	WLAN_EID_PEER_LINK = 55,
-	WLAN_EID_PREQ = 68,
-	WLAN_EID_PREP = 69,
-	WLAN_EID_PERR = 70,
-	WLAN_EID_RANN = 49,	/* compatible with FreeBSD */
+	WLAN_EID_PEER_LINK = 55,  /* no longer in 802.11s drafts */
+	WLAN_EID_PEER_MGMT = 117,
+	WLAN_EID_CHAN_SWITCH_PARAM = 118,
+	WLAN_EID_MESH_AWAKE_WINDOW = 119,
+	WLAN_EID_BEACON_TIMING = 120,
+	WLAN_EID_MCCAOP_SETUP_REQ = 121,
+	WLAN_EID_MCCAOP_SETUP_RESP = 122,
+	WLAN_EID_MCCAOP_ADVERT = 123,
+	WLAN_EID_MCCAOP_TEARDOWN = 124,
+	WLAN_EID_GANN = 125,
+	WLAN_EID_RANN = 126,
+	WLAN_EID_PREQ = 130,
+	WLAN_EID_PREP = 131,
+	WLAN_EID_PERR = 132,
+	WLAN_EID_PXU = 137,
+	WLAN_EID_PXUC = 138,
+	WLAN_EID_AUTH_MESH_PEER_EXCH = 139,
+	WLAN_EID_MIC = 140,
 
 	WLAN_EID_PWR_CONSTRAINT = 32,
 	WLAN_EID_PWR_CAPABILITY = 33,
@@ -1211,9 +1245,14 @@ enum ieee80211_category {
 	WLAN_CATEGORY_HT = 7,
 	WLAN_CATEGORY_SA_QUERY = 8,
 	WLAN_CATEGORY_PROTECTED_DUAL_OF_ACTION = 9,
+	WLAN_CATEGORY_MESH_ACTION = 13,
+	WLAN_CATEGORY_MULTIHOP_ACTION = 14,
+	WLAN_CATEGORY_SELF_PROTECTED = 15,
 	WLAN_CATEGORY_WMM = 17,
-	WLAN_CATEGORY_MESH_PLINK = 30,		/* Pending ANA approval */
-	WLAN_CATEGORY_MESH_PATH_SEL = 32,	/* Pending ANA approval */
+	/* TODO: remove MESH_PLINK and MESH_PATH_SEL after */
+	/*       mesh is updated to current 802.11s draft  */
+	WLAN_CATEGORY_MESH_PLINK = 30,
+	WLAN_CATEGORY_MESH_PATH_SEL = 32,
 	WLAN_CATEGORY_VENDOR_SPECIFIC_PROTECTED = 126,
 	WLAN_CATEGORY_VENDOR_SPECIFIC = 127,
 };
@@ -1351,6 +1390,8 @@ enum ieee80211_sa_query_action {
 /* AKM suite selectors */
 #define WLAN_AKM_SUITE_8021X		0x000FAC01
 #define WLAN_AKM_SUITE_PSK		0x000FAC02
+#define WLAN_AKM_SUITE_SAE			0x000FAC08
+#define WLAN_AKM_SUITE_FT_OVER_SAE	0x000FAC09
 
 #define WLAN_MAX_KEY_LEN		32
 
-- 
cgit v1.2.3


From 29e29f27486ed7074df259b3eda8656bb014e9b5 Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@stericsson.com>
Date: Fri, 1 Oct 2010 09:15:41 +0100
Subject: ARM: 6421/1: amba-pl011: add missing ST specific registers

The ST Micro derivates have several extra interesting registers
that we may soon use for something interesting so may just as
well define them in the header.

Signed-off-by: Jonas Aaberg <jonas.aberg@stericsson.com>
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/linux/amba/serial.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/amba/serial.h b/include/linux/amba/serial.h
index e1b634b635f2..6021588ba0a8 100644
--- a/include/linux/amba/serial.h
+++ b/include/linux/amba/serial.h
@@ -32,7 +32,9 @@
 #define UART01x_RSR		0x04	/* Receive status register (Read). */
 #define UART01x_ECR		0x04	/* Error clear register (Write). */
 #define UART010_LCRH		0x08	/* Line control register, high byte. */
+#define ST_UART011_DMAWM	0x08    /* DMA watermark configure register. */
 #define UART010_LCRM		0x0C	/* Line control register, middle byte. */
+#define ST_UART011_TIMEOUT	0x0C    /* Timeout period register. */
 #define UART010_LCRL		0x10	/* Line control register, low byte. */
 #define UART010_CR		0x14	/* Control register. */
 #define UART01x_FR		0x18	/* Flag register (Read only). */
@@ -51,6 +53,15 @@
 #define UART011_MIS		0x40	/* Masked interrupt status. */
 #define UART011_ICR		0x44	/* Interrupt clear register. */
 #define UART011_DMACR		0x48	/* DMA control register. */
+#define ST_UART011_XFCR		0x50	/* XON/XOFF control register. */
+#define ST_UART011_XON1		0x54	/* XON1 register. */
+#define ST_UART011_XON2		0x58	/* XON2 register. */
+#define ST_UART011_XOFF1	0x5C	/* XON1 register. */
+#define ST_UART011_XOFF2	0x60	/* XON2 register. */
+#define ST_UART011_ITCR		0x80	/* Integration test control register. */
+#define ST_UART011_ITIP		0x84	/* Integration test input register. */
+#define ST_UART011_ABCR		0x100	/* Autobaud control register. */
+#define ST_UART011_ABIMSC	0x15C	/* Autobaud interrupt mask/clear register. */
 
 #define UART011_DR_OE		(1 << 11)
 #define UART011_DR_BE		(1 << 10)
-- 
cgit v1.2.3


From 7c5347733dcc4ba0bac0baf86d99fae0561f33b7 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Mon, 11 Oct 2010 18:13:31 -0400
Subject: fanotify: disable fanotify syscalls

This patch disables the fanotify syscalls by just not building them and
letting the cond_syscall() statements in kernel/sys_ni.c redirect them
to sys_ni_syscall().

It was pointed out by Tvrtko Ursulin that the fanotify interface did not
include an explicit prioritization between groups.  This is necessary
for fanotify to be usable for hierarchical storage management software,
as they must get first access to the file, before inotify-like notifiers
see the file.

This feature can be added in an ABI compatible way in the next release
(by using a number of bits in the flags field to carry the info) but it
was suggested by Alan that maybe we should just hold off and do it in
the next cycle, likely with an (new) explicit argument to the syscall.
I don't like this approach best as I know people are already starting to
use the current interface, but Alan is all wise and noone on list backed
me up with just using what we have.  I feel this is needlessly ripping
the rug out from under people at the last minute, but if others think it
needs to be a new argument it might be the best way forward.

Three choices:
Go with what we got (and implement the new feature next cycle).  Add a
new field right now (and implement the new feature next cycle).  Wait
till next cycle to release the ABI (and implement the new feature next
cycle).  This is number 3.

Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/notify/Kconfig    | 2 +-
 include/linux/Kbuild | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/notify/Kconfig b/fs/notify/Kconfig
index 22c629eedd82..b388443c3a09 100644
--- a/fs/notify/Kconfig
+++ b/fs/notify/Kconfig
@@ -3,4 +3,4 @@ config FSNOTIFY
 
 source "fs/notify/dnotify/Kconfig"
 source "fs/notify/inotify/Kconfig"
-source "fs/notify/fanotify/Kconfig"
+#source "fs/notify/fanotify/Kconfig"
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 626b629429ff..4e8ea8c8ec1e 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -118,7 +118,6 @@ header-y += eventpoll.h
 header-y += ext2_fs.h
 header-y += fadvise.h
 header-y += falloc.h
-header-y += fanotify.h
 header-y += fb.h
 header-y += fcntl.h
 header-y += fd.h
-- 
cgit v1.2.3


From e144710b302525de5b90b9c3ba43562458d8957f Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 1 Oct 2010 16:03:45 +0200
Subject: genirq: Distangle irq.h

Move irq_desc and internal functions out of irq.h

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h     | 292 +++---------------------------------------------
 include/linux/irqdesc.h | 171 ++++++++++++++++++++++++++++
 kernel/irq/internals.h  | 100 +++++++++++++++++
 3 files changed, 284 insertions(+), 279 deletions(-)
 create mode 100644 include/linux/irqdesc.h

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 82ed8231394a..f5827abbc034 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -80,7 +80,6 @@ typedef	void (*irq_flow_handler_t)(unsigned int irq,
 # define IRQ_NO_BALANCING_MASK	IRQ_NO_BALANCING
 #endif
 
-struct proc_dir_entry;
 struct msi_desc;
 
 /**
@@ -202,152 +201,36 @@ struct irq_chip {
 #endif
 };
 
-struct timer_rand_state;
-struct irq_2_iommu;
-/**
- * struct irq_desc - interrupt descriptor
- * @irq_data:		per irq and chip data passed down to chip functions
- * @timer_rand_state:	pointer to timer rand state struct
- * @kstat_irqs:		irq stats per cpu
- * @handle_irq:		highlevel irq-events handler [if NULL, __do_IRQ()]
- * @action:		the irq action chain
- * @status:		status information
- * @depth:		disable-depth, for nested irq_disable() calls
- * @wake_depth:		enable depth, for multiple set_irq_wake() callers
- * @irq_count:		stats field to detect stalled irqs
- * @last_unhandled:	aging timer for unhandled count
- * @irqs_unhandled:	stats field for spurious unhandled interrupts
- * @lock:		locking for SMP
- * @pending_mask:	pending rebalanced interrupts
- * @threads_active:	number of irqaction threads currently running
- * @wait_for_threads:	wait queue for sync_irq to wait for threaded handlers
- * @dir:		/proc/irq/ procfs entry
- * @name:		flow handler name for /proc/interrupts output
- */
-struct irq_desc {
-
-#ifdef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
-	struct irq_data		irq_data;
-#else
-	/*
-	 * This union will go away, once we fixed the direct access to
-	 * irq_desc all over the place. The direct fields are a 1:1
-	 * overlay of irq_data.
-	 */
-	union {
-		struct irq_data		irq_data;
-		struct {
-			unsigned int		irq;
-			unsigned int		node;
-			struct irq_chip		*chip;
-			void			*handler_data;
-			void			*chip_data;
-			struct msi_desc		*msi_desc;
-#ifdef CONFIG_SMP
-			cpumask_var_t		affinity;
-#endif
-#ifdef CONFIG_INTR_REMAP
-			struct irq_2_iommu      *irq_2_iommu;
-#endif
-		};
-	};
-#endif
-
-	struct timer_rand_state *timer_rand_state;
-	unsigned int            *kstat_irqs;
-	irq_flow_handler_t	handle_irq;
-	struct irqaction	*action;	/* IRQ action list */
-	unsigned int		status;		/* IRQ status */
-
-	unsigned int		depth;		/* nested irq disables */
-	unsigned int		wake_depth;	/* nested wake enables */
-	unsigned int		irq_count;	/* For detecting broken IRQs */
-	unsigned long		last_unhandled;	/* Aging timer for unhandled count */
-	unsigned int		irqs_unhandled;
-	raw_spinlock_t		lock;
-#ifdef CONFIG_SMP
-	const struct cpumask	*affinity_hint;
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_var_t		pending_mask;
-#endif
-#endif
-	atomic_t		threads_active;
-	wait_queue_head_t       wait_for_threads;
-#ifdef CONFIG_PROC_FS
-	struct proc_dir_entry	*dir;
-#endif
-	const char		*name;
-} ____cacheline_internodealigned_in_smp;
-
-extern void arch_init_copy_chip_data(struct irq_desc *old_desc,
-					struct irq_desc *desc, int node);
-extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc);
-
-#ifndef CONFIG_SPARSE_IRQ
-extern struct irq_desc irq_desc[NR_IRQS];
-#endif
-
-#ifdef CONFIG_NUMA_IRQ_DESC
-extern struct irq_desc *move_irq_desc(struct irq_desc *old_desc, int node);
-#else
-static inline struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
-{
-	return desc;
-}
-#endif
-
-extern struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node);
+/* This include will go away once we isolated irq_desc usage to core code */
+#include <linux/irqdesc.h>
 
 /*
  * Pick up the arch-dependent methods:
  */
 #include <asm/hw_irq.h>
 
+struct irqaction;
 extern int setup_irq(unsigned int irq, struct irqaction *new);
 extern void remove_irq(unsigned int irq, struct irqaction *act);
 
 #ifdef CONFIG_GENERIC_HARDIRQS
 
 #ifdef CONFIG_SMP
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-
+# ifdef CONFIG_GENERIC_PENDING_IRQ
 void move_native_irq(int irq);
 void move_masked_irq(int irq);
-
-#else /* CONFIG_GENERIC_PENDING_IRQ */
-
-static inline void move_irq(int irq)
-{
-}
-
-static inline void move_native_irq(int irq)
-{
-}
-
-static inline void move_masked_irq(int irq)
-{
-}
-
-#endif /* CONFIG_GENERIC_PENDING_IRQ */
-
-#else /* CONFIG_SMP */
-
-#define move_native_irq(x)
-#define move_masked_irq(x)
-
-#endif /* CONFIG_SMP */
+# else
+static inline void move_irq(int irq) { }
+static inline void move_native_irq(int irq) { }
+static inline void move_masked_irq(int irq) { }
+# endif
+#else
+static inline void move_native_irq(int irq) { }
+static inline void move_masked_irq(int irq) { }
+#endif
 
 extern int no_irq_affinity;
 
-static inline int irq_balancing_disabled(unsigned int irq)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-	return desc->status & IRQ_NO_BALANCING_MASK;
-}
-
 /* Handle irq action chains: */
 extern irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action);
 
@@ -363,42 +246,10 @@ extern void handle_percpu_irq(unsigned int irq, struct irq_desc *desc);
 extern void handle_bad_irq(unsigned int irq, struct irq_desc *desc);
 extern void handle_nested_irq(unsigned int irq);
 
-/*
- * Monolithic do_IRQ implementation.
- */
-#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
-extern unsigned int __do_IRQ(unsigned int irq);
-#endif
-
-/*
- * Architectures call this to let the generic IRQ layer
- * handle an interrupt. If the descriptor is attached to an
- * irqchip-style controller then we call the ->handle_irq() handler,
- * and it calls __do_IRQ() if it's attached to an irqtype-style controller.
- */
-static inline void generic_handle_irq_desc(unsigned int irq, struct irq_desc *desc)
-{
-#ifdef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
-	desc->handle_irq(irq, desc);
-#else
-	if (likely(desc->handle_irq))
-		desc->handle_irq(irq, desc);
-	else
-		__do_IRQ(irq);
-#endif
-}
-
-static inline void generic_handle_irq(unsigned int irq)
-{
-	generic_handle_irq_desc(irq, irq_to_desc(irq));
-}
-
 /* Handling of unhandled and spurious interrupts: */
 extern void note_interrupt(unsigned int irq, struct irq_desc *desc,
 			   irqreturn_t action_ret);
 
-/* Resending of interrupts :*/
-void check_irq_resend(struct irq_desc *desc, unsigned int irq);
 
 /* Enable/disable irq debugging output: */
 extern int noirqdebug_setup(char *str);
@@ -421,16 +272,6 @@ extern void
 __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained,
 		  const char *name);
 
-/* caller has locked the irq_desc and both params are valid */
-static inline void __set_irq_handler_unlocked(int irq,
-					      irq_flow_handler_t handler)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-	desc->handle_irq = handler;
-}
-
 /*
  * Set a highlevel flow handler for a given IRQ:
  */
@@ -462,13 +303,6 @@ extern unsigned int create_irq_nr(unsigned int irq_want, int node);
 extern int create_irq(void);
 extern void destroy_irq(unsigned int irq);
 
-/* Test to see if a driver has successfully requested an irq */
-static inline int irq_has_action(unsigned int irq)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-	return desc->action != NULL;
-}
-
 /* Dynamic irq helper functions */
 extern void dynamic_irq_init(unsigned int irq);
 void dynamic_irq_init_keep_chip_data(unsigned int irq);
@@ -487,108 +321,8 @@ extern int set_irq_msi(unsigned int irq, struct msi_desc *entry);
 #define get_irq_data(irq)	(irq_to_desc(irq)->irq_data.handler_data)
 #define get_irq_msi(irq)	(irq_to_desc(irq)->irq_data.msi_desc)
 
-#define get_irq_desc_chip(desc)		((desc)->irq_data.chip)
-#define get_irq_desc_chip_data(desc)	((desc)->irq_data.chip_data)
-#define get_irq_desc_data(desc)		((desc)->irq_data.handler_data)
-#define get_irq_desc_msi(desc)		((desc)->irq_data.msi_desc)
-
 #endif /* CONFIG_GENERIC_HARDIRQS */
 
 #endif /* !CONFIG_S390 */
 
-#ifdef CONFIG_SMP
-/**
- * alloc_desc_masks - allocate cpumasks for irq_desc
- * @desc:	pointer to irq_desc struct
- * @node:	node which will be handling the cpumasks
- * @boot:	true if need bootmem
- *
- * Allocates affinity and pending_mask cpumask if required.
- * Returns true if successful (or not required).
- */
-static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
-							bool boot)
-{
-	gfp_t gfp = GFP_ATOMIC;
-
-	if (boot)
-		gfp = GFP_NOWAIT;
-
-#ifdef CONFIG_CPUMASK_OFFSTACK
-	if (!alloc_cpumask_var_node(&desc->irq_data.affinity, gfp, node))
-		return false;
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	if (!alloc_cpumask_var_node(&desc->pending_mask, gfp, node)) {
-		free_cpumask_var(desc->irq_data.affinity);
-		return false;
-	}
-#endif
-#endif
-	return true;
-}
-
-static inline void init_desc_masks(struct irq_desc *desc)
-{
-	cpumask_setall(desc->irq_data.affinity);
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_clear(desc->pending_mask);
-#endif
-}
-
-/**
- * init_copy_desc_masks - copy cpumasks for irq_desc
- * @old_desc:	pointer to old irq_desc struct
- * @new_desc:	pointer to new irq_desc struct
- *
- * Insures affinity and pending_masks are copied to new irq_desc.
- * If !CONFIG_CPUMASKS_OFFSTACK the cpumasks are embedded in the
- * irq_desc struct so the copy is redundant.
- */
-
-static inline void init_copy_desc_masks(struct irq_desc *old_desc,
-					struct irq_desc *new_desc)
-{
-#ifdef CONFIG_CPUMASK_OFFSTACK
-	cpumask_copy(new_desc->irq_data.affinity, old_desc->irq_data.affinity);
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_copy(new_desc->pending_mask, old_desc->pending_mask);
-#endif
-#endif
-}
-
-static inline void free_desc_masks(struct irq_desc *old_desc,
-				   struct irq_desc *new_desc)
-{
-	free_cpumask_var(old_desc->irq_data.affinity);
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	free_cpumask_var(old_desc->pending_mask);
-#endif
-}
-
-#else /* !CONFIG_SMP */
-
-static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
-								bool boot)
-{
-	return true;
-}
-
-static inline void init_desc_masks(struct irq_desc *desc)
-{
-}
-
-static inline void init_copy_desc_masks(struct irq_desc *old_desc,
-					struct irq_desc *new_desc)
-{
-}
-
-static inline void free_desc_masks(struct irq_desc *old_desc,
-				   struct irq_desc *new_desc)
-{
-}
-#endif	/* CONFIG_SMP */
-
 #endif /* _LINUX_IRQ_H */
diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h
new file mode 100644
index 000000000000..22e426fdd301
--- /dev/null
+++ b/include/linux/irqdesc.h
@@ -0,0 +1,171 @@
+#ifndef _LINUX_IRQDESC_H
+#define _LINUX_IRQDESC_H
+
+/*
+ * Core internal functions to deal with irq descriptors
+ *
+ * This include will move to kernel/irq once we cleaned up the tree.
+ * For now it's included from <linux/irq.h>
+ */
+
+struct proc_dir_entry;
+struct timer_rand_state;
+struct irq_2_iommu;
+/**
+ * struct irq_desc - interrupt descriptor
+ * @irq_data:		per irq and chip data passed down to chip functions
+ * @timer_rand_state:	pointer to timer rand state struct
+ * @kstat_irqs:		irq stats per cpu
+ * @handle_irq:		highlevel irq-events handler [if NULL, __do_IRQ()]
+ * @action:		the irq action chain
+ * @status:		status information
+ * @depth:		disable-depth, for nested irq_disable() calls
+ * @wake_depth:		enable depth, for multiple set_irq_wake() callers
+ * @irq_count:		stats field to detect stalled irqs
+ * @last_unhandled:	aging timer for unhandled count
+ * @irqs_unhandled:	stats field for spurious unhandled interrupts
+ * @lock:		locking for SMP
+ * @pending_mask:	pending rebalanced interrupts
+ * @threads_active:	number of irqaction threads currently running
+ * @wait_for_threads:	wait queue for sync_irq to wait for threaded handlers
+ * @dir:		/proc/irq/ procfs entry
+ * @name:		flow handler name for /proc/interrupts output
+ */
+struct irq_desc {
+
+#ifdef CONFIG_GENERIC_HARDIRQS_NO_DEPRECATED
+	struct irq_data		irq_data;
+#else
+	/*
+	 * This union will go away, once we fixed the direct access to
+	 * irq_desc all over the place. The direct fields are a 1:1
+	 * overlay of irq_data.
+	 */
+	union {
+		struct irq_data		irq_data;
+		struct {
+			unsigned int		irq;
+			unsigned int		node;
+			struct irq_chip		*chip;
+			void			*handler_data;
+			void			*chip_data;
+			struct msi_desc		*msi_desc;
+#ifdef CONFIG_SMP
+			cpumask_var_t		affinity;
+#endif
+#ifdef CONFIG_INTR_REMAP
+			struct irq_2_iommu      *irq_2_iommu;
+#endif
+		};
+	};
+#endif
+
+	struct timer_rand_state *timer_rand_state;
+	unsigned int		*kstat_irqs;
+	irq_flow_handler_t	handle_irq;
+	struct irqaction	*action;	/* IRQ action list */
+	unsigned int		status;		/* IRQ status */
+
+	unsigned int		depth;		/* nested irq disables */
+	unsigned int		wake_depth;	/* nested wake enables */
+	unsigned int		irq_count;	/* For detecting broken IRQs */
+	unsigned long		last_unhandled;	/* Aging timer for unhandled count */
+	unsigned int		irqs_unhandled;
+	raw_spinlock_t		lock;
+#ifdef CONFIG_SMP
+	const struct cpumask	*affinity_hint;
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	cpumask_var_t		pending_mask;
+#endif
+#endif
+	atomic_t		threads_active;
+	wait_queue_head_t       wait_for_threads;
+#ifdef CONFIG_PROC_FS
+	struct proc_dir_entry	*dir;
+#endif
+	const char		*name;
+} ____cacheline_internodealigned_in_smp;
+
+extern void arch_init_copy_chip_data(struct irq_desc *old_desc,
+					struct irq_desc *desc, int node);
+extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc);
+
+#ifndef CONFIG_SPARSE_IRQ
+extern struct irq_desc irq_desc[NR_IRQS];
+#endif
+
+#ifdef CONFIG_NUMA_IRQ_DESC
+extern struct irq_desc *move_irq_desc(struct irq_desc *old_desc, int node);
+#else
+static inline struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
+{
+	return desc;
+}
+#endif
+
+extern struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node);
+
+#ifdef CONFIG_GENERIC_HARDIRQS
+
+#define get_irq_desc_chip(desc)		((desc)->irq_data.chip)
+#define get_irq_desc_chip_data(desc)	((desc)->irq_data.chip_data)
+#define get_irq_desc_data(desc)		((desc)->irq_data.handler_data)
+#define get_irq_desc_msi(desc)		((desc)->irq_data.msi_desc)
+
+/*
+ * Monolithic do_IRQ implementation.
+ */
+#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
+extern unsigned int __do_IRQ(unsigned int irq);
+#endif
+
+/*
+ * Architectures call this to let the generic IRQ layer
+ * handle an interrupt. If the descriptor is attached to an
+ * irqchip-style controller then we call the ->handle_irq() handler,
+ * and it calls __do_IRQ() if it's attached to an irqtype-style controller.
+ */
+static inline void generic_handle_irq_desc(unsigned int irq, struct irq_desc *desc)
+{
+#ifdef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
+	desc->handle_irq(irq, desc);
+#else
+	if (likely(desc->handle_irq))
+		desc->handle_irq(irq, desc);
+	else
+		__do_IRQ(irq);
+#endif
+}
+
+static inline void generic_handle_irq(unsigned int irq)
+{
+	generic_handle_irq_desc(irq, irq_to_desc(irq));
+}
+
+/* Test to see if a driver has successfully requested an irq */
+static inline int irq_has_action(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+	return desc->action != NULL;
+}
+
+static inline int irq_balancing_disabled(unsigned int irq)
+{
+	struct irq_desc *desc;
+
+	desc = irq_to_desc(irq);
+	return desc->status & IRQ_NO_BALANCING_MASK;
+}
+
+/* caller has locked the irq_desc and both params are valid */
+static inline void __set_irq_handler_unlocked(int irq,
+					      irq_flow_handler_t handler)
+{
+	struct irq_desc *desc;
+
+	desc = irq_to_desc(irq);
+	desc->handle_irq = handler;
+}
+#endif
+
+#endif
diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h
index b905f0ab1bb2..e281e45fbb55 100644
--- a/kernel/irq/internals.h
+++ b/kernel/irq/internals.h
@@ -1,6 +1,7 @@
 /*
  * IRQ subsystem internal functions and variables:
  */
+#include <linux/irqdesc.h>
 
 extern int noirqdebug;
 
@@ -22,6 +23,9 @@ extern void init_kstat_irqs(struct irq_desc *desc, int node, int nr);
 extern void clear_kstat_irqs(struct irq_desc *desc);
 extern raw_spinlock_t sparse_irq_lock;
 
+/* Resending of interrupts :*/
+void check_irq_resend(struct irq_desc *desc, unsigned int irq);
+
 #ifdef CONFIG_SPARSE_IRQ
 void replace_irq_desc(unsigned int irq, struct irq_desc *desc);
 #endif
@@ -105,3 +109,99 @@ static inline void print_irq_desc(unsigned int irq, struct irq_desc *desc)
 
 #undef P
 
+/* Stuff below will be cleaned up after the sparse allocator is done */
+
+#ifdef CONFIG_SMP
+/**
+ * alloc_desc_masks - allocate cpumasks for irq_desc
+ * @desc:	pointer to irq_desc struct
+ * @node:	node which will be handling the cpumasks
+ * @boot:	true if need bootmem
+ *
+ * Allocates affinity and pending_mask cpumask if required.
+ * Returns true if successful (or not required).
+ */
+static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
+							bool boot)
+{
+	gfp_t gfp = GFP_ATOMIC;
+
+	if (boot)
+		gfp = GFP_NOWAIT;
+
+#ifdef CONFIG_CPUMASK_OFFSTACK
+	if (!alloc_cpumask_var_node(&desc->irq_data.affinity, gfp, node))
+		return false;
+
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	if (!alloc_cpumask_var_node(&desc->pending_mask, gfp, node)) {
+		free_cpumask_var(desc->irq_data.affinity);
+		return false;
+	}
+#endif
+#endif
+	return true;
+}
+
+static inline void init_desc_masks(struct irq_desc *desc)
+{
+	cpumask_setall(desc->irq_data.affinity);
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	cpumask_clear(desc->pending_mask);
+#endif
+}
+
+/**
+ * init_copy_desc_masks - copy cpumasks for irq_desc
+ * @old_desc:	pointer to old irq_desc struct
+ * @new_desc:	pointer to new irq_desc struct
+ *
+ * Insures affinity and pending_masks are copied to new irq_desc.
+ * If !CONFIG_CPUMASKS_OFFSTACK the cpumasks are embedded in the
+ * irq_desc struct so the copy is redundant.
+ */
+
+static inline void init_copy_desc_masks(struct irq_desc *old_desc,
+					struct irq_desc *new_desc)
+{
+#ifdef CONFIG_CPUMASK_OFFSTACK
+	cpumask_copy(new_desc->irq_data.affinity, old_desc->irq_data.affinity);
+
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	cpumask_copy(new_desc->pending_mask, old_desc->pending_mask);
+#endif
+#endif
+}
+
+static inline void free_desc_masks(struct irq_desc *old_desc,
+				   struct irq_desc *new_desc)
+{
+	free_cpumask_var(old_desc->irq_data.affinity);
+
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	free_cpumask_var(old_desc->pending_mask);
+#endif
+}
+
+#else /* !CONFIG_SMP */
+
+static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
+								bool boot)
+{
+	return true;
+}
+
+static inline void init_desc_masks(struct irq_desc *desc)
+{
+}
+
+static inline void init_copy_desc_masks(struct irq_desc *old_desc,
+					struct irq_desc *new_desc)
+{
+}
+
+static inline void free_desc_masks(struct irq_desc *old_desc,
+				   struct irq_desc *new_desc)
+{
+}
+#endif	/* CONFIG_SMP */
-- 
cgit v1.2.3


From 3a3856d00c74560a7b8d9f8a13c1ca94ee786b78 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 4 Oct 2010 13:47:12 +0200
Subject: genirq: Remove unsused inline

move_irq() has no users. Remove it and simplify the ifdef forrest while at it.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index f5827abbc034..80fdab208c13 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -215,15 +215,9 @@ extern void remove_irq(unsigned int irq, struct irqaction *act);
 
 #ifdef CONFIG_GENERIC_HARDIRQS
 
-#ifdef CONFIG_SMP
-# ifdef CONFIG_GENERIC_PENDING_IRQ
+#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_PENDING_IRQ)
 void move_native_irq(int irq);
 void move_masked_irq(int irq);
-# else
-static inline void move_irq(int irq) { }
-static inline void move_native_irq(int irq) { }
-static inline void move_masked_irq(int irq) { }
-# endif
 #else
 static inline void move_native_irq(int irq) { }
 static inline void move_masked_irq(int irq) { }
-- 
cgit v1.2.3


From 442471848f5abb55b99cba1229301655f67492b4 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Sep 2010 10:40:18 +0200
Subject: genirq: Provide status modifier

Provide a irq_desc.status modifier function to cleanup the direct
access to irq_desc in arch and driver code.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h | 27 +++++++++++++++++++++++++--
 kernel/irq/chip.c   | 26 +++++++-------------------
 2 files changed, 32 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 80fdab208c13..e7e7ac83edd8 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -72,6 +72,10 @@ typedef	void (*irq_flow_handler_t)(unsigned int irq,
 #define IRQ_ONESHOT		0x08000000	/* IRQ is not unmasked after hardirq */
 #define IRQ_NESTED_THREAD	0x10000000	/* IRQ is nested into another, no own handler thread */
 
+#define IRQF_MODIFY_MASK	\
+	(IRQ_TYPE_SENSE_MASK | IRQ_NOPROBE | IRQ_NOREQUEST | \
+	 IRQ_NOAUTOEN | IRQ_MOVE_PCNTXT | IRQ_LEVEL)
+
 #ifdef CONFIG_IRQ_PER_CPU
 # define CHECK_IRQ_PER_CPU(var) ((var) & IRQ_PER_CPU)
 # define IRQ_NO_BALANCING_MASK	(IRQ_PER_CPU | IRQ_NO_BALANCING)
@@ -289,8 +293,27 @@ set_irq_chained_handler(unsigned int irq,
 
 extern void set_irq_nested_thread(unsigned int irq, int nest);
 
-extern void set_irq_noprobe(unsigned int irq);
-extern void set_irq_probe(unsigned int irq);
+void irq_modify_status(unsigned int irq, unsigned long clr, unsigned long set);
+
+static inline void irq_set_status_flags(unsigned int irq, unsigned long set)
+{
+	irq_modify_status(irq, 0, set);
+}
+
+static inline void irq_clear_status_flags(unsigned int irq, unsigned long clr)
+{
+	irq_modify_status(irq, clr, 0);
+}
+
+static inline void set_irq_noprobe(unsigned int irq)
+{
+	irq_modify_status(irq, 0, IRQ_NOPROBE);
+}
+
+static inline void set_irq_probe(unsigned int irq)
+{
+	irq_modify_status(irq, IRQ_NOPROBE, 0);
+}
 
 /* Handle dynamic irq creation and destruction */
 extern unsigned int create_irq_nr(unsigned int irq_want, int node);
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 323547983f15..2b1f6906b824 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -851,32 +851,20 @@ set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip,
 	__set_irq_handler(irq, handle, 0, name);
 }
 
-void set_irq_noprobe(unsigned int irq)
+void irq_modify_status(unsigned int irq, unsigned long clr, unsigned long set)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	unsigned long flags;
 
-	if (!desc) {
-		printk(KERN_ERR "Trying to mark IRQ%d non-probeable\n", irq);
+	if (!desc)
 		return;
-	}
-
-	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->status |= IRQ_NOPROBE;
-	raw_spin_unlock_irqrestore(&desc->lock, flags);
-}
 
-void set_irq_probe(unsigned int irq)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-	unsigned long flags;
-
-	if (!desc) {
-		printk(KERN_ERR "Trying to mark IRQ%d probeable\n", irq);
-		return;
-	}
+	/* Sanitize flags */
+	set &= IRQF_MODIFY_MASK;
+	clr &= IRQF_MODIFY_MASK;
 
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->status &= ~IRQ_NOPROBE;
+	desc->status &= ~clr;
+	desc->status |= set;
 	raw_spin_unlock_irqrestore(&desc->lock, flags);
 }
-- 
cgit v1.2.3


From f303a6dd127b5ec6de90d1cd79ed19820c7e9658 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Sep 2010 17:34:01 +0200
Subject: genirq: Sanitize irq_data accessors

Get the data structure from the core and provide inline wrappers to
access the irq_data members.

Provide accessor inlines for irq_data as well.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h | 62 +++++++++++++++++++++++++++++++++++++++++++++++++----
 kernel/irq/chip.c   |  8 +++++++
 2 files changed, 66 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index e7e7ac83edd8..bea40556c5a6 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -85,6 +85,7 @@ typedef	void (*irq_flow_handler_t)(unsigned int irq,
 #endif
 
 struct msi_desc;
+struct irq_2_iommu;
 
 /**
  * struct irq_data - per irq and irq chip data passed down to chip functions
@@ -332,11 +333,64 @@ extern int set_irq_data(unsigned int irq, void *data);
 extern int set_irq_chip_data(unsigned int irq, void *data);
 extern int set_irq_type(unsigned int irq, unsigned int type);
 extern int set_irq_msi(unsigned int irq, struct msi_desc *entry);
+extern struct irq_data *irq_get_irq_data(unsigned int irq);
 
-#define get_irq_chip(irq)	(irq_to_desc(irq)->irq_data.chip)
-#define get_irq_chip_data(irq)	(irq_to_desc(irq)->irq_data.chip_data)
-#define get_irq_data(irq)	(irq_to_desc(irq)->irq_data.handler_data)
-#define get_irq_msi(irq)	(irq_to_desc(irq)->irq_data.msi_desc)
+static inline struct irq_chip *get_irq_chip(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->chip : NULL;
+}
+
+static inline struct irq_chip *irq_data_get_irq_chip(struct irq_data *d)
+{
+	return d->chip;
+}
+
+static inline void *get_irq_chip_data(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->chip_data : NULL;
+}
+
+static inline void *irq_data_get_irq_chip_data(struct irq_data *d)
+{
+	return d->chip_data;
+}
+
+static inline void *get_irq_data(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->handler_data : NULL;
+}
+
+static inline void *irq_data_get_irq_data(struct irq_data *d)
+{
+	return d->handler_data;
+}
+
+static inline struct msi_desc *get_irq_msi(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->msi_desc : NULL;
+}
+
+static inline struct msi_desc *irq_data_get_msi(struct irq_data *d)
+{
+	return d->msi_desc;
+}
+
+#ifdef CONFIG_INTR_REMAP
+static inline struct irq_2_iommu *get_irq_iommu(unsigned int irq)
+{
+	struct irq_data *d = irq_get_irq_data(irq);
+	return d ? d->irq_2_iommu : NULL;
+}
+
+static inline struct irq_2_iommu *irq_data_get_iommu(struct irq_data *d)
+{
+	return d->irq_2_iommu;
+}
+#endif
 
 #endif /* CONFIG_GENERIC_HARDIRQS */
 
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 2b1f6906b824..659be326c8e8 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -256,6 +256,14 @@ int set_irq_chip_data(unsigned int irq, void *data)
 }
 EXPORT_SYMBOL(set_irq_chip_data);
 
+struct irq_data *irq_get_irq_data(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+
+	return desc ? &desc->irq_data : NULL;
+}
+EXPORT_SYMBOL_GPL(irq_get_irq_data);
+
 /**
  *	set_irq_nested_thread - Set/Reset the IRQ_NESTED_THREAD flag of an irq
  *
-- 
cgit v1.2.3


From 154cd387cdf0e5566ce523cbddf92dd2a062dfd6 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 22 Sep 2010 15:58:45 +0200
Subject: genirq: Remove early_init_irq_lock_class()

early_init_irq_lock_class() is called way before anything touches the
irq descriptors. In case of SPARSE_IRQ=y this is a NOP operation
because the radix tree is empty at this point. For the SPARSE_IRQ=n
case it's sufficient to set the lock class in early_init_irq().

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/lockdep.h |  8 --------
 init/main.c             |  1 -
 kernel/irq/irqdesc.c    | 11 +----------
 3 files changed, 1 insertion(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 06aed8305bf3..17d050ce7ab8 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -424,14 +424,6 @@ do {								\
 
 #endif /* CONFIG_LOCKDEP */
 
-#ifdef CONFIG_GENERIC_HARDIRQS
-extern void early_init_irq_lock_class(void);
-#else
-static inline void early_init_irq_lock_class(void)
-{
-}
-#endif
-
 #ifdef CONFIG_TRACE_IRQFLAGS
 extern void early_boot_irqs_off(void);
 extern void early_boot_irqs_on(void);
diff --git a/init/main.c b/init/main.c
index 94ab488039aa..9684c9670b48 100644
--- a/init/main.c
+++ b/init/main.c
@@ -556,7 +556,6 @@ asmlinkage void __init start_kernel(void)
 
 	local_irq_disable();
 	early_boot_irqs_off();
-	early_init_irq_lock_class();
 
 /*
  * Interrupts are still disabled. Do necessary setups, then
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index fbf8cfa00510..0a7a0908afbc 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -231,6 +231,7 @@ int __init early_irq_init(void)
 		alloc_desc_masks(&desc[i], 0, true);
 		init_desc_masks(&desc[i]);
 		desc[i].kstat_irqs = kstat_irqs_all[i];
+		lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
 	}
 	return arch_early_irq_init();
 }
@@ -251,16 +252,6 @@ void clear_kstat_irqs(struct irq_desc *desc)
 	memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
 }
 
-void early_init_irq_lock_class(void)
-{
-	struct irq_desc *desc;
-	int i;
-
-	for_each_irq_desc(i, desc) {
-		lockdep_set_class(&desc->lock, &irq_desc_lock_class);
-	}
-}
-
 unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-- 
cgit v1.2.3


From 1318a481fc37c503a901b96ae06b692ca2b21af5 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 27 Sep 2010 21:01:37 +0200
Subject: genirq: Provide default irq init flags

Arch code sets it's own irq_desc.status flags right after boot and for
dynamically allocated interrupts. That might involve iterating over a
huge array.

Allow ARCH_IRQ_INIT_FLAGS to set separate flags aside of IRQ_DISABLED
which is the default.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h  | 6 ++++++
 kernel/irq/chip.c    | 2 +-
 kernel/irq/irqdesc.c | 6 +++---
 3 files changed, 10 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index bea40556c5a6..30a300991ed4 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -214,6 +214,12 @@ struct irq_chip {
  */
 #include <asm/hw_irq.h>
 
+#ifndef ARCH_IRQ_INIT_FLAGS
+# define ARCH_IRQ_INIT_FLAGS	0
+#endif
+
+#define IRQ_DEFAULT_INIT_FLAGS	(IRQ_DISABLED | ARCH_IRQ_INIT_FLAGS)
+
 struct irqaction;
 extern int setup_irq(unsigned int irq, struct irqaction *new);
 extern void remove_irq(unsigned int irq, struct irqaction *act);
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 659be326c8e8..3405761d6224 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -31,7 +31,7 @@ static void dynamic_irq_init_x(unsigned int irq, bool keep_chip_data)
 
 	/* Ensure we don't have left over values from a previous use of this irq */
 	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->status = IRQ_DISABLED;
+	desc->status = IRQ_DEFAULT_INIT_FLAGS;
 	desc->irq_data.chip = &no_irq_chip;
 	desc->handle_irq = handle_bad_irq;
 	desc->depth = 1;
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 78ff426a6cb7..29963f99f24d 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -39,7 +39,7 @@ EXPORT_SYMBOL_GPL(nr_irqs);
 #ifdef CONFIG_SPARSE_IRQ
 
 static struct irq_desc irq_desc_init = {
-	.status		= IRQ_DISABLED,
+	.status		= IRQ_DEFAULT_INIT_FLAGS,
 	.handle_irq	= handle_bad_irq,
 	.depth		= 1,
 	.lock		= __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
@@ -113,7 +113,7 @@ void replace_irq_desc(unsigned int irq, struct irq_desc *desc)
 
 static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_smp = {
 	[0 ... NR_IRQS_LEGACY-1] = {
-		.status		= IRQ_DISABLED,
+		.status		= IRQ_DEFAULT_INIT_FLAGS,
 		.handle_irq	= handle_bad_irq,
 		.depth		= 1,
 		.lock		= __RAW_SPIN_LOCK_UNLOCKED(irq_desc_init.lock),
@@ -204,7 +204,7 @@ out_unlock:
 
 struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
 	[0 ... NR_IRQS-1] = {
-		.status		= IRQ_DISABLED,
+		.status		= IRQ_DEFAULT_INIT_FLAGS,
 		.handle_irq	= handle_bad_irq,
 		.depth		= 1,
 		.lock		= __RAW_SPIN_LOCK_UNLOCKED(irq_desc->lock),
-- 
cgit v1.2.3


From 1f5a5b87f78fade3ae48dfd55e8765d1d622ea4e Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 27 Sep 2010 17:48:26 +0200
Subject: genirq: Implement a sane sparse_irq allocator

The current sparse_irq allocator has several short comings due to
failures in the design or the lack of it:

 - Requires iteration over the number of active irqs to find a free slot
   (Some architectures have grown their own workarounds for this)
 - Removal of entries is not possible
 - Racy between create_irq_nr and destroy_irq (plugged by horrible
   callbacks)
 - Migration of active irq descriptors is not possible
 - No bulk allocation of irq ranges
 - Sprinkeled irq_desc references all over the place outside of kernel/irq/
   (The previous chip functions series is addressing this issue)

Implement a sane allocator which fixes the above short comings (though
migration of active descriptors needs a full tree wide cleanup of the
direct and mostly unlocked access to irq_desc).

The new allocator still uses a radix_tree, but uses a bitmap for
keeping track of allocated irq numbers. That allows:

 - Fast lookup of a free slot
 - Allows the removal of descriptors
 - Prevents the create/destroy race
 - Bulk allocation of consecutive irq ranges
 - Basic design is ready for migration of life descriptors after
   further cleanups

The bitmap is also used in the SPARSE_IRQ=n case for lookup and
raceless (de)allocation of irq numbers. So it removes the requirement
for looping through the descriptor array to find slots.

Right now it uses sparse_irq_lock to protect the bitmap and the radix
tree, but after cleaning up all users we should be able convert that
to a mutex and to switch the radix_tree and decriptor allocations to
GFP_KERNEL.

[ Folded in a bugfix from Yinghai Lu ]

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h  |  23 +++++
 kernel/irq/irqdesc.c | 231 +++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 246 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 30a300991ed4..cefacf928b33 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -398,6 +398,29 @@ static inline struct irq_2_iommu *irq_data_get_iommu(struct irq_data *d)
 }
 #endif
 
+int irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node);
+void irq_free_descs(unsigned int irq, unsigned int cnt);
+
+static inline int irq_alloc_desc(int node)
+{
+	return irq_alloc_descs(-1, 0, 1, node);
+}
+
+static inline int irq_alloc_desc_at(unsigned int at, int node)
+{
+	return irq_alloc_descs(at, at, 1, node);
+}
+
+static inline int irq_alloc_desc_from(unsigned int from, int node)
+{
+	return irq_alloc_descs(-1, from, 1, node);
+}
+
+static inline void irq_free_desc(unsigned int irq)
+{
+	irq_free_descs(irq, 1);
+}
+
 #endif /* CONFIG_GENERIC_HARDIRQS */
 
 #endif /* !CONFIG_S390 */
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 29963f99f24d..4eea48b4f576 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -13,6 +13,7 @@
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
 #include <linux/radix-tree.h>
+#include <linux/bitmap.h>
 
 #include "internals.h"
 
@@ -33,9 +34,54 @@ static void __init init_irq_default_affinity(void)
 }
 #endif
 
+#ifdef CONFIG_SMP
+static int alloc_masks(struct irq_desc *desc, gfp_t gfp, int node)
+{
+	if (!zalloc_cpumask_var_node(&desc->irq_data.affinity, gfp, node))
+		return -ENOMEM;
+
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	if (!zalloc_cpumask_var_node(&desc->pending_mask, gfp, node)) {
+		free_cpumask_var(desc->irq_data.affinity);
+		return -ENOMEM;
+	}
+#endif
+	return 0;
+}
+
+static void desc_smp_init(struct irq_desc *desc, int node)
+{
+	desc->node = node;
+	cpumask_copy(desc->irq_data.affinity, irq_default_affinity);
+}
+
+#else
+static inline int
+alloc_masks(struct irq_desc *desc, gfp_t gfp, int node) { return 0; }
+static inline void desc_smp_init(struct irq_desc *desc, int node) { }
+#endif
+
+static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node)
+{
+	desc->irq_data.irq = irq;
+	desc->irq_data.chip = &no_irq_chip;
+	desc->irq_data.chip_data = NULL;
+	desc->irq_data.handler_data = NULL;
+	desc->irq_data.msi_desc = NULL;
+	desc->status = IRQ_DEFAULT_INIT_FLAGS;
+	desc->handle_irq = handle_bad_irq;
+	desc->depth = 1;
+	desc->name = NULL;
+	memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
+	desc_smp_init(desc, node);
+}
+
 int nr_irqs = NR_IRQS;
 EXPORT_SYMBOL_GPL(nr_irqs);
 
+DEFINE_RAW_SPINLOCK(sparse_irq_lock);
+static DECLARE_BITMAP(allocated_irqs, NR_IRQS);
+
 #ifdef CONFIG_SPARSE_IRQ
 
 static struct irq_desc irq_desc_init = {
@@ -85,14 +131,9 @@ static void init_one_irq_desc(int irq, struct irq_desc *desc, int node)
 	arch_init_chip_data(desc, node);
 }
 
-/*
- * Protect the sparse_irqs:
- */
-DEFINE_RAW_SPINLOCK(sparse_irq_lock);
-
 static RADIX_TREE(irq_desc_tree, GFP_ATOMIC);
 
-static void set_irq_desc(unsigned int irq, struct irq_desc *desc)
+static void irq_insert_desc(unsigned int irq, struct irq_desc *desc)
 {
 	radix_tree_insert(&irq_desc_tree, irq, desc);
 }
@@ -111,6 +152,94 @@ void replace_irq_desc(unsigned int irq, struct irq_desc *desc)
 		radix_tree_replace_slot(ptr, desc);
 }
 
+static void delete_irq_desc(unsigned int irq)
+{
+	radix_tree_delete(&irq_desc_tree, irq);
+}
+
+#ifdef CONFIG_SMP
+static void free_masks(struct irq_desc *desc)
+{
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	free_cpumask_var(desc->pending_mask);
+#endif
+	free_cpumask_var(desc->affinity);
+}
+#else
+static inline void free_masks(struct irq_desc *desc) { }
+#endif
+
+static struct irq_desc *alloc_desc(int irq, int node)
+{
+	struct irq_desc *desc;
+	gfp_t gfp = GFP_KERNEL;
+
+	desc = kzalloc_node(sizeof(*desc), gfp, node);
+	if (!desc)
+		return NULL;
+	/* allocate based on nr_cpu_ids */
+	desc->kstat_irqs = kzalloc_node(nr_cpu_ids * sizeof(*desc->kstat_irqs),
+					 gfp, node);
+	if (!desc->kstat_irqs)
+		goto err_desc;
+
+	if (alloc_masks(desc, gfp, node))
+		goto err_kstat;
+
+	raw_spin_lock_init(&desc->lock);
+	lockdep_set_class(&desc->lock, &irq_desc_lock_class);
+
+	desc_set_defaults(irq, desc, node);
+
+	return desc;
+
+err_kstat:
+	kfree(desc->kstat_irqs);
+err_desc:
+	kfree(desc);
+	return NULL;
+}
+
+static void free_desc(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
+	delete_irq_desc(irq);
+	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
+
+	free_masks(desc);
+	kfree(desc->kstat_irqs);
+	kfree(desc);
+}
+
+static int alloc_descs(unsigned int start, unsigned int cnt, int node)
+{
+	struct irq_desc *desc;
+	unsigned long flags;
+	int i;
+
+	for (i = 0; i < cnt; i++) {
+		desc = alloc_desc(start + i, node);
+		if (!desc)
+			goto err;
+		raw_spin_lock_irqsave(&sparse_irq_lock, flags);
+		irq_insert_desc(start + i, desc);
+		raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
+	}
+	return start;
+
+err:
+	for (i--; i >= 0; i--)
+		free_desc(start + i);
+
+	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
+	bitmap_clear(allocated_irqs, start, cnt);
+	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
+	return -ENOMEM;
+}
+
 static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_smp = {
 	[0 ... NR_IRQS_LEGACY-1] = {
 		.status		= IRQ_DEFAULT_INIT_FLAGS,
@@ -155,7 +284,7 @@ int __init early_irq_init(void)
 		lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
 		alloc_desc_masks(&desc[i], node, true);
 		init_desc_masks(&desc[i]);
-		set_irq_desc(i, &desc[i]);
+		irq_insert_desc(i, &desc[i]);
 	}
 
 	return arch_early_irq_init();
@@ -192,7 +321,7 @@ struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node)
 	}
 	init_one_irq_desc(irq, desc, node);
 
-	set_irq_desc(irq, desc);
+	irq_insert_desc(irq, desc);
 
 out_unlock:
 	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
@@ -245,8 +374,94 @@ struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node)
 {
 	return irq_to_desc(irq);
 }
+
+#ifdef CONFIG_SMP
+static inline int desc_node(struct irq_desc *desc)
+{
+	return desc->irq_data.node;
+}
+#else
+static inline int desc_node(struct irq_desc *desc) { return 0; }
+#endif
+
+static void free_desc(unsigned int irq)
+{
+	struct irq_desc *desc = irq_to_desc(irq);
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&desc->lock, flags);
+	desc_set_defaults(irq, desc, desc_node(desc));
+	raw_spin_unlock_irqrestore(&desc->lock, flags);
+}
+
+static inline int alloc_descs(unsigned int start, unsigned int cnt, int node)
+{
+	return start;
+}
 #endif /* !CONFIG_SPARSE_IRQ */
 
+/* Dynamic interrupt handling */
+
+/**
+ * irq_free_descs - free irq descriptors
+ * @from:	Start of descriptor range
+ * @cnt:	Number of consecutive irqs to free
+ */
+void irq_free_descs(unsigned int from, unsigned int cnt)
+{
+	unsigned long flags;
+	int i;
+
+	if (from >= nr_irqs || (from + cnt) > nr_irqs)
+		return;
+
+	for (i = 0; i < cnt; i++)
+		free_desc(from + i);
+
+	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
+	bitmap_clear(allocated_irqs, from, cnt);
+	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
+}
+
+/**
+ * irq_alloc_descs - allocate and initialize a range of irq descriptors
+ * @irq:	Allocate for specific irq number if irq >= 0
+ * @from:	Start the search from this irq number
+ * @cnt:	Number of consecutive irqs to allocate.
+ * @node:	Preferred node on which the irq descriptor should be allocated
+ *
+ * Returns the first irq number or error code
+ */
+int __ref
+irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node)
+{
+	unsigned long flags;
+	int start, ret;
+
+	if (!cnt)
+		return -EINVAL;
+
+	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
+
+	start = bitmap_find_next_zero_area(allocated_irqs, nr_irqs, from, cnt, 0);
+	ret = -EEXIST;
+	if (irq >=0 && start != irq)
+		goto err;
+
+	ret = -ENOMEM;
+	if (start >= nr_irqs)
+		goto err;
+
+	bitmap_set(allocated_irqs, start, cnt);
+	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
+	return alloc_descs(start, cnt, node);
+
+err:
+	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
+	return ret;
+}
+
+/* Statistics access */
 void clear_kstat_irqs(struct irq_desc *desc)
 {
 	memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
-- 
cgit v1.2.3


From a98d24b71b6e229965f18dc00d28dc71cb8fe324 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 30 Sep 2010 10:45:07 +0200
Subject: genirq: Implement sane enumeration

Use the allocator bitmap to lookup active interrupts.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irqnr.h |  5 +++++
 kernel/irq/irqdesc.c  | 11 +++++++++++
 2 files changed, 16 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h
index 7bf89bc8cbca..05aa8c23483f 100644
--- a/include/linux/irqnr.h
+++ b/include/linux/irqnr.h
@@ -25,6 +25,7 @@
 
 extern int nr_irqs;
 extern struct irq_desc *irq_to_desc(unsigned int irq);
+unsigned int irq_get_next_irq(unsigned int offset);
 
 # define for_each_irq_desc(irq, desc)					\
 	for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs;		\
@@ -47,6 +48,10 @@ extern struct irq_desc *irq_to_desc(unsigned int irq);
 #define irq_node(irq)	0
 #endif
 
+# define for_each_active_irq(irq)			\
+	for (irq = irq_get_next_irq(0); irq < nr_irqs;	\
+	     irq = irq_get_next_irq(irq + 1))
+
 #endif /* CONFIG_GENERIC_HARDIRQS */
 
 #define for_each_irq_nr(irq)                   \
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 6312a2c83971..2e7e94ef64da 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -463,6 +463,17 @@ err:
 	return ret;
 }
 
+/**
+ * irq_get_next_irq - get next allocated irq number
+ * @offset:	where to start the search
+ *
+ * Returns next irq number after offset or nr_irqs if none is found.
+ */
+unsigned int irq_get_next_irq(unsigned int offset)
+{
+	return find_next_bit(allocated_irqs, nr_irqs, offset);
+}
+
 /* Statistics access */
 void clear_kstat_irqs(struct irq_desc *desc)
 {
-- 
cgit v1.2.3


From 06f6c3399e9f9ff6eafc200e80f9226c3cee0eaf Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 12 Oct 2010 12:31:46 +0200
Subject: genirq: Implement irq reservation

Mark a range of interrupts as allocated. In the SPARSE_IRQ=n case we
need this to update the bitmap for the legacy irqs so the enumerator
via irq_get_next_irq() works.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/irq.h  |  1 +
 kernel/irq/irqdesc.c | 26 ++++++++++++++++++++++++++
 2 files changed, 27 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index cefacf928b33..096b74d5d0d7 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -400,6 +400,7 @@ static inline struct irq_2_iommu *irq_data_get_iommu(struct irq_data *d)
 
 int irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node);
 void irq_free_descs(unsigned int irq, unsigned int cnt);
+int irq_reserve_irqs(unsigned int from, unsigned int cnt);
 
 static inline int irq_alloc_desc(int node)
 {
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 2e7e94ef64da..35d9052901b9 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -463,6 +463,32 @@ err:
 	return ret;
 }
 
+/**
+ * irq_reserve_irqs - mark irqs allocated
+ * @from:	mark from irq number
+ * @cnt:	number of irqs to mark
+ *
+ * Returns 0 on success or an appropriate error code
+ */
+int irq_reserve_irqs(unsigned int from, unsigned int cnt)
+{
+	unsigned long flags;
+	unsigned int start;
+	int ret = 0;
+
+	if (!cnt || (from + cnt) > nr_irqs)
+		return -EINVAL;
+
+	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
+	start = bitmap_find_next_zero_area(allocated_irqs, nr_irqs, from, cnt, 0);
+	if (start == from)
+		bitmap_set(allocated_irqs, start, cnt);
+	else
+		ret = -EEXIST;
+	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
+	return ret;
+}
+
 /**
  * irq_get_next_irq - get next allocated irq number
  * @offset:	where to start the search
-- 
cgit v1.2.3


From b683de2b3cb17bb10fa6fd4af614dc75b5749fe0 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 27 Sep 2010 20:55:03 +0200
Subject: genirq: Query arch for number of early descriptors

sparse irq sets up NR_IRQS_LEGACY irq descriptors and archs then go
ahead and allocate more.

Use the unused return value of arch_probe_nr_irqs() to let the
architecture return the number of early allocations. Fix up all users.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 arch/arm/kernel/irq.c          |  6 ++----
 arch/sh/kernel/irq.c           |  2 +-
 arch/x86/kernel/apic/io_apic.c |  2 +-
 include/linux/irq.h            |  4 ++++
 kernel/irq/irqdesc.c           | 10 +++++-----
 kernel/softirq.c               |  4 +++-
 6 files changed, 16 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index c0d5c3b3a760..5456d11d6ae4 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -157,10 +157,8 @@ void __init init_IRQ(void)
 	struct irq_desc *desc;
 	int irq;
 
-	for (irq = 0; irq < nr_irqs; irq++) {
-		desc = irq_to_desc_alloc_node(irq, 0);
+	for (irq = 0; irq < nr_irqs; irq++)
 		desc->status |= IRQ_NOREQUEST | IRQ_NOPROBE;
-	}
 
 	init_arch_irq();
 }
@@ -169,7 +167,7 @@ void __init init_IRQ(void)
 int __init arch_probe_nr_irqs(void)
 {
 	nr_irqs = arch_nr_irqs ? arch_nr_irqs : NR_IRQS;
-	return 0;
+	return nr_irqs;
 }
 #endif
 
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index 257de1f0692b..ae5bac39b896 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -290,7 +290,7 @@ void __init init_IRQ(void)
 int __init arch_probe_nr_irqs(void)
 {
 	nr_irqs = sh_mv.mv_nr_irqs;
-	return 0;
+	return NR_IRQS_LEGACY;
 }
 #endif
 
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index f1efebaf5510..5aee1d1a306d 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -3880,7 +3880,7 @@ int __init arch_probe_nr_irqs(void)
 	if (nr < nr_irqs)
 		nr_irqs = nr;
 
-	return 0;
+	return NR_IRQS_LEGACY;
 }
 #endif
 
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 096b74d5d0d7..ef878823ee3b 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -214,6 +214,10 @@ struct irq_chip {
  */
 #include <asm/hw_irq.h>
 
+#ifndef NR_IRQS_LEGACY
+# define NR_IRQS_LEGACY 0
+#endif
+
 #ifndef ARCH_IRQ_INIT_FLAGS
 # define ARCH_IRQ_INIT_FLAGS	0
 #endif
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 7cbe4f93e2fb..a1fbd1d347af 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -226,16 +226,16 @@ struct irq_desc * __ref irq_to_desc_alloc_node(unsigned int irq, int node)
 
 int __init early_irq_init(void)
 {
-	int i, node = first_online_node;
+	int i, initcnt, node = first_online_node;
 	struct irq_desc *desc;
 
 	init_irq_default_affinity();
 
-	 /* initialize nr_irqs based on nr_cpu_ids */
-	arch_probe_nr_irqs();
-	printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d\n", NR_IRQS, nr_irqs);
+	/* Let arch update nr_irqs and return the nr of preallocated irqs */
+	initcnt = arch_probe_nr_irqs();
+	printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d %d\n", NR_IRQS, nr_irqs, initcnt);
 
-	for (i = 0; i < NR_IRQS_LEGACY; i++) {
+	for (i = 0; i < initcnt; i++) {
 		desc = alloc_desc(i, node);
 		set_bit(i, allocated_irqs);
 		irq_insert_desc(i, desc);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 07b4f1b1a73a..14a7b80b2cce 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -886,9 +886,10 @@ int __init __weak early_irq_init(void)
 	return 0;
 }
 
+#ifdef CONFIG_GENERIC_HARDIRQS
 int __init __weak arch_probe_nr_irqs(void)
 {
-	return 0;
+	return NR_IRQS_LEGACY;
 }
 
 int __init __weak arch_early_irq_init(void)
@@ -900,3 +901,4 @@ int __weak arch_init_chip_data(struct irq_desc *desc, int node)
 {
 	return 0;
 }
+#endif
-- 
cgit v1.2.3


From 1c9db52534a2c0e9776788cd34ccc193289fc18c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Sep 2010 16:46:51 +0200
Subject: pci: Convert msi to new irq_chip functions

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Russell King <linux@arm.linux.org.uk>
---
 arch/arm/mach-iop13xx/msi.c            |  8 ++++----
 arch/ia64/kernel/msi_ia64.c            |  4 ++--
 arch/ia64/sn/kernel/msi_sn.c           |  4 ++--
 arch/powerpc/platforms/cell/axon_msi.c |  6 +++---
 arch/powerpc/platforms/pseries/xics.c  |  2 +-
 arch/powerpc/sysdev/fsl_msi.c          |  4 ++--
 arch/powerpc/sysdev/mpic_pasemi_msi.c  | 22 +++++++++++-----------
 arch/powerpc/sysdev/mpic_u3msi.c       | 18 +++++++++---------
 arch/sparc/kernel/pci_msi.c            |  8 ++++----
 arch/x86/kernel/apic/io_apic.c         |  8 ++++----
 drivers/pci/msi.c                      | 14 +++++++-------
 include/linux/msi.h                    |  5 +++--
 12 files changed, 52 insertions(+), 51 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/mach-iop13xx/msi.c b/arch/arm/mach-iop13xx/msi.c
index f34b0ed80630..7149fcc16c8a 100644
--- a/arch/arm/mach-iop13xx/msi.c
+++ b/arch/arm/mach-iop13xx/msi.c
@@ -164,10 +164,10 @@ static void iop13xx_msi_nop(unsigned int irq)
 static struct irq_chip iop13xx_msi_chip = {
 	.name = "PCI-MSI",
 	.ack = iop13xx_msi_nop,
-	.enable = unmask_msi_irq,
-	.disable = mask_msi_irq,
-	.mask = mask_msi_irq,
-	.unmask = unmask_msi_irq,
+	.irq_enable = unmask_msi_irq,
+	.irq_disable = mask_msi_irq,
+	.irq_mask = mask_msi_irq,
+	.irq_unmask = unmask_msi_irq,
 };
 
 int arch_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc)
diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c
index 4a746ea838ff..dbb43424704c 100644
--- a/arch/ia64/kernel/msi_ia64.c
+++ b/arch/ia64/kernel/msi_ia64.c
@@ -104,8 +104,8 @@ static int ia64_msi_retrigger_irq(unsigned int irq)
  */
 static struct irq_chip ia64_msi_chip = {
 	.name		= "PCI-MSI",
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
 	.ack		= ia64_ack_msi_irq,
 #ifdef CONFIG_SMP
 	.set_affinity	= ia64_set_msi_irq_affinity,
diff --git a/arch/ia64/sn/kernel/msi_sn.c b/arch/ia64/sn/kernel/msi_sn.c
index 0c72dd463831..a5e500f02853 100644
--- a/arch/ia64/sn/kernel/msi_sn.c
+++ b/arch/ia64/sn/kernel/msi_sn.c
@@ -228,8 +228,8 @@ static int sn_msi_retrigger_irq(unsigned int irq)
 
 static struct irq_chip sn_msi_chip = {
 	.name		= "PCI-MSI",
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
 	.ack		= sn_ack_msi_irq,
 #ifdef CONFIG_SMP
 	.set_affinity	= sn_set_msi_irq_affinity,
diff --git a/arch/powerpc/platforms/cell/axon_msi.c b/arch/powerpc/platforms/cell/axon_msi.c
index 97085530aa63..e3e379c6caa7 100644
--- a/arch/powerpc/platforms/cell/axon_msi.c
+++ b/arch/powerpc/platforms/cell/axon_msi.c
@@ -310,9 +310,9 @@ static void axon_msi_teardown_msi_irqs(struct pci_dev *dev)
 }
 
 static struct irq_chip msic_irq_chip = {
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
-	.shutdown	= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
+	.irq_shutdown	= mask_msi_irq,
 	.name		= "AXON-MSI",
 };
 
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index 93834b0d8272..67e2c4bdac8f 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -243,7 +243,7 @@ static unsigned int xics_startup(unsigned int virq)
 	 * at that level, so we do it here by hand.
 	 */
 	if (irq_to_desc(virq)->msi_desc)
-		unmask_msi_irq(virq);
+		unmask_msi_irq(irq_get_irq_data(virq));
 
 	/* unmask it */
 	xics_unmask_irq(virq);
diff --git a/arch/powerpc/sysdev/fsl_msi.c b/arch/powerpc/sysdev/fsl_msi.c
index 87991d3abbab..bdbd896c89d8 100644
--- a/arch/powerpc/sysdev/fsl_msi.c
+++ b/arch/powerpc/sysdev/fsl_msi.c
@@ -51,8 +51,8 @@ static void fsl_msi_end_irq(unsigned int virq)
 }
 
 static struct irq_chip fsl_msi_chip = {
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
 	.ack		= fsl_msi_end_irq,
 	.name		= "FSL-MSI",
 };
diff --git a/arch/powerpc/sysdev/mpic_pasemi_msi.c b/arch/powerpc/sysdev/mpic_pasemi_msi.c
index 3b6a9a43718f..320ad5a9a25d 100644
--- a/arch/powerpc/sysdev/mpic_pasemi_msi.c
+++ b/arch/powerpc/sysdev/mpic_pasemi_msi.c
@@ -39,24 +39,24 @@
 static struct mpic *msi_mpic;
 
 
-static void mpic_pasemi_msi_mask_irq(unsigned int irq)
+static void mpic_pasemi_msi_mask_irq(struct irq_data *data)
 {
-	pr_debug("mpic_pasemi_msi_mask_irq %d\n", irq);
-	mask_msi_irq(irq);
-	mpic_mask_irq(irq);
+	pr_debug("mpic_pasemi_msi_mask_irq %d\n", data->irq);
+	mask_msi_irq(data);
+	mpic_mask_irq(data->irq);
 }
 
-static void mpic_pasemi_msi_unmask_irq(unsigned int irq)
+static void mpic_pasemi_msi_unmask_irq(struct irq_data *data)
 {
-	pr_debug("mpic_pasemi_msi_unmask_irq %d\n", irq);
-	mpic_unmask_irq(irq);
-	unmask_msi_irq(irq);
+	pr_debug("mpic_pasemi_msi_unmask_irq %d\n", data->irq);
+	mpic_unmask_irq(data->irq);
+	unmask_msi_irq(data);
 }
 
 static struct irq_chip mpic_pasemi_msi_chip = {
-	.shutdown	= mpic_pasemi_msi_mask_irq,
-	.mask		= mpic_pasemi_msi_mask_irq,
-	.unmask		= mpic_pasemi_msi_unmask_irq,
+	.irq_shutdown	= mpic_pasemi_msi_mask_irq,
+	.irq_mask	= mpic_pasemi_msi_mask_irq,
+	.irq_unmask	= mpic_pasemi_msi_unmask_irq,
 	.eoi		= mpic_end_irq,
 	.set_type	= mpic_set_irq_type,
 	.set_affinity	= mpic_set_affinity,
diff --git a/arch/powerpc/sysdev/mpic_u3msi.c b/arch/powerpc/sysdev/mpic_u3msi.c
index bcbfe79c704b..a2b028b4a202 100644
--- a/arch/powerpc/sysdev/mpic_u3msi.c
+++ b/arch/powerpc/sysdev/mpic_u3msi.c
@@ -23,22 +23,22 @@
 /* A bit ugly, can we get this from the pci_dev somehow? */
 static struct mpic *msi_mpic;
 
-static void mpic_u3msi_mask_irq(unsigned int irq)
+static void mpic_u3msi_mask_irq(struct irq_data *data)
 {
-	mask_msi_irq(irq);
-	mpic_mask_irq(irq);
+	mask_msi_irq(data);
+	mpic_mask_irq(data->irq);
 }
 
-static void mpic_u3msi_unmask_irq(unsigned int irq)
+static void mpic_u3msi_unmask_irq(struct irq_data *data)
 {
-	mpic_unmask_irq(irq);
-	unmask_msi_irq(irq);
+	mpic_unmask_irq(data->irq);
+	unmask_msi_irq(data);
 }
 
 static struct irq_chip mpic_u3msi_chip = {
-	.shutdown	= mpic_u3msi_mask_irq,
-	.mask		= mpic_u3msi_mask_irq,
-	.unmask		= mpic_u3msi_unmask_irq,
+	.irq_shutdown	= mpic_u3msi_mask_irq,
+	.irq_mask	= mpic_u3msi_mask_irq,
+	.irq_unmask	= mpic_u3msi_unmask_irq,
 	.eoi		= mpic_end_irq,
 	.set_type	= mpic_set_irq_type,
 	.set_affinity	= mpic_set_affinity,
diff --git a/arch/sparc/kernel/pci_msi.c b/arch/sparc/kernel/pci_msi.c
index 548b8ca9c210..b210416ace7b 100644
--- a/arch/sparc/kernel/pci_msi.c
+++ b/arch/sparc/kernel/pci_msi.c
@@ -114,10 +114,10 @@ static void free_msi(struct pci_pbm_info *pbm, int msi_num)
 
 static struct irq_chip msi_irq = {
 	.name		= "PCI-MSI",
-	.mask		= mask_msi_irq,
-	.unmask		= unmask_msi_irq,
-	.enable		= unmask_msi_irq,
-	.disable	= mask_msi_irq,
+	.irq_mask	= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
+	.irq_enable	= unmask_msi_irq,
+	.irq_disable	= mask_msi_irq,
 	/* XXX affinity XXX */
 };
 
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 7556eb7a1a47..b79938ff9bde 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -3441,8 +3441,8 @@ ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
  */
 static struct irq_chip msi_chip = {
 	.name		= "PCI-MSI",
-	.unmask		= unmask_msi_irq,
-	.mask		= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
 	.ack		= ack_apic_edge,
 #ifdef CONFIG_SMP
 	.set_affinity	= set_msi_irq_affinity,
@@ -3452,8 +3452,8 @@ static struct irq_chip msi_chip = {
 
 static struct irq_chip msi_ir_chip = {
 	.name		= "IR-PCI-MSI",
-	.unmask		= unmask_msi_irq,
-	.mask		= mask_msi_irq,
+	.irq_unmask	= unmask_msi_irq,
+	.irq_mask	= mask_msi_irq,
 #ifdef CONFIG_INTR_REMAP
 	.ack		= ir_ack_apic_edge,
 #ifdef CONFIG_SMP
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 69b7be33b3a2..55e0f9378dfe 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -170,27 +170,27 @@ static void msix_mask_irq(struct msi_desc *desc, u32 flag)
 	desc->masked = __msix_mask_irq(desc, flag);
 }
 
-static void msi_set_mask_bit(unsigned irq, u32 flag)
+static void msi_set_mask_bit(struct irq_data *data, u32 flag)
 {
-	struct msi_desc *desc = get_irq_msi(irq);
+	struct msi_desc *desc = irq_data_get_msi(data);
 
 	if (desc->msi_attrib.is_msix) {
 		msix_mask_irq(desc, flag);
 		readl(desc->mask_base);		/* Flush write to device */
 	} else {
-		unsigned offset = irq - desc->dev->irq;
+		unsigned offset = data->irq - desc->dev->irq;
 		msi_mask_irq(desc, 1 << offset, flag << offset);
 	}
 }
 
-void mask_msi_irq(unsigned int irq)
+void mask_msi_irq(struct irq_data *data)
 {
-	msi_set_mask_bit(irq, 1);
+	msi_set_mask_bit(data, 1);
 }
 
-void unmask_msi_irq(unsigned int irq)
+void unmask_msi_irq(struct irq_data *data)
 {
-	msi_set_mask_bit(irq, 0);
+	msi_set_mask_bit(data, 0);
 }
 
 void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
diff --git a/include/linux/msi.h b/include/linux/msi.h
index 91b05c171854..329d17c395a7 100644
--- a/include/linux/msi.h
+++ b/include/linux/msi.h
@@ -11,8 +11,9 @@ struct msi_msg {
 
 /* Helper functions */
 struct irq_desc;
-extern void mask_msi_irq(unsigned int irq);
-extern void unmask_msi_irq(unsigned int irq);
+struct irq_data;
+extern void mask_msi_irq(struct irq_data *data);
+extern void unmask_msi_irq(struct irq_data *data);
 extern void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
 extern void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
 extern void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
-- 
cgit v1.2.3


From 39431acb1a4c464e62471cb3058b8ffffb9244db Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Sep 2010 19:09:51 +0200
Subject: pci: Cleanup the irq_desc mess in msi

Handing down irq_desc to msi just so that msi can access
irq_desc.irq_data.msi_desc is a pretty stupid idea. The calling code
can hand down a pointer to msi_desc so msi code does not need to know
about the irq descriptor at all.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/kernel/apic/io_apic.c |  4 ++--
 drivers/pci/msi.c              | 24 +++++++++---------------
 include/linux/msi.h            |  8 ++++----
 3 files changed, 15 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index b79938ff9bde..74bb027b517e 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -3383,14 +3383,14 @@ static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 
 	cfg = desc->chip_data;
 
-	get_cached_msi_msg_desc(desc, &msg);
+	__get_cached_msi_msg(desc->irq_data.msi_desc, &msg);
 
 	msg.data &= ~MSI_DATA_VECTOR_MASK;
 	msg.data |= MSI_DATA_VECTOR(cfg->vector);
 	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
 	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
-	write_msi_msg_desc(desc, &msg);
+	__write_msi_msg(desc->irq_data.msi_desc, &msg);
 
 	return 0;
 }
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 55e0f9378dfe..5fcf5aec680f 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -193,10 +193,8 @@ void unmask_msi_irq(struct irq_data *data)
 	msi_set_mask_bit(data, 0);
 }
 
-void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __read_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	BUG_ON(entry->dev->current_state != PCI_D0);
 
 	if (entry->msi_attrib.is_msix) {
@@ -227,15 +225,13 @@ void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
 
 void read_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	read_msi_msg_desc(desc, msg);
+	__read_msi_msg(entry, msg);
 }
 
-void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __get_cached_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	/* Assert that the cache is valid, assuming that
 	 * valid messages are not all-zeroes. */
 	BUG_ON(!(entry->msg.address_hi | entry->msg.address_lo |
@@ -246,15 +242,13 @@ void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
 
 void get_cached_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	get_cached_msi_msg_desc(desc, msg);
+	__get_cached_msi_msg(entry, msg);
 }
 
-void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __write_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	if (entry->dev->current_state != PCI_D0) {
 		/* Don't touch the hardware now */
 	} else if (entry->msi_attrib.is_msix) {
@@ -292,9 +286,9 @@ void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
 
 void write_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	write_msi_msg_desc(desc, msg);
+	__write_msi_msg(entry, msg);
 }
 
 static void free_msi_irqs(struct pci_dev *dev)
diff --git a/include/linux/msi.h b/include/linux/msi.h
index 329d17c395a7..05acced439a3 100644
--- a/include/linux/msi.h
+++ b/include/linux/msi.h
@@ -10,13 +10,13 @@ struct msi_msg {
 };
 
 /* Helper functions */
-struct irq_desc;
 struct irq_data;
+struct msi_desc;
 extern void mask_msi_irq(struct irq_data *data);
 extern void unmask_msi_irq(struct irq_data *data);
-extern void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
-extern void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
-extern void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg);
+extern void __read_msi_msg(struct msi_desc *entry, struct msi_msg *msg);
+extern void __get_cached_msi_msg(struct msi_desc *entry, struct msi_msg *msg);
+extern void __write_msi_msg(struct msi_desc *entry, struct msi_msg *msg);
 extern void read_msi_msg(unsigned int irq, struct msi_msg *msg);
 extern void get_cached_msi_msg(unsigned int irq, struct msi_msg *msg);
 extern void write_msi_msg(unsigned int irq, struct msi_msg *msg);
-- 
cgit v1.2.3


From 5c2837fbaa609e615ef9a1c58a4cd26ce90be35b Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Sep 2010 17:15:11 +0200
Subject: dmar: Convert to new irq chip functions

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Acked-by: David Woodhouse <dwmw2@infradead.org>
---
 arch/ia64/kernel/msi_ia64.c    | 4 ++--
 arch/x86/kernel/apic/io_apic.c | 4 ++--
 drivers/pci/dmar.c             | 8 ++++----
 include/linux/dmar.h           | 5 +++--
 4 files changed, 11 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c
index dbb43424704c..00b19a416eab 100644
--- a/arch/ia64/kernel/msi_ia64.c
+++ b/arch/ia64/kernel/msi_ia64.c
@@ -160,8 +160,8 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 
 static struct irq_chip dmar_msi_type = {
 	.name = "DMAR_MSI",
-	.unmask = dmar_msi_unmask,
-	.mask = dmar_msi_mask,
+	.irq_unmask = dmar_msi_unmask,
+	.irq_mask = dmar_msi_mask,
 	.ack = ia64_ack_msi_irq,
 #ifdef CONFIG_SMP
 	.set_affinity = dmar_msi_set_affinity,
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 49aa857ff004..72b253ef310e 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -3578,8 +3578,8 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 
 static struct irq_chip dmar_msi_type = {
 	.name = "DMAR_MSI",
-	.unmask = dmar_msi_unmask,
-	.mask = dmar_msi_mask,
+	.irq_unmask = dmar_msi_unmask,
+	.irq_mask = dmar_msi_mask,
 	.irq_ack = ack_apic_edge,
 #ifdef CONFIG_SMP
 	.set_affinity = dmar_msi_set_affinity,
diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 0a19708074c2..3de3a436a432 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -1221,9 +1221,9 @@ const char *dmar_get_fault_reason(u8 fault_reason, int *fault_type)
 	}
 }
 
-void dmar_msi_unmask(unsigned int irq)
+void dmar_msi_unmask(struct irq_data *data)
 {
-	struct intel_iommu *iommu = get_irq_data(irq);
+	struct intel_iommu *iommu = irq_data_get_irq_data(data);
 	unsigned long flag;
 
 	/* unmask it */
@@ -1234,10 +1234,10 @@ void dmar_msi_unmask(unsigned int irq)
 	spin_unlock_irqrestore(&iommu->register_lock, flag);
 }
 
-void dmar_msi_mask(unsigned int irq)
+void dmar_msi_mask(struct irq_data *data)
 {
 	unsigned long flag;
-	struct intel_iommu *iommu = get_irq_data(irq);
+	struct intel_iommu *iommu = irq_data_get_irq_data(data);
 
 	/* mask it */
 	spin_lock_irqsave(&iommu->register_lock, flag);
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index d7cecc90ed34..cb86aa1ca436 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -187,8 +187,9 @@ static inline int set_msi_sid(struct irte *irte, struct pci_dev *dev)
 /* Can't use the common MSI interrupt functions
  * since DMAR is not a pci device
  */
-extern void dmar_msi_unmask(unsigned int irq);
-extern void dmar_msi_mask(unsigned int irq);
+struct irq_data;
+extern void dmar_msi_unmask(struct irq_data *data);
+extern void dmar_msi_mask(struct irq_data *data);
 extern void dmar_msi_read(int irq, struct msi_msg *msg);
 extern void dmar_msi_write(int irq, struct msi_msg *msg);
 extern int dmar_set_interrupt(struct intel_iommu *iommu);
-- 
cgit v1.2.3


From e9f7ac664bfc36685a8eb3315ec21c067d0cee36 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Sep 2010 17:22:09 +0200
Subject: ht: Convert to new irq_chip functions

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/kernel/apic/io_apic.c |  4 ++--
 drivers/pci/htirq.c            | 22 ++++++++--------------
 include/linux/htirq.h          |  5 +++--
 3 files changed, 13 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 72b253ef310e..5579f3f5943a 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -3730,8 +3730,8 @@ static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask)
 
 static struct irq_chip ht_irq_chip = {
 	.name		= "PCI-HT",
-	.mask		= mask_ht_irq,
-	.unmask		= unmask_ht_irq,
+	.irq_mask	= mask_ht_irq,
+	.irq_unmask	= unmask_ht_irq,
 	.irq_ack	= ack_apic_edge,
 #ifdef CONFIG_SMP
 	.set_affinity	= set_ht_irq_affinity,
diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c
index 98abf8b91294..834842aa5bbf 100644
--- a/drivers/pci/htirq.c
+++ b/drivers/pci/htirq.c
@@ -57,28 +57,22 @@ void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg)
 	*msg = cfg->msg;
 }
 
-void mask_ht_irq(unsigned int irq)
+void mask_ht_irq(struct irq_data *data)
 {
-	struct ht_irq_cfg *cfg;
-	struct ht_irq_msg msg;
-
-	cfg = get_irq_data(irq);
+	struct ht_irq_cfg *cfg = irq_data_get_irq_data(data);
+	struct ht_irq_msg msg = cfg->msg;
 
-	msg = cfg->msg;
 	msg.address_lo |= 1;
-	write_ht_irq_msg(irq, &msg);
+	write_ht_irq_msg(data->irq, &msg);
 }
 
-void unmask_ht_irq(unsigned int irq)
+void unmask_ht_irq(struct irq_data *data)
 {
-	struct ht_irq_cfg *cfg;
-	struct ht_irq_msg msg;
-
-	cfg = get_irq_data(irq);
+	struct ht_irq_cfg *cfg = irq_data_get_irq_data(data);
+	struct ht_irq_msg msg = cfg->msg;
 
-	msg = cfg->msg;
 	msg.address_lo &= ~1;
-	write_ht_irq_msg(irq, &msg);
+	write_ht_irq_msg(data->irq, &msg);
 }
 
 /**
diff --git a/include/linux/htirq.h b/include/linux/htirq.h
index c96ea46737d0..70a1dbbf2093 100644
--- a/include/linux/htirq.h
+++ b/include/linux/htirq.h
@@ -9,8 +9,9 @@ struct ht_irq_msg {
 /* Helper functions.. */
 void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg);
 void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg);
-void mask_ht_irq(unsigned int irq);
-void unmask_ht_irq(unsigned int irq);
+struct irq_data;
+void mask_ht_irq(struct irq_data *data);
+void unmask_ht_irq(struct irq_data *data);
 
 /* The arch hook for getting things started */
 int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev);
-- 
cgit v1.2.3


From d0ad63927c6d4d511e172c78ba4a623539ef6901 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 4 Oct 2010 18:41:37 +0200
Subject: pci: intr_remap: Remove unused functions

No users.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/intr_remapping.c | 47 --------------------------------------------
 include/linux/dmar.h         |  2 --
 2 files changed, 49 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index cb6252988546..343f7299c783 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -275,28 +275,6 @@ int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
 	return 0;
 }
 
-int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index)
-{
-	struct irq_2_iommu *irq_iommu;
-	unsigned long flags;
-
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
-
-	irq_iommu->iommu = NULL;
-	irq_iommu->irte_index = 0;
-	irq_iommu->sub_handle = 0;
-	irq_2_iommu(irq)->irte_mask = 0;
-
-	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-
-	return 0;
-}
-
 int modify_irte(int irq, struct irte *irte_modified)
 {
 	int rc;
@@ -328,31 +306,6 @@ int modify_irte(int irq, struct irte *irte_modified)
 	return rc;
 }
 
-int flush_irte(int irq)
-{
-	int rc;
-	int index;
-	struct intel_iommu *iommu;
-	struct irq_2_iommu *irq_iommu;
-	unsigned long flags;
-
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
-
-	iommu = irq_iommu->iommu;
-
-	index = irq_iommu->irte_index + irq_iommu->sub_handle;
-
-	rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask);
-	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-
-	return rc;
-}
-
 struct intel_iommu *map_hpet_to_ir(u8 hpet_id)
 {
 	int i;
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index cb86aa1ca436..200439ec7c49 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -119,8 +119,6 @@ extern int alloc_irte(struct intel_iommu *iommu, int irq, u16 count);
 extern int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index,
    			u16 sub_handle);
 extern int map_irq_to_irte_handle(int irq, u16 *sub_handle);
-extern int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index);
-extern int flush_irte(int irq);
 extern int free_irte(int irq);
 
 extern int irq_remapped(int irq);
-- 
cgit v1.2.3


From 423f085952fd7253407cb92984cc2d495a564481 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 10 Oct 2010 11:39:09 +0200
Subject: x86: Embedd irq_2_iommu into irq_cfg

That interrupt remapping code is x86 specific and tied to the io_apic
code. No need for separate allocator functions in the interrupt
remapping code. This allows to simplify the code and irq_2_iommu is
small (13 bytes on 64bit) so it's not a real problem even if interrupt
remapping is runtime disabled. If it's compile time disabled the
impact is zero.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/include/asm/hw_irq.h | 10 ++++++++++
 drivers/pci/intr_remapping.c  |  7 -------
 include/linux/dmar.h          |  1 +
 3 files changed, 11 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h
index 76848f27b1ac..e756c4bfed94 100644
--- a/arch/x86/include/asm/hw_irq.h
+++ b/arch/x86/include/asm/hw_irq.h
@@ -78,6 +78,13 @@ static inline void set_io_apic_irq_attr(struct io_apic_irq_attr *irq_attr,
 	irq_attr->polarity	= polarity;
 }
 
+struct irq_2_iommu {
+	struct intel_iommu *iommu;
+	u16 irte_index;
+	u16 sub_handle;
+	u8  irte_mask;
+};
+
 /*
  * This is performance-critical, we want to do it O(1)
  *
@@ -89,6 +96,9 @@ struct irq_cfg {
 	cpumask_var_t		old_domain;
 	u8			vector;
 	u8			move_in_progress : 1;
+#ifdef CONFIG_INTR_REMAP
+	struct irq_2_iommu	irq_2_iommu;
+#endif
 };
 
 extern struct irq_cfg *irq_cfg(unsigned int);
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index 343f7299c783..0f4691c5fab3 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -46,13 +46,6 @@ static __init int setup_intremap(char *str)
 }
 early_param("intremap", setup_intremap);
 
-struct irq_2_iommu {
-	struct intel_iommu *iommu;
-	u16 irte_index;
-	u16 sub_handle;
-	u8  irte_mask;
-};
-
 #ifdef CONFIG_GENERIC_HARDIRQS
 static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
 {
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index 200439ec7c49..4475f8cf7a62 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -106,6 +106,7 @@ struct irte {
 		__u64 high;
 	};
 };
+
 #ifdef CONFIG_INTR_REMAP
 extern int intr_remapping_enabled;
 extern int intr_remapping_supported(void);
-- 
cgit v1.2.3


From 1a0730d6649113c820217387a011a17dd4aff3ad Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 11 Oct 2010 11:55:37 +0200
Subject: x86: Speed up the irq_remapped check in hot pathes

irq_2_iommu is in struct irq_cfg, so we can do the irq_remapped check
based on irq_cfg instead of going through a lookup function. That's
especially interesting in the eoi_ioapic_irq() hotpath.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/include/asm/irq_remapping.h |  8 ++++++++
 arch/x86/kernel/apic/io_apic.c       | 12 ++++++------
 drivers/pci/intr_remapping.c         |  7 -------
 include/linux/dmar.h                 |  2 --
 4 files changed, 14 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/include/asm/irq_remapping.h b/arch/x86/include/asm/irq_remapping.h
index 8d841505344e..1c23360fb2d8 100644
--- a/arch/x86/include/asm/irq_remapping.h
+++ b/arch/x86/include/asm/irq_remapping.h
@@ -24,10 +24,18 @@ static inline void prepare_irte(struct irte *irte, int vector,
 	irte->dest_id = IRTE_DEST(dest);
 	irte->redir_hint = 1;
 }
+static inline bool irq_remapped(struct irq_cfg *cfg)
+{
+	return cfg->irq_2_iommu.iommu != NULL;
+}
 #else
 static void prepare_irte(struct irte *irte, int vector, unsigned int dest)
 {
 }
+static inline bool irq_remapped(struct irq_cfg *cfg)
+{
+	return false;
+}
 #endif
 
 #endif	/* _ASM_X86_IRQ_REMAPPING_H */
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 6ff6bb883c58..1b8e8a106120 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -1237,7 +1237,7 @@ static void ioapic_register_intr(unsigned int irq, unsigned long trigger)
 	else
 		irq_clear_status_flags(irq, IRQ_LEVEL);
 
-	if (irq_remapped(irq)) {
+	if (irq_remapped(get_irq_chip_data(irq))) {
 		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
 		if (trigger)
 			set_irq_chip_and_handler_name(irq, &ir_ioapic_chip,
@@ -2183,7 +2183,7 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq
 		 * With interrupt-remapping, destination information comes
 		 * from interrupt-remapping table entry.
 		 */
-		if (!irq_remapped(irq))
+		if (!irq_remapped(cfg))
 			io_apic_write(apic, 0x11 + pin*2, dest);
 		reg = io_apic_read(apic, 0x10 + pin*2);
 		reg &= ~IO_APIC_REDIR_VECTOR_MASK;
@@ -2415,7 +2415,7 @@ static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
 			 * intr-remapping table entry. Hence for the io-apic
 			 * EOI we use the pin number.
 			 */
-			if (irq_remapped(irq))
+			if (irq_remapped(cfg))
 				io_apic_eoi(entry->apic, entry->pin);
 			else
 				io_apic_eoi(entry->apic, cfg->vector);
@@ -3139,7 +3139,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq,
 
 	dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus());
 
-	if (irq_remapped(irq)) {
+	if (irq_remapped(get_irq_chip_data(irq))) {
 		struct irte irte;
 		int ir_index;
 		u16 sub_handle;
@@ -3321,7 +3321,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 	set_irq_msi(irq, msidesc);
 	write_msi_msg(irq, &msg);
 
-	if (irq_remapped(irq)) {
+	if (irq_remapped(get_irq_chip_data(irq))) {
 		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
 		set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge");
 	} else
@@ -3522,7 +3522,7 @@ int arch_setup_hpet_msi(unsigned int irq, unsigned int id)
 
 	hpet_msi_write(get_irq_data(irq), &msg);
 	irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
-	if (irq_remapped(irq))
+	if (irq_remapped(get_irq_chip_data(irq)))
 		set_irq_chip_and_handler_name(irq, &ir_hpet_msi_type,
 					      handle_edge_irq, "edge");
 	else
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index a620b8bd8f4b..ec87cd66f3eb 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -54,13 +54,6 @@ static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
 	return cfg ? &cfg->irq_2_iommu : NULL;
 }
 
-int irq_remapped(int irq)
-{
-	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
-
-	return irq_iommu ? irq_iommu->iommu != NULL : 0;
-}
-
 int get_irte(int irq, struct irte *entry)
 {
 	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index 4475f8cf7a62..51651b76d40f 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -122,7 +122,6 @@ extern int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index,
 extern int map_irq_to_irte_handle(int irq, u16 *sub_handle);
 extern int free_irte(int irq);
 
-extern int irq_remapped(int irq);
 extern struct intel_iommu *map_dev_to_ir(struct pci_dev *dev);
 extern struct intel_iommu *map_ioapic_to_ir(int apic);
 extern struct intel_iommu *map_hpet_to_ir(u8 id);
@@ -176,7 +175,6 @@ static inline int set_msi_sid(struct irte *irte, struct pci_dev *dev)
 	return 0;
 }
 
-#define irq_remapped(irq)		(0)
 #define enable_intr_remapping(mode)	(-1)
 #define disable_intr_remapping()	(0)
 #define reenable_intr_remapping(mode)	(0)
-- 
cgit v1.2.3


From 10ba1e0eeef6a3c9453d96364e28cb4d911e1ac3 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 11 Oct 2010 12:21:18 +0200
Subject: genirq: Remove irq_2_iommu

irq_2_iommu is now in the x86 code where it belongs. Remove all
leftovers.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 include/linux/irq.h     | 18 ------------------
 include/linux/irqdesc.h |  4 ----
 2 files changed, 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index ef878823ee3b..49702b22883e 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -85,7 +85,6 @@ typedef	void (*irq_flow_handler_t)(unsigned int irq,
 #endif
 
 struct msi_desc;
-struct irq_2_iommu;
 
 /**
  * struct irq_data - per irq and irq chip data passed down to chip functions
@@ -97,7 +96,6 @@ struct irq_2_iommu;
  *			methods, to allow shared chip implementations
  * @msi_desc:		MSI descriptor
  * @affinity:		IRQ affinity on SMP
- * @irq_2_iommu:	iommu with this irq
  *
  * The fields here need to overlay the ones in irq_desc until we
  * cleaned up the direct references and switched everything over to
@@ -113,9 +111,6 @@ struct irq_data {
 #ifdef CONFIG_SMP
 	cpumask_var_t		affinity;
 #endif
-#ifdef CONFIG_INTR_REMAP
-	struct irq_2_iommu      *irq_2_iommu;
-#endif
 };
 
 /**
@@ -389,19 +384,6 @@ static inline struct msi_desc *irq_data_get_msi(struct irq_data *d)
 	return d->msi_desc;
 }
 
-#ifdef CONFIG_INTR_REMAP
-static inline struct irq_2_iommu *get_irq_iommu(unsigned int irq)
-{
-	struct irq_data *d = irq_get_irq_data(irq);
-	return d ? d->irq_2_iommu : NULL;
-}
-
-static inline struct irq_2_iommu *irq_data_get_iommu(struct irq_data *d)
-{
-	return d->irq_2_iommu;
-}
-#endif
-
 int irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node);
 void irq_free_descs(unsigned int irq, unsigned int cnt);
 int irq_reserve_irqs(unsigned int from, unsigned int cnt);
diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h
index 22e426fdd301..f77dc5618d7e 100644
--- a/include/linux/irqdesc.h
+++ b/include/linux/irqdesc.h
@@ -10,7 +10,6 @@
 
 struct proc_dir_entry;
 struct timer_rand_state;
-struct irq_2_iommu;
 /**
  * struct irq_desc - interrupt descriptor
  * @irq_data:		per irq and chip data passed down to chip functions
@@ -52,9 +51,6 @@ struct irq_desc {
 			struct msi_desc		*msi_desc;
 #ifdef CONFIG_SMP
 			cpumask_var_t		affinity;
-#endif
-#ifdef CONFIG_INTR_REMAP
-			struct irq_2_iommu      *irq_2_iommu;
 #endif
 		};
 	};
-- 
cgit v1.2.3


From b7d0d8258a9f71949b810e0f82a3d75088f4d364 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 29 Sep 2010 18:44:23 +0200
Subject: genirq: Remove arch_init_chip_data()

This function should have not been there in the first place.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/interrupt.h | 3 ---
 kernel/irq/irqdesc.c      | 2 --
 kernel/softirq.c          | 5 -----
 3 files changed, 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index a0384a4d1e6f..19988983aeac 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -641,11 +641,8 @@ static inline void init_irq_proc(void)
 struct seq_file;
 int show_interrupts(struct seq_file *p, void *v);
 
-struct irq_desc;
-
 extern int early_irq_init(void);
 extern int arch_probe_nr_irqs(void);
 extern int arch_early_irq_init(void);
-extern int arch_init_chip_data(struct irq_desc *desc, int node);
 
 #endif
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index a1fbd1d347af..6c71f8ea5d7d 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -197,8 +197,6 @@ static int alloc_descs(unsigned int start, unsigned int cnt, int node)
 		desc = alloc_desc(start + i, node);
 		if (!desc)
 			goto err;
-		/* temporary until I fixed x86 madness */
-		arch_init_chip_data(desc, node);
 		raw_spin_lock_irqsave(&sparse_irq_lock, flags);
 		irq_insert_desc(start + i, desc);
 		raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 14a7b80b2cce..d19b1c9aa7c5 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -896,9 +896,4 @@ int __init __weak arch_early_irq_init(void)
 {
 	return 0;
 }
-
-int __weak arch_init_chip_data(struct irq_desc *desc, int node)
-{
-	return 0;
-}
 #endif
-- 
cgit v1.2.3


From b7b29338dc7111ed8bd4d6555d84afae13ebe752 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 29 Sep 2010 18:46:55 +0200
Subject: genirq: Sanitize dynamic irq handling

Use the cleanup functions of the dynamic allocator. No need to have
separate implementations.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h    |  12 ++++--
 kernel/irq/chip.c      | 102 -------------------------------------------------
 kernel/irq/internals.h |   1 -
 kernel/irq/irqdesc.c   |  41 +++++++++++---------
 4 files changed, 31 insertions(+), 125 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 49702b22883e..e9639115dff1 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -326,11 +326,15 @@ extern unsigned int create_irq_nr(unsigned int irq_want, int node);
 extern int create_irq(void);
 extern void destroy_irq(unsigned int irq);
 
-/* Dynamic irq helper functions */
-extern void dynamic_irq_init(unsigned int irq);
-void dynamic_irq_init_keep_chip_data(unsigned int irq);
+/*
+ * Dynamic irq helper functions. Obsolete. Use irq_alloc_desc* and
+ * irq_free_desc instead.
+ */
 extern void dynamic_irq_cleanup(unsigned int irq);
-void dynamic_irq_cleanup_keep_chip_data(unsigned int irq);
+static inline void dynamic_irq_init(unsigned int irq)
+{
+	dynamic_irq_cleanup(irq);
+}
 
 /* Set/get chip/data for an IRQ: */
 extern int set_irq_chip(unsigned int irq, struct irq_chip *chip);
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 3405761d6224..baa5c4acad83 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -18,108 +18,6 @@
 
 #include "internals.h"
 
-static void dynamic_irq_init_x(unsigned int irq, bool keep_chip_data)
-{
-	struct irq_desc *desc;
-	unsigned long flags;
-
-	desc = irq_to_desc(irq);
-	if (!desc) {
-		WARN(1, KERN_ERR "Trying to initialize invalid IRQ%d\n", irq);
-		return;
-	}
-
-	/* Ensure we don't have left over values from a previous use of this irq */
-	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc->status = IRQ_DEFAULT_INIT_FLAGS;
-	desc->irq_data.chip = &no_irq_chip;
-	desc->handle_irq = handle_bad_irq;
-	desc->depth = 1;
-	desc->irq_data.msi_desc = NULL;
-	desc->irq_data.handler_data = NULL;
-	if (!keep_chip_data)
-		desc->irq_data.chip_data = NULL;
-	desc->action = NULL;
-	desc->irq_count = 0;
-	desc->irqs_unhandled = 0;
-#ifdef CONFIG_SMP
-	cpumask_setall(desc->irq_data.affinity);
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_clear(desc->pending_mask);
-#endif
-#endif
-	raw_spin_unlock_irqrestore(&desc->lock, flags);
-}
-
-/**
- *	dynamic_irq_init - initialize a dynamically allocated irq
- *	@irq:	irq number to initialize
- */
-void dynamic_irq_init(unsigned int irq)
-{
-	dynamic_irq_init_x(irq, false);
-}
-
-/**
- *	dynamic_irq_init_keep_chip_data - initialize a dynamically allocated irq
- *	@irq:	irq number to initialize
- *
- *	does not set irq_to_desc(irq)->irq_data.chip_data to NULL
- */
-void dynamic_irq_init_keep_chip_data(unsigned int irq)
-{
-	dynamic_irq_init_x(irq, true);
-}
-
-static void dynamic_irq_cleanup_x(unsigned int irq, bool keep_chip_data)
-{
-	struct irq_desc *desc = irq_to_desc(irq);
-	unsigned long flags;
-
-	if (!desc) {
-		WARN(1, KERN_ERR "Trying to cleanup invalid IRQ%d\n", irq);
-		return;
-	}
-
-	raw_spin_lock_irqsave(&desc->lock, flags);
-	if (desc->action) {
-		raw_spin_unlock_irqrestore(&desc->lock, flags);
-		WARN(1, KERN_ERR "Destroying IRQ%d without calling free_irq\n",
-			irq);
-		return;
-	}
-	desc->irq_data.msi_desc = NULL;
-	desc->irq_data.handler_data = NULL;
-	if (!keep_chip_data)
-		desc->irq_data.chip_data = NULL;
-	desc->handle_irq = handle_bad_irq;
-	desc->irq_data.chip = &no_irq_chip;
-	desc->name = NULL;
-	clear_kstat_irqs(desc);
-	raw_spin_unlock_irqrestore(&desc->lock, flags);
-}
-
-/**
- *	dynamic_irq_cleanup - cleanup a dynamically allocated irq
- *	@irq:	irq number to initialize
- */
-void dynamic_irq_cleanup(unsigned int irq)
-{
-	dynamic_irq_cleanup_x(irq, false);
-}
-
-/**
- *	dynamic_irq_cleanup_keep_chip_data - cleanup a dynamically allocated irq
- *	@irq:	irq number to initialize
- *
- *	does not set irq_to_desc(irq)->irq_data.chip_data to NULL
- */
-void dynamic_irq_cleanup_keep_chip_data(unsigned int irq)
-{
-	dynamic_irq_cleanup_x(irq, true);
-}
-
-
 /**
  *	set_irq_chip - set the irq chip for an irq
  *	@irq:	irq number
diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h
index 8eb01e379ccc..f444203a772d 100644
--- a/kernel/irq/internals.h
+++ b/kernel/irq/internals.h
@@ -20,7 +20,6 @@ extern void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume);
 
 extern struct lock_class_key irq_desc_lock_class;
 extern void init_kstat_irqs(struct irq_desc *desc, int node, int nr);
-extern void clear_kstat_irqs(struct irq_desc *desc);
 extern raw_spinlock_t sparse_irq_lock;
 
 /* Resending of interrupts :*/
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 6c71f8ea5d7d..c9d5a1c12874 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -53,12 +53,21 @@ static void desc_smp_init(struct irq_desc *desc, int node)
 {
 	desc->irq_data.node = node;
 	cpumask_copy(desc->irq_data.affinity, irq_default_affinity);
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+	cpumask_clear(desc->pending_mask);
+#endif
+}
+
+static inline int desc_node(struct irq_desc *desc)
+{
+	return desc->irq_data.node;
 }
 
 #else
 static inline int
 alloc_masks(struct irq_desc *desc, gfp_t gfp, int node) { return 0; }
 static inline void desc_smp_init(struct irq_desc *desc, int node) { }
+static inline int desc_node(struct irq_desc *desc) { return 0; }
 #endif
 
 static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node)
@@ -71,6 +80,8 @@ static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node)
 	desc->status = IRQ_DEFAULT_INIT_FLAGS;
 	desc->handle_irq = handle_bad_irq;
 	desc->depth = 1;
+	desc->irq_count = 0;
+	desc->irqs_unhandled = 0;
 	desc->name = NULL;
 	memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
 	desc_smp_init(desc, node);
@@ -286,23 +297,9 @@ struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node)
 	return irq_to_desc(irq);
 }
 
-#ifdef CONFIG_SMP
-static inline int desc_node(struct irq_desc *desc)
-{
-	return desc->irq_data.node;
-}
-#else
-static inline int desc_node(struct irq_desc *desc) { return 0; }
-#endif
-
 static void free_desc(unsigned int irq)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
-	unsigned long flags;
-
-	raw_spin_lock_irqsave(&desc->lock, flags);
-	desc_set_defaults(irq, desc, desc_node(desc));
-	raw_spin_unlock_irqrestore(&desc->lock, flags);
+	dynamic_irq_cleanup(irq);
 }
 
 static inline int alloc_descs(unsigned int start, unsigned int cnt, int node)
@@ -409,10 +406,18 @@ unsigned int irq_get_next_irq(unsigned int offset)
 	return find_next_bit(allocated_irqs, nr_irqs, offset);
 }
 
-/* Statistics access */
-void clear_kstat_irqs(struct irq_desc *desc)
+/**
+ * dynamic_irq_cleanup - cleanup a dynamically allocated irq
+ * @irq:	irq number to initialize
+ */
+void dynamic_irq_cleanup(unsigned int irq)
 {
-	memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
+	struct irq_desc *desc = irq_to_desc(irq);
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&desc->lock, flags);
+	desc_set_defaults(irq, desc, desc_node(desc));
+	raw_spin_unlock_irqrestore(&desc->lock, flags);
 }
 
 unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
-- 
cgit v1.2.3


From 78f90d91f395cd0dc1ef3f21e0c5cd6fd50d202c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 29 Sep 2010 17:18:47 +0200
Subject: genirq: Remove the now unused sparse irq leftovers

The move_irq_desc() function was only used due to the problem that the
allocator did not free the old descriptors. So the descriptors had to
be moved in create_irq_nr(). That's history.

The code would have never been able to move active interrupt
descriptors on affinity settings. That can be done in a completely
different way w/o all this horror.

Remove all of it.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irqdesc.h   |  12 +----
 kernel/irq/Kconfig        |   5 --
 kernel/irq/Makefile       |   1 -
 kernel/irq/internals.h    | 102 ---------------------------------------
 kernel/irq/irqdesc.c      |  30 +-----------
 kernel/irq/numa_migrate.c | 120 ----------------------------------------------
 6 files changed, 4 insertions(+), 266 deletions(-)
 delete mode 100644 kernel/irq/numa_migrate.c

(limited to 'include/linux')

diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h
index f77dc5618d7e..979c68cc7458 100644
--- a/include/linux/irqdesc.h
+++ b/include/linux/irqdesc.h
@@ -82,24 +82,16 @@ struct irq_desc {
 	const char		*name;
 } ____cacheline_internodealigned_in_smp;
 
-extern void arch_init_copy_chip_data(struct irq_desc *old_desc,
-					struct irq_desc *desc, int node);
-extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc);
-
 #ifndef CONFIG_SPARSE_IRQ
 extern struct irq_desc irq_desc[NR_IRQS];
 #endif
 
-#ifdef CONFIG_NUMA_IRQ_DESC
-extern struct irq_desc *move_irq_desc(struct irq_desc *old_desc, int node);
-#else
+/* Will be removed once the last users in power and sh are gone */
+extern struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node);
 static inline struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
 {
 	return desc;
 }
-#endif
-
-extern struct irq_desc *irq_to_desc_alloc_node(unsigned int irq, int node);
 
 #ifdef CONFIG_GENERIC_HARDIRQS
 
diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig
index a42c0191d71a..31d766bf5d2e 100644
--- a/kernel/irq/Kconfig
+++ b/kernel/irq/Kconfig
@@ -26,11 +26,6 @@ config GENERIC_IRQ_PROBE
 config GENERIC_PENDING_IRQ
 	def_bool n
 
-if SPARSE_IRQ && NUMA
-config NUMA_IRQ_DESC
-	def_bool n
-endif
-
 config AUTO_IRQ_AFFINITY
        def_bool n
 
diff --git a/kernel/irq/Makefile b/kernel/irq/Makefile
index 1eaab0da56db..54329cd7b3ee 100644
--- a/kernel/irq/Makefile
+++ b/kernel/irq/Makefile
@@ -3,5 +3,4 @@ obj-y := irqdesc.o handle.o manage.o spurious.o resend.o chip.o dummychip.o devr
 obj-$(CONFIG_GENERIC_IRQ_PROBE) += autoprobe.o
 obj-$(CONFIG_PROC_FS) += proc.o
 obj-$(CONFIG_GENERIC_PENDING_IRQ) += migration.o
-obj-$(CONFIG_NUMA_IRQ_DESC) += numa_migrate.o
 obj-$(CONFIG_PM_SLEEP) += pm.o
diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h
index f444203a772d..4571ae7e085a 100644
--- a/kernel/irq/internals.h
+++ b/kernel/irq/internals.h
@@ -18,17 +18,11 @@ extern int __irq_set_trigger(struct irq_desc *desc, unsigned int irq,
 extern void __disable_irq(struct irq_desc *desc, unsigned int irq, bool susp);
 extern void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume);
 
-extern struct lock_class_key irq_desc_lock_class;
 extern void init_kstat_irqs(struct irq_desc *desc, int node, int nr);
-extern raw_spinlock_t sparse_irq_lock;
 
 /* Resending of interrupts :*/
 void check_irq_resend(struct irq_desc *desc, unsigned int irq);
 
-#ifdef CONFIG_SPARSE_IRQ
-void replace_irq_desc(unsigned int irq, struct irq_desc *desc);
-#endif
-
 #ifdef CONFIG_PROC_FS
 extern void register_irq_proc(unsigned int irq, struct irq_desc *desc);
 extern void unregister_irq_proc(unsigned int irq, struct irq_desc *desc);
@@ -110,99 +104,3 @@ static inline void print_irq_desc(unsigned int irq, struct irq_desc *desc)
 
 #undef P
 
-/* Stuff below will be cleaned up after the sparse allocator is done */
-
-#ifdef CONFIG_SMP
-/**
- * alloc_desc_masks - allocate cpumasks for irq_desc
- * @desc:	pointer to irq_desc struct
- * @node:	node which will be handling the cpumasks
- * @boot:	true if need bootmem
- *
- * Allocates affinity and pending_mask cpumask if required.
- * Returns true if successful (or not required).
- */
-static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
-							bool boot)
-{
-	gfp_t gfp = GFP_ATOMIC;
-
-	if (boot)
-		gfp = GFP_NOWAIT;
-
-#ifdef CONFIG_CPUMASK_OFFSTACK
-	if (!alloc_cpumask_var_node(&desc->irq_data.affinity, gfp, node))
-		return false;
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	if (!alloc_cpumask_var_node(&desc->pending_mask, gfp, node)) {
-		free_cpumask_var(desc->irq_data.affinity);
-		return false;
-	}
-#endif
-#endif
-	return true;
-}
-
-static inline void init_desc_masks(struct irq_desc *desc)
-{
-	cpumask_setall(desc->irq_data.affinity);
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_clear(desc->pending_mask);
-#endif
-}
-
-/**
- * init_copy_desc_masks - copy cpumasks for irq_desc
- * @old_desc:	pointer to old irq_desc struct
- * @new_desc:	pointer to new irq_desc struct
- *
- * Insures affinity and pending_masks are copied to new irq_desc.
- * If !CONFIG_CPUMASKS_OFFSTACK the cpumasks are embedded in the
- * irq_desc struct so the copy is redundant.
- */
-
-static inline void init_copy_desc_masks(struct irq_desc *old_desc,
-					struct irq_desc *new_desc)
-{
-#ifdef CONFIG_CPUMASK_OFFSTACK
-	cpumask_copy(new_desc->irq_data.affinity, old_desc->irq_data.affinity);
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	cpumask_copy(new_desc->pending_mask, old_desc->pending_mask);
-#endif
-#endif
-}
-
-static inline void free_desc_masks(struct irq_desc *old_desc,
-				   struct irq_desc *new_desc)
-{
-	free_cpumask_var(old_desc->irq_data.affinity);
-
-#ifdef CONFIG_GENERIC_PENDING_IRQ
-	free_cpumask_var(old_desc->pending_mask);
-#endif
-}
-
-#else /* !CONFIG_SMP */
-
-static inline bool alloc_desc_masks(struct irq_desc *desc, int node,
-								bool boot)
-{
-	return true;
-}
-
-static inline void init_desc_masks(struct irq_desc *desc)
-{
-}
-
-static inline void init_copy_desc_masks(struct irq_desc *old_desc,
-					struct irq_desc *new_desc)
-{
-}
-
-static inline void free_desc_masks(struct irq_desc *old_desc,
-				   struct irq_desc *new_desc)
-{
-}
-#endif	/* CONFIG_SMP */
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index c9d5a1c12874..4f0b9c9d5c46 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -20,7 +20,7 @@
 /*
  * lockdep: we want to handle all irq_desc locks as a single lock-class:
  */
-struct lock_class_key irq_desc_lock_class;
+static struct lock_class_key irq_desc_lock_class;
 
 #if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS)
 static void __init init_irq_default_affinity(void)
@@ -90,28 +90,11 @@ static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node)
 int nr_irqs = NR_IRQS;
 EXPORT_SYMBOL_GPL(nr_irqs);
 
-DEFINE_RAW_SPINLOCK(sparse_irq_lock);
+static DEFINE_RAW_SPINLOCK(sparse_irq_lock);
 static DECLARE_BITMAP(allocated_irqs, NR_IRQS);
 
 #ifdef CONFIG_SPARSE_IRQ
 
-void __ref init_kstat_irqs(struct irq_desc *desc, int node, int nr)
-{
-	void *ptr;
-
-	ptr = kzalloc_node(nr * sizeof(*desc->kstat_irqs),
-			   GFP_ATOMIC, node);
-
-	/*
-	 * don't overwite if can not get new one
-	 * init_copy_kstat_irqs() could still use old one
-	 */
-	if (ptr) {
-		printk(KERN_DEBUG "  alloc kstat_irqs on node %d\n", node);
-		desc->kstat_irqs = ptr;
-	}
-}
-
 static RADIX_TREE(irq_desc_tree, GFP_ATOMIC);
 
 static void irq_insert_desc(unsigned int irq, struct irq_desc *desc)
@@ -124,15 +107,6 @@ struct irq_desc *irq_to_desc(unsigned int irq)
 	return radix_tree_lookup(&irq_desc_tree, irq);
 }
 
-void replace_irq_desc(unsigned int irq, struct irq_desc *desc)
-{
-	void **ptr;
-
-	ptr = radix_tree_lookup_slot(&irq_desc_tree, irq);
-	if (ptr)
-		radix_tree_replace_slot(ptr, desc);
-}
-
 static void delete_irq_desc(unsigned int irq)
 {
 	radix_tree_delete(&irq_desc_tree, irq);
diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c
deleted file mode 100644
index e7f1f16402c1..000000000000
--- a/kernel/irq/numa_migrate.c
+++ /dev/null
@@ -1,120 +0,0 @@
-/*
- * NUMA irq-desc migration code
- *
- * Migrate IRQ data structures (irq_desc, chip_data, etc.) over to
- * the new "home node" of the IRQ.
- */
-
-#include <linux/irq.h>
-#include <linux/slab.h>
-#include <linux/module.h>
-#include <linux/random.h>
-#include <linux/interrupt.h>
-#include <linux/kernel_stat.h>
-
-#include "internals.h"
-
-static void init_copy_kstat_irqs(struct irq_desc *old_desc,
-				 struct irq_desc *desc,
-				 int node, int nr)
-{
-	init_kstat_irqs(desc, node, nr);
-
-	if (desc->kstat_irqs != old_desc->kstat_irqs)
-		memcpy(desc->kstat_irqs, old_desc->kstat_irqs,
-			 nr * sizeof(*desc->kstat_irqs));
-}
-
-static void free_kstat_irqs(struct irq_desc *old_desc, struct irq_desc *desc)
-{
-	if (old_desc->kstat_irqs == desc->kstat_irqs)
-		return;
-
-	kfree(old_desc->kstat_irqs);
-	old_desc->kstat_irqs = NULL;
-}
-
-static bool init_copy_one_irq_desc(int irq, struct irq_desc *old_desc,
-		 struct irq_desc *desc, int node)
-{
-	memcpy(desc, old_desc, sizeof(struct irq_desc));
-	if (!alloc_desc_masks(desc, node, false)) {
-		printk(KERN_ERR "irq %d: can not get new irq_desc cpumask "
-				"for migration.\n", irq);
-		return false;
-	}
-	raw_spin_lock_init(&desc->lock);
-	desc->irq_data.node = node;
-	lockdep_set_class(&desc->lock, &irq_desc_lock_class);
-	init_copy_kstat_irqs(old_desc, desc, node, nr_cpu_ids);
-	init_copy_desc_masks(old_desc, desc);
-	arch_init_copy_chip_data(old_desc, desc, node);
-	return true;
-}
-
-static void free_one_irq_desc(struct irq_desc *old_desc, struct irq_desc *desc)
-{
-	free_kstat_irqs(old_desc, desc);
-	free_desc_masks(old_desc, desc);
-	arch_free_chip_data(old_desc, desc);
-}
-
-static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc,
-						int node)
-{
-	struct irq_desc *desc;
-	unsigned int irq;
-	unsigned long flags;
-
-	irq = old_desc->irq_data.irq;
-
-	raw_spin_lock_irqsave(&sparse_irq_lock, flags);
-
-	/* We have to check it to avoid races with another CPU */
-	desc = irq_to_desc(irq);
-
-	if (desc && old_desc != desc)
-		goto out_unlock;
-
-	desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node);
-	if (!desc) {
-		printk(KERN_ERR "irq %d: can not get new irq_desc "
-				"for migration.\n", irq);
-		/* still use old one */
-		desc = old_desc;
-		goto out_unlock;
-	}
-	if (!init_copy_one_irq_desc(irq, old_desc, desc, node)) {
-		/* still use old one */
-		kfree(desc);
-		desc = old_desc;
-		goto out_unlock;
-	}
-
-	replace_irq_desc(irq, desc);
-	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
-
-	/* free the old one */
-	free_one_irq_desc(old_desc, desc);
-	kfree(old_desc);
-
-	return desc;
-
-out_unlock:
-	raw_spin_unlock_irqrestore(&sparse_irq_lock, flags);
-
-	return desc;
-}
-
-struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
-{
-	/* those static or target node is -1, do not move them */
-	if (desc->irq_data.irq < NR_IRQS_LEGACY || node == -1)
-		return desc;
-
-	if (desc->irq_data.node != node)
-		desc = __real_move_irq_desc(desc, node);
-
-	return desc;
-}
-
-- 
cgit v1.2.3


From fb3d8eb47ce377d6d7a8fc58b8046ea9eb376a28 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 9 Aug 2010 17:42:21 -0400
Subject: Bluetooth: Support SDIO devices that are AMP controllers

Signed-off-by: David Vrabel <david.vrabel@csr.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Gustavo F. Padovan <padovan@profusion.mobi>
---
 drivers/bluetooth/btsdio.c   | 8 ++++++++
 include/linux/mmc/sdio_ids.h | 1 +
 2 files changed, 9 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/bluetooth/btsdio.c b/drivers/bluetooth/btsdio.c
index 76e5127884f0..792e32d29a1d 100644
--- a/drivers/bluetooth/btsdio.c
+++ b/drivers/bluetooth/btsdio.c
@@ -46,6 +46,9 @@ static const struct sdio_device_id btsdio_table[] = {
 	/* Generic Bluetooth Type-B SDIO device */
 	{ SDIO_DEVICE_CLASS(SDIO_CLASS_BT_B) },
 
+	/* Generic Bluetooth AMP controller */
+	{ SDIO_DEVICE_CLASS(SDIO_CLASS_BT_AMP) },
+
 	{ }	/* Terminating entry */
 };
 
@@ -329,6 +332,11 @@ static int btsdio_probe(struct sdio_func *func,
 	hdev->bus = HCI_SDIO;
 	hdev->driver_data = data;
 
+	if (id->class == SDIO_CLASS_BT_AMP)
+		hdev->dev_type = HCI_AMP;
+	else
+		hdev->dev_type = HCI_BREDR;
+
 	data->hdev = hdev;
 
 	SET_HCIDEV_DEV(hdev, &func->dev);
diff --git a/include/linux/mmc/sdio_ids.h b/include/linux/mmc/sdio_ids.h
index 33b2ea09a4ad..a36ab3bc7b03 100644
--- a/include/linux/mmc/sdio_ids.h
+++ b/include/linux/mmc/sdio_ids.h
@@ -18,6 +18,7 @@
 #define SDIO_CLASS_PHS		0x06	/* PHS standard interface */
 #define SDIO_CLASS_WLAN		0x07	/* WLAN interface */
 #define SDIO_CLASS_ATA		0x08	/* Embedded SDIO-ATA std interface */
+#define SDIO_CLASS_BT_AMP	0x09	/* Type-A Bluetooth AMP interface */
 
 /*
  * Vendors and devices.  Sort key: vendor first, device next.
-- 
cgit v1.2.3


From 29b4433d991c88d86ca48a4c1cc33c671475be4b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 11 Oct 2010 10:22:12 +0000
Subject: net: percpu net_device refcount

We tried very hard to remove all possible dev_hold()/dev_put() pairs in
network stack, using RCU conversions.

There is still an unavoidable device refcount change for every dst we
create/destroy, and this can slow down some workloads (routers or some
app servers, mmap af_packet)

We can switch to a percpu refcount implementation, now dynamic per_cpu
infrastructure is mature. On a 64 cpus machine, this consumes 256 bytes
per device.

On x86, dev_hold(dev) code :

before
        lock    incl 0x280(%ebx)
after:
        movl    0x260(%ebx),%eax
        incl    fs:(%eax)

Stress bench :

(Sending 160.000.000 UDP frames,
IP route cache disabled, dual E5540 @2.53GHz,
32bit kernel, FIB_TRIE)

Before:

real    1m1.662s
user    0m14.373s
sys     12m55.960s

After:

real    0m51.179s
user    0m15.329s
sys     10m15.942s

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/hw/nes/nes_cm.c    |  4 ++--
 drivers/infiniband/hw/nes/nes_verbs.c |  4 ++--
 include/linux/netdevice.h             |  7 +++---
 net/core/dev.c                        | 40 +++++++++++++++++++++++++++++------
 4 files changed, 41 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c
index 61e0efd4ccfb..6220d9d75b58 100644
--- a/drivers/infiniband/hw/nes/nes_cm.c
+++ b/drivers/infiniband/hw/nes/nes_cm.c
@@ -2701,7 +2701,7 @@ static int nes_disconnect(struct nes_qp *nesqp, int abrupt)
 	nesibdev = nesvnic->nesibdev;
 
 	nes_debug(NES_DBG_CM, "netdev refcnt = %u.\n",
-			atomic_read(&nesvnic->netdev->refcnt));
+			netdev_refcnt_read(nesvnic->netdev));
 
 	if (nesqp->active_conn) {
 
@@ -2791,7 +2791,7 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	atomic_inc(&cm_accepts);
 
 	nes_debug(NES_DBG_CM, "netdev refcnt = %u.\n",
-			atomic_read(&nesvnic->netdev->refcnt));
+			netdev_refcnt_read(nesvnic->netdev));
 
 	/* allocate the ietf frame and space for private data */
 	nesqp->ietf_frame = pci_alloc_consistent(nesdev->pcidev,
diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c
index 9046e6675686..546fc22405fe 100644
--- a/drivers/infiniband/hw/nes/nes_verbs.c
+++ b/drivers/infiniband/hw/nes/nes_verbs.c
@@ -785,7 +785,7 @@ static struct ib_pd *nes_alloc_pd(struct ib_device *ibdev,
 
 	nes_debug(NES_DBG_PD, "nesvnic=%p, netdev=%p %s, ibdev=%p, context=%p, netdev refcnt=%u\n",
 			nesvnic, nesdev->netdev[0], nesdev->netdev[0]->name, ibdev, context,
-			atomic_read(&nesvnic->netdev->refcnt));
+			netdev_refcnt_read(nesvnic->netdev));
 
 	err = nes_alloc_resource(nesadapter, nesadapter->allocated_pds,
 			nesadapter->max_pd, &pd_num, &nesadapter->next_pd);
@@ -1416,7 +1416,7 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd,
 	/* update the QP table */
 	nesdev->nesadapter->qp_table[nesqp->hwqp.qp_id-NES_FIRST_QPN] = nesqp;
 	nes_debug(NES_DBG_QP, "netdev refcnt=%u\n",
-			atomic_read(&nesvnic->netdev->refcnt));
+			netdev_refcnt_read(nesvnic->netdev));
 
 	return &nesqp->ibqp;
 }
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 4160db3721ba..14fbb04c459d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1026,7 +1026,7 @@ struct net_device {
 	struct timer_list	watchdog_timer;
 
 	/* Number of references to this device */
-	atomic_t		refcnt ____cacheline_aligned_in_smp;
+	int __percpu		*pcpu_refcnt;
 
 	/* delayed register/unregister */
 	struct list_head	todo_list;
@@ -1330,6 +1330,7 @@ static inline void unregister_netdevice(struct net_device *dev)
 	unregister_netdevice_queue(dev, NULL);
 }
 
+extern int 		netdev_refcnt_read(const struct net_device *dev);
 extern void		free_netdev(struct net_device *dev);
 extern void		synchronize_net(void);
 extern int 		register_netdevice_notifier(struct notifier_block *nb);
@@ -1798,7 +1799,7 @@ extern void netdev_run_todo(void);
  */
 static inline void dev_put(struct net_device *dev)
 {
-	atomic_dec(&dev->refcnt);
+	irqsafe_cpu_dec(*dev->pcpu_refcnt);
 }
 
 /**
@@ -1809,7 +1810,7 @@ static inline void dev_put(struct net_device *dev)
  */
 static inline void dev_hold(struct net_device *dev)
 {
-	atomic_inc(&dev->refcnt);
+	irqsafe_cpu_inc(*dev->pcpu_refcnt);
 }
 
 /* Carrier loss detection, dial on demand. The functions netif_carrier_on
diff --git a/net/core/dev.c b/net/core/dev.c
index 193eafaabd88..04972a4783e2 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5192,9 +5192,6 @@ int init_dummy_netdev(struct net_device *dev)
 	 */
 	dev->reg_state = NETREG_DUMMY;
 
-	/* initialize the ref count */
-	atomic_set(&dev->refcnt, 1);
-
 	/* NAPI wants this */
 	INIT_LIST_HEAD(&dev->napi_list);
 
@@ -5202,6 +5199,11 @@ int init_dummy_netdev(struct net_device *dev)
 	set_bit(__LINK_STATE_PRESENT, &dev->state);
 	set_bit(__LINK_STATE_START, &dev->state);
 
+	/* Note : We dont allocate pcpu_refcnt for dummy devices,
+	 * because users of this 'device' dont need to change
+	 * its refcount.
+	 */
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(init_dummy_netdev);
@@ -5243,6 +5245,16 @@ out:
 }
 EXPORT_SYMBOL(register_netdev);
 
+int netdev_refcnt_read(const struct net_device *dev)
+{
+	int i, refcnt = 0;
+
+	for_each_possible_cpu(i)
+		refcnt += *per_cpu_ptr(dev->pcpu_refcnt, i);
+	return refcnt;
+}
+EXPORT_SYMBOL(netdev_refcnt_read);
+
 /*
  * netdev_wait_allrefs - wait until all references are gone.
  *
@@ -5257,11 +5269,14 @@ EXPORT_SYMBOL(register_netdev);
 static void netdev_wait_allrefs(struct net_device *dev)
 {
 	unsigned long rebroadcast_time, warning_time;
+	int refcnt;
 
 	linkwatch_forget_dev(dev);
 
 	rebroadcast_time = warning_time = jiffies;
-	while (atomic_read(&dev->refcnt) != 0) {
+	refcnt = netdev_refcnt_read(dev);
+
+	while (refcnt != 0) {
 		if (time_after(jiffies, rebroadcast_time + 1 * HZ)) {
 			rtnl_lock();
 
@@ -5288,11 +5303,13 @@ static void netdev_wait_allrefs(struct net_device *dev)
 
 		msleep(250);
 
+		refcnt = netdev_refcnt_read(dev);
+
 		if (time_after(jiffies, warning_time + 10 * HZ)) {
 			printk(KERN_EMERG "unregister_netdevice: "
 			       "waiting for %s to become free. Usage "
 			       "count = %d\n",
-			       dev->name, atomic_read(&dev->refcnt));
+			       dev->name, refcnt);
 			warning_time = jiffies;
 		}
 	}
@@ -5350,7 +5367,7 @@ void netdev_run_todo(void)
 		netdev_wait_allrefs(dev);
 
 		/* paranoia */
-		BUG_ON(atomic_read(&dev->refcnt));
+		BUG_ON(netdev_refcnt_read(dev));
 		WARN_ON(rcu_dereference_raw(dev->ip_ptr));
 		WARN_ON(dev->ip6_ptr);
 		WARN_ON(dev->dn_ptr);
@@ -5520,9 +5537,13 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 	dev = PTR_ALIGN(p, NETDEV_ALIGN);
 	dev->padded = (char *)dev - (char *)p;
 
-	if (dev_addr_init(dev))
+	dev->pcpu_refcnt = alloc_percpu(int);
+	if (!dev->pcpu_refcnt)
 		goto free_tx;
 
+	if (dev_addr_init(dev))
+		goto free_pcpu;
+
 	dev_mc_init(dev);
 	dev_uc_init(dev);
 
@@ -5553,6 +5574,8 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 
 free_tx:
 	kfree(tx);
+free_pcpu:
+	free_percpu(dev->pcpu_refcnt);
 free_p:
 	kfree(p);
 	return NULL;
@@ -5586,6 +5609,9 @@ void free_netdev(struct net_device *dev)
 	list_for_each_entry_safe(p, n, &dev->napi_list, dev_list)
 		netif_napi_del(p);
 
+	free_percpu(dev->pcpu_refcnt);
+	dev->pcpu_refcnt = NULL;
+
 	/*  Compatibility with error handling in drivers */
 	if (dev->reg_state == NETREG_UNINITIALIZED) {
 		kfree((char *)dev - dev->padded);
-- 
cgit v1.2.3


From c7fc2de0c83dbd2eaf759c5cd0e2b9cf1eb4df3a Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Tue, 12 Oct 2010 14:07:09 -0700
Subject: memblock, bootmem: Round pfn properly for memory and reserved regions

We need to round memory regions correctly -- specifically, we need to
round reserved region in the more expansive direction (lower limit
down, upper limit up) whereas usable memory regions need to be rounded
in the more restrictive direction (lower limit up, upper limit down).

This introduces two set of inlines:

	memblock_region_memory_base_pfn()
	memblock_region_memory_end_pfn()
	memblock_region_reserved_base_pfn()
	memblock_region_reserved_end_pfn()

Although they are antisymmetric (and therefore are technically
duplicates) the use of the different inlines explicitly documents the
programmer's intention.

The lack of proper rounding caused a bug on ARM, which was then found
to also affect other architectures.

Reported-by: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4CB4CDFD.4020105@kernel.org>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 arch/arm/mm/init.c       |  8 ++++----
 arch/powerpc/mm/mem.c    | 14 +++++++-------
 arch/powerpc/mm/numa.c   |  4 ++--
 arch/sh/mm/init.c        |  4 ++--
 arch/sparc/mm/init_64.c  |  4 ++--
 include/linux/memblock.h | 25 ++++++++++++-------------
 6 files changed, 29 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index d6022d1f51d1..63f441797c96 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -182,8 +182,8 @@ static void __init arm_bootmem_init(struct meminfo *mi,
 	 * Reserve the memblock reserved regions in bootmem.
 	 */
 	for_each_memblock(reserved, reg) {
-		phys_addr_t start = memblock_region_base_pfn(reg);
-		phys_addr_t end = memblock_region_end_pfn(reg);
+		phys_addr_t start = memblock_region_reserved_base_pfn(reg);
+		phys_addr_t end = memblock_region_reserved_end_pfn(reg);
 		if (start >= start_pfn && end <= end_pfn)
 			reserve_bootmem_node(pgdat, __pfn_to_phys(start),
 					     (end - start) << PAGE_SHIFT,
@@ -251,8 +251,8 @@ static void arm_memory_present(void)
 	struct memblock_region *reg;
 
 	for_each_memblock(memory, reg)
-		memory_present(0, memblock_region_base_pfn(reg),
-			       memblock_region_end_pfn(reg));
+		memory_present(0, memblock_region_memory_base_pfn(reg),
+			       memblock_region_memory_end_pfn(reg));
 }
 #endif
 
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index f661f6c527da..a66499650909 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -148,8 +148,8 @@ walk_system_ram_range(unsigned long start_pfn, unsigned long nr_pages,
 	int ret = -1;
 
 	for_each_memblock(memory, reg) {
-		tstart = max(start_pfn, memblock_region_base_pfn(reg));
-		tend = min(end_pfn, memblock_region_end_pfn(reg));
+		tstart = max(start_pfn, memblock_region_memory_base_pfn(reg));
+		tend = min(end_pfn, memblock_region_memory_end_pfn(reg));
 		if (tstart >= tend)
 			continue;
 		ret = (*func)(tstart, tend - tstart, arg);
@@ -195,8 +195,8 @@ void __init do_init_bootmem(void)
 	/* Add active regions with valid PFNs */
 	for_each_memblock(memory, reg) {
 		unsigned long start_pfn, end_pfn;
-		start_pfn = memblock_region_base_pfn(reg);
-		end_pfn = memblock_region_end_pfn(reg);
+		start_pfn = memblock_region_memory_base_pfn(reg);
+		end_pfn = memblock_region_memory_end_pfn(reg);
 		add_active_range(0, start_pfn, end_pfn);
 	}
 
@@ -236,9 +236,9 @@ static int __init mark_nonram_nosave(void)
 
 	for_each_memblock(memory, reg) {
 		if (prev &&
-		    memblock_region_end_pfn(prev) < memblock_region_base_pfn(reg))
-			register_nosave_region(memblock_region_end_pfn(prev),
-					       memblock_region_base_pfn(reg));
+		    memblock_region_memory_end_pfn(prev) < memblock_region_memory_base_pfn(reg))
+			register_nosave_region(memblock_region_memory_end_pfn(prev),
+					       memblock_region_memory_base_pfn(reg));
 		prev = reg;
 	}
 	return 0;
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index 066fb443ba5a..74505b245374 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -811,8 +811,8 @@ static void __init setup_nonnuma(void)
 	       (top_of_ram - total_ram) >> 20);
 
 	for_each_memblock(memory, reg) {
-		start_pfn = memblock_region_base_pfn(reg);
-		end_pfn = memblock_region_end_pfn(reg);
+		start_pfn = memblock_region_memory_base_pfn(reg);
+		end_pfn = memblock_region_memory_end_pfn(reg);
 
 		fake_numa_create_new_node(end_pfn, &nid);
 		add_active_range(nid, start_pfn, end_pfn);
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index b977475f7446..552bea5113f5 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -244,8 +244,8 @@ static void __init do_init_bootmem(void)
 	/* Add active regions with valid PFNs. */
 	for_each_memblock(memory, reg) {
 		unsigned long start_pfn, end_pfn;
-		start_pfn = memblock_region_base_pfn(reg);
-		end_pfn = memblock_region_end_pfn(reg);
+		start_pfn = memblock_region_memory_base_pfn(reg);
+		end_pfn = memblock_region_memory_end_pfn(reg);
 		__add_active_range(0, start_pfn, end_pfn);
 	}
 
diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index dc584d26d597..4c2572773b55 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -1294,8 +1294,8 @@ static void __init bootmem_init_nonnuma(void)
 		if (!reg->size)
 			continue;
 
-		start_pfn = memblock_region_base_pfn(reg);
-		end_pfn = memblock_region_end_pfn(reg);
+		start_pfn = memblock_region_memory_base_pfn(reg);
+		end_pfn = memblock_region_memory_end_pfn(reg);
 		add_active_range(0, start_pfn, end_pfn);
 	}
 
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 5096458c7535..62a10c2a11f2 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -111,40 +111,39 @@ extern void memblock_set_current_limit(phys_addr_t limit);
  */
 
 /**
- * memblock_region_base_pfn - Return the lowest pfn intersecting with the region
+ * memblock_region_memory_base_pfn - Return the lowest pfn intersecting with the memory region
  * @reg: memblock_region structure
  */
-static inline unsigned long memblock_region_base_pfn(const struct memblock_region *reg)
+static inline unsigned long memblock_region_memory_base_pfn(const struct memblock_region *reg)
 {
-	return reg->base >> PAGE_SHIFT;
+	return PFN_UP(reg->base);
 }
 
 /**
- * memblock_region_last_pfn - Return the highest pfn intersecting with the region
+ * memblock_region_memory_end_pfn - Return the end_pfn this region
  * @reg: memblock_region structure
  */
-static inline unsigned long memblock_region_last_pfn(const struct memblock_region *reg)
+static inline unsigned long memblock_region_memory_end_pfn(const struct memblock_region *reg)
 {
-	return (reg->base + reg->size - 1) >> PAGE_SHIFT;
+	return PFN_DOWN(reg->base + reg->size);
 }
 
 /**
- * memblock_region_end_pfn - Return the pfn of the first page following the region
- *                      but not intersecting it
+ * memblock_region_reserved_base_pfn - Return the lowest pfn intersecting with the reserved region
  * @reg: memblock_region structure
  */
-static inline unsigned long memblock_region_end_pfn(const struct memblock_region *reg)
+static inline unsigned long memblock_region_reserved_base_pfn(const struct memblock_region *reg)
 {
-	return memblock_region_last_pfn(reg) + 1;
+	return PFN_DOWN(reg->base);
 }
 
 /**
- * memblock_region_pages - Return the number of pages covering a region
+ * memblock_region_reserved_end_pfn - Return the end_pfn this region
  * @reg: memblock_region structure
  */
-static inline unsigned long memblock_region_pages(const struct memblock_region *reg)
+static inline unsigned long memblock_region_reserved_end_pfn(const struct memblock_region *reg)
 {
-	return memblock_region_end_pfn(reg) - memblock_region_end_pfn(reg);
+	return PFN_UP(reg->base + reg->size);
 }
 
 #define for_each_memblock(memblock_type, region)					\
-- 
cgit v1.2.3


From b1b6b9aa6fd32db97469e65d301ebc32dcd67992 Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@stericsson.com>
Date: Wed, 29 Sep 2010 17:31:35 +0900
Subject: spi/pl022: add PrimeCell generic DMA support

This extends the PL022 SSP/SPI driver with generic DMA engine
support using the PrimeCell DMA engine interface. Also fix up the
test code for the U300 platform.

Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 drivers/spi/amba-pl022.c   | 516 +++++++++++++++++++++++++++++++++++++--------
 include/linux/amba/pl022.h |   6 +
 2 files changed, 434 insertions(+), 88 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/spi/amba-pl022.c b/drivers/spi/amba-pl022.c
index 8cdddc97325c..db0c67908d2b 100644
--- a/drivers/spi/amba-pl022.c
+++ b/drivers/spi/amba-pl022.c
@@ -27,7 +27,6 @@
 /*
  * TODO:
  * - add timeout on polled transfers
- * - add generic DMA framework support
  */
 
 #include <linux/init.h>
@@ -45,6 +44,9 @@
 #include <linux/amba/pl022.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/scatterlist.h>
 
 /*
  * This macro is used to define some register default values.
@@ -381,6 +383,14 @@ struct pl022 {
 	enum ssp_reading		read;
 	enum ssp_writing		write;
 	u32				exp_fifo_level;
+	/* DMA settings */
+#ifdef CONFIG_DMA_ENGINE
+	struct dma_chan			*dma_rx_channel;
+	struct dma_chan			*dma_tx_channel;
+	struct sg_table			sgt_rx;
+	struct sg_table			sgt_tx;
+	char				*dummypage;
+#endif
 };
 
 /**
@@ -406,7 +416,7 @@ struct chip_data {
 	u16 dmacr;
 	u16 cpsr;
 	u8 n_bytes;
-	u8 enable_dma:1;
+	bool enable_dma;
 	enum ssp_reading read;
 	enum ssp_writing write;
 	void (*cs_control) (u32 command);
@@ -763,6 +773,371 @@ static void *next_transfer(struct pl022 *pl022)
 	}
 	return STATE_DONE;
 }
+
+/*
+ * This DMA functionality is only compiled in if we have
+ * access to the generic DMA devices/DMA engine.
+ */
+#ifdef CONFIG_DMA_ENGINE
+static void unmap_free_dma_scatter(struct pl022 *pl022)
+{
+	/* Unmap and free the SG tables */
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+		     pl022->sgt_tx.nents, DMA_TO_DEVICE);
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+		     pl022->sgt_rx.nents, DMA_FROM_DEVICE);
+	sg_free_table(&pl022->sgt_rx);
+	sg_free_table(&pl022->sgt_tx);
+}
+
+static void dma_callback(void *data)
+{
+	struct pl022 *pl022 = data;
+	struct spi_message *msg = pl022->cur_msg;
+
+	BUG_ON(!pl022->sgt_rx.sgl);
+
+#ifdef VERBOSE_DEBUG
+	/*
+	 * Optionally dump out buffers to inspect contents, this is
+	 * good if you want to convince yourself that the loopback
+	 * read/write contents are the same, when adopting to a new
+	 * DMA engine.
+	 */
+	{
+		struct scatterlist *sg;
+		unsigned int i;
+
+		dma_sync_sg_for_cpu(&pl022->adev->dev,
+				    pl022->sgt_rx.sgl,
+				    pl022->sgt_rx.nents,
+				    DMA_FROM_DEVICE);
+
+		for_each_sg(pl022->sgt_rx.sgl, sg, pl022->sgt_rx.nents, i) {
+			dev_dbg(&pl022->adev->dev, "SPI RX SG ENTRY: %d", i);
+			print_hex_dump(KERN_ERR, "SPI RX: ",
+				       DUMP_PREFIX_OFFSET,
+				       16,
+				       1,
+				       sg_virt(sg),
+				       sg_dma_len(sg),
+				       1);
+		}
+		for_each_sg(pl022->sgt_tx.sgl, sg, pl022->sgt_tx.nents, i) {
+			dev_dbg(&pl022->adev->dev, "SPI TX SG ENTRY: %d", i);
+			print_hex_dump(KERN_ERR, "SPI TX: ",
+				       DUMP_PREFIX_OFFSET,
+				       16,
+				       1,
+				       sg_virt(sg),
+				       sg_dma_len(sg),
+				       1);
+		}
+	}
+#endif
+
+	unmap_free_dma_scatter(pl022);
+
+	/* Update total bytes transfered */
+	msg->actual_length += pl022->cur_transfer->len;
+	if (pl022->cur_transfer->cs_change)
+		pl022->cur_chip->
+			cs_control(SSP_CHIP_DESELECT);
+
+	/* Move to next transfer */
+	msg->state = next_transfer(pl022);
+	tasklet_schedule(&pl022->pump_transfers);
+}
+
+static void setup_dma_scatter(struct pl022 *pl022,
+			      void *buffer,
+			      unsigned int length,
+			      struct sg_table *sgtab)
+{
+	struct scatterlist *sg;
+	int bytesleft = length;
+	void *bufp = buffer;
+	int mapbytes;
+	int i;
+
+	if (buffer) {
+		for_each_sg(sgtab->sgl, sg, sgtab->nents, i) {
+			/*
+			 * If there are less bytes left than what fits
+			 * in the current page (plus page alignment offset)
+			 * we just feed in this, else we stuff in as much
+			 * as we can.
+			 */
+			if (bytesleft < (PAGE_SIZE - offset_in_page(bufp)))
+				mapbytes = bytesleft;
+			else
+				mapbytes = PAGE_SIZE - offset_in_page(bufp);
+			sg_set_page(sg, virt_to_page(bufp),
+				    mapbytes, offset_in_page(bufp));
+			bufp += mapbytes;
+			bytesleft -= mapbytes;
+			dev_dbg(&pl022->adev->dev,
+				"set RX/TX target page @ %p, %d bytes, %d left\n",
+				bufp, mapbytes, bytesleft);
+		}
+	} else {
+		/* Map the dummy buffer on every page */
+		for_each_sg(sgtab->sgl, sg, sgtab->nents, i) {
+			if (bytesleft < PAGE_SIZE)
+				mapbytes = bytesleft;
+			else
+				mapbytes = PAGE_SIZE;
+			sg_set_page(sg, virt_to_page(pl022->dummypage),
+				    mapbytes, 0);
+			bytesleft -= mapbytes;
+			dev_dbg(&pl022->adev->dev,
+				"set RX/TX to dummy page %d bytes, %d left\n",
+				mapbytes, bytesleft);
+
+		}
+	}
+	BUG_ON(bytesleft);
+}
+
+/**
+ * configure_dma - configures the channels for the next transfer
+ * @pl022: SSP driver's private data structure
+ */
+static int configure_dma(struct pl022 *pl022)
+{
+	struct dma_slave_config rx_conf = {
+		.src_addr = SSP_DR(pl022->phybase),
+		.direction = DMA_FROM_DEVICE,
+		.src_maxburst = pl022->vendor->fifodepth >> 1,
+	};
+	struct dma_slave_config tx_conf = {
+		.dst_addr = SSP_DR(pl022->phybase),
+		.direction = DMA_TO_DEVICE,
+		.dst_maxburst = pl022->vendor->fifodepth >> 1,
+	};
+	unsigned int pages;
+	int ret;
+	int sglen;
+	struct dma_chan *rxchan = pl022->dma_rx_channel;
+	struct dma_chan *txchan = pl022->dma_tx_channel;
+	struct dma_async_tx_descriptor *rxdesc;
+	struct dma_async_tx_descriptor *txdesc;
+	dma_cookie_t cookie;
+
+	/* Check that the channels are available */
+	if (!rxchan || !txchan)
+		return -ENODEV;
+
+	switch (pl022->read) {
+	case READING_NULL:
+		/* Use the same as for writing */
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
+		break;
+	case READING_U8:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
+		break;
+	case READING_U16:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+		break;
+	case READING_U32:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+		break;
+	}
+
+	switch (pl022->write) {
+	case WRITING_NULL:
+		/* Use the same as for reading */
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
+		break;
+	case WRITING_U8:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
+		break;
+	case WRITING_U16:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+		break;
+	case WRITING_U32:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;;
+		break;
+	}
+
+	/* SPI pecularity: we need to read and write the same width */
+	if (rx_conf.src_addr_width == DMA_SLAVE_BUSWIDTH_UNDEFINED)
+		rx_conf.src_addr_width = tx_conf.dst_addr_width;
+	if (tx_conf.dst_addr_width == DMA_SLAVE_BUSWIDTH_UNDEFINED)
+		tx_conf.dst_addr_width = rx_conf.src_addr_width;
+	BUG_ON(rx_conf.src_addr_width != tx_conf.dst_addr_width);
+
+	rxchan->device->device_control(rxchan, DMA_SLAVE_CONFIG,
+				       (unsigned long) &rx_conf);
+	txchan->device->device_control(txchan, DMA_SLAVE_CONFIG,
+				       (unsigned long) &tx_conf);
+
+	/* Create sglists for the transfers */
+	pages = (pl022->cur_transfer->len >> PAGE_SHIFT) + 1;
+	dev_dbg(&pl022->adev->dev, "using %d pages for transfer\n", pages);
+
+	ret = sg_alloc_table(&pl022->sgt_rx, pages, GFP_KERNEL);
+	if (ret)
+		goto err_alloc_rx_sg;
+
+	ret = sg_alloc_table(&pl022->sgt_tx, pages, GFP_KERNEL);
+	if (ret)
+		goto err_alloc_tx_sg;
+
+	/* Fill in the scatterlists for the RX+TX buffers */
+	setup_dma_scatter(pl022, pl022->rx,
+			  pl022->cur_transfer->len, &pl022->sgt_rx);
+	setup_dma_scatter(pl022, pl022->tx,
+			  pl022->cur_transfer->len, &pl022->sgt_tx);
+
+	/* Map DMA buffers */
+	sglen = dma_map_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+			   pl022->sgt_rx.nents, DMA_FROM_DEVICE);
+	if (!sglen)
+		goto err_rx_sgmap;
+
+	sglen = dma_map_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+			   pl022->sgt_tx.nents, DMA_TO_DEVICE);
+	if (!sglen)
+		goto err_tx_sgmap;
+
+	/* Send both scatterlists */
+	rxdesc = rxchan->device->device_prep_slave_sg(rxchan,
+				      pl022->sgt_rx.sgl,
+				      pl022->sgt_rx.nents,
+				      DMA_FROM_DEVICE,
+				      DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!rxdesc)
+		goto err_rxdesc;
+
+	txdesc = txchan->device->device_prep_slave_sg(txchan,
+				      pl022->sgt_tx.sgl,
+				      pl022->sgt_tx.nents,
+				      DMA_TO_DEVICE,
+				      DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!txdesc)
+		goto err_txdesc;
+
+	/* Put the callback on the RX transfer only, that should finish last */
+	rxdesc->callback = dma_callback;
+	rxdesc->callback_param = pl022;
+
+	/* Submit and fire RX and TX with TX last so we're ready to read! */
+	cookie = rxdesc->tx_submit(rxdesc);
+	if (dma_submit_error(cookie))
+		goto err_submit_rx;
+	cookie = txdesc->tx_submit(txdesc);
+	if (dma_submit_error(cookie))
+		goto err_submit_tx;
+	rxchan->device->device_issue_pending(rxchan);
+	txchan->device->device_issue_pending(txchan);
+
+	return 0;
+
+err_submit_tx:
+err_submit_rx:
+err_txdesc:
+	txchan->device->device_control(txchan, DMA_TERMINATE_ALL, 0);
+err_rxdesc:
+	rxchan->device->device_control(rxchan, DMA_TERMINATE_ALL, 0);
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+		     pl022->sgt_tx.nents, DMA_TO_DEVICE);
+err_tx_sgmap:
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+		     pl022->sgt_tx.nents, DMA_FROM_DEVICE);
+err_rx_sgmap:
+	sg_free_table(&pl022->sgt_tx);
+err_alloc_tx_sg:
+	sg_free_table(&pl022->sgt_rx);
+err_alloc_rx_sg:
+	return -ENOMEM;
+}
+
+static int __init pl022_dma_probe(struct pl022 *pl022)
+{
+	dma_cap_mask_t mask;
+
+	/* Try to acquire a generic DMA engine slave channel */
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+	/*
+	 * We need both RX and TX channels to do DMA, else do none
+	 * of them.
+	 */
+	pl022->dma_rx_channel = dma_request_channel(mask,
+					    pl022->master_info->dma_filter,
+					    pl022->master_info->dma_rx_param);
+	if (!pl022->dma_rx_channel) {
+		dev_err(&pl022->adev->dev, "no RX DMA channel!\n");
+		goto err_no_rxchan;
+	}
+
+	pl022->dma_tx_channel = dma_request_channel(mask,
+					    pl022->master_info->dma_filter,
+					    pl022->master_info->dma_tx_param);
+	if (!pl022->dma_tx_channel) {
+		dev_err(&pl022->adev->dev, "no TX DMA channel!\n");
+		goto err_no_txchan;
+	}
+
+	pl022->dummypage = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!pl022->dummypage) {
+		dev_err(&pl022->adev->dev, "no DMA dummypage!\n");
+		goto err_no_dummypage;
+	}
+
+	dev_info(&pl022->adev->dev, "setup for DMA on RX %s, TX %s\n",
+		 dma_chan_name(pl022->dma_rx_channel),
+		 dma_chan_name(pl022->dma_tx_channel));
+
+	return 0;
+
+err_no_dummypage:
+	dma_release_channel(pl022->dma_tx_channel);
+err_no_txchan:
+	dma_release_channel(pl022->dma_rx_channel);
+	pl022->dma_rx_channel = NULL;
+err_no_rxchan:
+	return -ENODEV;
+}
+
+static void terminate_dma(struct pl022 *pl022)
+{
+	struct dma_chan *rxchan = pl022->dma_rx_channel;
+	struct dma_chan *txchan = pl022->dma_tx_channel;
+
+	rxchan->device->device_control(rxchan, DMA_TERMINATE_ALL, 0);
+	txchan->device->device_control(txchan, DMA_TERMINATE_ALL, 0);
+	unmap_free_dma_scatter(pl022);
+}
+
+static void pl022_dma_remove(struct pl022 *pl022)
+{
+	if (pl022->busy)
+		terminate_dma(pl022);
+	if (pl022->dma_tx_channel)
+		dma_release_channel(pl022->dma_tx_channel);
+	if (pl022->dma_rx_channel)
+		dma_release_channel(pl022->dma_rx_channel);
+	kfree(pl022->dummypage);
+}
+
+#else
+static inline int configure_dma(struct pl022 *pl022)
+{
+	return -ENODEV;
+}
+
+static inline int pl022_dma_probe(struct pl022 *pl022)
+{
+	return 0;
+}
+
+static inline void pl022_dma_remove(struct pl022 *pl022)
+{
+}
+#endif
+
 /**
  * pl022_interrupt_handler - Interrupt handler for SSP controller
  *
@@ -794,14 +1169,17 @@ static irqreturn_t pl022_interrupt_handler(int irq, void *dev_id)
 	if (unlikely(!irq_status))
 		return IRQ_NONE;
 
-	/* This handles the error code interrupts */
+	/*
+	 * This handles the FIFO interrupts, the timeout
+	 * interrupts are flatly ignored, they cannot be
+	 * trusted.
+	 */
 	if (unlikely(irq_status & SSP_MIS_MASK_RORMIS)) {
 		/*
 		 * Overrun interrupt - bail out since our Data has been
 		 * corrupted
 		 */
-		dev_err(&pl022->adev->dev,
-			"FIFO overrun\n");
+		dev_err(&pl022->adev->dev, "FIFO overrun\n");
 		if (readw(SSP_SR(pl022->virtbase)) & SSP_SR_MASK_RFF)
 			dev_err(&pl022->adev->dev,
 				"RXFIFO is full\n");
@@ -896,8 +1274,8 @@ static int set_up_next_transfer(struct pl022 *pl022,
 }
 
 /**
- * pump_transfers - Tasklet function which schedules next interrupt transfer
- * when running in interrupt transfer mode.
+ * pump_transfers - Tasklet function which schedules next transfer
+ * when running in interrupt or DMA transfer mode.
  * @data: SSP driver private data structure
  *
  */
@@ -954,65 +1332,23 @@ static void pump_transfers(unsigned long data)
 	}
 	/* Flush the FIFOs and let's go! */
 	flush(pl022);
-	writew(ENABLE_ALL_INTERRUPTS, SSP_IMSC(pl022->virtbase));
-}
-
-/**
- * NOT IMPLEMENTED
- * configure_dma - It configures the DMA pipes for DMA transfers
- * @data: SSP driver's private data structure
- *
- */
-static int configure_dma(void *data)
-{
-	struct pl022 *pl022 = data;
-	dev_dbg(&pl022->adev->dev, "configure DMA\n");
-	return -ENOTSUPP;
-}
-
-/**
- * do_dma_transfer - It handles transfers of the current message
- * if it is DMA xfer.
- * NOT FULLY IMPLEMENTED
- * @data: SSP driver's private data structure
- */
-static void do_dma_transfer(void *data)
-{
-	struct pl022 *pl022 = data;
-
-	if (configure_dma(data)) {
-		dev_dbg(&pl022->adev->dev, "configuration of DMA Failed!\n");
-		goto err_config_dma;
-	}
 
-	/* TODO: Implememt DMA setup of pipes here */
-
-	/* Enable target chip, set up transfer */
-	pl022->cur_chip->cs_control(SSP_CHIP_SELECT);
-	if (set_up_next_transfer(pl022, pl022->cur_transfer)) {
-		/* Error path */
-		pl022->cur_msg->state = STATE_ERROR;
-		pl022->cur_msg->status = -EIO;
-		giveback(pl022);
+	if (pl022->cur_chip->enable_dma) {
+		if (configure_dma(pl022)) {
+			dev_dbg(&pl022->adev->dev,
+				"configuration of DMA failed, fall back to interrupt mode\n");
+			goto err_config_dma;
+		}
 		return;
 	}
-	/* Enable SSP */
-	writew((readw(SSP_CR1(pl022->virtbase)) | SSP_CR1_MASK_SSE),
-	       SSP_CR1(pl022->virtbase));
-
-	/* TODO: Enable the DMA transfer here */
-	return;
 
- err_config_dma:
-	pl022->cur_msg->state = STATE_ERROR;
-	pl022->cur_msg->status = -EIO;
-	giveback(pl022);
-	return;
+err_config_dma:
+	writew(ENABLE_ALL_INTERRUPTS, SSP_IMSC(pl022->virtbase));
 }
 
-static void do_interrupt_transfer(void *data)
+static void do_interrupt_dma_transfer(struct pl022 *pl022)
 {
-	struct pl022 *pl022 = data;
+	u32 irqflags = ENABLE_ALL_INTERRUPTS;
 
 	/* Enable target chip */
 	pl022->cur_chip->cs_control(SSP_CHIP_SELECT);
@@ -1023,15 +1359,26 @@ static void do_interrupt_transfer(void *data)
 		giveback(pl022);
 		return;
 	}
+	/* If we're using DMA, set up DMA here */
+	if (pl022->cur_chip->enable_dma) {
+		/* Configure DMA transfer */
+		if (configure_dma(pl022)) {
+			dev_dbg(&pl022->adev->dev,
+				"configuration of DMA failed, fall back to interrupt mode\n");
+			goto err_config_dma;
+		}
+		/* Disable interrupts in DMA mode, IRQ from DMA controller */
+		irqflags = DISABLE_ALL_INTERRUPTS;
+	}
+err_config_dma:
 	/* Enable SSP, turn on interrupts */
 	writew((readw(SSP_CR1(pl022->virtbase)) | SSP_CR1_MASK_SSE),
 	       SSP_CR1(pl022->virtbase));
-	writew(ENABLE_ALL_INTERRUPTS, SSP_IMSC(pl022->virtbase));
+	writew(irqflags, SSP_IMSC(pl022->virtbase));
 }
 
-static void do_polling_transfer(void *data)
+static void do_polling_transfer(struct pl022 *pl022)
 {
-	struct pl022 *pl022 = data;
 	struct spi_message *message = NULL;
 	struct spi_transfer *transfer = NULL;
 	struct spi_transfer *previous = NULL;
@@ -1101,7 +1448,7 @@ static void do_polling_transfer(void *data)
  *
  * This function checks if there is any spi message in the queue that
  * needs processing and delegate control to appropriate function
- * do_polling_transfer()/do_interrupt_transfer()/do_dma_transfer()
+ * do_polling_transfer()/do_interrupt_dma_transfer()
  * based on the kind of the transfer
  *
  */
@@ -1150,10 +1497,8 @@ static void pump_messages(struct work_struct *work)
 
 	if (pl022->cur_chip->xfer_type == POLLING_TRANSFER)
 		do_polling_transfer(pl022);
-	else if (pl022->cur_chip->xfer_type == INTERRUPT_TRANSFER)
-		do_interrupt_transfer(pl022);
 	else
-		do_dma_transfer(pl022);
+		do_interrupt_dma_transfer(pl022);
 }
 
 
@@ -1468,23 +1813,6 @@ static int calculate_effective_freq(struct pl022 *pl022,
 	return 0;
 }
 
-/**
- * NOT IMPLEMENTED
- * process_dma_info - Processes the DMA info provided by client drivers
- * @chip_info: chip info provided by client device
- * @chip: Runtime state maintained by the SSP controller for each spi device
- *
- * This function processes and stores DMA config provided by client driver
- * into the runtime state maintained by the SSP controller driver
- */
-static int process_dma_info(struct pl022_config_chip *chip_info,
-			    struct chip_data *chip)
-{
-	dev_err(chip_info->dev,
-		"cannot process DMA info, DMA not implemented!\n");
-	return -ENOTSUPP;
-}
-
 /**
  * pl022_setup - setup function registered to SPI master framework
  * @spi: spi device which is requesting setup
@@ -1552,8 +1880,6 @@ static int pl022_setup(struct spi_device *spi)
 
 		dev_dbg(&spi->dev, "allocated memory for controller data\n");
 
-		/* Pointer back to the SPI device */
-		chip_info->dev = &spi->dev;
 		/*
 		 * Set controller data default values:
 		 * Polling is supported by default
@@ -1579,6 +1905,9 @@ static int pl022_setup(struct spi_device *spi)
 			"using user supplied controller_data settings\n");
 	}
 
+	/* Pointer back to the SPI device */
+	chip_info->dev = &spi->dev;
+
 	/*
 	 * We can override with custom divisors, else we use the board
 	 * frequency setting
@@ -1637,9 +1966,8 @@ static int pl022_setup(struct spi_device *spi)
 	chip->cpsr = 0;
 	if ((chip_info->com_mode == DMA_TRANSFER)
 	    && ((pl022->master_info)->enable_dma)) {
-		chip->enable_dma = 1;
+		chip->enable_dma = true;
 		dev_dbg(&spi->dev, "DMA mode set in controller state\n");
-		status = process_dma_info(chip_info, chip);
 		if (status < 0)
 			goto err_config_params;
 		SSP_WRITE_BITS(chip->dmacr, SSP_DMA_ENABLED,
@@ -1647,7 +1975,7 @@ static int pl022_setup(struct spi_device *spi)
 		SSP_WRITE_BITS(chip->dmacr, SSP_DMA_ENABLED,
 			       SSP_DMACR_MASK_TXDMAE, 1);
 	} else {
-		chip->enable_dma = 0;
+		chip->enable_dma = false;
 		dev_dbg(&spi->dev, "DMA mode NOT set in controller state\n");
 		SSP_WRITE_BITS(chip->dmacr, SSP_DMA_DISABLED,
 			       SSP_DMACR_MASK_RXDMAE, 0);
@@ -1773,6 +2101,7 @@ pl022_probe(struct amba_device *adev, struct amba_id *id)
 	if (status)
 		goto err_no_ioregion;
 
+	pl022->phybase = adev->res.start;
 	pl022->virtbase = ioremap(adev->res.start, resource_size(&adev->res));
 	if (pl022->virtbase == NULL) {
 		status = -ENOMEM;
@@ -1799,6 +2128,14 @@ pl022_probe(struct amba_device *adev, struct amba_id *id)
 		dev_err(&adev->dev, "probe - cannot get IRQ (%d)\n", status);
 		goto err_no_irq;
 	}
+
+	/* Get DMA channels */
+	if (platform_info->enable_dma) {
+		status = pl022_dma_probe(pl022);
+		if (status != 0)
+			goto err_no_dma;
+	}
+
 	/* Initialize and start queue */
 	status = init_queue(pl022);
 	if (status != 0) {
@@ -1827,6 +2164,8 @@ pl022_probe(struct amba_device *adev, struct amba_id *id)
  err_start_queue:
  err_init_queue:
 	destroy_queue(pl022);
+	pl022_dma_remove(pl022);
+ err_no_dma:
 	free_irq(adev->irq[0], pl022);
  err_no_irq:
 	clk_put(pl022->clk);
@@ -1857,6 +2196,7 @@ pl022_remove(struct amba_device *adev)
 		return status;
 	}
 	load_ssp_default_config(pl022);
+	pl022_dma_remove(pl022);
 	free_irq(adev->irq[0], pl022);
 	clk_disable(pl022->clk);
 	clk_put(pl022->clk);
diff --git a/include/linux/amba/pl022.h b/include/linux/amba/pl022.h
index abf26cc47a2b..db6a191ddcf7 100644
--- a/include/linux/amba/pl022.h
+++ b/include/linux/amba/pl022.h
@@ -228,6 +228,7 @@ enum ssp_chip_select {
 };
 
 
+struct dma_chan;
 /**
  * struct pl022_ssp_master - device.platform_data for SPI controller devices.
  * @num_chipselect: chipselects are used to distinguish individual
@@ -235,11 +236,16 @@ enum ssp_chip_select {
  *     each slave has a chipselect signal, but it's common that not
  *     every chipselect is connected to a slave.
  * @enable_dma: if true enables DMA driven transfers.
+ * @dma_rx_param: parameter to locate an RX DMA channel.
+ * @dma_tx_param: parameter to locate a TX DMA channel.
  */
 struct pl022_ssp_controller {
 	u16 bus_id;
 	u8 num_chipselect;
 	u8 enable_dma:1;
+	bool (*dma_filter)(struct dma_chan *chan, void *filter_param);
+	void *dma_rx_param;
+	void *dma_tx_param;
 };
 
 /**
-- 
cgit v1.2.3


From bde435a9ca376d0b7809768ca803dbf14416b9c1 Mon Sep 17 00:00:00 2001
From: Kevin Wells <wellsk40@gmail.com>
Date: Thu, 16 Sep 2010 06:18:50 -0700
Subject: spi/pl022: Add spi->mode support to AMBA SPI driver

This patch adds spi->mode support for the AMBA pl022 driver and
allows spidev to correctly alter SPI modes. Unused fields used in
the pl022 header file for the pl022_config_chip have been removed.

The ab8500 client driver selects the data transfer size instead
of the platform data.

For platforms that use the amba pl022 driver, the unused fields
in the controller data structure have been removed and the .mode
field in the SPI board info structure is used instead.

Signed-off-by: Kevin Wells <wellsk40@gmail.com>
Tested-by: Linus Walleij <linus.walleij@stericsson.com>
Acked-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 arch/arm/mach-lpc32xx/phy3250.c    |   7 +--
 arch/arm/mach-u300/dummyspichip.c  |   5 +-
 arch/arm/mach-u300/spi.c           |  10 +--
 arch/arm/mach-ux500/board-mop500.c |   8 +--
 drivers/mfd/ab8500-spi.c           |   5 ++
 drivers/spi/amba-pl022.c           | 121 ++++++++++++++++---------------------
 include/linux/amba/pl022.h         |   6 --
 7 files changed, 63 insertions(+), 99 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/mach-lpc32xx/phy3250.c b/arch/arm/mach-lpc32xx/phy3250.c
index bc9a42da2145..0c936cf5675a 100644
--- a/arch/arm/mach-lpc32xx/phy3250.c
+++ b/arch/arm/mach-lpc32xx/phy3250.c
@@ -172,18 +172,12 @@ static void phy3250_spi_cs_set(u32 control)
 }
 
 static struct pl022_config_chip spi0_chip_info = {
-	.lbm			= LOOPBACK_DISABLED,
 	.com_mode		= INTERRUPT_TRANSFER,
 	.iface			= SSP_INTERFACE_MOTOROLA_SPI,
 	.hierarchy		= SSP_MASTER,
 	.slave_tx_disable	= 0,
-	.endian_tx		= SSP_TX_LSB,
-	.endian_rx		= SSP_RX_LSB,
-	.data_size		= SSP_DATA_BITS_8,
 	.rx_lev_trig		= SSP_RX_4_OR_MORE_ELEM,
 	.tx_lev_trig		= SSP_TX_4_OR_MORE_EMPTY_LOC,
-	.clk_phase		= SSP_CLK_FIRST_EDGE,
-	.clk_pol		= SSP_CLK_POL_IDLE_LOW,
 	.ctrl_len		= SSP_BITS_8,
 	.wait_state		= SSP_MWIRE_WAIT_ZERO,
 	.duplex			= SSP_MICROWIRE_CHANNEL_FULL_DUPLEX,
@@ -239,6 +233,7 @@ static int __init phy3250_spi_board_register(void)
 			.max_speed_hz = 5000000,
 			.bus_num = 0,
 			.chip_select = 0,
+			.mode = SPI_MODE_0,
 			.platform_data = &eeprom,
 			.controller_data = &spi0_chip_info,
 		},
diff --git a/arch/arm/mach-u300/dummyspichip.c b/arch/arm/mach-u300/dummyspichip.c
index 5f55012b7c9e..03f793612594 100644
--- a/arch/arm/mach-u300/dummyspichip.c
+++ b/arch/arm/mach-u300/dummyspichip.c
@@ -46,7 +46,6 @@ static ssize_t dummy_looptest(struct device *dev,
 	 * struct, this is just used here to alter the behaviour of the chip
 	 * in order to perform tests.
 	 */
-	struct pl022_config_chip *chip_info = spi->controller_data;
 	int status;
 	u8 txbuf[14] = {0xDE, 0xAD, 0xBE, 0xEF, 0x2B, 0xAD,
 			0xCA, 0xFE, 0xBA, 0xBE, 0xB1, 0x05,
@@ -72,7 +71,7 @@ static ssize_t dummy_looptest(struct device *dev,
 	 * Force chip to 8 bit mode
 	 * WARNING: NEVER DO THIS IN REAL DRIVER CODE, THIS SHOULD BE STATIC!
 	 */
-	chip_info->data_size = SSP_DATA_BITS_8;
+	spi->bits_per_word = 8;
 	/* You should NOT DO THIS EITHER */
 	spi->master->setup(spi);
 
@@ -159,7 +158,7 @@ static ssize_t dummy_looptest(struct device *dev,
 	 * Force chip to 16 bit mode
 	 * WARNING: NEVER DO THIS IN REAL DRIVER CODE, THIS SHOULD BE STATIC!
 	 */
-	chip_info->data_size = SSP_DATA_BITS_16;
+	spi->bits_per_word = 16;
 	/* You should NOT DO THIS EITHER */
 	spi->master->setup(spi);
 
diff --git a/arch/arm/mach-u300/spi.c b/arch/arm/mach-u300/spi.c
index f0e887bea30e..edb2c0d255c2 100644
--- a/arch/arm/mach-u300/spi.c
+++ b/arch/arm/mach-u300/spi.c
@@ -30,8 +30,6 @@ static void select_dummy_chip(u32 chipselect)
 }
 
 struct pl022_config_chip dummy_chip_info = {
-	/* Nominally this is LOOPBACK_DISABLED, but this is our dummy chip! */
-	.lbm = LOOPBACK_ENABLED,
 	/*
 	 * available POLLING_TRANSFER and INTERRUPT_TRANSFER,
 	 * DMA_TRANSFER does not work
@@ -42,14 +40,8 @@ struct pl022_config_chip dummy_chip_info = {
 	.hierarchy = SSP_MASTER,
 	/* 0 = drive TX even as slave, 1 = do not drive TX as slave */
 	.slave_tx_disable = 0,
-	/* LSB first */
-	.endian_tx = SSP_TX_LSB,
-	.endian_rx = SSP_RX_LSB,
-	.data_size = SSP_DATA_BITS_8, /* used to be 12 in some default */
 	.rx_lev_trig = SSP_RX_1_OR_MORE_ELEM,
 	.tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC,
-	.clk_phase = SSP_CLK_SECOND_EDGE,
-	.clk_pol = SSP_CLK_POL_IDLE_LOW,
 	.ctrl_len = SSP_BITS_12,
 	.wait_state = SSP_MWIRE_WAIT_ZERO,
 	.duplex = SSP_MICROWIRE_CHANNEL_FULL_DUPLEX,
@@ -75,7 +67,7 @@ static struct spi_board_info u300_spi_devices[] = {
 		.bus_num        = 0, /* Only one bus on this chip */
 		.chip_select    = 0,
 		/* Means SPI_CS_HIGH, change if e.g low CS */
-		.mode           = 0,
+		.mode           = SPI_MODE_1 | SPI_LSB_FIRST | SPI_LOOP,
 	},
 #endif
 };
diff --git a/arch/arm/mach-ux500/board-mop500.c b/arch/arm/mach-ux500/board-mop500.c
index 0e8fd135a57d..219ae0ca4eef 100644
--- a/arch/arm/mach-ux500/board-mop500.c
+++ b/arch/arm/mach-ux500/board-mop500.c
@@ -55,19 +55,13 @@ static void ab4500_spi_cs_control(u32 command)
 }
 
 struct pl022_config_chip ab4500_chip_info = {
-	.lbm = LOOPBACK_DISABLED,
 	.com_mode = INTERRUPT_TRANSFER,
 	.iface = SSP_INTERFACE_MOTOROLA_SPI,
 	/* we can act as master only */
 	.hierarchy = SSP_MASTER,
 	.slave_tx_disable = 0,
-	.endian_rx = SSP_RX_MSB,
-	.endian_tx = SSP_TX_MSB,
-	.data_size = SSP_DATA_BITS_24,
 	.rx_lev_trig = SSP_RX_1_OR_MORE_ELEM,
 	.tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC,
-	.clk_phase = SSP_CLK_SECOND_EDGE,
-	.clk_pol = SSP_CLK_POL_IDLE_HIGH,
 	.cs_control = ab4500_spi_cs_control,
 };
 
@@ -83,7 +77,7 @@ static struct spi_board_info u8500_spi_devices[] = {
 		.max_speed_hz = 12000000,
 		.bus_num = 0,
 		.chip_select = 0,
-		.mode = SPI_MODE_0,
+		.mode = SPI_MODE_3,
 		.irq = IRQ_DB8500_AB8500,
 	},
 };
diff --git a/drivers/mfd/ab8500-spi.c b/drivers/mfd/ab8500-spi.c
index e1c8b62b086d..01b6d584442c 100644
--- a/drivers/mfd/ab8500-spi.c
+++ b/drivers/mfd/ab8500-spi.c
@@ -83,6 +83,11 @@ static int __devinit ab8500_spi_probe(struct spi_device *spi)
 	struct ab8500 *ab8500;
 	int ret;
 
+	spi->bits_per_word = 24;
+	ret = spi_setup(spi);
+	if (ret < 0)
+		return ret;
+
 	ab8500 = kzalloc(sizeof *ab8500, GFP_KERNEL);
 	if (!ab8500)
 		return -ENOMEM;
diff --git a/drivers/spi/amba-pl022.c b/drivers/spi/amba-pl022.c
index db0c67908d2b..59c90f3ccc26 100644
--- a/drivers/spi/amba-pl022.c
+++ b/drivers/spi/amba-pl022.c
@@ -1595,12 +1595,6 @@ static int destroy_queue(struct pl022 *pl022)
 static int verify_controller_parameters(struct pl022 *pl022,
 					struct pl022_config_chip *chip_info)
 {
-	if ((chip_info->lbm != LOOPBACK_ENABLED)
-	    && (chip_info->lbm != LOOPBACK_DISABLED)) {
-		dev_err(chip_info->dev,
-			"loopback Mode is configured incorrectly\n");
-		return -EINVAL;
-	}
 	if ((chip_info->iface < SSP_INTERFACE_MOTOROLA_SPI)
 	    || (chip_info->iface > SSP_INTERFACE_UNIDIRECTIONAL)) {
 		dev_err(chip_info->dev,
@@ -1626,24 +1620,6 @@ static int verify_controller_parameters(struct pl022 *pl022,
 			"cpsdvsr is configured incorrectly\n");
 		return -EINVAL;
 	}
-	if ((chip_info->endian_rx != SSP_RX_MSB)
-	    && (chip_info->endian_rx != SSP_RX_LSB)) {
-		dev_err(chip_info->dev,
-			"RX FIFO endianess is configured incorrectly\n");
-		return -EINVAL;
-	}
-	if ((chip_info->endian_tx != SSP_TX_MSB)
-	    && (chip_info->endian_tx != SSP_TX_LSB)) {
-		dev_err(chip_info->dev,
-			"TX FIFO endianess is configured incorrectly\n");
-		return -EINVAL;
-	}
-	if ((chip_info->data_size < SSP_DATA_BITS_4)
-	    || (chip_info->data_size > SSP_DATA_BITS_32)) {
-		dev_err(chip_info->dev,
-			"DATA Size is configured incorrectly\n");
-		return -EINVAL;
-	}
 	if ((chip_info->com_mode != INTERRUPT_TRANSFER)
 	    && (chip_info->com_mode != DMA_TRANSFER)
 	    && (chip_info->com_mode != POLLING_TRANSFER)) {
@@ -1663,20 +1639,6 @@ static int verify_controller_parameters(struct pl022 *pl022,
 			"TX FIFO Trigger Level is configured incorrectly\n");
 		return -EINVAL;
 	}
-	if (chip_info->iface == SSP_INTERFACE_MOTOROLA_SPI) {
-		if ((chip_info->clk_phase != SSP_CLK_FIRST_EDGE)
-		    && (chip_info->clk_phase != SSP_CLK_SECOND_EDGE)) {
-			dev_err(chip_info->dev,
-				"Clock Phase is configured incorrectly\n");
-			return -EINVAL;
-		}
-		if ((chip_info->clk_pol != SSP_CLK_POL_IDLE_LOW)
-		    && (chip_info->clk_pol != SSP_CLK_POL_IDLE_HIGH)) {
-			dev_err(chip_info->dev,
-				"Clock Polarity is configured incorrectly\n");
-			return -EINVAL;
-		}
-	}
 	if (chip_info->iface == SSP_INTERFACE_NATIONAL_MICROWIRE) {
 		if ((chip_info->ctrl_len < SSP_BITS_4)
 		    || (chip_info->ctrl_len > SSP_BITS_32)) {
@@ -1825,23 +1787,14 @@ static int calculate_effective_freq(struct pl022 *pl022,
  * controller hardware here, that is not done until the actual transfer
  * commence.
  */
-
-/* FIXME: JUST GUESSING the spi->mode bits understood by this driver */
-#define MODEBITS	(SPI_CPOL | SPI_CPHA | SPI_CS_HIGH \
-			| SPI_LSB_FIRST | SPI_LOOP)
-
 static int pl022_setup(struct spi_device *spi)
 {
 	struct pl022_config_chip *chip_info;
 	struct chip_data *chip;
 	int status = 0;
 	struct pl022 *pl022 = spi_master_get_devdata(spi->master);
-
-	if (spi->mode & ~MODEBITS) {
-		dev_dbg(&spi->dev, "unsupported mode bits %x\n",
-			spi->mode & ~MODEBITS);
-		return -EINVAL;
-	}
+	unsigned int bits = spi->bits_per_word;
+	u32 tmp;
 
 	if (!spi->max_speed_hz)
 		return -EINVAL;
@@ -1884,18 +1837,12 @@ static int pl022_setup(struct spi_device *spi)
 		 * Set controller data default values:
 		 * Polling is supported by default
 		 */
-		chip_info->lbm = LOOPBACK_DISABLED;
 		chip_info->com_mode = POLLING_TRANSFER;
 		chip_info->iface = SSP_INTERFACE_MOTOROLA_SPI;
 		chip_info->hierarchy = SSP_SLAVE;
 		chip_info->slave_tx_disable = DO_NOT_DRIVE_TX;
-		chip_info->endian_tx = SSP_TX_LSB;
-		chip_info->endian_rx = SSP_RX_LSB;
-		chip_info->data_size = SSP_DATA_BITS_12;
 		chip_info->rx_lev_trig = SSP_RX_1_OR_MORE_ELEM;
 		chip_info->tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC;
-		chip_info->clk_phase = SSP_CLK_SECOND_EDGE;
-		chip_info->clk_pol = SSP_CLK_POL_IDLE_LOW;
 		chip_info->ctrl_len = SSP_BITS_8;
 		chip_info->wait_state = SSP_MWIRE_WAIT_ZERO;
 		chip_info->duplex = SSP_MICROWIRE_CHANNEL_FULL_DUPLEX;
@@ -1933,12 +1880,16 @@ static int pl022_setup(struct spi_device *spi)
 	chip->xfer_type = chip_info->com_mode;
 	chip->cs_control = chip_info->cs_control;
 
-	if (chip_info->data_size <= 8) {
-		dev_dbg(&spi->dev, "1 <= n <=8 bits per word\n");
+	if (bits <= 3) {
+		/* PL022 doesn't support less than 4-bits */
+		status = -ENOTSUPP;
+		goto err_config_params;
+	} else if (bits <= 8) {
+		dev_dbg(&spi->dev, "4 <= n <=8 bits per word\n");
 		chip->n_bytes = 1;
 		chip->read = READING_U8;
 		chip->write = WRITING_U8;
-	} else if (chip_info->data_size <= 16) {
+	} else if (bits <= 16) {
 		dev_dbg(&spi->dev, "9 <= n <= 16 bits per word\n");
 		chip->n_bytes = 2;
 		chip->read = READING_U16;
@@ -1955,6 +1906,7 @@ static int pl022_setup(struct spi_device *spi)
 			dev_err(&spi->dev,
 				"a standard pl022 can only handle "
 				"1 <= n <= 16 bit words\n");
+			status = -ENOTSUPP;
 			goto err_config_params;
 		}
 	}
@@ -1987,6 +1939,8 @@ static int pl022_setup(struct spi_device *spi)
 
 	/* Special setup for the ST micro extended control registers */
 	if (pl022->vendor->extended_cr) {
+		u32 etx;
+
 		if (pl022->vendor->pl023) {
 			/* These bits are only in the PL023 */
 			SSP_WRITE_BITS(chip->cr1, chip_info->clkdelay,
@@ -2002,29 +1956,51 @@ static int pl022_setup(struct spi_device *spi)
 			SSP_WRITE_BITS(chip->cr1, chip_info->wait_state,
 				       SSP_CR1_MASK_MWAIT_ST, 6);
 		}
-		SSP_WRITE_BITS(chip->cr0, chip_info->data_size,
+		SSP_WRITE_BITS(chip->cr0, bits - 1,
 			       SSP_CR0_MASK_DSS_ST, 0);
-		SSP_WRITE_BITS(chip->cr1, chip_info->endian_rx,
-			       SSP_CR1_MASK_RENDN_ST, 4);
-		SSP_WRITE_BITS(chip->cr1, chip_info->endian_tx,
-			       SSP_CR1_MASK_TENDN_ST, 5);
+
+		if (spi->mode & SPI_LSB_FIRST) {
+			tmp = SSP_RX_LSB;
+			etx = SSP_TX_LSB;
+		} else {
+			tmp = SSP_RX_MSB;
+			etx = SSP_TX_MSB;
+		}
+		SSP_WRITE_BITS(chip->cr1, tmp, SSP_CR1_MASK_RENDN_ST, 4);
+		SSP_WRITE_BITS(chip->cr1, etx, SSP_CR1_MASK_TENDN_ST, 5);
 		SSP_WRITE_BITS(chip->cr1, chip_info->rx_lev_trig,
 			       SSP_CR1_MASK_RXIFLSEL_ST, 7);
 		SSP_WRITE_BITS(chip->cr1, chip_info->tx_lev_trig,
 			       SSP_CR1_MASK_TXIFLSEL_ST, 10);
 	} else {
-		SSP_WRITE_BITS(chip->cr0, chip_info->data_size,
+		SSP_WRITE_BITS(chip->cr0, bits - 1,
 			       SSP_CR0_MASK_DSS, 0);
 		SSP_WRITE_BITS(chip->cr0, chip_info->iface,
 			       SSP_CR0_MASK_FRF, 4);
 	}
+
 	/* Stuff that is common for all versions */
-	SSP_WRITE_BITS(chip->cr0, chip_info->clk_pol, SSP_CR0_MASK_SPO, 6);
-	SSP_WRITE_BITS(chip->cr0, chip_info->clk_phase, SSP_CR0_MASK_SPH, 7);
+	if (spi->mode & SPI_CPOL)
+		tmp = SSP_CLK_POL_IDLE_HIGH;
+	else
+		tmp = SSP_CLK_POL_IDLE_LOW;
+	SSP_WRITE_BITS(chip->cr0, tmp, SSP_CR0_MASK_SPO, 6);
+
+	if (spi->mode & SPI_CPHA)
+		tmp = SSP_CLK_SECOND_EDGE;
+	else
+		tmp = SSP_CLK_FIRST_EDGE;
+	SSP_WRITE_BITS(chip->cr0, tmp, SSP_CR0_MASK_SPH, 7);
+
 	SSP_WRITE_BITS(chip->cr0, chip_info->clk_freq.scr, SSP_CR0_MASK_SCR, 8);
 	/* Loopback is available on all versions except PL023 */
-	if (!pl022->vendor->pl023)
-		SSP_WRITE_BITS(chip->cr1, chip_info->lbm, SSP_CR1_MASK_LBM, 0);
+	if (!pl022->vendor->pl023) {
+		if (spi->mode & SPI_LOOP)
+			tmp = LOOPBACK_ENABLED;
+		else
+			tmp = LOOPBACK_DISABLED;
+		SSP_WRITE_BITS(chip->cr1, tmp, SSP_CR1_MASK_LBM, 0);
+	}
 	SSP_WRITE_BITS(chip->cr1, SSP_DISABLED, SSP_CR1_MASK_SSE, 1);
 	SSP_WRITE_BITS(chip->cr1, chip_info->hierarchy, SSP_CR1_MASK_MS, 2);
 	SSP_WRITE_BITS(chip->cr1, chip_info->slave_tx_disable, SSP_CR1_MASK_SOD, 3);
@@ -2033,6 +2009,7 @@ static int pl022_setup(struct spi_device *spi)
 	spi_set_ctldata(spi, chip);
 	return status;
  err_config_params:
+	spi_set_ctldata(spi, NULL);
  err_first_setup:
 	kfree(chip);
 	return status;
@@ -2095,6 +2072,14 @@ pl022_probe(struct amba_device *adev, struct amba_id *id)
 	master->setup = pl022_setup;
 	master->transfer = pl022_transfer;
 
+	/*
+	 * Supports mode 0-3, loopback, and active low CS. Transfers are
+	 * always MS bit first on the original pl022.
+	 */
+	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH | SPI_LOOP;
+	if (pl022->vendor->extended_cr)
+		master->mode_bits |= SPI_LSB_FIRST;
+
 	dev_dbg(&adev->dev, "BUSNO: %d\n", master->bus_num);
 
 	status = amba_request_regions(adev, NULL);
diff --git a/include/linux/amba/pl022.h b/include/linux/amba/pl022.h
index db6a191ddcf7..bf143663df81 100644
--- a/include/linux/amba/pl022.h
+++ b/include/linux/amba/pl022.h
@@ -277,19 +277,13 @@ struct pl022_ssp_controller {
  */
 struct pl022_config_chip {
 	struct device *dev;
-	enum ssp_loopback lbm;
 	enum ssp_interface iface;
 	enum ssp_hierarchy hierarchy;
 	bool slave_tx_disable;
 	struct ssp_clock_params clk_freq;
-	enum ssp_rx_endian endian_rx;
-	enum ssp_tx_endian endian_tx;
-	enum ssp_data_size data_size;
 	enum ssp_mode com_mode;
 	enum ssp_rx_level_trig rx_lev_trig;
 	enum ssp_tx_level_trig tx_lev_trig;
-	enum ssp_spi_clk_phase clk_phase;
-	enum ssp_spi_clk_pol clk_pol;
 	enum ssp_microwire_ctrl_len ctrl_len;
 	enum ssp_microwire_wait_state wait_state;
 	enum ssp_duplex duplex;
-- 
cgit v1.2.3


From 5a1c98be1de165c8ad1bd5343a5d779230669489 Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@stericsson.com>
Date: Fri, 1 Oct 2010 11:47:32 +0200
Subject: spi/pl022: get rid of chipinfo dev pointer

What is the dev pointer doing inside the platform data anyway.
We have another pointer to the actual device at hand, use that.

Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 drivers/spi/amba-pl022.c   | 27 ++++++++++++---------------
 include/linux/amba/pl022.h |  1 -
 2 files changed, 12 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/spi/amba-pl022.c b/drivers/spi/amba-pl022.c
index 59c90f3ccc26..19d54aa31a36 100644
--- a/drivers/spi/amba-pl022.c
+++ b/drivers/spi/amba-pl022.c
@@ -1597,58 +1597,58 @@ static int verify_controller_parameters(struct pl022 *pl022,
 {
 	if ((chip_info->iface < SSP_INTERFACE_MOTOROLA_SPI)
 	    || (chip_info->iface > SSP_INTERFACE_UNIDIRECTIONAL)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"interface is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if ((chip_info->iface == SSP_INTERFACE_UNIDIRECTIONAL) &&
 	    (!pl022->vendor->unidir)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"unidirectional mode not supported in this "
 			"hardware version\n");
 		return -EINVAL;
 	}
 	if ((chip_info->hierarchy != SSP_MASTER)
 	    && (chip_info->hierarchy != SSP_SLAVE)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"hierarchy is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if (((chip_info->clk_freq).cpsdvsr < CPSDVR_MIN)
 	    || ((chip_info->clk_freq).cpsdvsr > CPSDVR_MAX)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"cpsdvsr is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if ((chip_info->com_mode != INTERRUPT_TRANSFER)
 	    && (chip_info->com_mode != DMA_TRANSFER)
 	    && (chip_info->com_mode != POLLING_TRANSFER)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"Communication mode is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if ((chip_info->rx_lev_trig < SSP_RX_1_OR_MORE_ELEM)
 	    || (chip_info->rx_lev_trig > SSP_RX_32_OR_MORE_ELEM)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"RX FIFO Trigger Level is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if ((chip_info->tx_lev_trig < SSP_TX_1_OR_MORE_EMPTY_LOC)
 	    || (chip_info->tx_lev_trig > SSP_TX_32_OR_MORE_EMPTY_LOC)) {
-		dev_err(chip_info->dev,
+		dev_err(&pl022->adev->dev,
 			"TX FIFO Trigger Level is configured incorrectly\n");
 		return -EINVAL;
 	}
 	if (chip_info->iface == SSP_INTERFACE_NATIONAL_MICROWIRE) {
 		if ((chip_info->ctrl_len < SSP_BITS_4)
 		    || (chip_info->ctrl_len > SSP_BITS_32)) {
-			dev_err(chip_info->dev,
+			dev_err(&pl022->adev->dev,
 				"CTRL LEN is configured incorrectly\n");
 			return -EINVAL;
 		}
 		if ((chip_info->wait_state != SSP_MWIRE_WAIT_ZERO)
 		    && (chip_info->wait_state != SSP_MWIRE_WAIT_ONE)) {
-			dev_err(chip_info->dev,
+			dev_err(&pl022->adev->dev,
 				"Wait State is configured incorrectly\n");
 			return -EINVAL;
 		}
@@ -1658,13 +1658,13 @@ static int verify_controller_parameters(struct pl022 *pl022,
 			     SSP_MICROWIRE_CHANNEL_FULL_DUPLEX)
 			    && (chip_info->duplex !=
 				SSP_MICROWIRE_CHANNEL_HALF_DUPLEX)) {
-				dev_err(chip_info->dev,
+				dev_err(&pl022->adev->dev,
 					"Microwire duplex mode is configured incorrectly\n");
 				return -EINVAL;
 			}
 		} else {
 			if (chip_info->duplex != SSP_MICROWIRE_CHANNEL_FULL_DUPLEX)
-				dev_err(chip_info->dev,
+				dev_err(&pl022->adev->dev,
 					"Microwire half duplex mode requested,"
 					" but this is only available in the"
 					" ST version of PL022\n");
@@ -1672,7 +1672,7 @@ static int verify_controller_parameters(struct pl022 *pl022,
 		}
 	}
 	if (chip_info->cs_control == NULL) {
-		dev_warn(chip_info->dev,
+		dev_warn(&pl022->adev->dev,
 			"Chip Select Function is NULL for this chip\n");
 		chip_info->cs_control = null_cs_control;
 	}
@@ -1852,9 +1852,6 @@ static int pl022_setup(struct spi_device *spi)
 			"using user supplied controller_data settings\n");
 	}
 
-	/* Pointer back to the SPI device */
-	chip_info->dev = &spi->dev;
-
 	/*
 	 * We can override with custom divisors, else we use the board
 	 * frequency setting
diff --git a/include/linux/amba/pl022.h b/include/linux/amba/pl022.h
index bf143663df81..4ce98f54186b 100644
--- a/include/linux/amba/pl022.h
+++ b/include/linux/amba/pl022.h
@@ -276,7 +276,6 @@ struct pl022_ssp_controller {
  * @dma_config: DMA configuration for SSP controller and peripheral
  */
 struct pl022_config_chip {
-	struct device *dev;
 	enum ssp_interface iface;
 	enum ssp_hierarchy hierarchy;
 	bool slave_tx_disable;
-- 
cgit v1.2.3


From 12b00c2c025b8af697d9a022ea2e928cad889ef1 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@medozas.de>
Date: Wed, 13 Oct 2010 15:56:56 +0200
Subject: netfilter: xtables: resolve indirect macros 1/3

Many of the used macros are just there for userspace compatibility.
Substitute the in-kernel code to directly use the terminal macro
and stuff the defines into #ifndef __KERNEL__ sections.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
---
 include/linux/netfilter_arp/arp_tables.h  | 10 ++++++----
 include/linux/netfilter_ipv4/ip_tables.h  | 10 ++++++----
 include/linux/netfilter_ipv6/ip6_tables.h | 11 ++++++-----
 net/ipv4/netfilter/arp_tables.c           | 10 +++++-----
 net/ipv4/netfilter/ip_tables.c            | 12 ++++++------
 net/ipv6/netfilter/ip6_tables.c           | 12 ++++++------
 6 files changed, 35 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index e9948c0560f6..81938600470d 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -21,8 +21,10 @@
 
 #include <linux/netfilter/x_tables.h>
 
+#ifndef __KERNEL__
 #define ARPT_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN
 #define ARPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN
+#endif
 
 #define ARPT_DEV_ADDR_LEN_MAX 16
 
@@ -134,7 +136,7 @@ struct arpt_entry
 /* The argument to ARPT_SO_GET_INFO */
 struct arpt_getinfo {
 	/* Which table: caller fills this in. */
-	char name[ARPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* Kernel fills these in. */
 	/* Which hook entry points are valid: bitmask */
@@ -156,7 +158,7 @@ struct arpt_getinfo {
 /* The argument to ARPT_SO_SET_REPLACE. */
 struct arpt_replace {
 	/* Which table. */
-	char name[ARPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* Which hook entry points are valid: bitmask.  You can't
            change this. */
@@ -191,7 +193,7 @@ struct arpt_replace {
 /* The argument to ARPT_SO_GET_ENTRIES. */
 struct arpt_get_entries {
 	/* Which table: user fills this in. */
-	char name[ARPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* User fills this in: total entry size. */
 	unsigned int size;
@@ -230,7 +232,7 @@ struct arpt_standard {
 
 struct arpt_error_target {
 	struct arpt_entry_target target;
-	char errorname[ARPT_FUNCTION_MAXNAMELEN];
+	char errorname[XT_FUNCTION_MAXNAMELEN];
 };
 
 struct arpt_error {
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index 704a7b6e8169..1b7cdf1137e3 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -27,12 +27,14 @@
 
 #include <linux/netfilter/x_tables.h>
 
+#ifndef __KERNEL__
 #define IPT_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN
 #define IPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN
 #define ipt_match xt_match
 #define ipt_target xt_target
 #define ipt_table xt_table
 #define ipt_get_revision xt_get_revision
+#endif
 
 /* Yes, Virginia, you have to zero the padding. */
 struct ipt_ip {
@@ -146,7 +148,7 @@ struct ipt_icmp {
 /* The argument to IPT_SO_GET_INFO */
 struct ipt_getinfo {
 	/* Which table: caller fills this in. */
-	char name[IPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* Kernel fills these in. */
 	/* Which hook entry points are valid: bitmask */
@@ -168,7 +170,7 @@ struct ipt_getinfo {
 /* The argument to IPT_SO_SET_REPLACE. */
 struct ipt_replace {
 	/* Which table. */
-	char name[IPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* Which hook entry points are valid: bitmask.  You can't
            change this. */
@@ -202,7 +204,7 @@ struct ipt_replace {
 /* The argument to IPT_SO_GET_ENTRIES. */
 struct ipt_get_entries {
 	/* Which table: user fills this in. */
-	char name[IPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* User fills this in: total entry size. */
 	unsigned int size;
@@ -254,7 +256,7 @@ struct ipt_standard {
 
 struct ipt_error_target {
 	struct ipt_entry_target target;
-	char errorname[IPT_FUNCTION_MAXNAMELEN];
+	char errorname[XT_FUNCTION_MAXNAMELEN];
 };
 
 struct ipt_error {
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index 18442ff19c07..abe31d020e3c 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -27,13 +27,14 @@
 
 #include <linux/netfilter/x_tables.h>
 
+#ifndef __KERNEL__
 #define IP6T_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN
 #define IP6T_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN
-
 #define ip6t_match xt_match
 #define ip6t_target xt_target
 #define ip6t_table xt_table
 #define ip6t_get_revision xt_get_revision
+#endif
 
 /* Yes, Virginia, you have to zero the padding. */
 struct ip6t_ip6 {
@@ -117,7 +118,7 @@ struct ip6t_standard {
 
 struct ip6t_error_target {
 	struct ip6t_entry_target target;
-	char errorname[IP6T_FUNCTION_MAXNAMELEN];
+	char errorname[XT_FUNCTION_MAXNAMELEN];
 };
 
 struct ip6t_error {
@@ -203,7 +204,7 @@ struct ip6t_icmp {
 /* The argument to IP6T_SO_GET_INFO */
 struct ip6t_getinfo {
 	/* Which table: caller fills this in. */
-	char name[IP6T_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* Kernel fills these in. */
 	/* Which hook entry points are valid: bitmask */
@@ -225,7 +226,7 @@ struct ip6t_getinfo {
 /* The argument to IP6T_SO_SET_REPLACE. */
 struct ip6t_replace {
 	/* Which table. */
-	char name[IP6T_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* Which hook entry points are valid: bitmask.  You can't
            change this. */
@@ -259,7 +260,7 @@ struct ip6t_replace {
 /* The argument to IP6T_SO_GET_ENTRIES. */
 struct ip6t_get_entries {
 	/* Which table: user fills this in. */
-	char name[IP6T_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 
 	/* User fills this in: total entry size. */
 	unsigned int size;
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index e8f4f9a57f12..e427a9e3c489 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -895,7 +895,7 @@ static int compat_table_info(const struct xt_table_info *info,
 static int get_info(struct net *net, void __user *user,
                     const int *len, int compat)
 {
-	char name[ARPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 	struct xt_table *t;
 	int ret;
 
@@ -908,7 +908,7 @@ static int get_info(struct net *net, void __user *user,
 	if (copy_from_user(name, user, sizeof(name)) != 0)
 		return -EFAULT;
 
-	name[ARPT_TABLE_MAXNAMELEN-1] = '\0';
+	name[XT_TABLE_MAXNAMELEN-1] = '\0';
 #ifdef CONFIG_COMPAT
 	if (compat)
 		xt_compat_lock(NFPROTO_ARP);
@@ -1474,7 +1474,7 @@ out_unlock:
 }
 
 struct compat_arpt_replace {
-	char				name[ARPT_TABLE_MAXNAMELEN];
+	char				name[XT_TABLE_MAXNAMELEN];
 	u32				valid_hooks;
 	u32				num_entries;
 	u32				size;
@@ -1628,7 +1628,7 @@ static int compat_copy_entries_to_user(unsigned int total_size,
 }
 
 struct compat_arpt_get_entries {
-	char name[ARPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 	compat_uint_t size;
 	struct compat_arpt_entry entrytable[0];
 };
@@ -1840,7 +1840,7 @@ static struct xt_target arpt_builtin_tg[] __read_mostly = {
 	{
 		.name             = ARPT_ERROR_TARGET,
 		.target           = arpt_error,
-		.targetsize       = ARPT_FUNCTION_MAXNAMELEN,
+		.targetsize       = XT_FUNCTION_MAXNAMELEN,
 		.family           = NFPROTO_ARP,
 	},
 };
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index d163f2e3b2e9..2efd41bef452 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -1092,7 +1092,7 @@ static int compat_table_info(const struct xt_table_info *info,
 static int get_info(struct net *net, void __user *user,
                     const int *len, int compat)
 {
-	char name[IPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 	struct xt_table *t;
 	int ret;
 
@@ -1105,7 +1105,7 @@ static int get_info(struct net *net, void __user *user,
 	if (copy_from_user(name, user, sizeof(name)) != 0)
 		return -EFAULT;
 
-	name[IPT_TABLE_MAXNAMELEN-1] = '\0';
+	name[XT_TABLE_MAXNAMELEN-1] = '\0';
 #ifdef CONFIG_COMPAT
 	if (compat)
 		xt_compat_lock(AF_INET);
@@ -1400,7 +1400,7 @@ do_add_counters(struct net *net, const void __user *user,
 
 #ifdef CONFIG_COMPAT
 struct compat_ipt_replace {
-	char			name[IPT_TABLE_MAXNAMELEN];
+	char			name[XT_TABLE_MAXNAMELEN];
 	u32			valid_hooks;
 	u32			num_entries;
 	u32			size;
@@ -1884,7 +1884,7 @@ compat_do_ipt_set_ctl(struct sock *sk,	int cmd, void __user *user,
 }
 
 struct compat_ipt_get_entries {
-	char name[IPT_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 	compat_uint_t size;
 	struct compat_ipt_entry entrytable[0];
 };
@@ -2039,7 +2039,7 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 
 	case IPT_SO_GET_REVISION_MATCH:
 	case IPT_SO_GET_REVISION_TARGET: {
-		struct ipt_get_revision rev;
+		struct xt_get_revision rev;
 		int target;
 
 		if (*len != sizeof(rev)) {
@@ -2188,7 +2188,7 @@ static struct xt_target ipt_builtin_tg[] __read_mostly = {
 	{
 		.name             = IPT_ERROR_TARGET,
 		.target           = ipt_error,
-		.targetsize       = IPT_FUNCTION_MAXNAMELEN,
+		.targetsize       = XT_FUNCTION_MAXNAMELEN,
 		.family           = NFPROTO_IPV4,
 	},
 };
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 8e754be92c24..4b973e13952d 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -1105,7 +1105,7 @@ static int compat_table_info(const struct xt_table_info *info,
 static int get_info(struct net *net, void __user *user,
                     const int *len, int compat)
 {
-	char name[IP6T_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 	struct xt_table *t;
 	int ret;
 
@@ -1118,7 +1118,7 @@ static int get_info(struct net *net, void __user *user,
 	if (copy_from_user(name, user, sizeof(name)) != 0)
 		return -EFAULT;
 
-	name[IP6T_TABLE_MAXNAMELEN-1] = '\0';
+	name[XT_TABLE_MAXNAMELEN-1] = '\0';
 #ifdef CONFIG_COMPAT
 	if (compat)
 		xt_compat_lock(AF_INET6);
@@ -1415,7 +1415,7 @@ do_add_counters(struct net *net, const void __user *user, unsigned int len,
 
 #ifdef CONFIG_COMPAT
 struct compat_ip6t_replace {
-	char			name[IP6T_TABLE_MAXNAMELEN];
+	char			name[XT_TABLE_MAXNAMELEN];
 	u32			valid_hooks;
 	u32			num_entries;
 	u32			size;
@@ -1899,7 +1899,7 @@ compat_do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user,
 }
 
 struct compat_ip6t_get_entries {
-	char name[IP6T_TABLE_MAXNAMELEN];
+	char name[XT_TABLE_MAXNAMELEN];
 	compat_uint_t size;
 	struct compat_ip6t_entry entrytable[0];
 };
@@ -2054,7 +2054,7 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len)
 
 	case IP6T_SO_GET_REVISION_MATCH:
 	case IP6T_SO_GET_REVISION_TARGET: {
-		struct ip6t_get_revision rev;
+		struct xt_get_revision rev;
 		int target;
 
 		if (*len != sizeof(rev)) {
@@ -2203,7 +2203,7 @@ static struct xt_target ip6t_builtin_tg[] __read_mostly = {
 	{
 		.name             = IP6T_ERROR_TARGET,
 		.target           = ip6t_error,
-		.targetsize       = IP6T_FUNCTION_MAXNAMELEN,
+		.targetsize       = XT_FUNCTION_MAXNAMELEN,
 		.family           = NFPROTO_IPV6,
 	},
 };
-- 
cgit v1.2.3


From 87a2e70db62fec7348c6e5545eb7b7650c33d81b Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@medozas.de>
Date: Wed, 13 Oct 2010 16:11:22 +0200
Subject: netfilter: xtables: resolve indirect macros 2/3

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
---
 include/linux/netfilter_arp/arp_tables.h  | 15 ++++-----
 include/linux/netfilter_ipv4/ip_tables.h  | 18 +++++------
 include/linux/netfilter_ipv6/ip6_tables.h | 20 ++++++------
 net/ipv4/netfilter/arp_tables.c           | 38 +++++++++++-----------
 net/ipv4/netfilter/ip_tables.c            | 54 +++++++++++++++----------------
 net/ipv6/netfilter/ip6_tables.c           | 54 +++++++++++++++----------------
 net/sched/act_ipt.c                       | 12 +++----
 7 files changed, 103 insertions(+), 108 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 81938600470d..7e193c9241b3 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -24,6 +24,8 @@
 #ifndef __KERNEL__
 #define ARPT_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN
 #define ARPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN
+#define arpt_entry_target xt_entry_target
+#define arpt_standard_target xt_standard_target
 #endif
 
 #define ARPT_DEV_ADDR_LEN_MAX 16
@@ -65,9 +67,6 @@ struct arpt_arp {
 	u_int16_t invflags;
 };
 
-#define arpt_entry_target xt_entry_target
-#define arpt_standard_target xt_standard_target
-
 /* Values for "flag" field in struct arpt_ip (general arp structure).
  * No flags defined yet.
  */
@@ -208,7 +207,7 @@ struct arpt_get_entries {
 #define ARPT_ERROR_TARGET XT_ERROR_TARGET
 
 /* Helper functions */
-static __inline__ struct arpt_entry_target *arpt_get_target(struct arpt_entry *e)
+static __inline__ struct xt_entry_target *arpt_get_target(struct arpt_entry *e)
 {
 	return (void *)e + e->target_offset;
 }
@@ -227,11 +226,11 @@ static __inline__ struct arpt_entry_target *arpt_get_target(struct arpt_entry *e
 /* Standard entry. */
 struct arpt_standard {
 	struct arpt_entry entry;
-	struct arpt_standard_target target;
+	struct xt_standard_target target;
 };
 
 struct arpt_error_target {
-	struct arpt_entry_target target;
+	struct xt_entry_target target;
 	char errorname[XT_FUNCTION_MAXNAMELEN];
 };
 
@@ -250,7 +249,7 @@ struct arpt_error {
 {									       \
 	.entry		= ARPT_ENTRY_INIT(sizeof(struct arpt_standard)),       \
 	.target		= XT_TARGET_INIT(ARPT_STANDARD_TARGET,		       \
-					 sizeof(struct arpt_standard_target)), \
+					 sizeof(struct xt_standard_target)), \
 	.target.verdict	= -(__verdict) - 1,				       \
 }
 
@@ -287,7 +286,7 @@ struct compat_arpt_entry {
 	unsigned char elems[0];
 };
 
-static inline struct arpt_entry_target *
+static inline struct xt_entry_target *
 compat_arpt_get_target(struct compat_arpt_entry *e)
 {
 	return (void *)e + e->target_offset;
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index 1b7cdf1137e3..ec506918a9b9 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -34,6 +34,10 @@
 #define ipt_target xt_target
 #define ipt_table xt_table
 #define ipt_get_revision xt_get_revision
+#define ipt_entry_match xt_entry_match
+#define ipt_entry_target xt_entry_target
+#define ipt_standard_target xt_standard_target
+#define ipt_counters xt_counters
 #endif
 
 /* Yes, Virginia, you have to zero the padding. */
@@ -54,12 +58,6 @@ struct ipt_ip {
 	u_int8_t invflags;
 };
 
-#define ipt_entry_match xt_entry_match
-#define ipt_entry_target xt_entry_target
-#define ipt_standard_target xt_standard_target
-
-#define ipt_counters xt_counters
-
 /* Values for "flag" field in struct ipt_ip (general ip structure). */
 #define IPT_F_FRAG		0x01	/* Set if rule is a fragment rule */
 #define IPT_F_GOTO		0x02	/* Set if jump is a goto */
@@ -219,7 +217,7 @@ struct ipt_get_entries {
 #define IPT_ERROR_TARGET XT_ERROR_TARGET
 
 /* Helper functions */
-static __inline__ struct ipt_entry_target *
+static __inline__ struct xt_entry_target *
 ipt_get_target(struct ipt_entry *e)
 {
 	return (void *)e + e->target_offset;
@@ -251,11 +249,11 @@ extern void ipt_unregister_table(struct net *net, struct xt_table *table);
 /* Standard entry. */
 struct ipt_standard {
 	struct ipt_entry entry;
-	struct ipt_standard_target target;
+	struct xt_standard_target target;
 };
 
 struct ipt_error_target {
-	struct ipt_entry_target target;
+	struct xt_entry_target target;
 	char errorname[XT_FUNCTION_MAXNAMELEN];
 };
 
@@ -309,7 +307,7 @@ struct compat_ipt_entry {
 };
 
 /* Helper functions */
-static inline struct ipt_entry_target *
+static inline struct xt_entry_target *
 compat_ipt_get_target(struct compat_ipt_entry *e)
 {
 	return (void *)e + e->target_offset;
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index abe31d020e3c..40d11fa05840 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -34,6 +34,10 @@
 #define ip6t_target xt_target
 #define ip6t_table xt_table
 #define ip6t_get_revision xt_get_revision
+#define ip6t_entry_match xt_entry_match
+#define ip6t_entry_target xt_entry_target
+#define ip6t_standard_target xt_standard_target
+#define ip6t_counters xt_counters
 #endif
 
 /* Yes, Virginia, you have to zero the padding. */
@@ -63,12 +67,6 @@ struct ip6t_ip6 {
 	u_int8_t invflags;
 };
 
-#define ip6t_entry_match xt_entry_match
-#define ip6t_entry_target xt_entry_target
-#define ip6t_standard_target xt_standard_target
-
-#define ip6t_counters	xt_counters
-
 /* Values for "flag" field in struct ip6t_ip6 (general ip6 structure). */
 #define IP6T_F_PROTO		0x01	/* Set if rule cares about upper 
 					   protocols */
@@ -113,11 +111,11 @@ struct ip6t_entry {
 /* Standard entry */
 struct ip6t_standard {
 	struct ip6t_entry entry;
-	struct ip6t_standard_target target;
+	struct xt_standard_target target;
 };
 
 struct ip6t_error_target {
-	struct ip6t_entry_target target;
+	struct xt_entry_target target;
 	char errorname[XT_FUNCTION_MAXNAMELEN];
 };
 
@@ -136,7 +134,7 @@ struct ip6t_error {
 {									       \
 	.entry		= IP6T_ENTRY_INIT(sizeof(struct ip6t_standard)),       \
 	.target		= XT_TARGET_INIT(IP6T_STANDARD_TARGET,		       \
-					 sizeof(struct ip6t_standard_target)), \
+					 sizeof(struct xt_standard_target)),   \
 	.target.verdict	= -(__verdict) - 1,				       \
 }
 
@@ -275,7 +273,7 @@ struct ip6t_get_entries {
 #define IP6T_ERROR_TARGET XT_ERROR_TARGET
 
 /* Helper functions */
-static __inline__ struct ip6t_entry_target *
+static __inline__ struct xt_entry_target *
 ip6t_get_target(struct ip6t_entry *e)
 {
 	return (void *)e + e->target_offset;
@@ -332,7 +330,7 @@ struct compat_ip6t_entry {
 	unsigned char elems[0];
 };
 
-static inline struct ip6t_entry_target *
+static inline struct xt_entry_target *
 compat_ip6t_get_target(struct compat_ip6t_entry *e)
 {
 	return (void *)e + e->target_offset;
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index e427a9e3c489..ed178cbe6626 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -228,7 +228,7 @@ arpt_error(struct sk_buff *skb, const struct xt_action_param *par)
 	return NF_DROP;
 }
 
-static inline const struct arpt_entry_target *
+static inline const struct xt_entry_target *
 arpt_get_target_c(const struct arpt_entry *e)
 {
 	return arpt_get_target((struct arpt_entry *)e);
@@ -282,7 +282,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 
 	arp = arp_hdr(skb);
 	do {
-		const struct arpt_entry_target *t;
+		const struct xt_entry_target *t;
 
 		if (!arp_packet_match(arp, skb->dev, indev, outdev, &e->arp)) {
 			e = arpt_next_entry(e);
@@ -297,7 +297,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 		if (!t->u.kernel.target->target) {
 			int v;
 
-			v = ((struct arpt_standard_target *)t)->verdict;
+			v = ((struct xt_standard_target *)t)->verdict;
 			if (v < 0) {
 				/* Pop from stack? */
 				if (v != ARPT_RETURN) {
@@ -377,7 +377,7 @@ static int mark_source_chains(const struct xt_table_info *newinfo,
 		e->counters.pcnt = pos;
 
 		for (;;) {
-			const struct arpt_standard_target *t
+			const struct xt_standard_target *t
 				= (void *)arpt_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
@@ -464,14 +464,14 @@ static int mark_source_chains(const struct xt_table_info *newinfo,
 
 static inline int check_entry(const struct arpt_entry *e, const char *name)
 {
-	const struct arpt_entry_target *t;
+	const struct xt_entry_target *t;
 
 	if (!arp_checkentry(&e->arp)) {
 		duprintf("arp_tables: arp check failed %p %s.\n", e, name);
 		return -EINVAL;
 	}
 
-	if (e->target_offset + sizeof(struct arpt_entry_target) > e->next_offset)
+	if (e->target_offset + sizeof(struct xt_entry_target) > e->next_offset)
 		return -EINVAL;
 
 	t = arpt_get_target_c(e);
@@ -483,7 +483,7 @@ static inline int check_entry(const struct arpt_entry *e, const char *name)
 
 static inline int check_target(struct arpt_entry *e, const char *name)
 {
-	struct arpt_entry_target *t = arpt_get_target(e);
+	struct xt_entry_target *t = arpt_get_target(e);
 	int ret;
 	struct xt_tgchk_param par = {
 		.table     = name,
@@ -506,7 +506,7 @@ static inline int check_target(struct arpt_entry *e, const char *name)
 static inline int
 find_check_entry(struct arpt_entry *e, const char *name, unsigned int size)
 {
-	struct arpt_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	int ret;
 
@@ -536,7 +536,7 @@ out:
 
 static bool check_underflow(const struct arpt_entry *e)
 {
-	const struct arpt_entry_target *t;
+	const struct xt_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->arp))
@@ -544,7 +544,7 @@ static bool check_underflow(const struct arpt_entry *e)
 	t = arpt_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
-	verdict = ((struct arpt_standard_target *)t)->verdict;
+	verdict = ((struct xt_standard_target *)t)->verdict;
 	verdict = -verdict - 1;
 	return verdict == NF_DROP || verdict == NF_ACCEPT;
 }
@@ -566,7 +566,7 @@ static inline int check_entry_size_and_hooks(struct arpt_entry *e,
 	}
 
 	if (e->next_offset
-	    < sizeof(struct arpt_entry) + sizeof(struct arpt_entry_target)) {
+	    < sizeof(struct arpt_entry) + sizeof(struct xt_entry_target)) {
 		duprintf("checking: element %p size %u\n",
 			 e, e->next_offset);
 		return -EINVAL;
@@ -598,7 +598,7 @@ static inline int check_entry_size_and_hooks(struct arpt_entry *e,
 static inline void cleanup_entry(struct arpt_entry *e)
 {
 	struct xt_tgdtor_param par;
-	struct arpt_entry_target *t;
+	struct xt_entry_target *t;
 
 	t = arpt_get_target(e);
 	par.target   = t->u.kernel.target;
@@ -794,7 +794,7 @@ static int copy_entries_to_user(unsigned int total_size,
 	/* FIXME: use iterator macros --RR */
 	/* ... then go back and fix counters and names */
 	for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
-		const struct arpt_entry_target *t;
+		const struct xt_entry_target *t;
 
 		e = (struct arpt_entry *)(loc_cpu_entry + off);
 		if (copy_to_user(userptr + off
@@ -807,7 +807,7 @@ static int copy_entries_to_user(unsigned int total_size,
 
 		t = arpt_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
-				 + offsetof(struct arpt_entry_target,
+				 + offsetof(struct xt_entry_target,
 					    u.user.name),
 				 t->u.kernel.target->name,
 				 strlen(t->u.kernel.target->name)+1) != 0) {
@@ -844,7 +844,7 @@ static int compat_calc_entry(const struct arpt_entry *e,
 			     const struct xt_table_info *info,
 			     const void *base, struct xt_table_info *newinfo)
 {
-	const struct arpt_entry_target *t;
+	const struct xt_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
@@ -1204,7 +1204,7 @@ static int do_add_counters(struct net *net, const void __user *user,
 #ifdef CONFIG_COMPAT
 static inline void compat_release_entry(struct compat_arpt_entry *e)
 {
-	struct arpt_entry_target *t;
+	struct xt_entry_target *t;
 
 	t = compat_arpt_get_target(e);
 	module_put(t->u.kernel.target->me);
@@ -1220,7 +1220,7 @@ check_compat_entry_size_and_hooks(struct compat_arpt_entry *e,
 				  const unsigned int *underflows,
 				  const char *name)
 {
-	struct arpt_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	unsigned int entry_offset;
 	int ret, off, h;
@@ -1288,7 +1288,7 @@ compat_copy_entry_from_user(struct compat_arpt_entry *e, void **dstptr,
 			    unsigned int *size, const char *name,
 			    struct xt_table_info *newinfo, unsigned char *base)
 {
-	struct arpt_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	struct arpt_entry *de;
 	unsigned int origsize;
@@ -1567,7 +1567,7 @@ static int compat_copy_entry_to_user(struct arpt_entry *e, void __user **dstptr,
 				     struct xt_counters *counters,
 				     unsigned int i)
 {
-	struct arpt_entry_target *t;
+	struct xt_entry_target *t;
 	struct compat_arpt_entry __user *ce;
 	u_int16_t target_offset, next_offset;
 	compat_uint_t origsize;
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 2efd41bef452..cb108880050a 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -186,7 +186,7 @@ static inline bool unconditional(const struct ipt_ip *ip)
 }
 
 /* for const-correctness */
-static inline const struct ipt_entry_target *
+static inline const struct xt_entry_target *
 ipt_get_target_c(const struct ipt_entry *e)
 {
 	return ipt_get_target((struct ipt_entry *)e);
@@ -230,7 +230,7 @@ get_chainname_rulenum(const struct ipt_entry *s, const struct ipt_entry *e,
 		      const char *hookname, const char **chainname,
 		      const char **comment, unsigned int *rulenum)
 {
-	const struct ipt_standard_target *t = (void *)ipt_get_target_c(s);
+	const struct xt_standard_target *t = (void *)ipt_get_target_c(s);
 
 	if (strcmp(t->target.u.kernel.target->name, IPT_ERROR_TARGET) == 0) {
 		/* Head of user chain: ERROR target with chainname */
@@ -346,7 +346,7 @@ ipt_do_table(struct sk_buff *skb,
 		 get_entry(table_base, private->underflow[hook]));
 
 	do {
-		const struct ipt_entry_target *t;
+		const struct xt_entry_target *t;
 		const struct xt_entry_match *ematch;
 
 		IP_NF_ASSERT(e);
@@ -380,7 +380,7 @@ ipt_do_table(struct sk_buff *skb,
 		if (!t->u.kernel.target->target) {
 			int v;
 
-			v = ((struct ipt_standard_target *)t)->verdict;
+			v = ((struct xt_standard_target *)t)->verdict;
 			if (v < 0) {
 				/* Pop from stack? */
 				if (v != IPT_RETURN) {
@@ -461,7 +461,7 @@ mark_source_chains(const struct xt_table_info *newinfo,
 		e->counters.pcnt = pos;
 
 		for (;;) {
-			const struct ipt_standard_target *t
+			const struct xt_standard_target *t
 				= (void *)ipt_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
@@ -552,7 +552,7 @@ mark_source_chains(const struct xt_table_info *newinfo,
 	return 1;
 }
 
-static void cleanup_match(struct ipt_entry_match *m, struct net *net)
+static void cleanup_match(struct xt_entry_match *m, struct net *net)
 {
 	struct xt_mtdtor_param par;
 
@@ -568,14 +568,14 @@ static void cleanup_match(struct ipt_entry_match *m, struct net *net)
 static int
 check_entry(const struct ipt_entry *e, const char *name)
 {
-	const struct ipt_entry_target *t;
+	const struct xt_entry_target *t;
 
 	if (!ip_checkentry(&e->ip)) {
 		duprintf("ip check failed %p %s.\n", e, par->match->name);
 		return -EINVAL;
 	}
 
-	if (e->target_offset + sizeof(struct ipt_entry_target) >
+	if (e->target_offset + sizeof(struct xt_entry_target) >
 	    e->next_offset)
 		return -EINVAL;
 
@@ -587,7 +587,7 @@ check_entry(const struct ipt_entry *e, const char *name)
 }
 
 static int
-check_match(struct ipt_entry_match *m, struct xt_mtchk_param *par)
+check_match(struct xt_entry_match *m, struct xt_mtchk_param *par)
 {
 	const struct ipt_ip *ip = par->entryinfo;
 	int ret;
@@ -605,7 +605,7 @@ check_match(struct ipt_entry_match *m, struct xt_mtchk_param *par)
 }
 
 static int
-find_check_match(struct ipt_entry_match *m, struct xt_mtchk_param *par)
+find_check_match(struct xt_entry_match *m, struct xt_mtchk_param *par)
 {
 	struct xt_match *match;
 	int ret;
@@ -630,7 +630,7 @@ err:
 
 static int check_target(struct ipt_entry *e, struct net *net, const char *name)
 {
-	struct ipt_entry_target *t = ipt_get_target(e);
+	struct xt_entry_target *t = ipt_get_target(e);
 	struct xt_tgchk_param par = {
 		.net       = net,
 		.table     = name,
@@ -656,7 +656,7 @@ static int
 find_check_entry(struct ipt_entry *e, struct net *net, const char *name,
 		 unsigned int size)
 {
-	struct ipt_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	int ret;
 	unsigned int j;
@@ -707,7 +707,7 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name,
 
 static bool check_underflow(const struct ipt_entry *e)
 {
-	const struct ipt_entry_target *t;
+	const struct xt_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->ip))
@@ -715,7 +715,7 @@ static bool check_underflow(const struct ipt_entry *e)
 	t = ipt_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
-	verdict = ((struct ipt_standard_target *)t)->verdict;
+	verdict = ((struct xt_standard_target *)t)->verdict;
 	verdict = -verdict - 1;
 	return verdict == NF_DROP || verdict == NF_ACCEPT;
 }
@@ -738,7 +738,7 @@ check_entry_size_and_hooks(struct ipt_entry *e,
 	}
 
 	if (e->next_offset
-	    < sizeof(struct ipt_entry) + sizeof(struct ipt_entry_target)) {
+	    < sizeof(struct ipt_entry) + sizeof(struct xt_entry_target)) {
 		duprintf("checking: element %p size %u\n",
 			 e, e->next_offset);
 		return -EINVAL;
@@ -771,7 +771,7 @@ static void
 cleanup_entry(struct ipt_entry *e, struct net *net)
 {
 	struct xt_tgdtor_param par;
-	struct ipt_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_entry_match *ematch;
 
 	/* Cleanup all matches */
@@ -972,8 +972,8 @@ copy_entries_to_user(unsigned int total_size,
 	/* ... then go back and fix counters and names */
 	for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
 		unsigned int i;
-		const struct ipt_entry_match *m;
-		const struct ipt_entry_target *t;
+		const struct xt_entry_match *m;
+		const struct xt_entry_target *t;
 
 		e = (struct ipt_entry *)(loc_cpu_entry + off);
 		if (copy_to_user(userptr + off
@@ -990,7 +990,7 @@ copy_entries_to_user(unsigned int total_size,
 			m = (void *)e + i;
 
 			if (copy_to_user(userptr + off + i
-					 + offsetof(struct ipt_entry_match,
+					 + offsetof(struct xt_entry_match,
 						    u.user.name),
 					 m->u.kernel.match->name,
 					 strlen(m->u.kernel.match->name)+1)
@@ -1002,7 +1002,7 @@ copy_entries_to_user(unsigned int total_size,
 
 		t = ipt_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
-				 + offsetof(struct ipt_entry_target,
+				 + offsetof(struct xt_entry_target,
 					    u.user.name),
 				 t->u.kernel.target->name,
 				 strlen(t->u.kernel.target->name)+1) != 0) {
@@ -1040,7 +1040,7 @@ static int compat_calc_entry(const struct ipt_entry *e,
 			     const void *base, struct xt_table_info *newinfo)
 {
 	const struct xt_entry_match *ematch;
-	const struct ipt_entry_target *t;
+	const struct xt_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
@@ -1407,7 +1407,7 @@ struct compat_ipt_replace {
 	u32			hook_entry[NF_INET_NUMHOOKS];
 	u32			underflow[NF_INET_NUMHOOKS];
 	u32			num_counters;
-	compat_uptr_t		counters;	/* struct ipt_counters * */
+	compat_uptr_t		counters;	/* struct xt_counters * */
 	struct compat_ipt_entry	entries[0];
 };
 
@@ -1416,7 +1416,7 @@ compat_copy_entry_to_user(struct ipt_entry *e, void __user **dstptr,
 			  unsigned int *size, struct xt_counters *counters,
 			  unsigned int i)
 {
-	struct ipt_entry_target *t;
+	struct xt_entry_target *t;
 	struct compat_ipt_entry __user *ce;
 	u_int16_t target_offset, next_offset;
 	compat_uint_t origsize;
@@ -1451,7 +1451,7 @@ compat_copy_entry_to_user(struct ipt_entry *e, void __user **dstptr,
 }
 
 static int
-compat_find_calc_match(struct ipt_entry_match *m,
+compat_find_calc_match(struct xt_entry_match *m,
 		       const char *name,
 		       const struct ipt_ip *ip,
 		       unsigned int hookmask,
@@ -1473,7 +1473,7 @@ compat_find_calc_match(struct ipt_entry_match *m,
 
 static void compat_release_entry(struct compat_ipt_entry *e)
 {
-	struct ipt_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_entry_match *ematch;
 
 	/* Cleanup all matches */
@@ -1494,7 +1494,7 @@ check_compat_entry_size_and_hooks(struct compat_ipt_entry *e,
 				  const char *name)
 {
 	struct xt_entry_match *ematch;
-	struct ipt_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	unsigned int entry_offset;
 	unsigned int j;
@@ -1576,7 +1576,7 @@ compat_copy_entry_from_user(struct compat_ipt_entry *e, void **dstptr,
 			    unsigned int *size, const char *name,
 			    struct xt_table_info *newinfo, unsigned char *base)
 {
-	struct ipt_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	struct ipt_entry *de;
 	unsigned int origsize;
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 4b973e13952d..c7334c10a4b3 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -215,7 +215,7 @@ static inline bool unconditional(const struct ip6t_ip6 *ipv6)
 	return memcmp(ipv6, &uncond, sizeof(uncond)) == 0;
 }
 
-static inline const struct ip6t_entry_target *
+static inline const struct xt_entry_target *
 ip6t_get_target_c(const struct ip6t_entry *e)
 {
 	return ip6t_get_target((struct ip6t_entry *)e);
@@ -260,7 +260,7 @@ get_chainname_rulenum(const struct ip6t_entry *s, const struct ip6t_entry *e,
 		      const char *hookname, const char **chainname,
 		      const char **comment, unsigned int *rulenum)
 {
-	const struct ip6t_standard_target *t = (void *)ip6t_get_target_c(s);
+	const struct xt_standard_target *t = (void *)ip6t_get_target_c(s);
 
 	if (strcmp(t->target.u.kernel.target->name, IP6T_ERROR_TARGET) == 0) {
 		/* Head of user chain: ERROR target with chainname */
@@ -369,7 +369,7 @@ ip6t_do_table(struct sk_buff *skb,
 	e = get_entry(table_base, private->hook_entry[hook]);
 
 	do {
-		const struct ip6t_entry_target *t;
+		const struct xt_entry_target *t;
 		const struct xt_entry_match *ematch;
 
 		IP_NF_ASSERT(e);
@@ -403,7 +403,7 @@ ip6t_do_table(struct sk_buff *skb,
 		if (!t->u.kernel.target->target) {
 			int v;
 
-			v = ((struct ip6t_standard_target *)t)->verdict;
+			v = ((struct xt_standard_target *)t)->verdict;
 			if (v < 0) {
 				/* Pop from stack? */
 				if (v != IP6T_RETURN) {
@@ -474,7 +474,7 @@ mark_source_chains(const struct xt_table_info *newinfo,
 		e->counters.pcnt = pos;
 
 		for (;;) {
-			const struct ip6t_standard_target *t
+			const struct xt_standard_target *t
 				= (void *)ip6t_get_target_c(e);
 			int visited = e->comefrom & (1 << hook);
 
@@ -565,7 +565,7 @@ mark_source_chains(const struct xt_table_info *newinfo,
 	return 1;
 }
 
-static void cleanup_match(struct ip6t_entry_match *m, struct net *net)
+static void cleanup_match(struct xt_entry_match *m, struct net *net)
 {
 	struct xt_mtdtor_param par;
 
@@ -581,14 +581,14 @@ static void cleanup_match(struct ip6t_entry_match *m, struct net *net)
 static int
 check_entry(const struct ip6t_entry *e, const char *name)
 {
-	const struct ip6t_entry_target *t;
+	const struct xt_entry_target *t;
 
 	if (!ip6_checkentry(&e->ipv6)) {
 		duprintf("ip_tables: ip check failed %p %s.\n", e, name);
 		return -EINVAL;
 	}
 
-	if (e->target_offset + sizeof(struct ip6t_entry_target) >
+	if (e->target_offset + sizeof(struct xt_entry_target) >
 	    e->next_offset)
 		return -EINVAL;
 
@@ -599,7 +599,7 @@ check_entry(const struct ip6t_entry *e, const char *name)
 	return 0;
 }
 
-static int check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par)
+static int check_match(struct xt_entry_match *m, struct xt_mtchk_param *par)
 {
 	const struct ip6t_ip6 *ipv6 = par->entryinfo;
 	int ret;
@@ -618,7 +618,7 @@ static int check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par)
 }
 
 static int
-find_check_match(struct ip6t_entry_match *m, struct xt_mtchk_param *par)
+find_check_match(struct xt_entry_match *m, struct xt_mtchk_param *par)
 {
 	struct xt_match *match;
 	int ret;
@@ -643,7 +643,7 @@ err:
 
 static int check_target(struct ip6t_entry *e, struct net *net, const char *name)
 {
-	struct ip6t_entry_target *t = ip6t_get_target(e);
+	struct xt_entry_target *t = ip6t_get_target(e);
 	struct xt_tgchk_param par = {
 		.net       = net,
 		.table     = name,
@@ -670,7 +670,7 @@ static int
 find_check_entry(struct ip6t_entry *e, struct net *net, const char *name,
 		 unsigned int size)
 {
-	struct ip6t_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	int ret;
 	unsigned int j;
@@ -721,7 +721,7 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name,
 
 static bool check_underflow(const struct ip6t_entry *e)
 {
-	const struct ip6t_entry_target *t;
+	const struct xt_entry_target *t;
 	unsigned int verdict;
 
 	if (!unconditional(&e->ipv6))
@@ -729,7 +729,7 @@ static bool check_underflow(const struct ip6t_entry *e)
 	t = ip6t_get_target_c(e);
 	if (strcmp(t->u.user.name, XT_STANDARD_TARGET) != 0)
 		return false;
-	verdict = ((struct ip6t_standard_target *)t)->verdict;
+	verdict = ((struct xt_standard_target *)t)->verdict;
 	verdict = -verdict - 1;
 	return verdict == NF_DROP || verdict == NF_ACCEPT;
 }
@@ -752,7 +752,7 @@ check_entry_size_and_hooks(struct ip6t_entry *e,
 	}
 
 	if (e->next_offset
-	    < sizeof(struct ip6t_entry) + sizeof(struct ip6t_entry_target)) {
+	    < sizeof(struct ip6t_entry) + sizeof(struct xt_entry_target)) {
 		duprintf("checking: element %p size %u\n",
 			 e, e->next_offset);
 		return -EINVAL;
@@ -784,7 +784,7 @@ check_entry_size_and_hooks(struct ip6t_entry *e,
 static void cleanup_entry(struct ip6t_entry *e, struct net *net)
 {
 	struct xt_tgdtor_param par;
-	struct ip6t_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_entry_match *ematch;
 
 	/* Cleanup all matches */
@@ -985,8 +985,8 @@ copy_entries_to_user(unsigned int total_size,
 	/* ... then go back and fix counters and names */
 	for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){
 		unsigned int i;
-		const struct ip6t_entry_match *m;
-		const struct ip6t_entry_target *t;
+		const struct xt_entry_match *m;
+		const struct xt_entry_target *t;
 
 		e = (struct ip6t_entry *)(loc_cpu_entry + off);
 		if (copy_to_user(userptr + off
@@ -1003,7 +1003,7 @@ copy_entries_to_user(unsigned int total_size,
 			m = (void *)e + i;
 
 			if (copy_to_user(userptr + off + i
-					 + offsetof(struct ip6t_entry_match,
+					 + offsetof(struct xt_entry_match,
 						    u.user.name),
 					 m->u.kernel.match->name,
 					 strlen(m->u.kernel.match->name)+1)
@@ -1015,7 +1015,7 @@ copy_entries_to_user(unsigned int total_size,
 
 		t = ip6t_get_target_c(e);
 		if (copy_to_user(userptr + off + e->target_offset
-				 + offsetof(struct ip6t_entry_target,
+				 + offsetof(struct xt_entry_target,
 					    u.user.name),
 				 t->u.kernel.target->name,
 				 strlen(t->u.kernel.target->name)+1) != 0) {
@@ -1053,7 +1053,7 @@ static int compat_calc_entry(const struct ip6t_entry *e,
 			     const void *base, struct xt_table_info *newinfo)
 {
 	const struct xt_entry_match *ematch;
-	const struct ip6t_entry_target *t;
+	const struct xt_entry_target *t;
 	unsigned int entry_offset;
 	int off, i, ret;
 
@@ -1422,7 +1422,7 @@ struct compat_ip6t_replace {
 	u32			hook_entry[NF_INET_NUMHOOKS];
 	u32			underflow[NF_INET_NUMHOOKS];
 	u32			num_counters;
-	compat_uptr_t		counters;	/* struct ip6t_counters * */
+	compat_uptr_t		counters;	/* struct xt_counters * */
 	struct compat_ip6t_entry entries[0];
 };
 
@@ -1431,7 +1431,7 @@ compat_copy_entry_to_user(struct ip6t_entry *e, void __user **dstptr,
 			  unsigned int *size, struct xt_counters *counters,
 			  unsigned int i)
 {
-	struct ip6t_entry_target *t;
+	struct xt_entry_target *t;
 	struct compat_ip6t_entry __user *ce;
 	u_int16_t target_offset, next_offset;
 	compat_uint_t origsize;
@@ -1466,7 +1466,7 @@ compat_copy_entry_to_user(struct ip6t_entry *e, void __user **dstptr,
 }
 
 static int
-compat_find_calc_match(struct ip6t_entry_match *m,
+compat_find_calc_match(struct xt_entry_match *m,
 		       const char *name,
 		       const struct ip6t_ip6 *ipv6,
 		       unsigned int hookmask,
@@ -1488,7 +1488,7 @@ compat_find_calc_match(struct ip6t_entry_match *m,
 
 static void compat_release_entry(struct compat_ip6t_entry *e)
 {
-	struct ip6t_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_entry_match *ematch;
 
 	/* Cleanup all matches */
@@ -1509,7 +1509,7 @@ check_compat_entry_size_and_hooks(struct compat_ip6t_entry *e,
 				  const char *name)
 {
 	struct xt_entry_match *ematch;
-	struct ip6t_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	unsigned int entry_offset;
 	unsigned int j;
@@ -1591,7 +1591,7 @@ compat_copy_entry_from_user(struct compat_ip6t_entry *e, void **dstptr,
 			    unsigned int *size, const char *name,
 			    struct xt_table_info *newinfo, unsigned char *base)
 {
-	struct ip6t_entry_target *t;
+	struct xt_entry_target *t;
 	struct xt_target *target;
 	struct ip6t_entry *de;
 	unsigned int origsize;
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index c7e59e6ec349..f6d464f993ef 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -39,7 +39,7 @@ static struct tcf_hashinfo ipt_hash_info = {
 	.lock	=	&ipt_lock,
 };
 
-static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int hook)
+static int ipt_init_target(struct xt_entry_target *t, char *table, unsigned int hook)
 {
 	struct xt_tgchk_param par;
 	struct xt_target *target;
@@ -66,7 +66,7 @@ static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int
 	return 0;
 }
 
-static void ipt_destroy_target(struct ipt_entry_target *t)
+static void ipt_destroy_target(struct xt_entry_target *t)
 {
 	struct xt_tgdtor_param par = {
 		.target   = t->u.kernel.target,
@@ -99,7 +99,7 @@ static const struct nla_policy ipt_policy[TCA_IPT_MAX + 1] = {
 	[TCA_IPT_TABLE]	= { .type = NLA_STRING, .len = IFNAMSIZ },
 	[TCA_IPT_HOOK]	= { .type = NLA_U32 },
 	[TCA_IPT_INDEX]	= { .type = NLA_U32 },
-	[TCA_IPT_TARG]	= { .len = sizeof(struct ipt_entry_target) },
+	[TCA_IPT_TARG]	= { .len = sizeof(struct xt_entry_target) },
 };
 
 static int tcf_ipt_init(struct nlattr *nla, struct nlattr *est,
@@ -108,7 +108,7 @@ static int tcf_ipt_init(struct nlattr *nla, struct nlattr *est,
 	struct nlattr *tb[TCA_IPT_MAX + 1];
 	struct tcf_ipt *ipt;
 	struct tcf_common *pc;
-	struct ipt_entry_target *td, *t;
+	struct xt_entry_target *td, *t;
 	char *tname;
 	int ret = 0, err;
 	u32 hook = 0;
@@ -126,7 +126,7 @@ static int tcf_ipt_init(struct nlattr *nla, struct nlattr *est,
 	if (tb[TCA_IPT_TARG] == NULL)
 		return -EINVAL;
 
-	td = (struct ipt_entry_target *)nla_data(tb[TCA_IPT_TARG]);
+	td = (struct xt_entry_target *)nla_data(tb[TCA_IPT_TARG]);
 	if (nla_len(tb[TCA_IPT_TARG]) < td->u.target_size)
 		return -EINVAL;
 
@@ -249,7 +249,7 @@ static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int
 {
 	unsigned char *b = skb_tail_pointer(skb);
 	struct tcf_ipt *ipt = a->priv;
-	struct ipt_entry_target *t;
+	struct xt_entry_target *t;
 	struct tcf_t tm;
 	struct tc_cnt c;
 
-- 
cgit v1.2.3


From 243bf6e29eef642de0ff62f1ebf58bc2396d6d6e Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@medozas.de>
Date: Wed, 13 Oct 2010 16:28:00 +0200
Subject: netfilter: xtables: resolve indirect macros 3/3

---
 include/linux/netfilter_arp/arp_tables.h  | 33 +++++----------
 include/linux/netfilter_ipv4/ip_tables.h  | 69 ++++++++++++++-----------------
 include/linux/netfilter_ipv6/ip6_tables.h | 69 +++++++++++--------------------
 net/ipv4/netfilter/arp_tables.c           | 14 +++----
 net/ipv4/netfilter/arpt_mangle.c          |  2 +-
 net/ipv4/netfilter/ip_tables.c            | 18 ++++----
 net/ipv6/netfilter/ip6_tables.c           | 18 ++++----
 net/sched/act_ipt.c                       |  2 +-
 8 files changed, 94 insertions(+), 131 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 7e193c9241b3..6e2341a811d6 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -26,6 +26,14 @@
 #define ARPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN
 #define arpt_entry_target xt_entry_target
 #define arpt_standard_target xt_standard_target
+#define ARPT_CONTINUE XT_CONTINUE
+#define ARPT_RETURN XT_RETURN
+#define arpt_counters_info xt_counters_info
+#define arpt_counters xt_counters
+#define ARPT_STANDARD_TARGET XT_STANDARD_TARGET
+#define ARPT_ERROR_TARGET XT_ERROR_TARGET
+#define ARPT_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct arpt_entry, entries, size, fn, ## args)
 #endif
 
 #define ARPT_DEV_ADDR_LEN_MAX 16
@@ -126,12 +134,6 @@ struct arpt_entry
 #define ARPT_SO_GET_REVISION_TARGET	(ARPT_BASE_CTL + 3)
 #define ARPT_SO_GET_MAX			(ARPT_SO_GET_REVISION_TARGET)
 
-/* CONTINUE verdict for targets */
-#define ARPT_CONTINUE XT_CONTINUE
-
-/* For standard target */
-#define ARPT_RETURN XT_RETURN
-
 /* The argument to ARPT_SO_GET_INFO */
 struct arpt_getinfo {
 	/* Which table: caller fills this in. */
@@ -185,10 +187,6 @@ struct arpt_replace {
 	struct arpt_entry entries[0];
 };
 
-/* The argument to ARPT_SO_ADD_COUNTERS. */
-#define arpt_counters_info xt_counters_info
-#define arpt_counters xt_counters
-
 /* The argument to ARPT_SO_GET_ENTRIES. */
 struct arpt_get_entries {
 	/* Which table: user fills this in. */
@@ -201,23 +199,12 @@ struct arpt_get_entries {
 	struct arpt_entry entrytable[0];
 };
 
-/* Standard return verdict, or do jump. */
-#define ARPT_STANDARD_TARGET XT_STANDARD_TARGET
-/* Error verdict. */
-#define ARPT_ERROR_TARGET XT_ERROR_TARGET
-
 /* Helper functions */
 static __inline__ struct xt_entry_target *arpt_get_target(struct arpt_entry *e)
 {
 	return (void *)e + e->target_offset;
 }
 
-#ifndef __KERNEL__
-/* fn returns 0 to continue iteration */
-#define ARPT_ENTRY_ITERATE(entries, size, fn, args...) \
-	XT_ENTRY_ITERATE(struct arpt_entry, entries, size, fn, ## args)
-#endif
-
 /*
  *	Main firewall chains definitions and global var's definitions.
  */
@@ -248,7 +235,7 @@ struct arpt_error {
 #define ARPT_STANDARD_INIT(__verdict)					       \
 {									       \
 	.entry		= ARPT_ENTRY_INIT(sizeof(struct arpt_standard)),       \
-	.target		= XT_TARGET_INIT(ARPT_STANDARD_TARGET,		       \
+	.target		= XT_TARGET_INIT(XT_STANDARD_TARGET,		       \
 					 sizeof(struct xt_standard_target)), \
 	.target.verdict	= -(__verdict) - 1,				       \
 }
@@ -256,7 +243,7 @@ struct arpt_error {
 #define ARPT_ERROR_INIT							       \
 {									       \
 	.entry		= ARPT_ENTRY_INIT(sizeof(struct arpt_error)),	       \
-	.target		= XT_TARGET_INIT(ARPT_ERROR_TARGET,		       \
+	.target		= XT_TARGET_INIT(XT_ERROR_TARGET,		       \
 					 sizeof(struct arpt_error_target)),    \
 	.target.errorname = "ERROR",					       \
 }
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index ec506918a9b9..ee54b3b7e237 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -38,6 +38,36 @@
 #define ipt_entry_target xt_entry_target
 #define ipt_standard_target xt_standard_target
 #define ipt_counters xt_counters
+#define IPT_CONTINUE XT_CONTINUE
+#define IPT_RETURN XT_RETURN
+
+/* This group is older than old (iptables < v1.4.0-rc1~89) */
+#include <linux/netfilter/xt_tcpudp.h>
+#define ipt_udp xt_udp
+#define ipt_tcp xt_tcp
+#define IPT_TCP_INV_SRCPT	XT_TCP_INV_SRCPT
+#define IPT_TCP_INV_DSTPT	XT_TCP_INV_DSTPT
+#define IPT_TCP_INV_FLAGS	XT_TCP_INV_FLAGS
+#define IPT_TCP_INV_OPTION	XT_TCP_INV_OPTION
+#define IPT_TCP_INV_MASK	XT_TCP_INV_MASK
+#define IPT_UDP_INV_SRCPT	XT_UDP_INV_SRCPT
+#define IPT_UDP_INV_DSTPT	XT_UDP_INV_DSTPT
+#define IPT_UDP_INV_MASK	XT_UDP_INV_MASK
+
+/* The argument to IPT_SO_ADD_COUNTERS. */
+#define ipt_counters_info xt_counters_info
+/* Standard return verdict, or do jump. */
+#define IPT_STANDARD_TARGET XT_STANDARD_TARGET
+/* Error verdict. */
+#define IPT_ERROR_TARGET XT_ERROR_TARGET
+
+/* fn returns 0 to continue iteration */
+#define IPT_MATCH_ITERATE(e, fn, args...) \
+	XT_MATCH_ITERATE(struct ipt_entry, e, fn, ## args)
+
+/* fn returns 0 to continue iteration */
+#define IPT_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct ipt_entry, entries, size, fn, ## args)
 #endif
 
 /* Yes, Virginia, you have to zero the padding. */
@@ -116,23 +146,6 @@ struct ipt_entry {
 #define IPT_SO_GET_REVISION_TARGET	(IPT_BASE_CTL + 3)
 #define IPT_SO_GET_MAX			IPT_SO_GET_REVISION_TARGET
 
-#define IPT_CONTINUE XT_CONTINUE
-#define IPT_RETURN XT_RETURN
-
-#include <linux/netfilter/xt_tcpudp.h>
-#define ipt_udp xt_udp
-#define ipt_tcp xt_tcp
-
-#define IPT_TCP_INV_SRCPT	XT_TCP_INV_SRCPT
-#define IPT_TCP_INV_DSTPT	XT_TCP_INV_DSTPT
-#define IPT_TCP_INV_FLAGS	XT_TCP_INV_FLAGS
-#define IPT_TCP_INV_OPTION	XT_TCP_INV_OPTION
-#define IPT_TCP_INV_MASK	XT_TCP_INV_MASK
-
-#define IPT_UDP_INV_SRCPT	XT_UDP_INV_SRCPT
-#define IPT_UDP_INV_DSTPT	XT_UDP_INV_DSTPT
-#define IPT_UDP_INV_MASK	XT_UDP_INV_MASK
-
 /* ICMP matching stuff */
 struct ipt_icmp {
 	u_int8_t type;				/* type to match */
@@ -196,9 +209,6 @@ struct ipt_replace {
 	struct ipt_entry entries[0];
 };
 
-/* The argument to IPT_SO_ADD_COUNTERS. */
-#define ipt_counters_info xt_counters_info
-
 /* The argument to IPT_SO_GET_ENTRIES. */
 struct ipt_get_entries {
 	/* Which table: user fills this in. */
@@ -211,11 +221,6 @@ struct ipt_get_entries {
 	struct ipt_entry entrytable[0];
 };
 
-/* Standard return verdict, or do jump. */
-#define IPT_STANDARD_TARGET XT_STANDARD_TARGET
-/* Error verdict. */
-#define IPT_ERROR_TARGET XT_ERROR_TARGET
-
 /* Helper functions */
 static __inline__ struct xt_entry_target *
 ipt_get_target(struct ipt_entry *e)
@@ -223,16 +228,6 @@ ipt_get_target(struct ipt_entry *e)
 	return (void *)e + e->target_offset;
 }
 
-#ifndef __KERNEL__
-/* fn returns 0 to continue iteration */
-#define IPT_MATCH_ITERATE(e, fn, args...) \
-	XT_MATCH_ITERATE(struct ipt_entry, e, fn, ## args)
-
-/* fn returns 0 to continue iteration */
-#define IPT_ENTRY_ITERATE(entries, size, fn, args...) \
-	XT_ENTRY_ITERATE(struct ipt_entry, entries, size, fn, ## args)
-#endif
-
 /*
  *	Main firewall chains definitions and global var's definitions.
  */
@@ -271,7 +266,7 @@ struct ipt_error {
 #define IPT_STANDARD_INIT(__verdict)					       \
 {									       \
 	.entry		= IPT_ENTRY_INIT(sizeof(struct ipt_standard)),	       \
-	.target		= XT_TARGET_INIT(IPT_STANDARD_TARGET,		       \
+	.target		= XT_TARGET_INIT(XT_STANDARD_TARGET,		       \
 					 sizeof(struct xt_standard_target)),   \
 	.target.verdict	= -(__verdict) - 1,				       \
 }
@@ -279,7 +274,7 @@ struct ipt_error {
 #define IPT_ERROR_INIT							       \
 {									       \
 	.entry		= IPT_ENTRY_INIT(sizeof(struct ipt_error)),	       \
-	.target		= XT_TARGET_INIT(IPT_ERROR_TARGET,		       \
+	.target		= XT_TARGET_INIT(XT_ERROR_TARGET,		       \
 					 sizeof(struct ipt_error_target)),     \
 	.target.errorname = "ERROR",					       \
 }
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index 40d11fa05840..ac2b411ea63a 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -38,6 +38,29 @@
 #define ip6t_entry_target xt_entry_target
 #define ip6t_standard_target xt_standard_target
 #define ip6t_counters xt_counters
+#define IP6T_CONTINUE XT_CONTINUE
+#define IP6T_RETURN XT_RETURN
+
+/* Pre-iptables-1.4.0 */
+#include <linux/netfilter/xt_tcpudp.h>
+#define ip6t_tcp xt_tcp
+#define ip6t_udp xt_udp
+#define IP6T_TCP_INV_SRCPT	XT_TCP_INV_SRCPT
+#define IP6T_TCP_INV_DSTPT	XT_TCP_INV_DSTPT
+#define IP6T_TCP_INV_FLAGS	XT_TCP_INV_FLAGS
+#define IP6T_TCP_INV_OPTION	XT_TCP_INV_OPTION
+#define IP6T_TCP_INV_MASK	XT_TCP_INV_MASK
+#define IP6T_UDP_INV_SRCPT	XT_UDP_INV_SRCPT
+#define IP6T_UDP_INV_DSTPT	XT_UDP_INV_DSTPT
+#define IP6T_UDP_INV_MASK	XT_UDP_INV_MASK
+
+#define ip6t_counters_info xt_counters_info
+#define IP6T_STANDARD_TARGET XT_STANDARD_TARGET
+#define IP6T_ERROR_TARGET XT_ERROR_TARGET
+#define IP6T_MATCH_ITERATE(e, fn, args...) \
+	XT_MATCH_ITERATE(struct ip6t_entry, e, fn, ## args)
+#define IP6T_ENTRY_ITERATE(entries, size, fn, args...) \
+	XT_ENTRY_ITERATE(struct ip6t_entry, entries, size, fn, ## args)
 #endif
 
 /* Yes, Virginia, you have to zero the padding. */
@@ -133,7 +156,7 @@ struct ip6t_error {
 #define IP6T_STANDARD_INIT(__verdict)					       \
 {									       \
 	.entry		= IP6T_ENTRY_INIT(sizeof(struct ip6t_standard)),       \
-	.target		= XT_TARGET_INIT(IP6T_STANDARD_TARGET,		       \
+	.target		= XT_TARGET_INIT(XT_STANDARD_TARGET,		       \
 					 sizeof(struct xt_standard_target)),   \
 	.target.verdict	= -(__verdict) - 1,				       \
 }
@@ -141,7 +164,7 @@ struct ip6t_error {
 #define IP6T_ERROR_INIT							       \
 {									       \
 	.entry		= IP6T_ENTRY_INIT(sizeof(struct ip6t_error)),	       \
-	.target		= XT_TARGET_INIT(IP6T_ERROR_TARGET,		       \
+	.target		= XT_TARGET_INIT(XT_ERROR_TARGET,		       \
 					 sizeof(struct ip6t_error_target)),    \
 	.target.errorname = "ERROR",					       \
 }
@@ -165,30 +188,6 @@ struct ip6t_error {
 #define IP6T_SO_GET_REVISION_TARGET	(IP6T_BASE_CTL + 5)
 #define IP6T_SO_GET_MAX			IP6T_SO_GET_REVISION_TARGET
 
-/* CONTINUE verdict for targets */
-#define IP6T_CONTINUE XT_CONTINUE
-
-/* For standard target */
-#define IP6T_RETURN XT_RETURN
-
-/* TCP/UDP matching stuff */
-#include <linux/netfilter/xt_tcpudp.h>
-
-#define ip6t_tcp xt_tcp
-#define ip6t_udp xt_udp
-
-/* Values for "inv" field in struct ipt_tcp. */
-#define IP6T_TCP_INV_SRCPT	XT_TCP_INV_SRCPT
-#define IP6T_TCP_INV_DSTPT	XT_TCP_INV_DSTPT
-#define IP6T_TCP_INV_FLAGS	XT_TCP_INV_FLAGS
-#define IP6T_TCP_INV_OPTION	XT_TCP_INV_OPTION
-#define IP6T_TCP_INV_MASK	XT_TCP_INV_MASK
-
-/* Values for "invflags" field in struct ipt_udp. */
-#define IP6T_UDP_INV_SRCPT	XT_UDP_INV_SRCPT
-#define IP6T_UDP_INV_DSTPT	XT_UDP_INV_DSTPT
-#define IP6T_UDP_INV_MASK	XT_UDP_INV_MASK
-
 /* ICMP matching stuff */
 struct ip6t_icmp {
 	u_int8_t type;				/* type to match */
@@ -252,9 +251,6 @@ struct ip6t_replace {
 	struct ip6t_entry entries[0];
 };
 
-/* The argument to IP6T_SO_ADD_COUNTERS. */
-#define ip6t_counters_info xt_counters_info
-
 /* The argument to IP6T_SO_GET_ENTRIES. */
 struct ip6t_get_entries {
 	/* Which table: user fills this in. */
@@ -267,11 +263,6 @@ struct ip6t_get_entries {
 	struct ip6t_entry entrytable[0];
 };
 
-/* Standard return verdict, or do jump. */
-#define IP6T_STANDARD_TARGET XT_STANDARD_TARGET
-/* Error verdict. */
-#define IP6T_ERROR_TARGET XT_ERROR_TARGET
-
 /* Helper functions */
 static __inline__ struct xt_entry_target *
 ip6t_get_target(struct ip6t_entry *e)
@@ -279,16 +270,6 @@ ip6t_get_target(struct ip6t_entry *e)
 	return (void *)e + e->target_offset;
 }
 
-#ifndef __KERNEL__
-/* fn returns 0 to continue iteration */
-#define IP6T_MATCH_ITERATE(e, fn, args...) \
-	XT_MATCH_ITERATE(struct ip6t_entry, e, fn, ## args)
-
-/* fn returns 0 to continue iteration */
-#define IP6T_ENTRY_ITERATE(entries, size, fn, args...) \
-	XT_ENTRY_ITERATE(struct ip6t_entry, entries, size, fn, ## args)
-#endif
-
 /*
  *	Main firewall chains definitions and global var's definitions.
  */
diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c
index ed178cbe6626..d756edae59ec 100644
--- a/net/ipv4/netfilter/arp_tables.c
+++ b/net/ipv4/netfilter/arp_tables.c
@@ -300,7 +300,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 			v = ((struct xt_standard_target *)t)->verdict;
 			if (v < 0) {
 				/* Pop from stack? */
-				if (v != ARPT_RETURN) {
+				if (v != XT_RETURN) {
 					verdict = (unsigned)(-v) - 1;
 					break;
 				}
@@ -332,7 +332,7 @@ unsigned int arpt_do_table(struct sk_buff *skb,
 		/* Target might have changed stuff. */
 		arp = arp_hdr(skb);
 
-		if (verdict == ARPT_CONTINUE)
+		if (verdict == XT_CONTINUE)
 			e = arpt_next_entry(e);
 		else
 			/* Verdict */
@@ -392,13 +392,13 @@ static int mark_source_chains(const struct xt_table_info *newinfo,
 			/* Unconditional return/END. */
 			if ((e->target_offset == sizeof(struct arpt_entry) &&
 			     (strcmp(t->target.u.user.name,
-				     ARPT_STANDARD_TARGET) == 0) &&
+				     XT_STANDARD_TARGET) == 0) &&
 			     t->verdict < 0 && unconditional(&e->arp)) ||
 			    visited) {
 				unsigned int oldpos, size;
 
 				if ((strcmp(t->target.u.user.name,
-					    ARPT_STANDARD_TARGET) == 0) &&
+					    XT_STANDARD_TARGET) == 0) &&
 				    t->verdict < -NF_MAX_VERDICT - 1) {
 					duprintf("mark_source_chains: bad "
 						"negative verdict (%i)\n",
@@ -433,7 +433,7 @@ static int mark_source_chains(const struct xt_table_info *newinfo,
 				int newpos = t->verdict;
 
 				if (strcmp(t->target.u.user.name,
-					   ARPT_STANDARD_TARGET) == 0 &&
+					   XT_STANDARD_TARGET) == 0 &&
 				    newpos >= 0) {
 					if (newpos > newinfo->size -
 						sizeof(struct arpt_entry)) {
@@ -1828,7 +1828,7 @@ void arpt_unregister_table(struct xt_table *table)
 /* The built-in targets: standard (NULL) and error. */
 static struct xt_target arpt_builtin_tg[] __read_mostly = {
 	{
-		.name             = ARPT_STANDARD_TARGET,
+		.name             = XT_STANDARD_TARGET,
 		.targetsize       = sizeof(int),
 		.family           = NFPROTO_ARP,
 #ifdef CONFIG_COMPAT
@@ -1838,7 +1838,7 @@ static struct xt_target arpt_builtin_tg[] __read_mostly = {
 #endif
 	},
 	{
-		.name             = ARPT_ERROR_TARGET,
+		.name             = XT_ERROR_TARGET,
 		.target           = arpt_error,
 		.targetsize       = XT_FUNCTION_MAXNAMELEN,
 		.family           = NFPROTO_ARP,
diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c
index e1be7dd1171b..b8ddcc480ed9 100644
--- a/net/ipv4/netfilter/arpt_mangle.c
+++ b/net/ipv4/netfilter/arpt_mangle.c
@@ -63,7 +63,7 @@ static int checkentry(const struct xt_tgchk_param *par)
 		return false;
 
 	if (mangle->target != NF_DROP && mangle->target != NF_ACCEPT &&
-	   mangle->target != ARPT_CONTINUE)
+	   mangle->target != XT_CONTINUE)
 		return false;
 	return true;
 }
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index cb108880050a..d31b007a6d80 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -232,7 +232,7 @@ get_chainname_rulenum(const struct ipt_entry *s, const struct ipt_entry *e,
 {
 	const struct xt_standard_target *t = (void *)ipt_get_target_c(s);
 
-	if (strcmp(t->target.u.kernel.target->name, IPT_ERROR_TARGET) == 0) {
+	if (strcmp(t->target.u.kernel.target->name, XT_ERROR_TARGET) == 0) {
 		/* Head of user chain: ERROR target with chainname */
 		*chainname = t->target.data;
 		(*rulenum) = 0;
@@ -241,7 +241,7 @@ get_chainname_rulenum(const struct ipt_entry *s, const struct ipt_entry *e,
 
 		if (s->target_offset == sizeof(struct ipt_entry) &&
 		    strcmp(t->target.u.kernel.target->name,
-			   IPT_STANDARD_TARGET) == 0 &&
+			   XT_STANDARD_TARGET) == 0 &&
 		   t->verdict < 0 &&
 		   unconditional(&s->ip)) {
 			/* Tail of chains: STANDARD target (return/policy) */
@@ -383,7 +383,7 @@ ipt_do_table(struct sk_buff *skb,
 			v = ((struct xt_standard_target *)t)->verdict;
 			if (v < 0) {
 				/* Pop from stack? */
-				if (v != IPT_RETURN) {
+				if (v != XT_RETURN) {
 					verdict = (unsigned)(-v) - 1;
 					break;
 				}
@@ -421,7 +421,7 @@ ipt_do_table(struct sk_buff *skb,
 		verdict = t->u.kernel.target->target(skb, &acpar);
 		/* Target might have changed stuff. */
 		ip = ip_hdr(skb);
-		if (verdict == IPT_CONTINUE)
+		if (verdict == XT_CONTINUE)
 			e = ipt_next_entry(e);
 		else
 			/* Verdict */
@@ -475,13 +475,13 @@ mark_source_chains(const struct xt_table_info *newinfo,
 			/* Unconditional return/END. */
 			if ((e->target_offset == sizeof(struct ipt_entry) &&
 			     (strcmp(t->target.u.user.name,
-				     IPT_STANDARD_TARGET) == 0) &&
+				     XT_STANDARD_TARGET) == 0) &&
 			     t->verdict < 0 && unconditional(&e->ip)) ||
 			    visited) {
 				unsigned int oldpos, size;
 
 				if ((strcmp(t->target.u.user.name,
-			    		    IPT_STANDARD_TARGET) == 0) &&
+			    		    XT_STANDARD_TARGET) == 0) &&
 				    t->verdict < -NF_MAX_VERDICT - 1) {
 					duprintf("mark_source_chains: bad "
 						"negative verdict (%i)\n",
@@ -524,7 +524,7 @@ mark_source_chains(const struct xt_table_info *newinfo,
 				int newpos = t->verdict;
 
 				if (strcmp(t->target.u.user.name,
-					   IPT_STANDARD_TARGET) == 0 &&
+					   XT_STANDARD_TARGET) == 0 &&
 				    newpos >= 0) {
 					if (newpos > newinfo->size -
 						sizeof(struct ipt_entry)) {
@@ -2176,7 +2176,7 @@ static int icmp_checkentry(const struct xt_mtchk_param *par)
 
 static struct xt_target ipt_builtin_tg[] __read_mostly = {
 	{
-		.name             = IPT_STANDARD_TARGET,
+		.name             = XT_STANDARD_TARGET,
 		.targetsize       = sizeof(int),
 		.family           = NFPROTO_IPV4,
 #ifdef CONFIG_COMPAT
@@ -2186,7 +2186,7 @@ static struct xt_target ipt_builtin_tg[] __read_mostly = {
 #endif
 	},
 	{
-		.name             = IPT_ERROR_TARGET,
+		.name             = XT_ERROR_TARGET,
 		.target           = ipt_error,
 		.targetsize       = XT_FUNCTION_MAXNAMELEN,
 		.family           = NFPROTO_IPV4,
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index c7334c10a4b3..c683e9e7023b 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -262,7 +262,7 @@ get_chainname_rulenum(const struct ip6t_entry *s, const struct ip6t_entry *e,
 {
 	const struct xt_standard_target *t = (void *)ip6t_get_target_c(s);
 
-	if (strcmp(t->target.u.kernel.target->name, IP6T_ERROR_TARGET) == 0) {
+	if (strcmp(t->target.u.kernel.target->name, XT_ERROR_TARGET) == 0) {
 		/* Head of user chain: ERROR target with chainname */
 		*chainname = t->target.data;
 		(*rulenum) = 0;
@@ -271,7 +271,7 @@ get_chainname_rulenum(const struct ip6t_entry *s, const struct ip6t_entry *e,
 
 		if (s->target_offset == sizeof(struct ip6t_entry) &&
 		    strcmp(t->target.u.kernel.target->name,
-			   IP6T_STANDARD_TARGET) == 0 &&
+			   XT_STANDARD_TARGET) == 0 &&
 		    t->verdict < 0 &&
 		    unconditional(&s->ipv6)) {
 			/* Tail of chains: STANDARD target (return/policy) */
@@ -406,7 +406,7 @@ ip6t_do_table(struct sk_buff *skb,
 			v = ((struct xt_standard_target *)t)->verdict;
 			if (v < 0) {
 				/* Pop from stack? */
-				if (v != IP6T_RETURN) {
+				if (v != XT_RETURN) {
 					verdict = (unsigned)(-v) - 1;
 					break;
 				}
@@ -434,7 +434,7 @@ ip6t_do_table(struct sk_buff *skb,
 		acpar.targinfo = t->data;
 
 		verdict = t->u.kernel.target->target(skb, &acpar);
-		if (verdict == IP6T_CONTINUE)
+		if (verdict == XT_CONTINUE)
 			e = ip6t_next_entry(e);
 		else
 			/* Verdict */
@@ -488,13 +488,13 @@ mark_source_chains(const struct xt_table_info *newinfo,
 			/* Unconditional return/END. */
 			if ((e->target_offset == sizeof(struct ip6t_entry) &&
 			     (strcmp(t->target.u.user.name,
-				     IP6T_STANDARD_TARGET) == 0) &&
+				     XT_STANDARD_TARGET) == 0) &&
 			     t->verdict < 0 &&
 			     unconditional(&e->ipv6)) || visited) {
 				unsigned int oldpos, size;
 
 				if ((strcmp(t->target.u.user.name,
-					    IP6T_STANDARD_TARGET) == 0) &&
+					    XT_STANDARD_TARGET) == 0) &&
 				    t->verdict < -NF_MAX_VERDICT - 1) {
 					duprintf("mark_source_chains: bad "
 						"negative verdict (%i)\n",
@@ -537,7 +537,7 @@ mark_source_chains(const struct xt_table_info *newinfo,
 				int newpos = t->verdict;
 
 				if (strcmp(t->target.u.user.name,
-					   IP6T_STANDARD_TARGET) == 0 &&
+					   XT_STANDARD_TARGET) == 0 &&
 				    newpos >= 0) {
 					if (newpos > newinfo->size -
 						sizeof(struct ip6t_entry)) {
@@ -2191,7 +2191,7 @@ static int icmp6_checkentry(const struct xt_mtchk_param *par)
 /* The built-in targets: standard (NULL) and error. */
 static struct xt_target ip6t_builtin_tg[] __read_mostly = {
 	{
-		.name             = IP6T_STANDARD_TARGET,
+		.name             = XT_STANDARD_TARGET,
 		.targetsize       = sizeof(int),
 		.family           = NFPROTO_IPV6,
 #ifdef CONFIG_COMPAT
@@ -2201,7 +2201,7 @@ static struct xt_target ip6t_builtin_tg[] __read_mostly = {
 #endif
 	},
 	{
-		.name             = IP6T_ERROR_TARGET,
+		.name             = XT_ERROR_TARGET,
 		.target           = ip6t_error,
 		.targetsize       = XT_FUNCTION_MAXNAMELEN,
 		.family           = NFPROTO_IPV6,
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index f6d464f993ef..8daef9632255 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -230,7 +230,7 @@ static int tcf_ipt(struct sk_buff *skb, struct tc_action *a,
 		result = TC_ACT_SHOT;
 		ipt->tcf_qstats.drops++;
 		break;
-	case IPT_CONTINUE:
+	case XT_CONTINUE:
 		result = TC_ACT_PIPE;
 		break;
 	default:
-- 
cgit v1.2.3


From 75f0a0fd787bfa3ea1a916ca632a5b9e0007cbb7 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@medozas.de>
Date: Wed, 13 Oct 2010 16:37:45 +0200
Subject: netfilter: xtables: unify {ip,ip6,arp}t_error_target

Unification of struct *_error_target was forgotten in
v2.6.16-1689-g1e30a01.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
---
 include/linux/netfilter/x_tables.h        |  5 +++++
 include/linux/netfilter_arp/arp_tables.h  | 10 +++-------
 include/linux/netfilter_ipv4/ip_tables.h  | 10 +++-------
 include/linux/netfilter_ipv6/ip6_tables.h | 10 +++-------
 4 files changed, 14 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 24e5d01d27d0..742bec051440 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -66,6 +66,11 @@ struct xt_standard_target {
 	int verdict;
 };
 
+struct xt_error_target {
+	struct xt_entry_target target;
+	char errorname[XT_FUNCTION_MAXNAMELEN];
+};
+
 /* The argument to IPT_SO_GET_REVISION_*.  Returns highest revision
  * kernel supports, if >= revision. */
 struct xt_get_revision {
diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index 6e2341a811d6..f02d57436a34 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -26,6 +26,7 @@
 #define ARPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN
 #define arpt_entry_target xt_entry_target
 #define arpt_standard_target xt_standard_target
+#define arpt_error_target xt_error_target
 #define ARPT_CONTINUE XT_CONTINUE
 #define ARPT_RETURN XT_RETURN
 #define arpt_counters_info xt_counters_info
@@ -216,14 +217,9 @@ struct arpt_standard {
 	struct xt_standard_target target;
 };
 
-struct arpt_error_target {
-	struct xt_entry_target target;
-	char errorname[XT_FUNCTION_MAXNAMELEN];
-};
-
 struct arpt_error {
 	struct arpt_entry entry;
-	struct arpt_error_target target;
+	struct xt_error_target target;
 };
 
 #define ARPT_ENTRY_INIT(__size)						       \
@@ -244,7 +240,7 @@ struct arpt_error {
 {									       \
 	.entry		= ARPT_ENTRY_INIT(sizeof(struct arpt_error)),	       \
 	.target		= XT_TARGET_INIT(XT_ERROR_TARGET,		       \
-					 sizeof(struct arpt_error_target)),    \
+					 sizeof(struct xt_error_target)),      \
 	.target.errorname = "ERROR",					       \
 }
 
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index ee54b3b7e237..d0fef0a436f9 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -37,6 +37,7 @@
 #define ipt_entry_match xt_entry_match
 #define ipt_entry_target xt_entry_target
 #define ipt_standard_target xt_standard_target
+#define ipt_error_target xt_error_target
 #define ipt_counters xt_counters
 #define IPT_CONTINUE XT_CONTINUE
 #define IPT_RETURN XT_RETURN
@@ -247,14 +248,9 @@ struct ipt_standard {
 	struct xt_standard_target target;
 };
 
-struct ipt_error_target {
-	struct xt_entry_target target;
-	char errorname[XT_FUNCTION_MAXNAMELEN];
-};
-
 struct ipt_error {
 	struct ipt_entry entry;
-	struct ipt_error_target target;
+	struct xt_error_target target;
 };
 
 #define IPT_ENTRY_INIT(__size)						       \
@@ -275,7 +271,7 @@ struct ipt_error {
 {									       \
 	.entry		= IPT_ENTRY_INIT(sizeof(struct ipt_error)),	       \
 	.target		= XT_TARGET_INIT(XT_ERROR_TARGET,		       \
-					 sizeof(struct ipt_error_target)),     \
+					 sizeof(struct xt_error_target)),      \
 	.target.errorname = "ERROR",					       \
 }
 
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index ac2b411ea63a..dca11186e522 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -37,6 +37,7 @@
 #define ip6t_entry_match xt_entry_match
 #define ip6t_entry_target xt_entry_target
 #define ip6t_standard_target xt_standard_target
+#define ip6t_error_target xt_error_target
 #define ip6t_counters xt_counters
 #define IP6T_CONTINUE XT_CONTINUE
 #define IP6T_RETURN XT_RETURN
@@ -137,14 +138,9 @@ struct ip6t_standard {
 	struct xt_standard_target target;
 };
 
-struct ip6t_error_target {
-	struct xt_entry_target target;
-	char errorname[XT_FUNCTION_MAXNAMELEN];
-};
-
 struct ip6t_error {
 	struct ip6t_entry entry;
-	struct ip6t_error_target target;
+	struct xt_error_target target;
 };
 
 #define IP6T_ENTRY_INIT(__size)						       \
@@ -165,7 +161,7 @@ struct ip6t_error {
 {									       \
 	.entry		= IP6T_ENTRY_INIT(sizeof(struct ip6t_error)),	       \
 	.target		= XT_TARGET_INIT(XT_ERROR_TARGET,		       \
-					 sizeof(struct ip6t_error_target)),    \
+					 sizeof(struct xt_error_target)),      \
 	.target.errorname = "ERROR",					       \
 }
 
-- 
cgit v1.2.3


From 9ecdafd883db3c43296797382fc0b2c868144070 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@medozas.de>
Date: Wed, 13 Oct 2010 16:42:02 +0200
Subject: netfilter: xtables: remove unused defines

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
---
 include/linux/netfilter_arp/arp_tables.h  | 4 ----
 include/linux/netfilter_ipv4/ip_tables.h  | 4 ----
 include/linux/netfilter_ipv6/ip6_tables.h | 4 ----
 3 files changed, 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h
index f02d57436a34..adbf4bff87ed 100644
--- a/include/linux/netfilter_arp/arp_tables.h
+++ b/include/linux/netfilter_arp/arp_tables.h
@@ -255,8 +255,6 @@ extern unsigned int arpt_do_table(struct sk_buff *skb,
 				  const struct net_device *out,
 				  struct xt_table *table);
 
-#define ARPT_ALIGN(s) XT_ALIGN(s)
-
 #ifdef CONFIG_COMPAT
 #include <net/compat.h>
 
@@ -275,8 +273,6 @@ compat_arpt_get_target(struct compat_arpt_entry *e)
 	return (void *)e + e->target_offset;
 }
 
-#define COMPAT_ARPT_ALIGN(s)	COMPAT_XT_ALIGN(s)
-
 #endif /* CONFIG_COMPAT */
 #endif /*__KERNEL__*/
 #endif /* _ARPTABLES_H */
diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h
index d0fef0a436f9..64a5d95c58e8 100644
--- a/include/linux/netfilter_ipv4/ip_tables.h
+++ b/include/linux/netfilter_ipv4/ip_tables.h
@@ -282,8 +282,6 @@ extern unsigned int ipt_do_table(struct sk_buff *skb,
 				 const struct net_device *out,
 				 struct xt_table *table);
 
-#define IPT_ALIGN(s) XT_ALIGN(s)
-
 #ifdef CONFIG_COMPAT
 #include <net/compat.h>
 
@@ -304,8 +302,6 @@ compat_ipt_get_target(struct compat_ipt_entry *e)
 	return (void *)e + e->target_offset;
 }
 
-#define COMPAT_IPT_ALIGN(s) 	COMPAT_XT_ALIGN(s)
-
 #endif /* CONFIG_COMPAT */
 #endif /*__KERNEL__*/
 #endif /* _IPTABLES_H */
diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h
index dca11186e522..c9784f7a9c1f 100644
--- a/include/linux/netfilter_ipv6/ip6_tables.h
+++ b/include/linux/netfilter_ipv6/ip6_tables.h
@@ -292,8 +292,6 @@ extern int ip6t_ext_hdr(u8 nexthdr);
 extern int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
 			 int target, unsigned short *fragoff);
 
-#define IP6T_ALIGN(s) XT_ALIGN(s)
-
 #ifdef CONFIG_COMPAT
 #include <net/compat.h>
 
@@ -313,8 +311,6 @@ compat_ip6t_get_target(struct compat_ip6t_entry *e)
 	return (void *)e + e->target_offset;
 }
 
-#define COMPAT_IP6T_ALIGN(s)	COMPAT_XT_ALIGN(s)
-
 #endif /* CONFIG_COMPAT */
 #endif /*__KERNEL__*/
 #endif /* _IP6_TABLES_H */
-- 
cgit v1.2.3


From 892b6f90db81cccb723d5d92f4fddc2d68b206e1 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Wed, 13 Oct 2010 21:18:03 +0200
Subject: block: Ensure physical block size is unsigned int

Physical block size was declared unsigned int to accomodate the maximum
size reported by READ CAPACITY(16).  Make sure we use the right type in
the related functions.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Mike Snitzer <snitzer@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-settings.c   | 2 +-
 include/linux/blkdev.h | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-settings.c b/block/blk-settings.c
index a3600a7ab8bb..315b88c8cbbb 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -344,7 +344,7 @@ EXPORT_SYMBOL(blk_queue_logical_block_size);
  *   hardware can operate on without reverting to read-modify-write
  *   operations.
  */
-void blk_queue_physical_block_size(struct request_queue *q, unsigned short size)
+void blk_queue_physical_block_size(struct request_queue *q, unsigned int size)
 {
 	q->limits.physical_block_size = size;
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 1341df5806df..8f3dd981b973 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -860,7 +860,7 @@ extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
 extern void blk_queue_max_discard_sectors(struct request_queue *q,
 		unsigned int max_discard_sectors);
 extern void blk_queue_logical_block_size(struct request_queue *, unsigned short);
-extern void blk_queue_physical_block_size(struct request_queue *, unsigned short);
+extern void blk_queue_physical_block_size(struct request_queue *, unsigned int);
 extern void blk_queue_alignment_offset(struct request_queue *q,
 				       unsigned int alignment);
 extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min);
@@ -1013,7 +1013,7 @@ static inline unsigned int queue_physical_block_size(struct request_queue *q)
 	return q->limits.physical_block_size;
 }
 
-static inline int bdev_physical_block_size(struct block_device *bdev)
+static inline unsigned int bdev_physical_block_size(struct block_device *bdev)
 {
 	return queue_physical_block_size(bdev_get_queue(bdev));
 }
-- 
cgit v1.2.3


From 10d8dad8453f8648a448960d7a2d3d983dfe0ed3 Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Tue, 12 Oct 2010 07:07:42 +0200
Subject: wext: fix alignment problem in serializing 'struct iw_point'

wext: fix alignment problem in serializing 'struct iw_point'

This fixes a typo in the definition of the serialized length of struct iw_point:
 a) wireless.h is exported to userspace, the typo causes IW_EV_POINT_PK_LEN
    to be 12 on 64-bit, and 8 on 32-bit systems (causing misalignment);
 b) in compat-64 mode iwe_stream_add_point() memcpys overlap (see below).

The second case in  in compat-64 mode looks like (variable names are as in
include/net/iw_handler.h:iwe_stream_add_point()):

 point_len = IW_EV_COMPAT_POINT_LEN = 8
 lcp_len   = IW_EV_COMPAT_LCP_LEN   = 4
 2nd memcpy: IW_EV_POINT_PK_LEN - IW_EV_LCP_PK_LEN = 12 - 4 = 8

 IW_EV_LCP_PK_LEN
 <-------------->                *---> 'extra' data area
 +-------+-------+-------+-------+---------------+------- ...-+
 | len   | cmd   |length | flags |  (empty) -> extra      ... |
 +-------+-------+-------+-------+---------------+------- ...-+
    2       2       2       2          4

     lcp_len
 <-------------->                <-!! OVERLAP !!>
 <--1st memcpy--><------- 2nd memcpy ----------->
                                 <---- 3rd memcpy ------- ... >
 <--------- point_len ---------->

This case could cause overrun whenever iw_point.length < 4.
The other two cases are -
 * 32-bit systems: IW_EV_POINT_PK_LEN - IW_EV_LCP_PK_LEN =  8 - 4 = 4,
   the second memcpy copies exactly the 4 required bytes;
 * 64-bit systems: IW_EV_POINT_PK_LEN - IW_EV_LCP_PK_LEN = 12 - 4 = 8,
   the second memcpy copies a superfluous (but non overlapping) 4 bytes.

The patch changes IW_EV_POINT_PK_LEN to be 8, so that in all 3 cases always only
the requested iw_point.{length,flags} (both __u16) are copied, avoiding overrrun
(compat-64) and superfluous copy (64-bit). In addition, the userspace header is
sanitized (in agreement with version 30 of the wireless tools).

Many thanks to Johannes Berg for help and review with this patch.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/wireless.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index e6827eedf18b..4395b28bb86c 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -1157,6 +1157,6 @@ struct __compat_iw_event {
 #define IW_EV_PARAM_PK_LEN	(IW_EV_LCP_PK_LEN + sizeof(struct iw_param))
 #define IW_EV_ADDR_PK_LEN	(IW_EV_LCP_PK_LEN + sizeof(struct sockaddr))
 #define IW_EV_QUAL_PK_LEN	(IW_EV_LCP_PK_LEN + sizeof(struct iw_quality))
-#define IW_EV_POINT_PK_LEN	(IW_EV_LCP_LEN + 4)
+#define IW_EV_POINT_PK_LEN	(IW_EV_LCP_PK_LEN + 4)
 
 #endif	/* _LINUX_WIRELESS_H */
-- 
cgit v1.2.3


From b3d6255388de0680a14f0907deb7b7f4fa0d25d5 Mon Sep 17 00:00:00 2001
From: Kumar Sanghvi <kumar.sanghvi@stericsson.com>
Date: Tue, 12 Oct 2010 20:14:43 +0000
Subject: Phonet: 'connect' socket implementation for Pipe controller
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Based on suggestion by Rémi Denis-Courmont to implement 'connect'
for Pipe controller logic,  this patch implements 'connect' socket
call for the Pipe controller logic.
The patch does following:-
- Removes setsockopts for PNPIPE_CREATE and PNPIPE_DESTROY
- Adds setsockopt for setting the Pipe handle value
- Implements connect socket call
- Updates the Pipe controller logic

User-space should now follow below sequence with Pipe controller:-
-socket
-bind
-setsockopt for PNPIPE_PIPE_HANDLE
-connect
-setsockopt for PNPIPE_ENCAP_IP
-setsockopt for PNPIPE_ENABLE

GPRS/3G data has been tested working fine with this.

Signed-off-by: Kumar Sanghvi <kumar.sanghvi@stericsson.com>
Acked-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/phonet.h   |   3 +-
 include/net/phonet/pep.h |   4 +-
 net/phonet/pep.c         | 300 ++++++++++++++++++-----------------------------
 net/phonet/socket.c      |  99 ++++++++++++++++
 4 files changed, 215 insertions(+), 191 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/phonet.h b/include/linux/phonet.h
index e27cbf931740..26c8df786918 100644
--- a/include/linux/phonet.h
+++ b/include/linux/phonet.h
@@ -36,10 +36,9 @@
 /* Socket options for SOL_PNPIPE level */
 #define PNPIPE_ENCAP		1
 #define PNPIPE_IFINDEX		2
-#define PNPIPE_CREATE           3
+#define PNPIPE_PIPE_HANDLE	3
 #define PNPIPE_ENABLE           4
 /* unused slot */
-#define PNPIPE_DESTROY          6
 
 #define PNADDR_ANY		0
 #define PNADDR_BROADCAST	0xFC
diff --git a/include/net/phonet/pep.h b/include/net/phonet/pep.h
index def6cfa3f451..b60b28c99e87 100644
--- a/include/net/phonet/pep.h
+++ b/include/net/phonet/pep.h
@@ -46,8 +46,8 @@ struct pep_sock {
 	u8			init_enable;	/* auto-enable at creation */
 	u8			aligned;
 #ifdef CONFIG_PHONET_PIPECTRLR
-	u16                     remote_pep;
-	u8                      pipe_state;
+	u8			pipe_state;
+	struct sockaddr_pn	remote_pep;
 #endif
 };
 
diff --git a/net/phonet/pep.c b/net/phonet/pep.c
index f818f76d297d..9c903f9e5079 100644
--- a/net/phonet/pep.c
+++ b/net/phonet/pep.c
@@ -88,15 +88,6 @@ static int pep_reply(struct sock *sk, struct sk_buff *oskb,
 	const struct pnpipehdr *oph = pnp_hdr(oskb);
 	struct pnpipehdr *ph;
 	struct sk_buff *skb;
-#ifdef CONFIG_PHONET_PIPECTRLR
-	const struct phonethdr *hdr = pn_hdr(oskb);
-	struct sockaddr_pn spn = {
-		.spn_family = AF_PHONET,
-		.spn_resource = 0xD9,
-		.spn_dev = hdr->pn_sdev,
-		.spn_obj = hdr->pn_sobj,
-	};
-#endif
 
 	skb = alloc_skb(MAX_PNPIPE_HEADER + len, priority);
 	if (!skb)
@@ -114,11 +105,7 @@ static int pep_reply(struct sock *sk, struct sk_buff *oskb,
 	ph->pipe_handle = oph->pipe_handle;
 	ph->error_code = code;
 
-#ifdef CONFIG_PHONET_PIPECTRLR
-	return pn_skb_send(sk, skb, &spn);
-#else
 	return pn_skb_send(sk, skb, &pipe_srv);
-#endif
 }
 
 #define PAD 0x00
@@ -188,18 +175,13 @@ static int pipe_get_flow_info(struct sock *sk, struct sk_buff *skb,
 	return 0;
 }
 
-static int pipe_handler_send_req(struct sock *sk, u16 dobj, u8 utid,
-		u8 msg_id, u8 p_handle, gfp_t priority)
+static int pipe_handler_send_req(struct sock *sk, u8 utid,
+		u8 msg_id, gfp_t priority)
 {
 	int len;
 	struct pnpipehdr *ph;
 	struct sk_buff *skb;
-	struct sockaddr_pn spn = {
-		.spn_family = AF_PHONET,
-		.spn_resource = 0xD9,
-		.spn_dev = pn_dev(dobj),
-		.spn_obj = pn_obj(dobj),
-	};
+	struct pep_sock *pn = pep_sk(sk);
 
 	static const u8 data[4] = {
 		PAD, PAD, PAD, PAD,
@@ -235,30 +217,25 @@ static int pipe_handler_send_req(struct sock *sk, u16 dobj, u8 utid,
 	ph = pnp_hdr(skb);
 	ph->utid = utid;
 	ph->message_id = msg_id;
-	ph->pipe_handle = p_handle;
+	ph->pipe_handle = pn->pipe_handle;
 	ph->error_code = PN_PIPE_NO_ERROR;
 
-	return pn_skb_send(sk, skb, &spn);
+	return pn_skb_send(sk, skb, &pn->remote_pep);
 }
 
-static int pipe_handler_send_created_ind(struct sock *sk, u16 dobj,
-		u8 utid, u8 p_handle, u8 msg_id, u8 tx_fc, u8 rx_fc)
+static int pipe_handler_send_created_ind(struct sock *sk,
+		u8 utid, u8 msg_id)
 {
 	int err_code;
 	struct pnpipehdr *ph;
 	struct sk_buff *skb;
-	struct sockaddr_pn spn = {
-		.spn_family = AF_PHONET,
-		.spn_resource = 0xD9,
-		.spn_dev = pn_dev(dobj),
-		.spn_obj = pn_obj(dobj),
-	};
 
+	struct pep_sock *pn = pep_sk(sk);
 	static u8 data[4] = {
 		0x03, 0x04,
 	};
-	data[2] = tx_fc;
-	data[3] = rx_fc;
+	data[2] = pn->tx_fc;
+	data[3] = pn->rx_fc;
 
 	/*
 	 * actually, below is number of sub-blocks and not error code.
@@ -282,24 +259,18 @@ static int pipe_handler_send_created_ind(struct sock *sk, u16 dobj,
 	ph = pnp_hdr(skb);
 	ph->utid = utid;
 	ph->message_id = msg_id;
-	ph->pipe_handle = p_handle;
+	ph->pipe_handle = pn->pipe_handle;
 	ph->error_code = err_code;
 
-	return pn_skb_send(sk, skb, &spn);
+	return pn_skb_send(sk, skb, &pn->remote_pep);
 }
 
-static int pipe_handler_send_ind(struct sock *sk, u16 dobj, u8 utid,
-		u8 p_handle, u8 msg_id)
+static int pipe_handler_send_ind(struct sock *sk, u8 utid, u8 msg_id)
 {
 	int err_code;
 	struct pnpipehdr *ph;
 	struct sk_buff *skb;
-	struct sockaddr_pn spn = {
-		.spn_family = AF_PHONET,
-		.spn_resource = 0xD9,
-		.spn_dev = pn_dev(dobj),
-		.spn_obj = pn_obj(dobj),
-	};
+	struct pep_sock *pn = pep_sk(sk);
 
 	/*
 	 * actually, below is a filler.
@@ -321,10 +292,10 @@ static int pipe_handler_send_ind(struct sock *sk, u16 dobj, u8 utid,
 	ph = pnp_hdr(skb);
 	ph->utid = utid;
 	ph->message_id = msg_id;
-	ph->pipe_handle = p_handle;
+	ph->pipe_handle = pn->pipe_handle;
 	ph->error_code = err_code;
 
-	return pn_skb_send(sk, skb, &spn);
+	return pn_skb_send(sk, skb, &pn->remote_pep);
 }
 
 static int pipe_handler_enable_pipe(struct sock *sk, int enable)
@@ -339,34 +310,7 @@ static int pipe_handler_enable_pipe(struct sock *sk, int enable)
 		utid = PNS_PIPE_DISABLE_UTID;
 		req = PNS_PEP_DISABLE_REQ;
 	}
-	return pipe_handler_send_req(sk, pn->pn_sk.sobject, utid, req,
-			pn->pipe_handle, GFP_ATOMIC);
-}
-
-static int pipe_handler_create_pipe(struct sock *sk, int pipe_handle, int cmd)
-{
-	int ret;
-	struct pep_sock *pn = pep_sk(sk);
-
-	switch (cmd) {
-	case PNPIPE_CREATE:
-		ret = pipe_handler_send_req(sk, pn->pn_sk.sobject,
-				PNS_PEP_CONNECT_UTID, PNS_PEP_CONNECT_REQ,
-				pipe_handle, GFP_ATOMIC);
-		break;
-
-	case PNPIPE_DESTROY:
-		ret = pipe_handler_send_req(sk, pn->remote_pep,
-				PNS_PEP_DISCONNECT_UTID,
-				PNS_PEP_DISCONNECT_REQ,
-				pn->pipe_handle, GFP_ATOMIC);
-		break;
-
-	default:
-		ret = -EINVAL;
-	}
-
-	return ret;
+	return pipe_handler_send_req(sk, utid, req, GFP_ATOMIC);
 }
 #endif
 
@@ -434,14 +378,6 @@ static int pipe_snd_status(struct sock *sk, u8 type, u8 status, gfp_t priority)
 	struct pep_sock *pn = pep_sk(sk);
 	struct pnpipehdr *ph;
 	struct sk_buff *skb;
-#ifdef CONFIG_PHONET_PIPECTRLR
-	struct sockaddr_pn spn = {
-		.spn_family = AF_PHONET,
-		.spn_resource = 0xD9,
-		.spn_dev = pn_dev(pn->remote_pep),
-		.spn_obj = pn_obj(pn->remote_pep),
-	};
-#endif
 
 	skb = alloc_skb(MAX_PNPIPE_HEADER + 4, priority);
 	if (!skb)
@@ -462,7 +398,7 @@ static int pipe_snd_status(struct sock *sk, u8 type, u8 status, gfp_t priority)
 	ph->data[4] = status;
 
 #ifdef CONFIG_PHONET_PIPECTRLR
-	return pn_skb_send(sk, skb, &spn);
+	return pn_skb_send(sk, skb, &pn->remote_pep);
 #else
 	return pn_skb_send(sk, skb, &pipe_srv);
 #endif
@@ -582,12 +518,6 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 	struct pnpipehdr *hdr = pnp_hdr(skb);
 	struct sk_buff_head *queue;
 	int err = 0;
-#ifdef CONFIG_PHONET_PIPECTRLR
-	struct phonethdr *ph = pn_hdr(skb);
-	static u8 host_pref_rx_fc[3], host_req_tx_fc[3];
-	u8 remote_pref_rx_fc[3], remote_req_tx_fc[3];
-	u8 negotiated_rx_fc, negotiated_tx_fc;
-#endif
 
 	BUG_ON(sk->sk_state == TCP_CLOSE_WAIT);
 
@@ -596,40 +526,6 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 		pep_reject_conn(sk, skb, PN_PIPE_ERR_PEP_IN_USE);
 		break;
 
-#ifdef CONFIG_PHONET_PIPECTRLR
-	case PNS_PEP_CONNECT_RESP:
-		if ((ph->pn_sdev == pn_dev(pn->remote_pep)) &&
-				(ph->pn_sobj == pn_obj(pn->remote_pep))) {
-			pipe_get_flow_info(sk, skb, remote_pref_rx_fc,
-					remote_req_tx_fc);
-
-			 negotiated_tx_fc = pipe_negotiate_fc(remote_req_tx_fc,
-					 host_pref_rx_fc,
-					 sizeof(host_pref_rx_fc));
-			 negotiated_rx_fc = pipe_negotiate_fc(host_req_tx_fc,
-					 remote_pref_rx_fc,
-					 sizeof(host_pref_rx_fc));
-
-			pn->pipe_state = PIPE_DISABLED;
-			pipe_handler_send_created_ind(sk, pn->remote_pep,
-					PNS_PIPE_CREATED_IND_UTID,
-					pn->pipe_handle, PNS_PIPE_CREATED_IND,
-					negotiated_tx_fc, negotiated_rx_fc);
-			pipe_handler_send_created_ind(sk, pn->pn_sk.sobject,
-					PNS_PIPE_CREATED_IND_UTID,
-					pn->pipe_handle, PNS_PIPE_CREATED_IND,
-					negotiated_tx_fc, negotiated_rx_fc);
-		} else {
-			pipe_handler_send_req(sk, pn->remote_pep,
-					PNS_PEP_CONNECT_UTID,
-					PNS_PEP_CONNECT_REQ, pn->pipe_handle,
-					GFP_ATOMIC);
-			pipe_get_flow_info(sk, skb, host_pref_rx_fc,
-					host_req_tx_fc);
-		}
-		break;
-#endif
-
 	case PNS_PEP_DISCONNECT_REQ:
 		pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC);
 		sk->sk_state = TCP_CLOSE_WAIT;
@@ -640,10 +536,7 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 #ifdef CONFIG_PHONET_PIPECTRLR
 	case PNS_PEP_DISCONNECT_RESP:
 		pn->pipe_state = PIPE_IDLE;
-		pipe_handler_send_req(sk, pn->pn_sk.sobject,
-				PNS_PEP_DISCONNECT_UTID,
-				PNS_PEP_DISCONNECT_REQ, pn->pipe_handle,
-				GFP_KERNEL);
+		sk->sk_state = TCP_CLOSE;
 		break;
 #endif
 
@@ -654,21 +547,18 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 
 #ifdef CONFIG_PHONET_PIPECTRLR
 	case PNS_PEP_ENABLE_RESP:
-		if ((ph->pn_sdev == pn_dev(pn->remote_pep)) &&
-				(ph->pn_sobj == pn_obj(pn->remote_pep))) {
-			pn->pipe_state = PIPE_ENABLED;
-			pipe_handler_send_ind(sk, pn->remote_pep,
-					PNS_PIPE_ENABLED_IND_UTID,
-					pn->pipe_handle, PNS_PIPE_ENABLED_IND);
-			pipe_handler_send_ind(sk, pn->pn_sk.sobject,
-					PNS_PIPE_ENABLED_IND_UTID,
-					pn->pipe_handle, PNS_PIPE_ENABLED_IND);
-		} else
-			pipe_handler_send_req(sk, pn->remote_pep,
-					PNS_PIPE_ENABLE_UTID,
-					PNS_PEP_ENABLE_REQ, pn->pipe_handle,
-					GFP_KERNEL);
+		pn->pipe_state = PIPE_ENABLED;
+		pipe_handler_send_ind(sk, PNS_PIPE_ENABLED_IND_UTID,
+				PNS_PIPE_ENABLED_IND);
 
+		if (!pn_flow_safe(pn->tx_fc)) {
+			atomic_set(&pn->tx_credits, 1);
+			sk->sk_write_space(sk);
+		}
+		if (sk->sk_state == TCP_ESTABLISHED)
+			break; /* Nothing to do */
+		sk->sk_state = TCP_ESTABLISHED;
+		pipe_grant_credits(sk);
 		break;
 #endif
 
@@ -692,22 +582,12 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
 
 #ifdef CONFIG_PHONET_PIPECTRLR
 	case PNS_PEP_DISABLE_RESP:
-		if ((ph->pn_sdev == pn_dev(pn->remote_pep)) &&
-				(ph->pn_sobj == pn_obj(pn->remote_pep))) {
-			pn->pipe_state = PIPE_DISABLED;
-			pipe_handler_send_ind(sk, pn->remote_pep,
-					PNS_PIPE_DISABLED_IND_UTID,
-					pn->pipe_handle,
-					PNS_PIPE_DISABLED_IND);
-			pipe_handler_send_ind(sk, pn->pn_sk.sobject,
-					PNS_PIPE_DISABLED_IND_UTID,
-					pn->pipe_handle,
-					PNS_PIPE_DISABLED_IND);
-		} else
-			pipe_handler_send_req(sk, pn->remote_pep,
-					PNS_PIPE_DISABLE_UTID,
-					PNS_PEP_DISABLE_REQ, pn->pipe_handle,
-					GFP_KERNEL);
+		pn->pipe_state = PIPE_DISABLED;
+		atomic_set(&pn->tx_credits, 0);
+		pipe_handler_send_ind(sk, PNS_PIPE_DISABLED_IND_UTID,
+				PNS_PIPE_DISABLED_IND);
+		sk->sk_state = TCP_SYN_RECV;
+		pn->rx_credits = 0;
 		break;
 #endif
 
@@ -802,6 +682,42 @@ static void pipe_destruct(struct sock *sk)
 	skb_queue_purge(&pn->ctrlreq_queue);
 }
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+static int pep_connresp_rcv(struct sock *sk, struct sk_buff *skb)
+{
+	struct pep_sock *pn = pep_sk(sk);
+	u8 host_pref_rx_fc[3] = {3, 2, 1}, host_req_tx_fc[3] = {3, 2, 1};
+	u8 remote_pref_rx_fc[3], remote_req_tx_fc[3];
+	u8 negotiated_rx_fc, negotiated_tx_fc;
+	int ret;
+
+	pipe_get_flow_info(sk, skb, remote_pref_rx_fc,
+			remote_req_tx_fc);
+	negotiated_tx_fc = pipe_negotiate_fc(remote_req_tx_fc,
+			host_pref_rx_fc,
+			sizeof(host_pref_rx_fc));
+	negotiated_rx_fc = pipe_negotiate_fc(host_req_tx_fc,
+			remote_pref_rx_fc,
+			sizeof(host_pref_rx_fc));
+
+	pn->pipe_state = PIPE_DISABLED;
+	sk->sk_state = TCP_SYN_RECV;
+	sk->sk_backlog_rcv = pipe_do_rcv;
+	sk->sk_destruct = pipe_destruct;
+	pn->rx_credits = 0;
+	pn->rx_fc = negotiated_rx_fc;
+	pn->tx_fc = negotiated_tx_fc;
+	sk->sk_state_change(sk);
+
+	ret = pipe_handler_send_created_ind(sk,
+			PNS_PIPE_CREATED_IND_UTID,
+			PNS_PIPE_CREATED_IND
+			);
+
+	return ret;
+}
+#endif
+
 static int pep_connreq_rcv(struct sock *sk, struct sk_buff *skb)
 {
 	struct sock *newsk;
@@ -884,9 +800,6 @@ static int pep_connreq_rcv(struct sock *sk, struct sk_buff *skb)
 	newpn->rx_fc = newpn->tx_fc = PN_LEGACY_FLOW_CONTROL;
 	newpn->init_enable = enabled;
 	newpn->aligned = aligned;
-#ifdef CONFIG_PHONET_PIPECTRLR
-	newpn->remote_pep = pn->remote_pep;
-#endif
 
 	BUG_ON(!skb_queue_empty(&newsk->sk_receive_queue));
 	skb_queue_head(&newsk->sk_receive_queue, skb);
@@ -968,6 +881,12 @@ static int pep_do_rcv(struct sock *sk, struct sk_buff *skb)
 		err = pep_connreq_rcv(sk, skb);
 		break;
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+	case PNS_PEP_CONNECT_RESP:
+		err = pep_connresp_rcv(sk, skb);
+		break;
+#endif
+
 	case PNS_PEP_DISCONNECT_REQ:
 		pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC);
 		break;
@@ -1032,6 +951,18 @@ static void pep_sock_close(struct sock *sk, long timeout)
 		/* Forcefully remove dangling Phonet pipe */
 		pipe_do_remove(sk);
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+	if (pn->pipe_state != PIPE_IDLE) {
+		/* send pep disconnect request */
+		pipe_handler_send_req(sk,
+				PNS_PEP_DISCONNECT_UTID, PNS_PEP_DISCONNECT_REQ,
+				GFP_KERNEL);
+
+		pn->pipe_state = PIPE_IDLE;
+		sk->sk_state = TCP_CLOSE;
+	}
+#endif
+
 	ifindex = pn->ifindex;
 	pn->ifindex = 0;
 	release_sock(sk);
@@ -1108,6 +1039,20 @@ out:
 	return newsk;
 }
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+static int pep_sock_connect(struct sock *sk, struct sockaddr *addr, int len)
+{
+	struct pep_sock *pn = pep_sk(sk);
+	struct sockaddr_pn *spn =  (struct sockaddr_pn *)addr;
+
+	memcpy(&pn->remote_pep, spn, sizeof(struct sockaddr_pn));
+
+	return pipe_handler_send_req(sk,
+			PNS_PEP_CONNECT_UTID, PNS_PEP_CONNECT_REQ,
+			GFP_ATOMIC);
+}
+#endif
+
 static int pep_ioctl(struct sock *sk, int cmd, unsigned long arg)
 {
 	struct pep_sock *pn = pep_sk(sk);
@@ -1149,10 +1094,6 @@ static int pep_setsockopt(struct sock *sk, int level, int optname,
 {
 	struct pep_sock *pn = pep_sk(sk);
 	int val = 0, err = 0;
-#ifdef CONFIG_PHONET_PIPECTRLR
-	int remote_pep;
-	int pipe_handle;
-#endif
 
 	if (level != SOL_PNPIPE)
 		return -ENOPROTOOPT;
@@ -1164,28 +1105,15 @@ static int pep_setsockopt(struct sock *sk, int level, int optname,
 	lock_sock(sk);
 	switch (optname) {
 #ifdef CONFIG_PHONET_PIPECTRLR
-	case PNPIPE_CREATE:
+	case PNPIPE_PIPE_HANDLE:
 		if (val) {
 			if (pn->pipe_state > PIPE_IDLE) {
 				err = -EFAULT;
 				break;
 			}
-			remote_pep = val & 0xFFFF;
-			pipe_handle =  (val >> 16) & 0xFF;
-			pn->remote_pep = remote_pep;
-			err = pipe_handler_create_pipe(sk, pipe_handle,
-					PNPIPE_CREATE);
-			break;
-		}
-
-	case PNPIPE_DESTROY:
-		if (pn->pipe_state < PIPE_DISABLED) {
-			err = -EFAULT;
+			pn->pipe_handle = val;
 			break;
 		}
-
-		err = pipe_handler_create_pipe(sk, 0x0, PNPIPE_DESTROY);
-		break;
 #endif
 
 	case PNPIPE_ENCAP:
@@ -1278,14 +1206,6 @@ static int pipe_skb_send(struct sock *sk, struct sk_buff *skb)
 	struct pep_sock *pn = pep_sk(sk);
 	struct pnpipehdr *ph;
 	int err;
-#ifdef CONFIG_PHONET_PIPECTRLR
-	struct sockaddr_pn spn = {
-		.spn_family = AF_PHONET,
-		.spn_resource = 0xD9,
-		.spn_dev = pn_dev(pn->remote_pep),
-		.spn_obj = pn_obj(pn->remote_pep),
-	};
-#endif
 
 	if (pn_flow_safe(pn->tx_fc) &&
 	    !atomic_add_unless(&pn->tx_credits, -1, 0)) {
@@ -1304,7 +1224,7 @@ static int pipe_skb_send(struct sock *sk, struct sk_buff *skb)
 		ph->message_id = PNS_PIPE_DATA;
 	ph->pipe_handle = pn->pipe_handle;
 #ifdef CONFIG_PHONET_PIPECTRLR
-	err = pn_skb_send(sk, skb, &spn);
+	err = pn_skb_send(sk, skb, &pn->remote_pep);
 #else
 	err = pn_skb_send(sk, skb, &pipe_srv);
 #endif
@@ -1504,6 +1424,8 @@ static void pep_sock_unhash(struct sock *sk)
 	struct sock *skparent = NULL;
 
 	lock_sock(sk);
+
+#ifndef CONFIG_PHONET_PIPECTRLR
 	if ((1 << sk->sk_state) & ~(TCPF_CLOSE|TCPF_LISTEN)) {
 		skparent = pn->listener;
 		release_sock(sk);
@@ -1513,6 +1435,7 @@ static void pep_sock_unhash(struct sock *sk)
 		sk_del_node_init(sk);
 		sk = skparent;
 	}
+#endif
 	/* Unhash a listening sock only when it is closed
 	 * and all of its active connected pipes are closed. */
 	if (hlist_empty(&pn->hlist))
@@ -1526,6 +1449,9 @@ static void pep_sock_unhash(struct sock *sk)
 static struct proto pep_proto = {
 	.close		= pep_sock_close,
 	.accept		= pep_sock_accept,
+#ifdef CONFIG_PHONET_PIPECTRLR
+	.connect	= pep_sock_connect,
+#endif
 	.ioctl		= pep_ioctl,
 	.init		= pep_init,
 	.setsockopt	= pep_setsockopt,
diff --git a/net/phonet/socket.c b/net/phonet/socket.c
index aca8fba099e9..25f746d20c1f 100644
--- a/net/phonet/socket.c
+++ b/net/phonet/socket.c
@@ -225,6 +225,101 @@ static int pn_socket_autobind(struct socket *sock)
 	return 0; /* socket was already bound */
 }
 
+#ifdef CONFIG_PHONET_PIPECTRLR
+static int pn_socket_connect(struct socket *sock, struct sockaddr *addr,
+		int len, int flags)
+{
+	struct sock *sk = sock->sk;
+	struct sockaddr_pn *spn = (struct sockaddr_pn *)addr;
+	long timeo;
+	int err;
+
+	if (len < sizeof(struct sockaddr_pn))
+		return -EINVAL;
+	if (spn->spn_family != AF_PHONET)
+		return -EAFNOSUPPORT;
+
+	lock_sock(sk);
+
+	switch (sock->state) {
+	case SS_UNCONNECTED:
+		sk->sk_state = TCP_CLOSE;
+		break;
+	case SS_CONNECTING:
+		switch (sk->sk_state) {
+		case TCP_SYN_RECV:
+			sock->state = SS_CONNECTED;
+			err = -EISCONN;
+			goto out;
+		case TCP_CLOSE:
+			err = -EALREADY;
+			if (flags & O_NONBLOCK)
+				goto out;
+			goto wait_connect;
+		}
+		break;
+	case SS_CONNECTED:
+		switch (sk->sk_state) {
+		case TCP_SYN_RECV:
+			err = -EISCONN;
+			goto out;
+		case TCP_CLOSE:
+			sock->state = SS_UNCONNECTED;
+			break;
+		}
+		break;
+	case SS_DISCONNECTING:
+	case SS_FREE:
+		break;
+	}
+	sk->sk_state = TCP_CLOSE;
+	sk_stream_kill_queues(sk);
+
+	sock->state = SS_CONNECTING;
+	err = sk->sk_prot->connect(sk, addr, len);
+	if (err < 0) {
+		sock->state = SS_UNCONNECTED;
+		sk->sk_state = TCP_CLOSE;
+		goto out;
+	}
+
+	err = -EINPROGRESS;
+wait_connect:
+	if (sk->sk_state != TCP_SYN_RECV && (flags & O_NONBLOCK))
+		goto out;
+
+	timeo = sock_sndtimeo(sk, flags & O_NONBLOCK);
+	release_sock(sk);
+
+	err = -ERESTARTSYS;
+	timeo = wait_event_interruptible_timeout(*sk_sleep(sk),
+			sk->sk_state != TCP_CLOSE,
+			timeo);
+
+	lock_sock(sk);
+	if (timeo < 0)
+		goto out; /* -ERESTARTSYS */
+
+	err = -ETIMEDOUT;
+	if (timeo == 0 && sk->sk_state != TCP_SYN_RECV)
+		goto out;
+
+	if (sk->sk_state != TCP_SYN_RECV) {
+		sock->state = SS_UNCONNECTED;
+		err = sock_error(sk);
+		if (!err)
+			err = -ECONNREFUSED;
+		goto out;
+	}
+	sock->state = SS_CONNECTED;
+	err = 0;
+
+out:
+	release_sock(sk);
+	return err;
+}
+#endif
+
 static int pn_socket_accept(struct socket *sock, struct socket *newsock,
 				int flags)
 {
@@ -393,7 +488,11 @@ const struct proto_ops phonet_stream_ops = {
 	.owner		= THIS_MODULE,
 	.release	= pn_socket_release,
 	.bind		= pn_socket_bind,
+#ifdef CONFIG_PHONET_PIPECTRLR
+	.connect	= pn_socket_connect,
+#else
 	.connect	= sock_no_connect,
+#endif
 	.socketpair	= sock_no_socketpair,
 	.accept		= pn_socket_accept,
 	.getname	= pn_socket_getname,
-- 
cgit v1.2.3


From 4f0e332239e2b5f79757cb8f8f3db16c66f5d220 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Fri, 24 Sep 2010 13:34:42 -0500
Subject: powerpc/fsl-booke: Add PCI device ids for P2040/P3041/P5010/P5020
 QoirQ chips

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/sysdev/fsl_pci.c | 8 ++++++++
 include/linux/pci_ids.h       | 8 ++++++++
 2 files changed, 16 insertions(+)

(limited to 'include/linux')

diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c
index 505c8f0ece9b..818f7c6c8fa1 100644
--- a/arch/powerpc/sysdev/fsl_pci.c
+++ b/arch/powerpc/sysdev/fsl_pci.c
@@ -407,10 +407,18 @@ DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P2010E, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P2010, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P2020E, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P2020, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P2040E, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P2040, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P3041E, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P3041, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P4040E, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P4040, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P4080E, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P4080, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P5010E, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P5010, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P5020E, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P5020, quirk_fsl_pcie_header);
 #endif /* CONFIG_FSL_SOC_BOOKE || CONFIG_PPC_86xx */
 
 #if defined(CONFIG_PPC_83xx) || defined(CONFIG_PPC_MPC512x)
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 570fddeb0388..f69dfe5e01ef 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2315,6 +2315,14 @@
 #define PCI_DEVICE_ID_P4080		0x0401
 #define PCI_DEVICE_ID_P4040E		0x0408
 #define PCI_DEVICE_ID_P4040		0x0409
+#define PCI_DEVICE_ID_P2040E		0x0410
+#define PCI_DEVICE_ID_P2040		0x0411
+#define PCI_DEVICE_ID_P3041E		0x041E
+#define PCI_DEVICE_ID_P3041		0x041F
+#define PCI_DEVICE_ID_P5020E		0x0420
+#define PCI_DEVICE_ID_P5020		0x0421
+#define PCI_DEVICE_ID_P5010E		0x0428
+#define PCI_DEVICE_ID_P5010		0x0429
 #define PCI_DEVICE_ID_MPC8641		0x7010
 #define PCI_DEVICE_ID_MPC8641D		0x7011
 #define PCI_DEVICE_ID_MPC8610		0x7018
-- 
cgit v1.2.3


From 087a4eb55971dfcc8df18312faf9393d0a479f3a Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Date: Thu, 14 Oct 2010 12:10:30 +0900
Subject: stopmachine: Define __stop_machine when CONFIG_STOP_MACHINE=n

Define dummy __stop_machine() function even when
CONFIG_STOP_MACHINE=n. This getcpu-required version of
stop_machine() will be used from poke_text_smp().

Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: 2nddept-manager@sdl.hitachi.co.jp
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20101014031030.4100.34156.stgit@ltc236.sdl.hitachi.co.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/stop_machine.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/stop_machine.h b/include/linux/stop_machine.h
index 6b524a0d02e4..1808960c5059 100644
--- a/include/linux/stop_machine.h
+++ b/include/linux/stop_machine.h
@@ -126,8 +126,8 @@ int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus);
 
 #else	 /* CONFIG_STOP_MACHINE && CONFIG_SMP */
 
-static inline int stop_machine(int (*fn)(void *), void *data,
-			       const struct cpumask *cpus)
+static inline int __stop_machine(int (*fn)(void *), void *data,
+				 const struct cpumask *cpus)
 {
 	int ret;
 	local_irq_disable();
@@ -136,5 +136,11 @@ static inline int stop_machine(int (*fn)(void *), void *data,
 	return ret;
 }
 
+static inline int stop_machine(int (*fn)(void *), void *data,
+			       const struct cpumask *cpus)
+{
+	return __stop_machine(fn, data, cpus);
+}
+
 #endif	/* CONFIG_STOP_MACHINE && CONFIG_SMP */
 #endif	/* _LINUX_STOP_MACHINE */
-- 
cgit v1.2.3


From 265be2d09853d425ad14a61cda0ca63345613d0c Mon Sep 17 00:00:00 2001
From: Philipp Reisner <philipp.reisner@linbit.com>
Date: Mon, 31 May 2010 10:14:17 +0200
Subject: drbd: Finished the "on-no-data-accessible suspend-io;" functionality

When no data is accessible (no connection to the peer, nor a local disk)
allow the user to select to freeze all IO operations instead of getting
IO errors.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_int.h    |  1 +
 drivers/block/drbd/drbd_main.c   | 26 +++++++++++++++++++++++++-
 drivers/block/drbd/drbd_nl.c     | 13 +++++++++++++
 drivers/block/drbd/drbd_req.c    | 24 ++++++++++++++++++++++++
 drivers/block/drbd/drbd_req.h    |  2 ++
 drivers/block/drbd/drbd_worker.c | 18 ++++++++++++++++++
 include/linux/drbd.h             |  5 +++++
 include/linux/drbd_limits.h      |  1 +
 include/linux/drbd_nl.h          |  1 +
 9 files changed, 90 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index bef9138f1975..03cc975b9e6c 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -1469,6 +1469,7 @@ extern int w_send_barrier(struct drbd_conf *, struct drbd_work *, int);
 extern int w_send_read_req(struct drbd_conf *, struct drbd_work *, int);
 extern int w_prev_work_done(struct drbd_conf *, struct drbd_work *, int);
 extern int w_e_reissue(struct drbd_conf *, struct drbd_work *, int);
+extern int w_restart_disk_io(struct drbd_conf *, struct drbd_work *, int);
 
 extern void resync_timer_fn(unsigned long data);
 
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 7d359863ae32..106b9abdc430 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -925,7 +925,12 @@ static union drbd_state sanitize_state(struct drbd_conf *mdev, union drbd_state
 	if (fp == FP_STONITH &&
 	    (ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.pdsk > D_OUTDATED) &&
 	    !(os.role == R_PRIMARY && os.conn < C_CONNECTED && os.pdsk > D_OUTDATED))
-		ns.susp = 1;
+		ns.susp = 1; /* Suspend IO while fence-peer handler runs (peer lost) */
+
+	if (mdev->sync_conf.on_no_data == OND_SUSPEND_IO &&
+	    (ns.role == R_PRIMARY && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE) &&
+	    !(os.role == R_PRIMARY && os.disk < D_UP_TO_DATE && os.pdsk < D_UP_TO_DATE))
+		ns.susp = 1; /* Suspend IO while no data available (no accessible data available) */
 
 	if (ns.aftr_isp || ns.peer_isp || ns.user_isp) {
 		if (ns.conn == C_SYNC_SOURCE)
@@ -1236,6 +1241,25 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 	/* Here we have the actions that are performed after a
 	   state change. This function might sleep */
 
+	if (os.susp && ns.susp && mdev->sync_conf.on_no_data == OND_SUSPEND_IO) {
+		if (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED) {
+			if (ns.conn == C_CONNECTED) {
+				spin_lock_irq(&mdev->req_lock);
+				_tl_restart(mdev, resend);
+				_drbd_set_state(_NS(mdev, susp, 0), CS_VERBOSE, NULL);
+				spin_unlock_irq(&mdev->req_lock);
+			} else /* ns.conn > C_CONNECTED */
+				dev_err(DEV, "Unexpected Resynd going on!\n");
+		}
+
+		if (os.disk == D_ATTACHING && ns.disk > D_ATTACHING) {
+			spin_lock_irq(&mdev->req_lock);
+			_tl_restart(mdev, restart_frozen_disk_io);
+			_drbd_set_state(_NS(mdev, susp, 0), CS_VERBOSE, NULL);
+			spin_unlock_irq(&mdev->req_lock);
+		}
+	}
+
 	if (fp == FP_STONITH && ns.susp) {
 		/* case1: The outdate peer handler is successful:
 		 * case2: The connection was established again: */
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 73131c5ae339..563a6ade0179 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -33,6 +33,7 @@
 #include <linux/blkpg.h>
 #include <linux/cpumask.h>
 #include "drbd_int.h"
+#include "drbd_req.h"
 #include "drbd_wrappers.h"
 #include <asm/unaligned.h>
 #include <linux/drbd_tag_magic.h>
@@ -494,6 +495,8 @@ char *ppsize(char *buf, unsigned long long size)
 void drbd_suspend_io(struct drbd_conf *mdev)
 {
 	set_bit(SUSPEND_IO, &mdev->flags);
+	if (mdev->state.susp)
+		return;
 	wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_bio_cnt));
 }
 
@@ -1557,6 +1560,7 @@ static int drbd_nl_syncer_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *n
 		sc.rate       = DRBD_RATE_DEF;
 		sc.after      = DRBD_AFTER_DEF;
 		sc.al_extents = DRBD_AL_EXTENTS_DEF;
+		sc.on_no_data  = DRBD_ON_NO_DATA_DEF;
 	} else
 		memcpy(&sc, &mdev->sync_conf, sizeof(struct syncer_conf));
 
@@ -1765,7 +1769,16 @@ static int drbd_nl_suspend_io(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nl
 static int drbd_nl_resume_io(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
 			     struct drbd_nl_cfg_reply *reply)
 {
+	drbd_suspend_io(mdev);
 	reply->ret_code = drbd_request_state(mdev, NS(susp, 0));
+	if (reply->ret_code == SS_SUCCESS) {
+		if (mdev->state.conn < C_CONNECTED)
+			tl_clear(mdev);
+		if (mdev->state.disk == D_DISKLESS || mdev->state.disk == D_FAILED)
+			tl_restart(mdev, fail_frozen_disk_io);
+	}
+	drbd_resume_io(mdev);
+
 	return 0;
 }
 
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index 48647589aa0d..8259d4f77285 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -226,6 +226,8 @@ void _req_may_be_done(struct drbd_request *req, struct bio_and_error *m)
 		return;
 	if (s & RQ_LOCAL_PENDING)
 		return;
+	if (mdev->state.susp)
+		return;
 
 	if (req->master_bio) {
 		/* this is data_received (remote read)
@@ -634,6 +636,28 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what,
 		/* else: done by handed_over_to_network */
 		break;
 
+	case fail_frozen_disk_io:
+		if (!(req->rq_state & RQ_LOCAL_COMPLETED))
+			break;
+
+		_req_may_be_done(req, m);
+		break;
+
+	case restart_frozen_disk_io:
+		if (!(req->rq_state & RQ_LOCAL_COMPLETED))
+			break;
+
+		req->rq_state &= ~RQ_LOCAL_COMPLETED;
+
+		rv = MR_READ;
+		if (bio_data_dir(req->master_bio) == WRITE)
+			rv = MR_WRITE;
+
+		get_ldev(mdev);
+		req->w.cb = w_restart_disk_io;
+		drbd_queue_work(&mdev->data.work, &req->w);
+		break;
+
 	case resend:
 		/* If RQ_NET_OK is already set, we got a P_WRITE_ACK or P_RECV_ACK
 		   before the connection loss; only P_BARRIER_ACK was missing.
diff --git a/drivers/block/drbd/drbd_req.h b/drivers/block/drbd/drbd_req.h
index 07cb3b12edb4..f2e45aaa2cd5 100644
--- a/drivers/block/drbd/drbd_req.h
+++ b/drivers/block/drbd/drbd_req.h
@@ -105,6 +105,8 @@ enum drbd_req_event {
 	write_completed_with_error,
 	completed_ok,
 	resend,
+	fail_frozen_disk_io,
+	restart_frozen_disk_io,
 	nothing, /* for tracing only */
 };
 
diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index ca4a16cea2d8..3c1e88480d37 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -1173,6 +1173,24 @@ int w_send_read_req(struct drbd_conf *mdev, struct drbd_work *w, int cancel)
 	return ok;
 }
 
+int w_restart_disk_io(struct drbd_conf *mdev, struct drbd_work *w, int cancel)
+{
+	struct drbd_request *req = container_of(w, struct drbd_request, w);
+
+	if (bio_data_dir(req->master_bio) == WRITE)
+		drbd_al_begin_io(mdev, req->sector);
+	/* Calling drbd_al_begin_io() out of the worker might deadlocks
+	   theoretically. Practically it can not deadlock, since this is
+	   only used when unfreezing IOs. All the extents of the requests
+	   that made it into the TL are already active */
+
+	drbd_req_make_private_bio(req, req->master_bio);
+	req->private_bio->bi_bdev = mdev->ldev->backing_bdev;
+	generic_make_request(req->private_bio);
+
+	return 1;
+}
+
 static int _drbd_may_sync_now(struct drbd_conf *mdev)
 {
 	struct drbd_conf *odev = mdev;
diff --git a/include/linux/drbd.h b/include/linux/drbd.h
index 479ee3a1d901..7be069fcca57 100644
--- a/include/linux/drbd.h
+++ b/include/linux/drbd.h
@@ -91,6 +91,11 @@ enum drbd_after_sb_p {
 	ASB_VIOLENTLY
 };
 
+enum drbd_on_no_data {
+	OND_IO_ERROR,
+	OND_SUSPEND_IO
+};
+
 /* KEEP the order, do not delete or insert. Only append. */
 enum drbd_ret_codes {
 	ERR_CODE_BASE		= 100,
diff --git a/include/linux/drbd_limits.h b/include/linux/drbd_limits.h
index 440b42e38e89..7eb1e98009ec 100644
--- a/include/linux/drbd_limits.h
+++ b/include/linux/drbd_limits.h
@@ -128,6 +128,7 @@
 #define DRBD_AFTER_SB_1P_DEF ASB_DISCONNECT
 #define DRBD_AFTER_SB_2P_DEF ASB_DISCONNECT
 #define DRBD_RR_CONFLICT_DEF ASB_DISCONNECT
+#define DRBD_ON_NO_DATA_DEF OND_IO_ERROR
 
 #define DRBD_MAX_BIO_BVECS_MIN 0
 #define DRBD_MAX_BIO_BVECS_MAX 128
diff --git a/include/linux/drbd_nl.h b/include/linux/drbd_nl.h
index 5f042810a56c..9aebd0d80a5d 100644
--- a/include/linux/drbd_nl.h
+++ b/include/linux/drbd_nl.h
@@ -87,6 +87,7 @@ NL_PACKET(syncer_conf, 8,
 	NL_STRING(      51,     T_MAY_IGNORE,   cpu_mask,       32)
 	NL_STRING(	64,	T_MAY_IGNORE,	csums_alg,	SHARED_SECRET_MAX)
 	NL_BIT(         65,     T_MAY_IGNORE,   use_rle)
+	NL_INTEGER(	75,	T_MAY_IGNORE,	on_no_data)
 )
 
 NL_PACKET(invalidate, 9, )
-- 
cgit v1.2.3


From 47ff2d0a8e7ce87fed180729e8341f650bf585c8 Mon Sep 17 00:00:00 2001
From: Philipp Reisner <philipp.reisner@linbit.com>
Date: Fri, 18 Jun 2010 13:56:57 +0200
Subject: drbd: Do not allow a fencing-policy of resource-and-stonith with
 protocol A

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_nl.c  | 20 +++++++++++++++++++-
 drivers/block/drbd/drbd_req.c |  2 +-
 include/linux/drbd.h          |  1 +
 3 files changed, 21 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 563a6ade0179..5288bd72cd27 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -806,6 +806,15 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp
 		goto fail;
 	}
 
+	if (get_net_conf(mdev)) {
+		int prot = mdev->net_conf->wire_protocol;
+		put_net_conf(mdev);
+		if (nbc->dc.fencing == FP_STONITH && prot == DRBD_PROT_A) {
+			retcode = ERR_STONITH_AND_PROT_A;
+			goto fail;
+		}
+	}
+
 	nbc->lo_file = filp_open(nbc->dc.backing_dev, O_RDWR, 0);
 	if (IS_ERR(nbc->lo_file)) {
 		dev_err(DEV, "open(\"%s\") failed with %ld\n", nbc->dc.backing_dev,
@@ -1238,7 +1247,16 @@ static int drbd_nl_net_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
 	    && (new_conf->wire_protocol != DRBD_PROT_C)) {
 		retcode = ERR_NOT_PROTO_C;
 		goto fail;
-	};
+	}
+
+	if (get_ldev(mdev)) {
+		enum drbd_fencing_p fp = mdev->ldev->dc.fencing;
+		put_ldev(mdev);
+		if (new_conf->wire_protocol == DRBD_PROT_A && fp == FP_STONITH) {
+			retcode = ERR_STONITH_AND_PROT_A;
+			goto fail;
+		}
+	}
 
 	if (mdev->state.role == R_PRIMARY && new_conf->want_lose) {
 		retcode = ERR_DISCARD;
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index 8259d4f77285..fbe027886bad 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -660,7 +660,7 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what,
 
 	case resend:
 		/* If RQ_NET_OK is already set, we got a P_WRITE_ACK or P_RECV_ACK
-		   before the connection loss; only P_BARRIER_ACK was missing.
+		   before the connection loss (B&C only); only P_BARRIER_ACK was missing.
 		   Trowing them out of the TL here by pretending we got a BARRIER_ACK
 		   TODO: Either resync them, or ensure peer was not rebooted. */
 		if (!(req->rq_state & RQ_NET_OK)) {
diff --git a/include/linux/drbd.h b/include/linux/drbd.h
index 7be069fcca57..0b2bfb58d9c5 100644
--- a/include/linux/drbd.h
+++ b/include/linux/drbd.h
@@ -145,6 +145,7 @@ enum drbd_ret_codes {
 	ERR_CONNECTED		= 151, /* DRBD 8.3 only */
 	ERR_PERM		= 152,
 	ERR_NEED_APV_93		= 153,
+	ERR_STONITH_AND_PROT_A  = 154,
 
 	/* insert new ones above this line */
 	AFTER_LAST_ERR_CODE
-- 
cgit v1.2.3


From 9a31d7164d409ca59cfadb7957ac7b0acf4545b8 Mon Sep 17 00:00:00 2001
From: Philipp Reisner <philipp.reisner@linbit.com>
Date: Mon, 5 Jul 2010 13:42:03 +0200
Subject: drbd: New sync parameters for the smart resync rate controller

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_main.c |  6 +++++-
 drivers/block/drbd/drbd_nl.c   |  4 ++++
 include/linux/drbd_limits.h    | 24 ++++++++++++------------
 include/linux/drbd_nl.h        |  4 ++++
 4 files changed, 25 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 410d3d4f361e..5a484c1f5ce7 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -2690,7 +2690,11 @@ static void drbd_set_defaults(struct drbd_conf *mdev)
 		/* .cpu_mask = */	{}, 0,
 		/* .csums_alg = */	{}, 0,
 		/* .use_rle = */	0,
-		/* .on_no_data = */	DRBD_ON_NO_DATA_DEF
+		/* .on_no_data = */	DRBD_ON_NO_DATA_DEF,
+		/* .c_plan_ahead = */	DRBD_C_PLAN_AHEAD_DEF,
+		/* .c_delay_target = */	DRBD_C_DELAY_TARGET_DEF,
+		/* .c_fill_target = */	DRBD_C_FILL_TARGET_DEF,
+		/* .c_max_rate = */	DRBD_C_MAX_RATE_DEF
 	};
 
 	/* Have to use that way, because the layout differs between
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 6c08e637e25c..7d384fd39c16 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -1599,6 +1599,10 @@ static int drbd_nl_syncer_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *n
 		sc.after      = DRBD_AFTER_DEF;
 		sc.al_extents = DRBD_AL_EXTENTS_DEF;
 		sc.on_no_data  = DRBD_ON_NO_DATA_DEF;
+		sc.c_plan_ahead = DRBD_C_PLAN_AHEAD_DEF;
+		sc.c_delay_target = DRBD_C_DELAY_TARGET_DEF;
+		sc.c_fill_target = DRBD_C_FILL_TARGET_DEF;
+		sc.c_max_rate   = DRBD_C_MAX_RATE_DEF;
 	} else
 		memcpy(&sc, &mdev->sync_conf, sizeof(struct syncer_conf));
 
diff --git a/include/linux/drbd_limits.h b/include/linux/drbd_limits.h
index 7eb1e98009ec..06dbba47a8ef 100644
--- a/include/linux/drbd_limits.h
+++ b/include/linux/drbd_limits.h
@@ -134,21 +134,21 @@
 #define DRBD_MAX_BIO_BVECS_MAX 128
 #define DRBD_MAX_BIO_BVECS_DEF 0
 
-#define DRBD_DP_VOLUME_MIN 4
-#define DRBD_DP_VOLUME_MAX 1048576
-#define DRBD_DP_VOLUME_DEF 16384
+#define DRBD_C_PLAN_AHEAD_MIN  0
+#define DRBD_C_PLAN_AHEAD_MAX  300
+#define DRBD_C_PLAN_AHEAD_DEF  0 /* RS rate controller disabled by default */
 
-#define DRBD_DP_INTERVAL_MIN 1
-#define DRBD_DP_INTERVAL_MAX 600
-#define DRBD_DP_INTERVAL_DEF 5
+#define DRBD_C_DELAY_TARGET_MIN 1
+#define DRBD_C_DELAY_TARGET_MAX 100
+#define DRBD_C_DELAY_TARGET_DEF 10
 
-#define DRBD_RS_THROTTLE_TH_MIN 1
-#define DRBD_RS_THROTTLE_TH_MAX 600
-#define DRBD_RS_THROTTLE_TH_DEF 20
+#define DRBD_C_FILL_TARGET_MIN 0
+#define DRBD_C_FILL_TARGET_MAX 100000
+#define DRBD_C_FILL_TARGET_DEF 0 /* By default disabled -> controlled by delay_target */
 
-#define DRBD_RS_HOLD_OFF_TH_MIN 1
-#define DRBD_RS_HOLD_OFF_TH_MAX 6000
-#define DRBD_RS_HOLD_OFF_TH_DEF 100
+#define DRBD_C_MAX_RATE_MIN     250 /* kByte/sec */
+#define DRBD_C_MAX_RATE_MAX     (4 << 20)
+#define DRBD_C_MAX_RATE_DEF     102400
 
 #undef RANGE
 #endif
diff --git a/include/linux/drbd_nl.h b/include/linux/drbd_nl.h
index 9aebd0d80a5d..e23683c87ca1 100644
--- a/include/linux/drbd_nl.h
+++ b/include/linux/drbd_nl.h
@@ -88,6 +88,10 @@ NL_PACKET(syncer_conf, 8,
 	NL_STRING(	64,	T_MAY_IGNORE,	csums_alg,	SHARED_SECRET_MAX)
 	NL_BIT(         65,     T_MAY_IGNORE,   use_rle)
 	NL_INTEGER(	75,	T_MAY_IGNORE,	on_no_data)
+	NL_INTEGER(	76,	T_MAY_IGNORE,	c_plan_ahead)
+	NL_INTEGER(     77,	T_MAY_IGNORE,	c_delay_target)
+	NL_INTEGER(     78,	T_MAY_IGNORE,	c_fill_target)
+	NL_INTEGER(     79,	T_MAY_IGNORE,	c_max_rate)
 )
 
 NL_PACKET(invalidate, 9, )
-- 
cgit v1.2.3


From 0f0601f4ea2f53cfd8bcae060fb03d9bbde070ec Mon Sep 17 00:00:00 2001
From: Lars Ellenberg <lars.ellenberg@linbit.com>
Date: Wed, 11 Aug 2010 23:40:24 +0200
Subject: drbd: new configuration parameter c-min-rate

We now track the data rate of locally submitted resync related requests,
and can thus detect non-resync activity on the lower level device.

If the current sync rate is above c-min-rate, and the lower level device
appears to be busy, we throttle the resyncer.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_int.h      |  1 +
 drivers/block/drbd/drbd_main.c     |  7 ++-
 drivers/block/drbd/drbd_nl.c       |  3 +-
 drivers/block/drbd/drbd_receiver.c | 88 +++++++++++++++++++++++++++++++++++---
 drivers/block/drbd/drbd_worker.c   | 29 ++++++++-----
 include/linux/drbd_limits.h        |  4 ++
 include/linux/drbd_nl.h            |  1 +
 7 files changed, 116 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index 0fce3f36fc1c..0fedcc0b8dc9 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -1513,6 +1513,7 @@ extern int w_restart_disk_io(struct drbd_conf *, struct drbd_work *, int);
 extern void resync_timer_fn(unsigned long data);
 
 /* drbd_receiver.c */
+extern int drbd_rs_should_slow_down(struct drbd_conf *mdev);
 extern int drbd_submit_ee(struct drbd_conf *mdev, struct drbd_epoch_entry *e,
 		const unsigned rw, const int fault_type);
 extern int drbd_release_ee(struct drbd_conf *mdev, struct list_head *list);
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 1ff8418ae0fa..db93eee7e543 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -1098,6 +1098,8 @@ int __drbd_set_state(struct drbd_conf *mdev,
 		mdev->ov_left = mdev->rs_total
 			      - BM_SECT_TO_BIT(mdev->ov_position);
 		mdev->rs_start = now;
+		mdev->rs_last_events = 0;
+		mdev->rs_last_sect_ev = 0;
 		mdev->ov_last_oos_size = 0;
 		mdev->ov_last_oos_start = 0;
 
@@ -2706,7 +2708,8 @@ static void drbd_set_defaults(struct drbd_conf *mdev)
 		/* .c_plan_ahead = */	DRBD_C_PLAN_AHEAD_DEF,
 		/* .c_delay_target = */	DRBD_C_DELAY_TARGET_DEF,
 		/* .c_fill_target = */	DRBD_C_FILL_TARGET_DEF,
-		/* .c_max_rate = */	DRBD_C_MAX_RATE_DEF
+		/* .c_max_rate = */	DRBD_C_MAX_RATE_DEF,
+		/* .c_min_rate = */	DRBD_C_MIN_RATE_DEF
 	};
 
 	/* Have to use that way, because the layout differs between
@@ -2742,6 +2745,7 @@ void drbd_init_set_defaults(struct drbd_conf *mdev)
 	atomic_set(&mdev->packet_seq, 0);
 	atomic_set(&mdev->pp_in_use, 0);
 	atomic_set(&mdev->rs_sect_in, 0);
+	atomic_set(&mdev->rs_sect_ev, 0);
 
 	mutex_init(&mdev->md_io_mutex);
 	mutex_init(&mdev->data.mutex);
@@ -2819,6 +2823,7 @@ void drbd_mdev_cleanup(struct drbd_conf *mdev)
 	mdev->rs_total     =
 	mdev->rs_failed    = 0;
 	mdev->rs_last_events = 0;
+	mdev->rs_last_sect_ev = 0;
 	for (i = 0; i < DRBD_SYNC_MARKS; i++) {
 		mdev->rs_mark_left[i] = 0;
 		mdev->rs_mark_time[i] = 0;
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 295b8d593708..6b35d41706e4 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -1604,7 +1604,8 @@ static int drbd_nl_syncer_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *n
 		sc.c_plan_ahead = DRBD_C_PLAN_AHEAD_DEF;
 		sc.c_delay_target = DRBD_C_DELAY_TARGET_DEF;
 		sc.c_fill_target = DRBD_C_FILL_TARGET_DEF;
-		sc.c_max_rate   = DRBD_C_MAX_RATE_DEF;
+		sc.c_max_rate = DRBD_C_MAX_RATE_DEF;
+		sc.c_min_rate = DRBD_C_MIN_RATE_DEF;
 	} else
 		memcpy(&sc, &mdev->sync_conf, sizeof(struct syncer_conf));
 
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 346aed98027f..0d9967fef528 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -1561,6 +1561,7 @@ static int recv_resync_read(struct drbd_conf *mdev, sector_t sector, int data_si
 	list_add(&e->w.list, &mdev->sync_ee);
 	spin_unlock_irq(&mdev->req_lock);
 
+	atomic_add(data_size >> 9, &mdev->rs_sect_ev);
 	if (drbd_submit_ee(mdev, e, WRITE, DRBD_FAULT_RS_WR) == 0)
 		return TRUE;
 
@@ -2017,17 +2018,66 @@ out_interrupted:
 	return FALSE;
 }
 
+/* We may throttle resync, if the lower device seems to be busy,
+ * and current sync rate is above c_min_rate.
+ *
+ * To decide whether or not the lower device is busy, we use a scheme similar
+ * to MD RAID is_mddev_idle(): if the partition stats reveal "significant"
+ * (more than 64 sectors) of activity we cannot account for with our own resync
+ * activity, it obviously is "busy".
+ *
+ * The current sync rate used here uses only the most recent two step marks,
+ * to have a short time average so we can react faster.
+ */
+int drbd_rs_should_slow_down(struct drbd_conf *mdev)
+{
+	struct gendisk *disk = mdev->ldev->backing_bdev->bd_contains->bd_disk;
+	unsigned long db, dt, dbdt;
+	int curr_events;
+	int throttle = 0;
+
+	/* feature disabled? */
+	if (mdev->sync_conf.c_min_rate == 0)
+		return 0;
+
+	curr_events = (int)part_stat_read(&disk->part0, sectors[0]) +
+		      (int)part_stat_read(&disk->part0, sectors[1]) -
+			atomic_read(&mdev->rs_sect_ev);
+	if (!mdev->rs_last_events || curr_events - mdev->rs_last_events > 64) {
+		unsigned long rs_left;
+		int i;
+
+		mdev->rs_last_events = curr_events;
+
+		/* sync speed average over the last 2*DRBD_SYNC_MARK_STEP,
+		 * approx. */
+		i = (mdev->rs_last_mark + DRBD_SYNC_MARKS-2) % DRBD_SYNC_MARKS;
+		rs_left = drbd_bm_total_weight(mdev) - mdev->rs_failed;
+
+		dt = ((long)jiffies - (long)mdev->rs_mark_time[i]) / HZ;
+		if (!dt)
+			dt++;
+		db = mdev->rs_mark_left[i] - rs_left;
+		dbdt = Bit2KB(db/dt);
+
+		if (dbdt > mdev->sync_conf.c_min_rate)
+			throttle = 1;
+	}
+	return throttle;
+}
+
+
 static int receive_DataRequest(struct drbd_conf *mdev, struct p_header *h)
 {
 	sector_t sector;
 	const sector_t capacity = drbd_get_capacity(mdev->this_bdev);
 	struct drbd_epoch_entry *e;
 	struct digest_info *di = NULL;
+	struct p_block_req *p = (struct p_block_req *)h;
+	const int brps = sizeof(*p)-sizeof(*h);
 	int size, digest_size;
 	unsigned int fault_type;
-	struct p_block_req *p =
-		(struct p_block_req *)h;
-	const int brps = sizeof(*p)-sizeof(*h);
+
 
 	if (drbd_recv(mdev, h->payload, brps) != brps)
 		return FALSE;
@@ -2099,8 +2149,9 @@ static int receive_DataRequest(struct drbd_conf *mdev, struct p_header *h)
 		} else if (h->command == P_OV_REPLY) {
 			e->w.cb = w_e_end_ov_reply;
 			dec_rs_pending(mdev);
-			/* drbd_rs_begin_io done when we sent this request */
-			goto submit;
+			/* drbd_rs_begin_io done when we sent this request,
+			 * but accounting still needs to be done. */
+			goto submit_for_resync;
 		}
 		break;
 
@@ -2128,9 +2179,36 @@ static int receive_DataRequest(struct drbd_conf *mdev, struct p_header *h)
 		goto out_free_e;
 	}
 
+	/* Throttle, drbd_rs_begin_io and submit should become asynchronous
+	 * wrt the receiver, but it is not as straightforward as it may seem.
+	 * Various places in the resync start and stop logic assume resync
+	 * requests are processed in order, requeuing this on the worker thread
+	 * introduces a bunch of new code for synchronization between threads.
+	 *
+	 * Unlimited throttling before drbd_rs_begin_io may stall the resync
+	 * "forever", throttling after drbd_rs_begin_io will lock that extent
+	 * for application writes for the same time.  For now, just throttle
+	 * here, where the rest of the code expects the receiver to sleep for
+	 * a while, anyways.
+	 */
+
+	/* Throttle before drbd_rs_begin_io, as that locks out application IO;
+	 * this defers syncer requests for some time, before letting at least
+	 * on request through.  The resync controller on the receiving side
+	 * will adapt to the incoming rate accordingly.
+	 *
+	 * We cannot throttle here if remote is Primary/SyncTarget:
+	 * we would also throttle its application reads.
+	 * In that case, throttling is done on the SyncTarget only.
+	 */
+	if (mdev->state.peer != R_PRIMARY && drbd_rs_should_slow_down(mdev))
+		msleep(100);
 	if (drbd_rs_begin_io(mdev, e->sector))
 		goto out_free_e;
 
+submit_for_resync:
+	atomic_add(size >> 9, &mdev->rs_sect_ev);
+
 submit:
 	inc_unacked(mdev);
 	spin_lock_irq(&mdev->req_lock);
diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index f5d779b4d685..99c937acb471 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -215,10 +215,8 @@ void drbd_endio_sec(struct bio *bio, int error)
  */
 void drbd_endio_pri(struct bio *bio, int error)
 {
-	unsigned long flags;
 	struct drbd_request *req = bio->bi_private;
 	struct drbd_conf *mdev = req->mdev;
-	struct bio_and_error m;
 	enum drbd_req_event what;
 	int uptodate = bio_flagged(bio, BIO_UPTODATE);
 
@@ -244,12 +242,7 @@ void drbd_endio_pri(struct bio *bio, int error)
 	bio_put(req->private_bio);
 	req->private_bio = ERR_PTR(error);
 
-	spin_lock_irqsave(&mdev->req_lock, flags);
-	__req_mod(req, what, &m);
-	spin_unlock_irqrestore(&mdev->req_lock, flags);
-
-	if (m.bio)
-		complete_master_bio(mdev, &m);
+	req_mod(req, what);
 }
 
 int w_read_retry_remote(struct drbd_conf *mdev, struct drbd_work *w, int cancel)
@@ -376,6 +369,9 @@ static int read_for_csum(struct drbd_conf *mdev, sector_t sector, int size)
 	if (!get_ldev(mdev))
 		return -EIO;
 
+	if (drbd_rs_should_slow_down(mdev))
+		goto defer;
+
 	/* GFP_TRY, because if there is no memory available right now, this may
 	 * be rescheduled for later. It is "only" background resync, after all. */
 	e = drbd_alloc_ee(mdev, DRBD_MAGIC+0xbeef, sector, size, GFP_TRY);
@@ -387,6 +383,7 @@ static int read_for_csum(struct drbd_conf *mdev, sector_t sector, int size)
 	list_add(&e->w.list, &mdev->read_ee);
 	spin_unlock_irq(&mdev->req_lock);
 
+	atomic_add(size >> 9, &mdev->rs_sect_ev);
 	if (drbd_submit_ee(mdev, e, READ, DRBD_FAULT_RS_RD) == 0)
 		return 0;
 
@@ -512,8 +509,9 @@ int w_make_resync_request(struct drbd_conf *mdev,
 	sector_t sector;
 	const sector_t capacity = drbd_get_capacity(mdev->this_bdev);
 	int max_segment_size;
-	int number, i, rollback_i, size, pe, mx;
+	int number, rollback_i, size, pe, mx;
 	int align, queued, sndbuf;
+	int i = 0;
 
 	if (unlikely(cancel))
 		return 1;
@@ -549,7 +547,14 @@ int w_make_resync_request(struct drbd_conf *mdev,
 		mdev->c_sync_rate = mdev->sync_conf.rate;
 		number = SLEEP_TIME * mdev->c_sync_rate  / ((BM_BLOCK_SIZE / 1024) * HZ);
 	}
-	pe = atomic_read(&mdev->rs_pending_cnt);
+
+	/* Throttle resync on lower level disk activity, which may also be
+	 * caused by application IO on Primary/SyncTarget.
+	 * Keep this after the call to drbd_rs_controller, as that assumes
+	 * to be called as precisely as possible every SLEEP_TIME,
+	 * and would be confused otherwise. */
+	if (drbd_rs_should_slow_down(mdev))
+		goto requeue;
 
 	mutex_lock(&mdev->data.mutex);
 	if (mdev->data.socket)
@@ -563,6 +568,7 @@ int w_make_resync_request(struct drbd_conf *mdev,
 		mx = number;
 
 	/* Limit the number of pending RS requests to no more than the peer's receive buffer */
+	pe = atomic_read(&mdev->rs_pending_cnt);
 	if ((pe + number) > mx) {
 		number = mx - pe;
 	}
@@ -1492,6 +1498,8 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side)
 		mdev->rs_failed    = 0;
 		mdev->rs_paused    = 0;
 		mdev->rs_same_csum = 0;
+		mdev->rs_last_events = 0;
+		mdev->rs_last_sect_ev = 0;
 		mdev->rs_total     = tw;
 		mdev->rs_start     = now;
 		for (i = 0; i < DRBD_SYNC_MARKS; i++) {
@@ -1516,6 +1524,7 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side)
 		}
 
 		atomic_set(&mdev->rs_sect_in, 0);
+		atomic_set(&mdev->rs_sect_ev, 0);
 		mdev->rs_in_flight = 0;
 		mdev->rs_planed = 0;
 		spin_lock(&mdev->peer_seq_lock);
diff --git a/include/linux/drbd_limits.h b/include/linux/drbd_limits.h
index 06dbba47a8ef..0b24ded6fffd 100644
--- a/include/linux/drbd_limits.h
+++ b/include/linux/drbd_limits.h
@@ -150,5 +150,9 @@
 #define DRBD_C_MAX_RATE_MAX     (4 << 20)
 #define DRBD_C_MAX_RATE_DEF     102400
 
+#define DRBD_C_MIN_RATE_MIN     0 /* kByte/sec */
+#define DRBD_C_MIN_RATE_MAX     (4 << 20)
+#define DRBD_C_MIN_RATE_DEF     4096
+
 #undef RANGE
 #endif
diff --git a/include/linux/drbd_nl.h b/include/linux/drbd_nl.h
index e23683c87ca1..ade91107c9a5 100644
--- a/include/linux/drbd_nl.h
+++ b/include/linux/drbd_nl.h
@@ -92,6 +92,7 @@ NL_PACKET(syncer_conf, 8,
 	NL_INTEGER(     77,	T_MAY_IGNORE,	c_delay_target)
 	NL_INTEGER(     78,	T_MAY_IGNORE,	c_fill_target)
 	NL_INTEGER(     79,	T_MAY_IGNORE,	c_max_rate)
+	NL_INTEGER(     80,	T_MAY_IGNORE,	c_min_rate)
 )
 
 NL_PACKET(invalidate, 9, )
-- 
cgit v1.2.3


From 0b70a13dac014ec9274640b9e945bde493ba365e Mon Sep 17 00:00:00 2001
From: Philipp Reisner <philipp.reisner@linbit.com>
Date: Fri, 20 Aug 2010 13:36:10 +0200
Subject: drbd: Sending of big packets, for payloads from 64KByte to 4GByte

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_int.h      | 74 +++++++++++++++++-------------
 drivers/block/drbd/drbd_main.c     | 79 +++++++++++++++++++-------------
 drivers/block/drbd/drbd_receiver.c | 94 +++++++++++++++++++-------------------
 drivers/block/drbd/drbd_worker.c   |  2 +-
 include/linux/drbd.h               |  2 +
 5 files changed, 139 insertions(+), 112 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index 0fedcc0b8dc9..3f10efc2ac14 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -337,13 +337,25 @@ static inline void bm_xfer_ctx_bit_to_word_offset(struct bm_xfer_ctx *c)
  * NOTE that the payload starts at a long aligned offset,
  * regardless of 32 or 64 bit arch!
  */
-struct p_header {
+struct p_header80 {
 	u32	  magic;
 	u16	  command;
 	u16	  length;	/* bytes of data after this header */
 	u8	  payload[0];
 } __packed;
-/* 8 bytes. packet FIXED for the next century! */
+
+/* Header for big packets, Used for data packets exceeding 64kB */
+struct p_header95 {
+	u16	  magic;	/* use DRBD_MAGIC_BIG here */
+	u16	  command;
+	u32	  length;
+	u8	  payload[0];
+} __packed;
+
+union p_header {
+	struct p_header80 h80;
+	struct p_header95 h95;
+};
 
 /*
  * short commands, packets without payload, plain p_header:
@@ -367,7 +379,7 @@ struct p_header {
 #define DP_MAY_SET_IN_SYNC    4
 
 struct p_data {
-	struct p_header head;
+	union p_header head;
 	u64	    sector;    /* 64 bits sector number */
 	u64	    block_id;  /* to identify the request in protocol B&C */
 	u32	    seq_num;
@@ -383,7 +395,7 @@ struct p_data {
  *   P_DATA_REQUEST, P_RS_DATA_REQUEST
  */
 struct p_block_ack {
-	struct p_header head;
+	struct p_header80 head;
 	u64	    sector;
 	u64	    block_id;
 	u32	    blksize;
@@ -392,7 +404,7 @@ struct p_block_ack {
 
 
 struct p_block_req {
-	struct p_header head;
+	struct p_header80 head;
 	u64 sector;
 	u64 block_id;
 	u32 blksize;
@@ -409,7 +421,7 @@ struct p_block_req {
  */
 
 struct p_handshake {
-	struct p_header head;	/* 8 bytes */
+	struct p_header80 head;	/* 8 bytes */
 	u32 protocol_min;
 	u32 feature_flags;
 	u32 protocol_max;
@@ -424,19 +436,19 @@ struct p_handshake {
 /* 80 bytes, FIXED for the next century */
 
 struct p_barrier {
-	struct p_header head;
+	struct p_header80 head;
 	u32 barrier;	/* barrier number _handle_ only */
 	u32 pad;	/* to multiple of 8 Byte */
 } __packed;
 
 struct p_barrier_ack {
-	struct p_header head;
+	struct p_header80 head;
 	u32 barrier;
 	u32 set_size;
 } __packed;
 
 struct p_rs_param {
-	struct p_header head;
+	struct p_header80 head;
 	u32 rate;
 
 	      /* Since protocol version 88 and higher. */
@@ -444,7 +456,7 @@ struct p_rs_param {
 } __packed;
 
 struct p_rs_param_89 {
-	struct p_header head;
+	struct p_header80 head;
 	u32 rate;
         /* protocol version 89: */
 	char verify_alg[SHARED_SECRET_MAX];
@@ -452,7 +464,7 @@ struct p_rs_param_89 {
 } __packed;
 
 struct p_rs_param_95 {
-	struct p_header head;
+	struct p_header80 head;
 	u32 rate;
 	char verify_alg[SHARED_SECRET_MAX];
 	char csums_alg[SHARED_SECRET_MAX];
@@ -468,7 +480,7 @@ enum drbd_conn_flags {
 };
 
 struct p_protocol {
-	struct p_header head;
+	struct p_header80 head;
 	u32 protocol;
 	u32 after_sb_0p;
 	u32 after_sb_1p;
@@ -482,17 +494,17 @@ struct p_protocol {
 } __packed;
 
 struct p_uuids {
-	struct p_header head;
+	struct p_header80 head;
 	u64 uuid[UI_EXTENDED_SIZE];
 } __packed;
 
 struct p_rs_uuid {
-	struct p_header head;
+	struct p_header80 head;
 	u64	    uuid;
 } __packed;
 
 struct p_sizes {
-	struct p_header head;
+	struct p_header80 head;
 	u64	    d_size;  /* size of disk */
 	u64	    u_size;  /* user requested size */
 	u64	    c_size;  /* current exported size */
@@ -502,18 +514,18 @@ struct p_sizes {
 } __packed;
 
 struct p_state {
-	struct p_header head;
+	struct p_header80 head;
 	u32	    state;
 } __packed;
 
 struct p_req_state {
-	struct p_header head;
+	struct p_header80 head;
 	u32	    mask;
 	u32	    val;
 } __packed;
 
 struct p_req_state_reply {
-	struct p_header head;
+	struct p_header80 head;
 	u32	    retcode;
 } __packed;
 
@@ -528,7 +540,7 @@ struct p_drbd06_param {
 } __packed;
 
 struct p_discard {
-	struct p_header head;
+	struct p_header80 head;
 	u64	    block_id;
 	u32	    seq_num;
 	u32	    pad;
@@ -544,7 +556,7 @@ enum drbd_bitmap_code {
 };
 
 struct p_compressed_bm {
-	struct p_header head;
+	struct p_header80 head;
 	/* (encoding & 0x0f): actual encoding, see enum drbd_bitmap_code
 	 * (encoding & 0x80): polarity (set/unset) of first runlength
 	 * ((encoding >> 4) & 0x07): pad_bits, number of trailing zero bits
@@ -555,10 +567,10 @@ struct p_compressed_bm {
 	u8 code[0];
 } __packed;
 
-struct p_delay_probe {
-	struct p_header head;
-	u32	seq_num; /* sequence number to match the two probe packets */
-	u32	offset;	 /* usecs the probe got sent after the reference time point */
+struct p_delay_probe93 {
+	struct p_header80 head;
+	u32     seq_num; /* sequence number to match the two probe packets */
+	u32     offset;  /* usecs the probe got sent after the reference time point */
 } __packed;
 
 /* DCBP: Drbd Compressed Bitmap Packet ... */
@@ -605,7 +617,7 @@ DCBP_set_pad_bits(struct p_compressed_bm *p, int n)
  * so we need to use the fixed size 4KiB page size
  * most architechtures have used for a long time.
  */
-#define BM_PACKET_PAYLOAD_BYTES (4096 - sizeof(struct p_header))
+#define BM_PACKET_PAYLOAD_BYTES (4096 - sizeof(struct p_header80))
 #define BM_PACKET_WORDS (BM_PACKET_PAYLOAD_BYTES/sizeof(long))
 #define BM_PACKET_VLI_BYTES_MAX (4096 - sizeof(struct p_compressed_bm))
 #if (PAGE_SIZE < 4096)
@@ -614,7 +626,7 @@ DCBP_set_pad_bits(struct p_compressed_bm *p, int n)
 #endif
 
 union p_polymorph {
-        struct p_header          header;
+        struct p_header80        header;
         struct p_handshake       handshake;
         struct p_data            data;
         struct p_block_ack       block_ack;
@@ -1188,12 +1200,12 @@ extern int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_f
 extern int _drbd_send_state(struct drbd_conf *mdev);
 extern int drbd_send_state(struct drbd_conf *mdev);
 extern int _drbd_send_cmd(struct drbd_conf *mdev, struct socket *sock,
-			enum drbd_packets cmd, struct p_header *h,
+			enum drbd_packets cmd, struct p_header80 *h,
 			size_t size, unsigned msg_flags);
 #define USE_DATA_SOCKET 1
 #define USE_META_SOCKET 0
 extern int drbd_send_cmd(struct drbd_conf *mdev, int use_data_socket,
-			enum drbd_packets cmd, struct p_header *h,
+			enum drbd_packets cmd, struct p_header80 *h,
 			size_t size);
 extern int drbd_send_cmd2(struct drbd_conf *mdev, enum drbd_packets cmd,
 			char *data, size_t size);
@@ -1936,19 +1948,19 @@ static inline void request_ping(struct drbd_conf *mdev)
 static inline int drbd_send_short_cmd(struct drbd_conf *mdev,
 	enum drbd_packets cmd)
 {
-	struct p_header h;
+	struct p_header80 h;
 	return drbd_send_cmd(mdev, USE_DATA_SOCKET, cmd, &h, sizeof(h));
 }
 
 static inline int drbd_send_ping(struct drbd_conf *mdev)
 {
-	struct p_header h;
+	struct p_header80 h;
 	return drbd_send_cmd(mdev, USE_META_SOCKET, P_PING, &h, sizeof(h));
 }
 
 static inline int drbd_send_ping_ack(struct drbd_conf *mdev)
 {
-	struct p_header h;
+	struct p_header80 h;
 	return drbd_send_cmd(mdev, USE_META_SOCKET, P_PING_ACK, &h, sizeof(h));
 }
 
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index db93eee7e543..f3f4ea9c5eb9 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -1647,7 +1647,7 @@ void drbd_thread_current_set_cpu(struct drbd_conf *mdev)
 
 /* the appropriate socket mutex must be held already */
 int _drbd_send_cmd(struct drbd_conf *mdev, struct socket *sock,
-			  enum drbd_packets cmd, struct p_header *h,
+			  enum drbd_packets cmd, struct p_header80 *h,
 			  size_t size, unsigned msg_flags)
 {
 	int sent, ok;
@@ -1657,7 +1657,7 @@ int _drbd_send_cmd(struct drbd_conf *mdev, struct socket *sock,
 
 	h->magic   = BE_DRBD_MAGIC;
 	h->command = cpu_to_be16(cmd);
-	h->length  = cpu_to_be16(size-sizeof(struct p_header));
+	h->length  = cpu_to_be16(size-sizeof(struct p_header80));
 
 	sent = drbd_send(mdev, sock, h, size, msg_flags);
 
@@ -1672,7 +1672,7 @@ int _drbd_send_cmd(struct drbd_conf *mdev, struct socket *sock,
  * when we hold the appropriate socket mutex.
  */
 int drbd_send_cmd(struct drbd_conf *mdev, int use_data_socket,
-		  enum drbd_packets cmd, struct p_header *h, size_t size)
+		  enum drbd_packets cmd, struct p_header80 *h, size_t size)
 {
 	int ok = 0;
 	struct socket *sock;
@@ -1700,7 +1700,7 @@ int drbd_send_cmd(struct drbd_conf *mdev, int use_data_socket,
 int drbd_send_cmd2(struct drbd_conf *mdev, enum drbd_packets cmd, char *data,
 		   size_t size)
 {
-	struct p_header h;
+	struct p_header80 h;
 	int ok;
 
 	h.magic   = BE_DRBD_MAGIC;
@@ -1807,7 +1807,7 @@ int drbd_send_protocol(struct drbd_conf *mdev)
 		strcpy(p->integrity_alg, mdev->net_conf->integrity_alg);
 
 	rv = drbd_send_cmd(mdev, USE_DATA_SOCKET, P_PROTOCOL,
-			   (struct p_header *)p, size);
+			   (struct p_header80 *)p, size);
 	kfree(p);
 	return rv;
 }
@@ -1833,7 +1833,7 @@ int _drbd_send_uuids(struct drbd_conf *mdev, u64 uuid_flags)
 	put_ldev(mdev);
 
 	return drbd_send_cmd(mdev, USE_DATA_SOCKET, P_UUIDS,
-			     (struct p_header *)&p, sizeof(p));
+			     (struct p_header80 *)&p, sizeof(p));
 }
 
 int drbd_send_uuids(struct drbd_conf *mdev)
@@ -1854,7 +1854,7 @@ int drbd_send_sync_uuid(struct drbd_conf *mdev, u64 val)
 	p.uuid = cpu_to_be64(val);
 
 	return drbd_send_cmd(mdev, USE_DATA_SOCKET, P_SYNC_UUID,
-			     (struct p_header *)&p, sizeof(p));
+			     (struct p_header80 *)&p, sizeof(p));
 }
 
 int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags flags)
@@ -1884,7 +1884,7 @@ int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags fl
 	p.dds_flags = cpu_to_be16(flags);
 
 	ok = drbd_send_cmd(mdev, USE_DATA_SOCKET, P_SIZES,
-			   (struct p_header *)&p, sizeof(p));
+			   (struct p_header80 *)&p, sizeof(p));
 	return ok;
 }
 
@@ -1909,7 +1909,7 @@ int drbd_send_state(struct drbd_conf *mdev)
 
 	if (likely(sock != NULL)) {
 		ok = _drbd_send_cmd(mdev, sock, P_STATE,
-				    (struct p_header *)&p, sizeof(p), 0);
+				    (struct p_header80 *)&p, sizeof(p), 0);
 	}
 
 	mutex_unlock(&mdev->data.mutex);
@@ -1927,7 +1927,7 @@ int drbd_send_state_req(struct drbd_conf *mdev,
 	p.val     = cpu_to_be32(val.i);
 
 	return drbd_send_cmd(mdev, USE_DATA_SOCKET, P_STATE_CHG_REQ,
-			     (struct p_header *)&p, sizeof(p));
+			     (struct p_header80 *)&p, sizeof(p));
 }
 
 int drbd_send_sr_reply(struct drbd_conf *mdev, int retcode)
@@ -1937,7 +1937,7 @@ int drbd_send_sr_reply(struct drbd_conf *mdev, int retcode)
 	p.retcode    = cpu_to_be32(retcode);
 
 	return drbd_send_cmd(mdev, USE_META_SOCKET, P_STATE_CHG_REPLY,
-			     (struct p_header *)&p, sizeof(p));
+			     (struct p_header80 *)&p, sizeof(p));
 }
 
 int fill_bitmap_rle_bits(struct drbd_conf *mdev,
@@ -2036,7 +2036,7 @@ int fill_bitmap_rle_bits(struct drbd_conf *mdev,
 
 enum { OK, FAILED, DONE }
 send_bitmap_rle_or_plain(struct drbd_conf *mdev,
-	struct p_header *h, struct bm_xfer_ctx *c)
+	struct p_header80 *h, struct bm_xfer_ctx *c)
 {
 	struct p_compressed_bm *p = (void*)h;
 	unsigned long num_words;
@@ -2066,12 +2066,12 @@ send_bitmap_rle_or_plain(struct drbd_conf *mdev,
 		if (len)
 			drbd_bm_get_lel(mdev, c->word_offset, num_words, (unsigned long*)h->payload);
 		ok = _drbd_send_cmd(mdev, mdev->data.socket, P_BITMAP,
-				   h, sizeof(struct p_header) + len, 0);
+				   h, sizeof(struct p_header80) + len, 0);
 		c->word_offset += num_words;
 		c->bit_offset = c->word_offset * BITS_PER_LONG;
 
 		c->packets[1]++;
-		c->bytes[1] += sizeof(struct p_header) + len;
+		c->bytes[1] += sizeof(struct p_header80) + len;
 
 		if (c->bit_offset > c->bm_bits)
 			c->bit_offset = c->bm_bits;
@@ -2087,14 +2087,14 @@ send_bitmap_rle_or_plain(struct drbd_conf *mdev,
 int _drbd_send_bitmap(struct drbd_conf *mdev)
 {
 	struct bm_xfer_ctx c;
-	struct p_header *p;
+	struct p_header80 *p;
 	int ret;
 
 	ERR_IF(!mdev->bitmap) return FALSE;
 
 	/* maybe we should use some per thread scratch page,
 	 * and allocate that during initial device creation? */
-	p = (struct p_header *) __get_free_page(GFP_NOIO);
+	p = (struct p_header80 *) __get_free_page(GFP_NOIO);
 	if (!p) {
 		dev_err(DEV, "failed to allocate one page buffer in %s\n", __func__);
 		return FALSE;
@@ -2152,7 +2152,7 @@ int drbd_send_b_ack(struct drbd_conf *mdev, u32 barrier_nr, u32 set_size)
 	if (mdev->state.conn < C_CONNECTED)
 		return FALSE;
 	ok = drbd_send_cmd(mdev, USE_META_SOCKET, P_BARRIER_ACK,
-			(struct p_header *)&p, sizeof(p));
+			(struct p_header80 *)&p, sizeof(p));
 	return ok;
 }
 
@@ -2180,7 +2180,7 @@ static int _drbd_send_ack(struct drbd_conf *mdev, enum drbd_packets cmd,
 	if (!mdev->meta.socket || mdev->state.conn < C_CONNECTED)
 		return FALSE;
 	ok = drbd_send_cmd(mdev, USE_META_SOCKET, cmd,
-				(struct p_header *)&p, sizeof(p));
+				(struct p_header80 *)&p, sizeof(p));
 	return ok;
 }
 
@@ -2188,8 +2188,8 @@ int drbd_send_ack_dp(struct drbd_conf *mdev, enum drbd_packets cmd,
 		     struct p_data *dp)
 {
 	const int header_size = sizeof(struct p_data)
-			      - sizeof(struct p_header);
-	int data_size  = ((struct p_header *)dp)->length - header_size;
+			      - sizeof(struct p_header80);
+	int data_size  = ((struct p_header80 *)dp)->length - header_size;
 
 	return _drbd_send_ack(mdev, cmd, dp->sector, cpu_to_be32(data_size),
 			      dp->block_id);
@@ -2238,7 +2238,7 @@ int drbd_send_drequest(struct drbd_conf *mdev, int cmd,
 	p.blksize  = cpu_to_be32(size);
 
 	ok = drbd_send_cmd(mdev, USE_DATA_SOCKET, cmd,
-				(struct p_header *)&p, sizeof(p));
+				(struct p_header80 *)&p, sizeof(p));
 	return ok;
 }
 
@@ -2256,7 +2256,7 @@ int drbd_send_drequest_csum(struct drbd_conf *mdev,
 
 	p.head.magic   = BE_DRBD_MAGIC;
 	p.head.command = cpu_to_be16(cmd);
-	p.head.length  = cpu_to_be16(sizeof(p) - sizeof(struct p_header) + digest_size);
+	p.head.length  = cpu_to_be16(sizeof(p) - sizeof(struct p_header80) + digest_size);
 
 	mutex_lock(&mdev->data.mutex);
 
@@ -2278,7 +2278,7 @@ int drbd_send_ov_request(struct drbd_conf *mdev, sector_t sector, int size)
 	p.blksize  = cpu_to_be32(size);
 
 	ok = drbd_send_cmd(mdev, USE_DATA_SOCKET, P_OV_REQUEST,
-			   (struct p_header *)&p, sizeof(p));
+			   (struct p_header80 *)&p, sizeof(p));
 	return ok;
 }
 
@@ -2447,10 +2447,17 @@ int drbd_send_dblock(struct drbd_conf *mdev, struct drbd_request *req)
 	dgs = (mdev->agreed_pro_version >= 87 && mdev->integrity_w_tfm) ?
 		crypto_hash_digestsize(mdev->integrity_w_tfm) : 0;
 
-	p.head.magic   = BE_DRBD_MAGIC;
-	p.head.command = cpu_to_be16(P_DATA);
-	p.head.length  =
-		cpu_to_be16(sizeof(p) - sizeof(struct p_header) + dgs + req->size);
+	if (req->size <= (1 << 15)) {
+		p.head.h80.magic   = BE_DRBD_MAGIC;
+		p.head.h80.command = cpu_to_be16(P_DATA);
+		p.head.h80.length  =
+			cpu_to_be16(sizeof(p) - sizeof(union p_header) + dgs + req->size);
+	} else {
+		p.head.h95.magic   = BE_DRBD_MAGIC_BIG;
+		p.head.h95.command = cpu_to_be16(P_DATA);
+		p.head.h95.length  =
+			cpu_to_be32(sizeof(p) - sizeof(union p_header) + dgs + req->size);
+	}
 
 	p.sector   = cpu_to_be64(req->sector);
 	p.block_id = (unsigned long)req;
@@ -2511,10 +2518,17 @@ int drbd_send_block(struct drbd_conf *mdev, enum drbd_packets cmd,
 	dgs = (mdev->agreed_pro_version >= 87 && mdev->integrity_w_tfm) ?
 		crypto_hash_digestsize(mdev->integrity_w_tfm) : 0;
 
-	p.head.magic   = BE_DRBD_MAGIC;
-	p.head.command = cpu_to_be16(cmd);
-	p.head.length  =
-		cpu_to_be16(sizeof(p) - sizeof(struct p_header) + dgs + e->size);
+	if (e->size <= (1 << 15)) {
+		p.head.h80.magic   = BE_DRBD_MAGIC;
+		p.head.h80.command = cpu_to_be16(cmd);
+		p.head.h80.length  =
+			cpu_to_be16(sizeof(p) - sizeof(struct p_header80) + dgs + e->size);
+	} else {
+		p.head.h95.magic   = BE_DRBD_MAGIC_BIG;
+		p.head.h95.command = cpu_to_be16(cmd);
+		p.head.h95.length  =
+			cpu_to_be32(sizeof(p) - sizeof(struct p_header80) + dgs + e->size);
+	}
 
 	p.sector   = cpu_to_be64(e->sector);
 	p.block_id = e->block_id;
@@ -2527,8 +2541,7 @@ int drbd_send_block(struct drbd_conf *mdev, enum drbd_packets cmd,
 	if (!drbd_get_data_sock(mdev))
 		return 0;
 
-	ok = sizeof(p) == drbd_send(mdev, mdev->data.socket, &p,
-					sizeof(p), dgs ? MSG_MORE : 0);
+	ok = sizeof(p) == drbd_send(mdev, mdev->data.socket, &p, sizeof(p), dgs ? MSG_MORE : 0);
 	if (ok && dgs) {
 		dgb = mdev->int_dig_out;
 		drbd_csum_ee(mdev, mdev->integrity_w_tfm, e, dgb);
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 591a171291d9..9b3321e2c3cd 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -720,14 +720,14 @@ out:
 static int drbd_send_fp(struct drbd_conf *mdev,
 	struct socket *sock, enum drbd_packets cmd)
 {
-	struct p_header *h = (struct p_header *) &mdev->data.sbuf.header;
+	struct p_header80 *h = (struct p_header80 *) &mdev->data.sbuf.header;
 
 	return _drbd_send_cmd(mdev, sock, cmd, h, sizeof(*h), 0);
 }
 
 static enum drbd_packets drbd_recv_fp(struct drbd_conf *mdev, struct socket *sock)
 {
-	struct p_header *h = (struct p_header *) &mdev->data.sbuf.header;
+	struct p_header80 *h = (struct p_header80 *) &mdev->data.sbuf.header;
 	int rr;
 
 	rr = drbd_recv_short(mdev, sock, h, sizeof(*h), 0);
@@ -944,7 +944,7 @@ out_release_sockets:
 	return -1;
 }
 
-static int drbd_recv_header(struct drbd_conf *mdev, struct p_header *h)
+static int drbd_recv_header(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	int r;
 
@@ -1266,7 +1266,7 @@ int w_e_reissue(struct drbd_conf *mdev, struct drbd_work *w, int cancel) __relea
 	return 1;
 }
 
-static int receive_Barrier(struct drbd_conf *mdev, struct p_header *h)
+static int receive_Barrier(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	int rv, issue_flush;
 	struct p_barrier *p = (struct p_barrier *)h;
@@ -1570,7 +1570,7 @@ fail:
 	return FALSE;
 }
 
-static int receive_DataReply(struct drbd_conf *mdev, struct p_header *h)
+static int receive_DataReply(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct drbd_request *req;
 	sector_t sector;
@@ -1610,7 +1610,7 @@ static int receive_DataReply(struct drbd_conf *mdev, struct p_header *h)
 	return ok;
 }
 
-static int receive_RSDataReply(struct drbd_conf *mdev, struct p_header *h)
+static int receive_RSDataReply(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	sector_t sector;
 	unsigned int header_size, data_size;
@@ -1767,7 +1767,7 @@ static int drbd_wait_peer_seq(struct drbd_conf *mdev, const u32 packet_seq)
 }
 
 /* mirrored write */
-static int receive_Data(struct drbd_conf *mdev, struct p_header *h)
+static int receive_Data(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	sector_t sector;
 	struct drbd_epoch_entry *e;
@@ -2066,7 +2066,7 @@ int drbd_rs_should_slow_down(struct drbd_conf *mdev)
 }
 
 
-static int receive_DataRequest(struct drbd_conf *mdev, struct p_header *h)
+static int receive_DataRequest(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	sector_t sector;
 	const sector_t capacity = drbd_get_capacity(mdev->this_bdev);
@@ -2756,7 +2756,7 @@ static int cmp_after_sb(enum drbd_after_sb_p peer, enum drbd_after_sb_p self)
 	return 1;
 }
 
-static int receive_protocol(struct drbd_conf *mdev, struct p_header *h)
+static int receive_protocol(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_protocol *p = (struct p_protocol *)h;
 	int header_size, data_size;
@@ -2862,7 +2862,7 @@ struct crypto_hash *drbd_crypto_alloc_digest_safe(const struct drbd_conf *mdev,
 	return tfm;
 }
 
-static int receive_SyncParam(struct drbd_conf *mdev, struct p_header *h)
+static int receive_SyncParam(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	int ok = TRUE;
 	struct p_rs_param_95 *p = (struct p_rs_param_95 *)h;
@@ -3032,7 +3032,7 @@ static void warn_if_differ_considerably(struct drbd_conf *mdev,
 		     (unsigned long long)a, (unsigned long long)b);
 }
 
-static int receive_sizes(struct drbd_conf *mdev, struct p_header *h)
+static int receive_sizes(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_sizes *p = (struct p_sizes *)h;
 	enum determine_dev_size dd = unchanged;
@@ -3148,7 +3148,7 @@ static int receive_sizes(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-static int receive_uuids(struct drbd_conf *mdev, struct p_header *h)
+static int receive_uuids(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_uuids *p = (struct p_uuids *)h;
 	u64 *p_uuid;
@@ -3241,7 +3241,7 @@ static union drbd_state convert_state(union drbd_state ps)
 	return ms;
 }
 
-static int receive_req_state(struct drbd_conf *mdev, struct p_header *h)
+static int receive_req_state(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_req_state *p = (struct p_req_state *)h;
 	union drbd_state mask, val;
@@ -3271,7 +3271,7 @@ static int receive_req_state(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-static int receive_state(struct drbd_conf *mdev, struct p_header *h)
+static int receive_state(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_state *p = (struct p_state *)h;
 	enum drbd_conns nconn, oconn;
@@ -3395,7 +3395,7 @@ static int receive_state(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-static int receive_sync_uuid(struct drbd_conf *mdev, struct p_header *h)
+static int receive_sync_uuid(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_rs_uuid *p = (struct p_rs_uuid *)h;
 
@@ -3428,7 +3428,7 @@ static int receive_sync_uuid(struct drbd_conf *mdev, struct p_header *h)
 enum receive_bitmap_ret { OK, DONE, FAILED };
 
 static enum receive_bitmap_ret
-receive_bitmap_plain(struct drbd_conf *mdev, struct p_header *h,
+receive_bitmap_plain(struct drbd_conf *mdev, struct p_header80 *h,
 	unsigned long *buffer, struct bm_xfer_ctx *c)
 {
 	unsigned num_words = min_t(size_t, BM_PACKET_WORDS, c->bm_words - c->word_offset);
@@ -3533,7 +3533,7 @@ void INFO_bm_xfer_stats(struct drbd_conf *mdev,
 		const char *direction, struct bm_xfer_ctx *c)
 {
 	/* what would it take to transfer it "plaintext" */
-	unsigned plain = sizeof(struct p_header) *
+	unsigned plain = sizeof(struct p_header80) *
 		((c->bm_words+BM_PACKET_WORDS-1)/BM_PACKET_WORDS+1)
 		+ c->bm_words * sizeof(long);
 	unsigned total = c->bytes[0] + c->bytes[1];
@@ -3571,7 +3571,7 @@ void INFO_bm_xfer_stats(struct drbd_conf *mdev,
    in order to be agnostic to the 32 vs 64 bits issue.
 
    returns 0 on failure, 1 if we successfully received it. */
-static int receive_bitmap(struct drbd_conf *mdev, struct p_header *h)
+static int receive_bitmap(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct bm_xfer_ctx c;
 	void *buffer;
@@ -3623,7 +3623,7 @@ static int receive_bitmap(struct drbd_conf *mdev, struct p_header *h)
 		}
 
 		c.packets[h->command == P_BITMAP]++;
-		c.bytes[h->command == P_BITMAP] += sizeof(struct p_header) + h->length;
+		c.bytes[h->command == P_BITMAP] += sizeof(struct p_header80) + h->length;
 
 		if (ret != OK)
 			break;
@@ -3659,7 +3659,7 @@ static int receive_bitmap(struct drbd_conf *mdev, struct p_header *h)
 	return ok;
 }
 
-static int receive_skip_(struct drbd_conf *mdev, struct p_header *h, int silent)
+static int receive_skip_(struct drbd_conf *mdev, struct p_header80 *h, int silent)
 {
 	/* TODO zero copy sink :) */
 	static char sink[128];
@@ -3679,17 +3679,17 @@ static int receive_skip_(struct drbd_conf *mdev, struct p_header *h, int silent)
 	return size == 0;
 }
 
-static int receive_skip(struct drbd_conf *mdev, struct p_header *h)
+static int receive_skip(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	return receive_skip_(mdev, h, 0);
 }
 
-static int receive_skip_silent(struct drbd_conf *mdev, struct p_header *h)
+static int receive_skip_silent(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	return receive_skip_(mdev, h, 1);
 }
 
-static int receive_UnplugRemote(struct drbd_conf *mdev, struct p_header *h)
+static int receive_UnplugRemote(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	if (mdev->state.disk >= D_INCONSISTENT)
 		drbd_kick_lo(mdev);
@@ -3701,7 +3701,7 @@ static int receive_UnplugRemote(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-typedef int (*drbd_cmd_handler_f)(struct drbd_conf *, struct p_header *);
+typedef int (*drbd_cmd_handler_f)(struct drbd_conf *, struct p_header80 *);
 
 static drbd_cmd_handler_f drbd_default_handler[] = {
 	[P_DATA]	    = receive_Data,
@@ -3736,7 +3736,7 @@ static drbd_cmd_handler_f *drbd_opt_cmd_handler;
 static void drbdd(struct drbd_conf *mdev)
 {
 	drbd_cmd_handler_f handler;
-	struct p_header *header = &mdev->data.rbuf.header;
+	struct p_header80 *header = &mdev->data.rbuf.header;
 
 	while (get_t_state(&mdev->receiver) == Running) {
 		drbd_thread_current_set_cpu(mdev);
@@ -3964,7 +3964,7 @@ static int drbd_send_handshake(struct drbd_conf *mdev)
 	p->protocol_min = cpu_to_be32(PRO_VERSION_MIN);
 	p->protocol_max = cpu_to_be32(PRO_VERSION_MAX);
 	ok = _drbd_send_cmd( mdev, mdev->data.socket, P_HAND_SHAKE,
-			     (struct p_header *)p, sizeof(*p), 0 );
+			     (struct p_header80 *)p, sizeof(*p), 0 );
 	mutex_unlock(&mdev->data.mutex);
 	return ok;
 }
@@ -3981,7 +3981,7 @@ static int drbd_do_handshake(struct drbd_conf *mdev)
 	/* ASSERT current == mdev->receiver ... */
 	struct p_handshake *p = &mdev->data.rbuf.handshake;
 	const int expect = sizeof(struct p_handshake)
-			  -sizeof(struct p_header);
+			  -sizeof(struct p_header80);
 	int rv;
 
 	rv = drbd_send_handshake(mdev);
@@ -4058,7 +4058,7 @@ static int drbd_do_auth(struct drbd_conf *mdev)
 	char *response = NULL;
 	char *right_response = NULL;
 	char *peers_ch = NULL;
-	struct p_header p;
+	struct p_header80 p;
 	unsigned int key_len = strlen(mdev->net_conf->shared_secret);
 	unsigned int resp_size;
 	struct hash_desc desc;
@@ -4231,7 +4231,7 @@ int drbdd_init(struct drbd_thread *thi)
 
 /* ********* acknowledge sender ******** */
 
-static int got_RqSReply(struct drbd_conf *mdev, struct p_header *h)
+static int got_RqSReply(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_req_state_reply *p = (struct p_req_state_reply *)h;
 
@@ -4249,13 +4249,13 @@ static int got_RqSReply(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-static int got_Ping(struct drbd_conf *mdev, struct p_header *h)
+static int got_Ping(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	return drbd_send_ping_ack(mdev);
 
 }
 
-static int got_PingAck(struct drbd_conf *mdev, struct p_header *h)
+static int got_PingAck(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	/* restore idle timeout */
 	mdev->meta.socket->sk->sk_rcvtimeo = mdev->net_conf->ping_int*HZ;
@@ -4265,7 +4265,7 @@ static int got_PingAck(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-static int got_IsInSync(struct drbd_conf *mdev, struct p_header *h)
+static int got_IsInSync(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_block_ack *p = (struct p_block_ack *)h;
 	sector_t sector = be64_to_cpu(p->sector);
@@ -4336,7 +4336,7 @@ static int validate_req_change_req_state(struct drbd_conf *mdev,
 	return TRUE;
 }
 
-static int got_BlockAck(struct drbd_conf *mdev, struct p_header *h)
+static int got_BlockAck(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_block_ack *p = (struct p_block_ack *)h;
 	sector_t sector = be64_to_cpu(p->sector);
@@ -4376,7 +4376,7 @@ static int got_BlockAck(struct drbd_conf *mdev, struct p_header *h)
 		_ack_id_to_req, __func__ , what);
 }
 
-static int got_NegAck(struct drbd_conf *mdev, struct p_header *h)
+static int got_NegAck(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_block_ack *p = (struct p_block_ack *)h;
 	sector_t sector = be64_to_cpu(p->sector);
@@ -4396,7 +4396,7 @@ static int got_NegAck(struct drbd_conf *mdev, struct p_header *h)
 		_ack_id_to_req, __func__ , neg_acked);
 }
 
-static int got_NegDReply(struct drbd_conf *mdev, struct p_header *h)
+static int got_NegDReply(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_block_ack *p = (struct p_block_ack *)h;
 	sector_t sector = be64_to_cpu(p->sector);
@@ -4409,7 +4409,7 @@ static int got_NegDReply(struct drbd_conf *mdev, struct p_header *h)
 		_ar_id_to_req, __func__ , neg_acked);
 }
 
-static int got_NegRSDReply(struct drbd_conf *mdev, struct p_header *h)
+static int got_NegRSDReply(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	sector_t sector;
 	int size;
@@ -4431,7 +4431,7 @@ static int got_NegRSDReply(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-static int got_BarrierAck(struct drbd_conf *mdev, struct p_header *h)
+static int got_BarrierAck(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_barrier_ack *p = (struct p_barrier_ack *)h;
 
@@ -4440,7 +4440,7 @@ static int got_BarrierAck(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-static int got_OVResult(struct drbd_conf *mdev, struct p_header *h)
+static int got_OVResult(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	struct p_block_ack *p = (struct p_block_ack *)h;
 	struct drbd_work *w;
@@ -4474,7 +4474,7 @@ static int got_OVResult(struct drbd_conf *mdev, struct p_header *h)
 	return TRUE;
 }
 
-static int got_something_to_ignore_m(struct drbd_conf *mdev, struct p_header *h)
+static int got_something_to_ignore_m(struct drbd_conf *mdev, struct p_header80 *h)
 {
 	/* IGNORE */
 	return TRUE;
@@ -4482,7 +4482,7 @@ static int got_something_to_ignore_m(struct drbd_conf *mdev, struct p_header *h)
 
 struct asender_cmd {
 	size_t pkt_size;
-	int (*process)(struct drbd_conf *mdev, struct p_header *h);
+	int (*process)(struct drbd_conf *mdev, struct p_header80 *h);
 };
 
 static struct asender_cmd *get_asender_cmd(int cmd)
@@ -4491,8 +4491,8 @@ static struct asender_cmd *get_asender_cmd(int cmd)
 		/* anything missing from this table is in
 		 * the drbd_cmd_handler (drbd_default_handler) table,
 		 * see the beginning of drbdd() */
-	[P_PING]	    = { sizeof(struct p_header), got_Ping },
-	[P_PING_ACK]	    = { sizeof(struct p_header), got_PingAck },
+	[P_PING]	    = { sizeof(struct p_header80), got_Ping },
+	[P_PING_ACK]	    = { sizeof(struct p_header80), got_PingAck },
 	[P_RECV_ACK]	    = { sizeof(struct p_block_ack), got_BlockAck },
 	[P_WRITE_ACK]	    = { sizeof(struct p_block_ack), got_BlockAck },
 	[P_RS_WRITE_ACK]    = { sizeof(struct p_block_ack), got_BlockAck },
@@ -4504,7 +4504,7 @@ static struct asender_cmd *get_asender_cmd(int cmd)
 	[P_BARRIER_ACK]	    = { sizeof(struct p_barrier_ack), got_BarrierAck },
 	[P_STATE_CHG_REPLY] = { sizeof(struct p_req_state_reply), got_RqSReply },
 	[P_RS_IS_IN_SYNC]   = { sizeof(struct p_block_ack), got_IsInSync },
-	[P_DELAY_PROBE]     = { sizeof(struct p_delay_probe), got_something_to_ignore_m },
+	[P_DELAY_PROBE]     = { sizeof(struct p_delay_probe93), got_something_to_ignore_m },
 	[P_MAX_CMD]	    = { 0, NULL },
 	};
 	if (cmd > P_MAX_CMD || asender_tbl[cmd].process == NULL)
@@ -4515,13 +4515,13 @@ static struct asender_cmd *get_asender_cmd(int cmd)
 int drbd_asender(struct drbd_thread *thi)
 {
 	struct drbd_conf *mdev = thi->mdev;
-	struct p_header *h = &mdev->meta.rbuf.header;
+	struct p_header80 *h = &mdev->meta.rbuf.header;
 	struct asender_cmd *cmd = NULL;
 
 	int rv, len;
 	void *buf    = h;
 	int received = 0;
-	int expect   = sizeof(struct p_header);
+	int expect   = sizeof(struct p_header80);
 	int empty;
 
 	sprintf(current->comm, "drbd%d_asender", mdev_to_minor(mdev));
@@ -4621,7 +4621,7 @@ int drbd_asender(struct drbd_thread *thi)
 				goto disconnect;
 			}
 			expect = cmd->pkt_size;
-			ERR_IF(len != expect-sizeof(struct p_header))
+			ERR_IF(len != expect-sizeof(struct p_header80))
 				goto reconnect;
 		}
 		if (received == expect) {
@@ -4631,7 +4631,7 @@ int drbd_asender(struct drbd_thread *thi)
 
 			buf	 = h;
 			received = 0;
-			expect	 = sizeof(struct p_header);
+			expect	 = sizeof(struct p_header80);
 			cmd	 = NULL;
 		}
 	}
diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index 1eeb55423b3e..3d0e14e3ade3 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -1204,7 +1204,7 @@ int w_send_barrier(struct drbd_conf *mdev, struct drbd_work *w, int cancel)
 	 * dec_ap_pending will be done in got_BarrierAck
 	 * or (on connection loss) in w_clear_epoch.  */
 	ok = _drbd_send_cmd(mdev, mdev->data.socket, P_BARRIER,
-				(struct p_header *)p, sizeof(*p), 0);
+				(struct p_header80 *)p, sizeof(*p), 0);
 	drbd_put_data_sock(mdev);
 
 	return ok;
diff --git a/include/linux/drbd.h b/include/linux/drbd.h
index 0b2bfb58d9c5..89718a39791e 100644
--- a/include/linux/drbd.h
+++ b/include/linux/drbd.h
@@ -318,6 +318,8 @@ enum drbd_timeout_flag {
 
 #define DRBD_MAGIC 0x83740267
 #define BE_DRBD_MAGIC __constant_cpu_to_be32(DRBD_MAGIC)
+#define DRBD_MAGIC_BIG 0x835a
+#define BE_DRBD_MAGIC_BIG __constant_cpu_to_be16(DRBD_MAGIC_BIG)
 
 /* these are of type "int" */
 #define DRBD_MD_INDEX_INTERNAL -1
-- 
cgit v1.2.3


From fb22c402ffdf61dd121795b5809de587185d5240 Mon Sep 17 00:00:00 2001
From: Philipp Reisner <philipp.reisner@linbit.com>
Date: Wed, 8 Sep 2010 23:20:21 +0200
Subject: drbd: Track the reasons to suspend IO in dedicated state bits

There are three ways to get IO suspended:

 * Loss of any access to data
 * Fence-peer-handler running
 * User requested to suspend IO

Track those in different bits, so that one condition clearing its
state bit does not interfere with the other two conditions.

Only when the user resumes IO he overrules all three bits.

The fact is hidden from the user, he sees only a single suspend
bit.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_int.h      |  9 ++++++++-
 drivers/block/drbd/drbd_main.c     | 36 +++++++++++++++++++++++-------------
 drivers/block/drbd/drbd_nl.c       | 20 ++++++++++++++------
 drivers/block/drbd/drbd_proc.c     |  2 +-
 drivers/block/drbd/drbd_receiver.c |  6 +++---
 drivers/block/drbd/drbd_req.c      |  6 +++---
 include/linux/drbd.h               | 10 +++++++---
 7 files changed, 59 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index eb1273d04caf..ff7fffa00dac 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -1681,6 +1681,8 @@ void drbd_bcast_ee(struct drbd_conf *mdev,
 #define susp_MASK 1
 #define user_isp_MASK 1
 #define aftr_isp_MASK 1
+#define susp_nod_MASK 1
+#define susp_fen_MASK 1
 
 #define NS(T, S) \
 	({ union drbd_state mask; mask.i = 0; mask.T = T##_MASK; mask; }), \
@@ -2254,11 +2256,16 @@ static inline int drbd_state_is_stable(union drbd_state s)
 	return 1;
 }
 
+static inline int is_susp(union drbd_state s)
+{
+	return s.susp || s.susp_nod || s.susp_fen;
+}
+
 static inline int __inc_ap_bio_cond(struct drbd_conf *mdev)
 {
 	int mxb = drbd_get_max_buffers(mdev);
 
-	if (mdev->state.susp)
+	if (is_susp(mdev->state))
 		return 0;
 	if (test_bit(SUSPEND_IO, &mdev->flags))
 		return 0;
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 04c305d36f8e..4f33714fb3cd 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -654,7 +654,7 @@ static void print_st(struct drbd_conf *mdev, char *name, union drbd_state ns)
 	    drbd_role_str(ns.peer),
 	    drbd_disk_str(ns.disk),
 	    drbd_disk_str(ns.pdsk),
-	    ns.susp ? 's' : 'r',
+	    is_susp(ns) ? 's' : 'r',
 	    ns.aftr_isp ? 'a' : '-',
 	    ns.peer_isp ? 'p' : '-',
 	    ns.user_isp ? 'u' : '-'
@@ -925,12 +925,12 @@ static union drbd_state sanitize_state(struct drbd_conf *mdev, union drbd_state
 	if (fp == FP_STONITH &&
 	    (ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.pdsk > D_OUTDATED) &&
 	    !(os.role == R_PRIMARY && os.conn < C_CONNECTED && os.pdsk > D_OUTDATED))
-		ns.susp = 1; /* Suspend IO while fence-peer handler runs (peer lost) */
+		ns.susp_fen = 1; /* Suspend IO while fence-peer handler runs (peer lost) */
 
 	if (mdev->sync_conf.on_no_data == OND_SUSPEND_IO &&
 	    (ns.role == R_PRIMARY && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE) &&
 	    !(os.role == R_PRIMARY && os.disk < D_UP_TO_DATE && os.pdsk < D_UP_TO_DATE))
-		ns.susp = 1; /* Suspend IO while no data available (no accessible data available) */
+		ns.susp_nod = 1; /* Suspend IO while no data available (no accessible data available) */
 
 	if (ns.aftr_isp || ns.peer_isp || ns.user_isp) {
 		if (ns.conn == C_SYNC_SOURCE)
@@ -1030,7 +1030,10 @@ int __drbd_set_state(struct drbd_conf *mdev,
 		PSC(conn);
 		PSC(disk);
 		PSC(pdsk);
-		PSC(susp);
+		if (is_susp(ns) != is_susp(os))
+			pbp += sprintf(pbp, "susp( %s -> %s ) ",
+				       drbd_susp_str(is_susp(os)),
+				       drbd_susp_str(is_susp(ns)));
 		PSC(aftr_isp);
 		PSC(peer_isp);
 		PSC(user_isp);
@@ -1218,6 +1221,7 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 {
 	enum drbd_fencing_p fp;
 	enum drbd_req_event what = nothing;
+	union drbd_state nsm = (union drbd_state){ .i = -1 };
 
 	if (os.conn != C_CONNECTED && ns.conn == C_CONNECTED) {
 		clear_bit(CRASHED_PRIMARY, &mdev->flags);
@@ -1241,19 +1245,21 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 	/* Here we have the actions that are performed after a
 	   state change. This function might sleep */
 
-	if (os.susp && ns.susp && mdev->sync_conf.on_no_data == OND_SUSPEND_IO) {
+	nsm.i = -1;
+	if (ns.susp_nod) {
 		if (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED) {
 			if (ns.conn == C_CONNECTED)
-				what = resend;
+				what = resend, nsm.susp_nod = 0;
 			else /* ns.conn > C_CONNECTED */
 				dev_err(DEV, "Unexpected Resynd going on!\n");
 		}
 
 		if (os.disk == D_ATTACHING && ns.disk > D_ATTACHING)
-			what = restart_frozen_disk_io;
+			what = restart_frozen_disk_io, nsm.susp_nod = 0;
+
 	}
 
-	if (fp == FP_STONITH && ns.susp) {
+	if (ns.susp_fen) {
 		/* case1: The outdate peer handler is successful: */
 		if (os.pdsk > D_OUTDATED  && ns.pdsk <= D_OUTDATED) {
 			tl_clear(mdev);
@@ -1263,20 +1269,22 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 				drbd_md_sync(mdev);
 			}
 			spin_lock_irq(&mdev->req_lock);
-			_drbd_set_state(_NS(mdev, susp, 0), CS_VERBOSE, NULL);
+			_drbd_set_state(_NS(mdev, susp_fen, 0), CS_VERBOSE, NULL);
 			spin_unlock_irq(&mdev->req_lock);
 		}
 		/* case2: The connection was established again: */
 		if (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED) {
 			clear_bit(NEW_CUR_UUID, &mdev->flags);
 			what = resend;
+			nsm.susp_fen = 0;
 		}
 	}
 
 	if (what != nothing) {
 		spin_lock_irq(&mdev->req_lock);
 		_tl_restart(mdev, what);
-		_drbd_set_state(_NS(mdev, susp, 0), CS_VERBOSE, NULL);
+		nsm.i &= mdev->state.i;
+		_drbd_set_state(mdev, nsm, CS_VERBOSE, NULL);
 		spin_unlock_irq(&mdev->req_lock);
 	}
 
@@ -1298,7 +1306,7 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 		if (get_ldev(mdev)) {
 			if ((ns.role == R_PRIMARY || ns.peer == R_PRIMARY) &&
 			    mdev->ldev->md.uuid[UI_BITMAP] == 0 && ns.disk >= D_UP_TO_DATE) {
-				if (mdev->state.susp) {
+				if (is_susp(mdev->state)) {
 					set_bit(NEW_CUR_UUID, &mdev->flags);
 				} else {
 					drbd_uuid_new_current(mdev);
@@ -1417,7 +1425,7 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 		resume_next_sg(mdev);
 
 	/* free tl_hash if we Got thawed and are C_STANDALONE */
-	if (ns.conn == C_STANDALONE && ns.susp == 0 && mdev->tl_hash)
+	if (ns.conn == C_STANDALONE && !is_susp(ns) && mdev->tl_hash)
 		drbd_free_tl_hash(mdev);
 
 	/* Upon network connection, we need to start the receiver */
@@ -2732,7 +2740,9 @@ static void drbd_set_defaults(struct drbd_conf *mdev)
 		  .conn = C_STANDALONE,
 		  .disk = D_DISKLESS,
 		  .pdsk = D_UNKNOWN,
-		  .susp = 0
+		  .susp = 0,
+		  .susp_nod = 0,
+		  .susp_fen = 0
 		} };
 }
 
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 5b30f90cab3e..9ee44568dce3 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -209,7 +209,8 @@ enum drbd_disk_state drbd_try_outdate_peer(struct drbd_conf *mdev)
 		put_ldev(mdev);
 	} else {
 		dev_warn(DEV, "Not fencing peer, I'm not even Consistent myself.\n");
-		return mdev->state.pdsk;
+		nps = mdev->state.pdsk;
+		goto out;
 	}
 
 	r = drbd_khelper(mdev, "fence-peer");
@@ -256,6 +257,14 @@ enum drbd_disk_state drbd_try_outdate_peer(struct drbd_conf *mdev)
 
 	dev_info(DEV, "fence-peer helper returned %d (%s)\n",
 			(r>>8) & 0xff, ex_to_string);
+
+out:
+	if (mdev->state.susp_fen && nps >= D_UNKNOWN) {
+		/* The handler was not successful... unfreeze here, the
+		   state engine can not unfreeze... */
+		_drbd_request_state(mdev, NS(susp_fen, 0), CS_VERBOSE);
+	}
+
 	return nps;
 }
 
@@ -550,7 +559,7 @@ char *ppsize(char *buf, unsigned long long size)
 void drbd_suspend_io(struct drbd_conf *mdev)
 {
 	set_bit(SUSPEND_IO, &mdev->flags);
-	if (mdev->state.susp)
+	if (is_susp(mdev->state))
 		return;
 	wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_bio_cnt));
 }
@@ -1016,7 +1025,7 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp
 
 	drbd_suspend_io(mdev);
 	/* also wait for the last barrier ack. */
-	wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt) || mdev->state.susp);
+	wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt) || is_susp(mdev->state));
 	/* and for any other previously queued work */
 	drbd_flush_workqueue(mdev);
 
@@ -1114,8 +1123,7 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp
 		clear_bit(CRASHED_PRIMARY, &mdev->flags);
 
 	if (drbd_md_test_flag(mdev->ldev, MDF_PRIMARY_IND) &&
-	    !(mdev->state.role == R_PRIMARY && mdev->state.susp &&
-	      mdev->sync_conf.on_no_data == OND_SUSPEND_IO)) {
+	    !(mdev->state.role == R_PRIMARY && mdev->state.susp_nod)) {
 		set_bit(CRASHED_PRIMARY, &mdev->flags);
 		cp_discovered = 1;
 	}
@@ -1939,7 +1947,7 @@ static int drbd_nl_resume_io(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp
 		drbd_md_sync(mdev);
 	}
 	drbd_suspend_io(mdev);
-	reply->ret_code = drbd_request_state(mdev, NS(susp, 0));
+	reply->ret_code = drbd_request_state(mdev, NS3(susp, 0, susp_nod, 0, susp_fen, 0));
 	if (reply->ret_code == SS_SUCCESS) {
 		if (mdev->state.conn < C_CONNECTED)
 			tl_clear(mdev);
diff --git a/drivers/block/drbd/drbd_proc.c b/drivers/block/drbd/drbd_proc.c
index a4a4a06908c5..aec8426c1bf6 100644
--- a/drivers/block/drbd/drbd_proc.c
+++ b/drivers/block/drbd/drbd_proc.c
@@ -213,7 +213,7 @@ static int drbd_seq_show(struct seq_file *seq, void *v)
 			   drbd_disk_str(mdev->state.pdsk),
 			   (mdev->net_conf == NULL ? ' ' :
 			    (mdev->net_conf->wire_protocol - DRBD_PROT_A+'A')),
-			   mdev->state.susp ? 's' : 'r',
+			   is_susp(mdev->state) ? 's' : 'r',
 			   mdev->state.aftr_isp ? 'a' : '-',
 			   mdev->state.peer_isp ? 'p' : '-',
 			   mdev->state.user_isp ? 'u' : '-',
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 983e49cbd233..6b69b2f734dc 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -3315,7 +3315,7 @@ static int receive_state(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned
 	if ((nconn == C_CONNECTED || nconn == C_WF_BITMAP_S) && ns.disk == D_NEGOTIATING)
 		ns.disk = mdev->new_state_tmp.disk;
 	cs_flags = CS_VERBOSE + (oconn < C_CONNECTED && nconn >= C_CONNECTED ? 0 : CS_HARD);
-	if (ns.pdsk == D_CONSISTENT && ns.susp && nconn == C_CONNECTED && oconn < C_CONNECTED &&
+	if (ns.pdsk == D_CONSISTENT && is_susp(ns) && nconn == C_CONNECTED && oconn < C_CONNECTED &&
 	    test_bit(NEW_CUR_UUID, &mdev->flags)) {
 		/* Do not allow tl_restart(resend) for a rebooted peer. We can only allow this
 		   for temporal network outages! */
@@ -3829,7 +3829,7 @@ static void drbd_disconnect(struct drbd_conf *mdev)
 	kfree(mdev->p_uuid);
 	mdev->p_uuid = NULL;
 
-	if (!mdev->state.susp)
+	if (!is_susp(mdev->state))
 		tl_clear(mdev);
 
 	dev_info(DEV, "Connection closed\n");
@@ -3858,7 +3858,7 @@ static void drbd_disconnect(struct drbd_conf *mdev)
 	if (os.conn == C_DISCONNECTING) {
 		wait_event(mdev->net_cnt_wait, atomic_read(&mdev->net_cnt) == 0);
 
-		if (!mdev->state.susp) {
+		if (!is_susp(mdev->state)) {
 			/* we must not free the tl_hash
 			 * while application io is still on the fly */
 			wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_bio_cnt));
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index af608b39c4e0..9e91a2545fc8 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -287,7 +287,7 @@ static void _req_may_be_done_not_susp(struct drbd_request *req, struct bio_and_e
 {
 	struct drbd_conf *mdev = req->mdev;
 
-	if (!mdev->state.susp)
+	if (!is_susp(mdev->state))
 		_req_may_be_done(req, m);
 }
 
@@ -812,7 +812,7 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio)
 			    (mdev->state.pdsk == D_INCONSISTENT &&
 			     mdev->state.conn >= C_CONNECTED));
 
-	if (!(local || remote) && !mdev->state.susp) {
+	if (!(local || remote) && !is_susp(mdev->state)) {
 		dev_err(DEV, "IO ERROR: neither local nor remote disk\n");
 		goto fail_free_complete;
 	}
@@ -838,7 +838,7 @@ allocate_barrier:
 	/* GOOD, everything prepared, grab the spin_lock */
 	spin_lock_irq(&mdev->req_lock);
 
-	if (mdev->state.susp) {
+	if (is_susp(mdev->state)) {
 		/* If we got suspended, use the retry mechanism of
 		   generic_make_request() to restart processing of this
 		   bio. In the next call to drbd_make_request_26
diff --git a/include/linux/drbd.h b/include/linux/drbd.h
index 89718a39791e..5e72a5d3d48f 100644
--- a/include/linux/drbd.h
+++ b/include/linux/drbd.h
@@ -232,13 +232,17 @@ union drbd_state {
 		unsigned conn:5 ;   /* 17/32	 cstates */
 		unsigned disk:4 ;   /* 8/16	 from D_DISKLESS to D_UP_TO_DATE */
 		unsigned pdsk:4 ;   /* 8/16	 from D_DISKLESS to D_UP_TO_DATE */
-		unsigned susp:1 ;   /* 2/2	 IO suspended  no/yes */
+		unsigned susp:1 ;   /* 2/2	 IO suspended no/yes (by user) */
 		unsigned aftr_isp:1 ; /* isp .. imposed sync pause */
 		unsigned peer_isp:1 ;
 		unsigned user_isp:1 ;
-		unsigned _pad:11;   /* 0	 unused */
+		unsigned susp_nod:1 ; /* IO suspended because no data */
+		unsigned susp_fen:1 ; /* IO suspended because fence peer handler runs*/
+		unsigned _pad:9;   /* 0	 unused */
 #elif defined(__BIG_ENDIAN_BITFIELD)
-		unsigned _pad:11;   /* 0	 unused */
+		unsigned _pad:9;
+		unsigned susp_fen:1 ;
+		unsigned susp_nod:1 ;
 		unsigned user_isp:1 ;
 		unsigned peer_isp:1 ;
 		unsigned aftr_isp:1 ; /* isp .. imposed sync pause */
-- 
cgit v1.2.3


From 00b425377d60e67e86721d4ce6d7cbf131a5d0fd Mon Sep 17 00:00:00 2001
From: Philipp Reisner <philipp.reisner@linbit.com>
Date: Tue, 5 Oct 2010 11:19:39 +0200
Subject: drbd: Allow larger values for c-fill-target.

Connections through a compressing proxy might have more bits
on the fly. 500MByte instead of 50MByte

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_int.h | 2 +-
 include/linux/drbd_limits.h   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index ff7fffa00dac..1680939de101 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -348,7 +348,7 @@ struct p_header80 {
 struct p_header95 {
 	u16	  magic;	/* use DRBD_MAGIC_BIG here */
 	u16	  command;
-	u32	  length;
+	u32	  length;	/* Use only 24 bits of that. Ignore the highest 8 bit. */
 	u8	  payload[0];
 } __packed;
 
diff --git a/include/linux/drbd_limits.h b/include/linux/drbd_limits.h
index 0b24ded6fffd..4ac33f34b77e 100644
--- a/include/linux/drbd_limits.h
+++ b/include/linux/drbd_limits.h
@@ -143,7 +143,7 @@
 #define DRBD_C_DELAY_TARGET_DEF 10
 
 #define DRBD_C_FILL_TARGET_MIN 0
-#define DRBD_C_FILL_TARGET_MAX 100000
+#define DRBD_C_FILL_TARGET_MAX (1<<20) /* 500MByte in sec */
 #define DRBD_C_FILL_TARGET_DEF 0 /* By default disabled -> controlled by delay_target */
 
 #define DRBD_C_MAX_RATE_MIN     250 /* kByte/sec */
-- 
cgit v1.2.3


From 22cc37a943832c948808884604ec6f5ff2594c1d Mon Sep 17 00:00:00 2001
From: Lars Ellenberg <lars.ellenberg@linbit.com>
Date: Tue, 14 Sep 2010 20:40:41 +0200
Subject: drbd: fix unlikely access after free and list corruption

Various cleanup paths have been incomplete, for the very unlikely case
that we cannot allocate enough bios from process context when submitting
on behalf of the peer or resync process.

Never observed.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_receiver.c | 25 +++++++++++++++++++++++++
 drivers/block/drbd/drbd_worker.c   |  7 +++++++
 include/linux/drbd.h               |  4 ++--
 3 files changed, 34 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 990fe01afa50..71775a9de21d 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -1573,6 +1573,13 @@ static int recv_resync_read(struct drbd_conf *mdev, sector_t sector, int data_si
 	if (drbd_submit_ee(mdev, e, WRITE, DRBD_FAULT_RS_WR) == 0)
 		return TRUE;
 
+	/* drbd_submit_ee currently fails for one reason only:
+	 * not being able to allocate enough bios.
+	 * Is dropping the connection going to help? */
+	spin_lock_irq(&mdev->req_lock);
+	list_del(&e->w.list);
+	spin_unlock_irq(&mdev->req_lock);
+
 	drbd_free_ee(mdev, e);
 fail:
 	put_ldev(mdev);
@@ -1998,6 +2005,16 @@ static int receive_Data(struct drbd_conf *mdev, enum drbd_packets cmd, unsigned
 	if (drbd_submit_ee(mdev, e, rw, DRBD_FAULT_DT_WR) == 0)
 		return TRUE;
 
+	/* drbd_submit_ee currently fails for one reason only:
+	 * not being able to allocate enough bios.
+	 * Is dropping the connection going to help? */
+	spin_lock_irq(&mdev->req_lock);
+	list_del(&e->w.list);
+	hlist_del_init(&e->colision);
+	spin_unlock_irq(&mdev->req_lock);
+	if (e->flags & EE_CALL_AL_COMPLETE_IO)
+		drbd_al_complete_io(mdev, e->sector);
+
 out_interrupted:
 	/* yes, the epoch_size now is imbalanced.
 	 * but we drop the connection anyways, so we don't have a chance to
@@ -2202,6 +2219,14 @@ submit:
 	if (drbd_submit_ee(mdev, e, READ, fault_type) == 0)
 		return TRUE;
 
+	/* drbd_submit_ee currently fails for one reason only:
+	 * not being able to allocate enough bios.
+	 * Is dropping the connection going to help? */
+	spin_lock_irq(&mdev->req_lock);
+	list_del(&e->w.list);
+	spin_unlock_irq(&mdev->req_lock);
+	/* no drbd_rs_complete_io(), we are dropping the connection anyways */
+
 out_free_e:
 	put_ldev(mdev);
 	drbd_free_ee(mdev, e);
diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index 88be45ad84ed..f12822d53867 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -387,6 +387,13 @@ static int read_for_csum(struct drbd_conf *mdev, sector_t sector, int size)
 	if (drbd_submit_ee(mdev, e, READ, DRBD_FAULT_RS_RD) == 0)
 		return 0;
 
+	/* drbd_submit_ee currently fails for one reason only:
+	 * not being able to allocate enough bios.
+	 * Is dropping the connection going to help? */
+	spin_lock_irq(&mdev->req_lock);
+	list_del(&e->w.list);
+	spin_unlock_irq(&mdev->req_lock);
+
 	drbd_free_ee(mdev, e);
 defer:
 	put_ldev(mdev);
diff --git a/include/linux/drbd.h b/include/linux/drbd.h
index 5e72a5d3d48f..da7d9bd4f3f0 100644
--- a/include/linux/drbd.h
+++ b/include/linux/drbd.h
@@ -53,10 +53,10 @@
 
 
 extern const char *drbd_buildtag(void);
-#define REL_VERSION "8.3.8.1"
+#define REL_VERSION "8.3.9rc1"
 #define API_VERSION 88
 #define PRO_VERSION_MIN 86
-#define PRO_VERSION_MAX 94
+#define PRO_VERSION_MAX 95
 
 
 enum drbd_io_error_p {
-- 
cgit v1.2.3


From 0eead9ab41da33644ae2c97c57ad03da636a0422 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu, 14 Oct 2010 10:57:40 -0700
Subject: Don't dump task struct in a.out core-dumps

akiphie points out that a.out core-dumps have that odd task struct
dumping that was never used and was never really a good idea (it goes
back into the mists of history, probably the original core-dumping
code).  Just remove it.

Also do the access_ok() check on dump_write().  It probably doesn't
matter (since normal filesystems all seem to do it anyway), but he
points out that it's normally done by the VFS layer, so ...

[ I suspect that we should possibly do "vfs_write()" instead of
  calling ->write directly.  That also does the whole fsnotify and write
  statistics thing, which may or may not be a good idea. ]

And just to be anal, do this all for the x86-64 32-bit a.out emulation
code too, even though it's not enabled (and won't currently even
compile)

Reported-by: akiphie <akiphie@lavabit.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/ia32/ia32_aout.c | 22 +++++-----------------
 fs/binfmt_aout.c          |  4 ----
 include/linux/coredump.h  |  2 +-
 3 files changed, 6 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c
index 0350311906ae..2d93bdbc9ac0 100644
--- a/arch/x86/ia32/ia32_aout.c
+++ b/arch/x86/ia32/ia32_aout.c
@@ -34,7 +34,7 @@
 #include <asm/ia32.h>
 
 #undef WARN_OLD
-#undef CORE_DUMP /* probably broken */
+#undef CORE_DUMP /* definitely broken */
 
 static int load_aout_binary(struct linux_binprm *, struct pt_regs *regs);
 static int load_aout_library(struct file *);
@@ -131,21 +131,15 @@ static void set_brk(unsigned long start, unsigned long end)
  * macros to write out all the necessary info.
  */
 
-static int dump_write(struct file *file, const void *addr, int nr)
-{
-	return file->f_op->write(file, addr, nr, &file->f_pos) == nr;
-}
+#include <linux/coredump.h>
 
 #define DUMP_WRITE(addr, nr)			     \
 	if (!dump_write(file, (void *)(addr), (nr))) \
 		goto end_coredump;
 
-#define DUMP_SEEK(offset)						\
-	if (file->f_op->llseek) {					\
-		if (file->f_op->llseek(file, (offset), 0) != (offset))	\
-			goto end_coredump;				\
-	} else								\
-		file->f_pos = (offset)
+#define DUMP_SEEK(offset)		\
+	if (!dump_seek(file, offset))	\
+		goto end_coredump;
 
 #define START_DATA()	(u.u_tsize << PAGE_SHIFT)
 #define START_STACK(u)	(u.start_stack)
@@ -217,12 +211,6 @@ static int aout_core_dump(long signr, struct pt_regs *regs, struct file *file,
 		dump_size = dump.u_ssize << PAGE_SHIFT;
 		DUMP_WRITE(dump_start, dump_size);
 	}
-	/*
-	 * Finally dump the task struct.  Not be used by gdb, but
-	 * could be useful
-	 */
-	set_fs(KERNEL_DS);
-	DUMP_WRITE(current, sizeof(*current));
 end_coredump:
 	set_fs(fs);
 	return has_dumped;
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index f96eff04e11a..a6395bdb26ae 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -134,10 +134,6 @@ static int aout_core_dump(struct coredump_params *cprm)
 		if (!dump_write(file, dump_start, dump_size))
 			goto end_coredump;
 	}
-/* Finally dump the task struct.  Not be used by gdb, but could be useful */
-	set_fs(KERNEL_DS);
-	if (!dump_write(file, current, sizeof(*current)))
-		goto end_coredump;
 end_coredump:
 	set_fs(fs);
 	return has_dumped;
diff --git a/include/linux/coredump.h b/include/linux/coredump.h
index 8ba66a9d9022..59579cfee6a0 100644
--- a/include/linux/coredump.h
+++ b/include/linux/coredump.h
@@ -11,7 +11,7 @@
  */
 static inline int dump_write(struct file *file, const void *addr, int nr)
 {
-	return file->f_op->write(file, addr, nr, &file->f_pos) == nr;
+	return access_ok(VERIFY_READ, addr, nr) && file->f_op->write(file, addr, nr, &file->f_pos) == nr;
 }
 
 static inline int dump_seek(struct file *file, loff_t off)
-- 
cgit v1.2.3


From 3aa0ce825ade0cf5506e32ccf51d01fc8d22a9cf Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu, 14 Oct 2010 14:32:06 -0700
Subject: Un-inline the core-dump helper functions

Tony Luck reports that the addition of the access_ok() check in commit
0eead9ab41da ("Don't dump task struct in a.out core-dumps") broke the
ia64 compile due to missing the necessary header file includes.

Rather than add yet another include (<asm/unistd.h>) to make everything
happy, just uninline the silly core dump helper functions and move the
bodies to fs/exec.c where they make a lot more sense.

dump_seek() in particular was too big to be an inline function anyway,
and none of them are in any way performance-critical.  And we really
don't need to mess up our include file headers more than they already
are.

Reported-and-tested-by: Tony Luck <tony.luck@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 38 ++++++++++++++++++++++++++++++++++++++
 include/linux/coredump.h | 34 ++--------------------------------
 2 files changed, 40 insertions(+), 32 deletions(-)

(limited to 'include/linux')

diff --git a/fs/exec.c b/fs/exec.c
index 828dd2461d6b..03278c984ba0 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -2014,3 +2014,41 @@ fail_creds:
 fail:
 	return;
 }
+
+/*
+ * Core dumping helper functions.  These are the only things you should
+ * do on a core-file: use only these functions to write out all the
+ * necessary info.
+ */
+int dump_write(struct file *file, const void *addr, int nr)
+{
+	return access_ok(VERIFY_READ, addr, nr) && file->f_op->write(file, addr, nr, &file->f_pos) == nr;
+}
+
+int dump_seek(struct file *file, loff_t off)
+{
+	int ret = 1;
+
+	if (file->f_op->llseek && file->f_op->llseek != no_llseek) {
+		if (file->f_op->llseek(file, off, SEEK_CUR) < 0)
+			return 0;
+	} else {
+		char *buf = (char *)get_zeroed_page(GFP_KERNEL);
+
+		if (!buf)
+			return 0;
+		while (off > 0) {
+			unsigned long n = off;
+
+			if (n > PAGE_SIZE)
+				n = PAGE_SIZE;
+			if (!dump_write(file, buf, n)) {
+				ret = 0;
+				break;
+			}
+			off -= n;
+		}
+		free_page((unsigned long)buf);
+	}
+	return ret;
+}
diff --git a/include/linux/coredump.h b/include/linux/coredump.h
index 59579cfee6a0..ba4b85a6d9b8 100644
--- a/include/linux/coredump.h
+++ b/include/linux/coredump.h
@@ -9,37 +9,7 @@
  * These are the only things you should do on a core-file: use only these
  * functions to write out all the necessary info.
  */
-static inline int dump_write(struct file *file, const void *addr, int nr)
-{
-	return access_ok(VERIFY_READ, addr, nr) && file->f_op->write(file, addr, nr, &file->f_pos) == nr;
-}
-
-static inline int dump_seek(struct file *file, loff_t off)
-{
-	int ret = 1;
-
-	if (file->f_op->llseek && file->f_op->llseek != no_llseek) {
-		if (file->f_op->llseek(file, off, SEEK_CUR) < 0)
-			return 0;
-	} else {
-		char *buf = (char *)get_zeroed_page(GFP_KERNEL);
-
-		if (!buf)
-			return 0;
-		while (off > 0) {
-			unsigned long n = off;
-
-			if (n > PAGE_SIZE)
-				n = PAGE_SIZE;
-			if (!dump_write(file, buf, n)) {
-				ret = 0;
-				break;
-			}
-			off -= n;
-		}
-		free_page((unsigned long)buf);
-	}
-	return ret;
-}
+extern int dump_write(struct file *file, const void *addr, int nr);
+extern int dump_seek(struct file *file, loff_t off);
 
 #endif /* _LINUX_COREDUMP_H */
-- 
cgit v1.2.3


From be70e2671b95a8982ff133ebaafff6399ad393d4 Mon Sep 17 00:00:00 2001
From: Philipp Reisner <philipp.reisner@linbit.com>
Date: Thu, 14 Oct 2010 11:58:20 +0200
Subject: dynamic_debug.h: Fix dynamic_dev_dbg() macro if CONFIG_DYNAMIC_DEBUG
 not set

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
---
 include/linux/dynamic_debug.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/dynamic_debug.h b/include/linux/dynamic_debug.h
index 52c0da4bdd18..81bc20e36ec0 100644
--- a/include/linux/dynamic_debug.h
+++ b/include/linux/dynamic_debug.h
@@ -80,7 +80,7 @@ static inline int ddebug_remove_module(const char *mod)
 
 #define dynamic_pr_debug(fmt, ...)					\
 	do { if (0) printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__); } while (0)
-#define dynamic_dev_dbg(dev, format, ...)				\
+#define dynamic_dev_dbg(dev, fmt, ...)					\
 	do { if (0) dev_printk(KERN_DEBUG, dev, fmt, ##__VA_ARGS__); } while (0)
 #endif
 
-- 
cgit v1.2.3


From b3b3a9b63f2deacfd59137e3781211d21a568ca9 Mon Sep 17 00:00:00 2001
From: Anand Gadiyar <gadiyar@ti.com>
Date: Thu, 14 Oct 2010 11:31:42 -0400
Subject: oprofile: fix linker errors

Commit e9677b3ce (oprofile, ARM: Use oprofile_arch_exit() to
cleanup on failure) caused oprofile_perf_exit to be called
in the cleanup path of oprofile_perf_init. The __exit tag
for oprofile_perf_exit should therefore be dropped.

The same has to be done for exit_driverfs as well, as this
function is called from oprofile_perf_exit. Else, we get
the following two linker errors.

  LD      .tmp_vmlinux1
`oprofile_perf_exit' referenced in section `.init.text' of arch/arm/oprofile/built-in.o: defined in discarded section `.exit.text' of arch/arm/oprofile/built-in.o
make: *** [.tmp_vmlinux1] Error 1

  LD      .tmp_vmlinux1
`exit_driverfs' referenced in section `.text' of arch/arm/oprofile/built-in.o: defined in discarded section `.exit.text' of arch/arm/oprofile/built-in.o
make: *** [.tmp_vmlinux1] Error 1

Signed-off-by: Anand Gadiyar <gadiyar@ti.com>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 drivers/oprofile/oprofile_perf.c | 2 +-
 include/linux/oprofile.h         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/oprofile/oprofile_perf.c b/drivers/oprofile/oprofile_perf.c
index e707a72a3799..36ec67e3d91d 100644
--- a/drivers/oprofile/oprofile_perf.c
+++ b/drivers/oprofile/oprofile_perf.c
@@ -236,7 +236,7 @@ static int __init init_driverfs(void)
 	return ret;
 }
 
-static void __exit exit_driverfs(void)
+static void exit_driverfs(void)
 {
 	platform_device_unregister(oprofile_pdev);
 	platform_driver_unregister(&oprofile_driver);
diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h
index d67a8330b41e..32fb81212fd1 100644
--- a/include/linux/oprofile.h
+++ b/include/linux/oprofile.h
@@ -188,7 +188,7 @@ int oprofile_write_commit(struct op_entry *entry);
 
 #ifdef CONFIG_PERF_EVENTS
 int __init oprofile_perf_init(struct oprofile_operations *ops);
-void __exit oprofile_perf_exit(void);
+void oprofile_perf_exit(void);
 char *op_name_from_perf_id(void);
 #endif /* CONFIG_PERF_EVENTS */
 
-- 
cgit v1.2.3


From 5dbfe7aedf54aa7f62fd659e34371d4ea0e7bffe Mon Sep 17 00:00:00 2001
From: Lars Ellenberg <lars.ellenberg@linbit.com>
Date: Fri, 15 Oct 2010 09:52:46 +0200
Subject: drbd: add race-breaker to drbd_go_diskless

This adds a necessary race breaker to these commits:
    drbd: fix for possible deadlock on IO error during resync
    drbd: drop wrong debug asserts, fix recently introduced race

What we do is get a refcount, check the state, then depending on the
state and the requested minimum disk state, either hold it (success),
or give it back immediately (failed "try lock").

Some code paths (flushing of drbd metadata) may still grab and hold a
refcount even if we are D_FAILED (application IO won't).
So even if we hit local_cnt == 0 once after being D_FAILED,
we still need to wait for that again after we changed to D_DISKLESS.
Once local_cnt reaches 0 while we are D_DISKLESS, we can be sure that
no one will look at the protected members anymore, so only then is it
safe to free them.

We cannot easily convert to standard locking primitives here, as we want
to be able to use it in atomic context (we always do a "try lock"),
as well as hold references for a "long time" (from IO submission to
completion callback).

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
---
 drivers/block/drbd/drbd_main.c | 3 +++
 include/linux/drbd.h           | 2 +-
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index bbe3bff2cad6..8bfedc7164fa 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -3763,6 +3763,9 @@ static int w_go_diskless(struct drbd_conf *mdev, struct drbd_work *w, int unused
 	 * the protected members anymore, though, so in the after_state_ch work
 	 * it will be safe to free them. */
 	drbd_force_state(mdev, NS(disk, D_DISKLESS));
+	/* We need to wait for return of references checked out while we still
+	 * have been D_FAILED, though (drbd_md_sync, bitmap io). */
+	wait_event(mdev->misc_wait, !atomic_read(&mdev->local_cnt));
 
 	clear_bit(GO_DISKLESS, &mdev->flags);
 	return 1;
diff --git a/include/linux/drbd.h b/include/linux/drbd.h
index da7d9bd4f3f0..9b2a0158f399 100644
--- a/include/linux/drbd.h
+++ b/include/linux/drbd.h
@@ -53,7 +53,7 @@
 
 
 extern const char *drbd_buildtag(void);
-#define REL_VERSION "8.3.9rc1"
+#define REL_VERSION "8.3.9rc2"
 #define API_VERSION 88
 #define PRO_VERSION_MIN 86
 #define PRO_VERSION_MAX 95
-- 
cgit v1.2.3


From 495d2b3883682fcd1c3dee3a45e38fd00154ae25 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 15 Oct 2010 15:49:20 +0200
Subject: block: Make the integrity mapped property a bio flag

Previously we tracked whether the integrity metadata had been remapped
using a request flag. This was fine for low-level retries. However, if
an I/O was redriven by upper layers we would end up remapping again,
causing the retry to fail.

Deprecate the REQ_INTEGRITY flag and introduce BIO_MAPPED_INTEGRITY
which enables filesystems to notify lower layers that the bio in
question has already been remapped.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 drivers/scsi/sd_dif.c     | 11 ++++++-----
 include/linux/blk_types.h |  3 +--
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index 84be62149c6c..0cb39ff21171 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -375,21 +375,20 @@ int sd_dif_prepare(struct request *rq, sector_t hw_sector, unsigned int sector_s
 	unsigned int i, j;
 	u32 phys, virt;
 
-	/* Already remapped? */
-	if (rq->cmd_flags & REQ_INTEGRITY)
-		return 0;
-
 	sdkp = rq->bio->bi_bdev->bd_disk->private_data;
 
 	if (sdkp->protection_type == SD_DIF_TYPE3_PROTECTION)
 		return 0;
 
-	rq->cmd_flags |= REQ_INTEGRITY;
 	phys = hw_sector & 0xffffffff;
 
 	__rq_for_each_bio(bio, rq) {
 		struct bio_vec *iv;
 
+		/* Already remapped? */
+		if (bio_flagged(bio, BIO_MAPPED_INTEGRITY))
+			break;
+
 		virt = bio->bi_integrity->bip_sector & 0xffffffff;
 
 		bip_for_each_vec(iv, bio->bi_integrity, i) {
@@ -408,6 +407,8 @@ int sd_dif_prepare(struct request *rq, sector_t hw_sector, unsigned int sector_s
 
 			kunmap_atomic(sdt, KM_USER0);
 		}
+
+		bio->bi_flags |= BIO_MAPPED_INTEGRITY;
 	}
 
 	return 0;
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index 10a0c291b55a..d36629620a4f 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -97,6 +97,7 @@ struct bio {
 #define BIO_NULL_MAPPED 9	/* contains invalid user pages */
 #define BIO_FS_INTEGRITY 10	/* fs owns integrity data, not block layer */
 #define BIO_QUIET	11	/* Make BIO Quiet */
+#define BIO_MAPPED_INTEGRITY 12/* integrity metadata has been remapped */
 #define bio_flagged(bio, flag)	((bio)->bi_flags & (1 << (flag)))
 
 /*
@@ -148,7 +149,6 @@ enum rq_flag_bits {
 	__REQ_ORDERED_COLOR,	/* is before or after barrier */
 	__REQ_ALLOCED,		/* request came from our alloc pool */
 	__REQ_COPY_USER,	/* contains copies of user pages */
-	__REQ_INTEGRITY,	/* integrity metadata has been remapped */
 	__REQ_FLUSH,		/* request for cache flush */
 	__REQ_IO_STAT,		/* account I/O stat */
 	__REQ_MIXED_MERGE,	/* merge of different types, fail separately */
@@ -190,7 +190,6 @@ enum rq_flag_bits {
 #define REQ_ORDERED_COLOR	(1 << __REQ_ORDERED_COLOR)
 #define REQ_ALLOCED		(1 << __REQ_ALLOCED)
 #define REQ_COPY_USER		(1 << __REQ_COPY_USER)
-#define REQ_INTEGRITY		(1 << __REQ_INTEGRITY)
 #define REQ_FLUSH		(1 << __REQ_FLUSH)
 #define REQ_IO_STAT		(1 << __REQ_IO_STAT)
 #define REQ_MIXED_MERGE		(1 << __REQ_MIXED_MERGE)
-- 
cgit v1.2.3


From 40ffa93791985ab300fd488072e9f37ccf72e88c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 15 Oct 2010 21:08:14 +0200
Subject: x86: Remove stale pmtimer_64.c

This file is unused since the apic unification in 2.6.29, but nobody
noticed.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/Makefile     |  1 -
 arch/x86/kernel/pmtimer_64.c | 69 --------------------------------------------
 include/linux/acpi_pmtmr.h   |  2 --
 3 files changed, 72 deletions(-)
 delete mode 100644 arch/x86/kernel/pmtimer_64.c

(limited to 'include/linux')

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 0925676266bd..dd9a2e459c78 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -120,7 +120,6 @@ obj-$(CONFIG_SWIOTLB)			+= pci-swiotlb.o
 # 64 bit specific files
 ifeq ($(CONFIG_X86_64),y)
 	obj-$(CONFIG_X86_UV)		+= tlb_uv.o bios_uv.o uv_irq.o uv_sysfs.o uv_time.o
-	obj-$(CONFIG_X86_PM_TIMER)	+= pmtimer_64.o
 	obj-$(CONFIG_AUDIT)		+= audit_64.o
 
 	obj-$(CONFIG_GART_IOMMU)	+= pci-gart_64.o aperture_64.o
diff --git a/arch/x86/kernel/pmtimer_64.c b/arch/x86/kernel/pmtimer_64.c
deleted file mode 100644
index b112406f1996..000000000000
--- a/arch/x86/kernel/pmtimer_64.c
+++ /dev/null
@@ -1,69 +0,0 @@
-/* Ported over from i386 by AK, original copyright was:
- *
- * (C) Dominik Brodowski <linux@brodo.de> 2003
- *
- * Driver to use the Power Management Timer (PMTMR) available in some
- * southbridges as primary timing source for the Linux kernel.
- *
- * Based on parts of linux/drivers/acpi/hardware/hwtimer.c, timer_pit.c,
- * timer_hpet.c, and on Arjan van de Ven's implementation for 2.4.
- *
- * This file is licensed under the GPL v2.
- *
- * Dropped all the hardware bug workarounds for now. Hopefully they
- * are not needed on 64bit chipsets.
- */
-
-#include <linux/jiffies.h>
-#include <linux/kernel.h>
-#include <linux/time.h>
-#include <linux/init.h>
-#include <linux/cpumask.h>
-#include <linux/acpi_pmtmr.h>
-
-#include <asm/io.h>
-#include <asm/proto.h>
-#include <asm/msr.h>
-#include <asm/vsyscall.h>
-
-static inline u32 cyc2us(u32 cycles)
-{
-	/* The Power Management Timer ticks at 3.579545 ticks per microsecond.
-	 * 1 / PM_TIMER_FREQUENCY == 0.27936511 =~ 286/1024 [error: 0.024%]
-	 *
-	 * Even with HZ = 100, delta is at maximum 35796 ticks, so it can
-	 * easily be multiplied with 286 (=0x11E) without having to fear
-	 * u32 overflows.
-	 */
-	cycles *= 286;
-	return (cycles >> 10);
-}
-
-static unsigned pmtimer_wait_tick(void)
-{
-	u32 a, b;
-	for (a = b = inl(pmtmr_ioport) & ACPI_PM_MASK;
-	     a == b;
-	     b = inl(pmtmr_ioport) & ACPI_PM_MASK)
-		cpu_relax();
-	return b;
-}
-
-/* note: wait time is rounded up to one tick */
-void pmtimer_wait(unsigned us)
-{
-	u32 a, b;
-	a = pmtimer_wait_tick();
-	do {
-		b = inl(pmtmr_ioport);
-		cpu_relax();
-	} while (cyc2us(b - a) < us);
-}
-
-static int __init nopmtimer_setup(char *s)
-{
-	pmtmr_ioport = 0;
-	return 1;
-}
-
-__setup("nopmtimer", nopmtimer_setup);
diff --git a/include/linux/acpi_pmtmr.h b/include/linux/acpi_pmtmr.h
index 7e3d2859be50..1d0ef1ae8036 100644
--- a/include/linux/acpi_pmtmr.h
+++ b/include/linux/acpi_pmtmr.h
@@ -25,8 +25,6 @@ static inline u32 acpi_pm_read_early(void)
 	return acpi_pm_read_verified() & ACPI_PM_MASK;
 }
 
-extern void pmtimer_wait(unsigned);
-
 #else
 
 static inline u32 acpi_pm_read_early(void)
-- 
cgit v1.2.3


From 79b5dc0c64d88cda3da23b2e22a5cec0964372ac Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Fri, 15 Oct 2010 14:34:14 -0700
Subject: types.h: define __aligned_u64 and expose to userspace

We currently have a kernel internal type called aligned_u64 which aligns
__u64's on 8 bytes boundaries even on systems which would normally align
them on 4 byte boundaries.  This patch creates a new type __aligned_u64
which does the same thing but which is exposed to userspace rather than
being kernel internal.

[akpm: merge early as both the net and audit trees want this]

[akpm@linux-foundation.org: enhance the comment describing the reasons for using aligned_u64.  Via Andreas and Andi.]
Based-on-patch-by: Andreas Gruenbacher <agruen@suse.de>
Signed-off-by: Eric Paris <eparis@redhat.com>
Cc: Jan Engelhardt <jengelh@medozas.de>
Cc: David Miller <davem@davemloft.net>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/types.h | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/types.h b/include/linux/types.h
index 01a082f56ef4..357dbc19606f 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -121,7 +121,15 @@ typedef		__u64		u_int64_t;
 typedef		__s64		int64_t;
 #endif
 
-/* this is a special 64bit data type that is 8-byte aligned */
+/*
+ * aligned_u64 should be used in defining kernel<->userspace ABIs to avoid
+ * common 32/64-bit compat problems.
+ * 64-bit values align to 4-byte boundaries on x86_32 (and possibly other
+ * architectures) and to 8-byte boundaries on 64-bit architetures.  The new
+ * aligned_64 type enforces 8-byte alignment so that structs containing
+ * aligned_64 values have the same alignment on 32-bit and 64-bit architectures.
+ * No conversions are necessary between 32-bit user-space and a 64-bit kernel.
+ */
 #define aligned_u64 __u64 __attribute__((aligned(8)))
 #define aligned_be64 __be64 __attribute__((aligned(8)))
 #define aligned_le64 __le64 __attribute__((aligned(8)))
@@ -178,6 +186,11 @@ typedef __u64 __bitwise __be64;
 typedef __u16 __bitwise __sum16;
 typedef __u32 __bitwise __wsum;
 
+/* this is a special 64bit data type that is 8-byte aligned */
+#define __aligned_u64 __u64 __attribute__((aligned(8)))
+#define __aligned_be64 __be64 __attribute__((aligned(8)))
+#define __aligned_le64 __le64 __attribute__((aligned(8)))
+
 #ifdef __KERNEL__
 typedef unsigned __bitwise__ gfp_t;
 typedef unsigned __bitwise__ fmode_t;
-- 
cgit v1.2.3


From 564824b0c52c34692d804bb6ea214451615b0b50 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 11 Oct 2010 19:05:25 +0000
Subject: net: allocate skbs on local node

commit b30973f877 (node-aware skb allocation) spread a wrong habit of
allocating net drivers skbs on a given memory node : The one closest to
the NIC hardware. This is wrong because as soon as we try to scale
network stack, we need to use many cpus to handle traffic and hit
slub/slab management on cross-node allocations/frees when these cpus
have to alloc/free skbs bound to a central node.

skb allocated in RX path are ephemeral, they have a very short
lifetime : Extra cost to maintain NUMA affinity is too expensive. What
appeared as a nice idea four years ago is in fact a bad one.

In 2010, NIC hardwares are multiqueue, or we use RPS to spread the load,
and two 10Gb NIC might deliver more than 28 million packets per second,
needing all the available cpus.

Cost of cross-node handling in network and vm stacks outperforms the
small benefit hardware had when doing its DMA transfert in its 'local'
memory node at RX time. Even trying to differentiate the two allocations
done for one skb (the sk_buff on local node, the data part on NIC
hardware node) is not enough to bring good performance.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 20 ++++++++++++++++----
 net/core/skbuff.c      | 13 +------------
 2 files changed, 17 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 0b53c43ac92e..05a358f1ba11 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -496,13 +496,13 @@ extern struct sk_buff *__alloc_skb(unsigned int size,
 static inline struct sk_buff *alloc_skb(unsigned int size,
 					gfp_t priority)
 {
-	return __alloc_skb(size, priority, 0, -1);
+	return __alloc_skb(size, priority, 0, NUMA_NO_NODE);
 }
 
 static inline struct sk_buff *alloc_skb_fclone(unsigned int size,
 					       gfp_t priority)
 {
-	return __alloc_skb(size, priority, 1, -1);
+	return __alloc_skb(size, priority, 1, NUMA_NO_NODE);
 }
 
 extern bool skb_recycle_check(struct sk_buff *skb, int skb_size);
@@ -1563,13 +1563,25 @@ static inline struct sk_buff *netdev_alloc_skb_ip_align(struct net_device *dev,
 	return skb;
 }
 
-extern struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask);
+/**
+ *	__netdev_alloc_page - allocate a page for ps-rx on a specific device
+ *	@dev: network device to receive on
+ *	@gfp_mask: alloc_pages_node mask
+ *
+ * 	Allocate a new page. dev currently unused.
+ *
+ * 	%NULL is returned if there is no free memory.
+ */
+static inline struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask)
+{
+	return alloc_pages_node(NUMA_NO_NODE, gfp_mask, 0);
+}
 
 /**
  *	netdev_alloc_page - allocate a page for ps-rx on a specific device
  *	@dev: network device to receive on
  *
- * 	Allocate a new page node local to the specified device.
+ * 	Allocate a new page. dev currently unused.
  *
  * 	%NULL is returned if there is no free memory.
  */
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 752c1972b3a7..4e8b82e167d8 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -247,10 +247,9 @@ EXPORT_SYMBOL(__alloc_skb);
 struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
 		unsigned int length, gfp_t gfp_mask)
 {
-	int node = dev->dev.parent ? dev_to_node(dev->dev.parent) : -1;
 	struct sk_buff *skb;
 
-	skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, node);
+	skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, NUMA_NO_NODE);
 	if (likely(skb)) {
 		skb_reserve(skb, NET_SKB_PAD);
 		skb->dev = dev;
@@ -259,16 +258,6 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
 }
 EXPORT_SYMBOL(__netdev_alloc_skb);
 
-struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask)
-{
-	int node = dev->dev.parent ? dev_to_node(dev->dev.parent) : -1;
-	struct page *page;
-
-	page = alloc_pages_node(node, gfp_mask, 0);
-	return page;
-}
-EXPORT_SYMBOL(__netdev_alloc_page);
-
 void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page, int off,
 		int size)
 {
-- 
cgit v1.2.3


From 074037ec79bea73edf1b1ec72fef1010e83e3cc5 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 22 Sep 2010 22:09:10 +0200
Subject: PM / Wakeup: Introduce wakeup source objects and event statistics
 (v3)

Introduce struct wakeup_source for representing system wakeup sources
within the kernel and for collecting statistics related to them.
Make the recently introduced helper functions pm_wakeup_event(),
pm_stay_awake() and pm_relax() use struct wakeup_source objects
internally, so that wakeup statistics associated with wakeup devices
can be collected and reported in a consistent way (the definition of
pm_relax() is changed, which is harmless, because this function is
not called directly by anyone yet).  Introduce new wakeup-related
sysfs device attributes in /sys/devices/.../power for reporting the
device wakeup statistics.

Change the global wakeup events counters event_count and
events_in_progress into atomic variables, so that it is not necessary
to acquire a global spinlock in pm_wakeup_event(), pm_stay_awake()
and pm_relax(), which should allow us to avoid lock contention in
these functions on SMP systems with many wakeup devices.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/ABI/testing/sysfs-devices-power |  70 ++++
 drivers/base/power/main.c                     |   4 +-
 drivers/base/power/power.h                    |   1 +
 drivers/base/power/runtime.c                  |   2 -
 drivers/base/power/sysfs.c                    | 121 +++++-
 drivers/base/power/wakeup.c                   | 528 ++++++++++++++++++++++----
 include/linux/pm.h                            |  16 +-
 include/linux/pm_wakeup.h                     | 127 +++++--
 include/linux/suspend.h                       |   4 +-
 kernel/power/main.c                           |   8 +-
 10 files changed, 756 insertions(+), 125 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-devices-power b/Documentation/ABI/testing/sysfs-devices-power
index 6123c523bfd7..6bb2dd3c3a71 100644
--- a/Documentation/ABI/testing/sysfs-devices-power
+++ b/Documentation/ABI/testing/sysfs-devices-power
@@ -77,3 +77,73 @@ Description:
 		devices this attribute is set to "enabled" by bus type code or
 		device drivers and in that cases it should be safe to leave the
 		default value.
+
+What:		/sys/devices/.../power/wakeup_count
+Date:		September 2010
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/devices/.../wakeup_count attribute contains the number
+		of signaled wakeup events associated with the device.  This
+		attribute is read-only.  If the device is not enabled to wake up
+		the system from sleep states, this attribute is empty.
+
+What:		/sys/devices/.../power/wakeup_active_count
+Date:		September 2010
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/devices/.../wakeup_active_count attribute contains the
+		number of times the processing of wakeup events associated with
+		the device was completed (at the kernel level).  This attribute
+		is read-only.  If the device is not enabled to wake up the
+		system from sleep states, this attribute is empty.
+
+What:		/sys/devices/.../power/wakeup_hit_count
+Date:		September 2010
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/devices/.../wakeup_hit_count attribute contains the
+		number of times the processing of a wakeup event associated with
+		the device might prevent the system from entering a sleep state.
+		This attribute is read-only.  If the device is not enabled to
+		wake up the system from sleep states, this attribute is empty.
+
+What:		/sys/devices/.../power/wakeup_active
+Date:		September 2010
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/devices/.../wakeup_active attribute contains either 1,
+		or 0, depending on whether or not a wakeup event associated with
+		the device is being processed (1).  This attribute is read-only.
+		If the device is not enabled to wake up the system from sleep
+		states, this attribute is empty.
+
+What:		/sys/devices/.../power/wakeup_total_time_ms
+Date:		September 2010
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/devices/.../wakeup_total_time_ms attribute contains
+		the total time of processing wakeup events associated with the
+		device, in milliseconds.  This attribute is read-only.  If the
+		device is not enabled to wake up the system from sleep states,
+		this attribute is empty.
+
+What:		/sys/devices/.../power/wakeup_max_time_ms
+Date:		September 2010
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/devices/.../wakeup_max_time_ms attribute contains
+		the maximum time of processing a single wakeup event associated
+		with the device, in milliseconds.  This attribute is read-only.
+		If the device is not enabled to wake up the system from sleep
+		states, this attribute is empty.
+
+What:		/sys/devices/.../power/wakeup_last_time_ms
+Date:		September 2010
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/devices/.../wakeup_last_time_ms attribute contains
+		the value of the monotonic clock corresponding to the time of
+		signaling the last wakeup event associated with the device, in
+		milliseconds.  This attribute is read-only.  If the device is
+		not enabled to wake up the system from sleep states, this
+		attribute is empty.
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index db677199403c..7ae6fe414c38 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -60,7 +60,8 @@ void device_pm_init(struct device *dev)
 	dev->power.status = DPM_ON;
 	init_completion(&dev->power.completion);
 	complete_all(&dev->power.completion);
-	dev->power.wakeup_count = 0;
+	dev->power.wakeup = NULL;
+	spin_lock_init(&dev->power.lock);
 	pm_runtime_init(dev);
 }
 
@@ -120,6 +121,7 @@ void device_pm_remove(struct device *dev)
 	mutex_lock(&dpm_list_mtx);
 	list_del_init(&dev->power.entry);
 	mutex_unlock(&dpm_list_mtx);
+	device_wakeup_disable(dev);
 	pm_runtime_remove(dev);
 }
 
diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h
index c0bd03c83b9c..8b2745c69e2a 100644
--- a/drivers/base/power/power.h
+++ b/drivers/base/power/power.h
@@ -34,6 +34,7 @@ extern void device_pm_move_last(struct device *);
 
 static inline void device_pm_init(struct device *dev)
 {
+	spin_lock_init(&dev->power.lock);
 	pm_runtime_init(dev);
 }
 
diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c
index b78c401ffa73..e9520ad2d716 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -1099,8 +1099,6 @@ EXPORT_SYMBOL_GPL(pm_runtime_allow);
  */
 void pm_runtime_init(struct device *dev)
 {
-	spin_lock_init(&dev->power.lock);
-
 	dev->power.runtime_status = RPM_SUSPENDED;
 	dev->power.idle_notification = false;
 
diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c
index e56b4388fe61..8859780817e1 100644
--- a/drivers/base/power/sysfs.c
+++ b/drivers/base/power/sysfs.c
@@ -210,11 +210,122 @@ static DEVICE_ATTR(wakeup, 0644, wake_show, wake_store);
 static ssize_t wakeup_count_show(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
-	return sprintf(buf, "%lu\n", dev->power.wakeup_count);
+	unsigned long count = 0;
+	bool enabled = false;
+
+	spin_lock_irq(&dev->power.lock);
+	if (dev->power.wakeup) {
+		count = dev->power.wakeup->event_count;
+		enabled = true;
+	}
+	spin_unlock_irq(&dev->power.lock);
+	return enabled ? sprintf(buf, "%lu\n", count) : sprintf(buf, "\n");
 }
 
 static DEVICE_ATTR(wakeup_count, 0444, wakeup_count_show, NULL);
-#endif
+
+static ssize_t wakeup_active_count_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	unsigned long count = 0;
+	bool enabled = false;
+
+	spin_lock_irq(&dev->power.lock);
+	if (dev->power.wakeup) {
+		count = dev->power.wakeup->active_count;
+		enabled = true;
+	}
+	spin_unlock_irq(&dev->power.lock);
+	return enabled ? sprintf(buf, "%lu\n", count) : sprintf(buf, "\n");
+}
+
+static DEVICE_ATTR(wakeup_active_count, 0444, wakeup_active_count_show, NULL);
+
+static ssize_t wakeup_hit_count_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	unsigned long count = 0;
+	bool enabled = false;
+
+	spin_lock_irq(&dev->power.lock);
+	if (dev->power.wakeup) {
+		count = dev->power.wakeup->hit_count;
+		enabled = true;
+	}
+	spin_unlock_irq(&dev->power.lock);
+	return enabled ? sprintf(buf, "%lu\n", count) : sprintf(buf, "\n");
+}
+
+static DEVICE_ATTR(wakeup_hit_count, 0444, wakeup_hit_count_show, NULL);
+
+static ssize_t wakeup_active_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	unsigned int active = 0;
+	bool enabled = false;
+
+	spin_lock_irq(&dev->power.lock);
+	if (dev->power.wakeup) {
+		active = dev->power.wakeup->active;
+		enabled = true;
+	}
+	spin_unlock_irq(&dev->power.lock);
+	return enabled ? sprintf(buf, "%u\n", active) : sprintf(buf, "\n");
+}
+
+static DEVICE_ATTR(wakeup_active, 0444, wakeup_active_show, NULL);
+
+static ssize_t wakeup_total_time_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	s64 msec = 0;
+	bool enabled = false;
+
+	spin_lock_irq(&dev->power.lock);
+	if (dev->power.wakeup) {
+		msec = ktime_to_ms(dev->power.wakeup->total_time);
+		enabled = true;
+	}
+	spin_unlock_irq(&dev->power.lock);
+	return enabled ? sprintf(buf, "%lld\n", msec) : sprintf(buf, "\n");
+}
+
+static DEVICE_ATTR(wakeup_total_time_ms, 0444, wakeup_total_time_show, NULL);
+
+static ssize_t wakeup_max_time_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	s64 msec = 0;
+	bool enabled = false;
+
+	spin_lock_irq(&dev->power.lock);
+	if (dev->power.wakeup) {
+		msec = ktime_to_ms(dev->power.wakeup->max_time);
+		enabled = true;
+	}
+	spin_unlock_irq(&dev->power.lock);
+	return enabled ? sprintf(buf, "%lld\n", msec) : sprintf(buf, "\n");
+}
+
+static DEVICE_ATTR(wakeup_max_time_ms, 0444, wakeup_max_time_show, NULL);
+
+static ssize_t wakeup_last_time_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	s64 msec = 0;
+	bool enabled = false;
+
+	spin_lock_irq(&dev->power.lock);
+	if (dev->power.wakeup) {
+		msec = ktime_to_ms(dev->power.wakeup->last_time);
+		enabled = true;
+	}
+	spin_unlock_irq(&dev->power.lock);
+	return enabled ? sprintf(buf, "%lld\n", msec) : sprintf(buf, "\n");
+}
+
+static DEVICE_ATTR(wakeup_last_time_ms, 0444, wakeup_last_time_show, NULL);
+#endif /* CONFIG_PM_SLEEP */
 
 #ifdef CONFIG_PM_ADVANCED_DEBUG
 #ifdef CONFIG_PM_RUNTIME
@@ -288,6 +399,12 @@ static struct attribute * power_attrs[] = {
 	&dev_attr_wakeup.attr,
 #ifdef CONFIG_PM_SLEEP
 	&dev_attr_wakeup_count.attr,
+	&dev_attr_wakeup_active_count.attr,
+	&dev_attr_wakeup_hit_count.attr,
+	&dev_attr_wakeup_active.attr,
+	&dev_attr_wakeup_total_time_ms.attr,
+	&dev_attr_wakeup_max_time_ms.attr,
+	&dev_attr_wakeup_last_time_ms.attr,
 #endif
 #ifdef CONFIG_PM_ADVANCED_DEBUG
 	&dev_attr_async.attr,
diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c
index eb594facfc3f..03751a01dbad 100644
--- a/drivers/base/power/wakeup.c
+++ b/drivers/base/power/wakeup.c
@@ -11,7 +11,10 @@
 #include <linux/sched.h>
 #include <linux/capability.h>
 #include <linux/suspend.h>
-#include <linux/pm.h>
+
+#include "power.h"
+
+#define TIMEOUT		100
 
 /*
  * If set, the suspend/hibernate code will abort transitions to a sleep state
@@ -20,18 +23,244 @@
 bool events_check_enabled;
 
 /* The counter of registered wakeup events. */
-static unsigned long event_count;
+static atomic_t event_count = ATOMIC_INIT(0);
 /* A preserved old value of event_count. */
-static unsigned long saved_event_count;
+static unsigned int saved_count;
 /* The counter of wakeup events being processed. */
-static unsigned long events_in_progress;
+static atomic_t events_in_progress = ATOMIC_INIT(0);
 
 static DEFINE_SPINLOCK(events_lock);
 
 static void pm_wakeup_timer_fn(unsigned long data);
 
-static DEFINE_TIMER(events_timer, pm_wakeup_timer_fn, 0, 0);
-static unsigned long events_timer_expires;
+static LIST_HEAD(wakeup_sources);
+
+/**
+ * wakeup_source_create - Create a struct wakeup_source object.
+ * @name: Name of the new wakeup source.
+ */
+struct wakeup_source *wakeup_source_create(const char *name)
+{
+	struct wakeup_source *ws;
+
+	ws = kzalloc(sizeof(*ws), GFP_KERNEL);
+	if (!ws)
+		return NULL;
+
+	spin_lock_init(&ws->lock);
+	if (name)
+		ws->name = kstrdup(name, GFP_KERNEL);
+
+	return ws;
+}
+EXPORT_SYMBOL_GPL(wakeup_source_create);
+
+/**
+ * wakeup_source_destroy - Destroy a struct wakeup_source object.
+ * @ws: Wakeup source to destroy.
+ */
+void wakeup_source_destroy(struct wakeup_source *ws)
+{
+	if (!ws)
+		return;
+
+	spin_lock_irq(&ws->lock);
+	while (ws->active) {
+		spin_unlock_irq(&ws->lock);
+
+		schedule_timeout_interruptible(msecs_to_jiffies(TIMEOUT));
+
+		spin_lock_irq(&ws->lock);
+	}
+	spin_unlock_irq(&ws->lock);
+
+	kfree(ws->name);
+	kfree(ws);
+}
+EXPORT_SYMBOL_GPL(wakeup_source_destroy);
+
+/**
+ * wakeup_source_add - Add given object to the list of wakeup sources.
+ * @ws: Wakeup source object to add to the list.
+ */
+void wakeup_source_add(struct wakeup_source *ws)
+{
+	if (WARN_ON(!ws))
+		return;
+
+	setup_timer(&ws->timer, pm_wakeup_timer_fn, (unsigned long)ws);
+	ws->active = false;
+
+	spin_lock_irq(&events_lock);
+	list_add_rcu(&ws->entry, &wakeup_sources);
+	spin_unlock_irq(&events_lock);
+	synchronize_rcu();
+}
+EXPORT_SYMBOL_GPL(wakeup_source_add);
+
+/**
+ * wakeup_source_remove - Remove given object from the wakeup sources list.
+ * @ws: Wakeup source object to remove from the list.
+ */
+void wakeup_source_remove(struct wakeup_source *ws)
+{
+	if (WARN_ON(!ws))
+		return;
+
+	spin_lock_irq(&events_lock);
+	list_del_rcu(&ws->entry);
+	spin_unlock_irq(&events_lock);
+	synchronize_rcu();
+}
+EXPORT_SYMBOL_GPL(wakeup_source_remove);
+
+/**
+ * wakeup_source_register - Create wakeup source and add it to the list.
+ * @name: Name of the wakeup source to register.
+ */
+struct wakeup_source *wakeup_source_register(const char *name)
+{
+	struct wakeup_source *ws;
+
+	ws = wakeup_source_create(name);
+	if (ws)
+		wakeup_source_add(ws);
+
+	return ws;
+}
+EXPORT_SYMBOL_GPL(wakeup_source_register);
+
+/**
+ * wakeup_source_unregister - Remove wakeup source from the list and remove it.
+ * @ws: Wakeup source object to unregister.
+ */
+void wakeup_source_unregister(struct wakeup_source *ws)
+{
+	wakeup_source_remove(ws);
+	wakeup_source_destroy(ws);
+}
+EXPORT_SYMBOL_GPL(wakeup_source_unregister);
+
+/**
+ * device_wakeup_attach - Attach a wakeup source object to a device object.
+ * @dev: Device to handle.
+ * @ws: Wakeup source object to attach to @dev.
+ *
+ * This causes @dev to be treated as a wakeup device.
+ */
+static int device_wakeup_attach(struct device *dev, struct wakeup_source *ws)
+{
+	spin_lock_irq(&dev->power.lock);
+	if (dev->power.wakeup) {
+		spin_unlock_irq(&dev->power.lock);
+		return -EEXIST;
+	}
+	dev->power.wakeup = ws;
+	spin_unlock_irq(&dev->power.lock);
+	return 0;
+}
+
+/**
+ * device_wakeup_enable - Enable given device to be a wakeup source.
+ * @dev: Device to handle.
+ *
+ * Create a wakeup source object, register it and attach it to @dev.
+ */
+int device_wakeup_enable(struct device *dev)
+{
+	struct wakeup_source *ws;
+	int ret;
+
+	if (!dev || !dev->power.can_wakeup)
+		return -EINVAL;
+
+	ws = wakeup_source_register(dev_name(dev));
+	if (!ws)
+		return -ENOMEM;
+
+	ret = device_wakeup_attach(dev, ws);
+	if (ret)
+		wakeup_source_unregister(ws);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(device_wakeup_enable);
+
+/**
+ * device_wakeup_detach - Detach a device's wakeup source object from it.
+ * @dev: Device to detach the wakeup source object from.
+ *
+ * After it returns, @dev will not be treated as a wakeup device any more.
+ */
+static struct wakeup_source *device_wakeup_detach(struct device *dev)
+{
+	struct wakeup_source *ws;
+
+	spin_lock_irq(&dev->power.lock);
+	ws = dev->power.wakeup;
+	dev->power.wakeup = NULL;
+	spin_unlock_irq(&dev->power.lock);
+	return ws;
+}
+
+/**
+ * device_wakeup_disable - Do not regard a device as a wakeup source any more.
+ * @dev: Device to handle.
+ *
+ * Detach the @dev's wakeup source object from it, unregister this wakeup source
+ * object and destroy it.
+ */
+int device_wakeup_disable(struct device *dev)
+{
+	struct wakeup_source *ws;
+
+	if (!dev || !dev->power.can_wakeup)
+		return -EINVAL;
+
+	ws = device_wakeup_detach(dev);
+	if (ws)
+		wakeup_source_unregister(ws);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(device_wakeup_disable);
+
+/**
+ * device_init_wakeup - Device wakeup initialization.
+ * @dev: Device to handle.
+ * @enable: Whether or not to enable @dev as a wakeup device.
+ *
+ * By default, most devices should leave wakeup disabled.  The exceptions are
+ * devices that everyone expects to be wakeup sources: keyboards, power buttons,
+ * possibly network interfaces, etc.
+ */
+int device_init_wakeup(struct device *dev, bool enable)
+{
+	int ret = 0;
+
+	if (enable) {
+		device_set_wakeup_capable(dev, true);
+		ret = device_wakeup_enable(dev);
+	} else {
+		device_set_wakeup_capable(dev, false);
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(device_init_wakeup);
+
+/**
+ * device_set_wakeup_enable - Enable or disable a device to wake up the system.
+ * @dev: Device to handle.
+ */
+int device_set_wakeup_enable(struct device *dev, bool enable)
+{
+	if (!dev || !dev->power.can_wakeup)
+		return -EINVAL;
+
+	return enable ? device_wakeup_enable(dev) : device_wakeup_disable(dev);
+}
+EXPORT_SYMBOL_GPL(device_set_wakeup_enable);
 
 /*
  * The functions below use the observation that each wakeup event starts a
@@ -55,118 +284,259 @@ static unsigned long events_timer_expires;
  * knowledge, however, may not be available to it, so it can simply specify time
  * to wait before the system can be suspended and pass it as the second
  * argument of pm_wakeup_event().
+ *
+ * It is valid to call pm_relax() after pm_wakeup_event(), in which case the
+ * "no suspend" period will be ended either by the pm_relax(), or by the timer
+ * function executed when the timer expires, whichever comes first.
  */
 
+/**
+ * wakup_source_activate - Mark given wakeup source as active.
+ * @ws: Wakeup source to handle.
+ *
+ * Update the @ws' statistics and, if @ws has just been activated, notify the PM
+ * core of the event by incrementing the counter of of wakeup events being
+ * processed.
+ */
+static void wakeup_source_activate(struct wakeup_source *ws)
+{
+	ws->active = true;
+	ws->active_count++;
+	ws->timer_expires = jiffies;
+	ws->last_time = ktime_get();
+
+	atomic_inc(&events_in_progress);
+}
+
+/**
+ * __pm_stay_awake - Notify the PM core of a wakeup event.
+ * @ws: Wakeup source object associated with the source of the event.
+ *
+ * It is safe to call this function from interrupt context.
+ */
+void __pm_stay_awake(struct wakeup_source *ws)
+{
+	unsigned long flags;
+
+	if (!ws)
+		return;
+
+	spin_lock_irqsave(&ws->lock, flags);
+	ws->event_count++;
+	if (!ws->active)
+		wakeup_source_activate(ws);
+	spin_unlock_irqrestore(&ws->lock, flags);
+}
+EXPORT_SYMBOL_GPL(__pm_stay_awake);
+
 /**
  * pm_stay_awake - Notify the PM core that a wakeup event is being processed.
  * @dev: Device the wakeup event is related to.
  *
- * Notify the PM core of a wakeup event (signaled by @dev) by incrementing the
- * counter of wakeup events being processed.  If @dev is not NULL, the counter
- * of wakeup events related to @dev is incremented too.
+ * Notify the PM core of a wakeup event (signaled by @dev) by calling
+ * __pm_stay_awake for the @dev's wakeup source object.
  *
  * Call this function after detecting of a wakeup event if pm_relax() is going
  * to be called directly after processing the event (and possibly passing it to
  * user space for further processing).
- *
- * It is safe to call this function from interrupt context.
  */
 void pm_stay_awake(struct device *dev)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&events_lock, flags);
-	if (dev)
-		dev->power.wakeup_count++;
+	if (!dev)
+		return;
 
-	events_in_progress++;
-	spin_unlock_irqrestore(&events_lock, flags);
+	spin_lock_irqsave(&dev->power.lock, flags);
+	__pm_stay_awake(dev->power.wakeup);
+	spin_unlock_irqrestore(&dev->power.lock, flags);
 }
+EXPORT_SYMBOL_GPL(pm_stay_awake);
 
 /**
- * pm_relax - Notify the PM core that processing of a wakeup event has ended.
+ * wakup_source_deactivate - Mark given wakeup source as inactive.
+ * @ws: Wakeup source to handle.
  *
- * Notify the PM core that a wakeup event has been processed by decrementing
- * the counter of wakeup events being processed and incrementing the counter
- * of registered wakeup events.
+ * Update the @ws' statistics and notify the PM core that the wakeup source has
+ * become inactive by decrementing the counter of wakeup events being processed
+ * and incrementing the counter of registered wakeup events.
+ */
+static void wakeup_source_deactivate(struct wakeup_source *ws)
+{
+	ktime_t duration;
+	ktime_t now;
+
+	ws->relax_count++;
+	/*
+	 * __pm_relax() may be called directly or from a timer function.
+	 * If it is called directly right after the timer function has been
+	 * started, but before the timer function calls __pm_relax(), it is
+	 * possible that __pm_stay_awake() will be called in the meantime and
+	 * will set ws->active.  Then, ws->active may be cleared immediately
+	 * by the __pm_relax() called from the timer function, but in such a
+	 * case ws->relax_count will be different from ws->active_count.
+	 */
+	if (ws->relax_count != ws->active_count) {
+		ws->relax_count--;
+		return;
+	}
+
+	ws->active = false;
+
+	now = ktime_get();
+	duration = ktime_sub(now, ws->last_time);
+	ws->total_time = ktime_add(ws->total_time, duration);
+	if (ktime_to_ns(duration) > ktime_to_ns(ws->max_time))
+		ws->max_time = duration;
+
+	del_timer(&ws->timer);
+
+	/*
+	 * event_count has to be incremented before events_in_progress is
+	 * modified, so that the callers of pm_check_wakeup_events() and
+	 * pm_save_wakeup_count() don't see the old value of event_count and
+	 * events_in_progress equal to zero at the same time.
+	 */
+	atomic_inc(&event_count);
+	smp_mb__before_atomic_dec();
+	atomic_dec(&events_in_progress);
+}
+
+/**
+ * __pm_relax - Notify the PM core that processing of a wakeup event has ended.
+ * @ws: Wakeup source object associated with the source of the event.
  *
  * Call this function for wakeup events whose processing started with calling
- * pm_stay_awake().
+ * __pm_stay_awake().
  *
  * It is safe to call it from interrupt context.
  */
-void pm_relax(void)
+void __pm_relax(struct wakeup_source *ws)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&events_lock, flags);
-	if (events_in_progress) {
-		events_in_progress--;
-		event_count++;
-	}
-	spin_unlock_irqrestore(&events_lock, flags);
+	if (!ws)
+		return;
+
+	spin_lock_irqsave(&ws->lock, flags);
+	if (ws->active)
+		wakeup_source_deactivate(ws);
+	spin_unlock_irqrestore(&ws->lock, flags);
+}
+EXPORT_SYMBOL_GPL(__pm_relax);
+
+/**
+ * pm_relax - Notify the PM core that processing of a wakeup event has ended.
+ * @dev: Device that signaled the event.
+ *
+ * Execute __pm_relax() for the @dev's wakeup source object.
+ */
+void pm_relax(struct device *dev)
+{
+	unsigned long flags;
+
+	if (!dev)
+		return;
+
+	spin_lock_irqsave(&dev->power.lock, flags);
+	__pm_relax(dev->power.wakeup);
+	spin_unlock_irqrestore(&dev->power.lock, flags);
 }
+EXPORT_SYMBOL_GPL(pm_relax);
 
 /**
  * pm_wakeup_timer_fn - Delayed finalization of a wakeup event.
+ * @data: Address of the wakeup source object associated with the event source.
  *
- * Decrease the counter of wakeup events being processed after it was increased
- * by pm_wakeup_event().
+ * Call __pm_relax() for the wakeup source whose address is stored in @data.
  */
 static void pm_wakeup_timer_fn(unsigned long data)
+{
+	__pm_relax((struct wakeup_source *)data);
+}
+
+/**
+ * __pm_wakeup_event - Notify the PM core of a wakeup event.
+ * @ws: Wakeup source object associated with the event source.
+ * @msec: Anticipated event processing time (in milliseconds).
+ *
+ * Notify the PM core of a wakeup event whose source is @ws that will take
+ * approximately @msec milliseconds to be processed by the kernel.  If @ws is
+ * not active, activate it.  If @msec is nonzero, set up the @ws' timer to
+ * execute pm_wakeup_timer_fn() in future.
+ *
+ * It is safe to call this function from interrupt context.
+ */
+void __pm_wakeup_event(struct wakeup_source *ws, unsigned int msec)
 {
 	unsigned long flags;
+	unsigned long expires;
 
-	spin_lock_irqsave(&events_lock, flags);
-	if (events_timer_expires
-	    && time_before_eq(events_timer_expires, jiffies)) {
-		events_in_progress--;
-		events_timer_expires = 0;
+	if (!ws)
+		return;
+
+	spin_lock_irqsave(&ws->lock, flags);
+
+	ws->event_count++;
+	if (!ws->active)
+		wakeup_source_activate(ws);
+
+	if (!msec) {
+		wakeup_source_deactivate(ws);
+		goto unlock;
 	}
-	spin_unlock_irqrestore(&events_lock, flags);
+
+	expires = jiffies + msecs_to_jiffies(msec);
+	if (!expires)
+		expires = 1;
+
+	if (time_after(expires, ws->timer_expires)) {
+		mod_timer(&ws->timer, expires);
+		ws->timer_expires = expires;
+	}
+
+ unlock:
+	spin_unlock_irqrestore(&ws->lock, flags);
 }
+EXPORT_SYMBOL_GPL(__pm_wakeup_event);
+
 
 /**
  * pm_wakeup_event - Notify the PM core of a wakeup event.
  * @dev: Device the wakeup event is related to.
  * @msec: Anticipated event processing time (in milliseconds).
  *
- * Notify the PM core of a wakeup event (signaled by @dev) that will take
- * approximately @msec milliseconds to be processed by the kernel.  Increment
- * the counter of registered wakeup events and (if @msec is nonzero) set up
- * the wakeup events timer to execute pm_wakeup_timer_fn() in future (if the
- * timer has not been set up already, increment the counter of wakeup events
- * being processed).  If @dev is not NULL, the counter of wakeup events related
- * to @dev is incremented too.
- *
- * It is safe to call this function from interrupt context.
+ * Call __pm_wakeup_event() for the @dev's wakeup source object.
  */
 void pm_wakeup_event(struct device *dev, unsigned int msec)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&events_lock, flags);
-	event_count++;
-	if (dev)
-		dev->power.wakeup_count++;
-
-	if (msec) {
-		unsigned long expires;
+	if (!dev)
+		return;
 
-		expires = jiffies + msecs_to_jiffies(msec);
-		if (!expires)
-			expires = 1;
+	spin_lock_irqsave(&dev->power.lock, flags);
+	__pm_wakeup_event(dev->power.wakeup, msec);
+	spin_unlock_irqrestore(&dev->power.lock, flags);
+}
+EXPORT_SYMBOL_GPL(pm_wakeup_event);
 
-		if (!events_timer_expires
-		    || time_after(expires, events_timer_expires)) {
-			if (!events_timer_expires)
-				events_in_progress++;
+/**
+ * pm_wakeup_update_hit_counts - Update hit counts of all active wakeup sources.
+ */
+static void pm_wakeup_update_hit_counts(void)
+{
+	unsigned long flags;
+	struct wakeup_source *ws;
 
-			mod_timer(&events_timer, expires);
-			events_timer_expires = expires;
-		}
+	rcu_read_lock();
+	list_for_each_entry_rcu(ws, &wakeup_sources, entry) {
+		spin_lock_irqsave(&ws->lock, flags);
+		if (ws->active)
+			ws->hit_count++;
+		spin_unlock_irqrestore(&ws->lock, flags);
 	}
-	spin_unlock_irqrestore(&events_lock, flags);
+	rcu_read_unlock();
 }
 
 /**
@@ -184,10 +554,13 @@ bool pm_check_wakeup_events(void)
 
 	spin_lock_irqsave(&events_lock, flags);
 	if (events_check_enabled) {
-		ret = (event_count == saved_event_count) && !events_in_progress;
+		ret = ((unsigned int)atomic_read(&event_count) == saved_count)
+			&& !atomic_read(&events_in_progress);
 		events_check_enabled = ret;
 	}
 	spin_unlock_irqrestore(&events_lock, flags);
+	if (!ret)
+		pm_wakeup_update_hit_counts();
 	return ret;
 }
 
@@ -202,24 +575,20 @@ bool pm_check_wakeup_events(void)
  * drop down to zero has been interrupted by a signal (and the current number
  * of wakeup events being processed is still nonzero).  Otherwise return true.
  */
-bool pm_get_wakeup_count(unsigned long *count)
+bool pm_get_wakeup_count(unsigned int *count)
 {
 	bool ret;
 
-	spin_lock_irq(&events_lock);
 	if (capable(CAP_SYS_ADMIN))
 		events_check_enabled = false;
 
-	while (events_in_progress && !signal_pending(current)) {
-		spin_unlock_irq(&events_lock);
-
-		schedule_timeout_interruptible(msecs_to_jiffies(100));
-
-		spin_lock_irq(&events_lock);
+	while (atomic_read(&events_in_progress) && !signal_pending(current)) {
+		pm_wakeup_update_hit_counts();
+		schedule_timeout_interruptible(msecs_to_jiffies(TIMEOUT));
 	}
-	*count = event_count;
-	ret = !events_in_progress;
-	spin_unlock_irq(&events_lock);
+
+	ret = !atomic_read(&events_in_progress);
+	*count = atomic_read(&event_count);
 	return ret;
 }
 
@@ -232,16 +601,19 @@ bool pm_get_wakeup_count(unsigned long *count)
  * old number of registered wakeup events to be used by pm_check_wakeup_events()
  * and return true.  Otherwise return false.
  */
-bool pm_save_wakeup_count(unsigned long count)
+bool pm_save_wakeup_count(unsigned int count)
 {
 	bool ret = false;
 
 	spin_lock_irq(&events_lock);
-	if (count == event_count && !events_in_progress) {
-		saved_event_count = count;
+	if (count == (unsigned int)atomic_read(&event_count)
+	    && !atomic_read(&events_in_progress)) {
+		saved_count = count;
 		events_check_enabled = true;
 		ret = true;
 	}
 	spin_unlock_irq(&events_lock);
+	if (!ret)
+		pm_wakeup_update_hit_counts();
 	return ret;
 }
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 52e8c55ff314..a84118911ced 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -448,23 +448,24 @@ enum rpm_request {
 	RPM_REQ_RESUME,
 };
 
+struct wakeup_source;
+
 struct dev_pm_info {
 	pm_message_t		power_state;
 	unsigned int		can_wakeup:1;
-	unsigned int		should_wakeup:1;
 	unsigned		async_suspend:1;
 	enum dpm_state		status;		/* Owned by the PM core */
+	spinlock_t		lock;
 #ifdef CONFIG_PM_SLEEP
 	struct list_head	entry;
 	struct completion	completion;
-	unsigned long		wakeup_count;
+	struct wakeup_source	*wakeup;
 #endif
 #ifdef CONFIG_PM_RUNTIME
 	struct timer_list	suspend_timer;
 	unsigned long		timer_expires;
 	struct work_struct	work;
 	wait_queue_head_t	wait_queue;
-	spinlock_t		lock;
 	atomic_t		usage_count;
 	atomic_t		child_count;
 	unsigned int		disable_depth:3;
@@ -559,11 +560,6 @@ extern void __suspend_report_result(const char *function, void *fn, int ret);
 	} while (0)
 
 extern void device_pm_wait_for_dev(struct device *sub, struct device *dev);
-
-/* drivers/base/power/wakeup.c */
-extern void pm_wakeup_event(struct device *dev, unsigned int msec);
-extern void pm_stay_awake(struct device *dev);
-extern void pm_relax(void);
 #else /* !CONFIG_PM_SLEEP */
 
 #define device_pm_lock() do {} while (0)
@@ -577,10 +573,6 @@ static inline int dpm_suspend_start(pm_message_t state)
 #define suspend_report_result(fn, ret)		do {} while (0)
 
 static inline void device_pm_wait_for_dev(struct device *a, struct device *b) {}
-
-static inline void pm_wakeup_event(struct device *dev, unsigned int msec) {}
-static inline void pm_stay_awake(struct device *dev) {}
-static inline void pm_relax(void) {}
 #endif /* !CONFIG_PM_SLEEP */
 
 /* How to reorder dpm_list after device_move() */
diff --git a/include/linux/pm_wakeup.h b/include/linux/pm_wakeup.h
index 76aca48722ae..9cff00dd6b63 100644
--- a/include/linux/pm_wakeup.h
+++ b/include/linux/pm_wakeup.h
@@ -2,6 +2,7 @@
  *  pm_wakeup.h - Power management wakeup interface
  *
  *  Copyright (C) 2008 Alan Stern
+ *  Copyright (C) 2010 Rafael J. Wysocki, Novell Inc.
  *
  *  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
@@ -27,19 +28,77 @@
 
 #include <linux/types.h>
 
-#ifdef CONFIG_PM
-
-/* Changes to device_may_wakeup take effect on the next pm state change.
+/**
+ * struct wakeup_source - Representation of wakeup sources
  *
- * By default, most devices should leave wakeup disabled.  The exceptions
- * are devices that everyone expects to be wakeup sources: keyboards,
- * power buttons, possibly network interfaces, etc.
+ * @total_time: Total time this wakeup source has been active.
+ * @max_time: Maximum time this wakeup source has been continuously active.
+ * @last_time: Monotonic clock when the wakeup source's was activated last time.
+ * @event_count: Number of signaled wakeup events.
+ * @active_count: Number of times the wakeup sorce was activated.
+ * @relax_count: Number of times the wakeup sorce was deactivated.
+ * @hit_count: Number of times the wakeup sorce might abort system suspend.
+ * @active: Status of the wakeup source.
  */
-static inline void device_init_wakeup(struct device *dev, bool val)
+struct wakeup_source {
+	char 			*name;
+	struct list_head	entry;
+	spinlock_t		lock;
+	struct timer_list	timer;
+	unsigned long		timer_expires;
+	ktime_t total_time;
+	ktime_t max_time;
+	ktime_t last_time;
+	unsigned long		event_count;
+	unsigned long		active_count;
+	unsigned long		relax_count;
+	unsigned long		hit_count;
+	unsigned int		active:1;
+};
+
+#ifdef CONFIG_PM_SLEEP
+
+/*
+ * Changes to device_may_wakeup take effect on the next pm state change.
+ */
+
+static inline void device_set_wakeup_capable(struct device *dev, bool capable)
+{
+	dev->power.can_wakeup = capable;
+}
+
+static inline bool device_can_wakeup(struct device *dev)
+{
+	return dev->power.can_wakeup;
+}
+
+
+
+static inline bool device_may_wakeup(struct device *dev)
 {
-	dev->power.can_wakeup = dev->power.should_wakeup = val;
+	return dev->power.can_wakeup && !!dev->power.wakeup;
 }
 
+/* drivers/base/power/wakeup.c */
+extern struct wakeup_source *wakeup_source_create(const char *name);
+extern void wakeup_source_destroy(struct wakeup_source *ws);
+extern void wakeup_source_add(struct wakeup_source *ws);
+extern void wakeup_source_remove(struct wakeup_source *ws);
+extern struct wakeup_source *wakeup_source_register(const char *name);
+extern void wakeup_source_unregister(struct wakeup_source *ws);
+extern int device_wakeup_enable(struct device *dev);
+extern int device_wakeup_disable(struct device *dev);
+extern int device_init_wakeup(struct device *dev, bool val);
+extern int device_set_wakeup_enable(struct device *dev, bool enable);
+extern void __pm_stay_awake(struct wakeup_source *ws);
+extern void pm_stay_awake(struct device *dev);
+extern void __pm_relax(struct wakeup_source *ws);
+extern void pm_relax(struct device *dev);
+extern void __pm_wakeup_event(struct wakeup_source *ws, unsigned int msec);
+extern void pm_wakeup_event(struct device *dev, unsigned int msec);
+
+#else /* !CONFIG_PM_SLEEP */
+
 static inline void device_set_wakeup_capable(struct device *dev, bool capable)
 {
 	dev->power.can_wakeup = capable;
@@ -50,43 +109,63 @@ static inline bool device_can_wakeup(struct device *dev)
 	return dev->power.can_wakeup;
 }
 
-static inline void device_set_wakeup_enable(struct device *dev, bool enable)
+static inline bool device_may_wakeup(struct device *dev)
 {
-	dev->power.should_wakeup = enable;
+	return false;
 }
 
-static inline bool device_may_wakeup(struct device *dev)
+static inline struct wakeup_source *wakeup_source_create(const char *name)
 {
-	return dev->power.can_wakeup && dev->power.should_wakeup;
+	return NULL;
 }
 
-#else /* !CONFIG_PM */
+static inline void wakeup_source_destroy(struct wakeup_source *ws) {}
+
+static inline void wakeup_source_add(struct wakeup_source *ws) {}
 
-/* For some reason the following routines work even without CONFIG_PM */
-static inline void device_init_wakeup(struct device *dev, bool val)
+static inline void wakeup_source_remove(struct wakeup_source *ws) {}
+
+static inline struct wakeup_source *wakeup_source_register(const char *name)
 {
-	dev->power.can_wakeup = val;
+	return NULL;
 }
 
-static inline void device_set_wakeup_capable(struct device *dev, bool capable)
+static inline void wakeup_source_unregister(struct wakeup_source *ws) {}
+
+static inline int device_wakeup_enable(struct device *dev)
 {
-	dev->power.can_wakeup = capable;
+	return -EINVAL;
 }
 
-static inline bool device_can_wakeup(struct device *dev)
+static inline int device_wakeup_disable(struct device *dev)
 {
-	return dev->power.can_wakeup;
+	return 0;
 }
 
-static inline void device_set_wakeup_enable(struct device *dev, bool enable)
+static inline int device_init_wakeup(struct device *dev, bool val)
 {
+	dev->power.can_wakeup = val;
+	return val ? -EINVAL : 0;
 }
 
-static inline bool device_may_wakeup(struct device *dev)
+
+static inline int device_set_wakeup_enable(struct device *dev, bool enable)
 {
-	return false;
+	return -EINVAL;
 }
 
-#endif /* !CONFIG_PM */
+static inline void __pm_stay_awake(struct wakeup_source *ws) {}
+
+static inline void pm_stay_awake(struct device *dev) {}
+
+static inline void __pm_relax(struct wakeup_source *ws) {}
+
+static inline void pm_relax(struct device *dev) {}
+
+static inline void __pm_wakeup_event(struct wakeup_source *ws, unsigned int msec) {}
+
+static inline void pm_wakeup_event(struct device *dev, unsigned int msec) {}
+
+#endif /* !CONFIG_PM_SLEEP */
 
 #endif /* _LINUX_PM_WAKEUP_H */
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 4af270ec2204..6b1712c51102 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -293,8 +293,8 @@ extern int unregister_pm_notifier(struct notifier_block *nb);
 extern bool events_check_enabled;
 
 extern bool pm_check_wakeup_events(void);
-extern bool pm_get_wakeup_count(unsigned long *count);
-extern bool pm_save_wakeup_count(unsigned long count);
+extern bool pm_get_wakeup_count(unsigned int *count);
+extern bool pm_save_wakeup_count(unsigned int count);
 #else /* !CONFIG_PM_SLEEP */
 
 static inline int register_pm_notifier(struct notifier_block *nb)
diff --git a/kernel/power/main.c b/kernel/power/main.c
index f06ad6eff37a..6b12a0cf4d9f 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -237,18 +237,18 @@ static ssize_t wakeup_count_show(struct kobject *kobj,
 				struct kobj_attribute *attr,
 				char *buf)
 {
-	unsigned long val;
+	unsigned int val;
 
-	return pm_get_wakeup_count(&val) ? sprintf(buf, "%lu\n", val) : -EINTR;
+	return pm_get_wakeup_count(&val) ? sprintf(buf, "%u\n", val) : -EINTR;
 }
 
 static ssize_t wakeup_count_store(struct kobject *kobj,
 				struct kobj_attribute *attr,
 				const char *buf, size_t n)
 {
-	unsigned long val;
+	unsigned int val;
 
-	if (sscanf(buf, "%lu", &val) == 1) {
+	if (sscanf(buf, "%u", &val) == 1) {
 		if (pm_save_wakeup_count(val))
 			return n;
 	}
-- 
cgit v1.2.3


From 098dff738abbeaea15fc95c4f4fdaee1e9bbea75 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 22 Sep 2010 22:10:57 +0200
Subject: PM: Fix potential issue with failing asynchronous suspend

There is a potential issue with the asynchronous suspend code that
a device driver suspending asynchronously may not notice that it
should back off.  There are two failing scenarions, (1) when the
driver is waiting for a driver suspending synchronously to complete
and that second driver returns error code, in which case async_error
won't be set and the waiting driver will continue suspending and (2)
after the driver has called device_pm_wait_for_dev() and the waited
for driver returns error code, in which case the caller of
device_pm_wait_for_dev() will not know that there was an error and
will continue suspending.

To fix this issue make __device_suspend() set async_error, so
async_suspend() doesn't need to set it any more, and make
device_pm_wait_for_dev() return async_error, so that its callers
can check whether or not they should continue suspending.

No more changes are necessary, since device_pm_wait_for_dev() is
not used by any drivers' suspend routines.

Reported-by: Colin Cross <ccross@android.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/main.c | 15 +++++++++------
 include/linux/pm.h        |  7 +++++--
 2 files changed, 14 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 7ae6fe414c38..31b526661ec4 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -51,6 +51,8 @@ static pm_message_t pm_transition;
  */
 static bool transition_started;
 
+static int async_error;
+
 /**
  * device_pm_init - Initialize the PM-related part of a device object.
  * @dev: Device object being initialized.
@@ -602,6 +604,7 @@ static void dpm_resume(pm_message_t state)
 	INIT_LIST_HEAD(&list);
 	mutex_lock(&dpm_list_mtx);
 	pm_transition = state;
+	async_error = 0;
 
 	list_for_each_entry(dev, &dpm_list, power.entry) {
 		if (dev->power.status < DPM_OFF)
@@ -831,8 +834,6 @@ static int legacy_suspend(struct device *dev, pm_message_t state,
 	return error;
 }
 
-static int async_error;
-
 /**
  * device_suspend - Execute "suspend" callbacks for given device.
  * @dev: Device to handle.
@@ -887,6 +888,9 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async)
 	device_unlock(dev);
 	complete_all(&dev->power.completion);
 
+	if (error)
+		async_error = error;
+
 	return error;
 }
 
@@ -896,10 +900,8 @@ static void async_suspend(void *data, async_cookie_t cookie)
 	int error;
 
 	error = __device_suspend(dev, pm_transition, true);
-	if (error) {
+	if (error)
 		pm_dev_err(dev, pm_transition, " async", error);
-		async_error = error;
-	}
 
 	put_device(dev);
 }
@@ -1087,8 +1089,9 @@ EXPORT_SYMBOL_GPL(__suspend_report_result);
  * @dev: Device to wait for.
  * @subordinate: Device that needs to wait for @dev.
  */
-void device_pm_wait_for_dev(struct device *subordinate, struct device *dev)
+int device_pm_wait_for_dev(struct device *subordinate, struct device *dev)
 {
 	dpm_wait(dev, subordinate->power.async_suspend);
+	return async_error;
 }
 EXPORT_SYMBOL_GPL(device_pm_wait_for_dev);
diff --git a/include/linux/pm.h b/include/linux/pm.h
index a84118911ced..1abfe84f447d 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -559,7 +559,7 @@ extern void __suspend_report_result(const char *function, void *fn, int ret);
 		__suspend_report_result(__func__, fn, ret);		\
 	} while (0)
 
-extern void device_pm_wait_for_dev(struct device *sub, struct device *dev);
+extern int device_pm_wait_for_dev(struct device *sub, struct device *dev);
 #else /* !CONFIG_PM_SLEEP */
 
 #define device_pm_lock() do {} while (0)
@@ -572,7 +572,10 @@ static inline int dpm_suspend_start(pm_message_t state)
 
 #define suspend_report_result(fn, ret)		do {} while (0)
 
-static inline void device_pm_wait_for_dev(struct device *a, struct device *b) {}
+static inline int device_pm_wait_for_dev(struct device *a, struct device *b)
+{
+	return 0;
+}
 #endif /* !CONFIG_PM_SLEEP */
 
 /* How to reorder dpm_list after device_move() */
-- 
cgit v1.2.3


From 69d44ffbd772bede8c2a6d182e6e14f94826520b Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 25 Sep 2010 23:34:22 +0200
Subject: sysfs: Add sysfs_merge_group() and sysfs_unmerge_group()

This patch (as1420) adds sysfs_merge_group() and sysfs_unmerge_group()
functions, allowing drivers easily to add and remove sets of
attributes to a pre-existing attribute group directory.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 fs/sysfs/group.c      | 59 +++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/sysfs.h | 15 +++++++++++++
 2 files changed, 74 insertions(+)

(limited to 'include/linux')

diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c
index 23c1e598792a..442f34ff1af8 100644
--- a/fs/sysfs/group.c
+++ b/fs/sysfs/group.c
@@ -148,6 +148,65 @@ void sysfs_remove_group(struct kobject * kobj,
 	sysfs_put(sd);
 }
 
+/**
+ * sysfs_merge_group - merge files into a pre-existing attribute group.
+ * @kobj:	The kobject containing the group.
+ * @grp:	The files to create and the attribute group they belong to.
+ *
+ * This function returns an error if the group doesn't exist or any of the
+ * files already exist in that group, in which case none of the new files
+ * are created.
+ */
+int sysfs_merge_group(struct kobject *kobj,
+		       const struct attribute_group *grp)
+{
+	struct sysfs_dirent *dir_sd;
+	int error = 0;
+	struct attribute *const *attr;
+	int i;
+
+	if (grp)
+		dir_sd = sysfs_get_dirent(kobj->sd, NULL, grp->name);
+	else
+		dir_sd = sysfs_get(kobj->sd);
+	if (!dir_sd)
+		return -ENOENT;
+
+	for ((i = 0, attr = grp->attrs); *attr && !error; (++i, ++attr))
+		error = sysfs_add_file(dir_sd, *attr, SYSFS_KOBJ_ATTR);
+	if (error) {
+		while (--i >= 0)
+			sysfs_hash_and_remove(dir_sd, NULL, (*--attr)->name);
+	}
+	sysfs_put(dir_sd);
+
+	return error;
+}
+EXPORT_SYMBOL_GPL(sysfs_merge_group);
+
+/**
+ * sysfs_unmerge_group - remove files from a pre-existing attribute group.
+ * @kobj:	The kobject containing the group.
+ * @grp:	The files to remove and the attribute group they belong to.
+ */
+void sysfs_unmerge_group(struct kobject *kobj,
+		       const struct attribute_group *grp)
+{
+	struct sysfs_dirent *dir_sd;
+	struct attribute *const *attr;
+
+	if (grp)
+		dir_sd = sysfs_get_dirent(kobj->sd, NULL, grp->name);
+	else
+		dir_sd = sysfs_get(kobj->sd);
+	if (dir_sd) {
+		for (attr = grp->attrs; *attr; ++attr)
+			sysfs_hash_and_remove(dir_sd, NULL, (*attr)->name);
+		sysfs_put(dir_sd);
+	}
+}
+EXPORT_SYMBOL_GPL(sysfs_unmerge_group);
+
 
 EXPORT_SYMBOL_GPL(sysfs_create_group);
 EXPORT_SYMBOL_GPL(sysfs_update_group);
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 96eb576d82fd..30b881555fa5 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -164,6 +164,10 @@ int sysfs_add_file_to_group(struct kobject *kobj,
 			const struct attribute *attr, const char *group);
 void sysfs_remove_file_from_group(struct kobject *kobj,
 			const struct attribute *attr, const char *group);
+int sysfs_merge_group(struct kobject *kobj,
+		       const struct attribute_group *grp);
+void sysfs_unmerge_group(struct kobject *kobj,
+		       const struct attribute_group *grp);
 
 void sysfs_notify(struct kobject *kobj, const char *dir, const char *attr);
 void sysfs_notify_dirent(struct sysfs_dirent *sd);
@@ -302,6 +306,17 @@ static inline void sysfs_remove_file_from_group(struct kobject *kobj,
 {
 }
 
+static inline int sysfs_merge_group(struct kobject *kobj,
+		       const struct attribute_group *grp)
+{
+	return 0;
+}
+
+static inline void sysfs_unmerge_group(struct kobject *kobj,
+		       const struct attribute_group *grp)
+{
+}
+
 static inline void sysfs_notify(struct kobject *kobj, const char *dir,
 				const char *attr)
 {
-- 
cgit v1.2.3


From 3f9af0513ae5b1f185302c2d0ba656640926d970 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 25 Sep 2010 23:34:54 +0200
Subject: PM / Runtime: Replace boolean arguments with bitflags

The "from_wq" argument in __pm_runtime_suspend() and
__pm_runtime_resume() supposedly indicates whether or not the function
was called by the PM workqueue thread, but in fact it isn't always
used this way.  It really indicates whether or not the function should
return early if the requested operation is already in progress.

Along with this badly-named boolean argument, later patches in this
series will add several other boolean arguments to these functions and
others.  Therefore this patch (as1422) begins the conversion process
by replacing from_wq with a bitflag argument.  The same bitflags are
also used in __pm_runtime_get() and __pm_runtime_put(), where they
indicate whether or not the operation should be asynchronous.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 drivers/base/power/runtime.c | 75 +++++++++++++++++++++++---------------------
 include/linux/pm_runtime.h   | 23 +++++++++-----
 2 files changed, 54 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c
index ec08f1ae63f1..0c1db879544b 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -10,7 +10,7 @@
 #include <linux/pm_runtime.h>
 #include <linux/jiffies.h>
 
-static int __pm_runtime_resume(struct device *dev, bool from_wq);
+static int __pm_runtime_resume(struct device *dev, int rpmflags);
 static int __pm_request_idle(struct device *dev);
 static int __pm_request_resume(struct device *dev);
 
@@ -164,24 +164,24 @@ EXPORT_SYMBOL_GPL(pm_runtime_idle);
 /**
  * __pm_runtime_suspend - Carry out run-time suspend of given device.
  * @dev: Device to suspend.
- * @from_wq: If set, the function has been called via pm_wq.
+ * @rpmflags: Flag bits.
  *
  * Check if the device can be suspended and run the ->runtime_suspend() callback
- * provided by its bus type.  If another suspend has been started earlier, wait
- * for it to finish.  If an idle notification or suspend request is pending or
+ * provided by its bus type.  If another suspend has been started earlier,
+ * either return immediately or wait for it to finish, depending on the
+ * RPM_NOWAIT flag.  If an idle notification or suspend request is pending or
  * scheduled, cancel it.
  *
  * This function must be called under dev->power.lock with interrupts disabled.
  */
-int __pm_runtime_suspend(struct device *dev, bool from_wq)
+static int __pm_runtime_suspend(struct device *dev, int rpmflags)
 	__releases(&dev->power.lock) __acquires(&dev->power.lock)
 {
 	struct device *parent = NULL;
 	bool notify = false;
 	int retval = 0;
 
-	dev_dbg(dev, "__pm_runtime_suspend()%s!\n",
-		from_wq ? " from workqueue" : "");
+	dev_dbg(dev, "%s flags 0x%x\n", __func__, rpmflags);
 
  repeat:
 	if (dev->power.runtime_error) {
@@ -213,7 +213,7 @@ int __pm_runtime_suspend(struct device *dev, bool from_wq)
 	if (dev->power.runtime_status == RPM_SUSPENDING) {
 		DEFINE_WAIT(wait);
 
-		if (from_wq) {
+		if (rpmflags & RPM_NOWAIT) {
 			retval = -EINPROGRESS;
 			goto out;
 		}
@@ -286,7 +286,7 @@ int __pm_runtime_suspend(struct device *dev, bool from_wq)
 	wake_up_all(&dev->power.wait_queue);
 
 	if (dev->power.deferred_resume) {
-		__pm_runtime_resume(dev, false);
+		__pm_runtime_resume(dev, 0);
 		retval = -EAGAIN;
 		goto out;
 	}
@@ -303,7 +303,7 @@ int __pm_runtime_suspend(struct device *dev, bool from_wq)
 	}
 
  out:
-	dev_dbg(dev, "__pm_runtime_suspend() returns %d!\n", retval);
+	dev_dbg(dev, "%s returns %d\n", __func__, retval);
 
 	return retval;
 }
@@ -317,7 +317,7 @@ int pm_runtime_suspend(struct device *dev)
 	int retval;
 
 	spin_lock_irq(&dev->power.lock);
-	retval = __pm_runtime_suspend(dev, false);
+	retval = __pm_runtime_suspend(dev, 0);
 	spin_unlock_irq(&dev->power.lock);
 
 	return retval;
@@ -327,24 +327,25 @@ EXPORT_SYMBOL_GPL(pm_runtime_suspend);
 /**
  * __pm_runtime_resume - Carry out run-time resume of given device.
  * @dev: Device to resume.
- * @from_wq: If set, the function has been called via pm_wq.
+ * @rpmflags: Flag bits.
  *
  * Check if the device can be woken up and run the ->runtime_resume() callback
- * provided by its bus type.  If another resume has been started earlier, wait
- * for it to finish.  If there's a suspend running in parallel with this
- * function, wait for it to finish and resume the device.  Cancel any scheduled
- * or pending requests.
+ * provided by its bus type.  If another resume has been started earlier,
+ * either return imediately or wait for it to finish, depending on the
+ * RPM_NOWAIT flag.  If there's a suspend running in parallel with this
+ * function, either tell the other process to resume after suspending
+ * (deferred_resume) or wait for it to finish, depending on the RPM_NOWAIT
+ * flag.  Cancel any scheduled or pending requests.
  *
  * This function must be called under dev->power.lock with interrupts disabled.
  */
-int __pm_runtime_resume(struct device *dev, bool from_wq)
+static int __pm_runtime_resume(struct device *dev, int rpmflags)
 	__releases(&dev->power.lock) __acquires(&dev->power.lock)
 {
 	struct device *parent = NULL;
 	int retval = 0;
 
-	dev_dbg(dev, "__pm_runtime_resume()%s!\n",
-		from_wq ? " from workqueue" : "");
+	dev_dbg(dev, "%s flags 0x%x\n", __func__, rpmflags);
 
  repeat:
 	if (dev->power.runtime_error) {
@@ -365,7 +366,7 @@ int __pm_runtime_resume(struct device *dev, bool from_wq)
 	    || dev->power.runtime_status == RPM_SUSPENDING) {
 		DEFINE_WAIT(wait);
 
-		if (from_wq) {
+		if (rpmflags & RPM_NOWAIT) {
 			if (dev->power.runtime_status == RPM_SUSPENDING)
 				dev->power.deferred_resume = true;
 			retval = -EINPROGRESS;
@@ -407,7 +408,7 @@ int __pm_runtime_resume(struct device *dev, bool from_wq)
 		 */
 		if (!parent->power.disable_depth
 		    && !parent->power.ignore_children) {
-			__pm_runtime_resume(parent, false);
+			__pm_runtime_resume(parent, 0);
 			if (parent->power.runtime_status != RPM_ACTIVE)
 				retval = -EBUSY;
 		}
@@ -470,7 +471,7 @@ int __pm_runtime_resume(struct device *dev, bool from_wq)
 		spin_lock_irq(&dev->power.lock);
 	}
 
-	dev_dbg(dev, "__pm_runtime_resume() returns %d!\n", retval);
+	dev_dbg(dev, "%s returns %d\n", __func__, retval);
 
 	return retval;
 }
@@ -484,7 +485,7 @@ int pm_runtime_resume(struct device *dev)
 	int retval;
 
 	spin_lock_irq(&dev->power.lock);
-	retval = __pm_runtime_resume(dev, false);
+	retval = __pm_runtime_resume(dev, 0);
 	spin_unlock_irq(&dev->power.lock);
 
 	return retval;
@@ -519,10 +520,10 @@ static void pm_runtime_work(struct work_struct *work)
 		__pm_runtime_idle(dev);
 		break;
 	case RPM_REQ_SUSPEND:
-		__pm_runtime_suspend(dev, true);
+		__pm_runtime_suspend(dev, RPM_NOWAIT);
 		break;
 	case RPM_REQ_RESUME:
-		__pm_runtime_resume(dev, true);
+		__pm_runtime_resume(dev, RPM_NOWAIT);
 		break;
 	}
 
@@ -782,17 +783,18 @@ EXPORT_SYMBOL_GPL(pm_request_resume);
 /**
  * __pm_runtime_get - Reference count a device and wake it up, if necessary.
  * @dev: Device to handle.
- * @sync: If set and the device is suspended, resume it synchronously.
+ * @rpmflags: Flag bits.
  *
  * Increment the usage count of the device and resume it or submit a resume
- * request for it, depending on the value of @sync.
+ * request for it, depending on the RPM_ASYNC flag bit.
  */
-int __pm_runtime_get(struct device *dev, bool sync)
+int __pm_runtime_get(struct device *dev, int rpmflags)
 {
 	int retval;
 
 	atomic_inc(&dev->power.usage_count);
-	retval = sync ? pm_runtime_resume(dev) : pm_request_resume(dev);
+	retval = (rpmflags & RPM_ASYNC) ?
+	    pm_request_resume(dev) : pm_runtime_resume(dev);
 
 	return retval;
 }
@@ -801,18 +803,19 @@ EXPORT_SYMBOL_GPL(__pm_runtime_get);
 /**
  * __pm_runtime_put - Decrement the device's usage counter and notify its bus.
  * @dev: Device to handle.
- * @sync: If the device's bus type is to be notified, do that synchronously.
+ * @rpmflags: Flag bits.
  *
  * Decrement the usage count of the device and if it reaches zero, carry out a
  * synchronous idle notification or submit an idle notification request for it,
- * depending on the value of @sync.
+ * depending on the RPM_ASYNC flag bit.
  */
-int __pm_runtime_put(struct device *dev, bool sync)
+int __pm_runtime_put(struct device *dev, int rpmflags)
 {
 	int retval = 0;
 
 	if (atomic_dec_and_test(&dev->power.usage_count))
-		retval = sync ? pm_runtime_idle(dev) : pm_request_idle(dev);
+		retval = (rpmflags & RPM_ASYNC) ?
+		    pm_request_idle(dev) : pm_runtime_idle(dev);
 
 	return retval;
 }
@@ -967,7 +970,7 @@ int pm_runtime_barrier(struct device *dev)
 
 	if (dev->power.request_pending
 	    && dev->power.request == RPM_REQ_RESUME) {
-		__pm_runtime_resume(dev, false);
+		__pm_runtime_resume(dev, 0);
 		retval = 1;
 	}
 
@@ -1016,7 +1019,7 @@ void __pm_runtime_disable(struct device *dev, bool check_resume)
 		 */
 		pm_runtime_get_noresume(dev);
 
-		__pm_runtime_resume(dev, false);
+		__pm_runtime_resume(dev, 0);
 
 		pm_runtime_put_noidle(dev);
 	}
@@ -1064,7 +1067,7 @@ void pm_runtime_forbid(struct device *dev)
 
 	dev->power.runtime_auto = false;
 	atomic_inc(&dev->power.usage_count);
-	__pm_runtime_resume(dev, false);
+	__pm_runtime_resume(dev, 0);
 
  out:
 	spin_unlock_irq(&dev->power.lock);
diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h
index 6e81888c6222..c030cac59aac 100644
--- a/include/linux/pm_runtime.h
+++ b/include/linux/pm_runtime.h
@@ -12,6 +12,11 @@
 #include <linux/device.h>
 #include <linux/pm.h>
 
+/* Runtime PM flag argument bits */
+#define RPM_ASYNC		0x01	/* Request is asynchronous */
+#define RPM_NOWAIT		0x02	/* Don't wait for concurrent
+					    state change */
+
 #ifdef CONFIG_PM_RUNTIME
 
 extern struct workqueue_struct *pm_wq;
@@ -22,8 +27,8 @@ extern int pm_runtime_resume(struct device *dev);
 extern int pm_request_idle(struct device *dev);
 extern int pm_schedule_suspend(struct device *dev, unsigned int delay);
 extern int pm_request_resume(struct device *dev);
-extern int __pm_runtime_get(struct device *dev, bool sync);
-extern int __pm_runtime_put(struct device *dev, bool sync);
+extern int __pm_runtime_get(struct device *dev, int rpmflags);
+extern int __pm_runtime_put(struct device *dev, int rpmflags);
 extern int __pm_runtime_set_status(struct device *dev, unsigned int status);
 extern int pm_runtime_barrier(struct device *dev);
 extern void pm_runtime_enable(struct device *dev);
@@ -81,8 +86,10 @@ static inline int pm_schedule_suspend(struct device *dev, unsigned int delay)
 	return -ENOSYS;
 }
 static inline int pm_request_resume(struct device *dev) { return 0; }
-static inline int __pm_runtime_get(struct device *dev, bool sync) { return 1; }
-static inline int __pm_runtime_put(struct device *dev, bool sync) { return 0; }
+static inline int __pm_runtime_get(struct device *dev, int rpmflags)
+					{ return 1; }
+static inline int __pm_runtime_put(struct device *dev, int rpmflags)
+					{ return 0; }
 static inline int __pm_runtime_set_status(struct device *dev,
 					    unsigned int status) { return 0; }
 static inline int pm_runtime_barrier(struct device *dev) { return 0; }
@@ -107,22 +114,22 @@ static inline int pm_generic_runtime_resume(struct device *dev) { return 0; }
 
 static inline int pm_runtime_get(struct device *dev)
 {
-	return __pm_runtime_get(dev, false);
+	return __pm_runtime_get(dev, RPM_ASYNC);
 }
 
 static inline int pm_runtime_get_sync(struct device *dev)
 {
-	return __pm_runtime_get(dev, true);
+	return __pm_runtime_get(dev, 0);
 }
 
 static inline int pm_runtime_put(struct device *dev)
 {
-	return __pm_runtime_put(dev, false);
+	return __pm_runtime_put(dev, RPM_ASYNC);
 }
 
 static inline int pm_runtime_put_sync(struct device *dev)
 {
-	return __pm_runtime_put(dev, true);
+	return __pm_runtime_put(dev, 0);
 }
 
 static inline int pm_runtime_set_active(struct device *dev)
-- 
cgit v1.2.3


From 140a6c945211ee911dec776fafa52e03a7d7bb9a Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 25 Sep 2010 23:35:07 +0200
Subject: PM / Runtime: Combine runtime PM entry points

This patch (as1424) combines the various public entry points for the
runtime PM routines into three simple functions: one for idle, one for
suspend, and one for resume.  A new bitflag specifies whether or not
to increment or decrement the usage_count field.

The new entry points are named __pm_runtime_idle,
__pm_runtime_suspend, and __pm_runtime_resume, to reflect that they
are trampolines.  Simultaneously, the corresponding internal routines
are renamed to rpm_idle, rpm_suspend, and rpm_resume.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 drivers/base/power/runtime.c | 174 ++++++++++++++++---------------------------
 include/linux/pm_runtime.h   |  66 +++++++++++-----
 2 files changed, 110 insertions(+), 130 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c
index d7b5d84c235c..ed227b7c1bb5 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -11,7 +11,7 @@
 #include <linux/pm_runtime.h>
 #include <linux/jiffies.h>
 
-static int __pm_runtime_resume(struct device *dev, int rpmflags);
+static int rpm_resume(struct device *dev, int rpmflags);
 
 /**
  * update_pm_runtime_accounting - Update the time accounting of power states
@@ -107,7 +107,7 @@ static int rpm_check_suspend_allowed(struct device *dev)
 
 
 /**
- * __pm_runtime_idle - Notify device bus type if the device can be suspended.
+ * rpm_idle - Notify device bus type if the device can be suspended.
  * @dev: Device to notify the bus type about.
  * @rpmflags: Flag bits.
  *
@@ -118,7 +118,7 @@ static int rpm_check_suspend_allowed(struct device *dev)
  *
  * This function must be called under dev->power.lock with interrupts disabled.
  */
-static int __pm_runtime_idle(struct device *dev, int rpmflags)
+static int rpm_idle(struct device *dev, int rpmflags)
 	__releases(&dev->power.lock) __acquires(&dev->power.lock)
 {
 	int retval;
@@ -189,23 +189,7 @@ static int __pm_runtime_idle(struct device *dev, int rpmflags)
 }
 
 /**
- * pm_runtime_idle - Notify device bus type if the device can be suspended.
- * @dev: Device to notify the bus type about.
- */
-int pm_runtime_idle(struct device *dev)
-{
-	int retval;
-
-	spin_lock_irq(&dev->power.lock);
-	retval = __pm_runtime_idle(dev, 0);
-	spin_unlock_irq(&dev->power.lock);
-
-	return retval;
-}
-EXPORT_SYMBOL_GPL(pm_runtime_idle);
-
-/**
- * __pm_runtime_suspend - Carry out run-time suspend of given device.
+ * rpm_suspend - Carry out run-time suspend of given device.
  * @dev: Device to suspend.
  * @rpmflags: Flag bits.
  *
@@ -220,7 +204,7 @@ EXPORT_SYMBOL_GPL(pm_runtime_idle);
  *
  * This function must be called under dev->power.lock with interrupts disabled.
  */
-static int __pm_runtime_suspend(struct device *dev, int rpmflags)
+static int rpm_suspend(struct device *dev, int rpmflags)
 	__releases(&dev->power.lock) __acquires(&dev->power.lock)
 {
 	struct device *parent = NULL;
@@ -332,13 +316,13 @@ static int __pm_runtime_suspend(struct device *dev, int rpmflags)
 	wake_up_all(&dev->power.wait_queue);
 
 	if (dev->power.deferred_resume) {
-		__pm_runtime_resume(dev, 0);
+		rpm_resume(dev, 0);
 		retval = -EAGAIN;
 		goto out;
 	}
 
 	if (notify)
-		__pm_runtime_idle(dev, 0);
+		rpm_idle(dev, 0);
 
 	if (parent && !parent->power.ignore_children) {
 		spin_unlock_irq(&dev->power.lock);
@@ -355,23 +339,7 @@ static int __pm_runtime_suspend(struct device *dev, int rpmflags)
 }
 
 /**
- * pm_runtime_suspend - Carry out run-time suspend of given device.
- * @dev: Device to suspend.
- */
-int pm_runtime_suspend(struct device *dev)
-{
-	int retval;
-
-	spin_lock_irq(&dev->power.lock);
-	retval = __pm_runtime_suspend(dev, 0);
-	spin_unlock_irq(&dev->power.lock);
-
-	return retval;
-}
-EXPORT_SYMBOL_GPL(pm_runtime_suspend);
-
-/**
- * __pm_runtime_resume - Carry out run-time resume of given device.
+ * rpm_resume - Carry out run-time resume of given device.
  * @dev: Device to resume.
  * @rpmflags: Flag bits.
  *
@@ -387,7 +355,7 @@ EXPORT_SYMBOL_GPL(pm_runtime_suspend);
  *
  * This function must be called under dev->power.lock with interrupts disabled.
  */
-static int __pm_runtime_resume(struct device *dev, int rpmflags)
+static int rpm_resume(struct device *dev, int rpmflags)
 	__releases(&dev->power.lock) __acquires(&dev->power.lock)
 {
 	struct device *parent = NULL;
@@ -469,7 +437,7 @@ static int __pm_runtime_resume(struct device *dev, int rpmflags)
 		 */
 		if (!parent->power.disable_depth
 		    && !parent->power.ignore_children) {
-			__pm_runtime_resume(parent, 0);
+			rpm_resume(parent, 0);
 			if (parent->power.runtime_status != RPM_ACTIVE)
 				retval = -EBUSY;
 		}
@@ -521,7 +489,7 @@ static int __pm_runtime_resume(struct device *dev, int rpmflags)
 	wake_up_all(&dev->power.wait_queue);
 
 	if (!retval)
-		__pm_runtime_idle(dev, RPM_ASYNC);
+		rpm_idle(dev, RPM_ASYNC);
 
  out:
 	if (parent) {
@@ -537,22 +505,6 @@ static int __pm_runtime_resume(struct device *dev, int rpmflags)
 	return retval;
 }
 
-/**
- * pm_runtime_resume - Carry out run-time resume of given device.
- * @dev: Device to suspend.
- */
-int pm_runtime_resume(struct device *dev)
-{
-	int retval;
-
-	spin_lock_irq(&dev->power.lock);
-	retval = __pm_runtime_resume(dev, 0);
-	spin_unlock_irq(&dev->power.lock);
-
-	return retval;
-}
-EXPORT_SYMBOL_GPL(pm_runtime_resume);
-
 /**
  * pm_runtime_work - Universal run-time PM work function.
  * @work: Work structure used for scheduling the execution of this function.
@@ -578,13 +530,13 @@ static void pm_runtime_work(struct work_struct *work)
 	case RPM_REQ_NONE:
 		break;
 	case RPM_REQ_IDLE:
-		__pm_runtime_idle(dev, RPM_NOWAIT);
+		rpm_idle(dev, RPM_NOWAIT);
 		break;
 	case RPM_REQ_SUSPEND:
-		__pm_runtime_suspend(dev, RPM_NOWAIT);
+		rpm_suspend(dev, RPM_NOWAIT);
 		break;
 	case RPM_REQ_RESUME:
-		__pm_runtime_resume(dev, RPM_NOWAIT);
+		rpm_resume(dev, RPM_NOWAIT);
 		break;
 	}
 
@@ -592,23 +544,6 @@ static void pm_runtime_work(struct work_struct *work)
 	spin_unlock_irq(&dev->power.lock);
 }
 
-/**
- * pm_request_idle - Submit an idle notification request for given device.
- * @dev: Device to handle.
- */
-int pm_request_idle(struct device *dev)
-{
-	unsigned long flags;
-	int retval;
-
-	spin_lock_irqsave(&dev->power.lock, flags);
-	retval = __pm_runtime_idle(dev, RPM_ASYNC);
-	spin_unlock_irqrestore(&dev->power.lock, flags);
-
-	return retval;
-}
-EXPORT_SYMBOL_GPL(pm_request_idle);
-
 /**
  * pm_suspend_timer_fn - Timer function for pm_schedule_suspend().
  * @data: Device pointer passed by pm_schedule_suspend().
@@ -627,7 +562,7 @@ static void pm_suspend_timer_fn(unsigned long data)
 	/* If 'expire' is after 'jiffies' we've been called too early. */
 	if (expires > 0 && !time_after(expires, jiffies)) {
 		dev->power.timer_expires = 0;
-		__pm_runtime_suspend(dev, RPM_ASYNC);
+		rpm_suspend(dev, RPM_ASYNC);
 	}
 
 	spin_unlock_irqrestore(&dev->power.lock, flags);
@@ -646,7 +581,7 @@ int pm_schedule_suspend(struct device *dev, unsigned int delay)
 	spin_lock_irqsave(&dev->power.lock, flags);
 
 	if (!delay) {
-		retval = __pm_runtime_suspend(dev, RPM_ASYNC);
+		retval = rpm_suspend(dev, RPM_ASYNC);
 		goto out;
 	}
 
@@ -669,62 +604,81 @@ int pm_schedule_suspend(struct device *dev, unsigned int delay)
 EXPORT_SYMBOL_GPL(pm_schedule_suspend);
 
 /**
- * pm_request_resume - Submit a resume request for given device.
- * @dev: Device to resume.
+ * __pm_runtime_idle - Entry point for run-time idle operations.
+ * @dev: Device to send idle notification for.
+ * @rpmflags: Flag bits.
+ *
+ * If the RPM_GET_PUT flag is set, decrement the device's usage count and
+ * return immediately if it is larger than zero.  Then carry out an idle
+ * notification, either synchronous or asynchronous.
+ *
+ * This routine may be called in atomic context if the RPM_ASYNC flag is set.
  */
-int pm_request_resume(struct device *dev)
+int __pm_runtime_idle(struct device *dev, int rpmflags)
 {
 	unsigned long flags;
 	int retval;
 
+	if (rpmflags & RPM_GET_PUT) {
+		if (!atomic_dec_and_test(&dev->power.usage_count))
+			return 0;
+	}
+
 	spin_lock_irqsave(&dev->power.lock, flags);
-	retval = __pm_runtime_resume(dev, RPM_ASYNC);
+	retval = rpm_idle(dev, rpmflags);
 	spin_unlock_irqrestore(&dev->power.lock, flags);
 
 	return retval;
 }
-EXPORT_SYMBOL_GPL(pm_request_resume);
+EXPORT_SYMBOL_GPL(__pm_runtime_idle);
 
 /**
- * __pm_runtime_get - Reference count a device and wake it up, if necessary.
- * @dev: Device to handle.
+ * __pm_runtime_suspend - Entry point for run-time put/suspend operations.
+ * @dev: Device to suspend.
  * @rpmflags: Flag bits.
  *
- * Increment the usage count of the device and resume it or submit a resume
- * request for it, depending on the RPM_ASYNC flag bit.
+ * Carry out a suspend, either synchronous or asynchronous.
+ *
+ * This routine may be called in atomic context if the RPM_ASYNC flag is set.
  */
-int __pm_runtime_get(struct device *dev, int rpmflags)
+int __pm_runtime_suspend(struct device *dev, int rpmflags)
 {
+	unsigned long flags;
 	int retval;
 
-	atomic_inc(&dev->power.usage_count);
-	retval = (rpmflags & RPM_ASYNC) ?
-	    pm_request_resume(dev) : pm_runtime_resume(dev);
+	spin_lock_irqsave(&dev->power.lock, flags);
+	retval = rpm_suspend(dev, rpmflags);
+	spin_unlock_irqrestore(&dev->power.lock, flags);
 
 	return retval;
 }
-EXPORT_SYMBOL_GPL(__pm_runtime_get);
+EXPORT_SYMBOL_GPL(__pm_runtime_suspend);
 
 /**
- * __pm_runtime_put - Decrement the device's usage counter and notify its bus.
- * @dev: Device to handle.
+ * __pm_runtime_resume - Entry point for run-time resume operations.
+ * @dev: Device to resume.
  * @rpmflags: Flag bits.
  *
- * Decrement the usage count of the device and if it reaches zero, carry out a
- * synchronous idle notification or submit an idle notification request for it,
- * depending on the RPM_ASYNC flag bit.
+ * If the RPM_GET_PUT flag is set, increment the device's usage count.  Then
+ * carry out a resume, either synchronous or asynchronous.
+ *
+ * This routine may be called in atomic context if the RPM_ASYNC flag is set.
  */
-int __pm_runtime_put(struct device *dev, int rpmflags)
+int __pm_runtime_resume(struct device *dev, int rpmflags)
 {
-	int retval = 0;
+	unsigned long flags;
+	int retval;
 
-	if (atomic_dec_and_test(&dev->power.usage_count))
-		retval = (rpmflags & RPM_ASYNC) ?
-		    pm_request_idle(dev) : pm_runtime_idle(dev);
+	if (rpmflags & RPM_GET_PUT)
+		atomic_inc(&dev->power.usage_count);
+
+	spin_lock_irqsave(&dev->power.lock, flags);
+	retval = rpm_resume(dev, rpmflags);
+	spin_unlock_irqrestore(&dev->power.lock, flags);
 
 	return retval;
 }
-EXPORT_SYMBOL_GPL(__pm_runtime_put);
+EXPORT_SYMBOL_GPL(__pm_runtime_resume);
 
 /**
  * __pm_runtime_set_status - Set run-time PM status of a device.
@@ -875,7 +829,7 @@ int pm_runtime_barrier(struct device *dev)
 
 	if (dev->power.request_pending
 	    && dev->power.request == RPM_REQ_RESUME) {
-		__pm_runtime_resume(dev, 0);
+		rpm_resume(dev, 0);
 		retval = 1;
 	}
 
@@ -924,7 +878,7 @@ void __pm_runtime_disable(struct device *dev, bool check_resume)
 		 */
 		pm_runtime_get_noresume(dev);
 
-		__pm_runtime_resume(dev, 0);
+		rpm_resume(dev, 0);
 
 		pm_runtime_put_noidle(dev);
 	}
@@ -972,7 +926,7 @@ void pm_runtime_forbid(struct device *dev)
 
 	dev->power.runtime_auto = false;
 	atomic_inc(&dev->power.usage_count);
-	__pm_runtime_resume(dev, 0);
+	rpm_resume(dev, 0);
 
  out:
 	spin_unlock_irq(&dev->power.lock);
@@ -993,7 +947,7 @@ void pm_runtime_allow(struct device *dev)
 
 	dev->power.runtime_auto = true;
 	if (atomic_dec_and_test(&dev->power.usage_count))
-		__pm_runtime_idle(dev, 0);
+		rpm_idle(dev, 0);
 
  out:
 	spin_unlock_irq(&dev->power.lock);
diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h
index c030cac59aac..5869d87fffac 100644
--- a/include/linux/pm_runtime.h
+++ b/include/linux/pm_runtime.h
@@ -16,19 +16,17 @@
 #define RPM_ASYNC		0x01	/* Request is asynchronous */
 #define RPM_NOWAIT		0x02	/* Don't wait for concurrent
 					    state change */
+#define RPM_GET_PUT		0x04	/* Increment/decrement the
+					    usage_count */
 
 #ifdef CONFIG_PM_RUNTIME
 
 extern struct workqueue_struct *pm_wq;
 
-extern int pm_runtime_idle(struct device *dev);
-extern int pm_runtime_suspend(struct device *dev);
-extern int pm_runtime_resume(struct device *dev);
-extern int pm_request_idle(struct device *dev);
+extern int __pm_runtime_idle(struct device *dev, int rpmflags);
+extern int __pm_runtime_suspend(struct device *dev, int rpmflags);
+extern int __pm_runtime_resume(struct device *dev, int rpmflags);
 extern int pm_schedule_suspend(struct device *dev, unsigned int delay);
-extern int pm_request_resume(struct device *dev);
-extern int __pm_runtime_get(struct device *dev, int rpmflags);
-extern int __pm_runtime_put(struct device *dev, int rpmflags);
 extern int __pm_runtime_set_status(struct device *dev, unsigned int status);
 extern int pm_runtime_barrier(struct device *dev);
 extern void pm_runtime_enable(struct device *dev);
@@ -77,19 +75,22 @@ static inline bool pm_runtime_suspended(struct device *dev)
 
 #else /* !CONFIG_PM_RUNTIME */
 
-static inline int pm_runtime_idle(struct device *dev) { return -ENOSYS; }
-static inline int pm_runtime_suspend(struct device *dev) { return -ENOSYS; }
-static inline int pm_runtime_resume(struct device *dev) { return 0; }
-static inline int pm_request_idle(struct device *dev) { return -ENOSYS; }
+static inline int __pm_runtime_idle(struct device *dev, int rpmflags)
+{
+	return -ENOSYS;
+}
+static inline int __pm_runtime_suspend(struct device *dev, int rpmflags)
+{
+	return -ENOSYS;
+}
+static inline int __pm_runtime_resume(struct device *dev, int rpmflags)
+{
+	return 1;
+}
 static inline int pm_schedule_suspend(struct device *dev, unsigned int delay)
 {
 	return -ENOSYS;
 }
-static inline int pm_request_resume(struct device *dev) { return 0; }
-static inline int __pm_runtime_get(struct device *dev, int rpmflags)
-					{ return 1; }
-static inline int __pm_runtime_put(struct device *dev, int rpmflags)
-					{ return 0; }
 static inline int __pm_runtime_set_status(struct device *dev,
 					    unsigned int status) { return 0; }
 static inline int pm_runtime_barrier(struct device *dev) { return 0; }
@@ -112,24 +113,49 @@ static inline int pm_generic_runtime_resume(struct device *dev) { return 0; }
 
 #endif /* !CONFIG_PM_RUNTIME */
 
+static inline int pm_runtime_idle(struct device *dev)
+{
+	return __pm_runtime_idle(dev, 0);
+}
+
+static inline int pm_runtime_suspend(struct device *dev)
+{
+	return __pm_runtime_suspend(dev, 0);
+}
+
+static inline int pm_runtime_resume(struct device *dev)
+{
+	return __pm_runtime_resume(dev, 0);
+}
+
+static inline int pm_request_idle(struct device *dev)
+{
+	return __pm_runtime_idle(dev, RPM_ASYNC);
+}
+
+static inline int pm_request_resume(struct device *dev)
+{
+	return __pm_runtime_resume(dev, RPM_ASYNC);
+}
+
 static inline int pm_runtime_get(struct device *dev)
 {
-	return __pm_runtime_get(dev, RPM_ASYNC);
+	return __pm_runtime_resume(dev, RPM_GET_PUT | RPM_ASYNC);
 }
 
 static inline int pm_runtime_get_sync(struct device *dev)
 {
-	return __pm_runtime_get(dev, 0);
+	return __pm_runtime_resume(dev, RPM_GET_PUT);
 }
 
 static inline int pm_runtime_put(struct device *dev)
 {
-	return __pm_runtime_put(dev, RPM_ASYNC);
+	return __pm_runtime_idle(dev, RPM_GET_PUT | RPM_ASYNC);
 }
 
 static inline int pm_runtime_put_sync(struct device *dev)
 {
-	return __pm_runtime_put(dev, 0);
+	return __pm_runtime_idle(dev, RPM_GET_PUT);
 }
 
 static inline int pm_runtime_set_active(struct device *dev)
-- 
cgit v1.2.3


From 7490e44239e60293bca0c2663229050c36c660c2 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 25 Sep 2010 23:35:15 +0200
Subject: PM / Runtime: Add no_callbacks flag

Some devices, such as USB interfaces, cannot be power-managed
independently of their parents, i.e., they cannot be put in low power
while the parent remains at full power.  This patch (as1425) creates a
new "no_callbacks" flag, which tells the PM core not to invoke the
runtime-PM callback routines for the such devices but instead to
assume that the callbacks always succeed.  In addition, the
non-debugging runtime-PM sysfs attributes for the devices are removed,
since they are pretty much meaningless.

The advantage of this scheme comes not so much from avoiding the
callbacks themselves, but rather from the fact that without the need
for a process context in which to run the callbacks, more work can be
done in interrupt context.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 Documentation/power/runtime_pm.txt | 37 +++++++++++++++++++++++++
 drivers/base/power/power.h         |  1 +
 drivers/base/power/runtime.c       | 54 +++++++++++++++++++++++++++++++++++-
 drivers/base/power/sysfs.c         | 56 +++++++++++++++++++++++++++++++++-----
 include/linux/pm.h                 |  7 +++++
 include/linux/pm_runtime.h         |  2 ++
 6 files changed, 149 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/power/runtime_pm.txt b/Documentation/power/runtime_pm.txt
index 55b859b3bc72..9ba49b21ac86 100644
--- a/Documentation/power/runtime_pm.txt
+++ b/Documentation/power/runtime_pm.txt
@@ -1,6 +1,7 @@
 Run-time Power Management Framework for I/O Devices
 
 (C) 2009 Rafael J. Wysocki <rjw@sisk.pl>, Novell Inc.
+(C) 2010 Alan Stern <stern@rowland.harvard.edu>
 
 1. Introduction
 
@@ -230,6 +231,11 @@ defined in include/linux/pm.h:
       interface; it may only be modified with the help of the pm_runtime_allow()
       and pm_runtime_forbid() helper functions
 
+  unsigned int no_callbacks;
+    - indicates that the device does not use the run-time PM callbacks (see
+      Section 8); it may be modified only by the pm_runtime_no_callbacks()
+      helper function
+
 All of the above fields are members of the 'power' member of 'struct device'.
 
 4. Run-time PM Device Helper Functions
@@ -349,6 +355,11 @@ drivers/base/power/runtime.c and include/linux/pm_runtime.h:
       counter (used by the /sys/devices/.../power/control interface to
       effectively prevent the device from being power managed at run time)
 
+  void pm_runtime_no_callbacks(struct device *dev);
+    - set the power.no_callbacks flag for the device and remove the run-time
+      PM attributes from /sys/devices/.../power (or prevent them from being
+      added when the device is registered)
+
 It is safe to execute the following helper functions from interrupt context:
 
 pm_request_idle()
@@ -524,3 +535,29 @@ poweroff and run-time suspend callback, and similarly for system resume, thaw,
 restore, and run-time resume, can achieve this with the help of the
 UNIVERSAL_DEV_PM_OPS macro defined in include/linux/pm.h (possibly setting its
 last argument to NULL).
+
+8. "No-Callback" Devices
+
+Some "devices" are only logical sub-devices of their parent and cannot be
+power-managed on their own.  (The prototype example is a USB interface.  Entire
+USB devices can go into low-power mode or send wake-up requests, but neither is
+possible for individual interfaces.)  The drivers for these devices have no
+need of run-time PM callbacks; if the callbacks did exist, ->runtime_suspend()
+and ->runtime_resume() would always return 0 without doing anything else and
+->runtime_idle() would always call pm_runtime_suspend().
+
+Subsystems can tell the PM core about these devices by calling
+pm_runtime_no_callbacks().  This should be done after the device structure is
+initialized and before it is registered (although after device registration is
+also okay).  The routine will set the device's power.no_callbacks flag and
+prevent the non-debugging run-time PM sysfs attributes from being created.
+
+When power.no_callbacks is set, the PM core will not invoke the
+->runtime_idle(), ->runtime_suspend(), or ->runtime_resume() callbacks.
+Instead it will assume that suspends and resumes always succeed and that idle
+devices should be suspended.
+
+As a consequence, the PM core will never directly inform the device's subsystem
+or driver about run-time power changes.  Instead, the driver for the device's
+parent must take responsibility for telling the device's driver when the
+parent's power state changes.
diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h
index 8b2745c69e2a..698dde742587 100644
--- a/drivers/base/power/power.h
+++ b/drivers/base/power/power.h
@@ -60,6 +60,7 @@ static inline void device_pm_move_last(struct device *dev) {}
 
 extern int dpm_sysfs_add(struct device *);
 extern void dpm_sysfs_remove(struct device *);
+extern void rpm_sysfs_remove(struct device *);
 
 #else /* CONFIG_PM */
 
diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c
index ed227b7c1bb5..5bd4daa93ef1 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -10,8 +10,10 @@
 #include <linux/sched.h>
 #include <linux/pm_runtime.h>
 #include <linux/jiffies.h>
+#include "power.h"
 
 static int rpm_resume(struct device *dev, int rpmflags);
+static int rpm_suspend(struct device *dev, int rpmflags);
 
 /**
  * update_pm_runtime_accounting - Update the time accounting of power states
@@ -148,6 +150,12 @@ static int rpm_idle(struct device *dev, int rpmflags)
 	/* Pending requests need to be canceled. */
 	dev->power.request = RPM_REQ_NONE;
 
+	if (dev->power.no_callbacks) {
+		/* Assume ->runtime_idle() callback would have suspended. */
+		retval = rpm_suspend(dev, rpmflags);
+		goto out;
+	}
+
 	/* Carry out an asynchronous or a synchronous idle notification. */
 	if (rpmflags & RPM_ASYNC) {
 		dev->power.request = RPM_REQ_IDLE;
@@ -254,6 +262,10 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 		goto repeat;
 	}
 
+	dev->power.deferred_resume = false;
+	if (dev->power.no_callbacks)
+		goto no_callback;	/* Assume success. */
+
 	/* Carry out an asynchronous or a synchronous suspend. */
 	if (rpmflags & RPM_ASYNC) {
 		dev->power.request = RPM_REQ_SUSPEND;
@@ -265,7 +277,6 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 	}
 
 	__update_runtime_status(dev, RPM_SUSPENDING);
-	dev->power.deferred_resume = false;
 
 	if (dev->bus && dev->bus->pm && dev->bus->pm->runtime_suspend) {
 		spin_unlock_irq(&dev->power.lock);
@@ -305,6 +316,7 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 			pm_runtime_cancel_pending(dev);
 		}
 	} else {
+ no_callback:
 		__update_runtime_status(dev, RPM_SUSPENDED);
 		pm_runtime_deactivate_timer(dev);
 
@@ -409,6 +421,23 @@ static int rpm_resume(struct device *dev, int rpmflags)
 		goto repeat;
 	}
 
+	/*
+	 * See if we can skip waking up the parent.  This is safe only if
+	 * power.no_callbacks is set, because otherwise we don't know whether
+	 * the resume will actually succeed.
+	 */
+	if (dev->power.no_callbacks && !parent && dev->parent) {
+		spin_lock(&dev->parent->power.lock);
+		if (dev->parent->power.disable_depth > 0
+		    || dev->parent->power.ignore_children
+		    || dev->parent->power.runtime_status == RPM_ACTIVE) {
+			atomic_inc(&dev->parent->power.child_count);
+			spin_unlock(&dev->parent->power.lock);
+			goto no_callback;	/* Assume success. */
+		}
+		spin_unlock(&dev->parent->power.lock);
+	}
+
 	/* Carry out an asynchronous or a synchronous resume. */
 	if (rpmflags & RPM_ASYNC) {
 		dev->power.request = RPM_REQ_RESUME;
@@ -449,6 +478,9 @@ static int rpm_resume(struct device *dev, int rpmflags)
 		goto repeat;
 	}
 
+	if (dev->power.no_callbacks)
+		goto no_callback;	/* Assume success. */
+
 	__update_runtime_status(dev, RPM_RESUMING);
 
 	if (dev->bus && dev->bus->pm && dev->bus->pm->runtime_resume) {
@@ -482,6 +514,7 @@ static int rpm_resume(struct device *dev, int rpmflags)
 		__update_runtime_status(dev, RPM_SUSPENDED);
 		pm_runtime_cancel_pending(dev);
 	} else {
+ no_callback:
 		__update_runtime_status(dev, RPM_ACTIVE);
 		if (parent)
 			atomic_inc(&parent->power.child_count);
@@ -954,6 +987,25 @@ void pm_runtime_allow(struct device *dev)
 }
 EXPORT_SYMBOL_GPL(pm_runtime_allow);
 
+/**
+ * pm_runtime_no_callbacks - Ignore run-time PM callbacks for a device.
+ * @dev: Device to handle.
+ *
+ * Set the power.no_callbacks flag, which tells the PM core that this
+ * device is power-managed through its parent and has no run-time PM
+ * callbacks of its own.  The run-time sysfs attributes will be removed.
+ *
+ */
+void pm_runtime_no_callbacks(struct device *dev)
+{
+	spin_lock_irq(&dev->power.lock);
+	dev->power.no_callbacks = 1;
+	spin_unlock_irq(&dev->power.lock);
+	if (device_is_registered(dev))
+		rpm_sysfs_remove(dev);
+}
+EXPORT_SYMBOL_GPL(pm_runtime_no_callbacks);
+
 /**
  * pm_runtime_init - Initialize run-time PM fields in given device object.
  * @dev: Device object to initialize.
diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c
index 8859780817e1..b5708c47ce2d 100644
--- a/drivers/base/power/sysfs.c
+++ b/drivers/base/power/sysfs.c
@@ -81,6 +81,9 @@
 static const char enabled[] = "enabled";
 static const char disabled[] = "disabled";
 
+const char power_group_name[] = "power";
+EXPORT_SYMBOL_GPL(power_group_name);
+
 #ifdef CONFIG_PM_RUNTIME
 static const char ctrl_auto[] = "auto";
 static const char ctrl_on[] = "on";
@@ -390,12 +393,6 @@ static DEVICE_ATTR(async, 0644, async_show, async_store);
 #endif /* CONFIG_PM_ADVANCED_DEBUG */
 
 static struct attribute * power_attrs[] = {
-#ifdef CONFIG_PM_RUNTIME
-	&dev_attr_control.attr,
-	&dev_attr_runtime_status.attr,
-	&dev_attr_runtime_suspended_time.attr,
-	&dev_attr_runtime_active_time.attr,
-#endif
 	&dev_attr_wakeup.attr,
 #ifdef CONFIG_PM_SLEEP
 	&dev_attr_wakeup_count.attr,
@@ -409,6 +406,7 @@ static struct attribute * power_attrs[] = {
 #ifdef CONFIG_PM_ADVANCED_DEBUG
 	&dev_attr_async.attr,
 #ifdef CONFIG_PM_RUNTIME
+	&dev_attr_runtime_status.attr,
 	&dev_attr_runtime_usage.attr,
 	&dev_attr_runtime_active_kids.attr,
 	&dev_attr_runtime_enabled.attr,
@@ -417,10 +415,52 @@ static struct attribute * power_attrs[] = {
 	NULL,
 };
 static struct attribute_group pm_attr_group = {
-	.name	= "power",
+	.name	= power_group_name,
 	.attrs	= power_attrs,
 };
 
+#ifdef CONFIG_PM_RUNTIME
+
+static struct attribute *runtime_attrs[] = {
+#ifndef CONFIG_PM_ADVANCED_DEBUG
+	&dev_attr_runtime_status.attr,
+#endif
+	&dev_attr_control.attr,
+	&dev_attr_runtime_suspended_time.attr,
+	&dev_attr_runtime_active_time.attr,
+	NULL,
+};
+static struct attribute_group pm_runtime_attr_group = {
+	.name	= power_group_name,
+	.attrs	= runtime_attrs,
+};
+
+int dpm_sysfs_add(struct device *dev)
+{
+	int rc;
+
+	rc = sysfs_create_group(&dev->kobj, &pm_attr_group);
+	if (rc == 0 && !dev->power.no_callbacks) {
+		rc = sysfs_merge_group(&dev->kobj, &pm_runtime_attr_group);
+		if (rc)
+			sysfs_remove_group(&dev->kobj, &pm_attr_group);
+	}
+	return rc;
+}
+
+void rpm_sysfs_remove(struct device *dev)
+{
+	sysfs_unmerge_group(&dev->kobj, &pm_runtime_attr_group);
+}
+
+void dpm_sysfs_remove(struct device *dev)
+{
+	rpm_sysfs_remove(dev);
+	sysfs_remove_group(&dev->kobj, &pm_attr_group);
+}
+
+#else /* CONFIG_PM_RUNTIME */
+
 int dpm_sysfs_add(struct device * dev)
 {
 	return sysfs_create_group(&dev->kobj, &pm_attr_group);
@@ -430,3 +470,5 @@ void dpm_sysfs_remove(struct device * dev)
 {
 	sysfs_remove_group(&dev->kobj, &pm_attr_group);
 }
+
+#endif
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 1abfe84f447d..abd81ffaba3c 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -41,6 +41,12 @@ extern void (*pm_power_off_prepare)(void);
 
 struct device;
 
+#ifdef CONFIG_PM
+extern const char power_group_name[];		/* = "power" */
+#else
+#define power_group_name	NULL
+#endif
+
 typedef struct pm_message {
 	int event;
 } pm_message_t;
@@ -475,6 +481,7 @@ struct dev_pm_info {
 	unsigned int		deferred_resume:1;
 	unsigned int		run_wake:1;
 	unsigned int		runtime_auto:1;
+	unsigned int		no_callbacks:1;
 	enum rpm_request	request;
 	enum rpm_status		runtime_status;
 	int			runtime_error;
diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h
index 5869d87fffac..8ca52f7c357e 100644
--- a/include/linux/pm_runtime.h
+++ b/include/linux/pm_runtime.h
@@ -36,6 +36,7 @@ extern void pm_runtime_forbid(struct device *dev);
 extern int pm_generic_runtime_idle(struct device *dev);
 extern int pm_generic_runtime_suspend(struct device *dev);
 extern int pm_generic_runtime_resume(struct device *dev);
+extern void pm_runtime_no_callbacks(struct device *dev);
 
 static inline bool pm_children_suspended(struct device *dev)
 {
@@ -110,6 +111,7 @@ static inline bool pm_runtime_suspended(struct device *dev) { return false; }
 static inline int pm_generic_runtime_idle(struct device *dev) { return 0; }
 static inline int pm_generic_runtime_suspend(struct device *dev) { return 0; }
 static inline int pm_generic_runtime_resume(struct device *dev) { return 0; }
+static inline void pm_runtime_no_callbacks(struct device *dev) {}
 
 #endif /* !CONFIG_PM_RUNTIME */
 
-- 
cgit v1.2.3


From 15bcb91d7e607d8a2e060f01f7784a7454668da4 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 25 Sep 2010 23:35:21 +0200
Subject: PM / Runtime: Implement autosuspend support

This patch (as1427) implements the "autosuspend" facility for runtime
PM.  A few new fields are added to the dev_pm_info structure and
several new PM helper functions are defined, for telling the PM core
whether or not a device uses autosuspend, for setting the autosuspend
delay, and for marking periods of device activity.

Drivers that do not want to use autosuspend can continue using the
same helper functions as before; their behavior will not change.  In
addition, drivers supporting autosuspend can also call the old helper
functions to get the old behavior.

The details are all explained in Documentation/power/runtime_pm.txt
and Documentation/ABI/testing/sysfs-devices-power.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 Documentation/ABI/testing/sysfs-devices-power |  18 +++
 Documentation/power/runtime_pm.txt            | 190 +++++++++++++++++++++++++-
 drivers/base/power/runtime.c                  | 186 ++++++++++++++++++++++++-
 drivers/base/power/sysfs.c                    |  40 ++++++
 include/linux/pm.h                            |   8 ++
 include/linux/pm_runtime.h                    |  45 ++++++
 6 files changed, 473 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-devices-power b/Documentation/ABI/testing/sysfs-devices-power
index 6bb2dd3c3a71..7628cd1bc36a 100644
--- a/Documentation/ABI/testing/sysfs-devices-power
+++ b/Documentation/ABI/testing/sysfs-devices-power
@@ -147,3 +147,21 @@ Description:
 		milliseconds.  This attribute is read-only.  If the device is
 		not enabled to wake up the system from sleep states, this
 		attribute is empty.
+
+What:		/sys/devices/.../power/autosuspend_delay_ms
+Date:		September 2010
+Contact:	Alan Stern <stern@rowland.harvard.edu>
+Description:
+		The /sys/devices/.../power/autosuspend_delay_ms attribute
+		contains the autosuspend delay value (in milliseconds).  Some
+		drivers do not want their device to suspend as soon as it
+		becomes idle at run time; they want the device to remain
+		inactive for a certain minimum period of time first.  That
+		period is called the autosuspend delay.  Negative values will
+		prevent the device from being suspended at run time (similar
+		to writing "on" to the power/control attribute).  Values >=
+		1000 will cause the autosuspend timer expiration to be rounded
+		up to the nearest second.
+
+		Not all drivers support this attribute.  If it isn't supported,
+		attempts to read or write it will yield I/O errors.
diff --git a/Documentation/power/runtime_pm.txt b/Documentation/power/runtime_pm.txt
index 9ba49b21ac86..489e9bacd165 100644
--- a/Documentation/power/runtime_pm.txt
+++ b/Documentation/power/runtime_pm.txt
@@ -158,7 +158,8 @@ rules:
     to execute it, the other callbacks will not be executed for the same device.
 
   * A request to execute ->runtime_resume() will cancel any pending or
-    scheduled requests to execute the other callbacks for the same device.
+    scheduled requests to execute the other callbacks for the same device,
+    except for scheduled autosuspends.
 
 3. Run-time PM Device Fields
 
@@ -166,7 +167,7 @@ The following device run-time PM fields are present in 'struct dev_pm_info', as
 defined in include/linux/pm.h:
 
   struct timer_list suspend_timer;
-    - timer used for scheduling (delayed) suspend request
+    - timer used for scheduling (delayed) suspend and autosuspend requests
 
   unsigned long timer_expires;
     - timer expiration time, in jiffies (if this is different from zero, the
@@ -236,6 +237,23 @@ defined in include/linux/pm.h:
       Section 8); it may be modified only by the pm_runtime_no_callbacks()
       helper function
 
+  unsigned int use_autosuspend;
+    - indicates that the device's driver supports delayed autosuspend (see
+      Section 9); it may be modified only by the
+      pm_runtime{_dont}_use_autosuspend() helper functions
+
+  unsigned int timer_autosuspends;
+    - indicates that the PM core should attempt to carry out an autosuspend
+      when the timer expires rather than a normal suspend
+
+  int autosuspend_delay;
+    - the delay time (in milliseconds) to be used for autosuspend
+
+  unsigned long last_busy;
+    - the time (in jiffies) when the pm_runtime_mark_last_busy() helper
+      function was last called for this device; used in calculating inactivity
+      periods for autosuspend
+
 All of the above fields are members of the 'power' member of 'struct device'.
 
 4. Run-time PM Device Helper Functions
@@ -261,6 +279,12 @@ drivers/base/power/runtime.c and include/linux/pm_runtime.h:
       error code on failure, where -EAGAIN or -EBUSY means it is safe to attempt
       to suspend the device again in future
 
+  int pm_runtime_autosuspend(struct device *dev);
+    - same as pm_runtime_suspend() except that the autosuspend delay is taken
+      into account; if pm_runtime_autosuspend_expiration() says the delay has
+      not yet expired then an autosuspend is scheduled for the appropriate time
+      and 0 is returned
+
   int pm_runtime_resume(struct device *dev);
     - execute the subsystem-level resume callback for the device; returns 0 on
       success, 1 if the device's run-time PM status was already 'active' or
@@ -273,6 +297,11 @@ drivers/base/power/runtime.c and include/linux/pm_runtime.h:
       device (the request is represented by a work item in pm_wq); returns 0 on
       success or error code if the request has not been queued up
 
+  int pm_request_autosuspend(struct device *dev);
+    - schedule the execution of the subsystem-level suspend callback for the
+      device when the autosuspend delay has expired; if the delay has already
+      expired then the work item is queued up immediately
+
   int pm_schedule_suspend(struct device *dev, unsigned int delay);
     - schedule the execution of the subsystem-level suspend callback for the
       device in future, where 'delay' is the time to wait before queuing up a
@@ -304,12 +333,20 @@ drivers/base/power/runtime.c and include/linux/pm_runtime.h:
     - decrement the device's usage counter
 
   int pm_runtime_put(struct device *dev);
-    - decrement the device's usage counter, run pm_request_idle(dev) and return
-      its result
+    - decrement the device's usage counter; if the result is 0 then run
+      pm_request_idle(dev) and return its result
+
+  int pm_runtime_put_autosuspend(struct device *dev);
+    - decrement the device's usage counter; if the result is 0 then run
+      pm_request_autosuspend(dev) and return its result
 
   int pm_runtime_put_sync(struct device *dev);
-    - decrement the device's usage counter, run pm_runtime_idle(dev) and return
-      its result
+    - decrement the device's usage counter; if the result is 0 then run
+      pm_runtime_idle(dev) and return its result
+
+  int pm_runtime_put_sync_autosuspend(struct device *dev);
+    - decrement the device's usage counter; if the result is 0 then run
+      pm_runtime_autosuspend(dev) and return its result
 
   void pm_runtime_enable(struct device *dev);
     - enable the run-time PM helper functions to run the device bus type's
@@ -360,19 +397,46 @@ drivers/base/power/runtime.c and include/linux/pm_runtime.h:
       PM attributes from /sys/devices/.../power (or prevent them from being
       added when the device is registered)
 
+  void pm_runtime_mark_last_busy(struct device *dev);
+    - set the power.last_busy field to the current time
+
+  void pm_runtime_use_autosuspend(struct device *dev);
+    - set the power.use_autosuspend flag, enabling autosuspend delays
+
+  void pm_runtime_dont_use_autosuspend(struct device *dev);
+    - clear the power.use_autosuspend flag, disabling autosuspend delays
+
+  void pm_runtime_set_autosuspend_delay(struct device *dev, int delay);
+    - set the power.autosuspend_delay value to 'delay' (expressed in
+      milliseconds); if 'delay' is negative then run-time suspends are
+      prevented
+
+  unsigned long pm_runtime_autosuspend_expiration(struct device *dev);
+    - calculate the time when the current autosuspend delay period will expire,
+      based on power.last_busy and power.autosuspend_delay; if the delay time
+      is 1000 ms or larger then the expiration time is rounded up to the
+      nearest second; returns 0 if the delay period has already expired or
+      power.use_autosuspend isn't set, otherwise returns the expiration time
+      in jiffies
+
 It is safe to execute the following helper functions from interrupt context:
 
 pm_request_idle()
+pm_request_autosuspend()
 pm_schedule_suspend()
 pm_request_resume()
 pm_runtime_get_noresume()
 pm_runtime_get()
 pm_runtime_put_noidle()
 pm_runtime_put()
+pm_runtime_put_autosuspend()
+pm_runtime_enable()
 pm_suspend_ignore_children()
 pm_runtime_set_active()
 pm_runtime_set_suspended()
-pm_runtime_enable()
+pm_runtime_suspended()
+pm_runtime_mark_last_busy()
+pm_runtime_autosuspend_expiration()
 
 5. Run-time PM Initialization, Device Probing and Removal
 
@@ -561,3 +625,115 @@ As a consequence, the PM core will never directly inform the device's subsystem
 or driver about run-time power changes.  Instead, the driver for the device's
 parent must take responsibility for telling the device's driver when the
 parent's power state changes.
+
+9. Autosuspend, or automatically-delayed suspends
+
+Changing a device's power state isn't free; it requires both time and energy.
+A device should be put in a low-power state only when there's some reason to
+think it will remain in that state for a substantial time.  A common heuristic
+says that a device which hasn't been used for a while is liable to remain
+unused; following this advice, drivers should not allow devices to be suspended
+at run-time until they have been inactive for some minimum period.  Even when
+the heuristic ends up being non-optimal, it will still prevent devices from
+"bouncing" too rapidly between low-power and full-power states.
+
+The term "autosuspend" is an historical remnant.  It doesn't mean that the
+device is automatically suspended (the subsystem or driver still has to call
+the appropriate PM routines); rather it means that run-time suspends will
+automatically be delayed until the desired period of inactivity has elapsed.
+
+Inactivity is determined based on the power.last_busy field.  Drivers should
+call pm_runtime_mark_last_busy() to update this field after carrying out I/O,
+typically just before calling pm_runtime_put_autosuspend().  The desired length
+of the inactivity period is a matter of policy.  Subsystems can set this length
+initially by calling pm_runtime_set_autosuspend_delay(), but after device
+registration the length should be controlled by user space, using the
+/sys/devices/.../power/autosuspend_delay_ms attribute.
+
+In order to use autosuspend, subsystems or drivers must call
+pm_runtime_use_autosuspend() (preferably before registering the device), and
+thereafter they should use the various *_autosuspend() helper functions instead
+of the non-autosuspend counterparts:
+
+	Instead of: pm_runtime_suspend    use: pm_runtime_autosuspend;
+	Instead of: pm_schedule_suspend   use: pm_request_autosuspend;
+	Instead of: pm_runtime_put        use: pm_runtime_put_autosuspend;
+	Instead of: pm_runtime_put_sync   use: pm_runtime_put_sync_autosuspend.
+
+Drivers may also continue to use the non-autosuspend helper functions; they
+will behave normally, not taking the autosuspend delay into account.
+Similarly, if the power.use_autosuspend field isn't set then the autosuspend
+helper functions will behave just like the non-autosuspend counterparts.
+
+The implementation is well suited for asynchronous use in interrupt contexts.
+However such use inevitably involves races, because the PM core can't
+synchronize ->runtime_suspend() callbacks with the arrival of I/O requests.
+This synchronization must be handled by the driver, using its private lock.
+Here is a schematic pseudo-code example:
+
+	foo_read_or_write(struct foo_priv *foo, void *data)
+	{
+		lock(&foo->private_lock);
+		add_request_to_io_queue(foo, data);
+		if (foo->num_pending_requests++ == 0)
+			pm_runtime_get(&foo->dev);
+		if (!foo->is_suspended)
+			foo_process_next_request(foo);
+		unlock(&foo->private_lock);
+	}
+
+	foo_io_completion(struct foo_priv *foo, void *req)
+	{
+		lock(&foo->private_lock);
+		if (--foo->num_pending_requests == 0) {
+			pm_runtime_mark_last_busy(&foo->dev);
+			pm_runtime_put_autosuspend(&foo->dev);
+		} else {
+			foo_process_next_request(foo);
+		}
+		unlock(&foo->private_lock);
+		/* Send req result back to the user ... */
+	}
+
+	int foo_runtime_suspend(struct device *dev)
+	{
+		struct foo_priv foo = container_of(dev, ...);
+		int ret = 0;
+
+		lock(&foo->private_lock);
+		if (foo->num_pending_requests > 0) {
+			ret = -EBUSY;
+		} else {
+			/* ... suspend the device ... */
+			foo->is_suspended = 1;
+		}
+		unlock(&foo->private_lock);
+		return ret;
+	}
+
+	int foo_runtime_resume(struct device *dev)
+	{
+		struct foo_priv foo = container_of(dev, ...);
+
+		lock(&foo->private_lock);
+		/* ... resume the device ... */
+		foo->is_suspended = 0;
+		pm_runtime_mark_last_busy(&foo->dev);
+		if (foo->num_pending_requests > 0)
+			foo_process_requests(foo);
+		unlock(&foo->private_lock);
+		return 0;
+	}
+
+The important point is that after foo_io_completion() asks for an autosuspend,
+the foo_runtime_suspend() callback may race with foo_read_or_write().
+Therefore foo_runtime_suspend() has to check whether there are any pending I/O
+requests (while holding the private lock) before allowing the suspend to
+proceed.
+
+In addition, the power.autosuspend_delay field can be changed by user space at
+any time.  If a driver cares about this, it can call
+pm_runtime_autosuspend_expiration() from within the ->runtime_suspend()
+callback while holding its private lock.  If the function returns a nonzero
+value then the delay has not yet expired and the callback should return
+-EAGAIN.
diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c
index 5bd4daa93ef1..cd4e100a1362 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -9,7 +9,6 @@
 
 #include <linux/sched.h>
 #include <linux/pm_runtime.h>
-#include <linux/jiffies.h>
 #include "power.h"
 
 static int rpm_resume(struct device *dev, int rpmflags);
@@ -79,6 +78,53 @@ static void pm_runtime_cancel_pending(struct device *dev)
 	dev->power.request = RPM_REQ_NONE;
 }
 
+/*
+ * pm_runtime_autosuspend_expiration - Get a device's autosuspend-delay expiration time.
+ * @dev: Device to handle.
+ *
+ * Compute the autosuspend-delay expiration time based on the device's
+ * power.last_busy time.  If the delay has already expired or is disabled
+ * (negative) or the power.use_autosuspend flag isn't set, return 0.
+ * Otherwise return the expiration time in jiffies (adjusted to be nonzero).
+ *
+ * This function may be called either with or without dev->power.lock held.
+ * Either way it can be racy, since power.last_busy may be updated at any time.
+ */
+unsigned long pm_runtime_autosuspend_expiration(struct device *dev)
+{
+	int autosuspend_delay;
+	long elapsed;
+	unsigned long last_busy;
+	unsigned long expires = 0;
+
+	if (!dev->power.use_autosuspend)
+		goto out;
+
+	autosuspend_delay = ACCESS_ONCE(dev->power.autosuspend_delay);
+	if (autosuspend_delay < 0)
+		goto out;
+
+	last_busy = ACCESS_ONCE(dev->power.last_busy);
+	elapsed = jiffies - last_busy;
+	if (elapsed < 0)
+		goto out;	/* jiffies has wrapped around. */
+
+	/*
+	 * If the autosuspend_delay is >= 1 second, align the timer by rounding
+	 * up to the nearest second.
+	 */
+	expires = last_busy + msecs_to_jiffies(autosuspend_delay);
+	if (autosuspend_delay >= 1000)
+		expires = round_jiffies(expires);
+	expires += !expires;
+	if (elapsed >= expires - last_busy)
+		expires = 0;	/* Already expired. */
+
+ out:
+	return expires;
+}
+EXPORT_SYMBOL_GPL(pm_runtime_autosuspend_expiration);
+
 /**
  * rpm_check_suspend_allowed - Test whether a device may be suspended.
  * @dev: Device to test.
@@ -234,6 +280,32 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 	if (retval)
 		goto out;
 
+	/* If the autosuspend_delay time hasn't expired yet, reschedule. */
+	if ((rpmflags & RPM_AUTO)
+	    && dev->power.runtime_status != RPM_SUSPENDING) {
+		unsigned long expires = pm_runtime_autosuspend_expiration(dev);
+
+		if (expires != 0) {
+			/* Pending requests need to be canceled. */
+			dev->power.request = RPM_REQ_NONE;
+
+			/*
+			 * Optimization: If the timer is already running and is
+			 * set to expire at or before the autosuspend delay,
+			 * avoid the overhead of resetting it.  Just let it
+			 * expire; pm_suspend_timer_fn() will take care of the
+			 * rest.
+			 */
+			if (!(dev->power.timer_expires && time_before_eq(
+			    dev->power.timer_expires, expires))) {
+				dev->power.timer_expires = expires;
+				mod_timer(&dev->power.suspend_timer, expires);
+			}
+			dev->power.timer_autosuspends = 1;
+			goto out;
+		}
+	}
+
 	/* Other scheduled or pending requests need to be canceled. */
 	pm_runtime_cancel_pending(dev);
 
@@ -268,7 +340,8 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 
 	/* Carry out an asynchronous or a synchronous suspend. */
 	if (rpmflags & RPM_ASYNC) {
-		dev->power.request = RPM_REQ_SUSPEND;
+		dev->power.request = (rpmflags & RPM_AUTO) ?
+		    RPM_REQ_AUTOSUSPEND : RPM_REQ_SUSPEND;
 		if (!dev->power.request_pending) {
 			dev->power.request_pending = true;
 			queue_work(pm_wq, &dev->power.work);
@@ -383,8 +456,15 @@ static int rpm_resume(struct device *dev, int rpmflags)
 	if (retval)
 		goto out;
 
-	/* Other scheduled or pending requests need to be canceled. */
-	pm_runtime_cancel_pending(dev);
+	/*
+	 * Other scheduled or pending requests need to be canceled.  Small
+	 * optimization: If an autosuspend timer is running, leave it running
+	 * rather than cancelling it now only to restart it again in the near
+	 * future.
+	 */
+	dev->power.request = RPM_REQ_NONE;
+	if (!dev->power.timer_autosuspends)
+		pm_runtime_deactivate_timer(dev);
 
 	if (dev->power.runtime_status == RPM_ACTIVE) {
 		retval = 1;
@@ -568,6 +648,9 @@ static void pm_runtime_work(struct work_struct *work)
 	case RPM_REQ_SUSPEND:
 		rpm_suspend(dev, RPM_NOWAIT);
 		break;
+	case RPM_REQ_AUTOSUSPEND:
+		rpm_suspend(dev, RPM_NOWAIT | RPM_AUTO);
+		break;
 	case RPM_REQ_RESUME:
 		rpm_resume(dev, RPM_NOWAIT);
 		break;
@@ -595,7 +678,8 @@ static void pm_suspend_timer_fn(unsigned long data)
 	/* If 'expire' is after 'jiffies' we've been called too early. */
 	if (expires > 0 && !time_after(expires, jiffies)) {
 		dev->power.timer_expires = 0;
-		rpm_suspend(dev, RPM_ASYNC);
+		rpm_suspend(dev, dev->power.timer_autosuspends ?
+		    (RPM_ASYNC | RPM_AUTO) : RPM_ASYNC);
 	}
 
 	spin_unlock_irqrestore(&dev->power.lock, flags);
@@ -627,6 +711,7 @@ int pm_schedule_suspend(struct device *dev, unsigned int delay)
 
 	dev->power.timer_expires = jiffies + msecs_to_jiffies(delay);
 	dev->power.timer_expires += !dev->power.timer_expires;
+	dev->power.timer_autosuspends = 0;
 	mod_timer(&dev->power.suspend_timer, dev->power.timer_expires);
 
  out:
@@ -670,7 +755,9 @@ EXPORT_SYMBOL_GPL(__pm_runtime_idle);
  * @dev: Device to suspend.
  * @rpmflags: Flag bits.
  *
- * Carry out a suspend, either synchronous or asynchronous.
+ * If the RPM_GET_PUT flag is set, decrement the device's usage count and
+ * return immediately if it is larger than zero.  Then carry out a suspend,
+ * either synchronous or asynchronous.
  *
  * This routine may be called in atomic context if the RPM_ASYNC flag is set.
  */
@@ -679,6 +766,11 @@ int __pm_runtime_suspend(struct device *dev, int rpmflags)
 	unsigned long flags;
 	int retval;
 
+	if (rpmflags & RPM_GET_PUT) {
+		if (!atomic_dec_and_test(&dev->power.usage_count))
+			return 0;
+	}
+
 	spin_lock_irqsave(&dev->power.lock, flags);
 	retval = rpm_suspend(dev, rpmflags);
 	spin_unlock_irqrestore(&dev->power.lock, flags);
@@ -980,7 +1072,7 @@ void pm_runtime_allow(struct device *dev)
 
 	dev->power.runtime_auto = true;
 	if (atomic_dec_and_test(&dev->power.usage_count))
-		rpm_idle(dev, 0);
+		rpm_idle(dev, RPM_AUTO);
 
  out:
 	spin_unlock_irq(&dev->power.lock);
@@ -1006,6 +1098,86 @@ void pm_runtime_no_callbacks(struct device *dev)
 }
 EXPORT_SYMBOL_GPL(pm_runtime_no_callbacks);
 
+/**
+ * update_autosuspend - Handle a change to a device's autosuspend settings.
+ * @dev: Device to handle.
+ * @old_delay: The former autosuspend_delay value.
+ * @old_use: The former use_autosuspend value.
+ *
+ * Prevent runtime suspend if the new delay is negative and use_autosuspend is
+ * set; otherwise allow it.  Send an idle notification if suspends are allowed.
+ *
+ * This function must be called under dev->power.lock with interrupts disabled.
+ */
+static void update_autosuspend(struct device *dev, int old_delay, int old_use)
+{
+	int delay = dev->power.autosuspend_delay;
+
+	/* Should runtime suspend be prevented now? */
+	if (dev->power.use_autosuspend && delay < 0) {
+
+		/* If it used to be allowed then prevent it. */
+		if (!old_use || old_delay >= 0) {
+			atomic_inc(&dev->power.usage_count);
+			rpm_resume(dev, 0);
+		}
+	}
+
+	/* Runtime suspend should be allowed now. */
+	else {
+
+		/* If it used to be prevented then allow it. */
+		if (old_use && old_delay < 0)
+			atomic_dec(&dev->power.usage_count);
+
+		/* Maybe we can autosuspend now. */
+		rpm_idle(dev, RPM_AUTO);
+	}
+}
+
+/**
+ * pm_runtime_set_autosuspend_delay - Set a device's autosuspend_delay value.
+ * @dev: Device to handle.
+ * @delay: Value of the new delay in milliseconds.
+ *
+ * Set the device's power.autosuspend_delay value.  If it changes to negative
+ * and the power.use_autosuspend flag is set, prevent run-time suspends.  If it
+ * changes the other way, allow run-time suspends.
+ */
+void pm_runtime_set_autosuspend_delay(struct device *dev, int delay)
+{
+	int old_delay, old_use;
+
+	spin_lock_irq(&dev->power.lock);
+	old_delay = dev->power.autosuspend_delay;
+	old_use = dev->power.use_autosuspend;
+	dev->power.autosuspend_delay = delay;
+	update_autosuspend(dev, old_delay, old_use);
+	spin_unlock_irq(&dev->power.lock);
+}
+EXPORT_SYMBOL_GPL(pm_runtime_set_autosuspend_delay);
+
+/**
+ * __pm_runtime_use_autosuspend - Set a device's use_autosuspend flag.
+ * @dev: Device to handle.
+ * @use: New value for use_autosuspend.
+ *
+ * Set the device's power.use_autosuspend flag, and allow or prevent run-time
+ * suspends as needed.
+ */
+void __pm_runtime_use_autosuspend(struct device *dev, bool use)
+{
+	int old_delay, old_use;
+
+	spin_lock_irq(&dev->power.lock);
+	old_delay = dev->power.autosuspend_delay;
+	old_use = dev->power.use_autosuspend;
+	dev->power.use_autosuspend = use;
+	update_autosuspend(dev, old_delay, old_use);
+	spin_unlock_irq(&dev->power.lock);
+}
+EXPORT_SYMBOL_GPL(__pm_runtime_use_autosuspend);
+
 /**
  * pm_runtime_init - Initialize run-time PM fields in given device object.
  * @dev: Device object to initialize.
diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c
index b5708c47ce2d..0b1e46bf3e56 100644
--- a/drivers/base/power/sysfs.c
+++ b/drivers/base/power/sysfs.c
@@ -75,6 +75,18 @@
  *	attribute is set to "enabled" by bus type code or device drivers and in
  *	that cases it should be safe to leave the default value.
  *
+ *	autosuspend_delay_ms - Report/change a device's autosuspend_delay value
+ *
+ *	Some drivers don't want to carry out a runtime suspend as soon as a
+ *	device becomes idle; they want it always to remain idle for some period
+ *	of time before suspending it.  This period is the autosuspend_delay
+ *	value (expressed in milliseconds) and it can be controlled by the user.
+ *	If the value is negative then the device will never be runtime
+ *	suspended.
+ *
+ *	NOTE: The autosuspend_delay_ms attribute and the autosuspend_delay
+ *	value are used only if the driver calls pm_runtime_use_autosuspend().
+ *
  *	wakeup_count - Report the number of wakeup events related to the device
  */
 
@@ -173,6 +185,33 @@ static ssize_t rtpm_status_show(struct device *dev,
 }
 
 static DEVICE_ATTR(runtime_status, 0444, rtpm_status_show, NULL);
+
+static ssize_t autosuspend_delay_ms_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	if (!dev->power.use_autosuspend)
+		return -EIO;
+	return sprintf(buf, "%d\n", dev->power.autosuspend_delay);
+}
+
+static ssize_t autosuspend_delay_ms_store(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t n)
+{
+	long delay;
+
+	if (!dev->power.use_autosuspend)
+		return -EIO;
+
+	if (strict_strtol(buf, 10, &delay) != 0 || delay != (int) delay)
+		return -EINVAL;
+
+	pm_runtime_set_autosuspend_delay(dev, delay);
+	return n;
+}
+
+static DEVICE_ATTR(autosuspend_delay_ms, 0644, autosuspend_delay_ms_show,
+		autosuspend_delay_ms_store);
+
 #endif
 
 static ssize_t
@@ -428,6 +467,7 @@ static struct attribute *runtime_attrs[] = {
 	&dev_attr_control.attr,
 	&dev_attr_runtime_suspended_time.attr,
 	&dev_attr_runtime_active_time.attr,
+	&dev_attr_autosuspend_delay_ms.attr,
 	NULL,
 };
 static struct attribute_group pm_runtime_attr_group = {
diff --git a/include/linux/pm.h b/include/linux/pm.h
index abd81ffaba3c..40f3f45702ba 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -444,6 +444,9 @@ enum rpm_status {
  *
  * RPM_REQ_SUSPEND	Run the device bus type's ->runtime_suspend() callback
  *
+ * RPM_REQ_AUTOSUSPEND	Same as RPM_REQ_SUSPEND, but not until the device has
+ *			been inactive for as long as power.autosuspend_delay
+ *
  * RPM_REQ_RESUME	Run the device bus type's ->runtime_resume() callback
  */
 
@@ -451,6 +454,7 @@ enum rpm_request {
 	RPM_REQ_NONE = 0,
 	RPM_REQ_IDLE,
 	RPM_REQ_SUSPEND,
+	RPM_REQ_AUTOSUSPEND,
 	RPM_REQ_RESUME,
 };
 
@@ -482,9 +486,13 @@ struct dev_pm_info {
 	unsigned int		run_wake:1;
 	unsigned int		runtime_auto:1;
 	unsigned int		no_callbacks:1;
+	unsigned int		use_autosuspend:1;
+	unsigned int		timer_autosuspends:1;
 	enum rpm_request	request;
 	enum rpm_status		runtime_status;
 	int			runtime_error;
+	int			autosuspend_delay;
+	unsigned long		last_busy;
 	unsigned long		active_jiffies;
 	unsigned long		suspended_jiffies;
 	unsigned long		accounting_timestamp;
diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h
index 8ca52f7c357e..99ed1aa8f933 100644
--- a/include/linux/pm_runtime.h
+++ b/include/linux/pm_runtime.h
@@ -12,12 +12,15 @@
 #include <linux/device.h>
 #include <linux/pm.h>
 
+#include <linux/jiffies.h>
+
 /* Runtime PM flag argument bits */
 #define RPM_ASYNC		0x01	/* Request is asynchronous */
 #define RPM_NOWAIT		0x02	/* Don't wait for concurrent
 					    state change */
 #define RPM_GET_PUT		0x04	/* Increment/decrement the
 					    usage_count */
+#define RPM_AUTO		0x08	/* Use autosuspend_delay */
 
 #ifdef CONFIG_PM_RUNTIME
 
@@ -37,6 +40,9 @@ extern int pm_generic_runtime_idle(struct device *dev);
 extern int pm_generic_runtime_suspend(struct device *dev);
 extern int pm_generic_runtime_resume(struct device *dev);
 extern void pm_runtime_no_callbacks(struct device *dev);
+extern void __pm_runtime_use_autosuspend(struct device *dev, bool use);
+extern void pm_runtime_set_autosuspend_delay(struct device *dev, int delay);
+extern unsigned long pm_runtime_autosuspend_expiration(struct device *dev);
 
 static inline bool pm_children_suspended(struct device *dev)
 {
@@ -74,6 +80,11 @@ static inline bool pm_runtime_suspended(struct device *dev)
 	return dev->power.runtime_status == RPM_SUSPENDED;
 }
 
+static inline void pm_runtime_mark_last_busy(struct device *dev)
+{
+	ACCESS_ONCE(dev->power.last_busy) = jiffies;
+}
+
 #else /* !CONFIG_PM_RUNTIME */
 
 static inline int __pm_runtime_idle(struct device *dev, int rpmflags)
@@ -113,6 +124,14 @@ static inline int pm_generic_runtime_suspend(struct device *dev) { return 0; }
 static inline int pm_generic_runtime_resume(struct device *dev) { return 0; }
 static inline void pm_runtime_no_callbacks(struct device *dev) {}
 
+static inline void pm_runtime_mark_last_busy(struct device *dev) {}
+static inline void __pm_runtime_use_autosuspend(struct device *dev,
+						bool use) {}
+static inline void pm_runtime_set_autosuspend_delay(struct device *dev,
+						int delay) {}
+static inline unsigned long pm_runtime_autosuspend_expiration(
+				struct device *dev) { return 0; }
+
 #endif /* !CONFIG_PM_RUNTIME */
 
 static inline int pm_runtime_idle(struct device *dev)
@@ -125,6 +144,11 @@ static inline int pm_runtime_suspend(struct device *dev)
 	return __pm_runtime_suspend(dev, 0);
 }
 
+static inline int pm_runtime_autosuspend(struct device *dev)
+{
+	return __pm_runtime_suspend(dev, RPM_AUTO);
+}
+
 static inline int pm_runtime_resume(struct device *dev)
 {
 	return __pm_runtime_resume(dev, 0);
@@ -155,11 +179,22 @@ static inline int pm_runtime_put(struct device *dev)
 	return __pm_runtime_idle(dev, RPM_GET_PUT | RPM_ASYNC);
 }
 
+static inline int pm_runtime_put_autosuspend(struct device *dev)
+{
+	return __pm_runtime_suspend(dev,
+	    RPM_GET_PUT | RPM_ASYNC | RPM_AUTO);
+}
+
 static inline int pm_runtime_put_sync(struct device *dev)
 {
 	return __pm_runtime_idle(dev, RPM_GET_PUT);
 }
 
+static inline int pm_runtime_put_sync_autosuspend(struct device *dev)
+{
+	return __pm_runtime_suspend(dev, RPM_GET_PUT | RPM_AUTO);
+}
+
 static inline int pm_runtime_set_active(struct device *dev)
 {
 	return __pm_runtime_set_status(dev, RPM_ACTIVE);
@@ -175,4 +210,14 @@ static inline void pm_runtime_disable(struct device *dev)
 	__pm_runtime_disable(dev, true);
 }
 
+static inline void pm_runtime_use_autosuspend(struct device *dev)
+{
+	__pm_runtime_use_autosuspend(dev, true);
+}
+
+static inline void pm_runtime_dont_use_autosuspend(struct device *dev)
+{
+	__pm_runtime_use_autosuspend(dev, false);
+}
+
 #endif
-- 
cgit v1.2.3


From 5fc62aad4e7779c2f04691e48b351d08c050c1f1 Mon Sep 17 00:00:00 2001
From: Ming Lei <tom.leiming@gmail.com>
Date: Wed, 29 Sep 2010 00:12:22 +0200
Subject: PM: runtime: add missed pm_request_autosuspend

The patch "PM / Runtime: Implement autosuspend support" introduces
"autosuspend" facility for runtime PM, but misses helper function
of pm_request_autosuspend, so add it.

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 include/linux/pm_runtime.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h
index 99ed1aa8f933..3ec2358f8692 100644
--- a/include/linux/pm_runtime.h
+++ b/include/linux/pm_runtime.h
@@ -164,6 +164,11 @@ static inline int pm_request_resume(struct device *dev)
 	return __pm_runtime_resume(dev, RPM_ASYNC);
 }
 
+static inline int pm_request_autosuspend(struct device *dev)
+{
+	return __pm_runtime_suspend(dev, RPM_ASYNC | RPM_AUTO);
+}
+
 static inline int pm_runtime_get(struct device *dev)
 {
 	return __pm_runtime_resume(dev, RPM_GET_PUT | RPM_ASYNC);
-- 
cgit v1.2.3


From dbeeec5fe868f2e2e92fe94daa2c5a047240fdc4 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 4 Oct 2010 22:07:32 +0200
Subject: PM: Allow wakeup events to abort freezing of tasks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If there is a wakeup event during the freezing of tasks, suspend or
hibernation will fail anyway.  Since try_to_freeze_tasks() can take
up to 20 seconds to complete or fail, aborting it as soon as a wakeup
event is detected improves the worst case wakeup latency.

Based on a patch from Arve Hjønnevåg.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
---
 include/linux/suspend.h |  2 ++
 kernel/power/process.c  | 11 +++++++++--
 2 files changed, 11 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 6b1712c51102..26697514c5ec 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -308,6 +308,8 @@ static inline int unregister_pm_notifier(struct notifier_block *nb)
 }
 
 #define pm_notifier(fn, pri)	do { (void)(fn); } while (0)
+
+static inline bool pm_check_wakeup_events(void) { return true; }
 #endif /* !CONFIG_PM_SLEEP */
 
 extern struct mutex pm_mutex;
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 028a99598f49..e50b4c1b2a0f 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -40,6 +40,7 @@ static int try_to_freeze_tasks(bool sig_only)
 	struct timeval start, end;
 	u64 elapsed_csecs64;
 	unsigned int elapsed_csecs;
+	bool wakeup = false;
 
 	do_gettimeofday(&start);
 
@@ -78,6 +79,11 @@ static int try_to_freeze_tasks(bool sig_only)
 		if (!todo || time_after(jiffies, end_time))
 			break;
 
+		if (!pm_check_wakeup_events()) {
+			wakeup = true;
+			break;
+		}
+
 		/*
 		 * We need to retry, but first give the freezing tasks some
 		 * time to enter the regrigerator.
@@ -97,8 +103,9 @@ static int try_to_freeze_tasks(bool sig_only)
 		 * but it cleans up leftover PF_FREEZE requests.
 		 */
 		printk("\n");
-		printk(KERN_ERR "Freezing of tasks failed after %d.%02d seconds "
+		printk(KERN_ERR "Freezing of tasks %s after %d.%02d seconds "
 		       "(%d tasks refusing to freeze, wq_busy=%d):\n",
+		       wakeup ? "aborted" : "failed",
 		       elapsed_csecs / 100, elapsed_csecs % 100,
 		       todo - wq_busy, wq_busy);
 
@@ -107,7 +114,7 @@ static int try_to_freeze_tasks(bool sig_only)
 		read_lock(&tasklist_lock);
 		do_each_thread(g, p) {
 			task_lock(p);
-			if (freezing(p) && !freezer_should_skip(p))
+			if (!wakeup && freezing(p) && !freezer_should_skip(p))
 				sched_show_task(p);
 			cancel_freezing(p);
 			task_unlock(p);
-- 
cgit v1.2.3


From d33ac60beaf2c7dee5cd90aba7c1eb385dd70937 Mon Sep 17 00:00:00 2001
From: James Hogan <james@albanarts.com>
Date: Tue, 12 Oct 2010 00:00:25 +0200
Subject: PM: Add sysfs attr for rechecking dev hash from PM trace

If the device which fails to resume is part of a loadable kernel module
it won't be checked at startup against the magic number stored in the
RTC.

Add a read-only sysfs attribute /sys/power/pm_trace_dev_match which
contains a list of newline separated devices (usually just the one)
which currently match the last magic number. This allows the device
which is failing to resume to be found after the modules are loaded
again.

Signed-off-by: James Hogan <james@albanarts.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 Documentation/ABI/testing/sysfs-power | 29 +++++++++++++++++++++++++++++
 Documentation/power/s2ram.txt         |  7 +++++++
 drivers/base/power/trace.c            | 31 +++++++++++++++++++++++++++++++
 include/linux/resume-trace.h          |  2 ++
 kernel/power/main.c                   | 18 ++++++++++++++++++
 5 files changed, 87 insertions(+)

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-power b/Documentation/ABI/testing/sysfs-power
index 2875f1f74a07..194ca446ac28 100644
--- a/Documentation/ABI/testing/sysfs-power
+++ b/Documentation/ABI/testing/sysfs-power
@@ -99,9 +99,38 @@ Description:
 
 		dmesg -s 1000000 | grep 'hash matches'
 
+		If you do not get any matches (or they appear to be false
+		positives), it is possible that the last PM event point
+		referred to a device created by a loadable kernel module.  In
+		this case cat /sys/power/pm_trace_dev_match (see below) after
+		your system is started up and the kernel modules are loaded.
+
 		CAUTION: Using it will cause your machine's real-time (CMOS)
 		clock to be set to a random invalid time after a resume.
 
+What;		/sys/power/pm_trace_dev_match
+Date:		October 2010
+Contact:	James Hogan <james@albanarts.com>
+Description:
+		The /sys/power/pm_trace_dev_match file contains the name of the
+		device associated with the last PM event point saved in the RTC
+		across reboots when pm_trace has been used.  More precisely it
+		contains the list of current devices (including those
+		registered by loadable kernel modules since boot) which match
+		the device hash in the RTC at boot, with a newline after each
+		one.
+
+		The advantage of this file over the hash matches printed to the
+		kernel log (see /sys/power/pm_trace), is that it includes
+		devices created after boot by loadable kernel modules.
+
+		Due to the small hash size necessary to fit in the RTC, it is
+		possible that more than one device matches the hash, in which
+		case further investigation is required to determine which
+		device is causing the problem.  Note that genuine RTC clock
+		values (such as when pm_trace has not been used), can still
+		match a device and output it's name here.
+
 What:		/sys/power/pm_async
 Date:		January 2009
 Contact:	Rafael J. Wysocki <rjw@sisk.pl>
diff --git a/Documentation/power/s2ram.txt b/Documentation/power/s2ram.txt
index 514b94fc931e..1bdfa0443773 100644
--- a/Documentation/power/s2ram.txt
+++ b/Documentation/power/s2ram.txt
@@ -49,6 +49,13 @@ machine that doesn't boot) is:
    device (lspci and /sys/devices/pci* is your friend), and see if you can
    fix it, disable it, or trace into its resume function.
 
+   If no device matches the hash (or any matches appear to be false positives),
+   the culprit may be a device from a loadable kernel module that is not loaded
+   until after the hash is checked. You can check the hash against the current
+   devices again after more modules are loaded using sysfs:
+
+	cat /sys/power/pm_trace_dev_match
+
 For example, the above happens to be the VGA device on my EVO, which I
 used to run with "radeonfb" (it's an ATI Radeon mobility). It turns out
 that "radeonfb" simply cannot resume that device - it tries to set the
diff --git a/drivers/base/power/trace.c b/drivers/base/power/trace.c
index 17e24e3f4422..9f4258df4cfd 100644
--- a/drivers/base/power/trace.c
+++ b/drivers/base/power/trace.c
@@ -207,6 +207,37 @@ static int show_dev_hash(unsigned int value)
 
 static unsigned int hash_value_early_read;
 
+int show_trace_dev_match(char *buf, size_t size)
+{
+	unsigned int value = hash_value_early_read / (USERHASH * FILEHASH);
+	int ret = 0;
+	struct list_head *entry;
+
+	/*
+	 * It's possible that multiple devices will match the hash and we can't
+	 * tell which is the culprit, so it's best to output them all.
+	 */
+	device_pm_lock();
+	entry = dpm_list.prev;
+	while (size && entry != &dpm_list) {
+		struct device *dev = to_device(entry);
+		unsigned int hash = hash_string(DEVSEED, dev_name(dev),
+						DEVHASH);
+		if (hash == value) {
+			int len = snprintf(buf, size, "%s\n",
+					    dev_driver_string(dev));
+			if (len > size)
+				len = size;
+			buf += len;
+			ret += len;
+			size -= len;
+		}
+		entry = entry->prev;
+	}
+	device_pm_unlock();
+	return ret;
+}
+
 static int early_resume_init(void)
 {
 	hash_value_early_read = read_magic_time();
diff --git a/include/linux/resume-trace.h b/include/linux/resume-trace.h
index bc8c3881c729..f31db2368782 100644
--- a/include/linux/resume-trace.h
+++ b/include/linux/resume-trace.h
@@ -3,6 +3,7 @@
 
 #ifdef CONFIG_PM_TRACE
 #include <asm/resume-trace.h>
+#include <linux/types.h>
 
 extern int pm_trace_enabled;
 
@@ -14,6 +15,7 @@ static inline int pm_trace_is_enabled(void)
 struct device;
 extern void set_trace_device(struct device *);
 extern void generate_resume_trace(const void *tracedata, unsigned int user);
+extern int show_trace_dev_match(char *buf, size_t size);
 
 #define TRACE_DEVICE(dev) do { \
 	if (pm_trace_enabled) \
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 6b12a0cf4d9f..7b5db6a8561e 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -281,12 +281,30 @@ pm_trace_store(struct kobject *kobj, struct kobj_attribute *attr,
 }
 
 power_attr(pm_trace);
+
+static ssize_t pm_trace_dev_match_show(struct kobject *kobj,
+				       struct kobj_attribute *attr,
+				       char *buf)
+{
+	return show_trace_dev_match(buf, PAGE_SIZE);
+}
+
+static ssize_t
+pm_trace_dev_match_store(struct kobject *kobj, struct kobj_attribute *attr,
+			 const char *buf, size_t n)
+{
+	return -EINVAL;
+}
+
+power_attr(pm_trace_dev_match);
+
 #endif /* CONFIG_PM_TRACE */
 
 static struct attribute * g[] = {
 	&state_attr.attr,
 #ifdef CONFIG_PM_TRACE
 	&pm_trace_attr.attr,
+	&pm_trace_dev_match_attr.attr,
 #endif
 #ifdef CONFIG_PM_SLEEP
 	&pm_async_attr.attr,
-- 
cgit v1.2.3


From e1f60b292ffd61151403327aa19ff7a1871820bd Mon Sep 17 00:00:00 2001
From: Nishanth Menon <nm@ti.com>
Date: Wed, 13 Oct 2010 00:13:10 +0200
Subject: PM: Introduce library for device-specific OPPs (v7)

SoCs have a standard set of tuples consisting of frequency and
voltage pairs that the device will support per voltage domain. These
are called Operating Performance Points or OPPs. The actual
definitions of OPP varies over silicon versions. For a specific domain,
we can have a set of {frequency, voltage} pairs. As the kernel boots
and more information is available, a default set of these are activated
based on the precise nature of device. Further on operation, based on
conditions prevailing in the system (such as temperature), some OPP
availability may be temporarily controlled by the SoC frameworks.

To implement an OPP, some sort of power management support is necessary
hence this library depends on CONFIG_PM.

Contributions include:
Sanjeev Premi for the initial concept:
	http://patchwork.kernel.org/patch/50998/
Kevin Hilman for converting original design to device-based.
Kevin Hilman and Paul Walmsey for cleaning up many of the function
abstractions, improvements and data structure handling.
Romit Dasgupta for using enums instead of opp pointers.
Thara Gopinath, Eduardo Valentin and Vishwanath BS for fixes and
cleanups.
Linus Walleij for recommending this layer be made generic for usage
in other architectures beyond OMAP and ARM.
Mark Brown, Andrew Morton, Rafael J. Wysocki, Paul E. McKenney for
valuable improvements.

Discussions and comments from:
http://marc.info/?l=linux-omap&m=126033945313269&w=2
http://marc.info/?l=linux-omap&m=125482970102327&w=2
http://marc.info/?t=125809247500002&r=1&w=2
http://marc.info/?l=linux-omap&m=126025973426007&w=2
http://marc.info/?t=128152609200064&r=1&w=2
http://marc.info/?t=128468723000002&r=1&w=2
incorporated.

v1: http://marc.info/?t=128468723000002&r=1&w=2

Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 Documentation/power/00-INDEX |   2 +
 Documentation/power/opp.txt  | 375 ++++++++++++++++++++++++++
 drivers/base/power/Makefile  |   1 +
 drivers/base/power/opp.c     | 628 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/opp.h          | 105 ++++++++
 kernel/power/Kconfig         |  14 +
 6 files changed, 1125 insertions(+)
 create mode 100644 Documentation/power/opp.txt
 create mode 100644 drivers/base/power/opp.c
 create mode 100644 include/linux/opp.h

(limited to 'include/linux')

diff --git a/Documentation/power/00-INDEX b/Documentation/power/00-INDEX
index fb742c213c9e..45e9d4a91284 100644
--- a/Documentation/power/00-INDEX
+++ b/Documentation/power/00-INDEX
@@ -14,6 +14,8 @@ interface.txt
 	- Power management user interface in /sys/power
 notifiers.txt
 	- Registering suspend notifiers in device drivers
+opp.txt
+	- Operating Performance Point library
 pci.txt
 	- How the PCI Subsystem Does Power Management
 pm_qos_interface.txt
diff --git a/Documentation/power/opp.txt b/Documentation/power/opp.txt
new file mode 100644
index 000000000000..44d87ad3cea9
--- /dev/null
+++ b/Documentation/power/opp.txt
@@ -0,0 +1,375 @@
+*=============*
+* OPP Library *
+*=============*
+
+(C) 2009-2010 Nishanth Menon <nm@ti.com>, Texas Instruments Incorporated
+
+Contents
+--------
+1. Introduction
+2. Initial OPP List Registration
+3. OPP Search Functions
+4. OPP Availability Control Functions
+5. OPP Data Retrieval Functions
+6. Cpufreq Table Generation
+7. Data Structures
+
+1. Introduction
+===============
+Complex SoCs of today consists of a multiple sub-modules working in conjunction.
+In an operational system executing varied use cases, not all modules in the SoC
+need to function at their highest performing frequency all the time. To
+facilitate this, sub-modules in a SoC are grouped into domains, allowing some
+domains to run at lower voltage and frequency while other domains are loaded
+more. The set of discrete tuples consisting of frequency and voltage pairs that
+the device will support per domain are called Operating Performance Points or
+OPPs.
+
+OPP library provides a set of helper functions to organize and query the OPP
+information. The library is located in drivers/base/power/opp.c and the header
+is located in include/linux/opp.h. OPP library can be enabled by enabling
+CONFIG_PM_OPP from power management menuconfig menu. OPP library depends on
+CONFIG_PM as certain SoCs such as Texas Instrument's OMAP framework allows to
+optionally boot at a certain OPP without needing cpufreq.
+
+Typical usage of the OPP library is as follows:
+(users)		-> registers a set of default OPPs		-> (library)
+SoC framework	-> modifies on required cases certain OPPs	-> OPP layer
+		-> queries to search/retrieve information	->
+
+OPP layer expects each domain to be represented by a unique device pointer. SoC
+framework registers a set of initial OPPs per device with the OPP layer. This
+list is expected to be an optimally small number typically around 5 per device.
+This initial list contains a set of OPPs that the framework expects to be safely
+enabled by default in the system.
+
+Note on OPP Availability:
+------------------------
+As the system proceeds to operate, SoC framework may choose to make certain
+OPPs available or not available on each device based on various external
+factors. Example usage: Thermal management or other exceptional situations where
+SoC framework might choose to disable a higher frequency OPP to safely continue
+operations until that OPP could be re-enabled if possible.
+
+OPP library facilitates this concept in it's implementation. The following
+operational functions operate only on available opps:
+opp_find_freq_{ceil, floor}, opp_get_voltage, opp_get_freq, opp_get_opp_count
+and opp_init_cpufreq_table
+
+opp_find_freq_exact is meant to be used to find the opp pointer which can then
+be used for opp_enable/disable functions to make an opp available as required.
+
+WARNING: Users of OPP library should refresh their availability count using
+get_opp_count if opp_enable/disable functions are invoked for a device, the
+exact mechanism to trigger these or the notification mechanism to other
+dependent subsystems such as cpufreq are left to the discretion of the SoC
+specific framework which uses the OPP library. Similar care needs to be taken
+care to refresh the cpufreq table in cases of these operations.
+
+WARNING on OPP List locking mechanism:
+-------------------------------------------------
+OPP library uses RCU for exclusivity. RCU allows the query functions to operate
+in multiple contexts and this synchronization mechanism is optimal for a read
+intensive operations on data structure as the OPP library caters to.
+
+To ensure that the data retrieved are sane, the users such as SoC framework
+should ensure that the section of code operating on OPP queries are locked
+using RCU read locks. The opp_find_freq_{exact,ceil,floor},
+opp_get_{voltage, freq, opp_count} fall into this category.
+
+opp_{add,enable,disable} are updaters which use mutex and implement it's own
+RCU locking mechanisms. opp_init_cpufreq_table acts as an updater and uses
+mutex to implment RCU updater strategy. These functions should *NOT* be called
+under RCU locks and other contexts that prevent blocking functions in RCU or
+mutex operations from working.
+
+2. Initial OPP List Registration
+================================
+The SoC implementation calls opp_add function iteratively to add OPPs per
+device. It is expected that the SoC framework will register the OPP entries
+optimally- typical numbers range to be less than 5. The list generated by
+registering the OPPs is maintained by OPP library throughout the device
+operation. The SoC framework can subsequently control the availability of the
+OPPs dynamically using the opp_enable / disable functions.
+
+opp_add - Add a new OPP for a specific domain represented by the device pointer.
+	The OPP is defined using the frequency and voltage. Once added, the OPP
+	is assumed to be available and control of it's availability can be done
+	with the opp_enable/disable functions. OPP library internally stores
+	and manages this information in the opp struct. This function may be
+	used by SoC framework to define a optimal list as per the demands of
+	SoC usage environment.
+
+	WARNING: Do not use this function in interrupt context.
+
+	Example:
+	 soc_pm_init()
+	 {
+		/* Do things */
+		r = opp_add(mpu_dev, 1000000, 900000);
+		if (!r) {
+			pr_err("%s: unable to register mpu opp(%d)\n", r);
+			goto no_cpufreq;
+		}
+		/* Do cpufreq things */
+	 no_cpufreq:
+		/* Do remaining things */
+	 }
+
+3. OPP Search Functions
+=======================
+High level framework such as cpufreq operates on frequencies. To map the
+frequency back to the corresponding OPP, OPP library provides handy functions
+to search the OPP list that OPP library internally manages. These search
+functions return the matching pointer representing the opp if a match is
+found, else returns error. These errors are expected to be handled by standard
+error checks such as IS_ERR() and appropriate actions taken by the caller.
+
+opp_find_freq_exact - Search for an OPP based on an *exact* frequency and
+	availability. This function is especially useful to enable an OPP which
+	is not available by default.
+	Example: In a case when SoC framework detects a situation where a
+	higher frequency could be made available, it can use this function to
+	find the OPP prior to call the opp_enable to actually make it available.
+	 rcu_read_lock();
+	 opp = opp_find_freq_exact(dev, 1000000000, false);
+	 rcu_read_unlock();
+	 /* dont operate on the pointer.. just do a sanity check.. */
+	 if (IS_ERR(opp)) {
+		pr_err("frequency not disabled!\n");
+		/* trigger appropriate actions.. */
+	 } else {
+		opp_enable(dev,1000000000);
+	 }
+
+	NOTE: This is the only search function that operates on OPPs which are
+	not available.
+
+opp_find_freq_floor - Search for an available OPP which is *at most* the
+	provided frequency. This function is useful while searching for a lesser
+	match OR operating on OPP information in the order of decreasing
+	frequency.
+	Example: To find the highest opp for a device:
+	 freq = ULONG_MAX;
+	 rcu_read_lock();
+	 opp_find_freq_floor(dev, &freq);
+	 rcu_read_unlock();
+
+opp_find_freq_ceil - Search for an available OPP which is *at least* the
+	provided frequency. This function is useful while searching for a
+	higher match OR operating on OPP information in the order of increasing
+	frequency.
+	Example 1: To find the lowest opp for a device:
+	 freq = 0;
+	 rcu_read_lock();
+	 opp_find_freq_ceil(dev, &freq);
+	 rcu_read_unlock();
+	Example 2: A simplified implementation of a SoC cpufreq_driver->target:
+	 soc_cpufreq_target(..)
+	 {
+		/* Do stuff like policy checks etc. */
+		/* Find the best frequency match for the req */
+		rcu_read_lock();
+		opp = opp_find_freq_ceil(dev, &freq);
+		rcu_read_unlock();
+		if (!IS_ERR(opp))
+			soc_switch_to_freq_voltage(freq);
+		else
+			/* do something when we cant satisfy the req */
+		/* do other stuff */
+	 }
+
+4. OPP Availability Control Functions
+=====================================
+A default OPP list registered with the OPP library may not cater to all possible
+situation. The OPP library provides a set of functions to modify the
+availability of a OPP within the OPP list. This allows SoC frameworks to have
+fine grained dynamic control of which sets of OPPs are operationally available.
+These functions are intended to *temporarily* remove an OPP in conditions such
+as thermal considerations (e.g. don't use OPPx until the temperature drops).
+
+WARNING: Do not use these functions in interrupt context.
+
+opp_enable - Make a OPP available for operation.
+	Example: Lets say that 1GHz OPP is to be made available only if the
+	SoC temperature is lower than a certain threshold. The SoC framework
+	implementation might choose to do something as follows:
+	 if (cur_temp < temp_low_thresh) {
+		/* Enable 1GHz if it was disabled */
+		rcu_read_lock();
+		opp = opp_find_freq_exact(dev, 1000000000, false);
+		rcu_read_unlock();
+		/* just error check */
+		if (!IS_ERR(opp))
+			ret = opp_enable(dev, 1000000000);
+		else
+			goto try_something_else;
+	 }
+
+opp_disable - Make an OPP to be not available for operation
+	Example: Lets say that 1GHz OPP is to be disabled if the temperature
+	exceeds a threshold value. The SoC framework implementation might
+	choose to do something as follows:
+	 if (cur_temp > temp_high_thresh) {
+		/* Disable 1GHz if it was enabled */
+		rcu_read_lock();
+		opp = opp_find_freq_exact(dev, 1000000000, true);
+		rcu_read_unlock();
+		/* just error check */
+		if (!IS_ERR(opp))
+			ret = opp_disable(dev, 1000000000);
+		else
+			goto try_something_else;
+	 }
+
+5. OPP Data Retrieval Functions
+===============================
+Since OPP library abstracts away the OPP information, a set of functions to pull
+information from the OPP structure is necessary. Once an OPP pointer is
+retrieved using the search functions, the following functions can be used by SoC
+framework to retrieve the information represented inside the OPP layer.
+
+opp_get_voltage - Retrieve the voltage represented by the opp pointer.
+	Example: At a cpufreq transition to a different frequency, SoC
+	framework requires to set the voltage represented by the OPP using
+	the regulator framework to the Power Management chip providing the
+	voltage.
+	 soc_switch_to_freq_voltage(freq)
+	 {
+		/* do things */
+		rcu_read_lock();
+		opp = opp_find_freq_ceil(dev, &freq);
+		v = opp_get_voltage(opp);
+		rcu_read_unlock();
+		if (v)
+			regulator_set_voltage(.., v);
+		/* do other things */
+	 }
+
+opp_get_freq - Retrieve the freq represented by the opp pointer.
+	Example: Lets say the SoC framework uses a couple of helper functions
+	we could pass opp pointers instead of doing additional parameters to
+	handle quiet a bit of data parameters.
+	 soc_cpufreq_target(..)
+	 {
+		/* do things.. */
+		 max_freq = ULONG_MAX;
+		 rcu_read_lock();
+		 max_opp = opp_find_freq_floor(dev,&max_freq);
+		 requested_opp = opp_find_freq_ceil(dev,&freq);
+		 if (!IS_ERR(max_opp) && !IS_ERR(requested_opp))
+			r = soc_test_validity(max_opp, requested_opp);
+		 rcu_read_unlock();
+		/* do other things */
+	 }
+	 soc_test_validity(..)
+	 {
+		 if(opp_get_voltage(max_opp) < opp_get_voltage(requested_opp))
+			 return -EINVAL;
+		 if(opp_get_freq(max_opp) < opp_get_freq(requested_opp))
+			 return -EINVAL;
+		/* do things.. */
+	 }
+
+opp_get_opp_count - Retrieve the number of available opps for a device
+	Example: Lets say a co-processor in the SoC needs to know the available
+	frequencies in a table, the main processor can notify as following:
+	 soc_notify_coproc_available_frequencies()
+	 {
+		/* Do things */
+		rcu_read_lock();
+		num_available = opp_get_opp_count(dev);
+		speeds = kzalloc(sizeof(u32) * num_available, GFP_KERNEL);
+		/* populate the table in increasing order */
+		freq = 0;
+		while (!IS_ERR(opp = opp_find_freq_ceil(dev, &freq))) {
+			speeds[i] = freq;
+			freq++;
+			i++;
+		}
+		rcu_read_unlock();
+
+		soc_notify_coproc(AVAILABLE_FREQs, speeds, num_available);
+		/* Do other things */
+	 }
+
+6. Cpufreq Table Generation
+===========================
+opp_init_cpufreq_table - cpufreq framework typically is initialized with
+	cpufreq_frequency_table_cpuinfo which is provided with the list of
+	frequencies that are available for operation. This function provides
+	a ready to use conversion routine to translate the OPP layer's internal
+	information about the available frequencies into a format readily
+	providable to cpufreq.
+
+	WARNING: Do not use this function in interrupt context.
+
+	Example:
+	 soc_pm_init()
+	 {
+		/* Do things */
+		r = opp_init_cpufreq_table(dev, &freq_table);
+		if (!r)
+			cpufreq_frequency_table_cpuinfo(policy, freq_table);
+		/* Do other things */
+	 }
+
+	NOTE: This function is available only if CONFIG_CPU_FREQ is enabled in
+	addition to CONFIG_PM as power management feature is required to
+	dynamically scale voltage and frequency in a system.
+
+7. Data Structures
+==================
+Typically an SoC contains multiple voltage domains which are variable. Each
+domain is represented by a device pointer. The relationship to OPP can be
+represented as follows:
+SoC
+ |- device 1
+ |	|- opp 1 (availability, freq, voltage)
+ |	|- opp 2 ..
+ ...	...
+ |	`- opp n ..
+ |- device 2
+ ...
+ `- device m
+
+OPP library maintains a internal list that the SoC framework populates and
+accessed by various functions as described above. However, the structures
+representing the actual OPPs and domains are internal to the OPP library itself
+to allow for suitable abstraction reusable across systems.
+
+struct opp - The internal data structure of OPP library which is used to
+	represent an OPP. In addition to the freq, voltage, availability
+	information, it also contains internal book keeping information required
+	for the OPP library to operate on.  Pointer to this structure is
+	provided back to the users such as SoC framework to be used as a
+	identifier for OPP in the interactions with OPP layer.
+
+	WARNING: The struct opp pointer should not be parsed or modified by the
+	users. The defaults of for an instance is populated by opp_add, but the
+	availability of the OPP can be modified by opp_enable/disable functions.
+
+struct device - This is used to identify a domain to the OPP layer. The
+	nature of the device and it's implementation is left to the user of
+	OPP library such as the SoC framework.
+
+Overall, in a simplistic view, the data structure operations is represented as
+following:
+
+Initialization / modification:
+            +-----+        /- opp_enable
+opp_add --> | opp | <-------
+  |         +-----+        \- opp_disable
+  \-------> domain_info(device)
+
+Search functions:
+             /-- opp_find_freq_ceil  ---\   +-----+
+domain_info<---- opp_find_freq_exact -----> | opp |
+             \-- opp_find_freq_floor ---/   +-----+
+
+Retrieval functions:
++-----+     /- opp_get_voltage
+| opp | <---
++-----+     \- opp_get_freq
+
+domain_info <- opp_get_opp_count
diff --git a/drivers/base/power/Makefile b/drivers/base/power/Makefile
index cbccf9a3cee4..abe46edfe5b4 100644
--- a/drivers/base/power/Makefile
+++ b/drivers/base/power/Makefile
@@ -3,6 +3,7 @@ obj-$(CONFIG_PM_SLEEP)	+= main.o wakeup.o
 obj-$(CONFIG_PM_RUNTIME)	+= runtime.o
 obj-$(CONFIG_PM_OPS)	+= generic_ops.o
 obj-$(CONFIG_PM_TRACE_RTC)	+= trace.o
+obj-$(CONFIG_PM_OPP)	+= opp.o
 
 ccflags-$(CONFIG_DEBUG_DRIVER) := -DDEBUG
 ccflags-$(CONFIG_PM_VERBOSE)   += -DDEBUG
diff --git a/drivers/base/power/opp.c b/drivers/base/power/opp.c
new file mode 100644
index 000000000000..2bb9b4cf59d7
--- /dev/null
+++ b/drivers/base/power/opp.c
@@ -0,0 +1,628 @@
+/*
+ * Generic OPP Interface
+ *
+ * Copyright (C) 2009-2010 Texas Instruments Incorporated.
+ *	Nishanth Menon
+ *	Romit Dasgupta
+ *	Kevin Hilman
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/cpufreq.h>
+#include <linux/list.h>
+#include <linux/rculist.h>
+#include <linux/rcupdate.h>
+#include <linux/opp.h>
+
+/*
+ * Internal data structure organization with the OPP layer library is as
+ * follows:
+ * dev_opp_list (root)
+ *	|- device 1 (represents voltage domain 1)
+ *	|	|- opp 1 (availability, freq, voltage)
+ *	|	|- opp 2 ..
+ *	...	...
+ *	|	`- opp n ..
+ *	|- device 2 (represents the next voltage domain)
+ *	...
+ *	`- device m (represents mth voltage domain)
+ * device 1, 2.. are represented by dev_opp structure while each opp
+ * is represented by the opp structure.
+ */
+
+/**
+ * struct opp - Generic OPP description structure
+ * @node:	opp list node. The nodes are maintained throughout the lifetime
+ *		of boot. It is expected only an optimal set of OPPs are
+ *		added to the library by the SoC framework.
+ *		RCU usage: opp list is traversed with RCU locks. node
+ *		modification is possible realtime, hence the modifications
+ *		are protected by the dev_opp_list_lock for integrity.
+ *		IMPORTANT: the opp nodes should be maintained in increasing
+ *		order.
+ * @available:	true/false - marks if this OPP as available or not
+ * @rate:	Frequency in hertz
+ * @u_volt:	Nominal voltage in microvolts corresponding to this OPP
+ * @dev_opp:	points back to the device_opp struct this opp belongs to
+ *
+ * This structure stores the OPP information for a given device.
+ */
+struct opp {
+	struct list_head node;
+
+	bool available;
+	unsigned long rate;
+	unsigned long u_volt;
+
+	struct device_opp *dev_opp;
+};
+
+/**
+ * struct device_opp - Device opp structure
+ * @node:	list node - contains the devices with OPPs that
+ *		have been registered. Nodes once added are not modified in this
+ *		list.
+ *		RCU usage: nodes are not modified in the list of device_opp,
+ *		however addition is possible and is secured by dev_opp_list_lock
+ * @dev:	device pointer
+ * @opp_list:	list of opps
+ *
+ * This is an internal data structure maintaining the link to opps attached to
+ * a device. This structure is not meant to be shared to users as it is
+ * meant for book keeping and private to OPP library
+ */
+struct device_opp {
+	struct list_head node;
+
+	struct device *dev;
+	struct list_head opp_list;
+};
+
+/*
+ * The root of the list of all devices. All device_opp structures branch off
+ * from here, with each device_opp containing the list of opp it supports in
+ * various states of availability.
+ */
+static LIST_HEAD(dev_opp_list);
+/* Lock to allow exclusive modification to the device and opp lists */
+static DEFINE_MUTEX(dev_opp_list_lock);
+
+/**
+ * find_device_opp() - find device_opp struct using device pointer
+ * @dev:	device pointer used to lookup device OPPs
+ *
+ * Search list of device OPPs for one containing matching device. Does a RCU
+ * reader operation to grab the pointer needed.
+ *
+ * Returns pointer to 'struct device_opp' if found, otherwise -ENODEV or
+ * -EINVAL based on type of error.
+ *
+ * Locking: This function must be called under rcu_read_lock(). device_opp
+ * is a RCU protected pointer. This means that device_opp is valid as long
+ * as we are under RCU lock.
+ */
+static struct device_opp *find_device_opp(struct device *dev)
+{
+	struct device_opp *tmp_dev_opp, *dev_opp = ERR_PTR(-ENODEV);
+
+	if (unlikely(IS_ERR_OR_NULL(dev))) {
+		pr_err("%s: Invalid parameters\n", __func__);
+		return ERR_PTR(-EINVAL);
+	}
+
+	list_for_each_entry_rcu(tmp_dev_opp, &dev_opp_list, node) {
+		if (tmp_dev_opp->dev == dev) {
+			dev_opp = tmp_dev_opp;
+			break;
+		}
+	}
+
+	return dev_opp;
+}
+
+/**
+ * opp_get_voltage() - Gets the voltage corresponding to an available opp
+ * @opp:	opp for which voltage has to be returned for
+ *
+ * Return voltage in micro volt corresponding to the opp, else
+ * return 0
+ *
+ * Locking: This function must be called under rcu_read_lock(). opp is a rcu
+ * protected pointer. This means that opp which could have been fetched by
+ * opp_find_freq_{exact,ceil,floor} functions is valid as long as we are
+ * under RCU lock. The pointer returned by the opp_find_freq family must be
+ * used in the same section as the usage of this function with the pointer
+ * prior to unlocking with rcu_read_unlock() to maintain the integrity of the
+ * pointer.
+ */
+unsigned long opp_get_voltage(struct opp *opp)
+{
+	struct opp *tmp_opp;
+	unsigned long v = 0;
+
+	tmp_opp = rcu_dereference(opp);
+	if (unlikely(IS_ERR_OR_NULL(tmp_opp)) || !tmp_opp->available)
+		pr_err("%s: Invalid parameters\n", __func__);
+	else
+		v = tmp_opp->u_volt;
+
+	return v;
+}
+
+/**
+ * opp_get_freq() - Gets the frequency corresponding to an available opp
+ * @opp:	opp for which frequency has to be returned for
+ *
+ * Return frequency in hertz corresponding to the opp, else
+ * return 0
+ *
+ * Locking: This function must be called under rcu_read_lock(). opp is a rcu
+ * protected pointer. This means that opp which could have been fetched by
+ * opp_find_freq_{exact,ceil,floor} functions is valid as long as we are
+ * under RCU lock. The pointer returned by the opp_find_freq family must be
+ * used in the same section as the usage of this function with the pointer
+ * prior to unlocking with rcu_read_unlock() to maintain the integrity of the
+ * pointer.
+ */
+unsigned long opp_get_freq(struct opp *opp)
+{
+	struct opp *tmp_opp;
+	unsigned long f = 0;
+
+	tmp_opp = rcu_dereference(opp);
+	if (unlikely(IS_ERR_OR_NULL(tmp_opp)) || !tmp_opp->available)
+		pr_err("%s: Invalid parameters\n", __func__);
+	else
+		f = tmp_opp->rate;
+
+	return f;
+}
+
+/**
+ * opp_get_opp_count() - Get number of opps available in the opp list
+ * @dev:	device for which we do this operation
+ *
+ * This function returns the number of available opps if there are any,
+ * else returns 0 if none or the corresponding error value.
+ *
+ * Locking: This function must be called under rcu_read_lock(). This function
+ * internally references two RCU protected structures: device_opp and opp which
+ * are safe as long as we are under a common RCU locked section.
+ */
+int opp_get_opp_count(struct device *dev)
+{
+	struct device_opp *dev_opp;
+	struct opp *temp_opp;
+	int count = 0;
+
+	dev_opp = find_device_opp(dev);
+	if (IS_ERR(dev_opp)) {
+		int r = PTR_ERR(dev_opp);
+		dev_err(dev, "%s: device OPP not found (%d)\n", __func__, r);
+		return r;
+	}
+
+	list_for_each_entry_rcu(temp_opp, &dev_opp->opp_list, node) {
+		if (temp_opp->available)
+			count++;
+	}
+
+	return count;
+}
+
+/**
+ * opp_find_freq_exact() - search for an exact frequency
+ * @dev:		device for which we do this operation
+ * @freq:		frequency to search for
+ * @is_available:	true/false - match for available opp
+ *
+ * Searches for exact match in the opp list and returns pointer to the matching
+ * opp if found, else returns ERR_PTR in case of error and should be handled
+ * using IS_ERR.
+ *
+ * Note: available is a modifier for the search. if available=true, then the
+ * match is for exact matching frequency and is available in the stored OPP
+ * table. if false, the match is for exact frequency which is not available.
+ *
+ * This provides a mechanism to enable an opp which is not available currently
+ * or the opposite as well.
+ *
+ * Locking: This function must be called under rcu_read_lock(). opp is a rcu
+ * protected pointer. The reason for the same is that the opp pointer which is
+ * returned will remain valid for use with opp_get_{voltage, freq} only while
+ * under the locked area. The pointer returned must be used prior to unlocking
+ * with rcu_read_unlock() to maintain the integrity of the pointer.
+ */
+struct opp *opp_find_freq_exact(struct device *dev, unsigned long freq,
+				bool available)
+{
+	struct device_opp *dev_opp;
+	struct opp *temp_opp, *opp = ERR_PTR(-ENODEV);
+
+	dev_opp = find_device_opp(dev);
+	if (IS_ERR(dev_opp)) {
+		int r = PTR_ERR(dev_opp);
+		dev_err(dev, "%s: device OPP not found (%d)\n", __func__, r);
+		return ERR_PTR(r);
+	}
+
+	list_for_each_entry_rcu(temp_opp, &dev_opp->opp_list, node) {
+		if (temp_opp->available == available &&
+				temp_opp->rate == freq) {
+			opp = temp_opp;
+			break;
+		}
+	}
+
+	return opp;
+}
+
+/**
+ * opp_find_freq_ceil() - Search for an rounded ceil freq
+ * @dev:	device for which we do this operation
+ * @freq:	Start frequency
+ *
+ * Search for the matching ceil *available* OPP from a starting freq
+ * for a device.
+ *
+ * Returns matching *opp and refreshes *freq accordingly, else returns
+ * ERR_PTR in case of error and should be handled using IS_ERR.
+ *
+ * Locking: This function must be called under rcu_read_lock(). opp is a rcu
+ * protected pointer. The reason for the same is that the opp pointer which is
+ * returned will remain valid for use with opp_get_{voltage, freq} only while
+ * under the locked area. The pointer returned must be used prior to unlocking
+ * with rcu_read_unlock() to maintain the integrity of the pointer.
+ */
+struct opp *opp_find_freq_ceil(struct device *dev, unsigned long *freq)
+{
+	struct device_opp *dev_opp;
+	struct opp *temp_opp, *opp = ERR_PTR(-ENODEV);
+
+	if (!dev || !freq) {
+		dev_err(dev, "%s: Invalid argument freq=%p\n", __func__, freq);
+		return ERR_PTR(-EINVAL);
+	}
+
+	dev_opp = find_device_opp(dev);
+	if (IS_ERR(dev_opp))
+		return opp;
+
+	list_for_each_entry_rcu(temp_opp, &dev_opp->opp_list, node) {
+		if (temp_opp->available && temp_opp->rate >= *freq) {
+			opp = temp_opp;
+			*freq = opp->rate;
+			break;
+		}
+	}
+
+	return opp;
+}
+
+/**
+ * opp_find_freq_floor() - Search for a rounded floor freq
+ * @dev:	device for which we do this operation
+ * @freq:	Start frequency
+ *
+ * Search for the matching floor *available* OPP from a starting freq
+ * for a device.
+ *
+ * Returns matching *opp and refreshes *freq accordingly, else returns
+ * ERR_PTR in case of error and should be handled using IS_ERR.
+ *
+ * Locking: This function must be called under rcu_read_lock(). opp is a rcu
+ * protected pointer. The reason for the same is that the opp pointer which is
+ * returned will remain valid for use with opp_get_{voltage, freq} only while
+ * under the locked area. The pointer returned must be used prior to unlocking
+ * with rcu_read_unlock() to maintain the integrity of the pointer.
+ */
+struct opp *opp_find_freq_floor(struct device *dev, unsigned long *freq)
+{
+	struct device_opp *dev_opp;
+	struct opp *temp_opp, *opp = ERR_PTR(-ENODEV);
+
+	if (!dev || !freq) {
+		dev_err(dev, "%s: Invalid argument freq=%p\n", __func__, freq);
+		return ERR_PTR(-EINVAL);
+	}
+
+	dev_opp = find_device_opp(dev);
+	if (IS_ERR(dev_opp))
+		return opp;
+
+	list_for_each_entry_rcu(temp_opp, &dev_opp->opp_list, node) {
+		if (temp_opp->available) {
+			/* go to the next node, before choosing prev */
+			if (temp_opp->rate > *freq)
+				break;
+			else
+				opp = temp_opp;
+		}
+	}
+	if (!IS_ERR(opp))
+		*freq = opp->rate;
+
+	return opp;
+}
+
+/**
+ * opp_add()  - Add an OPP table from a table definitions
+ * @dev:	device for which we do this operation
+ * @freq:	Frequency in Hz for this OPP
+ * @u_volt:	Voltage in uVolts for this OPP
+ *
+ * This function adds an opp definition to the opp list and returns status.
+ * The opp is made available by default and it can be controlled using
+ * opp_enable/disable functions.
+ *
+ * Locking: The internal device_opp and opp structures are RCU protected.
+ * Hence this function internally uses RCU updater strategy with mutex locks
+ * to keep the integrity of the internal data structures. Callers should ensure
+ * that this function is *NOT* called under RCU protection or in contexts where
+ * mutex cannot be locked.
+ */
+int opp_add(struct device *dev, unsigned long freq, unsigned long u_volt)
+{
+	struct device_opp *dev_opp = NULL;
+	struct opp *opp, *new_opp;
+	struct list_head *head;
+
+	/* allocate new OPP node */
+	new_opp = kzalloc(sizeof(struct opp), GFP_KERNEL);
+	if (!new_opp) {
+		dev_warn(dev, "%s: Unable to create new OPP node\n", __func__);
+		return -ENOMEM;
+	}
+
+	/* Hold our list modification lock here */
+	mutex_lock(&dev_opp_list_lock);
+
+	/* Check for existing list for 'dev' */
+	dev_opp = find_device_opp(dev);
+	if (IS_ERR(dev_opp)) {
+		/*
+		 * Allocate a new device OPP table. In the infrequent case
+		 * where a new device is needed to be added, we pay this
+		 * penalty.
+		 */
+		dev_opp = kzalloc(sizeof(struct device_opp), GFP_KERNEL);
+		if (!dev_opp) {
+			mutex_unlock(&dev_opp_list_lock);
+			kfree(new_opp);
+			dev_warn(dev,
+				"%s: Unable to create device OPP structure\n",
+				__func__);
+			return -ENOMEM;
+		}
+
+		dev_opp->dev = dev;
+		INIT_LIST_HEAD(&dev_opp->opp_list);
+
+		/* Secure the device list modification */
+		list_add_rcu(&dev_opp->node, &dev_opp_list);
+	}
+
+	/* populate the opp table */
+	new_opp->dev_opp = dev_opp;
+	new_opp->rate = freq;
+	new_opp->u_volt = u_volt;
+	new_opp->available = true;
+
+	/* Insert new OPP in order of increasing frequency */
+	head = &dev_opp->opp_list;
+	list_for_each_entry_rcu(opp, &dev_opp->opp_list, node) {
+		if (new_opp->rate < opp->rate)
+			break;
+		else
+			head = &opp->node;
+	}
+
+	list_add_rcu(&new_opp->node, head);
+	mutex_unlock(&dev_opp_list_lock);
+
+	return 0;
+}
+
+/**
+ * opp_set_availability() - helper to set the availability of an opp
+ * @dev:		device for which we do this operation
+ * @freq:		OPP frequency to modify availability
+ * @availability_req:	availability status requested for this opp
+ *
+ * Set the availability of an OPP with an RCU operation, opp_{enable,disable}
+ * share a common logic which is isolated here.
+ *
+ * Returns -EINVAL for bad pointers, -ENOMEM if no memory available for the
+ * copy operation, returns 0 if no modifcation was done OR modification was
+ * successful.
+ *
+ * Locking: The internal device_opp and opp structures are RCU protected.
+ * Hence this function internally uses RCU updater strategy with mutex locks to
+ * keep the integrity of the internal data structures. Callers should ensure
+ * that this function is *NOT* called under RCU protection or in contexts where
+ * mutex locking or synchronize_rcu() blocking calls cannot be used.
+ */
+static int opp_set_availability(struct device *dev, unsigned long freq,
+		bool availability_req)
+{
+	struct device_opp *tmp_dev_opp, *dev_opp = NULL;
+	struct opp *new_opp, *tmp_opp, *opp = ERR_PTR(-ENODEV);
+	int r = 0;
+
+	/* keep the node allocated */
+	new_opp = kmalloc(sizeof(struct opp), GFP_KERNEL);
+	if (!new_opp) {
+		dev_warn(dev, "%s: Unable to create OPP\n", __func__);
+		return -ENOMEM;
+	}
+
+	mutex_lock(&dev_opp_list_lock);
+
+	/* Find the device_opp */
+	list_for_each_entry(tmp_dev_opp, &dev_opp_list, node) {
+		if (dev == tmp_dev_opp->dev) {
+			dev_opp = tmp_dev_opp;
+			break;
+		}
+	}
+	if (IS_ERR(dev_opp)) {
+		r = PTR_ERR(dev_opp);
+		dev_warn(dev, "%s: Device OPP not found (%d)\n", __func__, r);
+		goto unlock;
+	}
+
+	/* Do we have the frequency? */
+	list_for_each_entry(tmp_opp, &dev_opp->opp_list, node) {
+		if (tmp_opp->rate == freq) {
+			opp = tmp_opp;
+			break;
+		}
+	}
+	if (IS_ERR(opp)) {
+		r = PTR_ERR(opp);
+		goto unlock;
+	}
+
+	/* Is update really needed? */
+	if (opp->available == availability_req)
+		goto unlock;
+	/* copy the old data over */
+	*new_opp = *opp;
+
+	/* plug in new node */
+	new_opp->available = availability_req;
+
+	list_replace_rcu(&opp->node, &new_opp->node);
+	mutex_unlock(&dev_opp_list_lock);
+	synchronize_rcu();
+
+	/* clean up old opp */
+	new_opp = opp;
+	goto out;
+
+unlock:
+	mutex_unlock(&dev_opp_list_lock);
+out:
+	kfree(new_opp);
+	return r;
+}
+
+/**
+ * opp_enable() - Enable a specific OPP
+ * @dev:	device for which we do this operation
+ * @freq:	OPP frequency to enable
+ *
+ * Enables a provided opp. If the operation is valid, this returns 0, else the
+ * corresponding error value. It is meant to be used for users an OPP available
+ * after being temporarily made unavailable with opp_disable.
+ *
+ * Locking: The internal device_opp and opp structures are RCU protected.
+ * Hence this function indirectly uses RCU and mutex locks to keep the
+ * integrity of the internal data structures. Callers should ensure that
+ * this function is *NOT* called under RCU protection or in contexts where
+ * mutex locking or synchronize_rcu() blocking calls cannot be used.
+ */
+int opp_enable(struct device *dev, unsigned long freq)
+{
+	return opp_set_availability(dev, freq, true);
+}
+
+/**
+ * opp_disable() - Disable a specific OPP
+ * @dev:	device for which we do this operation
+ * @freq:	OPP frequency to disable
+ *
+ * Disables a provided opp. If the operation is valid, this returns
+ * 0, else the corresponding error value. It is meant to be a temporary
+ * control by users to make this OPP not available until the circumstances are
+ * right to make it available again (with a call to opp_enable).
+ *
+ * Locking: The internal device_opp and opp structures are RCU protected.
+ * Hence this function indirectly uses RCU and mutex locks to keep the
+ * integrity of the internal data structures. Callers should ensure that
+ * this function is *NOT* called under RCU protection or in contexts where
+ * mutex locking or synchronize_rcu() blocking calls cannot be used.
+ */
+int opp_disable(struct device *dev, unsigned long freq)
+{
+	return opp_set_availability(dev, freq, false);
+}
+
+#ifdef CONFIG_CPU_FREQ
+/**
+ * opp_init_cpufreq_table() - create a cpufreq table for a device
+ * @dev:	device for which we do this operation
+ * @table:	Cpufreq table returned back to caller
+ *
+ * Generate a cpufreq table for a provided device- this assumes that the
+ * opp list is already initialized and ready for usage.
+ *
+ * This function allocates required memory for the cpufreq table. It is
+ * expected that the caller does the required maintenance such as freeing
+ * the table as required.
+ *
+ * Returns -EINVAL for bad pointers, -ENODEV if the device is not found, -ENOMEM
+ * if no memory available for the operation (table is not populated), returns 0
+ * if successful and table is populated.
+ *
+ * WARNING: It is  important for the callers to ensure refreshing their copy of
+ * the table if any of the mentioned functions have been invoked in the interim.
+ *
+ * Locking: The internal device_opp and opp structures are RCU protected.
+ * To simplify the logic, we pretend we are updater and hold relevant mutex here
+ * Callers should ensure that this function is *NOT* called under RCU protection
+ * or in contexts where mutex locking cannot be used.
+ */
+int opp_init_cpufreq_table(struct device *dev,
+			    struct cpufreq_frequency_table **table)
+{
+	struct device_opp *dev_opp;
+	struct opp *opp;
+	struct cpufreq_frequency_table *freq_table;
+	int i = 0;
+
+	/* Pretend as if I am an updater */
+	mutex_lock(&dev_opp_list_lock);
+
+	dev_opp = find_device_opp(dev);
+	if (IS_ERR(dev_opp)) {
+		int r = PTR_ERR(dev_opp);
+		mutex_unlock(&dev_opp_list_lock);
+		dev_err(dev, "%s: Device OPP not found (%d)\n", __func__, r);
+		return r;
+	}
+
+	freq_table = kzalloc(sizeof(struct cpufreq_frequency_table) *
+			     (opp_get_opp_count(dev) + 1), GFP_KERNEL);
+	if (!freq_table) {
+		mutex_unlock(&dev_opp_list_lock);
+		dev_warn(dev, "%s: Unable to allocate frequency table\n",
+			__func__);
+		return -ENOMEM;
+	}
+
+	list_for_each_entry(opp, &dev_opp->opp_list, node) {
+		if (opp->available) {
+			freq_table[i].index = i;
+			freq_table[i].frequency = opp->rate / 1000;
+			i++;
+		}
+	}
+	mutex_unlock(&dev_opp_list_lock);
+
+	freq_table[i].index = i;
+	freq_table[i].frequency = CPUFREQ_TABLE_END;
+
+	*table = &freq_table[0];
+
+	return 0;
+}
+#endif		/* CONFIG_CPU_FREQ */
diff --git a/include/linux/opp.h b/include/linux/opp.h
new file mode 100644
index 000000000000..5449945d589f
--- /dev/null
+++ b/include/linux/opp.h
@@ -0,0 +1,105 @@
+/*
+ * Generic OPP Interface
+ *
+ * Copyright (C) 2009-2010 Texas Instruments Incorporated.
+ *	Nishanth Menon
+ *	Romit Dasgupta
+ *	Kevin Hilman
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __LINUX_OPP_H__
+#define __LINUX_OPP_H__
+
+#include <linux/err.h>
+#include <linux/cpufreq.h>
+
+struct opp;
+
+#if defined(CONFIG_PM_OPP)
+
+unsigned long opp_get_voltage(struct opp *opp);
+
+unsigned long opp_get_freq(struct opp *opp);
+
+int opp_get_opp_count(struct device *dev);
+
+struct opp *opp_find_freq_exact(struct device *dev, unsigned long freq,
+				bool available);
+
+struct opp *opp_find_freq_floor(struct device *dev, unsigned long *freq);
+
+struct opp *opp_find_freq_ceil(struct device *dev, unsigned long *freq);
+
+int opp_add(struct device *dev, unsigned long freq, unsigned long u_volt);
+
+int opp_enable(struct device *dev, unsigned long freq);
+
+int opp_disable(struct device *dev, unsigned long freq);
+
+#else
+static inline unsigned long opp_get_voltage(struct opp *opp)
+{
+	return 0;
+}
+
+static inline unsigned long opp_get_freq(struct opp *opp)
+{
+	return 0;
+}
+
+static inline int opp_get_opp_count(struct device *dev)
+{
+	return 0;
+}
+
+static inline struct opp *opp_find_freq_exact(struct device *dev,
+					unsigned long freq, bool available)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static inline struct opp *opp_find_freq_floor(struct device *dev,
+					unsigned long *freq)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static inline struct opp *opp_find_freq_ceil(struct device *dev,
+					unsigned long *freq)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static inline int opp_add(struct device *dev, unsigned long freq,
+					unsigned long u_volt)
+{
+	return -EINVAL;
+}
+
+static inline int opp_enable(struct device *dev, unsigned long freq)
+{
+	return 0;
+}
+
+static inline int opp_disable(struct device *dev, unsigned long freq)
+{
+	return 0;
+}
+#endif		/* CONFIG_PM */
+
+#if defined(CONFIG_CPU_FREQ) && defined(CONFIG_PM_OPP)
+int opp_init_cpufreq_table(struct device *dev,
+			    struct cpufreq_frequency_table **table);
+#else
+static inline int opp_init_cpufreq_table(struct device *dev,
+			    struct cpufreq_frequency_table **table)
+{
+	return -EINVAL;
+}
+#endif		/* CONFIG_CPU_FREQ */
+
+#endif		/* __LINUX_OPP_H__ */
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index e45894c696ee..29bff6117abc 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -245,3 +245,17 @@ config PM_OPS
 	bool
 	depends on PM_SLEEP || PM_RUNTIME
 	default y
+
+config PM_OPP
+	bool "Operating Performance Point (OPP) Layer library"
+	depends on PM
+	---help---
+	  SOCs have a standard set of tuples consisting of frequency and
+	  voltage pairs that the device will support per voltage domain. This
+	  is called Operating Performance Point or OPP. The actual definitions
+	  of OPP varies over silicon within the same family of devices.
+
+	  OPP layer organizes the data internally using device pointers
+	  representing individual voltage domains and provides SOC
+	  implementations a ready to use framework to manage OPPs.
+	  For more information, read <file:Documentation/power/opp.txt>
-- 
cgit v1.2.3


From 43f974cdb4ab6d65f849610deb9ef738d62b2e65 Mon Sep 17 00:00:00 2001
From: Nick Bowler <nbowler@elliptictech.com>
Date: Mon, 18 Oct 2010 11:22:05 +0200
Subject: netfilter: install missing ebtables headers for userspace

The ebt_ip6.h and ebt_nflog.h headers are not not known to Kbuild and
therefore not installed by make headers_install.  Fix that up.

Signed-off-by: Nick Bowler <nbowler@elliptictech.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter_bridge/Kbuild | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netfilter_bridge/Kbuild b/include/linux/netfilter_bridge/Kbuild
index d4d78672873e..e48f1a3f5a4a 100644
--- a/include/linux/netfilter_bridge/Kbuild
+++ b/include/linux/netfilter_bridge/Kbuild
@@ -3,11 +3,13 @@ header-y += ebt_among.h
 header-y += ebt_arp.h
 header-y += ebt_arpreply.h
 header-y += ebt_ip.h
+header-y += ebt_ip6.h
 header-y += ebt_limit.h
 header-y += ebt_log.h
 header-y += ebt_mark_m.h
 header-y += ebt_mark_t.h
 header-y += ebt_nat.h
+header-y += ebt_nflog.h
 header-y += ebt_pkttype.h
 header-y += ebt_redirect.h
 header-y += ebt_stp.h
-- 
cgit v1.2.3


From f1f8c6cbe6f08f93ac2a4ca19625891d8a82b7f8 Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Mon, 18 Oct 2010 15:00:18 +0200
Subject: can: mcp251x: Don't use pdata->model for chip selection anymore

Since commit e446630c960946b5c1762e4eadb618becef599e7, i.e. v2.6.35-rc1,
the mcp251x chip model can be selected via the modalias member in the
struct spi_board_info. The driver stores the actual model in the
struct mcp251x_platform_data.

From the driver point of view the platform_data should be read only.
Since all in-tree users of the mcp251x have already been converted to
the modalias method, this patch moves the "model" member from the
struct mcp251x_platform_data to the driver's private data structure.

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Christian Pellegrin <chripell@fsfe.org>
Cc: Marc Zyngier <maz@misterjones.org>
---
 drivers/net/can/mcp251x.c            | 24 ++++++++++++------------
 include/linux/can/platform/mcp251x.h |  4 ----
 2 files changed, 12 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/can/mcp251x.c b/drivers/net/can/mcp251x.c
index f5e2edd09ce4..0386bed43551 100644
--- a/drivers/net/can/mcp251x.c
+++ b/drivers/net/can/mcp251x.c
@@ -38,14 +38,14 @@
  * static struct mcp251x_platform_data mcp251x_info = {
  *         .oscillator_frequency = 8000000,
  *         .board_specific_setup = &mcp251x_setup,
- *         .model = CAN_MCP251X_MCP2510,
  *         .power_enable = mcp251x_power_enable,
  *         .transceiver_enable = NULL,
  * };
  *
  * static struct spi_board_info spi_board_info[] = {
  *         {
- *                 .modalias = "mcp251x",
+ *                 .modalias = "mcp2510",
+ *			// or "mcp2515" depending on your controller
  *                 .platform_data = &mcp251x_info,
  *                 .irq = IRQ_EINT13,
  *                 .max_speed_hz = 2*1000*1000,
@@ -224,10 +224,16 @@ static struct can_bittiming_const mcp251x_bittiming_const = {
 	.brp_inc = 1,
 };
 
+enum mcp251x_model {
+	CAN_MCP251X_MCP2510	= 0x2510,
+	CAN_MCP251X_MCP2515	= 0x2515,
+};
+
 struct mcp251x_priv {
 	struct can_priv	   can;
 	struct net_device *net;
 	struct spi_device *spi;
+	enum mcp251x_model model;
 
 	struct mutex mcp_lock; /* SPI device lock */
 
@@ -362,10 +368,9 @@ static void mcp251x_write_bits(struct spi_device *spi, u8 reg,
 static void mcp251x_hw_tx_frame(struct spi_device *spi, u8 *buf,
 				int len, int tx_buf_idx)
 {
-	struct mcp251x_platform_data *pdata = spi->dev.platform_data;
 	struct mcp251x_priv *priv = dev_get_drvdata(&spi->dev);
 
-	if (pdata->model == CAN_MCP251X_MCP2510) {
+	if (priv->model == CAN_MCP251X_MCP2510) {
 		int i;
 
 		for (i = 1; i < TXBDAT_OFF + len; i++)
@@ -408,9 +413,8 @@ static void mcp251x_hw_rx_frame(struct spi_device *spi, u8 *buf,
 				int buf_idx)
 {
 	struct mcp251x_priv *priv = dev_get_drvdata(&spi->dev);
-	struct mcp251x_platform_data *pdata = spi->dev.platform_data;
 
-	if (pdata->model == CAN_MCP251X_MCP2510) {
+	if (priv->model == CAN_MCP251X_MCP2510) {
 		int i, len;
 
 		for (i = 1; i < RXBDAT_OFF; i++)
@@ -951,16 +955,12 @@ static int __devinit mcp251x_can_probe(struct spi_device *spi)
 	struct net_device *net;
 	struct mcp251x_priv *priv;
 	struct mcp251x_platform_data *pdata = spi->dev.platform_data;
-	int model = spi_get_device_id(spi)->driver_data;
 	int ret = -ENODEV;
 
 	if (!pdata)
 		/* Platform data is required for osc freq */
 		goto error_out;
 
-	if (model)
-		pdata->model = model;
-
 	/* Allocate can/net device */
 	net = alloc_candev(sizeof(struct mcp251x_priv), TX_ECHO_SKB_MAX);
 	if (!net) {
@@ -977,6 +977,7 @@ static int __devinit mcp251x_can_probe(struct spi_device *spi)
 	priv->can.clock.freq = pdata->oscillator_frequency / 2;
 	priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES |
 		CAN_CTRLMODE_LOOPBACK | CAN_CTRLMODE_LISTENONLY;
+	priv->model = spi_get_device_id(spi)->driver_data;
 	priv->net = net;
 	dev_set_drvdata(&spi->dev, priv);
 
@@ -1150,8 +1151,7 @@ static int mcp251x_can_resume(struct spi_device *spi)
 #define mcp251x_can_resume NULL
 #endif
 
-static struct spi_device_id mcp251x_id_table[] = {
-	{ "mcp251x", 	0 /* Use pdata.model */ },
+static const struct spi_device_id mcp251x_id_table[] = {
 	{ "mcp2510",	CAN_MCP251X_MCP2510 },
 	{ "mcp2515",	CAN_MCP251X_MCP2515 },
 	{ },
diff --git a/include/linux/can/platform/mcp251x.h b/include/linux/can/platform/mcp251x.h
index dba28268e651..8e20540043f5 100644
--- a/include/linux/can/platform/mcp251x.h
+++ b/include/linux/can/platform/mcp251x.h
@@ -12,7 +12,6 @@
 /**
  * struct mcp251x_platform_data - MCP251X SPI CAN controller platform data
  * @oscillator_frequency:       - oscillator frequency in Hz
- * @model:                      - actual type of chip
  * @board_specific_setup:       - called before probing the chip (power,reset)
  * @transceiver_enable:         - called to power on/off the transceiver
  * @power_enable:               - called to power on/off the mcp *and* the
@@ -25,9 +24,6 @@
 
 struct mcp251x_platform_data {
 	unsigned long oscillator_frequency;
-	int model;
-#define CAN_MCP251X_MCP2510 0x2510
-#define CAN_MCP251X_MCP2515 0x2515
 	int (*board_specific_setup)(struct spi_device *spi);
 	int (*transceiver_enable)(int enable);
 	int (*power_enable) (int enable);
-- 
cgit v1.2.3


From c2355e1ab910278a94d487b78590ee3c8eecd08a Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@tuxdriver.com>
Date: Wed, 13 Oct 2010 16:01:49 +0000
Subject: bonding: Fix bonding drivers improper modification of netpoll
 structure

The bonding driver currently modifies the netpoll structure in its xmit path
while sending frames from netpoll.  This is racy, as other cpus can access the
netpoll structure in parallel. Since the bonding driver points np->dev to a
slave device, other cpus can inadvertently attempt to send data directly to
slave devices, leading to improper locking with the bonding master, lost frames,
and deadlocks.  This patch fixes that up.

This patch also removes the real_dev pointer from the netpoll structure as that
data is really only used by bonding in the poll_controller, and we can emulate
its behavior by check each slave for IS_UP.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bonding/bond_main.c | 15 +++++++++------
 include/linux/netpoll.h         |  9 +++++++--
 net/core/netpoll.c              |  6 +++---
 3 files changed, 19 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 7703d35de65d..813cc2f8edd6 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -449,11 +449,9 @@ int bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb,
 	if (unlikely(bond->dev->priv_flags & IFF_IN_NETPOLL)) {
 		struct netpoll *np = bond->dev->npinfo->netpoll;
 		slave_dev->npinfo = bond->dev->npinfo;
-		np->real_dev = np->dev = skb->dev;
 		slave_dev->priv_flags |= IFF_IN_NETPOLL;
-		netpoll_send_skb(np, skb);
+		netpoll_send_skb_on_dev(np, skb, slave_dev);
 		slave_dev->priv_flags &= ~IFF_IN_NETPOLL;
-		np->dev = bond->dev;
 	} else
 #endif
 		dev_queue_xmit(skb);
@@ -1332,9 +1330,14 @@ static bool slaves_support_netpoll(struct net_device *bond_dev)
 
 static void bond_poll_controller(struct net_device *bond_dev)
 {
-	struct net_device *dev = bond_dev->npinfo->netpoll->real_dev;
-	if (dev != bond_dev)
-		netpoll_poll_dev(dev);
+	struct bonding *bond = netdev_priv(bond_dev);
+	struct slave *slave;
+	int i;
+
+	bond_for_each_slave(bond, slave, i) {
+		if (slave->dev && IS_UP(slave->dev))
+			netpoll_poll_dev(slave->dev);
+	}
 }
 
 static void bond_netpoll_cleanup(struct net_device *bond_dev)
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 50d8009be86c..79358bb712c6 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -14,7 +14,6 @@
 
 struct netpoll {
 	struct net_device *dev;
-	struct net_device *real_dev;
 	char dev_name[IFNAMSIZ];
 	const char *name;
 	void (*rx_hook)(struct netpoll *, int, char *, int);
@@ -53,7 +52,13 @@ void netpoll_set_trap(int trap);
 void __netpoll_cleanup(struct netpoll *np);
 void netpoll_cleanup(struct netpoll *np);
 int __netpoll_rx(struct sk_buff *skb);
-void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb);
+void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb,
+			     struct net_device *dev);
+static inline void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
+{
+	netpoll_send_skb_on_dev(np, skb, np->dev);
+}
+
 
 
 #ifdef CONFIG_NETPOLL
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 537e01afd81b..4e98ffac3af0 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -288,11 +288,11 @@ static int netpoll_owner_active(struct net_device *dev)
 	return 0;
 }
 
-void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
+void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb,
+			     struct net_device *dev)
 {
 	int status = NETDEV_TX_BUSY;
 	unsigned long tries;
-	struct net_device *dev = np->dev;
 	const struct net_device_ops *ops = dev->netdev_ops;
 	/* It is up to the caller to keep npinfo alive. */
 	struct netpoll_info *npinfo = np->dev->npinfo;
@@ -346,7 +346,7 @@ void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
 		schedule_delayed_work(&npinfo->tx_work,0);
 	}
 }
-EXPORT_SYMBOL(netpoll_send_skb);
+EXPORT_SYMBOL(netpoll_send_skb_on_dev);
 
 void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
 {
-- 
cgit v1.2.3


From 620162505e5d46bc4494b1761743e4b0b3bf8e16 Mon Sep 17 00:00:00 2001
From: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Date: Tue, 5 Oct 2010 18:01:51 +0900
Subject: lockdep: Add improved subclass caching

Current lockdep_map only caches one class with subclass == 0,
and looks up hash table of classes when subclass != 0.

It seems that this has no problem because the case of
subclass != 0 is rare. But locks of struct rq are
acquired with subclass == 1 when task migration is executed.
Task migration is high frequent event, so I modified lockdep
to cache subclasses.

I measured the score of perf bench sched messaging.
This patch has slightly but certain (order of milli seconds
or 10 milli seconds) effect when lots of tasks are running.
I'll show the result in the tail of this description.

NR_LOCKDEP_CACHING_CLASSES specifies how many classes can be
cached in the instances of lockdep_map.
I discussed with Peter Zijlstra in LinuxCon Japan about
this approach and he taught me that caching every subclasses(8)
is cleary waste of memory. So number of cached classes
should be configurable.

=== Score comparison of benchmarks ===
# "min" means best score, and "max" means worst score

for i in `seq 1 10`; do ./perf bench -f simple sched messaging; done

before: min: 0.565000, max: 0.583000, avg: 0.572500
after:  min: 0.559000, max: 0.568000, avg: 0.563300

# with more processes
for i in `seq 1 10`; do ./perf bench -f simple sched messaging -g 40; done

before: min: 2.274000, max: 2.298000, avg: 2.286300
after:  min: 2.242000, max: 2.270000, avg: 2.259700

Signed-off-by: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286269311-28336-2-git-send-email-mitake@dcl.info.waseda.ac.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/lockdep.h | 13 ++++++++++++-
 kernel/lockdep.c        | 25 ++++++++++++++++++-------
 2 files changed, 30 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 06aed8305bf3..2186a64ee4b5 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -31,6 +31,17 @@ extern int lock_stat;
 
 #define MAX_LOCKDEP_SUBCLASSES		8UL
 
+/*
+ * NR_LOCKDEP_CACHING_CLASSES ... Number of classes
+ * cached in the instance of lockdep_map
+ *
+ * Currently main class (subclass == 0) and signle depth subclass
+ * are cached in lockdep_map. This optimization is mainly targeting
+ * on rq->lock. double_rq_lock() acquires this highly competitive with
+ * single depth.
+ */
+#define NR_LOCKDEP_CACHING_CLASSES	2
+
 /*
  * Lock-classes are keyed via unique addresses, by embedding the
  * lockclass-key into the kernel (or module) .data section. (For
@@ -138,7 +149,7 @@ void clear_lock_stats(struct lock_class *class);
  */
 struct lockdep_map {
 	struct lock_class_key		*key;
-	struct lock_class		*class_cache;
+	struct lock_class		*class_cache[NR_LOCKDEP_CACHING_CLASSES];
 	const char			*name;
 #ifdef CONFIG_LOCK_STAT
 	int				cpu;
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 84baa71cfda5..bc4d32871f9a 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -774,7 +774,9 @@ out_unlock_set:
 	raw_local_irq_restore(flags);
 
 	if (!subclass || force)
-		lock->class_cache = class;
+		lock->class_cache[0] = class;
+	else if (subclass < NR_LOCKDEP_CACHING_CLASSES)
+		lock->class_cache[subclass] = class;
 
 	if (DEBUG_LOCKS_WARN_ON(class->subclass != subclass))
 		return NULL;
@@ -2679,7 +2681,11 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this,
 void lockdep_init_map(struct lockdep_map *lock, const char *name,
 		      struct lock_class_key *key, int subclass)
 {
-	lock->class_cache = NULL;
+	int i;
+
+	for (i = 0; i < NR_LOCKDEP_CACHING_CLASSES; i++)
+		lock->class_cache[i] = NULL;
+
 #ifdef CONFIG_LOCK_STAT
 	lock->cpu = raw_smp_processor_id();
 #endif
@@ -2750,10 +2756,10 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass,
 	if (lock->key == &__lockdep_no_validate__)
 		check = 1;
 
-	if (!subclass)
-		class = lock->class_cache;
+	if (subclass < NR_LOCKDEP_CACHING_CLASSES)
+		class = lock->class_cache[subclass];
 	/*
-	 * Not cached yet or subclass?
+	 * Not cached?
 	 */
 	if (unlikely(!class)) {
 		class = register_lock_class(lock, subclass, 0);
@@ -2918,7 +2924,7 @@ static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock)
 		return 1;
 
 	if (hlock->references) {
-		struct lock_class *class = lock->class_cache;
+		struct lock_class *class = lock->class_cache[0];
 
 		if (!class)
 			class = look_up_lock_class(lock, 0);
@@ -3559,7 +3565,12 @@ void lockdep_reset_lock(struct lockdep_map *lock)
 		if (list_empty(head))
 			continue;
 		list_for_each_entry_safe(class, next, head, hash_entry) {
-			if (unlikely(class == lock->class_cache)) {
+			int match = 0;
+
+			for (j = 0; j < NR_LOCKDEP_CACHING_CLASSES; j++)
+				match |= class == lock->class_cache[j];
+
+			if (unlikely(match)) {
 				if (debug_locks_off_graph_unlock())
 					WARN_ON(1);
 				goto out_restore;
-- 
cgit v1.2.3


From e360adbe29241a0194e10e20595360dd7b98a2b3 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 14 Oct 2010 14:01:34 +0800
Subject: irq_work: Add generic hardirq context callbacks

Provide a mechanism that allows running code in IRQ context. It is
most useful for NMI code that needs to interact with the rest of the
system -- like wakeup a task to drain buffers.

Perf currently has such a mechanism, so extract that and provide it as
a generic feature, independent of perf so that others may also
benefit.

The IRQ context callback is generated through self-IPIs where
possible, or on architectures like powerpc the decrementer (the
built-in timer facility) is set to generate an interrupt immediately.

Architectures that don't have anything like this get to do with a
callback from the timer tick. These architectures can call
irq_work_run() at the tail of any IRQ handlers that might enqueue such
work (like the perf IRQ handler) to avoid undue latencies in
processing the work.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Kyle McMartin <kyle@mcmartin.ca>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
[ various fixes ]
Signed-off-by: Huang Ying <ying.huang@intel.com>
LKML-Reference: <1287036094.7768.291.camel@yhuang-dev>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/alpha/Kconfig                   |   1 +
 arch/alpha/include/asm/perf_event.h  |   5 --
 arch/alpha/kernel/time.c             |  30 +++----
 arch/arm/Kconfig                     |   1 +
 arch/arm/include/asm/perf_event.h    |  12 ---
 arch/arm/kernel/perf_event.c         |   8 +-
 arch/frv/Kconfig                     |   1 +
 arch/frv/lib/Makefile                |   2 +-
 arch/frv/lib/perf_event.c            |  19 ----
 arch/parisc/Kconfig                  |   1 +
 arch/parisc/include/asm/perf_event.h |   3 +-
 arch/powerpc/Kconfig                 |   1 +
 arch/powerpc/include/asm/paca.h      |   2 +-
 arch/powerpc/kernel/time.c           |  42 ++++-----
 arch/s390/Kconfig                    |   1 +
 arch/s390/include/asm/perf_event.h   |   3 +-
 arch/sh/Kconfig                      |   1 +
 arch/sh/include/asm/perf_event.h     |   7 --
 arch/sparc/Kconfig                   |   2 +
 arch/sparc/include/asm/perf_event.h  |   4 -
 arch/sparc/kernel/pcr.c              |   8 +-
 arch/x86/Kconfig                     |   1 +
 arch/x86/include/asm/entry_arch.h    |   4 +-
 arch/x86/include/asm/hardirq.h       |   2 +-
 arch/x86/include/asm/hw_irq.h        |   2 +-
 arch/x86/include/asm/irq_vectors.h   |   4 +-
 arch/x86/kernel/Makefile             |   1 +
 arch/x86/kernel/cpu/perf_event.c     |  19 ----
 arch/x86/kernel/entry_64.S           |   6 +-
 arch/x86/kernel/irq.c                |   8 +-
 arch/x86/kernel/irq_work.c           |  30 +++++++
 arch/x86/kernel/irqinit.c            |   6 +-
 include/linux/irq_work.h             |  20 +++++
 include/linux/perf_event.h           |  11 +--
 init/Kconfig                         |   8 ++
 kernel/Makefile                      |   2 +
 kernel/irq_work.c                    | 164 +++++++++++++++++++++++++++++++++++
 kernel/perf_event.c                  | 104 ++--------------------
 kernel/timer.c                       |   7 +-
 39 files changed, 311 insertions(+), 242 deletions(-)
 delete mode 100644 arch/frv/lib/perf_event.c
 create mode 100644 arch/x86/kernel/irq_work.c
 create mode 100644 include/linux/irq_work.h
 create mode 100644 kernel/irq_work.c

(limited to 'include/linux')

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index b9647bb66d13..d04ccd73af45 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -9,6 +9,7 @@ config ALPHA
 	select HAVE_IDE
 	select HAVE_OPROFILE
 	select HAVE_SYSCALL_WRAPPERS
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select HAVE_DMA_ATTRS
 	help
diff --git a/arch/alpha/include/asm/perf_event.h b/arch/alpha/include/asm/perf_event.h
index 4157cd3c44a9..fe792ca818f6 100644
--- a/arch/alpha/include/asm/perf_event.h
+++ b/arch/alpha/include/asm/perf_event.h
@@ -1,11 +1,6 @@
 #ifndef __ASM_ALPHA_PERF_EVENT_H
 #define __ASM_ALPHA_PERF_EVENT_H
 
-/* Alpha only supports software events through this interface. */
-extern void set_perf_event_pending(void);
-
-#define PERF_EVENT_INDEX_OFFSET 0
-
 #ifdef CONFIG_PERF_EVENTS
 extern void init_hw_perf_events(void);
 #else
diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c
index 396af1799ea4..0f1d8493cfca 100644
--- a/arch/alpha/kernel/time.c
+++ b/arch/alpha/kernel/time.c
@@ -41,7 +41,7 @@
 #include <linux/init.h>
 #include <linux/bcd.h>
 #include <linux/profile.h>
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -83,25 +83,25 @@ static struct {
 
 unsigned long est_cycle_freq;
 
-#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_IRQ_WORK
 
-DEFINE_PER_CPU(u8, perf_event_pending);
+DEFINE_PER_CPU(u8, irq_work_pending);
 
-#define set_perf_event_pending_flag()  __get_cpu_var(perf_event_pending) = 1
-#define test_perf_event_pending()      __get_cpu_var(perf_event_pending)
-#define clear_perf_event_pending()     __get_cpu_var(perf_event_pending) = 0
+#define set_irq_work_pending_flag()  __get_cpu_var(irq_work_pending) = 1
+#define test_irq_work_pending()      __get_cpu_var(irq_work_pending)
+#define clear_irq_work_pending()     __get_cpu_var(irq_work_pending) = 0
 
-void set_perf_event_pending(void)
+void set_irq_work_pending(void)
 {
-	set_perf_event_pending_flag();
+	set_irq_work_pending_flag();
 }
 
-#else  /* CONFIG_PERF_EVENTS */
+#else  /* CONFIG_IRQ_WORK */
 
-#define test_perf_event_pending()      0
-#define clear_perf_event_pending()
+#define test_irq_work_pending()      0
+#define clear_irq_work_pending()
 
-#endif /* CONFIG_PERF_EVENTS */
+#endif /* CONFIG_IRQ_WORK */
 
 
 static inline __u32 rpcc(void)
@@ -191,9 +191,9 @@ irqreturn_t timer_interrupt(int irq, void *dev)
 
 	write_sequnlock(&xtime_lock);
 
-	if (test_perf_event_pending()) {
-		clear_perf_event_pending();
-		perf_event_do_pending();
+	if (test_irq_work_pending()) {
+		clear_irq_work_pending();
+		irq_work_run();
 	}
 
 #ifndef CONFIG_SMP
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 88c97bc7a6f5..7c0dfccd05bd 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -23,6 +23,7 @@ config ARM
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_LZO
 	select HAVE_KERNEL_LZMA
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 	select HAVE_REGS_AND_STACK_ACCESS_API
diff --git a/arch/arm/include/asm/perf_event.h b/arch/arm/include/asm/perf_event.h
index b5799a3b7117..c4aa4e8c6af9 100644
--- a/arch/arm/include/asm/perf_event.h
+++ b/arch/arm/include/asm/perf_event.h
@@ -12,18 +12,6 @@
 #ifndef __ARM_PERF_EVENT_H__
 #define __ARM_PERF_EVENT_H__
 
-/*
- * NOP: on *most* (read: all supported) ARM platforms, the performance
- * counter interrupts are regular interrupts and not an NMI. This
- * means that when we receive the interrupt we can call
- * perf_event_do_pending() that handles all of the work with
- * interrupts disabled.
- */
-static inline void
-set_perf_event_pending(void)
-{
-}
-
 /* ARM performance counters start from 1 (in the cp15 accesses) so use the
  * same indexes here for consistency. */
 #define PERF_EVENT_INDEX_OFFSET 1
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 6cc6521881aa..49643b1467e6 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -1092,7 +1092,7 @@ armv6pmu_handle_irq(int irq_num,
 	 * platforms that can have the PMU interrupts raised as an NMI, this
 	 * will not work.
 	 */
-	perf_event_do_pending();
+	irq_work_run();
 
 	return IRQ_HANDLED;
 }
@@ -2068,7 +2068,7 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev)
 	 * platforms that can have the PMU interrupts raised as an NMI, this
 	 * will not work.
 	 */
-	perf_event_do_pending();
+	irq_work_run();
 
 	return IRQ_HANDLED;
 }
@@ -2436,7 +2436,7 @@ xscale1pmu_handle_irq(int irq_num, void *dev)
 			armpmu->disable(hwc, idx);
 	}
 
-	perf_event_do_pending();
+	irq_work_run();
 
 	/*
 	 * Re-enable the PMU.
@@ -2763,7 +2763,7 @@ xscale2pmu_handle_irq(int irq_num, void *dev)
 			armpmu->disable(hwc, idx);
 	}
 
-	perf_event_do_pending();
+	irq_work_run();
 
 	/*
 	 * Re-enable the PMU.
diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index 16399bd24993..0f2417df6323 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -7,6 +7,7 @@ config FRV
 	default y
 	select HAVE_IDE
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 
 config ZONE_DMA
diff --git a/arch/frv/lib/Makefile b/arch/frv/lib/Makefile
index f4709756d0d9..4ff2fb1e6b16 100644
--- a/arch/frv/lib/Makefile
+++ b/arch/frv/lib/Makefile
@@ -5,4 +5,4 @@
 lib-y := \
 	__ashldi3.o __lshrdi3.o __muldi3.o __ashrdi3.o __negdi2.o __ucmpdi2.o \
 	checksum.o memcpy.o memset.o atomic-ops.o atomic64-ops.o \
-	outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o perf_event.o
+	outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o
diff --git a/arch/frv/lib/perf_event.c b/arch/frv/lib/perf_event.c
deleted file mode 100644
index 9ac5acfd2e91..000000000000
--- a/arch/frv/lib/perf_event.c
+++ /dev/null
@@ -1,19 +0,0 @@
-/* Performance event handling
- *
- * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public Licence
- * as published by the Free Software Foundation; either version
- * 2 of the Licence, or (at your option) any later version.
- */
-
-#include <linux/perf_event.h>
-
-/*
- * mark the performance event as pending
- */
-void set_perf_event_pending(void)
-{
-}
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 907417d187e1..79a04a9394d5 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -16,6 +16,7 @@ config PARISC
 	select RTC_DRV_GENERIC
 	select INIT_ALL_POSSIBLE
 	select BUG
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select GENERIC_ATOMIC64 if !64BIT
 	help
diff --git a/arch/parisc/include/asm/perf_event.h b/arch/parisc/include/asm/perf_event.h
index cc146427d8f9..1e0fd8ba6c03 100644
--- a/arch/parisc/include/asm/perf_event.h
+++ b/arch/parisc/include/asm/perf_event.h
@@ -1,7 +1,6 @@
 #ifndef __ASM_PARISC_PERF_EVENT_H
 #define __ASM_PARISC_PERF_EVENT_H
 
-/* parisc only supports software events through this interface. */
-static inline void set_perf_event_pending(void) { }
+/* Empty, just to avoid compiling error */
 
 #endif /* __ASM_PARISC_PERF_EVENT_H */
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 631e5a0fb6ab..4b1e521d966f 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -138,6 +138,7 @@ config PPC
 	select HAVE_OPROFILE
 	select HAVE_SYSCALL_WRAPPERS if PPC64
 	select GENERIC_ATOMIC64 if PPC32
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_HW_BREAKPOINT if PERF_EVENTS && PPC_BOOK3S_64
diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h
index 1ff6662f7faf..9b287fdd8ea3 100644
--- a/arch/powerpc/include/asm/paca.h
+++ b/arch/powerpc/include/asm/paca.h
@@ -129,7 +129,7 @@ struct paca_struct {
 	u8 soft_enabled;		/* irq soft-enable flag */
 	u8 hard_enabled;		/* set if irqs are enabled in MSR */
 	u8 io_sync;			/* writel() needs spin_unlock sync */
-	u8 perf_event_pending;		/* PM interrupt while soft-disabled */
+	u8 irq_work_pending;		/* IRQ_WORK interrupt while soft-disable */
 
 	/* Stuff for accurate time accounting */
 	u64 user_time;			/* accumulated usermode TB ticks */
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 8533b3b83f5d..54888eb10c3b 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -53,7 +53,7 @@
 #include <linux/posix-timers.h>
 #include <linux/irq.h>
 #include <linux/delay.h>
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 #include <asm/trace.h>
 
 #include <asm/io.h>
@@ -493,60 +493,60 @@ void __init iSeries_time_init_early(void)
 }
 #endif /* CONFIG_PPC_ISERIES */
 
-#ifdef CONFIG_PERF_EVENTS
+#ifdef CONFIG_IRQ_WORK
 
 /*
  * 64-bit uses a byte in the PACA, 32-bit uses a per-cpu variable...
  */
 #ifdef CONFIG_PPC64
-static inline unsigned long test_perf_event_pending(void)
+static inline unsigned long test_irq_work_pending(void)
 {
 	unsigned long x;
 
 	asm volatile("lbz %0,%1(13)"
 		: "=r" (x)
-		: "i" (offsetof(struct paca_struct, perf_event_pending)));
+		: "i" (offsetof(struct paca_struct, irq_work_pending)));
 	return x;
 }
 
-static inline void set_perf_event_pending_flag(void)
+static inline void set_irq_work_pending_flag(void)
 {
 	asm volatile("stb %0,%1(13)" : :
 		"r" (1),
-		"i" (offsetof(struct paca_struct, perf_event_pending)));
+		"i" (offsetof(struct paca_struct, irq_work_pending)));
 }
 
-static inline void clear_perf_event_pending(void)
+static inline void clear_irq_work_pending(void)
 {
 	asm volatile("stb %0,%1(13)" : :
 		"r" (0),
-		"i" (offsetof(struct paca_struct, perf_event_pending)));
+		"i" (offsetof(struct paca_struct, irq_work_pending)));
 }
 
 #else /* 32-bit */
 
-DEFINE_PER_CPU(u8, perf_event_pending);
+DEFINE_PER_CPU(u8, irq_work_pending);
 
-#define set_perf_event_pending_flag()	__get_cpu_var(perf_event_pending) = 1
-#define test_perf_event_pending()	__get_cpu_var(perf_event_pending)
-#define clear_perf_event_pending()	__get_cpu_var(perf_event_pending) = 0
+#define set_irq_work_pending_flag()	__get_cpu_var(irq_work_pending) = 1
+#define test_irq_work_pending()		__get_cpu_var(irq_work_pending)
+#define clear_irq_work_pending()	__get_cpu_var(irq_work_pending) = 0
 
 #endif /* 32 vs 64 bit */
 
-void set_perf_event_pending(void)
+void set_irq_work_pending(void)
 {
 	preempt_disable();
-	set_perf_event_pending_flag();
+	set_irq_work_pending_flag();
 	set_dec(1);
 	preempt_enable();
 }
 
-#else  /* CONFIG_PERF_EVENTS */
+#else  /* CONFIG_IRQ_WORK */
 
-#define test_perf_event_pending()	0
-#define clear_perf_event_pending()
+#define test_irq_work_pending()	0
+#define clear_irq_work_pending()
 
-#endif /* CONFIG_PERF_EVENTS */
+#endif /* CONFIG_IRQ_WORK */
 
 /*
  * For iSeries shared processors, we have to let the hypervisor
@@ -587,9 +587,9 @@ void timer_interrupt(struct pt_regs * regs)
 
 	calculate_steal_time();
 
-	if (test_perf_event_pending()) {
-		clear_perf_event_pending();
-		perf_event_do_pending();
+	if (test_irq_work_pending()) {
+		clear_irq_work_pending();
+		irq_work_run();
 	}
 
 #ifdef CONFIG_PPC_ISERIES
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index f0777a47e3a5..958f0dadeadf 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -95,6 +95,7 @@ config S390
 	select HAVE_KVM if 64BIT
 	select HAVE_ARCH_TRACEHOOK
 	select INIT_ALL_POSSIBLE
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_BZIP2
diff --git a/arch/s390/include/asm/perf_event.h b/arch/s390/include/asm/perf_event.h
index 3840cbe77637..a75f168d2718 100644
--- a/arch/s390/include/asm/perf_event.h
+++ b/arch/s390/include/asm/perf_event.h
@@ -4,7 +4,6 @@
  * Copyright 2009 Martin Schwidefsky, IBM Corporation.
  */
 
-static inline void set_perf_event_pending(void) {}
-static inline void clear_perf_event_pending(void) {}
+/* Empty, just to avoid compiling error */
 
 #define PERF_EVENT_INDEX_OFFSET 0
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 35b6c3f85173..35b6879628a0 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -16,6 +16,7 @@ config SUPERH
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_DMA_API_DEBUG
 	select HAVE_DMA_ATTRS
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 	select HAVE_KERNEL_GZIP
diff --git a/arch/sh/include/asm/perf_event.h b/arch/sh/include/asm/perf_event.h
index 3d0c9f36d150..14308bed7ea5 100644
--- a/arch/sh/include/asm/perf_event.h
+++ b/arch/sh/include/asm/perf_event.h
@@ -26,11 +26,4 @@ extern int register_sh_pmu(struct sh_pmu *);
 extern int reserve_pmc_hardware(void);
 extern void release_pmc_hardware(void);
 
-static inline void set_perf_event_pending(void)
-{
-	/* Nothing to see here, move along. */
-}
-
-#define PERF_EVENT_INDEX_OFFSET	0
-
 #endif /* __ASM_SH_PERF_EVENT_H */
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 9212cd42a832..3e9d31401fb2 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -26,6 +26,7 @@ config SPARC
 	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select RTC_CLASS
 	select RTC_DRV_M48T59
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 	select HAVE_DMA_ATTRS
@@ -54,6 +55,7 @@ config SPARC64
 	select RTC_DRV_BQ4802
 	select RTC_DRV_SUN4V
 	select RTC_DRV_STARFIRE
+	select HAVE_IRQ_WORK
 	select HAVE_PERF_EVENTS
 	select PERF_USE_VMALLOC
 
diff --git a/arch/sparc/include/asm/perf_event.h b/arch/sparc/include/asm/perf_event.h
index 727af70646cb..6e8bfa1786da 100644
--- a/arch/sparc/include/asm/perf_event.h
+++ b/arch/sparc/include/asm/perf_event.h
@@ -1,10 +1,6 @@
 #ifndef __ASM_SPARC_PERF_EVENT_H
 #define __ASM_SPARC_PERF_EVENT_H
 
-extern void set_perf_event_pending(void);
-
-#define	PERF_EVENT_INDEX_OFFSET	0
-
 #ifdef CONFIG_PERF_EVENTS
 #include <asm/ptrace.h>
 
diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c
index c4a6a50b4849..b87873c0e8ea 100644
--- a/arch/sparc/kernel/pcr.c
+++ b/arch/sparc/kernel/pcr.c
@@ -7,7 +7,7 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 #include <linux/ftrace.h>
 
 #include <asm/pil.h>
@@ -43,14 +43,14 @@ void __irq_entry deferred_pcr_work_irq(int irq, struct pt_regs *regs)
 
 	old_regs = set_irq_regs(regs);
 	irq_enter();
-#ifdef CONFIG_PERF_EVENTS
-	perf_event_do_pending();
+#ifdef CONFIG_IRQ_WORK
+	irq_work_run();
 #endif
 	irq_exit();
 	set_irq_regs(old_regs);
 }
 
-void set_perf_event_pending(void)
+void arch_irq_work_raise(void)
 {
 	set_softint(1 << PIL_DEFERRED_PCR_WORK);
 }
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 9815221976a7..fd227d6b8d9c 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -25,6 +25,7 @@ config X86
 	select HAVE_IDE
 	select HAVE_OPROFILE
 	select HAVE_PERF_EVENTS if (!M386 && !M486)
+	select HAVE_IRQ_WORK
 	select HAVE_IOREMAP_PROT
 	select HAVE_KPROBES
 	select ARCH_WANT_OPTIONAL_GPIOLIB
diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h
index 8e8ec663a98f..b8e96a18676b 100644
--- a/arch/x86/include/asm/entry_arch.h
+++ b/arch/x86/include/asm/entry_arch.h
@@ -49,8 +49,8 @@ BUILD_INTERRUPT(apic_timer_interrupt,LOCAL_TIMER_VECTOR)
 BUILD_INTERRUPT(error_interrupt,ERROR_APIC_VECTOR)
 BUILD_INTERRUPT(spurious_interrupt,SPURIOUS_APIC_VECTOR)
 
-#ifdef CONFIG_PERF_EVENTS
-BUILD_INTERRUPT(perf_pending_interrupt, LOCAL_PENDING_VECTOR)
+#ifdef CONFIG_IRQ_WORK
+BUILD_INTERRUPT(irq_work_interrupt, IRQ_WORK_VECTOR)
 #endif
 
 #ifdef CONFIG_X86_THERMAL_VECTOR
diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h
index aeab29aee617..55e4de613f0e 100644
--- a/arch/x86/include/asm/hardirq.h
+++ b/arch/x86/include/asm/hardirq.h
@@ -14,7 +14,7 @@ typedef struct {
 #endif
 	unsigned int x86_platform_ipis;	/* arch dependent */
 	unsigned int apic_perf_irqs;
-	unsigned int apic_pending_irqs;
+	unsigned int apic_irq_work_irqs;
 #ifdef CONFIG_SMP
 	unsigned int irq_resched_count;
 	unsigned int irq_call_count;
diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h
index 46c0fe05f230..3a54a1ca1a02 100644
--- a/arch/x86/include/asm/hw_irq.h
+++ b/arch/x86/include/asm/hw_irq.h
@@ -29,7 +29,7 @@
 extern void apic_timer_interrupt(void);
 extern void x86_platform_ipi(void);
 extern void error_interrupt(void);
-extern void perf_pending_interrupt(void);
+extern void irq_work_interrupt(void);
 
 extern void spurious_interrupt(void);
 extern void thermal_interrupt(void);
diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h
index e2ca30092557..6af0894dafb4 100644
--- a/arch/x86/include/asm/irq_vectors.h
+++ b/arch/x86/include/asm/irq_vectors.h
@@ -114,9 +114,9 @@
 #define X86_PLATFORM_IPI_VECTOR		0xed
 
 /*
- * Performance monitoring pending work vector:
+ * IRQ work vector:
  */
-#define LOCAL_PENDING_VECTOR		0xec
+#define IRQ_WORK_VECTOR			0xec
 
 #define UV_BAU_MESSAGE			0xea
 
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 9d3f485e5dd0..7490bf8d1459 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -35,6 +35,7 @@ obj-y			:= process_$(BITS).o signal.o entry_$(BITS).o
 obj-y			+= traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o
 obj-y			+= time.o ioport.o ldt.o dumpstack.o
 obj-y			+= setup.o x86_init.o i8259.o irqinit.o jump_label.o
+obj-$(CONFIG_IRQ_WORK)  += irq_work.o
 obj-$(CONFIG_X86_VISWS)	+= visws_quirks.o
 obj-$(CONFIG_X86_32)	+= probe_roms_32.o
 obj-$(CONFIG_X86_32)	+= sys_i386_32.o i386_ksyms_32.o
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index e2513f26ba8b..fe73c1844a9a 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1196,25 +1196,6 @@ static int x86_pmu_handle_irq(struct pt_regs *regs)
 	return handled;
 }
 
-void smp_perf_pending_interrupt(struct pt_regs *regs)
-{
-	irq_enter();
-	ack_APIC_irq();
-	inc_irq_stat(apic_pending_irqs);
-	perf_event_do_pending();
-	irq_exit();
-}
-
-void set_perf_event_pending(void)
-{
-#ifdef CONFIG_X86_LOCAL_APIC
-	if (!x86_pmu.apic || !x86_pmu_initialized())
-		return;
-
-	apic->send_IPI_self(LOCAL_PENDING_VECTOR);
-#endif
-}
-
 void perf_events_lapic_init(void)
 {
 	if (!x86_pmu.apic || !x86_pmu_initialized())
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 17be5ec7cbba..c375c79065f8 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -1023,9 +1023,9 @@ apicinterrupt ERROR_APIC_VECTOR \
 apicinterrupt SPURIOUS_APIC_VECTOR \
 	spurious_interrupt smp_spurious_interrupt
 
-#ifdef CONFIG_PERF_EVENTS
-apicinterrupt LOCAL_PENDING_VECTOR \
-	perf_pending_interrupt smp_perf_pending_interrupt
+#ifdef CONFIG_IRQ_WORK
+apicinterrupt IRQ_WORK_VECTOR \
+	irq_work_interrupt smp_irq_work_interrupt
 #endif
 
 /*
diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index 91fd0c70a18a..44edb03fc9ec 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -67,10 +67,10 @@ static int show_other_interrupts(struct seq_file *p, int prec)
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->apic_perf_irqs);
 	seq_printf(p, "  Performance monitoring interrupts\n");
-	seq_printf(p, "%*s: ", prec, "PND");
+	seq_printf(p, "%*s: ", prec, "IWI");
 	for_each_online_cpu(j)
-		seq_printf(p, "%10u ", irq_stats(j)->apic_pending_irqs);
-	seq_printf(p, "  Performance pending work\n");
+		seq_printf(p, "%10u ", irq_stats(j)->apic_irq_work_irqs);
+	seq_printf(p, "  IRQ work interrupts\n");
 #endif
 	if (x86_platform_ipi_callback) {
 		seq_printf(p, "%*s: ", prec, "PLT");
@@ -185,7 +185,7 @@ u64 arch_irq_stat_cpu(unsigned int cpu)
 	sum += irq_stats(cpu)->apic_timer_irqs;
 	sum += irq_stats(cpu)->irq_spurious_count;
 	sum += irq_stats(cpu)->apic_perf_irqs;
-	sum += irq_stats(cpu)->apic_pending_irqs;
+	sum += irq_stats(cpu)->apic_irq_work_irqs;
 #endif
 	if (x86_platform_ipi_callback)
 		sum += irq_stats(cpu)->x86_platform_ipis;
diff --git a/arch/x86/kernel/irq_work.c b/arch/x86/kernel/irq_work.c
new file mode 100644
index 000000000000..ca8f703a1e70
--- /dev/null
+++ b/arch/x86/kernel/irq_work.c
@@ -0,0 +1,30 @@
+/*
+ * x86 specific code for irq_work
+ *
+ * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/irq_work.h>
+#include <linux/hardirq.h>
+#include <asm/apic.h>
+
+void smp_irq_work_interrupt(struct pt_regs *regs)
+{
+	irq_enter();
+	ack_APIC_irq();
+	inc_irq_stat(apic_irq_work_irqs);
+	irq_work_run();
+	irq_exit();
+}
+
+void arch_irq_work_raise(void)
+{
+#ifdef CONFIG_X86_LOCAL_APIC
+	if (!cpu_has_apic)
+		return;
+
+	apic->send_IPI_self(IRQ_WORK_VECTOR);
+	apic_wait_icr_idle();
+#endif
+}
diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c
index 990ae7cfc578..713969b9266b 100644
--- a/arch/x86/kernel/irqinit.c
+++ b/arch/x86/kernel/irqinit.c
@@ -224,9 +224,9 @@ static void __init apic_intr_init(void)
 	alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
 	alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt);
 
-	/* Performance monitoring interrupts: */
-# ifdef CONFIG_PERF_EVENTS
-	alloc_intr_gate(LOCAL_PENDING_VECTOR, perf_pending_interrupt);
+	/* IRQ work interrupts: */
+# ifdef CONFIG_IRQ_WORK
+	alloc_intr_gate(IRQ_WORK_VECTOR, irq_work_interrupt);
 # endif
 
 #endif
diff --git a/include/linux/irq_work.h b/include/linux/irq_work.h
new file mode 100644
index 000000000000..4fa09d4d0b71
--- /dev/null
+++ b/include/linux/irq_work.h
@@ -0,0 +1,20 @@
+#ifndef _LINUX_IRQ_WORK_H
+#define _LINUX_IRQ_WORK_H
+
+struct irq_work {
+	struct irq_work *next;
+	void (*func)(struct irq_work *);
+};
+
+static inline
+void init_irq_work(struct irq_work *entry, void (*func)(struct irq_work *))
+{
+	entry->next = NULL;
+	entry->func = func;
+}
+
+bool irq_work_queue(struct irq_work *entry);
+void irq_work_run(void);
+void irq_work_sync(struct irq_work *entry);
+
+#endif /* _LINUX_IRQ_WORK_H */
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index a9227e985207..2ebfc9ae4755 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -486,6 +486,7 @@ struct perf_guest_info_callbacks {
 #include <linux/workqueue.h>
 #include <linux/ftrace.h>
 #include <linux/cpu.h>
+#include <linux/irq_work.h>
 #include <asm/atomic.h>
 #include <asm/local.h>
 
@@ -672,11 +673,6 @@ struct perf_buffer {
 	void				*data_pages[0];
 };
 
-struct perf_pending_entry {
-	struct perf_pending_entry *next;
-	void (*func)(struct perf_pending_entry *);
-};
-
 struct perf_sample_data;
 
 typedef void (*perf_overflow_handler_t)(struct perf_event *, int,
@@ -784,7 +780,7 @@ struct perf_event {
 	int				pending_wakeup;
 	int				pending_kill;
 	int				pending_disable;
-	struct perf_pending_entry	pending;
+	struct irq_work			pending;
 
 	atomic_t			event_limit;
 
@@ -898,8 +894,6 @@ extern int perf_event_init_task(struct task_struct *child);
 extern void perf_event_exit_task(struct task_struct *child);
 extern void perf_event_free_task(struct task_struct *task);
 extern void perf_event_delayed_put(struct task_struct *task);
-extern void set_perf_event_pending(void);
-extern void perf_event_do_pending(void);
 extern void perf_event_print_debug(void);
 extern void perf_pmu_disable(struct pmu *pmu);
 extern void perf_pmu_enable(struct pmu *pmu);
@@ -1078,7 +1072,6 @@ static inline int perf_event_init_task(struct task_struct *child)	{ return 0; }
 static inline void perf_event_exit_task(struct task_struct *child)	{ }
 static inline void perf_event_free_task(struct task_struct *task)	{ }
 static inline void perf_event_delayed_put(struct task_struct *task)	{ }
-static inline void perf_event_do_pending(void)				{ }
 static inline void perf_event_print_debug(void)				{ }
 static inline int perf_event_task_disable(void)				{ return -EINVAL; }
 static inline int perf_event_task_enable(void)				{ return -EINVAL; }
diff --git a/init/Kconfig b/init/Kconfig
index 2de5b1cbadd9..1ef0b439908e 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -21,6 +21,13 @@ config CONSTRUCTORS
 	depends on !UML
 	default y
 
+config HAVE_IRQ_WORK
+	bool
+
+config IRQ_WORK
+	bool
+	depends on HAVE_IRQ_WORK
+
 menu "General setup"
 
 config EXPERIMENTAL
@@ -987,6 +994,7 @@ config PERF_EVENTS
 	default y if (PROFILING || PERF_COUNTERS)
 	depends on HAVE_PERF_EVENTS
 	select ANON_INODES
+	select IRQ_WORK
 	help
 	  Enable kernel support for various performance events provided
 	  by software and hardware.
diff --git a/kernel/Makefile b/kernel/Makefile
index d52b473c99a1..4d9bf5f8531f 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -23,6 +23,7 @@ CFLAGS_REMOVE_rtmutex-debug.o = -pg
 CFLAGS_REMOVE_cgroup-debug.o = -pg
 CFLAGS_REMOVE_sched_clock.o = -pg
 CFLAGS_REMOVE_perf_event.o = -pg
+CFLAGS_REMOVE_irq_work.o = -pg
 endif
 
 obj-$(CONFIG_FREEZER) += freezer.o
@@ -100,6 +101,7 @@ obj-$(CONFIG_TRACING) += trace/
 obj-$(CONFIG_X86_DS) += trace/
 obj-$(CONFIG_RING_BUFFER) += trace/
 obj-$(CONFIG_SMP) += sched_cpupri.o
+obj-$(CONFIG_IRQ_WORK) += irq_work.o
 obj-$(CONFIG_PERF_EVENTS) += perf_event.o
 obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o
 obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o
diff --git a/kernel/irq_work.c b/kernel/irq_work.c
new file mode 100644
index 000000000000..f16763ff8481
--- /dev/null
+++ b/kernel/irq_work.c
@@ -0,0 +1,164 @@
+/*
+ * Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ *
+ * Provides a framework for enqueueing and running callbacks from hardirq
+ * context. The enqueueing is NMI-safe.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/irq_work.h>
+#include <linux/hardirq.h>
+
+/*
+ * An entry can be in one of four states:
+ *
+ * free	     NULL, 0 -> {claimed}       : free to be used
+ * claimed   NULL, 3 -> {pending}       : claimed to be enqueued
+ * pending   next, 3 -> {busy}          : queued, pending callback
+ * busy      NULL, 2 -> {free, claimed} : callback in progress, can be claimed
+ *
+ * We use the lower two bits of the next pointer to keep PENDING and BUSY
+ * flags.
+ */
+
+#define IRQ_WORK_PENDING	1UL
+#define IRQ_WORK_BUSY		2UL
+#define IRQ_WORK_FLAGS		3UL
+
+static inline bool irq_work_is_set(struct irq_work *entry, int flags)
+{
+	return (unsigned long)entry->next & flags;
+}
+
+static inline struct irq_work *irq_work_next(struct irq_work *entry)
+{
+	unsigned long next = (unsigned long)entry->next;
+	next &= ~IRQ_WORK_FLAGS;
+	return (struct irq_work *)next;
+}
+
+static inline struct irq_work *next_flags(struct irq_work *entry, int flags)
+{
+	unsigned long next = (unsigned long)entry;
+	next |= flags;
+	return (struct irq_work *)next;
+}
+
+static DEFINE_PER_CPU(struct irq_work *, irq_work_list);
+
+/*
+ * Claim the entry so that no one else will poke at it.
+ */
+static bool irq_work_claim(struct irq_work *entry)
+{
+	struct irq_work *next, *nflags;
+
+	do {
+		next = entry->next;
+		if ((unsigned long)next & IRQ_WORK_PENDING)
+			return false;
+		nflags = next_flags(next, IRQ_WORK_FLAGS);
+	} while (cmpxchg(&entry->next, next, nflags) != next);
+
+	return true;
+}
+
+
+void __weak arch_irq_work_raise(void)
+{
+	/*
+	 * Lame architectures will get the timer tick callback
+	 */
+}
+
+/*
+ * Queue the entry and raise the IPI if needed.
+ */
+static void __irq_work_queue(struct irq_work *entry)
+{
+	struct irq_work **head, *next;
+
+	head = &get_cpu_var(irq_work_list);
+
+	do {
+		next = *head;
+		/* Can assign non-atomic because we keep the flags set. */
+		entry->next = next_flags(next, IRQ_WORK_FLAGS);
+	} while (cmpxchg(head, next, entry) != next);
+
+	/* The list was empty, raise self-interrupt to start processing. */
+	if (!irq_work_next(entry))
+		arch_irq_work_raise();
+
+	put_cpu_var(irq_work_list);
+}
+
+/*
+ * Enqueue the irq_work @entry, returns true on success, failure when the
+ * @entry was already enqueued by someone else.
+ *
+ * Can be re-enqueued while the callback is still in progress.
+ */
+bool irq_work_queue(struct irq_work *entry)
+{
+	if (!irq_work_claim(entry)) {
+		/*
+		 * Already enqueued, can't do!
+		 */
+		return false;
+	}
+
+	__irq_work_queue(entry);
+	return true;
+}
+EXPORT_SYMBOL_GPL(irq_work_queue);
+
+/*
+ * Run the irq_work entries on this cpu. Requires to be ran from hardirq
+ * context with local IRQs disabled.
+ */
+void irq_work_run(void)
+{
+	struct irq_work *list, **head;
+
+	head = &__get_cpu_var(irq_work_list);
+	if (*head == NULL)
+		return;
+
+	BUG_ON(!in_irq());
+	BUG_ON(!irqs_disabled());
+
+	list = xchg(head, NULL);
+	while (list != NULL) {
+		struct irq_work *entry = list;
+
+		list = irq_work_next(list);
+
+		/*
+		 * Clear the PENDING bit, after this point the @entry
+		 * can be re-used.
+		 */
+		entry->next = next_flags(NULL, IRQ_WORK_BUSY);
+		entry->func(entry);
+		/*
+		 * Clear the BUSY bit and return to the free state if
+		 * no-one else claimed it meanwhile.
+		 */
+		cmpxchg(&entry->next, next_flags(NULL, IRQ_WORK_BUSY), NULL);
+	}
+}
+EXPORT_SYMBOL_GPL(irq_work_run);
+
+/*
+ * Synchronize against the irq_work @entry, ensures the entry is not
+ * currently in use.
+ */
+void irq_work_sync(struct irq_work *entry)
+{
+	WARN_ON_ONCE(irqs_disabled());
+
+	while (irq_work_is_set(entry, IRQ_WORK_BUSY))
+		cpu_relax();
+}
+EXPORT_SYMBOL_GPL(irq_work_sync);
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 634f86a4b2f9..99b9700e74d0 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -2206,12 +2206,11 @@ static void free_event_rcu(struct rcu_head *head)
 	kfree(event);
 }
 
-static void perf_pending_sync(struct perf_event *event);
 static void perf_buffer_put(struct perf_buffer *buffer);
 
 static void free_event(struct perf_event *event)
 {
-	perf_pending_sync(event);
+	irq_work_sync(&event->pending);
 
 	if (!event->parent) {
 		atomic_dec(&nr_events);
@@ -3162,16 +3161,7 @@ void perf_event_wakeup(struct perf_event *event)
 	}
 }
 
-/*
- * Pending wakeups
- *
- * Handle the case where we need to wakeup up from NMI (or rq->lock) context.
- *
- * The NMI bit means we cannot possibly take locks. Therefore, maintain a
- * single linked list and use cmpxchg() to add entries lockless.
- */
-
-static void perf_pending_event(struct perf_pending_entry *entry)
+static void perf_pending_event(struct irq_work *entry)
 {
 	struct perf_event *event = container_of(entry,
 			struct perf_event, pending);
@@ -3187,89 +3177,6 @@ static void perf_pending_event(struct perf_pending_entry *entry)
 	}
 }
 
-#define PENDING_TAIL ((struct perf_pending_entry *)-1UL)
-
-static DEFINE_PER_CPU(struct perf_pending_entry *, perf_pending_head) = {
-	PENDING_TAIL,
-};
-
-static void perf_pending_queue(struct perf_pending_entry *entry,
-			       void (*func)(struct perf_pending_entry *))
-{
-	struct perf_pending_entry **head;
-
-	if (cmpxchg(&entry->next, NULL, PENDING_TAIL) != NULL)
-		return;
-
-	entry->func = func;
-
-	head = &get_cpu_var(perf_pending_head);
-
-	do {
-		entry->next = *head;
-	} while (cmpxchg(head, entry->next, entry) != entry->next);
-
-	set_perf_event_pending();
-
-	put_cpu_var(perf_pending_head);
-}
-
-static int __perf_pending_run(void)
-{
-	struct perf_pending_entry *list;
-	int nr = 0;
-
-	list = xchg(&__get_cpu_var(perf_pending_head), PENDING_TAIL);
-	while (list != PENDING_TAIL) {
-		void (*func)(struct perf_pending_entry *);
-		struct perf_pending_entry *entry = list;
-
-		list = list->next;
-
-		func = entry->func;
-		entry->next = NULL;
-		/*
-		 * Ensure we observe the unqueue before we issue the wakeup,
-		 * so that we won't be waiting forever.
-		 * -- see perf_not_pending().
-		 */
-		smp_wmb();
-
-		func(entry);
-		nr++;
-	}
-
-	return nr;
-}
-
-static inline int perf_not_pending(struct perf_event *event)
-{
-	/*
-	 * If we flush on whatever cpu we run, there is a chance we don't
-	 * need to wait.
-	 */
-	get_cpu();
-	__perf_pending_run();
-	put_cpu();
-
-	/*
-	 * Ensure we see the proper queue state before going to sleep
-	 * so that we do not miss the wakeup. -- see perf_pending_handle()
-	 */
-	smp_rmb();
-	return event->pending.next == NULL;
-}
-
-static void perf_pending_sync(struct perf_event *event)
-{
-	wait_event(event->waitq, perf_not_pending(event));
-}
-
-void perf_event_do_pending(void)
-{
-	__perf_pending_run();
-}
-
 /*
  * We assume there is only KVM supporting the callbacks.
  * Later on, we might change it to a list if there is
@@ -3319,8 +3226,7 @@ static void perf_output_wakeup(struct perf_output_handle *handle)
 
 	if (handle->nmi) {
 		handle->event->pending_wakeup = 1;
-		perf_pending_queue(&handle->event->pending,
-				   perf_pending_event);
+		irq_work_queue(&handle->event->pending);
 	} else
 		perf_event_wakeup(handle->event);
 }
@@ -4356,8 +4262,7 @@ static int __perf_event_overflow(struct perf_event *event, int nmi,
 		event->pending_kill = POLL_HUP;
 		if (nmi) {
 			event->pending_disable = 1;
-			perf_pending_queue(&event->pending,
-					   perf_pending_event);
+			irq_work_queue(&event->pending);
 		} else
 			perf_event_disable(event);
 	}
@@ -5374,6 +5279,7 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 	INIT_LIST_HEAD(&event->event_entry);
 	INIT_LIST_HEAD(&event->sibling_list);
 	init_waitqueue_head(&event->waitq);
+	init_irq_work(&event->pending, perf_pending_event);
 
 	mutex_init(&event->mmap_mutex);
 
diff --git a/kernel/timer.c b/kernel/timer.c
index 97bf05baade7..68a9ae7679b7 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -37,7 +37,7 @@
 #include <linux/delay.h>
 #include <linux/tick.h>
 #include <linux/kallsyms.h>
-#include <linux/perf_event.h>
+#include <linux/irq_work.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
 
@@ -1279,7 +1279,10 @@ void update_process_times(int user_tick)
 	run_local_timers();
 	rcu_check_callbacks(cpu, user_tick);
 	printk_tick();
-	perf_event_do_pending();
+#ifdef CONFIG_IRQ_WORK
+	if (in_irq())
+		irq_work_run();
+#endif
 	scheduler_tick();
 	run_posix_cpu_timers(p);
 }
-- 
cgit v1.2.3


From d580ff8699e8811a9af37e9de4dea375401bdeec Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 14 Oct 2010 17:43:23 +0200
Subject: perf, hw_breakpoint: Fix crash in hw_breakpoint creation

hw_breakpoint creation needs to account stuff per-task to ensure there
is always sufficient hardware resources to back these things due to
ptrace.

With the perf per pmu context changes the event initialization no
longer has access to the event context, for the simple reason that we
need to first find the pmu (result of initialization) before we can
find the context.

This makes hw_breakpoints unhappy, because it can no longer do per
task accounting, cure this by frobbing a task pointer in the event::hw
bits for now...

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101014203625.391543667@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h |  7 +++++++
 kernel/hw_breakpoint.c     |  8 ++++----
 kernel/perf_event.c        | 23 ++++++++++++++++++-----
 3 files changed, 29 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 2ebfc9ae4755..97965fac55fe 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -536,6 +536,12 @@ struct hw_perf_event {
 		struct { /* breakpoint */
 			struct arch_hw_breakpoint	info;
 			struct list_head		bp_list;
+			/*
+			 * Crufty hack to avoid the chicken and egg
+			 * problem hw_breakpoint has with context
+			 * creation and event initalization.
+			 */
+			struct task_struct		*bp_target;
 		};
 #endif
 	};
@@ -693,6 +699,7 @@ struct swevent_hlist {
 
 #define PERF_ATTACH_CONTEXT	0x01
 #define PERF_ATTACH_GROUP	0x02
+#define PERF_ATTACH_TASK	0x04
 
 /**
  * struct perf_event - performance event kernel representation:
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index 3b714e839c10..2c9120f0afca 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -113,12 +113,12 @@ static unsigned int max_task_bp_pinned(int cpu, enum bp_type_idx type)
  */
 static int task_bp_pinned(struct perf_event *bp, enum bp_type_idx type)
 {
-	struct perf_event_context *ctx = bp->ctx;
+	struct task_struct *tsk = bp->hw.bp_target;
 	struct perf_event *iter;
 	int count = 0;
 
 	list_for_each_entry(iter, &bp_task_head, hw.bp_list) {
-		if (iter->ctx == ctx && find_slot_idx(iter) == type)
+		if (iter->hw.bp_target == tsk && find_slot_idx(iter) == type)
 			count += hw_breakpoint_weight(iter);
 	}
 
@@ -134,7 +134,7 @@ fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp,
 		    enum bp_type_idx type)
 {
 	int cpu = bp->cpu;
-	struct task_struct *tsk = bp->ctx->task;
+	struct task_struct *tsk = bp->hw.bp_target;
 
 	if (cpu >= 0) {
 		slots->pinned = per_cpu(nr_cpu_bp_pinned[type], cpu);
@@ -213,7 +213,7 @@ toggle_bp_slot(struct perf_event *bp, bool enable, enum bp_type_idx type,
 	       int weight)
 {
 	int cpu = bp->cpu;
-	struct task_struct *tsk = bp->ctx->task;
+	struct task_struct *tsk = bp->hw.bp_target;
 
 	/* Pinned counter cpu profiling */
 	if (!tsk) {
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index b21d06aaef60..856e20baf13f 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -5255,9 +5255,10 @@ unlock:
  */
 static struct perf_event *
 perf_event_alloc(struct perf_event_attr *attr, int cpu,
-		   struct perf_event *group_leader,
-		   struct perf_event *parent_event,
-		   perf_overflow_handler_t overflow_handler)
+		 struct task_struct *task,
+		 struct perf_event *group_leader,
+		 struct perf_event *parent_event,
+		 perf_overflow_handler_t overflow_handler)
 {
 	struct pmu *pmu;
 	struct perf_event *event;
@@ -5299,6 +5300,17 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 
 	event->state		= PERF_EVENT_STATE_INACTIVE;
 
+	if (task) {
+		event->attach_state = PERF_ATTACH_TASK;
+#ifdef CONFIG_HAVE_HW_BREAKPOINT
+		/*
+		 * hw_breakpoint is a bit difficult here..
+		 */
+		if (attr->type == PERF_TYPE_BREAKPOINT)
+			event->hw.bp_target = task;
+#endif
+	}
+
 	if (!overflow_handler && parent_event)
 		overflow_handler = parent_event->overflow_handler;
 	
@@ -5559,7 +5571,7 @@ SYSCALL_DEFINE5(perf_event_open,
 		}
 	}
 
-	event = perf_event_alloc(&attr, cpu, group_leader, NULL, NULL);
+	event = perf_event_alloc(&attr, cpu, task, group_leader, NULL, NULL);
 	if (IS_ERR(event)) {
 		err = PTR_ERR(event);
 		goto err_task;
@@ -5728,7 +5740,7 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
 	 * Get the target context (task or percpu):
 	 */
 
-	event = perf_event_alloc(attr, cpu, NULL, NULL, overflow_handler);
+	event = perf_event_alloc(attr, cpu, task, NULL, NULL, overflow_handler);
 	if (IS_ERR(event)) {
 		err = PTR_ERR(event);
 		goto err;
@@ -5996,6 +6008,7 @@ inherit_event(struct perf_event *parent_event,
 
 	child_event = perf_event_alloc(&parent_event->attr,
 					   parent_event->cpu,
+					   child,
 					   group_leader, parent_event,
 					   NULL);
 	if (IS_ERR(child_event))
-- 
cgit v1.2.3


From 3b6e901f839f42afb40f614418df82c08b01320a Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 14 Oct 2010 21:10:38 +0200
Subject: jump_label: Use more consistent naming

Now that there's still only a few users around, rename things to make
them more consistent.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101014203625.448565169@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/jump_label.h | 8 ++++----
 kernel/tracepoint.c        | 6 +++---
 lib/dynamic_debug.c        | 4 ++--
 3 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h
index b72cd9f92c2e..81be4962b7a1 100644
--- a/include/linux/jump_label.h
+++ b/include/linux/jump_label.h
@@ -25,10 +25,10 @@ extern void jump_label_update(unsigned long key, enum jump_label_type type);
 extern void jump_label_apply_nops(struct module *mod);
 extern int jump_label_text_reserved(void *start, void *end);
 
-#define enable_jump_label(key) \
+#define jump_label_enable(key) \
 	jump_label_update((unsigned long)key, JUMP_LABEL_ENABLE);
 
-#define disable_jump_label(key) \
+#define jump_label_disable(key) \
 	jump_label_update((unsigned long)key, JUMP_LABEL_DISABLE);
 
 #else
@@ -39,12 +39,12 @@ do {						\
 		goto label;			\
 } while (0)
 
-#define enable_jump_label(cond_var)	\
+#define jump_label_enable(cond_var)	\
 do {					\
        *(cond_var) = 1;			\
 } while (0)
 
-#define disable_jump_label(cond_var)	\
+#define jump_label_disable(cond_var)	\
 do {					\
        *(cond_var) = 0;			\
 } while (0)
diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c
index d6073a50a6ca..e95ee7f31d43 100644
--- a/kernel/tracepoint.c
+++ b/kernel/tracepoint.c
@@ -265,10 +265,10 @@ static void set_tracepoint(struct tracepoint_entry **entry,
 	 */
 	rcu_assign_pointer(elem->funcs, (*entry)->funcs);
 	if (!elem->state && active) {
-		enable_jump_label(&elem->state);
+		jump_label_enable(&elem->state);
 		elem->state = active;
 	} else if (elem->state && !active) {
-		disable_jump_label(&elem->state);
+		jump_label_disable(&elem->state);
 		elem->state = active;
 	}
 }
@@ -285,7 +285,7 @@ static void disable_tracepoint(struct tracepoint *elem)
 		elem->unregfunc();
 
 	if (elem->state) {
-		disable_jump_label(&elem->state);
+		jump_label_disable(&elem->state);
 		elem->state = 0;
 	}
 	rcu_assign_pointer(elem->funcs, NULL);
diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c
index e925c7b960f1..7bd6df781ce5 100644
--- a/lib/dynamic_debug.c
+++ b/lib/dynamic_debug.c
@@ -142,9 +142,9 @@ static void ddebug_change(const struct ddebug_query *query,
 				dt->num_enabled++;
 			dp->flags = newflags;
 			if (newflags) {
-				enable_jump_label(&dp->enabled);
+				jump_label_enable(&dp->enabled);
 			} else {
-				disable_jump_label(&dp->enabled);
+				jump_label_disable(&dp->enabled);
 			}
 			if (verbose)
 				printk(KERN_INFO
-- 
cgit v1.2.3


From 8b92538d84e50062560ba33adbaed7887b6e4a42 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 14 Oct 2010 21:39:02 +0200
Subject: jump_label: Add atomic_t interface

Add an interface to allow usage of jump_labels with atomic counters.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101014203625.501657727@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/jump_label_ref.h | 44 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)
 create mode 100644 include/linux/jump_label_ref.h

(limited to 'include/linux')

diff --git a/include/linux/jump_label_ref.h b/include/linux/jump_label_ref.h
new file mode 100644
index 000000000000..e5d012ad92c6
--- /dev/null
+++ b/include/linux/jump_label_ref.h
@@ -0,0 +1,44 @@
+#ifndef _LINUX_JUMP_LABEL_REF_H
+#define _LINUX_JUMP_LABEL_REF_H
+
+#include <linux/jump_label.h>
+#include <asm/atomic.h>
+
+#ifdef HAVE_JUMP_LABEL
+
+static inline void jump_label_inc(atomic_t *key)
+{
+	if (atomic_add_return(1, key) == 1)
+		jump_label_enable(key);
+}
+
+static inline void jump_label_dec(atomic_t *key)
+{
+	if (atomic_dec_and_test(key))
+		jump_label_disable(key);
+}
+
+#else /* !HAVE_JUMP_LABEL */
+
+static inline void jump_label_inc(atomic_t *key)
+{
+	atomic_inc(key);
+}
+
+static inline void jump_label_dec(atomic_t *key)
+{
+	atomic_dec(key);
+}
+
+#undef JUMP_LABEL
+#define JUMP_LABEL(key, label)						\
+do {									\
+	if (unlikely(__builtin_choose_expr(				\
+	      __builtin_types_compatible_p(typeof(key), atomic_t *),	\
+	      atomic_read((atomic_t *)(key)), *(key))))			\
+		goto label;						\
+} while (0)
+
+#endif /* HAVE_JUMP_LABEL */
+
+#endif /* _LINUX_JUMP_LABEL_REF_H */
-- 
cgit v1.2.3


From 82cd6def9806dcb6a325fb6abbc1d61388a15f6a Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 14 Oct 2010 17:57:23 +0200
Subject: perf: Use jump_labels to optimize the scheduler hooks

Trades a call + conditional + ret for an unconditional jmp.

Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20101014203625.501657727@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h | 27 +++++++++++++++++++++++++--
 kernel/perf_event.c        | 24 +++++++++---------------
 2 files changed, 34 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 97965fac55fe..7f0e7f52af8b 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -487,6 +487,7 @@ struct perf_guest_info_callbacks {
 #include <linux/ftrace.h>
 #include <linux/cpu.h>
 #include <linux/irq_work.h>
+#include <linux/jump_label_ref.h>
 #include <asm/atomic.h>
 #include <asm/local.h>
 
@@ -895,8 +896,30 @@ extern void perf_pmu_unregister(struct pmu *pmu);
 
 extern int perf_num_counters(void);
 extern const char *perf_pmu_name(void);
-extern void perf_event_task_sched_in(struct task_struct *task);
-extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
+extern void __perf_event_task_sched_in(struct task_struct *task);
+extern void __perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
+
+extern atomic_t perf_task_events;
+
+static inline void perf_event_task_sched_in(struct task_struct *task)
+{
+	JUMP_LABEL(&perf_task_events, have_events);
+	return;
+
+have_events:
+	__perf_event_task_sched_in(task);
+}
+
+static inline
+void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next)
+{
+	JUMP_LABEL(&perf_task_events, have_events);
+	return;
+
+have_events:
+	__perf_event_task_sched_out(task, next);
+}
+
 extern int perf_event_init_task(struct task_struct *child);
 extern void perf_event_exit_task(struct task_struct *child);
 extern void perf_event_free_task(struct task_struct *task);
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 856e20baf13f..f7febb02ab97 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -34,7 +34,7 @@
 
 #include <asm/irq_regs.h>
 
-static atomic_t nr_events __read_mostly;
+atomic_t perf_task_events __read_mostly;
 static atomic_t nr_mmap_events __read_mostly;
 static atomic_t nr_comm_events __read_mostly;
 static atomic_t nr_task_events __read_mostly;
@@ -1311,8 +1311,8 @@ void perf_event_context_sched_out(struct task_struct *task, int ctxn,
  * accessing the event control register. If a NMI hits, then it will
  * not restart the event.
  */
-void perf_event_task_sched_out(struct task_struct *task,
-			       struct task_struct *next)
+void __perf_event_task_sched_out(struct task_struct *task,
+				 struct task_struct *next)
 {
 	int ctxn;
 
@@ -1337,14 +1337,6 @@ static void task_ctx_sched_out(struct perf_event_context *ctx,
 	cpuctx->task_ctx = NULL;
 }
 
-/*
- * Called with IRQs disabled
- */
-static void __perf_event_task_sched_out(struct perf_event_context *ctx)
-{
-	task_ctx_sched_out(ctx, EVENT_ALL);
-}
-
 /*
  * Called with IRQs disabled
  */
@@ -1494,7 +1486,7 @@ void perf_event_context_sched_in(struct perf_event_context *ctx)
  * accessing the event control register. If a NMI hits, then it will
  * keep the event running.
  */
-void perf_event_task_sched_in(struct task_struct *task)
+void __perf_event_task_sched_in(struct task_struct *task)
 {
 	struct perf_event_context *ctx;
 	int ctxn;
@@ -2216,7 +2208,8 @@ static void free_event(struct perf_event *event)
 	irq_work_sync(&event->pending);
 
 	if (!event->parent) {
-		atomic_dec(&nr_events);
+		if (event->attach_state & PERF_ATTACH_TASK)
+			jump_label_dec(&perf_task_events);
 		if (event->attr.mmap || event->attr.mmap_data)
 			atomic_dec(&nr_mmap_events);
 		if (event->attr.comm)
@@ -5354,7 +5347,8 @@ done:
 	event->pmu = pmu;
 
 	if (!event->parent) {
-		atomic_inc(&nr_events);
+		if (event->attach_state & PERF_ATTACH_TASK)
+			jump_label_inc(&perf_task_events);
 		if (event->attr.mmap || event->attr.mmap_data)
 			atomic_inc(&nr_mmap_events);
 		if (event->attr.comm)
@@ -5849,7 +5843,7 @@ static void perf_event_exit_task_context(struct task_struct *child, int ctxn)
 	 * our context.
 	 */
 	child_ctx = child->perf_event_ctxp[ctxn];
-	__perf_event_task_sched_out(child_ctx);
+	task_ctx_sched_out(child_ctx, EVENT_ALL);
 
 	/*
 	 * Take the context lock here so that if find_get_context is
-- 
cgit v1.2.3


From 7e54a5a0b655734326dc78c2b5efc1eb35497bb6 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 14 Oct 2010 22:32:45 +0200
Subject: perf: Optimize sw events

Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_event.h | 20 +++++++++++---------
 kernel/perf_event.c        |  4 ++--
 2 files changed, 13 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 7f0e7f52af8b..3b80cbf509ef 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -1012,18 +1012,20 @@ static inline void perf_fetch_caller_regs(struct pt_regs *regs)
 	perf_arch_fetch_caller_regs(regs, CALLER_ADDR0);
 }
 
-static inline void
+static __always_inline void
 perf_sw_event(u32 event_id, u64 nr, int nmi, struct pt_regs *regs, u64 addr)
 {
-	if (atomic_read(&perf_swevent_enabled[event_id])) {
-		struct pt_regs hot_regs;
-
-		if (!regs) {
-			perf_fetch_caller_regs(&hot_regs);
-			regs = &hot_regs;
-		}
-		__perf_sw_event(event_id, nr, nmi, regs, addr);
+	struct pt_regs hot_regs;
+
+	JUMP_LABEL(&perf_swevent_enabled[event_id], have_event);
+	return;
+
+have_event:
+	if (!regs) {
+		perf_fetch_caller_regs(&hot_regs);
+		regs = &hot_regs;
 	}
+	__perf_sw_event(event_id, nr, nmi, regs, addr);
 }
 
 extern void perf_event_mmap(struct vm_area_struct *vma);
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index f7febb02ab97..05ecf6f7c672 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -4669,7 +4669,7 @@ static void sw_perf_event_destroy(struct perf_event *event)
 
 	WARN_ON(event->parent);
 
-	atomic_dec(&perf_swevent_enabled[event_id]);
+	jump_label_dec(&perf_swevent_enabled[event_id]);
 	swevent_hlist_put(event);
 }
 
@@ -4699,7 +4699,7 @@ static int perf_swevent_init(struct perf_event *event)
 		if (err)
 			return err;
 
-		atomic_inc(&perf_swevent_enabled[event_id]);
+		jump_label_inc(&perf_swevent_enabled[event_id]);
 		event->destroy = sw_perf_event_destroy;
 	}
 
-- 
cgit v1.2.3


From ebf31f502492527e2b6b5e5cf85a4ebc7fc8a52e Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Sun, 17 Oct 2010 12:15:00 +0200
Subject: jump_label: Add COND_STMT(), reducer wrappery

The use of the JUMP_LABEL() construct ends up creating endless silly
wrappers, create a higher level construct to reduce this clutter.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/jump_label.h | 10 ++++++++++
 include/linux/perf_event.h | 12 ++----------
 2 files changed, 12 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h
index 81be4962b7a1..b67cb180e6e9 100644
--- a/include/linux/jump_label.h
+++ b/include/linux/jump_label.h
@@ -61,4 +61,14 @@ static inline int jump_label_text_reserved(void *start, void *end)
 
 #endif
 
+#define COND_STMT(key, stmt)					\
+do {								\
+	__label__ jl_enabled;					\
+	JUMP_LABEL(key, jl_enabled);				\
+	if (0) {						\
+jl_enabled:							\
+		stmt;						\
+	}							\
+} while (0)
+
 #endif
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 3b80cbf509ef..057bf22a8323 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -903,21 +903,13 @@ extern atomic_t perf_task_events;
 
 static inline void perf_event_task_sched_in(struct task_struct *task)
 {
-	JUMP_LABEL(&perf_task_events, have_events);
-	return;
-
-have_events:
-	__perf_event_task_sched_in(task);
+	COND_STMT(&perf_task_events, __perf_event_task_sched_in(task));
 }
 
 static inline
 void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next)
 {
-	JUMP_LABEL(&perf_task_events, have_events);
-	return;
-
-have_events:
-	__perf_event_task_sched_out(task, next);
+	COND_STMT(&perf_task_events, __perf_event_task_sched_out(task, next));
 }
 
 extern int perf_event_init_task(struct task_struct *child);
-- 
cgit v1.2.3


From 75e1056f5c57050415b64cb761a3acc35d91f013 Mon Sep 17 00:00:00 2001
From: Venkatesh Pallipadi <venki@google.com>
Date: Mon, 4 Oct 2010 17:03:16 -0700
Subject: sched: Fix softirq time accounting

Peter Zijlstra found a bug in the way softirq time is accounted in
VIRT_CPU_ACCOUNTING on this thread:

   http://lkml.indiana.edu/hypermail//linux/kernel/1009.2/01366.html

The problem is, softirq processing uses local_bh_disable internally. There
is no way, later in the flow, to differentiate between whether softirq is
being processed or is it just that bh has been disabled. So, a hardirq when bh
is disabled results in time being wrongly accounted as softirq.

Looking at the code a bit more, the problem exists in !VIRT_CPU_ACCOUNTING
as well. As account_system_time() in normal tick based accouting also uses
softirq_count, which will be set even when not in softirq with bh disabled.

Peter also suggested solution of using 2*SOFTIRQ_OFFSET as irq count
for local_bh_{disable,enable} and using just SOFTIRQ_OFFSET while softirq
processing. The patch below does that and adds API in_serving_softirq() which
returns whether we are currently processing softirq or not.

Also changes one of the usages of softirq_count in net/sched/cls_cgroup.c
to in_serving_softirq.

Looks like many usages of in_softirq really want in_serving_softirq. Those
changes can be made individually on a case by case basis.

Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-2-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/hardirq.h |  5 +++++
 include/linux/sched.h   |  6 +++---
 kernel/sched.c          |  2 +-
 kernel/softirq.c        | 51 ++++++++++++++++++++++++++++++++-----------------
 net/sched/cls_cgroup.c  |  2 +-
 5 files changed, 44 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index d5b387669dab..e37a77cbd588 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -64,6 +64,8 @@
 #define HARDIRQ_OFFSET	(1UL << HARDIRQ_SHIFT)
 #define NMI_OFFSET	(1UL << NMI_SHIFT)
 
+#define SOFTIRQ_DISABLE_OFFSET	(2 * SOFTIRQ_OFFSET)
+
 #ifndef PREEMPT_ACTIVE
 #define PREEMPT_ACTIVE_BITS	1
 #define PREEMPT_ACTIVE_SHIFT	(NMI_SHIFT + NMI_BITS)
@@ -82,10 +84,13 @@
 /*
  * Are we doing bottom half or hardware interrupt processing?
  * Are we in a softirq context? Interrupt context?
+ * in_softirq - Are we currently processing softirq or have bh disabled?
+ * in_serving_softirq - Are we currently processing softirq?
  */
 #define in_irq()		(hardirq_count())
 #define in_softirq()		(softirq_count())
 #define in_interrupt()		(irq_count())
+#define in_serving_softirq()	(softirq_count() & SOFTIRQ_OFFSET)
 
 /*
  * Are we in NMI context?
diff --git a/include/linux/sched.h b/include/linux/sched.h
index cdf56693ecbf..8744e50cb083 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2366,9 +2366,9 @@ extern int __cond_resched_lock(spinlock_t *lock);
 
 extern int __cond_resched_softirq(void);
 
-#define cond_resched_softirq() ({				\
-	__might_sleep(__FILE__, __LINE__, SOFTIRQ_OFFSET);	\
-	__cond_resched_softirq();				\
+#define cond_resched_softirq() ({					\
+	__might_sleep(__FILE__, __LINE__, SOFTIRQ_DISABLE_OFFSET);	\
+	__cond_resched_softirq();					\
 })
 
 /*
diff --git a/kernel/sched.c b/kernel/sched.c
index 771b518e5f1f..089be8adb074 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3422,7 +3422,7 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
 	tmp = cputime_to_cputime64(cputime);
 	if (hardirq_count() - hardirq_offset)
 		cpustat->irq = cputime64_add(cpustat->irq, tmp);
-	else if (softirq_count())
+	else if (in_serving_softirq())
 		cpustat->softirq = cputime64_add(cpustat->softirq, tmp);
 	else
 		cpustat->system = cputime64_add(cpustat->system, tmp);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 07b4f1b1a73a..988dfbe6bbe8 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -76,12 +76,22 @@ void wakeup_softirqd(void)
 		wake_up_process(tsk);
 }
 
+/*
+ * preempt_count and SOFTIRQ_OFFSET usage:
+ * - preempt_count is changed by SOFTIRQ_OFFSET on entering or leaving
+ *   softirq processing.
+ * - preempt_count is changed by SOFTIRQ_DISABLE_OFFSET (= 2 * SOFTIRQ_OFFSET)
+ *   on local_bh_disable or local_bh_enable.
+ * This lets us distinguish between whether we are currently processing
+ * softirq and whether we just have bh disabled.
+ */
+
 /*
  * This one is for softirq.c-internal use,
  * where hardirqs are disabled legitimately:
  */
 #ifdef CONFIG_TRACE_IRQFLAGS
-static void __local_bh_disable(unsigned long ip)
+static void __local_bh_disable(unsigned long ip, unsigned int cnt)
 {
 	unsigned long flags;
 
@@ -95,32 +105,43 @@ static void __local_bh_disable(unsigned long ip)
 	 * We must manually increment preempt_count here and manually
 	 * call the trace_preempt_off later.
 	 */
-	preempt_count() += SOFTIRQ_OFFSET;
+	preempt_count() += cnt;
 	/*
 	 * Were softirqs turned off above:
 	 */
-	if (softirq_count() == SOFTIRQ_OFFSET)
+	if (softirq_count() == cnt)
 		trace_softirqs_off(ip);
 	raw_local_irq_restore(flags);
 
-	if (preempt_count() == SOFTIRQ_OFFSET)
+	if (preempt_count() == cnt)
 		trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1));
 }
 #else /* !CONFIG_TRACE_IRQFLAGS */
-static inline void __local_bh_disable(unsigned long ip)
+static inline void __local_bh_disable(unsigned long ip, unsigned int cnt)
 {
-	add_preempt_count(SOFTIRQ_OFFSET);
+	add_preempt_count(cnt);
 	barrier();
 }
 #endif /* CONFIG_TRACE_IRQFLAGS */
 
 void local_bh_disable(void)
 {
-	__local_bh_disable((unsigned long)__builtin_return_address(0));
+	__local_bh_disable((unsigned long)__builtin_return_address(0),
+				SOFTIRQ_DISABLE_OFFSET);
 }
 
 EXPORT_SYMBOL(local_bh_disable);
 
+static void __local_bh_enable(unsigned int cnt)
+{
+	WARN_ON_ONCE(in_irq());
+	WARN_ON_ONCE(!irqs_disabled());
+
+	if (softirq_count() == cnt)
+		trace_softirqs_on((unsigned long)__builtin_return_address(0));
+	sub_preempt_count(cnt);
+}
+
 /*
  * Special-case - softirqs can safely be enabled in
  * cond_resched_softirq(), or by __do_softirq(),
@@ -128,12 +149,7 @@ EXPORT_SYMBOL(local_bh_disable);
  */
 void _local_bh_enable(void)
 {
-	WARN_ON_ONCE(in_irq());
-	WARN_ON_ONCE(!irqs_disabled());
-
-	if (softirq_count() == SOFTIRQ_OFFSET)
-		trace_softirqs_on((unsigned long)__builtin_return_address(0));
-	sub_preempt_count(SOFTIRQ_OFFSET);
+	__local_bh_enable(SOFTIRQ_DISABLE_OFFSET);
 }
 
 EXPORT_SYMBOL(_local_bh_enable);
@@ -147,13 +163,13 @@ static inline void _local_bh_enable_ip(unsigned long ip)
 	/*
 	 * Are softirqs going to be turned on now:
 	 */
-	if (softirq_count() == SOFTIRQ_OFFSET)
+	if (softirq_count() == SOFTIRQ_DISABLE_OFFSET)
 		trace_softirqs_on(ip);
 	/*
 	 * Keep preemption disabled until we are done with
 	 * softirq processing:
  	 */
- 	sub_preempt_count(SOFTIRQ_OFFSET - 1);
+	sub_preempt_count(SOFTIRQ_DISABLE_OFFSET - 1);
 
 	if (unlikely(!in_interrupt() && local_softirq_pending()))
 		do_softirq();
@@ -198,7 +214,8 @@ asmlinkage void __do_softirq(void)
 	pending = local_softirq_pending();
 	account_system_vtime(current);
 
-	__local_bh_disable((unsigned long)__builtin_return_address(0));
+	__local_bh_disable((unsigned long)__builtin_return_address(0),
+				SOFTIRQ_OFFSET);
 	lockdep_softirq_enter();
 
 	cpu = smp_processor_id();
@@ -245,7 +262,7 @@ restart:
 	lockdep_softirq_exit();
 
 	account_system_vtime(current);
-	_local_bh_enable();
+	__local_bh_enable(SOFTIRQ_OFFSET);
 }
 
 #ifndef __ARCH_HAS_DO_SOFTIRQ
diff --git a/net/sched/cls_cgroup.c b/net/sched/cls_cgroup.c
index 78ef2c5e130b..37dff78e9cb1 100644
--- a/net/sched/cls_cgroup.c
+++ b/net/sched/cls_cgroup.c
@@ -123,7 +123,7 @@ static int cls_cgroup_classify(struct sk_buff *skb, struct tcf_proto *tp,
 	 * calls by looking at the number of nested bh disable calls because
 	 * softirqs always disables bh.
 	 */
-	if (softirq_count() != SOFTIRQ_OFFSET) {
+	if (in_serving_softirq()) {
 		/* If there is an sk_classid we'll use that. */
 		if (!skb->sk)
 			return -1;
-- 
cgit v1.2.3


From e1e10a265d28273ab8c70be19d43dcbdeead6c5a Mon Sep 17 00:00:00 2001
From: Venkatesh Pallipadi <venki@google.com>
Date: Mon, 4 Oct 2010 17:03:17 -0700
Subject: sched: Consolidate account_system_vtime extern declaration

Just a minor cleanup patch that makes things easier to the following patches.
No functionality change in this patch.

Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-3-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/ia64/include/asm/system.h    | 4 ----
 arch/powerpc/include/asm/system.h | 4 ----
 arch/s390/include/asm/system.h    | 1 -
 include/linux/hardirq.h           | 2 ++
 4 files changed, 2 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/include/asm/system.h b/arch/ia64/include/asm/system.h
index 9f342a574ce8..dd028f2b13b3 100644
--- a/arch/ia64/include/asm/system.h
+++ b/arch/ia64/include/asm/system.h
@@ -272,10 +272,6 @@ void cpu_idle_wait(void);
 
 void default_idle(void);
 
-#ifdef CONFIG_VIRT_CPU_ACCOUNTING
-extern void account_system_vtime(struct task_struct *);
-#endif
-
 #endif /* __KERNEL__ */
 
 #endif /* __ASSEMBLY__ */
diff --git a/arch/powerpc/include/asm/system.h b/arch/powerpc/include/asm/system.h
index 6c294acac848..9c3d160670b4 100644
--- a/arch/powerpc/include/asm/system.h
+++ b/arch/powerpc/include/asm/system.h
@@ -542,10 +542,6 @@ extern void reloc_got2(unsigned long);
 
 #define PTRRELOC(x)	((typeof(x)) add_reloc_offset((unsigned long)(x)))
 
-#ifdef CONFIG_VIRT_CPU_ACCOUNTING
-extern void account_system_vtime(struct task_struct *);
-#endif
-
 extern struct dentry *powerpc_debugfs_root;
 
 #endif /* __KERNEL__ */
diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h
index cef66210c846..38ddd8a9a9e8 100644
--- a/arch/s390/include/asm/system.h
+++ b/arch/s390/include/asm/system.h
@@ -97,7 +97,6 @@ static inline void restore_access_regs(unsigned int *acrs)
 
 extern void account_vtime(struct task_struct *, struct task_struct *);
 extern void account_tick_vtime(struct task_struct *);
-extern void account_system_vtime(struct task_struct *);
 
 #ifdef CONFIG_PFAULT
 extern void pfault_irq_init(void);
diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index e37a77cbd588..41367c5c3c68 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -141,6 +141,8 @@ struct task_struct;
 static inline void account_system_vtime(struct task_struct *tsk)
 {
 }
+#else
+extern void account_system_vtime(struct task_struct *tsk);
 #endif
 
 #if defined(CONFIG_NO_HZ)
-- 
cgit v1.2.3


From 6cdd5199daf0cb7b0fcc8dca941af08492612887 Mon Sep 17 00:00:00 2001
From: Venkatesh Pallipadi <venki@google.com>
Date: Mon, 4 Oct 2010 17:03:18 -0700
Subject: sched: Add a PF flag for ksoftirqd identification

To account softirq time cleanly in scheduler, we need to identify whether
softirq is invoked in ksoftirqd context or softirq at hardirq tail context.
Add PF_KSOFTIRQD for that purpose.

As all PF flag bits are currently taken, create space by moving one of the
infrequently used bits (PF_THREAD_BOUND) down in task_struct to be along
with some other state fields.

Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-4-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 1 +
 kernel/softirq.c      | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 8744e50cb083..aca0ce675939 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1682,6 +1682,7 @@ extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *
 /*
  * Per process flags
  */
+#define PF_KSOFTIRQD	0x00000001	/* I am ksoftirqd */
 #define PF_STARTING	0x00000002	/* being created */
 #define PF_EXITING	0x00000004	/* getting shut down */
 #define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 988dfbe6bbe8..267f7b763ebb 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -713,6 +713,7 @@ static int run_ksoftirqd(void * __bind_cpu)
 {
 	set_current_state(TASK_INTERRUPTIBLE);
 
+	current->flags |= PF_KSOFTIRQD;
 	while (!kthread_should_stop()) {
 		preempt_disable();
 		if (!local_softirq_pending()) {
-- 
cgit v1.2.3


From b52bfee445d315549d41eacf2fa7c156e7d153d5 Mon Sep 17 00:00:00 2001
From: Venkatesh Pallipadi <venki@google.com>
Date: Mon, 4 Oct 2010 17:03:19 -0700
Subject: sched: Add IRQ_TIME_ACCOUNTING, finer accounting of irq time

s390/powerpc/ia64 have support for CONFIG_VIRT_CPU_ACCOUNTING which does
the fine granularity accounting of user, system, hardirq, softirq times.
Adding that option on archs like x86 will be challenging however, given the
state of TSC reliability on various platforms and also the overhead it will
add in syscall entry exit.

Instead, add a lighter variant that only does finer accounting of
hardirq and softirq times, providing precise irq times (instead of timer tick
based samples). This accounting is added with a new config option
CONFIG_IRQ_TIME_ACCOUNTING so that there won't be any overhead for users not
interested in paying the perf penalty.

This accounting is based on sched_clock, with the code being generic.
So, other archs may find it useful as well.

This patch just adds the core logic and does not enable this logic yet.

Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1286237003-12406-5-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/hardirq.h |  2 +-
 include/linux/sched.h   | 13 +++++++++++++
 kernel/sched.c          | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 63 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index 41367c5c3c68..ff43e9268449 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -137,7 +137,7 @@ extern void synchronize_irq(unsigned int irq);
 
 struct task_struct;
 
-#ifndef CONFIG_VIRT_CPU_ACCOUNTING
+#if !defined(CONFIG_VIRT_CPU_ACCOUNTING) && !defined(CONFIG_IRQ_TIME_ACCOUNTING)
 static inline void account_system_vtime(struct task_struct *tsk)
 {
 }
diff --git a/include/linux/sched.h b/include/linux/sched.h
index aca0ce675939..2cca9a92f5e5 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1826,6 +1826,19 @@ extern void sched_clock_idle_sleep_event(void);
 extern void sched_clock_idle_wakeup_event(u64 delta_ns);
 #endif
 
+#ifdef CONFIG_IRQ_TIME_ACCOUNTING
+/*
+ * An i/f to runtime opt-in for irq time accounting based off of sched_clock.
+ * The reason for this explicit opt-in is not to have perf penalty with
+ * slow sched_clocks.
+ */
+extern void enable_sched_clock_irqtime(void);
+extern void disable_sched_clock_irqtime(void);
+#else
+static inline void enable_sched_clock_irqtime(void) {}
+static inline void disable_sched_clock_irqtime(void) {}
+#endif
+
 extern unsigned long long
 task_sched_runtime(struct task_struct *task);
 extern unsigned long long thread_group_sched_runtime(struct task_struct *task);
diff --git a/kernel/sched.c b/kernel/sched.c
index 089be8adb074..9b302e355791 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1908,6 +1908,55 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int flags)
 	dec_nr_running(rq);
 }
 
+#ifdef CONFIG_IRQ_TIME_ACCOUNTING
+
+static DEFINE_PER_CPU(u64, cpu_hardirq_time);
+static DEFINE_PER_CPU(u64, cpu_softirq_time);
+
+static DEFINE_PER_CPU(u64, irq_start_time);
+static int sched_clock_irqtime;
+
+void enable_sched_clock_irqtime(void)
+{
+	sched_clock_irqtime = 1;
+}
+
+void disable_sched_clock_irqtime(void)
+{
+	sched_clock_irqtime = 0;
+}
+
+void account_system_vtime(struct task_struct *curr)
+{
+	unsigned long flags;
+	int cpu;
+	u64 now, delta;
+
+	if (!sched_clock_irqtime)
+		return;
+
+	local_irq_save(flags);
+
+	now = sched_clock();
+	cpu = smp_processor_id();
+	delta = now - per_cpu(irq_start_time, cpu);
+	per_cpu(irq_start_time, cpu) = now;
+	/*
+	 * We do not account for softirq time from ksoftirqd here.
+	 * We want to continue accounting softirq time to ksoftirqd thread
+	 * in that case, so as not to confuse scheduler with a special task
+	 * that do not consume any time, but still wants to run.
+	 */
+	if (hardirq_count())
+		per_cpu(cpu_hardirq_time, cpu) += delta;
+	else if (in_serving_softirq() && !(curr->flags & PF_KSOFTIRQD))
+		per_cpu(cpu_softirq_time, cpu) += delta;
+
+	local_irq_restore(flags);
+}
+
+#endif
+
 #include "sched_idletask.c"
 #include "sched_fair.c"
 #include "sched_rt.c"
-- 
cgit v1.2.3


From 7681bfeeccff5efa9eb29bf09249a3c400b15327 Mon Sep 17 00:00:00 2001
From: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Date: Tue, 19 Oct 2010 09:05:00 +0200
Subject: block: fix accounting bug on cross partition merges

/proc/diskstats would display a strange output as follows.

$ cat /proc/diskstats |grep sda
   8       0 sda 90524 7579 102154 20464 0 0 0 0 0 14096 20089
   8       1 sda1 19085 1352 21841 4209 0 0 0 0 4294967064 15689 4293424691
                                                ~~~~~~~~~~
   8       2 sda2 71252 3624 74891 15950 0 0 0 0 232 23995 1562390
   8       3 sda3 54 487 2188 92 0 0 0 0 0 88 92
   8       4 sda4 4 0 8 0 0 0 0 0 0 0 0
   8       5 sda5 81 2027 2130 138 0 0 0 0 0 87 137

Its reason is the wrong way of accounting hd_struct->in_flight. When a bio is
merged into a request belongs to different partition by ELEVATOR_FRONT_MERGE.

The detailed root cause is as follows.

Assuming that there are two partition, sda1 and sda2.

1. A request for sda2 is in request_queue. Hence sda1's hd_struct->in_flight
   is 0 and sda2's one is 1.

        | hd_struct->in_flight
   ---------------------------
   sda1 |          0
   sda2 |          1
   ---------------------------

2. A bio belongs to sda1 is issued and is merged into the request mentioned on
   step1 by ELEVATOR_BACK_MERGE. The first sector of the request is changed
   from sda2 region to sda1 region. However the two partition's
   hd_struct->in_flight are not changed.

        | hd_struct->in_flight
   ---------------------------
   sda1 |          0
   sda2 |          1
   ---------------------------

3. The request is finished and blk_account_io_done() is called. In this case,
   sda2's hd_struct->in_flight, not a sda1's one, is decremented.

        | hd_struct->in_flight
   ---------------------------
   sda1 |         -1
   sda2 |          1
   ---------------------------

The patch fixes the problem by caching the partition lookup
inside the request structure, hence making sure that the increment
and decrement will always happen on the same partition struct. This
also speeds up IO with accounting enabled, since it cuts down on
the number of lookups we have to do.

When reloading partition tables, quiesce IO to ensure that no
request references to the partition struct exists. When it is safe
to free the partition table, the IO for that device is restarted
again.

Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 block/blk-core.c         | 24 ++++++++++++++++--------
 block/blk-merge.c        |  2 +-
 block/blk.h              |  4 ----
 block/genhd.c            | 14 ++++++++++++++
 fs/partitions/check.c    | 12 ++++++++++++
 include/linux/blkdev.h   |  1 +
 include/linux/elevator.h |  2 ++
 include/linux/genhd.h    |  1 +
 8 files changed, 47 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-core.c b/block/blk-core.c
index 797d5095eb83..ddc68332d655 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -64,13 +64,15 @@ static void drive_stat_acct(struct request *rq, int new_io)
 		return;
 
 	cpu = part_stat_lock();
-	part = disk_map_sector_rcu(rq->rq_disk, blk_rq_pos(rq));
 
-	if (!new_io)
+	if (!new_io) {
+		part = rq->part;
 		part_stat_inc(cpu, part, merges[rw]);
-	else {
+	} else {
+		part = disk_map_sector_rcu(rq->rq_disk, blk_rq_pos(rq));
 		part_round_stats(cpu, part);
 		part_inc_in_flight(part, rw);
+		rq->part = part;
 	}
 
 	part_stat_unlock();
@@ -128,6 +130,7 @@ void blk_rq_init(struct request_queue *q, struct request *rq)
 	rq->ref_count = 1;
 	rq->start_time = jiffies;
 	set_start_time_ns(rq);
+	rq->part = NULL;
 }
 EXPORT_SYMBOL(blk_rq_init);
 
@@ -804,11 +807,16 @@ static struct request *get_request(struct request_queue *q, int rw_flags,
 	rl->starved[is_sync] = 0;
 
 	priv = !test_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
-	if (priv)
+	if (priv) {
 		rl->elvpriv++;
 
-	if (blk_queue_io_stat(q))
-		rw_flags |= REQ_IO_STAT;
+		/*
+		 * Don't do stats for non-priv requests
+		 */
+		if (blk_queue_io_stat(q))
+			rw_flags |= REQ_IO_STAT;
+	}
+
 	spin_unlock_irq(q->queue_lock);
 
 	rq = blk_alloc_request(q, rw_flags, priv, gfp_mask);
@@ -1777,7 +1785,7 @@ static void blk_account_io_completion(struct request *req, unsigned int bytes)
 		int cpu;
 
 		cpu = part_stat_lock();
-		part = disk_map_sector_rcu(req->rq_disk, blk_rq_pos(req));
+		part = req->part;
 		part_stat_add(cpu, part, sectors[rw], bytes >> 9);
 		part_stat_unlock();
 	}
@@ -1797,7 +1805,7 @@ static void blk_account_io_done(struct request *req)
 		int cpu;
 
 		cpu = part_stat_lock();
-		part = disk_map_sector_rcu(req->rq_disk, blk_rq_pos(req));
+		part = req->part;
 
 		part_stat_inc(cpu, part, ios[rw]);
 		part_stat_add(cpu, part, ticks[rw], duration);
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 6a725461654d..38ff234012a4 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -351,7 +351,7 @@ static void blk_account_io_merge(struct request *req)
 		int cpu;
 
 		cpu = part_stat_lock();
-		part = disk_map_sector_rcu(req->rq_disk, blk_rq_pos(req));
+		part = req->part;
 
 		part_round_stats(cpu, part);
 		part_dec_in_flight(part, rq_data_dir(req));
diff --git a/block/blk.h b/block/blk.h
index 6738831ba447..1340cce5721a 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -110,10 +110,6 @@ void blk_queue_congestion_threshold(struct request_queue *q);
 
 int blk_dev_init(void);
 
-void elv_quiesce_start(struct request_queue *q);
-void elv_quiesce_end(struct request_queue *q);
-
-
 /*
  * Return the threshold (number of used requests) at which the queue is
  * considered to be congested.  It include a little hysteresis to keep the
diff --git a/block/genhd.c b/block/genhd.c
index 7923e720ddf5..8313834596db 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -932,8 +932,15 @@ static void disk_free_ptbl_rcu_cb(struct rcu_head *head)
 {
 	struct disk_part_tbl *ptbl =
 		container_of(head, struct disk_part_tbl, rcu_head);
+	struct gendisk *disk = ptbl->disk;
+	struct request_queue *q = disk->queue;
+	unsigned long flags;
 
 	kfree(ptbl);
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	elv_quiesce_end(q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 
 /**
@@ -951,11 +958,17 @@ static void disk_replace_part_tbl(struct gendisk *disk,
 				  struct disk_part_tbl *new_ptbl)
 {
 	struct disk_part_tbl *old_ptbl = disk->part_tbl;
+	struct request_queue *q = disk->queue;
 
 	rcu_assign_pointer(disk->part_tbl, new_ptbl);
 
 	if (old_ptbl) {
 		rcu_assign_pointer(old_ptbl->last_lookup, NULL);
+
+		spin_lock_irq(q->queue_lock);
+		elv_quiesce_start(q);
+		spin_unlock_irq(q->queue_lock);
+
 		call_rcu(&old_ptbl->rcu_head, disk_free_ptbl_rcu_cb);
 	}
 }
@@ -996,6 +1009,7 @@ int disk_expand_part_tbl(struct gendisk *disk, int partno)
 		return -ENOMEM;
 
 	new_ptbl->len = target;
+	new_ptbl->disk = disk;
 
 	for (i = 0; i < len; i++)
 		rcu_assign_pointer(new_ptbl->part[i], old_ptbl->part[i]);
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 6dfbee03ccc6..30f46c2cb9d5 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -365,17 +365,25 @@ struct device_type part_type = {
 static void delete_partition_rcu_cb(struct rcu_head *head)
 {
 	struct hd_struct *part = container_of(head, struct hd_struct, rcu_head);
+	struct gendisk *disk = part_to_disk(part);
+	struct request_queue *q = disk->queue;
+	unsigned long flags;
 
 	part->start_sect = 0;
 	part->nr_sects = 0;
 	part_stat_set_all(part, 0);
 	put_device(part_to_dev(part));
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	elv_quiesce_end(q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 
 void delete_partition(struct gendisk *disk, int partno)
 {
 	struct disk_part_tbl *ptbl = disk->part_tbl;
 	struct hd_struct *part;
+	struct request_queue *q = disk->queue;
 
 	if (partno >= ptbl->len)
 		return;
@@ -390,6 +398,10 @@ void delete_partition(struct gendisk *disk, int partno)
 	kobject_put(part->holder_dir);
 	device_del(part_to_dev(part));
 
+	spin_lock_irq(q->queue_lock);
+	elv_quiesce_start(q);
+	spin_unlock_irq(q->queue_lock);
+
 	call_rcu(&part->rcu_head, delete_partition_rcu_cb);
 }
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 8f3dd981b973..16f7f1be1acf 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -115,6 +115,7 @@ struct request {
 	void *elevator_private3;
 
 	struct gendisk *rq_disk;
+	struct hd_struct *part;
 	unsigned long start_time;
 #ifdef CONFIG_BLK_CGROUP
 	unsigned long long start_time_ns;
diff --git a/include/linux/elevator.h b/include/linux/elevator.h
index 2c958f4fce1e..df1ee866d715 100644
--- a/include/linux/elevator.h
+++ b/include/linux/elevator.h
@@ -121,6 +121,8 @@ extern void elv_completed_request(struct request_queue *, struct request *);
 extern int elv_set_request(struct request_queue *, struct request *, gfp_t);
 extern void elv_put_request(struct request_queue *, struct request *);
 extern void elv_drain_elevator(struct request_queue *);
+extern void elv_quiesce_start(struct request_queue *);
+extern void elv_quiesce_end(struct request_queue *);
 
 /*
  * io scheduler registration
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 66e26b5a1537..57647ecfc1bd 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -140,6 +140,7 @@ struct disk_part_tbl {
 	struct rcu_head rcu_head;
 	int len;
 	struct hd_struct *last_lookup;
+	struct gendisk *disk;
 	struct hd_struct *part[];
 };
 
-- 
cgit v1.2.3


From 6362beea8914cbd4630ccde3617d944aeca2d48f Mon Sep 17 00:00:00 2001
From: Mike Miller <mike.miller@hp.com>
Date: Tue, 19 Oct 2010 09:40:34 +0200
Subject: cciss: fix PCI IDs for new Smart Array controllers

cciss: fix PCI IDs for new controllers

This patch fixes the botched up PCI IDs of new controllers. Please consider
this patch for inclusion.

Signed-off-by: Mike Miller <mike.miller@hp.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 drivers/block/cciss.c   | 22 ++++++++++++----------
 include/linux/pci_ids.h |  1 +
 2 files changed, 13 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index df2ed4d6684d..41f9acb8ecd7 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -105,11 +105,12 @@ static const struct pci_device_id cciss_pci_device_id[] = {
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSE,     0x103C, 0x3249},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSE,     0x103C, 0x324A},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSE,     0x103C, 0x324B},
-	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSE,     0x103C, 0x3250},
-	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSE,     0x103C, 0x3251},
-	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSE,     0x103C, 0x3252},
-	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSE,     0x103C, 0x3253},
-	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSE,     0x103C, 0x3254},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3350},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3351},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3352},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3353},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3354},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3355},
 	{0,}
 };
 
@@ -149,11 +150,12 @@ static struct board_type products[] = {
 	{0x3249103C, "Smart Array P812", &SA5_access},
 	{0x324A103C, "Smart Array P712m", &SA5_access},
 	{0x324B103C, "Smart Array P711m", &SA5_access},
-	{0x3250103C, "Smart Array", &SA5_access},
-	{0x3251103C, "Smart Array", &SA5_access},
-	{0x3252103C, "Smart Array", &SA5_access},
-	{0x3253103C, "Smart Array", &SA5_access},
-	{0x3254103C, "Smart Array", &SA5_access},
+	{0x3350103C, "Smart Array", &SA5_access},
+	{0x3351103C, "Smart Array", &SA5_access},
+	{0x3352103C, "Smart Array", &SA5_access},
+	{0x3353103C, "Smart Array", &SA5_access},
+	{0x3354103C, "Smart Array", &SA5_access},
+	{0x3355103C, "Smart Array", &SA5_access},
 };
 
 /* How long to wait (in milliseconds) for board to go into simple mode */
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index f6a3b2d36cad..fe7d72ce33af 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -739,6 +739,7 @@
 #define PCI_DEVICE_ID_HP_CISSC		0x3230
 #define PCI_DEVICE_ID_HP_CISSD		0x3238
 #define PCI_DEVICE_ID_HP_CISSE		0x323a
+#define PCI_DEVICE_ID_HP_CISSF		0x323b
 #define PCI_DEVICE_ID_HP_ZX2_IOC	0x4031
 
 #define PCI_VENDOR_ID_PCTECH		0x1042
-- 
cgit v1.2.3


From ebbf41df4aabb6d506fa18ea8cb4c2b4388a18b9 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Tue, 19 Oct 2010 10:19:06 +0200
Subject: netfilter: ctnetlink: add expectation deletion events

This patch allows to listen to events that inform about
expectations destroyed.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/nf_conntrack_common.h |  1 +
 include/net/netfilter/nf_conntrack_expect.h   |  8 ++++++-
 net/netfilter/nf_conntrack_expect.c           |  6 ++++--
 net/netfilter/nf_conntrack_netlink.c          | 30 +++++++++++++++++++--------
 4 files changed, 33 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h
index 23a1a08578a8..50cdc2559a5a 100644
--- a/include/linux/netfilter/nf_conntrack_common.h
+++ b/include/linux/netfilter/nf_conntrack_common.h
@@ -98,6 +98,7 @@ enum ip_conntrack_events {
 
 enum ip_conntrack_expect_events {
 	IPEXP_NEW,		/* new expectation */
+	IPEXP_DESTROY,		/* destroyed expectation */
 };
 
 /* expectation flags */
diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index 416b83844485..0f8a8c587532 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -82,7 +82,13 @@ struct nf_conntrack_expect *
 nf_ct_find_expectation(struct net *net, u16 zone,
 		       const struct nf_conntrack_tuple *tuple);
 
-void nf_ct_unlink_expect(struct nf_conntrack_expect *exp);
+void nf_ct_unlink_expect_report(struct nf_conntrack_expect *exp,
+				u32 pid, int report);
+static inline void nf_ct_unlink_expect(struct nf_conntrack_expect *exp)
+{
+	nf_ct_unlink_expect_report(exp, 0, 0);
+}
+
 void nf_ct_remove_expectations(struct nf_conn *ct);
 void nf_ct_unexpect_related(struct nf_conntrack_expect *exp);
 void nf_ct_remove_userspace_expectations(void);
diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index b30a1f2aac00..46e8966912b1 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -41,7 +41,8 @@ static struct kmem_cache *nf_ct_expect_cachep __read_mostly;
 static HLIST_HEAD(nf_ct_userspace_expect_list);
 
 /* nf_conntrack_expect helper functions */
-void nf_ct_unlink_expect(struct nf_conntrack_expect *exp)
+void nf_ct_unlink_expect_report(struct nf_conntrack_expect *exp,
+				u32 pid, int report)
 {
 	struct nf_conn_help *master_help = nfct_help(exp->master);
 	struct net *net = nf_ct_exp_net(exp);
@@ -55,11 +56,12 @@ void nf_ct_unlink_expect(struct nf_conntrack_expect *exp)
 	if (!(exp->flags & NF_CT_EXPECT_USERSPACE))
 		master_help->expecting[exp->class]--;
 
+	nf_ct_expect_event_report(IPEXP_DESTROY, exp, pid, report);
 	nf_ct_expect_put(exp);
 
 	NF_CT_STAT_INC(net, expect_delete);
 }
-EXPORT_SYMBOL_GPL(nf_ct_unlink_expect);
+EXPORT_SYMBOL_GPL(nf_ct_unlink_expect_report);
 
 static void nf_ct_expectation_timed_out(unsigned long ul_expect)
 {
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index b4077be5f663..62bad229106b 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -1632,17 +1632,20 @@ ctnetlink_expect_event(unsigned int events, struct nf_exp_event *item)
 	struct nlmsghdr *nlh;
 	struct nfgenmsg *nfmsg;
 	struct sk_buff *skb;
-	unsigned int type;
+	unsigned int type, group;
 	int flags = 0;
 
-	if (events & (1 << IPEXP_NEW)) {
+	if (events & (1 << IPEXP_DESTROY)) {
+		type = IPCTNL_MSG_EXP_DELETE;
+		group = NFNLGRP_CONNTRACK_EXP_DESTROY;
+	} else if (events & (1 << IPEXP_NEW)) {
 		type = IPCTNL_MSG_EXP_NEW;
 		flags = NLM_F_CREATE|NLM_F_EXCL;
+		group = NFNLGRP_CONNTRACK_EXP_NEW;
 	} else
 		return 0;
 
-	if (!item->report &&
-	    !nfnetlink_has_listeners(net, NFNLGRP_CONNTRACK_EXP_NEW))
+	if (!item->report && !nfnetlink_has_listeners(net, group))
 		return 0;
 
 	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
@@ -1665,8 +1668,7 @@ ctnetlink_expect_event(unsigned int events, struct nf_exp_event *item)
 	rcu_read_unlock();
 
 	nlmsg_end(skb, nlh);
-	nfnetlink_send(skb, net, item->pid, NFNLGRP_CONNTRACK_EXP_NEW,
-		       item->report, GFP_ATOMIC);
+	nfnetlink_send(skb, net, item->pid, group, item->report, GFP_ATOMIC);
 	return 0;
 
 nla_put_failure:
@@ -1849,7 +1851,13 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb,
 		}
 
 		/* after list removal, usage count == 1 */
-		nf_ct_unexpect_related(exp);
+		spin_lock_bh(&nf_conntrack_lock);
+		if (del_timer(&exp->timeout)) {
+			nf_ct_unlink_expect_report(exp, NETLINK_CB(skb).pid,
+						   nlmsg_report(nlh));
+			nf_ct_expect_put(exp);
+		}
+		spin_unlock_bh(&nf_conntrack_lock);
 		/* have to put what we 'get' above.
 		 * after this line usage count == 0 */
 		nf_ct_expect_put(exp);
@@ -1866,7 +1874,9 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb,
 				m_help = nfct_help(exp->master);
 				if (!strcmp(m_help->helper->name, name) &&
 				    del_timer(&exp->timeout)) {
-					nf_ct_unlink_expect(exp);
+					nf_ct_unlink_expect_report(exp,
+							NETLINK_CB(skb).pid,
+							nlmsg_report(nlh));
 					nf_ct_expect_put(exp);
 				}
 			}
@@ -1880,7 +1890,9 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb,
 						  &net->ct.expect_hash[i],
 						  hnode) {
 				if (del_timer(&exp->timeout)) {
-					nf_ct_unlink_expect(exp);
+					nf_ct_unlink_expect_report(exp,
+							NETLINK_CB(skb).pid,
+							nlmsg_report(nlh));
 					nf_ct_expect_put(exp);
 				}
 			}
-- 
cgit v1.2.3


From daaae6b010ac0f60c9c35e481589966f9f1fcc22 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Tue, 19 Oct 2010 11:28:15 +0200
Subject: workqueue: remove in_workqueue_context()

Commit a25909a4 (lockdep: Add an in_workqueue_context() lockdep-based
test function) added in_workqueue_context() but there hasn't been any
in-kernel user and the lockdep annotation in workqueue is scheduled to
change.  Remove the unused function.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 include/linux/workqueue.h |  4 ----
 kernel/workqueue.c        | 15 ---------------
 2 files changed, 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 03bbe903e5ce..070bb7a88936 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -430,8 +430,4 @@ extern bool freeze_workqueues_busy(void);
 extern void thaw_workqueues(void);
 #endif /* CONFIG_FREEZER */
 
-#ifdef CONFIG_LOCKDEP
-int in_workqueue_context(struct workqueue_struct *wq);
-#endif
-
 #endif
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index eb5c1972443a..30acdb74cc23 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -310,21 +310,6 @@ static inline int __next_wq_cpu(int cpu, const struct cpumask *mask,
 	     (cpu) < WORK_CPU_NONE;					\
 	     (cpu) = __next_wq_cpu((cpu), cpu_possible_mask, (wq)))
 
-#ifdef CONFIG_LOCKDEP
-/**
- * in_workqueue_context() - in context of specified workqueue?
- * @wq: the workqueue of interest
- *
- * Checks lockdep state to see if the current task is executing from
- * within a workqueue item.  This function exists only if lockdep is
- * enabled.
- */
-int in_workqueue_context(struct workqueue_struct *wq)
-{
-	return lock_is_held(&wq->lockdep_map);
-}
-#endif
-
 #ifdef CONFIG_DEBUG_OBJECTS_WORK
 
 static struct debug_obj_descr work_debug_descr;
-- 
cgit v1.2.3


From e6484930d7c73d324bccda7d43d131088da697b9 Mon Sep 17 00:00:00 2001
From: Tom Herbert <therbert@google.com>
Date: Mon, 18 Oct 2010 18:04:39 +0000
Subject: net: allocate tx queues in register_netdevice

This patch introduces netif_alloc_netdev_queues which is called from
register_device instead of alloc_netdev_mq.  This makes TX queue
allocation symmetric with RX allocation.  Also, queue locks allocation
is done in netdev_init_one_queue.  Change set_real_num_tx_queues to
fail if requested number < 1 or greater than number of allocated
queues.

Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |   4 +-
 net/core/dev.c            | 106 +++++++++++++++++++++++-----------------------
 2 files changed, 55 insertions(+), 55 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 14fbb04c459d..880d56565828 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1696,8 +1696,8 @@ static inline int netif_is_multiqueue(const struct net_device *dev)
 	return dev->num_tx_queues > 1;
 }
 
-extern void netif_set_real_num_tx_queues(struct net_device *dev,
-					 unsigned int txq);
+extern int netif_set_real_num_tx_queues(struct net_device *dev,
+					unsigned int txq);
 
 #ifdef CONFIG_RPS
 extern int netif_set_real_num_rx_queues(struct net_device *dev,
diff --git a/net/core/dev.c b/net/core/dev.c
index d33adecec44b..4c3ac53e4b16 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1553,18 +1553,20 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
  * Routine to help set real_num_tx_queues. To avoid skbs mapped to queues
  * greater then real_num_tx_queues stale skbs on the qdisc must be flushed.
  */
-void netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq)
+int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq)
 {
-	unsigned int real_num = dev->real_num_tx_queues;
+	if (txq < 1 || txq > dev->num_tx_queues)
+		return -EINVAL;
 
-	if (unlikely(txq > dev->num_tx_queues))
-		;
-	else if (txq > real_num)
-		dev->real_num_tx_queues = txq;
-	else if (txq < real_num) {
-		dev->real_num_tx_queues = txq;
-		qdisc_reset_all_tx_gt(dev, txq);
+	if (dev->reg_state == NETREG_REGISTERED) {
+		ASSERT_RTNL();
+
+		if (txq < dev->real_num_tx_queues)
+			qdisc_reset_all_tx_gt(dev, txq);
 	}
+
+	dev->real_num_tx_queues = txq;
+	return 0;
 }
 EXPORT_SYMBOL(netif_set_real_num_tx_queues);
 
@@ -4928,20 +4930,6 @@ static void rollback_registered(struct net_device *dev)
 	rollback_registered_many(&single);
 }
 
-static void __netdev_init_queue_locks_one(struct net_device *dev,
-					  struct netdev_queue *dev_queue,
-					  void *_unused)
-{
-	spin_lock_init(&dev_queue->_xmit_lock);
-	netdev_set_xmit_lockdep_class(&dev_queue->_xmit_lock, dev->type);
-	dev_queue->xmit_lock_owner = -1;
-}
-
-static void netdev_init_queue_locks(struct net_device *dev)
-{
-	netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL);
-}
-
 unsigned long netdev_fix_features(unsigned long features, const char *name)
 {
 	/* Fix illegal SG+CSUM combinations. */
@@ -5034,6 +5022,41 @@ static int netif_alloc_rx_queues(struct net_device *dev)
 	return 0;
 }
 
+static int netif_alloc_netdev_queues(struct net_device *dev)
+{
+	unsigned int count = dev->num_tx_queues;
+	struct netdev_queue *tx;
+
+	BUG_ON(count < 1);
+
+	tx = kcalloc(count, sizeof(struct netdev_queue), GFP_KERNEL);
+	if (!tx) {
+		pr_err("netdev: Unable to allocate %u tx queues.\n",
+		       count);
+		return -ENOMEM;
+	}
+	dev->_tx = tx;
+	return 0;
+}
+
+static void netdev_init_one_queue(struct net_device *dev,
+				  struct netdev_queue *queue,
+				  void *_unused)
+{
+	queue->dev = dev;
+
+	/* Initialize queue lock */
+	spin_lock_init(&queue->_xmit_lock);
+	netdev_set_xmit_lockdep_class(&queue->_xmit_lock, dev->type);
+	queue->xmit_lock_owner = -1;
+}
+
+static void netdev_init_queues(struct net_device *dev)
+{
+	netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL);
+	spin_lock_init(&dev->tx_global_lock);
+}
+
 /**
  *	register_netdevice	- register a network device
  *	@dev: device to register
@@ -5067,7 +5090,6 @@ int register_netdevice(struct net_device *dev)
 
 	spin_lock_init(&dev->addr_list_lock);
 	netdev_set_addr_lockdep_class(dev);
-	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
 
@@ -5075,6 +5097,12 @@ int register_netdevice(struct net_device *dev)
 	if (ret)
 		goto out;
 
+	ret = netif_alloc_netdev_queues(dev);
+	if (ret)
+		goto out;
+
+	netdev_init_queues(dev);
+
 	/* Init, if this function is available */
 	if (dev->netdev_ops->ndo_init) {
 		ret = dev->netdev_ops->ndo_init(dev);
@@ -5456,19 +5484,6 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
 }
 EXPORT_SYMBOL(dev_get_stats);
 
-static void netdev_init_one_queue(struct net_device *dev,
-				  struct netdev_queue *queue,
-				  void *_unused)
-{
-	queue->dev = dev;
-}
-
-static void netdev_init_queues(struct net_device *dev)
-{
-	netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL);
-	spin_lock_init(&dev->tx_global_lock);
-}
-
 struct netdev_queue *dev_ingress_queue_create(struct net_device *dev)
 {
 	struct netdev_queue *queue = dev_ingress_queue(dev);
@@ -5480,7 +5495,6 @@ struct netdev_queue *dev_ingress_queue_create(struct net_device *dev)
 	if (!queue)
 		return NULL;
 	netdev_init_one_queue(dev, queue, NULL);
-	__netdev_init_queue_locks_one(dev, queue, NULL);
 	queue->qdisc = &noop_qdisc;
 	queue->qdisc_sleeping = &noop_qdisc;
 	rcu_assign_pointer(dev->ingress_queue, queue);
@@ -5502,7 +5516,6 @@ struct netdev_queue *dev_ingress_queue_create(struct net_device *dev)
 struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 		void (*setup)(struct net_device *), unsigned int queue_count)
 {
-	struct netdev_queue *tx;
 	struct net_device *dev;
 	size_t alloc_size;
 	struct net_device *p;
@@ -5530,20 +5543,12 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 		return NULL;
 	}
 
-	tx = kcalloc(queue_count, sizeof(struct netdev_queue), GFP_KERNEL);
-	if (!tx) {
-		printk(KERN_ERR "alloc_netdev: Unable to allocate "
-		       "tx qdiscs.\n");
-		goto free_p;
-	}
-
-
 	dev = PTR_ALIGN(p, NETDEV_ALIGN);
 	dev->padded = (char *)dev - (char *)p;
 
 	dev->pcpu_refcnt = alloc_percpu(int);
 	if (!dev->pcpu_refcnt)
-		goto free_tx;
+		goto free_p;
 
 	if (dev_addr_init(dev))
 		goto free_pcpu;
@@ -5553,7 +5558,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 
 	dev_net_set(dev, &init_net);
 
-	dev->_tx = tx;
 	dev->num_tx_queues = queue_count;
 	dev->real_num_tx_queues = queue_count;
 
@@ -5564,8 +5568,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 
 	dev->gso_max_size = GSO_MAX_SIZE;
 
-	netdev_init_queues(dev);
-
 	INIT_LIST_HEAD(&dev->ethtool_ntuple_list.list);
 	dev->ethtool_ntuple_list.count = 0;
 	INIT_LIST_HEAD(&dev->napi_list);
@@ -5576,8 +5578,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 	strcpy(dev->name, name);
 	return dev;
 
-free_tx:
-	kfree(tx);
 free_pcpu:
 	free_percpu(dev->pcpu_refcnt);
 free_p:
-- 
cgit v1.2.3


From 27b75c95f10d249574d9c4cb9dab878107faede8 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Fri, 15 Oct 2010 05:44:11 +0000
Subject: net: avoid RCU for NOCACHE dst

There is no point using RCU for dst we allocate for a very short time
(used once).

Change dst_release() to take DST_NOCACHE into account, but also change
skb_dst_set_noref() to force a refcount increment for such dst.

This is a _huge_ gain, because we dont waste memory to store xx thousand
of dsts. Instead of queueing them to RCU, we can free them instantly.

CPU caches can stay hot, re-using same memory blocks to hold temporary
dsts.

Note : remove unneeded smp_mb__before_atomic_dec(); in dst_release(),
since atomic_dec_return() implies a full memory barrier.

Stress test, 160.000.000 udp frames sent, IP route cache disabled
(DDOS).

Before:

real    0m38.091s
user    0m13.189s
sys     7m53.018s

After:

real	0m29.946s
user	0m12.157s
sys	7m40.605s

For reference, if IP route cache was enabled :

real	0m32.030s
user	0m10.521s
sys	8m15.243s

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 14 +-------------
 net/core/dst.c         | 29 ++++++++++++++++++++++++++++-
 net/ipv4/route.c       |  9 ++++-----
 3 files changed, 33 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 05a358f1ba11..e6ba898de61c 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -460,19 +460,7 @@ static inline void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)
 	skb->_skb_refdst = (unsigned long)dst;
 }
 
-/**
- * skb_dst_set_noref - sets skb dst, without a reference
- * @skb: buffer
- * @dst: dst entry
- *
- * Sets skb dst, assuming a reference was not taken on dst
- * skb_dst_drop() should not dst_release() this dst
- */
-static inline void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst)
-{
-	WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held());
-	skb->_skb_refdst = (unsigned long)dst | SKB_DST_NOREF;
-}
+extern void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst);
 
 /**
  * skb_dst_is_noref - Test if skb dst isnt refcounted
diff --git a/net/core/dst.c b/net/core/dst.c
index 32e542d7f472..8abe628b79f1 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -271,13 +271,40 @@ void dst_release(struct dst_entry *dst)
 	if (dst) {
 		int newrefcnt;
 
-		smp_mb__before_atomic_dec();
 		newrefcnt = atomic_dec_return(&dst->__refcnt);
 		WARN_ON(newrefcnt < 0);
+		if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt) {
+			dst = dst_destroy(dst);
+			if (dst)
+				__dst_free(dst);
+		}
 	}
 }
 EXPORT_SYMBOL(dst_release);
 
+/**
+ * skb_dst_set_noref - sets skb dst, without a reference
+ * @skb: buffer
+ * @dst: dst entry
+ *
+ * Sets skb dst, assuming a reference was not taken on dst
+ * skb_dst_drop() should not dst_release() this dst
+ */
+void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst)
+{
+	WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held());
+	/* If dst not in cache, we must take a reference, because
+	 * dst_release() will destroy dst as soon as its refcount becomes zero
+	 */
+	if (unlikely(dst->flags & DST_NOCACHE)) {
+		dst_hold(dst);
+		skb_dst_set(skb, dst);
+	} else {
+		skb->_skb_refdst = (unsigned long)dst | SKB_DST_NOREF;
+	}
+}
+EXPORT_SYMBOL(skb_dst_set_noref);
+
 /* Dirty hack. We did it in 2.2 (in __dst_free),
  * we have _very_ good reasons not to repeat
  * this mistake in 2.3, but we have no choice
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index ff98983d2a45..d6cb2bfcd8e1 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1105,9 +1105,9 @@ restart:
 		 * Note that we do rt_free on this new route entry, so that
 		 * once its refcount hits zero, we are still able to reap it
 		 * (Thanks Alexey)
-		 * Note also the rt_free uses call_rcu.  We don't actually
-		 * need rcu protection here, this is just our path to get
-		 * on the route gc list.
+		 * Note: To avoid expensive rcu stuff for this uncached dst,
+		 * we set DST_NOCACHE so that dst_release() can free dst without
+		 * waiting a grace period.
 		 */
 
 		rt->dst.flags |= DST_NOCACHE;
@@ -1117,12 +1117,11 @@ restart:
 				if (net_ratelimit())
 					printk(KERN_WARNING
 					    "Neighbour table failure & not caching routes.\n");
-				rt_drop(rt);
+				ip_rt_put(rt);
 				return err;
 			}
 		}
 
-		rt_free(rt);
 		goto skip_hashing;
 	}
 
-- 
cgit v1.2.3


From c957ef2c59e952803766ddc22e89981ab534606f Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Wed, 20 Oct 2010 11:07:02 +0800
Subject: percpu: Introduce a read-mostly percpu API

Add a new readmostly percpu section and API.  This can be used to
avoid dirtying data lines which are generally not written to, which is
especially important for data which may be accessed by processors
other than the one for which the percpu area belongs to.

[ hpa: moved it *after* the page-aligned section, for obvious
  reasons. ]

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
LKML-Reference: <1287544022.4571.7.camel@sli10-conroe.sh.intel.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 include/asm-generic/vmlinux.lds.h | 4 ++++
 include/linux/percpu-defs.h       | 9 +++++++++
 2 files changed, 13 insertions(+)

(limited to 'include/linux')

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 8a92a170fb7d..d7e7b21511b1 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -677,7 +677,9 @@
 				- LOAD_OFFSET) {			\
 		VMLINUX_SYMBOL(__per_cpu_start) = .;			\
 		*(.data..percpu..first)					\
+		. = ALIGN(PAGE_SIZE);					\
 		*(.data..percpu..page_aligned)				\
+		*(.data..percpu..readmostly)				\
 		*(.data..percpu)					\
 		*(.data..percpu..shared_aligned)			\
 		VMLINUX_SYMBOL(__per_cpu_end) = .;			\
@@ -703,6 +705,8 @@
 		VMLINUX_SYMBOL(__per_cpu_load) = .;			\
 		VMLINUX_SYMBOL(__per_cpu_start) = .;			\
 		*(.data..percpu..first)					\
+		. = ALIGN(PAGE_SIZE);					\
+		*(.data..percpu..readmostly)				\
 		*(.data..percpu..page_aligned)				\
 		*(.data..percpu)					\
 		*(.data..percpu..shared_aligned)			\
diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h
index ce2dc655cd1d..27ef6b190ea6 100644
--- a/include/linux/percpu-defs.h
+++ b/include/linux/percpu-defs.h
@@ -138,6 +138,15 @@
 	DEFINE_PER_CPU_SECTION(type, name, "..page_aligned")		\
 	__aligned(PAGE_SIZE)
 
+/*
+ * Declaration/definition used for per-CPU variables that must be read mostly.
+ */
+#define DECLARE_PER_CPU_READ_MOSTLY(type, name)			\
+	DECLARE_PER_CPU_SECTION(type, name, "..readmostly")
+
+#define DEFINE_PER_CPU_READ_MOSTLY(type, name)				\
+	DEFINE_PER_CPU_SECTION(type, name, "..readmostly")
+
 /*
  * Intermodule exports for per-CPU variables.  sparse forgets about
  * address space across EXPORT_SYMBOL(), change EXPORT_SYMBOL() to
-- 
cgit v1.2.3


From 3d14c5d2b6e15c21d8e5467dc62d33127c23a644 Mon Sep 17 00:00:00 2001
From: Yehuda Sadeh <yehuda@hq.newdream.net>
Date: Tue, 6 Apr 2010 15:14:15 -0700
Subject: ceph: factor out libceph from Ceph file system

This factors out protocol and low-level storage parts of ceph into a
separate libceph module living in net/ceph and include/linux/ceph.  This
is mostly a matter of moving files around.  However, a few key pieces
of the interface change as well:

 - ceph_client becomes ceph_fs_client and ceph_client, where the latter
   captures the mon and osd clients, and the fs_client gets the mds client
   and file system specific pieces.
 - Mount option parsing and debugfs setup is correspondingly broken into
   two pieces.
 - The mon client gets a generic handler callback for otherwise unknown
   messages (mds map, in this case).
 - The basic supported/required feature bits can be expanded (and are by
   ceph_fs_client).

No functional change, aside from some subtle error handling cases that got
cleaned up in the refactoring process.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 MAINTAINERS                     |    2 +
 fs/ceph/Kconfig                 |   14 +-
 fs/ceph/Makefile                |   11 +-
 fs/ceph/README                  |   20 -
 fs/ceph/addr.c                  |   65 +-
 fs/ceph/armor.c                 |  103 --
 fs/ceph/auth.c                  |  259 -----
 fs/ceph/auth.h                  |   92 --
 fs/ceph/auth_none.c             |  131 ---
 fs/ceph/auth_none.h             |   30 -
 fs/ceph/auth_x.c                |  687 -----------
 fs/ceph/auth_x.h                |   49 -
 fs/ceph/auth_x_protocol.h       |   90 --
 fs/ceph/buffer.c                |   65 --
 fs/ceph/buffer.h                |   39 -
 fs/ceph/caps.c                  |   35 +-
 fs/ceph/ceph_debug.h            |   37 -
 fs/ceph/ceph_frag.c             |    3 +-
 fs/ceph/ceph_frag.h             |  109 --
 fs/ceph/ceph_fs.c               |   72 --
 fs/ceph/ceph_fs.h               |  728 ------------
 fs/ceph/ceph_hash.c             |  118 --
 fs/ceph/ceph_hash.h             |   13 -
 fs/ceph/ceph_strings.c          |  193 ---
 fs/ceph/crush/crush.c           |  151 ---
 fs/ceph/crush/crush.h           |  180 ---
 fs/ceph/crush/hash.c            |  149 ---
 fs/ceph/crush/hash.h            |   17 -
 fs/ceph/crush/mapper.c          |  609 ----------
 fs/ceph/crush/mapper.h          |   20 -
 fs/ceph/crypto.c                |  412 -------
 fs/ceph/crypto.h                |   48 -
 fs/ceph/debugfs.c               |  407 ++-----
 fs/ceph/decode.h                |  201 ----
 fs/ceph/dir.c                   |   55 +-
 fs/ceph/export.c                |    5 +-
 fs/ceph/file.c                  |  207 +---
 fs/ceph/inode.c                 |   19 +-
 fs/ceph/ioctl.c                 |   11 +-
 fs/ceph/locks.c                 |    6 +-
 fs/ceph/mds_client.c            |   86 +-
 fs/ceph/mds_client.h            |   20 +-
 fs/ceph/mdsmap.c                |   11 +-
 fs/ceph/mdsmap.h                |   62 -
 fs/ceph/messenger.c             | 2432 --------------------------------------
 fs/ceph/messenger.h             |  257 ----
 fs/ceph/mon_client.c            | 1018 ----------------
 fs/ceph/mon_client.h            |  121 --
 fs/ceph/msgpool.c               |   64 -
 fs/ceph/msgpool.h               |   25 -
 fs/ceph/msgr.h                  |  175 ---
 fs/ceph/osd_client.c            | 1771 ----------------------------
 fs/ceph/osd_client.h            |  233 ----
 fs/ceph/osdmap.c                | 1123 ------------------
 fs/ceph/osdmap.h                |  130 ---
 fs/ceph/pagelist.c              |   63 -
 fs/ceph/pagelist.h              |   54 -
 fs/ceph/rados.h                 |  405 -------
 fs/ceph/snap.c                  |   10 +-
 fs/ceph/strings.c               |  117 ++
 fs/ceph/super.c                 | 1154 ++++++++----------
 fs/ceph/super.h                 |  397 +++----
 fs/ceph/types.h                 |   29 -
 fs/ceph/xattr.c                 |   15 +-
 include/linux/ceph/auth.h       |   92 ++
 include/linux/ceph/buffer.h     |   39 +
 include/linux/ceph/ceph_debug.h |   38 +
 include/linux/ceph/ceph_frag.h  |  109 ++
 include/linux/ceph/ceph_fs.h    |  728 ++++++++++++
 include/linux/ceph/ceph_hash.h  |   13 +
 include/linux/ceph/debugfs.h    |   33 +
 include/linux/ceph/decode.h     |  201 ++++
 include/linux/ceph/libceph.h    |  249 ++++
 include/linux/ceph/mdsmap.h     |   62 +
 include/linux/ceph/messenger.h  |  261 +++++
 include/linux/ceph/mon_client.h |  122 ++
 include/linux/ceph/msgpool.h    |   25 +
 include/linux/ceph/msgr.h       |  175 +++
 include/linux/ceph/osd_client.h |  234 ++++
 include/linux/ceph/osdmap.h     |  130 +++
 include/linux/ceph/pagelist.h   |   54 +
 include/linux/ceph/rados.h      |  405 +++++++
 include/linux/ceph/types.h      |   29 +
 include/linux/crush/crush.h     |  180 +++
 include/linux/crush/hash.h      |   17 +
 include/linux/crush/mapper.h    |   20 +
 net/Kconfig                     |    1 +
 net/Makefile                    |    1 +
 net/ceph/Kconfig                |   28 +
 net/ceph/Makefile               |   37 +
 net/ceph/armor.c                |  103 ++
 net/ceph/auth.c                 |  259 +++++
 net/ceph/auth_none.c            |  132 +++
 net/ceph/auth_none.h            |   29 +
 net/ceph/auth_x.c               |  688 +++++++++++
 net/ceph/auth_x.h               |   50 +
 net/ceph/auth_x_protocol.h      |   90 ++
 net/ceph/buffer.c               |   68 ++
 net/ceph/ceph_common.c          |  529 +++++++++
 net/ceph/ceph_fs.c              |   75 ++
 net/ceph/ceph_hash.c            |  118 ++
 net/ceph/ceph_strings.c         |   84 ++
 net/ceph/crush/crush.c          |  151 +++
 net/ceph/crush/hash.c           |  149 +++
 net/ceph/crush/mapper.c         |  609 ++++++++++
 net/ceph/crypto.c               |  412 +++++++
 net/ceph/crypto.h               |   48 +
 net/ceph/debugfs.c              |  268 +++++
 net/ceph/messenger.c            | 2453 +++++++++++++++++++++++++++++++++++++++
 net/ceph/mon_client.c           | 1027 ++++++++++++++++
 net/ceph/msgpool.c              |   64 +
 net/ceph/osd_client.c           | 1773 ++++++++++++++++++++++++++++
 net/ceph/osdmap.c               | 1128 ++++++++++++++++++
 net/ceph/pagelist.c             |   65 ++
 net/ceph/pagevec.c              |  223 ++++
 115 files changed, 14920 insertions(+), 14192 deletions(-)
 delete mode 100644 fs/ceph/README
 delete mode 100644 fs/ceph/armor.c
 delete mode 100644 fs/ceph/auth.c
 delete mode 100644 fs/ceph/auth.h
 delete mode 100644 fs/ceph/auth_none.c
 delete mode 100644 fs/ceph/auth_none.h
 delete mode 100644 fs/ceph/auth_x.c
 delete mode 100644 fs/ceph/auth_x.h
 delete mode 100644 fs/ceph/auth_x_protocol.h
 delete mode 100644 fs/ceph/buffer.c
 delete mode 100644 fs/ceph/buffer.h
 delete mode 100644 fs/ceph/ceph_debug.h
 delete mode 100644 fs/ceph/ceph_frag.h
 delete mode 100644 fs/ceph/ceph_fs.c
 delete mode 100644 fs/ceph/ceph_fs.h
 delete mode 100644 fs/ceph/ceph_hash.c
 delete mode 100644 fs/ceph/ceph_hash.h
 delete mode 100644 fs/ceph/ceph_strings.c
 delete mode 100644 fs/ceph/crush/crush.c
 delete mode 100644 fs/ceph/crush/crush.h
 delete mode 100644 fs/ceph/crush/hash.c
 delete mode 100644 fs/ceph/crush/hash.h
 delete mode 100644 fs/ceph/crush/mapper.c
 delete mode 100644 fs/ceph/crush/mapper.h
 delete mode 100644 fs/ceph/crypto.c
 delete mode 100644 fs/ceph/crypto.h
 delete mode 100644 fs/ceph/decode.h
 delete mode 100644 fs/ceph/mdsmap.h
 delete mode 100644 fs/ceph/messenger.c
 delete mode 100644 fs/ceph/messenger.h
 delete mode 100644 fs/ceph/mon_client.c
 delete mode 100644 fs/ceph/mon_client.h
 delete mode 100644 fs/ceph/msgpool.c
 delete mode 100644 fs/ceph/msgpool.h
 delete mode 100644 fs/ceph/msgr.h
 delete mode 100644 fs/ceph/osd_client.c
 delete mode 100644 fs/ceph/osd_client.h
 delete mode 100644 fs/ceph/osdmap.c
 delete mode 100644 fs/ceph/osdmap.h
 delete mode 100644 fs/ceph/pagelist.c
 delete mode 100644 fs/ceph/pagelist.h
 delete mode 100644 fs/ceph/rados.h
 create mode 100644 fs/ceph/strings.c
 delete mode 100644 fs/ceph/types.h
 create mode 100644 include/linux/ceph/auth.h
 create mode 100644 include/linux/ceph/buffer.h
 create mode 100644 include/linux/ceph/ceph_debug.h
 create mode 100644 include/linux/ceph/ceph_frag.h
 create mode 100644 include/linux/ceph/ceph_fs.h
 create mode 100644 include/linux/ceph/ceph_hash.h
 create mode 100644 include/linux/ceph/debugfs.h
 create mode 100644 include/linux/ceph/decode.h
 create mode 100644 include/linux/ceph/libceph.h
 create mode 100644 include/linux/ceph/mdsmap.h
 create mode 100644 include/linux/ceph/messenger.h
 create mode 100644 include/linux/ceph/mon_client.h
 create mode 100644 include/linux/ceph/msgpool.h
 create mode 100644 include/linux/ceph/msgr.h
 create mode 100644 include/linux/ceph/osd_client.h
 create mode 100644 include/linux/ceph/osdmap.h
 create mode 100644 include/linux/ceph/pagelist.h
 create mode 100644 include/linux/ceph/rados.h
 create mode 100644 include/linux/ceph/types.h
 create mode 100644 include/linux/crush/crush.h
 create mode 100644 include/linux/crush/hash.h
 create mode 100644 include/linux/crush/mapper.h
 create mode 100644 net/ceph/Kconfig
 create mode 100644 net/ceph/Makefile
 create mode 100644 net/ceph/armor.c
 create mode 100644 net/ceph/auth.c
 create mode 100644 net/ceph/auth_none.c
 create mode 100644 net/ceph/auth_none.h
 create mode 100644 net/ceph/auth_x.c
 create mode 100644 net/ceph/auth_x.h
 create mode 100644 net/ceph/auth_x_protocol.h
 create mode 100644 net/ceph/buffer.c
 create mode 100644 net/ceph/ceph_common.c
 create mode 100644 net/ceph/ceph_fs.c
 create mode 100644 net/ceph/ceph_hash.c
 create mode 100644 net/ceph/ceph_strings.c
 create mode 100644 net/ceph/crush/crush.c
 create mode 100644 net/ceph/crush/hash.c
 create mode 100644 net/ceph/crush/mapper.c
 create mode 100644 net/ceph/crypto.c
 create mode 100644 net/ceph/crypto.h
 create mode 100644 net/ceph/debugfs.c
 create mode 100644 net/ceph/messenger.c
 create mode 100644 net/ceph/mon_client.c
 create mode 100644 net/ceph/msgpool.c
 create mode 100644 net/ceph/osd_client.c
 create mode 100644 net/ceph/osdmap.c
 create mode 100644 net/ceph/pagelist.c
 create mode 100644 net/ceph/pagevec.c

(limited to 'include/linux')

diff --git a/MAINTAINERS b/MAINTAINERS
index 7679bf32f7bb..48d05654d456 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1527,6 +1527,8 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client.git
 S:	Supported
 F:	Documentation/filesystems/ceph.txt
 F:	fs/ceph
+F:	net/ceph
+F:	include/linux/ceph
 
 CERTIFIED WIRELESS USB (WUSB) SUBSYSTEM:
 M:	David Vrabel <david.vrabel@csr.com>
diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig
index 0fcd2640c23f..9eb134ea6eb2 100644
--- a/fs/ceph/Kconfig
+++ b/fs/ceph/Kconfig
@@ -1,9 +1,11 @@
 config CEPH_FS
         tristate "Ceph distributed file system (EXPERIMENTAL)"
 	depends on INET && EXPERIMENTAL
+	select CEPH_LIB
 	select LIBCRC32C
 	select CRYPTO_AES
 	select CRYPTO
+	default n
 	help
 	  Choose Y or M here to include support for mounting the
 	  experimental Ceph distributed file system.  Ceph is an extremely
@@ -14,15 +16,3 @@ config CEPH_FS
 
 	  If unsure, say N.
 
-config CEPH_FS_PRETTYDEBUG
-	bool "Include file:line in ceph debug output"
-	depends on CEPH_FS
-	default n
-	help
-	  If you say Y here, debug output will include a filename and
-	  line to aid debugging.  This icnreases kernel size and slows
-	  execution slightly when debug call sites are enabled (e.g.,
-	  via CONFIG_DYNAMIC_DEBUG).
-
-	  If unsure, say N.
-
diff --git a/fs/ceph/Makefile b/fs/ceph/Makefile
index 278e1172600d..9e6c4f2e8ff1 100644
--- a/fs/ceph/Makefile
+++ b/fs/ceph/Makefile
@@ -8,15 +8,8 @@ obj-$(CONFIG_CEPH_FS) += ceph.o
 
 ceph-objs := super.o inode.o dir.o file.o locks.o addr.o ioctl.o \
 	export.o caps.o snap.o xattr.o \
-	messenger.o msgpool.o buffer.o pagelist.o \
-	mds_client.o mdsmap.o \
-	mon_client.o \
-	osd_client.o osdmap.o crush/crush.o crush/mapper.o crush/hash.o \
-	debugfs.o \
-	auth.o auth_none.o \
-	crypto.o armor.o \
-	auth_x.o \
-	ceph_fs.o ceph_strings.o ceph_hash.o ceph_frag.o
+	mds_client.o mdsmap.o strings.o ceph_frag.o \
+	debugfs.o
 
 else
 #Otherwise we were called directly from the command
diff --git a/fs/ceph/README b/fs/ceph/README
deleted file mode 100644
index 18352fab37c0..000000000000
--- a/fs/ceph/README
+++ /dev/null
@@ -1,20 +0,0 @@
-#
-# The following files are shared by (and manually synchronized
-# between) the Ceph userland and kernel client.
-#
-# userland                  kernel
-src/include/ceph_fs.h	    fs/ceph/ceph_fs.h
-src/include/ceph_fs.cc	    fs/ceph/ceph_fs.c
-src/include/msgr.h	    fs/ceph/msgr.h
-src/include/rados.h	    fs/ceph/rados.h
-src/include/ceph_strings.cc fs/ceph/ceph_strings.c
-src/include/ceph_frag.h	    fs/ceph/ceph_frag.h
-src/include/ceph_frag.cc    fs/ceph/ceph_frag.c
-src/include/ceph_hash.h	    fs/ceph/ceph_hash.h
-src/include/ceph_hash.cc    fs/ceph/ceph_hash.c
-src/crush/crush.c	    fs/ceph/crush/crush.c
-src/crush/crush.h	    fs/ceph/crush/crush.h
-src/crush/mapper.c	    fs/ceph/crush/mapper.c
-src/crush/mapper.h	    fs/ceph/crush/mapper.h
-src/crush/hash.h	    fs/ceph/crush/hash.h
-src/crush/hash.c	    fs/ceph/crush/hash.c
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index efbc604001c8..51bcc5ce3230 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/backing-dev.h>
 #include <linux/fs.h>
@@ -10,7 +10,8 @@
 #include <linux/task_io_accounting_ops.h>
 
 #include "super.h"
-#include "osd_client.h"
+#include "mds_client.h"
+#include <linux/ceph/osd_client.h>
 
 /*
  * Ceph address space ops.
@@ -193,7 +194,8 @@ static int readpage_nounlock(struct file *filp, struct page *page)
 {
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_inode_to_client(inode)->osdc;
+	struct ceph_osd_client *osdc = 
+		&ceph_inode_to_client(inode)->client->osdc;
 	int err = 0;
 	u64 len = PAGE_CACHE_SIZE;
 
@@ -265,7 +267,8 @@ static int ceph_readpages(struct file *file, struct address_space *mapping,
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_inode_to_client(inode)->osdc;
+	struct ceph_osd_client *osdc =
+		&ceph_inode_to_client(inode)->client->osdc;
 	int rc = 0;
 	struct page **pages;
 	loff_t offset;
@@ -365,7 +368,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 {
 	struct inode *inode;
 	struct ceph_inode_info *ci;
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc;
 	struct ceph_osd_client *osdc;
 	loff_t page_off = page->index << PAGE_CACHE_SHIFT;
 	int len = PAGE_CACHE_SIZE;
@@ -383,8 +386,8 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	}
 	inode = page->mapping->host;
 	ci = ceph_inode(inode);
-	client = ceph_inode_to_client(inode);
-	osdc = &client->osdc;
+	fsc = ceph_inode_to_client(inode);
+	osdc = &fsc->client->osdc;
 
 	/* verify this is a writeable snap context */
 	snapc = (void *)page->private;
@@ -414,10 +417,10 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	dout("writepage %p page %p index %lu on %llu~%u snapc %p\n",
 	     inode, page, page->index, page_off, len, snapc);
 
-	writeback_stat = atomic_long_inc_return(&client->writeback_count);
+	writeback_stat = atomic_long_inc_return(&fsc->writeback_count);
 	if (writeback_stat >
-	    CONGESTION_ON_THRESH(client->mount_args->congestion_kb))
-		set_bdi_congested(&client->backing_dev_info, BLK_RW_ASYNC);
+	    CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb))
+		set_bdi_congested(&fsc->backing_dev_info, BLK_RW_ASYNC);
 
 	set_page_writeback(page);
 	err = ceph_osdc_writepages(osdc, ceph_vino(inode),
@@ -496,7 +499,7 @@ static void writepages_finish(struct ceph_osd_request *req,
 	struct address_space *mapping = inode->i_mapping;
 	__s32 rc = -EIO;
 	u64 bytes = 0;
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	long writeback_stat;
 	unsigned issued = ceph_caps_issued(ci);
 
@@ -529,10 +532,10 @@ static void writepages_finish(struct ceph_osd_request *req,
 		WARN_ON(!PageUptodate(page));
 
 		writeback_stat =
-			atomic_long_dec_return(&client->writeback_count);
+			atomic_long_dec_return(&fsc->writeback_count);
 		if (writeback_stat <
-		    CONGESTION_OFF_THRESH(client->mount_args->congestion_kb))
-			clear_bdi_congested(&client->backing_dev_info,
+		    CONGESTION_OFF_THRESH(fsc->mount_options->congestion_kb))
+			clear_bdi_congested(&fsc->backing_dev_info,
 					    BLK_RW_ASYNC);
 
 		ceph_put_snap_context((void *)page->private);
@@ -569,13 +572,13 @@ static void writepages_finish(struct ceph_osd_request *req,
  * mempool.  we avoid the mempool if we can because req->r_num_pages
  * may be less than the maximum write size.
  */
-static void alloc_page_vec(struct ceph_client *client,
+static void alloc_page_vec(struct ceph_fs_client *fsc,
 			   struct ceph_osd_request *req)
 {
 	req->r_pages = kmalloc(sizeof(struct page *) * req->r_num_pages,
 			       GFP_NOFS);
 	if (!req->r_pages) {
-		req->r_pages = mempool_alloc(client->wb_pagevec_pool, GFP_NOFS);
+		req->r_pages = mempool_alloc(fsc->wb_pagevec_pool, GFP_NOFS);
 		req->r_pages_from_pool = 1;
 		WARN_ON(!req->r_pages);
 	}
@@ -590,7 +593,7 @@ static int ceph_writepages_start(struct address_space *mapping,
 	struct inode *inode = mapping->host;
 	struct backing_dev_info *bdi = mapping->backing_dev_info;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc;
 	pgoff_t index, start, end;
 	int range_whole = 0;
 	int should_loop = 1;
@@ -617,13 +620,13 @@ static int ceph_writepages_start(struct address_space *mapping,
 	     wbc->sync_mode == WB_SYNC_NONE ? "NONE" :
 	     (wbc->sync_mode == WB_SYNC_ALL ? "ALL" : "HOLD"));
 
-	client = ceph_inode_to_client(inode);
-	if (client->mount_state == CEPH_MOUNT_SHUTDOWN) {
+	fsc = ceph_inode_to_client(inode);
+	if (fsc->mount_state == CEPH_MOUNT_SHUTDOWN) {
 		pr_warning("writepage_start %p on forced umount\n", inode);
 		return -EIO; /* we're in a forced umount, don't write! */
 	}
-	if (client->mount_args->wsize && client->mount_args->wsize < wsize)
-		wsize = client->mount_args->wsize;
+	if (fsc->mount_options->wsize && fsc->mount_options->wsize < wsize)
+		wsize = fsc->mount_options->wsize;
 	if (wsize < PAGE_CACHE_SIZE)
 		wsize = PAGE_CACHE_SIZE;
 	max_pages_ever = wsize >> PAGE_CACHE_SHIFT;
@@ -769,7 +772,7 @@ get_more_pages:
 				offset = (unsigned long long)page->index
 					<< PAGE_CACHE_SHIFT;
 				len = wsize;
-				req = ceph_osdc_new_request(&client->osdc,
+				req = ceph_osdc_new_request(&fsc->client->osdc,
 					    &ci->i_layout,
 					    ceph_vino(inode),
 					    offset, &len,
@@ -782,7 +785,7 @@ get_more_pages:
 					    &inode->i_mtime, true, 1);
 				max_pages = req->r_num_pages;
 
-				alloc_page_vec(client, req);
+				alloc_page_vec(fsc, req);
 				req->r_callback = writepages_finish;
 				req->r_inode = inode;
 			}
@@ -794,10 +797,10 @@ get_more_pages:
 			     inode, page, page->index);
 
 			writeback_stat =
-			       atomic_long_inc_return(&client->writeback_count);
+			       atomic_long_inc_return(&fsc->writeback_count);
 			if (writeback_stat > CONGESTION_ON_THRESH(
-				    client->mount_args->congestion_kb)) {
-				set_bdi_congested(&client->backing_dev_info,
+				    fsc->mount_options->congestion_kb)) {
+				set_bdi_congested(&fsc->backing_dev_info,
 						  BLK_RW_ASYNC);
 			}
 
@@ -846,7 +849,7 @@ get_more_pages:
 		op->payload_len = cpu_to_le32(len);
 		req->r_request->hdr.data_len = cpu_to_le32(len);
 
-		ceph_osdc_start_request(&client->osdc, req, true);
+		ceph_osdc_start_request(&fsc->client->osdc, req, true);
 		req = NULL;
 
 		/* continue? */
@@ -915,7 +918,7 @@ static int ceph_update_writeable_page(struct file *file,
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	loff_t page_off = pos & PAGE_CACHE_MASK;
 	int pos_in_page = pos & ~PAGE_CACHE_MASK;
 	int end_in_page = pos_in_page + len;
@@ -1053,8 +1056,8 @@ static int ceph_write_end(struct file *file, struct address_space *mapping,
 			  struct page *page, void *fsdata)
 {
 	struct inode *inode = file->f_dentry->d_inode;
-	struct ceph_client *client = ceph_inode_to_client(inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	unsigned from = pos & (PAGE_CACHE_SIZE - 1);
 	int check_cap = 0;
 
@@ -1123,7 +1126,7 @@ static int ceph_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct inode *inode = vma->vm_file->f_dentry->d_inode;
 	struct page *page = vmf->page;
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	loff_t off = page->index << PAGE_CACHE_SHIFT;
 	loff_t size, len;
 	int ret;
diff --git a/fs/ceph/armor.c b/fs/ceph/armor.c
deleted file mode 100644
index eb2a666b0be7..000000000000
--- a/fs/ceph/armor.c
+++ /dev/null
@@ -1,103 +0,0 @@
-
-#include <linux/errno.h>
-
-int ceph_armor(char *dst, const char *src, const char *end);
-int ceph_unarmor(char *dst, const char *src, const char *end);
-
-/*
- * base64 encode/decode.
- */
-
-static const char *pem_key =
-	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
-
-static int encode_bits(int c)
-{
-	return pem_key[c];
-}
-
-static int decode_bits(char c)
-{
-	if (c >= 'A' && c <= 'Z')
-		return c - 'A';
-	if (c >= 'a' && c <= 'z')
-		return c - 'a' + 26;
-	if (c >= '0' && c <= '9')
-		return c - '0' + 52;
-	if (c == '+')
-		return 62;
-	if (c == '/')
-		return 63;
-	if (c == '=')
-		return 0; /* just non-negative, please */
-	return -EINVAL;
-}
-
-int ceph_armor(char *dst, const char *src, const char *end)
-{
-	int olen = 0;
-	int line = 0;
-
-	while (src < end) {
-		unsigned char a, b, c;
-
-		a = *src++;
-		*dst++ = encode_bits(a >> 2);
-		if (src < end) {
-			b = *src++;
-			*dst++ = encode_bits(((a & 3) << 4) | (b >> 4));
-			if (src < end) {
-				c = *src++;
-				*dst++ = encode_bits(((b & 15) << 2) |
-						     (c >> 6));
-				*dst++ = encode_bits(c & 63);
-			} else {
-				*dst++ = encode_bits((b & 15) << 2);
-				*dst++ = '=';
-			}
-		} else {
-			*dst++ = encode_bits(((a & 3) << 4));
-			*dst++ = '=';
-			*dst++ = '=';
-		}
-		olen += 4;
-		line += 4;
-		if (line == 64) {
-			line = 0;
-			*(dst++) = '\n';
-			olen++;
-		}
-	}
-	return olen;
-}
-
-int ceph_unarmor(char *dst, const char *src, const char *end)
-{
-	int olen = 0;
-
-	while (src < end) {
-		int a, b, c, d;
-
-		if (src < end && src[0] == '\n')
-			src++;
-		if (src + 4 > end)
-			return -EINVAL;
-		a = decode_bits(src[0]);
-		b = decode_bits(src[1]);
-		c = decode_bits(src[2]);
-		d = decode_bits(src[3]);
-		if (a < 0 || b < 0 || c < 0 || d < 0)
-			return -EINVAL;
-
-		*dst++ = (a << 2) | (b >> 4);
-		if (src[2] == '=')
-			return olen + 1;
-		*dst++ = ((b & 15) << 4) | (c >> 2);
-		if (src[3] == '=')
-			return olen + 2;
-		*dst++ = ((c & 3) << 6) | d;
-		olen += 3;
-		src += 4;
-	}
-	return olen;
-}
diff --git a/fs/ceph/auth.c b/fs/ceph/auth.c
deleted file mode 100644
index 6d2e30600627..000000000000
--- a/fs/ceph/auth.c
+++ /dev/null
@@ -1,259 +0,0 @@
-#include "ceph_debug.h"
-
-#include <linux/module.h>
-#include <linux/err.h>
-#include <linux/slab.h>
-
-#include "types.h"
-#include "auth_none.h"
-#include "auth_x.h"
-#include "decode.h"
-#include "super.h"
-
-#include "messenger.h"
-
-/*
- * get protocol handler
- */
-static u32 supported_protocols[] = {
-	CEPH_AUTH_NONE,
-	CEPH_AUTH_CEPHX
-};
-
-static int ceph_auth_init_protocol(struct ceph_auth_client *ac, int protocol)
-{
-	switch (protocol) {
-	case CEPH_AUTH_NONE:
-		return ceph_auth_none_init(ac);
-	case CEPH_AUTH_CEPHX:
-		return ceph_x_init(ac);
-	default:
-		return -ENOENT;
-	}
-}
-
-/*
- * setup, teardown.
- */
-struct ceph_auth_client *ceph_auth_init(const char *name, const char *secret)
-{
-	struct ceph_auth_client *ac;
-	int ret;
-
-	dout("auth_init name '%s' secret '%s'\n", name, secret);
-
-	ret = -ENOMEM;
-	ac = kzalloc(sizeof(*ac), GFP_NOFS);
-	if (!ac)
-		goto out;
-
-	ac->negotiating = true;
-	if (name)
-		ac->name = name;
-	else
-		ac->name = CEPH_AUTH_NAME_DEFAULT;
-	dout("auth_init name %s secret %s\n", ac->name, secret);
-	ac->secret = secret;
-	return ac;
-
-out:
-	return ERR_PTR(ret);
-}
-
-void ceph_auth_destroy(struct ceph_auth_client *ac)
-{
-	dout("auth_destroy %p\n", ac);
-	if (ac->ops)
-		ac->ops->destroy(ac);
-	kfree(ac);
-}
-
-/*
- * Reset occurs when reconnecting to the monitor.
- */
-void ceph_auth_reset(struct ceph_auth_client *ac)
-{
-	dout("auth_reset %p\n", ac);
-	if (ac->ops && !ac->negotiating)
-		ac->ops->reset(ac);
-	ac->negotiating = true;
-}
-
-int ceph_entity_name_encode(const char *name, void **p, void *end)
-{
-	int len = strlen(name);
-
-	if (*p + 2*sizeof(u32) + len > end)
-		return -ERANGE;
-	ceph_encode_32(p, CEPH_ENTITY_TYPE_CLIENT);
-	ceph_encode_32(p, len);
-	ceph_encode_copy(p, name, len);
-	return 0;
-}
-
-/*
- * Initiate protocol negotiation with monitor.  Include entity name
- * and list supported protocols.
- */
-int ceph_auth_build_hello(struct ceph_auth_client *ac, void *buf, size_t len)
-{
-	struct ceph_mon_request_header *monhdr = buf;
-	void *p = monhdr + 1, *end = buf + len, *lenp;
-	int i, num;
-	int ret;
-
-	dout("auth_build_hello\n");
-	monhdr->have_version = 0;
-	monhdr->session_mon = cpu_to_le16(-1);
-	monhdr->session_mon_tid = 0;
-
-	ceph_encode_32(&p, 0);  /* no protocol, yet */
-
-	lenp = p;
-	p += sizeof(u32);
-
-	ceph_decode_need(&p, end, 1 + sizeof(u32), bad);
-	ceph_encode_8(&p, 1);
-	num = ARRAY_SIZE(supported_protocols);
-	ceph_encode_32(&p, num);
-	ceph_decode_need(&p, end, num * sizeof(u32), bad);
-	for (i = 0; i < num; i++)
-		ceph_encode_32(&p, supported_protocols[i]);
-
-	ret = ceph_entity_name_encode(ac->name, &p, end);
-	if (ret < 0)
-		return ret;
-	ceph_decode_need(&p, end, sizeof(u64), bad);
-	ceph_encode_64(&p, ac->global_id);
-
-	ceph_encode_32(&lenp, p - lenp - sizeof(u32));
-	return p - buf;
-
-bad:
-	return -ERANGE;
-}
-
-static int ceph_build_auth_request(struct ceph_auth_client *ac,
-				   void *msg_buf, size_t msg_len)
-{
-	struct ceph_mon_request_header *monhdr = msg_buf;
-	void *p = monhdr + 1;
-	void *end = msg_buf + msg_len;
-	int ret;
-
-	monhdr->have_version = 0;
-	monhdr->session_mon = cpu_to_le16(-1);
-	monhdr->session_mon_tid = 0;
-
-	ceph_encode_32(&p, ac->protocol);
-
-	ret = ac->ops->build_request(ac, p + sizeof(u32), end);
-	if (ret < 0) {
-		pr_err("error %d building auth method %s request\n", ret,
-		       ac->ops->name);
-		return ret;
-	}
-	dout(" built request %d bytes\n", ret);
-	ceph_encode_32(&p, ret);
-	return p + ret - msg_buf;
-}
-
-/*
- * Handle auth message from monitor.
- */
-int ceph_handle_auth_reply(struct ceph_auth_client *ac,
-			   void *buf, size_t len,
-			   void *reply_buf, size_t reply_len)
-{
-	void *p = buf;
-	void *end = buf + len;
-	int protocol;
-	s32 result;
-	u64 global_id;
-	void *payload, *payload_end;
-	int payload_len;
-	char *result_msg;
-	int result_msg_len;
-	int ret = -EINVAL;
-
-	dout("handle_auth_reply %p %p\n", p, end);
-	ceph_decode_need(&p, end, sizeof(u32) * 3 + sizeof(u64), bad);
-	protocol = ceph_decode_32(&p);
-	result = ceph_decode_32(&p);
-	global_id = ceph_decode_64(&p);
-	payload_len = ceph_decode_32(&p);
-	payload = p;
-	p += payload_len;
-	ceph_decode_need(&p, end, sizeof(u32), bad);
-	result_msg_len = ceph_decode_32(&p);
-	result_msg = p;
-	p += result_msg_len;
-	if (p != end)
-		goto bad;
-
-	dout(" result %d '%.*s' gid %llu len %d\n", result, result_msg_len,
-	     result_msg, global_id, payload_len);
-
-	payload_end = payload + payload_len;
-
-	if (global_id && ac->global_id != global_id) {
-		dout(" set global_id %lld -> %lld\n", ac->global_id, global_id);
-		ac->global_id = global_id;
-	}
-
-	if (ac->negotiating) {
-		/* server does not support our protocols? */
-		if (!protocol && result < 0) {
-			ret = result;
-			goto out;
-		}
-		/* set up (new) protocol handler? */
-		if (ac->protocol && ac->protocol != protocol) {
-			ac->ops->destroy(ac);
-			ac->protocol = 0;
-			ac->ops = NULL;
-		}
-		if (ac->protocol != protocol) {
-			ret = ceph_auth_init_protocol(ac, protocol);
-			if (ret) {
-				pr_err("error %d on auth protocol %d init\n",
-				       ret, protocol);
-				goto out;
-			}
-		}
-
-		ac->negotiating = false;
-	}
-
-	ret = ac->ops->handle_reply(ac, result, payload, payload_end);
-	if (ret == -EAGAIN) {
-		return ceph_build_auth_request(ac, reply_buf, reply_len);
-	} else if (ret) {
-		pr_err("auth method '%s' error %d\n", ac->ops->name, ret);
-		return ret;
-	}
-	return 0;
-
-bad:
-	pr_err("failed to decode auth msg\n");
-out:
-	return ret;
-}
-
-int ceph_build_auth(struct ceph_auth_client *ac,
-		    void *msg_buf, size_t msg_len)
-{
-	if (!ac->protocol)
-		return ceph_auth_build_hello(ac, msg_buf, msg_len);
-	BUG_ON(!ac->ops);
-	if (ac->ops->should_authenticate(ac))
-		return ceph_build_auth_request(ac, msg_buf, msg_len);
-	return 0;
-}
-
-int ceph_auth_is_authenticated(struct ceph_auth_client *ac)
-{
-	if (!ac->ops)
-		return 0;
-	return ac->ops->is_authenticated(ac);
-}
diff --git a/fs/ceph/auth.h b/fs/ceph/auth.h
deleted file mode 100644
index d38a2fb4a137..000000000000
--- a/fs/ceph/auth.h
+++ /dev/null
@@ -1,92 +0,0 @@
-#ifndef _FS_CEPH_AUTH_H
-#define _FS_CEPH_AUTH_H
-
-#include "types.h"
-#include "buffer.h"
-
-/*
- * Abstract interface for communicating with the authenticate module.
- * There is some handshake that takes place between us and the monitor
- * to acquire the necessary keys.  These are used to generate an
- * 'authorizer' that we use when connecting to a service (mds, osd).
- */
-
-struct ceph_auth_client;
-struct ceph_authorizer;
-
-struct ceph_auth_client_ops {
-	const char *name;
-
-	/*
-	 * true if we are authenticated and can connect to
-	 * services.
-	 */
-	int (*is_authenticated)(struct ceph_auth_client *ac);
-
-	/*
-	 * true if we should (re)authenticate, e.g., when our tickets
-	 * are getting old and crusty.
-	 */
-	int (*should_authenticate)(struct ceph_auth_client *ac);
-
-	/*
-	 * build requests and process replies during monitor
-	 * handshake.  if handle_reply returns -EAGAIN, we build
-	 * another request.
-	 */
-	int (*build_request)(struct ceph_auth_client *ac, void *buf, void *end);
-	int (*handle_reply)(struct ceph_auth_client *ac, int result,
-			    void *buf, void *end);
-
-	/*
-	 * Create authorizer for connecting to a service, and verify
-	 * the response to authenticate the service.
-	 */
-	int (*create_authorizer)(struct ceph_auth_client *ac, int peer_type,
-				 struct ceph_authorizer **a,
-				 void **buf, size_t *len,
-				 void **reply_buf, size_t *reply_len);
-	int (*verify_authorizer_reply)(struct ceph_auth_client *ac,
-				       struct ceph_authorizer *a, size_t len);
-	void (*destroy_authorizer)(struct ceph_auth_client *ac,
-				   struct ceph_authorizer *a);
-	void (*invalidate_authorizer)(struct ceph_auth_client *ac,
-				      int peer_type);
-
-	/* reset when we (re)connect to a monitor */
-	void (*reset)(struct ceph_auth_client *ac);
-
-	void (*destroy)(struct ceph_auth_client *ac);
-};
-
-struct ceph_auth_client {
-	u32 protocol;           /* CEPH_AUTH_* */
-	void *private;          /* for use by protocol implementation */
-	const struct ceph_auth_client_ops *ops;  /* null iff protocol==0 */
-
-	bool negotiating;       /* true if negotiating protocol */
-	const char *name;       /* entity name */
-	u64 global_id;          /* our unique id in system */
-	const char *secret;     /* our secret key */
-	unsigned want_keys;     /* which services we want */
-};
-
-extern struct ceph_auth_client *ceph_auth_init(const char *name,
-					       const char *secret);
-extern void ceph_auth_destroy(struct ceph_auth_client *ac);
-
-extern void ceph_auth_reset(struct ceph_auth_client *ac);
-
-extern int ceph_auth_build_hello(struct ceph_auth_client *ac,
-				 void *buf, size_t len);
-extern int ceph_handle_auth_reply(struct ceph_auth_client *ac,
-				  void *buf, size_t len,
-				  void *reply_buf, size_t reply_len);
-extern int ceph_entity_name_encode(const char *name, void **p, void *end);
-
-extern int ceph_build_auth(struct ceph_auth_client *ac,
-		    void *msg_buf, size_t msg_len);
-
-extern int ceph_auth_is_authenticated(struct ceph_auth_client *ac);
-
-#endif
diff --git a/fs/ceph/auth_none.c b/fs/ceph/auth_none.c
deleted file mode 100644
index ad1dc21286c7..000000000000
--- a/fs/ceph/auth_none.c
+++ /dev/null
@@ -1,131 +0,0 @@
-
-#include "ceph_debug.h"
-
-#include <linux/err.h>
-#include <linux/module.h>
-#include <linux/random.h>
-#include <linux/slab.h>
-
-#include "auth_none.h"
-#include "auth.h"
-#include "decode.h"
-
-static void reset(struct ceph_auth_client *ac)
-{
-	struct ceph_auth_none_info *xi = ac->private;
-
-	xi->starting = true;
-	xi->built_authorizer = false;
-}
-
-static void destroy(struct ceph_auth_client *ac)
-{
-	kfree(ac->private);
-	ac->private = NULL;
-}
-
-static int is_authenticated(struct ceph_auth_client *ac)
-{
-	struct ceph_auth_none_info *xi = ac->private;
-
-	return !xi->starting;
-}
-
-static int should_authenticate(struct ceph_auth_client *ac)
-{
-	struct ceph_auth_none_info *xi = ac->private;
-
-	return xi->starting;
-}
-
-/*
- * the generic auth code decode the global_id, and we carry no actual
- * authenticate state, so nothing happens here.
- */
-static int handle_reply(struct ceph_auth_client *ac, int result,
-			void *buf, void *end)
-{
-	struct ceph_auth_none_info *xi = ac->private;
-
-	xi->starting = false;
-	return result;
-}
-
-/*
- * build an 'authorizer' with our entity_name and global_id.  we can
- * reuse a single static copy since it is identical for all services
- * we connect to.
- */
-static int ceph_auth_none_create_authorizer(
-	struct ceph_auth_client *ac, int peer_type,
-	struct ceph_authorizer **a,
-	void **buf, size_t *len,
-	void **reply_buf, size_t *reply_len)
-{
-	struct ceph_auth_none_info *ai = ac->private;
-	struct ceph_none_authorizer *au = &ai->au;
-	void *p, *end;
-	int ret;
-
-	if (!ai->built_authorizer) {
-		p = au->buf;
-		end = p + sizeof(au->buf);
-		ceph_encode_8(&p, 1);
-		ret = ceph_entity_name_encode(ac->name, &p, end - 8);
-		if (ret < 0)
-			goto bad;
-		ceph_decode_need(&p, end, sizeof(u64), bad2);
-		ceph_encode_64(&p, ac->global_id);
-		au->buf_len = p - (void *)au->buf;
-		ai->built_authorizer = true;
-		dout("built authorizer len %d\n", au->buf_len);
-	}
-
-	*a = (struct ceph_authorizer *)au;
-	*buf = au->buf;
-	*len = au->buf_len;
-	*reply_buf = au->reply_buf;
-	*reply_len = sizeof(au->reply_buf);
-	return 0;
-
-bad2:
-	ret = -ERANGE;
-bad:
-	return ret;
-}
-
-static void ceph_auth_none_destroy_authorizer(struct ceph_auth_client *ac,
-				      struct ceph_authorizer *a)
-{
-	/* nothing to do */
-}
-
-static const struct ceph_auth_client_ops ceph_auth_none_ops = {
-	.name = "none",
-	.reset = reset,
-	.destroy = destroy,
-	.is_authenticated = is_authenticated,
-	.should_authenticate = should_authenticate,
-	.handle_reply = handle_reply,
-	.create_authorizer = ceph_auth_none_create_authorizer,
-	.destroy_authorizer = ceph_auth_none_destroy_authorizer,
-};
-
-int ceph_auth_none_init(struct ceph_auth_client *ac)
-{
-	struct ceph_auth_none_info *xi;
-
-	dout("ceph_auth_none_init %p\n", ac);
-	xi = kzalloc(sizeof(*xi), GFP_NOFS);
-	if (!xi)
-		return -ENOMEM;
-
-	xi->starting = true;
-	xi->built_authorizer = false;
-
-	ac->protocol = CEPH_AUTH_NONE;
-	ac->private = xi;
-	ac->ops = &ceph_auth_none_ops;
-	return 0;
-}
-
diff --git a/fs/ceph/auth_none.h b/fs/ceph/auth_none.h
deleted file mode 100644
index 8164df1a08be..000000000000
--- a/fs/ceph/auth_none.h
+++ /dev/null
@@ -1,30 +0,0 @@
-#ifndef _FS_CEPH_AUTH_NONE_H
-#define _FS_CEPH_AUTH_NONE_H
-
-#include <linux/slab.h>
-
-#include "auth.h"
-
-/*
- * null security mode.
- *
- * we use a single static authorizer that simply encodes our entity name
- * and global id.
- */
-
-struct ceph_none_authorizer {
-	char buf[128];
-	int buf_len;
-	char reply_buf[0];
-};
-
-struct ceph_auth_none_info {
-	bool starting;
-	bool built_authorizer;
-	struct ceph_none_authorizer au;   /* we only need one; it's static */
-};
-
-extern int ceph_auth_none_init(struct ceph_auth_client *ac);
-
-#endif
-
diff --git a/fs/ceph/auth_x.c b/fs/ceph/auth_x.c
deleted file mode 100644
index a2d002cbdec2..000000000000
--- a/fs/ceph/auth_x.c
+++ /dev/null
@@ -1,687 +0,0 @@
-
-#include "ceph_debug.h"
-
-#include <linux/err.h>
-#include <linux/module.h>
-#include <linux/random.h>
-#include <linux/slab.h>
-
-#include "auth_x.h"
-#include "auth_x_protocol.h"
-#include "crypto.h"
-#include "auth.h"
-#include "decode.h"
-
-#define TEMP_TICKET_BUF_LEN	256
-
-static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed);
-
-static int ceph_x_is_authenticated(struct ceph_auth_client *ac)
-{
-	struct ceph_x_info *xi = ac->private;
-	int need;
-
-	ceph_x_validate_tickets(ac, &need);
-	dout("ceph_x_is_authenticated want=%d need=%d have=%d\n",
-	     ac->want_keys, need, xi->have_keys);
-	return (ac->want_keys & xi->have_keys) == ac->want_keys;
-}
-
-static int ceph_x_should_authenticate(struct ceph_auth_client *ac)
-{
-	struct ceph_x_info *xi = ac->private;
-	int need;
-
-	ceph_x_validate_tickets(ac, &need);
-	dout("ceph_x_should_authenticate want=%d need=%d have=%d\n",
-	     ac->want_keys, need, xi->have_keys);
-	return need != 0;
-}
-
-static int ceph_x_encrypt_buflen(int ilen)
-{
-	return sizeof(struct ceph_x_encrypt_header) + ilen + 16 +
-		sizeof(u32);
-}
-
-static int ceph_x_encrypt(struct ceph_crypto_key *secret,
-			  void *ibuf, int ilen, void *obuf, size_t olen)
-{
-	struct ceph_x_encrypt_header head = {
-		.struct_v = 1,
-		.magic = cpu_to_le64(CEPHX_ENC_MAGIC)
-	};
-	size_t len = olen - sizeof(u32);
-	int ret;
-
-	ret = ceph_encrypt2(secret, obuf + sizeof(u32), &len,
-			    &head, sizeof(head), ibuf, ilen);
-	if (ret)
-		return ret;
-	ceph_encode_32(&obuf, len);
-	return len + sizeof(u32);
-}
-
-static int ceph_x_decrypt(struct ceph_crypto_key *secret,
-			  void **p, void *end, void *obuf, size_t olen)
-{
-	struct ceph_x_encrypt_header head;
-	size_t head_len = sizeof(head);
-	int len, ret;
-
-	len = ceph_decode_32(p);
-	if (*p + len > end)
-		return -EINVAL;
-
-	dout("ceph_x_decrypt len %d\n", len);
-	ret = ceph_decrypt2(secret, &head, &head_len, obuf, &olen,
-			    *p, len);
-	if (ret)
-		return ret;
-	if (head.struct_v != 1 || le64_to_cpu(head.magic) != CEPHX_ENC_MAGIC)
-		return -EPERM;
-	*p += len;
-	return olen;
-}
-
-/*
- * get existing (or insert new) ticket handler
- */
-static struct ceph_x_ticket_handler *
-get_ticket_handler(struct ceph_auth_client *ac, int service)
-{
-	struct ceph_x_ticket_handler *th;
-	struct ceph_x_info *xi = ac->private;
-	struct rb_node *parent = NULL, **p = &xi->ticket_handlers.rb_node;
-
-	while (*p) {
-		parent = *p;
-		th = rb_entry(parent, struct ceph_x_ticket_handler, node);
-		if (service < th->service)
-			p = &(*p)->rb_left;
-		else if (service > th->service)
-			p = &(*p)->rb_right;
-		else
-			return th;
-	}
-
-	/* add it */
-	th = kzalloc(sizeof(*th), GFP_NOFS);
-	if (!th)
-		return ERR_PTR(-ENOMEM);
-	th->service = service;
-	rb_link_node(&th->node, parent, p);
-	rb_insert_color(&th->node, &xi->ticket_handlers);
-	return th;
-}
-
-static void remove_ticket_handler(struct ceph_auth_client *ac,
-				  struct ceph_x_ticket_handler *th)
-{
-	struct ceph_x_info *xi = ac->private;
-
-	dout("remove_ticket_handler %p %d\n", th, th->service);
-	rb_erase(&th->node, &xi->ticket_handlers);
-	ceph_crypto_key_destroy(&th->session_key);
-	if (th->ticket_blob)
-		ceph_buffer_put(th->ticket_blob);
-	kfree(th);
-}
-
-static int ceph_x_proc_ticket_reply(struct ceph_auth_client *ac,
-				    struct ceph_crypto_key *secret,
-				    void *buf, void *end)
-{
-	struct ceph_x_info *xi = ac->private;
-	int num;
-	void *p = buf;
-	int ret;
-	char *dbuf;
-	char *ticket_buf;
-	u8 reply_struct_v;
-
-	dbuf = kmalloc(TEMP_TICKET_BUF_LEN, GFP_NOFS);
-	if (!dbuf)
-		return -ENOMEM;
-
-	ret = -ENOMEM;
-	ticket_buf = kmalloc(TEMP_TICKET_BUF_LEN, GFP_NOFS);
-	if (!ticket_buf)
-		goto out_dbuf;
-
-	ceph_decode_need(&p, end, 1 + sizeof(u32), bad);
-	reply_struct_v = ceph_decode_8(&p);
-	if (reply_struct_v != 1)
-		goto bad;
-	num = ceph_decode_32(&p);
-	dout("%d tickets\n", num);
-	while (num--) {
-		int type;
-		u8 tkt_struct_v, blob_struct_v;
-		struct ceph_x_ticket_handler *th;
-		void *dp, *dend;
-		int dlen;
-		char is_enc;
-		struct timespec validity;
-		struct ceph_crypto_key old_key;
-		void *tp, *tpend;
-		struct ceph_timespec new_validity;
-		struct ceph_crypto_key new_session_key;
-		struct ceph_buffer *new_ticket_blob;
-		unsigned long new_expires, new_renew_after;
-		u64 new_secret_id;
-
-		ceph_decode_need(&p, end, sizeof(u32) + 1, bad);
-
-		type = ceph_decode_32(&p);
-		dout(" ticket type %d %s\n", type, ceph_entity_type_name(type));
-
-		tkt_struct_v = ceph_decode_8(&p);
-		if (tkt_struct_v != 1)
-			goto bad;
-
-		th = get_ticket_handler(ac, type);
-		if (IS_ERR(th)) {
-			ret = PTR_ERR(th);
-			goto out;
-		}
-
-		/* blob for me */
-		dlen = ceph_x_decrypt(secret, &p, end, dbuf,
-				      TEMP_TICKET_BUF_LEN);
-		if (dlen <= 0) {
-			ret = dlen;
-			goto out;
-		}
-		dout(" decrypted %d bytes\n", dlen);
-		dend = dbuf + dlen;
-		dp = dbuf;
-
-		tkt_struct_v = ceph_decode_8(&dp);
-		if (tkt_struct_v != 1)
-			goto bad;
-
-		memcpy(&old_key, &th->session_key, sizeof(old_key));
-		ret = ceph_crypto_key_decode(&new_session_key, &dp, dend);
-		if (ret)
-			goto out;
-
-		ceph_decode_copy(&dp, &new_validity, sizeof(new_validity));
-		ceph_decode_timespec(&validity, &new_validity);
-		new_expires = get_seconds() + validity.tv_sec;
-		new_renew_after = new_expires - (validity.tv_sec / 4);
-		dout(" expires=%lu renew_after=%lu\n", new_expires,
-		     new_renew_after);
-
-		/* ticket blob for service */
-		ceph_decode_8_safe(&p, end, is_enc, bad);
-		tp = ticket_buf;
-		if (is_enc) {
-			/* encrypted */
-			dout(" encrypted ticket\n");
-			dlen = ceph_x_decrypt(&old_key, &p, end, ticket_buf,
-					      TEMP_TICKET_BUF_LEN);
-			if (dlen < 0) {
-				ret = dlen;
-				goto out;
-			}
-			dlen = ceph_decode_32(&tp);
-		} else {
-			/* unencrypted */
-			ceph_decode_32_safe(&p, end, dlen, bad);
-			ceph_decode_need(&p, end, dlen, bad);
-			ceph_decode_copy(&p, ticket_buf, dlen);
-		}
-		tpend = tp + dlen;
-		dout(" ticket blob is %d bytes\n", dlen);
-		ceph_decode_need(&tp, tpend, 1 + sizeof(u64), bad);
-		blob_struct_v = ceph_decode_8(&tp);
-		new_secret_id = ceph_decode_64(&tp);
-		ret = ceph_decode_buffer(&new_ticket_blob, &tp, tpend);
-		if (ret)
-			goto out;
-
-		/* all is well, update our ticket */
-		ceph_crypto_key_destroy(&th->session_key);
-		if (th->ticket_blob)
-			ceph_buffer_put(th->ticket_blob);
-		th->session_key = new_session_key;
-		th->ticket_blob = new_ticket_blob;
-		th->validity = new_validity;
-		th->secret_id = new_secret_id;
-		th->expires = new_expires;
-		th->renew_after = new_renew_after;
-		dout(" got ticket service %d (%s) secret_id %lld len %d\n",
-		     type, ceph_entity_type_name(type), th->secret_id,
-		     (int)th->ticket_blob->vec.iov_len);
-		xi->have_keys |= th->service;
-	}
-
-	ret = 0;
-out:
-	kfree(ticket_buf);
-out_dbuf:
-	kfree(dbuf);
-	return ret;
-
-bad:
-	ret = -EINVAL;
-	goto out;
-}
-
-static int ceph_x_build_authorizer(struct ceph_auth_client *ac,
-				   struct ceph_x_ticket_handler *th,
-				   struct ceph_x_authorizer *au)
-{
-	int maxlen;
-	struct ceph_x_authorize_a *msg_a;
-	struct ceph_x_authorize_b msg_b;
-	void *p, *end;
-	int ret;
-	int ticket_blob_len =
-		(th->ticket_blob ? th->ticket_blob->vec.iov_len : 0);
-
-	dout("build_authorizer for %s %p\n",
-	     ceph_entity_type_name(th->service), au);
-
-	maxlen = sizeof(*msg_a) + sizeof(msg_b) +
-		ceph_x_encrypt_buflen(ticket_blob_len);
-	dout("  need len %d\n", maxlen);
-	if (au->buf && au->buf->alloc_len < maxlen) {
-		ceph_buffer_put(au->buf);
-		au->buf = NULL;
-	}
-	if (!au->buf) {
-		au->buf = ceph_buffer_new(maxlen, GFP_NOFS);
-		if (!au->buf)
-			return -ENOMEM;
-	}
-	au->service = th->service;
-
-	msg_a = au->buf->vec.iov_base;
-	msg_a->struct_v = 1;
-	msg_a->global_id = cpu_to_le64(ac->global_id);
-	msg_a->service_id = cpu_to_le32(th->service);
-	msg_a->ticket_blob.struct_v = 1;
-	msg_a->ticket_blob.secret_id = cpu_to_le64(th->secret_id);
-	msg_a->ticket_blob.blob_len = cpu_to_le32(ticket_blob_len);
-	if (ticket_blob_len) {
-		memcpy(msg_a->ticket_blob.blob, th->ticket_blob->vec.iov_base,
-		       th->ticket_blob->vec.iov_len);
-	}
-	dout(" th %p secret_id %lld %lld\n", th, th->secret_id,
-	     le64_to_cpu(msg_a->ticket_blob.secret_id));
-
-	p = msg_a + 1;
-	p += ticket_blob_len;
-	end = au->buf->vec.iov_base + au->buf->vec.iov_len;
-
-	get_random_bytes(&au->nonce, sizeof(au->nonce));
-	msg_b.struct_v = 1;
-	msg_b.nonce = cpu_to_le64(au->nonce);
-	ret = ceph_x_encrypt(&th->session_key, &msg_b, sizeof(msg_b),
-			     p, end - p);
-	if (ret < 0)
-		goto out_buf;
-	p += ret;
-	au->buf->vec.iov_len = p - au->buf->vec.iov_base;
-	dout(" built authorizer nonce %llx len %d\n", au->nonce,
-	     (int)au->buf->vec.iov_len);
-	BUG_ON(au->buf->vec.iov_len > maxlen);
-	return 0;
-
-out_buf:
-	ceph_buffer_put(au->buf);
-	au->buf = NULL;
-	return ret;
-}
-
-static int ceph_x_encode_ticket(struct ceph_x_ticket_handler *th,
-				void **p, void *end)
-{
-	ceph_decode_need(p, end, 1 + sizeof(u64), bad);
-	ceph_encode_8(p, 1);
-	ceph_encode_64(p, th->secret_id);
-	if (th->ticket_blob) {
-		const char *buf = th->ticket_blob->vec.iov_base;
-		u32 len = th->ticket_blob->vec.iov_len;
-
-		ceph_encode_32_safe(p, end, len, bad);
-		ceph_encode_copy_safe(p, end, buf, len, bad);
-	} else {
-		ceph_encode_32_safe(p, end, 0, bad);
-	}
-
-	return 0;
-bad:
-	return -ERANGE;
-}
-
-static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed)
-{
-	int want = ac->want_keys;
-	struct ceph_x_info *xi = ac->private;
-	int service;
-
-	*pneed = ac->want_keys & ~(xi->have_keys);
-
-	for (service = 1; service <= want; service <<= 1) {
-		struct ceph_x_ticket_handler *th;
-
-		if (!(ac->want_keys & service))
-			continue;
-
-		if (*pneed & service)
-			continue;
-
-		th = get_ticket_handler(ac, service);
-
-		if (IS_ERR(th)) {
-			*pneed |= service;
-			continue;
-		}
-
-		if (get_seconds() >= th->renew_after)
-			*pneed |= service;
-		if (get_seconds() >= th->expires)
-			xi->have_keys &= ~service;
-	}
-}
-
-
-static int ceph_x_build_request(struct ceph_auth_client *ac,
-				void *buf, void *end)
-{
-	struct ceph_x_info *xi = ac->private;
-	int need;
-	struct ceph_x_request_header *head = buf;
-	int ret;
-	struct ceph_x_ticket_handler *th =
-		get_ticket_handler(ac, CEPH_ENTITY_TYPE_AUTH);
-
-	if (IS_ERR(th))
-		return PTR_ERR(th);
-
-	ceph_x_validate_tickets(ac, &need);
-
-	dout("build_request want %x have %x need %x\n",
-	     ac->want_keys, xi->have_keys, need);
-
-	if (need & CEPH_ENTITY_TYPE_AUTH) {
-		struct ceph_x_authenticate *auth = (void *)(head + 1);
-		void *p = auth + 1;
-		struct ceph_x_challenge_blob tmp;
-		char tmp_enc[40];
-		u64 *u;
-
-		if (p > end)
-			return -ERANGE;
-
-		dout(" get_auth_session_key\n");
-		head->op = cpu_to_le16(CEPHX_GET_AUTH_SESSION_KEY);
-
-		/* encrypt and hash */
-		get_random_bytes(&auth->client_challenge, sizeof(u64));
-		tmp.client_challenge = auth->client_challenge;
-		tmp.server_challenge = cpu_to_le64(xi->server_challenge);
-		ret = ceph_x_encrypt(&xi->secret, &tmp, sizeof(tmp),
-				     tmp_enc, sizeof(tmp_enc));
-		if (ret < 0)
-			return ret;
-
-		auth->struct_v = 1;
-		auth->key = 0;
-		for (u = (u64 *)tmp_enc; u + 1 <= (u64 *)(tmp_enc + ret); u++)
-			auth->key ^= *(__le64 *)u;
-		dout(" server_challenge %llx client_challenge %llx key %llx\n",
-		     xi->server_challenge, le64_to_cpu(auth->client_challenge),
-		     le64_to_cpu(auth->key));
-
-		/* now encode the old ticket if exists */
-		ret = ceph_x_encode_ticket(th, &p, end);
-		if (ret < 0)
-			return ret;
-
-		return p - buf;
-	}
-
-	if (need) {
-		void *p = head + 1;
-		struct ceph_x_service_ticket_request *req;
-
-		if (p > end)
-			return -ERANGE;
-		head->op = cpu_to_le16(CEPHX_GET_PRINCIPAL_SESSION_KEY);
-
-		ret = ceph_x_build_authorizer(ac, th, &xi->auth_authorizer);
-		if (ret)
-			return ret;
-		ceph_encode_copy(&p, xi->auth_authorizer.buf->vec.iov_base,
-				 xi->auth_authorizer.buf->vec.iov_len);
-
-		req = p;
-		req->keys = cpu_to_le32(need);
-		p += sizeof(*req);
-		return p - buf;
-	}
-
-	return 0;
-}
-
-static int ceph_x_handle_reply(struct ceph_auth_client *ac, int result,
-			       void *buf, void *end)
-{
-	struct ceph_x_info *xi = ac->private;
-	struct ceph_x_reply_header *head = buf;
-	struct ceph_x_ticket_handler *th;
-	int len = end - buf;
-	int op;
-	int ret;
-
-	if (result)
-		return result;  /* XXX hmm? */
-
-	if (xi->starting) {
-		/* it's a hello */
-		struct ceph_x_server_challenge *sc = buf;
-
-		if (len != sizeof(*sc))
-			return -EINVAL;
-		xi->server_challenge = le64_to_cpu(sc->server_challenge);
-		dout("handle_reply got server challenge %llx\n",
-		     xi->server_challenge);
-		xi->starting = false;
-		xi->have_keys &= ~CEPH_ENTITY_TYPE_AUTH;
-		return -EAGAIN;
-	}
-
-	op = le16_to_cpu(head->op);
-	result = le32_to_cpu(head->result);
-	dout("handle_reply op %d result %d\n", op, result);
-	switch (op) {
-	case CEPHX_GET_AUTH_SESSION_KEY:
-		/* verify auth key */
-		ret = ceph_x_proc_ticket_reply(ac, &xi->secret,
-					       buf + sizeof(*head), end);
-		break;
-
-	case CEPHX_GET_PRINCIPAL_SESSION_KEY:
-		th = get_ticket_handler(ac, CEPH_ENTITY_TYPE_AUTH);
-		if (IS_ERR(th))
-			return PTR_ERR(th);
-		ret = ceph_x_proc_ticket_reply(ac, &th->session_key,
-					       buf + sizeof(*head), end);
-		break;
-
-	default:
-		return -EINVAL;
-	}
-	if (ret)
-		return ret;
-	if (ac->want_keys == xi->have_keys)
-		return 0;
-	return -EAGAIN;
-}
-
-static int ceph_x_create_authorizer(
-	struct ceph_auth_client *ac, int peer_type,
-	struct ceph_authorizer **a,
-	void **buf, size_t *len,
-	void **reply_buf, size_t *reply_len)
-{
-	struct ceph_x_authorizer *au;
-	struct ceph_x_ticket_handler *th;
-	int ret;
-
-	th = get_ticket_handler(ac, peer_type);
-	if (IS_ERR(th))
-		return PTR_ERR(th);
-
-	au = kzalloc(sizeof(*au), GFP_NOFS);
-	if (!au)
-		return -ENOMEM;
-
-	ret = ceph_x_build_authorizer(ac, th, au);
-	if (ret) {
-		kfree(au);
-		return ret;
-	}
-
-	*a = (struct ceph_authorizer *)au;
-	*buf = au->buf->vec.iov_base;
-	*len = au->buf->vec.iov_len;
-	*reply_buf = au->reply_buf;
-	*reply_len = sizeof(au->reply_buf);
-	return 0;
-}
-
-static int ceph_x_verify_authorizer_reply(struct ceph_auth_client *ac,
-					  struct ceph_authorizer *a, size_t len)
-{
-	struct ceph_x_authorizer *au = (void *)a;
-	struct ceph_x_ticket_handler *th;
-	int ret = 0;
-	struct ceph_x_authorize_reply reply;
-	void *p = au->reply_buf;
-	void *end = p + sizeof(au->reply_buf);
-
-	th = get_ticket_handler(ac, au->service);
-	if (IS_ERR(th))
-		return PTR_ERR(th);
-	ret = ceph_x_decrypt(&th->session_key, &p, end, &reply, sizeof(reply));
-	if (ret < 0)
-		return ret;
-	if (ret != sizeof(reply))
-		return -EPERM;
-
-	if (au->nonce + 1 != le64_to_cpu(reply.nonce_plus_one))
-		ret = -EPERM;
-	else
-		ret = 0;
-	dout("verify_authorizer_reply nonce %llx got %llx ret %d\n",
-	     au->nonce, le64_to_cpu(reply.nonce_plus_one), ret);
-	return ret;
-}
-
-static void ceph_x_destroy_authorizer(struct ceph_auth_client *ac,
-				      struct ceph_authorizer *a)
-{
-	struct ceph_x_authorizer *au = (void *)a;
-
-	ceph_buffer_put(au->buf);
-	kfree(au);
-}
-
-
-static void ceph_x_reset(struct ceph_auth_client *ac)
-{
-	struct ceph_x_info *xi = ac->private;
-
-	dout("reset\n");
-	xi->starting = true;
-	xi->server_challenge = 0;
-}
-
-static void ceph_x_destroy(struct ceph_auth_client *ac)
-{
-	struct ceph_x_info *xi = ac->private;
-	struct rb_node *p;
-
-	dout("ceph_x_destroy %p\n", ac);
-	ceph_crypto_key_destroy(&xi->secret);
-
-	while ((p = rb_first(&xi->ticket_handlers)) != NULL) {
-		struct ceph_x_ticket_handler *th =
-			rb_entry(p, struct ceph_x_ticket_handler, node);
-		remove_ticket_handler(ac, th);
-	}
-
-	if (xi->auth_authorizer.buf)
-		ceph_buffer_put(xi->auth_authorizer.buf);
-
-	kfree(ac->private);
-	ac->private = NULL;
-}
-
-static void ceph_x_invalidate_authorizer(struct ceph_auth_client *ac,
-				   int peer_type)
-{
-	struct ceph_x_ticket_handler *th;
-
-	th = get_ticket_handler(ac, peer_type);
-	if (!IS_ERR(th))
-		remove_ticket_handler(ac, th);
-}
-
-
-static const struct ceph_auth_client_ops ceph_x_ops = {
-	.name = "x",
-	.is_authenticated = ceph_x_is_authenticated,
-	.should_authenticate = ceph_x_should_authenticate,
-	.build_request = ceph_x_build_request,
-	.handle_reply = ceph_x_handle_reply,
-	.create_authorizer = ceph_x_create_authorizer,
-	.verify_authorizer_reply = ceph_x_verify_authorizer_reply,
-	.destroy_authorizer = ceph_x_destroy_authorizer,
-	.invalidate_authorizer = ceph_x_invalidate_authorizer,
-	.reset =  ceph_x_reset,
-	.destroy = ceph_x_destroy,
-};
-
-
-int ceph_x_init(struct ceph_auth_client *ac)
-{
-	struct ceph_x_info *xi;
-	int ret;
-
-	dout("ceph_x_init %p\n", ac);
-	ret = -ENOMEM;
-	xi = kzalloc(sizeof(*xi), GFP_NOFS);
-	if (!xi)
-		goto out;
-
-	ret = -EINVAL;
-	if (!ac->secret) {
-		pr_err("no secret set (for auth_x protocol)\n");
-		goto out_nomem;
-	}
-
-	ret = ceph_crypto_key_unarmor(&xi->secret, ac->secret);
-	if (ret)
-		goto out_nomem;
-
-	xi->starting = true;
-	xi->ticket_handlers = RB_ROOT;
-
-	ac->protocol = CEPH_AUTH_CEPHX;
-	ac->private = xi;
-	ac->ops = &ceph_x_ops;
-	return 0;
-
-out_nomem:
-	kfree(xi);
-out:
-	return ret;
-}
-
-
diff --git a/fs/ceph/auth_x.h b/fs/ceph/auth_x.h
deleted file mode 100644
index ff6f8180e681..000000000000
--- a/fs/ceph/auth_x.h
+++ /dev/null
@@ -1,49 +0,0 @@
-#ifndef _FS_CEPH_AUTH_X_H
-#define _FS_CEPH_AUTH_X_H
-
-#include <linux/rbtree.h>
-
-#include "crypto.h"
-#include "auth.h"
-#include "auth_x_protocol.h"
-
-/*
- * Handle ticket for a single service.
- */
-struct ceph_x_ticket_handler {
-	struct rb_node node;
-	unsigned service;
-
-	struct ceph_crypto_key session_key;
-	struct ceph_timespec validity;
-
-	u64 secret_id;
-	struct ceph_buffer *ticket_blob;
-
-	unsigned long renew_after, expires;
-};
-
-
-struct ceph_x_authorizer {
-	struct ceph_buffer *buf;
-	unsigned service;
-	u64 nonce;
-	char reply_buf[128];  /* big enough for encrypted blob */
-};
-
-struct ceph_x_info {
-	struct ceph_crypto_key secret;
-
-	bool starting;
-	u64 server_challenge;
-
-	unsigned have_keys;
-	struct rb_root ticket_handlers;
-
-	struct ceph_x_authorizer auth_authorizer;
-};
-
-extern int ceph_x_init(struct ceph_auth_client *ac);
-
-#endif
-
diff --git a/fs/ceph/auth_x_protocol.h b/fs/ceph/auth_x_protocol.h
deleted file mode 100644
index 671d30576c4f..000000000000
--- a/fs/ceph/auth_x_protocol.h
+++ /dev/null
@@ -1,90 +0,0 @@
-#ifndef __FS_CEPH_AUTH_X_PROTOCOL
-#define __FS_CEPH_AUTH_X_PROTOCOL
-
-#define CEPHX_GET_AUTH_SESSION_KEY      0x0100
-#define CEPHX_GET_PRINCIPAL_SESSION_KEY 0x0200
-#define CEPHX_GET_ROTATING_KEY          0x0400
-
-/* common bits */
-struct ceph_x_ticket_blob {
-	__u8 struct_v;
-	__le64 secret_id;
-	__le32 blob_len;
-	char blob[];
-} __attribute__ ((packed));
-
-
-/* common request/reply headers */
-struct ceph_x_request_header {
-	__le16 op;
-} __attribute__ ((packed));
-
-struct ceph_x_reply_header {
-	__le16 op;
-	__le32 result;
-} __attribute__ ((packed));
-
-
-/* authenticate handshake */
-
-/* initial hello (no reply header) */
-struct ceph_x_server_challenge {
-	__u8 struct_v;
-	__le64 server_challenge;
-} __attribute__ ((packed));
-
-struct ceph_x_authenticate {
-	__u8 struct_v;
-	__le64 client_challenge;
-	__le64 key;
-	/* ticket blob */
-} __attribute__ ((packed));
-
-struct ceph_x_service_ticket_request {
-	__u8 struct_v;
-	__le32 keys;
-} __attribute__ ((packed));
-
-struct ceph_x_challenge_blob {
-	__le64 server_challenge;
-	__le64 client_challenge;
-} __attribute__ ((packed));
-
-
-
-/* authorize handshake */
-
-/*
- * The authorizer consists of two pieces:
- *  a - service id, ticket blob
- *  b - encrypted with session key
- */
-struct ceph_x_authorize_a {
-	__u8 struct_v;
-	__le64 global_id;
-	__le32 service_id;
-	struct ceph_x_ticket_blob ticket_blob;
-} __attribute__ ((packed));
-
-struct ceph_x_authorize_b {
-	__u8 struct_v;
-	__le64 nonce;
-} __attribute__ ((packed));
-
-struct ceph_x_authorize_reply {
-	__u8 struct_v;
-	__le64 nonce_plus_one;
-} __attribute__ ((packed));
-
-
-/*
- * encyption bundle
- */
-#define CEPHX_ENC_MAGIC 0xff009cad8826aa55ull
-
-struct ceph_x_encrypt_header {
-	__u8 struct_v;
-	__le64 magic;
-} __attribute__ ((packed));
-
-#endif
diff --git a/fs/ceph/buffer.c b/fs/ceph/buffer.c
deleted file mode 100644
index cd39f17021de..000000000000
--- a/fs/ceph/buffer.c
+++ /dev/null
@@ -1,65 +0,0 @@
-
-#include "ceph_debug.h"
-
-#include <linux/slab.h>
-
-#include "buffer.h"
-#include "decode.h"
-
-struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp)
-{
-	struct ceph_buffer *b;
-
-	b = kmalloc(sizeof(*b), gfp);
-	if (!b)
-		return NULL;
-
-	b->vec.iov_base = kmalloc(len, gfp | __GFP_NOWARN);
-	if (b->vec.iov_base) {
-		b->is_vmalloc = false;
-	} else {
-		b->vec.iov_base = __vmalloc(len, gfp, PAGE_KERNEL);
-		if (!b->vec.iov_base) {
-			kfree(b);
-			return NULL;
-		}
-		b->is_vmalloc = true;
-	}
-
-	kref_init(&b->kref);
-	b->alloc_len = len;
-	b->vec.iov_len = len;
-	dout("buffer_new %p\n", b);
-	return b;
-}
-
-void ceph_buffer_release(struct kref *kref)
-{
-	struct ceph_buffer *b = container_of(kref, struct ceph_buffer, kref);
-
-	dout("buffer_release %p\n", b);
-	if (b->vec.iov_base) {
-		if (b->is_vmalloc)
-			vfree(b->vec.iov_base);
-		else
-			kfree(b->vec.iov_base);
-	}
-	kfree(b);
-}
-
-int ceph_decode_buffer(struct ceph_buffer **b, void **p, void *end)
-{
-	size_t len;
-
-	ceph_decode_need(p, end, sizeof(u32), bad);
-	len = ceph_decode_32(p);
-	dout("decode_buffer len %d\n", (int)len);
-	ceph_decode_need(p, end, len, bad);
-	*b = ceph_buffer_new(len, GFP_NOFS);
-	if (!*b)
-		return -ENOMEM;
-	ceph_decode_copy(p, (*b)->vec.iov_base, len);
-	return 0;
-bad:
-	return -EINVAL;
-}
diff --git a/fs/ceph/buffer.h b/fs/ceph/buffer.h
deleted file mode 100644
index 58d19014068f..000000000000
--- a/fs/ceph/buffer.h
+++ /dev/null
@@ -1,39 +0,0 @@
-#ifndef __FS_CEPH_BUFFER_H
-#define __FS_CEPH_BUFFER_H
-
-#include <linux/kref.h>
-#include <linux/mm.h>
-#include <linux/vmalloc.h>
-#include <linux/types.h>
-#include <linux/uio.h>
-
-/*
- * a simple reference counted buffer.
- *
- * use kmalloc for small sizes (<= one page), vmalloc for larger
- * sizes.
- */
-struct ceph_buffer {
-	struct kref kref;
-	struct kvec vec;
-	size_t alloc_len;
-	bool is_vmalloc;
-};
-
-extern struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp);
-extern void ceph_buffer_release(struct kref *kref);
-
-static inline struct ceph_buffer *ceph_buffer_get(struct ceph_buffer *b)
-{
-	kref_get(&b->kref);
-	return b;
-}
-
-static inline void ceph_buffer_put(struct ceph_buffer *b)
-{
-	kref_put(&b->kref, ceph_buffer_release);
-}
-
-extern int ceph_decode_buffer(struct ceph_buffer **b, void **p, void *end);
-
-#endif
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 5e9da996a151..3cff67cbb9c0 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/fs.h>
 #include <linux/kernel.h>
@@ -9,8 +9,9 @@
 #include <linux/writeback.h>
 
 #include "super.h"
-#include "decode.h"
-#include "messenger.h"
+#include "mds_client.h"
+#include <linux/ceph/decode.h>
+#include <linux/ceph/messenger.h>
 
 /*
  * Capability management
@@ -287,11 +288,11 @@ void ceph_put_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap)
 	spin_unlock(&mdsc->caps_list_lock);
 }
 
-void ceph_reservation_status(struct ceph_client *client,
+void ceph_reservation_status(struct ceph_fs_client *fsc,
 			     int *total, int *avail, int *used, int *reserved,
 			     int *min)
 {
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 
 	if (total)
 		*total = mdsc->caps_total_count;
@@ -399,7 +400,7 @@ static void __insert_cap_node(struct ceph_inode_info *ci,
 static void __cap_set_timeouts(struct ceph_mds_client *mdsc,
 			       struct ceph_inode_info *ci)
 {
-	struct ceph_mount_args *ma = mdsc->client->mount_args;
+	struct ceph_mount_options *ma = mdsc->fsc->mount_options;
 
 	ci->i_hold_caps_min = round_jiffies(jiffies +
 					    ma->caps_wanted_delay_min * HZ);
@@ -515,7 +516,7 @@ int ceph_add_cap(struct inode *inode,
 		 unsigned seq, unsigned mseq, u64 realmino, int flags,
 		 struct ceph_cap_reservation *caps_reservation)
 {
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_cap *new_cap = NULL;
 	struct ceph_cap *cap;
@@ -873,7 +874,7 @@ void __ceph_remove_cap(struct ceph_cap *cap)
 	struct ceph_mds_session *session = cap->session;
 	struct ceph_inode_info *ci = cap->ci;
 	struct ceph_mds_client *mdsc =
-		&ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
+		ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 	int removed = 0;
 
 	dout("__ceph_remove_cap %p from %p\n", cap, &ci->vfs_inode);
@@ -1210,7 +1211,7 @@ void __ceph_flush_snaps(struct ceph_inode_info *ci,
 	int mds;
 	struct ceph_cap_snap *capsnap;
 	u32 mseq;
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	struct ceph_mds_session *session = NULL; /* if session != NULL, we hold
 						    session->s_mutex */
 	u64 next_follows = 0;  /* keep track of how far we've gotten through the
@@ -1336,7 +1337,7 @@ static void ceph_flush_snaps(struct ceph_inode_info *ci)
 void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
 {
 	struct ceph_mds_client *mdsc =
-		&ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
+		ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 	struct inode *inode = &ci->vfs_inode;
 	int was = ci->i_dirty_caps;
 	int dirty = 0;
@@ -1378,7 +1379,7 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
 static int __mark_caps_flushing(struct inode *inode,
 				 struct ceph_mds_session *session)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int flushing;
 
@@ -1462,8 +1463,8 @@ static int try_nonblocking_invalidate(struct inode *inode)
 void ceph_check_caps(struct ceph_inode_info *ci, int flags,
 		     struct ceph_mds_session *session)
 {
-	struct ceph_client *client = ceph_inode_to_client(&ci->vfs_inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(&ci->vfs_inode);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = &ci->vfs_inode;
 	struct ceph_cap *cap;
 	int file_wanted, used;
@@ -1706,7 +1707,7 @@ ack:
 static int try_flush_caps(struct inode *inode, struct ceph_mds_session *session,
 			  unsigned *flush_tid)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int unlock_session = session ? 0 : 1;
 	int flushing = 0;
@@ -1872,7 +1873,7 @@ int ceph_write_inode(struct inode *inode, struct writeback_control *wbc)
 				       caps_are_flushed(inode, flush_tid));
 	} else {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(inode->i_sb)->mdsc;
+			ceph_sb_to_client(inode->i_sb)->mdsc;
 
 		spin_lock(&inode->i_lock);
 		if (__ceph_caps_dirty(ci))
@@ -2465,7 +2466,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid,
 	__releases(inode->i_lock)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	unsigned seq = le32_to_cpu(m->seq);
 	int dirty = le32_to_cpu(m->dirty);
 	int cleaned = 0;
@@ -2713,7 +2714,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 		      struct ceph_msg *msg)
 {
 	struct ceph_mds_client *mdsc = session->s_mdsc;
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	struct inode *inode;
 	struct ceph_cap *cap;
 	struct ceph_mds_caps *h;
diff --git a/fs/ceph/ceph_debug.h b/fs/ceph/ceph_debug.h
deleted file mode 100644
index 1818c2305610..000000000000
--- a/fs/ceph/ceph_debug.h
+++ /dev/null
@@ -1,37 +0,0 @@
-#ifndef _FS_CEPH_DEBUG_H
-#define _FS_CEPH_DEBUG_H
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#ifdef CONFIG_CEPH_FS_PRETTYDEBUG
-
-/*
- * wrap pr_debug to include a filename:lineno prefix on each line.
- * this incurs some overhead (kernel size and execution time) due to
- * the extra function call at each call site.
- */
-
-# if defined(DEBUG) || defined(CONFIG_DYNAMIC_DEBUG)
-extern const char *ceph_file_part(const char *s, int len);
-#  define dout(fmt, ...)						\
-	pr_debug(" %12.12s:%-4d : " fmt,				\
-		 ceph_file_part(__FILE__, sizeof(__FILE__)),		\
-		 __LINE__, ##__VA_ARGS__)
-# else
-/* faux printk call just to see any compiler warnings. */
-#  define dout(fmt, ...)	do {				\
-		if (0)						\
-			printk(KERN_DEBUG fmt, ##__VA_ARGS__);	\
-	} while (0)
-# endif
-
-#else
-
-/*
- * or, just wrap pr_debug
- */
-# define dout(fmt, ...)	pr_debug(" " fmt, ##__VA_ARGS__)
-
-#endif
-
-#endif
diff --git a/fs/ceph/ceph_frag.c b/fs/ceph/ceph_frag.c
index ab6cf35c4091..bdce8b1fbd06 100644
--- a/fs/ceph/ceph_frag.c
+++ b/fs/ceph/ceph_frag.c
@@ -1,7 +1,8 @@
 /*
  * Ceph 'frag' type
  */
-#include "types.h"
+#include <linux/module.h>
+#include <linux/ceph/types.h>
 
 int ceph_frag_compare(__u32 a, __u32 b)
 {
diff --git a/fs/ceph/ceph_frag.h b/fs/ceph/ceph_frag.h
deleted file mode 100644
index 5babb8e95352..000000000000
--- a/fs/ceph/ceph_frag.h
+++ /dev/null
@@ -1,109 +0,0 @@
-#ifndef FS_CEPH_FRAG_H
-#define FS_CEPH_FRAG_H
-
-/*
- * "Frags" are a way to describe a subset of a 32-bit number space,
- * using a mask and a value to match against that mask.  Any given frag
- * (subset of the number space) can be partitioned into 2^n sub-frags.
- *
- * Frags are encoded into a 32-bit word:
- *   8 upper bits = "bits"
- *  24 lower bits = "value"
- * (We could go to 5+27 bits, but who cares.)
- *
- * We use the _most_ significant bits of the 24 bit value.  This makes
- * values logically sort.
- *
- * Unfortunately, because the "bits" field is still in the high bits, we
- * can't sort encoded frags numerically.  However, it does allow you
- * to feed encoded frags as values into frag_contains_value.
- */
-static inline __u32 ceph_frag_make(__u32 b, __u32 v)
-{
-	return (b << 24) |
-		(v & (0xffffffu << (24-b)) & 0xffffffu);
-}
-static inline __u32 ceph_frag_bits(__u32 f)
-{
-	return f >> 24;
-}
-static inline __u32 ceph_frag_value(__u32 f)
-{
-	return f & 0xffffffu;
-}
-static inline __u32 ceph_frag_mask(__u32 f)
-{
-	return (0xffffffu << (24-ceph_frag_bits(f))) & 0xffffffu;
-}
-static inline __u32 ceph_frag_mask_shift(__u32 f)
-{
-	return 24 - ceph_frag_bits(f);
-}
-
-static inline int ceph_frag_contains_value(__u32 f, __u32 v)
-{
-	return (v & ceph_frag_mask(f)) == ceph_frag_value(f);
-}
-static inline int ceph_frag_contains_frag(__u32 f, __u32 sub)
-{
-	/* is sub as specific as us, and contained by us? */
-	return ceph_frag_bits(sub) >= ceph_frag_bits(f) &&
-	       (ceph_frag_value(sub) & ceph_frag_mask(f)) == ceph_frag_value(f);
-}
-
-static inline __u32 ceph_frag_parent(__u32 f)
-{
-	return ceph_frag_make(ceph_frag_bits(f) - 1,
-			 ceph_frag_value(f) & (ceph_frag_mask(f) << 1));
-}
-static inline int ceph_frag_is_left_child(__u32 f)
-{
-	return ceph_frag_bits(f) > 0 &&
-		(ceph_frag_value(f) & (0x1000000 >> ceph_frag_bits(f))) == 0;
-}
-static inline int ceph_frag_is_right_child(__u32 f)
-{
-	return ceph_frag_bits(f) > 0 &&
-		(ceph_frag_value(f) & (0x1000000 >> ceph_frag_bits(f))) == 1;
-}
-static inline __u32 ceph_frag_sibling(__u32 f)
-{
-	return ceph_frag_make(ceph_frag_bits(f),
-		      ceph_frag_value(f) ^ (0x1000000 >> ceph_frag_bits(f)));
-}
-static inline __u32 ceph_frag_left_child(__u32 f)
-{
-	return ceph_frag_make(ceph_frag_bits(f)+1, ceph_frag_value(f));
-}
-static inline __u32 ceph_frag_right_child(__u32 f)
-{
-	return ceph_frag_make(ceph_frag_bits(f)+1,
-	      ceph_frag_value(f) | (0x1000000 >> (1+ceph_frag_bits(f))));
-}
-static inline __u32 ceph_frag_make_child(__u32 f, int by, int i)
-{
-	int newbits = ceph_frag_bits(f) + by;
-	return ceph_frag_make(newbits,
-			 ceph_frag_value(f) | (i << (24 - newbits)));
-}
-static inline int ceph_frag_is_leftmost(__u32 f)
-{
-	return ceph_frag_value(f) == 0;
-}
-static inline int ceph_frag_is_rightmost(__u32 f)
-{
-	return ceph_frag_value(f) == ceph_frag_mask(f);
-}
-static inline __u32 ceph_frag_next(__u32 f)
-{
-	return ceph_frag_make(ceph_frag_bits(f),
-			 ceph_frag_value(f) + (0x1000000 >> ceph_frag_bits(f)));
-}
-
-/*
- * comparator to sort frags logically, as when traversing the
- * number space in ascending order...
- */
-int ceph_frag_compare(__u32 a, __u32 b);
-
-#endif
diff --git a/fs/ceph/ceph_fs.c b/fs/ceph/ceph_fs.c
deleted file mode 100644
index 3ac6cc7c1156..000000000000
--- a/fs/ceph/ceph_fs.c
+++ /dev/null
@@ -1,72 +0,0 @@
-/*
- * Some non-inline ceph helpers
- */
-#include "types.h"
-
-/*
- * return true if @layout appears to be valid
- */
-int ceph_file_layout_is_valid(const struct ceph_file_layout *layout)
-{
-	__u32 su = le32_to_cpu(layout->fl_stripe_unit);
-	__u32 sc = le32_to_cpu(layout->fl_stripe_count);
-	__u32 os = le32_to_cpu(layout->fl_object_size);
-
-	/* stripe unit, object size must be non-zero, 64k increment */
-	if (!su || (su & (CEPH_MIN_STRIPE_UNIT-1)))
-		return 0;
-	if (!os || (os & (CEPH_MIN_STRIPE_UNIT-1)))
-		return 0;
-	/* object size must be a multiple of stripe unit */
-	if (os < su || os % su)
-		return 0;
-	/* stripe count must be non-zero */
-	if (!sc)
-		return 0;
-	return 1;
-}
-
-
-int ceph_flags_to_mode(int flags)
-{
-	int mode;
-
-#ifdef O_DIRECTORY  /* fixme */
-	if ((flags & O_DIRECTORY) == O_DIRECTORY)
-		return CEPH_FILE_MODE_PIN;
-#endif
-	if ((flags & O_APPEND) == O_APPEND)
-		flags |= O_WRONLY;
-
-	if ((flags & O_ACCMODE) == O_RDWR)
-		mode = CEPH_FILE_MODE_RDWR;
-	else if ((flags & O_ACCMODE) == O_WRONLY)
-		mode = CEPH_FILE_MODE_WR;
-	else
-		mode = CEPH_FILE_MODE_RD;
-
-#ifdef O_LAZY
-	if (flags & O_LAZY)
-		mode |= CEPH_FILE_MODE_LAZY;
-#endif
-
-	return mode;
-}
-
-int ceph_caps_for_mode(int mode)
-{
-	int caps = CEPH_CAP_PIN;
-
-	if (mode & CEPH_FILE_MODE_RD)
-		caps |= CEPH_CAP_FILE_SHARED |
-			CEPH_CAP_FILE_RD | CEPH_CAP_FILE_CACHE;
-	if (mode & CEPH_FILE_MODE_WR)
-		caps |= CEPH_CAP_FILE_EXCL |
-			CEPH_CAP_FILE_WR | CEPH_CAP_FILE_BUFFER |
-			CEPH_CAP_AUTH_SHARED | CEPH_CAP_AUTH_EXCL |
-			CEPH_CAP_XATTR_SHARED | CEPH_CAP_XATTR_EXCL;
-	if (mode & CEPH_FILE_MODE_LAZY)
-		caps |= CEPH_CAP_FILE_LAZYIO;
-
-	return caps;
-}
diff --git a/fs/ceph/ceph_fs.h b/fs/ceph/ceph_fs.h
deleted file mode 100644
index d5619ac86711..000000000000
--- a/fs/ceph/ceph_fs.h
+++ /dev/null
@@ -1,728 +0,0 @@
-/*
- * ceph_fs.h - Ceph constants and data types to share between kernel and
- * user space.
- *
- * Most types in this file are defined as little-endian, and are
- * primarily intended to describe data structures that pass over the
- * wire or that are stored on disk.
- *
- * LGPL2
- */
-
-#ifndef CEPH_FS_H
-#define CEPH_FS_H
-
-#include "msgr.h"
-#include "rados.h"
-
-/*
- * subprotocol versions.  when specific messages types or high-level
- * protocols change, bump the affected components.  we keep rev
- * internal cluster protocols separately from the public,
- * client-facing protocol.
- */
-#define CEPH_OSD_PROTOCOL     8 /* cluster internal */
-#define CEPH_MDS_PROTOCOL    12 /* cluster internal */
-#define CEPH_MON_PROTOCOL     5 /* cluster internal */
-#define CEPH_OSDC_PROTOCOL   24 /* server/client */
-#define CEPH_MDSC_PROTOCOL   32 /* server/client */
-#define CEPH_MONC_PROTOCOL   15 /* server/client */
-
-
-#define CEPH_INO_ROOT  1
-#define CEPH_INO_CEPH  2        /* hidden .ceph dir */
-
-/* arbitrary limit on max # of monitors (cluster of 3 is typical) */
-#define CEPH_MAX_MON   31
-
-
-/*
- * feature bits
- */
-#define CEPH_FEATURE_UID            (1<<0)
-#define CEPH_FEATURE_NOSRCADDR      (1<<1)
-#define CEPH_FEATURE_MONCLOCKCHECK  (1<<2)
-#define CEPH_FEATURE_FLOCK          (1<<3)
-
-
-/*
- * ceph_file_layout - describe data layout for a file/inode
- */
-struct ceph_file_layout {
-	/* file -> object mapping */
-	__le32 fl_stripe_unit;     /* stripe unit, in bytes.  must be multiple
-				      of page size. */
-	__le32 fl_stripe_count;    /* over this many objects */
-	__le32 fl_object_size;     /* until objects are this big, then move to
-				      new objects */
-	__le32 fl_cas_hash;        /* 0 = none; 1 = sha256 */
-
-	/* pg -> disk layout */
-	__le32 fl_object_stripe_unit;  /* for per-object parity, if any */
-
-	/* object -> pg layout */
-	__le32 fl_pg_preferred; /* preferred primary for pg (-1 for none) */
-	__le32 fl_pg_pool;      /* namespace, crush ruleset, rep level */
-} __attribute__ ((packed));
-
-#define CEPH_MIN_STRIPE_UNIT 65536
-
-int ceph_file_layout_is_valid(const struct ceph_file_layout *layout);
-
-
-/* crypto algorithms */
-#define CEPH_CRYPTO_NONE 0x0
-#define CEPH_CRYPTO_AES  0x1
-
-#define CEPH_AES_IV "cephsageyudagreg"
-
-/* security/authentication protocols */
-#define CEPH_AUTH_UNKNOWN	0x0
-#define CEPH_AUTH_NONE	 	0x1
-#define CEPH_AUTH_CEPHX	 	0x2
-
-#define CEPH_AUTH_UID_DEFAULT ((__u64) -1)
-
-
-/*********************************************
- * message layer
- */
-
-/*
- * message types
- */
-
-/* misc */
-#define CEPH_MSG_SHUTDOWN               1
-#define CEPH_MSG_PING                   2
-
-/* client <-> monitor */
-#define CEPH_MSG_MON_MAP                4
-#define CEPH_MSG_MON_GET_MAP            5
-#define CEPH_MSG_STATFS                 13
-#define CEPH_MSG_STATFS_REPLY           14
-#define CEPH_MSG_MON_SUBSCRIBE          15
-#define CEPH_MSG_MON_SUBSCRIBE_ACK      16
-#define CEPH_MSG_AUTH			17
-#define CEPH_MSG_AUTH_REPLY		18
-
-/* client <-> mds */
-#define CEPH_MSG_MDS_MAP                21
-
-#define CEPH_MSG_CLIENT_SESSION         22
-#define CEPH_MSG_CLIENT_RECONNECT       23
-
-#define CEPH_MSG_CLIENT_REQUEST         24
-#define CEPH_MSG_CLIENT_REQUEST_FORWARD 25
-#define CEPH_MSG_CLIENT_REPLY           26
-#define CEPH_MSG_CLIENT_CAPS            0x310
-#define CEPH_MSG_CLIENT_LEASE           0x311
-#define CEPH_MSG_CLIENT_SNAP            0x312
-#define CEPH_MSG_CLIENT_CAPRELEASE      0x313
-
-/* pool ops */
-#define CEPH_MSG_POOLOP_REPLY           48
-#define CEPH_MSG_POOLOP                 49
-
-
-/* osd */
-#define CEPH_MSG_OSD_MAP          41
-#define CEPH_MSG_OSD_OP           42
-#define CEPH_MSG_OSD_OPREPLY      43
-
-/* pool operations */
-enum {
-  POOL_OP_CREATE			= 0x01,
-  POOL_OP_DELETE			= 0x02,
-  POOL_OP_AUID_CHANGE			= 0x03,
-  POOL_OP_CREATE_SNAP			= 0x11,
-  POOL_OP_DELETE_SNAP			= 0x12,
-  POOL_OP_CREATE_UNMANAGED_SNAP		= 0x21,
-  POOL_OP_DELETE_UNMANAGED_SNAP		= 0x22,
-};
-
-struct ceph_mon_request_header {
-	__le64 have_version;
-	__le16 session_mon;
-	__le64 session_mon_tid;
-} __attribute__ ((packed));
-
-struct ceph_mon_statfs {
-	struct ceph_mon_request_header monhdr;
-	struct ceph_fsid fsid;
-} __attribute__ ((packed));
-
-struct ceph_statfs {
-	__le64 kb, kb_used, kb_avail;
-	__le64 num_objects;
-} __attribute__ ((packed));
-
-struct ceph_mon_statfs_reply {
-	struct ceph_fsid fsid;
-	__le64 version;
-	struct ceph_statfs st;
-} __attribute__ ((packed));
-
-const char *ceph_pool_op_name(int op);
-
-struct ceph_mon_poolop {
-	struct ceph_mon_request_header monhdr;
-	struct ceph_fsid fsid;
-	__le32 pool;
-	__le32 op;
-	__le64 auid;
-	__le64 snapid;
-	__le32 name_len;
-} __attribute__ ((packed));
-
-struct ceph_mon_poolop_reply {
-	struct ceph_mon_request_header monhdr;
-	struct ceph_fsid fsid;
-	__le32 reply_code;
-	__le32 epoch;
-	char has_data;
-	char data[0];
-} __attribute__ ((packed));
-
-struct ceph_mon_unmanaged_snap {
-	__le64 snapid;
-} __attribute__ ((packed));
-
-struct ceph_osd_getmap {
-	struct ceph_mon_request_header monhdr;
-	struct ceph_fsid fsid;
-	__le32 start;
-} __attribute__ ((packed));
-
-struct ceph_mds_getmap {
-	struct ceph_mon_request_header monhdr;
-	struct ceph_fsid fsid;
-} __attribute__ ((packed));
-
-struct ceph_client_mount {
-	struct ceph_mon_request_header monhdr;
-} __attribute__ ((packed));
-
-struct ceph_mon_subscribe_item {
-	__le64 have_version;	__le64 have;
-	__u8 onetime;
-} __attribute__ ((packed));
-
-struct ceph_mon_subscribe_ack {
-	__le32 duration;         /* seconds */
-	struct ceph_fsid fsid;
-} __attribute__ ((packed));
-
-/*
- * mds states
- *   > 0 -> in
- *  <= 0 -> out
- */
-#define CEPH_MDS_STATE_DNE          0  /* down, does not exist. */
-#define CEPH_MDS_STATE_STOPPED     -1  /* down, once existed, but no subtrees.
-					  empty log. */
-#define CEPH_MDS_STATE_BOOT        -4  /* up, boot announcement. */
-#define CEPH_MDS_STATE_STANDBY     -5  /* up, idle.  waiting for assignment. */
-#define CEPH_MDS_STATE_CREATING    -6  /* up, creating MDS instance. */
-#define CEPH_MDS_STATE_STARTING    -7  /* up, starting previously stopped mds */
-#define CEPH_MDS_STATE_STANDBY_REPLAY -8 /* up, tailing active node's journal */
-
-#define CEPH_MDS_STATE_REPLAY       8  /* up, replaying journal. */
-#define CEPH_MDS_STATE_RESOLVE      9  /* up, disambiguating distributed
-					  operations (import, rename, etc.) */
-#define CEPH_MDS_STATE_RECONNECT    10 /* up, reconnect to clients */
-#define CEPH_MDS_STATE_REJOIN       11 /* up, rejoining distributed cache */
-#define CEPH_MDS_STATE_CLIENTREPLAY 12 /* up, replaying client operations */
-#define CEPH_MDS_STATE_ACTIVE       13 /* up, active */
-#define CEPH_MDS_STATE_STOPPING     14 /* up, but exporting metadata */
-
-extern const char *ceph_mds_state_name(int s);
-
-
-/*
- * metadata lock types.
- *  - these are bitmasks.. we can compose them
- *  - they also define the lock ordering by the MDS
- *  - a few of these are internal to the mds
- */
-#define CEPH_LOCK_DVERSION    1
-#define CEPH_LOCK_DN          2
-#define CEPH_LOCK_ISNAP       16
-#define CEPH_LOCK_IVERSION    32    /* mds internal */
-#define CEPH_LOCK_IFILE       64
-#define CEPH_LOCK_IAUTH       128
-#define CEPH_LOCK_ILINK       256
-#define CEPH_LOCK_IDFT        512   /* dir frag tree */
-#define CEPH_LOCK_INEST       1024  /* mds internal */
-#define CEPH_LOCK_IXATTR      2048
-#define CEPH_LOCK_IFLOCK      4096  /* advisory file locks */
-#define CEPH_LOCK_INO         8192  /* immutable inode bits; not a lock */
-
-/* client_session ops */
-enum {
-	CEPH_SESSION_REQUEST_OPEN,
-	CEPH_SESSION_OPEN,
-	CEPH_SESSION_REQUEST_CLOSE,
-	CEPH_SESSION_CLOSE,
-	CEPH_SESSION_REQUEST_RENEWCAPS,
-	CEPH_SESSION_RENEWCAPS,
-	CEPH_SESSION_STALE,
-	CEPH_SESSION_RECALL_STATE,
-};
-
-extern const char *ceph_session_op_name(int op);
-
-struct ceph_mds_session_head {
-	__le32 op;
-	__le64 seq;
-	struct ceph_timespec stamp;
-	__le32 max_caps, max_leases;
-} __attribute__ ((packed));
-
-/* client_request */
-/*
- * metadata ops.
- *  & 0x001000 -> write op
- *  & 0x010000 -> follow symlink (e.g. stat(), not lstat()).
- &  & 0x100000 -> use weird ino/path trace
- */
-#define CEPH_MDS_OP_WRITE        0x001000
-enum {
-	CEPH_MDS_OP_LOOKUP     = 0x00100,
-	CEPH_MDS_OP_GETATTR    = 0x00101,
-	CEPH_MDS_OP_LOOKUPHASH = 0x00102,
-	CEPH_MDS_OP_LOOKUPPARENT = 0x00103,
-
-	CEPH_MDS_OP_SETXATTR   = 0x01105,
-	CEPH_MDS_OP_RMXATTR    = 0x01106,
-	CEPH_MDS_OP_SETLAYOUT  = 0x01107,
-	CEPH_MDS_OP_SETATTR    = 0x01108,
-	CEPH_MDS_OP_SETFILELOCK= 0x01109,
-	CEPH_MDS_OP_GETFILELOCK= 0x00110,
-
-	CEPH_MDS_OP_MKNOD      = 0x01201,
-	CEPH_MDS_OP_LINK       = 0x01202,
-	CEPH_MDS_OP_UNLINK     = 0x01203,
-	CEPH_MDS_OP_RENAME     = 0x01204,
-	CEPH_MDS_OP_MKDIR      = 0x01220,
-	CEPH_MDS_OP_RMDIR      = 0x01221,
-	CEPH_MDS_OP_SYMLINK    = 0x01222,
-
-	CEPH_MDS_OP_CREATE     = 0x01301,
-	CEPH_MDS_OP_OPEN       = 0x00302,
-	CEPH_MDS_OP_READDIR    = 0x00305,
-
-	CEPH_MDS_OP_LOOKUPSNAP = 0x00400,
-	CEPH_MDS_OP_MKSNAP     = 0x01400,
-	CEPH_MDS_OP_RMSNAP     = 0x01401,
-	CEPH_MDS_OP_LSSNAP     = 0x00402,
-};
-
-extern const char *ceph_mds_op_name(int op);
-
-
-#define CEPH_SETATTR_MODE   1
-#define CEPH_SETATTR_UID    2
-#define CEPH_SETATTR_GID    4
-#define CEPH_SETATTR_MTIME  8
-#define CEPH_SETATTR_ATIME 16
-#define CEPH_SETATTR_SIZE  32
-#define CEPH_SETATTR_CTIME 64
-
-union ceph_mds_request_args {
-	struct {
-		__le32 mask;                 /* CEPH_CAP_* */
-	} __attribute__ ((packed)) getattr;
-	struct {
-		__le32 mode;
-		__le32 uid;
-		__le32 gid;
-		struct ceph_timespec mtime;
-		struct ceph_timespec atime;
-		__le64 size, old_size;       /* old_size needed by truncate */
-		__le32 mask;                 /* CEPH_SETATTR_* */
-	} __attribute__ ((packed)) setattr;
-	struct {
-		__le32 frag;                 /* which dir fragment */
-		__le32 max_entries;          /* how many dentries to grab */
-		__le32 max_bytes;
-	} __attribute__ ((packed)) readdir;
-	struct {
-		__le32 mode;
-		__le32 rdev;
-	} __attribute__ ((packed)) mknod;
-	struct {
-		__le32 mode;
-	} __attribute__ ((packed)) mkdir;
-	struct {
-		__le32 flags;
-		__le32 mode;
-		__le32 stripe_unit;          /* layout for newly created file */
-		__le32 stripe_count;         /* ... */
-		__le32 object_size;
-		__le32 file_replication;
-		__le32 preferred;
-	} __attribute__ ((packed)) open;
-	struct {
-		__le32 flags;
-	} __attribute__ ((packed)) setxattr;
-	struct {
-		struct ceph_file_layout layout;
-	} __attribute__ ((packed)) setlayout;
-	struct {
-		__u8 rule; /* currently fcntl or flock */
-		__u8 type; /* shared, exclusive, remove*/
-		__le64 pid; /* process id requesting the lock */
-		__le64 pid_namespace;
-		__le64 start; /* initial location to lock */
-		__le64 length; /* num bytes to lock from start */
-		__u8 wait; /* will caller wait for lock to become available? */
-	} __attribute__ ((packed)) filelock_change;
-} __attribute__ ((packed));
-
-#define CEPH_MDS_FLAG_REPLAY        1  /* this is a replayed op */
-#define CEPH_MDS_FLAG_WANT_DENTRY   2  /* want dentry in reply */
-
-struct ceph_mds_request_head {
-	__le64 oldest_client_tid;
-	__le32 mdsmap_epoch;           /* on client */
-	__le32 flags;                  /* CEPH_MDS_FLAG_* */
-	__u8 num_retry, num_fwd;       /* count retry, fwd attempts */
-	__le16 num_releases;           /* # include cap/lease release records */
-	__le32 op;                     /* mds op code */
-	__le32 caller_uid, caller_gid;
-	__le64 ino;                    /* use this ino for openc, mkdir, mknod,
-					  etc. (if replaying) */
-	union ceph_mds_request_args args;
-} __attribute__ ((packed));
-
-/* cap/lease release record */
-struct ceph_mds_request_release {
-	__le64 ino, cap_id;            /* ino and unique cap id */
-	__le32 caps, wanted;           /* new issued, wanted */
-	__le32 seq, issue_seq, mseq;
-	__le32 dname_seq;              /* if releasing a dentry lease, a */
-	__le32 dname_len;              /* string follows. */
-} __attribute__ ((packed));
-
-/* client reply */
-struct ceph_mds_reply_head {
-	__le32 op;
-	__le32 result;
-	__le32 mdsmap_epoch;
-	__u8 safe;                     /* true if committed to disk */
-	__u8 is_dentry, is_target;     /* true if dentry, target inode records
-					  are included with reply */
-} __attribute__ ((packed));
-
-/* one for each node split */
-struct ceph_frag_tree_split {
-	__le32 frag;                   /* this frag splits... */
-	__le32 by;                     /* ...by this many bits */
-} __attribute__ ((packed));
-
-struct ceph_frag_tree_head {
-	__le32 nsplits;                /* num ceph_frag_tree_split records */
-	struct ceph_frag_tree_split splits[];
-} __attribute__ ((packed));
-
-/* capability issue, for bundling with mds reply */
-struct ceph_mds_reply_cap {
-	__le32 caps, wanted;           /* caps issued, wanted */
-	__le64 cap_id;
-	__le32 seq, mseq;
-	__le64 realm;                  /* snap realm */
-	__u8 flags;                    /* CEPH_CAP_FLAG_* */
-} __attribute__ ((packed));
-
-#define CEPH_CAP_FLAG_AUTH  1          /* cap is issued by auth mds */
-
-/* inode record, for bundling with mds reply */
-struct ceph_mds_reply_inode {
-	__le64 ino;
-	__le64 snapid;
-	__le32 rdev;
-	__le64 version;                /* inode version */
-	__le64 xattr_version;          /* version for xattr blob */
-	struct ceph_mds_reply_cap cap; /* caps issued for this inode */
-	struct ceph_file_layout layout;
-	struct ceph_timespec ctime, mtime, atime;
-	__le32 time_warp_seq;
-	__le64 size, max_size, truncate_size;
-	__le32 truncate_seq;
-	__le32 mode, uid, gid;
-	__le32 nlink;
-	__le64 files, subdirs, rbytes, rfiles, rsubdirs;  /* dir stats */
-	struct ceph_timespec rctime;
-	struct ceph_frag_tree_head fragtree;  /* (must be at end of struct) */
-} __attribute__ ((packed));
-/* followed by frag array, then symlink string, then xattr blob */
-
-/* reply_lease follows dname, and reply_inode */
-struct ceph_mds_reply_lease {
-	__le16 mask;            /* lease type(s) */
-	__le32 duration_ms;     /* lease duration */
-	__le32 seq;
-} __attribute__ ((packed));
-
-struct ceph_mds_reply_dirfrag {
-	__le32 frag;            /* fragment */
-	__le32 auth;            /* auth mds, if this is a delegation point */
-	__le32 ndist;           /* number of mds' this is replicated on */
-	__le32 dist[];
-} __attribute__ ((packed));
-
-#define CEPH_LOCK_FCNTL    1
-#define CEPH_LOCK_FLOCK    2
-
-#define CEPH_LOCK_SHARED   1
-#define CEPH_LOCK_EXCL     2
-#define CEPH_LOCK_UNLOCK   4
-
-struct ceph_filelock {
-	__le64 start;/* file offset to start lock at */
-	__le64 length; /* num bytes to lock; 0 for all following start */
-	__le64 client; /* which client holds the lock */
-	__le64 pid; /* process id holding the lock on the client */
-	__le64 pid_namespace;
-	__u8 type; /* shared lock, exclusive lock, or unlock */
-} __attribute__ ((packed));
-
-
-/* file access modes */
-#define CEPH_FILE_MODE_PIN        0
-#define CEPH_FILE_MODE_RD         1
-#define CEPH_FILE_MODE_WR         2
-#define CEPH_FILE_MODE_RDWR       3  /* RD | WR */
-#define CEPH_FILE_MODE_LAZY       4  /* lazy io */
-#define CEPH_FILE_MODE_NUM        8  /* bc these are bit fields.. mostly */
-
-int ceph_flags_to_mode(int flags);
-
-
-/* capability bits */
-#define CEPH_CAP_PIN         1  /* no specific capabilities beyond the pin */
-
-/* generic cap bits */
-#define CEPH_CAP_GSHARED     1  /* client can reads */
-#define CEPH_CAP_GEXCL       2  /* client can read and update */
-#define CEPH_CAP_GCACHE      4  /* (file) client can cache reads */
-#define CEPH_CAP_GRD         8  /* (file) client can read */
-#define CEPH_CAP_GWR        16  /* (file) client can write */
-#define CEPH_CAP_GBUFFER    32  /* (file) client can buffer writes */
-#define CEPH_CAP_GWREXTEND  64  /* (file) client can extend EOF */
-#define CEPH_CAP_GLAZYIO   128  /* (file) client can perform lazy io */
-
-/* per-lock shift */
-#define CEPH_CAP_SAUTH      2
-#define CEPH_CAP_SLINK      4
-#define CEPH_CAP_SXATTR     6
-#define CEPH_CAP_SFILE      8
-#define CEPH_CAP_SFLOCK    20 
-
-#define CEPH_CAP_BITS       22
-
-/* composed values */
-#define CEPH_CAP_AUTH_SHARED  (CEPH_CAP_GSHARED  << CEPH_CAP_SAUTH)
-#define CEPH_CAP_AUTH_EXCL     (CEPH_CAP_GEXCL     << CEPH_CAP_SAUTH)
-#define CEPH_CAP_LINK_SHARED  (CEPH_CAP_GSHARED  << CEPH_CAP_SLINK)
-#define CEPH_CAP_LINK_EXCL     (CEPH_CAP_GEXCL     << CEPH_CAP_SLINK)
-#define CEPH_CAP_XATTR_SHARED (CEPH_CAP_GSHARED  << CEPH_CAP_SXATTR)
-#define CEPH_CAP_XATTR_EXCL    (CEPH_CAP_GEXCL     << CEPH_CAP_SXATTR)
-#define CEPH_CAP_FILE(x)    (x << CEPH_CAP_SFILE)
-#define CEPH_CAP_FILE_SHARED   (CEPH_CAP_GSHARED   << CEPH_CAP_SFILE)
-#define CEPH_CAP_FILE_EXCL     (CEPH_CAP_GEXCL     << CEPH_CAP_SFILE)
-#define CEPH_CAP_FILE_CACHE    (CEPH_CAP_GCACHE    << CEPH_CAP_SFILE)
-#define CEPH_CAP_FILE_RD       (CEPH_CAP_GRD       << CEPH_CAP_SFILE)
-#define CEPH_CAP_FILE_WR       (CEPH_CAP_GWR       << CEPH_CAP_SFILE)
-#define CEPH_CAP_FILE_BUFFER   (CEPH_CAP_GBUFFER   << CEPH_CAP_SFILE)
-#define CEPH_CAP_FILE_WREXTEND (CEPH_CAP_GWREXTEND << CEPH_CAP_SFILE)
-#define CEPH_CAP_FILE_LAZYIO   (CEPH_CAP_GLAZYIO   << CEPH_CAP_SFILE)
-#define CEPH_CAP_FLOCK_SHARED  (CEPH_CAP_GSHARED   << CEPH_CAP_SFLOCK)
-#define CEPH_CAP_FLOCK_EXCL    (CEPH_CAP_GEXCL     << CEPH_CAP_SFLOCK)
-
-
-/* cap masks (for getattr) */
-#define CEPH_STAT_CAP_INODE    CEPH_CAP_PIN
-#define CEPH_STAT_CAP_TYPE     CEPH_CAP_PIN  /* mode >> 12 */
-#define CEPH_STAT_CAP_SYMLINK  CEPH_CAP_PIN
-#define CEPH_STAT_CAP_UID      CEPH_CAP_AUTH_SHARED
-#define CEPH_STAT_CAP_GID      CEPH_CAP_AUTH_SHARED
-#define CEPH_STAT_CAP_MODE     CEPH_CAP_AUTH_SHARED
-#define CEPH_STAT_CAP_NLINK    CEPH_CAP_LINK_SHARED
-#define CEPH_STAT_CAP_LAYOUT   CEPH_CAP_FILE_SHARED
-#define CEPH_STAT_CAP_MTIME    CEPH_CAP_FILE_SHARED
-#define CEPH_STAT_CAP_SIZE     CEPH_CAP_FILE_SHARED
-#define CEPH_STAT_CAP_ATIME    CEPH_CAP_FILE_SHARED  /* fixme */
-#define CEPH_STAT_CAP_XATTR    CEPH_CAP_XATTR_SHARED
-#define CEPH_STAT_CAP_INODE_ALL (CEPH_CAP_PIN |			\
-				 CEPH_CAP_AUTH_SHARED |	\
-				 CEPH_CAP_LINK_SHARED |	\
-				 CEPH_CAP_FILE_SHARED |	\
-				 CEPH_CAP_XATTR_SHARED)
-
-#define CEPH_CAP_ANY_SHARED (CEPH_CAP_AUTH_SHARED |			\
-			      CEPH_CAP_LINK_SHARED |			\
-			      CEPH_CAP_XATTR_SHARED |			\
-			      CEPH_CAP_FILE_SHARED)
-#define CEPH_CAP_ANY_RD   (CEPH_CAP_ANY_SHARED | CEPH_CAP_FILE_RD |	\
-			   CEPH_CAP_FILE_CACHE)
-
-#define CEPH_CAP_ANY_EXCL (CEPH_CAP_AUTH_EXCL |		\
-			   CEPH_CAP_LINK_EXCL |		\
-			   CEPH_CAP_XATTR_EXCL |	\
-			   CEPH_CAP_FILE_EXCL)
-#define CEPH_CAP_ANY_FILE_WR (CEPH_CAP_FILE_WR | CEPH_CAP_FILE_BUFFER |	\
-			      CEPH_CAP_FILE_EXCL)
-#define CEPH_CAP_ANY_WR   (CEPH_CAP_ANY_EXCL | CEPH_CAP_ANY_FILE_WR)
-#define CEPH_CAP_ANY      (CEPH_CAP_ANY_RD | CEPH_CAP_ANY_EXCL | \
-			   CEPH_CAP_ANY_FILE_WR | CEPH_CAP_FILE_LAZYIO | \
-			   CEPH_CAP_PIN)
-
-#define CEPH_CAP_LOCKS (CEPH_LOCK_IFILE | CEPH_LOCK_IAUTH | CEPH_LOCK_ILINK | \
-			CEPH_LOCK_IXATTR)
-
-int ceph_caps_for_mode(int mode);
-
-enum {
-	CEPH_CAP_OP_GRANT,         /* mds->client grant */
-	CEPH_CAP_OP_REVOKE,        /* mds->client revoke */
-	CEPH_CAP_OP_TRUNC,         /* mds->client trunc notify */
-	CEPH_CAP_OP_EXPORT,        /* mds has exported the cap */
-	CEPH_CAP_OP_IMPORT,        /* mds has imported the cap */
-	CEPH_CAP_OP_UPDATE,        /* client->mds update */
-	CEPH_CAP_OP_DROP,          /* client->mds drop cap bits */
-	CEPH_CAP_OP_FLUSH,         /* client->mds cap writeback */
-	CEPH_CAP_OP_FLUSH_ACK,     /* mds->client flushed */
-	CEPH_CAP_OP_FLUSHSNAP,     /* client->mds flush snapped metadata */
-	CEPH_CAP_OP_FLUSHSNAP_ACK, /* mds->client flushed snapped metadata */
-	CEPH_CAP_OP_RELEASE,       /* client->mds release (clean) cap */
-	CEPH_CAP_OP_RENEW,         /* client->mds renewal request */
-};
-
-extern const char *ceph_cap_op_name(int op);
-
-/*
- * caps message, used for capability callbacks, acks, requests, etc.
- */
-struct ceph_mds_caps {
-	__le32 op;                  /* CEPH_CAP_OP_* */
-	__le64 ino, realm;
-	__le64 cap_id;
-	__le32 seq, issue_seq;
-	__le32 caps, wanted, dirty; /* latest issued/wanted/dirty */
-	__le32 migrate_seq;
-	__le64 snap_follows;
-	__le32 snap_trace_len;
-
-	/* authlock */
-	__le32 uid, gid, mode;
-
-	/* linklock */
-	__le32 nlink;
-
-	/* xattrlock */
-	__le32 xattr_len;
-	__le64 xattr_version;
-
-	/* filelock */
-	__le64 size, max_size, truncate_size;
-	__le32 truncate_seq;
-	struct ceph_timespec mtime, atime, ctime;
-	struct ceph_file_layout layout;
-	__le32 time_warp_seq;
-} __attribute__ ((packed));
-
-/* cap release msg head */
-struct ceph_mds_cap_release {
-	__le32 num;                /* number of cap_items that follow */
-} __attribute__ ((packed));
-
-struct ceph_mds_cap_item {
-	__le64 ino;
-	__le64 cap_id;
-	__le32 migrate_seq, seq;
-} __attribute__ ((packed));
-
-#define CEPH_MDS_LEASE_REVOKE           1  /*    mds  -> client */
-#define CEPH_MDS_LEASE_RELEASE          2  /* client  -> mds    */
-#define CEPH_MDS_LEASE_RENEW            3  /* client <-> mds    */
-#define CEPH_MDS_LEASE_REVOKE_ACK       4  /* client  -> mds    */
-
-extern const char *ceph_lease_op_name(int o);
-
-/* lease msg header */
-struct ceph_mds_lease {
-	__u8 action;            /* CEPH_MDS_LEASE_* */
-	__le16 mask;            /* which lease */
-	__le64 ino;
-	__le64 first, last;     /* snap range */
-	__le32 seq;
-	__le32 duration_ms;     /* duration of renewal */
-} __attribute__ ((packed));
-/* followed by a __le32+string for dname */
-
-/* client reconnect */
-struct ceph_mds_cap_reconnect {
-	__le64 cap_id;
-	__le32 wanted;
-	__le32 issued;
-	__le64 snaprealm;
-	__le64 pathbase;        /* base ino for our path to this ino */
-	__le32 flock_len;       /* size of flock state blob, if any */
-} __attribute__ ((packed));
-/* followed by flock blob */
-
-struct ceph_mds_cap_reconnect_v1 {
-	__le64 cap_id;
-	__le32 wanted;
-	__le32 issued;
-	__le64 size;
-	struct ceph_timespec mtime, atime;
-	__le64 snaprealm;
-	__le64 pathbase;        /* base ino for our path to this ino */
-} __attribute__ ((packed));
-
-struct ceph_mds_snaprealm_reconnect {
-	__le64 ino;     /* snap realm base */
-	__le64 seq;     /* snap seq for this snap realm */
-	__le64 parent;  /* parent realm */
-} __attribute__ ((packed));
-
-/*
- * snaps
- */
-enum {
-	CEPH_SNAP_OP_UPDATE,  /* CREATE or DESTROY */
-	CEPH_SNAP_OP_CREATE,
-	CEPH_SNAP_OP_DESTROY,
-	CEPH_SNAP_OP_SPLIT,
-};
-
-extern const char *ceph_snap_op_name(int o);
-
-/* snap msg header */
-struct ceph_mds_snap_head {
-	__le32 op;                /* CEPH_SNAP_OP_* */
-	__le64 split;             /* ino to split off, if any */
-	__le32 num_split_inos;    /* # inos belonging to new child realm */
-	__le32 num_split_realms;  /* # child realms udner new child realm */
-	__le32 trace_len;         /* size of snap trace blob */
-} __attribute__ ((packed));
-/* followed by split ino list, then split realms, then the trace blob */
-
-/*
- * encode info about a snaprealm, as viewed by a client
- */
-struct ceph_mds_snap_realm {
-	__le64 ino;           /* ino */
-	__le64 created;       /* snap: when created */
-	__le64 parent;        /* ino: parent realm */
-	__le64 parent_since;  /* snap: same parent since */
-	__le64 seq;           /* snap: version */
-	__le32 num_snaps;
-	__le32 num_prior_parent_snaps;
-} __attribute__ ((packed));
-/* followed by my snap list, then prior parent snap list */
-
-#endif
diff --git a/fs/ceph/ceph_hash.c b/fs/ceph/ceph_hash.c
deleted file mode 100644
index bd570015d147..000000000000
--- a/fs/ceph/ceph_hash.c
+++ /dev/null
@@ -1,118 +0,0 @@
-
-#include "types.h"
-
-/*
- * Robert Jenkin's hash function.
- * http://burtleburtle.net/bob/hash/evahash.html
- * This is in the public domain.
- */
-#define mix(a, b, c)						\
-	do {							\
-		a = a - b;  a = a - c;  a = a ^ (c >> 13);	\
-		b = b - c;  b = b - a;  b = b ^ (a << 8);	\
-		c = c - a;  c = c - b;  c = c ^ (b >> 13);	\
-		a = a - b;  a = a - c;  a = a ^ (c >> 12);	\
-		b = b - c;  b = b - a;  b = b ^ (a << 16);	\
-		c = c - a;  c = c - b;  c = c ^ (b >> 5);	\
-		a = a - b;  a = a - c;  a = a ^ (c >> 3);	\
-		b = b - c;  b = b - a;  b = b ^ (a << 10);	\
-		c = c - a;  c = c - b;  c = c ^ (b >> 15);	\
-	} while (0)
-
-unsigned ceph_str_hash_rjenkins(const char *str, unsigned length)
-{
-	const unsigned char *k = (const unsigned char *)str;
-	__u32 a, b, c;  /* the internal state */
-	__u32 len;      /* how many key bytes still need mixing */
-
-	/* Set up the internal state */
-	len = length;
-	a = 0x9e3779b9;      /* the golden ratio; an arbitrary value */
-	b = a;
-	c = 0;               /* variable initialization of internal state */
-
-	/* handle most of the key */
-	while (len >= 12) {
-		a = a + (k[0] + ((__u32)k[1] << 8) + ((__u32)k[2] << 16) +
-			 ((__u32)k[3] << 24));
-		b = b + (k[4] + ((__u32)k[5] << 8) + ((__u32)k[6] << 16) +
-			 ((__u32)k[7] << 24));
-		c = c + (k[8] + ((__u32)k[9] << 8) + ((__u32)k[10] << 16) +
-			 ((__u32)k[11] << 24));
-		mix(a, b, c);
-		k = k + 12;
-		len = len - 12;
-	}
-
-	/* handle the last 11 bytes */
-	c = c + length;
-	switch (len) {            /* all the case statements fall through */
-	case 11:
-		c = c + ((__u32)k[10] << 24);
-	case 10:
-		c = c + ((__u32)k[9] << 16);
-	case 9:
-		c = c + ((__u32)k[8] << 8);
-		/* the first byte of c is reserved for the length */
-	case 8:
-		b = b + ((__u32)k[7] << 24);
-	case 7:
-		b = b + ((__u32)k[6] << 16);
-	case 6:
-		b = b + ((__u32)k[5] << 8);
-	case 5:
-		b = b + k[4];
-	case 4:
-		a = a + ((__u32)k[3] << 24);
-	case 3:
-		a = a + ((__u32)k[2] << 16);
-	case 2:
-		a = a + ((__u32)k[1] << 8);
-	case 1:
-		a = a + k[0];
-		/* case 0: nothing left to add */
-	}
-	mix(a, b, c);
-
-	return c;
-}
-
-/*
- * linux dcache hash
- */
-unsigned ceph_str_hash_linux(const char *str, unsigned length)
-{
-	unsigned long hash = 0;
-	unsigned char c;
-
-	while (length--) {
-		c = *str++;
-		hash = (hash + (c << 4) + (c >> 4)) * 11;
-	}
-	return hash;
-}
-
-
-unsigned ceph_str_hash(int type, const char *s, unsigned len)
-{
-	switch (type) {
-	case CEPH_STR_HASH_LINUX:
-		return ceph_str_hash_linux(s, len);
-	case CEPH_STR_HASH_RJENKINS:
-		return ceph_str_hash_rjenkins(s, len);
-	default:
-		return -1;
-	}
-}
-
-const char *ceph_str_hash_name(int type)
-{
-	switch (type) {
-	case CEPH_STR_HASH_LINUX:
-		return "linux";
-	case CEPH_STR_HASH_RJENKINS:
-		return "rjenkins";
-	default:
-		return "unknown";
-	}
-}
diff --git a/fs/ceph/ceph_hash.h b/fs/ceph/ceph_hash.h
deleted file mode 100644
index d099c3f90236..000000000000
--- a/fs/ceph/ceph_hash.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#ifndef FS_CEPH_HASH_H
-#define FS_CEPH_HASH_H
-
-#define CEPH_STR_HASH_LINUX      0x1  /* linux dcache hash */
-#define CEPH_STR_HASH_RJENKINS   0x2  /* robert jenkins' */
-
-extern unsigned ceph_str_hash_linux(const char *s, unsigned len);
-extern unsigned ceph_str_hash_rjenkins(const char *s, unsigned len);
-
-extern unsigned ceph_str_hash(int type, const char *s, unsigned len);
-extern const char *ceph_str_hash_name(int type);
-
-#endif
diff --git a/fs/ceph/ceph_strings.c b/fs/ceph/ceph_strings.c
deleted file mode 100644
index c6179d3a26a2..000000000000
--- a/fs/ceph/ceph_strings.c
+++ /dev/null
@@ -1,193 +0,0 @@
-/*
- * Ceph string constants
- */
-#include "types.h"
-
-const char *ceph_entity_type_name(int type)
-{
-	switch (type) {
-	case CEPH_ENTITY_TYPE_MDS: return "mds";
-	case CEPH_ENTITY_TYPE_OSD: return "osd";
-	case CEPH_ENTITY_TYPE_MON: return "mon";
-	case CEPH_ENTITY_TYPE_CLIENT: return "client";
-	case CEPH_ENTITY_TYPE_AUTH: return "auth";
-	default: return "unknown";
-	}
-}
-
-const char *ceph_osd_op_name(int op)
-{
-	switch (op) {
-	case CEPH_OSD_OP_READ: return "read";
-	case CEPH_OSD_OP_STAT: return "stat";
-
-	case CEPH_OSD_OP_MASKTRUNC: return "masktrunc";
-
-	case CEPH_OSD_OP_WRITE: return "write";
-	case CEPH_OSD_OP_DELETE: return "delete";
-	case CEPH_OSD_OP_TRUNCATE: return "truncate";
-	case CEPH_OSD_OP_ZERO: return "zero";
-	case CEPH_OSD_OP_WRITEFULL: return "writefull";
-	case CEPH_OSD_OP_ROLLBACK: return "rollback";
-
-	case CEPH_OSD_OP_APPEND: return "append";
-	case CEPH_OSD_OP_STARTSYNC: return "startsync";
-	case CEPH_OSD_OP_SETTRUNC: return "settrunc";
-	case CEPH_OSD_OP_TRIMTRUNC: return "trimtrunc";
-
-	case CEPH_OSD_OP_TMAPUP: return "tmapup";
-	case CEPH_OSD_OP_TMAPGET: return "tmapget";
-	case CEPH_OSD_OP_TMAPPUT: return "tmapput";
-
-	case CEPH_OSD_OP_GETXATTR: return "getxattr";
-	case CEPH_OSD_OP_GETXATTRS: return "getxattrs";
-	case CEPH_OSD_OP_SETXATTR: return "setxattr";
-	case CEPH_OSD_OP_SETXATTRS: return "setxattrs";
-	case CEPH_OSD_OP_RESETXATTRS: return "resetxattrs";
-	case CEPH_OSD_OP_RMXATTR: return "rmxattr";
-	case CEPH_OSD_OP_CMPXATTR: return "cmpxattr";
-
-	case CEPH_OSD_OP_PULL: return "pull";
-	case CEPH_OSD_OP_PUSH: return "push";
-	case CEPH_OSD_OP_BALANCEREADS: return "balance-reads";
-	case CEPH_OSD_OP_UNBALANCEREADS: return "unbalance-reads";
-	case CEPH_OSD_OP_SCRUB: return "scrub";
-
-	case CEPH_OSD_OP_WRLOCK: return "wrlock";
-	case CEPH_OSD_OP_WRUNLOCK: return "wrunlock";
-	case CEPH_OSD_OP_RDLOCK: return "rdlock";
-	case CEPH_OSD_OP_RDUNLOCK: return "rdunlock";
-	case CEPH_OSD_OP_UPLOCK: return "uplock";
-	case CEPH_OSD_OP_DNLOCK: return "dnlock";
-
-	case CEPH_OSD_OP_CALL: return "call";
-
-	case CEPH_OSD_OP_PGLS: return "pgls";
-	}
-	return "???";
-}
-
-const char *ceph_mds_state_name(int s)
-{
-	switch (s) {
-		/* down and out */
-	case CEPH_MDS_STATE_DNE:        return "down:dne";
-	case CEPH_MDS_STATE_STOPPED:    return "down:stopped";
-		/* up and out */
-	case CEPH_MDS_STATE_BOOT:       return "up:boot";
-	case CEPH_MDS_STATE_STANDBY:    return "up:standby";
-	case CEPH_MDS_STATE_STANDBY_REPLAY:    return "up:standby-replay";
-	case CEPH_MDS_STATE_CREATING:   return "up:creating";
-	case CEPH_MDS_STATE_STARTING:   return "up:starting";
-		/* up and in */
-	case CEPH_MDS_STATE_REPLAY:     return "up:replay";
-	case CEPH_MDS_STATE_RESOLVE:    return "up:resolve";
-	case CEPH_MDS_STATE_RECONNECT:  return "up:reconnect";
-	case CEPH_MDS_STATE_REJOIN:     return "up:rejoin";
-	case CEPH_MDS_STATE_CLIENTREPLAY: return "up:clientreplay";
-	case CEPH_MDS_STATE_ACTIVE:     return "up:active";
-	case CEPH_MDS_STATE_STOPPING:   return "up:stopping";
-	}
-	return "???";
-}
-
-const char *ceph_session_op_name(int op)
-{
-	switch (op) {
-	case CEPH_SESSION_REQUEST_OPEN: return "request_open";
-	case CEPH_SESSION_OPEN: return "open";
-	case CEPH_SESSION_REQUEST_CLOSE: return "request_close";
-	case CEPH_SESSION_CLOSE: return "close";
-	case CEPH_SESSION_REQUEST_RENEWCAPS: return "request_renewcaps";
-	case CEPH_SESSION_RENEWCAPS: return "renewcaps";
-	case CEPH_SESSION_STALE: return "stale";
-	case CEPH_SESSION_RECALL_STATE: return "recall_state";
-	}
-	return "???";
-}
-
-const char *ceph_mds_op_name(int op)
-{
-	switch (op) {
-	case CEPH_MDS_OP_LOOKUP:  return "lookup";
-	case CEPH_MDS_OP_LOOKUPHASH:  return "lookuphash";
-	case CEPH_MDS_OP_LOOKUPPARENT:  return "lookupparent";
-	case CEPH_MDS_OP_GETATTR:  return "getattr";
-	case CEPH_MDS_OP_SETXATTR: return "setxattr";
-	case CEPH_MDS_OP_SETATTR: return "setattr";
-	case CEPH_MDS_OP_RMXATTR: return "rmxattr";
-	case CEPH_MDS_OP_READDIR: return "readdir";
-	case CEPH_MDS_OP_MKNOD: return "mknod";
-	case CEPH_MDS_OP_LINK: return "link";
-	case CEPH_MDS_OP_UNLINK: return "unlink";
-	case CEPH_MDS_OP_RENAME: return "rename";
-	case CEPH_MDS_OP_MKDIR: return "mkdir";
-	case CEPH_MDS_OP_RMDIR: return "rmdir";
-	case CEPH_MDS_OP_SYMLINK: return "symlink";
-	case CEPH_MDS_OP_CREATE: return "create";
-	case CEPH_MDS_OP_OPEN: return "open";
-	case CEPH_MDS_OP_LOOKUPSNAP: return "lookupsnap";
-	case CEPH_MDS_OP_LSSNAP: return "lssnap";
-	case CEPH_MDS_OP_MKSNAP: return "mksnap";
-	case CEPH_MDS_OP_RMSNAP: return "rmsnap";
-	case CEPH_MDS_OP_SETFILELOCK: return "setfilelock";
-	case CEPH_MDS_OP_GETFILELOCK: return "getfilelock";
-	}
-	return "???";
-}
-
-const char *ceph_cap_op_name(int op)
-{
-	switch (op) {
-	case CEPH_CAP_OP_GRANT: return "grant";
-	case CEPH_CAP_OP_REVOKE: return "revoke";
-	case CEPH_CAP_OP_TRUNC: return "trunc";
-	case CEPH_CAP_OP_EXPORT: return "export";
-	case CEPH_CAP_OP_IMPORT: return "import";
-	case CEPH_CAP_OP_UPDATE: return "update";
-	case CEPH_CAP_OP_DROP: return "drop";
-	case CEPH_CAP_OP_FLUSH: return "flush";
-	case CEPH_CAP_OP_FLUSH_ACK: return "flush_ack";
-	case CEPH_CAP_OP_FLUSHSNAP: return "flushsnap";
-	case CEPH_CAP_OP_FLUSHSNAP_ACK: return "flushsnap_ack";
-	case CEPH_CAP_OP_RELEASE: return "release";
-	case CEPH_CAP_OP_RENEW: return "renew";
-	}
-	return "???";
-}
-
-const char *ceph_lease_op_name(int o)
-{
-	switch (o) {
-	case CEPH_MDS_LEASE_REVOKE: return "revoke";
-	case CEPH_MDS_LEASE_RELEASE: return "release";
-	case CEPH_MDS_LEASE_RENEW: return "renew";
-	case CEPH_MDS_LEASE_REVOKE_ACK: return "revoke_ack";
-	}
-	return "???";
-}
-
-const char *ceph_snap_op_name(int o)
-{
-	switch (o) {
-	case CEPH_SNAP_OP_UPDATE: return "update";
-	case CEPH_SNAP_OP_CREATE: return "create";
-	case CEPH_SNAP_OP_DESTROY: return "destroy";
-	case CEPH_SNAP_OP_SPLIT: return "split";
-	}
-	return "???";
-}
-
-const char *ceph_pool_op_name(int op)
-{
-	switch (op) {
-	case POOL_OP_CREATE: return "create";
-	case POOL_OP_DELETE: return "delete";
-	case POOL_OP_AUID_CHANGE: return "auid change";
-	case POOL_OP_CREATE_SNAP: return "create snap";
-	case POOL_OP_DELETE_SNAP: return "delete snap";
-	case POOL_OP_CREATE_UNMANAGED_SNAP: return "create unmanaged snap";
-	case POOL_OP_DELETE_UNMANAGED_SNAP: return "delete unmanaged snap";
-	}
-	return "???";
-}
diff --git a/fs/ceph/crush/crush.c b/fs/ceph/crush/crush.c
deleted file mode 100644
index fabd302e5779..000000000000
--- a/fs/ceph/crush/crush.c
+++ /dev/null
@@ -1,151 +0,0 @@
-
-#ifdef __KERNEL__
-# include <linux/slab.h>
-#else
-# include <stdlib.h>
-# include <assert.h>
-# define kfree(x) do { if (x) free(x); } while (0)
-# define BUG_ON(x) assert(!(x))
-#endif
-
-#include "crush.h"
-
-const char *crush_bucket_alg_name(int alg)
-{
-	switch (alg) {
-	case CRUSH_BUCKET_UNIFORM: return "uniform";
-	case CRUSH_BUCKET_LIST: return "list";
-	case CRUSH_BUCKET_TREE: return "tree";
-	case CRUSH_BUCKET_STRAW: return "straw";
-	default: return "unknown";
-	}
-}
-
-/**
- * crush_get_bucket_item_weight - Get weight of an item in given bucket
- * @b: bucket pointer
- * @p: item index in bucket
- */
-int crush_get_bucket_item_weight(struct crush_bucket *b, int p)
-{
-	if (p >= b->size)
-		return 0;
-
-	switch (b->alg) {
-	case CRUSH_BUCKET_UNIFORM:
-		return ((struct crush_bucket_uniform *)b)->item_weight;
-	case CRUSH_BUCKET_LIST:
-		return ((struct crush_bucket_list *)b)->item_weights[p];
-	case CRUSH_BUCKET_TREE:
-		if (p & 1)
-			return ((struct crush_bucket_tree *)b)->node_weights[p];
-		return 0;
-	case CRUSH_BUCKET_STRAW:
-		return ((struct crush_bucket_straw *)b)->item_weights[p];
-	}
-	return 0;
-}
-
-/**
- * crush_calc_parents - Calculate parent vectors for the given crush map.
- * @map: crush_map pointer
- */
-void crush_calc_parents(struct crush_map *map)
-{
-	int i, b, c;
-
-	for (b = 0; b < map->max_buckets; b++) {
-		if (map->buckets[b] == NULL)
-			continue;
-		for (i = 0; i < map->buckets[b]->size; i++) {
-			c = map->buckets[b]->items[i];
-			BUG_ON(c >= map->max_devices ||
-			       c < -map->max_buckets);
-			if (c >= 0)
-				map->device_parents[c] = map->buckets[b]->id;
-			else
-				map->bucket_parents[-1-c] = map->buckets[b]->id;
-		}
-	}
-}
-
-void crush_destroy_bucket_uniform(struct crush_bucket_uniform *b)
-{
-	kfree(b->h.perm);
-	kfree(b->h.items);
-	kfree(b);
-}
-
-void crush_destroy_bucket_list(struct crush_bucket_list *b)
-{
-	kfree(b->item_weights);
-	kfree(b->sum_weights);
-	kfree(b->h.perm);
-	kfree(b->h.items);
-	kfree(b);
-}
-
-void crush_destroy_bucket_tree(struct crush_bucket_tree *b)
-{
-	kfree(b->node_weights);
-	kfree(b);
-}
-
-void crush_destroy_bucket_straw(struct crush_bucket_straw *b)
-{
-	kfree(b->straws);
-	kfree(b->item_weights);
-	kfree(b->h.perm);
-	kfree(b->h.items);
-	kfree(b);
-}
-
-void crush_destroy_bucket(struct crush_bucket *b)
-{
-	switch (b->alg) {
-	case CRUSH_BUCKET_UNIFORM:
-		crush_destroy_bucket_uniform((struct crush_bucket_uniform *)b);
-		break;
-	case CRUSH_BUCKET_LIST:
-		crush_destroy_bucket_list((struct crush_bucket_list *)b);
-		break;
-	case CRUSH_BUCKET_TREE:
-		crush_destroy_bucket_tree((struct crush_bucket_tree *)b);
-		break;
-	case CRUSH_BUCKET_STRAW:
-		crush_destroy_bucket_straw((struct crush_bucket_straw *)b);
-		break;
-	}
-}
-
-/**
- * crush_destroy - Destroy a crush_map
- * @map: crush_map pointer
- */
-void crush_destroy(struct crush_map *map)
-{
-	int b;
-
-	/* buckets */
-	if (map->buckets) {
-		for (b = 0; b < map->max_buckets; b++) {
-			if (map->buckets[b] == NULL)
-				continue;
-			crush_destroy_bucket(map->buckets[b]);
-		}
-		kfree(map->buckets);
-	}
-
-	/* rules */
-	if (map->rules) {
-		for (b = 0; b < map->max_rules; b++)
-			kfree(map->rules[b]);
-		kfree(map->rules);
-	}
-
-	kfree(map->bucket_parents);
-	kfree(map->device_parents);
-	kfree(map);
-}
-
-
diff --git a/fs/ceph/crush/crush.h b/fs/ceph/crush/crush.h
deleted file mode 100644
index 97e435b191f4..000000000000
--- a/fs/ceph/crush/crush.h
+++ /dev/null
@@ -1,180 +0,0 @@
-#ifndef CEPH_CRUSH_CRUSH_H
-#define CEPH_CRUSH_CRUSH_H
-
-#include <linux/types.h>
-
-/*
- * CRUSH is a pseudo-random data distribution algorithm that
- * efficiently distributes input values (typically, data objects)
- * across a heterogeneous, structured storage cluster.
- *
- * The algorithm was originally described in detail in this paper
- * (although the algorithm has evolved somewhat since then):
- *
- *     http://www.ssrc.ucsc.edu/Papers/weil-sc06.pdf
- *
- * LGPL2
- */
-
-
-#define CRUSH_MAGIC 0x00010000ul   /* for detecting algorithm revisions */
-
-
-#define CRUSH_MAX_DEPTH 10  /* max crush hierarchy depth */
-#define CRUSH_MAX_SET   10  /* max size of a mapping result */
-
-
-/*
- * CRUSH uses user-defined "rules" to describe how inputs should be
- * mapped to devices.  A rule consists of sequence of steps to perform
- * to generate the set of output devices.
- */
-struct crush_rule_step {
-	__u32 op;
-	__s32 arg1;
-	__s32 arg2;
-};
-
-/* step op codes */
-enum {
-	CRUSH_RULE_NOOP = 0,
-	CRUSH_RULE_TAKE = 1,          /* arg1 = value to start with */
-	CRUSH_RULE_CHOOSE_FIRSTN = 2, /* arg1 = num items to pick */
-				      /* arg2 = type */
-	CRUSH_RULE_CHOOSE_INDEP = 3,  /* same */
-	CRUSH_RULE_EMIT = 4,          /* no args */
-	CRUSH_RULE_CHOOSE_LEAF_FIRSTN = 6,
-	CRUSH_RULE_CHOOSE_LEAF_INDEP = 7,
-};
-
-/*
- * for specifying choose num (arg1) relative to the max parameter
- * passed to do_rule
- */
-#define CRUSH_CHOOSE_N            0
-#define CRUSH_CHOOSE_N_MINUS(x)   (-(x))
-
-/*
- * The rule mask is used to describe what the rule is intended for.
- * Given a ruleset and size of output set, we search through the
- * rule list for a matching rule_mask.
- */
-struct crush_rule_mask {
-	__u8 ruleset;
-	__u8 type;
-	__u8 min_size;
-	__u8 max_size;
-};
-
-struct crush_rule {
-	__u32 len;
-	struct crush_rule_mask mask;
-	struct crush_rule_step steps[0];
-};
-
-#define crush_rule_size(len) (sizeof(struct crush_rule) + \
-			      (len)*sizeof(struct crush_rule_step))
-
-
-
-/*
- * A bucket is a named container of other items (either devices or
- * other buckets).  Items within a bucket are chosen using one of a
- * few different algorithms.  The table summarizes how the speed of
- * each option measures up against mapping stability when items are
- * added or removed.
- *
- *  Bucket Alg     Speed       Additions    Removals
- *  ------------------------------------------------
- *  uniform         O(1)       poor         poor
- *  list            O(n)       optimal      poor
- *  tree            O(log n)   good         good
- *  straw           O(n)       optimal      optimal
- */
-enum {
-	CRUSH_BUCKET_UNIFORM = 1,
-	CRUSH_BUCKET_LIST = 2,
-	CRUSH_BUCKET_TREE = 3,
-	CRUSH_BUCKET_STRAW = 4
-};
-extern const char *crush_bucket_alg_name(int alg);
-
-struct crush_bucket {
-	__s32 id;        /* this'll be negative */
-	__u16 type;      /* non-zero; type=0 is reserved for devices */
-	__u8 alg;        /* one of CRUSH_BUCKET_* */
-	__u8 hash;       /* which hash function to use, CRUSH_HASH_* */
-	__u32 weight;    /* 16-bit fixed point */
-	__u32 size;      /* num items */
-	__s32 *items;
-
-	/*
-	 * cached random permutation: used for uniform bucket and for
-	 * the linear search fallback for the other bucket types.
-	 */
-	__u32 perm_x;  /* @x for which *perm is defined */
-	__u32 perm_n;  /* num elements of *perm that are permuted/defined */
-	__u32 *perm;
-};
-
-struct crush_bucket_uniform {
-	struct crush_bucket h;
-	__u32 item_weight;  /* 16-bit fixed point; all items equally weighted */
-};
-
-struct crush_bucket_list {
-	struct crush_bucket h;
-	__u32 *item_weights;  /* 16-bit fixed point */
-	__u32 *sum_weights;   /* 16-bit fixed point.  element i is sum
-				 of weights 0..i, inclusive */
-};
-
-struct crush_bucket_tree {
-	struct crush_bucket h;  /* note: h.size is _tree_ size, not number of
-				   actual items */
-	__u8 num_nodes;
-	__u32 *node_weights;
-};
-
-struct crush_bucket_straw {
-	struct crush_bucket h;
-	__u32 *item_weights;   /* 16-bit fixed point */
-	__u32 *straws;         /* 16-bit fixed point */
-};
-
-
-
-/*
- * CRUSH map includes all buckets, rules, etc.
- */
-struct crush_map {
-	struct crush_bucket **buckets;
-	struct crush_rule **rules;
-
-	/*
-	 * Parent pointers to identify the parent bucket a device or
-	 * bucket in the hierarchy.  If an item appears more than
-	 * once, this is the _last_ time it appeared (where buckets
-	 * are processed in bucket id order, from -1 on down to
-	 * -max_buckets.
-	 */
-	__u32 *bucket_parents;
-	__u32 *device_parents;
-
-	__s32 max_buckets;
-	__u32 max_rules;
-	__s32 max_devices;
-};
-
-
-/* crush.c */
-extern int crush_get_bucket_item_weight(struct crush_bucket *b, int pos);
-extern void crush_calc_parents(struct crush_map *map);
-extern void crush_destroy_bucket_uniform(struct crush_bucket_uniform *b);
-extern void crush_destroy_bucket_list(struct crush_bucket_list *b);
-extern void crush_destroy_bucket_tree(struct crush_bucket_tree *b);
-extern void crush_destroy_bucket_straw(struct crush_bucket_straw *b);
-extern void crush_destroy_bucket(struct crush_bucket *b);
-extern void crush_destroy(struct crush_map *map);
-
-#endif
diff --git a/fs/ceph/crush/hash.c b/fs/ceph/crush/hash.c
deleted file mode 100644
index 5873aed694bf..000000000000
--- a/fs/ceph/crush/hash.c
+++ /dev/null
@@ -1,149 +0,0 @@
-
-#include <linux/types.h>
-#include "hash.h"
-
-/*
- * Robert Jenkins' function for mixing 32-bit values
- * http://burtleburtle.net/bob/hash/evahash.html
- * a, b = random bits, c = input and output
- */
-#define crush_hashmix(a, b, c) do {			\
-		a = a-b;  a = a-c;  a = a^(c>>13);	\
-		b = b-c;  b = b-a;  b = b^(a<<8);	\
-		c = c-a;  c = c-b;  c = c^(b>>13);	\
-		a = a-b;  a = a-c;  a = a^(c>>12);	\
-		b = b-c;  b = b-a;  b = b^(a<<16);	\
-		c = c-a;  c = c-b;  c = c^(b>>5);	\
-		a = a-b;  a = a-c;  a = a^(c>>3);	\
-		b = b-c;  b = b-a;  b = b^(a<<10);	\
-		c = c-a;  c = c-b;  c = c^(b>>15);	\
-	} while (0)
-
-#define crush_hash_seed 1315423911
-
-static __u32 crush_hash32_rjenkins1(__u32 a)
-{
-	__u32 hash = crush_hash_seed ^ a;
-	__u32 b = a;
-	__u32 x = 231232;
-	__u32 y = 1232;
-	crush_hashmix(b, x, hash);
-	crush_hashmix(y, a, hash);
-	return hash;
-}
-
-static __u32 crush_hash32_rjenkins1_2(__u32 a, __u32 b)
-{
-	__u32 hash = crush_hash_seed ^ a ^ b;
-	__u32 x = 231232;
-	__u32 y = 1232;
-	crush_hashmix(a, b, hash);
-	crush_hashmix(x, a, hash);
-	crush_hashmix(b, y, hash);
-	return hash;
-}
-
-static __u32 crush_hash32_rjenkins1_3(__u32 a, __u32 b, __u32 c)
-{
-	__u32 hash = crush_hash_seed ^ a ^ b ^ c;
-	__u32 x = 231232;
-	__u32 y = 1232;
-	crush_hashmix(a, b, hash);
-	crush_hashmix(c, x, hash);
-	crush_hashmix(y, a, hash);
-	crush_hashmix(b, x, hash);
-	crush_hashmix(y, c, hash);
-	return hash;
-}
-
-static __u32 crush_hash32_rjenkins1_4(__u32 a, __u32 b, __u32 c, __u32 d)
-{
-	__u32 hash = crush_hash_seed ^ a ^ b ^ c ^ d;
-	__u32 x = 231232;
-	__u32 y = 1232;
-	crush_hashmix(a, b, hash);
-	crush_hashmix(c, d, hash);
-	crush_hashmix(a, x, hash);
-	crush_hashmix(y, b, hash);
-	crush_hashmix(c, x, hash);
-	crush_hashmix(y, d, hash);
-	return hash;
-}
-
-static __u32 crush_hash32_rjenkins1_5(__u32 a, __u32 b, __u32 c, __u32 d,
-				      __u32 e)
-{
-	__u32 hash = crush_hash_seed ^ a ^ b ^ c ^ d ^ e;
-	__u32 x = 231232;
-	__u32 y = 1232;
-	crush_hashmix(a, b, hash);
-	crush_hashmix(c, d, hash);
-	crush_hashmix(e, x, hash);
-	crush_hashmix(y, a, hash);
-	crush_hashmix(b, x, hash);
-	crush_hashmix(y, c, hash);
-	crush_hashmix(d, x, hash);
-	crush_hashmix(y, e, hash);
-	return hash;
-}
-
-
-__u32 crush_hash32(int type, __u32 a)
-{
-	switch (type) {
-	case CRUSH_HASH_RJENKINS1:
-		return crush_hash32_rjenkins1(a);
-	default:
-		return 0;
-	}
-}
-
-__u32 crush_hash32_2(int type, __u32 a, __u32 b)
-{
-	switch (type) {
-	case CRUSH_HASH_RJENKINS1:
-		return crush_hash32_rjenkins1_2(a, b);
-	default:
-		return 0;
-	}
-}
-
-__u32 crush_hash32_3(int type, __u32 a, __u32 b, __u32 c)
-{
-	switch (type) {
-	case CRUSH_HASH_RJENKINS1:
-		return crush_hash32_rjenkins1_3(a, b, c);
-	default:
-		return 0;
-	}
-}
-
-__u32 crush_hash32_4(int type, __u32 a, __u32 b, __u32 c, __u32 d)
-{
-	switch (type) {
-	case CRUSH_HASH_RJENKINS1:
-		return crush_hash32_rjenkins1_4(a, b, c, d);
-	default:
-		return 0;
-	}
-}
-
-__u32 crush_hash32_5(int type, __u32 a, __u32 b, __u32 c, __u32 d, __u32 e)
-{
-	switch (type) {
-	case CRUSH_HASH_RJENKINS1:
-		return crush_hash32_rjenkins1_5(a, b, c, d, e);
-	default:
-		return 0;
-	}
-}
-
-const char *crush_hash_name(int type)
-{
-	switch (type) {
-	case CRUSH_HASH_RJENKINS1:
-		return "rjenkins1";
-	default:
-		return "unknown";
-	}
-}
diff --git a/fs/ceph/crush/hash.h b/fs/ceph/crush/hash.h
deleted file mode 100644
index 91e884230d5d..000000000000
--- a/fs/ceph/crush/hash.h
+++ /dev/null
@@ -1,17 +0,0 @@
-#ifndef CEPH_CRUSH_HASH_H
-#define CEPH_CRUSH_HASH_H
-
-#define CRUSH_HASH_RJENKINS1   0
-
-#define CRUSH_HASH_DEFAULT CRUSH_HASH_RJENKINS1
-
-extern const char *crush_hash_name(int type);
-
-extern __u32 crush_hash32(int type, __u32 a);
-extern __u32 crush_hash32_2(int type, __u32 a, __u32 b);
-extern __u32 crush_hash32_3(int type, __u32 a, __u32 b, __u32 c);
-extern __u32 crush_hash32_4(int type, __u32 a, __u32 b, __u32 c, __u32 d);
-extern __u32 crush_hash32_5(int type, __u32 a, __u32 b, __u32 c, __u32 d,
-			    __u32 e);
-
-#endif
diff --git a/fs/ceph/crush/mapper.c b/fs/ceph/crush/mapper.c
deleted file mode 100644
index a4eec133258e..000000000000
--- a/fs/ceph/crush/mapper.c
+++ /dev/null
@@ -1,609 +0,0 @@
-
-#ifdef __KERNEL__
-# include <linux/string.h>
-# include <linux/slab.h>
-# include <linux/bug.h>
-# include <linux/kernel.h>
-# ifndef dprintk
-#  define dprintk(args...)
-# endif
-#else
-# include <string.h>
-# include <stdio.h>
-# include <stdlib.h>
-# include <assert.h>
-# define BUG_ON(x) assert(!(x))
-# define dprintk(args...) /* printf(args) */
-# define kmalloc(x, f) malloc(x)
-# define kfree(x) free(x)
-#endif
-
-#include "crush.h"
-#include "hash.h"
-
-/*
- * Implement the core CRUSH mapping algorithm.
- */
-
-/**
- * crush_find_rule - find a crush_rule id for a given ruleset, type, and size.
- * @map: the crush_map
- * @ruleset: the storage ruleset id (user defined)
- * @type: storage ruleset type (user defined)
- * @size: output set size
- */
-int crush_find_rule(struct crush_map *map, int ruleset, int type, int size)
-{
-	int i;
-
-	for (i = 0; i < map->max_rules; i++) {
-		if (map->rules[i] &&
-		    map->rules[i]->mask.ruleset == ruleset &&
-		    map->rules[i]->mask.type == type &&
-		    map->rules[i]->mask.min_size <= size &&
-		    map->rules[i]->mask.max_size >= size)
-			return i;
-	}
-	return -1;
-}
-
-
-/*
- * bucket choose methods
- *
- * For each bucket algorithm, we have a "choose" method that, given a
- * crush input @x and replica position (usually, position in output set) @r,
- * will produce an item in the bucket.
- */
-
-/*
- * Choose based on a random permutation of the bucket.
- *
- * We used to use some prime number arithmetic to do this, but it
- * wasn't very random, and had some other bad behaviors.  Instead, we
- * calculate an actual random permutation of the bucket members.
- * Since this is expensive, we optimize for the r=0 case, which
- * captures the vast majority of calls.
- */
-static int bucket_perm_choose(struct crush_bucket *bucket,
-			      int x, int r)
-{
-	unsigned pr = r % bucket->size;
-	unsigned i, s;
-
-	/* start a new permutation if @x has changed */
-	if (bucket->perm_x != x || bucket->perm_n == 0) {
-		dprintk("bucket %d new x=%d\n", bucket->id, x);
-		bucket->perm_x = x;
-
-		/* optimize common r=0 case */
-		if (pr == 0) {
-			s = crush_hash32_3(bucket->hash, x, bucket->id, 0) %
-				bucket->size;
-			bucket->perm[0] = s;
-			bucket->perm_n = 0xffff;   /* magic value, see below */
-			goto out;
-		}
-
-		for (i = 0; i < bucket->size; i++)
-			bucket->perm[i] = i;
-		bucket->perm_n = 0;
-	} else if (bucket->perm_n == 0xffff) {
-		/* clean up after the r=0 case above */
-		for (i = 1; i < bucket->size; i++)
-			bucket->perm[i] = i;
-		bucket->perm[bucket->perm[0]] = 0;
-		bucket->perm_n = 1;
-	}
-
-	/* calculate permutation up to pr */
-	for (i = 0; i < bucket->perm_n; i++)
-		dprintk(" perm_choose have %d: %d\n", i, bucket->perm[i]);
-	while (bucket->perm_n <= pr) {
-		unsigned p = bucket->perm_n;
-		/* no point in swapping the final entry */
-		if (p < bucket->size - 1) {
-			i = crush_hash32_3(bucket->hash, x, bucket->id, p) %
-				(bucket->size - p);
-			if (i) {
-				unsigned t = bucket->perm[p + i];
-				bucket->perm[p + i] = bucket->perm[p];
-				bucket->perm[p] = t;
-			}
-			dprintk(" perm_choose swap %d with %d\n", p, p+i);
-		}
-		bucket->perm_n++;
-	}
-	for (i = 0; i < bucket->size; i++)
-		dprintk(" perm_choose  %d: %d\n", i, bucket->perm[i]);
-
-	s = bucket->perm[pr];
-out:
-	dprintk(" perm_choose %d sz=%d x=%d r=%d (%d) s=%d\n", bucket->id,
-		bucket->size, x, r, pr, s);
-	return bucket->items[s];
-}
-
-/* uniform */
-static int bucket_uniform_choose(struct crush_bucket_uniform *bucket,
-				 int x, int r)
-{
-	return bucket_perm_choose(&bucket->h, x, r);
-}
-
-/* list */
-static int bucket_list_choose(struct crush_bucket_list *bucket,
-			      int x, int r)
-{
-	int i;
-
-	for (i = bucket->h.size-1; i >= 0; i--) {
-		__u64 w = crush_hash32_4(bucket->h.hash,x, bucket->h.items[i],
-					 r, bucket->h.id);
-		w &= 0xffff;
-		dprintk("list_choose i=%d x=%d r=%d item %d weight %x "
-			"sw %x rand %llx",
-			i, x, r, bucket->h.items[i], bucket->item_weights[i],
-			bucket->sum_weights[i], w);
-		w *= bucket->sum_weights[i];
-		w = w >> 16;
-		/*dprintk(" scaled %llx\n", w);*/
-		if (w < bucket->item_weights[i])
-			return bucket->h.items[i];
-	}
-
-	BUG_ON(1);
-	return 0;
-}
-
-
-/* (binary) tree */
-static int height(int n)
-{
-	int h = 0;
-	while ((n & 1) == 0) {
-		h++;
-		n = n >> 1;
-	}
-	return h;
-}
-
-static int left(int x)
-{
-	int h = height(x);
-	return x - (1 << (h-1));
-}
-
-static int right(int x)
-{
-	int h = height(x);
-	return x + (1 << (h-1));
-}
-
-static int terminal(int x)
-{
-	return x & 1;
-}
-
-static int bucket_tree_choose(struct crush_bucket_tree *bucket,
-			      int x, int r)
-{
-	int n, l;
-	__u32 w;
-	__u64 t;
-
-	/* start at root */
-	n = bucket->num_nodes >> 1;
-
-	while (!terminal(n)) {
-		/* pick point in [0, w) */
-		w = bucket->node_weights[n];
-		t = (__u64)crush_hash32_4(bucket->h.hash, x, n, r,
-					  bucket->h.id) * (__u64)w;
-		t = t >> 32;
-
-		/* descend to the left or right? */
-		l = left(n);
-		if (t < bucket->node_weights[l])
-			n = l;
-		else
-			n = right(n);
-	}
-
-	return bucket->h.items[n >> 1];
-}
-
-
-/* straw */
-
-static int bucket_straw_choose(struct crush_bucket_straw *bucket,
-			       int x, int r)
-{
-	int i;
-	int high = 0;
-	__u64 high_draw = 0;
-	__u64 draw;
-
-	for (i = 0; i < bucket->h.size; i++) {
-		draw = crush_hash32_3(bucket->h.hash, x, bucket->h.items[i], r);
-		draw &= 0xffff;
-		draw *= bucket->straws[i];
-		if (i == 0 || draw > high_draw) {
-			high = i;
-			high_draw = draw;
-		}
-	}
-	return bucket->h.items[high];
-}
-
-static int crush_bucket_choose(struct crush_bucket *in, int x, int r)
-{
-	dprintk(" crush_bucket_choose %d x=%d r=%d\n", in->id, x, r);
-	switch (in->alg) {
-	case CRUSH_BUCKET_UNIFORM:
-		return bucket_uniform_choose((struct crush_bucket_uniform *)in,
-					  x, r);
-	case CRUSH_BUCKET_LIST:
-		return bucket_list_choose((struct crush_bucket_list *)in,
-					  x, r);
-	case CRUSH_BUCKET_TREE:
-		return bucket_tree_choose((struct crush_bucket_tree *)in,
-					  x, r);
-	case CRUSH_BUCKET_STRAW:
-		return bucket_straw_choose((struct crush_bucket_straw *)in,
-					   x, r);
-	default:
-		BUG_ON(1);
-		return in->items[0];
-	}
-}
-
-/*
- * true if device is marked "out" (failed, fully offloaded)
- * of the cluster
- */
-static int is_out(struct crush_map *map, __u32 *weight, int item, int x)
-{
-	if (weight[item] >= 0x10000)
-		return 0;
-	if (weight[item] == 0)
-		return 1;
-	if ((crush_hash32_2(CRUSH_HASH_RJENKINS1, x, item) & 0xffff)
-	    < weight[item])
-		return 0;
-	return 1;
-}
-
-/**
- * crush_choose - choose numrep distinct items of given type
- * @map: the crush_map
- * @bucket: the bucket we are choose an item from
- * @x: crush input value
- * @numrep: the number of items to choose
- * @type: the type of item to choose
- * @out: pointer to output vector
- * @outpos: our position in that vector
- * @firstn: true if choosing "first n" items, false if choosing "indep"
- * @recurse_to_leaf: true if we want one device under each item of given type
- * @out2: second output vector for leaf items (if @recurse_to_leaf)
- */
-static int crush_choose(struct crush_map *map,
-			struct crush_bucket *bucket,
-			__u32 *weight,
-			int x, int numrep, int type,
-			int *out, int outpos,
-			int firstn, int recurse_to_leaf,
-			int *out2)
-{
-	int rep;
-	int ftotal, flocal;
-	int retry_descent, retry_bucket, skip_rep;
-	struct crush_bucket *in = bucket;
-	int r;
-	int i;
-	int item = 0;
-	int itemtype;
-	int collide, reject;
-	const int orig_tries = 5; /* attempts before we fall back to search */
-
-	dprintk("CHOOSE%s bucket %d x %d outpos %d numrep %d\n", recurse_to_leaf ? "_LEAF" : "",
-		bucket->id, x, outpos, numrep);
-
-	for (rep = outpos; rep < numrep; rep++) {
-		/* keep trying until we get a non-out, non-colliding item */
-		ftotal = 0;
-		skip_rep = 0;
-		do {
-			retry_descent = 0;
-			in = bucket;               /* initial bucket */
-
-			/* choose through intervening buckets */
-			flocal = 0;
-			do {
-				collide = 0;
-				retry_bucket = 0;
-				r = rep;
-				if (in->alg == CRUSH_BUCKET_UNIFORM) {
-					/* be careful */
-					if (firstn || numrep >= in->size)
-						/* r' = r + f_total */
-						r += ftotal;
-					else if (in->size % numrep == 0)
-						/* r'=r+(n+1)*f_local */
-						r += (numrep+1) *
-							(flocal+ftotal);
-					else
-						/* r' = r + n*f_local */
-						r += numrep * (flocal+ftotal);
-				} else {
-					if (firstn)
-						/* r' = r + f_total */
-						r += ftotal;
-					else
-						/* r' = r + n*f_local */
-						r += numrep * (flocal+ftotal);
-				}
-
-				/* bucket choose */
-				if (in->size == 0) {
-					reject = 1;
-					goto reject;
-				}
-				if (flocal >= (in->size>>1) &&
-				    flocal > orig_tries)
-					item = bucket_perm_choose(in, x, r);
-				else
-					item = crush_bucket_choose(in, x, r);
-				BUG_ON(item >= map->max_devices);
-
-				/* desired type? */
-				if (item < 0)
-					itemtype = map->buckets[-1-item]->type;
-				else
-					itemtype = 0;
-				dprintk("  item %d type %d\n", item, itemtype);
-
-				/* keep going? */
-				if (itemtype != type) {
-					BUG_ON(item >= 0 ||
-					       (-1-item) >= map->max_buckets);
-					in = map->buckets[-1-item];
-					retry_bucket = 1;
-					continue;
-				}
-
-				/* collision? */
-				for (i = 0; i < outpos; i++) {
-					if (out[i] == item) {
-						collide = 1;
-						break;
-					}
-				}
-
-				reject = 0;
-				if (recurse_to_leaf) {
-					if (item < 0) {
-						if (crush_choose(map,
-							 map->buckets[-1-item],
-							 weight,
-							 x, outpos+1, 0,
-							 out2, outpos,
-							 firstn, 0,
-							 NULL) <= outpos)
-							/* didn't get leaf */
-							reject = 1;
-					} else {
-						/* we already have a leaf! */
-						out2[outpos] = item;
-					}
-				}
-
-				if (!reject) {
-					/* out? */
-					if (itemtype == 0)
-						reject = is_out(map, weight,
-								item, x);
-					else
-						reject = 0;
-				}
-
-reject:
-				if (reject || collide) {
-					ftotal++;
-					flocal++;
-
-					if (collide && flocal < 3)
-						/* retry locally a few times */
-						retry_bucket = 1;
-					else if (flocal < in->size + orig_tries)
-						/* exhaustive bucket search */
-						retry_bucket = 1;
-					else if (ftotal < 20)
-						/* then retry descent */
-						retry_descent = 1;
-					else
-						/* else give up */
-						skip_rep = 1;
-					dprintk("  reject %d  collide %d  "
-						"ftotal %d  flocal %d\n",
-						reject, collide, ftotal,
-						flocal);
-				}
-			} while (retry_bucket);
-		} while (retry_descent);
-
-		if (skip_rep) {
-			dprintk("skip rep\n");
-			continue;
-		}
-
-		dprintk("CHOOSE got %d\n", item);
-		out[outpos] = item;
-		outpos++;
-	}
-
-	dprintk("CHOOSE returns %d\n", outpos);
-	return outpos;
-}
-
-
-/**
- * crush_do_rule - calculate a mapping with the given input and rule
- * @map: the crush_map
- * @ruleno: the rule id
- * @x: hash input
- * @result: pointer to result vector
- * @result_max: maximum result size
- * @force: force initial replica choice; -1 for none
- */
-int crush_do_rule(struct crush_map *map,
-		  int ruleno, int x, int *result, int result_max,
-		  int force, __u32 *weight)
-{
-	int result_len;
-	int force_context[CRUSH_MAX_DEPTH];
-	int force_pos = -1;
-	int a[CRUSH_MAX_SET];
-	int b[CRUSH_MAX_SET];
-	int c[CRUSH_MAX_SET];
-	int recurse_to_leaf;
-	int *w;
-	int wsize = 0;
-	int *o;
-	int osize;
-	int *tmp;
-	struct crush_rule *rule;
-	int step;
-	int i, j;
-	int numrep;
-	int firstn;
-	int rc = -1;
-
-	BUG_ON(ruleno >= map->max_rules);
-
-	rule = map->rules[ruleno];
-	result_len = 0;
-	w = a;
-	o = b;
-
-	/*
-	 * determine hierarchical context of force, if any.  note
-	 * that this may or may not correspond to the specific types
-	 * referenced by the crush rule.
-	 */
-	if (force >= 0) {
-		if (force >= map->max_devices ||
-		    map->device_parents[force] == 0) {
-			/*dprintk("CRUSH: forcefed device dne\n");*/
-			rc = -1;  /* force fed device dne */
-			goto out;
-		}
-		if (!is_out(map, weight, force, x)) {
-			while (1) {
-				force_context[++force_pos] = force;
-				if (force >= 0)
-					force = map->device_parents[force];
-				else
-					force = map->bucket_parents[-1-force];
-				if (force == 0)
-					break;
-			}
-		}
-	}
-
-	for (step = 0; step < rule->len; step++) {
-		firstn = 0;
-		switch (rule->steps[step].op) {
-		case CRUSH_RULE_TAKE:
-			w[0] = rule->steps[step].arg1;
-			if (force_pos >= 0) {
-				BUG_ON(force_context[force_pos] != w[0]);
-				force_pos--;
-			}
-			wsize = 1;
-			break;
-
-		case CRUSH_RULE_CHOOSE_LEAF_FIRSTN:
-		case CRUSH_RULE_CHOOSE_FIRSTN:
-			firstn = 1;
-		case CRUSH_RULE_CHOOSE_LEAF_INDEP:
-		case CRUSH_RULE_CHOOSE_INDEP:
-			BUG_ON(wsize == 0);
-
-			recurse_to_leaf =
-				rule->steps[step].op ==
-				 CRUSH_RULE_CHOOSE_LEAF_FIRSTN ||
-				rule->steps[step].op ==
-				CRUSH_RULE_CHOOSE_LEAF_INDEP;
-
-			/* reset output */
-			osize = 0;
-
-			for (i = 0; i < wsize; i++) {
-				/*
-				 * see CRUSH_N, CRUSH_N_MINUS macros.
-				 * basically, numrep <= 0 means relative to
-				 * the provided result_max
-				 */
-				numrep = rule->steps[step].arg1;
-				if (numrep <= 0) {
-					numrep += result_max;
-					if (numrep <= 0)
-						continue;
-				}
-				j = 0;
-				if (osize == 0 && force_pos >= 0) {
-					/* skip any intermediate types */
-					while (force_pos &&
-					       force_context[force_pos] < 0 &&
-					       rule->steps[step].arg2 !=
-					       map->buckets[-1 -
-					       force_context[force_pos]]->type)
-						force_pos--;
-					o[osize] = force_context[force_pos];
-					if (recurse_to_leaf)
-						c[osize] = force_context[0];
-					j++;
-					force_pos--;
-				}
-				osize += crush_choose(map,
-						      map->buckets[-1-w[i]],
-						      weight,
-						      x, numrep,
-						      rule->steps[step].arg2,
-						      o+osize, j,
-						      firstn,
-						      recurse_to_leaf, c+osize);
-			}
-
-			if (recurse_to_leaf)
-				/* copy final _leaf_ values to output set */
-				memcpy(o, c, osize*sizeof(*o));
-
-			/* swap t and w arrays */
-			tmp = o;
-			o = w;
-			w = tmp;
-			wsize = osize;
-			break;
-
-
-		case CRUSH_RULE_EMIT:
-			for (i = 0; i < wsize && result_len < result_max; i++) {
-				result[result_len] = w[i];
-				result_len++;
-			}
-			wsize = 0;
-			break;
-
-		default:
-			BUG_ON(1);
-		}
-	}
-	rc = result_len;
-
-out:
-	return rc;
-}
-
-
diff --git a/fs/ceph/crush/mapper.h b/fs/ceph/crush/mapper.h
deleted file mode 100644
index c46b99c18bb0..000000000000
--- a/fs/ceph/crush/mapper.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef CEPH_CRUSH_MAPPER_H
-#define CEPH_CRUSH_MAPPER_H
-
-/*
- * CRUSH functions for find rules and then mapping an input to an
- * output set.
- *
- * LGPL2
- */
-
-#include "crush.h"
-
-extern int crush_find_rule(struct crush_map *map, int pool, int type, int size);
-extern int crush_do_rule(struct crush_map *map,
-			 int ruleno,
-			 int x, int *result, int result_max,
-			 int forcefeed,    /* -1 for none */
-			 __u32 *weights);
-
-#endif
diff --git a/fs/ceph/crypto.c b/fs/ceph/crypto.c
deleted file mode 100644
index a3e627f63293..000000000000
--- a/fs/ceph/crypto.c
+++ /dev/null
@@ -1,412 +0,0 @@
-
-#include "ceph_debug.h"
-
-#include <linux/err.h>
-#include <linux/scatterlist.h>
-#include <linux/slab.h>
-#include <crypto/hash.h>
-
-#include "crypto.h"
-#include "decode.h"
-
-int ceph_crypto_key_encode(struct ceph_crypto_key *key, void **p, void *end)
-{
-	if (*p + sizeof(u16) + sizeof(key->created) +
-	    sizeof(u16) + key->len > end)
-		return -ERANGE;
-	ceph_encode_16(p, key->type);
-	ceph_encode_copy(p, &key->created, sizeof(key->created));
-	ceph_encode_16(p, key->len);
-	ceph_encode_copy(p, key->key, key->len);
-	return 0;
-}
-
-int ceph_crypto_key_decode(struct ceph_crypto_key *key, void **p, void *end)
-{
-	ceph_decode_need(p, end, 2*sizeof(u16) + sizeof(key->created), bad);
-	key->type = ceph_decode_16(p);
-	ceph_decode_copy(p, &key->created, sizeof(key->created));
-	key->len = ceph_decode_16(p);
-	ceph_decode_need(p, end, key->len, bad);
-	key->key = kmalloc(key->len, GFP_NOFS);
-	if (!key->key)
-		return -ENOMEM;
-	ceph_decode_copy(p, key->key, key->len);
-	return 0;
-
-bad:
-	dout("failed to decode crypto key\n");
-	return -EINVAL;
-}
-
-int ceph_crypto_key_unarmor(struct ceph_crypto_key *key, const char *inkey)
-{
-	int inlen = strlen(inkey);
-	int blen = inlen * 3 / 4;
-	void *buf, *p;
-	int ret;
-
-	dout("crypto_key_unarmor %s\n", inkey);
-	buf = kmalloc(blen, GFP_NOFS);
-	if (!buf)
-		return -ENOMEM;
-	blen = ceph_unarmor(buf, inkey, inkey+inlen);
-	if (blen < 0) {
-		kfree(buf);
-		return blen;
-	}
-
-	p = buf;
-	ret = ceph_crypto_key_decode(key, &p, p + blen);
-	kfree(buf);
-	if (ret)
-		return ret;
-	dout("crypto_key_unarmor key %p type %d len %d\n", key,
-	     key->type, key->len);
-	return 0;
-}
-
-
-
-#define AES_KEY_SIZE 16
-
-static struct crypto_blkcipher *ceph_crypto_alloc_cipher(void)
-{
-	return crypto_alloc_blkcipher("cbc(aes)", 0, CRYPTO_ALG_ASYNC);
-}
-
-static const u8 *aes_iv = (u8 *)CEPH_AES_IV;
-
-static int ceph_aes_encrypt(const void *key, int key_len,
-			    void *dst, size_t *dst_len,
-			    const void *src, size_t src_len)
-{
-	struct scatterlist sg_in[2], sg_out[1];
-	struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher();
-	struct blkcipher_desc desc = { .tfm = tfm, .flags = 0 };
-	int ret;
-	void *iv;
-	int ivsize;
-	size_t zero_padding = (0x10 - (src_len & 0x0f));
-	char pad[16];
-
-	if (IS_ERR(tfm))
-		return PTR_ERR(tfm);
-
-	memset(pad, zero_padding, zero_padding);
-
-	*dst_len = src_len + zero_padding;
-
-	crypto_blkcipher_setkey((void *)tfm, key, key_len);
-	sg_init_table(sg_in, 2);
-	sg_set_buf(&sg_in[0], src, src_len);
-	sg_set_buf(&sg_in[1], pad, zero_padding);
-	sg_init_table(sg_out, 1);
-	sg_set_buf(sg_out, dst, *dst_len);
-	iv = crypto_blkcipher_crt(tfm)->iv;
-	ivsize = crypto_blkcipher_ivsize(tfm);
-
-	memcpy(iv, aes_iv, ivsize);
-	/*
-	print_hex_dump(KERN_ERR, "enc key: ", DUMP_PREFIX_NONE, 16, 1,
-		       key, key_len, 1);
-	print_hex_dump(KERN_ERR, "enc src: ", DUMP_PREFIX_NONE, 16, 1,
-			src, src_len, 1);
-	print_hex_dump(KERN_ERR, "enc pad: ", DUMP_PREFIX_NONE, 16, 1,
-			pad, zero_padding, 1);
-	*/
-	ret = crypto_blkcipher_encrypt(&desc, sg_out, sg_in,
-				     src_len + zero_padding);
-	crypto_free_blkcipher(tfm);
-	if (ret < 0)
-		pr_err("ceph_aes_crypt failed %d\n", ret);
-	/*
-	print_hex_dump(KERN_ERR, "enc out: ", DUMP_PREFIX_NONE, 16, 1,
-		       dst, *dst_len, 1);
-	*/
-	return 0;
-}
-
-static int ceph_aes_encrypt2(const void *key, int key_len, void *dst,
-			     size_t *dst_len,
-			     const void *src1, size_t src1_len,
-			     const void *src2, size_t src2_len)
-{
-	struct scatterlist sg_in[3], sg_out[1];
-	struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher();
-	struct blkcipher_desc desc = { .tfm = tfm, .flags = 0 };
-	int ret;
-	void *iv;
-	int ivsize;
-	size_t zero_padding = (0x10 - ((src1_len + src2_len) & 0x0f));
-	char pad[16];
-
-	if (IS_ERR(tfm))
-		return PTR_ERR(tfm);
-
-	memset(pad, zero_padding, zero_padding);
-
-	*dst_len = src1_len + src2_len + zero_padding;
-
-	crypto_blkcipher_setkey((void *)tfm, key, key_len);
-	sg_init_table(sg_in, 3);
-	sg_set_buf(&sg_in[0], src1, src1_len);
-	sg_set_buf(&sg_in[1], src2, src2_len);
-	sg_set_buf(&sg_in[2], pad, zero_padding);
-	sg_init_table(sg_out, 1);
-	sg_set_buf(sg_out, dst, *dst_len);
-	iv = crypto_blkcipher_crt(tfm)->iv;
-	ivsize = crypto_blkcipher_ivsize(tfm);
-
-	memcpy(iv, aes_iv, ivsize);
-	/*
-	print_hex_dump(KERN_ERR, "enc  key: ", DUMP_PREFIX_NONE, 16, 1,
-		       key, key_len, 1);
-	print_hex_dump(KERN_ERR, "enc src1: ", DUMP_PREFIX_NONE, 16, 1,
-			src1, src1_len, 1);
-	print_hex_dump(KERN_ERR, "enc src2: ", DUMP_PREFIX_NONE, 16, 1,
-			src2, src2_len, 1);
-	print_hex_dump(KERN_ERR, "enc  pad: ", DUMP_PREFIX_NONE, 16, 1,
-			pad, zero_padding, 1);
-	*/
-	ret = crypto_blkcipher_encrypt(&desc, sg_out, sg_in,
-				     src1_len + src2_len + zero_padding);
-	crypto_free_blkcipher(tfm);
-	if (ret < 0)
-		pr_err("ceph_aes_crypt2 failed %d\n", ret);
-	/*
-	print_hex_dump(KERN_ERR, "enc  out: ", DUMP_PREFIX_NONE, 16, 1,
-		       dst, *dst_len, 1);
-	*/
-	return 0;
-}
-
-static int ceph_aes_decrypt(const void *key, int key_len,
-			    void *dst, size_t *dst_len,
-			    const void *src, size_t src_len)
-{
-	struct scatterlist sg_in[1], sg_out[2];
-	struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher();
-	struct blkcipher_desc desc = { .tfm = tfm };
-	char pad[16];
-	void *iv;
-	int ivsize;
-	int ret;
-	int last_byte;
-
-	if (IS_ERR(tfm))
-		return PTR_ERR(tfm);
-
-	crypto_blkcipher_setkey((void *)tfm, key, key_len);
-	sg_init_table(sg_in, 1);
-	sg_init_table(sg_out, 2);
-	sg_set_buf(sg_in, src, src_len);
-	sg_set_buf(&sg_out[0], dst, *dst_len);
-	sg_set_buf(&sg_out[1], pad, sizeof(pad));
-
-	iv = crypto_blkcipher_crt(tfm)->iv;
-	ivsize = crypto_blkcipher_ivsize(tfm);
-
-	memcpy(iv, aes_iv, ivsize);
-
-	/*
-	print_hex_dump(KERN_ERR, "dec key: ", DUMP_PREFIX_NONE, 16, 1,
-		       key, key_len, 1);
-	print_hex_dump(KERN_ERR, "dec  in: ", DUMP_PREFIX_NONE, 16, 1,
-		       src, src_len, 1);
-	*/
-
-	ret = crypto_blkcipher_decrypt(&desc, sg_out, sg_in, src_len);
-	crypto_free_blkcipher(tfm);
-	if (ret < 0) {
-		pr_err("ceph_aes_decrypt failed %d\n", ret);
-		return ret;
-	}
-
-	if (src_len <= *dst_len)
-		last_byte = ((char *)dst)[src_len - 1];
-	else
-		last_byte = pad[src_len - *dst_len - 1];
-	if (last_byte <= 16 && src_len >= last_byte) {
-		*dst_len = src_len - last_byte;
-	} else {
-		pr_err("ceph_aes_decrypt got bad padding %d on src len %d\n",
-		       last_byte, (int)src_len);
-		return -EPERM;  /* bad padding */
-	}
-	/*
-	print_hex_dump(KERN_ERR, "dec out: ", DUMP_PREFIX_NONE, 16, 1,
-		       dst, *dst_len, 1);
-	*/
-	return 0;
-}
-
-static int ceph_aes_decrypt2(const void *key, int key_len,
-			     void *dst1, size_t *dst1_len,
-			     void *dst2, size_t *dst2_len,
-			     const void *src, size_t src_len)
-{
-	struct scatterlist sg_in[1], sg_out[3];
-	struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher();
-	struct blkcipher_desc desc = { .tfm = tfm };
-	char pad[16];
-	void *iv;
-	int ivsize;
-	int ret;
-	int last_byte;
-
-	if (IS_ERR(tfm))
-		return PTR_ERR(tfm);
-
-	sg_init_table(sg_in, 1);
-	sg_set_buf(sg_in, src, src_len);
-	sg_init_table(sg_out, 3);
-	sg_set_buf(&sg_out[0], dst1, *dst1_len);
-	sg_set_buf(&sg_out[1], dst2, *dst2_len);
-	sg_set_buf(&sg_out[2], pad, sizeof(pad));
-
-	crypto_blkcipher_setkey((void *)tfm, key, key_len);
-	iv = crypto_blkcipher_crt(tfm)->iv;
-	ivsize = crypto_blkcipher_ivsize(tfm);
-
-	memcpy(iv, aes_iv, ivsize);
-
-	/*
-	print_hex_dump(KERN_ERR, "dec  key: ", DUMP_PREFIX_NONE, 16, 1,
-		       key, key_len, 1);
-	print_hex_dump(KERN_ERR, "dec   in: ", DUMP_PREFIX_NONE, 16, 1,
-		       src, src_len, 1);
-	*/
-
-	ret = crypto_blkcipher_decrypt(&desc, sg_out, sg_in, src_len);
-	crypto_free_blkcipher(tfm);
-	if (ret < 0) {
-		pr_err("ceph_aes_decrypt failed %d\n", ret);
-		return ret;
-	}
-
-	if (src_len <= *dst1_len)
-		last_byte = ((char *)dst1)[src_len - 1];
-	else if (src_len <= *dst1_len + *dst2_len)
-		last_byte = ((char *)dst2)[src_len - *dst1_len - 1];
-	else
-		last_byte = pad[src_len - *dst1_len - *dst2_len - 1];
-	if (last_byte <= 16 && src_len >= last_byte) {
-		src_len -= last_byte;
-	} else {
-		pr_err("ceph_aes_decrypt got bad padding %d on src len %d\n",
-		       last_byte, (int)src_len);
-		return -EPERM;  /* bad padding */
-	}
-
-	if (src_len < *dst1_len) {
-		*dst1_len = src_len;
-		*dst2_len = 0;
-	} else {
-		*dst2_len = src_len - *dst1_len;
-	}
-	/*
-	print_hex_dump(KERN_ERR, "dec  out1: ", DUMP_PREFIX_NONE, 16, 1,
-		       dst1, *dst1_len, 1);
-	print_hex_dump(KERN_ERR, "dec  out2: ", DUMP_PREFIX_NONE, 16, 1,
-		       dst2, *dst2_len, 1);
-	*/
-
-	return 0;
-}
-
-
-int ceph_decrypt(struct ceph_crypto_key *secret, void *dst, size_t *dst_len,
-		 const void *src, size_t src_len)
-{
-	switch (secret->type) {
-	case CEPH_CRYPTO_NONE:
-		if (*dst_len < src_len)
-			return -ERANGE;
-		memcpy(dst, src, src_len);
-		*dst_len = src_len;
-		return 0;
-
-	case CEPH_CRYPTO_AES:
-		return ceph_aes_decrypt(secret->key, secret->len, dst,
-					dst_len, src, src_len);
-
-	default:
-		return -EINVAL;
-	}
-}
-
-int ceph_decrypt2(struct ceph_crypto_key *secret,
-			void *dst1, size_t *dst1_len,
-			void *dst2, size_t *dst2_len,
-			const void *src, size_t src_len)
-{
-	size_t t;
-
-	switch (secret->type) {
-	case CEPH_CRYPTO_NONE:
-		if (*dst1_len + *dst2_len < src_len)
-			return -ERANGE;
-		t = min(*dst1_len, src_len);
-		memcpy(dst1, src, t);
-		*dst1_len = t;
-		src += t;
-		src_len -= t;
-		if (src_len) {
-			t = min(*dst2_len, src_len);
-			memcpy(dst2, src, t);
-			*dst2_len = t;
-		}
-		return 0;
-
-	case CEPH_CRYPTO_AES:
-		return ceph_aes_decrypt2(secret->key, secret->len,
-					 dst1, dst1_len, dst2, dst2_len,
-					 src, src_len);
-
-	default:
-		return -EINVAL;
-	}
-}
-
-int ceph_encrypt(struct ceph_crypto_key *secret, void *dst, size_t *dst_len,
-		 const void *src, size_t src_len)
-{
-	switch (secret->type) {
-	case CEPH_CRYPTO_NONE:
-		if (*dst_len < src_len)
-			return -ERANGE;
-		memcpy(dst, src, src_len);
-		*dst_len = src_len;
-		return 0;
-
-	case CEPH_CRYPTO_AES:
-		return ceph_aes_encrypt(secret->key, secret->len, dst,
-					dst_len, src, src_len);
-
-	default:
-		return -EINVAL;
-	}
-}
-
-int ceph_encrypt2(struct ceph_crypto_key *secret, void *dst, size_t *dst_len,
-		  const void *src1, size_t src1_len,
-		  const void *src2, size_t src2_len)
-{
-	switch (secret->type) {
-	case CEPH_CRYPTO_NONE:
-		if (*dst_len < src1_len + src2_len)
-			return -ERANGE;
-		memcpy(dst, src1, src1_len);
-		memcpy(dst + src1_len, src2, src2_len);
-		*dst_len = src1_len + src2_len;
-		return 0;
-
-	case CEPH_CRYPTO_AES:
-		return ceph_aes_encrypt2(secret->key, secret->len, dst, dst_len,
-					 src1, src1_len, src2, src2_len);
-
-	default:
-		return -EINVAL;
-	}
-}
diff --git a/fs/ceph/crypto.h b/fs/ceph/crypto.h
deleted file mode 100644
index bdf38607323c..000000000000
--- a/fs/ceph/crypto.h
+++ /dev/null
@@ -1,48 +0,0 @@
-#ifndef _FS_CEPH_CRYPTO_H
-#define _FS_CEPH_CRYPTO_H
-
-#include "types.h"
-#include "buffer.h"
-
-/*
- * cryptographic secret
- */
-struct ceph_crypto_key {
-	int type;
-	struct ceph_timespec created;
-	int len;
-	void *key;
-};
-
-static inline void ceph_crypto_key_destroy(struct ceph_crypto_key *key)
-{
-	kfree(key->key);
-}
-
-extern int ceph_crypto_key_encode(struct ceph_crypto_key *key,
-				  void **p, void *end);
-extern int ceph_crypto_key_decode(struct ceph_crypto_key *key,
-				  void **p, void *end);
-extern int ceph_crypto_key_unarmor(struct ceph_crypto_key *key, const char *in);
-
-/* crypto.c */
-extern int ceph_decrypt(struct ceph_crypto_key *secret,
-			void *dst, size_t *dst_len,
-			const void *src, size_t src_len);
-extern int ceph_encrypt(struct ceph_crypto_key *secret,
-			void *dst, size_t *dst_len,
-			const void *src, size_t src_len);
-extern int ceph_decrypt2(struct ceph_crypto_key *secret,
-			void *dst1, size_t *dst1_len,
-			void *dst2, size_t *dst2_len,
-			const void *src, size_t src_len);
-extern int ceph_encrypt2(struct ceph_crypto_key *secret,
-			 void *dst, size_t *dst_len,
-			 const void *src1, size_t src1_len,
-			 const void *src2, size_t src2_len);
-
-/* armor.c */
-extern int ceph_armor(char *dst, const char *src, const char *end);
-extern int ceph_unarmor(char *dst, const char *src, const char *end);
-
-#endif
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 6fd8b20a8611..94fe1d284c3a 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/device.h>
 #include <linux/slab.h>
@@ -7,143 +7,48 @@
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
 
-#include "super.h"
-#include "mds_client.h"
-#include "mon_client.h"
-#include "auth.h"
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
 
 #ifdef CONFIG_DEBUG_FS
 
-/*
- * Implement /sys/kernel/debug/ceph fun
- *
- * /sys/kernel/debug/ceph/client*  - an instance of the ceph client
- *      .../osdmap      - current osdmap
- *      .../mdsmap      - current mdsmap
- *      .../monmap      - current monmap
- *      .../osdc        - active osd requests
- *      .../mdsc        - active mds requests
- *      .../monc        - mon client state
- *      .../dentry_lru  - dump contents of dentry lru
- *      .../caps        - expose cap (reservation) stats
- *      .../bdi         - symlink to ../../bdi/something
- */
-
-static struct dentry *ceph_debugfs_dir;
-
-static int monmap_show(struct seq_file *s, void *p)
-{
-	int i;
-	struct ceph_client *client = s->private;
-
-	if (client->monc.monmap == NULL)
-		return 0;
-
-	seq_printf(s, "epoch %d\n", client->monc.monmap->epoch);
-	for (i = 0; i < client->monc.monmap->num_mon; i++) {
-		struct ceph_entity_inst *inst =
-			&client->monc.monmap->mon_inst[i];
-
-		seq_printf(s, "\t%s%lld\t%s\n",
-			   ENTITY_NAME(inst->name),
-			   pr_addr(&inst->addr.in_addr));
-	}
-	return 0;
-}
+#include "super.h"
+#include "mds_client.h"
 
 static int mdsmap_show(struct seq_file *s, void *p)
 {
 	int i;
-	struct ceph_client *client = s->private;
+	struct ceph_fs_client *fsc = s->private;
 
-	if (client->mdsc.mdsmap == NULL)
+	if (fsc->mdsc == NULL || fsc->mdsc->mdsmap == NULL)
 		return 0;
-	seq_printf(s, "epoch %d\n", client->mdsc.mdsmap->m_epoch);
-	seq_printf(s, "root %d\n", client->mdsc.mdsmap->m_root);
+	seq_printf(s, "epoch %d\n", fsc->mdsc->mdsmap->m_epoch);
+	seq_printf(s, "root %d\n", fsc->mdsc->mdsmap->m_root);
 	seq_printf(s, "session_timeout %d\n",
-		       client->mdsc.mdsmap->m_session_timeout);
+		       fsc->mdsc->mdsmap->m_session_timeout);
 	seq_printf(s, "session_autoclose %d\n",
-		       client->mdsc.mdsmap->m_session_autoclose);
-	for (i = 0; i < client->mdsc.mdsmap->m_max_mds; i++) {
+		       fsc->mdsc->mdsmap->m_session_autoclose);
+	for (i = 0; i < fsc->mdsc->mdsmap->m_max_mds; i++) {
 		struct ceph_entity_addr *addr =
-			&client->mdsc.mdsmap->m_info[i].addr;
-		int state = client->mdsc.mdsmap->m_info[i].state;
+			&fsc->mdsc->mdsmap->m_info[i].addr;
+		int state = fsc->mdsc->mdsmap->m_info[i].state;
 
-		seq_printf(s, "\tmds%d\t%s\t(%s)\n", i, pr_addr(&addr->in_addr),
+		seq_printf(s, "\tmds%d\t%s\t(%s)\n", i,
+			       ceph_pr_addr(&addr->in_addr),
 			       ceph_mds_state_name(state));
 	}
 	return 0;
 }
 
-static int osdmap_show(struct seq_file *s, void *p)
-{
-	int i;
-	struct ceph_client *client = s->private;
-	struct rb_node *n;
-
-	if (client->osdc.osdmap == NULL)
-		return 0;
-	seq_printf(s, "epoch %d\n", client->osdc.osdmap->epoch);
-	seq_printf(s, "flags%s%s\n",
-		   (client->osdc.osdmap->flags & CEPH_OSDMAP_NEARFULL) ?
-		   " NEARFULL" : "",
-		   (client->osdc.osdmap->flags & CEPH_OSDMAP_FULL) ?
-		   " FULL" : "");
-	for (n = rb_first(&client->osdc.osdmap->pg_pools); n; n = rb_next(n)) {
-		struct ceph_pg_pool_info *pool =
-			rb_entry(n, struct ceph_pg_pool_info, node);
-		seq_printf(s, "pg_pool %d pg_num %d / %d, lpg_num %d / %d\n",
-			   pool->id, pool->v.pg_num, pool->pg_num_mask,
-			   pool->v.lpg_num, pool->lpg_num_mask);
-	}
-	for (i = 0; i < client->osdc.osdmap->max_osd; i++) {
-		struct ceph_entity_addr *addr =
-			&client->osdc.osdmap->osd_addr[i];
-		int state = client->osdc.osdmap->osd_state[i];
-		char sb[64];
-
-		seq_printf(s, "\tosd%d\t%s\t%3d%%\t(%s)\n",
-			   i, pr_addr(&addr->in_addr),
-			   ((client->osdc.osdmap->osd_weight[i]*100) >> 16),
-			   ceph_osdmap_state_str(sb, sizeof(sb), state));
-	}
-	return 0;
-}
-
-static int monc_show(struct seq_file *s, void *p)
-{
-	struct ceph_client *client = s->private;
-	struct ceph_mon_generic_request *req;
-	struct ceph_mon_client *monc = &client->monc;
-	struct rb_node *rp;
-
-	mutex_lock(&monc->mutex);
-
-	if (monc->have_mdsmap)
-		seq_printf(s, "have mdsmap %u\n", (unsigned)monc->have_mdsmap);
-	if (monc->have_osdmap)
-		seq_printf(s, "have osdmap %u\n", (unsigned)monc->have_osdmap);
-	if (monc->want_next_osdmap)
-		seq_printf(s, "want next osdmap\n");
-
-	for (rp = rb_first(&monc->generic_request_tree); rp; rp = rb_next(rp)) {
-		__u16 op;
-		req = rb_entry(rp, struct ceph_mon_generic_request, node);
-		op = le16_to_cpu(req->request->hdr.type);
-		if (op == CEPH_MSG_STATFS)
-			seq_printf(s, "%lld statfs\n", req->tid);
-		else
-			seq_printf(s, "%lld unknown\n", req->tid);
-	}
-
-	mutex_unlock(&monc->mutex);
-	return 0;
-}
-
+/*
+ * mdsc debugfs
+ */
 static int mdsc_show(struct seq_file *s, void *p)
 {
-	struct ceph_client *client = s->private;
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = s->private;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	struct rb_node *rp;
 	int pathlen;
@@ -214,61 +119,12 @@ static int mdsc_show(struct seq_file *s, void *p)
 	return 0;
 }
 
-static int osdc_show(struct seq_file *s, void *pp)
-{
-	struct ceph_client *client = s->private;
-	struct ceph_osd_client *osdc = &client->osdc;
-	struct rb_node *p;
-
-	mutex_lock(&osdc->request_mutex);
-	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
-		struct ceph_osd_request *req;
-		struct ceph_osd_request_head *head;
-		struct ceph_osd_op *op;
-		int num_ops;
-		int opcode, olen;
-		int i;
-
-		req = rb_entry(p, struct ceph_osd_request, r_node);
-
-		seq_printf(s, "%lld\tosd%d\t%d.%x\t", req->r_tid,
-			   req->r_osd ? req->r_osd->o_osd : -1,
-			   le32_to_cpu(req->r_pgid.pool),
-			   le16_to_cpu(req->r_pgid.ps));
-
-		head = req->r_request->front.iov_base;
-		op = (void *)(head + 1);
-
-		num_ops = le16_to_cpu(head->num_ops);
-		olen = le32_to_cpu(head->object_len);
-		seq_printf(s, "%.*s", olen,
-			   (const char *)(head->ops + num_ops));
-
-		if (req->r_reassert_version.epoch)
-			seq_printf(s, "\t%u'%llu",
-			   (unsigned)le32_to_cpu(req->r_reassert_version.epoch),
-			   le64_to_cpu(req->r_reassert_version.version));
-		else
-			seq_printf(s, "\t");
-
-		for (i = 0; i < num_ops; i++) {
-			opcode = le16_to_cpu(op->op);
-			seq_printf(s, "\t%s", ceph_osd_op_name(opcode));
-			op++;
-		}
-
-		seq_printf(s, "\n");
-	}
-	mutex_unlock(&osdc->request_mutex);
-	return 0;
-}
-
 static int caps_show(struct seq_file *s, void *p)
 {
-	struct ceph_client *client = s->private;
+	struct ceph_fs_client *fsc = s->private;
 	int total, avail, used, reserved, min;
 
-	ceph_reservation_status(client, &total, &avail, &used, &reserved, &min);
+	ceph_reservation_status(fsc, &total, &avail, &used, &reserved, &min);
 	seq_printf(s, "total\t\t%d\n"
 		   "avail\t\t%d\n"
 		   "used\t\t%d\n"
@@ -280,8 +136,8 @@ static int caps_show(struct seq_file *s, void *p)
 
 static int dentry_lru_show(struct seq_file *s, void *ptr)
 {
-	struct ceph_client *client = s->private;
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = s->private;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_dentry_info *di;
 
 	spin_lock(&mdsc->dentry_lru_lock);
@@ -295,199 +151,124 @@ static int dentry_lru_show(struct seq_file *s, void *ptr)
 	return 0;
 }
 
-#define DEFINE_SHOW_FUNC(name)						\
-static int name##_open(struct inode *inode, struct file *file)		\
-{									\
-	struct seq_file *sf;						\
-	int ret;							\
-									\
-	ret = single_open(file, name, NULL);				\
-	sf = file->private_data;					\
-	sf->private = inode->i_private;					\
-	return ret;							\
-}									\
-									\
-static const struct file_operations name##_fops = {			\
-	.open		= name##_open,					\
-	.read		= seq_read,					\
-	.llseek		= seq_lseek,					\
-	.release	= single_release,				\
-};
-
-DEFINE_SHOW_FUNC(monmap_show)
-DEFINE_SHOW_FUNC(mdsmap_show)
-DEFINE_SHOW_FUNC(osdmap_show)
-DEFINE_SHOW_FUNC(monc_show)
-DEFINE_SHOW_FUNC(mdsc_show)
-DEFINE_SHOW_FUNC(osdc_show)
-DEFINE_SHOW_FUNC(dentry_lru_show)
-DEFINE_SHOW_FUNC(caps_show)
+CEPH_DEFINE_SHOW_FUNC(mdsmap_show)
+CEPH_DEFINE_SHOW_FUNC(mdsc_show)
+CEPH_DEFINE_SHOW_FUNC(caps_show)
+CEPH_DEFINE_SHOW_FUNC(dentry_lru_show)
+
 
+/*
+ * debugfs
+ */
 static int congestion_kb_set(void *data, u64 val)
 {
-	struct ceph_client *client = (struct ceph_client *)data;
-
-	if (client)
-		client->mount_args->congestion_kb = (int)val;
+	struct ceph_fs_client *fsc = (struct ceph_fs_client *)data;
 
+	fsc->mount_options->congestion_kb = (int)val;
 	return 0;
 }
 
 static int congestion_kb_get(void *data, u64 *val)
 {
-	struct ceph_client *client = (struct ceph_client *)data;
-
-	if (client)
-		*val = (u64)client->mount_args->congestion_kb;
+	struct ceph_fs_client *fsc = (struct ceph_fs_client *)data;
 
+	*val = (u64)fsc->mount_options->congestion_kb;
 	return 0;
 }
 
-
 DEFINE_SIMPLE_ATTRIBUTE(congestion_kb_fops, congestion_kb_get,
 			congestion_kb_set, "%llu\n");
 
-int __init ceph_debugfs_init(void)
-{
-	ceph_debugfs_dir = debugfs_create_dir("ceph", NULL);
-	if (!ceph_debugfs_dir)
-		return -ENOMEM;
-	return 0;
-}
 
-void ceph_debugfs_cleanup(void)
+void ceph_fs_debugfs_cleanup(struct ceph_fs_client *fsc)
 {
-	debugfs_remove(ceph_debugfs_dir);
+	dout("ceph_fs_debugfs_cleanup\n");
+	debugfs_remove(fsc->debugfs_bdi);
+	debugfs_remove(fsc->debugfs_congestion_kb);
+	debugfs_remove(fsc->debugfs_mdsmap);
+	debugfs_remove(fsc->debugfs_caps);
+	debugfs_remove(fsc->debugfs_mdsc);
+	debugfs_remove(fsc->debugfs_dentry_lru);
 }
 
-int ceph_debugfs_client_init(struct ceph_client *client)
+int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 {
-	int ret = 0;
-	char name[80];
-
-	snprintf(name, sizeof(name), "%pU.client%lld", &client->fsid,
-		 client->monc.auth->global_id);
-
-	client->debugfs_dir = debugfs_create_dir(name, ceph_debugfs_dir);
-	if (!client->debugfs_dir)
-		goto out;
+	char name[100];
+	int err = -ENOMEM;
 
-	client->monc.debugfs_file = debugfs_create_file("monc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &monc_show_fops);
-	if (!client->monc.debugfs_file)
-		goto out;
-
-	client->mdsc.debugfs_file = debugfs_create_file("mdsc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &mdsc_show_fops);
-	if (!client->mdsc.debugfs_file)
+	dout("ceph_fs_debugfs_init\n");
+	fsc->debugfs_congestion_kb =
+		debugfs_create_file("writeback_congestion_kb",
+				    0600,
+				    fsc->client->debugfs_dir,
+				    fsc,
+				    &congestion_kb_fops);
+	if (!fsc->debugfs_congestion_kb)
 		goto out;
 
-	client->osdc.debugfs_file = debugfs_create_file("osdc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &osdc_show_fops);
-	if (!client->osdc.debugfs_file)
-		goto out;
+	dout("a\n");
 
-	client->debugfs_monmap = debugfs_create_file("monmap",
-					0600,
-					client->debugfs_dir,
-					client,
-					&monmap_show_fops);
-	if (!client->debugfs_monmap)
+	snprintf(name, sizeof(name), "../../bdi/%s",
+		 dev_name(fsc->backing_dev_info.dev));
+	fsc->debugfs_bdi =
+		debugfs_create_symlink("bdi",
+				       fsc->client->debugfs_dir,
+				       name);
+	if (!fsc->debugfs_bdi)
 		goto out;
 
-	client->debugfs_mdsmap = debugfs_create_file("mdsmap",
+	dout("b\n");
+	fsc->debugfs_mdsmap = debugfs_create_file("mdsmap",
 					0600,
-					client->debugfs_dir,
-					client,
+					fsc->client->debugfs_dir,
+					fsc,
 					&mdsmap_show_fops);
-	if (!client->debugfs_mdsmap)
+	if (!fsc->debugfs_mdsmap)
 		goto out;
 
-	client->debugfs_osdmap = debugfs_create_file("osdmap",
-					0600,
-					client->debugfs_dir,
-					client,
-					&osdmap_show_fops);
-	if (!client->debugfs_osdmap)
+	dout("ca\n");
+	fsc->debugfs_mdsc = debugfs_create_file("mdsc",
+						0600,
+						fsc->client->debugfs_dir,
+						fsc,
+						&mdsc_show_fops);
+	if (!fsc->debugfs_mdsc)
 		goto out;
 
-	client->debugfs_dentry_lru = debugfs_create_file("dentry_lru",
-					0600,
-					client->debugfs_dir,
-					client,
-					&dentry_lru_show_fops);
-	if (!client->debugfs_dentry_lru)
-		goto out;
-
-	client->debugfs_caps = debugfs_create_file("caps",
+	dout("da\n");
+	fsc->debugfs_caps = debugfs_create_file("caps",
 						   0400,
-						   client->debugfs_dir,
-						   client,
+						   fsc->client->debugfs_dir,
+						   fsc,
 						   &caps_show_fops);
-	if (!client->debugfs_caps)
+	if (!fsc->debugfs_caps)
 		goto out;
 
-	client->debugfs_congestion_kb =
-		debugfs_create_file("writeback_congestion_kb",
-				    0600,
-				    client->debugfs_dir,
-				    client,
-				    &congestion_kb_fops);
-	if (!client->debugfs_congestion_kb)
+	dout("ea\n");
+	fsc->debugfs_dentry_lru = debugfs_create_file("dentry_lru",
+					0600,
+					fsc->client->debugfs_dir,
+					fsc,
+					&dentry_lru_show_fops);
+	if (!fsc->debugfs_dentry_lru)
 		goto out;
 
-	sprintf(name, "../../bdi/%s", dev_name(client->sb->s_bdi->dev));
-	client->debugfs_bdi = debugfs_create_symlink("bdi", client->debugfs_dir,
-						     name);
-
 	return 0;
 
 out:
-	ceph_debugfs_client_cleanup(client);
-	return ret;
+	ceph_fs_debugfs_cleanup(fsc);
+	return err;
 }
 
-void ceph_debugfs_client_cleanup(struct ceph_client *client)
-{
-	debugfs_remove(client->debugfs_bdi);
-	debugfs_remove(client->debugfs_caps);
-	debugfs_remove(client->debugfs_dentry_lru);
-	debugfs_remove(client->debugfs_osdmap);
-	debugfs_remove(client->debugfs_mdsmap);
-	debugfs_remove(client->debugfs_monmap);
-	debugfs_remove(client->osdc.debugfs_file);
-	debugfs_remove(client->mdsc.debugfs_file);
-	debugfs_remove(client->monc.debugfs_file);
-	debugfs_remove(client->debugfs_congestion_kb);
-	debugfs_remove(client->debugfs_dir);
-}
 
 #else  /* CONFIG_DEBUG_FS */
 
-int __init ceph_debugfs_init(void)
-{
-	return 0;
-}
-
-void ceph_debugfs_cleanup(void)
-{
-}
-
-int ceph_debugfs_client_init(struct ceph_client *client)
+int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 {
 	return 0;
 }
 
-void ceph_debugfs_client_cleanup(struct ceph_client *client)
+void ceph_fs_debugfs_cleanup(struct ceph_fs_client *fsc)
 {
 }
 
diff --git a/fs/ceph/decode.h b/fs/ceph/decode.h
deleted file mode 100644
index c5b6939fb32a..000000000000
--- a/fs/ceph/decode.h
+++ /dev/null
@@ -1,201 +0,0 @@
-#ifndef __CEPH_DECODE_H
-#define __CEPH_DECODE_H
-
-#include <asm/unaligned.h>
-#include <linux/time.h>
-
-#include "types.h"
-
-/*
- * in all cases,
- *   void **p     pointer to position pointer
- *   void *end    pointer to end of buffer (last byte + 1)
- */
-
-static inline u64 ceph_decode_64(void **p)
-{
-	u64 v = get_unaligned_le64(*p);
-	*p += sizeof(u64);
-	return v;
-}
-static inline u32 ceph_decode_32(void **p)
-{
-	u32 v = get_unaligned_le32(*p);
-	*p += sizeof(u32);
-	return v;
-}
-static inline u16 ceph_decode_16(void **p)
-{
-	u16 v = get_unaligned_le16(*p);
-	*p += sizeof(u16);
-	return v;
-}
-static inline u8 ceph_decode_8(void **p)
-{
-	u8 v = *(u8 *)*p;
-	(*p)++;
-	return v;
-}
-static inline void ceph_decode_copy(void **p, void *pv, size_t n)
-{
-	memcpy(pv, *p, n);
-	*p += n;
-}
-
-/*
- * bounds check input.
- */
-#define ceph_decode_need(p, end, n, bad)		\
-	do {						\
-		if (unlikely(*(p) + (n) > (end))) 	\
-			goto bad;			\
-	} while (0)
-
-#define ceph_decode_64_safe(p, end, v, bad)			\
-	do {							\
-		ceph_decode_need(p, end, sizeof(u64), bad);	\
-		v = ceph_decode_64(p);				\
-	} while (0)
-#define ceph_decode_32_safe(p, end, v, bad)			\
-	do {							\
-		ceph_decode_need(p, end, sizeof(u32), bad);	\
-		v = ceph_decode_32(p);				\
-	} while (0)
-#define ceph_decode_16_safe(p, end, v, bad)			\
-	do {							\
-		ceph_decode_need(p, end, sizeof(u16), bad);	\
-		v = ceph_decode_16(p);				\
-	} while (0)
-#define ceph_decode_8_safe(p, end, v, bad)			\
-	do {							\
-		ceph_decode_need(p, end, sizeof(u8), bad);	\
-		v = ceph_decode_8(p);				\
-	} while (0)
-
-#define ceph_decode_copy_safe(p, end, pv, n, bad)		\
-	do {							\
-		ceph_decode_need(p, end, n, bad);		\
-		ceph_decode_copy(p, pv, n);			\
-	} while (0)
-
-/*
- * struct ceph_timespec <-> struct timespec
- */
-static inline void ceph_decode_timespec(struct timespec *ts,
-					const struct ceph_timespec *tv)
-{
-	ts->tv_sec = le32_to_cpu(tv->tv_sec);
-	ts->tv_nsec = le32_to_cpu(tv->tv_nsec);
-}
-static inline void ceph_encode_timespec(struct ceph_timespec *tv,
-					const struct timespec *ts)
-{
-	tv->tv_sec = cpu_to_le32(ts->tv_sec);
-	tv->tv_nsec = cpu_to_le32(ts->tv_nsec);
-}
-
-/*
- * sockaddr_storage <-> ceph_sockaddr
- */
-static inline void ceph_encode_addr(struct ceph_entity_addr *a)
-{
-	__be16 ss_family = htons(a->in_addr.ss_family);
-	a->in_addr.ss_family = *(__u16 *)&ss_family;
-}
-static inline void ceph_decode_addr(struct ceph_entity_addr *a)
-{
-	__be16 ss_family = *(__be16 *)&a->in_addr.ss_family;
-	a->in_addr.ss_family = ntohs(ss_family);
-	WARN_ON(a->in_addr.ss_family == 512);
-}
-
-/*
- * encoders
- */
-static inline void ceph_encode_64(void **p, u64 v)
-{
-	put_unaligned_le64(v, (__le64 *)*p);
-	*p += sizeof(u64);
-}
-static inline void ceph_encode_32(void **p, u32 v)
-{
-	put_unaligned_le32(v, (__le32 *)*p);
-	*p += sizeof(u32);
-}
-static inline void ceph_encode_16(void **p, u16 v)
-{
-	put_unaligned_le16(v, (__le16 *)*p);
-	*p += sizeof(u16);
-}
-static inline void ceph_encode_8(void **p, u8 v)
-{
-	*(u8 *)*p = v;
-	(*p)++;
-}
-static inline void ceph_encode_copy(void **p, const void *s, int len)
-{
-	memcpy(*p, s, len);
-	*p += len;
-}
-
-/*
- * filepath, string encoders
- */
-static inline void ceph_encode_filepath(void **p, void *end,
-					u64 ino, const char *path)
-{
-	u32 len = path ? strlen(path) : 0;
-	BUG_ON(*p + sizeof(ino) + sizeof(len) + len > end);
-	ceph_encode_8(p, 1);
-	ceph_encode_64(p, ino);
-	ceph_encode_32(p, len);
-	if (len)
-		memcpy(*p, path, len);
-	*p += len;
-}
-
-static inline void ceph_encode_string(void **p, void *end,
-				      const char *s, u32 len)
-{
-	BUG_ON(*p + sizeof(len) + len > end);
-	ceph_encode_32(p, len);
-	if (len)
-		memcpy(*p, s, len);
-	*p += len;
-}
-
-#define ceph_encode_need(p, end, n, bad)		\
-	do {						\
-		if (unlikely(*(p) + (n) > (end))) 	\
-			goto bad;			\
-	} while (0)
-
-#define ceph_encode_64_safe(p, end, v, bad)			\
-	do {							\
-		ceph_encode_need(p, end, sizeof(u64), bad);	\
-		ceph_encode_64(p, v);				\
-	} while (0)
-#define ceph_encode_32_safe(p, end, v, bad)			\
-	do {							\
-		ceph_encode_need(p, end, sizeof(u32), bad);	\
-		ceph_encode_32(p, v);			\
-	} while (0)
-#define ceph_encode_16_safe(p, end, v, bad)			\
-	do {							\
-		ceph_encode_need(p, end, sizeof(u16), bad);	\
-		ceph_encode_16(p, v);			\
-	} while (0)
-
-#define ceph_encode_copy_safe(p, end, pv, n, bad)		\
-	do {							\
-		ceph_encode_need(p, end, n, bad);		\
-		ceph_encode_copy(p, pv, n);			\
-	} while (0)
-#define ceph_encode_string_safe(p, end, s, n, bad)		\
-	do {							\
-		ceph_encode_need(p, end, n, bad);		\
-		ceph_encode_string(p, end, s, n);		\
-	} while (0)
-
-
-#endif
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index a1986eb52045..a8d2aacc612b 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/spinlock.h>
 #include <linux/fs_struct.h>
@@ -7,6 +7,7 @@
 #include <linux/sched.h>
 
 #include "super.h"
+#include "mds_client.h"
 
 /*
  * Directory operations: readdir, lookup, create, link, unlink,
@@ -227,15 +228,15 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	struct ceph_file_info *fi = filp->private_data;
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_inode_to_client(inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	unsigned frag = fpos_frag(filp->f_pos);
 	int off = fpos_off(filp->f_pos);
 	int err;
 	u32 ftype;
 	struct ceph_mds_reply_info_parsed *rinfo;
-	const int max_entries = client->mount_args->max_readdir;
-	const int max_bytes = client->mount_args->max_readdir_bytes;
+	const int max_entries = fsc->mount_options->max_readdir;
+	const int max_bytes = fsc->mount_options->max_readdir_bytes;
 
 	dout("readdir %p filp %p frag %u off %u\n", inode, filp, frag, off);
 	if (fi->at_end)
@@ -267,7 +268,7 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	/* can we use the dcache? */
 	spin_lock(&inode->i_lock);
 	if ((filp->f_pos == 2 || fi->dentry) &&
-	    !ceph_test_opt(client, NOASYNCREADDIR) &&
+	    !ceph_test_mount_opt(fsc, NOASYNCREADDIR) &&
 	    ceph_snap(inode) != CEPH_SNAPDIR &&
 	    (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
 	    __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) {
@@ -487,14 +488,14 @@ static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int origin)
 struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
 				  struct dentry *dentry, int err)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
 	struct inode *parent = dentry->d_parent->d_inode;
 
 	/* .snap dir? */
 	if (err == -ENOENT &&
 	    ceph_vino(parent).ino != CEPH_INO_ROOT && /* no .snap in root dir */
 	    strcmp(dentry->d_name.name,
-		   client->mount_args->snapdir_name) == 0) {
+		   fsc->mount_options->snapdir_name) == 0) {
 		struct inode *inode = ceph_get_snapdir(parent);
 		dout("ENOENT on snapdir %p '%.*s', linking to snapdir %p\n",
 		     dentry, dentry->d_name.len, dentry->d_name.name, inode);
@@ -539,8 +540,8 @@ static int is_root_ceph_dentry(struct inode *inode, struct dentry *dentry)
 static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
 				  struct nameidata *nd)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int op;
 	int err;
@@ -572,7 +573,7 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
 		spin_lock(&dir->i_lock);
 		dout(" dir %p flags are %d\n", dir, ci->i_ceph_flags);
 		if (strncmp(dentry->d_name.name,
-			    client->mount_args->snapdir_name,
+			    fsc->mount_options->snapdir_name,
 			    dentry->d_name.len) &&
 		    !is_root_ceph_dentry(dir, dentry) &&
 		    (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
@@ -629,8 +630,8 @@ int ceph_handle_notrace_create(struct inode *dir, struct dentry *dentry)
 static int ceph_mknod(struct inode *dir, struct dentry *dentry,
 		      int mode, dev_t rdev)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -685,8 +686,8 @@ static int ceph_create(struct inode *dir, struct dentry *dentry, int mode,
 static int ceph_symlink(struct inode *dir, struct dentry *dentry,
 			    const char *dest)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -716,8 +717,8 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry,
 
 static int ceph_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err = -EROFS;
 	int op;
@@ -758,8 +759,8 @@ out:
 static int ceph_link(struct dentry *old_dentry, struct inode *dir,
 		     struct dentry *dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -813,8 +814,8 @@ static int drop_caps_for_unlink(struct inode *inode)
  */
 static int ceph_unlink(struct inode *dir, struct dentry *dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = dentry->d_inode;
 	struct ceph_mds_request *req;
 	int err = -EROFS;
@@ -854,8 +855,8 @@ out:
 static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry,
 		       struct inode *new_dir, struct dentry *new_dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(old_dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(old_dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -1076,7 +1077,7 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int left;
 
-	if (!ceph_test_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT))
+	if (!ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT))
 		return -EISDIR;
 
 	if (!cf->dir_info) {
@@ -1177,7 +1178,7 @@ void ceph_dentry_lru_add(struct dentry *dn)
 	dout("dentry_lru_add %p %p '%.*s'\n", di, dn,
 	     dn->d_name.len, dn->d_name.name);
 	if (di) {
-		mdsc = &ceph_sb_to_client(dn->d_sb)->mdsc;
+		mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
 		spin_lock(&mdsc->dentry_lru_lock);
 		list_add_tail(&di->lru, &mdsc->dentry_lru);
 		mdsc->num_dentry++;
@@ -1193,7 +1194,7 @@ void ceph_dentry_lru_touch(struct dentry *dn)
 	dout("dentry_lru_touch %p %p '%.*s' (offset %lld)\n", di, dn,
 	     dn->d_name.len, dn->d_name.name, di->offset);
 	if (di) {
-		mdsc = &ceph_sb_to_client(dn->d_sb)->mdsc;
+		mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
 		spin_lock(&mdsc->dentry_lru_lock);
 		list_move_tail(&di->lru, &mdsc->dentry_lru);
 		spin_unlock(&mdsc->dentry_lru_lock);
@@ -1208,7 +1209,7 @@ void ceph_dentry_lru_del(struct dentry *dn)
 	dout("dentry_lru_del %p %p '%.*s'\n", di, dn,
 	     dn->d_name.len, dn->d_name.name);
 	if (di) {
-		mdsc = &ceph_sb_to_client(dn->d_sb)->mdsc;
+		mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
 		spin_lock(&mdsc->dentry_lru_lock);
 		list_del_init(&di->lru);
 		mdsc->num_dentry--;
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index e38423e82f2e..2297d9426992 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -1,10 +1,11 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/exportfs.h>
 #include <linux/slab.h>
 #include <asm/unaligned.h>
 
 #include "super.h"
+#include "mds_client.h"
 
 /*
  * NFS export support
@@ -120,7 +121,7 @@ static struct dentry *__fh_to_dentry(struct super_block *sb,
 static struct dentry *__cfh_to_dentry(struct super_block *sb,
 				      struct ceph_nfs_confh *cfh)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc;
 	struct inode *inode;
 	struct dentry *dentry;
 	struct ceph_vino vino;
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 66e4da6dba22..e77c28cf3690 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -1,5 +1,6 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/file.h>
@@ -38,8 +39,8 @@
 static struct ceph_mds_request *
 prepare_open_request(struct super_block *sb, int flags, int create_mode)
 {
-	struct ceph_client *client = ceph_sb_to_client(sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int want_auth = USE_ANY_MDS;
 	int op = (flags & O_CREAT) ? CEPH_MDS_OP_CREATE : CEPH_MDS_OP_OPEN;
@@ -117,8 +118,8 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
 int ceph_open(struct inode *inode, struct file *file)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_sb_to_client(inode->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	struct ceph_file_info *cf = file->private_data;
 	struct inode *parent_inode = file->f_dentry->d_parent->d_inode;
@@ -216,8 +217,8 @@ struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry,
 				struct nameidata *nd, int mode,
 				int locked_dir)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct file *file = nd->intent.open.file;
 	struct inode *parent_inode = get_dentry_parent_inode(file->f_dentry);
 	struct ceph_mds_request *req;
@@ -269,163 +270,6 @@ int ceph_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-/*
- * build a vector of user pages
- */
-static struct page **get_direct_page_vector(const char __user *data,
-					    int num_pages,
-					    loff_t off, size_t len)
-{
-	struct page **pages;
-	int rc;
-
-	pages = kmalloc(sizeof(*pages) * num_pages, GFP_NOFS);
-	if (!pages)
-		return ERR_PTR(-ENOMEM);
-
-	down_read(&current->mm->mmap_sem);
-	rc = get_user_pages(current, current->mm, (unsigned long)data,
-			    num_pages, 0, 0, pages, NULL);
-	up_read(&current->mm->mmap_sem);
-	if (rc < 0)
-		goto fail;
-	return pages;
-
-fail:
-	kfree(pages);
-	return ERR_PTR(rc);
-}
-
-static void put_page_vector(struct page **pages, int num_pages)
-{
-	int i;
-
-	for (i = 0; i < num_pages; i++)
-		put_page(pages[i]);
-	kfree(pages);
-}
-
-void ceph_release_page_vector(struct page **pages, int num_pages)
-{
-	int i;
-
-	for (i = 0; i < num_pages; i++)
-		__free_pages(pages[i], 0);
-	kfree(pages);
-}
-
-/*
- * allocate a vector new pages
- */
-static struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags)
-{
-	struct page **pages;
-	int i;
-
-	pages = kmalloc(sizeof(*pages) * num_pages, flags);
-	if (!pages)
-		return ERR_PTR(-ENOMEM);
-	for (i = 0; i < num_pages; i++) {
-		pages[i] = __page_cache_alloc(flags);
-		if (pages[i] == NULL) {
-			ceph_release_page_vector(pages, i);
-			return ERR_PTR(-ENOMEM);
-		}
-	}
-	return pages;
-}
-
-/*
- * copy user data into a page vector
- */
-static int copy_user_to_page_vector(struct page **pages,
-				    const char __user *data,
-				    loff_t off, size_t len)
-{
-	int i = 0;
-	int po = off & ~PAGE_CACHE_MASK;
-	int left = len;
-	int l, bad;
-
-	while (left > 0) {
-		l = min_t(int, PAGE_CACHE_SIZE-po, left);
-		bad = copy_from_user(page_address(pages[i]) + po, data, l);
-		if (bad == l)
-			return -EFAULT;
-		data += l - bad;
-		left -= l - bad;
-		po += l - bad;
-		if (po == PAGE_CACHE_SIZE) {
-			po = 0;
-			i++;
-		}
-	}
-	return len;
-}
-
-/*
- * copy user data from a page vector into a user pointer
- */
-static int copy_page_vector_to_user(struct page **pages, char __user *data,
-				    loff_t off, size_t len)
-{
-	int i = 0;
-	int po = off & ~PAGE_CACHE_MASK;
-	int left = len;
-	int l, bad;
-
-	while (left > 0) {
-		l = min_t(int, left, PAGE_CACHE_SIZE-po);
-		bad = copy_to_user(data, page_address(pages[i]) + po, l);
-		if (bad == l)
-			return -EFAULT;
-		data += l - bad;
-		left -= l - bad;
-		if (po) {
-			po += l - bad;
-			if (po == PAGE_CACHE_SIZE)
-				po = 0;
-		}
-		i++;
-	}
-	return len;
-}
-
-/*
- * Zero an extent within a page vector.  Offset is relative to the
- * start of the first page.
- */
-static void zero_page_vector_range(int off, int len, struct page **pages)
-{
-	int i = off >> PAGE_CACHE_SHIFT;
-
-	off &= ~PAGE_CACHE_MASK;
-
-	dout("zero_page_vector_page %u~%u\n", off, len);
-
-	/* leading partial page? */
-	if (off) {
-		int end = min((int)PAGE_CACHE_SIZE, off + len);
-		dout("zeroing %d %p head from %d\n", i, pages[i],
-		     (int)off);
-		zero_user_segment(pages[i], off, end);
-		len -= (end - off);
-		i++;
-	}
-	while (len >= PAGE_CACHE_SIZE) {
-		dout("zeroing %d %p len=%d\n", i, pages[i], len);
-		zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE);
-		len -= PAGE_CACHE_SIZE;
-		i++;
-	}
-	/* trailing partial page? */
-	if (len) {
-		dout("zeroing %d %p tail to %d\n", i, pages[i], (int)len);
-		zero_user_segment(pages[i], 0, len);
-	}
-}
-
-
 /*
  * Read a range of bytes striped over one or more objects.  Iterate over
  * objects we stripe over.  (That's not atomic, but good enough for now.)
@@ -438,7 +282,7 @@ static int striped_read(struct inode *inode,
 			struct page **pages, int num_pages,
 			int *checkeof)
 {
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	u64 pos, this_len;
 	int page_off = off & ~PAGE_CACHE_MASK; /* first byte's offset in page */
@@ -459,7 +303,7 @@ static int striped_read(struct inode *inode,
 
 more:
 	this_len = left;
-	ret = ceph_osdc_readpages(&client->osdc, ceph_vino(inode),
+	ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode),
 				  &ci->i_layout, pos, &this_len,
 				  ci->i_truncate_seq,
 				  ci->i_truncate_size,
@@ -477,8 +321,8 @@ more:
 
 		if (read < pos - off) {
 			dout(" zero gap %llu to %llu\n", off + read, pos);
-			zero_page_vector_range(page_off + read,
-					       pos - off - read, pages);
+			ceph_zero_page_vector_range(page_off + read,
+						    pos - off - read, pages);
 		}
 		pos += ret;
 		read = pos - off;
@@ -495,8 +339,8 @@ more:
 		/* was original extent fully inside i_size? */
 		if (pos + left <= inode->i_size) {
 			dout("zero tail\n");
-			zero_page_vector_range(page_off + read, len - read,
-					       pages);
+			ceph_zero_page_vector_range(page_off + read, len - read,
+						    pages);
 			read = len;
 			goto out;
 		}
@@ -531,7 +375,7 @@ static ssize_t ceph_sync_read(struct file *file, char __user *data,
 	     (file->f_flags & O_DIRECT) ? "O_DIRECT" : "");
 
 	if (file->f_flags & O_DIRECT) {
-		pages = get_direct_page_vector(data, num_pages, off, len);
+		pages = ceph_get_direct_page_vector(data, num_pages, off, len);
 
 		/*
 		 * flush any page cache pages in this range.  this
@@ -552,13 +396,13 @@ static ssize_t ceph_sync_read(struct file *file, char __user *data,
 	ret = striped_read(inode, off, len, pages, num_pages, checkeof);
 
 	if (ret >= 0 && (file->f_flags & O_DIRECT) == 0)
-		ret = copy_page_vector_to_user(pages, data, off, ret);
+		ret = ceph_copy_page_vector_to_user(pages, data, off, ret);
 	if (ret >= 0)
 		*poff = off + ret;
 
 done:
 	if (file->f_flags & O_DIRECT)
-		put_page_vector(pages, num_pages);
+		ceph_put_page_vector(pages, num_pages);
 	else
 		ceph_release_page_vector(pages, num_pages);
 	dout("sync_read result %d\n", ret);
@@ -594,7 +438,7 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	struct ceph_osd_request *req;
 	struct page **pages;
 	int num_pages;
@@ -642,7 +486,7 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
 	 */
 more:
 	len = left;
-	req = ceph_osdc_new_request(&client->osdc, &ci->i_layout,
+	req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
 				    ceph_vino(inode), pos, &len,
 				    CEPH_OSD_OP_WRITE, flags,
 				    ci->i_snap_realm->cached_context,
@@ -655,7 +499,7 @@ more:
 	num_pages = calc_pages_for(pos, len);
 
 	if (file->f_flags & O_DIRECT) {
-		pages = get_direct_page_vector(data, num_pages, pos, len);
+		pages = ceph_get_direct_page_vector(data, num_pages, pos, len);
 		if (IS_ERR(pages)) {
 			ret = PTR_ERR(pages);
 			goto out;
@@ -673,7 +517,7 @@ more:
 			ret = PTR_ERR(pages);
 			goto out;
 		}
-		ret = copy_user_to_page_vector(pages, data, pos, len);
+		ret = ceph_copy_user_to_page_vector(pages, data, pos, len);
 		if (ret < 0) {
 			ceph_release_page_vector(pages, num_pages);
 			goto out;
@@ -689,7 +533,7 @@ more:
 	req->r_num_pages = num_pages;
 	req->r_inode = inode;
 
-	ret = ceph_osdc_start_request(&client->osdc, req, false);
+	ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!ret) {
 		if (req->r_safe_callback) {
 			/*
@@ -701,11 +545,11 @@ more:
 			spin_unlock(&ci->i_unsafe_lock);
 			ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR);
 		}
-		ret = ceph_osdc_wait_request(&client->osdc, req);
+		ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
 	}
 
 	if (file->f_flags & O_DIRECT)
-		put_page_vector(pages, num_pages);
+		ceph_put_page_vector(pages, num_pages);
 	else if (file->f_flags & O_SYNC)
 		ceph_release_page_vector(pages, num_pages);
 
@@ -814,7 +658,8 @@ static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov,
 	struct ceph_file_info *fi = file->private_data;
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->osdc;
+	struct ceph_osd_client *osdc =
+		&ceph_sb_to_client(inode->i_sb)->client->osdc;
 	loff_t endoff = pos + iov->iov_len;
 	int want, got = 0;
 	int ret, err;
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 62377ec37edf..1d6a45b5a04c 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/module.h>
 #include <linux/fs.h>
@@ -13,7 +13,8 @@
 #include <linux/pagevec.h>
 
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+#include <linux/ceph/decode.h>
 
 /*
  * Ceph inode operations
@@ -384,7 +385,7 @@ void ceph_destroy_inode(struct inode *inode)
 	 */
 	if (ci->i_snap_realm) {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
+			ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 		struct ceph_snap_realm *realm = ci->i_snap_realm;
 
 		dout(" dropping residual ref to snap realm %p\n", realm);
@@ -685,7 +686,7 @@ static int fill_inode(struct inode *inode,
 		}
 
 		/* it may be better to set st_size in getattr instead? */
-		if (ceph_test_opt(ceph_sb_to_client(inode->i_sb), RBYTES))
+		if (ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), RBYTES))
 			inode->i_size = ci->i_rbytes;
 		break;
 	default:
@@ -901,7 +902,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 	struct inode *in = NULL;
 	struct ceph_mds_reply_inode *ininfo;
 	struct ceph_vino vino;
-	struct ceph_client *client = ceph_sb_to_client(sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
 	int i = 0;
 	int err = 0;
 
@@ -965,7 +966,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 	 */
 	if (rinfo->head->is_dentry && !req->r_aborted &&
 	    (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
-					       client->mount_args->snapdir_name,
+					       fsc->mount_options->snapdir_name,
 					       req->r_dentry->d_name.len))) {
 		/*
 		 * lookup link rename   : null -> possibly existing inode
@@ -1533,7 +1534,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
 	struct inode *parent_inode = dentry->d_parent->d_inode;
 	const unsigned int ia_valid = attr->ia_valid;
 	struct ceph_mds_request *req;
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(dentry->d_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(dentry->d_sb)->mdsc;
 	int issued;
 	int release = 0, dirtied = 0;
 	int mask = 0;
@@ -1728,8 +1729,8 @@ out:
  */
 int ceph_do_getattr(struct inode *inode, int mask)
 {
-	struct ceph_client *client = ceph_sb_to_client(inode->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 76e307d2aba1..899578b0c46b 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -1,8 +1,10 @@
 #include <linux/in.h>
 
-#include "ioctl.h"
 #include "super.h"
-#include "ceph_debug.h"
+#include "mds_client.h"
+#include <linux/ceph/ceph_debug.h>
+
+#include "ioctl.h"
 
 
 /*
@@ -37,7 +39,7 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg)
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct inode *parent_inode = file->f_dentry->d_parent->d_inode;
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_mds_request *req;
 	struct ceph_ioctl_layout l;
 	int err, i;
@@ -98,7 +100,8 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg)
 	struct ceph_ioctl_dataloc dl;
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->osdc;
+	struct ceph_osd_client *osdc =
+		&ceph_sb_to_client(inode->i_sb)->client->osdc;
 	u64 len = 1, olen;
 	u64 tmp;
 	struct ceph_object_layout ol;
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index ff4e753aae92..087afe9ca367 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -1,11 +1,11 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/file.h>
 #include <linux/namei.h>
 
 #include "super.h"
 #include "mds_client.h"
-#include "pagelist.h"
+#include <linux/ceph/pagelist.h>
 
 /**
  * Implement fcntl and flock locking functions.
@@ -16,7 +16,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_mds_client *mdsc =
-		&ceph_sb_to_client(inode->i_sb)->mdsc;
+		ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index fad95f8f2608..33568239a08e 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1,17 +1,20 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/wait.h>
 #include <linux/slab.h>
 #include <linux/sched.h>
 #include <linux/smp_lock.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
 
-#include "mds_client.h"
-#include "mon_client.h"
 #include "super.h"
-#include "messenger.h"
-#include "decode.h"
-#include "auth.h"
-#include "pagelist.h"
+#include "mds_client.h"
+
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/pagelist.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
 
 /*
  * A cluster of MDS (metadata server) daemons is responsible for
@@ -286,8 +289,9 @@ void ceph_put_mds_session(struct ceph_mds_session *s)
 	     atomic_read(&s->s_ref), atomic_read(&s->s_ref)-1);
 	if (atomic_dec_and_test(&s->s_ref)) {
 		if (s->s_authorizer)
-			s->s_mdsc->client->monc.auth->ops->destroy_authorizer(
-				s->s_mdsc->client->monc.auth, s->s_authorizer);
+		     s->s_mdsc->fsc->client->monc.auth->ops->destroy_authorizer(
+			     s->s_mdsc->fsc->client->monc.auth,
+			     s->s_authorizer);
 		kfree(s);
 	}
 }
@@ -344,7 +348,7 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
 	s->s_seq = 0;
 	mutex_init(&s->s_mutex);
 
-	ceph_con_init(mdsc->client->msgr, &s->s_con);
+	ceph_con_init(mdsc->fsc->client->msgr, &s->s_con);
 	s->s_con.private = s;
 	s->s_con.ops = &mds_con_ops;
 	s->s_con.peer_name.type = CEPH_ENTITY_TYPE_MDS;
@@ -599,7 +603,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 	} else if (req->r_dentry) {
 		struct inode *dir = req->r_dentry->d_parent->d_inode;
 
-		if (dir->i_sb != mdsc->client->sb) {
+		if (dir->i_sb != mdsc->fsc->sb) {
 			/* not this fs! */
 			inode = req->r_dentry->d_inode;
 		} else if (ceph_snap(dir) != CEPH_NOSNAP) {
@@ -884,7 +888,7 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap,
 	__ceph_remove_cap(cap);
 	if (!__ceph_is_any_real_caps(ci)) {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(inode->i_sb)->mdsc;
+			ceph_sb_to_client(inode->i_sb)->mdsc;
 
 		spin_lock(&mdsc->cap_dirty_lock);
 		if (!list_empty(&ci->i_dirty_item)) {
@@ -1146,7 +1150,7 @@ int ceph_add_cap_releases(struct ceph_mds_client *mdsc,
 	struct ceph_msg *msg, *partial = NULL;
 	struct ceph_mds_cap_release *head;
 	int err = -ENOMEM;
-	int extra = mdsc->client->mount_args->cap_release_safety;
+	int extra = mdsc->fsc->mount_options->cap_release_safety;
 	int num;
 
 	dout("add_cap_releases %p mds%d extra %d\n", session, session->s_mds,
@@ -2085,7 +2089,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 
 	/* insert trace into our cache */
 	mutex_lock(&req->r_fill_mutex);
-	err = ceph_fill_trace(mdsc->client->sb, req, req->r_session);
+	err = ceph_fill_trace(mdsc->fsc->sb, req, req->r_session);
 	if (err == 0) {
 		if (result == 0 && rinfo->dir_nr)
 			ceph_readdir_prepopulate(req, req->r_session);
@@ -2613,7 +2617,7 @@ static void handle_lease(struct ceph_mds_client *mdsc,
 			 struct ceph_mds_session *session,
 			 struct ceph_msg *msg)
 {
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	struct inode *inode;
 	struct ceph_inode_info *ci;
 	struct dentry *parent, *dentry;
@@ -2891,10 +2895,16 @@ static void delayed_work(struct work_struct *work)
 	schedule_delayed(mdsc);
 }
 
+int ceph_mdsc_init(struct ceph_fs_client *fsc)
 
-int ceph_mdsc_init(struct ceph_mds_client *mdsc, struct ceph_client *client)
 {
-	mdsc->client = client;
+	struct ceph_mds_client *mdsc;
+
+	mdsc = kzalloc(sizeof(struct ceph_mds_client), GFP_NOFS);
+	if (!mdsc)
+		return -ENOMEM;
+	mdsc->fsc = fsc;
+	fsc->mdsc = mdsc;
 	mutex_init(&mdsc->mutex);
 	mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS);
 	if (mdsc->mdsmap == NULL)
@@ -2927,7 +2937,7 @@ int ceph_mdsc_init(struct ceph_mds_client *mdsc, struct ceph_client *client)
 	INIT_LIST_HEAD(&mdsc->dentry_lru);
 
 	ceph_caps_init(mdsc);
-	ceph_adjust_min_caps(mdsc, client->min_caps);
+	ceph_adjust_min_caps(mdsc, fsc->min_caps);
 
 	return 0;
 }
@@ -2939,7 +2949,7 @@ int ceph_mdsc_init(struct ceph_mds_client *mdsc, struct ceph_client *client)
 static void wait_requests(struct ceph_mds_client *mdsc)
 {
 	struct ceph_mds_request *req;
-	struct ceph_client *client = mdsc->client;
+	struct ceph_fs_client *fsc = mdsc->fsc;
 
 	mutex_lock(&mdsc->mutex);
 	if (__get_oldest_req(mdsc)) {
@@ -2947,7 +2957,7 @@ static void wait_requests(struct ceph_mds_client *mdsc)
 
 		dout("wait_requests waiting for requests\n");
 		wait_for_completion_timeout(&mdsc->safe_umount_waiters,
-				    client->mount_args->mount_timeout * HZ);
+				    fsc->client->options->mount_timeout * HZ);
 
 		/* tear down remaining requests */
 		mutex_lock(&mdsc->mutex);
@@ -3030,7 +3040,7 @@ void ceph_mdsc_sync(struct ceph_mds_client *mdsc)
 {
 	u64 want_tid, want_flush;
 
-	if (mdsc->client->mount_state == CEPH_MOUNT_SHUTDOWN)
+	if (mdsc->fsc->mount_state == CEPH_MOUNT_SHUTDOWN)
 		return;
 
 	dout("sync\n");
@@ -3053,7 +3063,7 @@ bool done_closing_sessions(struct ceph_mds_client *mdsc)
 {
 	int i, n = 0;
 
-	if (mdsc->client->mount_state == CEPH_MOUNT_SHUTDOWN)
+	if (mdsc->fsc->mount_state == CEPH_MOUNT_SHUTDOWN)
 		return true;
 
 	mutex_lock(&mdsc->mutex);
@@ -3071,8 +3081,8 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
 {
 	struct ceph_mds_session *session;
 	int i;
-	struct ceph_client *client = mdsc->client;
-	unsigned long timeout = client->mount_args->mount_timeout * HZ;
+	struct ceph_fs_client *fsc = mdsc->fsc;
+	unsigned long timeout = fsc->client->options->mount_timeout * HZ;
 
 	dout("close_sessions\n");
 
@@ -3119,7 +3129,7 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
 	dout("stopped\n");
 }
 
-void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
+static void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
 {
 	dout("stop\n");
 	cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */
@@ -3129,6 +3139,15 @@ void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
 	ceph_caps_finalize(mdsc);
 }
 
+void ceph_mdsc_destroy(struct ceph_fs_client *fsc)
+{
+	struct ceph_mds_client *mdsc = fsc->mdsc;
+
+	ceph_mdsc_stop(mdsc);
+	fsc->mdsc = NULL;
+	kfree(mdsc);
+}
+
 
 /*
  * handle mds map update.
@@ -3145,14 +3164,14 @@ void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
 
 	ceph_decode_need(&p, end, sizeof(fsid)+2*sizeof(u32), bad);
 	ceph_decode_copy(&p, &fsid, sizeof(fsid));
-	if (ceph_check_fsid(mdsc->client, &fsid) < 0)
+	if (ceph_check_fsid(mdsc->fsc->client, &fsid) < 0)
 		return;
 	epoch = ceph_decode_32(&p);
 	maplen = ceph_decode_32(&p);
 	dout("handle_map epoch %u len %d\n", epoch, (int)maplen);
 
 	/* do we need it? */
-	ceph_monc_got_mdsmap(&mdsc->client->monc, epoch);
+	ceph_monc_got_mdsmap(&mdsc->fsc->client->monc, epoch);
 	mutex_lock(&mdsc->mutex);
 	if (mdsc->mdsmap && epoch <= mdsc->mdsmap->m_epoch) {
 		dout("handle_map epoch %u <= our %u\n",
@@ -3176,7 +3195,7 @@ void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
 	} else {
 		mdsc->mdsmap = newmap;  /* first mds map */
 	}
-	mdsc->client->sb->s_maxbytes = mdsc->mdsmap->m_max_file_size;
+	mdsc->fsc->sb->s_maxbytes = mdsc->mdsmap->m_max_file_size;
 
 	__wake_requests(mdsc, &mdsc->waiting_for_map);
 
@@ -3277,7 +3296,7 @@ static int get_authorizer(struct ceph_connection *con,
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 	int ret = 0;
 
 	if (force_new && s->s_authorizer) {
@@ -3311,7 +3330,7 @@ static int verify_authorizer_reply(struct ceph_connection *con, int len)
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 
 	return ac->ops->verify_authorizer_reply(ac, s->s_authorizer, len);
 }
@@ -3320,12 +3339,12 @@ static int invalidate_authorizer(struct ceph_connection *con)
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 
 	if (ac->ops->invalidate_authorizer)
 		ac->ops->invalidate_authorizer(ac, CEPH_ENTITY_TYPE_MDS);
 
-	return ceph_monc_validate_auth(&mdsc->client->monc);
+	return ceph_monc_validate_auth(&mdsc->fsc->client->monc);
 }
 
 static const struct ceph_connection_operations mds_con_ops = {
@@ -3338,7 +3357,4 @@ static const struct ceph_connection_operations mds_con_ops = {
 	.peer_reset = peer_reset,
 };
 
-
-
-
 /* eof */
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index c98267ce6d2a..d66d63c72355 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -8,9 +8,9 @@
 #include <linux/rbtree.h>
 #include <linux/spinlock.h>
 
-#include "types.h"
-#include "messenger.h"
-#include "mdsmap.h"
+#include <linux/ceph/types.h>
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/mdsmap.h>
 
 /*
  * Some lock dependencies:
@@ -26,7 +26,7 @@
  *
  */
 
-struct ceph_client;
+struct ceph_fs_client;
 struct ceph_cap;
 
 /*
@@ -230,7 +230,7 @@ struct ceph_mds_request {
  * mds client state
  */
 struct ceph_mds_client {
-	struct ceph_client      *client;
+	struct ceph_fs_client  *fsc;
 	struct mutex            mutex;         /* all nested structures */
 
 	struct ceph_mdsmap      *mdsmap;
@@ -289,11 +289,6 @@ struct ceph_mds_client {
 	int		caps_avail_count;    /* unused, unreserved */
 	int		caps_min_count;      /* keep at least this many
 						(unreserved) */
-
-#ifdef CONFIG_DEBUG_FS
-	struct dentry 	  *debugfs_file;
-#endif
-
 	spinlock_t	  dentry_lru_lock;
 	struct list_head  dentry_lru;
 	int		  num_dentry;
@@ -316,10 +311,9 @@ extern void ceph_put_mds_session(struct ceph_mds_session *s);
 extern int ceph_send_msg_mds(struct ceph_mds_client *mdsc,
 			     struct ceph_msg *msg, int mds);
 
-extern int ceph_mdsc_init(struct ceph_mds_client *mdsc,
-			   struct ceph_client *client);
+extern int ceph_mdsc_init(struct ceph_fs_client *fsc);
 extern void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc);
-extern void ceph_mdsc_stop(struct ceph_mds_client *mdsc);
+extern void ceph_mdsc_destroy(struct ceph_fs_client *fsc);
 
 extern void ceph_mdsc_sync(struct ceph_mds_client *mdsc);
 
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index 040be6d1150b..73b7d44e8a35 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/bug.h>
 #include <linux/err.h>
@@ -6,9 +6,9 @@
 #include <linux/slab.h>
 #include <linux/types.h>
 
-#include "mdsmap.h"
-#include "messenger.h"
-#include "decode.h"
+#include <linux/ceph/mdsmap.h>
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/decode.h>
 
 #include "super.h"
 
@@ -117,7 +117,8 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		}
 
 		dout("mdsmap_decode %d/%d %lld mds%d.%d %s %s\n",
-		     i+1, n, global_id, mds, inc, pr_addr(&addr.in_addr),
+		     i+1, n, global_id, mds, inc,
+		     ceph_pr_addr(&addr.in_addr),
 		     ceph_mds_state_name(state));
 		if (mds >= 0 && mds < m->m_max_mds && state > 0) {
 			m->m_info[mds].global_id = global_id;
diff --git a/fs/ceph/mdsmap.h b/fs/ceph/mdsmap.h
deleted file mode 100644
index 4c5cb0880bba..000000000000
--- a/fs/ceph/mdsmap.h
+++ /dev/null
@@ -1,62 +0,0 @@
-#ifndef _FS_CEPH_MDSMAP_H
-#define _FS_CEPH_MDSMAP_H
-
-#include "types.h"
-
-/*
- * mds map - describe servers in the mds cluster.
- *
- * we limit fields to those the client actually xcares about
- */
-struct ceph_mds_info {
-	u64 global_id;
-	struct ceph_entity_addr addr;
-	s32 state;
-	int num_export_targets;
-	bool laggy;
-	u32 *export_targets;
-};
-
-struct ceph_mdsmap {
-	u32 m_epoch, m_client_epoch, m_last_failure;
-	u32 m_root;
-	u32 m_session_timeout;          /* seconds */
-	u32 m_session_autoclose;        /* seconds */
-	u64 m_max_file_size;
-	u32 m_max_mds;                  /* size of m_addr, m_state arrays */
-	struct ceph_mds_info *m_info;
-
-	/* which object pools file data can be stored in */
-	int m_num_data_pg_pools;
-	u32 *m_data_pg_pools;
-	u32 m_cas_pg_pool;
-};
-
-static inline struct ceph_entity_addr *
-ceph_mdsmap_get_addr(struct ceph_mdsmap *m, int w)
-{
-	if (w >= m->m_max_mds)
-		return NULL;
-	return &m->m_info[w].addr;
-}
-
-static inline int ceph_mdsmap_get_state(struct ceph_mdsmap *m, int w)
-{
-	BUG_ON(w < 0);
-	if (w >= m->m_max_mds)
-		return CEPH_MDS_STATE_DNE;
-	return m->m_info[w].state;
-}
-
-static inline bool ceph_mdsmap_is_laggy(struct ceph_mdsmap *m, int w)
-{
-	if (w >= 0 && w < m->m_max_mds)
-		return m->m_info[w].laggy;
-	return false;
-}
-
-extern int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m);
-extern struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end);
-extern void ceph_mdsmap_destroy(struct ceph_mdsmap *m);
-
-#endif
diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c
deleted file mode 100644
index 17a09b32a591..000000000000
--- a/fs/ceph/messenger.c
+++ /dev/null
@@ -1,2432 +0,0 @@
-#include "ceph_debug.h"
-
-#include <linux/crc32c.h>
-#include <linux/ctype.h>
-#include <linux/highmem.h>
-#include <linux/inet.h>
-#include <linux/kthread.h>
-#include <linux/net.h>
-#include <linux/slab.h>
-#include <linux/socket.h>
-#include <linux/string.h>
-#include <linux/bio.h>
-#include <linux/blkdev.h>
-#include <net/tcp.h>
-
-#include "super.h"
-#include "messenger.h"
-#include "decode.h"
-#include "pagelist.h"
-
-/*
- * Ceph uses the messenger to exchange ceph_msg messages with other
- * hosts in the system.  The messenger provides ordered and reliable
- * delivery.  We tolerate TCP disconnects by reconnecting (with
- * exponential backoff) in the case of a fault (disconnection, bad
- * crc, protocol error).  Acks allow sent messages to be discarded by
- * the sender.
- */
-
-/* static tag bytes (protocol control messages) */
-static char tag_msg = CEPH_MSGR_TAG_MSG;
-static char tag_ack = CEPH_MSGR_TAG_ACK;
-static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE;
-
-#ifdef CONFIG_LOCKDEP
-static struct lock_class_key socket_class;
-#endif
-
-
-static void queue_con(struct ceph_connection *con);
-static void con_work(struct work_struct *);
-static void ceph_fault(struct ceph_connection *con);
-
-/*
- * nicely render a sockaddr as a string.
- */
-#define MAX_ADDR_STR 20
-#define MAX_ADDR_STR_LEN 60
-static char addr_str[MAX_ADDR_STR][MAX_ADDR_STR_LEN];
-static DEFINE_SPINLOCK(addr_str_lock);
-static int last_addr_str;
-
-const char *pr_addr(const struct sockaddr_storage *ss)
-{
-	int i;
-	char *s;
-	struct sockaddr_in *in4 = (void *)ss;
-	struct sockaddr_in6 *in6 = (void *)ss;
-
-	spin_lock(&addr_str_lock);
-	i = last_addr_str++;
-	if (last_addr_str == MAX_ADDR_STR)
-		last_addr_str = 0;
-	spin_unlock(&addr_str_lock);
-	s = addr_str[i];
-
-	switch (ss->ss_family) {
-	case AF_INET:
-		snprintf(s, MAX_ADDR_STR_LEN, "%pI4:%u", &in4->sin_addr,
-			 (unsigned int)ntohs(in4->sin_port));
-		break;
-
-	case AF_INET6:
-		snprintf(s, MAX_ADDR_STR_LEN, "[%pI6c]:%u", &in6->sin6_addr,
-			 (unsigned int)ntohs(in6->sin6_port));
-		break;
-
-	default:
-		sprintf(s, "(unknown sockaddr family %d)", (int)ss->ss_family);
-	}
-
-	return s;
-}
-
-static void encode_my_addr(struct ceph_messenger *msgr)
-{
-	memcpy(&msgr->my_enc_addr, &msgr->inst.addr, sizeof(msgr->my_enc_addr));
-	ceph_encode_addr(&msgr->my_enc_addr);
-}
-
-/*
- * work queue for all reading and writing to/from the socket.
- */
-struct workqueue_struct *ceph_msgr_wq;
-
-int __init ceph_msgr_init(void)
-{
-	ceph_msgr_wq = create_workqueue("ceph-msgr");
-	if (IS_ERR(ceph_msgr_wq)) {
-		int ret = PTR_ERR(ceph_msgr_wq);
-		pr_err("msgr_init failed to create workqueue: %d\n", ret);
-		ceph_msgr_wq = NULL;
-		return ret;
-	}
-	return 0;
-}
-
-void ceph_msgr_exit(void)
-{
-	destroy_workqueue(ceph_msgr_wq);
-}
-
-void ceph_msgr_flush(void)
-{
-	flush_workqueue(ceph_msgr_wq);
-}
-
-
-/*
- * socket callback functions
- */
-
-/* data available on socket, or listen socket received a connect */
-static void ceph_data_ready(struct sock *sk, int count_unused)
-{
-	struct ceph_connection *con =
-		(struct ceph_connection *)sk->sk_user_data;
-	if (sk->sk_state != TCP_CLOSE_WAIT) {
-		dout("ceph_data_ready on %p state = %lu, queueing work\n",
-		     con, con->state);
-		queue_con(con);
-	}
-}
-
-/* socket has buffer space for writing */
-static void ceph_write_space(struct sock *sk)
-{
-	struct ceph_connection *con =
-		(struct ceph_connection *)sk->sk_user_data;
-
-	/* only queue to workqueue if there is data we want to write. */
-	if (test_bit(WRITE_PENDING, &con->state)) {
-		dout("ceph_write_space %p queueing write work\n", con);
-		queue_con(con);
-	} else {
-		dout("ceph_write_space %p nothing to write\n", con);
-	}
-
-	/* since we have our own write_space, clear the SOCK_NOSPACE flag */
-	clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
-}
-
-/* socket's state has changed */
-static void ceph_state_change(struct sock *sk)
-{
-	struct ceph_connection *con =
-		(struct ceph_connection *)sk->sk_user_data;
-
-	dout("ceph_state_change %p state = %lu sk_state = %u\n",
-	     con, con->state, sk->sk_state);
-
-	if (test_bit(CLOSED, &con->state))
-		return;
-
-	switch (sk->sk_state) {
-	case TCP_CLOSE:
-		dout("ceph_state_change TCP_CLOSE\n");
-	case TCP_CLOSE_WAIT:
-		dout("ceph_state_change TCP_CLOSE_WAIT\n");
-		if (test_and_set_bit(SOCK_CLOSED, &con->state) == 0) {
-			if (test_bit(CONNECTING, &con->state))
-				con->error_msg = "connection failed";
-			else
-				con->error_msg = "socket closed";
-			queue_con(con);
-		}
-		break;
-	case TCP_ESTABLISHED:
-		dout("ceph_state_change TCP_ESTABLISHED\n");
-		queue_con(con);
-		break;
-	}
-}
-
-/*
- * set up socket callbacks
- */
-static void set_sock_callbacks(struct socket *sock,
-			       struct ceph_connection *con)
-{
-	struct sock *sk = sock->sk;
-	sk->sk_user_data = (void *)con;
-	sk->sk_data_ready = ceph_data_ready;
-	sk->sk_write_space = ceph_write_space;
-	sk->sk_state_change = ceph_state_change;
-}
-
-
-/*
- * socket helpers
- */
-
-/*
- * initiate connection to a remote socket.
- */
-static struct socket *ceph_tcp_connect(struct ceph_connection *con)
-{
-	struct sockaddr_storage *paddr = &con->peer_addr.in_addr;
-	struct socket *sock;
-	int ret;
-
-	BUG_ON(con->sock);
-	ret = sock_create_kern(con->peer_addr.in_addr.ss_family, SOCK_STREAM,
-			       IPPROTO_TCP, &sock);
-	if (ret)
-		return ERR_PTR(ret);
-	con->sock = sock;
-	sock->sk->sk_allocation = GFP_NOFS;
-
-#ifdef CONFIG_LOCKDEP
-	lockdep_set_class(&sock->sk->sk_lock, &socket_class);
-#endif
-
-	set_sock_callbacks(sock, con);
-
-	dout("connect %s\n", pr_addr(&con->peer_addr.in_addr));
-
-	ret = sock->ops->connect(sock, (struct sockaddr *)paddr, sizeof(*paddr),
-				 O_NONBLOCK);
-	if (ret == -EINPROGRESS) {
-		dout("connect %s EINPROGRESS sk_state = %u\n",
-		     pr_addr(&con->peer_addr.in_addr),
-		     sock->sk->sk_state);
-		ret = 0;
-	}
-	if (ret < 0) {
-		pr_err("connect %s error %d\n",
-		       pr_addr(&con->peer_addr.in_addr), ret);
-		sock_release(sock);
-		con->sock = NULL;
-		con->error_msg = "connect error";
-	}
-
-	if (ret < 0)
-		return ERR_PTR(ret);
-	return sock;
-}
-
-static int ceph_tcp_recvmsg(struct socket *sock, void *buf, size_t len)
-{
-	struct kvec iov = {buf, len};
-	struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL };
-
-	return kernel_recvmsg(sock, &msg, &iov, 1, len, msg.msg_flags);
-}
-
-/*
- * write something.  @more is true if caller will be sending more data
- * shortly.
- */
-static int ceph_tcp_sendmsg(struct socket *sock, struct kvec *iov,
-		     size_t kvlen, size_t len, int more)
-{
-	struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL };
-
-	if (more)
-		msg.msg_flags |= MSG_MORE;
-	else
-		msg.msg_flags |= MSG_EOR;  /* superfluous, but what the hell */
-
-	return kernel_sendmsg(sock, &msg, iov, kvlen, len);
-}
-
-
-/*
- * Shutdown/close the socket for the given connection.
- */
-static int con_close_socket(struct ceph_connection *con)
-{
-	int rc;
-
-	dout("con_close_socket on %p sock %p\n", con, con->sock);
-	if (!con->sock)
-		return 0;
-	set_bit(SOCK_CLOSED, &con->state);
-	rc = con->sock->ops->shutdown(con->sock, SHUT_RDWR);
-	sock_release(con->sock);
-	con->sock = NULL;
-	clear_bit(SOCK_CLOSED, &con->state);
-	return rc;
-}
-
-/*
- * Reset a connection.  Discard all incoming and outgoing messages
- * and clear *_seq state.
- */
-static void ceph_msg_remove(struct ceph_msg *msg)
-{
-	list_del_init(&msg->list_head);
-	ceph_msg_put(msg);
-}
-static void ceph_msg_remove_list(struct list_head *head)
-{
-	while (!list_empty(head)) {
-		struct ceph_msg *msg = list_first_entry(head, struct ceph_msg,
-							list_head);
-		ceph_msg_remove(msg);
-	}
-}
-
-static void reset_connection(struct ceph_connection *con)
-{
-	/* reset connection, out_queue, msg_ and connect_seq */
-	/* discard existing out_queue and msg_seq */
-	ceph_msg_remove_list(&con->out_queue);
-	ceph_msg_remove_list(&con->out_sent);
-
-	if (con->in_msg) {
-		ceph_msg_put(con->in_msg);
-		con->in_msg = NULL;
-	}
-
-	con->connect_seq = 0;
-	con->out_seq = 0;
-	if (con->out_msg) {
-		ceph_msg_put(con->out_msg);
-		con->out_msg = NULL;
-	}
-	con->out_keepalive_pending = false;
-	con->in_seq = 0;
-	con->in_seq_acked = 0;
-}
-
-/*
- * mark a peer down.  drop any open connections.
- */
-void ceph_con_close(struct ceph_connection *con)
-{
-	dout("con_close %p peer %s\n", con, pr_addr(&con->peer_addr.in_addr));
-	set_bit(CLOSED, &con->state);  /* in case there's queued work */
-	clear_bit(STANDBY, &con->state);  /* avoid connect_seq bump */
-	clear_bit(LOSSYTX, &con->state);  /* so we retry next connect */
-	clear_bit(KEEPALIVE_PENDING, &con->state);
-	clear_bit(WRITE_PENDING, &con->state);
-	mutex_lock(&con->mutex);
-	reset_connection(con);
-	con->peer_global_seq = 0;
-	cancel_delayed_work(&con->work);
-	mutex_unlock(&con->mutex);
-	queue_con(con);
-}
-
-/*
- * Reopen a closed connection, with a new peer address.
- */
-void ceph_con_open(struct ceph_connection *con, struct ceph_entity_addr *addr)
-{
-	dout("con_open %p %s\n", con, pr_addr(&addr->in_addr));
-	set_bit(OPENING, &con->state);
-	clear_bit(CLOSED, &con->state);
-	memcpy(&con->peer_addr, addr, sizeof(*addr));
-	con->delay = 0;      /* reset backoff memory */
-	queue_con(con);
-}
-
-/*
- * return true if this connection ever successfully opened
- */
-bool ceph_con_opened(struct ceph_connection *con)
-{
-	return con->connect_seq > 0;
-}
-
-/*
- * generic get/put
- */
-struct ceph_connection *ceph_con_get(struct ceph_connection *con)
-{
-	dout("con_get %p nref = %d -> %d\n", con,
-	     atomic_read(&con->nref), atomic_read(&con->nref) + 1);
-	if (atomic_inc_not_zero(&con->nref))
-		return con;
-	return NULL;
-}
-
-void ceph_con_put(struct ceph_connection *con)
-{
-	dout("con_put %p nref = %d -> %d\n", con,
-	     atomic_read(&con->nref), atomic_read(&con->nref) - 1);
-	BUG_ON(atomic_read(&con->nref) == 0);
-	if (atomic_dec_and_test(&con->nref)) {
-		BUG_ON(con->sock);
-		kfree(con);
-	}
-}
-
-/*
- * initialize a new connection.
- */
-void ceph_con_init(struct ceph_messenger *msgr, struct ceph_connection *con)
-{
-	dout("con_init %p\n", con);
-	memset(con, 0, sizeof(*con));
-	atomic_set(&con->nref, 1);
-	con->msgr = msgr;
-	mutex_init(&con->mutex);
-	INIT_LIST_HEAD(&con->out_queue);
-	INIT_LIST_HEAD(&con->out_sent);
-	INIT_DELAYED_WORK(&con->work, con_work);
-}
-
-
-/*
- * We maintain a global counter to order connection attempts.  Get
- * a unique seq greater than @gt.
- */
-static u32 get_global_seq(struct ceph_messenger *msgr, u32 gt)
-{
-	u32 ret;
-
-	spin_lock(&msgr->global_seq_lock);
-	if (msgr->global_seq < gt)
-		msgr->global_seq = gt;
-	ret = ++msgr->global_seq;
-	spin_unlock(&msgr->global_seq_lock);
-	return ret;
-}
-
-
-/*
- * Prepare footer for currently outgoing message, and finish things
- * off.  Assumes out_kvec* are already valid.. we just add on to the end.
- */
-static void prepare_write_message_footer(struct ceph_connection *con, int v)
-{
-	struct ceph_msg *m = con->out_msg;
-
-	dout("prepare_write_message_footer %p\n", con);
-	con->out_kvec_is_msg = true;
-	con->out_kvec[v].iov_base = &m->footer;
-	con->out_kvec[v].iov_len = sizeof(m->footer);
-	con->out_kvec_bytes += sizeof(m->footer);
-	con->out_kvec_left++;
-	con->out_more = m->more_to_follow;
-	con->out_msg_done = true;
-}
-
-/*
- * Prepare headers for the next outgoing message.
- */
-static void prepare_write_message(struct ceph_connection *con)
-{
-	struct ceph_msg *m;
-	int v = 0;
-
-	con->out_kvec_bytes = 0;
-	con->out_kvec_is_msg = true;
-	con->out_msg_done = false;
-
-	/* Sneak an ack in there first?  If we can get it into the same
-	 * TCP packet that's a good thing. */
-	if (con->in_seq > con->in_seq_acked) {
-		con->in_seq_acked = con->in_seq;
-		con->out_kvec[v].iov_base = &tag_ack;
-		con->out_kvec[v++].iov_len = 1;
-		con->out_temp_ack = cpu_to_le64(con->in_seq_acked);
-		con->out_kvec[v].iov_base = &con->out_temp_ack;
-		con->out_kvec[v++].iov_len = sizeof(con->out_temp_ack);
-		con->out_kvec_bytes = 1 + sizeof(con->out_temp_ack);
-	}
-
-	m = list_first_entry(&con->out_queue,
-		       struct ceph_msg, list_head);
-	con->out_msg = m;
-	if (test_bit(LOSSYTX, &con->state)) {
-		list_del_init(&m->list_head);
-	} else {
-		/* put message on sent list */
-		ceph_msg_get(m);
-		list_move_tail(&m->list_head, &con->out_sent);
-	}
-
-	/*
-	 * only assign outgoing seq # if we haven't sent this message
-	 * yet.  if it is requeued, resend with it's original seq.
-	 */
-	if (m->needs_out_seq) {
-		m->hdr.seq = cpu_to_le64(++con->out_seq);
-		m->needs_out_seq = false;
-	}
-
-	dout("prepare_write_message %p seq %lld type %d len %d+%d+%d %d pgs\n",
-	     m, con->out_seq, le16_to_cpu(m->hdr.type),
-	     le32_to_cpu(m->hdr.front_len), le32_to_cpu(m->hdr.middle_len),
-	     le32_to_cpu(m->hdr.data_len),
-	     m->nr_pages);
-	BUG_ON(le32_to_cpu(m->hdr.front_len) != m->front.iov_len);
-
-	/* tag + hdr + front + middle */
-	con->out_kvec[v].iov_base = &tag_msg;
-	con->out_kvec[v++].iov_len = 1;
-	con->out_kvec[v].iov_base = &m->hdr;
-	con->out_kvec[v++].iov_len = sizeof(m->hdr);
-	con->out_kvec[v++] = m->front;
-	if (m->middle)
-		con->out_kvec[v++] = m->middle->vec;
-	con->out_kvec_left = v;
-	con->out_kvec_bytes += 1 + sizeof(m->hdr) + m->front.iov_len +
-		(m->middle ? m->middle->vec.iov_len : 0);
-	con->out_kvec_cur = con->out_kvec;
-
-	/* fill in crc (except data pages), footer */
-	con->out_msg->hdr.crc =
-		cpu_to_le32(crc32c(0, (void *)&m->hdr,
-				      sizeof(m->hdr) - sizeof(m->hdr.crc)));
-	con->out_msg->footer.flags = CEPH_MSG_FOOTER_COMPLETE;
-	con->out_msg->footer.front_crc =
-		cpu_to_le32(crc32c(0, m->front.iov_base, m->front.iov_len));
-	if (m->middle)
-		con->out_msg->footer.middle_crc =
-			cpu_to_le32(crc32c(0, m->middle->vec.iov_base,
-					   m->middle->vec.iov_len));
-	else
-		con->out_msg->footer.middle_crc = 0;
-	con->out_msg->footer.data_crc = 0;
-	dout("prepare_write_message front_crc %u data_crc %u\n",
-	     le32_to_cpu(con->out_msg->footer.front_crc),
-	     le32_to_cpu(con->out_msg->footer.middle_crc));
-
-	/* is there a data payload? */
-	if (le32_to_cpu(m->hdr.data_len) > 0) {
-		/* initialize page iterator */
-		con->out_msg_pos.page = 0;
-		if (m->pages)
-			con->out_msg_pos.page_pos =
-				le16_to_cpu(m->hdr.data_off) & ~PAGE_MASK;
-		else
-			con->out_msg_pos.page_pos = 0;
-		con->out_msg_pos.data_pos = 0;
-		con->out_msg_pos.did_page_crc = 0;
-		con->out_more = 1;  /* data + footer will follow */
-	} else {
-		/* no, queue up footer too and be done */
-		prepare_write_message_footer(con, v);
-	}
-
-	set_bit(WRITE_PENDING, &con->state);
-}
-
-/*
- * Prepare an ack.
- */
-static void prepare_write_ack(struct ceph_connection *con)
-{
-	dout("prepare_write_ack %p %llu -> %llu\n", con,
-	     con->in_seq_acked, con->in_seq);
-	con->in_seq_acked = con->in_seq;
-
-	con->out_kvec[0].iov_base = &tag_ack;
-	con->out_kvec[0].iov_len = 1;
-	con->out_temp_ack = cpu_to_le64(con->in_seq_acked);
-	con->out_kvec[1].iov_base = &con->out_temp_ack;
-	con->out_kvec[1].iov_len = sizeof(con->out_temp_ack);
-	con->out_kvec_left = 2;
-	con->out_kvec_bytes = 1 + sizeof(con->out_temp_ack);
-	con->out_kvec_cur = con->out_kvec;
-	con->out_more = 1;  /* more will follow.. eventually.. */
-	set_bit(WRITE_PENDING, &con->state);
-}
-
-/*
- * Prepare to write keepalive byte.
- */
-static void prepare_write_keepalive(struct ceph_connection *con)
-{
-	dout("prepare_write_keepalive %p\n", con);
-	con->out_kvec[0].iov_base = &tag_keepalive;
-	con->out_kvec[0].iov_len = 1;
-	con->out_kvec_left = 1;
-	con->out_kvec_bytes = 1;
-	con->out_kvec_cur = con->out_kvec;
-	set_bit(WRITE_PENDING, &con->state);
-}
-
-/*
- * Connection negotiation.
- */
-
-static void prepare_connect_authorizer(struct ceph_connection *con)
-{
-	void *auth_buf;
-	int auth_len = 0;
-	int auth_protocol = 0;
-
-	mutex_unlock(&con->mutex);
-	if (con->ops->get_authorizer)
-		con->ops->get_authorizer(con, &auth_buf, &auth_len,
-					 &auth_protocol, &con->auth_reply_buf,
-					 &con->auth_reply_buf_len,
-					 con->auth_retry);
-	mutex_lock(&con->mutex);
-
-	con->out_connect.authorizer_protocol = cpu_to_le32(auth_protocol);
-	con->out_connect.authorizer_len = cpu_to_le32(auth_len);
-
-	con->out_kvec[con->out_kvec_left].iov_base = auth_buf;
-	con->out_kvec[con->out_kvec_left].iov_len = auth_len;
-	con->out_kvec_left++;
-	con->out_kvec_bytes += auth_len;
-}
-
-/*
- * We connected to a peer and are saying hello.
- */
-static void prepare_write_banner(struct ceph_messenger *msgr,
-				 struct ceph_connection *con)
-{
-	int len = strlen(CEPH_BANNER);
-
-	con->out_kvec[0].iov_base = CEPH_BANNER;
-	con->out_kvec[0].iov_len = len;
-	con->out_kvec[1].iov_base = &msgr->my_enc_addr;
-	con->out_kvec[1].iov_len = sizeof(msgr->my_enc_addr);
-	con->out_kvec_left = 2;
-	con->out_kvec_bytes = len + sizeof(msgr->my_enc_addr);
-	con->out_kvec_cur = con->out_kvec;
-	con->out_more = 0;
-	set_bit(WRITE_PENDING, &con->state);
-}
-
-static void prepare_write_connect(struct ceph_messenger *msgr,
-				  struct ceph_connection *con,
-				  int after_banner)
-{
-	unsigned global_seq = get_global_seq(con->msgr, 0);
-	int proto;
-
-	switch (con->peer_name.type) {
-	case CEPH_ENTITY_TYPE_MON:
-		proto = CEPH_MONC_PROTOCOL;
-		break;
-	case CEPH_ENTITY_TYPE_OSD:
-		proto = CEPH_OSDC_PROTOCOL;
-		break;
-	case CEPH_ENTITY_TYPE_MDS:
-		proto = CEPH_MDSC_PROTOCOL;
-		break;
-	default:
-		BUG();
-	}
-
-	dout("prepare_write_connect %p cseq=%d gseq=%d proto=%d\n", con,
-	     con->connect_seq, global_seq, proto);
-
-	con->out_connect.features = cpu_to_le64(CEPH_FEATURE_SUPPORTED);
-	con->out_connect.host_type = cpu_to_le32(CEPH_ENTITY_TYPE_CLIENT);
-	con->out_connect.connect_seq = cpu_to_le32(con->connect_seq);
-	con->out_connect.global_seq = cpu_to_le32(global_seq);
-	con->out_connect.protocol_version = cpu_to_le32(proto);
-	con->out_connect.flags = 0;
-
-	if (!after_banner) {
-		con->out_kvec_left = 0;
-		con->out_kvec_bytes = 0;
-	}
-	con->out_kvec[con->out_kvec_left].iov_base = &con->out_connect;
-	con->out_kvec[con->out_kvec_left].iov_len = sizeof(con->out_connect);
-	con->out_kvec_left++;
-	con->out_kvec_bytes += sizeof(con->out_connect);
-	con->out_kvec_cur = con->out_kvec;
-	con->out_more = 0;
-	set_bit(WRITE_PENDING, &con->state);
-
-	prepare_connect_authorizer(con);
-}
-
-
-/*
- * write as much of pending kvecs to the socket as we can.
- *  1 -> done
- *  0 -> socket full, but more to do
- * <0 -> error
- */
-static int write_partial_kvec(struct ceph_connection *con)
-{
-	int ret;
-
-	dout("write_partial_kvec %p %d left\n", con, con->out_kvec_bytes);
-	while (con->out_kvec_bytes > 0) {
-		ret = ceph_tcp_sendmsg(con->sock, con->out_kvec_cur,
-				       con->out_kvec_left, con->out_kvec_bytes,
-				       con->out_more);
-		if (ret <= 0)
-			goto out;
-		con->out_kvec_bytes -= ret;
-		if (con->out_kvec_bytes == 0)
-			break;            /* done */
-		while (ret > 0) {
-			if (ret >= con->out_kvec_cur->iov_len) {
-				ret -= con->out_kvec_cur->iov_len;
-				con->out_kvec_cur++;
-				con->out_kvec_left--;
-			} else {
-				con->out_kvec_cur->iov_len -= ret;
-				con->out_kvec_cur->iov_base += ret;
-				ret = 0;
-				break;
-			}
-		}
-	}
-	con->out_kvec_left = 0;
-	con->out_kvec_is_msg = false;
-	ret = 1;
-out:
-	dout("write_partial_kvec %p %d left in %d kvecs ret = %d\n", con,
-	     con->out_kvec_bytes, con->out_kvec_left, ret);
-	return ret;  /* done! */
-}
-
-#ifdef CONFIG_BLOCK
-static void init_bio_iter(struct bio *bio, struct bio **iter, int *seg)
-{
-	if (!bio) {
-		*iter = NULL;
-		*seg = 0;
-		return;
-	}
-	*iter = bio;
-	*seg = bio->bi_idx;
-}
-
-static void iter_bio_next(struct bio **bio_iter, int *seg)
-{
-	if (*bio_iter == NULL)
-		return;
-
-	BUG_ON(*seg >= (*bio_iter)->bi_vcnt);
-
-	(*seg)++;
-	if (*seg == (*bio_iter)->bi_vcnt)
-		init_bio_iter((*bio_iter)->bi_next, bio_iter, seg);
-}
-#endif
-
-/*
- * Write as much message data payload as we can.  If we finish, queue
- * up the footer.
- *  1 -> done, footer is now queued in out_kvec[].
- *  0 -> socket full, but more to do
- * <0 -> error
- */
-static int write_partial_msg_pages(struct ceph_connection *con)
-{
-	struct ceph_msg *msg = con->out_msg;
-	unsigned data_len = le32_to_cpu(msg->hdr.data_len);
-	size_t len;
-	int crc = con->msgr->nocrc;
-	int ret;
-	int total_max_write;
-	int in_trail = 0;
-	size_t trail_len = (msg->trail ? msg->trail->length : 0);
-
-	dout("write_partial_msg_pages %p msg %p page %d/%d offset %d\n",
-	     con, con->out_msg, con->out_msg_pos.page, con->out_msg->nr_pages,
-	     con->out_msg_pos.page_pos);
-
-#ifdef CONFIG_BLOCK
-	if (msg->bio && !msg->bio_iter)
-		init_bio_iter(msg->bio, &msg->bio_iter, &msg->bio_seg);
-#endif
-
-	while (data_len > con->out_msg_pos.data_pos) {
-		struct page *page = NULL;
-		void *kaddr = NULL;
-		int max_write = PAGE_SIZE;
-		int page_shift = 0;
-
-		total_max_write = data_len - trail_len -
-			con->out_msg_pos.data_pos;
-
-		/*
-		 * if we are calculating the data crc (the default), we need
-		 * to map the page.  if our pages[] has been revoked, use the
-		 * zero page.
-		 */
-
-		/* have we reached the trail part of the data? */
-		if (con->out_msg_pos.data_pos >= data_len - trail_len) {
-			in_trail = 1;
-
-			total_max_write = data_len - con->out_msg_pos.data_pos;
-
-			page = list_first_entry(&msg->trail->head,
-						struct page, lru);
-			if (crc)
-				kaddr = kmap(page);
-			max_write = PAGE_SIZE;
-		} else if (msg->pages) {
-			page = msg->pages[con->out_msg_pos.page];
-			if (crc)
-				kaddr = kmap(page);
-		} else if (msg->pagelist) {
-			page = list_first_entry(&msg->pagelist->head,
-						struct page, lru);
-			if (crc)
-				kaddr = kmap(page);
-#ifdef CONFIG_BLOCK
-		} else if (msg->bio) {
-			struct bio_vec *bv;
-
-			bv = bio_iovec_idx(msg->bio_iter, msg->bio_seg);
-			page = bv->bv_page;
-			page_shift = bv->bv_offset;
-			if (crc)
-				kaddr = kmap(page) + page_shift;
-			max_write = bv->bv_len;
-#endif
-		} else {
-			page = con->msgr->zero_page;
-			if (crc)
-				kaddr = page_address(con->msgr->zero_page);
-		}
-		len = min_t(int, max_write - con->out_msg_pos.page_pos,
-			    total_max_write);
-
-		if (crc && !con->out_msg_pos.did_page_crc) {
-			void *base = kaddr + con->out_msg_pos.page_pos;
-			u32 tmpcrc = le32_to_cpu(con->out_msg->footer.data_crc);
-
-			BUG_ON(kaddr == NULL);
-			con->out_msg->footer.data_crc =
-				cpu_to_le32(crc32c(tmpcrc, base, len));
-			con->out_msg_pos.did_page_crc = 1;
-		}
-		ret = kernel_sendpage(con->sock, page,
-				      con->out_msg_pos.page_pos + page_shift,
-				      len,
-				      MSG_DONTWAIT | MSG_NOSIGNAL |
-				      MSG_MORE);
-
-		if (crc &&
-		    (msg->pages || msg->pagelist || msg->bio || in_trail))
-			kunmap(page);
-
-		if (ret <= 0)
-			goto out;
-
-		con->out_msg_pos.data_pos += ret;
-		con->out_msg_pos.page_pos += ret;
-		if (ret == len) {
-			con->out_msg_pos.page_pos = 0;
-			con->out_msg_pos.page++;
-			con->out_msg_pos.did_page_crc = 0;
-			if (in_trail)
-				list_move_tail(&page->lru,
-					       &msg->trail->head);
-			else if (msg->pagelist)
-				list_move_tail(&page->lru,
-					       &msg->pagelist->head);
-#ifdef CONFIG_BLOCK
-			else if (msg->bio)
-				iter_bio_next(&msg->bio_iter, &msg->bio_seg);
-#endif
-		}
-	}
-
-	dout("write_partial_msg_pages %p msg %p done\n", con, msg);
-
-	/* prepare and queue up footer, too */
-	if (!crc)
-		con->out_msg->footer.flags |= CEPH_MSG_FOOTER_NOCRC;
-	con->out_kvec_bytes = 0;
-	con->out_kvec_left = 0;
-	con->out_kvec_cur = con->out_kvec;
-	prepare_write_message_footer(con, 0);
-	ret = 1;
-out:
-	return ret;
-}
-
-/*
- * write some zeros
- */
-static int write_partial_skip(struct ceph_connection *con)
-{
-	int ret;
-
-	while (con->out_skip > 0) {
-		struct kvec iov = {
-			.iov_base = page_address(con->msgr->zero_page),
-			.iov_len = min(con->out_skip, (int)PAGE_CACHE_SIZE)
-		};
-
-		ret = ceph_tcp_sendmsg(con->sock, &iov, 1, iov.iov_len, 1);
-		if (ret <= 0)
-			goto out;
-		con->out_skip -= ret;
-	}
-	ret = 1;
-out:
-	return ret;
-}
-
-/*
- * Prepare to read connection handshake, or an ack.
- */
-static void prepare_read_banner(struct ceph_connection *con)
-{
-	dout("prepare_read_banner %p\n", con);
-	con->in_base_pos = 0;
-}
-
-static void prepare_read_connect(struct ceph_connection *con)
-{
-	dout("prepare_read_connect %p\n", con);
-	con->in_base_pos = 0;
-}
-
-static void prepare_read_ack(struct ceph_connection *con)
-{
-	dout("prepare_read_ack %p\n", con);
-	con->in_base_pos = 0;
-}
-
-static void prepare_read_tag(struct ceph_connection *con)
-{
-	dout("prepare_read_tag %p\n", con);
-	con->in_base_pos = 0;
-	con->in_tag = CEPH_MSGR_TAG_READY;
-}
-
-/*
- * Prepare to read a message.
- */
-static int prepare_read_message(struct ceph_connection *con)
-{
-	dout("prepare_read_message %p\n", con);
-	BUG_ON(con->in_msg != NULL);
-	con->in_base_pos = 0;
-	con->in_front_crc = con->in_middle_crc = con->in_data_crc = 0;
-	return 0;
-}
-
-
-static int read_partial(struct ceph_connection *con,
-			int *to, int size, void *object)
-{
-	*to += size;
-	while (con->in_base_pos < *to) {
-		int left = *to - con->in_base_pos;
-		int have = size - left;
-		int ret = ceph_tcp_recvmsg(con->sock, object + have, left);
-		if (ret <= 0)
-			return ret;
-		con->in_base_pos += ret;
-	}
-	return 1;
-}
-
-
-/*
- * Read all or part of the connect-side handshake on a new connection
- */
-static int read_partial_banner(struct ceph_connection *con)
-{
-	int ret, to = 0;
-
-	dout("read_partial_banner %p at %d\n", con, con->in_base_pos);
-
-	/* peer's banner */
-	ret = read_partial(con, &to, strlen(CEPH_BANNER), con->in_banner);
-	if (ret <= 0)
-		goto out;
-	ret = read_partial(con, &to, sizeof(con->actual_peer_addr),
-			   &con->actual_peer_addr);
-	if (ret <= 0)
-		goto out;
-	ret = read_partial(con, &to, sizeof(con->peer_addr_for_me),
-			   &con->peer_addr_for_me);
-	if (ret <= 0)
-		goto out;
-out:
-	return ret;
-}
-
-static int read_partial_connect(struct ceph_connection *con)
-{
-	int ret, to = 0;
-
-	dout("read_partial_connect %p at %d\n", con, con->in_base_pos);
-
-	ret = read_partial(con, &to, sizeof(con->in_reply), &con->in_reply);
-	if (ret <= 0)
-		goto out;
-	ret = read_partial(con, &to, le32_to_cpu(con->in_reply.authorizer_len),
-			   con->auth_reply_buf);
-	if (ret <= 0)
-		goto out;
-
-	dout("read_partial_connect %p tag %d, con_seq = %u, g_seq = %u\n",
-	     con, (int)con->in_reply.tag,
-	     le32_to_cpu(con->in_reply.connect_seq),
-	     le32_to_cpu(con->in_reply.global_seq));
-out:
-	return ret;
-
-}
-
-/*
- * Verify the hello banner looks okay.
- */
-static int verify_hello(struct ceph_connection *con)
-{
-	if (memcmp(con->in_banner, CEPH_BANNER, strlen(CEPH_BANNER))) {
-		pr_err("connect to %s got bad banner\n",
-		       pr_addr(&con->peer_addr.in_addr));
-		con->error_msg = "protocol error, bad banner";
-		return -1;
-	}
-	return 0;
-}
-
-static bool addr_is_blank(struct sockaddr_storage *ss)
-{
-	switch (ss->ss_family) {
-	case AF_INET:
-		return ((struct sockaddr_in *)ss)->sin_addr.s_addr == 0;
-	case AF_INET6:
-		return
-		     ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[0] == 0 &&
-		     ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[1] == 0 &&
-		     ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[2] == 0 &&
-		     ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[3] == 0;
-	}
-	return false;
-}
-
-static int addr_port(struct sockaddr_storage *ss)
-{
-	switch (ss->ss_family) {
-	case AF_INET:
-		return ntohs(((struct sockaddr_in *)ss)->sin_port);
-	case AF_INET6:
-		return ntohs(((struct sockaddr_in6 *)ss)->sin6_port);
-	}
-	return 0;
-}
-
-static void addr_set_port(struct sockaddr_storage *ss, int p)
-{
-	switch (ss->ss_family) {
-	case AF_INET:
-		((struct sockaddr_in *)ss)->sin_port = htons(p);
-	case AF_INET6:
-		((struct sockaddr_in6 *)ss)->sin6_port = htons(p);
-	}
-}
-
-/*
- * Parse an ip[:port] list into an addr array.  Use the default
- * monitor port if a port isn't specified.
- */
-int ceph_parse_ips(const char *c, const char *end,
-		   struct ceph_entity_addr *addr,
-		   int max_count, int *count)
-{
-	int i;
-	const char *p = c;
-
-	dout("parse_ips on '%.*s'\n", (int)(end-c), c);
-	for (i = 0; i < max_count; i++) {
-		const char *ipend;
-		struct sockaddr_storage *ss = &addr[i].in_addr;
-		struct sockaddr_in *in4 = (void *)ss;
-		struct sockaddr_in6 *in6 = (void *)ss;
-		int port;
-		char delim = ',';
-
-		if (*p == '[') {
-			delim = ']';
-			p++;
-		}
-
-		memset(ss, 0, sizeof(*ss));
-		if (in4_pton(p, end - p, (u8 *)&in4->sin_addr.s_addr,
-			     delim, &ipend))
-			ss->ss_family = AF_INET;
-		else if (in6_pton(p, end - p, (u8 *)&in6->sin6_addr.s6_addr,
-				  delim, &ipend))
-			ss->ss_family = AF_INET6;
-		else
-			goto bad;
-		p = ipend;
-
-		if (delim == ']') {
-			if (*p != ']') {
-				dout("missing matching ']'\n");
-				goto bad;
-			}
-			p++;
-		}
-
-		/* port? */
-		if (p < end && *p == ':') {
-			port = 0;
-			p++;
-			while (p < end && *p >= '0' && *p <= '9') {
-				port = (port * 10) + (*p - '0');
-				p++;
-			}
-			if (port > 65535 || port == 0)
-				goto bad;
-		} else {
-			port = CEPH_MON_PORT;
-		}
-
-		addr_set_port(ss, port);
-
-		dout("parse_ips got %s\n", pr_addr(ss));
-
-		if (p == end)
-			break;
-		if (*p != ',')
-			goto bad;
-		p++;
-	}
-
-	if (p != end)
-		goto bad;
-
-	if (count)
-		*count = i + 1;
-	return 0;
-
-bad:
-	pr_err("parse_ips bad ip '%.*s'\n", (int)(end - c), c);
-	return -EINVAL;
-}
-
-static int process_banner(struct ceph_connection *con)
-{
-	dout("process_banner on %p\n", con);
-
-	if (verify_hello(con) < 0)
-		return -1;
-
-	ceph_decode_addr(&con->actual_peer_addr);
-	ceph_decode_addr(&con->peer_addr_for_me);
-
-	/*
-	 * Make sure the other end is who we wanted.  note that the other
-	 * end may not yet know their ip address, so if it's 0.0.0.0, give
-	 * them the benefit of the doubt.
-	 */
-	if (memcmp(&con->peer_addr, &con->actual_peer_addr,
-		   sizeof(con->peer_addr)) != 0 &&
-	    !(addr_is_blank(&con->actual_peer_addr.in_addr) &&
-	      con->actual_peer_addr.nonce == con->peer_addr.nonce)) {
-		pr_warning("wrong peer, want %s/%d, got %s/%d\n",
-			   pr_addr(&con->peer_addr.in_addr),
-			   (int)le32_to_cpu(con->peer_addr.nonce),
-			   pr_addr(&con->actual_peer_addr.in_addr),
-			   (int)le32_to_cpu(con->actual_peer_addr.nonce));
-		con->error_msg = "wrong peer at address";
-		return -1;
-	}
-
-	/*
-	 * did we learn our address?
-	 */
-	if (addr_is_blank(&con->msgr->inst.addr.in_addr)) {
-		int port = addr_port(&con->msgr->inst.addr.in_addr);
-
-		memcpy(&con->msgr->inst.addr.in_addr,
-		       &con->peer_addr_for_me.in_addr,
-		       sizeof(con->peer_addr_for_me.in_addr));
-		addr_set_port(&con->msgr->inst.addr.in_addr, port);
-		encode_my_addr(con->msgr);
-		dout("process_banner learned my addr is %s\n",
-		     pr_addr(&con->msgr->inst.addr.in_addr));
-	}
-
-	set_bit(NEGOTIATING, &con->state);
-	prepare_read_connect(con);
-	return 0;
-}
-
-static void fail_protocol(struct ceph_connection *con)
-{
-	reset_connection(con);
-	set_bit(CLOSED, &con->state);  /* in case there's queued work */
-
-	mutex_unlock(&con->mutex);
-	if (con->ops->bad_proto)
-		con->ops->bad_proto(con);
-	mutex_lock(&con->mutex);
-}
-
-static int process_connect(struct ceph_connection *con)
-{
-	u64 sup_feat = CEPH_FEATURE_SUPPORTED;
-	u64 req_feat = CEPH_FEATURE_REQUIRED;
-	u64 server_feat = le64_to_cpu(con->in_reply.features);
-
-	dout("process_connect on %p tag %d\n", con, (int)con->in_tag);
-
-	switch (con->in_reply.tag) {
-	case CEPH_MSGR_TAG_FEATURES:
-		pr_err("%s%lld %s feature set mismatch,"
-		       " my %llx < server's %llx, missing %llx\n",
-		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr),
-		       sup_feat, server_feat, server_feat & ~sup_feat);
-		con->error_msg = "missing required protocol features";
-		fail_protocol(con);
-		return -1;
-
-	case CEPH_MSGR_TAG_BADPROTOVER:
-		pr_err("%s%lld %s protocol version mismatch,"
-		       " my %d != server's %d\n",
-		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr),
-		       le32_to_cpu(con->out_connect.protocol_version),
-		       le32_to_cpu(con->in_reply.protocol_version));
-		con->error_msg = "protocol version mismatch";
-		fail_protocol(con);
-		return -1;
-
-	case CEPH_MSGR_TAG_BADAUTHORIZER:
-		con->auth_retry++;
-		dout("process_connect %p got BADAUTHORIZER attempt %d\n", con,
-		     con->auth_retry);
-		if (con->auth_retry == 2) {
-			con->error_msg = "connect authorization failure";
-			reset_connection(con);
-			set_bit(CLOSED, &con->state);
-			return -1;
-		}
-		con->auth_retry = 1;
-		prepare_write_connect(con->msgr, con, 0);
-		prepare_read_connect(con);
-		break;
-
-	case CEPH_MSGR_TAG_RESETSESSION:
-		/*
-		 * If we connected with a large connect_seq but the peer
-		 * has no record of a session with us (no connection, or
-		 * connect_seq == 0), they will send RESETSESION to indicate
-		 * that they must have reset their session, and may have
-		 * dropped messages.
-		 */
-		dout("process_connect got RESET peer seq %u\n",
-		     le32_to_cpu(con->in_connect.connect_seq));
-		pr_err("%s%lld %s connection reset\n",
-		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr));
-		reset_connection(con);
-		prepare_write_connect(con->msgr, con, 0);
-		prepare_read_connect(con);
-
-		/* Tell ceph about it. */
-		mutex_unlock(&con->mutex);
-		pr_info("reset on %s%lld\n", ENTITY_NAME(con->peer_name));
-		if (con->ops->peer_reset)
-			con->ops->peer_reset(con);
-		mutex_lock(&con->mutex);
-		break;
-
-	case CEPH_MSGR_TAG_RETRY_SESSION:
-		/*
-		 * If we sent a smaller connect_seq than the peer has, try
-		 * again with a larger value.
-		 */
-		dout("process_connect got RETRY my seq = %u, peer_seq = %u\n",
-		     le32_to_cpu(con->out_connect.connect_seq),
-		     le32_to_cpu(con->in_connect.connect_seq));
-		con->connect_seq = le32_to_cpu(con->in_connect.connect_seq);
-		prepare_write_connect(con->msgr, con, 0);
-		prepare_read_connect(con);
-		break;
-
-	case CEPH_MSGR_TAG_RETRY_GLOBAL:
-		/*
-		 * If we sent a smaller global_seq than the peer has, try
-		 * again with a larger value.
-		 */
-		dout("process_connect got RETRY_GLOBAL my %u peer_gseq %u\n",
-		     con->peer_global_seq,
-		     le32_to_cpu(con->in_connect.global_seq));
-		get_global_seq(con->msgr,
-			       le32_to_cpu(con->in_connect.global_seq));
-		prepare_write_connect(con->msgr, con, 0);
-		prepare_read_connect(con);
-		break;
-
-	case CEPH_MSGR_TAG_READY:
-		if (req_feat & ~server_feat) {
-			pr_err("%s%lld %s protocol feature mismatch,"
-			       " my required %llx > server's %llx, need %llx\n",
-			       ENTITY_NAME(con->peer_name),
-			       pr_addr(&con->peer_addr.in_addr),
-			       req_feat, server_feat, req_feat & ~server_feat);
-			con->error_msg = "missing required protocol features";
-			fail_protocol(con);
-			return -1;
-		}
-		clear_bit(CONNECTING, &con->state);
-		con->peer_global_seq = le32_to_cpu(con->in_reply.global_seq);
-		con->connect_seq++;
-		con->peer_features = server_feat;
-		dout("process_connect got READY gseq %d cseq %d (%d)\n",
-		     con->peer_global_seq,
-		     le32_to_cpu(con->in_reply.connect_seq),
-		     con->connect_seq);
-		WARN_ON(con->connect_seq !=
-			le32_to_cpu(con->in_reply.connect_seq));
-
-		if (con->in_reply.flags & CEPH_MSG_CONNECT_LOSSY)
-			set_bit(LOSSYTX, &con->state);
-
-		prepare_read_tag(con);
-		break;
-
-	case CEPH_MSGR_TAG_WAIT:
-		/*
-		 * If there is a connection race (we are opening
-		 * connections to each other), one of us may just have
-		 * to WAIT.  This shouldn't happen if we are the
-		 * client.
-		 */
-		pr_err("process_connect peer connecting WAIT\n");
-
-	default:
-		pr_err("connect protocol error, will retry\n");
-		con->error_msg = "protocol error, garbage tag during connect";
-		return -1;
-	}
-	return 0;
-}
-
-
-/*
- * read (part of) an ack
- */
-static int read_partial_ack(struct ceph_connection *con)
-{
-	int to = 0;
-
-	return read_partial(con, &to, sizeof(con->in_temp_ack),
-			    &con->in_temp_ack);
-}
-
-
-/*
- * We can finally discard anything that's been acked.
- */
-static void process_ack(struct ceph_connection *con)
-{
-	struct ceph_msg *m;
-	u64 ack = le64_to_cpu(con->in_temp_ack);
-	u64 seq;
-
-	while (!list_empty(&con->out_sent)) {
-		m = list_first_entry(&con->out_sent, struct ceph_msg,
-				     list_head);
-		seq = le64_to_cpu(m->hdr.seq);
-		if (seq > ack)
-			break;
-		dout("got ack for seq %llu type %d at %p\n", seq,
-		     le16_to_cpu(m->hdr.type), m);
-		ceph_msg_remove(m);
-	}
-	prepare_read_tag(con);
-}
-
-
-
-
-static int read_partial_message_section(struct ceph_connection *con,
-					struct kvec *section,
-					unsigned int sec_len, u32 *crc)
-{
-	int ret, left;
-
-	BUG_ON(!section);
-
-	while (section->iov_len < sec_len) {
-		BUG_ON(section->iov_base == NULL);
-		left = sec_len - section->iov_len;
-		ret = ceph_tcp_recvmsg(con->sock, (char *)section->iov_base +
-				       section->iov_len, left);
-		if (ret <= 0)
-			return ret;
-		section->iov_len += ret;
-		if (section->iov_len == sec_len)
-			*crc = crc32c(0, section->iov_base,
-				      section->iov_len);
-	}
-
-	return 1;
-}
-
-static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con,
-				struct ceph_msg_header *hdr,
-				int *skip);
-
-
-static int read_partial_message_pages(struct ceph_connection *con,
-				      struct page **pages,
-				      unsigned data_len, int datacrc)
-{
-	void *p;
-	int ret;
-	int left;
-
-	left = min((int)(data_len - con->in_msg_pos.data_pos),
-		   (int)(PAGE_SIZE - con->in_msg_pos.page_pos));
-	/* (page) data */
-	BUG_ON(pages == NULL);
-	p = kmap(pages[con->in_msg_pos.page]);
-	ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
-			       left);
-	if (ret > 0 && datacrc)
-		con->in_data_crc =
-			crc32c(con->in_data_crc,
-				  p + con->in_msg_pos.page_pos, ret);
-	kunmap(pages[con->in_msg_pos.page]);
-	if (ret <= 0)
-		return ret;
-	con->in_msg_pos.data_pos += ret;
-	con->in_msg_pos.page_pos += ret;
-	if (con->in_msg_pos.page_pos == PAGE_SIZE) {
-		con->in_msg_pos.page_pos = 0;
-		con->in_msg_pos.page++;
-	}
-
-	return ret;
-}
-
-#ifdef CONFIG_BLOCK
-static int read_partial_message_bio(struct ceph_connection *con,
-				    struct bio **bio_iter, int *bio_seg,
-				    unsigned data_len, int datacrc)
-{
-	struct bio_vec *bv = bio_iovec_idx(*bio_iter, *bio_seg);
-	void *p;
-	int ret, left;
-
-	if (IS_ERR(bv))
-		return PTR_ERR(bv);
-
-	left = min((int)(data_len - con->in_msg_pos.data_pos),
-		   (int)(bv->bv_len - con->in_msg_pos.page_pos));
-
-	p = kmap(bv->bv_page) + bv->bv_offset;
-
-	ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
-			       left);
-	if (ret > 0 && datacrc)
-		con->in_data_crc =
-			crc32c(con->in_data_crc,
-				  p + con->in_msg_pos.page_pos, ret);
-	kunmap(bv->bv_page);
-	if (ret <= 0)
-		return ret;
-	con->in_msg_pos.data_pos += ret;
-	con->in_msg_pos.page_pos += ret;
-	if (con->in_msg_pos.page_pos == bv->bv_len) {
-		con->in_msg_pos.page_pos = 0;
-		iter_bio_next(bio_iter, bio_seg);
-	}
-
-	return ret;
-}
-#endif
-
-/*
- * read (part of) a message.
- */
-static int read_partial_message(struct ceph_connection *con)
-{
-	struct ceph_msg *m = con->in_msg;
-	int ret;
-	int to, left;
-	unsigned front_len, middle_len, data_len, data_off;
-	int datacrc = con->msgr->nocrc;
-	int skip;
-	u64 seq;
-
-	dout("read_partial_message con %p msg %p\n", con, m);
-
-	/* header */
-	while (con->in_base_pos < sizeof(con->in_hdr)) {
-		left = sizeof(con->in_hdr) - con->in_base_pos;
-		ret = ceph_tcp_recvmsg(con->sock,
-				       (char *)&con->in_hdr + con->in_base_pos,
-				       left);
-		if (ret <= 0)
-			return ret;
-		con->in_base_pos += ret;
-		if (con->in_base_pos == sizeof(con->in_hdr)) {
-			u32 crc = crc32c(0, (void *)&con->in_hdr,
-				 sizeof(con->in_hdr) - sizeof(con->in_hdr.crc));
-			if (crc != le32_to_cpu(con->in_hdr.crc)) {
-				pr_err("read_partial_message bad hdr "
-				       " crc %u != expected %u\n",
-				       crc, con->in_hdr.crc);
-				return -EBADMSG;
-			}
-		}
-	}
-	front_len = le32_to_cpu(con->in_hdr.front_len);
-	if (front_len > CEPH_MSG_MAX_FRONT_LEN)
-		return -EIO;
-	middle_len = le32_to_cpu(con->in_hdr.middle_len);
-	if (middle_len > CEPH_MSG_MAX_DATA_LEN)
-		return -EIO;
-	data_len = le32_to_cpu(con->in_hdr.data_len);
-	if (data_len > CEPH_MSG_MAX_DATA_LEN)
-		return -EIO;
-	data_off = le16_to_cpu(con->in_hdr.data_off);
-
-	/* verify seq# */
-	seq = le64_to_cpu(con->in_hdr.seq);
-	if ((s64)seq - (s64)con->in_seq < 1) {
-		pr_info("skipping %s%lld %s seq %lld, expected %lld\n",
-			ENTITY_NAME(con->peer_name),
-			pr_addr(&con->peer_addr.in_addr),
-			seq, con->in_seq + 1);
-		con->in_base_pos = -front_len - middle_len - data_len -
-			sizeof(m->footer);
-		con->in_tag = CEPH_MSGR_TAG_READY;
-		con->in_seq++;
-		return 0;
-	} else if ((s64)seq - (s64)con->in_seq > 1) {
-		pr_err("read_partial_message bad seq %lld expected %lld\n",
-		       seq, con->in_seq + 1);
-		con->error_msg = "bad message sequence # for incoming message";
-		return -EBADMSG;
-	}
-
-	/* allocate message? */
-	if (!con->in_msg) {
-		dout("got hdr type %d front %d data %d\n", con->in_hdr.type,
-		     con->in_hdr.front_len, con->in_hdr.data_len);
-		skip = 0;
-		con->in_msg = ceph_alloc_msg(con, &con->in_hdr, &skip);
-		if (skip) {
-			/* skip this message */
-			dout("alloc_msg said skip message\n");
-			BUG_ON(con->in_msg);
-			con->in_base_pos = -front_len - middle_len - data_len -
-				sizeof(m->footer);
-			con->in_tag = CEPH_MSGR_TAG_READY;
-			con->in_seq++;
-			return 0;
-		}
-		if (!con->in_msg) {
-			con->error_msg =
-				"error allocating memory for incoming message";
-			return -ENOMEM;
-		}
-		m = con->in_msg;
-		m->front.iov_len = 0;    /* haven't read it yet */
-		if (m->middle)
-			m->middle->vec.iov_len = 0;
-
-		con->in_msg_pos.page = 0;
-		if (m->pages)
-			con->in_msg_pos.page_pos = data_off & ~PAGE_MASK;
-		else
-			con->in_msg_pos.page_pos = 0;
-		con->in_msg_pos.data_pos = 0;
-	}
-
-	/* front */
-	ret = read_partial_message_section(con, &m->front, front_len,
-					   &con->in_front_crc);
-	if (ret <= 0)
-		return ret;
-
-	/* middle */
-	if (m->middle) {
-		ret = read_partial_message_section(con, &m->middle->vec,
-						   middle_len,
-						   &con->in_middle_crc);
-		if (ret <= 0)
-			return ret;
-	}
-#ifdef CONFIG_BLOCK
-	if (m->bio && !m->bio_iter)
-		init_bio_iter(m->bio, &m->bio_iter, &m->bio_seg);
-#endif
-
-	/* (page) data */
-	while (con->in_msg_pos.data_pos < data_len) {
-		if (m->pages) {
-			ret = read_partial_message_pages(con, m->pages,
-						 data_len, datacrc);
-			if (ret <= 0)
-				return ret;
-#ifdef CONFIG_BLOCK
-		} else if (m->bio) {
-
-			ret = read_partial_message_bio(con,
-						 &m->bio_iter, &m->bio_seg,
-						 data_len, datacrc);
-			if (ret <= 0)
-				return ret;
-#endif
-		} else {
-			BUG_ON(1);
-		}
-	}
-
-	/* footer */
-	to = sizeof(m->hdr) + sizeof(m->footer);
-	while (con->in_base_pos < to) {
-		left = to - con->in_base_pos;
-		ret = ceph_tcp_recvmsg(con->sock, (char *)&m->footer +
-				       (con->in_base_pos - sizeof(m->hdr)),
-				       left);
-		if (ret <= 0)
-			return ret;
-		con->in_base_pos += ret;
-	}
-	dout("read_partial_message got msg %p %d (%u) + %d (%u) + %d (%u)\n",
-	     m, front_len, m->footer.front_crc, middle_len,
-	     m->footer.middle_crc, data_len, m->footer.data_crc);
-
-	/* crc ok? */
-	if (con->in_front_crc != le32_to_cpu(m->footer.front_crc)) {
-		pr_err("read_partial_message %p front crc %u != exp. %u\n",
-		       m, con->in_front_crc, m->footer.front_crc);
-		return -EBADMSG;
-	}
-	if (con->in_middle_crc != le32_to_cpu(m->footer.middle_crc)) {
-		pr_err("read_partial_message %p middle crc %u != exp %u\n",
-		       m, con->in_middle_crc, m->footer.middle_crc);
-		return -EBADMSG;
-	}
-	if (datacrc &&
-	    (m->footer.flags & CEPH_MSG_FOOTER_NOCRC) == 0 &&
-	    con->in_data_crc != le32_to_cpu(m->footer.data_crc)) {
-		pr_err("read_partial_message %p data crc %u != exp. %u\n", m,
-		       con->in_data_crc, le32_to_cpu(m->footer.data_crc));
-		return -EBADMSG;
-	}
-
-	return 1; /* done! */
-}
-
-/*
- * Process message.  This happens in the worker thread.  The callback should
- * be careful not to do anything that waits on other incoming messages or it
- * may deadlock.
- */
-static void process_message(struct ceph_connection *con)
-{
-	struct ceph_msg *msg;
-
-	msg = con->in_msg;
-	con->in_msg = NULL;
-
-	/* if first message, set peer_name */
-	if (con->peer_name.type == 0)
-		con->peer_name = msg->hdr.src;
-
-	con->in_seq++;
-	mutex_unlock(&con->mutex);
-
-	dout("===== %p %llu from %s%lld %d=%s len %d+%d (%u %u %u) =====\n",
-	     msg, le64_to_cpu(msg->hdr.seq),
-	     ENTITY_NAME(msg->hdr.src),
-	     le16_to_cpu(msg->hdr.type),
-	     ceph_msg_type_name(le16_to_cpu(msg->hdr.type)),
-	     le32_to_cpu(msg->hdr.front_len),
-	     le32_to_cpu(msg->hdr.data_len),
-	     con->in_front_crc, con->in_middle_crc, con->in_data_crc);
-	con->ops->dispatch(con, msg);
-
-	mutex_lock(&con->mutex);
-	prepare_read_tag(con);
-}
-
-
-/*
- * Write something to the socket.  Called in a worker thread when the
- * socket appears to be writeable and we have something ready to send.
- */
-static int try_write(struct ceph_connection *con)
-{
-	struct ceph_messenger *msgr = con->msgr;
-	int ret = 1;
-
-	dout("try_write start %p state %lu nref %d\n", con, con->state,
-	     atomic_read(&con->nref));
-
-more:
-	dout("try_write out_kvec_bytes %d\n", con->out_kvec_bytes);
-
-	/* open the socket first? */
-	if (con->sock == NULL) {
-		/*
-		 * if we were STANDBY and are reconnecting _this_
-		 * connection, bump connect_seq now.  Always bump
-		 * global_seq.
-		 */
-		if (test_and_clear_bit(STANDBY, &con->state))
-			con->connect_seq++;
-
-		prepare_write_banner(msgr, con);
-		prepare_write_connect(msgr, con, 1);
-		prepare_read_banner(con);
-		set_bit(CONNECTING, &con->state);
-		clear_bit(NEGOTIATING, &con->state);
-
-		BUG_ON(con->in_msg);
-		con->in_tag = CEPH_MSGR_TAG_READY;
-		dout("try_write initiating connect on %p new state %lu\n",
-		     con, con->state);
-		con->sock = ceph_tcp_connect(con);
-		if (IS_ERR(con->sock)) {
-			con->sock = NULL;
-			con->error_msg = "connect error";
-			ret = -1;
-			goto out;
-		}
-	}
-
-more_kvec:
-	/* kvec data queued? */
-	if (con->out_skip) {
-		ret = write_partial_skip(con);
-		if (ret <= 0)
-			goto done;
-		if (ret < 0) {
-			dout("try_write write_partial_skip err %d\n", ret);
-			goto done;
-		}
-	}
-	if (con->out_kvec_left) {
-		ret = write_partial_kvec(con);
-		if (ret <= 0)
-			goto done;
-	}
-
-	/* msg pages? */
-	if (con->out_msg) {
-		if (con->out_msg_done) {
-			ceph_msg_put(con->out_msg);
-			con->out_msg = NULL;   /* we're done with this one */
-			goto do_next;
-		}
-
-		ret = write_partial_msg_pages(con);
-		if (ret == 1)
-			goto more_kvec;  /* we need to send the footer, too! */
-		if (ret == 0)
-			goto done;
-		if (ret < 0) {
-			dout("try_write write_partial_msg_pages err %d\n",
-			     ret);
-			goto done;
-		}
-	}
-
-do_next:
-	if (!test_bit(CONNECTING, &con->state)) {
-		/* is anything else pending? */
-		if (!list_empty(&con->out_queue)) {
-			prepare_write_message(con);
-			goto more;
-		}
-		if (con->in_seq > con->in_seq_acked) {
-			prepare_write_ack(con);
-			goto more;
-		}
-		if (test_and_clear_bit(KEEPALIVE_PENDING, &con->state)) {
-			prepare_write_keepalive(con);
-			goto more;
-		}
-	}
-
-	/* Nothing to do! */
-	clear_bit(WRITE_PENDING, &con->state);
-	dout("try_write nothing else to write.\n");
-done:
-	ret = 0;
-out:
-	dout("try_write done on %p\n", con);
-	return ret;
-}
-
-
-
-/*
- * Read what we can from the socket.
- */
-static int try_read(struct ceph_connection *con)
-{
-	int ret = -1;
-
-	if (!con->sock)
-		return 0;
-
-	if (test_bit(STANDBY, &con->state))
-		return 0;
-
-	dout("try_read start on %p\n", con);
-
-more:
-	dout("try_read tag %d in_base_pos %d\n", (int)con->in_tag,
-	     con->in_base_pos);
-	if (test_bit(CONNECTING, &con->state)) {
-		if (!test_bit(NEGOTIATING, &con->state)) {
-			dout("try_read connecting\n");
-			ret = read_partial_banner(con);
-			if (ret <= 0)
-				goto done;
-			if (process_banner(con) < 0) {
-				ret = -1;
-				goto out;
-			}
-		}
-		ret = read_partial_connect(con);
-		if (ret <= 0)
-			goto done;
-		if (process_connect(con) < 0) {
-			ret = -1;
-			goto out;
-		}
-		goto more;
-	}
-
-	if (con->in_base_pos < 0) {
-		/*
-		 * skipping + discarding content.
-		 *
-		 * FIXME: there must be a better way to do this!
-		 */
-		static char buf[1024];
-		int skip = min(1024, -con->in_base_pos);
-		dout("skipping %d / %d bytes\n", skip, -con->in_base_pos);
-		ret = ceph_tcp_recvmsg(con->sock, buf, skip);
-		if (ret <= 0)
-			goto done;
-		con->in_base_pos += ret;
-		if (con->in_base_pos)
-			goto more;
-	}
-	if (con->in_tag == CEPH_MSGR_TAG_READY) {
-		/*
-		 * what's next?
-		 */
-		ret = ceph_tcp_recvmsg(con->sock, &con->in_tag, 1);
-		if (ret <= 0)
-			goto done;
-		dout("try_read got tag %d\n", (int)con->in_tag);
-		switch (con->in_tag) {
-		case CEPH_MSGR_TAG_MSG:
-			prepare_read_message(con);
-			break;
-		case CEPH_MSGR_TAG_ACK:
-			prepare_read_ack(con);
-			break;
-		case CEPH_MSGR_TAG_CLOSE:
-			set_bit(CLOSED, &con->state);   /* fixme */
-			goto done;
-		default:
-			goto bad_tag;
-		}
-	}
-	if (con->in_tag == CEPH_MSGR_TAG_MSG) {
-		ret = read_partial_message(con);
-		if (ret <= 0) {
-			switch (ret) {
-			case -EBADMSG:
-				con->error_msg = "bad crc";
-				ret = -EIO;
-				goto out;
-			case -EIO:
-				con->error_msg = "io error";
-				goto out;
-			default:
-				goto done;
-			}
-		}
-		if (con->in_tag == CEPH_MSGR_TAG_READY)
-			goto more;
-		process_message(con);
-		goto more;
-	}
-	if (con->in_tag == CEPH_MSGR_TAG_ACK) {
-		ret = read_partial_ack(con);
-		if (ret <= 0)
-			goto done;
-		process_ack(con);
-		goto more;
-	}
-
-done:
-	ret = 0;
-out:
-	dout("try_read done on %p\n", con);
-	return ret;
-
-bad_tag:
-	pr_err("try_read bad con->in_tag = %d\n", (int)con->in_tag);
-	con->error_msg = "protocol error, garbage tag";
-	ret = -1;
-	goto out;
-}
-
-
-/*
- * Atomically queue work on a connection.  Bump @con reference to
- * avoid races with connection teardown.
- *
- * There is some trickery going on with QUEUED and BUSY because we
- * only want a _single_ thread operating on each connection at any
- * point in time, but we want to use all available CPUs.
- *
- * The worker thread only proceeds if it can atomically set BUSY.  It
- * clears QUEUED and does it's thing.  When it thinks it's done, it
- * clears BUSY, then rechecks QUEUED.. if it's set again, it loops
- * (tries again to set BUSY).
- *
- * To queue work, we first set QUEUED, _then_ if BUSY isn't set, we
- * try to queue work.  If that fails (work is already queued, or BUSY)
- * we give up (work also already being done or is queued) but leave QUEUED
- * set so that the worker thread will loop if necessary.
- */
-static void queue_con(struct ceph_connection *con)
-{
-	if (test_bit(DEAD, &con->state)) {
-		dout("queue_con %p ignoring: DEAD\n",
-		     con);
-		return;
-	}
-
-	if (!con->ops->get(con)) {
-		dout("queue_con %p ref count 0\n", con);
-		return;
-	}
-
-	set_bit(QUEUED, &con->state);
-	if (test_bit(BUSY, &con->state)) {
-		dout("queue_con %p - already BUSY\n", con);
-		con->ops->put(con);
-	} else if (!queue_work(ceph_msgr_wq, &con->work.work)) {
-		dout("queue_con %p - already queued\n", con);
-		con->ops->put(con);
-	} else {
-		dout("queue_con %p\n", con);
-	}
-}
-
-/*
- * Do some work on a connection.  Drop a connection ref when we're done.
- */
-static void con_work(struct work_struct *work)
-{
-	struct ceph_connection *con = container_of(work, struct ceph_connection,
-						   work.work);
-	int backoff = 0;
-
-more:
-	if (test_and_set_bit(BUSY, &con->state) != 0) {
-		dout("con_work %p BUSY already set\n", con);
-		goto out;
-	}
-	dout("con_work %p start, clearing QUEUED\n", con);
-	clear_bit(QUEUED, &con->state);
-
-	mutex_lock(&con->mutex);
-
-	if (test_bit(CLOSED, &con->state)) { /* e.g. if we are replaced */
-		dout("con_work CLOSED\n");
-		con_close_socket(con);
-		goto done;
-	}
-	if (test_and_clear_bit(OPENING, &con->state)) {
-		/* reopen w/ new peer */
-		dout("con_work OPENING\n");
-		con_close_socket(con);
-	}
-
-	if (test_and_clear_bit(SOCK_CLOSED, &con->state) ||
-	    try_read(con) < 0 ||
-	    try_write(con) < 0) {
-		mutex_unlock(&con->mutex);
-		backoff = 1;
-		ceph_fault(con);     /* error/fault path */
-		goto done_unlocked;
-	}
-
-done:
-	mutex_unlock(&con->mutex);
-
-done_unlocked:
-	clear_bit(BUSY, &con->state);
-	dout("con->state=%lu\n", con->state);
-	if (test_bit(QUEUED, &con->state)) {
-		if (!backoff || test_bit(OPENING, &con->state)) {
-			dout("con_work %p QUEUED reset, looping\n", con);
-			goto more;
-		}
-		dout("con_work %p QUEUED reset, but just faulted\n", con);
-		clear_bit(QUEUED, &con->state);
-	}
-	dout("con_work %p done\n", con);
-
-out:
-	con->ops->put(con);
-}
-
-
-/*
- * Generic error/fault handler.  A retry mechanism is used with
- * exponential backoff
- */
-static void ceph_fault(struct ceph_connection *con)
-{
-	pr_err("%s%lld %s %s\n", ENTITY_NAME(con->peer_name),
-	       pr_addr(&con->peer_addr.in_addr), con->error_msg);
-	dout("fault %p state %lu to peer %s\n",
-	     con, con->state, pr_addr(&con->peer_addr.in_addr));
-
-	if (test_bit(LOSSYTX, &con->state)) {
-		dout("fault on LOSSYTX channel\n");
-		goto out;
-	}
-
-	mutex_lock(&con->mutex);
-	if (test_bit(CLOSED, &con->state))
-		goto out_unlock;
-
-	con_close_socket(con);
-
-	if (con->in_msg) {
-		ceph_msg_put(con->in_msg);
-		con->in_msg = NULL;
-	}
-
-	/* Requeue anything that hasn't been acked */
-	list_splice_init(&con->out_sent, &con->out_queue);
-
-	/* If there are no messages in the queue, place the connection
-	 * in a STANDBY state (i.e., don't try to reconnect just yet). */
-	if (list_empty(&con->out_queue) && !con->out_keepalive_pending) {
-		dout("fault setting STANDBY\n");
-		set_bit(STANDBY, &con->state);
-	} else {
-		/* retry after a delay. */
-		if (con->delay == 0)
-			con->delay = BASE_DELAY_INTERVAL;
-		else if (con->delay < MAX_DELAY_INTERVAL)
-			con->delay *= 2;
-		dout("fault queueing %p delay %lu\n", con, con->delay);
-		con->ops->get(con);
-		if (queue_delayed_work(ceph_msgr_wq, &con->work,
-				       round_jiffies_relative(con->delay)) == 0)
-			con->ops->put(con);
-	}
-
-out_unlock:
-	mutex_unlock(&con->mutex);
-out:
-	/*
-	 * in case we faulted due to authentication, invalidate our
-	 * current tickets so that we can get new ones.
-	 */
-	if (con->auth_retry && con->ops->invalidate_authorizer) {
-		dout("calling invalidate_authorizer()\n");
-		con->ops->invalidate_authorizer(con);
-	}
-
-	if (con->ops->fault)
-		con->ops->fault(con);
-}
-
-
-
-/*
- * create a new messenger instance
- */
-struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr)
-{
-	struct ceph_messenger *msgr;
-
-	msgr = kzalloc(sizeof(*msgr), GFP_KERNEL);
-	if (msgr == NULL)
-		return ERR_PTR(-ENOMEM);
-
-	spin_lock_init(&msgr->global_seq_lock);
-
-	/* the zero page is needed if a request is "canceled" while the message
-	 * is being written over the socket */
-	msgr->zero_page = __page_cache_alloc(GFP_KERNEL | __GFP_ZERO);
-	if (!msgr->zero_page) {
-		kfree(msgr);
-		return ERR_PTR(-ENOMEM);
-	}
-	kmap(msgr->zero_page);
-
-	if (myaddr)
-		msgr->inst.addr = *myaddr;
-
-	/* select a random nonce */
-	msgr->inst.addr.type = 0;
-	get_random_bytes(&msgr->inst.addr.nonce, sizeof(msgr->inst.addr.nonce));
-	encode_my_addr(msgr);
-
-	dout("messenger_create %p\n", msgr);
-	return msgr;
-}
-
-void ceph_messenger_destroy(struct ceph_messenger *msgr)
-{
-	dout("destroy %p\n", msgr);
-	kunmap(msgr->zero_page);
-	__free_page(msgr->zero_page);
-	kfree(msgr);
-	dout("destroyed messenger %p\n", msgr);
-}
-
-/*
- * Queue up an outgoing message on the given connection.
- */
-void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
-{
-	if (test_bit(CLOSED, &con->state)) {
-		dout("con_send %p closed, dropping %p\n", con, msg);
-		ceph_msg_put(msg);
-		return;
-	}
-
-	/* set src+dst */
-	msg->hdr.src = con->msgr->inst.name;
-
-	BUG_ON(msg->front.iov_len != le32_to_cpu(msg->hdr.front_len));
-
-	msg->needs_out_seq = true;
-
-	/* queue */
-	mutex_lock(&con->mutex);
-	BUG_ON(!list_empty(&msg->list_head));
-	list_add_tail(&msg->list_head, &con->out_queue);
-	dout("----- %p to %s%lld %d=%s len %d+%d+%d -----\n", msg,
-	     ENTITY_NAME(con->peer_name), le16_to_cpu(msg->hdr.type),
-	     ceph_msg_type_name(le16_to_cpu(msg->hdr.type)),
-	     le32_to_cpu(msg->hdr.front_len),
-	     le32_to_cpu(msg->hdr.middle_len),
-	     le32_to_cpu(msg->hdr.data_len));
-	mutex_unlock(&con->mutex);
-
-	/* if there wasn't anything waiting to send before, queue
-	 * new work */
-	if (test_and_set_bit(WRITE_PENDING, &con->state) == 0)
-		queue_con(con);
-}
-
-/*
- * Revoke a message that was previously queued for send
- */
-void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg)
-{
-	mutex_lock(&con->mutex);
-	if (!list_empty(&msg->list_head)) {
-		dout("con_revoke %p msg %p - was on queue\n", con, msg);
-		list_del_init(&msg->list_head);
-		ceph_msg_put(msg);
-		msg->hdr.seq = 0;
-	}
-	if (con->out_msg == msg) {
-		dout("con_revoke %p msg %p - was sending\n", con, msg);
-		con->out_msg = NULL;
-		if (con->out_kvec_is_msg) {
-			con->out_skip = con->out_kvec_bytes;
-			con->out_kvec_is_msg = false;
-		}
-		ceph_msg_put(msg);
-		msg->hdr.seq = 0;
-	}
-	mutex_unlock(&con->mutex);
-}
-
-/*
- * Revoke a message that we may be reading data into
- */
-void ceph_con_revoke_message(struct ceph_connection *con, struct ceph_msg *msg)
-{
-	mutex_lock(&con->mutex);
-	if (con->in_msg && con->in_msg == msg) {
-		unsigned front_len = le32_to_cpu(con->in_hdr.front_len);
-		unsigned middle_len = le32_to_cpu(con->in_hdr.middle_len);
-		unsigned data_len = le32_to_cpu(con->in_hdr.data_len);
-
-		/* skip rest of message */
-		dout("con_revoke_pages %p msg %p revoked\n", con, msg);
-			con->in_base_pos = con->in_base_pos -
-				sizeof(struct ceph_msg_header) -
-				front_len -
-				middle_len -
-				data_len -
-				sizeof(struct ceph_msg_footer);
-		ceph_msg_put(con->in_msg);
-		con->in_msg = NULL;
-		con->in_tag = CEPH_MSGR_TAG_READY;
-		con->in_seq++;
-	} else {
-		dout("con_revoke_pages %p msg %p pages %p no-op\n",
-		     con, con->in_msg, msg);
-	}
-	mutex_unlock(&con->mutex);
-}
-
-/*
- * Queue a keepalive byte to ensure the tcp connection is alive.
- */
-void ceph_con_keepalive(struct ceph_connection *con)
-{
-	if (test_and_set_bit(KEEPALIVE_PENDING, &con->state) == 0 &&
-	    test_and_set_bit(WRITE_PENDING, &con->state) == 0)
-		queue_con(con);
-}
-
-
-/*
- * construct a new message with given type, size
- * the new msg has a ref count of 1.
- */
-struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags)
-{
-	struct ceph_msg *m;
-
-	m = kmalloc(sizeof(*m), flags);
-	if (m == NULL)
-		goto out;
-	kref_init(&m->kref);
-	INIT_LIST_HEAD(&m->list_head);
-
-	m->hdr.tid = 0;
-	m->hdr.type = cpu_to_le16(type);
-	m->hdr.priority = cpu_to_le16(CEPH_MSG_PRIO_DEFAULT);
-	m->hdr.version = 0;
-	m->hdr.front_len = cpu_to_le32(front_len);
-	m->hdr.middle_len = 0;
-	m->hdr.data_len = 0;
-	m->hdr.data_off = 0;
-	m->hdr.reserved = 0;
-	m->footer.front_crc = 0;
-	m->footer.middle_crc = 0;
-	m->footer.data_crc = 0;
-	m->footer.flags = 0;
-	m->front_max = front_len;
-	m->front_is_vmalloc = false;
-	m->more_to_follow = false;
-	m->pool = NULL;
-
-	/* front */
-	if (front_len) {
-		if (front_len > PAGE_CACHE_SIZE) {
-			m->front.iov_base = __vmalloc(front_len, flags,
-						      PAGE_KERNEL);
-			m->front_is_vmalloc = true;
-		} else {
-			m->front.iov_base = kmalloc(front_len, flags);
-		}
-		if (m->front.iov_base == NULL) {
-			pr_err("msg_new can't allocate %d bytes\n",
-			     front_len);
-			goto out2;
-		}
-	} else {
-		m->front.iov_base = NULL;
-	}
-	m->front.iov_len = front_len;
-
-	/* middle */
-	m->middle = NULL;
-
-	/* data */
-	m->nr_pages = 0;
-	m->pages = NULL;
-	m->pagelist = NULL;
-	m->bio = NULL;
-	m->bio_iter = NULL;
-	m->bio_seg = 0;
-	m->trail = NULL;
-
-	dout("ceph_msg_new %p front %d\n", m, front_len);
-	return m;
-
-out2:
-	ceph_msg_put(m);
-out:
-	pr_err("msg_new can't create type %d front %d\n", type, front_len);
-	return NULL;
-}
-
-/*
- * Allocate "middle" portion of a message, if it is needed and wasn't
- * allocated by alloc_msg.  This allows us to read a small fixed-size
- * per-type header in the front and then gracefully fail (i.e.,
- * propagate the error to the caller based on info in the front) when
- * the middle is too large.
- */
-static int ceph_alloc_middle(struct ceph_connection *con, struct ceph_msg *msg)
-{
-	int type = le16_to_cpu(msg->hdr.type);
-	int middle_len = le32_to_cpu(msg->hdr.middle_len);
-
-	dout("alloc_middle %p type %d %s middle_len %d\n", msg, type,
-	     ceph_msg_type_name(type), middle_len);
-	BUG_ON(!middle_len);
-	BUG_ON(msg->middle);
-
-	msg->middle = ceph_buffer_new(middle_len, GFP_NOFS);
-	if (!msg->middle)
-		return -ENOMEM;
-	return 0;
-}
-
-/*
- * Generic message allocator, for incoming messages.
- */
-static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con,
-				struct ceph_msg_header *hdr,
-				int *skip)
-{
-	int type = le16_to_cpu(hdr->type);
-	int front_len = le32_to_cpu(hdr->front_len);
-	int middle_len = le32_to_cpu(hdr->middle_len);
-	struct ceph_msg *msg = NULL;
-	int ret;
-
-	if (con->ops->alloc_msg) {
-		mutex_unlock(&con->mutex);
-		msg = con->ops->alloc_msg(con, hdr, skip);
-		mutex_lock(&con->mutex);
-		if (!msg || *skip)
-			return NULL;
-	}
-	if (!msg) {
-		*skip = 0;
-		msg = ceph_msg_new(type, front_len, GFP_NOFS);
-		if (!msg) {
-			pr_err("unable to allocate msg type %d len %d\n",
-			       type, front_len);
-			return NULL;
-		}
-	}
-	memcpy(&msg->hdr, &con->in_hdr, sizeof(con->in_hdr));
-
-	if (middle_len && !msg->middle) {
-		ret = ceph_alloc_middle(con, msg);
-		if (ret < 0) {
-			ceph_msg_put(msg);
-			return NULL;
-		}
-	}
-
-	return msg;
-}
-
-
-/*
- * Free a generically kmalloc'd message.
- */
-void ceph_msg_kfree(struct ceph_msg *m)
-{
-	dout("msg_kfree %p\n", m);
-	if (m->front_is_vmalloc)
-		vfree(m->front.iov_base);
-	else
-		kfree(m->front.iov_base);
-	kfree(m);
-}
-
-/*
- * Drop a msg ref.  Destroy as needed.
- */
-void ceph_msg_last_put(struct kref *kref)
-{
-	struct ceph_msg *m = container_of(kref, struct ceph_msg, kref);
-
-	dout("ceph_msg_put last one on %p\n", m);
-	WARN_ON(!list_empty(&m->list_head));
-
-	/* drop middle, data, if any */
-	if (m->middle) {
-		ceph_buffer_put(m->middle);
-		m->middle = NULL;
-	}
-	m->nr_pages = 0;
-	m->pages = NULL;
-
-	if (m->pagelist) {
-		ceph_pagelist_release(m->pagelist);
-		kfree(m->pagelist);
-		m->pagelist = NULL;
-	}
-
-	m->trail = NULL;
-
-	if (m->pool)
-		ceph_msgpool_put(m->pool, m);
-	else
-		ceph_msg_kfree(m);
-}
-
-void ceph_msg_dump(struct ceph_msg *msg)
-{
-	pr_debug("msg_dump %p (front_max %d nr_pages %d)\n", msg,
-		 msg->front_max, msg->nr_pages);
-	print_hex_dump(KERN_DEBUG, "header: ",
-		       DUMP_PREFIX_OFFSET, 16, 1,
-		       &msg->hdr, sizeof(msg->hdr), true);
-	print_hex_dump(KERN_DEBUG, " front: ",
-		       DUMP_PREFIX_OFFSET, 16, 1,
-		       msg->front.iov_base, msg->front.iov_len, true);
-	if (msg->middle)
-		print_hex_dump(KERN_DEBUG, "middle: ",
-			       DUMP_PREFIX_OFFSET, 16, 1,
-			       msg->middle->vec.iov_base,
-			       msg->middle->vec.iov_len, true);
-	print_hex_dump(KERN_DEBUG, "footer: ",
-		       DUMP_PREFIX_OFFSET, 16, 1,
-		       &msg->footer, sizeof(msg->footer), true);
-}
diff --git a/fs/ceph/messenger.h b/fs/ceph/messenger.h
deleted file mode 100644
index 5a79450604ef..000000000000
--- a/fs/ceph/messenger.h
+++ /dev/null
@@ -1,257 +0,0 @@
-#ifndef __FS_CEPH_MESSENGER_H
-#define __FS_CEPH_MESSENGER_H
-
-#include <linux/kref.h>
-#include <linux/mutex.h>
-#include <linux/net.h>
-#include <linux/radix-tree.h>
-#include <linux/uio.h>
-#include <linux/version.h>
-#include <linux/workqueue.h>
-
-#include "types.h"
-#include "buffer.h"
-
-struct ceph_msg;
-struct ceph_connection;
-
-extern struct workqueue_struct *ceph_msgr_wq;       /* receive work queue */
-
-/*
- * Ceph defines these callbacks for handling connection events.
- */
-struct ceph_connection_operations {
-	struct ceph_connection *(*get)(struct ceph_connection *);
-	void (*put)(struct ceph_connection *);
-
-	/* handle an incoming message. */
-	void (*dispatch) (struct ceph_connection *con, struct ceph_msg *m);
-
-	/* authorize an outgoing connection */
-	int (*get_authorizer) (struct ceph_connection *con,
-			       void **buf, int *len, int *proto,
-			       void **reply_buf, int *reply_len, int force_new);
-	int (*verify_authorizer_reply) (struct ceph_connection *con, int len);
-	int (*invalidate_authorizer)(struct ceph_connection *con);
-
-	/* protocol version mismatch */
-	void (*bad_proto) (struct ceph_connection *con);
-
-	/* there was some error on the socket (disconnect, whatever) */
-	void (*fault) (struct ceph_connection *con);
-
-	/* a remote host as terminated a message exchange session, and messages
-	 * we sent (or they tried to send us) may be lost. */
-	void (*peer_reset) (struct ceph_connection *con);
-
-	struct ceph_msg * (*alloc_msg) (struct ceph_connection *con,
-					struct ceph_msg_header *hdr,
-					int *skip);
-};
-
-/* use format string %s%d */
-#define ENTITY_NAME(n) ceph_entity_type_name((n).type), le64_to_cpu((n).num)
-
-struct ceph_messenger {
-	struct ceph_entity_inst inst;    /* my name+address */
-	struct ceph_entity_addr my_enc_addr;
-	struct page *zero_page;          /* used in certain error cases */
-
-	bool nocrc;
-
-	/*
-	 * the global_seq counts connections i (attempt to) initiate
-	 * in order to disambiguate certain connect race conditions.
-	 */
-	u32 global_seq;
-	spinlock_t global_seq_lock;
-};
-
-/*
- * a single message.  it contains a header (src, dest, message type, etc.),
- * footer (crc values, mainly), a "front" message body, and possibly a
- * data payload (stored in some number of pages).
- */
-struct ceph_msg {
-	struct ceph_msg_header hdr;	/* header */
-	struct ceph_msg_footer footer;	/* footer */
-	struct kvec front;              /* unaligned blobs of message */
-	struct ceph_buffer *middle;
-	struct page **pages;            /* data payload.  NOT OWNER. */
-	unsigned nr_pages;              /* size of page array */
-	struct ceph_pagelist *pagelist; /* instead of pages */
-	struct list_head list_head;
-	struct kref kref;
-	struct bio  *bio;		/* instead of pages/pagelist */
-	struct bio  *bio_iter;		/* bio iterator */
-	int bio_seg;			/* current bio segment */
-	struct ceph_pagelist *trail;	/* the trailing part of the data */
-	bool front_is_vmalloc;
-	bool more_to_follow;
-	bool needs_out_seq;
-	int front_max;
-
-	struct ceph_msgpool *pool;
-};
-
-struct ceph_msg_pos {
-	int page, page_pos;  /* which page; offset in page */
-	int data_pos;        /* offset in data payload */
-	int did_page_crc;    /* true if we've calculated crc for current page */
-};
-
-/* ceph connection fault delay defaults, for exponential backoff */
-#define BASE_DELAY_INTERVAL	(HZ/2)
-#define MAX_DELAY_INTERVAL	(5 * 60 * HZ)
-
-/*
- * ceph_connection state bit flags
- *
- * QUEUED and BUSY are used together to ensure that only a single
- * thread is currently opening, reading or writing data to the socket.
- */
-#define LOSSYTX         0  /* we can close channel or drop messages on errors */
-#define CONNECTING	1
-#define NEGOTIATING	2
-#define KEEPALIVE_PENDING      3
-#define WRITE_PENDING	4  /* we have data ready to send */
-#define QUEUED          5  /* there is work queued on this connection */
-#define BUSY            6  /* work is being done */
-#define STANDBY		8  /* no outgoing messages, socket closed.  we keep
-			    * the ceph_connection around to maintain shared
-			    * state with the peer. */
-#define CLOSED		10 /* we've closed the connection */
-#define SOCK_CLOSED	11 /* socket state changed to closed */
-#define OPENING         13 /* open connection w/ (possibly new) peer */
-#define DEAD            14 /* dead, about to kfree */
-
-/*
- * A single connection with another host.
- *
- * We maintain a queue of outgoing messages, and some session state to
- * ensure that we can preserve the lossless, ordered delivery of
- * messages in the case of a TCP disconnect.
- */
-struct ceph_connection {
-	void *private;
-	atomic_t nref;
-
-	const struct ceph_connection_operations *ops;
-
-	struct ceph_messenger *msgr;
-	struct socket *sock;
-	unsigned long state;	/* connection state (see flags above) */
-	const char *error_msg;  /* error message, if any */
-
-	struct ceph_entity_addr peer_addr; /* peer address */
-	struct ceph_entity_name peer_name; /* peer name */
-	struct ceph_entity_addr peer_addr_for_me;
-	unsigned peer_features;
-	u32 connect_seq;      /* identify the most recent connection
-				 attempt for this connection, client */
-	u32 peer_global_seq;  /* peer's global seq for this connection */
-
-	int auth_retry;       /* true if we need a newer authorizer */
-	void *auth_reply_buf;   /* where to put the authorizer reply */
-	int auth_reply_buf_len;
-
-	struct mutex mutex;
-
-	/* out queue */
-	struct list_head out_queue;
-	struct list_head out_sent;   /* sending or sent but unacked */
-	u64 out_seq;		     /* last message queued for send */
-	bool out_keepalive_pending;
-
-	u64 in_seq, in_seq_acked;  /* last message received, acked */
-
-	/* connection negotiation temps */
-	char in_banner[CEPH_BANNER_MAX_LEN];
-	union {
-		struct {  /* outgoing connection */
-			struct ceph_msg_connect out_connect;
-			struct ceph_msg_connect_reply in_reply;
-		};
-		struct {  /* incoming */
-			struct ceph_msg_connect in_connect;
-			struct ceph_msg_connect_reply out_reply;
-		};
-	};
-	struct ceph_entity_addr actual_peer_addr;
-
-	/* message out temps */
-	struct ceph_msg *out_msg;        /* sending message (== tail of
-					    out_sent) */
-	bool out_msg_done;
-	struct ceph_msg_pos out_msg_pos;
-
-	struct kvec out_kvec[8],         /* sending header/footer data */
-		*out_kvec_cur;
-	int out_kvec_left;   /* kvec's left in out_kvec */
-	int out_skip;        /* skip this many bytes */
-	int out_kvec_bytes;  /* total bytes left */
-	bool out_kvec_is_msg; /* kvec refers to out_msg */
-	int out_more;        /* there is more data after the kvecs */
-	__le64 out_temp_ack; /* for writing an ack */
-
-	/* message in temps */
-	struct ceph_msg_header in_hdr;
-	struct ceph_msg *in_msg;
-	struct ceph_msg_pos in_msg_pos;
-	u32 in_front_crc, in_middle_crc, in_data_crc;  /* calculated crc */
-
-	char in_tag;         /* protocol control byte */
-	int in_base_pos;     /* bytes read */
-	__le64 in_temp_ack;  /* for reading an ack */
-
-	struct delayed_work work;	    /* send|recv work */
-	unsigned long       delay;          /* current delay interval */
-};
-
-
-extern const char *pr_addr(const struct sockaddr_storage *ss);
-extern int ceph_parse_ips(const char *c, const char *end,
-			  struct ceph_entity_addr *addr,
-			  int max_count, int *count);
-
-
-extern int ceph_msgr_init(void);
-extern void ceph_msgr_exit(void);
-extern void ceph_msgr_flush(void);
-
-extern struct ceph_messenger *ceph_messenger_create(
-	struct ceph_entity_addr *myaddr);
-extern void ceph_messenger_destroy(struct ceph_messenger *);
-
-extern void ceph_con_init(struct ceph_messenger *msgr,
-			  struct ceph_connection *con);
-extern void ceph_con_open(struct ceph_connection *con,
-			  struct ceph_entity_addr *addr);
-extern bool ceph_con_opened(struct ceph_connection *con);
-extern void ceph_con_close(struct ceph_connection *con);
-extern void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg);
-extern void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg);
-extern void ceph_con_revoke_message(struct ceph_connection *con,
-				  struct ceph_msg *msg);
-extern void ceph_con_keepalive(struct ceph_connection *con);
-extern struct ceph_connection *ceph_con_get(struct ceph_connection *con);
-extern void ceph_con_put(struct ceph_connection *con);
-
-extern struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags);
-extern void ceph_msg_kfree(struct ceph_msg *m);
-
-
-static inline struct ceph_msg *ceph_msg_get(struct ceph_msg *msg)
-{
-	kref_get(&msg->kref);
-	return msg;
-}
-extern void ceph_msg_last_put(struct kref *kref);
-static inline void ceph_msg_put(struct ceph_msg *msg)
-{
-	kref_put(&msg->kref, ceph_msg_last_put);
-}
-
-extern void ceph_msg_dump(struct ceph_msg *msg);
-
-#endif
diff --git a/fs/ceph/mon_client.c b/fs/ceph/mon_client.c
deleted file mode 100644
index b2a5a3e4a671..000000000000
--- a/fs/ceph/mon_client.c
+++ /dev/null
@@ -1,1018 +0,0 @@
-#include "ceph_debug.h"
-
-#include <linux/types.h>
-#include <linux/slab.h>
-#include <linux/random.h>
-#include <linux/sched.h>
-
-#include "mon_client.h"
-#include "super.h"
-#include "auth.h"
-#include "decode.h"
-
-/*
- * Interact with Ceph monitor cluster.  Handle requests for new map
- * versions, and periodically resend as needed.  Also implement
- * statfs() and umount().
- *
- * A small cluster of Ceph "monitors" are responsible for managing critical
- * cluster configuration and state information.  An odd number (e.g., 3, 5)
- * of cmon daemons use a modified version of the Paxos part-time parliament
- * algorithm to manage the MDS map (mds cluster membership), OSD map, and
- * list of clients who have mounted the file system.
- *
- * We maintain an open, active session with a monitor at all times in order to
- * receive timely MDSMap updates.  We periodically send a keepalive byte on the
- * TCP socket to ensure we detect a failure.  If the connection does break, we
- * randomly hunt for a new monitor.  Once the connection is reestablished, we
- * resend any outstanding requests.
- */
-
-static const struct ceph_connection_operations mon_con_ops;
-
-static int __validate_auth(struct ceph_mon_client *monc);
-
-/*
- * Decode a monmap blob (e.g., during mount).
- */
-struct ceph_monmap *ceph_monmap_decode(void *p, void *end)
-{
-	struct ceph_monmap *m = NULL;
-	int i, err = -EINVAL;
-	struct ceph_fsid fsid;
-	u32 epoch, num_mon;
-	u16 version;
-	u32 len;
-
-	ceph_decode_32_safe(&p, end, len, bad);
-	ceph_decode_need(&p, end, len, bad);
-
-	dout("monmap_decode %p %p len %d\n", p, end, (int)(end-p));
-
-	ceph_decode_16_safe(&p, end, version, bad);
-
-	ceph_decode_need(&p, end, sizeof(fsid) + 2*sizeof(u32), bad);
-	ceph_decode_copy(&p, &fsid, sizeof(fsid));
-	epoch = ceph_decode_32(&p);
-
-	num_mon = ceph_decode_32(&p);
-	ceph_decode_need(&p, end, num_mon*sizeof(m->mon_inst[0]), bad);
-
-	if (num_mon >= CEPH_MAX_MON)
-		goto bad;
-	m = kmalloc(sizeof(*m) + sizeof(m->mon_inst[0])*num_mon, GFP_NOFS);
-	if (m == NULL)
-		return ERR_PTR(-ENOMEM);
-	m->fsid = fsid;
-	m->epoch = epoch;
-	m->num_mon = num_mon;
-	ceph_decode_copy(&p, m->mon_inst, num_mon*sizeof(m->mon_inst[0]));
-	for (i = 0; i < num_mon; i++)
-		ceph_decode_addr(&m->mon_inst[i].addr);
-
-	dout("monmap_decode epoch %d, num_mon %d\n", m->epoch,
-	     m->num_mon);
-	for (i = 0; i < m->num_mon; i++)
-		dout("monmap_decode  mon%d is %s\n", i,
-		     pr_addr(&m->mon_inst[i].addr.in_addr));
-	return m;
-
-bad:
-	dout("monmap_decode failed with %d\n", err);
-	kfree(m);
-	return ERR_PTR(err);
-}
-
-/*
- * return true if *addr is included in the monmap.
- */
-int ceph_monmap_contains(struct ceph_monmap *m, struct ceph_entity_addr *addr)
-{
-	int i;
-
-	for (i = 0; i < m->num_mon; i++)
-		if (memcmp(addr, &m->mon_inst[i].addr, sizeof(*addr)) == 0)
-			return 1;
-	return 0;
-}
-
-/*
- * Send an auth request.
- */
-static void __send_prepared_auth_request(struct ceph_mon_client *monc, int len)
-{
-	monc->pending_auth = 1;
-	monc->m_auth->front.iov_len = len;
-	monc->m_auth->hdr.front_len = cpu_to_le32(len);
-	ceph_con_revoke(monc->con, monc->m_auth);
-	ceph_msg_get(monc->m_auth);  /* keep our ref */
-	ceph_con_send(monc->con, monc->m_auth);
-}
-
-/*
- * Close monitor session, if any.
- */
-static void __close_session(struct ceph_mon_client *monc)
-{
-	if (monc->con) {
-		dout("__close_session closing mon%d\n", monc->cur_mon);
-		ceph_con_revoke(monc->con, monc->m_auth);
-		ceph_con_close(monc->con);
-		monc->cur_mon = -1;
-		monc->pending_auth = 0;
-		ceph_auth_reset(monc->auth);
-	}
-}
-
-/*
- * Open a session with a (new) monitor.
- */
-static int __open_session(struct ceph_mon_client *monc)
-{
-	char r;
-	int ret;
-
-	if (monc->cur_mon < 0) {
-		get_random_bytes(&r, 1);
-		monc->cur_mon = r % monc->monmap->num_mon;
-		dout("open_session num=%d r=%d -> mon%d\n",
-		     monc->monmap->num_mon, r, monc->cur_mon);
-		monc->sub_sent = 0;
-		monc->sub_renew_after = jiffies;  /* i.e., expired */
-		monc->want_next_osdmap = !!monc->want_next_osdmap;
-
-		dout("open_session mon%d opening\n", monc->cur_mon);
-		monc->con->peer_name.type = CEPH_ENTITY_TYPE_MON;
-		monc->con->peer_name.num = cpu_to_le64(monc->cur_mon);
-		ceph_con_open(monc->con,
-			      &monc->monmap->mon_inst[monc->cur_mon].addr);
-
-		/* initiatiate authentication handshake */
-		ret = ceph_auth_build_hello(monc->auth,
-					    monc->m_auth->front.iov_base,
-					    monc->m_auth->front_max);
-		__send_prepared_auth_request(monc, ret);
-	} else {
-		dout("open_session mon%d already open\n", monc->cur_mon);
-	}
-	return 0;
-}
-
-static bool __sub_expired(struct ceph_mon_client *monc)
-{
-	return time_after_eq(jiffies, monc->sub_renew_after);
-}
-
-/*
- * Reschedule delayed work timer.
- */
-static void __schedule_delayed(struct ceph_mon_client *monc)
-{
-	unsigned delay;
-
-	if (monc->cur_mon < 0 || __sub_expired(monc))
-		delay = 10 * HZ;
-	else
-		delay = 20 * HZ;
-	dout("__schedule_delayed after %u\n", delay);
-	schedule_delayed_work(&monc->delayed_work, delay);
-}
-
-/*
- * Send subscribe request for mdsmap and/or osdmap.
- */
-static void __send_subscribe(struct ceph_mon_client *monc)
-{
-	dout("__send_subscribe sub_sent=%u exp=%u want_osd=%d\n",
-	     (unsigned)monc->sub_sent, __sub_expired(monc),
-	     monc->want_next_osdmap);
-	if ((__sub_expired(monc) && !monc->sub_sent) ||
-	    monc->want_next_osdmap == 1) {
-		struct ceph_msg *msg = monc->m_subscribe;
-		struct ceph_mon_subscribe_item *i;
-		void *p, *end;
-
-		p = msg->front.iov_base;
-		end = p + msg->front_max;
-
-		dout("__send_subscribe to 'mdsmap' %u+\n",
-		     (unsigned)monc->have_mdsmap);
-		if (monc->want_next_osdmap) {
-			dout("__send_subscribe to 'osdmap' %u\n",
-			     (unsigned)monc->have_osdmap);
-			ceph_encode_32(&p, 3);
-			ceph_encode_string(&p, end, "osdmap", 6);
-			i = p;
-			i->have = cpu_to_le64(monc->have_osdmap);
-			i->onetime = 1;
-			p += sizeof(*i);
-			monc->want_next_osdmap = 2;  /* requested */
-		} else {
-			ceph_encode_32(&p, 2);
-		}
-		ceph_encode_string(&p, end, "mdsmap", 6);
-		i = p;
-		i->have = cpu_to_le64(monc->have_mdsmap);
-		i->onetime = 0;
-		p += sizeof(*i);
-		ceph_encode_string(&p, end, "monmap", 6);
-		i = p;
-		i->have = 0;
-		i->onetime = 0;
-		p += sizeof(*i);
-
-		msg->front.iov_len = p - msg->front.iov_base;
-		msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
-		ceph_con_revoke(monc->con, msg);
-		ceph_con_send(monc->con, ceph_msg_get(msg));
-
-		monc->sub_sent = jiffies | 1;  /* never 0 */
-	}
-}
-
-static void handle_subscribe_ack(struct ceph_mon_client *monc,
-				 struct ceph_msg *msg)
-{
-	unsigned seconds;
-	struct ceph_mon_subscribe_ack *h = msg->front.iov_base;
-
-	if (msg->front.iov_len < sizeof(*h))
-		goto bad;
-	seconds = le32_to_cpu(h->duration);
-
-	mutex_lock(&monc->mutex);
-	if (monc->hunting) {
-		pr_info("mon%d %s session established\n",
-			monc->cur_mon, pr_addr(&monc->con->peer_addr.in_addr));
-		monc->hunting = false;
-	}
-	dout("handle_subscribe_ack after %d seconds\n", seconds);
-	monc->sub_renew_after = monc->sub_sent + (seconds >> 1)*HZ - 1;
-	monc->sub_sent = 0;
-	mutex_unlock(&monc->mutex);
-	return;
-bad:
-	pr_err("got corrupt subscribe-ack msg\n");
-	ceph_msg_dump(msg);
-}
-
-/*
- * Keep track of which maps we have
- */
-int ceph_monc_got_mdsmap(struct ceph_mon_client *monc, u32 got)
-{
-	mutex_lock(&monc->mutex);
-	monc->have_mdsmap = got;
-	mutex_unlock(&monc->mutex);
-	return 0;
-}
-
-int ceph_monc_got_osdmap(struct ceph_mon_client *monc, u32 got)
-{
-	mutex_lock(&monc->mutex);
-	monc->have_osdmap = got;
-	monc->want_next_osdmap = 0;
-	mutex_unlock(&monc->mutex);
-	return 0;
-}
-
-/*
- * Register interest in the next osdmap
- */
-void ceph_monc_request_next_osdmap(struct ceph_mon_client *monc)
-{
-	dout("request_next_osdmap have %u\n", monc->have_osdmap);
-	mutex_lock(&monc->mutex);
-	if (!monc->want_next_osdmap)
-		monc->want_next_osdmap = 1;
-	if (monc->want_next_osdmap < 2)
-		__send_subscribe(monc);
-	mutex_unlock(&monc->mutex);
-}
-
-/*
- *
- */
-int ceph_monc_open_session(struct ceph_mon_client *monc)
-{
-	if (!monc->con) {
-		monc->con = kmalloc(sizeof(*monc->con), GFP_KERNEL);
-		if (!monc->con)
-			return -ENOMEM;
-		ceph_con_init(monc->client->msgr, monc->con);
-		monc->con->private = monc;
-		monc->con->ops = &mon_con_ops;
-	}
-
-	mutex_lock(&monc->mutex);
-	__open_session(monc);
-	__schedule_delayed(monc);
-	mutex_unlock(&monc->mutex);
-	return 0;
-}
-
-/*
- * The monitor responds with mount ack indicate mount success.  The
- * included client ticket allows the client to talk to MDSs and OSDs.
- */
-static void ceph_monc_handle_map(struct ceph_mon_client *monc,
-				 struct ceph_msg *msg)
-{
-	struct ceph_client *client = monc->client;
-	struct ceph_monmap *monmap = NULL, *old = monc->monmap;
-	void *p, *end;
-
-	mutex_lock(&monc->mutex);
-
-	dout("handle_monmap\n");
-	p = msg->front.iov_base;
-	end = p + msg->front.iov_len;
-
-	monmap = ceph_monmap_decode(p, end);
-	if (IS_ERR(monmap)) {
-		pr_err("problem decoding monmap, %d\n",
-		       (int)PTR_ERR(monmap));
-		goto out;
-	}
-
-	if (ceph_check_fsid(monc->client, &monmap->fsid) < 0) {
-		kfree(monmap);
-		goto out;
-	}
-
-	client->monc.monmap = monmap;
-	kfree(old);
-
-out:
-	mutex_unlock(&monc->mutex);
-	wake_up_all(&client->auth_wq);
-}
-
-/*
- * generic requests (e.g., statfs, poolop)
- */
-static struct ceph_mon_generic_request *__lookup_generic_req(
-	struct ceph_mon_client *monc, u64 tid)
-{
-	struct ceph_mon_generic_request *req;
-	struct rb_node *n = monc->generic_request_tree.rb_node;
-
-	while (n) {
-		req = rb_entry(n, struct ceph_mon_generic_request, node);
-		if (tid < req->tid)
-			n = n->rb_left;
-		else if (tid > req->tid)
-			n = n->rb_right;
-		else
-			return req;
-	}
-	return NULL;
-}
-
-static void __insert_generic_request(struct ceph_mon_client *monc,
-			    struct ceph_mon_generic_request *new)
-{
-	struct rb_node **p = &monc->generic_request_tree.rb_node;
-	struct rb_node *parent = NULL;
-	struct ceph_mon_generic_request *req = NULL;
-
-	while (*p) {
-		parent = *p;
-		req = rb_entry(parent, struct ceph_mon_generic_request, node);
-		if (new->tid < req->tid)
-			p = &(*p)->rb_left;
-		else if (new->tid > req->tid)
-			p = &(*p)->rb_right;
-		else
-			BUG();
-	}
-
-	rb_link_node(&new->node, parent, p);
-	rb_insert_color(&new->node, &monc->generic_request_tree);
-}
-
-static void release_generic_request(struct kref *kref)
-{
-	struct ceph_mon_generic_request *req =
-		container_of(kref, struct ceph_mon_generic_request, kref);
-
-	if (req->reply)
-		ceph_msg_put(req->reply);
-	if (req->request)
-		ceph_msg_put(req->request);
-
-	kfree(req);
-}
-
-static void put_generic_request(struct ceph_mon_generic_request *req)
-{
-	kref_put(&req->kref, release_generic_request);
-}
-
-static void get_generic_request(struct ceph_mon_generic_request *req)
-{
-	kref_get(&req->kref);
-}
-
-static struct ceph_msg *get_generic_reply(struct ceph_connection *con,
-					 struct ceph_msg_header *hdr,
-					 int *skip)
-{
-	struct ceph_mon_client *monc = con->private;
-	struct ceph_mon_generic_request *req;
-	u64 tid = le64_to_cpu(hdr->tid);
-	struct ceph_msg *m;
-
-	mutex_lock(&monc->mutex);
-	req = __lookup_generic_req(monc, tid);
-	if (!req) {
-		dout("get_generic_reply %lld dne\n", tid);
-		*skip = 1;
-		m = NULL;
-	} else {
-		dout("get_generic_reply %lld got %p\n", tid, req->reply);
-		m = ceph_msg_get(req->reply);
-		/*
-		 * we don't need to track the connection reading into
-		 * this reply because we only have one open connection
-		 * at a time, ever.
-		 */
-	}
-	mutex_unlock(&monc->mutex);
-	return m;
-}
-
-static int do_generic_request(struct ceph_mon_client *monc,
-			      struct ceph_mon_generic_request *req)
-{
-	int err;
-
-	/* register request */
-	mutex_lock(&monc->mutex);
-	req->tid = ++monc->last_tid;
-	req->request->hdr.tid = cpu_to_le64(req->tid);
-	__insert_generic_request(monc, req);
-	monc->num_generic_requests++;
-	ceph_con_send(monc->con, ceph_msg_get(req->request));
-	mutex_unlock(&monc->mutex);
-
-	err = wait_for_completion_interruptible(&req->completion);
-
-	mutex_lock(&monc->mutex);
-	rb_erase(&req->node, &monc->generic_request_tree);
-	monc->num_generic_requests--;
-	mutex_unlock(&monc->mutex);
-
-	if (!err)
-		err = req->result;
-	return err;
-}
-
-/*
- * statfs
- */
-static void handle_statfs_reply(struct ceph_mon_client *monc,
-				struct ceph_msg *msg)
-{
-	struct ceph_mon_generic_request *req;
-	struct ceph_mon_statfs_reply *reply = msg->front.iov_base;
-	u64 tid = le64_to_cpu(msg->hdr.tid);
-
-	if (msg->front.iov_len != sizeof(*reply))
-		goto bad;
-	dout("handle_statfs_reply %p tid %llu\n", msg, tid);
-
-	mutex_lock(&monc->mutex);
-	req = __lookup_generic_req(monc, tid);
-	if (req) {
-		*(struct ceph_statfs *)req->buf = reply->st;
-		req->result = 0;
-		get_generic_request(req);
-	}
-	mutex_unlock(&monc->mutex);
-	if (req) {
-		complete_all(&req->completion);
-		put_generic_request(req);
-	}
-	return;
-
-bad:
-	pr_err("corrupt generic reply, tid %llu\n", tid);
-	ceph_msg_dump(msg);
-}
-
-/*
- * Do a synchronous statfs().
- */
-int ceph_monc_do_statfs(struct ceph_mon_client *monc, struct ceph_statfs *buf)
-{
-	struct ceph_mon_generic_request *req;
-	struct ceph_mon_statfs *h;
-	int err;
-
-	req = kzalloc(sizeof(*req), GFP_NOFS);
-	if (!req)
-		return -ENOMEM;
-
-	kref_init(&req->kref);
-	req->buf = buf;
-	req->buf_len = sizeof(*buf);
-	init_completion(&req->completion);
-
-	err = -ENOMEM;
-	req->request = ceph_msg_new(CEPH_MSG_STATFS, sizeof(*h), GFP_NOFS);
-	if (!req->request)
-		goto out;
-	req->reply = ceph_msg_new(CEPH_MSG_STATFS_REPLY, 1024, GFP_NOFS);
-	if (!req->reply)
-		goto out;
-
-	/* fill out request */
-	h = req->request->front.iov_base;
-	h->monhdr.have_version = 0;
-	h->monhdr.session_mon = cpu_to_le16(-1);
-	h->monhdr.session_mon_tid = 0;
-	h->fsid = monc->monmap->fsid;
-
-	err = do_generic_request(monc, req);
-
-out:
-	kref_put(&req->kref, release_generic_request);
-	return err;
-}
-
-/*
- * pool ops
- */
-static int get_poolop_reply_buf(const char *src, size_t src_len,
-				char *dst, size_t dst_len)
-{
-	u32 buf_len;
-
-	if (src_len != sizeof(u32) + dst_len)
-		return -EINVAL;
-
-	buf_len = le32_to_cpu(*(u32 *)src);
-	if (buf_len != dst_len)
-		return -EINVAL;
-
-	memcpy(dst, src + sizeof(u32), dst_len);
-	return 0;
-}
-
-static void handle_poolop_reply(struct ceph_mon_client *monc,
-				struct ceph_msg *msg)
-{
-	struct ceph_mon_generic_request *req;
-	struct ceph_mon_poolop_reply *reply = msg->front.iov_base;
-	u64 tid = le64_to_cpu(msg->hdr.tid);
-
-	if (msg->front.iov_len < sizeof(*reply))
-		goto bad;
-	dout("handle_poolop_reply %p tid %llu\n", msg, tid);
-
-	mutex_lock(&monc->mutex);
-	req = __lookup_generic_req(monc, tid);
-	if (req) {
-		if (req->buf_len &&
-		    get_poolop_reply_buf(msg->front.iov_base + sizeof(*reply),
-				     msg->front.iov_len - sizeof(*reply),
-				     req->buf, req->buf_len) < 0) {
-			mutex_unlock(&monc->mutex);
-			goto bad;
-		}
-		req->result = le32_to_cpu(reply->reply_code);
-		get_generic_request(req);
-	}
-	mutex_unlock(&monc->mutex);
-	if (req) {
-		complete(&req->completion);
-		put_generic_request(req);
-	}
-	return;
-
-bad:
-	pr_err("corrupt generic reply, tid %llu\n", tid);
-	ceph_msg_dump(msg);
-}
-
-/*
- * Do a synchronous pool op.
- */
-int ceph_monc_do_poolop(struct ceph_mon_client *monc, u32 op,
-			u32 pool, u64 snapid,
-			char *buf, int len)
-{
-	struct ceph_mon_generic_request *req;
-	struct ceph_mon_poolop *h;
-	int err;
-
-	req = kzalloc(sizeof(*req), GFP_NOFS);
-	if (!req)
-		return -ENOMEM;
-
-	kref_init(&req->kref);
-	req->buf = buf;
-	req->buf_len = len;
-	init_completion(&req->completion);
-
-	err = -ENOMEM;
-	req->request = ceph_msg_new(CEPH_MSG_POOLOP, sizeof(*h), GFP_NOFS);
-	if (!req->request)
-		goto out;
-	req->reply = ceph_msg_new(CEPH_MSG_POOLOP_REPLY, 1024, GFP_NOFS);
-	if (!req->reply)
-		goto out;
-
-	/* fill out request */
-	req->request->hdr.version = cpu_to_le16(2);
-	h = req->request->front.iov_base;
-	h->monhdr.have_version = 0;
-	h->monhdr.session_mon = cpu_to_le16(-1);
-	h->monhdr.session_mon_tid = 0;
-	h->fsid = monc->monmap->fsid;
-	h->pool = cpu_to_le32(pool);
-	h->op = cpu_to_le32(op);
-	h->auid = 0;
-	h->snapid = cpu_to_le64(snapid);
-	h->name_len = 0;
-
-	err = do_generic_request(monc, req);
-
-out:
-	kref_put(&req->kref, release_generic_request);
-	return err;
-}
-
-int ceph_monc_create_snapid(struct ceph_mon_client *monc,
-			    u32 pool, u64 *snapid)
-{
-	return ceph_monc_do_poolop(monc,  POOL_OP_CREATE_UNMANAGED_SNAP,
-				   pool, 0, (char *)snapid, sizeof(*snapid));
-
-}
-
-int ceph_monc_delete_snapid(struct ceph_mon_client *monc,
-			    u32 pool, u64 snapid)
-{
-	return ceph_monc_do_poolop(monc,  POOL_OP_CREATE_UNMANAGED_SNAP,
-				   pool, snapid, 0, 0);
-
-}
-
-/*
- * Resend pending generic requests.
- */
-static void __resend_generic_request(struct ceph_mon_client *monc)
-{
-	struct ceph_mon_generic_request *req;
-	struct rb_node *p;
-
-	for (p = rb_first(&monc->generic_request_tree); p; p = rb_next(p)) {
-		req = rb_entry(p, struct ceph_mon_generic_request, node);
-		ceph_con_revoke(monc->con, req->request);
-		ceph_con_send(monc->con, ceph_msg_get(req->request));
-	}
-}
-
-/*
- * Delayed work.  If we haven't mounted yet, retry.  Otherwise,
- * renew/retry subscription as needed (in case it is timing out, or we
- * got an ENOMEM).  And keep the monitor connection alive.
- */
-static void delayed_work(struct work_struct *work)
-{
-	struct ceph_mon_client *monc =
-		container_of(work, struct ceph_mon_client, delayed_work.work);
-
-	dout("monc delayed_work\n");
-	mutex_lock(&monc->mutex);
-	if (monc->hunting) {
-		__close_session(monc);
-		__open_session(monc);  /* continue hunting */
-	} else {
-		ceph_con_keepalive(monc->con);
-
-		__validate_auth(monc);
-
-		if (monc->auth->ops->is_authenticated(monc->auth))
-			__send_subscribe(monc);
-	}
-	__schedule_delayed(monc);
-	mutex_unlock(&monc->mutex);
-}
-
-/*
- * On startup, we build a temporary monmap populated with the IPs
- * provided by mount(2).
- */
-static int build_initial_monmap(struct ceph_mon_client *monc)
-{
-	struct ceph_mount_args *args = monc->client->mount_args;
-	struct ceph_entity_addr *mon_addr = args->mon_addr;
-	int num_mon = args->num_mon;
-	int i;
-
-	/* build initial monmap */
-	monc->monmap = kzalloc(sizeof(*monc->monmap) +
-			       num_mon*sizeof(monc->monmap->mon_inst[0]),
-			       GFP_KERNEL);
-	if (!monc->monmap)
-		return -ENOMEM;
-	for (i = 0; i < num_mon; i++) {
-		monc->monmap->mon_inst[i].addr = mon_addr[i];
-		monc->monmap->mon_inst[i].addr.nonce = 0;
-		monc->monmap->mon_inst[i].name.type =
-			CEPH_ENTITY_TYPE_MON;
-		monc->monmap->mon_inst[i].name.num = cpu_to_le64(i);
-	}
-	monc->monmap->num_mon = num_mon;
-	monc->have_fsid = false;
-
-	/* release addr memory */
-	kfree(args->mon_addr);
-	args->mon_addr = NULL;
-	args->num_mon = 0;
-	return 0;
-}
-
-int ceph_monc_init(struct ceph_mon_client *monc, struct ceph_client *cl)
-{
-	int err = 0;
-
-	dout("init\n");
-	memset(monc, 0, sizeof(*monc));
-	monc->client = cl;
-	monc->monmap = NULL;
-	mutex_init(&monc->mutex);
-
-	err = build_initial_monmap(monc);
-	if (err)
-		goto out;
-
-	monc->con = NULL;
-
-	/* authentication */
-	monc->auth = ceph_auth_init(cl->mount_args->name,
-				    cl->mount_args->secret);
-	if (IS_ERR(monc->auth))
-		return PTR_ERR(monc->auth);
-	monc->auth->want_keys =
-		CEPH_ENTITY_TYPE_AUTH | CEPH_ENTITY_TYPE_MON |
-		CEPH_ENTITY_TYPE_OSD | CEPH_ENTITY_TYPE_MDS;
-
-	/* msgs */
-	err = -ENOMEM;
-	monc->m_subscribe_ack = ceph_msg_new(CEPH_MSG_MON_SUBSCRIBE_ACK,
-				     sizeof(struct ceph_mon_subscribe_ack),
-				     GFP_NOFS);
-	if (!monc->m_subscribe_ack)
-		goto out_monmap;
-
-	monc->m_subscribe = ceph_msg_new(CEPH_MSG_MON_SUBSCRIBE, 96, GFP_NOFS);
-	if (!monc->m_subscribe)
-		goto out_subscribe_ack;
-
-	monc->m_auth_reply = ceph_msg_new(CEPH_MSG_AUTH_REPLY, 4096, GFP_NOFS);
-	if (!monc->m_auth_reply)
-		goto out_subscribe;
-
-	monc->m_auth = ceph_msg_new(CEPH_MSG_AUTH, 4096, GFP_NOFS);
-	monc->pending_auth = 0;
-	if (!monc->m_auth)
-		goto out_auth_reply;
-
-	monc->cur_mon = -1;
-	monc->hunting = true;
-	monc->sub_renew_after = jiffies;
-	monc->sub_sent = 0;
-
-	INIT_DELAYED_WORK(&monc->delayed_work, delayed_work);
-	monc->generic_request_tree = RB_ROOT;
-	monc->num_generic_requests = 0;
-	monc->last_tid = 0;
-
-	monc->have_mdsmap = 0;
-	monc->have_osdmap = 0;
-	monc->want_next_osdmap = 1;
-	return 0;
-
-out_auth_reply:
-	ceph_msg_put(monc->m_auth_reply);
-out_subscribe:
-	ceph_msg_put(monc->m_subscribe);
-out_subscribe_ack:
-	ceph_msg_put(monc->m_subscribe_ack);
-out_monmap:
-	kfree(monc->monmap);
-out:
-	return err;
-}
-
-void ceph_monc_stop(struct ceph_mon_client *monc)
-{
-	dout("stop\n");
-	cancel_delayed_work_sync(&monc->delayed_work);
-
-	mutex_lock(&monc->mutex);
-	__close_session(monc);
-	if (monc->con) {
-		monc->con->private = NULL;
-		monc->con->ops->put(monc->con);
-		monc->con = NULL;
-	}
-	mutex_unlock(&monc->mutex);
-
-	ceph_auth_destroy(monc->auth);
-
-	ceph_msg_put(monc->m_auth);
-	ceph_msg_put(monc->m_auth_reply);
-	ceph_msg_put(monc->m_subscribe);
-	ceph_msg_put(monc->m_subscribe_ack);
-
-	kfree(monc->monmap);
-}
-
-static void handle_auth_reply(struct ceph_mon_client *monc,
-			      struct ceph_msg *msg)
-{
-	int ret;
-	int was_auth = 0;
-
-	mutex_lock(&monc->mutex);
-	if (monc->auth->ops)
-		was_auth = monc->auth->ops->is_authenticated(monc->auth);
-	monc->pending_auth = 0;
-	ret = ceph_handle_auth_reply(monc->auth, msg->front.iov_base,
-				     msg->front.iov_len,
-				     monc->m_auth->front.iov_base,
-				     monc->m_auth->front_max);
-	if (ret < 0) {
-		monc->client->auth_err = ret;
-		wake_up_all(&monc->client->auth_wq);
-	} else if (ret > 0) {
-		__send_prepared_auth_request(monc, ret);
-	} else if (!was_auth && monc->auth->ops->is_authenticated(monc->auth)) {
-		dout("authenticated, starting session\n");
-
-		monc->client->msgr->inst.name.type = CEPH_ENTITY_TYPE_CLIENT;
-		monc->client->msgr->inst.name.num =
-					cpu_to_le64(monc->auth->global_id);
-
-		__send_subscribe(monc);
-		__resend_generic_request(monc);
-	}
-	mutex_unlock(&monc->mutex);
-}
-
-static int __validate_auth(struct ceph_mon_client *monc)
-{
-	int ret;
-
-	if (monc->pending_auth)
-		return 0;
-
-	ret = ceph_build_auth(monc->auth, monc->m_auth->front.iov_base,
-			      monc->m_auth->front_max);
-	if (ret <= 0)
-		return ret; /* either an error, or no need to authenticate */
-	__send_prepared_auth_request(monc, ret);
-	return 0;
-}
-
-int ceph_monc_validate_auth(struct ceph_mon_client *monc)
-{
-	int ret;
-
-	mutex_lock(&monc->mutex);
-	ret = __validate_auth(monc);
-	mutex_unlock(&monc->mutex);
-	return ret;
-}
-
-/*
- * handle incoming message
- */
-static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
-{
-	struct ceph_mon_client *monc = con->private;
-	int type = le16_to_cpu(msg->hdr.type);
-
-	if (!monc)
-		return;
-
-	switch (type) {
-	case CEPH_MSG_AUTH_REPLY:
-		handle_auth_reply(monc, msg);
-		break;
-
-	case CEPH_MSG_MON_SUBSCRIBE_ACK:
-		handle_subscribe_ack(monc, msg);
-		break;
-
-	case CEPH_MSG_STATFS_REPLY:
-		handle_statfs_reply(monc, msg);
-		break;
-
-	case CEPH_MSG_POOLOP_REPLY:
-		handle_poolop_reply(monc, msg);
-		break;
-
-	case CEPH_MSG_MON_MAP:
-		ceph_monc_handle_map(monc, msg);
-		break;
-
-	case CEPH_MSG_MDS_MAP:
-		ceph_mdsc_handle_map(&monc->client->mdsc, msg);
-		break;
-
-	case CEPH_MSG_OSD_MAP:
-		ceph_osdc_handle_map(&monc->client->osdc, msg);
-		break;
-
-	default:
-		pr_err("received unknown message type %d %s\n", type,
-		       ceph_msg_type_name(type));
-	}
-	ceph_msg_put(msg);
-}
-
-/*
- * Allocate memory for incoming message
- */
-static struct ceph_msg *mon_alloc_msg(struct ceph_connection *con,
-				      struct ceph_msg_header *hdr,
-				      int *skip)
-{
-	struct ceph_mon_client *monc = con->private;
-	int type = le16_to_cpu(hdr->type);
-	int front_len = le32_to_cpu(hdr->front_len);
-	struct ceph_msg *m = NULL;
-
-	*skip = 0;
-
-	switch (type) {
-	case CEPH_MSG_MON_SUBSCRIBE_ACK:
-		m = ceph_msg_get(monc->m_subscribe_ack);
-		break;
-	case CEPH_MSG_POOLOP_REPLY:
-	case CEPH_MSG_STATFS_REPLY:
-		return get_generic_reply(con, hdr, skip);
-	case CEPH_MSG_AUTH_REPLY:
-		m = ceph_msg_get(monc->m_auth_reply);
-		break;
-	case CEPH_MSG_MON_MAP:
-	case CEPH_MSG_MDS_MAP:
-	case CEPH_MSG_OSD_MAP:
-		m = ceph_msg_new(type, front_len, GFP_NOFS);
-		break;
-	}
-
-	if (!m) {
-		pr_info("alloc_msg unknown type %d\n", type);
-		*skip = 1;
-	}
-	return m;
-}
-
-/*
- * If the monitor connection resets, pick a new monitor and resubmit
- * any pending requests.
- */
-static void mon_fault(struct ceph_connection *con)
-{
-	struct ceph_mon_client *monc = con->private;
-
-	if (!monc)
-		return;
-
-	dout("mon_fault\n");
-	mutex_lock(&monc->mutex);
-	if (!con->private)
-		goto out;
-
-	if (monc->con && !monc->hunting)
-		pr_info("mon%d %s session lost, "
-			"hunting for new mon\n", monc->cur_mon,
-			pr_addr(&monc->con->peer_addr.in_addr));
-
-	__close_session(monc);
-	if (!monc->hunting) {
-		/* start hunting */
-		monc->hunting = true;
-		__open_session(monc);
-	} else {
-		/* already hunting, let's wait a bit */
-		__schedule_delayed(monc);
-	}
-out:
-	mutex_unlock(&monc->mutex);
-}
-
-static const struct ceph_connection_operations mon_con_ops = {
-	.get = ceph_con_get,
-	.put = ceph_con_put,
-	.dispatch = dispatch,
-	.fault = mon_fault,
-	.alloc_msg = mon_alloc_msg,
-};
diff --git a/fs/ceph/mon_client.h b/fs/ceph/mon_client.h
deleted file mode 100644
index 8e396f2c0963..000000000000
--- a/fs/ceph/mon_client.h
+++ /dev/null
@@ -1,121 +0,0 @@
-#ifndef _FS_CEPH_MON_CLIENT_H
-#define _FS_CEPH_MON_CLIENT_H
-
-#include <linux/completion.h>
-#include <linux/kref.h>
-#include <linux/rbtree.h>
-
-#include "messenger.h"
-
-struct ceph_client;
-struct ceph_mount_args;
-struct ceph_auth_client;
-
-/*
- * The monitor map enumerates the set of all monitors.
- */
-struct ceph_monmap {
-	struct ceph_fsid fsid;
-	u32 epoch;
-	u32 num_mon;
-	struct ceph_entity_inst mon_inst[0];
-};
-
-struct ceph_mon_client;
-struct ceph_mon_generic_request;
-
-
-/*
- * Generic mechanism for resending monitor requests.
- */
-typedef void (*ceph_monc_request_func_t)(struct ceph_mon_client *monc,
-					 int newmon);
-
-/* a pending monitor request */
-struct ceph_mon_request {
-	struct ceph_mon_client *monc;
-	struct delayed_work delayed_work;
-	unsigned long delay;
-	ceph_monc_request_func_t do_request;
-};
-
-/*
- * ceph_mon_generic_request is being used for the statfs and poolop requests
- * which are bening done a bit differently because we need to get data back
- * to the caller
- */
-struct ceph_mon_generic_request {
-	struct kref kref;
-	u64 tid;
-	struct rb_node node;
-	int result;
-	void *buf;
-	int buf_len;
-	struct completion completion;
-	struct ceph_msg *request;  /* original request */
-	struct ceph_msg *reply;    /* and reply */
-};
-
-struct ceph_mon_client {
-	struct ceph_client *client;
-	struct ceph_monmap *monmap;
-
-	struct mutex mutex;
-	struct delayed_work delayed_work;
-
-	struct ceph_auth_client *auth;
-	struct ceph_msg *m_auth, *m_auth_reply, *m_subscribe, *m_subscribe_ack;
-	int pending_auth;
-
-	bool hunting;
-	int cur_mon;                       /* last monitor i contacted */
-	unsigned long sub_sent, sub_renew_after;
-	struct ceph_connection *con;
-	bool have_fsid;
-
-	/* pending generic requests */
-	struct rb_root generic_request_tree;
-	int num_generic_requests;
-	u64 last_tid;
-
-	/* mds/osd map */
-	int want_next_osdmap; /* 1 = want, 2 = want+asked */
-	u32 have_osdmap, have_mdsmap;
-
-#ifdef CONFIG_DEBUG_FS
-	struct dentry *debugfs_file;
-#endif
-};
-
-extern struct ceph_monmap *ceph_monmap_decode(void *p, void *end);
-extern int ceph_monmap_contains(struct ceph_monmap *m,
-				struct ceph_entity_addr *addr);
-
-extern int ceph_monc_init(struct ceph_mon_client *monc, struct ceph_client *cl);
-extern void ceph_monc_stop(struct ceph_mon_client *monc);
-
-/*
- * The model here is to indicate that we need a new map of at least
- * epoch @want, and also call in when we receive a map.  We will
- * periodically rerequest the map from the monitor cluster until we
- * get what we want.
- */
-extern int ceph_monc_got_mdsmap(struct ceph_mon_client *monc, u32 have);
-extern int ceph_monc_got_osdmap(struct ceph_mon_client *monc, u32 have);
-
-extern void ceph_monc_request_next_osdmap(struct ceph_mon_client *monc);
-
-extern int ceph_monc_do_statfs(struct ceph_mon_client *monc,
-			       struct ceph_statfs *buf);
-
-extern int ceph_monc_open_session(struct ceph_mon_client *monc);
-
-extern int ceph_monc_validate_auth(struct ceph_mon_client *monc);
-
-extern int ceph_monc_create_snapid(struct ceph_mon_client *monc,
-				   u32 pool, u64 *snapid);
-
-extern int ceph_monc_delete_snapid(struct ceph_mon_client *monc,
-				   u32 pool, u64 snapid);
-
-#endif
diff --git a/fs/ceph/msgpool.c b/fs/ceph/msgpool.c
deleted file mode 100644
index dd65a6438131..000000000000
--- a/fs/ceph/msgpool.c
+++ /dev/null
@@ -1,64 +0,0 @@
-#include "ceph_debug.h"
-
-#include <linux/err.h>
-#include <linux/sched.h>
-#include <linux/types.h>
-#include <linux/vmalloc.h>
-
-#include "msgpool.h"
-
-static void *alloc_fn(gfp_t gfp_mask, void *arg)
-{
-	struct ceph_msgpool *pool = arg;
-	void *p;
-
-	p = ceph_msg_new(0, pool->front_len, gfp_mask);
-	if (!p)
-		pr_err("msgpool %s alloc failed\n", pool->name);
-	return p;
-}
-
-static void free_fn(void *element, void *arg)
-{
-	ceph_msg_put(element);
-}
-
-int ceph_msgpool_init(struct ceph_msgpool *pool,
-		      int front_len, int size, bool blocking, const char *name)
-{
-	pool->front_len = front_len;
-	pool->pool = mempool_create(size, alloc_fn, free_fn, pool);
-	if (!pool->pool)
-		return -ENOMEM;
-	pool->name = name;
-	return 0;
-}
-
-void ceph_msgpool_destroy(struct ceph_msgpool *pool)
-{
-	mempool_destroy(pool->pool);
-}
-
-struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool,
-				  int front_len)
-{
-	if (front_len > pool->front_len) {
-		pr_err("msgpool_get pool %s need front %d, pool size is %d\n",
-		       pool->name, front_len, pool->front_len);
-		WARN_ON(1);
-
-		/* try to alloc a fresh message */
-		return ceph_msg_new(0, front_len, GFP_NOFS);
-	}
-
-	return mempool_alloc(pool->pool, GFP_NOFS);
-}
-
-void ceph_msgpool_put(struct ceph_msgpool *pool, struct ceph_msg *msg)
-{
-	/* reset msg front_len; user may have changed it */
-	msg->front.iov_len = pool->front_len;
-	msg->hdr.front_len = cpu_to_le32(pool->front_len);
-
-	kref_init(&msg->kref);  /* retake single ref */
-}
diff --git a/fs/ceph/msgpool.h b/fs/ceph/msgpool.h
deleted file mode 100644
index a362605f9368..000000000000
--- a/fs/ceph/msgpool.h
+++ /dev/null
@@ -1,25 +0,0 @@
-#ifndef _FS_CEPH_MSGPOOL
-#define _FS_CEPH_MSGPOOL
-
-#include <linux/mempool.h>
-#include "messenger.h"
-
-/*
- * we use memory pools for preallocating messages we may receive, to
- * avoid unexpected OOM conditions.
- */
-struct ceph_msgpool {
-	const char *name;
-	mempool_t *pool;
-	int front_len;          /* preallocated payload size */
-};
-
-extern int ceph_msgpool_init(struct ceph_msgpool *pool,
-			     int front_len, int size, bool blocking,
-			     const char *name);
-extern void ceph_msgpool_destroy(struct ceph_msgpool *pool);
-extern struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *,
-					 int front_len);
-extern void ceph_msgpool_put(struct ceph_msgpool *, struct ceph_msg *);
-
-#endif
diff --git a/fs/ceph/msgr.h b/fs/ceph/msgr.h
deleted file mode 100644
index 680d3d648cac..000000000000
--- a/fs/ceph/msgr.h
+++ /dev/null
@@ -1,175 +0,0 @@
-#ifndef CEPH_MSGR_H
-#define CEPH_MSGR_H
-
-/*
- * Data types for message passing layer used by Ceph.
- */
-
-#define CEPH_MON_PORT    6789  /* default monitor port */
-
-/*
- * client-side processes will try to bind to ports in this
- * range, simply for the benefit of tools like nmap or wireshark
- * that would like to identify the protocol.
- */
-#define CEPH_PORT_FIRST  6789
-#define CEPH_PORT_START  6800  /* non-monitors start here */
-#define CEPH_PORT_LAST   6900
-
-/*
- * tcp connection banner.  include a protocol version. and adjust
- * whenever the wire protocol changes.  try to keep this string length
- * constant.
- */
-#define CEPH_BANNER "ceph v027"
-#define CEPH_BANNER_MAX_LEN 30
-
-
-/*
- * Rollover-safe type and comparator for 32-bit sequence numbers.
- * Comparator returns -1, 0, or 1.
- */
-typedef __u32 ceph_seq_t;
-
-static inline __s32 ceph_seq_cmp(__u32 a, __u32 b)
-{
-       return (__s32)a - (__s32)b;
-}
-
-
-/*
- * entity_name -- logical name for a process participating in the
- * network, e.g. 'mds0' or 'osd3'.
- */
-struct ceph_entity_name {
-	__u8 type;      /* CEPH_ENTITY_TYPE_* */
-	__le64 num;
-} __attribute__ ((packed));
-
-#define CEPH_ENTITY_TYPE_MON    0x01
-#define CEPH_ENTITY_TYPE_MDS    0x02
-#define CEPH_ENTITY_TYPE_OSD    0x04
-#define CEPH_ENTITY_TYPE_CLIENT 0x08
-#define CEPH_ENTITY_TYPE_AUTH   0x20
-
-#define CEPH_ENTITY_TYPE_ANY    0xFF
-
-extern const char *ceph_entity_type_name(int type);
-
-/*
- * entity_addr -- network address
- */
-struct ceph_entity_addr {
-	__le32 type;
-	__le32 nonce;  /* unique id for process (e.g. pid) */
-	struct sockaddr_storage in_addr;
-} __attribute__ ((packed));
-
-struct ceph_entity_inst {
-	struct ceph_entity_name name;
-	struct ceph_entity_addr addr;
-} __attribute__ ((packed));
-
-
-/* used by message exchange protocol */
-#define CEPH_MSGR_TAG_READY         1  /* server->client: ready for messages */
-#define CEPH_MSGR_TAG_RESETSESSION  2  /* server->client: reset, try again */
-#define CEPH_MSGR_TAG_WAIT          3  /* server->client: wait for racing
-					  incoming connection */
-#define CEPH_MSGR_TAG_RETRY_SESSION 4  /* server->client + cseq: try again
-					  with higher cseq */
-#define CEPH_MSGR_TAG_RETRY_GLOBAL  5  /* server->client + gseq: try again
-					  with higher gseq */
-#define CEPH_MSGR_TAG_CLOSE         6  /* closing pipe */
-#define CEPH_MSGR_TAG_MSG           7  /* message */
-#define CEPH_MSGR_TAG_ACK           8  /* message ack */
-#define CEPH_MSGR_TAG_KEEPALIVE     9  /* just a keepalive byte! */
-#define CEPH_MSGR_TAG_BADPROTOVER  10  /* bad protocol version */
-#define CEPH_MSGR_TAG_BADAUTHORIZER 11 /* bad authorizer */
-#define CEPH_MSGR_TAG_FEATURES      12 /* insufficient features */
-
-
-/*
- * connection negotiation
- */
-struct ceph_msg_connect {
-	__le64 features;     /* supported feature bits */
-	__le32 host_type;    /* CEPH_ENTITY_TYPE_* */
-	__le32 global_seq;   /* count connections initiated by this host */
-	__le32 connect_seq;  /* count connections initiated in this session */
-	__le32 protocol_version;
-	__le32 authorizer_protocol;
-	__le32 authorizer_len;
-	__u8  flags;         /* CEPH_MSG_CONNECT_* */
-} __attribute__ ((packed));
-
-struct ceph_msg_connect_reply {
-	__u8 tag;
-	__le64 features;     /* feature bits for this session */
-	__le32 global_seq;
-	__le32 connect_seq;
-	__le32 protocol_version;
-	__le32 authorizer_len;
-	__u8 flags;
-} __attribute__ ((packed));
-
-#define CEPH_MSG_CONNECT_LOSSY  1  /* messages i send may be safely dropped */
-
-
-/*
- * message header
- */
-struct ceph_msg_header_old {
-	__le64 seq;       /* message seq# for this session */
-	__le64 tid;       /* transaction id */
-	__le16 type;      /* message type */
-	__le16 priority;  /* priority.  higher value == higher priority */
-	__le16 version;   /* version of message encoding */
-
-	__le32 front_len; /* bytes in main payload */
-	__le32 middle_len;/* bytes in middle payload */
-	__le32 data_len;  /* bytes of data payload */
-	__le16 data_off;  /* sender: include full offset;
-			     receiver: mask against ~PAGE_MASK */
-
-	struct ceph_entity_inst src, orig_src;
-	__le32 reserved;
-	__le32 crc;       /* header crc32c */
-} __attribute__ ((packed));
-
-struct ceph_msg_header {
-	__le64 seq;       /* message seq# for this session */
-	__le64 tid;       /* transaction id */
-	__le16 type;      /* message type */
-	__le16 priority;  /* priority.  higher value == higher priority */
-	__le16 version;   /* version of message encoding */
-
-	__le32 front_len; /* bytes in main payload */
-	__le32 middle_len;/* bytes in middle payload */
-	__le32 data_len;  /* bytes of data payload */
-	__le16 data_off;  /* sender: include full offset;
-			     receiver: mask against ~PAGE_MASK */
-
-	struct ceph_entity_name src;
-	__le32 reserved;
-	__le32 crc;       /* header crc32c */
-} __attribute__ ((packed));
-
-#define CEPH_MSG_PRIO_LOW     64
-#define CEPH_MSG_PRIO_DEFAULT 127
-#define CEPH_MSG_PRIO_HIGH    196
-#define CEPH_MSG_PRIO_HIGHEST 255
-
-/*
- * follows data payload
- */
-struct ceph_msg_footer {
-	__le32 front_crc, middle_crc, data_crc;
-	__u8 flags;
-} __attribute__ ((packed));
-
-#define CEPH_MSG_FOOTER_COMPLETE  (1<<0)   /* msg wasn't aborted */
-#define CEPH_MSG_FOOTER_NOCRC     (1<<1)   /* no data crc */
-
-
-#endif
diff --git a/fs/ceph/osd_client.c b/fs/ceph/osd_client.c
deleted file mode 100644
index 0edb43f1ef26..000000000000
--- a/fs/ceph/osd_client.c
+++ /dev/null
@@ -1,1771 +0,0 @@
-#include "ceph_debug.h"
-
-#include <linux/err.h>
-#include <linux/highmem.h>
-#include <linux/mm.h>
-#include <linux/pagemap.h>
-#include <linux/slab.h>
-#include <linux/uaccess.h>
-#ifdef CONFIG_BLOCK
-#include <linux/bio.h>
-#endif
-
-#include "super.h"
-#include "osd_client.h"
-#include "messenger.h"
-#include "decode.h"
-#include "auth.h"
-#include "pagelist.h"
-
-#define OSD_OP_FRONT_LEN	4096
-#define OSD_OPREPLY_FRONT_LEN	512
-
-static const struct ceph_connection_operations osd_con_ops;
-static int __kick_requests(struct ceph_osd_client *osdc,
-			  struct ceph_osd *kickosd);
-
-static void kick_requests(struct ceph_osd_client *osdc, struct ceph_osd *osd);
-
-static int op_needs_trail(int op)
-{
-	switch (op) {
-	case CEPH_OSD_OP_GETXATTR:
-	case CEPH_OSD_OP_SETXATTR:
-	case CEPH_OSD_OP_CMPXATTR:
-	case CEPH_OSD_OP_CALL:
-		return 1;
-	default:
-		return 0;
-	}
-}
-
-static int op_has_extent(int op)
-{
-	return (op == CEPH_OSD_OP_READ ||
-		op == CEPH_OSD_OP_WRITE);
-}
-
-void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
-			struct ceph_file_layout *layout,
-			u64 snapid,
-			u64 off, u64 *plen, u64 *bno,
-			struct ceph_osd_request *req,
-			struct ceph_osd_req_op *op)
-{
-	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
-	u64 orig_len = *plen;
-	u64 objoff, objlen;    /* extent in object */
-
-	reqhead->snapid = cpu_to_le64(snapid);
-
-	/* object extent? */
-	ceph_calc_file_object_mapping(layout, off, plen, bno,
-				      &objoff, &objlen);
-	if (*plen < orig_len)
-		dout(" skipping last %llu, final file extent %llu~%llu\n",
-		     orig_len - *plen, off, *plen);
-
-	if (op_has_extent(op->op)) {
-		op->extent.offset = objoff;
-		op->extent.length = objlen;
-	}
-	req->r_num_pages = calc_pages_for(off, *plen);
-
-	dout("calc_layout bno=%llx %llu~%llu (%d pages)\n",
-	     *bno, objoff, objlen, req->r_num_pages);
-
-}
-
-/*
- * Implement client access to distributed object storage cluster.
- *
- * All data objects are stored within a cluster/cloud of OSDs, or
- * "object storage devices."  (Note that Ceph OSDs have _nothing_ to
- * do with the T10 OSD extensions to SCSI.)  Ceph OSDs are simply
- * remote daemons serving up and coordinating consistent and safe
- * access to storage.
- *
- * Cluster membership and the mapping of data objects onto storage devices
- * are described by the osd map.
- *
- * We keep track of pending OSD requests (read, write), resubmit
- * requests to different OSDs when the cluster topology/data layout
- * change, or retry the affected requests when the communications
- * channel with an OSD is reset.
- */
-
-/*
- * calculate the mapping of a file extent onto an object, and fill out the
- * request accordingly.  shorten extent as necessary if it crosses an
- * object boundary.
- *
- * fill osd op in request message.
- */
-static void calc_layout(struct ceph_osd_client *osdc,
-			struct ceph_vino vino,
-			struct ceph_file_layout *layout,
-			u64 off, u64 *plen,
-			struct ceph_osd_request *req,
-			struct ceph_osd_req_op *op)
-{
-	u64 bno;
-
-	ceph_calc_raw_layout(osdc, layout, vino.snap, off,
-			     plen, &bno, req, op);
-
-	sprintf(req->r_oid, "%llx.%08llx", vino.ino, bno);
-	req->r_oid_len = strlen(req->r_oid);
-}
-
-/*
- * requests
- */
-void ceph_osdc_release_request(struct kref *kref)
-{
-	struct ceph_osd_request *req = container_of(kref,
-						    struct ceph_osd_request,
-						    r_kref);
-
-	if (req->r_request)
-		ceph_msg_put(req->r_request);
-	if (req->r_reply)
-		ceph_msg_put(req->r_reply);
-	if (req->r_con_filling_msg) {
-		dout("release_request revoking pages %p from con %p\n",
-		     req->r_pages, req->r_con_filling_msg);
-		ceph_con_revoke_message(req->r_con_filling_msg,
-				      req->r_reply);
-		ceph_con_put(req->r_con_filling_msg);
-	}
-	if (req->r_own_pages)
-		ceph_release_page_vector(req->r_pages,
-					 req->r_num_pages);
-#ifdef CONFIG_BLOCK
-	if (req->r_bio)
-		bio_put(req->r_bio);
-#endif
-	ceph_put_snap_context(req->r_snapc);
-	if (req->r_trail) {
-		ceph_pagelist_release(req->r_trail);
-		kfree(req->r_trail);
-	}
-	if (req->r_mempool)
-		mempool_free(req, req->r_osdc->req_mempool);
-	else
-		kfree(req);
-}
-
-static int op_needs_trail(int op)
-{
-	switch (op) {
-	case CEPH_OSD_OP_GETXATTR:
-	case CEPH_OSD_OP_SETXATTR:
-	case CEPH_OSD_OP_CMPXATTR:
-	case CEPH_OSD_OP_CALL:
-		return 1;
-	default:
-		return 0;
-	}
-}
-
-static int get_num_ops(struct ceph_osd_req_op *ops, int *needs_trail)
-{
-	int i = 0;
-
-	if (needs_trail)
-		*needs_trail = 0;
-	while (ops[i].op) {
-		if (needs_trail && op_needs_trail(ops[i].op))
-			*needs_trail = 1;
-		i++;
-	}
-
-	return i;
-}
-
-struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
-					       int flags,
-					       struct ceph_snap_context *snapc,
-					       struct ceph_osd_req_op *ops,
-					       bool use_mempool,
-					       gfp_t gfp_flags,
-					       struct page **pages,
-					       struct bio *bio)
-{
-	struct ceph_osd_request *req;
-	struct ceph_msg *msg;
-	int needs_trail;
-	int num_op = get_num_ops(ops, &needs_trail);
-	size_t msg_size = sizeof(struct ceph_osd_request_head);
-
-	msg_size += num_op*sizeof(struct ceph_osd_op);
-
-	if (use_mempool) {
-		req = mempool_alloc(osdc->req_mempool, gfp_flags);
-		memset(req, 0, sizeof(*req));
-	} else {
-		req = kzalloc(sizeof(*req), gfp_flags);
-	}
-	if (req == NULL)
-		return NULL;
-
-	req->r_osdc = osdc;
-	req->r_mempool = use_mempool;
-
-	kref_init(&req->r_kref);
-	init_completion(&req->r_completion);
-	init_completion(&req->r_safe_completion);
-	INIT_LIST_HEAD(&req->r_unsafe_item);
-	req->r_flags = flags;
-
-	WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0);
-
-	/* create reply message */
-	if (use_mempool)
-		msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0);
-	else
-		msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY,
-				   OSD_OPREPLY_FRONT_LEN, gfp_flags);
-	if (!msg) {
-		ceph_osdc_put_request(req);
-		return NULL;
-	}
-	req->r_reply = msg;
-
-	/* allocate space for the trailing data */
-	if (needs_trail) {
-		req->r_trail = kmalloc(sizeof(struct ceph_pagelist), gfp_flags);
-		if (!req->r_trail) {
-			ceph_osdc_put_request(req);
-			return NULL;
-		}
-		ceph_pagelist_init(req->r_trail);
-	}
-	/* create request message; allow space for oid */
-	msg_size += 40;
-	if (snapc)
-		msg_size += sizeof(u64) * snapc->num_snaps;
-	if (use_mempool)
-		msg = ceph_msgpool_get(&osdc->msgpool_op, 0);
-	else
-		msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, gfp_flags);
-	if (!msg) {
-		ceph_osdc_put_request(req);
-		return NULL;
-	}
-
-	msg->hdr.type = cpu_to_le16(CEPH_MSG_OSD_OP);
-	memset(msg->front.iov_base, 0, msg->front.iov_len);
-
-	req->r_request = msg;
-	req->r_pages = pages;
-#ifdef CONFIG_BLOCK
-	if (bio) {
-		req->r_bio = bio;
-		bio_get(req->r_bio);
-	}
-#endif
-
-	return req;
-}
-
-static void osd_req_encode_op(struct ceph_osd_request *req,
-			      struct ceph_osd_op *dst,
-			      struct ceph_osd_req_op *src)
-{
-	dst->op = cpu_to_le16(src->op);
-
-	switch (dst->op) {
-	case CEPH_OSD_OP_READ:
-	case CEPH_OSD_OP_WRITE:
-		dst->extent.offset =
-			cpu_to_le64(src->extent.offset);
-		dst->extent.length =
-			cpu_to_le64(src->extent.length);
-		dst->extent.truncate_size =
-			cpu_to_le64(src->extent.truncate_size);
-		dst->extent.truncate_seq =
-			cpu_to_le32(src->extent.truncate_seq);
-		break;
-
-	case CEPH_OSD_OP_GETXATTR:
-	case CEPH_OSD_OP_SETXATTR:
-	case CEPH_OSD_OP_CMPXATTR:
-		BUG_ON(!req->r_trail);
-
-		dst->xattr.name_len = cpu_to_le32(src->xattr.name_len);
-		dst->xattr.value_len = cpu_to_le32(src->xattr.value_len);
-		dst->xattr.cmp_op = src->xattr.cmp_op;
-		dst->xattr.cmp_mode = src->xattr.cmp_mode;
-		ceph_pagelist_append(req->r_trail, src->xattr.name,
-				     src->xattr.name_len);
-		ceph_pagelist_append(req->r_trail, src->xattr.val,
-				     src->xattr.value_len);
-		break;
-	case CEPH_OSD_OP_CALL:
-		BUG_ON(!req->r_trail);
-
-		dst->cls.class_len = src->cls.class_len;
-		dst->cls.method_len = src->cls.method_len;
-		dst->cls.indata_len = cpu_to_le32(src->cls.indata_len);
-
-		ceph_pagelist_append(req->r_trail, src->cls.class_name,
-				     src->cls.class_len);
-		ceph_pagelist_append(req->r_trail, src->cls.method_name,
-				     src->cls.method_len);
-		ceph_pagelist_append(req->r_trail, src->cls.indata,
-				     src->cls.indata_len);
-		break;
-	case CEPH_OSD_OP_ROLLBACK:
-		dst->snap.snapid = cpu_to_le64(src->snap.snapid);
-		break;
-	case CEPH_OSD_OP_STARTSYNC:
-		break;
-	default:
-		pr_err("unrecognized osd opcode %d\n", dst->op);
-		WARN_ON(1);
-		break;
-	}
-	dst->payload_len = cpu_to_le32(src->payload_len);
-}
-
-/*
- * build new request AND message
- *
- */
-void ceph_osdc_build_request(struct ceph_osd_request *req,
-			     u64 off, u64 *plen,
-			     struct ceph_osd_req_op *src_ops,
-			     struct ceph_snap_context *snapc,
-			     struct timespec *mtime,
-			     const char *oid,
-			     int oid_len)
-{
-	struct ceph_msg *msg = req->r_request;
-	struct ceph_osd_request_head *head;
-	struct ceph_osd_req_op *src_op;
-	struct ceph_osd_op *op;
-	void *p;
-	int num_op = get_num_ops(src_ops, NULL);
-	size_t msg_size = sizeof(*head) + num_op*sizeof(*op);
-	int flags = req->r_flags;
-	u64 data_len = 0;
-	int i;
-
-	head = msg->front.iov_base;
-	op = (void *)(head + 1);
-	p = (void *)(op + num_op);
-
-	req->r_snapc = ceph_get_snap_context(snapc);
-
-	head->client_inc = cpu_to_le32(1); /* always, for now. */
-	head->flags = cpu_to_le32(flags);
-	if (flags & CEPH_OSD_FLAG_WRITE)
-		ceph_encode_timespec(&head->mtime, mtime);
-	head->num_ops = cpu_to_le16(num_op);
-
-
-	/* fill in oid */
-	head->object_len = cpu_to_le32(oid_len);
-	memcpy(p, oid, oid_len);
-	p += oid_len;
-
-	src_op = src_ops;
-	while (src_op->op) {
-		osd_req_encode_op(req, op, src_op);
-		src_op++;
-		op++;
-	}
-
-	if (req->r_trail)
-		data_len += req->r_trail->length;
-
-	if (snapc) {
-		head->snap_seq = cpu_to_le64(snapc->seq);
-		head->num_snaps = cpu_to_le32(snapc->num_snaps);
-		for (i = 0; i < snapc->num_snaps; i++) {
-			put_unaligned_le64(snapc->snaps[i], p);
-			p += sizeof(u64);
-		}
-	}
-
-	if (flags & CEPH_OSD_FLAG_WRITE) {
-		req->r_request->hdr.data_off = cpu_to_le16(off);
-		req->r_request->hdr.data_len = cpu_to_le32(*plen + data_len);
-	} else if (data_len) {
-		req->r_request->hdr.data_off = 0;
-		req->r_request->hdr.data_len = cpu_to_le32(data_len);
-	}
-
-	BUG_ON(p > msg->front.iov_base + msg->front.iov_len);
-	msg_size = p - msg->front.iov_base;
-	msg->front.iov_len = msg_size;
-	msg->hdr.front_len = cpu_to_le32(msg_size);
-	return;
-}
-
-/*
- * build new request AND message, calculate layout, and adjust file
- * extent as needed.
- *
- * if the file was recently truncated, we include information about its
- * old and new size so that the object can be updated appropriately.  (we
- * avoid synchronously deleting truncated objects because it's slow.)
- *
- * if @do_sync, include a 'startsync' command so that the osd will flush
- * data quickly.
- */
-struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *osdc,
-					       struct ceph_file_layout *layout,
-					       struct ceph_vino vino,
-					       u64 off, u64 *plen,
-					       int opcode, int flags,
-					       struct ceph_snap_context *snapc,
-					       int do_sync,
-					       u32 truncate_seq,
-					       u64 truncate_size,
-					       struct timespec *mtime,
-					       bool use_mempool, int num_reply)
-{
-	struct ceph_osd_req_op ops[3];
-	struct ceph_osd_request *req;
-
-	ops[0].op = opcode;
-	ops[0].extent.truncate_seq = truncate_seq;
-	ops[0].extent.truncate_size = truncate_size;
-	ops[0].payload_len = 0;
-
-	if (do_sync) {
-		ops[1].op = CEPH_OSD_OP_STARTSYNC;
-		ops[1].payload_len = 0;
-		ops[2].op = 0;
-	} else
-		ops[1].op = 0;
-
-	req = ceph_osdc_alloc_request(osdc, flags,
-					 snapc, ops,
-					 use_mempool,
-					 GFP_NOFS, NULL, NULL);
-	if (IS_ERR(req))
-		return req;
-
-	/* calculate max write size */
-	calc_layout(osdc, vino, layout, off, plen, req, ops);
-	req->r_file_layout = *layout;  /* keep a copy */
-
-	ceph_osdc_build_request(req, off, plen, ops,
-				snapc,
-				mtime,
-				req->r_oid, req->r_oid_len);
-
-	return req;
-}
-
-/*
- * We keep osd requests in an rbtree, sorted by ->r_tid.
- */
-static void __insert_request(struct ceph_osd_client *osdc,
-			     struct ceph_osd_request *new)
-{
-	struct rb_node **p = &osdc->requests.rb_node;
-	struct rb_node *parent = NULL;
-	struct ceph_osd_request *req = NULL;
-
-	while (*p) {
-		parent = *p;
-		req = rb_entry(parent, struct ceph_osd_request, r_node);
-		if (new->r_tid < req->r_tid)
-			p = &(*p)->rb_left;
-		else if (new->r_tid > req->r_tid)
-			p = &(*p)->rb_right;
-		else
-			BUG();
-	}
-
-	rb_link_node(&new->r_node, parent, p);
-	rb_insert_color(&new->r_node, &osdc->requests);
-}
-
-static struct ceph_osd_request *__lookup_request(struct ceph_osd_client *osdc,
-						 u64 tid)
-{
-	struct ceph_osd_request *req;
-	struct rb_node *n = osdc->requests.rb_node;
-
-	while (n) {
-		req = rb_entry(n, struct ceph_osd_request, r_node);
-		if (tid < req->r_tid)
-			n = n->rb_left;
-		else if (tid > req->r_tid)
-			n = n->rb_right;
-		else
-			return req;
-	}
-	return NULL;
-}
-
-static struct ceph_osd_request *
-__lookup_request_ge(struct ceph_osd_client *osdc,
-		    u64 tid)
-{
-	struct ceph_osd_request *req;
-	struct rb_node *n = osdc->requests.rb_node;
-
-	while (n) {
-		req = rb_entry(n, struct ceph_osd_request, r_node);
-		if (tid < req->r_tid) {
-			if (!n->rb_left)
-				return req;
-			n = n->rb_left;
-		} else if (tid > req->r_tid) {
-			n = n->rb_right;
-		} else {
-			return req;
-		}
-	}
-	return NULL;
-}
-
-
-/*
- * If the osd connection drops, we need to resubmit all requests.
- */
-static void osd_reset(struct ceph_connection *con)
-{
-	struct ceph_osd *osd = con->private;
-	struct ceph_osd_client *osdc;
-
-	if (!osd)
-		return;
-	dout("osd_reset osd%d\n", osd->o_osd);
-	osdc = osd->o_osdc;
-	down_read(&osdc->map_sem);
-	kick_requests(osdc, osd);
-	up_read(&osdc->map_sem);
-}
-
-/*
- * Track open sessions with osds.
- */
-static struct ceph_osd *create_osd(struct ceph_osd_client *osdc)
-{
-	struct ceph_osd *osd;
-
-	osd = kzalloc(sizeof(*osd), GFP_NOFS);
-	if (!osd)
-		return NULL;
-
-	atomic_set(&osd->o_ref, 1);
-	osd->o_osdc = osdc;
-	INIT_LIST_HEAD(&osd->o_requests);
-	INIT_LIST_HEAD(&osd->o_osd_lru);
-	osd->o_incarnation = 1;
-
-	ceph_con_init(osdc->client->msgr, &osd->o_con);
-	osd->o_con.private = osd;
-	osd->o_con.ops = &osd_con_ops;
-	osd->o_con.peer_name.type = CEPH_ENTITY_TYPE_OSD;
-
-	INIT_LIST_HEAD(&osd->o_keepalive_item);
-	return osd;
-}
-
-static struct ceph_osd *get_osd(struct ceph_osd *osd)
-{
-	if (atomic_inc_not_zero(&osd->o_ref)) {
-		dout("get_osd %p %d -> %d\n", osd, atomic_read(&osd->o_ref)-1,
-		     atomic_read(&osd->o_ref));
-		return osd;
-	} else {
-		dout("get_osd %p FAIL\n", osd);
-		return NULL;
-	}
-}
-
-static void put_osd(struct ceph_osd *osd)
-{
-	dout("put_osd %p %d -> %d\n", osd, atomic_read(&osd->o_ref),
-	     atomic_read(&osd->o_ref) - 1);
-	if (atomic_dec_and_test(&osd->o_ref)) {
-		struct ceph_auth_client *ac = osd->o_osdc->client->monc.auth;
-
-		if (osd->o_authorizer)
-			ac->ops->destroy_authorizer(ac, osd->o_authorizer);
-		kfree(osd);
-	}
-}
-
-/*
- * remove an osd from our map
- */
-static void __remove_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd)
-{
-	dout("__remove_osd %p\n", osd);
-	BUG_ON(!list_empty(&osd->o_requests));
-	rb_erase(&osd->o_node, &osdc->osds);
-	list_del_init(&osd->o_osd_lru);
-	ceph_con_close(&osd->o_con);
-	put_osd(osd);
-}
-
-static void __move_osd_to_lru(struct ceph_osd_client *osdc,
-			      struct ceph_osd *osd)
-{
-	dout("__move_osd_to_lru %p\n", osd);
-	BUG_ON(!list_empty(&osd->o_osd_lru));
-	list_add_tail(&osd->o_osd_lru, &osdc->osd_lru);
-	osd->lru_ttl = jiffies + osdc->client->mount_args->osd_idle_ttl * HZ;
-}
-
-static void __remove_osd_from_lru(struct ceph_osd *osd)
-{
-	dout("__remove_osd_from_lru %p\n", osd);
-	if (!list_empty(&osd->o_osd_lru))
-		list_del_init(&osd->o_osd_lru);
-}
-
-static void remove_old_osds(struct ceph_osd_client *osdc, int remove_all)
-{
-	struct ceph_osd *osd, *nosd;
-
-	dout("__remove_old_osds %p\n", osdc);
-	mutex_lock(&osdc->request_mutex);
-	list_for_each_entry_safe(osd, nosd, &osdc->osd_lru, o_osd_lru) {
-		if (!remove_all && time_before(jiffies, osd->lru_ttl))
-			break;
-		__remove_osd(osdc, osd);
-	}
-	mutex_unlock(&osdc->request_mutex);
-}
-
-/*
- * reset osd connect
- */
-static int __reset_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd)
-{
-	struct ceph_osd_request *req;
-	int ret = 0;
-
-	dout("__reset_osd %p osd%d\n", osd, osd->o_osd);
-	if (list_empty(&osd->o_requests)) {
-		__remove_osd(osdc, osd);
-	} else if (memcmp(&osdc->osdmap->osd_addr[osd->o_osd],
-			  &osd->o_con.peer_addr,
-			  sizeof(osd->o_con.peer_addr)) == 0 &&
-		   !ceph_con_opened(&osd->o_con)) {
-		dout(" osd addr hasn't changed and connection never opened,"
-		     " letting msgr retry");
-		/* touch each r_stamp for handle_timeout()'s benfit */
-		list_for_each_entry(req, &osd->o_requests, r_osd_item)
-			req->r_stamp = jiffies;
-		ret = -EAGAIN;
-	} else {
-		ceph_con_close(&osd->o_con);
-		ceph_con_open(&osd->o_con, &osdc->osdmap->osd_addr[osd->o_osd]);
-		osd->o_incarnation++;
-	}
-	return ret;
-}
-
-static void __insert_osd(struct ceph_osd_client *osdc, struct ceph_osd *new)
-{
-	struct rb_node **p = &osdc->osds.rb_node;
-	struct rb_node *parent = NULL;
-	struct ceph_osd *osd = NULL;
-
-	while (*p) {
-		parent = *p;
-		osd = rb_entry(parent, struct ceph_osd, o_node);
-		if (new->o_osd < osd->o_osd)
-			p = &(*p)->rb_left;
-		else if (new->o_osd > osd->o_osd)
-			p = &(*p)->rb_right;
-		else
-			BUG();
-	}
-
-	rb_link_node(&new->o_node, parent, p);
-	rb_insert_color(&new->o_node, &osdc->osds);
-}
-
-static struct ceph_osd *__lookup_osd(struct ceph_osd_client *osdc, int o)
-{
-	struct ceph_osd *osd;
-	struct rb_node *n = osdc->osds.rb_node;
-
-	while (n) {
-		osd = rb_entry(n, struct ceph_osd, o_node);
-		if (o < osd->o_osd)
-			n = n->rb_left;
-		else if (o > osd->o_osd)
-			n = n->rb_right;
-		else
-			return osd;
-	}
-	return NULL;
-}
-
-static void __schedule_osd_timeout(struct ceph_osd_client *osdc)
-{
-	schedule_delayed_work(&osdc->timeout_work,
-			osdc->client->mount_args->osd_keepalive_timeout * HZ);
-}
-
-static void __cancel_osd_timeout(struct ceph_osd_client *osdc)
-{
-	cancel_delayed_work(&osdc->timeout_work);
-}
-
-/*
- * Register request, assign tid.  If this is the first request, set up
- * the timeout event.
- */
-static void register_request(struct ceph_osd_client *osdc,
-			     struct ceph_osd_request *req)
-{
-	mutex_lock(&osdc->request_mutex);
-	req->r_tid = ++osdc->last_tid;
-	req->r_request->hdr.tid = cpu_to_le64(req->r_tid);
-	INIT_LIST_HEAD(&req->r_req_lru_item);
-
-	dout("register_request %p tid %lld\n", req, req->r_tid);
-	__insert_request(osdc, req);
-	ceph_osdc_get_request(req);
-	osdc->num_requests++;
-
-	if (osdc->num_requests == 1) {
-		dout(" first request, scheduling timeout\n");
-		__schedule_osd_timeout(osdc);
-	}
-	mutex_unlock(&osdc->request_mutex);
-}
-
-/*
- * called under osdc->request_mutex
- */
-static void __unregister_request(struct ceph_osd_client *osdc,
-				 struct ceph_osd_request *req)
-{
-	dout("__unregister_request %p tid %lld\n", req, req->r_tid);
-	rb_erase(&req->r_node, &osdc->requests);
-	osdc->num_requests--;
-
-	if (req->r_osd) {
-		/* make sure the original request isn't in flight. */
-		ceph_con_revoke(&req->r_osd->o_con, req->r_request);
-
-		list_del_init(&req->r_osd_item);
-		if (list_empty(&req->r_osd->o_requests))
-			__move_osd_to_lru(osdc, req->r_osd);
-		req->r_osd = NULL;
-	}
-
-	ceph_osdc_put_request(req);
-
-	list_del_init(&req->r_req_lru_item);
-	if (osdc->num_requests == 0) {
-		dout(" no requests, canceling timeout\n");
-		__cancel_osd_timeout(osdc);
-	}
-}
-
-/*
- * Cancel a previously queued request message
- */
-static void __cancel_request(struct ceph_osd_request *req)
-{
-	if (req->r_sent && req->r_osd) {
-		ceph_con_revoke(&req->r_osd->o_con, req->r_request);
-		req->r_sent = 0;
-	}
-	list_del_init(&req->r_req_lru_item);
-}
-
-/*
- * Pick an osd (the first 'up' osd in the pg), allocate the osd struct
- * (as needed), and set the request r_osd appropriately.  If there is
- * no up osd, set r_osd to NULL.
- *
- * Return 0 if unchanged, 1 if changed, or negative on error.
- *
- * Caller should hold map_sem for read and request_mutex.
- */
-static int __map_osds(struct ceph_osd_client *osdc,
-		      struct ceph_osd_request *req)
-{
-	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
-	struct ceph_pg pgid;
-	int acting[CEPH_PG_MAX_SIZE];
-	int o = -1, num = 0;
-	int err;
-
-	dout("map_osds %p tid %lld\n", req, req->r_tid);
-	err = ceph_calc_object_layout(&reqhead->layout, req->r_oid,
-				      &req->r_file_layout, osdc->osdmap);
-	if (err)
-		return err;
-	pgid = reqhead->layout.ol_pgid;
-	req->r_pgid = pgid;
-
-	err = ceph_calc_pg_acting(osdc->osdmap, pgid, acting);
-	if (err > 0) {
-		o = acting[0];
-		num = err;
-	}
-
-	if ((req->r_osd && req->r_osd->o_osd == o &&
-	     req->r_sent >= req->r_osd->o_incarnation &&
-	     req->r_num_pg_osds == num &&
-	     memcmp(req->r_pg_osds, acting, sizeof(acting[0])*num) == 0) ||
-	    (req->r_osd == NULL && o == -1))
-		return 0;  /* no change */
-
-	dout("map_osds tid %llu pgid %d.%x osd%d (was osd%d)\n",
-	     req->r_tid, le32_to_cpu(pgid.pool), le16_to_cpu(pgid.ps), o,
-	     req->r_osd ? req->r_osd->o_osd : -1);
-
-	/* record full pg acting set */
-	memcpy(req->r_pg_osds, acting, sizeof(acting[0]) * num);
-	req->r_num_pg_osds = num;
-
-	if (req->r_osd) {
-		__cancel_request(req);
-		list_del_init(&req->r_osd_item);
-		req->r_osd = NULL;
-	}
-
-	req->r_osd = __lookup_osd(osdc, o);
-	if (!req->r_osd && o >= 0) {
-		err = -ENOMEM;
-		req->r_osd = create_osd(osdc);
-		if (!req->r_osd)
-			goto out;
-
-		dout("map_osds osd %p is osd%d\n", req->r_osd, o);
-		req->r_osd->o_osd = o;
-		req->r_osd->o_con.peer_name.num = cpu_to_le64(o);
-		__insert_osd(osdc, req->r_osd);
-
-		ceph_con_open(&req->r_osd->o_con, &osdc->osdmap->osd_addr[o]);
-	}
-
-	if (req->r_osd) {
-		__remove_osd_from_lru(req->r_osd);
-		list_add(&req->r_osd_item, &req->r_osd->o_requests);
-	}
-	err = 1;   /* osd or pg changed */
-
-out:
-	return err;
-}
-
-/*
- * caller should hold map_sem (for read) and request_mutex
- */
-static int __send_request(struct ceph_osd_client *osdc,
-			  struct ceph_osd_request *req)
-{
-	struct ceph_osd_request_head *reqhead;
-	int err;
-
-	err = __map_osds(osdc, req);
-	if (err < 0)
-		return err;
-	if (req->r_osd == NULL) {
-		dout("send_request %p no up osds in pg\n", req);
-		ceph_monc_request_next_osdmap(&osdc->client->monc);
-		return 0;
-	}
-
-	dout("send_request %p tid %llu to osd%d flags %d\n",
-	     req, req->r_tid, req->r_osd->o_osd, req->r_flags);
-
-	reqhead = req->r_request->front.iov_base;
-	reqhead->osdmap_epoch = cpu_to_le32(osdc->osdmap->epoch);
-	reqhead->flags |= cpu_to_le32(req->r_flags);  /* e.g., RETRY */
-	reqhead->reassert_version = req->r_reassert_version;
-
-	req->r_stamp = jiffies;
-	list_move_tail(&req->r_req_lru_item, &osdc->req_lru);
-
-	ceph_msg_get(req->r_request); /* send consumes a ref */
-	ceph_con_send(&req->r_osd->o_con, req->r_request);
-	req->r_sent = req->r_osd->o_incarnation;
-	return 0;
-}
-
-/*
- * Timeout callback, called every N seconds when 1 or more osd
- * requests has been active for more than N seconds.  When this
- * happens, we ping all OSDs with requests who have timed out to
- * ensure any communications channel reset is detected.  Reset the
- * request timeouts another N seconds in the future as we go.
- * Reschedule the timeout event another N seconds in future (unless
- * there are no open requests).
- */
-static void handle_timeout(struct work_struct *work)
-{
-	struct ceph_osd_client *osdc =
-		container_of(work, struct ceph_osd_client, timeout_work.work);
-	struct ceph_osd_request *req, *last_req = NULL;
-	struct ceph_osd *osd;
-	unsigned long timeout = osdc->client->mount_args->osd_timeout * HZ;
-	unsigned long keepalive =
-		osdc->client->mount_args->osd_keepalive_timeout * HZ;
-	unsigned long last_stamp = 0;
-	struct rb_node *p;
-	struct list_head slow_osds;
-
-	dout("timeout\n");
-	down_read(&osdc->map_sem);
-
-	ceph_monc_request_next_osdmap(&osdc->client->monc);
-
-	mutex_lock(&osdc->request_mutex);
-	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
-		req = rb_entry(p, struct ceph_osd_request, r_node);
-
-		if (req->r_resend) {
-			int err;
-
-			dout("osdc resending prev failed %lld\n", req->r_tid);
-			err = __send_request(osdc, req);
-			if (err)
-				dout("osdc failed again on %lld\n", req->r_tid);
-			else
-				req->r_resend = false;
-			continue;
-		}
-	}
-
-	/*
-	 * reset osds that appear to be _really_ unresponsive.  this
-	 * is a failsafe measure.. we really shouldn't be getting to
-	 * this point if the system is working properly.  the monitors
-	 * should mark the osd as failed and we should find out about
-	 * it from an updated osd map.
-	 */
-	while (timeout && !list_empty(&osdc->req_lru)) {
-		req = list_entry(osdc->req_lru.next, struct ceph_osd_request,
-				 r_req_lru_item);
-
-		if (time_before(jiffies, req->r_stamp + timeout))
-			break;
-
-		BUG_ON(req == last_req && req->r_stamp == last_stamp);
-		last_req = req;
-		last_stamp = req->r_stamp;
-
-		osd = req->r_osd;
-		BUG_ON(!osd);
-		pr_warning(" tid %llu timed out on osd%d, will reset osd\n",
-			   req->r_tid, osd->o_osd);
-		__kick_requests(osdc, osd);
-	}
-
-	/*
-	 * ping osds that are a bit slow.  this ensures that if there
-	 * is a break in the TCP connection we will notice, and reopen
-	 * a connection with that osd (from the fault callback).
-	 */
-	INIT_LIST_HEAD(&slow_osds);
-	list_for_each_entry(req, &osdc->req_lru, r_req_lru_item) {
-		if (time_before(jiffies, req->r_stamp + keepalive))
-			break;
-
-		osd = req->r_osd;
-		BUG_ON(!osd);
-		dout(" tid %llu is slow, will send keepalive on osd%d\n",
-		     req->r_tid, osd->o_osd);
-		list_move_tail(&osd->o_keepalive_item, &slow_osds);
-	}
-	while (!list_empty(&slow_osds)) {
-		osd = list_entry(slow_osds.next, struct ceph_osd,
-				 o_keepalive_item);
-		list_del_init(&osd->o_keepalive_item);
-		ceph_con_keepalive(&osd->o_con);
-	}
-
-	__schedule_osd_timeout(osdc);
-	mutex_unlock(&osdc->request_mutex);
-
-	up_read(&osdc->map_sem);
-}
-
-static void handle_osds_timeout(struct work_struct *work)
-{
-	struct ceph_osd_client *osdc =
-		container_of(work, struct ceph_osd_client,
-			     osds_timeout_work.work);
-	unsigned long delay =
-		osdc->client->mount_args->osd_idle_ttl * HZ >> 2;
-
-	dout("osds timeout\n");
-	down_read(&osdc->map_sem);
-	remove_old_osds(osdc, 0);
-	up_read(&osdc->map_sem);
-
-	schedule_delayed_work(&osdc->osds_timeout_work,
-			      round_jiffies_relative(delay));
-}
-
-/*
- * handle osd op reply.  either call the callback if it is specified,
- * or do the completion to wake up the waiting thread.
- */
-static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg,
-			 struct ceph_connection *con)
-{
-	struct ceph_osd_reply_head *rhead = msg->front.iov_base;
-	struct ceph_osd_request *req;
-	u64 tid;
-	int numops, object_len, flags;
-	s32 result;
-
-	tid = le64_to_cpu(msg->hdr.tid);
-	if (msg->front.iov_len < sizeof(*rhead))
-		goto bad;
-	numops = le32_to_cpu(rhead->num_ops);
-	object_len = le32_to_cpu(rhead->object_len);
-	result = le32_to_cpu(rhead->result);
-	if (msg->front.iov_len != sizeof(*rhead) + object_len +
-	    numops * sizeof(struct ceph_osd_op))
-		goto bad;
-	dout("handle_reply %p tid %llu result %d\n", msg, tid, (int)result);
-
-	/* lookup */
-	mutex_lock(&osdc->request_mutex);
-	req = __lookup_request(osdc, tid);
-	if (req == NULL) {
-		dout("handle_reply tid %llu dne\n", tid);
-		mutex_unlock(&osdc->request_mutex);
-		return;
-	}
-	ceph_osdc_get_request(req);
-	flags = le32_to_cpu(rhead->flags);
-
-	/*
-	 * if this connection filled our message, drop our reference now, to
-	 * avoid a (safe but slower) revoke later.
-	 */
-	if (req->r_con_filling_msg == con && req->r_reply == msg) {
-		dout(" dropping con_filling_msg ref %p\n", con);
-		req->r_con_filling_msg = NULL;
-		ceph_con_put(con);
-	}
-
-	if (!req->r_got_reply) {
-		unsigned bytes;
-
-		req->r_result = le32_to_cpu(rhead->result);
-		bytes = le32_to_cpu(msg->hdr.data_len);
-		dout("handle_reply result %d bytes %d\n", req->r_result,
-		     bytes);
-		if (req->r_result == 0)
-			req->r_result = bytes;
-
-		/* in case this is a write and we need to replay, */
-		req->r_reassert_version = rhead->reassert_version;
-
-		req->r_got_reply = 1;
-	} else if ((flags & CEPH_OSD_FLAG_ONDISK) == 0) {
-		dout("handle_reply tid %llu dup ack\n", tid);
-		mutex_unlock(&osdc->request_mutex);
-		goto done;
-	}
-
-	dout("handle_reply tid %llu flags %d\n", tid, flags);
-
-	/* either this is a read, or we got the safe response */
-	if (result < 0 ||
-	    (flags & CEPH_OSD_FLAG_ONDISK) ||
-	    ((flags & CEPH_OSD_FLAG_WRITE) == 0))
-		__unregister_request(osdc, req);
-
-	mutex_unlock(&osdc->request_mutex);
-
-	if (req->r_callback)
-		req->r_callback(req, msg);
-	else
-		complete_all(&req->r_completion);
-
-	if (flags & CEPH_OSD_FLAG_ONDISK) {
-		if (req->r_safe_callback)
-			req->r_safe_callback(req, msg);
-		complete_all(&req->r_safe_completion);  /* fsync waiter */
-	}
-
-done:
-	ceph_osdc_put_request(req);
-	return;
-
-bad:
-	pr_err("corrupt osd_op_reply got %d %d expected %d\n",
-	       (int)msg->front.iov_len, le32_to_cpu(msg->hdr.front_len),
-	       (int)sizeof(*rhead));
-	ceph_msg_dump(msg);
-}
-
-
-static int __kick_requests(struct ceph_osd_client *osdc,
-			  struct ceph_osd *kickosd)
-{
-	struct ceph_osd_request *req;
-	struct rb_node *p, *n;
-	int needmap = 0;
-	int err;
-
-	dout("kick_requests osd%d\n", kickosd ? kickosd->o_osd : -1);
-	if (kickosd) {
-		err = __reset_osd(osdc, kickosd);
-		if (err == -EAGAIN)
-			return 1;
-	} else {
-		for (p = rb_first(&osdc->osds); p; p = n) {
-			struct ceph_osd *osd =
-				rb_entry(p, struct ceph_osd, o_node);
-
-			n = rb_next(p);
-			if (!ceph_osd_is_up(osdc->osdmap, osd->o_osd) ||
-			    memcmp(&osd->o_con.peer_addr,
-				   ceph_osd_addr(osdc->osdmap,
-						 osd->o_osd),
-				   sizeof(struct ceph_entity_addr)) != 0)
-				__reset_osd(osdc, osd);
-		}
-	}
-
-	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
-		req = rb_entry(p, struct ceph_osd_request, r_node);
-
-		if (req->r_resend) {
-			dout(" r_resend set on tid %llu\n", req->r_tid);
-			__cancel_request(req);
-			goto kick;
-		}
-		if (req->r_osd && kickosd == req->r_osd) {
-			__cancel_request(req);
-			goto kick;
-		}
-
-		err = __map_osds(osdc, req);
-		if (err == 0)
-			continue;  /* no change */
-		if (err < 0) {
-			/*
-			 * FIXME: really, we should set the request
-			 * error and fail if this isn't a 'nofail'
-			 * request, but that's a fair bit more
-			 * complicated to do.  So retry!
-			 */
-			dout(" setting r_resend on %llu\n", req->r_tid);
-			req->r_resend = true;
-			continue;
-		}
-		if (req->r_osd == NULL) {
-			dout("tid %llu maps to no valid osd\n", req->r_tid);
-			needmap++;  /* request a newer map */
-			continue;
-		}
-
-kick:
-		dout("kicking %p tid %llu osd%d\n", req, req->r_tid,
-		     req->r_osd ? req->r_osd->o_osd : -1);
-		req->r_flags |= CEPH_OSD_FLAG_RETRY;
-		err = __send_request(osdc, req);
-		if (err) {
-			dout(" setting r_resend on %llu\n", req->r_tid);
-			req->r_resend = true;
-		}
-	}
-
-	return needmap;
-}
-
-/*
- * Resubmit osd requests whose osd or osd address has changed.  Request
- * a new osd map if osds are down, or we are otherwise unable to determine
- * how to direct a request.
- *
- * Close connections to down osds.
- *
- * If @who is specified, resubmit requests for that specific osd.
- *
- * Caller should hold map_sem for read and request_mutex.
- */
-static void kick_requests(struct ceph_osd_client *osdc,
-			  struct ceph_osd *kickosd)
-{
-	int needmap;
-
-	mutex_lock(&osdc->request_mutex);
-	needmap = __kick_requests(osdc, kickosd);
-	mutex_unlock(&osdc->request_mutex);
-
-	if (needmap) {
-		dout("%d requests for down osds, need new map\n", needmap);
-		ceph_monc_request_next_osdmap(&osdc->client->monc);
-	}
-
-}
-/*
- * Process updated osd map.
- *
- * The message contains any number of incremental and full maps, normally
- * indicating some sort of topology change in the cluster.  Kick requests
- * off to different OSDs as needed.
- */
-void ceph_osdc_handle_map(struct ceph_osd_client *osdc, struct ceph_msg *msg)
-{
-	void *p, *end, *next;
-	u32 nr_maps, maplen;
-	u32 epoch;
-	struct ceph_osdmap *newmap = NULL, *oldmap;
-	int err;
-	struct ceph_fsid fsid;
-
-	dout("handle_map have %u\n", osdc->osdmap ? osdc->osdmap->epoch : 0);
-	p = msg->front.iov_base;
-	end = p + msg->front.iov_len;
-
-	/* verify fsid */
-	ceph_decode_need(&p, end, sizeof(fsid), bad);
-	ceph_decode_copy(&p, &fsid, sizeof(fsid));
-	if (ceph_check_fsid(osdc->client, &fsid) < 0)
-		return;
-
-	down_write(&osdc->map_sem);
-
-	/* incremental maps */
-	ceph_decode_32_safe(&p, end, nr_maps, bad);
-	dout(" %d inc maps\n", nr_maps);
-	while (nr_maps > 0) {
-		ceph_decode_need(&p, end, 2*sizeof(u32), bad);
-		epoch = ceph_decode_32(&p);
-		maplen = ceph_decode_32(&p);
-		ceph_decode_need(&p, end, maplen, bad);
-		next = p + maplen;
-		if (osdc->osdmap && osdc->osdmap->epoch+1 == epoch) {
-			dout("applying incremental map %u len %d\n",
-			     epoch, maplen);
-			newmap = osdmap_apply_incremental(&p, next,
-							  osdc->osdmap,
-							  osdc->client->msgr);
-			if (IS_ERR(newmap)) {
-				err = PTR_ERR(newmap);
-				goto bad;
-			}
-			BUG_ON(!newmap);
-			if (newmap != osdc->osdmap) {
-				ceph_osdmap_destroy(osdc->osdmap);
-				osdc->osdmap = newmap;
-			}
-		} else {
-			dout("ignoring incremental map %u len %d\n",
-			     epoch, maplen);
-		}
-		p = next;
-		nr_maps--;
-	}
-	if (newmap)
-		goto done;
-
-	/* full maps */
-	ceph_decode_32_safe(&p, end, nr_maps, bad);
-	dout(" %d full maps\n", nr_maps);
-	while (nr_maps) {
-		ceph_decode_need(&p, end, 2*sizeof(u32), bad);
-		epoch = ceph_decode_32(&p);
-		maplen = ceph_decode_32(&p);
-		ceph_decode_need(&p, end, maplen, bad);
-		if (nr_maps > 1) {
-			dout("skipping non-latest full map %u len %d\n",
-			     epoch, maplen);
-		} else if (osdc->osdmap && osdc->osdmap->epoch >= epoch) {
-			dout("skipping full map %u len %d, "
-			     "older than our %u\n", epoch, maplen,
-			     osdc->osdmap->epoch);
-		} else {
-			dout("taking full map %u len %d\n", epoch, maplen);
-			newmap = osdmap_decode(&p, p+maplen);
-			if (IS_ERR(newmap)) {
-				err = PTR_ERR(newmap);
-				goto bad;
-			}
-			BUG_ON(!newmap);
-			oldmap = osdc->osdmap;
-			osdc->osdmap = newmap;
-			if (oldmap)
-				ceph_osdmap_destroy(oldmap);
-		}
-		p += maplen;
-		nr_maps--;
-	}
-
-done:
-	downgrade_write(&osdc->map_sem);
-	ceph_monc_got_osdmap(&osdc->client->monc, osdc->osdmap->epoch);
-	if (newmap)
-		kick_requests(osdc, NULL);
-	up_read(&osdc->map_sem);
-	wake_up_all(&osdc->client->auth_wq);
-	return;
-
-bad:
-	pr_err("osdc handle_map corrupt msg\n");
-	ceph_msg_dump(msg);
-	up_write(&osdc->map_sem);
-	return;
-}
-
-/*
- * Register request, send initial attempt.
- */
-int ceph_osdc_start_request(struct ceph_osd_client *osdc,
-			    struct ceph_osd_request *req,
-			    bool nofail)
-{
-	int rc = 0;
-
-	req->r_request->pages = req->r_pages;
-	req->r_request->nr_pages = req->r_num_pages;
-#ifdef CONFIG_BLOCK
-	req->r_request->bio = req->r_bio;
-#endif
-	req->r_request->trail = req->r_trail;
-
-	register_request(osdc, req);
-
-	down_read(&osdc->map_sem);
-	mutex_lock(&osdc->request_mutex);
-	/*
-	 * a racing kick_requests() may have sent the message for us
-	 * while we dropped request_mutex above, so only send now if
-	 * the request still han't been touched yet.
-	 */
-	if (req->r_sent == 0) {
-		rc = __send_request(osdc, req);
-		if (rc) {
-			if (nofail) {
-				dout("osdc_start_request failed send, "
-				     " marking %lld\n", req->r_tid);
-				req->r_resend = true;
-				rc = 0;
-			} else {
-				__unregister_request(osdc, req);
-			}
-		}
-	}
-	mutex_unlock(&osdc->request_mutex);
-	up_read(&osdc->map_sem);
-	return rc;
-}
-
-/*
- * wait for a request to complete
- */
-int ceph_osdc_wait_request(struct ceph_osd_client *osdc,
-			   struct ceph_osd_request *req)
-{
-	int rc;
-
-	rc = wait_for_completion_interruptible(&req->r_completion);
-	if (rc < 0) {
-		mutex_lock(&osdc->request_mutex);
-		__cancel_request(req);
-		__unregister_request(osdc, req);
-		mutex_unlock(&osdc->request_mutex);
-		dout("wait_request tid %llu canceled/timed out\n", req->r_tid);
-		return rc;
-	}
-
-	dout("wait_request tid %llu result %d\n", req->r_tid, req->r_result);
-	return req->r_result;
-}
-
-/*
- * sync - wait for all in-flight requests to flush.  avoid starvation.
- */
-void ceph_osdc_sync(struct ceph_osd_client *osdc)
-{
-	struct ceph_osd_request *req;
-	u64 last_tid, next_tid = 0;
-
-	mutex_lock(&osdc->request_mutex);
-	last_tid = osdc->last_tid;
-	while (1) {
-		req = __lookup_request_ge(osdc, next_tid);
-		if (!req)
-			break;
-		if (req->r_tid > last_tid)
-			break;
-
-		next_tid = req->r_tid + 1;
-		if ((req->r_flags & CEPH_OSD_FLAG_WRITE) == 0)
-			continue;
-
-		ceph_osdc_get_request(req);
-		mutex_unlock(&osdc->request_mutex);
-		dout("sync waiting on tid %llu (last is %llu)\n",
-		     req->r_tid, last_tid);
-		wait_for_completion(&req->r_safe_completion);
-		mutex_lock(&osdc->request_mutex);
-		ceph_osdc_put_request(req);
-	}
-	mutex_unlock(&osdc->request_mutex);
-	dout("sync done (thru tid %llu)\n", last_tid);
-}
-
-/*
- * init, shutdown
- */
-int ceph_osdc_init(struct ceph_osd_client *osdc, struct ceph_client *client)
-{
-	int err;
-
-	dout("init\n");
-	osdc->client = client;
-	osdc->osdmap = NULL;
-	init_rwsem(&osdc->map_sem);
-	init_completion(&osdc->map_waiters);
-	osdc->last_requested_map = 0;
-	mutex_init(&osdc->request_mutex);
-	osdc->last_tid = 0;
-	osdc->osds = RB_ROOT;
-	INIT_LIST_HEAD(&osdc->osd_lru);
-	osdc->requests = RB_ROOT;
-	INIT_LIST_HEAD(&osdc->req_lru);
-	osdc->num_requests = 0;
-	INIT_DELAYED_WORK(&osdc->timeout_work, handle_timeout);
-	INIT_DELAYED_WORK(&osdc->osds_timeout_work, handle_osds_timeout);
-
-	schedule_delayed_work(&osdc->osds_timeout_work,
-	   round_jiffies_relative(osdc->client->mount_args->osd_idle_ttl * HZ));
-
-	err = -ENOMEM;
-	osdc->req_mempool = mempool_create_kmalloc_pool(10,
-					sizeof(struct ceph_osd_request));
-	if (!osdc->req_mempool)
-		goto out;
-
-	err = ceph_msgpool_init(&osdc->msgpool_op, OSD_OP_FRONT_LEN, 10, true,
-				"osd_op");
-	if (err < 0)
-		goto out_mempool;
-	err = ceph_msgpool_init(&osdc->msgpool_op_reply,
-				OSD_OPREPLY_FRONT_LEN, 10, true,
-				"osd_op_reply");
-	if (err < 0)
-		goto out_msgpool;
-	return 0;
-
-out_msgpool:
-	ceph_msgpool_destroy(&osdc->msgpool_op);
-out_mempool:
-	mempool_destroy(osdc->req_mempool);
-out:
-	return err;
-}
-
-void ceph_osdc_stop(struct ceph_osd_client *osdc)
-{
-	cancel_delayed_work_sync(&osdc->timeout_work);
-	cancel_delayed_work_sync(&osdc->osds_timeout_work);
-	if (osdc->osdmap) {
-		ceph_osdmap_destroy(osdc->osdmap);
-		osdc->osdmap = NULL;
-	}
-	remove_old_osds(osdc, 1);
-	mempool_destroy(osdc->req_mempool);
-	ceph_msgpool_destroy(&osdc->msgpool_op);
-	ceph_msgpool_destroy(&osdc->msgpool_op_reply);
-}
-
-/*
- * Read some contiguous pages.  If we cross a stripe boundary, shorten
- * *plen.  Return number of bytes read, or error.
- */
-int ceph_osdc_readpages(struct ceph_osd_client *osdc,
-			struct ceph_vino vino, struct ceph_file_layout *layout,
-			u64 off, u64 *plen,
-			u32 truncate_seq, u64 truncate_size,
-			struct page **pages, int num_pages)
-{
-	struct ceph_osd_request *req;
-	int rc = 0;
-
-	dout("readpages on ino %llx.%llx on %llu~%llu\n", vino.ino,
-	     vino.snap, off, *plen);
-	req = ceph_osdc_new_request(osdc, layout, vino, off, plen,
-				    CEPH_OSD_OP_READ, CEPH_OSD_FLAG_READ,
-				    NULL, 0, truncate_seq, truncate_size, NULL,
-				    false, 1);
-	if (!req)
-		return -ENOMEM;
-
-	/* it may be a short read due to an object boundary */
-	req->r_pages = pages;
-
-	dout("readpages  final extent is %llu~%llu (%d pages)\n",
-	     off, *plen, req->r_num_pages);
-
-	rc = ceph_osdc_start_request(osdc, req, false);
-	if (!rc)
-		rc = ceph_osdc_wait_request(osdc, req);
-
-	ceph_osdc_put_request(req);
-	dout("readpages result %d\n", rc);
-	return rc;
-}
-
-/*
- * do a synchronous write on N pages
- */
-int ceph_osdc_writepages(struct ceph_osd_client *osdc, struct ceph_vino vino,
-			 struct ceph_file_layout *layout,
-			 struct ceph_snap_context *snapc,
-			 u64 off, u64 len,
-			 u32 truncate_seq, u64 truncate_size,
-			 struct timespec *mtime,
-			 struct page **pages, int num_pages,
-			 int flags, int do_sync, bool nofail)
-{
-	struct ceph_osd_request *req;
-	int rc = 0;
-
-	BUG_ON(vino.snap != CEPH_NOSNAP);
-	req = ceph_osdc_new_request(osdc, layout, vino, off, &len,
-				    CEPH_OSD_OP_WRITE,
-				    flags | CEPH_OSD_FLAG_ONDISK |
-					    CEPH_OSD_FLAG_WRITE,
-				    snapc, do_sync,
-				    truncate_seq, truncate_size, mtime,
-				    nofail, 1);
-	if (!req)
-		return -ENOMEM;
-
-	/* it may be a short write due to an object boundary */
-	req->r_pages = pages;
-	dout("writepages %llu~%llu (%d pages)\n", off, len,
-	     req->r_num_pages);
-
-	rc = ceph_osdc_start_request(osdc, req, nofail);
-	if (!rc)
-		rc = ceph_osdc_wait_request(osdc, req);
-
-	ceph_osdc_put_request(req);
-	if (rc == 0)
-		rc = len;
-	dout("writepages result %d\n", rc);
-	return rc;
-}
-
-/*
- * handle incoming message
- */
-static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
-{
-	struct ceph_osd *osd = con->private;
-	struct ceph_osd_client *osdc;
-	int type = le16_to_cpu(msg->hdr.type);
-
-	if (!osd)
-		goto out;
-	osdc = osd->o_osdc;
-
-	switch (type) {
-	case CEPH_MSG_OSD_MAP:
-		ceph_osdc_handle_map(osdc, msg);
-		break;
-	case CEPH_MSG_OSD_OPREPLY:
-		handle_reply(osdc, msg, con);
-		break;
-
-	default:
-		pr_err("received unknown message type %d %s\n", type,
-		       ceph_msg_type_name(type));
-	}
-out:
-	ceph_msg_put(msg);
-}
-
-/*
- * lookup and return message for incoming reply.  set up reply message
- * pages.
- */
-static struct ceph_msg *get_reply(struct ceph_connection *con,
-				  struct ceph_msg_header *hdr,
-				  int *skip)
-{
-	struct ceph_osd *osd = con->private;
-	struct ceph_osd_client *osdc = osd->o_osdc;
-	struct ceph_msg *m;
-	struct ceph_osd_request *req;
-	int front = le32_to_cpu(hdr->front_len);
-	int data_len = le32_to_cpu(hdr->data_len);
-	u64 tid;
-
-	tid = le64_to_cpu(hdr->tid);
-	mutex_lock(&osdc->request_mutex);
-	req = __lookup_request(osdc, tid);
-	if (!req) {
-		*skip = 1;
-		m = NULL;
-		pr_info("get_reply unknown tid %llu from osd%d\n", tid,
-			osd->o_osd);
-		goto out;
-	}
-
-	if (req->r_con_filling_msg) {
-		dout("get_reply revoking msg %p from old con %p\n",
-		     req->r_reply, req->r_con_filling_msg);
-		ceph_con_revoke_message(req->r_con_filling_msg, req->r_reply);
-		ceph_con_put(req->r_con_filling_msg);
-		req->r_con_filling_msg = NULL;
-	}
-
-	if (front > req->r_reply->front.iov_len) {
-		pr_warning("get_reply front %d > preallocated %d\n",
-			   front, (int)req->r_reply->front.iov_len);
-		m = ceph_msg_new(CEPH_MSG_OSD_OPREPLY, front, GFP_NOFS);
-		if (!m)
-			goto out;
-		ceph_msg_put(req->r_reply);
-		req->r_reply = m;
-	}
-	m = ceph_msg_get(req->r_reply);
-
-	if (data_len > 0) {
-		unsigned data_off = le16_to_cpu(hdr->data_off);
-		int want = calc_pages_for(data_off & ~PAGE_MASK, data_len);
-
-		if (unlikely(req->r_num_pages < want)) {
-			pr_warning("tid %lld reply %d > expected %d pages\n",
-				   tid, want, m->nr_pages);
-			*skip = 1;
-			ceph_msg_put(m);
-			m = NULL;
-			goto out;
-		}
-		m->pages = req->r_pages;
-		m->nr_pages = req->r_num_pages;
-#ifdef CONFIG_BLOCK
-		m->bio = req->r_bio;
-#endif
-	}
-	*skip = 0;
-	req->r_con_filling_msg = ceph_con_get(con);
-	dout("get_reply tid %lld %p\n", tid, m);
-
-out:
-	mutex_unlock(&osdc->request_mutex);
-	return m;
-
-}
-
-static struct ceph_msg *alloc_msg(struct ceph_connection *con,
-				  struct ceph_msg_header *hdr,
-				  int *skip)
-{
-	struct ceph_osd *osd = con->private;
-	int type = le16_to_cpu(hdr->type);
-	int front = le32_to_cpu(hdr->front_len);
-
-	switch (type) {
-	case CEPH_MSG_OSD_MAP:
-		return ceph_msg_new(type, front, GFP_NOFS);
-	case CEPH_MSG_OSD_OPREPLY:
-		return get_reply(con, hdr, skip);
-	default:
-		pr_info("alloc_msg unexpected msg type %d from osd%d\n", type,
-			osd->o_osd);
-		*skip = 1;
-		return NULL;
-	}
-}
-
-/*
- * Wrappers to refcount containing ceph_osd struct
- */
-static struct ceph_connection *get_osd_con(struct ceph_connection *con)
-{
-	struct ceph_osd *osd = con->private;
-	if (get_osd(osd))
-		return con;
-	return NULL;
-}
-
-static void put_osd_con(struct ceph_connection *con)
-{
-	struct ceph_osd *osd = con->private;
-	put_osd(osd);
-}
-
-/*
- * authentication
- */
-static int get_authorizer(struct ceph_connection *con,
-			  void **buf, int *len, int *proto,
-			  void **reply_buf, int *reply_len, int force_new)
-{
-	struct ceph_osd *o = con->private;
-	struct ceph_osd_client *osdc = o->o_osdc;
-	struct ceph_auth_client *ac = osdc->client->monc.auth;
-	int ret = 0;
-
-	if (force_new && o->o_authorizer) {
-		ac->ops->destroy_authorizer(ac, o->o_authorizer);
-		o->o_authorizer = NULL;
-	}
-	if (o->o_authorizer == NULL) {
-		ret = ac->ops->create_authorizer(
-			ac, CEPH_ENTITY_TYPE_OSD,
-			&o->o_authorizer,
-			&o->o_authorizer_buf,
-			&o->o_authorizer_buf_len,
-			&o->o_authorizer_reply_buf,
-			&o->o_authorizer_reply_buf_len);
-		if (ret)
-			return ret;
-	}
-
-	*proto = ac->protocol;
-	*buf = o->o_authorizer_buf;
-	*len = o->o_authorizer_buf_len;
-	*reply_buf = o->o_authorizer_reply_buf;
-	*reply_len = o->o_authorizer_reply_buf_len;
-	return 0;
-}
-
-
-static int verify_authorizer_reply(struct ceph_connection *con, int len)
-{
-	struct ceph_osd *o = con->private;
-	struct ceph_osd_client *osdc = o->o_osdc;
-	struct ceph_auth_client *ac = osdc->client->monc.auth;
-
-	return ac->ops->verify_authorizer_reply(ac, o->o_authorizer, len);
-}
-
-static int invalidate_authorizer(struct ceph_connection *con)
-{
-	struct ceph_osd *o = con->private;
-	struct ceph_osd_client *osdc = o->o_osdc;
-	struct ceph_auth_client *ac = osdc->client->monc.auth;
-
-	if (ac->ops->invalidate_authorizer)
-		ac->ops->invalidate_authorizer(ac, CEPH_ENTITY_TYPE_OSD);
-
-	return ceph_monc_validate_auth(&osdc->client->monc);
-}
-
-static const struct ceph_connection_operations osd_con_ops = {
-	.get = get_osd_con,
-	.put = put_osd_con,
-	.dispatch = dispatch,
-	.get_authorizer = get_authorizer,
-	.verify_authorizer_reply = verify_authorizer_reply,
-	.invalidate_authorizer = invalidate_authorizer,
-	.alloc_msg = alloc_msg,
-	.fault = osd_reset,
-};
diff --git a/fs/ceph/osd_client.h b/fs/ceph/osd_client.h
deleted file mode 100644
index 65c9c560f1ac..000000000000
--- a/fs/ceph/osd_client.h
+++ /dev/null
@@ -1,233 +0,0 @@
-#ifndef _FS_CEPH_OSD_CLIENT_H
-#define _FS_CEPH_OSD_CLIENT_H
-
-#include <linux/completion.h>
-#include <linux/kref.h>
-#include <linux/mempool.h>
-#include <linux/rbtree.h>
-
-#include "types.h"
-#include "osdmap.h"
-#include "messenger.h"
-
-struct ceph_msg;
-struct ceph_snap_context;
-struct ceph_osd_request;
-struct ceph_osd_client;
-struct ceph_authorizer;
-struct ceph_pagelist;
-
-/*
- * completion callback for async writepages
- */
-typedef void (*ceph_osdc_callback_t)(struct ceph_osd_request *,
-				     struct ceph_msg *);
-
-/* a given osd we're communicating with */
-struct ceph_osd {
-	atomic_t o_ref;
-	struct ceph_osd_client *o_osdc;
-	int o_osd;
-	int o_incarnation;
-	struct rb_node o_node;
-	struct ceph_connection o_con;
-	struct list_head o_requests;
-	struct list_head o_osd_lru;
-	struct ceph_authorizer *o_authorizer;
-	void *o_authorizer_buf, *o_authorizer_reply_buf;
-	size_t o_authorizer_buf_len, o_authorizer_reply_buf_len;
-	unsigned long lru_ttl;
-	int o_marked_for_keepalive;
-	struct list_head o_keepalive_item;
-};
-
-/* an in-flight request */
-struct ceph_osd_request {
-	u64             r_tid;              /* unique for this client */
-	struct rb_node  r_node;
-	struct list_head r_req_lru_item;
-	struct list_head r_osd_item;
-	struct ceph_osd *r_osd;
-	struct ceph_pg   r_pgid;
-	int              r_pg_osds[CEPH_PG_MAX_SIZE];
-	int              r_num_pg_osds;
-
-	struct ceph_connection *r_con_filling_msg;
-
-	struct ceph_msg  *r_request, *r_reply;
-	int               r_result;
-	int               r_flags;     /* any additional flags for the osd */
-	u32               r_sent;      /* >0 if r_request is sending/sent */
-	int               r_got_reply;
-
-	struct ceph_osd_client *r_osdc;
-	struct kref       r_kref;
-	bool              r_mempool;
-	struct completion r_completion, r_safe_completion;
-	ceph_osdc_callback_t r_callback, r_safe_callback;
-	struct ceph_eversion r_reassert_version;
-	struct list_head  r_unsafe_item;
-
-	struct inode *r_inode;         	      /* for use by callbacks */
-
-	char              r_oid[40];          /* object name */
-	int               r_oid_len;
-	unsigned long     r_stamp;            /* send OR check time */
-	bool              r_resend;           /* msg send failed, needs retry */
-
-	struct ceph_file_layout r_file_layout;
-	struct ceph_snap_context *r_snapc;    /* snap context for writes */
-	unsigned          r_num_pages;        /* size of page array (follows) */
-	struct page     **r_pages;            /* pages for data payload */
-	int               r_pages_from_pool;
-	int               r_own_pages;        /* if true, i own page list */
-#ifdef CONFIG_BLOCK
-	struct bio       *r_bio;	      /* instead of pages */
-#endif
-
-	struct ceph_pagelist *r_trail;	      /* trailing part of the data */
-};
-
-struct ceph_osd_client {
-	struct ceph_client     *client;
-
-	struct ceph_osdmap     *osdmap;       /* current map */
-	struct rw_semaphore    map_sem;
-	struct completion      map_waiters;
-	u64                    last_requested_map;
-
-	struct mutex           request_mutex;
-	struct rb_root         osds;          /* osds */
-	struct list_head       osd_lru;       /* idle osds */
-	u64                    timeout_tid;   /* tid of timeout triggering rq */
-	u64                    last_tid;      /* tid of last request */
-	struct rb_root         requests;      /* pending requests */
-	struct list_head       req_lru;	      /* pending requests lru */
-	int                    num_requests;
-	struct delayed_work    timeout_work;
-	struct delayed_work    osds_timeout_work;
-#ifdef CONFIG_DEBUG_FS
-	struct dentry 	       *debugfs_file;
-#endif
-
-	mempool_t              *req_mempool;
-
-	struct ceph_msgpool	msgpool_op;
-	struct ceph_msgpool	msgpool_op_reply;
-};
-
-struct ceph_osd_req_op {
-	u16 op;           /* CEPH_OSD_OP_* */
-	u32 flags;        /* CEPH_OSD_FLAG_* */
-	union {
-		struct {
-			u64 offset, length;
-			u64 truncate_size;
-			u32 truncate_seq;
-		} extent;
-		struct {
-			const char *name;
-			u32 name_len;
-			const char  *val;
-			u32 value_len;
-			__u8 cmp_op;       /* CEPH_OSD_CMPXATTR_OP_* */
-			__u8 cmp_mode;     /* CEPH_OSD_CMPXATTR_MODE_* */
-		} xattr;
-		struct {
-			const char *class_name;
-			__u8 class_len;
-			const char *method_name;
-			__u8 method_len;
-			__u8 argc;
-			const char *indata;
-			u32 indata_len;
-		} cls;
-		struct {
-			u64 cookie, count;
-		} pgls;
-	        struct {
-		        u64 snapid;
-	        } snap;
-	};
-	u32 payload_len;
-};
-
-extern int ceph_osdc_init(struct ceph_osd_client *osdc,
-			  struct ceph_client *client);
-extern void ceph_osdc_stop(struct ceph_osd_client *osdc);
-
-extern void ceph_osdc_handle_reply(struct ceph_osd_client *osdc,
-				   struct ceph_msg *msg);
-extern void ceph_osdc_handle_map(struct ceph_osd_client *osdc,
-				 struct ceph_msg *msg);
-
-extern void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
-			struct ceph_file_layout *layout,
-			u64 snapid,
-			u64 off, u64 *plen, u64 *bno,
-			struct ceph_osd_request *req,
-			struct ceph_osd_req_op *op);
-
-extern struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
-					       int flags,
-					       struct ceph_snap_context *snapc,
-					       struct ceph_osd_req_op *ops,
-					       bool use_mempool,
-					       gfp_t gfp_flags,
-					       struct page **pages,
-					       struct bio *bio);
-
-extern void ceph_osdc_build_request(struct ceph_osd_request *req,
-				    u64 off, u64 *plen,
-				    struct ceph_osd_req_op *src_ops,
-				    struct ceph_snap_context *snapc,
-				    struct timespec *mtime,
-				    const char *oid,
-				    int oid_len);
-
-extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *,
-				      struct ceph_file_layout *layout,
-				      struct ceph_vino vino,
-				      u64 offset, u64 *len, int op, int flags,
-				      struct ceph_snap_context *snapc,
-				      int do_sync, u32 truncate_seq,
-				      u64 truncate_size,
-				      struct timespec *mtime,
-				      bool use_mempool, int num_reply);
-
-static inline void ceph_osdc_get_request(struct ceph_osd_request *req)
-{
-	kref_get(&req->r_kref);
-}
-extern void ceph_osdc_release_request(struct kref *kref);
-static inline void ceph_osdc_put_request(struct ceph_osd_request *req)
-{
-	kref_put(&req->r_kref, ceph_osdc_release_request);
-}
-
-extern int ceph_osdc_start_request(struct ceph_osd_client *osdc,
-				   struct ceph_osd_request *req,
-				   bool nofail);
-extern int ceph_osdc_wait_request(struct ceph_osd_client *osdc,
-				  struct ceph_osd_request *req);
-extern void ceph_osdc_sync(struct ceph_osd_client *osdc);
-
-extern int ceph_osdc_readpages(struct ceph_osd_client *osdc,
-			       struct ceph_vino vino,
-			       struct ceph_file_layout *layout,
-			       u64 off, u64 *plen,
-			       u32 truncate_seq, u64 truncate_size,
-			       struct page **pages, int nr_pages);
-
-extern int ceph_osdc_writepages(struct ceph_osd_client *osdc,
-				struct ceph_vino vino,
-				struct ceph_file_layout *layout,
-				struct ceph_snap_context *sc,
-				u64 off, u64 len,
-				u32 truncate_seq, u64 truncate_size,
-				struct timespec *mtime,
-				struct page **pages, int nr_pages,
-				int flags, int do_sync, bool nofail);
-
-#endif
-
diff --git a/fs/ceph/osdmap.c b/fs/ceph/osdmap.c
deleted file mode 100644
index 3ccd11761cb6..000000000000
--- a/fs/ceph/osdmap.c
+++ /dev/null
@@ -1,1123 +0,0 @@
-
-#include "ceph_debug.h"
-
-#include <linux/slab.h>
-#include <asm/div64.h>
-
-#include "super.h"
-#include "osdmap.h"
-#include "crush/hash.h"
-#include "crush/mapper.h"
-#include "decode.h"
-
-char *ceph_osdmap_state_str(char *str, int len, int state)
-{
-	int flag = 0;
-
-	if (!len)
-		goto done;
-
-	*str = '\0';
-	if (state) {
-		if (state & CEPH_OSD_EXISTS) {
-			snprintf(str, len, "exists");
-			flag = 1;
-		}
-		if (state & CEPH_OSD_UP) {
-			snprintf(str, len, "%s%s%s", str, (flag ? ", " : ""),
-				 "up");
-			flag = 1;
-		}
-	} else {
-		snprintf(str, len, "doesn't exist");
-	}
-done:
-	return str;
-}
-
-/* maps */
-
-static int calc_bits_of(unsigned t)
-{
-	int b = 0;
-	while (t) {
-		t = t >> 1;
-		b++;
-	}
-	return b;
-}
-
-/*
- * the foo_mask is the smallest value 2^n-1 that is >= foo.
- */
-static void calc_pg_masks(struct ceph_pg_pool_info *pi)
-{
-	pi->pg_num_mask = (1 << calc_bits_of(le32_to_cpu(pi->v.pg_num)-1)) - 1;
-	pi->pgp_num_mask =
-		(1 << calc_bits_of(le32_to_cpu(pi->v.pgp_num)-1)) - 1;
-	pi->lpg_num_mask =
-		(1 << calc_bits_of(le32_to_cpu(pi->v.lpg_num)-1)) - 1;
-	pi->lpgp_num_mask =
-		(1 << calc_bits_of(le32_to_cpu(pi->v.lpgp_num)-1)) - 1;
-}
-
-/*
- * decode crush map
- */
-static int crush_decode_uniform_bucket(void **p, void *end,
-				       struct crush_bucket_uniform *b)
-{
-	dout("crush_decode_uniform_bucket %p to %p\n", *p, end);
-	ceph_decode_need(p, end, (1+b->h.size) * sizeof(u32), bad);
-	b->item_weight = ceph_decode_32(p);
-	return 0;
-bad:
-	return -EINVAL;
-}
-
-static int crush_decode_list_bucket(void **p, void *end,
-				    struct crush_bucket_list *b)
-{
-	int j;
-	dout("crush_decode_list_bucket %p to %p\n", *p, end);
-	b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
-	if (b->item_weights == NULL)
-		return -ENOMEM;
-	b->sum_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
-	if (b->sum_weights == NULL)
-		return -ENOMEM;
-	ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad);
-	for (j = 0; j < b->h.size; j++) {
-		b->item_weights[j] = ceph_decode_32(p);
-		b->sum_weights[j] = ceph_decode_32(p);
-	}
-	return 0;
-bad:
-	return -EINVAL;
-}
-
-static int crush_decode_tree_bucket(void **p, void *end,
-				    struct crush_bucket_tree *b)
-{
-	int j;
-	dout("crush_decode_tree_bucket %p to %p\n", *p, end);
-	ceph_decode_32_safe(p, end, b->num_nodes, bad);
-	b->node_weights = kcalloc(b->num_nodes, sizeof(u32), GFP_NOFS);
-	if (b->node_weights == NULL)
-		return -ENOMEM;
-	ceph_decode_need(p, end, b->num_nodes * sizeof(u32), bad);
-	for (j = 0; j < b->num_nodes; j++)
-		b->node_weights[j] = ceph_decode_32(p);
-	return 0;
-bad:
-	return -EINVAL;
-}
-
-static int crush_decode_straw_bucket(void **p, void *end,
-				     struct crush_bucket_straw *b)
-{
-	int j;
-	dout("crush_decode_straw_bucket %p to %p\n", *p, end);
-	b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
-	if (b->item_weights == NULL)
-		return -ENOMEM;
-	b->straws = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
-	if (b->straws == NULL)
-		return -ENOMEM;
-	ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad);
-	for (j = 0; j < b->h.size; j++) {
-		b->item_weights[j] = ceph_decode_32(p);
-		b->straws[j] = ceph_decode_32(p);
-	}
-	return 0;
-bad:
-	return -EINVAL;
-}
-
-static struct crush_map *crush_decode(void *pbyval, void *end)
-{
-	struct crush_map *c;
-	int err = -EINVAL;
-	int i, j;
-	void **p = &pbyval;
-	void *start = pbyval;
-	u32 magic;
-
-	dout("crush_decode %p to %p len %d\n", *p, end, (int)(end - *p));
-
-	c = kzalloc(sizeof(*c), GFP_NOFS);
-	if (c == NULL)
-		return ERR_PTR(-ENOMEM);
-
-	ceph_decode_need(p, end, 4*sizeof(u32), bad);
-	magic = ceph_decode_32(p);
-	if (magic != CRUSH_MAGIC) {
-		pr_err("crush_decode magic %x != current %x\n",
-		       (unsigned)magic, (unsigned)CRUSH_MAGIC);
-		goto bad;
-	}
-	c->max_buckets = ceph_decode_32(p);
-	c->max_rules = ceph_decode_32(p);
-	c->max_devices = ceph_decode_32(p);
-
-	c->device_parents = kcalloc(c->max_devices, sizeof(u32), GFP_NOFS);
-	if (c->device_parents == NULL)
-		goto badmem;
-	c->bucket_parents = kcalloc(c->max_buckets, sizeof(u32), GFP_NOFS);
-	if (c->bucket_parents == NULL)
-		goto badmem;
-
-	c->buckets = kcalloc(c->max_buckets, sizeof(*c->buckets), GFP_NOFS);
-	if (c->buckets == NULL)
-		goto badmem;
-	c->rules = kcalloc(c->max_rules, sizeof(*c->rules), GFP_NOFS);
-	if (c->rules == NULL)
-		goto badmem;
-
-	/* buckets */
-	for (i = 0; i < c->max_buckets; i++) {
-		int size = 0;
-		u32 alg;
-		struct crush_bucket *b;
-
-		ceph_decode_32_safe(p, end, alg, bad);
-		if (alg == 0) {
-			c->buckets[i] = NULL;
-			continue;
-		}
-		dout("crush_decode bucket %d off %x %p to %p\n",
-		     i, (int)(*p-start), *p, end);
-
-		switch (alg) {
-		case CRUSH_BUCKET_UNIFORM:
-			size = sizeof(struct crush_bucket_uniform);
-			break;
-		case CRUSH_BUCKET_LIST:
-			size = sizeof(struct crush_bucket_list);
-			break;
-		case CRUSH_BUCKET_TREE:
-			size = sizeof(struct crush_bucket_tree);
-			break;
-		case CRUSH_BUCKET_STRAW:
-			size = sizeof(struct crush_bucket_straw);
-			break;
-		default:
-			err = -EINVAL;
-			goto bad;
-		}
-		BUG_ON(size == 0);
-		b = c->buckets[i] = kzalloc(size, GFP_NOFS);
-		if (b == NULL)
-			goto badmem;
-
-		ceph_decode_need(p, end, 4*sizeof(u32), bad);
-		b->id = ceph_decode_32(p);
-		b->type = ceph_decode_16(p);
-		b->alg = ceph_decode_8(p);
-		b->hash = ceph_decode_8(p);
-		b->weight = ceph_decode_32(p);
-		b->size = ceph_decode_32(p);
-
-		dout("crush_decode bucket size %d off %x %p to %p\n",
-		     b->size, (int)(*p-start), *p, end);
-
-		b->items = kcalloc(b->size, sizeof(__s32), GFP_NOFS);
-		if (b->items == NULL)
-			goto badmem;
-		b->perm = kcalloc(b->size, sizeof(u32), GFP_NOFS);
-		if (b->perm == NULL)
-			goto badmem;
-		b->perm_n = 0;
-
-		ceph_decode_need(p, end, b->size*sizeof(u32), bad);
-		for (j = 0; j < b->size; j++)
-			b->items[j] = ceph_decode_32(p);
-
-		switch (b->alg) {
-		case CRUSH_BUCKET_UNIFORM:
-			err = crush_decode_uniform_bucket(p, end,
-				  (struct crush_bucket_uniform *)b);
-			if (err < 0)
-				goto bad;
-			break;
-		case CRUSH_BUCKET_LIST:
-			err = crush_decode_list_bucket(p, end,
-			       (struct crush_bucket_list *)b);
-			if (err < 0)
-				goto bad;
-			break;
-		case CRUSH_BUCKET_TREE:
-			err = crush_decode_tree_bucket(p, end,
-				(struct crush_bucket_tree *)b);
-			if (err < 0)
-				goto bad;
-			break;
-		case CRUSH_BUCKET_STRAW:
-			err = crush_decode_straw_bucket(p, end,
-				(struct crush_bucket_straw *)b);
-			if (err < 0)
-				goto bad;
-			break;
-		}
-	}
-
-	/* rules */
-	dout("rule vec is %p\n", c->rules);
-	for (i = 0; i < c->max_rules; i++) {
-		u32 yes;
-		struct crush_rule *r;
-
-		ceph_decode_32_safe(p, end, yes, bad);
-		if (!yes) {
-			dout("crush_decode NO rule %d off %x %p to %p\n",
-			     i, (int)(*p-start), *p, end);
-			c->rules[i] = NULL;
-			continue;
-		}
-
-		dout("crush_decode rule %d off %x %p to %p\n",
-		     i, (int)(*p-start), *p, end);
-
-		/* len */
-		ceph_decode_32_safe(p, end, yes, bad);
-#if BITS_PER_LONG == 32
-		err = -EINVAL;
-		if (yes > ULONG_MAX / sizeof(struct crush_rule_step))
-			goto bad;
-#endif
-		r = c->rules[i] = kmalloc(sizeof(*r) +
-					  yes*sizeof(struct crush_rule_step),
-					  GFP_NOFS);
-		if (r == NULL)
-			goto badmem;
-		dout(" rule %d is at %p\n", i, r);
-		r->len = yes;
-		ceph_decode_copy_safe(p, end, &r->mask, 4, bad); /* 4 u8's */
-		ceph_decode_need(p, end, r->len*3*sizeof(u32), bad);
-		for (j = 0; j < r->len; j++) {
-			r->steps[j].op = ceph_decode_32(p);
-			r->steps[j].arg1 = ceph_decode_32(p);
-			r->steps[j].arg2 = ceph_decode_32(p);
-		}
-	}
-
-	/* ignore trailing name maps. */
-
-	dout("crush_decode success\n");
-	return c;
-
-badmem:
-	err = -ENOMEM;
-bad:
-	dout("crush_decode fail %d\n", err);
-	crush_destroy(c);
-	return ERR_PTR(err);
-}
-
-/*
- * rbtree of pg_mapping for handling pg_temp (explicit mapping of pgid
- * to a set of osds)
- */
-static int pgid_cmp(struct ceph_pg l, struct ceph_pg r)
-{
-	u64 a = *(u64 *)&l;
-	u64 b = *(u64 *)&r;
-
-	if (a < b)
-		return -1;
-	if (a > b)
-		return 1;
-	return 0;
-}
-
-static int __insert_pg_mapping(struct ceph_pg_mapping *new,
-			       struct rb_root *root)
-{
-	struct rb_node **p = &root->rb_node;
-	struct rb_node *parent = NULL;
-	struct ceph_pg_mapping *pg = NULL;
-	int c;
-
-	while (*p) {
-		parent = *p;
-		pg = rb_entry(parent, struct ceph_pg_mapping, node);
-		c = pgid_cmp(new->pgid, pg->pgid);
-		if (c < 0)
-			p = &(*p)->rb_left;
-		else if (c > 0)
-			p = &(*p)->rb_right;
-		else
-			return -EEXIST;
-	}
-
-	rb_link_node(&new->node, parent, p);
-	rb_insert_color(&new->node, root);
-	return 0;
-}
-
-static struct ceph_pg_mapping *__lookup_pg_mapping(struct rb_root *root,
-						   struct ceph_pg pgid)
-{
-	struct rb_node *n = root->rb_node;
-	struct ceph_pg_mapping *pg;
-	int c;
-
-	while (n) {
-		pg = rb_entry(n, struct ceph_pg_mapping, node);
-		c = pgid_cmp(pgid, pg->pgid);
-		if (c < 0)
-			n = n->rb_left;
-		else if (c > 0)
-			n = n->rb_right;
-		else
-			return pg;
-	}
-	return NULL;
-}
-
-/*
- * rbtree of pg pool info
- */
-static int __insert_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *new)
-{
-	struct rb_node **p = &root->rb_node;
-	struct rb_node *parent = NULL;
-	struct ceph_pg_pool_info *pi = NULL;
-
-	while (*p) {
-		parent = *p;
-		pi = rb_entry(parent, struct ceph_pg_pool_info, node);
-		if (new->id < pi->id)
-			p = &(*p)->rb_left;
-		else if (new->id > pi->id)
-			p = &(*p)->rb_right;
-		else
-			return -EEXIST;
-	}
-
-	rb_link_node(&new->node, parent, p);
-	rb_insert_color(&new->node, root);
-	return 0;
-}
-
-static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id)
-{
-	struct ceph_pg_pool_info *pi;
-	struct rb_node *n = root->rb_node;
-
-	while (n) {
-		pi = rb_entry(n, struct ceph_pg_pool_info, node);
-		if (id < pi->id)
-			n = n->rb_left;
-		else if (id > pi->id)
-			n = n->rb_right;
-		else
-			return pi;
-	}
-	return NULL;
-}
-
-int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name)
-{
-	struct rb_node *rbp;
-
-	for (rbp = rb_first(&map->pg_pools); rbp; rbp = rb_next(rbp)) {
-		struct ceph_pg_pool_info *pi =
-			rb_entry(rbp, struct ceph_pg_pool_info, node);
-		if (pi->name && strcmp(pi->name, name) == 0)
-			return pi->id;
-	}
-	return -ENOENT;
-}
-
-static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi)
-{
-	rb_erase(&pi->node, root);
-	kfree(pi->name);
-	kfree(pi);
-}
-
-static int __decode_pool(void **p, void *end, struct ceph_pg_pool_info *pi)
-{
-	unsigned n, m;
-
-	ceph_decode_copy(p, &pi->v, sizeof(pi->v));
-	calc_pg_masks(pi);
-
-	/* num_snaps * snap_info_t */
-	n = le32_to_cpu(pi->v.num_snaps);
-	while (n--) {
-		ceph_decode_need(p, end, sizeof(u64) + 1 + sizeof(u64) +
-				 sizeof(struct ceph_timespec), bad);
-		*p += sizeof(u64) +       /* key */
-			1 + sizeof(u64) + /* u8, snapid */
-			sizeof(struct ceph_timespec);
-		m = ceph_decode_32(p);    /* snap name */
-		*p += m;
-	}
-
-	*p += le32_to_cpu(pi->v.num_removed_snap_intervals) * sizeof(u64) * 2;
-	return 0;
-
-bad:
-	return -EINVAL;
-}
-
-static int __decode_pool_names(void **p, void *end, struct ceph_osdmap *map)
-{
-	struct ceph_pg_pool_info *pi;
-	u32 num, len, pool;
-
-	ceph_decode_32_safe(p, end, num, bad);
-	dout(" %d pool names\n", num);
-	while (num--) {
-		ceph_decode_32_safe(p, end, pool, bad);
-		ceph_decode_32_safe(p, end, len, bad);
-		dout("  pool %d len %d\n", pool, len);
-		pi = __lookup_pg_pool(&map->pg_pools, pool);
-		if (pi) {
-			kfree(pi->name);
-			pi->name = kmalloc(len + 1, GFP_NOFS);
-			if (pi->name) {
-				memcpy(pi->name, *p, len);
-				pi->name[len] = '\0';
-				dout("  name is %s\n", pi->name);
-			}
-		}
-		*p += len;
-	}
-	return 0;
-
-bad:
-	return -EINVAL;
-}
-
-/*
- * osd map
- */
-void ceph_osdmap_destroy(struct ceph_osdmap *map)
-{
-	dout("osdmap_destroy %p\n", map);
-	if (map->crush)
-		crush_destroy(map->crush);
-	while (!RB_EMPTY_ROOT(&map->pg_temp)) {
-		struct ceph_pg_mapping *pg =
-			rb_entry(rb_first(&map->pg_temp),
-				 struct ceph_pg_mapping, node);
-		rb_erase(&pg->node, &map->pg_temp);
-		kfree(pg);
-	}
-	while (!RB_EMPTY_ROOT(&map->pg_pools)) {
-		struct ceph_pg_pool_info *pi =
-			rb_entry(rb_first(&map->pg_pools),
-				 struct ceph_pg_pool_info, node);
-		__remove_pg_pool(&map->pg_pools, pi);
-	}
-	kfree(map->osd_state);
-	kfree(map->osd_weight);
-	kfree(map->osd_addr);
-	kfree(map);
-}
-
-/*
- * adjust max osd value.  reallocate arrays.
- */
-static int osdmap_set_max_osd(struct ceph_osdmap *map, int max)
-{
-	u8 *state;
-	struct ceph_entity_addr *addr;
-	u32 *weight;
-
-	state = kcalloc(max, sizeof(*state), GFP_NOFS);
-	addr = kcalloc(max, sizeof(*addr), GFP_NOFS);
-	weight = kcalloc(max, sizeof(*weight), GFP_NOFS);
-	if (state == NULL || addr == NULL || weight == NULL) {
-		kfree(state);
-		kfree(addr);
-		kfree(weight);
-		return -ENOMEM;
-	}
-
-	/* copy old? */
-	if (map->osd_state) {
-		memcpy(state, map->osd_state, map->max_osd*sizeof(*state));
-		memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr));
-		memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight));
-		kfree(map->osd_state);
-		kfree(map->osd_addr);
-		kfree(map->osd_weight);
-	}
-
-	map->osd_state = state;
-	map->osd_weight = weight;
-	map->osd_addr = addr;
-	map->max_osd = max;
-	return 0;
-}
-
-/*
- * decode a full map.
- */
-struct ceph_osdmap *osdmap_decode(void **p, void *end)
-{
-	struct ceph_osdmap *map;
-	u16 version;
-	u32 len, max, i;
-	u8 ev;
-	int err = -EINVAL;
-	void *start = *p;
-	struct ceph_pg_pool_info *pi;
-
-	dout("osdmap_decode %p to %p len %d\n", *p, end, (int)(end - *p));
-
-	map = kzalloc(sizeof(*map), GFP_NOFS);
-	if (map == NULL)
-		return ERR_PTR(-ENOMEM);
-	map->pg_temp = RB_ROOT;
-
-	ceph_decode_16_safe(p, end, version, bad);
-	if (version > CEPH_OSDMAP_VERSION) {
-		pr_warning("got unknown v %d > %d of osdmap\n", version,
-			   CEPH_OSDMAP_VERSION);
-		goto bad;
-	}
-
-	ceph_decode_need(p, end, 2*sizeof(u64)+6*sizeof(u32), bad);
-	ceph_decode_copy(p, &map->fsid, sizeof(map->fsid));
-	map->epoch = ceph_decode_32(p);
-	ceph_decode_copy(p, &map->created, sizeof(map->created));
-	ceph_decode_copy(p, &map->modified, sizeof(map->modified));
-
-	ceph_decode_32_safe(p, end, max, bad);
-	while (max--) {
-		ceph_decode_need(p, end, 4 + 1 + sizeof(pi->v), bad);
-		pi = kzalloc(sizeof(*pi), GFP_NOFS);
-		if (!pi)
-			goto bad;
-		pi->id = ceph_decode_32(p);
-		ev = ceph_decode_8(p); /* encoding version */
-		if (ev > CEPH_PG_POOL_VERSION) {
-			pr_warning("got unknown v %d > %d of ceph_pg_pool\n",
-				   ev, CEPH_PG_POOL_VERSION);
-			kfree(pi);
-			goto bad;
-		}
-		err = __decode_pool(p, end, pi);
-		if (err < 0)
-			goto bad;
-		__insert_pg_pool(&map->pg_pools, pi);
-	}
-
-	if (version >= 5 && __decode_pool_names(p, end, map) < 0)
-		goto bad;
-
-	ceph_decode_32_safe(p, end, map->pool_max, bad);
-
-	ceph_decode_32_safe(p, end, map->flags, bad);
-
-	max = ceph_decode_32(p);
-
-	/* (re)alloc osd arrays */
-	err = osdmap_set_max_osd(map, max);
-	if (err < 0)
-		goto bad;
-	dout("osdmap_decode max_osd = %d\n", map->max_osd);
-
-	/* osds */
-	err = -EINVAL;
-	ceph_decode_need(p, end, 3*sizeof(u32) +
-			 map->max_osd*(1 + sizeof(*map->osd_weight) +
-				       sizeof(*map->osd_addr)), bad);
-	*p += 4; /* skip length field (should match max) */
-	ceph_decode_copy(p, map->osd_state, map->max_osd);
-
-	*p += 4; /* skip length field (should match max) */
-	for (i = 0; i < map->max_osd; i++)
-		map->osd_weight[i] = ceph_decode_32(p);
-
-	*p += 4; /* skip length field (should match max) */
-	ceph_decode_copy(p, map->osd_addr, map->max_osd*sizeof(*map->osd_addr));
-	for (i = 0; i < map->max_osd; i++)
-		ceph_decode_addr(&map->osd_addr[i]);
-
-	/* pg_temp */
-	ceph_decode_32_safe(p, end, len, bad);
-	for (i = 0; i < len; i++) {
-		int n, j;
-		struct ceph_pg pgid;
-		struct ceph_pg_mapping *pg;
-
-		ceph_decode_need(p, end, sizeof(u32) + sizeof(u64), bad);
-		ceph_decode_copy(p, &pgid, sizeof(pgid));
-		n = ceph_decode_32(p);
-		ceph_decode_need(p, end, n * sizeof(u32), bad);
-		err = -ENOMEM;
-		pg = kmalloc(sizeof(*pg) + n*sizeof(u32), GFP_NOFS);
-		if (!pg)
-			goto bad;
-		pg->pgid = pgid;
-		pg->len = n;
-		for (j = 0; j < n; j++)
-			pg->osds[j] = ceph_decode_32(p);
-
-		err = __insert_pg_mapping(pg, &map->pg_temp);
-		if (err)
-			goto bad;
-		dout(" added pg_temp %llx len %d\n", *(u64 *)&pgid, len);
-	}
-
-	/* crush */
-	ceph_decode_32_safe(p, end, len, bad);
-	dout("osdmap_decode crush len %d from off 0x%x\n", len,
-	     (int)(*p - start));
-	ceph_decode_need(p, end, len, bad);
-	map->crush = crush_decode(*p, end);
-	*p += len;
-	if (IS_ERR(map->crush)) {
-		err = PTR_ERR(map->crush);
-		map->crush = NULL;
-		goto bad;
-	}
-
-	/* ignore the rest of the map */
-	*p = end;
-
-	dout("osdmap_decode done %p %p\n", *p, end);
-	return map;
-
-bad:
-	dout("osdmap_decode fail\n");
-	ceph_osdmap_destroy(map);
-	return ERR_PTR(err);
-}
-
-/*
- * decode and apply an incremental map update.
- */
-struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
-					     struct ceph_osdmap *map,
-					     struct ceph_messenger *msgr)
-{
-	struct crush_map *newcrush = NULL;
-	struct ceph_fsid fsid;
-	u32 epoch = 0;
-	struct ceph_timespec modified;
-	u32 len, pool;
-	__s32 new_pool_max, new_flags, max;
-	void *start = *p;
-	int err = -EINVAL;
-	u16 version;
-	struct rb_node *rbp;
-
-	ceph_decode_16_safe(p, end, version, bad);
-	if (version > CEPH_OSDMAP_INC_VERSION) {
-		pr_warning("got unknown v %d > %d of inc osdmap\n", version,
-			   CEPH_OSDMAP_INC_VERSION);
-		goto bad;
-	}
-
-	ceph_decode_need(p, end, sizeof(fsid)+sizeof(modified)+2*sizeof(u32),
-			 bad);
-	ceph_decode_copy(p, &fsid, sizeof(fsid));
-	epoch = ceph_decode_32(p);
-	BUG_ON(epoch != map->epoch+1);
-	ceph_decode_copy(p, &modified, sizeof(modified));
-	new_pool_max = ceph_decode_32(p);
-	new_flags = ceph_decode_32(p);
-
-	/* full map? */
-	ceph_decode_32_safe(p, end, len, bad);
-	if (len > 0) {
-		dout("apply_incremental full map len %d, %p to %p\n",
-		     len, *p, end);
-		return osdmap_decode(p, min(*p+len, end));
-	}
-
-	/* new crush? */
-	ceph_decode_32_safe(p, end, len, bad);
-	if (len > 0) {
-		dout("apply_incremental new crush map len %d, %p to %p\n",
-		     len, *p, end);
-		newcrush = crush_decode(*p, min(*p+len, end));
-		if (IS_ERR(newcrush))
-			return ERR_CAST(newcrush);
-		*p += len;
-	}
-
-	/* new flags? */
-	if (new_flags >= 0)
-		map->flags = new_flags;
-	if (new_pool_max >= 0)
-		map->pool_max = new_pool_max;
-
-	ceph_decode_need(p, end, 5*sizeof(u32), bad);
-
-	/* new max? */
-	max = ceph_decode_32(p);
-	if (max >= 0) {
-		err = osdmap_set_max_osd(map, max);
-		if (err < 0)
-			goto bad;
-	}
-
-	map->epoch++;
-	map->modified = map->modified;
-	if (newcrush) {
-		if (map->crush)
-			crush_destroy(map->crush);
-		map->crush = newcrush;
-		newcrush = NULL;
-	}
-
-	/* new_pool */
-	ceph_decode_32_safe(p, end, len, bad);
-	while (len--) {
-		__u8 ev;
-		struct ceph_pg_pool_info *pi;
-
-		ceph_decode_32_safe(p, end, pool, bad);
-		ceph_decode_need(p, end, 1 + sizeof(pi->v), bad);
-		ev = ceph_decode_8(p);  /* encoding version */
-		if (ev > CEPH_PG_POOL_VERSION) {
-			pr_warning("got unknown v %d > %d of ceph_pg_pool\n",
-				   ev, CEPH_PG_POOL_VERSION);
-			goto bad;
-		}
-		pi = __lookup_pg_pool(&map->pg_pools, pool);
-		if (!pi) {
-			pi = kzalloc(sizeof(*pi), GFP_NOFS);
-			if (!pi) {
-				err = -ENOMEM;
-				goto bad;
-			}
-			pi->id = pool;
-			__insert_pg_pool(&map->pg_pools, pi);
-		}
-		err = __decode_pool(p, end, pi);
-		if (err < 0)
-			goto bad;
-	}
-	if (version >= 5 && __decode_pool_names(p, end, map) < 0)
-		goto bad;
-
-	/* old_pool */
-	ceph_decode_32_safe(p, end, len, bad);
-	while (len--) {
-		struct ceph_pg_pool_info *pi;
-
-		ceph_decode_32_safe(p, end, pool, bad);
-		pi = __lookup_pg_pool(&map->pg_pools, pool);
-		if (pi)
-			__remove_pg_pool(&map->pg_pools, pi);
-	}
-
-	/* new_up */
-	err = -EINVAL;
-	ceph_decode_32_safe(p, end, len, bad);
-	while (len--) {
-		u32 osd;
-		struct ceph_entity_addr addr;
-		ceph_decode_32_safe(p, end, osd, bad);
-		ceph_decode_copy_safe(p, end, &addr, sizeof(addr), bad);
-		ceph_decode_addr(&addr);
-		pr_info("osd%d up\n", osd);
-		BUG_ON(osd >= map->max_osd);
-		map->osd_state[osd] |= CEPH_OSD_UP;
-		map->osd_addr[osd] = addr;
-	}
-
-	/* new_down */
-	ceph_decode_32_safe(p, end, len, bad);
-	while (len--) {
-		u32 osd;
-		ceph_decode_32_safe(p, end, osd, bad);
-		(*p)++;  /* clean flag */
-		pr_info("osd%d down\n", osd);
-		if (osd < map->max_osd)
-			map->osd_state[osd] &= ~CEPH_OSD_UP;
-	}
-
-	/* new_weight */
-	ceph_decode_32_safe(p, end, len, bad);
-	while (len--) {
-		u32 osd, off;
-		ceph_decode_need(p, end, sizeof(u32)*2, bad);
-		osd = ceph_decode_32(p);
-		off = ceph_decode_32(p);
-		pr_info("osd%d weight 0x%x %s\n", osd, off,
-		     off == CEPH_OSD_IN ? "(in)" :
-		     (off == CEPH_OSD_OUT ? "(out)" : ""));
-		if (osd < map->max_osd)
-			map->osd_weight[osd] = off;
-	}
-
-	/* new_pg_temp */
-	rbp = rb_first(&map->pg_temp);
-	ceph_decode_32_safe(p, end, len, bad);
-	while (len--) {
-		struct ceph_pg_mapping *pg;
-		int j;
-		struct ceph_pg pgid;
-		u32 pglen;
-		ceph_decode_need(p, end, sizeof(u64) + sizeof(u32), bad);
-		ceph_decode_copy(p, &pgid, sizeof(pgid));
-		pglen = ceph_decode_32(p);
-
-		/* remove any? */
-		while (rbp && pgid_cmp(rb_entry(rbp, struct ceph_pg_mapping,
-						node)->pgid, pgid) <= 0) {
-			struct ceph_pg_mapping *cur =
-				rb_entry(rbp, struct ceph_pg_mapping, node);
-
-			rbp = rb_next(rbp);
-			dout(" removed pg_temp %llx\n", *(u64 *)&cur->pgid);
-			rb_erase(&cur->node, &map->pg_temp);
-			kfree(cur);
-		}
-
-		if (pglen) {
-			/* insert */
-			ceph_decode_need(p, end, pglen*sizeof(u32), bad);
-			pg = kmalloc(sizeof(*pg) + sizeof(u32)*pglen, GFP_NOFS);
-			if (!pg) {
-				err = -ENOMEM;
-				goto bad;
-			}
-			pg->pgid = pgid;
-			pg->len = pglen;
-			for (j = 0; j < pglen; j++)
-				pg->osds[j] = ceph_decode_32(p);
-			err = __insert_pg_mapping(pg, &map->pg_temp);
-			if (err) {
-				kfree(pg);
-				goto bad;
-			}
-			dout(" added pg_temp %llx len %d\n", *(u64 *)&pgid,
-			     pglen);
-		}
-	}
-	while (rbp) {
-		struct ceph_pg_mapping *cur =
-			rb_entry(rbp, struct ceph_pg_mapping, node);
-
-		rbp = rb_next(rbp);
-		dout(" removed pg_temp %llx\n", *(u64 *)&cur->pgid);
-		rb_erase(&cur->node, &map->pg_temp);
-		kfree(cur);
-	}
-
-	/* ignore the rest */
-	*p = end;
-	return map;
-
-bad:
-	pr_err("corrupt inc osdmap epoch %d off %d (%p of %p-%p)\n",
-	       epoch, (int)(*p - start), *p, start, end);
-	print_hex_dump(KERN_DEBUG, "osdmap: ",
-		       DUMP_PREFIX_OFFSET, 16, 1,
-		       start, end - start, true);
-	if (newcrush)
-		crush_destroy(newcrush);
-	return ERR_PTR(err);
-}
-
-
-
-
-/*
- * calculate file layout from given offset, length.
- * fill in correct oid, logical length, and object extent
- * offset, length.
- *
- * for now, we write only a single su, until we can
- * pass a stride back to the caller.
- */
-void ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
-				   u64 off, u64 *plen,
-				   u64 *ono,
-				   u64 *oxoff, u64 *oxlen)
-{
-	u32 osize = le32_to_cpu(layout->fl_object_size);
-	u32 su = le32_to_cpu(layout->fl_stripe_unit);
-	u32 sc = le32_to_cpu(layout->fl_stripe_count);
-	u32 bl, stripeno, stripepos, objsetno;
-	u32 su_per_object;
-	u64 t, su_offset;
-
-	dout("mapping %llu~%llu  osize %u fl_su %u\n", off, *plen,
-	     osize, su);
-	su_per_object = osize / su;
-	dout("osize %u / su %u = su_per_object %u\n", osize, su,
-	     su_per_object);
-
-	BUG_ON((su & ~PAGE_MASK) != 0);
-	/* bl = *off / su; */
-	t = off;
-	do_div(t, su);
-	bl = t;
-	dout("off %llu / su %u = bl %u\n", off, su, bl);
-
-	stripeno = bl / sc;
-	stripepos = bl % sc;
-	objsetno = stripeno / su_per_object;
-
-	*ono = objsetno * sc + stripepos;
-	dout("objset %u * sc %u = ono %u\n", objsetno, sc, (unsigned)*ono);
-
-	/* *oxoff = *off % layout->fl_stripe_unit;  # offset in su */
-	t = off;
-	su_offset = do_div(t, su);
-	*oxoff = su_offset + (stripeno % su_per_object) * su;
-
-	/*
-	 * Calculate the length of the extent being written to the selected
-	 * object. This is the minimum of the full length requested (plen) or
-	 * the remainder of the current stripe being written to.
-	 */
-	*oxlen = min_t(u64, *plen, su - su_offset);
-	*plen = *oxlen;
-
-	dout(" obj extent %llu~%llu\n", *oxoff, *oxlen);
-}
-
-/*
- * calculate an object layout (i.e. pgid) from an oid,
- * file_layout, and osdmap
- */
-int ceph_calc_object_layout(struct ceph_object_layout *ol,
-			    const char *oid,
-			    struct ceph_file_layout *fl,
-			    struct ceph_osdmap *osdmap)
-{
-	unsigned num, num_mask;
-	struct ceph_pg pgid;
-	s32 preferred = (s32)le32_to_cpu(fl->fl_pg_preferred);
-	int poolid = le32_to_cpu(fl->fl_pg_pool);
-	struct ceph_pg_pool_info *pool;
-	unsigned ps;
-
-	BUG_ON(!osdmap);
-
-	pool = __lookup_pg_pool(&osdmap->pg_pools, poolid);
-	if (!pool)
-		return -EIO;
-	ps = ceph_str_hash(pool->v.object_hash, oid, strlen(oid));
-	if (preferred >= 0) {
-		ps += preferred;
-		num = le32_to_cpu(pool->v.lpg_num);
-		num_mask = pool->lpg_num_mask;
-	} else {
-		num = le32_to_cpu(pool->v.pg_num);
-		num_mask = pool->pg_num_mask;
-	}
-
-	pgid.ps = cpu_to_le16(ps);
-	pgid.preferred = cpu_to_le16(preferred);
-	pgid.pool = fl->fl_pg_pool;
-	if (preferred >= 0)
-		dout("calc_object_layout '%s' pgid %d.%xp%d\n", oid, poolid, ps,
-		     (int)preferred);
-	else
-		dout("calc_object_layout '%s' pgid %d.%x\n", oid, poolid, ps);
-
-	ol->ol_pgid = pgid;
-	ol->ol_stripe_unit = fl->fl_object_stripe_unit;
-	return 0;
-}
-
-/*
- * Calculate raw osd vector for the given pgid.  Return pointer to osd
- * array, or NULL on failure.
- */
-static int *calc_pg_raw(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
-			int *osds, int *num)
-{
-	struct ceph_pg_mapping *pg;
-	struct ceph_pg_pool_info *pool;
-	int ruleno;
-	unsigned poolid, ps, pps;
-	int preferred;
-
-	/* pg_temp? */
-	pg = __lookup_pg_mapping(&osdmap->pg_temp, pgid);
-	if (pg) {
-		*num = pg->len;
-		return pg->osds;
-	}
-
-	/* crush */
-	poolid = le32_to_cpu(pgid.pool);
-	ps = le16_to_cpu(pgid.ps);
-	preferred = (s16)le16_to_cpu(pgid.preferred);
-
-	/* don't forcefeed bad device ids to crush */
-	if (preferred >= osdmap->max_osd ||
-	    preferred >= osdmap->crush->max_devices)
-		preferred = -1;
-
-	pool = __lookup_pg_pool(&osdmap->pg_pools, poolid);
-	if (!pool)
-		return NULL;
-	ruleno = crush_find_rule(osdmap->crush, pool->v.crush_ruleset,
-				 pool->v.type, pool->v.size);
-	if (ruleno < 0) {
-		pr_err("no crush rule pool %d ruleset %d type %d size %d\n",
-		       poolid, pool->v.crush_ruleset, pool->v.type,
-		       pool->v.size);
-		return NULL;
-	}
-
-	if (preferred >= 0)
-		pps = ceph_stable_mod(ps,
-				      le32_to_cpu(pool->v.lpgp_num),
-				      pool->lpgp_num_mask);
-	else
-		pps = ceph_stable_mod(ps,
-				      le32_to_cpu(pool->v.pgp_num),
-				      pool->pgp_num_mask);
-	pps += poolid;
-	*num = crush_do_rule(osdmap->crush, ruleno, pps, osds,
-			     min_t(int, pool->v.size, *num),
-			     preferred, osdmap->osd_weight);
-	return osds;
-}
-
-/*
- * Return acting set for given pgid.
- */
-int ceph_calc_pg_acting(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
-			int *acting)
-{
-	int rawosds[CEPH_PG_MAX_SIZE], *osds;
-	int i, o, num = CEPH_PG_MAX_SIZE;
-
-	osds = calc_pg_raw(osdmap, pgid, rawosds, &num);
-	if (!osds)
-		return -1;
-
-	/* primary is first up osd */
-	o = 0;
-	for (i = 0; i < num; i++)
-		if (ceph_osd_is_up(osdmap, osds[i]))
-			acting[o++] = osds[i];
-	return o;
-}
-
-/*
- * Return primary osd for given pgid, or -1 if none.
- */
-int ceph_calc_pg_primary(struct ceph_osdmap *osdmap, struct ceph_pg pgid)
-{
-	int rawosds[CEPH_PG_MAX_SIZE], *osds;
-	int i, num = CEPH_PG_MAX_SIZE;
-
-	osds = calc_pg_raw(osdmap, pgid, rawosds, &num);
-	if (!osds)
-		return -1;
-
-	/* primary is first up osd */
-	for (i = 0; i < num; i++)
-		if (ceph_osd_is_up(osdmap, osds[i]))
-			return osds[i];
-	return -1;
-}
diff --git a/fs/ceph/osdmap.h b/fs/ceph/osdmap.h
deleted file mode 100644
index a592b211be39..000000000000
--- a/fs/ceph/osdmap.h
+++ /dev/null
@@ -1,130 +0,0 @@
-#ifndef _FS_CEPH_OSDMAP_H
-#define _FS_CEPH_OSDMAP_H
-
-#include <linux/rbtree.h>
-#include "types.h"
-#include "ceph_fs.h"
-#include "crush/crush.h"
-
-/*
- * The osd map describes the current membership of the osd cluster and
- * specifies the mapping of objects to placement groups and placement
- * groups to (sets of) osds.  That is, it completely specifies the
- * (desired) distribution of all data objects in the system at some
- * point in time.
- *
- * Each map version is identified by an epoch, which increases monotonically.
- *
- * The map can be updated either via an incremental map (diff) describing
- * the change between two successive epochs, or as a fully encoded map.
- */
-struct ceph_pg_pool_info {
-	struct rb_node node;
-	int id;
-	struct ceph_pg_pool v;
-	int pg_num_mask, pgp_num_mask, lpg_num_mask, lpgp_num_mask;
-	char *name;
-};
-
-struct ceph_pg_mapping {
-	struct rb_node node;
-	struct ceph_pg pgid;
-	int len;
-	int osds[];
-};
-
-struct ceph_osdmap {
-	struct ceph_fsid fsid;
-	u32 epoch;
-	u32 mkfs_epoch;
-	struct ceph_timespec created, modified;
-
-	u32 flags;         /* CEPH_OSDMAP_* */
-
-	u32 max_osd;       /* size of osd_state, _offload, _addr arrays */
-	u8 *osd_state;     /* CEPH_OSD_* */
-	u32 *osd_weight;   /* 0 = failed, 0x10000 = 100% normal */
-	struct ceph_entity_addr *osd_addr;
-
-	struct rb_root pg_temp;
-	struct rb_root pg_pools;
-	u32 pool_max;
-
-	/* the CRUSH map specifies the mapping of placement groups to
-	 * the list of osds that store+replicate them. */
-	struct crush_map *crush;
-};
-
-/*
- * file layout helpers
- */
-#define ceph_file_layout_su(l) ((__s32)le32_to_cpu((l).fl_stripe_unit))
-#define ceph_file_layout_stripe_count(l) \
-	((__s32)le32_to_cpu((l).fl_stripe_count))
-#define ceph_file_layout_object_size(l) ((__s32)le32_to_cpu((l).fl_object_size))
-#define ceph_file_layout_cas_hash(l) ((__s32)le32_to_cpu((l).fl_cas_hash))
-#define ceph_file_layout_object_su(l) \
-	((__s32)le32_to_cpu((l).fl_object_stripe_unit))
-#define ceph_file_layout_pg_preferred(l) \
-	((__s32)le32_to_cpu((l).fl_pg_preferred))
-#define ceph_file_layout_pg_pool(l) \
-	((__s32)le32_to_cpu((l).fl_pg_pool))
-
-static inline unsigned ceph_file_layout_stripe_width(struct ceph_file_layout *l)
-{
-	return le32_to_cpu(l->fl_stripe_unit) *
-		le32_to_cpu(l->fl_stripe_count);
-}
-
-/* "period" == bytes before i start on a new set of objects */
-static inline unsigned ceph_file_layout_period(struct ceph_file_layout *l)
-{
-	return le32_to_cpu(l->fl_object_size) *
-		le32_to_cpu(l->fl_stripe_count);
-}
-
-
-static inline int ceph_osd_is_up(struct ceph_osdmap *map, int osd)
-{
-	return (osd < map->max_osd) && (map->osd_state[osd] & CEPH_OSD_UP);
-}
-
-static inline bool ceph_osdmap_flag(struct ceph_osdmap *map, int flag)
-{
-	return map && (map->flags & flag);
-}
-
-extern char *ceph_osdmap_state_str(char *str, int len, int state);
-
-static inline struct ceph_entity_addr *ceph_osd_addr(struct ceph_osdmap *map,
-						     int osd)
-{
-	if (osd >= map->max_osd)
-		return NULL;
-	return &map->osd_addr[osd];
-}
-
-extern struct ceph_osdmap *osdmap_decode(void **p, void *end);
-extern struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
-					    struct ceph_osdmap *map,
-					    struct ceph_messenger *msgr);
-extern void ceph_osdmap_destroy(struct ceph_osdmap *map);
-
-/* calculate mapping of a file extent to an object */
-extern void ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
-					  u64 off, u64 *plen,
-					  u64 *bno, u64 *oxoff, u64 *oxlen);
-
-/* calculate mapping of object to a placement group */
-extern int ceph_calc_object_layout(struct ceph_object_layout *ol,
-				   const char *oid,
-				   struct ceph_file_layout *fl,
-				   struct ceph_osdmap *osdmap);
-extern int ceph_calc_pg_acting(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
-			       int *acting);
-extern int ceph_calc_pg_primary(struct ceph_osdmap *osdmap,
-				struct ceph_pg pgid);
-
-extern int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name);
-
-#endif
diff --git a/fs/ceph/pagelist.c b/fs/ceph/pagelist.c
deleted file mode 100644
index 326e1c04176f..000000000000
--- a/fs/ceph/pagelist.c
+++ /dev/null
@@ -1,63 +0,0 @@
-
-#include <linux/gfp.h>
-#include <linux/pagemap.h>
-#include <linux/highmem.h>
-
-#include "pagelist.h"
-
-static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl)
-{
-	struct page *page = list_entry(pl->head.prev, struct page,
-				       lru);
-	kunmap(page);
-}
-
-int ceph_pagelist_release(struct ceph_pagelist *pl)
-{
-	if (pl->mapped_tail)
-		ceph_pagelist_unmap_tail(pl);
-
-	while (!list_empty(&pl->head)) {
-		struct page *page = list_first_entry(&pl->head, struct page,
-						     lru);
-		list_del(&page->lru);
-		__free_page(page);
-	}
-	return 0;
-}
-
-static int ceph_pagelist_addpage(struct ceph_pagelist *pl)
-{
-	struct page *page = __page_cache_alloc(GFP_NOFS);
-	if (!page)
-		return -ENOMEM;
-	pl->room += PAGE_SIZE;
-	list_add_tail(&page->lru, &pl->head);
-	if (pl->mapped_tail)
-		ceph_pagelist_unmap_tail(pl);
-	pl->mapped_tail = kmap(page);
-	return 0;
-}
-
-int ceph_pagelist_append(struct ceph_pagelist *pl, const void *buf, size_t len)
-{
-	while (pl->room < len) {
-		size_t bit = pl->room;
-		int ret;
-
-		memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK),
-		       buf, bit);
-		pl->length += bit;
-		pl->room -= bit;
-		buf += bit;
-		len -= bit;
-		ret = ceph_pagelist_addpage(pl);
-		if (ret)
-			return ret;
-	}
-
-	memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK), buf, len);
-	pl->length += len;
-	pl->room -= len;
-	return 0;
-}
diff --git a/fs/ceph/pagelist.h b/fs/ceph/pagelist.h
deleted file mode 100644
index cc9327aa1c98..000000000000
--- a/fs/ceph/pagelist.h
+++ /dev/null
@@ -1,54 +0,0 @@
-#ifndef __FS_CEPH_PAGELIST_H
-#define __FS_CEPH_PAGELIST_H
-
-#include <linux/list.h>
-
-struct ceph_pagelist {
-	struct list_head head;
-	void *mapped_tail;
-	size_t length;
-	size_t room;
-};
-
-static inline void ceph_pagelist_init(struct ceph_pagelist *pl)
-{
-	INIT_LIST_HEAD(&pl->head);
-	pl->mapped_tail = NULL;
-	pl->length = 0;
-	pl->room = 0;
-}
-extern int ceph_pagelist_release(struct ceph_pagelist *pl);
-
-extern int ceph_pagelist_append(struct ceph_pagelist *pl, const void *d, size_t l);
-
-static inline int ceph_pagelist_encode_64(struct ceph_pagelist *pl, u64 v)
-{
-	__le64 ev = cpu_to_le64(v);
-	return ceph_pagelist_append(pl, &ev, sizeof(ev));
-}
-static inline int ceph_pagelist_encode_32(struct ceph_pagelist *pl, u32 v)
-{
-	__le32 ev = cpu_to_le32(v);
-	return ceph_pagelist_append(pl, &ev, sizeof(ev));
-}
-static inline int ceph_pagelist_encode_16(struct ceph_pagelist *pl, u16 v)
-{
-	__le16 ev = cpu_to_le16(v);
-	return ceph_pagelist_append(pl, &ev, sizeof(ev));
-}
-static inline int ceph_pagelist_encode_8(struct ceph_pagelist *pl, u8 v)
-{
-	return ceph_pagelist_append(pl, &v, 1);
-}
-static inline int ceph_pagelist_encode_string(struct ceph_pagelist *pl,
-					      char *s, size_t len)
-{
-	int ret = ceph_pagelist_encode_32(pl, len);
-	if (ret)
-		return ret;
-	if (len)
-		return ceph_pagelist_append(pl, s, len);
-	return 0;
-}
-
-#endif
diff --git a/fs/ceph/rados.h b/fs/ceph/rados.h
deleted file mode 100644
index 6d5247f2e81b..000000000000
--- a/fs/ceph/rados.h
+++ /dev/null
@@ -1,405 +0,0 @@
-#ifndef CEPH_RADOS_H
-#define CEPH_RADOS_H
-
-/*
- * Data types for the Ceph distributed object storage layer RADOS
- * (Reliable Autonomic Distributed Object Store).
- */
-
-#include "msgr.h"
-
-/*
- * osdmap encoding versions
- */
-#define CEPH_OSDMAP_INC_VERSION     5
-#define CEPH_OSDMAP_INC_VERSION_EXT 5
-#define CEPH_OSDMAP_VERSION         5
-#define CEPH_OSDMAP_VERSION_EXT     5
-
-/*
- * fs id
- */
-struct ceph_fsid {
-	unsigned char fsid[16];
-};
-
-static inline int ceph_fsid_compare(const struct ceph_fsid *a,
-				    const struct ceph_fsid *b)
-{
-	return memcmp(a, b, sizeof(*a));
-}
-
-/*
- * ino, object, etc.
- */
-typedef __le64 ceph_snapid_t;
-#define CEPH_SNAPDIR ((__u64)(-1))  /* reserved for hidden .snap dir */
-#define CEPH_NOSNAP  ((__u64)(-2))  /* "head", "live" revision */
-#define CEPH_MAXSNAP ((__u64)(-3))  /* largest valid snapid */
-
-struct ceph_timespec {
-	__le32 tv_sec;
-	__le32 tv_nsec;
-} __attribute__ ((packed));
-
-
-/*
- * object layout - how objects are mapped into PGs
- */
-#define CEPH_OBJECT_LAYOUT_HASH     1
-#define CEPH_OBJECT_LAYOUT_LINEAR   2
-#define CEPH_OBJECT_LAYOUT_HASHINO  3
-
-/*
- * pg layout -- how PGs are mapped onto (sets of) OSDs
- */
-#define CEPH_PG_LAYOUT_CRUSH  0
-#define CEPH_PG_LAYOUT_HASH   1
-#define CEPH_PG_LAYOUT_LINEAR 2
-#define CEPH_PG_LAYOUT_HYBRID 3
-
-#define CEPH_PG_MAX_SIZE      16  /* max # osds in a single pg */
-
-/*
- * placement group.
- * we encode this into one __le64.
- */
-struct ceph_pg {
-	__le16 preferred; /* preferred primary osd */
-	__le16 ps;        /* placement seed */
-	__le32 pool;      /* object pool */
-} __attribute__ ((packed));
-
-/*
- * pg_pool is a set of pgs storing a pool of objects
- *
- *  pg_num -- base number of pseudorandomly placed pgs
- *
- *  pgp_num -- effective number when calculating pg placement.  this
- * is used for pg_num increases.  new pgs result in data being "split"
- * into new pgs.  for this to proceed smoothly, new pgs are intiially
- * colocated with their parents; that is, pgp_num doesn't increase
- * until the new pgs have successfully split.  only _then_ are the new
- * pgs placed independently.
- *
- *  lpg_num -- localized pg count (per device).  replicas are randomly
- * selected.
- *
- *  lpgp_num -- as above.
- */
-#define CEPH_PG_TYPE_REP     1
-#define CEPH_PG_TYPE_RAID4   2
-#define CEPH_PG_POOL_VERSION 2
-struct ceph_pg_pool {
-	__u8 type;                /* CEPH_PG_TYPE_* */
-	__u8 size;                /* number of osds in each pg */
-	__u8 crush_ruleset;       /* crush placement rule */
-	__u8 object_hash;         /* hash mapping object name to ps */
-	__le32 pg_num, pgp_num;   /* number of pg's */
-	__le32 lpg_num, lpgp_num; /* number of localized pg's */
-	__le32 last_change;       /* most recent epoch changed */
-	__le64 snap_seq;          /* seq for per-pool snapshot */
-	__le32 snap_epoch;        /* epoch of last snap */
-	__le32 num_snaps;
-	__le32 num_removed_snap_intervals; /* if non-empty, NO per-pool snaps */
-	__le64 auid;               /* who owns the pg */
-} __attribute__ ((packed));
-
-/*
- * stable_mod func is used to control number of placement groups.
- * similar to straight-up modulo, but produces a stable mapping as b
- * increases over time.  b is the number of bins, and bmask is the
- * containing power of 2 minus 1.
- *
- * b <= bmask and bmask=(2**n)-1
- * e.g., b=12 -> bmask=15, b=123 -> bmask=127
- */
-static inline int ceph_stable_mod(int x, int b, int bmask)
-{
-	if ((x & bmask) < b)
-		return x & bmask;
-	else
-		return x & (bmask >> 1);
-}
-
-/*
- * object layout - how a given object should be stored.
- */
-struct ceph_object_layout {
-	struct ceph_pg ol_pgid;   /* raw pg, with _full_ ps precision. */
-	__le32 ol_stripe_unit;    /* for per-object parity, if any */
-} __attribute__ ((packed));
-
-/*
- * compound epoch+version, used by storage layer to serialize mutations
- */
-struct ceph_eversion {
-	__le32 epoch;
-	__le64 version;
-} __attribute__ ((packed));
-
-/*
- * osd map bits
- */
-
-/* status bits */
-#define CEPH_OSD_EXISTS 1
-#define CEPH_OSD_UP     2
-
-/* osd weights.  fixed point value: 0x10000 == 1.0 ("in"), 0 == "out" */
-#define CEPH_OSD_IN  0x10000
-#define CEPH_OSD_OUT 0
-
-
-/*
- * osd map flag bits
- */
-#define CEPH_OSDMAP_NEARFULL (1<<0)  /* sync writes (near ENOSPC) */
-#define CEPH_OSDMAP_FULL     (1<<1)  /* no data writes (ENOSPC) */
-#define CEPH_OSDMAP_PAUSERD  (1<<2)  /* pause all reads */
-#define CEPH_OSDMAP_PAUSEWR  (1<<3)  /* pause all writes */
-#define CEPH_OSDMAP_PAUSEREC (1<<4)  /* pause recovery */
-
-/*
- * osd ops
- */
-#define CEPH_OSD_OP_MODE       0xf000
-#define CEPH_OSD_OP_MODE_RD    0x1000
-#define CEPH_OSD_OP_MODE_WR    0x2000
-#define CEPH_OSD_OP_MODE_RMW   0x3000
-#define CEPH_OSD_OP_MODE_SUB   0x4000
-
-#define CEPH_OSD_OP_TYPE       0x0f00
-#define CEPH_OSD_OP_TYPE_LOCK  0x0100
-#define CEPH_OSD_OP_TYPE_DATA  0x0200
-#define CEPH_OSD_OP_TYPE_ATTR  0x0300
-#define CEPH_OSD_OP_TYPE_EXEC  0x0400
-#define CEPH_OSD_OP_TYPE_PG    0x0500
-
-enum {
-	/** data **/
-	/* read */
-	CEPH_OSD_OP_READ      = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 1,
-	CEPH_OSD_OP_STAT      = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 2,
-
-	/* fancy read */
-	CEPH_OSD_OP_MASKTRUNC = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 4,
-
-	/* write */
-	CEPH_OSD_OP_WRITE     = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 1,
-	CEPH_OSD_OP_WRITEFULL = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 2,
-	CEPH_OSD_OP_TRUNCATE  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 3,
-	CEPH_OSD_OP_ZERO      = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 4,
-	CEPH_OSD_OP_DELETE    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 5,
-
-	/* fancy write */
-	CEPH_OSD_OP_APPEND    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 6,
-	CEPH_OSD_OP_STARTSYNC = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 7,
-	CEPH_OSD_OP_SETTRUNC  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 8,
-	CEPH_OSD_OP_TRIMTRUNC = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 9,
-
-	CEPH_OSD_OP_TMAPUP  = CEPH_OSD_OP_MODE_RMW | CEPH_OSD_OP_TYPE_DATA | 10,
-	CEPH_OSD_OP_TMAPPUT = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 11,
-	CEPH_OSD_OP_TMAPGET = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 12,
-
-	CEPH_OSD_OP_CREATE  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 13,
-	CEPH_OSD_OP_ROLLBACK= CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 14,
-
-	/** attrs **/
-	/* read */
-	CEPH_OSD_OP_GETXATTR  = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 1,
-	CEPH_OSD_OP_GETXATTRS = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 2,
-	CEPH_OSD_OP_CMPXATTR  = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 3,
-
-	/* write */
-	CEPH_OSD_OP_SETXATTR  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 1,
-	CEPH_OSD_OP_SETXATTRS = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 2,
-	CEPH_OSD_OP_RESETXATTRS = CEPH_OSD_OP_MODE_WR|CEPH_OSD_OP_TYPE_ATTR | 3,
-	CEPH_OSD_OP_RMXATTR   = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 4,
-
-	/** subop **/
-	CEPH_OSD_OP_PULL           = CEPH_OSD_OP_MODE_SUB | 1,
-	CEPH_OSD_OP_PUSH           = CEPH_OSD_OP_MODE_SUB | 2,
-	CEPH_OSD_OP_BALANCEREADS   = CEPH_OSD_OP_MODE_SUB | 3,
-	CEPH_OSD_OP_UNBALANCEREADS = CEPH_OSD_OP_MODE_SUB | 4,
-	CEPH_OSD_OP_SCRUB          = CEPH_OSD_OP_MODE_SUB | 5,
-
-	/** lock **/
-	CEPH_OSD_OP_WRLOCK    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 1,
-	CEPH_OSD_OP_WRUNLOCK  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 2,
-	CEPH_OSD_OP_RDLOCK    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 3,
-	CEPH_OSD_OP_RDUNLOCK  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 4,
-	CEPH_OSD_OP_UPLOCK    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 5,
-	CEPH_OSD_OP_DNLOCK    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 6,
-
-	/** exec **/
-	CEPH_OSD_OP_CALL    = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_EXEC | 1,
-
-	/** pg **/
-	CEPH_OSD_OP_PGLS      = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_PG | 1,
-};
-
-static inline int ceph_osd_op_type_lock(int op)
-{
-	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_LOCK;
-}
-static inline int ceph_osd_op_type_data(int op)
-{
-	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_DATA;
-}
-static inline int ceph_osd_op_type_attr(int op)
-{
-	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_ATTR;
-}
-static inline int ceph_osd_op_type_exec(int op)
-{
-	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_EXEC;
-}
-static inline int ceph_osd_op_type_pg(int op)
-{
-	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_PG;
-}
-
-static inline int ceph_osd_op_mode_subop(int op)
-{
-	return (op & CEPH_OSD_OP_MODE) == CEPH_OSD_OP_MODE_SUB;
-}
-static inline int ceph_osd_op_mode_read(int op)
-{
-	return (op & CEPH_OSD_OP_MODE) == CEPH_OSD_OP_MODE_RD;
-}
-static inline int ceph_osd_op_mode_modify(int op)
-{
-	return (op & CEPH_OSD_OP_MODE) == CEPH_OSD_OP_MODE_WR;
-}
-
-/*
- * note that the following tmap stuff is also defined in the ceph librados.h
- * any modification here needs to be updated there
- */
-#define CEPH_OSD_TMAP_HDR 'h'
-#define CEPH_OSD_TMAP_SET 's'
-#define CEPH_OSD_TMAP_RM  'r'
-
-extern const char *ceph_osd_op_name(int op);
-
-
-/*
- * osd op flags
- *
- * An op may be READ, WRITE, or READ|WRITE.
- */
-enum {
-	CEPH_OSD_FLAG_ACK = 1,          /* want (or is) "ack" ack */
-	CEPH_OSD_FLAG_ONNVRAM = 2,      /* want (or is) "onnvram" ack */
-	CEPH_OSD_FLAG_ONDISK = 4,       /* want (or is) "ondisk" ack */
-	CEPH_OSD_FLAG_RETRY = 8,        /* resend attempt */
-	CEPH_OSD_FLAG_READ = 16,        /* op may read */
-	CEPH_OSD_FLAG_WRITE = 32,       /* op may write */
-	CEPH_OSD_FLAG_ORDERSNAP = 64,   /* EOLDSNAP if snapc is out of order */
-	CEPH_OSD_FLAG_PEERSTAT = 128,   /* msg includes osd_peer_stat */
-	CEPH_OSD_FLAG_BALANCE_READS = 256,
-	CEPH_OSD_FLAG_PARALLELEXEC = 512, /* execute op in parallel */
-	CEPH_OSD_FLAG_PGOP = 1024,      /* pg op, no object */
-	CEPH_OSD_FLAG_EXEC = 2048,      /* op may exec */
-	CEPH_OSD_FLAG_EXEC_PUBLIC = 4096, /* op may exec (public) */
-};
-
-enum {
-	CEPH_OSD_OP_FLAG_EXCL = 1,      /* EXCL object create */
-};
-
-#define EOLDSNAPC    ERESTART  /* ORDERSNAP flag set; writer has old snapc*/
-#define EBLACKLISTED ESHUTDOWN /* blacklisted */
-
-/* xattr comparison */
-enum {
-	CEPH_OSD_CMPXATTR_OP_NOP = 0,
-	CEPH_OSD_CMPXATTR_OP_EQ  = 1,
-	CEPH_OSD_CMPXATTR_OP_NE  = 2,
-	CEPH_OSD_CMPXATTR_OP_GT  = 3,
-	CEPH_OSD_CMPXATTR_OP_GTE = 4,
-	CEPH_OSD_CMPXATTR_OP_LT  = 5,
-	CEPH_OSD_CMPXATTR_OP_LTE = 6
-};
-
-enum {
-	CEPH_OSD_CMPXATTR_MODE_STRING = 1,
-	CEPH_OSD_CMPXATTR_MODE_U64    = 2
-};
-
-/*
- * an individual object operation.  each may be accompanied by some data
- * payload
- */
-struct ceph_osd_op {
-	__le16 op;           /* CEPH_OSD_OP_* */
-	__le32 flags;        /* CEPH_OSD_FLAG_* */
-	union {
-		struct {
-			__le64 offset, length;
-			__le64 truncate_size;
-			__le32 truncate_seq;
-		} __attribute__ ((packed)) extent;
-		struct {
-			__le32 name_len;
-			__le32 value_len;
-			__u8 cmp_op;       /* CEPH_OSD_CMPXATTR_OP_* */
-			__u8 cmp_mode;     /* CEPH_OSD_CMPXATTR_MODE_* */
-		} __attribute__ ((packed)) xattr;
-		struct {
-			__u8 class_len;
-			__u8 method_len;
-			__u8 argc;
-			__le32 indata_len;
-		} __attribute__ ((packed)) cls;
-		struct {
-			__le64 cookie, count;
-		} __attribute__ ((packed)) pgls;
-	        struct {
-		        __le64 snapid;
-	        } __attribute__ ((packed)) snap;
-	};
-	__le32 payload_len;
-} __attribute__ ((packed));
-
-/*
- * osd request message header.  each request may include multiple
- * ceph_osd_op object operations.
- */
-struct ceph_osd_request_head {
-	__le32 client_inc;                 /* client incarnation */
-	struct ceph_object_layout layout;  /* pgid */
-	__le32 osdmap_epoch;               /* client's osdmap epoch */
-
-	__le32 flags;
-
-	struct ceph_timespec mtime;        /* for mutations only */
-	struct ceph_eversion reassert_version; /* if we are replaying op */
-
-	__le32 object_len;     /* length of object name */
-
-	__le64 snapid;         /* snapid to read */
-	__le64 snap_seq;       /* writer's snap context */
-	__le32 num_snaps;
-
-	__le16 num_ops;
-	struct ceph_osd_op ops[];  /* followed by ops[], obj, ticket, snaps */
-} __attribute__ ((packed));
-
-struct ceph_osd_reply_head {
-	__le32 client_inc;                /* client incarnation */
-	__le32 flags;
-	struct ceph_object_layout layout;
-	__le32 osdmap_epoch;
-	struct ceph_eversion reassert_version; /* for replaying uncommitted */
-
-	__le32 result;                    /* result code */
-
-	__le32 object_len;                /* length of object name */
-	__le32 num_ops;
-	struct ceph_osd_op ops[0];  /* ops[], object */
-} __attribute__ ((packed));
-
-
-#endif
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index 190b6c4a6f2b..39c243acd062 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -1,10 +1,12 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/sort.h>
 #include <linux/slab.h>
 
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+
+#include <linux/ceph/decode.h>
 
 /*
  * Snapshots in ceph are driven in large part by cooperation from the
@@ -526,7 +528,7 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
 			    struct ceph_cap_snap *capsnap)
 {
 	struct inode *inode = &ci->vfs_inode;
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 
 	BUG_ON(capsnap->writing);
 	capsnap->size = inode->i_size;
@@ -747,7 +749,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
 		      struct ceph_mds_session *session,
 		      struct ceph_msg *msg)
 {
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	int mds = session->s_mds;
 	u64 split;
 	int op;
diff --git a/fs/ceph/strings.c b/fs/ceph/strings.c
new file mode 100644
index 000000000000..cd5097d7c804
--- /dev/null
+++ b/fs/ceph/strings.c
@@ -0,0 +1,117 @@
+/*
+ * Ceph fs string constants
+ */
+#include <linux/module.h>
+#include <linux/ceph/types.h>
+
+
+const char *ceph_mds_state_name(int s)
+{
+	switch (s) {
+		/* down and out */
+	case CEPH_MDS_STATE_DNE:        return "down:dne";
+	case CEPH_MDS_STATE_STOPPED:    return "down:stopped";
+		/* up and out */
+	case CEPH_MDS_STATE_BOOT:       return "up:boot";
+	case CEPH_MDS_STATE_STANDBY:    return "up:standby";
+	case CEPH_MDS_STATE_STANDBY_REPLAY:    return "up:standby-replay";
+	case CEPH_MDS_STATE_CREATING:   return "up:creating";
+	case CEPH_MDS_STATE_STARTING:   return "up:starting";
+		/* up and in */
+	case CEPH_MDS_STATE_REPLAY:     return "up:replay";
+	case CEPH_MDS_STATE_RESOLVE:    return "up:resolve";
+	case CEPH_MDS_STATE_RECONNECT:  return "up:reconnect";
+	case CEPH_MDS_STATE_REJOIN:     return "up:rejoin";
+	case CEPH_MDS_STATE_CLIENTREPLAY: return "up:clientreplay";
+	case CEPH_MDS_STATE_ACTIVE:     return "up:active";
+	case CEPH_MDS_STATE_STOPPING:   return "up:stopping";
+	}
+	return "???";
+}
+
+const char *ceph_session_op_name(int op)
+{
+	switch (op) {
+	case CEPH_SESSION_REQUEST_OPEN: return "request_open";
+	case CEPH_SESSION_OPEN: return "open";
+	case CEPH_SESSION_REQUEST_CLOSE: return "request_close";
+	case CEPH_SESSION_CLOSE: return "close";
+	case CEPH_SESSION_REQUEST_RENEWCAPS: return "request_renewcaps";
+	case CEPH_SESSION_RENEWCAPS: return "renewcaps";
+	case CEPH_SESSION_STALE: return "stale";
+	case CEPH_SESSION_RECALL_STATE: return "recall_state";
+	}
+	return "???";
+}
+
+const char *ceph_mds_op_name(int op)
+{
+	switch (op) {
+	case CEPH_MDS_OP_LOOKUP:  return "lookup";
+	case CEPH_MDS_OP_LOOKUPHASH:  return "lookuphash";
+	case CEPH_MDS_OP_LOOKUPPARENT:  return "lookupparent";
+	case CEPH_MDS_OP_GETATTR:  return "getattr";
+	case CEPH_MDS_OP_SETXATTR: return "setxattr";
+	case CEPH_MDS_OP_SETATTR: return "setattr";
+	case CEPH_MDS_OP_RMXATTR: return "rmxattr";
+	case CEPH_MDS_OP_READDIR: return "readdir";
+	case CEPH_MDS_OP_MKNOD: return "mknod";
+	case CEPH_MDS_OP_LINK: return "link";
+	case CEPH_MDS_OP_UNLINK: return "unlink";
+	case CEPH_MDS_OP_RENAME: return "rename";
+	case CEPH_MDS_OP_MKDIR: return "mkdir";
+	case CEPH_MDS_OP_RMDIR: return "rmdir";
+	case CEPH_MDS_OP_SYMLINK: return "symlink";
+	case CEPH_MDS_OP_CREATE: return "create";
+	case CEPH_MDS_OP_OPEN: return "open";
+	case CEPH_MDS_OP_LOOKUPSNAP: return "lookupsnap";
+	case CEPH_MDS_OP_LSSNAP: return "lssnap";
+	case CEPH_MDS_OP_MKSNAP: return "mksnap";
+	case CEPH_MDS_OP_RMSNAP: return "rmsnap";
+	case CEPH_MDS_OP_SETFILELOCK: return "setfilelock";
+	case CEPH_MDS_OP_GETFILELOCK: return "getfilelock";
+	}
+	return "???";
+}
+
+const char *ceph_cap_op_name(int op)
+{
+	switch (op) {
+	case CEPH_CAP_OP_GRANT: return "grant";
+	case CEPH_CAP_OP_REVOKE: return "revoke";
+	case CEPH_CAP_OP_TRUNC: return "trunc";
+	case CEPH_CAP_OP_EXPORT: return "export";
+	case CEPH_CAP_OP_IMPORT: return "import";
+	case CEPH_CAP_OP_UPDATE: return "update";
+	case CEPH_CAP_OP_DROP: return "drop";
+	case CEPH_CAP_OP_FLUSH: return "flush";
+	case CEPH_CAP_OP_FLUSH_ACK: return "flush_ack";
+	case CEPH_CAP_OP_FLUSHSNAP: return "flushsnap";
+	case CEPH_CAP_OP_FLUSHSNAP_ACK: return "flushsnap_ack";
+	case CEPH_CAP_OP_RELEASE: return "release";
+	case CEPH_CAP_OP_RENEW: return "renew";
+	}
+	return "???";
+}
+
+const char *ceph_lease_op_name(int o)
+{
+	switch (o) {
+	case CEPH_MDS_LEASE_REVOKE: return "revoke";
+	case CEPH_MDS_LEASE_RELEASE: return "release";
+	case CEPH_MDS_LEASE_RENEW: return "renew";
+	case CEPH_MDS_LEASE_REVOKE_ACK: return "revoke_ack";
+	}
+	return "???";
+}
+
+const char *ceph_snap_op_name(int o)
+{
+	switch (o) {
+	case CEPH_SNAP_OP_UPDATE: return "update";
+	case CEPH_SNAP_OP_CREATE: return "create";
+	case CEPH_SNAP_OP_DESTROY: return "destroy";
+	case CEPH_SNAP_OP_SPLIT: return "split";
+	}
+	return "???";
+}
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index 9922628532b2..d6e0e0421891 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -1,5 +1,5 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/backing-dev.h>
 #include <linux/ctype.h>
@@ -15,10 +15,13 @@
 #include <linux/statfs.h>
 #include <linux/string.h>
 
-#include "decode.h"
 #include "super.h"
-#include "mon_client.h"
-#include "auth.h"
+#include "mds_client.h"
+
+#include <linux/ceph/decode.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
 
 /*
  * Ceph superblock operations
@@ -26,36 +29,22 @@
  * Handle the basics of mounting, unmounting.
  */
 
-
-/*
- * find filename portion of a path (/foo/bar/baz -> baz)
- */
-const char *ceph_file_part(const char *s, int len)
-{
-	const char *e = s + len;
-
-	while (e != s && *(e-1) != '/')
-		e--;
-	return e;
-}
-
-
 /*
  * super ops
  */
 static void ceph_put_super(struct super_block *s)
 {
-	struct ceph_client *client = ceph_sb_to_client(s);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(s);
 
 	dout("put_super\n");
-	ceph_mdsc_close_sessions(&client->mdsc);
+	ceph_mdsc_close_sessions(fsc->mdsc);
 
 	/*
 	 * ensure we release the bdi before put_anon_super releases
 	 * the device name.
 	 */
-	if (s->s_bdi == &client->backing_dev_info) {
-		bdi_unregister(&client->backing_dev_info);
+	if (s->s_bdi == &fsc->backing_dev_info) {
+		bdi_unregister(&fsc->backing_dev_info);
 		s->s_bdi = NULL;
 	}
 
@@ -64,14 +53,14 @@ static void ceph_put_super(struct super_block *s)
 
 static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
-	struct ceph_client *client = ceph_inode_to_client(dentry->d_inode);
-	struct ceph_monmap *monmap = client->monc.monmap;
+	struct ceph_fs_client *fsc = ceph_inode_to_client(dentry->d_inode);
+	struct ceph_monmap *monmap = fsc->client->monc.monmap;
 	struct ceph_statfs st;
 	u64 fsid;
 	int err;
 
 	dout("statfs\n");
-	err = ceph_monc_do_statfs(&client->monc, &st);
+	err = ceph_monc_do_statfs(&fsc->client->monc, &st);
 	if (err < 0)
 		return err;
 
@@ -104,238 +93,28 @@ static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf)
 
 static int ceph_sync_fs(struct super_block *sb, int wait)
 {
-	struct ceph_client *client = ceph_sb_to_client(sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
 
 	if (!wait) {
 		dout("sync_fs (non-blocking)\n");
-		ceph_flush_dirty_caps(&client->mdsc);
+		ceph_flush_dirty_caps(fsc->mdsc);
 		dout("sync_fs (non-blocking) done\n");
 		return 0;
 	}
 
 	dout("sync_fs (blocking)\n");
-	ceph_osdc_sync(&ceph_sb_to_client(sb)->osdc);
-	ceph_mdsc_sync(&ceph_sb_to_client(sb)->mdsc);
+	ceph_osdc_sync(&fsc->client->osdc);
+	ceph_mdsc_sync(fsc->mdsc);
 	dout("sync_fs (blocking) done\n");
 	return 0;
 }
 
-static int default_congestion_kb(void)
-{
-	int congestion_kb;
-
-	/*
-	 * Copied from NFS
-	 *
-	 * congestion size, scale with available memory.
-	 *
-	 *  64MB:    8192k
-	 * 128MB:   11585k
-	 * 256MB:   16384k
-	 * 512MB:   23170k
-	 *   1GB:   32768k
-	 *   2GB:   46340k
-	 *   4GB:   65536k
-	 *   8GB:   92681k
-	 *  16GB:  131072k
-	 *
-	 * This allows larger machines to have larger/more transfers.
-	 * Limit the default to 256M
-	 */
-	congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
-	if (congestion_kb > 256*1024)
-		congestion_kb = 256*1024;
-
-	return congestion_kb;
-}
-
-/**
- * ceph_show_options - Show mount options in /proc/mounts
- * @m: seq_file to write to
- * @mnt: mount descriptor
- */
-static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt)
-{
-	struct ceph_client *client = ceph_sb_to_client(mnt->mnt_sb);
-	struct ceph_mount_args *args = client->mount_args;
-
-	if (args->flags & CEPH_OPT_FSID)
-		seq_printf(m, ",fsid=%pU", &args->fsid);
-	if (args->flags & CEPH_OPT_NOSHARE)
-		seq_puts(m, ",noshare");
-	if (args->flags & CEPH_OPT_DIRSTAT)
-		seq_puts(m, ",dirstat");
-	if ((args->flags & CEPH_OPT_RBYTES) == 0)
-		seq_puts(m, ",norbytes");
-	if (args->flags & CEPH_OPT_NOCRC)
-		seq_puts(m, ",nocrc");
-	if (args->flags & CEPH_OPT_NOASYNCREADDIR)
-		seq_puts(m, ",noasyncreaddir");
-
-	if (args->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT)
-		seq_printf(m, ",mount_timeout=%d", args->mount_timeout);
-	if (args->osd_idle_ttl != CEPH_OSD_IDLE_TTL_DEFAULT)
-		seq_printf(m, ",osd_idle_ttl=%d", args->osd_idle_ttl);
-	if (args->osd_timeout != CEPH_OSD_TIMEOUT_DEFAULT)
-		seq_printf(m, ",osdtimeout=%d", args->osd_timeout);
-	if (args->osd_keepalive_timeout != CEPH_OSD_KEEPALIVE_DEFAULT)
-		seq_printf(m, ",osdkeepalivetimeout=%d",
-			 args->osd_keepalive_timeout);
-	if (args->wsize)
-		seq_printf(m, ",wsize=%d", args->wsize);
-	if (args->rsize != CEPH_MOUNT_RSIZE_DEFAULT)
-		seq_printf(m, ",rsize=%d", args->rsize);
-	if (args->congestion_kb != default_congestion_kb())
-		seq_printf(m, ",write_congestion_kb=%d", args->congestion_kb);
-	if (args->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT)
-		seq_printf(m, ",caps_wanted_delay_min=%d",
-			 args->caps_wanted_delay_min);
-	if (args->caps_wanted_delay_max != CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT)
-		seq_printf(m, ",caps_wanted_delay_max=%d",
-			   args->caps_wanted_delay_max);
-	if (args->cap_release_safety != CEPH_CAP_RELEASE_SAFETY_DEFAULT)
-		seq_printf(m, ",cap_release_safety=%d",
-			   args->cap_release_safety);
-	if (args->max_readdir != CEPH_MAX_READDIR_DEFAULT)
-		seq_printf(m, ",readdir_max_entries=%d", args->max_readdir);
-	if (args->max_readdir_bytes != CEPH_MAX_READDIR_BYTES_DEFAULT)
-		seq_printf(m, ",readdir_max_bytes=%d", args->max_readdir_bytes);
-	if (strcmp(args->snapdir_name, CEPH_SNAPDIRNAME_DEFAULT))
-		seq_printf(m, ",snapdirname=%s", args->snapdir_name);
-	if (args->name)
-		seq_printf(m, ",name=%s", args->name);
-	if (args->secret)
-		seq_puts(m, ",secret=<hidden>");
-	return 0;
-}
-
-/*
- * caches
- */
-struct kmem_cache *ceph_inode_cachep;
-struct kmem_cache *ceph_cap_cachep;
-struct kmem_cache *ceph_dentry_cachep;
-struct kmem_cache *ceph_file_cachep;
-
-static void ceph_inode_init_once(void *foo)
-{
-	struct ceph_inode_info *ci = foo;
-	inode_init_once(&ci->vfs_inode);
-}
-
-static int __init init_caches(void)
-{
-	ceph_inode_cachep = kmem_cache_create("ceph_inode_info",
-				      sizeof(struct ceph_inode_info),
-				      __alignof__(struct ceph_inode_info),
-				      (SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD),
-				      ceph_inode_init_once);
-	if (ceph_inode_cachep == NULL)
-		return -ENOMEM;
-
-	ceph_cap_cachep = KMEM_CACHE(ceph_cap,
-				     SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
-	if (ceph_cap_cachep == NULL)
-		goto bad_cap;
-
-	ceph_dentry_cachep = KMEM_CACHE(ceph_dentry_info,
-					SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
-	if (ceph_dentry_cachep == NULL)
-		goto bad_dentry;
-
-	ceph_file_cachep = KMEM_CACHE(ceph_file_info,
-				      SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
-	if (ceph_file_cachep == NULL)
-		goto bad_file;
-
-	return 0;
-
-bad_file:
-	kmem_cache_destroy(ceph_dentry_cachep);
-bad_dentry:
-	kmem_cache_destroy(ceph_cap_cachep);
-bad_cap:
-	kmem_cache_destroy(ceph_inode_cachep);
-	return -ENOMEM;
-}
-
-static void destroy_caches(void)
-{
-	kmem_cache_destroy(ceph_inode_cachep);
-	kmem_cache_destroy(ceph_cap_cachep);
-	kmem_cache_destroy(ceph_dentry_cachep);
-	kmem_cache_destroy(ceph_file_cachep);
-}
-
-
-/*
- * ceph_umount_begin - initiate forced umount.  Tear down down the
- * mount, skipping steps that may hang while waiting for server(s).
- */
-static void ceph_umount_begin(struct super_block *sb)
-{
-	struct ceph_client *client = ceph_sb_to_client(sb);
-
-	dout("ceph_umount_begin - starting forced umount\n");
-	if (!client)
-		return;
-	client->mount_state = CEPH_MOUNT_SHUTDOWN;
-	return;
-}
-
-static const struct super_operations ceph_super_ops = {
-	.alloc_inode	= ceph_alloc_inode,
-	.destroy_inode	= ceph_destroy_inode,
-	.write_inode    = ceph_write_inode,
-	.sync_fs        = ceph_sync_fs,
-	.put_super	= ceph_put_super,
-	.show_options   = ceph_show_options,
-	.statfs		= ceph_statfs,
-	.umount_begin   = ceph_umount_begin,
-};
-
-
-const char *ceph_msg_type_name(int type)
-{
-	switch (type) {
-	case CEPH_MSG_SHUTDOWN: return "shutdown";
-	case CEPH_MSG_PING: return "ping";
-	case CEPH_MSG_AUTH: return "auth";
-	case CEPH_MSG_AUTH_REPLY: return "auth_reply";
-	case CEPH_MSG_MON_MAP: return "mon_map";
-	case CEPH_MSG_MON_GET_MAP: return "mon_get_map";
-	case CEPH_MSG_MON_SUBSCRIBE: return "mon_subscribe";
-	case CEPH_MSG_MON_SUBSCRIBE_ACK: return "mon_subscribe_ack";
-	case CEPH_MSG_STATFS: return "statfs";
-	case CEPH_MSG_STATFS_REPLY: return "statfs_reply";
-	case CEPH_MSG_MDS_MAP: return "mds_map";
-	case CEPH_MSG_CLIENT_SESSION: return "client_session";
-	case CEPH_MSG_CLIENT_RECONNECT: return "client_reconnect";
-	case CEPH_MSG_CLIENT_REQUEST: return "client_request";
-	case CEPH_MSG_CLIENT_REQUEST_FORWARD: return "client_request_forward";
-	case CEPH_MSG_CLIENT_REPLY: return "client_reply";
-	case CEPH_MSG_CLIENT_CAPS: return "client_caps";
-	case CEPH_MSG_CLIENT_CAPRELEASE: return "client_cap_release";
-	case CEPH_MSG_CLIENT_SNAP: return "client_snap";
-	case CEPH_MSG_CLIENT_LEASE: return "client_lease";
-	case CEPH_MSG_OSD_MAP: return "osd_map";
-	case CEPH_MSG_OSD_OP: return "osd_op";
-	case CEPH_MSG_OSD_OPREPLY: return "osd_opreply";
-	default: return "unknown";
-	}
-}
-
-
 /*
  * mount options
  */
 enum {
 	Opt_wsize,
 	Opt_rsize,
-	Opt_osdtimeout,
-	Opt_osdkeepalivetimeout,
-	Opt_mount_timeout,
-	Opt_osd_idle_ttl,
 	Opt_caps_wanted_delay_min,
 	Opt_caps_wanted_delay_max,
 	Opt_cap_release_safety,
@@ -344,29 +123,19 @@ enum {
 	Opt_congestion_kb,
 	Opt_last_int,
 	/* int args above */
-	Opt_fsid,
 	Opt_snapdirname,
-	Opt_name,
-	Opt_secret,
 	Opt_last_string,
 	/* string args above */
-	Opt_ip,
-	Opt_noshare,
 	Opt_dirstat,
 	Opt_nodirstat,
 	Opt_rbytes,
 	Opt_norbytes,
-	Opt_nocrc,
 	Opt_noasyncreaddir,
 };
 
-static match_table_t arg_tokens = {
+static match_table_t fsopt_tokens = {
 	{Opt_wsize, "wsize=%d"},
 	{Opt_rsize, "rsize=%d"},
-	{Opt_osdtimeout, "osdtimeout=%d"},
-	{Opt_osdkeepalivetimeout, "osdkeepalive=%d"},
-	{Opt_mount_timeout, "mount_timeout=%d"},
-	{Opt_osd_idle_ttl, "osd_idle_ttl=%d"},
 	{Opt_caps_wanted_delay_min, "caps_wanted_delay_min=%d"},
 	{Opt_caps_wanted_delay_max, "caps_wanted_delay_max=%d"},
 	{Opt_cap_release_safety, "cap_release_safety=%d"},
@@ -374,403 +143,459 @@ static match_table_t arg_tokens = {
 	{Opt_readdir_max_bytes, "readdir_max_bytes=%d"},
 	{Opt_congestion_kb, "write_congestion_kb=%d"},
 	/* int args above */
-	{Opt_fsid, "fsid=%s"},
 	{Opt_snapdirname, "snapdirname=%s"},
-	{Opt_name, "name=%s"},
-	{Opt_secret, "secret=%s"},
 	/* string args above */
-	{Opt_ip, "ip=%s"},
-	{Opt_noshare, "noshare"},
 	{Opt_dirstat, "dirstat"},
 	{Opt_nodirstat, "nodirstat"},
 	{Opt_rbytes, "rbytes"},
 	{Opt_norbytes, "norbytes"},
-	{Opt_nocrc, "nocrc"},
 	{Opt_noasyncreaddir, "noasyncreaddir"},
 	{-1, NULL}
 };
 
-static int parse_fsid(const char *str, struct ceph_fsid *fsid)
+static int parse_fsopt_token(char *c, void *private)
 {
-	int i = 0;
-	char tmp[3];
-	int err = -EINVAL;
-	int d;
-
-	dout("parse_fsid '%s'\n", str);
-	tmp[2] = 0;
-	while (*str && i < 16) {
-		if (ispunct(*str)) {
-			str++;
-			continue;
+	struct ceph_mount_options *fsopt = private;
+	substring_t argstr[MAX_OPT_ARGS];
+	int token, intval, ret;
+
+	token = match_token((char *)c, fsopt_tokens, argstr);
+	if (token < 0)
+		return -EINVAL;
+
+	if (token < Opt_last_int) {
+		ret = match_int(&argstr[0], &intval);
+		if (ret < 0) {
+			pr_err("bad mount option arg (not int) "
+			       "at '%s'\n", c);
+			return ret;
 		}
-		if (!isxdigit(str[0]) || !isxdigit(str[1]))
-			break;
-		tmp[0] = str[0];
-		tmp[1] = str[1];
-		if (sscanf(tmp, "%x", &d) < 1)
-			break;
-		fsid->fsid[i] = d & 0xff;
-		i++;
-		str += 2;
+		dout("got int token %d val %d\n", token, intval);
+	} else if (token > Opt_last_int && token < Opt_last_string) {
+		dout("got string token %d val %s\n", token,
+		     argstr[0].from);
+	} else {
+		dout("got token %d\n", token);
 	}
 
-	if (i == 16)
-		err = 0;
-	dout("parse_fsid ret %d got fsid %pU", err, fsid);
-	return err;
+	switch (token) {
+	case Opt_snapdirname:
+		kfree(fsopt->snapdir_name);
+		fsopt->snapdir_name = kstrndup(argstr[0].from,
+					       argstr[0].to-argstr[0].from,
+					       GFP_KERNEL);
+		if (!fsopt->snapdir_name)
+			return -ENOMEM;
+		break;
+
+		/* misc */
+	case Opt_wsize:
+		fsopt->wsize = intval;
+		break;
+	case Opt_rsize:
+		fsopt->rsize = intval;
+		break;
+	case Opt_caps_wanted_delay_min:
+		fsopt->caps_wanted_delay_min = intval;
+		break;
+	case Opt_caps_wanted_delay_max:
+		fsopt->caps_wanted_delay_max = intval;
+		break;
+	case Opt_readdir_max_entries:
+		fsopt->max_readdir = intval;
+		break;
+	case Opt_readdir_max_bytes:
+		fsopt->max_readdir_bytes = intval;
+		break;
+	case Opt_congestion_kb:
+		fsopt->congestion_kb = intval;
+		break;
+	case Opt_dirstat:
+		fsopt->flags |= CEPH_MOUNT_OPT_DIRSTAT;
+		break;
+	case Opt_nodirstat:
+		fsopt->flags &= ~CEPH_MOUNT_OPT_DIRSTAT;
+		break;
+	case Opt_rbytes:
+		fsopt->flags |= CEPH_MOUNT_OPT_RBYTES;
+		break;
+	case Opt_norbytes:
+		fsopt->flags &= ~CEPH_MOUNT_OPT_RBYTES;
+		break;
+	case Opt_noasyncreaddir:
+		fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR;
+		break;
+	default:
+		BUG_ON(token);
+	}
+	return 0;
 }
 
-static struct ceph_mount_args *parse_mount_args(int flags, char *options,
-						const char *dev_name,
-						const char **path)
+static void destroy_mount_options(struct ceph_mount_options *args)
 {
-	struct ceph_mount_args *args;
-	const char *c;
-	int err = -ENOMEM;
-	substring_t argstr[MAX_OPT_ARGS];
+	dout("destroy_mount_options %p\n", args);
+	kfree(args->snapdir_name);
+	kfree(args);
+}
 
-	args = kzalloc(sizeof(*args), GFP_KERNEL);
-	if (!args)
-		return ERR_PTR(-ENOMEM);
-	args->mon_addr = kcalloc(CEPH_MAX_MON, sizeof(*args->mon_addr),
-				 GFP_KERNEL);
-	if (!args->mon_addr)
-		goto out;
+static int strcmp_null(const char *s1, const char *s2)
+{
+	if (!s1 && !s2)
+		return 0;
+	if (s1 && !s2)
+		return -1;
+	if (!s1 && s2)
+		return 1;
+	return strcmp(s1, s2);
+}
 
-	dout("parse_mount_args %p, dev_name '%s'\n", args, dev_name);
-
-	/* start with defaults */
-	args->sb_flags = flags;
-	args->flags = CEPH_OPT_DEFAULT;
-	args->osd_timeout = CEPH_OSD_TIMEOUT_DEFAULT;
-	args->osd_keepalive_timeout = CEPH_OSD_KEEPALIVE_DEFAULT;
-	args->mount_timeout = CEPH_MOUNT_TIMEOUT_DEFAULT; /* seconds */
-	args->osd_idle_ttl = CEPH_OSD_IDLE_TTL_DEFAULT;   /* seconds */
-	args->caps_wanted_delay_min = CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT;
-	args->caps_wanted_delay_max = CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT;
-	args->rsize = CEPH_MOUNT_RSIZE_DEFAULT;
-	args->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
-	args->cap_release_safety = CEPH_CAP_RELEASE_SAFETY_DEFAULT;
-	args->max_readdir = CEPH_MAX_READDIR_DEFAULT;
-	args->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
-	args->congestion_kb = default_congestion_kb();
-
-	/* ip1[:port1][,ip2[:port2]...]:/subdir/in/fs */
-	err = -EINVAL;
-	if (!dev_name)
-		goto out;
-	*path = strstr(dev_name, ":/");
-	if (*path == NULL) {
-		pr_err("device name is missing path (no :/ in %s)\n",
-		       dev_name);
-		goto out;
-	}
+static int compare_mount_options(struct ceph_mount_options *new_fsopt,
+				 struct ceph_options *new_opt,
+				 struct ceph_fs_client *fsc)
+{
+	struct ceph_mount_options *fsopt1 = new_fsopt;
+	struct ceph_mount_options *fsopt2 = fsc->mount_options;
+	int ofs = offsetof(struct ceph_mount_options, snapdir_name);
+	int ret;
 
-	/* get mon ip(s) */
-	err = ceph_parse_ips(dev_name, *path, args->mon_addr,
-			     CEPH_MAX_MON, &args->num_mon);
-	if (err < 0)
-		goto out;
+	ret = memcmp(fsopt1, fsopt2, ofs);
+	if (ret)
+		return ret;
+
+	ret = strcmp_null(fsopt1->snapdir_name, fsopt2->snapdir_name);
+	if (ret)
+		return ret;
+
+	return ceph_compare_options(new_opt, fsc->client);
+}
+
+static int parse_mount_options(struct ceph_mount_options **pfsopt,
+			       struct ceph_options **popt,
+			       int flags, char *options,
+			       const char *dev_name,
+			       const char **path)
+{
+	struct ceph_mount_options *fsopt;
+	const char *dev_name_end;
+	int err = -ENOMEM;
+
+	fsopt = kzalloc(sizeof(*fsopt), GFP_KERNEL);
+	if (!fsopt)
+		return -ENOMEM;
+
+	dout("parse_mount_options %p, dev_name '%s'\n", fsopt, dev_name);
+
+        fsopt->sb_flags = flags;
+        fsopt->flags = CEPH_MOUNT_OPT_DEFAULT;
+
+        fsopt->rsize = CEPH_MOUNT_RSIZE_DEFAULT;
+        fsopt->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
+        fsopt->cap_release_safety = CEPH_CAP_RELEASE_SAFETY_DEFAULT;
+        fsopt->max_readdir = CEPH_MAX_READDIR_DEFAULT;
+        fsopt->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
+        fsopt->congestion_kb = default_congestion_kb();
+	
+        /* ip1[:port1][,ip2[:port2]...]:/subdir/in/fs */
+        err = -EINVAL;
+        if (!dev_name)
+                goto out;
+        *path = strstr(dev_name, ":/");
+        if (*path == NULL) {
+                pr_err("device name is missing path (no :/ in %s)\n",
+                       dev_name);
+                goto out;
+        }
+	dev_name_end = *path;
+	dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
 
 	/* path on server */
 	*path += 2;
 	dout("server path '%s'\n", *path);
 
-	/* parse mount options */
-	while ((c = strsep(&options, ",")) != NULL) {
-		int token, intval, ret;
-		if (!*c)
-			continue;
-		err = -EINVAL;
-		token = match_token((char *)c, arg_tokens, argstr);
-		if (token < 0) {
-			pr_err("bad mount option at '%s'\n", c);
-			goto out;
-		}
-		if (token < Opt_last_int) {
-			ret = match_int(&argstr[0], &intval);
-			if (ret < 0) {
-				pr_err("bad mount option arg (not int) "
-				       "at '%s'\n", c);
-				continue;
-			}
-			dout("got int token %d val %d\n", token, intval);
-		} else if (token > Opt_last_int && token < Opt_last_string) {
-			dout("got string token %d val %s\n", token,
-			     argstr[0].from);
-		} else {
-			dout("got token %d\n", token);
-		}
-		switch (token) {
-		case Opt_ip:
-			err = ceph_parse_ips(argstr[0].from,
-					     argstr[0].to,
-					     &args->my_addr,
-					     1, NULL);
-			if (err < 0)
-				goto out;
-			args->flags |= CEPH_OPT_MYIP;
-			break;
-
-		case Opt_fsid:
-			err = parse_fsid(argstr[0].from, &args->fsid);
-			if (err == 0)
-				args->flags |= CEPH_OPT_FSID;
-			break;
-		case Opt_snapdirname:
-			kfree(args->snapdir_name);
-			args->snapdir_name = kstrndup(argstr[0].from,
-					      argstr[0].to-argstr[0].from,
-					      GFP_KERNEL);
-			break;
-		case Opt_name:
-			args->name = kstrndup(argstr[0].from,
-					      argstr[0].to-argstr[0].from,
-					      GFP_KERNEL);
-			break;
-		case Opt_secret:
-			args->secret = kstrndup(argstr[0].from,
-						argstr[0].to-argstr[0].from,
-						GFP_KERNEL);
-			break;
-
-			/* misc */
-		case Opt_wsize:
-			args->wsize = intval;
-			break;
-		case Opt_rsize:
-			args->rsize = intval;
-			break;
-		case Opt_osdtimeout:
-			args->osd_timeout = intval;
-			break;
-		case Opt_osdkeepalivetimeout:
-			args->osd_keepalive_timeout = intval;
-			break;
-		case Opt_osd_idle_ttl:
-			args->osd_idle_ttl = intval;
-			break;
-		case Opt_mount_timeout:
-			args->mount_timeout = intval;
-			break;
-		case Opt_caps_wanted_delay_min:
-			args->caps_wanted_delay_min = intval;
-			break;
-		case Opt_caps_wanted_delay_max:
-			args->caps_wanted_delay_max = intval;
-			break;
-		case Opt_readdir_max_entries:
-			args->max_readdir = intval;
-			break;
-		case Opt_readdir_max_bytes:
-			args->max_readdir_bytes = intval;
-			break;
-		case Opt_congestion_kb:
-			args->congestion_kb = intval;
-			break;
-
-		case Opt_noshare:
-			args->flags |= CEPH_OPT_NOSHARE;
-			break;
-
-		case Opt_dirstat:
-			args->flags |= CEPH_OPT_DIRSTAT;
-			break;
-		case Opt_nodirstat:
-			args->flags &= ~CEPH_OPT_DIRSTAT;
-			break;
-		case Opt_rbytes:
-			args->flags |= CEPH_OPT_RBYTES;
-			break;
-		case Opt_norbytes:
-			args->flags &= ~CEPH_OPT_RBYTES;
-			break;
-		case Opt_nocrc:
-			args->flags |= CEPH_OPT_NOCRC;
-			break;
-		case Opt_noasyncreaddir:
-			args->flags |= CEPH_OPT_NOASYNCREADDIR;
-			break;
-
-		default:
-			BUG_ON(token);
-		}
-	}
-	return args;
+	err = ceph_parse_options(popt, options, dev_name, dev_name_end,
+				 parse_fsopt_token, (void *)fsopt);
+	if (err)
+		goto out;
+
+	/* success */
+	*pfsopt = fsopt;
+	return 0;
 
 out:
-	kfree(args->mon_addr);
-	kfree(args);
-	return ERR_PTR(err);
+	destroy_mount_options(fsopt);
+	return err;
 }
 
-static void destroy_mount_args(struct ceph_mount_args *args)
+/**
+ * ceph_show_options - Show mount options in /proc/mounts
+ * @m: seq_file to write to
+ * @mnt: mount descriptor
+ */
+static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt)
 {
-	dout("destroy_mount_args %p\n", args);
-	kfree(args->snapdir_name);
-	args->snapdir_name = NULL;
-	kfree(args->name);
-	args->name = NULL;
-	kfree(args->secret);
-	args->secret = NULL;
-	kfree(args);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(mnt->mnt_sb);
+	struct ceph_mount_options *fsopt = fsc->mount_options;
+	struct ceph_options *opt = fsc->client->options;
+
+	if (opt->flags & CEPH_OPT_FSID)
+		seq_printf(m, ",fsid=%pU", &opt->fsid);
+	if (opt->flags & CEPH_OPT_NOSHARE)
+		seq_puts(m, ",noshare");
+	if (opt->flags & CEPH_OPT_NOCRC)
+		seq_puts(m, ",nocrc");
+
+	if (opt->name)
+		seq_printf(m, ",name=%s", opt->name);
+	if (opt->secret)
+		seq_puts(m, ",secret=<hidden>");
+
+	if (opt->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT)
+		seq_printf(m, ",mount_timeout=%d", opt->mount_timeout);
+	if (opt->osd_idle_ttl != CEPH_OSD_IDLE_TTL_DEFAULT)
+		seq_printf(m, ",osd_idle_ttl=%d", opt->osd_idle_ttl);
+	if (opt->osd_timeout != CEPH_OSD_TIMEOUT_DEFAULT)
+		seq_printf(m, ",osdtimeout=%d", opt->osd_timeout);
+	if (opt->osd_keepalive_timeout != CEPH_OSD_KEEPALIVE_DEFAULT)
+		seq_printf(m, ",osdkeepalivetimeout=%d",
+			   opt->osd_keepalive_timeout);
+
+	if (fsopt->flags & CEPH_MOUNT_OPT_DIRSTAT)
+		seq_puts(m, ",dirstat");
+	if ((fsopt->flags & CEPH_MOUNT_OPT_RBYTES) == 0)
+		seq_puts(m, ",norbytes");
+	if (fsopt->flags & CEPH_MOUNT_OPT_NOASYNCREADDIR)
+		seq_puts(m, ",noasyncreaddir");
+
+	if (fsopt->wsize)
+		seq_printf(m, ",wsize=%d", fsopt->wsize);
+	if (fsopt->rsize != CEPH_MOUNT_RSIZE_DEFAULT)
+		seq_printf(m, ",rsize=%d", fsopt->rsize);
+	if (fsopt->congestion_kb != default_congestion_kb())
+		seq_printf(m, ",write_congestion_kb=%d", fsopt->congestion_kb);
+	if (fsopt->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT)
+		seq_printf(m, ",caps_wanted_delay_min=%d",
+			 fsopt->caps_wanted_delay_min);
+	if (fsopt->caps_wanted_delay_max != CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT)
+		seq_printf(m, ",caps_wanted_delay_max=%d",
+			   fsopt->caps_wanted_delay_max);
+	if (fsopt->cap_release_safety != CEPH_CAP_RELEASE_SAFETY_DEFAULT)
+		seq_printf(m, ",cap_release_safety=%d",
+			   fsopt->cap_release_safety);
+	if (fsopt->max_readdir != CEPH_MAX_READDIR_DEFAULT)
+		seq_printf(m, ",readdir_max_entries=%d", fsopt->max_readdir);
+	if (fsopt->max_readdir_bytes != CEPH_MAX_READDIR_BYTES_DEFAULT)
+		seq_printf(m, ",readdir_max_bytes=%d", fsopt->max_readdir_bytes);
+	if (strcmp(fsopt->snapdir_name, CEPH_SNAPDIRNAME_DEFAULT))
+		seq_printf(m, ",snapdirname=%s", fsopt->snapdir_name);
+	return 0;
 }
 
 /*
- * create a fresh client instance
+ * handle any mon messages the standard library doesn't understand.
+ * return error if we don't either.
  */
-static struct ceph_client *ceph_create_client(struct ceph_mount_args *args)
+static int extra_mon_dispatch(struct ceph_client *client, struct ceph_msg *msg)
 {
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc = client->private;
+	int type = le16_to_cpu(msg->hdr.type);
+
+	switch (type) {
+	case CEPH_MSG_MDS_MAP:
+		ceph_mdsc_handle_map(fsc->mdsc, msg);
+		return 0;
+
+	default:
+		return -1;
+	}
+}
+
+/*
+ * create a new fs client
+ */
+struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
+					struct ceph_options *opt)
+{
+	struct ceph_fs_client *fsc;
 	int err = -ENOMEM;
 
-	client = kzalloc(sizeof(*client), GFP_KERNEL);
-	if (client == NULL)
+	fsc = kzalloc(sizeof(*fsc), GFP_KERNEL);
+	if (!fsc)
 		return ERR_PTR(-ENOMEM);
 
-	mutex_init(&client->mount_mutex);
-
-	init_waitqueue_head(&client->auth_wq);
+	fsc->client = ceph_create_client(opt, fsc);
+	if (IS_ERR(fsc->client)) {
+		err = PTR_ERR(fsc->client);
+		goto fail;
+	}
+	fsc->client->extra_mon_dispatch = extra_mon_dispatch;
+	fsc->client->supported_features |= CEPH_FEATURE_FLOCK;
+	fsc->client->monc.want_mdsmap = 1;
 
-	client->sb = NULL;
-	client->mount_state = CEPH_MOUNT_MOUNTING;
-	client->mount_args = args;
+	fsc->mount_options = fsopt;
 
-	client->msgr = NULL;
+	fsc->sb = NULL;
+	fsc->mount_state = CEPH_MOUNT_MOUNTING;
 
-	client->auth_err = 0;
-	atomic_long_set(&client->writeback_count, 0);
+	atomic_long_set(&fsc->writeback_count, 0);
 
-	err = bdi_init(&client->backing_dev_info);
+	err = bdi_init(&fsc->backing_dev_info);
 	if (err < 0)
-		goto fail;
+		goto fail_client;
 
 	err = -ENOMEM;
-	client->wb_wq = create_workqueue("ceph-writeback");
-	if (client->wb_wq == NULL)
+	fsc->wb_wq = create_workqueue("ceph-writeback");
+	if (fsc->wb_wq == NULL)
 		goto fail_bdi;
-	client->pg_inv_wq = create_singlethread_workqueue("ceph-pg-invalid");
-	if (client->pg_inv_wq == NULL)
+	fsc->pg_inv_wq = create_singlethread_workqueue("ceph-pg-invalid");
+	if (fsc->pg_inv_wq == NULL)
 		goto fail_wb_wq;
-	client->trunc_wq = create_singlethread_workqueue("ceph-trunc");
-	if (client->trunc_wq == NULL)
+	fsc->trunc_wq = create_singlethread_workqueue("ceph-trunc");
+	if (fsc->trunc_wq == NULL)
 		goto fail_pg_inv_wq;
 
 	/* set up mempools */
 	err = -ENOMEM;
-	client->wb_pagevec_pool = mempool_create_kmalloc_pool(10,
-			      client->mount_args->wsize >> PAGE_CACHE_SHIFT);
-	if (!client->wb_pagevec_pool)
+	fsc->wb_pagevec_pool = mempool_create_kmalloc_pool(10,
+			      fsc->mount_options->wsize >> PAGE_CACHE_SHIFT);
+	if (!fsc->wb_pagevec_pool)
 		goto fail_trunc_wq;
 
 	/* caps */
-	client->min_caps = args->max_readdir;
+	fsc->min_caps = fsopt->max_readdir;
+
+	return fsc;
 
-	/* subsystems */
-	err = ceph_monc_init(&client->monc, client);
-	if (err < 0)
-		goto fail_mempool;
-	err = ceph_osdc_init(&client->osdc, client);
-	if (err < 0)
-		goto fail_monc;
-	err = ceph_mdsc_init(&client->mdsc, client);
-	if (err < 0)
-		goto fail_osdc;
-	return client;
-
-fail_osdc:
-	ceph_osdc_stop(&client->osdc);
-fail_monc:
-	ceph_monc_stop(&client->monc);
-fail_mempool:
-	mempool_destroy(client->wb_pagevec_pool);
 fail_trunc_wq:
-	destroy_workqueue(client->trunc_wq);
+	destroy_workqueue(fsc->trunc_wq);
 fail_pg_inv_wq:
-	destroy_workqueue(client->pg_inv_wq);
+	destroy_workqueue(fsc->pg_inv_wq);
 fail_wb_wq:
-	destroy_workqueue(client->wb_wq);
+	destroy_workqueue(fsc->wb_wq);
 fail_bdi:
-	bdi_destroy(&client->backing_dev_info);
+	bdi_destroy(&fsc->backing_dev_info);
+fail_client:
+	ceph_destroy_client(fsc->client);
 fail:
-	kfree(client);
+	kfree(fsc);
 	return ERR_PTR(err);
 }
 
-static void ceph_destroy_client(struct ceph_client *client)
+void destroy_fs_client(struct ceph_fs_client *fsc)
 {
-	dout("destroy_client %p\n", client);
+	dout("destroy_fs_client %p\n", fsc);
 
-	/* unmount */
-	ceph_mdsc_stop(&client->mdsc);
-	ceph_osdc_stop(&client->osdc);
+	destroy_workqueue(fsc->wb_wq);
+	destroy_workqueue(fsc->pg_inv_wq);
+	destroy_workqueue(fsc->trunc_wq);
 
-	/*
-	 * make sure mds and osd connections close out before destroying
-	 * the auth module, which is needed to free those connections'
-	 * ceph_authorizers.
-	 */
-	ceph_msgr_flush();
-
-	ceph_monc_stop(&client->monc);
+	bdi_destroy(&fsc->backing_dev_info);
 
-	ceph_debugfs_client_cleanup(client);
-	destroy_workqueue(client->wb_wq);
-	destroy_workqueue(client->pg_inv_wq);
-	destroy_workqueue(client->trunc_wq);
+	mempool_destroy(fsc->wb_pagevec_pool);
 
-	bdi_destroy(&client->backing_dev_info);
+	destroy_mount_options(fsc->mount_options);
 
-	if (client->msgr)
-		ceph_messenger_destroy(client->msgr);
-	mempool_destroy(client->wb_pagevec_pool);
+	ceph_fs_debugfs_cleanup(fsc);
 
-	destroy_mount_args(client->mount_args);
+	ceph_destroy_client(fsc->client);
 
-	kfree(client);
-	dout("destroy_client %p done\n", client);
+	kfree(fsc);
+	dout("destroy_fs_client %p done\n", fsc);
 }
 
 /*
- * Initially learn our fsid, or verify an fsid matches.
+ * caches
  */
-int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid)
+struct kmem_cache *ceph_inode_cachep;
+struct kmem_cache *ceph_cap_cachep;
+struct kmem_cache *ceph_dentry_cachep;
+struct kmem_cache *ceph_file_cachep;
+
+static void ceph_inode_init_once(void *foo)
 {
-	if (client->have_fsid) {
-		if (ceph_fsid_compare(&client->fsid, fsid)) {
-			pr_err("bad fsid, had %pU got %pU",
-			       &client->fsid, fsid);
-			return -1;
-		}
-	} else {
-		pr_info("client%lld fsid %pU\n", client->monc.auth->global_id,
-			fsid);
-		memcpy(&client->fsid, fsid, sizeof(*fsid));
-		ceph_debugfs_client_init(client);
-		client->have_fsid = true;
-	}
+	struct ceph_inode_info *ci = foo;
+	inode_init_once(&ci->vfs_inode);
+}
+
+static int __init init_caches(void)
+{
+	ceph_inode_cachep = kmem_cache_create("ceph_inode_info",
+				      sizeof(struct ceph_inode_info),
+				      __alignof__(struct ceph_inode_info),
+				      (SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD),
+				      ceph_inode_init_once);
+	if (ceph_inode_cachep == NULL)
+		return -ENOMEM;
+
+	ceph_cap_cachep = KMEM_CACHE(ceph_cap,
+				     SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
+	if (ceph_cap_cachep == NULL)
+		goto bad_cap;
+
+	ceph_dentry_cachep = KMEM_CACHE(ceph_dentry_info,
+					SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
+	if (ceph_dentry_cachep == NULL)
+		goto bad_dentry;
+
+	ceph_file_cachep = KMEM_CACHE(ceph_file_info,
+				      SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
+	if (ceph_file_cachep == NULL)
+		goto bad_file;
+
 	return 0;
+
+bad_file:
+	kmem_cache_destroy(ceph_dentry_cachep);
+bad_dentry:
+	kmem_cache_destroy(ceph_cap_cachep);
+bad_cap:
+	kmem_cache_destroy(ceph_inode_cachep);
+	return -ENOMEM;
 }
 
+static void destroy_caches(void)
+{
+	kmem_cache_destroy(ceph_inode_cachep);
+	kmem_cache_destroy(ceph_cap_cachep);
+	kmem_cache_destroy(ceph_dentry_cachep);
+	kmem_cache_destroy(ceph_file_cachep);
+}
+
+
 /*
- * true if we have the mon map (and have thus joined the cluster)
+ * ceph_umount_begin - initiate forced umount.  Tear down down the
+ * mount, skipping steps that may hang while waiting for server(s).
  */
-static int have_mon_and_osd_map(struct ceph_client *client)
+static void ceph_umount_begin(struct super_block *sb)
 {
-	return client->monc.monmap && client->monc.monmap->epoch &&
-	       client->osdc.osdmap && client->osdc.osdmap->epoch;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
+
+	dout("ceph_umount_begin - starting forced umount\n");
+	if (!fsc)
+		return;
+	fsc->mount_state = CEPH_MOUNT_SHUTDOWN;
+	return;
 }
 
+static const struct super_operations ceph_super_ops = {
+	.alloc_inode	= ceph_alloc_inode,
+	.destroy_inode	= ceph_destroy_inode,
+	.write_inode    = ceph_write_inode,
+	.sync_fs        = ceph_sync_fs,
+	.put_super	= ceph_put_super,
+	.show_options   = ceph_show_options,
+	.statfs		= ceph_statfs,
+	.umount_begin   = ceph_umount_begin,
+};
+
 /*
  * Bootstrap mount by opening the root directory.  Note the mount
  * @started time from caller, and time out if this takes too long.
  */
-static struct dentry *open_root_dentry(struct ceph_client *client,
+static struct dentry *open_root_dentry(struct ceph_fs_client *fsc,
 				       const char *path,
 				       unsigned long started)
 {
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req = NULL;
 	int err;
 	struct dentry *root;
@@ -784,14 +609,14 @@ static struct dentry *open_root_dentry(struct ceph_client *client,
 	req->r_ino1.ino = CEPH_INO_ROOT;
 	req->r_ino1.snap = CEPH_NOSNAP;
 	req->r_started = started;
-	req->r_timeout = client->mount_args->mount_timeout * HZ;
+	req->r_timeout = fsc->client->options->mount_timeout * HZ;
 	req->r_args.getattr.mask = cpu_to_le32(CEPH_STAT_CAP_INODE);
 	req->r_num_caps = 2;
 	err = ceph_mdsc_do_request(mdsc, NULL, req);
 	if (err == 0) {
 		dout("open_root_inode success\n");
 		if (ceph_ino(req->r_target_inode) == CEPH_INO_ROOT &&
-		    client->sb->s_root == NULL)
+		    fsc->sb->s_root == NULL)
 			root = d_alloc_root(req->r_target_inode);
 		else
 			root = d_obtain_alias(req->r_target_inode);
@@ -804,105 +629,86 @@ static struct dentry *open_root_dentry(struct ceph_client *client,
 	return root;
 }
 
+
+
+
 /*
  * mount: join the ceph cluster, and open root directory.
  */
-static int ceph_mount(struct ceph_client *client, struct vfsmount *mnt,
+static int ceph_mount(struct ceph_fs_client *fsc, struct vfsmount *mnt,
 		      const char *path)
 {
-	struct ceph_entity_addr *myaddr = NULL;
 	int err;
-	unsigned long timeout = client->mount_args->mount_timeout * HZ;
 	unsigned long started = jiffies;  /* note the start time */
 	struct dentry *root;
+	int first = 0;   /* first vfsmount for this super_block */
 
 	dout("mount start\n");
-	mutex_lock(&client->mount_mutex);
-
-	/* initialize the messenger */
-	if (client->msgr == NULL) {
-		if (ceph_test_opt(client, MYIP))
-			myaddr = &client->mount_args->my_addr;
-		client->msgr = ceph_messenger_create(myaddr);
-		if (IS_ERR(client->msgr)) {
-			err = PTR_ERR(client->msgr);
-			client->msgr = NULL;
-			goto out;
-		}
-		client->msgr->nocrc = ceph_test_opt(client, NOCRC);
-	}
+	mutex_lock(&fsc->client->mount_mutex);
 
-	/* open session, and wait for mon, mds, and osd maps */
-	err = ceph_monc_open_session(&client->monc);
+	err = __ceph_open_session(fsc->client, started);
 	if (err < 0)
 		goto out;
 
-	while (!have_mon_and_osd_map(client)) {
-		err = -EIO;
-		if (timeout && time_after_eq(jiffies, started + timeout))
-			goto out;
-
-		/* wait */
-		dout("mount waiting for mon_map\n");
-		err = wait_event_interruptible_timeout(client->auth_wq,
-		       have_mon_and_osd_map(client) || (client->auth_err < 0),
-		       timeout);
-		if (err == -EINTR || err == -ERESTARTSYS)
-			goto out;
-		if (client->auth_err < 0) {
-			err = client->auth_err;
-			goto out;
-		}
-	}
-
 	dout("mount opening root\n");
-	root = open_root_dentry(client, "", started);
+	root = open_root_dentry(fsc, "", started);
 	if (IS_ERR(root)) {
 		err = PTR_ERR(root);
 		goto out;
 	}
-	if (client->sb->s_root)
+	if (fsc->sb->s_root) {
 		dput(root);
-	else
-		client->sb->s_root = root;
+	} else {
+		fsc->sb->s_root = root;
+		first = 1;
+
+		err = ceph_fs_debugfs_init(fsc);
+		if (err < 0)
+			goto fail;
+	}
 
 	if (path[0] == 0) {
 		dget(root);
 	} else {
 		dout("mount opening base mountpoint\n");
-		root = open_root_dentry(client, path, started);
+		root = open_root_dentry(fsc, path, started);
 		if (IS_ERR(root)) {
 			err = PTR_ERR(root);
-			dput(client->sb->s_root);
-			client->sb->s_root = NULL;
-			goto out;
+			goto fail;
 		}
 	}
 
 	mnt->mnt_root = root;
-	mnt->mnt_sb = client->sb;
+	mnt->mnt_sb = fsc->sb;
 
-	client->mount_state = CEPH_MOUNT_MOUNTED;
+	fsc->mount_state = CEPH_MOUNT_MOUNTED;
 	dout("mount success\n");
 	err = 0;
 
 out:
-	mutex_unlock(&client->mount_mutex);
+	mutex_unlock(&fsc->client->mount_mutex);
 	return err;
+
+fail:
+	if (first) {
+		dput(fsc->sb->s_root);
+		fsc->sb->s_root = NULL;
+	}
+	goto out;
 }
 
 static int ceph_set_super(struct super_block *s, void *data)
 {
-	struct ceph_client *client = data;
+	struct ceph_fs_client *fsc = data;
 	int ret;
 
 	dout("set_super %p data %p\n", s, data);
 
-	s->s_flags = client->mount_args->sb_flags;
+	s->s_flags = fsc->mount_options->sb_flags;
 	s->s_maxbytes = 1ULL << 40;  /* temp value until we get mdsmap */
 
-	s->s_fs_info = client;
-	client->sb = s;
+	s->s_fs_info = fsc;
+	fsc->sb = s;
 
 	s->s_op = &ceph_super_ops;
 	s->s_export_op = &ceph_export_ops;
@@ -917,7 +723,7 @@ static int ceph_set_super(struct super_block *s, void *data)
 
 fail:
 	s->s_fs_info = NULL;
-	client->sb = NULL;
+	fsc->sb = NULL;
 	return ret;
 }
 
@@ -926,30 +732,23 @@ fail:
  */
 static int ceph_compare_super(struct super_block *sb, void *data)
 {
-	struct ceph_client *new = data;
-	struct ceph_mount_args *args = new->mount_args;
-	struct ceph_client *other = ceph_sb_to_client(sb);
-	int i;
+	struct ceph_fs_client *new = data;
+	struct ceph_mount_options *fsopt = new->mount_options;
+	struct ceph_options *opt = new->client->options;
+	struct ceph_fs_client *other = ceph_sb_to_client(sb);
 
 	dout("ceph_compare_super %p\n", sb);
-	if (args->flags & CEPH_OPT_FSID) {
-		if (ceph_fsid_compare(&args->fsid, &other->fsid)) {
-			dout("fsid doesn't match\n");
-			return 0;
-		}
-	} else {
-		/* do we share (a) monitor? */
-		for (i = 0; i < new->monc.monmap->num_mon; i++)
-			if (ceph_monmap_contains(other->monc.monmap,
-					 &new->monc.monmap->mon_inst[i].addr))
-				break;
-		if (i == new->monc.monmap->num_mon) {
-			dout("mon ip not part of monmap\n");
-			return 0;
-		}
-		dout("mon ip matches existing sb %p\n", sb);
+
+	if (compare_mount_options(fsopt, opt, other)) {
+		dout("monitor(s)/mount options don't match\n");
+		return 0;
 	}
-	if (args->sb_flags != other->mount_args->sb_flags) {
+	if ((opt->flags & CEPH_OPT_FSID) &&
+	    ceph_fsid_compare(&opt->fsid, &other->client->fsid)) {
+		dout("fsid doesn't match\n");
+		return 0;
+	}
+	if (fsopt->sb_flags != other->mount_options->sb_flags) {
 		dout("flags differ\n");
 		return 0;
 	}
@@ -961,19 +760,20 @@ static int ceph_compare_super(struct super_block *sb, void *data)
  */
 static atomic_long_t bdi_seq = ATOMIC_LONG_INIT(0);
 
-static int ceph_register_bdi(struct super_block *sb, struct ceph_client *client)
+static int ceph_register_bdi(struct super_block *sb,
+			     struct ceph_fs_client *fsc)
 {
 	int err;
 
 	/* set ra_pages based on rsize mount option? */
-	if (client->mount_args->rsize >= PAGE_CACHE_SIZE)
-		client->backing_dev_info.ra_pages =
-			(client->mount_args->rsize + PAGE_CACHE_SIZE - 1)
+	if (fsc->mount_options->rsize >= PAGE_CACHE_SIZE)
+		fsc->backing_dev_info.ra_pages =
+			(fsc->mount_options->rsize + PAGE_CACHE_SIZE - 1)
 			>> PAGE_SHIFT;
-	err = bdi_register(&client->backing_dev_info, NULL, "ceph-%d",
+	err = bdi_register(&fsc->backing_dev_info, NULL, "ceph-%d",
 			   atomic_long_inc_return(&bdi_seq));
 	if (!err)
-		sb->s_bdi = &client->backing_dev_info;
+		sb->s_bdi = &fsc->backing_dev_info;
 	return err;
 }
 
@@ -982,46 +782,52 @@ static int ceph_get_sb(struct file_system_type *fs_type,
 		       struct vfsmount *mnt)
 {
 	struct super_block *sb;
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc;
 	int err;
 	int (*compare_super)(struct super_block *, void *) = ceph_compare_super;
 	const char *path = NULL;
-	struct ceph_mount_args *args;
+	struct ceph_mount_options *fsopt = NULL;
+	struct ceph_options *opt = NULL;
 
 	dout("ceph_get_sb\n");
-	args = parse_mount_args(flags, data, dev_name, &path);
-	if (IS_ERR(args)) {
-		err = PTR_ERR(args);
+	err = parse_mount_options(&fsopt, &opt, flags, data, dev_name, &path);
+	if (err < 0)
 		goto out_final;
-	}
 
 	/* create client (which we may/may not use) */
-	client = ceph_create_client(args);
-	if (IS_ERR(client)) {
-		err = PTR_ERR(client);
+	fsc = create_fs_client(fsopt, opt);
+	if (IS_ERR(fsc)) {
+		err = PTR_ERR(fsc);
+		kfree(fsopt);
+		kfree(opt);
 		goto out_final;
 	}
 
-	if (client->mount_args->flags & CEPH_OPT_NOSHARE)
+	err = ceph_mdsc_init(fsc);
+	if (err < 0)
+		goto out;
+
+	if (ceph_test_opt(fsc->client, NOSHARE))
 		compare_super = NULL;
-	sb = sget(fs_type, compare_super, ceph_set_super, client);
+	sb = sget(fs_type, compare_super, ceph_set_super, fsc);
 	if (IS_ERR(sb)) {
 		err = PTR_ERR(sb);
 		goto out;
 	}
 
-	if (ceph_sb_to_client(sb) != client) {
-		ceph_destroy_client(client);
-		client = ceph_sb_to_client(sb);
-		dout("get_sb got existing client %p\n", client);
+	if (ceph_sb_to_client(sb) != fsc) {
+		ceph_mdsc_destroy(fsc);
+		destroy_fs_client(fsc);
+		fsc = ceph_sb_to_client(sb);
+		dout("get_sb got existing client %p\n", fsc);
 	} else {
-		dout("get_sb using new client %p\n", client);
-		err = ceph_register_bdi(sb, client);
+		dout("get_sb using new client %p\n", fsc);
+		err = ceph_register_bdi(sb, fsc);
 		if (err < 0)
 			goto out_splat;
 	}
 
-	err = ceph_mount(client, mnt, path);
+	err = ceph_mount(fsc, mnt, path);
 	if (err < 0)
 		goto out_splat;
 	dout("root %p inode %p ino %llx.%llx\n", mnt->mnt_root,
@@ -1029,12 +835,13 @@ static int ceph_get_sb(struct file_system_type *fs_type,
 	return 0;
 
 out_splat:
-	ceph_mdsc_close_sessions(&client->mdsc);
+	ceph_mdsc_close_sessions(fsc->mdsc);
 	deactivate_locked_super(sb);
 	goto out_final;
 
 out:
-	ceph_destroy_client(client);
+	ceph_mdsc_destroy(fsc);
+	destroy_fs_client(fsc);
 out_final:
 	dout("ceph_get_sb fail %d\n", err);
 	return err;
@@ -1042,11 +849,12 @@ out_final:
 
 static void ceph_kill_sb(struct super_block *s)
 {
-	struct ceph_client *client = ceph_sb_to_client(s);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(s);
 	dout("kill_sb %p\n", s);
-	ceph_mdsc_pre_umount(&client->mdsc);
+	ceph_mdsc_pre_umount(fsc->mdsc);
 	kill_anon_super(s);    /* will call put_super after sb is r/o */
-	ceph_destroy_client(client);
+	ceph_mdsc_destroy(fsc);
+	destroy_fs_client(fsc);
 }
 
 static struct file_system_type ceph_fs_type = {
@@ -1062,36 +870,20 @@ static struct file_system_type ceph_fs_type = {
 
 static int __init init_ceph(void)
 {
-	int ret = 0;
-
-	ret = ceph_debugfs_init();
-	if (ret < 0)
-		goto out;
-
-	ret = ceph_msgr_init();
-	if (ret < 0)
-		goto out_debugfs;
-
-	ret = init_caches();
+	int ret = init_caches();
 	if (ret)
-		goto out_msgr;
+		goto out;
 
 	ret = register_filesystem(&ceph_fs_type);
 	if (ret)
 		goto out_icache;
 
-	pr_info("loaded (mon/mds/osd proto %d/%d/%d, osdmap %d/%d %d/%d)\n",
-		CEPH_MONC_PROTOCOL, CEPH_MDSC_PROTOCOL, CEPH_OSDC_PROTOCOL,
-		CEPH_OSDMAP_VERSION, CEPH_OSDMAP_VERSION_EXT,
-		CEPH_OSDMAP_INC_VERSION, CEPH_OSDMAP_INC_VERSION_EXT);
+	pr_info("loaded (mds proto %d)\n", CEPH_MDSC_PROTOCOL);
+
 	return 0;
 
 out_icache:
 	destroy_caches();
-out_msgr:
-	ceph_msgr_exit();
-out_debugfs:
-	ceph_debugfs_cleanup();
 out:
 	return ret;
 }
@@ -1101,8 +893,6 @@ static void __exit exit_ceph(void)
 	dout("exit_ceph\n");
 	unregister_filesystem(&ceph_fs_type);
 	destroy_caches();
-	ceph_msgr_exit();
-	ceph_debugfs_cleanup();
 }
 
 module_init(init_ceph);
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index b87638e84c4b..e2e904442ce2 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -1,7 +1,7 @@
 #ifndef _FS_CEPH_SUPER_H
 #define _FS_CEPH_SUPER_H
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <asm/unaligned.h>
 #include <linux/backing-dev.h>
@@ -14,13 +14,7 @@
 #include <linux/writeback.h>
 #include <linux/slab.h>
 
-#include "types.h"
-#include "messenger.h"
-#include "msgpool.h"
-#include "mon_client.h"
-#include "mds_client.h"
-#include "osd_client.h"
-#include "ceph_fs.h"
+#include <linux/ceph/libceph.h>
 
 /* f_type in struct statfs */
 #define CEPH_SUPER_MAGIC 0x00c36400
@@ -30,42 +24,25 @@
 #define CEPH_BLOCK_SHIFT   20  /* 1 MB */
 #define CEPH_BLOCK         (1 << CEPH_BLOCK_SHIFT)
 
-/*
- * Supported features
- */
-#define CEPH_FEATURE_SUPPORTED CEPH_FEATURE_NOSRCADDR | CEPH_FEATURE_FLOCK
-#define CEPH_FEATURE_REQUIRED  CEPH_FEATURE_NOSRCADDR
+#define CEPH_MOUNT_OPT_DIRSTAT         (1<<4) /* `cat dirname` for stats */
+#define CEPH_MOUNT_OPT_RBYTES          (1<<5) /* dir st_bytes = rbytes */
+#define CEPH_MOUNT_OPT_NOASYNCREADDIR  (1<<7) /* no dcache readdir */
 
-/*
- * mount options
- */
-#define CEPH_OPT_FSID             (1<<0)
-#define CEPH_OPT_NOSHARE          (1<<1) /* don't share client with other sbs */
-#define CEPH_OPT_MYIP             (1<<2) /* specified my ip */
-#define CEPH_OPT_DIRSTAT          (1<<4) /* funky `cat dirname` for stats */
-#define CEPH_OPT_RBYTES           (1<<5) /* dir st_bytes = rbytes */
-#define CEPH_OPT_NOCRC            (1<<6) /* no data crc on writes */
-#define CEPH_OPT_NOASYNCREADDIR   (1<<7) /* no dcache readdir */
+#define CEPH_MOUNT_OPT_DEFAULT    (CEPH_MOUNT_OPT_RBYTES)
 
-#define CEPH_OPT_DEFAULT   (CEPH_OPT_RBYTES)
+#define ceph_set_mount_opt(fsc, opt) \
+	(fsc)->mount_options->flags |= CEPH_MOUNT_OPT_##opt;
+#define ceph_test_mount_opt(fsc, opt) \
+	(!!((fsc)->mount_options->flags & CEPH_MOUNT_OPT_##opt))
 
-#define ceph_set_opt(client, opt) \
-	(client)->mount_args->flags |= CEPH_OPT_##opt;
-#define ceph_test_opt(client, opt) \
-	(!!((client)->mount_args->flags & CEPH_OPT_##opt))
+#define CEPH_MAX_READDIR_DEFAULT        1024
+#define CEPH_MAX_READDIR_BYTES_DEFAULT  (512*1024)
+#define CEPH_SNAPDIRNAME_DEFAULT        ".snap"
 
-
-struct ceph_mount_args {
-	int sb_flags;
+struct ceph_mount_options {
 	int flags;
-	struct ceph_fsid fsid;
-	struct ceph_entity_addr my_addr;
-	int num_mon;
-	struct ceph_entity_addr *mon_addr;
-	int mount_timeout;
-	int osd_idle_ttl;
-	int osd_timeout;
-	int osd_keepalive_timeout;
+	int sb_flags;
+
 	int wsize;
 	int rsize;            /* max readahead */
 	int congestion_kb;    /* max writeback in flight */
@@ -73,82 +50,25 @@ struct ceph_mount_args {
 	int cap_release_safety;
 	int max_readdir;       /* max readdir result (entires) */
 	int max_readdir_bytes; /* max readdir result (bytes) */
-	char *snapdir_name;   /* default ".snap" */
-	char *name;
-	char *secret;
-};
-
-/*
- * defaults
- */
-#define CEPH_MOUNT_TIMEOUT_DEFAULT  60
-#define CEPH_OSD_TIMEOUT_DEFAULT    60  /* seconds */
-#define CEPH_OSD_KEEPALIVE_DEFAULT  5
-#define CEPH_OSD_IDLE_TTL_DEFAULT    60
-#define CEPH_MOUNT_RSIZE_DEFAULT    (512*1024) /* readahead */
-#define CEPH_MAX_READDIR_DEFAULT    1024
-#define CEPH_MAX_READDIR_BYTES_DEFAULT    (512*1024)
-
-#define CEPH_MSG_MAX_FRONT_LEN	(16*1024*1024)
-#define CEPH_MSG_MAX_DATA_LEN	(16*1024*1024)
-
-#define CEPH_SNAPDIRNAME_DEFAULT ".snap"
-#define CEPH_AUTH_NAME_DEFAULT   "guest"
-/*
- * Delay telling the MDS we no longer want caps, in case we reopen
- * the file.  Delay a minimum amount of time, even if we send a cap
- * message for some other reason.  Otherwise, take the oppotunity to
- * update the mds to avoid sending another message later.
- */
-#define CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT      5  /* cap release delay */
-#define CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT     60  /* cap release delay */
-
-#define CEPH_CAP_RELEASE_SAFETY_DEFAULT        (CEPH_CAPS_PER_RELEASE * 4)
-
-/* mount state */
-enum {
-	CEPH_MOUNT_MOUNTING,
-	CEPH_MOUNT_MOUNTED,
-	CEPH_MOUNT_UNMOUNTING,
-	CEPH_MOUNT_UNMOUNTED,
-	CEPH_MOUNT_SHUTDOWN,
-};
 
-/*
- * subtract jiffies
- */
-static inline unsigned long time_sub(unsigned long a, unsigned long b)
-{
-	BUG_ON(time_after(b, a));
-	return (long)a - (long)b;
-}
-
-/*
- * per-filesystem client state
- *
- * possibly shared by multiple mount points, if they are
- * mounting the same ceph filesystem/cluster.
- */
-struct ceph_client {
-	struct ceph_fsid fsid;
-	bool have_fsid;
+	/*
+	 * everything above this point can be memcmp'd; everything below
+	 * is handled in compare_mount_options()
+	 */
 
-	struct mutex mount_mutex;       /* serialize mount attempts */
-	struct ceph_mount_args *mount_args;
+	char *snapdir_name;   /* default ".snap" */
+};
 
+struct ceph_fs_client {
 	struct super_block *sb;
 
-	unsigned long mount_state;
-	wait_queue_head_t auth_wq;
-
-	int auth_err;
+	struct ceph_mount_options *mount_options;
+	struct ceph_client *client;
 
+	unsigned long mount_state;
 	int min_caps;                  /* min caps i added */
 
-	struct ceph_messenger *msgr;   /* messenger instance */
-	struct ceph_mon_client monc;
-	struct ceph_mds_client mdsc;
-	struct ceph_osd_client osdc;
+	struct ceph_mds_client *mdsc;
 
 	/* writeback */
 	mempool_t *wb_pagevec_pool;
@@ -160,14 +80,14 @@ struct ceph_client {
 	struct backing_dev_info backing_dev_info;
 
 #ifdef CONFIG_DEBUG_FS
-	struct dentry *debugfs_monmap;
-	struct dentry *debugfs_mdsmap, *debugfs_osdmap;
-	struct dentry *debugfs_dir, *debugfs_dentry_lru, *debugfs_caps;
+	struct dentry *debugfs_dentry_lru, *debugfs_caps;
 	struct dentry *debugfs_congestion_kb;
 	struct dentry *debugfs_bdi;
+	struct dentry *debugfs_mdsc, *debugfs_mdsmap;
 #endif
 };
 
+
 /*
  * File i/o capability.  This tracks shared state with the metadata
  * server that allows us to cache or writeback attributes or to read
@@ -275,6 +195,20 @@ struct ceph_inode_xattr {
 	int should_free_val;
 };
 
+/*
+ * Ceph dentry state
+ */
+struct ceph_dentry_info {
+	struct ceph_mds_session *lease_session;
+	u32 lease_gen, lease_shared_gen;
+	u32 lease_seq;
+	unsigned long lease_renew_after, lease_renew_from;
+	struct list_head lru;
+	struct dentry *dentry;
+	u64 time;
+	u64 offset;
+};
+
 struct ceph_inode_xattrs_info {
 	/*
 	 * (still encoded) xattr blob. we avoid the overhead of parsing
@@ -296,11 +230,6 @@ struct ceph_inode_xattrs_info {
 /*
  * Ceph inode.
  */
-#define CEPH_I_COMPLETE  1  /* we have complete directory cached */
-#define CEPH_I_NODELAY   4  /* do not delay cap release */
-#define CEPH_I_FLUSH     8  /* do not delay flush of dirty metadata */
-#define CEPH_I_NOFLUSH  16  /* do not flush dirty caps */
-
 struct ceph_inode_info {
 	struct ceph_vino i_vino;   /* ceph ino + snap */
 
@@ -391,6 +320,63 @@ static inline struct ceph_inode_info *ceph_inode(struct inode *inode)
 	return container_of(inode, struct ceph_inode_info, vfs_inode);
 }
 
+static inline struct ceph_vino ceph_vino(struct inode *inode)
+{
+	return ceph_inode(inode)->i_vino;
+}
+
+/*
+ * ino_t is <64 bits on many architectures, blech.
+ *
+ * don't include snap in ino hash, at least for now.
+ */
+static inline ino_t ceph_vino_to_ino(struct ceph_vino vino)
+{
+	ino_t ino = (ino_t)vino.ino;  /* ^ (vino.snap << 20); */
+#if BITS_PER_LONG == 32
+	ino ^= vino.ino >> (sizeof(u64)-sizeof(ino_t)) * 8;
+	if (!ino)
+		ino = 1;
+#endif
+	return ino;
+}
+
+/* for printf-style formatting */
+#define ceph_vinop(i) ceph_inode(i)->i_vino.ino, ceph_inode(i)->i_vino.snap
+
+static inline u64 ceph_ino(struct inode *inode)
+{
+	return ceph_inode(inode)->i_vino.ino;
+}
+static inline u64 ceph_snap(struct inode *inode)
+{
+	return ceph_inode(inode)->i_vino.snap;
+}
+
+static inline int ceph_ino_compare(struct inode *inode, void *data)
+{
+	struct ceph_vino *pvino = (struct ceph_vino *)data;
+	struct ceph_inode_info *ci = ceph_inode(inode);
+	return ci->i_vino.ino == pvino->ino &&
+		ci->i_vino.snap == pvino->snap;
+}
+
+static inline struct inode *ceph_find_inode(struct super_block *sb,
+					    struct ceph_vino vino)
+{
+	ino_t t = ceph_vino_to_ino(vino);
+	return ilookup5(sb, t, ceph_ino_compare, &vino);
+}
+
+
+/*
+ * Ceph inode.
+ */
+#define CEPH_I_COMPLETE  1  /* we have complete directory cached */
+#define CEPH_I_NODELAY   4  /* do not delay cap release */
+#define CEPH_I_FLUSH     8  /* do not delay flush of dirty metadata */
+#define CEPH_I_NOFLUSH  16  /* do not flush dirty caps */
+
 static inline void ceph_i_clear(struct inode *inode, unsigned mask)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
@@ -432,20 +418,6 @@ extern u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
 			    struct ceph_inode_frag *pfrag,
 			    int *found);
 
-/*
- * Ceph dentry state
- */
-struct ceph_dentry_info {
-	struct ceph_mds_session *lease_session;
-	u32 lease_gen, lease_shared_gen;
-	u32 lease_seq;
-	unsigned long lease_renew_after, lease_renew_from;
-	struct list_head lru;
-	struct dentry *dentry;
-	u64 time;
-	u64 offset;
-};
-
 static inline struct ceph_dentry_info *ceph_dentry(struct dentry *dentry)
 {
 	return (struct ceph_dentry_info *)dentry->d_fsdata;
@@ -456,22 +428,6 @@ static inline loff_t ceph_make_fpos(unsigned frag, unsigned off)
 	return ((loff_t)frag << 32) | (loff_t)off;
 }
 
-/*
- * ino_t is <64 bits on many architectures, blech.
- *
- * don't include snap in ino hash, at least for now.
- */
-static inline ino_t ceph_vino_to_ino(struct ceph_vino vino)
-{
-	ino_t ino = (ino_t)vino.ino;  /* ^ (vino.snap << 20); */
-#if BITS_PER_LONG == 32
-	ino ^= vino.ino >> (sizeof(u64)-sizeof(ino_t)) * 8;
-	if (!ino)
-		ino = 1;
-#endif
-	return ino;
-}
-
 static inline int ceph_set_ino_cb(struct inode *inode, void *data)
 {
 	ceph_inode(inode)->i_vino = *(struct ceph_vino *)data;
@@ -479,39 +435,6 @@ static inline int ceph_set_ino_cb(struct inode *inode, void *data)
 	return 0;
 }
 
-static inline struct ceph_vino ceph_vino(struct inode *inode)
-{
-	return ceph_inode(inode)->i_vino;
-}
-
-/* for printf-style formatting */
-#define ceph_vinop(i) ceph_inode(i)->i_vino.ino, ceph_inode(i)->i_vino.snap
-
-static inline u64 ceph_ino(struct inode *inode)
-{
-	return ceph_inode(inode)->i_vino.ino;
-}
-static inline u64 ceph_snap(struct inode *inode)
-{
-	return ceph_inode(inode)->i_vino.snap;
-}
-
-static inline int ceph_ino_compare(struct inode *inode, void *data)
-{
-	struct ceph_vino *pvino = (struct ceph_vino *)data;
-	struct ceph_inode_info *ci = ceph_inode(inode);
-	return ci->i_vino.ino == pvino->ino &&
-		ci->i_vino.snap == pvino->snap;
-}
-
-static inline struct inode *ceph_find_inode(struct super_block *sb,
-					    struct ceph_vino vino)
-{
-	ino_t t = ceph_vino_to_ino(vino);
-	return ilookup5(sb, t, ceph_ino_compare, &vino);
-}
-
-
 /*
  * caps helpers
  */
@@ -576,18 +499,18 @@ extern int ceph_reserve_caps(struct ceph_mds_client *mdsc,
 			     struct ceph_cap_reservation *ctx, int need);
 extern int ceph_unreserve_caps(struct ceph_mds_client *mdsc,
 			       struct ceph_cap_reservation *ctx);
-extern void ceph_reservation_status(struct ceph_client *client,
+extern void ceph_reservation_status(struct ceph_fs_client *client,
 				    int *total, int *avail, int *used,
 				    int *reserved, int *min);
 
-static inline struct ceph_client *ceph_inode_to_client(struct inode *inode)
+static inline struct ceph_fs_client *ceph_inode_to_client(struct inode *inode)
 {
-	return (struct ceph_client *)inode->i_sb->s_fs_info;
+	return (struct ceph_fs_client *)inode->i_sb->s_fs_info;
 }
 
-static inline struct ceph_client *ceph_sb_to_client(struct super_block *sb)
+static inline struct ceph_fs_client *ceph_sb_to_client(struct super_block *sb)
 {
-	return (struct ceph_client *)sb->s_fs_info;
+	return (struct ceph_fs_client *)sb->s_fs_info;
 }
 
 
@@ -616,51 +539,6 @@ struct ceph_file_info {
 
 
 
-/*
- * snapshots
- */
-
-/*
- * A "snap context" is the set of existing snapshots when we
- * write data.  It is used by the OSD to guide its COW behavior.
- *
- * The ceph_snap_context is refcounted, and attached to each dirty
- * page, indicating which context the dirty data belonged when it was
- * dirtied.
- */
-struct ceph_snap_context {
-	atomic_t nref;
-	u64 seq;
-	int num_snaps;
-	u64 snaps[];
-};
-
-static inline struct ceph_snap_context *
-ceph_get_snap_context(struct ceph_snap_context *sc)
-{
-	/*
-	printk("get_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
-	       atomic_read(&sc->nref)+1);
-	*/
-	if (sc)
-		atomic_inc(&sc->nref);
-	return sc;
-}
-
-static inline void ceph_put_snap_context(struct ceph_snap_context *sc)
-{
-	if (!sc)
-		return;
-	/*
-	printk("put_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
-	       atomic_read(&sc->nref)-1);
-	*/
-	if (atomic_dec_and_test(&sc->nref)) {
-		/*printk(" deleting snap_context %p\n", sc);*/
-		kfree(sc);
-	}
-}
-
 /*
  * A "snap realm" describes a subset of the file hierarchy sharing
  * the same set of snapshots that apply to it.  The realms themselves
@@ -699,16 +577,33 @@ struct ceph_snap_realm {
 	spinlock_t inodes_with_caps_lock;
 };
 
-
-
-/*
- * calculate the number of pages a given length and offset map onto,
- * if we align the data.
- */
-static inline int calc_pages_for(u64 off, u64 len)
+static inline int default_congestion_kb(void)
 {
-	return ((off+len+PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT) -
-		(off >> PAGE_CACHE_SHIFT);
+	int congestion_kb;
+
+	/*
+	 * Copied from NFS
+	 *
+	 * congestion size, scale with available memory.
+	 *
+	 *  64MB:    8192k
+	 * 128MB:   11585k
+	 * 256MB:   16384k
+	 * 512MB:   23170k
+	 *   1GB:   32768k
+	 *   2GB:   46340k
+	 *   4GB:   65536k
+	 *   8GB:   92681k
+	 *  16GB:  131072k
+	 *
+	 * This allows larger machines to have larger/more transfers.
+	 * Limit the default to 256M
+	 */
+	congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
+	if (congestion_kb > 256*1024)
+		congestion_kb = 256*1024;
+
+	return congestion_kb;
 }
 
 
@@ -741,16 +636,6 @@ static inline bool __ceph_have_pending_cap_snap(struct ceph_inode_info *ci)
 			   ci_item)->writing;
 }
 
-
-/* super.c */
-extern struct kmem_cache *ceph_inode_cachep;
-extern struct kmem_cache *ceph_cap_cachep;
-extern struct kmem_cache *ceph_dentry_cachep;
-extern struct kmem_cache *ceph_file_cachep;
-
-extern const char *ceph_msg_type_name(int type);
-extern int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid);
-
 /* inode.c */
 extern const struct inode_operations ceph_file_iops;
 
@@ -857,12 +742,18 @@ extern int ceph_mmap(struct file *file, struct vm_area_struct *vma);
 /* file.c */
 extern const struct file_operations ceph_file_fops;
 extern const struct address_space_operations ceph_aops;
+extern int ceph_copy_to_page_vector(struct page **pages,
+				    const char *data,
+				    loff_t off, size_t len);
+extern int ceph_copy_from_page_vector(struct page **pages,
+				    char *data,
+				    loff_t off, size_t len);
+extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags);
 extern int ceph_open(struct inode *inode, struct file *file);
 extern struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry,
 				       struct nameidata *nd, int mode,
 				       int locked_dir);
 extern int ceph_release(struct inode *inode, struct file *filp);
-extern void ceph_release_page_vector(struct page **pages, int num_pages);
 
 /* dir.c */
 extern const struct file_operations ceph_dir_fops;
@@ -892,12 +783,6 @@ extern long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 /* export.c */
 extern const struct export_operations ceph_export_ops;
 
-/* debugfs.c */
-extern int ceph_debugfs_init(void);
-extern void ceph_debugfs_cleanup(void);
-extern int ceph_debugfs_client_init(struct ceph_client *client);
-extern void ceph_debugfs_client_cleanup(struct ceph_client *client);
-
 /* locks.c */
 extern int ceph_lock(struct file *file, int cmd, struct file_lock *fl);
 extern int ceph_flock(struct file *file, int cmd, struct file_lock *fl);
@@ -914,4 +799,8 @@ static inline struct inode *get_dentry_parent_inode(struct dentry *dentry)
 	return NULL;
 }
 
+/* debugfs.c */
+extern int ceph_fs_debugfs_init(struct ceph_fs_client *client);
+extern void ceph_fs_debugfs_cleanup(struct ceph_fs_client *client);
+
 #endif /* _FS_CEPH_SUPER_H */
diff --git a/fs/ceph/types.h b/fs/ceph/types.h
deleted file mode 100644
index 28b35a005ec2..000000000000
--- a/fs/ceph/types.h
+++ /dev/null
@@ -1,29 +0,0 @@
-#ifndef _FS_CEPH_TYPES_H
-#define _FS_CEPH_TYPES_H
-
-/* needed before including ceph_fs.h */
-#include <linux/in.h>
-#include <linux/types.h>
-#include <linux/fcntl.h>
-#include <linux/string.h>
-
-#include "ceph_fs.h"
-#include "ceph_frag.h"
-#include "ceph_hash.h"
-
-/*
- * Identify inodes by both their ino AND snapshot id (a u64).
- */
-struct ceph_vino {
-	u64 ino;
-	u64 snap;
-};
-
-
-/* context for the caps reservation mechanism */
-struct ceph_cap_reservation {
-	int count;
-};
-
-
-#endif
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index 9578af610b73..70e919960acb 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -1,6 +1,9 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
+
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+
+#include <linux/ceph/decode.h>
 
 #include <linux/xattr.h>
 #include <linux/slab.h>
@@ -620,12 +623,12 @@ out:
 static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
 			      const char *value, size_t size, int flags)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
 	struct inode *inode = dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct inode *parent_inode = dentry->d_parent->d_inode;
 	struct ceph_mds_request *req;
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	int err;
 	int i, nr_pages;
 	struct page **pages = NULL;
@@ -777,8 +780,8 @@ out:
 
 static int ceph_send_removexattr(struct dentry *dentry, const char *name)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = dentry->d_inode;
 	struct inode *parent_inode = dentry->d_parent->d_inode;
 	struct ceph_mds_request *req;
diff --git a/include/linux/ceph/auth.h b/include/linux/ceph/auth.h
new file mode 100644
index 000000000000..7fff521d7eb5
--- /dev/null
+++ b/include/linux/ceph/auth.h
@@ -0,0 +1,92 @@
+#ifndef _FS_CEPH_AUTH_H
+#define _FS_CEPH_AUTH_H
+
+#include <linux/ceph/types.h>
+#include <linux/ceph/buffer.h>
+
+/*
+ * Abstract interface for communicating with the authenticate module.
+ * There is some handshake that takes place between us and the monitor
+ * to acquire the necessary keys.  These are used to generate an
+ * 'authorizer' that we use when connecting to a service (mds, osd).
+ */
+
+struct ceph_auth_client;
+struct ceph_authorizer;
+
+struct ceph_auth_client_ops {
+	const char *name;
+
+	/*
+	 * true if we are authenticated and can connect to
+	 * services.
+	 */
+	int (*is_authenticated)(struct ceph_auth_client *ac);
+
+	/*
+	 * true if we should (re)authenticate, e.g., when our tickets
+	 * are getting old and crusty.
+	 */
+	int (*should_authenticate)(struct ceph_auth_client *ac);
+
+	/*
+	 * build requests and process replies during monitor
+	 * handshake.  if handle_reply returns -EAGAIN, we build
+	 * another request.
+	 */
+	int (*build_request)(struct ceph_auth_client *ac, void *buf, void *end);
+	int (*handle_reply)(struct ceph_auth_client *ac, int result,
+			    void *buf, void *end);
+
+	/*
+	 * Create authorizer for connecting to a service, and verify
+	 * the response to authenticate the service.
+	 */
+	int (*create_authorizer)(struct ceph_auth_client *ac, int peer_type,
+				 struct ceph_authorizer **a,
+				 void **buf, size_t *len,
+				 void **reply_buf, size_t *reply_len);
+	int (*verify_authorizer_reply)(struct ceph_auth_client *ac,
+				       struct ceph_authorizer *a, size_t len);
+	void (*destroy_authorizer)(struct ceph_auth_client *ac,
+				   struct ceph_authorizer *a);
+	void (*invalidate_authorizer)(struct ceph_auth_client *ac,
+				      int peer_type);
+
+	/* reset when we (re)connect to a monitor */
+	void (*reset)(struct ceph_auth_client *ac);
+
+	void (*destroy)(struct ceph_auth_client *ac);
+};
+
+struct ceph_auth_client {
+	u32 protocol;           /* CEPH_AUTH_* */
+	void *private;          /* for use by protocol implementation */
+	const struct ceph_auth_client_ops *ops;  /* null iff protocol==0 */
+
+	bool negotiating;       /* true if negotiating protocol */
+	const char *name;       /* entity name */
+	u64 global_id;          /* our unique id in system */
+	const char *secret;     /* our secret key */
+	unsigned want_keys;     /* which services we want */
+};
+
+extern struct ceph_auth_client *ceph_auth_init(const char *name,
+					       const char *secret);
+extern void ceph_auth_destroy(struct ceph_auth_client *ac);
+
+extern void ceph_auth_reset(struct ceph_auth_client *ac);
+
+extern int ceph_auth_build_hello(struct ceph_auth_client *ac,
+				 void *buf, size_t len);
+extern int ceph_handle_auth_reply(struct ceph_auth_client *ac,
+				  void *buf, size_t len,
+				  void *reply_buf, size_t reply_len);
+extern int ceph_entity_name_encode(const char *name, void **p, void *end);
+
+extern int ceph_build_auth(struct ceph_auth_client *ac,
+		    void *msg_buf, size_t msg_len);
+
+extern int ceph_auth_is_authenticated(struct ceph_auth_client *ac);
+
+#endif
diff --git a/include/linux/ceph/buffer.h b/include/linux/ceph/buffer.h
new file mode 100644
index 000000000000..58d19014068f
--- /dev/null
+++ b/include/linux/ceph/buffer.h
@@ -0,0 +1,39 @@
+#ifndef __FS_CEPH_BUFFER_H
+#define __FS_CEPH_BUFFER_H
+
+#include <linux/kref.h>
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
+#include <linux/types.h>
+#include <linux/uio.h>
+
+/*
+ * a simple reference counted buffer.
+ *
+ * use kmalloc for small sizes (<= one page), vmalloc for larger
+ * sizes.
+ */
+struct ceph_buffer {
+	struct kref kref;
+	struct kvec vec;
+	size_t alloc_len;
+	bool is_vmalloc;
+};
+
+extern struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp);
+extern void ceph_buffer_release(struct kref *kref);
+
+static inline struct ceph_buffer *ceph_buffer_get(struct ceph_buffer *b)
+{
+	kref_get(&b->kref);
+	return b;
+}
+
+static inline void ceph_buffer_put(struct ceph_buffer *b)
+{
+	kref_put(&b->kref, ceph_buffer_release);
+}
+
+extern int ceph_decode_buffer(struct ceph_buffer **b, void **p, void *end);
+
+#endif
diff --git a/include/linux/ceph/ceph_debug.h b/include/linux/ceph/ceph_debug.h
new file mode 100644
index 000000000000..aa2e19182d99
--- /dev/null
+++ b/include/linux/ceph/ceph_debug.h
@@ -0,0 +1,38 @@
+#ifndef _FS_CEPH_DEBUG_H
+#define _FS_CEPH_DEBUG_H
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#ifdef CONFIG_CEPH_LIB_PRETTYDEBUG
+
+/*
+ * wrap pr_debug to include a filename:lineno prefix on each line.
+ * this incurs some overhead (kernel size and execution time) due to
+ * the extra function call at each call site.
+ */
+
+# if defined(DEBUG) || defined(CONFIG_DYNAMIC_DEBUG)
+extern const char *ceph_file_part(const char *s, int len);
+#  define dout(fmt, ...)						\
+	pr_debug("%.*s %12.12s:%-4d : " fmt,				\
+		 8 - (int)sizeof(KBUILD_MODNAME), "    ",		\
+		 ceph_file_part(__FILE__, sizeof(__FILE__)),		\
+		 __LINE__, ##__VA_ARGS__)
+# else
+/* faux printk call just to see any compiler warnings. */
+#  define dout(fmt, ...)	do {				\
+		if (0)						\
+			printk(KERN_DEBUG fmt, ##__VA_ARGS__);	\
+	} while (0)
+# endif
+
+#else
+
+/*
+ * or, just wrap pr_debug
+ */
+# define dout(fmt, ...)	pr_debug(" " fmt, ##__VA_ARGS__)
+
+#endif
+
+#endif
diff --git a/include/linux/ceph/ceph_frag.h b/include/linux/ceph/ceph_frag.h
new file mode 100644
index 000000000000..5babb8e95352
--- /dev/null
+++ b/include/linux/ceph/ceph_frag.h
@@ -0,0 +1,109 @@
+#ifndef FS_CEPH_FRAG_H
+#define FS_CEPH_FRAG_H
+
+/*
+ * "Frags" are a way to describe a subset of a 32-bit number space,
+ * using a mask and a value to match against that mask.  Any given frag
+ * (subset of the number space) can be partitioned into 2^n sub-frags.
+ *
+ * Frags are encoded into a 32-bit word:
+ *   8 upper bits = "bits"
+ *  24 lower bits = "value"
+ * (We could go to 5+27 bits, but who cares.)
+ *
+ * We use the _most_ significant bits of the 24 bit value.  This makes
+ * values logically sort.
+ *
+ * Unfortunately, because the "bits" field is still in the high bits, we
+ * can't sort encoded frags numerically.  However, it does allow you
+ * to feed encoded frags as values into frag_contains_value.
+ */
+static inline __u32 ceph_frag_make(__u32 b, __u32 v)
+{
+	return (b << 24) |
+		(v & (0xffffffu << (24-b)) & 0xffffffu);
+}
+static inline __u32 ceph_frag_bits(__u32 f)
+{
+	return f >> 24;
+}
+static inline __u32 ceph_frag_value(__u32 f)
+{
+	return f & 0xffffffu;
+}
+static inline __u32 ceph_frag_mask(__u32 f)
+{
+	return (0xffffffu << (24-ceph_frag_bits(f))) & 0xffffffu;
+}
+static inline __u32 ceph_frag_mask_shift(__u32 f)
+{
+	return 24 - ceph_frag_bits(f);
+}
+
+static inline int ceph_frag_contains_value(__u32 f, __u32 v)
+{
+	return (v & ceph_frag_mask(f)) == ceph_frag_value(f);
+}
+static inline int ceph_frag_contains_frag(__u32 f, __u32 sub)
+{
+	/* is sub as specific as us, and contained by us? */
+	return ceph_frag_bits(sub) >= ceph_frag_bits(f) &&
+	       (ceph_frag_value(sub) & ceph_frag_mask(f)) == ceph_frag_value(f);
+}
+
+static inline __u32 ceph_frag_parent(__u32 f)
+{
+	return ceph_frag_make(ceph_frag_bits(f) - 1,
+			 ceph_frag_value(f) & (ceph_frag_mask(f) << 1));
+}
+static inline int ceph_frag_is_left_child(__u32 f)
+{
+	return ceph_frag_bits(f) > 0 &&
+		(ceph_frag_value(f) & (0x1000000 >> ceph_frag_bits(f))) == 0;
+}
+static inline int ceph_frag_is_right_child(__u32 f)
+{
+	return ceph_frag_bits(f) > 0 &&
+		(ceph_frag_value(f) & (0x1000000 >> ceph_frag_bits(f))) == 1;
+}
+static inline __u32 ceph_frag_sibling(__u32 f)
+{
+	return ceph_frag_make(ceph_frag_bits(f),
+		      ceph_frag_value(f) ^ (0x1000000 >> ceph_frag_bits(f)));
+}
+static inline __u32 ceph_frag_left_child(__u32 f)
+{
+	return ceph_frag_make(ceph_frag_bits(f)+1, ceph_frag_value(f));
+}
+static inline __u32 ceph_frag_right_child(__u32 f)
+{
+	return ceph_frag_make(ceph_frag_bits(f)+1,
+	      ceph_frag_value(f) | (0x1000000 >> (1+ceph_frag_bits(f))));
+}
+static inline __u32 ceph_frag_make_child(__u32 f, int by, int i)
+{
+	int newbits = ceph_frag_bits(f) + by;
+	return ceph_frag_make(newbits,
+			 ceph_frag_value(f) | (i << (24 - newbits)));
+}
+static inline int ceph_frag_is_leftmost(__u32 f)
+{
+	return ceph_frag_value(f) == 0;
+}
+static inline int ceph_frag_is_rightmost(__u32 f)
+{
+	return ceph_frag_value(f) == ceph_frag_mask(f);
+}
+static inline __u32 ceph_frag_next(__u32 f)
+{
+	return ceph_frag_make(ceph_frag_bits(f),
+			 ceph_frag_value(f) + (0x1000000 >> ceph_frag_bits(f)));
+}
+
+/*
+ * comparator to sort frags logically, as when traversing the
+ * number space in ascending order...
+ */
+int ceph_frag_compare(__u32 a, __u32 b);
+
+#endif
diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h
new file mode 100644
index 000000000000..d5619ac86711
--- /dev/null
+++ b/include/linux/ceph/ceph_fs.h
@@ -0,0 +1,728 @@
+/*
+ * ceph_fs.h - Ceph constants and data types to share between kernel and
+ * user space.
+ *
+ * Most types in this file are defined as little-endian, and are
+ * primarily intended to describe data structures that pass over the
+ * wire or that are stored on disk.
+ *
+ * LGPL2
+ */
+
+#ifndef CEPH_FS_H
+#define CEPH_FS_H
+
+#include "msgr.h"
+#include "rados.h"
+
+/*
+ * subprotocol versions.  when specific messages types or high-level
+ * protocols change, bump the affected components.  we keep rev
+ * internal cluster protocols separately from the public,
+ * client-facing protocol.
+ */
+#define CEPH_OSD_PROTOCOL     8 /* cluster internal */
+#define CEPH_MDS_PROTOCOL    12 /* cluster internal */
+#define CEPH_MON_PROTOCOL     5 /* cluster internal */
+#define CEPH_OSDC_PROTOCOL   24 /* server/client */
+#define CEPH_MDSC_PROTOCOL   32 /* server/client */
+#define CEPH_MONC_PROTOCOL   15 /* server/client */
+
+
+#define CEPH_INO_ROOT  1
+#define CEPH_INO_CEPH  2        /* hidden .ceph dir */
+
+/* arbitrary limit on max # of monitors (cluster of 3 is typical) */
+#define CEPH_MAX_MON   31
+
+
+/*
+ * feature bits
+ */
+#define CEPH_FEATURE_UID            (1<<0)
+#define CEPH_FEATURE_NOSRCADDR      (1<<1)
+#define CEPH_FEATURE_MONCLOCKCHECK  (1<<2)
+#define CEPH_FEATURE_FLOCK          (1<<3)
+
+
+/*
+ * ceph_file_layout - describe data layout for a file/inode
+ */
+struct ceph_file_layout {
+	/* file -> object mapping */
+	__le32 fl_stripe_unit;     /* stripe unit, in bytes.  must be multiple
+				      of page size. */
+	__le32 fl_stripe_count;    /* over this many objects */
+	__le32 fl_object_size;     /* until objects are this big, then move to
+				      new objects */
+	__le32 fl_cas_hash;        /* 0 = none; 1 = sha256 */
+
+	/* pg -> disk layout */
+	__le32 fl_object_stripe_unit;  /* for per-object parity, if any */
+
+	/* object -> pg layout */
+	__le32 fl_pg_preferred; /* preferred primary for pg (-1 for none) */
+	__le32 fl_pg_pool;      /* namespace, crush ruleset, rep level */
+} __attribute__ ((packed));
+
+#define CEPH_MIN_STRIPE_UNIT 65536
+
+int ceph_file_layout_is_valid(const struct ceph_file_layout *layout);
+
+
+/* crypto algorithms */
+#define CEPH_CRYPTO_NONE 0x0
+#define CEPH_CRYPTO_AES  0x1
+
+#define CEPH_AES_IV "cephsageyudagreg"
+
+/* security/authentication protocols */
+#define CEPH_AUTH_UNKNOWN	0x0
+#define CEPH_AUTH_NONE	 	0x1
+#define CEPH_AUTH_CEPHX	 	0x2
+
+#define CEPH_AUTH_UID_DEFAULT ((__u64) -1)
+
+
+/*********************************************
+ * message layer
+ */
+
+/*
+ * message types
+ */
+
+/* misc */
+#define CEPH_MSG_SHUTDOWN               1
+#define CEPH_MSG_PING                   2
+
+/* client <-> monitor */
+#define CEPH_MSG_MON_MAP                4
+#define CEPH_MSG_MON_GET_MAP            5
+#define CEPH_MSG_STATFS                 13
+#define CEPH_MSG_STATFS_REPLY           14
+#define CEPH_MSG_MON_SUBSCRIBE          15
+#define CEPH_MSG_MON_SUBSCRIBE_ACK      16
+#define CEPH_MSG_AUTH			17
+#define CEPH_MSG_AUTH_REPLY		18
+
+/* client <-> mds */
+#define CEPH_MSG_MDS_MAP                21
+
+#define CEPH_MSG_CLIENT_SESSION         22
+#define CEPH_MSG_CLIENT_RECONNECT       23
+
+#define CEPH_MSG_CLIENT_REQUEST         24
+#define CEPH_MSG_CLIENT_REQUEST_FORWARD 25
+#define CEPH_MSG_CLIENT_REPLY           26
+#define CEPH_MSG_CLIENT_CAPS            0x310
+#define CEPH_MSG_CLIENT_LEASE           0x311
+#define CEPH_MSG_CLIENT_SNAP            0x312
+#define CEPH_MSG_CLIENT_CAPRELEASE      0x313
+
+/* pool ops */
+#define CEPH_MSG_POOLOP_REPLY           48
+#define CEPH_MSG_POOLOP                 49
+
+
+/* osd */
+#define CEPH_MSG_OSD_MAP          41
+#define CEPH_MSG_OSD_OP           42
+#define CEPH_MSG_OSD_OPREPLY      43
+
+/* pool operations */
+enum {
+  POOL_OP_CREATE			= 0x01,
+  POOL_OP_DELETE			= 0x02,
+  POOL_OP_AUID_CHANGE			= 0x03,
+  POOL_OP_CREATE_SNAP			= 0x11,
+  POOL_OP_DELETE_SNAP			= 0x12,
+  POOL_OP_CREATE_UNMANAGED_SNAP		= 0x21,
+  POOL_OP_DELETE_UNMANAGED_SNAP		= 0x22,
+};
+
+struct ceph_mon_request_header {
+	__le64 have_version;
+	__le16 session_mon;
+	__le64 session_mon_tid;
+} __attribute__ ((packed));
+
+struct ceph_mon_statfs {
+	struct ceph_mon_request_header monhdr;
+	struct ceph_fsid fsid;
+} __attribute__ ((packed));
+
+struct ceph_statfs {
+	__le64 kb, kb_used, kb_avail;
+	__le64 num_objects;
+} __attribute__ ((packed));
+
+struct ceph_mon_statfs_reply {
+	struct ceph_fsid fsid;
+	__le64 version;
+	struct ceph_statfs st;
+} __attribute__ ((packed));
+
+const char *ceph_pool_op_name(int op);
+
+struct ceph_mon_poolop {
+	struct ceph_mon_request_header monhdr;
+	struct ceph_fsid fsid;
+	__le32 pool;
+	__le32 op;
+	__le64 auid;
+	__le64 snapid;
+	__le32 name_len;
+} __attribute__ ((packed));
+
+struct ceph_mon_poolop_reply {
+	struct ceph_mon_request_header monhdr;
+	struct ceph_fsid fsid;
+	__le32 reply_code;
+	__le32 epoch;
+	char has_data;
+	char data[0];
+} __attribute__ ((packed));
+
+struct ceph_mon_unmanaged_snap {
+	__le64 snapid;
+} __attribute__ ((packed));
+
+struct ceph_osd_getmap {
+	struct ceph_mon_request_header monhdr;
+	struct ceph_fsid fsid;
+	__le32 start;
+} __attribute__ ((packed));
+
+struct ceph_mds_getmap {
+	struct ceph_mon_request_header monhdr;
+	struct ceph_fsid fsid;
+} __attribute__ ((packed));
+
+struct ceph_client_mount {
+	struct ceph_mon_request_header monhdr;
+} __attribute__ ((packed));
+
+struct ceph_mon_subscribe_item {
+	__le64 have_version;	__le64 have;
+	__u8 onetime;
+} __attribute__ ((packed));
+
+struct ceph_mon_subscribe_ack {
+	__le32 duration;         /* seconds */
+	struct ceph_fsid fsid;
+} __attribute__ ((packed));
+
+/*
+ * mds states
+ *   > 0 -> in
+ *  <= 0 -> out
+ */
+#define CEPH_MDS_STATE_DNE          0  /* down, does not exist. */
+#define CEPH_MDS_STATE_STOPPED     -1  /* down, once existed, but no subtrees.
+					  empty log. */
+#define CEPH_MDS_STATE_BOOT        -4  /* up, boot announcement. */
+#define CEPH_MDS_STATE_STANDBY     -5  /* up, idle.  waiting for assignment. */
+#define CEPH_MDS_STATE_CREATING    -6  /* up, creating MDS instance. */
+#define CEPH_MDS_STATE_STARTING    -7  /* up, starting previously stopped mds */
+#define CEPH_MDS_STATE_STANDBY_REPLAY -8 /* up, tailing active node's journal */
+
+#define CEPH_MDS_STATE_REPLAY       8  /* up, replaying journal. */
+#define CEPH_MDS_STATE_RESOLVE      9  /* up, disambiguating distributed
+					  operations (import, rename, etc.) */
+#define CEPH_MDS_STATE_RECONNECT    10 /* up, reconnect to clients */
+#define CEPH_MDS_STATE_REJOIN       11 /* up, rejoining distributed cache */
+#define CEPH_MDS_STATE_CLIENTREPLAY 12 /* up, replaying client operations */
+#define CEPH_MDS_STATE_ACTIVE       13 /* up, active */
+#define CEPH_MDS_STATE_STOPPING     14 /* up, but exporting metadata */
+
+extern const char *ceph_mds_state_name(int s);
+
+
+/*
+ * metadata lock types.
+ *  - these are bitmasks.. we can compose them
+ *  - they also define the lock ordering by the MDS
+ *  - a few of these are internal to the mds
+ */
+#define CEPH_LOCK_DVERSION    1
+#define CEPH_LOCK_DN          2
+#define CEPH_LOCK_ISNAP       16
+#define CEPH_LOCK_IVERSION    32    /* mds internal */
+#define CEPH_LOCK_IFILE       64
+#define CEPH_LOCK_IAUTH       128
+#define CEPH_LOCK_ILINK       256
+#define CEPH_LOCK_IDFT        512   /* dir frag tree */
+#define CEPH_LOCK_INEST       1024  /* mds internal */
+#define CEPH_LOCK_IXATTR      2048
+#define CEPH_LOCK_IFLOCK      4096  /* advisory file locks */
+#define CEPH_LOCK_INO         8192  /* immutable inode bits; not a lock */
+
+/* client_session ops */
+enum {
+	CEPH_SESSION_REQUEST_OPEN,
+	CEPH_SESSION_OPEN,
+	CEPH_SESSION_REQUEST_CLOSE,
+	CEPH_SESSION_CLOSE,
+	CEPH_SESSION_REQUEST_RENEWCAPS,
+	CEPH_SESSION_RENEWCAPS,
+	CEPH_SESSION_STALE,
+	CEPH_SESSION_RECALL_STATE,
+};
+
+extern const char *ceph_session_op_name(int op);
+
+struct ceph_mds_session_head {
+	__le32 op;
+	__le64 seq;
+	struct ceph_timespec stamp;
+	__le32 max_caps, max_leases;
+} __attribute__ ((packed));
+
+/* client_request */
+/*
+ * metadata ops.
+ *  & 0x001000 -> write op
+ *  & 0x010000 -> follow symlink (e.g. stat(), not lstat()).
+ &  & 0x100000 -> use weird ino/path trace
+ */
+#define CEPH_MDS_OP_WRITE        0x001000
+enum {
+	CEPH_MDS_OP_LOOKUP     = 0x00100,
+	CEPH_MDS_OP_GETATTR    = 0x00101,
+	CEPH_MDS_OP_LOOKUPHASH = 0x00102,
+	CEPH_MDS_OP_LOOKUPPARENT = 0x00103,
+
+	CEPH_MDS_OP_SETXATTR   = 0x01105,
+	CEPH_MDS_OP_RMXATTR    = 0x01106,
+	CEPH_MDS_OP_SETLAYOUT  = 0x01107,
+	CEPH_MDS_OP_SETATTR    = 0x01108,
+	CEPH_MDS_OP_SETFILELOCK= 0x01109,
+	CEPH_MDS_OP_GETFILELOCK= 0x00110,
+
+	CEPH_MDS_OP_MKNOD      = 0x01201,
+	CEPH_MDS_OP_LINK       = 0x01202,
+	CEPH_MDS_OP_UNLINK     = 0x01203,
+	CEPH_MDS_OP_RENAME     = 0x01204,
+	CEPH_MDS_OP_MKDIR      = 0x01220,
+	CEPH_MDS_OP_RMDIR      = 0x01221,
+	CEPH_MDS_OP_SYMLINK    = 0x01222,
+
+	CEPH_MDS_OP_CREATE     = 0x01301,
+	CEPH_MDS_OP_OPEN       = 0x00302,
+	CEPH_MDS_OP_READDIR    = 0x00305,
+
+	CEPH_MDS_OP_LOOKUPSNAP = 0x00400,
+	CEPH_MDS_OP_MKSNAP     = 0x01400,
+	CEPH_MDS_OP_RMSNAP     = 0x01401,
+	CEPH_MDS_OP_LSSNAP     = 0x00402,
+};
+
+extern const char *ceph_mds_op_name(int op);
+
+
+#define CEPH_SETATTR_MODE   1
+#define CEPH_SETATTR_UID    2
+#define CEPH_SETATTR_GID    4
+#define CEPH_SETATTR_MTIME  8
+#define CEPH_SETATTR_ATIME 16
+#define CEPH_SETATTR_SIZE  32
+#define CEPH_SETATTR_CTIME 64
+
+union ceph_mds_request_args {
+	struct {
+		__le32 mask;                 /* CEPH_CAP_* */
+	} __attribute__ ((packed)) getattr;
+	struct {
+		__le32 mode;
+		__le32 uid;
+		__le32 gid;
+		struct ceph_timespec mtime;
+		struct ceph_timespec atime;
+		__le64 size, old_size;       /* old_size needed by truncate */
+		__le32 mask;                 /* CEPH_SETATTR_* */
+	} __attribute__ ((packed)) setattr;
+	struct {
+		__le32 frag;                 /* which dir fragment */
+		__le32 max_entries;          /* how many dentries to grab */
+		__le32 max_bytes;
+	} __attribute__ ((packed)) readdir;
+	struct {
+		__le32 mode;
+		__le32 rdev;
+	} __attribute__ ((packed)) mknod;
+	struct {
+		__le32 mode;
+	} __attribute__ ((packed)) mkdir;
+	struct {
+		__le32 flags;
+		__le32 mode;
+		__le32 stripe_unit;          /* layout for newly created file */
+		__le32 stripe_count;         /* ... */
+		__le32 object_size;
+		__le32 file_replication;
+		__le32 preferred;
+	} __attribute__ ((packed)) open;
+	struct {
+		__le32 flags;
+	} __attribute__ ((packed)) setxattr;
+	struct {
+		struct ceph_file_layout layout;
+	} __attribute__ ((packed)) setlayout;
+	struct {
+		__u8 rule; /* currently fcntl or flock */
+		__u8 type; /* shared, exclusive, remove*/
+		__le64 pid; /* process id requesting the lock */
+		__le64 pid_namespace;
+		__le64 start; /* initial location to lock */
+		__le64 length; /* num bytes to lock from start */
+		__u8 wait; /* will caller wait for lock to become available? */
+	} __attribute__ ((packed)) filelock_change;
+} __attribute__ ((packed));
+
+#define CEPH_MDS_FLAG_REPLAY        1  /* this is a replayed op */
+#define CEPH_MDS_FLAG_WANT_DENTRY   2  /* want dentry in reply */
+
+struct ceph_mds_request_head {
+	__le64 oldest_client_tid;
+	__le32 mdsmap_epoch;           /* on client */
+	__le32 flags;                  /* CEPH_MDS_FLAG_* */
+	__u8 num_retry, num_fwd;       /* count retry, fwd attempts */
+	__le16 num_releases;           /* # include cap/lease release records */
+	__le32 op;                     /* mds op code */
+	__le32 caller_uid, caller_gid;
+	__le64 ino;                    /* use this ino for openc, mkdir, mknod,
+					  etc. (if replaying) */
+	union ceph_mds_request_args args;
+} __attribute__ ((packed));
+
+/* cap/lease release record */
+struct ceph_mds_request_release {
+	__le64 ino, cap_id;            /* ino and unique cap id */
+	__le32 caps, wanted;           /* new issued, wanted */
+	__le32 seq, issue_seq, mseq;
+	__le32 dname_seq;              /* if releasing a dentry lease, a */
+	__le32 dname_len;              /* string follows. */
+} __attribute__ ((packed));
+
+/* client reply */
+struct ceph_mds_reply_head {
+	__le32 op;
+	__le32 result;
+	__le32 mdsmap_epoch;
+	__u8 safe;                     /* true if committed to disk */
+	__u8 is_dentry, is_target;     /* true if dentry, target inode records
+					  are included with reply */
+} __attribute__ ((packed));
+
+/* one for each node split */
+struct ceph_frag_tree_split {
+	__le32 frag;                   /* this frag splits... */
+	__le32 by;                     /* ...by this many bits */
+} __attribute__ ((packed));
+
+struct ceph_frag_tree_head {
+	__le32 nsplits;                /* num ceph_frag_tree_split records */
+	struct ceph_frag_tree_split splits[];
+} __attribute__ ((packed));
+
+/* capability issue, for bundling with mds reply */
+struct ceph_mds_reply_cap {
+	__le32 caps, wanted;           /* caps issued, wanted */
+	__le64 cap_id;
+	__le32 seq, mseq;
+	__le64 realm;                  /* snap realm */
+	__u8 flags;                    /* CEPH_CAP_FLAG_* */
+} __attribute__ ((packed));
+
+#define CEPH_CAP_FLAG_AUTH  1          /* cap is issued by auth mds */
+
+/* inode record, for bundling with mds reply */
+struct ceph_mds_reply_inode {
+	__le64 ino;
+	__le64 snapid;
+	__le32 rdev;
+	__le64 version;                /* inode version */
+	__le64 xattr_version;          /* version for xattr blob */
+	struct ceph_mds_reply_cap cap; /* caps issued for this inode */
+	struct ceph_file_layout layout;
+	struct ceph_timespec ctime, mtime, atime;
+	__le32 time_warp_seq;
+	__le64 size, max_size, truncate_size;
+	__le32 truncate_seq;
+	__le32 mode, uid, gid;
+	__le32 nlink;
+	__le64 files, subdirs, rbytes, rfiles, rsubdirs;  /* dir stats */
+	struct ceph_timespec rctime;
+	struct ceph_frag_tree_head fragtree;  /* (must be at end of struct) */
+} __attribute__ ((packed));
+/* followed by frag array, then symlink string, then xattr blob */
+
+/* reply_lease follows dname, and reply_inode */
+struct ceph_mds_reply_lease {
+	__le16 mask;            /* lease type(s) */
+	__le32 duration_ms;     /* lease duration */
+	__le32 seq;
+} __attribute__ ((packed));
+
+struct ceph_mds_reply_dirfrag {
+	__le32 frag;            /* fragment */
+	__le32 auth;            /* auth mds, if this is a delegation point */
+	__le32 ndist;           /* number of mds' this is replicated on */
+	__le32 dist[];
+} __attribute__ ((packed));
+
+#define CEPH_LOCK_FCNTL    1
+#define CEPH_LOCK_FLOCK    2
+
+#define CEPH_LOCK_SHARED   1
+#define CEPH_LOCK_EXCL     2
+#define CEPH_LOCK_UNLOCK   4
+
+struct ceph_filelock {
+	__le64 start;/* file offset to start lock at */
+	__le64 length; /* num bytes to lock; 0 for all following start */
+	__le64 client; /* which client holds the lock */
+	__le64 pid; /* process id holding the lock on the client */
+	__le64 pid_namespace;
+	__u8 type; /* shared lock, exclusive lock, or unlock */
+} __attribute__ ((packed));
+
+
+/* file access modes */
+#define CEPH_FILE_MODE_PIN        0
+#define CEPH_FILE_MODE_RD         1
+#define CEPH_FILE_MODE_WR         2
+#define CEPH_FILE_MODE_RDWR       3  /* RD | WR */
+#define CEPH_FILE_MODE_LAZY       4  /* lazy io */
+#define CEPH_FILE_MODE_NUM        8  /* bc these are bit fields.. mostly */
+
+int ceph_flags_to_mode(int flags);
+
+
+/* capability bits */
+#define CEPH_CAP_PIN         1  /* no specific capabilities beyond the pin */
+
+/* generic cap bits */
+#define CEPH_CAP_GSHARED     1  /* client can reads */
+#define CEPH_CAP_GEXCL       2  /* client can read and update */
+#define CEPH_CAP_GCACHE      4  /* (file) client can cache reads */
+#define CEPH_CAP_GRD         8  /* (file) client can read */
+#define CEPH_CAP_GWR        16  /* (file) client can write */
+#define CEPH_CAP_GBUFFER    32  /* (file) client can buffer writes */
+#define CEPH_CAP_GWREXTEND  64  /* (file) client can extend EOF */
+#define CEPH_CAP_GLAZYIO   128  /* (file) client can perform lazy io */
+
+/* per-lock shift */
+#define CEPH_CAP_SAUTH      2
+#define CEPH_CAP_SLINK      4
+#define CEPH_CAP_SXATTR     6
+#define CEPH_CAP_SFILE      8
+#define CEPH_CAP_SFLOCK    20 
+
+#define CEPH_CAP_BITS       22
+
+/* composed values */
+#define CEPH_CAP_AUTH_SHARED  (CEPH_CAP_GSHARED  << CEPH_CAP_SAUTH)
+#define CEPH_CAP_AUTH_EXCL     (CEPH_CAP_GEXCL     << CEPH_CAP_SAUTH)
+#define CEPH_CAP_LINK_SHARED  (CEPH_CAP_GSHARED  << CEPH_CAP_SLINK)
+#define CEPH_CAP_LINK_EXCL     (CEPH_CAP_GEXCL     << CEPH_CAP_SLINK)
+#define CEPH_CAP_XATTR_SHARED (CEPH_CAP_GSHARED  << CEPH_CAP_SXATTR)
+#define CEPH_CAP_XATTR_EXCL    (CEPH_CAP_GEXCL     << CEPH_CAP_SXATTR)
+#define CEPH_CAP_FILE(x)    (x << CEPH_CAP_SFILE)
+#define CEPH_CAP_FILE_SHARED   (CEPH_CAP_GSHARED   << CEPH_CAP_SFILE)
+#define CEPH_CAP_FILE_EXCL     (CEPH_CAP_GEXCL     << CEPH_CAP_SFILE)
+#define CEPH_CAP_FILE_CACHE    (CEPH_CAP_GCACHE    << CEPH_CAP_SFILE)
+#define CEPH_CAP_FILE_RD       (CEPH_CAP_GRD       << CEPH_CAP_SFILE)
+#define CEPH_CAP_FILE_WR       (CEPH_CAP_GWR       << CEPH_CAP_SFILE)
+#define CEPH_CAP_FILE_BUFFER   (CEPH_CAP_GBUFFER   << CEPH_CAP_SFILE)
+#define CEPH_CAP_FILE_WREXTEND (CEPH_CAP_GWREXTEND << CEPH_CAP_SFILE)
+#define CEPH_CAP_FILE_LAZYIO   (CEPH_CAP_GLAZYIO   << CEPH_CAP_SFILE)
+#define CEPH_CAP_FLOCK_SHARED  (CEPH_CAP_GSHARED   << CEPH_CAP_SFLOCK)
+#define CEPH_CAP_FLOCK_EXCL    (CEPH_CAP_GEXCL     << CEPH_CAP_SFLOCK)
+
+
+/* cap masks (for getattr) */
+#define CEPH_STAT_CAP_INODE    CEPH_CAP_PIN
+#define CEPH_STAT_CAP_TYPE     CEPH_CAP_PIN  /* mode >> 12 */
+#define CEPH_STAT_CAP_SYMLINK  CEPH_CAP_PIN
+#define CEPH_STAT_CAP_UID      CEPH_CAP_AUTH_SHARED
+#define CEPH_STAT_CAP_GID      CEPH_CAP_AUTH_SHARED
+#define CEPH_STAT_CAP_MODE     CEPH_CAP_AUTH_SHARED
+#define CEPH_STAT_CAP_NLINK    CEPH_CAP_LINK_SHARED
+#define CEPH_STAT_CAP_LAYOUT   CEPH_CAP_FILE_SHARED
+#define CEPH_STAT_CAP_MTIME    CEPH_CAP_FILE_SHARED
+#define CEPH_STAT_CAP_SIZE     CEPH_CAP_FILE_SHARED
+#define CEPH_STAT_CAP_ATIME    CEPH_CAP_FILE_SHARED  /* fixme */
+#define CEPH_STAT_CAP_XATTR    CEPH_CAP_XATTR_SHARED
+#define CEPH_STAT_CAP_INODE_ALL (CEPH_CAP_PIN |			\
+				 CEPH_CAP_AUTH_SHARED |	\
+				 CEPH_CAP_LINK_SHARED |	\
+				 CEPH_CAP_FILE_SHARED |	\
+				 CEPH_CAP_XATTR_SHARED)
+
+#define CEPH_CAP_ANY_SHARED (CEPH_CAP_AUTH_SHARED |			\
+			      CEPH_CAP_LINK_SHARED |			\
+			      CEPH_CAP_XATTR_SHARED |			\
+			      CEPH_CAP_FILE_SHARED)
+#define CEPH_CAP_ANY_RD   (CEPH_CAP_ANY_SHARED | CEPH_CAP_FILE_RD |	\
+			   CEPH_CAP_FILE_CACHE)
+
+#define CEPH_CAP_ANY_EXCL (CEPH_CAP_AUTH_EXCL |		\
+			   CEPH_CAP_LINK_EXCL |		\
+			   CEPH_CAP_XATTR_EXCL |	\
+			   CEPH_CAP_FILE_EXCL)
+#define CEPH_CAP_ANY_FILE_WR (CEPH_CAP_FILE_WR | CEPH_CAP_FILE_BUFFER |	\
+			      CEPH_CAP_FILE_EXCL)
+#define CEPH_CAP_ANY_WR   (CEPH_CAP_ANY_EXCL | CEPH_CAP_ANY_FILE_WR)
+#define CEPH_CAP_ANY      (CEPH_CAP_ANY_RD | CEPH_CAP_ANY_EXCL | \
+			   CEPH_CAP_ANY_FILE_WR | CEPH_CAP_FILE_LAZYIO | \
+			   CEPH_CAP_PIN)
+
+#define CEPH_CAP_LOCKS (CEPH_LOCK_IFILE | CEPH_LOCK_IAUTH | CEPH_LOCK_ILINK | \
+			CEPH_LOCK_IXATTR)
+
+int ceph_caps_for_mode(int mode);
+
+enum {
+	CEPH_CAP_OP_GRANT,         /* mds->client grant */
+	CEPH_CAP_OP_REVOKE,        /* mds->client revoke */
+	CEPH_CAP_OP_TRUNC,         /* mds->client trunc notify */
+	CEPH_CAP_OP_EXPORT,        /* mds has exported the cap */
+	CEPH_CAP_OP_IMPORT,        /* mds has imported the cap */
+	CEPH_CAP_OP_UPDATE,        /* client->mds update */
+	CEPH_CAP_OP_DROP,          /* client->mds drop cap bits */
+	CEPH_CAP_OP_FLUSH,         /* client->mds cap writeback */
+	CEPH_CAP_OP_FLUSH_ACK,     /* mds->client flushed */
+	CEPH_CAP_OP_FLUSHSNAP,     /* client->mds flush snapped metadata */
+	CEPH_CAP_OP_FLUSHSNAP_ACK, /* mds->client flushed snapped metadata */
+	CEPH_CAP_OP_RELEASE,       /* client->mds release (clean) cap */
+	CEPH_CAP_OP_RENEW,         /* client->mds renewal request */
+};
+
+extern const char *ceph_cap_op_name(int op);
+
+/*
+ * caps message, used for capability callbacks, acks, requests, etc.
+ */
+struct ceph_mds_caps {
+	__le32 op;                  /* CEPH_CAP_OP_* */
+	__le64 ino, realm;
+	__le64 cap_id;
+	__le32 seq, issue_seq;
+	__le32 caps, wanted, dirty; /* latest issued/wanted/dirty */
+	__le32 migrate_seq;
+	__le64 snap_follows;
+	__le32 snap_trace_len;
+
+	/* authlock */
+	__le32 uid, gid, mode;
+
+	/* linklock */
+	__le32 nlink;
+
+	/* xattrlock */
+	__le32 xattr_len;
+	__le64 xattr_version;
+
+	/* filelock */
+	__le64 size, max_size, truncate_size;
+	__le32 truncate_seq;
+	struct ceph_timespec mtime, atime, ctime;
+	struct ceph_file_layout layout;
+	__le32 time_warp_seq;
+} __attribute__ ((packed));
+
+/* cap release msg head */
+struct ceph_mds_cap_release {
+	__le32 num;                /* number of cap_items that follow */
+} __attribute__ ((packed));
+
+struct ceph_mds_cap_item {
+	__le64 ino;
+	__le64 cap_id;
+	__le32 migrate_seq, seq;
+} __attribute__ ((packed));
+
+#define CEPH_MDS_LEASE_REVOKE           1  /*    mds  -> client */
+#define CEPH_MDS_LEASE_RELEASE          2  /* client  -> mds    */
+#define CEPH_MDS_LEASE_RENEW            3  /* client <-> mds    */
+#define CEPH_MDS_LEASE_REVOKE_ACK       4  /* client  -> mds    */
+
+extern const char *ceph_lease_op_name(int o);
+
+/* lease msg header */
+struct ceph_mds_lease {
+	__u8 action;            /* CEPH_MDS_LEASE_* */
+	__le16 mask;            /* which lease */
+	__le64 ino;
+	__le64 first, last;     /* snap range */
+	__le32 seq;
+	__le32 duration_ms;     /* duration of renewal */
+} __attribute__ ((packed));
+/* followed by a __le32+string for dname */
+
+/* client reconnect */
+struct ceph_mds_cap_reconnect {
+	__le64 cap_id;
+	__le32 wanted;
+	__le32 issued;
+	__le64 snaprealm;
+	__le64 pathbase;        /* base ino for our path to this ino */
+	__le32 flock_len;       /* size of flock state blob, if any */
+} __attribute__ ((packed));
+/* followed by flock blob */
+
+struct ceph_mds_cap_reconnect_v1 {
+	__le64 cap_id;
+	__le32 wanted;
+	__le32 issued;
+	__le64 size;
+	struct ceph_timespec mtime, atime;
+	__le64 snaprealm;
+	__le64 pathbase;        /* base ino for our path to this ino */
+} __attribute__ ((packed));
+
+struct ceph_mds_snaprealm_reconnect {
+	__le64 ino;     /* snap realm base */
+	__le64 seq;     /* snap seq for this snap realm */
+	__le64 parent;  /* parent realm */
+} __attribute__ ((packed));
+
+/*
+ * snaps
+ */
+enum {
+	CEPH_SNAP_OP_UPDATE,  /* CREATE or DESTROY */
+	CEPH_SNAP_OP_CREATE,
+	CEPH_SNAP_OP_DESTROY,
+	CEPH_SNAP_OP_SPLIT,
+};
+
+extern const char *ceph_snap_op_name(int o);
+
+/* snap msg header */
+struct ceph_mds_snap_head {
+	__le32 op;                /* CEPH_SNAP_OP_* */
+	__le64 split;             /* ino to split off, if any */
+	__le32 num_split_inos;    /* # inos belonging to new child realm */
+	__le32 num_split_realms;  /* # child realms udner new child realm */
+	__le32 trace_len;         /* size of snap trace blob */
+} __attribute__ ((packed));
+/* followed by split ino list, then split realms, then the trace blob */
+
+/*
+ * encode info about a snaprealm, as viewed by a client
+ */
+struct ceph_mds_snap_realm {
+	__le64 ino;           /* ino */
+	__le64 created;       /* snap: when created */
+	__le64 parent;        /* ino: parent realm */
+	__le64 parent_since;  /* snap: same parent since */
+	__le64 seq;           /* snap: version */
+	__le32 num_snaps;
+	__le32 num_prior_parent_snaps;
+} __attribute__ ((packed));
+/* followed by my snap list, then prior parent snap list */
+
+#endif
diff --git a/include/linux/ceph/ceph_hash.h b/include/linux/ceph/ceph_hash.h
new file mode 100644
index 000000000000..d099c3f90236
--- /dev/null
+++ b/include/linux/ceph/ceph_hash.h
@@ -0,0 +1,13 @@
+#ifndef FS_CEPH_HASH_H
+#define FS_CEPH_HASH_H
+
+#define CEPH_STR_HASH_LINUX      0x1  /* linux dcache hash */
+#define CEPH_STR_HASH_RJENKINS   0x2  /* robert jenkins' */
+
+extern unsigned ceph_str_hash_linux(const char *s, unsigned len);
+extern unsigned ceph_str_hash_rjenkins(const char *s, unsigned len);
+
+extern unsigned ceph_str_hash(int type, const char *s, unsigned len);
+extern const char *ceph_str_hash_name(int type);
+
+#endif
diff --git a/include/linux/ceph/debugfs.h b/include/linux/ceph/debugfs.h
new file mode 100644
index 000000000000..2a79702e092b
--- /dev/null
+++ b/include/linux/ceph/debugfs.h
@@ -0,0 +1,33 @@
+#ifndef _FS_CEPH_DEBUGFS_H
+#define _FS_CEPH_DEBUGFS_H
+
+#include "ceph_debug.h"
+#include "types.h"
+
+#define CEPH_DEFINE_SHOW_FUNC(name)					\
+static int name##_open(struct inode *inode, struct file *file)		\
+{									\
+	struct seq_file *sf;						\
+	int ret;							\
+									\
+	ret = single_open(file, name, NULL);				\
+	sf = file->private_data;					\
+	sf->private = inode->i_private;					\
+	return ret;							\
+}									\
+									\
+static const struct file_operations name##_fops = {			\
+	.open		= name##_open,					\
+	.read		= seq_read,					\
+	.llseek		= seq_lseek,					\
+	.release	= single_release,				\
+};
+
+/* debugfs.c */
+extern int ceph_debugfs_init(void);
+extern void ceph_debugfs_cleanup(void);
+extern int ceph_debugfs_client_init(struct ceph_client *client);
+extern void ceph_debugfs_client_cleanup(struct ceph_client *client);
+
+#endif
+
diff --git a/include/linux/ceph/decode.h b/include/linux/ceph/decode.h
new file mode 100644
index 000000000000..c5b6939fb32a
--- /dev/null
+++ b/include/linux/ceph/decode.h
@@ -0,0 +1,201 @@
+#ifndef __CEPH_DECODE_H
+#define __CEPH_DECODE_H
+
+#include <asm/unaligned.h>
+#include <linux/time.h>
+
+#include "types.h"
+
+/*
+ * in all cases,
+ *   void **p     pointer to position pointer
+ *   void *end    pointer to end of buffer (last byte + 1)
+ */
+
+static inline u64 ceph_decode_64(void **p)
+{
+	u64 v = get_unaligned_le64(*p);
+	*p += sizeof(u64);
+	return v;
+}
+static inline u32 ceph_decode_32(void **p)
+{
+	u32 v = get_unaligned_le32(*p);
+	*p += sizeof(u32);
+	return v;
+}
+static inline u16 ceph_decode_16(void **p)
+{
+	u16 v = get_unaligned_le16(*p);
+	*p += sizeof(u16);
+	return v;
+}
+static inline u8 ceph_decode_8(void **p)
+{
+	u8 v = *(u8 *)*p;
+	(*p)++;
+	return v;
+}
+static inline void ceph_decode_copy(void **p, void *pv, size_t n)
+{
+	memcpy(pv, *p, n);
+	*p += n;
+}
+
+/*
+ * bounds check input.
+ */
+#define ceph_decode_need(p, end, n, bad)		\
+	do {						\
+		if (unlikely(*(p) + (n) > (end))) 	\
+			goto bad;			\
+	} while (0)
+
+#define ceph_decode_64_safe(p, end, v, bad)			\
+	do {							\
+		ceph_decode_need(p, end, sizeof(u64), bad);	\
+		v = ceph_decode_64(p);				\
+	} while (0)
+#define ceph_decode_32_safe(p, end, v, bad)			\
+	do {							\
+		ceph_decode_need(p, end, sizeof(u32), bad);	\
+		v = ceph_decode_32(p);				\
+	} while (0)
+#define ceph_decode_16_safe(p, end, v, bad)			\
+	do {							\
+		ceph_decode_need(p, end, sizeof(u16), bad);	\
+		v = ceph_decode_16(p);				\
+	} while (0)
+#define ceph_decode_8_safe(p, end, v, bad)			\
+	do {							\
+		ceph_decode_need(p, end, sizeof(u8), bad);	\
+		v = ceph_decode_8(p);				\
+	} while (0)
+
+#define ceph_decode_copy_safe(p, end, pv, n, bad)		\
+	do {							\
+		ceph_decode_need(p, end, n, bad);		\
+		ceph_decode_copy(p, pv, n);			\
+	} while (0)
+
+/*
+ * struct ceph_timespec <-> struct timespec
+ */
+static inline void ceph_decode_timespec(struct timespec *ts,
+					const struct ceph_timespec *tv)
+{
+	ts->tv_sec = le32_to_cpu(tv->tv_sec);
+	ts->tv_nsec = le32_to_cpu(tv->tv_nsec);
+}
+static inline void ceph_encode_timespec(struct ceph_timespec *tv,
+					const struct timespec *ts)
+{
+	tv->tv_sec = cpu_to_le32(ts->tv_sec);
+	tv->tv_nsec = cpu_to_le32(ts->tv_nsec);
+}
+
+/*
+ * sockaddr_storage <-> ceph_sockaddr
+ */
+static inline void ceph_encode_addr(struct ceph_entity_addr *a)
+{
+	__be16 ss_family = htons(a->in_addr.ss_family);
+	a->in_addr.ss_family = *(__u16 *)&ss_family;
+}
+static inline void ceph_decode_addr(struct ceph_entity_addr *a)
+{
+	__be16 ss_family = *(__be16 *)&a->in_addr.ss_family;
+	a->in_addr.ss_family = ntohs(ss_family);
+	WARN_ON(a->in_addr.ss_family == 512);
+}
+
+/*
+ * encoders
+ */
+static inline void ceph_encode_64(void **p, u64 v)
+{
+	put_unaligned_le64(v, (__le64 *)*p);
+	*p += sizeof(u64);
+}
+static inline void ceph_encode_32(void **p, u32 v)
+{
+	put_unaligned_le32(v, (__le32 *)*p);
+	*p += sizeof(u32);
+}
+static inline void ceph_encode_16(void **p, u16 v)
+{
+	put_unaligned_le16(v, (__le16 *)*p);
+	*p += sizeof(u16);
+}
+static inline void ceph_encode_8(void **p, u8 v)
+{
+	*(u8 *)*p = v;
+	(*p)++;
+}
+static inline void ceph_encode_copy(void **p, const void *s, int len)
+{
+	memcpy(*p, s, len);
+	*p += len;
+}
+
+/*
+ * filepath, string encoders
+ */
+static inline void ceph_encode_filepath(void **p, void *end,
+					u64 ino, const char *path)
+{
+	u32 len = path ? strlen(path) : 0;
+	BUG_ON(*p + sizeof(ino) + sizeof(len) + len > end);
+	ceph_encode_8(p, 1);
+	ceph_encode_64(p, ino);
+	ceph_encode_32(p, len);
+	if (len)
+		memcpy(*p, path, len);
+	*p += len;
+}
+
+static inline void ceph_encode_string(void **p, void *end,
+				      const char *s, u32 len)
+{
+	BUG_ON(*p + sizeof(len) + len > end);
+	ceph_encode_32(p, len);
+	if (len)
+		memcpy(*p, s, len);
+	*p += len;
+}
+
+#define ceph_encode_need(p, end, n, bad)		\
+	do {						\
+		if (unlikely(*(p) + (n) > (end))) 	\
+			goto bad;			\
+	} while (0)
+
+#define ceph_encode_64_safe(p, end, v, bad)			\
+	do {							\
+		ceph_encode_need(p, end, sizeof(u64), bad);	\
+		ceph_encode_64(p, v);				\
+	} while (0)
+#define ceph_encode_32_safe(p, end, v, bad)			\
+	do {							\
+		ceph_encode_need(p, end, sizeof(u32), bad);	\
+		ceph_encode_32(p, v);			\
+	} while (0)
+#define ceph_encode_16_safe(p, end, v, bad)			\
+	do {							\
+		ceph_encode_need(p, end, sizeof(u16), bad);	\
+		ceph_encode_16(p, v);			\
+	} while (0)
+
+#define ceph_encode_copy_safe(p, end, pv, n, bad)		\
+	do {							\
+		ceph_encode_need(p, end, n, bad);		\
+		ceph_encode_copy(p, pv, n);			\
+	} while (0)
+#define ceph_encode_string_safe(p, end, s, n, bad)		\
+	do {							\
+		ceph_encode_need(p, end, n, bad);		\
+		ceph_encode_string(p, end, s, n);		\
+	} while (0)
+
+
+#endif
diff --git a/include/linux/ceph/libceph.h b/include/linux/ceph/libceph.h
new file mode 100644
index 000000000000..f22b2e941686
--- /dev/null
+++ b/include/linux/ceph/libceph.h
@@ -0,0 +1,249 @@
+#ifndef _FS_CEPH_LIBCEPH_H
+#define _FS_CEPH_LIBCEPH_H
+
+#include "ceph_debug.h"
+
+#include <asm/unaligned.h>
+#include <linux/backing-dev.h>
+#include <linux/completion.h>
+#include <linux/exportfs.h>
+#include <linux/fs.h>
+#include <linux/mempool.h>
+#include <linux/pagemap.h>
+#include <linux/wait.h>
+#include <linux/writeback.h>
+#include <linux/slab.h>
+
+#include "types.h"
+#include "messenger.h"
+#include "msgpool.h"
+#include "mon_client.h"
+#include "osd_client.h"
+#include "ceph_fs.h"
+
+/*
+ * Supported features
+ */
+#define CEPH_FEATURE_SUPPORTED_DEFAULT CEPH_FEATURE_NOSRCADDR
+#define CEPH_FEATURE_REQUIRED_DEFAULT  CEPH_FEATURE_NOSRCADDR
+
+/*
+ * mount options
+ */
+#define CEPH_OPT_FSID             (1<<0)
+#define CEPH_OPT_NOSHARE          (1<<1) /* don't share client with other sbs */
+#define CEPH_OPT_MYIP             (1<<2) /* specified my ip */
+#define CEPH_OPT_NOCRC            (1<<3) /* no data crc on writes */
+
+#define CEPH_OPT_DEFAULT   (0);
+
+#define ceph_set_opt(client, opt) \
+	(client)->options->flags |= CEPH_OPT_##opt;
+#define ceph_test_opt(client, opt) \
+	(!!((client)->options->flags & CEPH_OPT_##opt))
+
+struct ceph_options {
+	int flags;
+	struct ceph_fsid fsid;
+	struct ceph_entity_addr my_addr;
+	int mount_timeout;
+	int osd_idle_ttl;
+	int osd_timeout;
+	int osd_keepalive_timeout;
+
+	/*
+	 * any type that can't be simply compared or doesn't need need
+	 * to be compared should go beyond this point,
+	 * ceph_compare_options() should be updated accordingly
+	 */
+
+	struct ceph_entity_addr *mon_addr; /* should be the first
+					      pointer type of args */
+	int num_mon;
+	char *name;
+	char *secret;
+};
+
+/*
+ * defaults
+ */
+#define CEPH_MOUNT_TIMEOUT_DEFAULT  60
+#define CEPH_OSD_TIMEOUT_DEFAULT    60  /* seconds */
+#define CEPH_OSD_KEEPALIVE_DEFAULT  5
+#define CEPH_OSD_IDLE_TTL_DEFAULT    60
+#define CEPH_MOUNT_RSIZE_DEFAULT    (512*1024) /* readahead */
+
+#define CEPH_MSG_MAX_FRONT_LEN	(16*1024*1024)
+#define CEPH_MSG_MAX_DATA_LEN	(16*1024*1024)
+
+#define CEPH_AUTH_NAME_DEFAULT   "guest"
+
+/*
+ * Delay telling the MDS we no longer want caps, in case we reopen
+ * the file.  Delay a minimum amount of time, even if we send a cap
+ * message for some other reason.  Otherwise, take the oppotunity to
+ * update the mds to avoid sending another message later.
+ */
+#define CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT      5  /* cap release delay */
+#define CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT     60  /* cap release delay */
+
+#define CEPH_CAP_RELEASE_SAFETY_DEFAULT        (CEPH_CAPS_PER_RELEASE * 4)
+
+/* mount state */
+enum {
+	CEPH_MOUNT_MOUNTING,
+	CEPH_MOUNT_MOUNTED,
+	CEPH_MOUNT_UNMOUNTING,
+	CEPH_MOUNT_UNMOUNTED,
+	CEPH_MOUNT_SHUTDOWN,
+};
+
+/*
+ * subtract jiffies
+ */
+static inline unsigned long time_sub(unsigned long a, unsigned long b)
+{
+	BUG_ON(time_after(b, a));
+	return (long)a - (long)b;
+}
+
+struct ceph_mds_client;
+
+/*
+ * per client state
+ *
+ * possibly shared by multiple mount points, if they are
+ * mounting the same ceph filesystem/cluster.
+ */
+struct ceph_client {
+	struct ceph_fsid fsid;
+	bool have_fsid;
+
+	void *private;
+
+	struct ceph_options *options;
+
+	struct mutex mount_mutex;      /* serialize mount attempts */
+	wait_queue_head_t auth_wq;
+	int auth_err;
+
+	int (*extra_mon_dispatch)(struct ceph_client *, struct ceph_msg *);
+
+	u32 supported_features;
+	u32 required_features;
+
+	struct ceph_messenger *msgr;   /* messenger instance */
+	struct ceph_mon_client monc;
+	struct ceph_osd_client osdc;
+
+#ifdef CONFIG_DEBUG_FS
+	struct dentry *debugfs_dir;
+	struct dentry *debugfs_monmap;
+	struct dentry *debugfs_osdmap;
+#endif
+};
+
+
+
+/*
+ * snapshots
+ */
+
+/*
+ * A "snap context" is the set of existing snapshots when we
+ * write data.  It is used by the OSD to guide its COW behavior.
+ *
+ * The ceph_snap_context is refcounted, and attached to each dirty
+ * page, indicating which context the dirty data belonged when it was
+ * dirtied.
+ */
+struct ceph_snap_context {
+	atomic_t nref;
+	u64 seq;
+	int num_snaps;
+	u64 snaps[];
+};
+
+static inline struct ceph_snap_context *
+ceph_get_snap_context(struct ceph_snap_context *sc)
+{
+	/*
+	printk("get_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
+	       atomic_read(&sc->nref)+1);
+	*/
+	if (sc)
+		atomic_inc(&sc->nref);
+	return sc;
+}
+
+static inline void ceph_put_snap_context(struct ceph_snap_context *sc)
+{
+	if (!sc)
+		return;
+	/*
+	printk("put_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
+	       atomic_read(&sc->nref)-1);
+	*/
+	if (atomic_dec_and_test(&sc->nref)) {
+		/*printk(" deleting snap_context %p\n", sc);*/
+		kfree(sc);
+	}
+}
+
+/*
+ * calculate the number of pages a given length and offset map onto,
+ * if we align the data.
+ */
+static inline int calc_pages_for(u64 off, u64 len)
+{
+	return ((off+len+PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT) -
+		(off >> PAGE_CACHE_SHIFT);
+}
+
+/* ceph_common.c */
+extern const char *ceph_msg_type_name(int type);
+extern int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid);
+extern struct kmem_cache *ceph_inode_cachep;
+extern struct kmem_cache *ceph_cap_cachep;
+extern struct kmem_cache *ceph_dentry_cachep;
+extern struct kmem_cache *ceph_file_cachep;
+
+extern int ceph_parse_options(struct ceph_options **popt, char *options,
+			      const char *dev_name, const char *dev_name_end,
+			      int (*parse_extra_token)(char *c, void *private),
+			      void *private);
+extern void ceph_destroy_options(struct ceph_options *opt);
+extern int ceph_compare_options(struct ceph_options *new_opt,
+				struct ceph_client *client);
+extern struct ceph_client *ceph_create_client(struct ceph_options *opt,
+					      void *private);
+extern u64 ceph_client_id(struct ceph_client *client);
+extern void ceph_destroy_client(struct ceph_client *client);
+extern int __ceph_open_session(struct ceph_client *client,
+			       unsigned long started);
+extern int ceph_open_session(struct ceph_client *client);
+
+/* pagevec.c */
+extern void ceph_release_page_vector(struct page **pages, int num_pages);
+
+extern struct page **ceph_get_direct_page_vector(const char __user *data,
+					    int num_pages,
+					    loff_t off, size_t len);
+extern void ceph_put_page_vector(struct page **pages, int num_pages);
+extern void ceph_release_page_vector(struct page **pages, int num_pages);
+extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags);
+extern int ceph_copy_user_to_page_vector(struct page **pages,
+					 const char __user *data,
+					 loff_t off, size_t len);
+extern int ceph_copy_to_page_vector(struct page **pages,
+				    const char *data,
+				    loff_t off, size_t len);
+extern int ceph_copy_from_page_vector(struct page **pages,
+				    char *data,
+				    loff_t off, size_t len);
+extern int ceph_copy_page_vector_to_user(struct page **pages, char __user *data,
+				    loff_t off, size_t len);
+extern void ceph_zero_page_vector_range(int off, int len, struct page **pages);
+
+
+#endif /* _FS_CEPH_SUPER_H */
diff --git a/include/linux/ceph/mdsmap.h b/include/linux/ceph/mdsmap.h
new file mode 100644
index 000000000000..4c5cb0880bba
--- /dev/null
+++ b/include/linux/ceph/mdsmap.h
@@ -0,0 +1,62 @@
+#ifndef _FS_CEPH_MDSMAP_H
+#define _FS_CEPH_MDSMAP_H
+
+#include "types.h"
+
+/*
+ * mds map - describe servers in the mds cluster.
+ *
+ * we limit fields to those the client actually xcares about
+ */
+struct ceph_mds_info {
+	u64 global_id;
+	struct ceph_entity_addr addr;
+	s32 state;
+	int num_export_targets;
+	bool laggy;
+	u32 *export_targets;
+};
+
+struct ceph_mdsmap {
+	u32 m_epoch, m_client_epoch, m_last_failure;
+	u32 m_root;
+	u32 m_session_timeout;          /* seconds */
+	u32 m_session_autoclose;        /* seconds */
+	u64 m_max_file_size;
+	u32 m_max_mds;                  /* size of m_addr, m_state arrays */
+	struct ceph_mds_info *m_info;
+
+	/* which object pools file data can be stored in */
+	int m_num_data_pg_pools;
+	u32 *m_data_pg_pools;
+	u32 m_cas_pg_pool;
+};
+
+static inline struct ceph_entity_addr *
+ceph_mdsmap_get_addr(struct ceph_mdsmap *m, int w)
+{
+	if (w >= m->m_max_mds)
+		return NULL;
+	return &m->m_info[w].addr;
+}
+
+static inline int ceph_mdsmap_get_state(struct ceph_mdsmap *m, int w)
+{
+	BUG_ON(w < 0);
+	if (w >= m->m_max_mds)
+		return CEPH_MDS_STATE_DNE;
+	return m->m_info[w].state;
+}
+
+static inline bool ceph_mdsmap_is_laggy(struct ceph_mdsmap *m, int w)
+{
+	if (w >= 0 && w < m->m_max_mds)
+		return m->m_info[w].laggy;
+	return false;
+}
+
+extern int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m);
+extern struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end);
+extern void ceph_mdsmap_destroy(struct ceph_mdsmap *m);
+
+#endif
diff --git a/include/linux/ceph/messenger.h b/include/linux/ceph/messenger.h
new file mode 100644
index 000000000000..5956d62c3057
--- /dev/null
+++ b/include/linux/ceph/messenger.h
@@ -0,0 +1,261 @@
+#ifndef __FS_CEPH_MESSENGER_H
+#define __FS_CEPH_MESSENGER_H
+
+#include <linux/kref.h>
+#include <linux/mutex.h>
+#include <linux/net.h>
+#include <linux/radix-tree.h>
+#include <linux/uio.h>
+#include <linux/version.h>
+#include <linux/workqueue.h>
+
+#include "types.h"
+#include "buffer.h"
+
+struct ceph_msg;
+struct ceph_connection;
+
+extern struct workqueue_struct *ceph_msgr_wq;       /* receive work queue */
+
+/*
+ * Ceph defines these callbacks for handling connection events.
+ */
+struct ceph_connection_operations {
+	struct ceph_connection *(*get)(struct ceph_connection *);
+	void (*put)(struct ceph_connection *);
+
+	/* handle an incoming message. */
+	void (*dispatch) (struct ceph_connection *con, struct ceph_msg *m);
+
+	/* authorize an outgoing connection */
+	int (*get_authorizer) (struct ceph_connection *con,
+			       void **buf, int *len, int *proto,
+			       void **reply_buf, int *reply_len, int force_new);
+	int (*verify_authorizer_reply) (struct ceph_connection *con, int len);
+	int (*invalidate_authorizer)(struct ceph_connection *con);
+
+	/* protocol version mismatch */
+	void (*bad_proto) (struct ceph_connection *con);
+
+	/* there was some error on the socket (disconnect, whatever) */
+	void (*fault) (struct ceph_connection *con);
+
+	/* a remote host as terminated a message exchange session, and messages
+	 * we sent (or they tried to send us) may be lost. */
+	void (*peer_reset) (struct ceph_connection *con);
+
+	struct ceph_msg * (*alloc_msg) (struct ceph_connection *con,
+					struct ceph_msg_header *hdr,
+					int *skip);
+};
+
+/* use format string %s%d */
+#define ENTITY_NAME(n) ceph_entity_type_name((n).type), le64_to_cpu((n).num)
+
+struct ceph_messenger {
+	struct ceph_entity_inst inst;    /* my name+address */
+	struct ceph_entity_addr my_enc_addr;
+	struct page *zero_page;          /* used in certain error cases */
+
+	bool nocrc;
+
+	/*
+	 * the global_seq counts connections i (attempt to) initiate
+	 * in order to disambiguate certain connect race conditions.
+	 */
+	u32 global_seq;
+	spinlock_t global_seq_lock;
+
+	u32 supported_features;
+	u32 required_features;
+};
+
+/*
+ * a single message.  it contains a header (src, dest, message type, etc.),
+ * footer (crc values, mainly), a "front" message body, and possibly a
+ * data payload (stored in some number of pages).
+ */
+struct ceph_msg {
+	struct ceph_msg_header hdr;	/* header */
+	struct ceph_msg_footer footer;	/* footer */
+	struct kvec front;              /* unaligned blobs of message */
+	struct ceph_buffer *middle;
+	struct page **pages;            /* data payload.  NOT OWNER. */
+	unsigned nr_pages;              /* size of page array */
+	struct ceph_pagelist *pagelist; /* instead of pages */
+	struct list_head list_head;
+	struct kref kref;
+	struct bio  *bio;		/* instead of pages/pagelist */
+	struct bio  *bio_iter;		/* bio iterator */
+	int bio_seg;			/* current bio segment */
+	struct ceph_pagelist *trail;	/* the trailing part of the data */
+	bool front_is_vmalloc;
+	bool more_to_follow;
+	bool needs_out_seq;
+	int front_max;
+
+	struct ceph_msgpool *pool;
+};
+
+struct ceph_msg_pos {
+	int page, page_pos;  /* which page; offset in page */
+	int data_pos;        /* offset in data payload */
+	int did_page_crc;    /* true if we've calculated crc for current page */
+};
+
+/* ceph connection fault delay defaults, for exponential backoff */
+#define BASE_DELAY_INTERVAL	(HZ/2)
+#define MAX_DELAY_INTERVAL	(5 * 60 * HZ)
+
+/*
+ * ceph_connection state bit flags
+ *
+ * QUEUED and BUSY are used together to ensure that only a single
+ * thread is currently opening, reading or writing data to the socket.
+ */
+#define LOSSYTX         0  /* we can close channel or drop messages on errors */
+#define CONNECTING	1
+#define NEGOTIATING	2
+#define KEEPALIVE_PENDING      3
+#define WRITE_PENDING	4  /* we have data ready to send */
+#define QUEUED          5  /* there is work queued on this connection */
+#define BUSY            6  /* work is being done */
+#define STANDBY		8  /* no outgoing messages, socket closed.  we keep
+			    * the ceph_connection around to maintain shared
+			    * state with the peer. */
+#define CLOSED		10 /* we've closed the connection */
+#define SOCK_CLOSED	11 /* socket state changed to closed */
+#define OPENING         13 /* open connection w/ (possibly new) peer */
+#define DEAD            14 /* dead, about to kfree */
+
+/*
+ * A single connection with another host.
+ *
+ * We maintain a queue of outgoing messages, and some session state to
+ * ensure that we can preserve the lossless, ordered delivery of
+ * messages in the case of a TCP disconnect.
+ */
+struct ceph_connection {
+	void *private;
+	atomic_t nref;
+
+	const struct ceph_connection_operations *ops;
+
+	struct ceph_messenger *msgr;
+	struct socket *sock;
+	unsigned long state;	/* connection state (see flags above) */
+	const char *error_msg;  /* error message, if any */
+
+	struct ceph_entity_addr peer_addr; /* peer address */
+	struct ceph_entity_name peer_name; /* peer name */
+	struct ceph_entity_addr peer_addr_for_me;
+	unsigned peer_features;
+	u32 connect_seq;      /* identify the most recent connection
+				 attempt for this connection, client */
+	u32 peer_global_seq;  /* peer's global seq for this connection */
+
+	int auth_retry;       /* true if we need a newer authorizer */
+	void *auth_reply_buf;   /* where to put the authorizer reply */
+	int auth_reply_buf_len;
+
+	struct mutex mutex;
+
+	/* out queue */
+	struct list_head out_queue;
+	struct list_head out_sent;   /* sending or sent but unacked */
+	u64 out_seq;		     /* last message queued for send */
+	bool out_keepalive_pending;
+
+	u64 in_seq, in_seq_acked;  /* last message received, acked */
+
+	/* connection negotiation temps */
+	char in_banner[CEPH_BANNER_MAX_LEN];
+	union {
+		struct {  /* outgoing connection */
+			struct ceph_msg_connect out_connect;
+			struct ceph_msg_connect_reply in_reply;
+		};
+		struct {  /* incoming */
+			struct ceph_msg_connect in_connect;
+			struct ceph_msg_connect_reply out_reply;
+		};
+	};
+	struct ceph_entity_addr actual_peer_addr;
+
+	/* message out temps */
+	struct ceph_msg *out_msg;        /* sending message (== tail of
+					    out_sent) */
+	bool out_msg_done;
+	struct ceph_msg_pos out_msg_pos;
+
+	struct kvec out_kvec[8],         /* sending header/footer data */
+		*out_kvec_cur;
+	int out_kvec_left;   /* kvec's left in out_kvec */
+	int out_skip;        /* skip this many bytes */
+	int out_kvec_bytes;  /* total bytes left */
+	bool out_kvec_is_msg; /* kvec refers to out_msg */
+	int out_more;        /* there is more data after the kvecs */
+	__le64 out_temp_ack; /* for writing an ack */
+
+	/* message in temps */
+	struct ceph_msg_header in_hdr;
+	struct ceph_msg *in_msg;
+	struct ceph_msg_pos in_msg_pos;
+	u32 in_front_crc, in_middle_crc, in_data_crc;  /* calculated crc */
+
+	char in_tag;         /* protocol control byte */
+	int in_base_pos;     /* bytes read */
+	__le64 in_temp_ack;  /* for reading an ack */
+
+	struct delayed_work work;	    /* send|recv work */
+	unsigned long       delay;          /* current delay interval */
+};
+
+
+extern const char *ceph_pr_addr(const struct sockaddr_storage *ss);
+extern int ceph_parse_ips(const char *c, const char *end,
+			  struct ceph_entity_addr *addr,
+			  int max_count, int *count);
+
+
+extern int ceph_msgr_init(void);
+extern void ceph_msgr_exit(void);
+extern void ceph_msgr_flush(void);
+
+extern struct ceph_messenger *ceph_messenger_create(
+	struct ceph_entity_addr *myaddr,
+	u32 features, u32 required);
+extern void ceph_messenger_destroy(struct ceph_messenger *);
+
+extern void ceph_con_init(struct ceph_messenger *msgr,
+			  struct ceph_connection *con);
+extern void ceph_con_open(struct ceph_connection *con,
+			  struct ceph_entity_addr *addr);
+extern bool ceph_con_opened(struct ceph_connection *con);
+extern void ceph_con_close(struct ceph_connection *con);
+extern void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg);
+extern void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg);
+extern void ceph_con_revoke_message(struct ceph_connection *con,
+				  struct ceph_msg *msg);
+extern void ceph_con_keepalive(struct ceph_connection *con);
+extern struct ceph_connection *ceph_con_get(struct ceph_connection *con);
+extern void ceph_con_put(struct ceph_connection *con);
+
+extern struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags);
+extern void ceph_msg_kfree(struct ceph_msg *m);
+
+
+static inline struct ceph_msg *ceph_msg_get(struct ceph_msg *msg)
+{
+	kref_get(&msg->kref);
+	return msg;
+}
+extern void ceph_msg_last_put(struct kref *kref);
+static inline void ceph_msg_put(struct ceph_msg *msg)
+{
+	kref_put(&msg->kref, ceph_msg_last_put);
+}
+
+extern void ceph_msg_dump(struct ceph_msg *msg);
+
+#endif
diff --git a/include/linux/ceph/mon_client.h b/include/linux/ceph/mon_client.h
new file mode 100644
index 000000000000..545f85917780
--- /dev/null
+++ b/include/linux/ceph/mon_client.h
@@ -0,0 +1,122 @@
+#ifndef _FS_CEPH_MON_CLIENT_H
+#define _FS_CEPH_MON_CLIENT_H
+
+#include <linux/completion.h>
+#include <linux/kref.h>
+#include <linux/rbtree.h>
+
+#include "messenger.h"
+
+struct ceph_client;
+struct ceph_mount_args;
+struct ceph_auth_client;
+
+/*
+ * The monitor map enumerates the set of all monitors.
+ */
+struct ceph_monmap {
+	struct ceph_fsid fsid;
+	u32 epoch;
+	u32 num_mon;
+	struct ceph_entity_inst mon_inst[0];
+};
+
+struct ceph_mon_client;
+struct ceph_mon_generic_request;
+
+
+/*
+ * Generic mechanism for resending monitor requests.
+ */
+typedef void (*ceph_monc_request_func_t)(struct ceph_mon_client *monc,
+					 int newmon);
+
+/* a pending monitor request */
+struct ceph_mon_request {
+	struct ceph_mon_client *monc;
+	struct delayed_work delayed_work;
+	unsigned long delay;
+	ceph_monc_request_func_t do_request;
+};
+
+/*
+ * ceph_mon_generic_request is being used for the statfs and poolop requests
+ * which are bening done a bit differently because we need to get data back
+ * to the caller
+ */
+struct ceph_mon_generic_request {
+	struct kref kref;
+	u64 tid;
+	struct rb_node node;
+	int result;
+	void *buf;
+	int buf_len;
+	struct completion completion;
+	struct ceph_msg *request;  /* original request */
+	struct ceph_msg *reply;    /* and reply */
+};
+
+struct ceph_mon_client {
+	struct ceph_client *client;
+	struct ceph_monmap *monmap;
+
+	struct mutex mutex;
+	struct delayed_work delayed_work;
+
+	struct ceph_auth_client *auth;
+	struct ceph_msg *m_auth, *m_auth_reply, *m_subscribe, *m_subscribe_ack;
+	int pending_auth;
+
+	bool hunting;
+	int cur_mon;                       /* last monitor i contacted */
+	unsigned long sub_sent, sub_renew_after;
+	struct ceph_connection *con;
+	bool have_fsid;
+
+	/* pending generic requests */
+	struct rb_root generic_request_tree;
+	int num_generic_requests;
+	u64 last_tid;
+
+	/* mds/osd map */
+	int want_mdsmap;
+	int want_next_osdmap; /* 1 = want, 2 = want+asked */
+	u32 have_osdmap, have_mdsmap;
+
+#ifdef CONFIG_DEBUG_FS
+	struct dentry *debugfs_file;
+#endif
+};
+
+extern struct ceph_monmap *ceph_monmap_decode(void *p, void *end);
+extern int ceph_monmap_contains(struct ceph_monmap *m,
+				struct ceph_entity_addr *addr);
+
+extern int ceph_monc_init(struct ceph_mon_client *monc, struct ceph_client *cl);
+extern void ceph_monc_stop(struct ceph_mon_client *monc);
+
+/*
+ * The model here is to indicate that we need a new map of at least
+ * epoch @want, and also call in when we receive a map.  We will
+ * periodically rerequest the map from the monitor cluster until we
+ * get what we want.
+ */
+extern int ceph_monc_got_mdsmap(struct ceph_mon_client *monc, u32 have);
+extern int ceph_monc_got_osdmap(struct ceph_mon_client *monc, u32 have);
+
+extern void ceph_monc_request_next_osdmap(struct ceph_mon_client *monc);
+
+extern int ceph_monc_do_statfs(struct ceph_mon_client *monc,
+			       struct ceph_statfs *buf);
+
+extern int ceph_monc_open_session(struct ceph_mon_client *monc);
+
+extern int ceph_monc_validate_auth(struct ceph_mon_client *monc);
+
+extern int ceph_monc_create_snapid(struct ceph_mon_client *monc,
+				   u32 pool, u64 *snapid);
+
+extern int ceph_monc_delete_snapid(struct ceph_mon_client *monc,
+				   u32 pool, u64 snapid);
+
+#endif
diff --git a/include/linux/ceph/msgpool.h b/include/linux/ceph/msgpool.h
new file mode 100644
index 000000000000..a362605f9368
--- /dev/null
+++ b/include/linux/ceph/msgpool.h
@@ -0,0 +1,25 @@
+#ifndef _FS_CEPH_MSGPOOL
+#define _FS_CEPH_MSGPOOL
+
+#include <linux/mempool.h>
+#include "messenger.h"
+
+/*
+ * we use memory pools for preallocating messages we may receive, to
+ * avoid unexpected OOM conditions.
+ */
+struct ceph_msgpool {
+	const char *name;
+	mempool_t *pool;
+	int front_len;          /* preallocated payload size */
+};
+
+extern int ceph_msgpool_init(struct ceph_msgpool *pool,
+			     int front_len, int size, bool blocking,
+			     const char *name);
+extern void ceph_msgpool_destroy(struct ceph_msgpool *pool);
+extern struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *,
+					 int front_len);
+extern void ceph_msgpool_put(struct ceph_msgpool *, struct ceph_msg *);
+
+#endif
diff --git a/include/linux/ceph/msgr.h b/include/linux/ceph/msgr.h
new file mode 100644
index 000000000000..680d3d648cac
--- /dev/null
+++ b/include/linux/ceph/msgr.h
@@ -0,0 +1,175 @@
+#ifndef CEPH_MSGR_H
+#define CEPH_MSGR_H
+
+/*
+ * Data types for message passing layer used by Ceph.
+ */
+
+#define CEPH_MON_PORT    6789  /* default monitor port */
+
+/*
+ * client-side processes will try to bind to ports in this
+ * range, simply for the benefit of tools like nmap or wireshark
+ * that would like to identify the protocol.
+ */
+#define CEPH_PORT_FIRST  6789
+#define CEPH_PORT_START  6800  /* non-monitors start here */
+#define CEPH_PORT_LAST   6900
+
+/*
+ * tcp connection banner.  include a protocol version. and adjust
+ * whenever the wire protocol changes.  try to keep this string length
+ * constant.
+ */
+#define CEPH_BANNER "ceph v027"
+#define CEPH_BANNER_MAX_LEN 30
+
+
+/*
+ * Rollover-safe type and comparator for 32-bit sequence numbers.
+ * Comparator returns -1, 0, or 1.
+ */
+typedef __u32 ceph_seq_t;
+
+static inline __s32 ceph_seq_cmp(__u32 a, __u32 b)
+{
+       return (__s32)a - (__s32)b;
+}
+
+
+/*
+ * entity_name -- logical name for a process participating in the
+ * network, e.g. 'mds0' or 'osd3'.
+ */
+struct ceph_entity_name {
+	__u8 type;      /* CEPH_ENTITY_TYPE_* */
+	__le64 num;
+} __attribute__ ((packed));
+
+#define CEPH_ENTITY_TYPE_MON    0x01
+#define CEPH_ENTITY_TYPE_MDS    0x02
+#define CEPH_ENTITY_TYPE_OSD    0x04
+#define CEPH_ENTITY_TYPE_CLIENT 0x08
+#define CEPH_ENTITY_TYPE_AUTH   0x20
+
+#define CEPH_ENTITY_TYPE_ANY    0xFF
+
+extern const char *ceph_entity_type_name(int type);
+
+/*
+ * entity_addr -- network address
+ */
+struct ceph_entity_addr {
+	__le32 type;
+	__le32 nonce;  /* unique id for process (e.g. pid) */
+	struct sockaddr_storage in_addr;
+} __attribute__ ((packed));
+
+struct ceph_entity_inst {
+	struct ceph_entity_name name;
+	struct ceph_entity_addr addr;
+} __attribute__ ((packed));
+
+
+/* used by message exchange protocol */
+#define CEPH_MSGR_TAG_READY         1  /* server->client: ready for messages */
+#define CEPH_MSGR_TAG_RESETSESSION  2  /* server->client: reset, try again */
+#define CEPH_MSGR_TAG_WAIT          3  /* server->client: wait for racing
+					  incoming connection */
+#define CEPH_MSGR_TAG_RETRY_SESSION 4  /* server->client + cseq: try again
+					  with higher cseq */
+#define CEPH_MSGR_TAG_RETRY_GLOBAL  5  /* server->client + gseq: try again
+					  with higher gseq */
+#define CEPH_MSGR_TAG_CLOSE         6  /* closing pipe */
+#define CEPH_MSGR_TAG_MSG           7  /* message */
+#define CEPH_MSGR_TAG_ACK           8  /* message ack */
+#define CEPH_MSGR_TAG_KEEPALIVE     9  /* just a keepalive byte! */
+#define CEPH_MSGR_TAG_BADPROTOVER  10  /* bad protocol version */
+#define CEPH_MSGR_TAG_BADAUTHORIZER 11 /* bad authorizer */
+#define CEPH_MSGR_TAG_FEATURES      12 /* insufficient features */
+
+
+/*
+ * connection negotiation
+ */
+struct ceph_msg_connect {
+	__le64 features;     /* supported feature bits */
+	__le32 host_type;    /* CEPH_ENTITY_TYPE_* */
+	__le32 global_seq;   /* count connections initiated by this host */
+	__le32 connect_seq;  /* count connections initiated in this session */
+	__le32 protocol_version;
+	__le32 authorizer_protocol;
+	__le32 authorizer_len;
+	__u8  flags;         /* CEPH_MSG_CONNECT_* */
+} __attribute__ ((packed));
+
+struct ceph_msg_connect_reply {
+	__u8 tag;
+	__le64 features;     /* feature bits for this session */
+	__le32 global_seq;
+	__le32 connect_seq;
+	__le32 protocol_version;
+	__le32 authorizer_len;
+	__u8 flags;
+} __attribute__ ((packed));
+
+#define CEPH_MSG_CONNECT_LOSSY  1  /* messages i send may be safely dropped */
+
+
+/*
+ * message header
+ */
+struct ceph_msg_header_old {
+	__le64 seq;       /* message seq# for this session */
+	__le64 tid;       /* transaction id */
+	__le16 type;      /* message type */
+	__le16 priority;  /* priority.  higher value == higher priority */
+	__le16 version;   /* version of message encoding */
+
+	__le32 front_len; /* bytes in main payload */
+	__le32 middle_len;/* bytes in middle payload */
+	__le32 data_len;  /* bytes of data payload */
+	__le16 data_off;  /* sender: include full offset;
+			     receiver: mask against ~PAGE_MASK */
+
+	struct ceph_entity_inst src, orig_src;
+	__le32 reserved;
+	__le32 crc;       /* header crc32c */
+} __attribute__ ((packed));
+
+struct ceph_msg_header {
+	__le64 seq;       /* message seq# for this session */
+	__le64 tid;       /* transaction id */
+	__le16 type;      /* message type */
+	__le16 priority;  /* priority.  higher value == higher priority */
+	__le16 version;   /* version of message encoding */
+
+	__le32 front_len; /* bytes in main payload */
+	__le32 middle_len;/* bytes in middle payload */
+	__le32 data_len;  /* bytes of data payload */
+	__le16 data_off;  /* sender: include full offset;
+			     receiver: mask against ~PAGE_MASK */
+
+	struct ceph_entity_name src;
+	__le32 reserved;
+	__le32 crc;       /* header crc32c */
+} __attribute__ ((packed));
+
+#define CEPH_MSG_PRIO_LOW     64
+#define CEPH_MSG_PRIO_DEFAULT 127
+#define CEPH_MSG_PRIO_HIGH    196
+#define CEPH_MSG_PRIO_HIGHEST 255
+
+/*
+ * follows data payload
+ */
+struct ceph_msg_footer {
+	__le32 front_crc, middle_crc, data_crc;
+	__u8 flags;
+} __attribute__ ((packed));
+
+#define CEPH_MSG_FOOTER_COMPLETE  (1<<0)   /* msg wasn't aborted */
+#define CEPH_MSG_FOOTER_NOCRC     (1<<1)   /* no data crc */
+
+
+#endif
diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h
new file mode 100644
index 000000000000..6c91fb032c39
--- /dev/null
+++ b/include/linux/ceph/osd_client.h
@@ -0,0 +1,234 @@
+#ifndef _FS_CEPH_OSD_CLIENT_H
+#define _FS_CEPH_OSD_CLIENT_H
+
+#include <linux/completion.h>
+#include <linux/kref.h>
+#include <linux/mempool.h>
+#include <linux/rbtree.h>
+
+#include "types.h"
+#include "osdmap.h"
+#include "messenger.h"
+
+struct ceph_msg;
+struct ceph_snap_context;
+struct ceph_osd_request;
+struct ceph_osd_client;
+struct ceph_authorizer;
+struct ceph_pagelist;
+
+/*
+ * completion callback for async writepages
+ */
+typedef void (*ceph_osdc_callback_t)(struct ceph_osd_request *,
+				     struct ceph_msg *);
+
+/* a given osd we're communicating with */
+struct ceph_osd {
+	atomic_t o_ref;
+	struct ceph_osd_client *o_osdc;
+	int o_osd;
+	int o_incarnation;
+	struct rb_node o_node;
+	struct ceph_connection o_con;
+	struct list_head o_requests;
+	struct list_head o_osd_lru;
+	struct ceph_authorizer *o_authorizer;
+	void *o_authorizer_buf, *o_authorizer_reply_buf;
+	size_t o_authorizer_buf_len, o_authorizer_reply_buf_len;
+	unsigned long lru_ttl;
+	int o_marked_for_keepalive;
+	struct list_head o_keepalive_item;
+};
+
+/* an in-flight request */
+struct ceph_osd_request {
+	u64             r_tid;              /* unique for this client */
+	struct rb_node  r_node;
+	struct list_head r_req_lru_item;
+	struct list_head r_osd_item;
+	struct ceph_osd *r_osd;
+	struct ceph_pg   r_pgid;
+	int              r_pg_osds[CEPH_PG_MAX_SIZE];
+	int              r_num_pg_osds;
+
+	struct ceph_connection *r_con_filling_msg;
+
+	struct ceph_msg  *r_request, *r_reply;
+	int               r_result;
+	int               r_flags;     /* any additional flags for the osd */
+	u32               r_sent;      /* >0 if r_request is sending/sent */
+	int               r_got_reply;
+
+	struct ceph_osd_client *r_osdc;
+	struct kref       r_kref;
+	bool              r_mempool;
+	struct completion r_completion, r_safe_completion;
+	ceph_osdc_callback_t r_callback, r_safe_callback;
+	struct ceph_eversion r_reassert_version;
+	struct list_head  r_unsafe_item;
+
+	struct inode *r_inode;         	      /* for use by callbacks */
+	void *r_priv;			      /* ditto */
+
+	char              r_oid[40];          /* object name */
+	int               r_oid_len;
+	unsigned long     r_stamp;            /* send OR check time */
+	bool              r_resend;           /* msg send failed, needs retry */
+
+	struct ceph_file_layout r_file_layout;
+	struct ceph_snap_context *r_snapc;    /* snap context for writes */
+	unsigned          r_num_pages;        /* size of page array (follows) */
+	struct page     **r_pages;            /* pages for data payload */
+	int               r_pages_from_pool;
+	int               r_own_pages;        /* if true, i own page list */
+#ifdef CONFIG_BLOCK
+	struct bio       *r_bio;	      /* instead of pages */
+#endif
+
+	struct ceph_pagelist *r_trail;	      /* trailing part of the data */
+};
+
+struct ceph_osd_client {
+	struct ceph_client     *client;
+
+	struct ceph_osdmap     *osdmap;       /* current map */
+	struct rw_semaphore    map_sem;
+	struct completion      map_waiters;
+	u64                    last_requested_map;
+
+	struct mutex           request_mutex;
+	struct rb_root         osds;          /* osds */
+	struct list_head       osd_lru;       /* idle osds */
+	u64                    timeout_tid;   /* tid of timeout triggering rq */
+	u64                    last_tid;      /* tid of last request */
+	struct rb_root         requests;      /* pending requests */
+	struct list_head       req_lru;	      /* pending requests lru */
+	int                    num_requests;
+	struct delayed_work    timeout_work;
+	struct delayed_work    osds_timeout_work;
+#ifdef CONFIG_DEBUG_FS
+	struct dentry 	       *debugfs_file;
+#endif
+
+	mempool_t              *req_mempool;
+
+	struct ceph_msgpool	msgpool_op;
+	struct ceph_msgpool	msgpool_op_reply;
+};
+
+struct ceph_osd_req_op {
+	u16 op;           /* CEPH_OSD_OP_* */
+	u32 flags;        /* CEPH_OSD_FLAG_* */
+	union {
+		struct {
+			u64 offset, length;
+			u64 truncate_size;
+			u32 truncate_seq;
+		} extent;
+		struct {
+			const char *name;
+			u32 name_len;
+			const char  *val;
+			u32 value_len;
+			__u8 cmp_op;       /* CEPH_OSD_CMPXATTR_OP_* */
+			__u8 cmp_mode;     /* CEPH_OSD_CMPXATTR_MODE_* */
+		} xattr;
+		struct {
+			const char *class_name;
+			__u8 class_len;
+			const char *method_name;
+			__u8 method_len;
+			__u8 argc;
+			const char *indata;
+			u32 indata_len;
+		} cls;
+		struct {
+			u64 cookie, count;
+		} pgls;
+	        struct {
+		        u64 snapid;
+	        } snap;
+	};
+	u32 payload_len;
+};
+
+extern int ceph_osdc_init(struct ceph_osd_client *osdc,
+			  struct ceph_client *client);
+extern void ceph_osdc_stop(struct ceph_osd_client *osdc);
+
+extern void ceph_osdc_handle_reply(struct ceph_osd_client *osdc,
+				   struct ceph_msg *msg);
+extern void ceph_osdc_handle_map(struct ceph_osd_client *osdc,
+				 struct ceph_msg *msg);
+
+extern void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
+			struct ceph_file_layout *layout,
+			u64 snapid,
+			u64 off, u64 *plen, u64 *bno,
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op);
+
+extern struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
+					       int flags,
+					       struct ceph_snap_context *snapc,
+					       struct ceph_osd_req_op *ops,
+					       bool use_mempool,
+					       gfp_t gfp_flags,
+					       struct page **pages,
+					       struct bio *bio);
+
+extern void ceph_osdc_build_request(struct ceph_osd_request *req,
+				    u64 off, u64 *plen,
+				    struct ceph_osd_req_op *src_ops,
+				    struct ceph_snap_context *snapc,
+				    struct timespec *mtime,
+				    const char *oid,
+				    int oid_len);
+
+extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *,
+				      struct ceph_file_layout *layout,
+				      struct ceph_vino vino,
+				      u64 offset, u64 *len, int op, int flags,
+				      struct ceph_snap_context *snapc,
+				      int do_sync, u32 truncate_seq,
+				      u64 truncate_size,
+				      struct timespec *mtime,
+				      bool use_mempool, int num_reply);
+
+static inline void ceph_osdc_get_request(struct ceph_osd_request *req)
+{
+	kref_get(&req->r_kref);
+}
+extern void ceph_osdc_release_request(struct kref *kref);
+static inline void ceph_osdc_put_request(struct ceph_osd_request *req)
+{
+	kref_put(&req->r_kref, ceph_osdc_release_request);
+}
+
+extern int ceph_osdc_start_request(struct ceph_osd_client *osdc,
+				   struct ceph_osd_request *req,
+				   bool nofail);
+extern int ceph_osdc_wait_request(struct ceph_osd_client *osdc,
+				  struct ceph_osd_request *req);
+extern void ceph_osdc_sync(struct ceph_osd_client *osdc);
+
+extern int ceph_osdc_readpages(struct ceph_osd_client *osdc,
+			       struct ceph_vino vino,
+			       struct ceph_file_layout *layout,
+			       u64 off, u64 *plen,
+			       u32 truncate_seq, u64 truncate_size,
+			       struct page **pages, int nr_pages);
+
+extern int ceph_osdc_writepages(struct ceph_osd_client *osdc,
+				struct ceph_vino vino,
+				struct ceph_file_layout *layout,
+				struct ceph_snap_context *sc,
+				u64 off, u64 len,
+				u32 truncate_seq, u64 truncate_size,
+				struct timespec *mtime,
+				struct page **pages, int nr_pages,
+				int flags, int do_sync, bool nofail);
+
+#endif
+
diff --git a/include/linux/ceph/osdmap.h b/include/linux/ceph/osdmap.h
new file mode 100644
index 000000000000..ba4c205cbb01
--- /dev/null
+++ b/include/linux/ceph/osdmap.h
@@ -0,0 +1,130 @@
+#ifndef _FS_CEPH_OSDMAP_H
+#define _FS_CEPH_OSDMAP_H
+
+#include <linux/rbtree.h>
+#include "types.h"
+#include "ceph_fs.h"
+#include <linux/crush/crush.h>
+
+/*
+ * The osd map describes the current membership of the osd cluster and
+ * specifies the mapping of objects to placement groups and placement
+ * groups to (sets of) osds.  That is, it completely specifies the
+ * (desired) distribution of all data objects in the system at some
+ * point in time.
+ *
+ * Each map version is identified by an epoch, which increases monotonically.
+ *
+ * The map can be updated either via an incremental map (diff) describing
+ * the change between two successive epochs, or as a fully encoded map.
+ */
+struct ceph_pg_pool_info {
+	struct rb_node node;
+	int id;
+	struct ceph_pg_pool v;
+	int pg_num_mask, pgp_num_mask, lpg_num_mask, lpgp_num_mask;
+	char *name;
+};
+
+struct ceph_pg_mapping {
+	struct rb_node node;
+	struct ceph_pg pgid;
+	int len;
+	int osds[];
+};
+
+struct ceph_osdmap {
+	struct ceph_fsid fsid;
+	u32 epoch;
+	u32 mkfs_epoch;
+	struct ceph_timespec created, modified;
+
+	u32 flags;         /* CEPH_OSDMAP_* */
+
+	u32 max_osd;       /* size of osd_state, _offload, _addr arrays */
+	u8 *osd_state;     /* CEPH_OSD_* */
+	u32 *osd_weight;   /* 0 = failed, 0x10000 = 100% normal */
+	struct ceph_entity_addr *osd_addr;
+
+	struct rb_root pg_temp;
+	struct rb_root pg_pools;
+	u32 pool_max;
+
+	/* the CRUSH map specifies the mapping of placement groups to
+	 * the list of osds that store+replicate them. */
+	struct crush_map *crush;
+};
+
+/*
+ * file layout helpers
+ */
+#define ceph_file_layout_su(l) ((__s32)le32_to_cpu((l).fl_stripe_unit))
+#define ceph_file_layout_stripe_count(l) \
+	((__s32)le32_to_cpu((l).fl_stripe_count))
+#define ceph_file_layout_object_size(l) ((__s32)le32_to_cpu((l).fl_object_size))
+#define ceph_file_layout_cas_hash(l) ((__s32)le32_to_cpu((l).fl_cas_hash))
+#define ceph_file_layout_object_su(l) \
+	((__s32)le32_to_cpu((l).fl_object_stripe_unit))
+#define ceph_file_layout_pg_preferred(l) \
+	((__s32)le32_to_cpu((l).fl_pg_preferred))
+#define ceph_file_layout_pg_pool(l) \
+	((__s32)le32_to_cpu((l).fl_pg_pool))
+
+static inline unsigned ceph_file_layout_stripe_width(struct ceph_file_layout *l)
+{
+	return le32_to_cpu(l->fl_stripe_unit) *
+		le32_to_cpu(l->fl_stripe_count);
+}
+
+/* "period" == bytes before i start on a new set of objects */
+static inline unsigned ceph_file_layout_period(struct ceph_file_layout *l)
+{
+	return le32_to_cpu(l->fl_object_size) *
+		le32_to_cpu(l->fl_stripe_count);
+}
+
+
+static inline int ceph_osd_is_up(struct ceph_osdmap *map, int osd)
+{
+	return (osd < map->max_osd) && (map->osd_state[osd] & CEPH_OSD_UP);
+}
+
+static inline bool ceph_osdmap_flag(struct ceph_osdmap *map, int flag)
+{
+	return map && (map->flags & flag);
+}
+
+extern char *ceph_osdmap_state_str(char *str, int len, int state);
+
+static inline struct ceph_entity_addr *ceph_osd_addr(struct ceph_osdmap *map,
+						     int osd)
+{
+	if (osd >= map->max_osd)
+		return NULL;
+	return &map->osd_addr[osd];
+}
+
+extern struct ceph_osdmap *osdmap_decode(void **p, void *end);
+extern struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
+					    struct ceph_osdmap *map,
+					    struct ceph_messenger *msgr);
+extern void ceph_osdmap_destroy(struct ceph_osdmap *map);
+
+/* calculate mapping of a file extent to an object */
+extern void ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
+					  u64 off, u64 *plen,
+					  u64 *bno, u64 *oxoff, u64 *oxlen);
+
+/* calculate mapping of object to a placement group */
+extern int ceph_calc_object_layout(struct ceph_object_layout *ol,
+				   const char *oid,
+				   struct ceph_file_layout *fl,
+				   struct ceph_osdmap *osdmap);
+extern int ceph_calc_pg_acting(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
+			       int *acting);
+extern int ceph_calc_pg_primary(struct ceph_osdmap *osdmap,
+				struct ceph_pg pgid);
+
+extern int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name);
+
+#endif
diff --git a/include/linux/ceph/pagelist.h b/include/linux/ceph/pagelist.h
new file mode 100644
index 000000000000..cc9327aa1c98
--- /dev/null
+++ b/include/linux/ceph/pagelist.h
@@ -0,0 +1,54 @@
+#ifndef __FS_CEPH_PAGELIST_H
+#define __FS_CEPH_PAGELIST_H
+
+#include <linux/list.h>
+
+struct ceph_pagelist {
+	struct list_head head;
+	void *mapped_tail;
+	size_t length;
+	size_t room;
+};
+
+static inline void ceph_pagelist_init(struct ceph_pagelist *pl)
+{
+	INIT_LIST_HEAD(&pl->head);
+	pl->mapped_tail = NULL;
+	pl->length = 0;
+	pl->room = 0;
+}
+extern int ceph_pagelist_release(struct ceph_pagelist *pl);
+
+extern int ceph_pagelist_append(struct ceph_pagelist *pl, const void *d, size_t l);
+
+static inline int ceph_pagelist_encode_64(struct ceph_pagelist *pl, u64 v)
+{
+	__le64 ev = cpu_to_le64(v);
+	return ceph_pagelist_append(pl, &ev, sizeof(ev));
+}
+static inline int ceph_pagelist_encode_32(struct ceph_pagelist *pl, u32 v)
+{
+	__le32 ev = cpu_to_le32(v);
+	return ceph_pagelist_append(pl, &ev, sizeof(ev));
+}
+static inline int ceph_pagelist_encode_16(struct ceph_pagelist *pl, u16 v)
+{
+	__le16 ev = cpu_to_le16(v);
+	return ceph_pagelist_append(pl, &ev, sizeof(ev));
+}
+static inline int ceph_pagelist_encode_8(struct ceph_pagelist *pl, u8 v)
+{
+	return ceph_pagelist_append(pl, &v, 1);
+}
+static inline int ceph_pagelist_encode_string(struct ceph_pagelist *pl,
+					      char *s, size_t len)
+{
+	int ret = ceph_pagelist_encode_32(pl, len);
+	if (ret)
+		return ret;
+	if (len)
+		return ceph_pagelist_append(pl, s, len);
+	return 0;
+}
+
+#endif
diff --git a/include/linux/ceph/rados.h b/include/linux/ceph/rados.h
new file mode 100644
index 000000000000..6d5247f2e81b
--- /dev/null
+++ b/include/linux/ceph/rados.h
@@ -0,0 +1,405 @@
+#ifndef CEPH_RADOS_H
+#define CEPH_RADOS_H
+
+/*
+ * Data types for the Ceph distributed object storage layer RADOS
+ * (Reliable Autonomic Distributed Object Store).
+ */
+
+#include "msgr.h"
+
+/*
+ * osdmap encoding versions
+ */
+#define CEPH_OSDMAP_INC_VERSION     5
+#define CEPH_OSDMAP_INC_VERSION_EXT 5
+#define CEPH_OSDMAP_VERSION         5
+#define CEPH_OSDMAP_VERSION_EXT     5
+
+/*
+ * fs id
+ */
+struct ceph_fsid {
+	unsigned char fsid[16];
+};
+
+static inline int ceph_fsid_compare(const struct ceph_fsid *a,
+				    const struct ceph_fsid *b)
+{
+	return memcmp(a, b, sizeof(*a));
+}
+
+/*
+ * ino, object, etc.
+ */
+typedef __le64 ceph_snapid_t;
+#define CEPH_SNAPDIR ((__u64)(-1))  /* reserved for hidden .snap dir */
+#define CEPH_NOSNAP  ((__u64)(-2))  /* "head", "live" revision */
+#define CEPH_MAXSNAP ((__u64)(-3))  /* largest valid snapid */
+
+struct ceph_timespec {
+	__le32 tv_sec;
+	__le32 tv_nsec;
+} __attribute__ ((packed));
+
+
+/*
+ * object layout - how objects are mapped into PGs
+ */
+#define CEPH_OBJECT_LAYOUT_HASH     1
+#define CEPH_OBJECT_LAYOUT_LINEAR   2
+#define CEPH_OBJECT_LAYOUT_HASHINO  3
+
+/*
+ * pg layout -- how PGs are mapped onto (sets of) OSDs
+ */
+#define CEPH_PG_LAYOUT_CRUSH  0
+#define CEPH_PG_LAYOUT_HASH   1
+#define CEPH_PG_LAYOUT_LINEAR 2
+#define CEPH_PG_LAYOUT_HYBRID 3
+
+#define CEPH_PG_MAX_SIZE      16  /* max # osds in a single pg */
+
+/*
+ * placement group.
+ * we encode this into one __le64.
+ */
+struct ceph_pg {
+	__le16 preferred; /* preferred primary osd */
+	__le16 ps;        /* placement seed */
+	__le32 pool;      /* object pool */
+} __attribute__ ((packed));
+
+/*
+ * pg_pool is a set of pgs storing a pool of objects
+ *
+ *  pg_num -- base number of pseudorandomly placed pgs
+ *
+ *  pgp_num -- effective number when calculating pg placement.  this
+ * is used for pg_num increases.  new pgs result in data being "split"
+ * into new pgs.  for this to proceed smoothly, new pgs are intiially
+ * colocated with their parents; that is, pgp_num doesn't increase
+ * until the new pgs have successfully split.  only _then_ are the new
+ * pgs placed independently.
+ *
+ *  lpg_num -- localized pg count (per device).  replicas are randomly
+ * selected.
+ *
+ *  lpgp_num -- as above.
+ */
+#define CEPH_PG_TYPE_REP     1
+#define CEPH_PG_TYPE_RAID4   2
+#define CEPH_PG_POOL_VERSION 2
+struct ceph_pg_pool {
+	__u8 type;                /* CEPH_PG_TYPE_* */
+	__u8 size;                /* number of osds in each pg */
+	__u8 crush_ruleset;       /* crush placement rule */
+	__u8 object_hash;         /* hash mapping object name to ps */
+	__le32 pg_num, pgp_num;   /* number of pg's */
+	__le32 lpg_num, lpgp_num; /* number of localized pg's */
+	__le32 last_change;       /* most recent epoch changed */
+	__le64 snap_seq;          /* seq for per-pool snapshot */
+	__le32 snap_epoch;        /* epoch of last snap */
+	__le32 num_snaps;
+	__le32 num_removed_snap_intervals; /* if non-empty, NO per-pool snaps */
+	__le64 auid;               /* who owns the pg */
+} __attribute__ ((packed));
+
+/*
+ * stable_mod func is used to control number of placement groups.
+ * similar to straight-up modulo, but produces a stable mapping as b
+ * increases over time.  b is the number of bins, and bmask is the
+ * containing power of 2 minus 1.
+ *
+ * b <= bmask and bmask=(2**n)-1
+ * e.g., b=12 -> bmask=15, b=123 -> bmask=127
+ */
+static inline int ceph_stable_mod(int x, int b, int bmask)
+{
+	if ((x & bmask) < b)
+		return x & bmask;
+	else
+		return x & (bmask >> 1);
+}
+
+/*
+ * object layout - how a given object should be stored.
+ */
+struct ceph_object_layout {
+	struct ceph_pg ol_pgid;   /* raw pg, with _full_ ps precision. */
+	__le32 ol_stripe_unit;    /* for per-object parity, if any */
+} __attribute__ ((packed));
+
+/*
+ * compound epoch+version, used by storage layer to serialize mutations
+ */
+struct ceph_eversion {
+	__le32 epoch;
+	__le64 version;
+} __attribute__ ((packed));
+
+/*
+ * osd map bits
+ */
+
+/* status bits */
+#define CEPH_OSD_EXISTS 1
+#define CEPH_OSD_UP     2
+
+/* osd weights.  fixed point value: 0x10000 == 1.0 ("in"), 0 == "out" */
+#define CEPH_OSD_IN  0x10000
+#define CEPH_OSD_OUT 0
+
+
+/*
+ * osd map flag bits
+ */
+#define CEPH_OSDMAP_NEARFULL (1<<0)  /* sync writes (near ENOSPC) */
+#define CEPH_OSDMAP_FULL     (1<<1)  /* no data writes (ENOSPC) */
+#define CEPH_OSDMAP_PAUSERD  (1<<2)  /* pause all reads */
+#define CEPH_OSDMAP_PAUSEWR  (1<<3)  /* pause all writes */
+#define CEPH_OSDMAP_PAUSEREC (1<<4)  /* pause recovery */
+
+/*
+ * osd ops
+ */
+#define CEPH_OSD_OP_MODE       0xf000
+#define CEPH_OSD_OP_MODE_RD    0x1000
+#define CEPH_OSD_OP_MODE_WR    0x2000
+#define CEPH_OSD_OP_MODE_RMW   0x3000
+#define CEPH_OSD_OP_MODE_SUB   0x4000
+
+#define CEPH_OSD_OP_TYPE       0x0f00
+#define CEPH_OSD_OP_TYPE_LOCK  0x0100
+#define CEPH_OSD_OP_TYPE_DATA  0x0200
+#define CEPH_OSD_OP_TYPE_ATTR  0x0300
+#define CEPH_OSD_OP_TYPE_EXEC  0x0400
+#define CEPH_OSD_OP_TYPE_PG    0x0500
+
+enum {
+	/** data **/
+	/* read */
+	CEPH_OSD_OP_READ      = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 1,
+	CEPH_OSD_OP_STAT      = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 2,
+
+	/* fancy read */
+	CEPH_OSD_OP_MASKTRUNC = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 4,
+
+	/* write */
+	CEPH_OSD_OP_WRITE     = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 1,
+	CEPH_OSD_OP_WRITEFULL = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 2,
+	CEPH_OSD_OP_TRUNCATE  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 3,
+	CEPH_OSD_OP_ZERO      = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 4,
+	CEPH_OSD_OP_DELETE    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 5,
+
+	/* fancy write */
+	CEPH_OSD_OP_APPEND    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 6,
+	CEPH_OSD_OP_STARTSYNC = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 7,
+	CEPH_OSD_OP_SETTRUNC  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 8,
+	CEPH_OSD_OP_TRIMTRUNC = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 9,
+
+	CEPH_OSD_OP_TMAPUP  = CEPH_OSD_OP_MODE_RMW | CEPH_OSD_OP_TYPE_DATA | 10,
+	CEPH_OSD_OP_TMAPPUT = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 11,
+	CEPH_OSD_OP_TMAPGET = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_DATA | 12,
+
+	CEPH_OSD_OP_CREATE  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 13,
+	CEPH_OSD_OP_ROLLBACK= CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_DATA | 14,
+
+	/** attrs **/
+	/* read */
+	CEPH_OSD_OP_GETXATTR  = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 1,
+	CEPH_OSD_OP_GETXATTRS = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 2,
+	CEPH_OSD_OP_CMPXATTR  = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_ATTR | 3,
+
+	/* write */
+	CEPH_OSD_OP_SETXATTR  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 1,
+	CEPH_OSD_OP_SETXATTRS = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 2,
+	CEPH_OSD_OP_RESETXATTRS = CEPH_OSD_OP_MODE_WR|CEPH_OSD_OP_TYPE_ATTR | 3,
+	CEPH_OSD_OP_RMXATTR   = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_ATTR | 4,
+
+	/** subop **/
+	CEPH_OSD_OP_PULL           = CEPH_OSD_OP_MODE_SUB | 1,
+	CEPH_OSD_OP_PUSH           = CEPH_OSD_OP_MODE_SUB | 2,
+	CEPH_OSD_OP_BALANCEREADS   = CEPH_OSD_OP_MODE_SUB | 3,
+	CEPH_OSD_OP_UNBALANCEREADS = CEPH_OSD_OP_MODE_SUB | 4,
+	CEPH_OSD_OP_SCRUB          = CEPH_OSD_OP_MODE_SUB | 5,
+
+	/** lock **/
+	CEPH_OSD_OP_WRLOCK    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 1,
+	CEPH_OSD_OP_WRUNLOCK  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 2,
+	CEPH_OSD_OP_RDLOCK    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 3,
+	CEPH_OSD_OP_RDUNLOCK  = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 4,
+	CEPH_OSD_OP_UPLOCK    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 5,
+	CEPH_OSD_OP_DNLOCK    = CEPH_OSD_OP_MODE_WR | CEPH_OSD_OP_TYPE_LOCK | 6,
+
+	/** exec **/
+	CEPH_OSD_OP_CALL    = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_EXEC | 1,
+
+	/** pg **/
+	CEPH_OSD_OP_PGLS      = CEPH_OSD_OP_MODE_RD | CEPH_OSD_OP_TYPE_PG | 1,
+};
+
+static inline int ceph_osd_op_type_lock(int op)
+{
+	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_LOCK;
+}
+static inline int ceph_osd_op_type_data(int op)
+{
+	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_DATA;
+}
+static inline int ceph_osd_op_type_attr(int op)
+{
+	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_ATTR;
+}
+static inline int ceph_osd_op_type_exec(int op)
+{
+	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_EXEC;
+}
+static inline int ceph_osd_op_type_pg(int op)
+{
+	return (op & CEPH_OSD_OP_TYPE) == CEPH_OSD_OP_TYPE_PG;
+}
+
+static inline int ceph_osd_op_mode_subop(int op)
+{
+	return (op & CEPH_OSD_OP_MODE) == CEPH_OSD_OP_MODE_SUB;
+}
+static inline int ceph_osd_op_mode_read(int op)
+{
+	return (op & CEPH_OSD_OP_MODE) == CEPH_OSD_OP_MODE_RD;
+}
+static inline int ceph_osd_op_mode_modify(int op)
+{
+	return (op & CEPH_OSD_OP_MODE) == CEPH_OSD_OP_MODE_WR;
+}
+
+/*
+ * note that the following tmap stuff is also defined in the ceph librados.h
+ * any modification here needs to be updated there
+ */
+#define CEPH_OSD_TMAP_HDR 'h'
+#define CEPH_OSD_TMAP_SET 's'
+#define CEPH_OSD_TMAP_RM  'r'
+
+extern const char *ceph_osd_op_name(int op);
+
+
+/*
+ * osd op flags
+ *
+ * An op may be READ, WRITE, or READ|WRITE.
+ */
+enum {
+	CEPH_OSD_FLAG_ACK = 1,          /* want (or is) "ack" ack */
+	CEPH_OSD_FLAG_ONNVRAM = 2,      /* want (or is) "onnvram" ack */
+	CEPH_OSD_FLAG_ONDISK = 4,       /* want (or is) "ondisk" ack */
+	CEPH_OSD_FLAG_RETRY = 8,        /* resend attempt */
+	CEPH_OSD_FLAG_READ = 16,        /* op may read */
+	CEPH_OSD_FLAG_WRITE = 32,       /* op may write */
+	CEPH_OSD_FLAG_ORDERSNAP = 64,   /* EOLDSNAP if snapc is out of order */
+	CEPH_OSD_FLAG_PEERSTAT = 128,   /* msg includes osd_peer_stat */
+	CEPH_OSD_FLAG_BALANCE_READS = 256,
+	CEPH_OSD_FLAG_PARALLELEXEC = 512, /* execute op in parallel */
+	CEPH_OSD_FLAG_PGOP = 1024,      /* pg op, no object */
+	CEPH_OSD_FLAG_EXEC = 2048,      /* op may exec */
+	CEPH_OSD_FLAG_EXEC_PUBLIC = 4096, /* op may exec (public) */
+};
+
+enum {
+	CEPH_OSD_OP_FLAG_EXCL = 1,      /* EXCL object create */
+};
+
+#define EOLDSNAPC    ERESTART  /* ORDERSNAP flag set; writer has old snapc*/
+#define EBLACKLISTED ESHUTDOWN /* blacklisted */
+
+/* xattr comparison */
+enum {
+	CEPH_OSD_CMPXATTR_OP_NOP = 0,
+	CEPH_OSD_CMPXATTR_OP_EQ  = 1,
+	CEPH_OSD_CMPXATTR_OP_NE  = 2,
+	CEPH_OSD_CMPXATTR_OP_GT  = 3,
+	CEPH_OSD_CMPXATTR_OP_GTE = 4,
+	CEPH_OSD_CMPXATTR_OP_LT  = 5,
+	CEPH_OSD_CMPXATTR_OP_LTE = 6
+};
+
+enum {
+	CEPH_OSD_CMPXATTR_MODE_STRING = 1,
+	CEPH_OSD_CMPXATTR_MODE_U64    = 2
+};
+
+/*
+ * an individual object operation.  each may be accompanied by some data
+ * payload
+ */
+struct ceph_osd_op {
+	__le16 op;           /* CEPH_OSD_OP_* */
+	__le32 flags;        /* CEPH_OSD_FLAG_* */
+	union {
+		struct {
+			__le64 offset, length;
+			__le64 truncate_size;
+			__le32 truncate_seq;
+		} __attribute__ ((packed)) extent;
+		struct {
+			__le32 name_len;
+			__le32 value_len;
+			__u8 cmp_op;       /* CEPH_OSD_CMPXATTR_OP_* */
+			__u8 cmp_mode;     /* CEPH_OSD_CMPXATTR_MODE_* */
+		} __attribute__ ((packed)) xattr;
+		struct {
+			__u8 class_len;
+			__u8 method_len;
+			__u8 argc;
+			__le32 indata_len;
+		} __attribute__ ((packed)) cls;
+		struct {
+			__le64 cookie, count;
+		} __attribute__ ((packed)) pgls;
+	        struct {
+		        __le64 snapid;
+	        } __attribute__ ((packed)) snap;
+	};
+	__le32 payload_len;
+} __attribute__ ((packed));
+
+/*
+ * osd request message header.  each request may include multiple
+ * ceph_osd_op object operations.
+ */
+struct ceph_osd_request_head {
+	__le32 client_inc;                 /* client incarnation */
+	struct ceph_object_layout layout;  /* pgid */
+	__le32 osdmap_epoch;               /* client's osdmap epoch */
+
+	__le32 flags;
+
+	struct ceph_timespec mtime;        /* for mutations only */
+	struct ceph_eversion reassert_version; /* if we are replaying op */
+
+	__le32 object_len;     /* length of object name */
+
+	__le64 snapid;         /* snapid to read */
+	__le64 snap_seq;       /* writer's snap context */
+	__le32 num_snaps;
+
+	__le16 num_ops;
+	struct ceph_osd_op ops[];  /* followed by ops[], obj, ticket, snaps */
+} __attribute__ ((packed));
+
+struct ceph_osd_reply_head {
+	__le32 client_inc;                /* client incarnation */
+	__le32 flags;
+	struct ceph_object_layout layout;
+	__le32 osdmap_epoch;
+	struct ceph_eversion reassert_version; /* for replaying uncommitted */
+
+	__le32 result;                    /* result code */
+
+	__le32 object_len;                /* length of object name */
+	__le32 num_ops;
+	struct ceph_osd_op ops[0];  /* ops[], object */
+} __attribute__ ((packed));
+
+
+#endif
diff --git a/include/linux/ceph/types.h b/include/linux/ceph/types.h
new file mode 100644
index 000000000000..28b35a005ec2
--- /dev/null
+++ b/include/linux/ceph/types.h
@@ -0,0 +1,29 @@
+#ifndef _FS_CEPH_TYPES_H
+#define _FS_CEPH_TYPES_H
+
+/* needed before including ceph_fs.h */
+#include <linux/in.h>
+#include <linux/types.h>
+#include <linux/fcntl.h>
+#include <linux/string.h>
+
+#include "ceph_fs.h"
+#include "ceph_frag.h"
+#include "ceph_hash.h"
+
+/*
+ * Identify inodes by both their ino AND snapshot id (a u64).
+ */
+struct ceph_vino {
+	u64 ino;
+	u64 snap;
+};
+
+
+/* context for the caps reservation mechanism */
+struct ceph_cap_reservation {
+	int count;
+};
+
+
+#endif
diff --git a/include/linux/crush/crush.h b/include/linux/crush/crush.h
new file mode 100644
index 000000000000..97e435b191f4
--- /dev/null
+++ b/include/linux/crush/crush.h
@@ -0,0 +1,180 @@
+#ifndef CEPH_CRUSH_CRUSH_H
+#define CEPH_CRUSH_CRUSH_H
+
+#include <linux/types.h>
+
+/*
+ * CRUSH is a pseudo-random data distribution algorithm that
+ * efficiently distributes input values (typically, data objects)
+ * across a heterogeneous, structured storage cluster.
+ *
+ * The algorithm was originally described in detail in this paper
+ * (although the algorithm has evolved somewhat since then):
+ *
+ *     http://www.ssrc.ucsc.edu/Papers/weil-sc06.pdf
+ *
+ * LGPL2
+ */
+
+
+#define CRUSH_MAGIC 0x00010000ul   /* for detecting algorithm revisions */
+
+
+#define CRUSH_MAX_DEPTH 10  /* max crush hierarchy depth */
+#define CRUSH_MAX_SET   10  /* max size of a mapping result */
+
+
+/*
+ * CRUSH uses user-defined "rules" to describe how inputs should be
+ * mapped to devices.  A rule consists of sequence of steps to perform
+ * to generate the set of output devices.
+ */
+struct crush_rule_step {
+	__u32 op;
+	__s32 arg1;
+	__s32 arg2;
+};
+
+/* step op codes */
+enum {
+	CRUSH_RULE_NOOP = 0,
+	CRUSH_RULE_TAKE = 1,          /* arg1 = value to start with */
+	CRUSH_RULE_CHOOSE_FIRSTN = 2, /* arg1 = num items to pick */
+				      /* arg2 = type */
+	CRUSH_RULE_CHOOSE_INDEP = 3,  /* same */
+	CRUSH_RULE_EMIT = 4,          /* no args */
+	CRUSH_RULE_CHOOSE_LEAF_FIRSTN = 6,
+	CRUSH_RULE_CHOOSE_LEAF_INDEP = 7,
+};
+
+/*
+ * for specifying choose num (arg1) relative to the max parameter
+ * passed to do_rule
+ */
+#define CRUSH_CHOOSE_N            0
+#define CRUSH_CHOOSE_N_MINUS(x)   (-(x))
+
+/*
+ * The rule mask is used to describe what the rule is intended for.
+ * Given a ruleset and size of output set, we search through the
+ * rule list for a matching rule_mask.
+ */
+struct crush_rule_mask {
+	__u8 ruleset;
+	__u8 type;
+	__u8 min_size;
+	__u8 max_size;
+};
+
+struct crush_rule {
+	__u32 len;
+	struct crush_rule_mask mask;
+	struct crush_rule_step steps[0];
+};
+
+#define crush_rule_size(len) (sizeof(struct crush_rule) + \
+			      (len)*sizeof(struct crush_rule_step))
+
+
+
+/*
+ * A bucket is a named container of other items (either devices or
+ * other buckets).  Items within a bucket are chosen using one of a
+ * few different algorithms.  The table summarizes how the speed of
+ * each option measures up against mapping stability when items are
+ * added or removed.
+ *
+ *  Bucket Alg     Speed       Additions    Removals
+ *  ------------------------------------------------
+ *  uniform         O(1)       poor         poor
+ *  list            O(n)       optimal      poor
+ *  tree            O(log n)   good         good
+ *  straw           O(n)       optimal      optimal
+ */
+enum {
+	CRUSH_BUCKET_UNIFORM = 1,
+	CRUSH_BUCKET_LIST = 2,
+	CRUSH_BUCKET_TREE = 3,
+	CRUSH_BUCKET_STRAW = 4
+};
+extern const char *crush_bucket_alg_name(int alg);
+
+struct crush_bucket {
+	__s32 id;        /* this'll be negative */
+	__u16 type;      /* non-zero; type=0 is reserved for devices */
+	__u8 alg;        /* one of CRUSH_BUCKET_* */
+	__u8 hash;       /* which hash function to use, CRUSH_HASH_* */
+	__u32 weight;    /* 16-bit fixed point */
+	__u32 size;      /* num items */
+	__s32 *items;
+
+	/*
+	 * cached random permutation: used for uniform bucket and for
+	 * the linear search fallback for the other bucket types.
+	 */
+	__u32 perm_x;  /* @x for which *perm is defined */
+	__u32 perm_n;  /* num elements of *perm that are permuted/defined */
+	__u32 *perm;
+};
+
+struct crush_bucket_uniform {
+	struct crush_bucket h;
+	__u32 item_weight;  /* 16-bit fixed point; all items equally weighted */
+};
+
+struct crush_bucket_list {
+	struct crush_bucket h;
+	__u32 *item_weights;  /* 16-bit fixed point */
+	__u32 *sum_weights;   /* 16-bit fixed point.  element i is sum
+				 of weights 0..i, inclusive */
+};
+
+struct crush_bucket_tree {
+	struct crush_bucket h;  /* note: h.size is _tree_ size, not number of
+				   actual items */
+	__u8 num_nodes;
+	__u32 *node_weights;
+};
+
+struct crush_bucket_straw {
+	struct crush_bucket h;
+	__u32 *item_weights;   /* 16-bit fixed point */
+	__u32 *straws;         /* 16-bit fixed point */
+};
+
+
+
+/*
+ * CRUSH map includes all buckets, rules, etc.
+ */
+struct crush_map {
+	struct crush_bucket **buckets;
+	struct crush_rule **rules;
+
+	/*
+	 * Parent pointers to identify the parent bucket a device or
+	 * bucket in the hierarchy.  If an item appears more than
+	 * once, this is the _last_ time it appeared (where buckets
+	 * are processed in bucket id order, from -1 on down to
+	 * -max_buckets.
+	 */
+	__u32 *bucket_parents;
+	__u32 *device_parents;
+
+	__s32 max_buckets;
+	__u32 max_rules;
+	__s32 max_devices;
+};
+
+
+/* crush.c */
+extern int crush_get_bucket_item_weight(struct crush_bucket *b, int pos);
+extern void crush_calc_parents(struct crush_map *map);
+extern void crush_destroy_bucket_uniform(struct crush_bucket_uniform *b);
+extern void crush_destroy_bucket_list(struct crush_bucket_list *b);
+extern void crush_destroy_bucket_tree(struct crush_bucket_tree *b);
+extern void crush_destroy_bucket_straw(struct crush_bucket_straw *b);
+extern void crush_destroy_bucket(struct crush_bucket *b);
+extern void crush_destroy(struct crush_map *map);
+
+#endif
diff --git a/include/linux/crush/hash.h b/include/linux/crush/hash.h
new file mode 100644
index 000000000000..91e884230d5d
--- /dev/null
+++ b/include/linux/crush/hash.h
@@ -0,0 +1,17 @@
+#ifndef CEPH_CRUSH_HASH_H
+#define CEPH_CRUSH_HASH_H
+
+#define CRUSH_HASH_RJENKINS1   0
+
+#define CRUSH_HASH_DEFAULT CRUSH_HASH_RJENKINS1
+
+extern const char *crush_hash_name(int type);
+
+extern __u32 crush_hash32(int type, __u32 a);
+extern __u32 crush_hash32_2(int type, __u32 a, __u32 b);
+extern __u32 crush_hash32_3(int type, __u32 a, __u32 b, __u32 c);
+extern __u32 crush_hash32_4(int type, __u32 a, __u32 b, __u32 c, __u32 d);
+extern __u32 crush_hash32_5(int type, __u32 a, __u32 b, __u32 c, __u32 d,
+			    __u32 e);
+
+#endif
diff --git a/include/linux/crush/mapper.h b/include/linux/crush/mapper.h
new file mode 100644
index 000000000000..c46b99c18bb0
--- /dev/null
+++ b/include/linux/crush/mapper.h
@@ -0,0 +1,20 @@
+#ifndef CEPH_CRUSH_MAPPER_H
+#define CEPH_CRUSH_MAPPER_H
+
+/*
+ * CRUSH functions for find rules and then mapping an input to an
+ * output set.
+ *
+ * LGPL2
+ */
+
+#include "crush.h"
+
+extern int crush_find_rule(struct crush_map *map, int pool, int type, int size);
+extern int crush_do_rule(struct crush_map *map,
+			 int ruleno,
+			 int x, int *result, int result_max,
+			 int forcefeed,    /* -1 for none */
+			 __u32 *weights);
+
+#endif
diff --git a/net/Kconfig b/net/Kconfig
index e926884c1675..55fd82e9ffd9 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -293,6 +293,7 @@ source "net/wimax/Kconfig"
 source "net/rfkill/Kconfig"
 source "net/9p/Kconfig"
 source "net/caif/Kconfig"
+source "net/ceph/Kconfig"
 
 
 endif   # if NET
diff --git a/net/Makefile b/net/Makefile
index ea60fbce9b1b..6b7bfd7f1416 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -68,3 +68,4 @@ obj-$(CONFIG_SYSCTL)		+= sysctl_net.o
 endif
 obj-$(CONFIG_WIMAX)		+= wimax/
 obj-$(CONFIG_DNS_RESOLVER)	+= dns_resolver/
+obj-$(CONFIG_CEPH_LIB)		+= ceph/
diff --git a/net/ceph/Kconfig b/net/ceph/Kconfig
new file mode 100644
index 000000000000..ad424049b0cf
--- /dev/null
+++ b/net/ceph/Kconfig
@@ -0,0 +1,28 @@
+config CEPH_LIB
+        tristate "Ceph core library (EXPERIMENTAL)"
+	depends on INET && EXPERIMENTAL
+	select LIBCRC32C
+	select CRYPTO_AES
+	select CRYPTO
+	default n
+	help
+	  Choose Y or M here to include cephlib, which provides the
+	  common functionality to both the Ceph filesystem and
+	  to the rados block device (rbd).
+
+	  More information at http://ceph.newdream.net/.
+
+	  If unsure, say N.
+
+config CEPH_LIB_PRETTYDEBUG
+	bool "Include file:line in ceph debug output"
+	depends on CEPH_LIB
+	default n
+	help
+	  If you say Y here, debug output will include a filename and
+	  line to aid debugging.  This increases kernel size and slows
+	  execution slightly when debug call sites are enabled (e.g.,
+	  via CONFIG_DYNAMIC_DEBUG).
+
+	  If unsure, say N.
+
diff --git a/net/ceph/Makefile b/net/ceph/Makefile
new file mode 100644
index 000000000000..aab1cabb8035
--- /dev/null
+++ b/net/ceph/Makefile
@@ -0,0 +1,37 @@
+#
+# Makefile for CEPH filesystem.
+#
+
+ifneq ($(KERNELRELEASE),)
+
+obj-$(CONFIG_CEPH_LIB) += libceph.o
+
+libceph-objs := ceph_common.o messenger.o msgpool.o buffer.o pagelist.o \
+	mon_client.o \
+	osd_client.o osdmap.o crush/crush.o crush/mapper.o crush/hash.o \
+	debugfs.o \
+	auth.o auth_none.o \
+	crypto.o armor.o \
+	auth_x.o \
+	ceph_fs.o ceph_strings.o ceph_hash.o \
+	pagevec.o
+
+else
+#Otherwise we were called directly from the command
+# line; invoke the kernel build system.
+
+KERNELDIR ?= /lib/modules/$(shell uname -r)/build
+PWD := $(shell pwd)
+
+default: all
+
+all:
+	$(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules
+
+modules_install:
+	$(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules_install
+
+clean:
+	$(MAKE) -C $(KERNELDIR) M=$(PWD) clean
+
+endif
diff --git a/net/ceph/armor.c b/net/ceph/armor.c
new file mode 100644
index 000000000000..eb2a666b0be7
--- /dev/null
+++ b/net/ceph/armor.c
@@ -0,0 +1,103 @@
+
+#include <linux/errno.h>
+
+int ceph_armor(char *dst, const char *src, const char *end);
+int ceph_unarmor(char *dst, const char *src, const char *end);
+
+/*
+ * base64 encode/decode.
+ */
+
+static const char *pem_key =
+	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+
+static int encode_bits(int c)
+{
+	return pem_key[c];
+}
+
+static int decode_bits(char c)
+{
+	if (c >= 'A' && c <= 'Z')
+		return c - 'A';
+	if (c >= 'a' && c <= 'z')
+		return c - 'a' + 26;
+	if (c >= '0' && c <= '9')
+		return c - '0' + 52;
+	if (c == '+')
+		return 62;
+	if (c == '/')
+		return 63;
+	if (c == '=')
+		return 0; /* just non-negative, please */
+	return -EINVAL;
+}
+
+int ceph_armor(char *dst, const char *src, const char *end)
+{
+	int olen = 0;
+	int line = 0;
+
+	while (src < end) {
+		unsigned char a, b, c;
+
+		a = *src++;
+		*dst++ = encode_bits(a >> 2);
+		if (src < end) {
+			b = *src++;
+			*dst++ = encode_bits(((a & 3) << 4) | (b >> 4));
+			if (src < end) {
+				c = *src++;
+				*dst++ = encode_bits(((b & 15) << 2) |
+						     (c >> 6));
+				*dst++ = encode_bits(c & 63);
+			} else {
+				*dst++ = encode_bits((b & 15) << 2);
+				*dst++ = '=';
+			}
+		} else {
+			*dst++ = encode_bits(((a & 3) << 4));
+			*dst++ = '=';
+			*dst++ = '=';
+		}
+		olen += 4;
+		line += 4;
+		if (line == 64) {
+			line = 0;
+			*(dst++) = '\n';
+			olen++;
+		}
+	}
+	return olen;
+}
+
+int ceph_unarmor(char *dst, const char *src, const char *end)
+{
+	int olen = 0;
+
+	while (src < end) {
+		int a, b, c, d;
+
+		if (src < end && src[0] == '\n')
+			src++;
+		if (src + 4 > end)
+			return -EINVAL;
+		a = decode_bits(src[0]);
+		b = decode_bits(src[1]);
+		c = decode_bits(src[2]);
+		d = decode_bits(src[3]);
+		if (a < 0 || b < 0 || c < 0 || d < 0)
+			return -EINVAL;
+
+		*dst++ = (a << 2) | (b >> 4);
+		if (src[2] == '=')
+			return olen + 1;
+		*dst++ = ((b & 15) << 4) | (c >> 2);
+		if (src[3] == '=')
+			return olen + 2;
+		*dst++ = ((c & 3) << 6) | d;
+		olen += 3;
+		src += 4;
+	}
+	return olen;
+}
diff --git a/net/ceph/auth.c b/net/ceph/auth.c
new file mode 100644
index 000000000000..549c1f43e1d5
--- /dev/null
+++ b/net/ceph/auth.c
@@ -0,0 +1,259 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/slab.h>
+
+#include <linux/ceph/types.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/messenger.h>
+#include "auth_none.h"
+#include "auth_x.h"
+
+
+/*
+ * get protocol handler
+ */
+static u32 supported_protocols[] = {
+	CEPH_AUTH_NONE,
+	CEPH_AUTH_CEPHX
+};
+
+static int ceph_auth_init_protocol(struct ceph_auth_client *ac, int protocol)
+{
+	switch (protocol) {
+	case CEPH_AUTH_NONE:
+		return ceph_auth_none_init(ac);
+	case CEPH_AUTH_CEPHX:
+		return ceph_x_init(ac);
+	default:
+		return -ENOENT;
+	}
+}
+
+/*
+ * setup, teardown.
+ */
+struct ceph_auth_client *ceph_auth_init(const char *name, const char *secret)
+{
+	struct ceph_auth_client *ac;
+	int ret;
+
+	dout("auth_init name '%s' secret '%s'\n", name, secret);
+
+	ret = -ENOMEM;
+	ac = kzalloc(sizeof(*ac), GFP_NOFS);
+	if (!ac)
+		goto out;
+
+	ac->negotiating = true;
+	if (name)
+		ac->name = name;
+	else
+		ac->name = CEPH_AUTH_NAME_DEFAULT;
+	dout("auth_init name %s secret %s\n", ac->name, secret);
+	ac->secret = secret;
+	return ac;
+
+out:
+	return ERR_PTR(ret);
+}
+
+void ceph_auth_destroy(struct ceph_auth_client *ac)
+{
+	dout("auth_destroy %p\n", ac);
+	if (ac->ops)
+		ac->ops->destroy(ac);
+	kfree(ac);
+}
+
+/*
+ * Reset occurs when reconnecting to the monitor.
+ */
+void ceph_auth_reset(struct ceph_auth_client *ac)
+{
+	dout("auth_reset %p\n", ac);
+	if (ac->ops && !ac->negotiating)
+		ac->ops->reset(ac);
+	ac->negotiating = true;
+}
+
+int ceph_entity_name_encode(const char *name, void **p, void *end)
+{
+	int len = strlen(name);
+
+	if (*p + 2*sizeof(u32) + len > end)
+		return -ERANGE;
+	ceph_encode_32(p, CEPH_ENTITY_TYPE_CLIENT);
+	ceph_encode_32(p, len);
+	ceph_encode_copy(p, name, len);
+	return 0;
+}
+
+/*
+ * Initiate protocol negotiation with monitor.  Include entity name
+ * and list supported protocols.
+ */
+int ceph_auth_build_hello(struct ceph_auth_client *ac, void *buf, size_t len)
+{
+	struct ceph_mon_request_header *monhdr = buf;
+	void *p = monhdr + 1, *end = buf + len, *lenp;
+	int i, num;
+	int ret;
+
+	dout("auth_build_hello\n");
+	monhdr->have_version = 0;
+	monhdr->session_mon = cpu_to_le16(-1);
+	monhdr->session_mon_tid = 0;
+
+	ceph_encode_32(&p, 0);  /* no protocol, yet */
+
+	lenp = p;
+	p += sizeof(u32);
+
+	ceph_decode_need(&p, end, 1 + sizeof(u32), bad);
+	ceph_encode_8(&p, 1);
+	num = ARRAY_SIZE(supported_protocols);
+	ceph_encode_32(&p, num);
+	ceph_decode_need(&p, end, num * sizeof(u32), bad);
+	for (i = 0; i < num; i++)
+		ceph_encode_32(&p, supported_protocols[i]);
+
+	ret = ceph_entity_name_encode(ac->name, &p, end);
+	if (ret < 0)
+		return ret;
+	ceph_decode_need(&p, end, sizeof(u64), bad);
+	ceph_encode_64(&p, ac->global_id);
+
+	ceph_encode_32(&lenp, p - lenp - sizeof(u32));
+	return p - buf;
+
+bad:
+	return -ERANGE;
+}
+
+static int ceph_build_auth_request(struct ceph_auth_client *ac,
+				   void *msg_buf, size_t msg_len)
+{
+	struct ceph_mon_request_header *monhdr = msg_buf;
+	void *p = monhdr + 1;
+	void *end = msg_buf + msg_len;
+	int ret;
+
+	monhdr->have_version = 0;
+	monhdr->session_mon = cpu_to_le16(-1);
+	monhdr->session_mon_tid = 0;
+
+	ceph_encode_32(&p, ac->protocol);
+
+	ret = ac->ops->build_request(ac, p + sizeof(u32), end);
+	if (ret < 0) {
+		pr_err("error %d building auth method %s request\n", ret,
+		       ac->ops->name);
+		return ret;
+	}
+	dout(" built request %d bytes\n", ret);
+	ceph_encode_32(&p, ret);
+	return p + ret - msg_buf;
+}
+
+/*
+ * Handle auth message from monitor.
+ */
+int ceph_handle_auth_reply(struct ceph_auth_client *ac,
+			   void *buf, size_t len,
+			   void *reply_buf, size_t reply_len)
+{
+	void *p = buf;
+	void *end = buf + len;
+	int protocol;
+	s32 result;
+	u64 global_id;
+	void *payload, *payload_end;
+	int payload_len;
+	char *result_msg;
+	int result_msg_len;
+	int ret = -EINVAL;
+
+	dout("handle_auth_reply %p %p\n", p, end);
+	ceph_decode_need(&p, end, sizeof(u32) * 3 + sizeof(u64), bad);
+	protocol = ceph_decode_32(&p);
+	result = ceph_decode_32(&p);
+	global_id = ceph_decode_64(&p);
+	payload_len = ceph_decode_32(&p);
+	payload = p;
+	p += payload_len;
+	ceph_decode_need(&p, end, sizeof(u32), bad);
+	result_msg_len = ceph_decode_32(&p);
+	result_msg = p;
+	p += result_msg_len;
+	if (p != end)
+		goto bad;
+
+	dout(" result %d '%.*s' gid %llu len %d\n", result, result_msg_len,
+	     result_msg, global_id, payload_len);
+
+	payload_end = payload + payload_len;
+
+	if (global_id && ac->global_id != global_id) {
+		dout(" set global_id %lld -> %lld\n", ac->global_id, global_id);
+		ac->global_id = global_id;
+	}
+
+	if (ac->negotiating) {
+		/* server does not support our protocols? */
+		if (!protocol && result < 0) {
+			ret = result;
+			goto out;
+		}
+		/* set up (new) protocol handler? */
+		if (ac->protocol && ac->protocol != protocol) {
+			ac->ops->destroy(ac);
+			ac->protocol = 0;
+			ac->ops = NULL;
+		}
+		if (ac->protocol != protocol) {
+			ret = ceph_auth_init_protocol(ac, protocol);
+			if (ret) {
+				pr_err("error %d on auth protocol %d init\n",
+				       ret, protocol);
+				goto out;
+			}
+		}
+
+		ac->negotiating = false;
+	}
+
+	ret = ac->ops->handle_reply(ac, result, payload, payload_end);
+	if (ret == -EAGAIN) {
+		return ceph_build_auth_request(ac, reply_buf, reply_len);
+	} else if (ret) {
+		pr_err("auth method '%s' error %d\n", ac->ops->name, ret);
+		return ret;
+	}
+	return 0;
+
+bad:
+	pr_err("failed to decode auth msg\n");
+out:
+	return ret;
+}
+
+int ceph_build_auth(struct ceph_auth_client *ac,
+		    void *msg_buf, size_t msg_len)
+{
+	if (!ac->protocol)
+		return ceph_auth_build_hello(ac, msg_buf, msg_len);
+	BUG_ON(!ac->ops);
+	if (ac->ops->should_authenticate(ac))
+		return ceph_build_auth_request(ac, msg_buf, msg_len);
+	return 0;
+}
+
+int ceph_auth_is_authenticated(struct ceph_auth_client *ac)
+{
+	if (!ac->ops)
+		return 0;
+	return ac->ops->is_authenticated(ac);
+}
diff --git a/net/ceph/auth_none.c b/net/ceph/auth_none.c
new file mode 100644
index 000000000000..214c2bb43d62
--- /dev/null
+++ b/net/ceph/auth_none.c
@@ -0,0 +1,132 @@
+
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/err.h>
+#include <linux/module.h>
+#include <linux/random.h>
+#include <linux/slab.h>
+
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+
+#include "auth_none.h"
+
+static void reset(struct ceph_auth_client *ac)
+{
+	struct ceph_auth_none_info *xi = ac->private;
+
+	xi->starting = true;
+	xi->built_authorizer = false;
+}
+
+static void destroy(struct ceph_auth_client *ac)
+{
+	kfree(ac->private);
+	ac->private = NULL;
+}
+
+static int is_authenticated(struct ceph_auth_client *ac)
+{
+	struct ceph_auth_none_info *xi = ac->private;
+
+	return !xi->starting;
+}
+
+static int should_authenticate(struct ceph_auth_client *ac)
+{
+	struct ceph_auth_none_info *xi = ac->private;
+
+	return xi->starting;
+}
+
+/*
+ * the generic auth code decode the global_id, and we carry no actual
+ * authenticate state, so nothing happens here.
+ */
+static int handle_reply(struct ceph_auth_client *ac, int result,
+			void *buf, void *end)
+{
+	struct ceph_auth_none_info *xi = ac->private;
+
+	xi->starting = false;
+	return result;
+}
+
+/*
+ * build an 'authorizer' with our entity_name and global_id.  we can
+ * reuse a single static copy since it is identical for all services
+ * we connect to.
+ */
+static int ceph_auth_none_create_authorizer(
+	struct ceph_auth_client *ac, int peer_type,
+	struct ceph_authorizer **a,
+	void **buf, size_t *len,
+	void **reply_buf, size_t *reply_len)
+{
+	struct ceph_auth_none_info *ai = ac->private;
+	struct ceph_none_authorizer *au = &ai->au;
+	void *p, *end;
+	int ret;
+
+	if (!ai->built_authorizer) {
+		p = au->buf;
+		end = p + sizeof(au->buf);
+		ceph_encode_8(&p, 1);
+		ret = ceph_entity_name_encode(ac->name, &p, end - 8);
+		if (ret < 0)
+			goto bad;
+		ceph_decode_need(&p, end, sizeof(u64), bad2);
+		ceph_encode_64(&p, ac->global_id);
+		au->buf_len = p - (void *)au->buf;
+		ai->built_authorizer = true;
+		dout("built authorizer len %d\n", au->buf_len);
+	}
+
+	*a = (struct ceph_authorizer *)au;
+	*buf = au->buf;
+	*len = au->buf_len;
+	*reply_buf = au->reply_buf;
+	*reply_len = sizeof(au->reply_buf);
+	return 0;
+
+bad2:
+	ret = -ERANGE;
+bad:
+	return ret;
+}
+
+static void ceph_auth_none_destroy_authorizer(struct ceph_auth_client *ac,
+				      struct ceph_authorizer *a)
+{
+	/* nothing to do */
+}
+
+static const struct ceph_auth_client_ops ceph_auth_none_ops = {
+	.name = "none",
+	.reset = reset,
+	.destroy = destroy,
+	.is_authenticated = is_authenticated,
+	.should_authenticate = should_authenticate,
+	.handle_reply = handle_reply,
+	.create_authorizer = ceph_auth_none_create_authorizer,
+	.destroy_authorizer = ceph_auth_none_destroy_authorizer,
+};
+
+int ceph_auth_none_init(struct ceph_auth_client *ac)
+{
+	struct ceph_auth_none_info *xi;
+
+	dout("ceph_auth_none_init %p\n", ac);
+	xi = kzalloc(sizeof(*xi), GFP_NOFS);
+	if (!xi)
+		return -ENOMEM;
+
+	xi->starting = true;
+	xi->built_authorizer = false;
+
+	ac->protocol = CEPH_AUTH_NONE;
+	ac->private = xi;
+	ac->ops = &ceph_auth_none_ops;
+	return 0;
+}
+
diff --git a/net/ceph/auth_none.h b/net/ceph/auth_none.h
new file mode 100644
index 000000000000..ed7d088b1bc9
--- /dev/null
+++ b/net/ceph/auth_none.h
@@ -0,0 +1,29 @@
+#ifndef _FS_CEPH_AUTH_NONE_H
+#define _FS_CEPH_AUTH_NONE_H
+
+#include <linux/slab.h>
+#include <linux/ceph/auth.h>
+
+/*
+ * null security mode.
+ *
+ * we use a single static authorizer that simply encodes our entity name
+ * and global id.
+ */
+
+struct ceph_none_authorizer {
+	char buf[128];
+	int buf_len;
+	char reply_buf[0];
+};
+
+struct ceph_auth_none_info {
+	bool starting;
+	bool built_authorizer;
+	struct ceph_none_authorizer au;   /* we only need one; it's static */
+};
+
+extern int ceph_auth_none_init(struct ceph_auth_client *ac);
+
+#endif
+
diff --git a/net/ceph/auth_x.c b/net/ceph/auth_x.c
new file mode 100644
index 000000000000..7fd5dfcf6e18
--- /dev/null
+++ b/net/ceph/auth_x.c
@@ -0,0 +1,688 @@
+
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/err.h>
+#include <linux/module.h>
+#include <linux/random.h>
+#include <linux/slab.h>
+
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+
+#include "crypto.h"
+#include "auth_x.h"
+#include "auth_x_protocol.h"
+
+#define TEMP_TICKET_BUF_LEN	256
+
+static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed);
+
+static int ceph_x_is_authenticated(struct ceph_auth_client *ac)
+{
+	struct ceph_x_info *xi = ac->private;
+	int need;
+
+	ceph_x_validate_tickets(ac, &need);
+	dout("ceph_x_is_authenticated want=%d need=%d have=%d\n",
+	     ac->want_keys, need, xi->have_keys);
+	return (ac->want_keys & xi->have_keys) == ac->want_keys;
+}
+
+static int ceph_x_should_authenticate(struct ceph_auth_client *ac)
+{
+	struct ceph_x_info *xi = ac->private;
+	int need;
+
+	ceph_x_validate_tickets(ac, &need);
+	dout("ceph_x_should_authenticate want=%d need=%d have=%d\n",
+	     ac->want_keys, need, xi->have_keys);
+	return need != 0;
+}
+
+static int ceph_x_encrypt_buflen(int ilen)
+{
+	return sizeof(struct ceph_x_encrypt_header) + ilen + 16 +
+		sizeof(u32);
+}
+
+static int ceph_x_encrypt(struct ceph_crypto_key *secret,
+			  void *ibuf, int ilen, void *obuf, size_t olen)
+{
+	struct ceph_x_encrypt_header head = {
+		.struct_v = 1,
+		.magic = cpu_to_le64(CEPHX_ENC_MAGIC)
+	};
+	size_t len = olen - sizeof(u32);
+	int ret;
+
+	ret = ceph_encrypt2(secret, obuf + sizeof(u32), &len,
+			    &head, sizeof(head), ibuf, ilen);
+	if (ret)
+		return ret;
+	ceph_encode_32(&obuf, len);
+	return len + sizeof(u32);
+}
+
+static int ceph_x_decrypt(struct ceph_crypto_key *secret,
+			  void **p, void *end, void *obuf, size_t olen)
+{
+	struct ceph_x_encrypt_header head;
+	size_t head_len = sizeof(head);
+	int len, ret;
+
+	len = ceph_decode_32(p);
+	if (*p + len > end)
+		return -EINVAL;
+
+	dout("ceph_x_decrypt len %d\n", len);
+	ret = ceph_decrypt2(secret, &head, &head_len, obuf, &olen,
+			    *p, len);
+	if (ret)
+		return ret;
+	if (head.struct_v != 1 || le64_to_cpu(head.magic) != CEPHX_ENC_MAGIC)
+		return -EPERM;
+	*p += len;
+	return olen;
+}
+
+/*
+ * get existing (or insert new) ticket handler
+ */
+static struct ceph_x_ticket_handler *
+get_ticket_handler(struct ceph_auth_client *ac, int service)
+{
+	struct ceph_x_ticket_handler *th;
+	struct ceph_x_info *xi = ac->private;
+	struct rb_node *parent = NULL, **p = &xi->ticket_handlers.rb_node;
+
+	while (*p) {
+		parent = *p;
+		th = rb_entry(parent, struct ceph_x_ticket_handler, node);
+		if (service < th->service)
+			p = &(*p)->rb_left;
+		else if (service > th->service)
+			p = &(*p)->rb_right;
+		else
+			return th;
+	}
+
+	/* add it */
+	th = kzalloc(sizeof(*th), GFP_NOFS);
+	if (!th)
+		return ERR_PTR(-ENOMEM);
+	th->service = service;
+	rb_link_node(&th->node, parent, p);
+	rb_insert_color(&th->node, &xi->ticket_handlers);
+	return th;
+}
+
+static void remove_ticket_handler(struct ceph_auth_client *ac,
+				  struct ceph_x_ticket_handler *th)
+{
+	struct ceph_x_info *xi = ac->private;
+
+	dout("remove_ticket_handler %p %d\n", th, th->service);
+	rb_erase(&th->node, &xi->ticket_handlers);
+	ceph_crypto_key_destroy(&th->session_key);
+	if (th->ticket_blob)
+		ceph_buffer_put(th->ticket_blob);
+	kfree(th);
+}
+
+static int ceph_x_proc_ticket_reply(struct ceph_auth_client *ac,
+				    struct ceph_crypto_key *secret,
+				    void *buf, void *end)
+{
+	struct ceph_x_info *xi = ac->private;
+	int num;
+	void *p = buf;
+	int ret;
+	char *dbuf;
+	char *ticket_buf;
+	u8 reply_struct_v;
+
+	dbuf = kmalloc(TEMP_TICKET_BUF_LEN, GFP_NOFS);
+	if (!dbuf)
+		return -ENOMEM;
+
+	ret = -ENOMEM;
+	ticket_buf = kmalloc(TEMP_TICKET_BUF_LEN, GFP_NOFS);
+	if (!ticket_buf)
+		goto out_dbuf;
+
+	ceph_decode_need(&p, end, 1 + sizeof(u32), bad);
+	reply_struct_v = ceph_decode_8(&p);
+	if (reply_struct_v != 1)
+		goto bad;
+	num = ceph_decode_32(&p);
+	dout("%d tickets\n", num);
+	while (num--) {
+		int type;
+		u8 tkt_struct_v, blob_struct_v;
+		struct ceph_x_ticket_handler *th;
+		void *dp, *dend;
+		int dlen;
+		char is_enc;
+		struct timespec validity;
+		struct ceph_crypto_key old_key;
+		void *tp, *tpend;
+		struct ceph_timespec new_validity;
+		struct ceph_crypto_key new_session_key;
+		struct ceph_buffer *new_ticket_blob;
+		unsigned long new_expires, new_renew_after;
+		u64 new_secret_id;
+
+		ceph_decode_need(&p, end, sizeof(u32) + 1, bad);
+
+		type = ceph_decode_32(&p);
+		dout(" ticket type %d %s\n", type, ceph_entity_type_name(type));
+
+		tkt_struct_v = ceph_decode_8(&p);
+		if (tkt_struct_v != 1)
+			goto bad;
+
+		th = get_ticket_handler(ac, type);
+		if (IS_ERR(th)) {
+			ret = PTR_ERR(th);
+			goto out;
+		}
+
+		/* blob for me */
+		dlen = ceph_x_decrypt(secret, &p, end, dbuf,
+				      TEMP_TICKET_BUF_LEN);
+		if (dlen <= 0) {
+			ret = dlen;
+			goto out;
+		}
+		dout(" decrypted %d bytes\n", dlen);
+		dend = dbuf + dlen;
+		dp = dbuf;
+
+		tkt_struct_v = ceph_decode_8(&dp);
+		if (tkt_struct_v != 1)
+			goto bad;
+
+		memcpy(&old_key, &th->session_key, sizeof(old_key));
+		ret = ceph_crypto_key_decode(&new_session_key, &dp, dend);
+		if (ret)
+			goto out;
+
+		ceph_decode_copy(&dp, &new_validity, sizeof(new_validity));
+		ceph_decode_timespec(&validity, &new_validity);
+		new_expires = get_seconds() + validity.tv_sec;
+		new_renew_after = new_expires - (validity.tv_sec / 4);
+		dout(" expires=%lu renew_after=%lu\n", new_expires,
+		     new_renew_after);
+
+		/* ticket blob for service */
+		ceph_decode_8_safe(&p, end, is_enc, bad);
+		tp = ticket_buf;
+		if (is_enc) {
+			/* encrypted */
+			dout(" encrypted ticket\n");
+			dlen = ceph_x_decrypt(&old_key, &p, end, ticket_buf,
+					      TEMP_TICKET_BUF_LEN);
+			if (dlen < 0) {
+				ret = dlen;
+				goto out;
+			}
+			dlen = ceph_decode_32(&tp);
+		} else {
+			/* unencrypted */
+			ceph_decode_32_safe(&p, end, dlen, bad);
+			ceph_decode_need(&p, end, dlen, bad);
+			ceph_decode_copy(&p, ticket_buf, dlen);
+		}
+		tpend = tp + dlen;
+		dout(" ticket blob is %d bytes\n", dlen);
+		ceph_decode_need(&tp, tpend, 1 + sizeof(u64), bad);
+		blob_struct_v = ceph_decode_8(&tp);
+		new_secret_id = ceph_decode_64(&tp);
+		ret = ceph_decode_buffer(&new_ticket_blob, &tp, tpend);
+		if (ret)
+			goto out;
+
+		/* all is well, update our ticket */
+		ceph_crypto_key_destroy(&th->session_key);
+		if (th->ticket_blob)
+			ceph_buffer_put(th->ticket_blob);
+		th->session_key = new_session_key;
+		th->ticket_blob = new_ticket_blob;
+		th->validity = new_validity;
+		th->secret_id = new_secret_id;
+		th->expires = new_expires;
+		th->renew_after = new_renew_after;
+		dout(" got ticket service %d (%s) secret_id %lld len %d\n",
+		     type, ceph_entity_type_name(type), th->secret_id,
+		     (int)th->ticket_blob->vec.iov_len);
+		xi->have_keys |= th->service;
+	}
+
+	ret = 0;
+out:
+	kfree(ticket_buf);
+out_dbuf:
+	kfree(dbuf);
+	return ret;
+
+bad:
+	ret = -EINVAL;
+	goto out;
+}
+
+static int ceph_x_build_authorizer(struct ceph_auth_client *ac,
+				   struct ceph_x_ticket_handler *th,
+				   struct ceph_x_authorizer *au)
+{
+	int maxlen;
+	struct ceph_x_authorize_a *msg_a;
+	struct ceph_x_authorize_b msg_b;
+	void *p, *end;
+	int ret;
+	int ticket_blob_len =
+		(th->ticket_blob ? th->ticket_blob->vec.iov_len : 0);
+
+	dout("build_authorizer for %s %p\n",
+	     ceph_entity_type_name(th->service), au);
+
+	maxlen = sizeof(*msg_a) + sizeof(msg_b) +
+		ceph_x_encrypt_buflen(ticket_blob_len);
+	dout("  need len %d\n", maxlen);
+	if (au->buf && au->buf->alloc_len < maxlen) {
+		ceph_buffer_put(au->buf);
+		au->buf = NULL;
+	}
+	if (!au->buf) {
+		au->buf = ceph_buffer_new(maxlen, GFP_NOFS);
+		if (!au->buf)
+			return -ENOMEM;
+	}
+	au->service = th->service;
+
+	msg_a = au->buf->vec.iov_base;
+	msg_a->struct_v = 1;
+	msg_a->global_id = cpu_to_le64(ac->global_id);
+	msg_a->service_id = cpu_to_le32(th->service);
+	msg_a->ticket_blob.struct_v = 1;
+	msg_a->ticket_blob.secret_id = cpu_to_le64(th->secret_id);
+	msg_a->ticket_blob.blob_len = cpu_to_le32(ticket_blob_len);
+	if (ticket_blob_len) {
+		memcpy(msg_a->ticket_blob.blob, th->ticket_blob->vec.iov_base,
+		       th->ticket_blob->vec.iov_len);
+	}
+	dout(" th %p secret_id %lld %lld\n", th, th->secret_id,
+	     le64_to_cpu(msg_a->ticket_blob.secret_id));
+
+	p = msg_a + 1;
+	p += ticket_blob_len;
+	end = au->buf->vec.iov_base + au->buf->vec.iov_len;
+
+	get_random_bytes(&au->nonce, sizeof(au->nonce));
+	msg_b.struct_v = 1;
+	msg_b.nonce = cpu_to_le64(au->nonce);
+	ret = ceph_x_encrypt(&th->session_key, &msg_b, sizeof(msg_b),
+			     p, end - p);
+	if (ret < 0)
+		goto out_buf;
+	p += ret;
+	au->buf->vec.iov_len = p - au->buf->vec.iov_base;
+	dout(" built authorizer nonce %llx len %d\n", au->nonce,
+	     (int)au->buf->vec.iov_len);
+	BUG_ON(au->buf->vec.iov_len > maxlen);
+	return 0;
+
+out_buf:
+	ceph_buffer_put(au->buf);
+	au->buf = NULL;
+	return ret;
+}
+
+static int ceph_x_encode_ticket(struct ceph_x_ticket_handler *th,
+				void **p, void *end)
+{
+	ceph_decode_need(p, end, 1 + sizeof(u64), bad);
+	ceph_encode_8(p, 1);
+	ceph_encode_64(p, th->secret_id);
+	if (th->ticket_blob) {
+		const char *buf = th->ticket_blob->vec.iov_base;
+		u32 len = th->ticket_blob->vec.iov_len;
+
+		ceph_encode_32_safe(p, end, len, bad);
+		ceph_encode_copy_safe(p, end, buf, len, bad);
+	} else {
+		ceph_encode_32_safe(p, end, 0, bad);
+	}
+
+	return 0;
+bad:
+	return -ERANGE;
+}
+
+static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed)
+{
+	int want = ac->want_keys;
+	struct ceph_x_info *xi = ac->private;
+	int service;
+
+	*pneed = ac->want_keys & ~(xi->have_keys);
+
+	for (service = 1; service <= want; service <<= 1) {
+		struct ceph_x_ticket_handler *th;
+
+		if (!(ac->want_keys & service))
+			continue;
+
+		if (*pneed & service)
+			continue;
+
+		th = get_ticket_handler(ac, service);
+
+		if (IS_ERR(th)) {
+			*pneed |= service;
+			continue;
+		}
+
+		if (get_seconds() >= th->renew_after)
+			*pneed |= service;
+		if (get_seconds() >= th->expires)
+			xi->have_keys &= ~service;
+	}
+}
+
+
+static int ceph_x_build_request(struct ceph_auth_client *ac,
+				void *buf, void *end)
+{
+	struct ceph_x_info *xi = ac->private;
+	int need;
+	struct ceph_x_request_header *head = buf;
+	int ret;
+	struct ceph_x_ticket_handler *th =
+		get_ticket_handler(ac, CEPH_ENTITY_TYPE_AUTH);
+
+	if (IS_ERR(th))
+		return PTR_ERR(th);
+
+	ceph_x_validate_tickets(ac, &need);
+
+	dout("build_request want %x have %x need %x\n",
+	     ac->want_keys, xi->have_keys, need);
+
+	if (need & CEPH_ENTITY_TYPE_AUTH) {
+		struct ceph_x_authenticate *auth = (void *)(head + 1);
+		void *p = auth + 1;
+		struct ceph_x_challenge_blob tmp;
+		char tmp_enc[40];
+		u64 *u;
+
+		if (p > end)
+			return -ERANGE;
+
+		dout(" get_auth_session_key\n");
+		head->op = cpu_to_le16(CEPHX_GET_AUTH_SESSION_KEY);
+
+		/* encrypt and hash */
+		get_random_bytes(&auth->client_challenge, sizeof(u64));
+		tmp.client_challenge = auth->client_challenge;
+		tmp.server_challenge = cpu_to_le64(xi->server_challenge);
+		ret = ceph_x_encrypt(&xi->secret, &tmp, sizeof(tmp),
+				     tmp_enc, sizeof(tmp_enc));
+		if (ret < 0)
+			return ret;
+
+		auth->struct_v = 1;
+		auth->key = 0;
+		for (u = (u64 *)tmp_enc; u + 1 <= (u64 *)(tmp_enc + ret); u++)
+			auth->key ^= *(__le64 *)u;
+		dout(" server_challenge %llx client_challenge %llx key %llx\n",
+		     xi->server_challenge, le64_to_cpu(auth->client_challenge),
+		     le64_to_cpu(auth->key));
+
+		/* now encode the old ticket if exists */
+		ret = ceph_x_encode_ticket(th, &p, end);
+		if (ret < 0)
+			return ret;
+
+		return p - buf;
+	}
+
+	if (need) {
+		void *p = head + 1;
+		struct ceph_x_service_ticket_request *req;
+
+		if (p > end)
+			return -ERANGE;
+		head->op = cpu_to_le16(CEPHX_GET_PRINCIPAL_SESSION_KEY);
+
+		ret = ceph_x_build_authorizer(ac, th, &xi->auth_authorizer);
+		if (ret)
+			return ret;
+		ceph_encode_copy(&p, xi->auth_authorizer.buf->vec.iov_base,
+				 xi->auth_authorizer.buf->vec.iov_len);
+
+		req = p;
+		req->keys = cpu_to_le32(need);
+		p += sizeof(*req);
+		return p - buf;
+	}
+
+	return 0;
+}
+
+static int ceph_x_handle_reply(struct ceph_auth_client *ac, int result,
+			       void *buf, void *end)
+{
+	struct ceph_x_info *xi = ac->private;
+	struct ceph_x_reply_header *head = buf;
+	struct ceph_x_ticket_handler *th;
+	int len = end - buf;
+	int op;
+	int ret;
+
+	if (result)
+		return result;  /* XXX hmm? */
+
+	if (xi->starting) {
+		/* it's a hello */
+		struct ceph_x_server_challenge *sc = buf;
+
+		if (len != sizeof(*sc))
+			return -EINVAL;
+		xi->server_challenge = le64_to_cpu(sc->server_challenge);
+		dout("handle_reply got server challenge %llx\n",
+		     xi->server_challenge);
+		xi->starting = false;
+		xi->have_keys &= ~CEPH_ENTITY_TYPE_AUTH;
+		return -EAGAIN;
+	}
+
+	op = le16_to_cpu(head->op);
+	result = le32_to_cpu(head->result);
+	dout("handle_reply op %d result %d\n", op, result);
+	switch (op) {
+	case CEPHX_GET_AUTH_SESSION_KEY:
+		/* verify auth key */
+		ret = ceph_x_proc_ticket_reply(ac, &xi->secret,
+					       buf + sizeof(*head), end);
+		break;
+
+	case CEPHX_GET_PRINCIPAL_SESSION_KEY:
+		th = get_ticket_handler(ac, CEPH_ENTITY_TYPE_AUTH);
+		if (IS_ERR(th))
+			return PTR_ERR(th);
+		ret = ceph_x_proc_ticket_reply(ac, &th->session_key,
+					       buf + sizeof(*head), end);
+		break;
+
+	default:
+		return -EINVAL;
+	}
+	if (ret)
+		return ret;
+	if (ac->want_keys == xi->have_keys)
+		return 0;
+	return -EAGAIN;
+}
+
+static int ceph_x_create_authorizer(
+	struct ceph_auth_client *ac, int peer_type,
+	struct ceph_authorizer **a,
+	void **buf, size_t *len,
+	void **reply_buf, size_t *reply_len)
+{
+	struct ceph_x_authorizer *au;
+	struct ceph_x_ticket_handler *th;
+	int ret;
+
+	th = get_ticket_handler(ac, peer_type);
+	if (IS_ERR(th))
+		return PTR_ERR(th);
+
+	au = kzalloc(sizeof(*au), GFP_NOFS);
+	if (!au)
+		return -ENOMEM;
+
+	ret = ceph_x_build_authorizer(ac, th, au);
+	if (ret) {
+		kfree(au);
+		return ret;
+	}
+
+	*a = (struct ceph_authorizer *)au;
+	*buf = au->buf->vec.iov_base;
+	*len = au->buf->vec.iov_len;
+	*reply_buf = au->reply_buf;
+	*reply_len = sizeof(au->reply_buf);
+	return 0;
+}
+
+static int ceph_x_verify_authorizer_reply(struct ceph_auth_client *ac,
+					  struct ceph_authorizer *a, size_t len)
+{
+	struct ceph_x_authorizer *au = (void *)a;
+	struct ceph_x_ticket_handler *th;
+	int ret = 0;
+	struct ceph_x_authorize_reply reply;
+	void *p = au->reply_buf;
+	void *end = p + sizeof(au->reply_buf);
+
+	th = get_ticket_handler(ac, au->service);
+	if (IS_ERR(th))
+		return PTR_ERR(th);
+	ret = ceph_x_decrypt(&th->session_key, &p, end, &reply, sizeof(reply));
+	if (ret < 0)
+		return ret;
+	if (ret != sizeof(reply))
+		return -EPERM;
+
+	if (au->nonce + 1 != le64_to_cpu(reply.nonce_plus_one))
+		ret = -EPERM;
+	else
+		ret = 0;
+	dout("verify_authorizer_reply nonce %llx got %llx ret %d\n",
+	     au->nonce, le64_to_cpu(reply.nonce_plus_one), ret);
+	return ret;
+}
+
+static void ceph_x_destroy_authorizer(struct ceph_auth_client *ac,
+				      struct ceph_authorizer *a)
+{
+	struct ceph_x_authorizer *au = (void *)a;
+
+	ceph_buffer_put(au->buf);
+	kfree(au);
+}
+
+
+static void ceph_x_reset(struct ceph_auth_client *ac)
+{
+	struct ceph_x_info *xi = ac->private;
+
+	dout("reset\n");
+	xi->starting = true;
+	xi->server_challenge = 0;
+}
+
+static void ceph_x_destroy(struct ceph_auth_client *ac)
+{
+	struct ceph_x_info *xi = ac->private;
+	struct rb_node *p;
+
+	dout("ceph_x_destroy %p\n", ac);
+	ceph_crypto_key_destroy(&xi->secret);
+
+	while ((p = rb_first(&xi->ticket_handlers)) != NULL) {
+		struct ceph_x_ticket_handler *th =
+			rb_entry(p, struct ceph_x_ticket_handler, node);
+		remove_ticket_handler(ac, th);
+	}
+
+	if (xi->auth_authorizer.buf)
+		ceph_buffer_put(xi->auth_authorizer.buf);
+
+	kfree(ac->private);
+	ac->private = NULL;
+}
+
+static void ceph_x_invalidate_authorizer(struct ceph_auth_client *ac,
+				   int peer_type)
+{
+	struct ceph_x_ticket_handler *th;
+
+	th = get_ticket_handler(ac, peer_type);
+	if (!IS_ERR(th))
+		remove_ticket_handler(ac, th);
+}
+
+
+static const struct ceph_auth_client_ops ceph_x_ops = {
+	.name = "x",
+	.is_authenticated = ceph_x_is_authenticated,
+	.should_authenticate = ceph_x_should_authenticate,
+	.build_request = ceph_x_build_request,
+	.handle_reply = ceph_x_handle_reply,
+	.create_authorizer = ceph_x_create_authorizer,
+	.verify_authorizer_reply = ceph_x_verify_authorizer_reply,
+	.destroy_authorizer = ceph_x_destroy_authorizer,
+	.invalidate_authorizer = ceph_x_invalidate_authorizer,
+	.reset =  ceph_x_reset,
+	.destroy = ceph_x_destroy,
+};
+
+
+int ceph_x_init(struct ceph_auth_client *ac)
+{
+	struct ceph_x_info *xi;
+	int ret;
+
+	dout("ceph_x_init %p\n", ac);
+	ret = -ENOMEM;
+	xi = kzalloc(sizeof(*xi), GFP_NOFS);
+	if (!xi)
+		goto out;
+
+	ret = -EINVAL;
+	if (!ac->secret) {
+		pr_err("no secret set (for auth_x protocol)\n");
+		goto out_nomem;
+	}
+
+	ret = ceph_crypto_key_unarmor(&xi->secret, ac->secret);
+	if (ret)
+		goto out_nomem;
+
+	xi->starting = true;
+	xi->ticket_handlers = RB_ROOT;
+
+	ac->protocol = CEPH_AUTH_CEPHX;
+	ac->private = xi;
+	ac->ops = &ceph_x_ops;
+	return 0;
+
+out_nomem:
+	kfree(xi);
+out:
+	return ret;
+}
+
+
diff --git a/net/ceph/auth_x.h b/net/ceph/auth_x.h
new file mode 100644
index 000000000000..e02da7a5c5a1
--- /dev/null
+++ b/net/ceph/auth_x.h
@@ -0,0 +1,50 @@
+#ifndef _FS_CEPH_AUTH_X_H
+#define _FS_CEPH_AUTH_X_H
+
+#include <linux/rbtree.h>
+
+#include <linux/ceph/auth.h>
+
+#include "crypto.h"
+#include "auth_x_protocol.h"
+
+/*
+ * Handle ticket for a single service.
+ */
+struct ceph_x_ticket_handler {
+	struct rb_node node;
+	unsigned service;
+
+	struct ceph_crypto_key session_key;
+	struct ceph_timespec validity;
+
+	u64 secret_id;
+	struct ceph_buffer *ticket_blob;
+
+	unsigned long renew_after, expires;
+};
+
+
+struct ceph_x_authorizer {
+	struct ceph_buffer *buf;
+	unsigned service;
+	u64 nonce;
+	char reply_buf[128];  /* big enough for encrypted blob */
+};
+
+struct ceph_x_info {
+	struct ceph_crypto_key secret;
+
+	bool starting;
+	u64 server_challenge;
+
+	unsigned have_keys;
+	struct rb_root ticket_handlers;
+
+	struct ceph_x_authorizer auth_authorizer;
+};
+
+extern int ceph_x_init(struct ceph_auth_client *ac);
+
+#endif
+
diff --git a/net/ceph/auth_x_protocol.h b/net/ceph/auth_x_protocol.h
new file mode 100644
index 000000000000..671d30576c4f
--- /dev/null
+++ b/net/ceph/auth_x_protocol.h
@@ -0,0 +1,90 @@
+#ifndef __FS_CEPH_AUTH_X_PROTOCOL
+#define __FS_CEPH_AUTH_X_PROTOCOL
+
+#define CEPHX_GET_AUTH_SESSION_KEY      0x0100
+#define CEPHX_GET_PRINCIPAL_SESSION_KEY 0x0200
+#define CEPHX_GET_ROTATING_KEY          0x0400
+
+/* common bits */
+struct ceph_x_ticket_blob {
+	__u8 struct_v;
+	__le64 secret_id;
+	__le32 blob_len;
+	char blob[];
+} __attribute__ ((packed));
+
+
+/* common request/reply headers */
+struct ceph_x_request_header {
+	__le16 op;
+} __attribute__ ((packed));
+
+struct ceph_x_reply_header {
+	__le16 op;
+	__le32 result;
+} __attribute__ ((packed));
+
+
+/* authenticate handshake */
+
+/* initial hello (no reply header) */
+struct ceph_x_server_challenge {
+	__u8 struct_v;
+	__le64 server_challenge;
+} __attribute__ ((packed));
+
+struct ceph_x_authenticate {
+	__u8 struct_v;
+	__le64 client_challenge;
+	__le64 key;
+	/* ticket blob */
+} __attribute__ ((packed));
+
+struct ceph_x_service_ticket_request {
+	__u8 struct_v;
+	__le32 keys;
+} __attribute__ ((packed));
+
+struct ceph_x_challenge_blob {
+	__le64 server_challenge;
+	__le64 client_challenge;
+} __attribute__ ((packed));
+
+
+
+/* authorize handshake */
+
+/*
+ * The authorizer consists of two pieces:
+ *  a - service id, ticket blob
+ *  b - encrypted with session key
+ */
+struct ceph_x_authorize_a {
+	__u8 struct_v;
+	__le64 global_id;
+	__le32 service_id;
+	struct ceph_x_ticket_blob ticket_blob;
+} __attribute__ ((packed));
+
+struct ceph_x_authorize_b {
+	__u8 struct_v;
+	__le64 nonce;
+} __attribute__ ((packed));
+
+struct ceph_x_authorize_reply {
+	__u8 struct_v;
+	__le64 nonce_plus_one;
+} __attribute__ ((packed));
+
+
+/*
+ * encyption bundle
+ */
+#define CEPHX_ENC_MAGIC 0xff009cad8826aa55ull
+
+struct ceph_x_encrypt_header {
+	__u8 struct_v;
+	__le64 magic;
+} __attribute__ ((packed));
+
+#endif
diff --git a/net/ceph/buffer.c b/net/ceph/buffer.c
new file mode 100644
index 000000000000..53d8abfa25d5
--- /dev/null
+++ b/net/ceph/buffer.c
@@ -0,0 +1,68 @@
+
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/module.h>
+#include <linux/slab.h>
+
+#include <linux/ceph/buffer.h>
+#include <linux/ceph/decode.h>
+
+struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp)
+{
+	struct ceph_buffer *b;
+
+	b = kmalloc(sizeof(*b), gfp);
+	if (!b)
+		return NULL;
+
+	b->vec.iov_base = kmalloc(len, gfp | __GFP_NOWARN);
+	if (b->vec.iov_base) {
+		b->is_vmalloc = false;
+	} else {
+		b->vec.iov_base = __vmalloc(len, gfp, PAGE_KERNEL);
+		if (!b->vec.iov_base) {
+			kfree(b);
+			return NULL;
+		}
+		b->is_vmalloc = true;
+	}
+
+	kref_init(&b->kref);
+	b->alloc_len = len;
+	b->vec.iov_len = len;
+	dout("buffer_new %p\n", b);
+	return b;
+}
+EXPORT_SYMBOL(ceph_buffer_new);
+
+void ceph_buffer_release(struct kref *kref)
+{
+	struct ceph_buffer *b = container_of(kref, struct ceph_buffer, kref);
+
+	dout("buffer_release %p\n", b);
+	if (b->vec.iov_base) {
+		if (b->is_vmalloc)
+			vfree(b->vec.iov_base);
+		else
+			kfree(b->vec.iov_base);
+	}
+	kfree(b);
+}
+EXPORT_SYMBOL(ceph_buffer_release);
+
+int ceph_decode_buffer(struct ceph_buffer **b, void **p, void *end)
+{
+	size_t len;
+
+	ceph_decode_need(p, end, sizeof(u32), bad);
+	len = ceph_decode_32(p);
+	dout("decode_buffer len %d\n", (int)len);
+	ceph_decode_need(p, end, len, bad);
+	*b = ceph_buffer_new(len, GFP_NOFS);
+	if (!*b)
+		return -ENOMEM;
+	ceph_decode_copy(p, (*b)->vec.iov_base, len);
+	return 0;
+bad:
+	return -EINVAL;
+}
diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c
new file mode 100644
index 000000000000..f6f2eebc0767
--- /dev/null
+++ b/net/ceph/ceph_common.c
@@ -0,0 +1,529 @@
+
+#include <linux/ceph/ceph_debug.h>
+#include <linux/backing-dev.h>
+#include <linux/ctype.h>
+#include <linux/fs.h>
+#include <linux/inet.h>
+#include <linux/in6.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+#include <linux/parser.h>
+#include <linux/sched.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include <linux/statfs.h>
+#include <linux/string.h>
+
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/debugfs.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+
+
+
+/*
+ * find filename portion of a path (/foo/bar/baz -> baz)
+ */
+const char *ceph_file_part(const char *s, int len)
+{
+	const char *e = s + len;
+
+	while (e != s && *(e-1) != '/')
+		e--;
+	return e;
+}
+EXPORT_SYMBOL(ceph_file_part);
+
+const char *ceph_msg_type_name(int type)
+{
+	switch (type) {
+	case CEPH_MSG_SHUTDOWN: return "shutdown";
+	case CEPH_MSG_PING: return "ping";
+	case CEPH_MSG_AUTH: return "auth";
+	case CEPH_MSG_AUTH_REPLY: return "auth_reply";
+	case CEPH_MSG_MON_MAP: return "mon_map";
+	case CEPH_MSG_MON_GET_MAP: return "mon_get_map";
+	case CEPH_MSG_MON_SUBSCRIBE: return "mon_subscribe";
+	case CEPH_MSG_MON_SUBSCRIBE_ACK: return "mon_subscribe_ack";
+	case CEPH_MSG_STATFS: return "statfs";
+	case CEPH_MSG_STATFS_REPLY: return "statfs_reply";
+	case CEPH_MSG_MDS_MAP: return "mds_map";
+	case CEPH_MSG_CLIENT_SESSION: return "client_session";
+	case CEPH_MSG_CLIENT_RECONNECT: return "client_reconnect";
+	case CEPH_MSG_CLIENT_REQUEST: return "client_request";
+	case CEPH_MSG_CLIENT_REQUEST_FORWARD: return "client_request_forward";
+	case CEPH_MSG_CLIENT_REPLY: return "client_reply";
+	case CEPH_MSG_CLIENT_CAPS: return "client_caps";
+	case CEPH_MSG_CLIENT_CAPRELEASE: return "client_cap_release";
+	case CEPH_MSG_CLIENT_SNAP: return "client_snap";
+	case CEPH_MSG_CLIENT_LEASE: return "client_lease";
+	case CEPH_MSG_OSD_MAP: return "osd_map";
+	case CEPH_MSG_OSD_OP: return "osd_op";
+	case CEPH_MSG_OSD_OPREPLY: return "osd_opreply";
+	default: return "unknown";
+	}
+}
+EXPORT_SYMBOL(ceph_msg_type_name);
+
+/*
+ * Initially learn our fsid, or verify an fsid matches.
+ */
+int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid)
+{
+	if (client->have_fsid) {
+		if (ceph_fsid_compare(&client->fsid, fsid)) {
+			pr_err("bad fsid, had %pU got %pU",
+			       &client->fsid, fsid);
+			return -1;
+		}
+	} else {
+		pr_info("client%lld fsid %pU\n", ceph_client_id(client), fsid);
+		memcpy(&client->fsid, fsid, sizeof(*fsid));
+		ceph_debugfs_client_init(client);
+		client->have_fsid = true;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_check_fsid);
+
+static int strcmp_null(const char *s1, const char *s2)
+{
+	if (!s1 && !s2)
+		return 0;
+	if (s1 && !s2)
+		return -1;
+	if (!s1 && s2)
+		return 1;
+	return strcmp(s1, s2);
+}
+
+int ceph_compare_options(struct ceph_options *new_opt,
+			 struct ceph_client *client)
+{
+	struct ceph_options *opt1 = new_opt;
+	struct ceph_options *opt2 = client->options;
+	int ofs = offsetof(struct ceph_options, mon_addr);
+	int i;
+	int ret;
+
+	ret = memcmp(opt1, opt2, ofs);
+	if (ret)
+		return ret;
+
+	ret = strcmp_null(opt1->name, opt2->name);
+	if (ret)
+		return ret;
+
+	ret = strcmp_null(opt1->secret, opt2->secret);
+	if (ret)
+		return ret;
+
+	/* any matching mon ip implies a match */
+	for (i = 0; i < opt1->num_mon; i++) {
+		if (ceph_monmap_contains(client->monc.monmap,
+				 &opt1->mon_addr[i]))
+			return 0;
+	}
+	return -1;
+}
+EXPORT_SYMBOL(ceph_compare_options);
+
+
+static int parse_fsid(const char *str, struct ceph_fsid *fsid)
+{
+	int i = 0;
+	char tmp[3];
+	int err = -EINVAL;
+	int d;
+
+	dout("parse_fsid '%s'\n", str);
+	tmp[2] = 0;
+	while (*str && i < 16) {
+		if (ispunct(*str)) {
+			str++;
+			continue;
+		}
+		if (!isxdigit(str[0]) || !isxdigit(str[1]))
+			break;
+		tmp[0] = str[0];
+		tmp[1] = str[1];
+		if (sscanf(tmp, "%x", &d) < 1)
+			break;
+		fsid->fsid[i] = d & 0xff;
+		i++;
+		str += 2;
+	}
+
+	if (i == 16)
+		err = 0;
+	dout("parse_fsid ret %d got fsid %pU", err, fsid);
+	return err;
+}
+
+/*
+ * ceph options
+ */
+enum {
+	Opt_osdtimeout,
+	Opt_osdkeepalivetimeout,
+	Opt_mount_timeout,
+	Opt_osd_idle_ttl,
+	Opt_last_int,
+	/* int args above */
+	Opt_fsid,
+	Opt_name,
+	Opt_secret,
+	Opt_ip,
+	Opt_last_string,
+	/* string args above */
+	Opt_noshare,
+	Opt_nocrc,
+};
+
+static match_table_t opt_tokens = {
+	{Opt_osdtimeout, "osdtimeout=%d"},
+	{Opt_osdkeepalivetimeout, "osdkeepalive=%d"},
+	{Opt_mount_timeout, "mount_timeout=%d"},
+	{Opt_osd_idle_ttl, "osd_idle_ttl=%d"},
+	/* int args above */
+	{Opt_fsid, "fsid=%s"},
+	{Opt_name, "name=%s"},
+	{Opt_secret, "secret=%s"},
+	{Opt_ip, "ip=%s"},
+	/* string args above */
+	{Opt_noshare, "noshare"},
+	{Opt_nocrc, "nocrc"},
+	{-1, NULL}
+};
+
+void ceph_destroy_options(struct ceph_options *opt)
+{
+	dout("destroy_options %p\n", opt);
+	kfree(opt->name);
+	kfree(opt->secret);
+	kfree(opt);
+}
+EXPORT_SYMBOL(ceph_destroy_options);
+
+int ceph_parse_options(struct ceph_options **popt, char *options,
+		       const char *dev_name, const char *dev_name_end,
+		       int (*parse_extra_token)(char *c, void *private),
+		       void *private)
+{
+	struct ceph_options *opt;
+	const char *c;
+	int err = -ENOMEM;
+	substring_t argstr[MAX_OPT_ARGS];
+
+	opt = kzalloc(sizeof(*opt), GFP_KERNEL);
+	if (!opt)
+		return err;
+	opt->mon_addr = kcalloc(CEPH_MAX_MON, sizeof(*opt->mon_addr),
+				GFP_KERNEL);
+	if (!opt->mon_addr)
+		goto out;
+
+	dout("parse_options %p options '%s' dev_name '%s'\n", opt, options,
+	     dev_name);
+
+	/* start with defaults */
+	opt->flags = CEPH_OPT_DEFAULT;
+	opt->osd_timeout = CEPH_OSD_TIMEOUT_DEFAULT;
+	opt->osd_keepalive_timeout = CEPH_OSD_KEEPALIVE_DEFAULT;
+	opt->mount_timeout = CEPH_MOUNT_TIMEOUT_DEFAULT; /* seconds */
+	opt->osd_idle_ttl = CEPH_OSD_IDLE_TTL_DEFAULT;   /* seconds */
+
+	/* get mon ip(s) */
+	/* ip1[:port1][,ip2[:port2]...] */
+	err = ceph_parse_ips(dev_name, dev_name_end, opt->mon_addr,
+			     CEPH_MAX_MON, &opt->num_mon);
+	if (err < 0)
+		goto out;
+
+	/* parse mount options */
+	while ((c = strsep(&options, ",")) != NULL) {
+		int token, intval, ret;
+		if (!*c)
+			continue;
+		err = -EINVAL;
+		token = match_token((char *)c, opt_tokens, argstr);
+		if (token < 0) {
+			/* extra? */
+			err = parse_extra_token((char *)c, private);
+			if (err < 0) {
+				pr_err("bad option at '%s'\n", c);
+				goto out;
+			}
+			continue;
+		}
+		if (token < Opt_last_int) {
+			ret = match_int(&argstr[0], &intval);
+			if (ret < 0) {
+				pr_err("bad mount option arg (not int) "
+				       "at '%s'\n", c);
+				continue;
+			}
+			dout("got int token %d val %d\n", token, intval);
+		} else if (token > Opt_last_int && token < Opt_last_string) {
+			dout("got string token %d val %s\n", token,
+			     argstr[0].from);
+		} else {
+			dout("got token %d\n", token);
+		}
+		switch (token) {
+		case Opt_ip:
+			err = ceph_parse_ips(argstr[0].from,
+					     argstr[0].to,
+					     &opt->my_addr,
+					     1, NULL);
+			if (err < 0)
+				goto out;
+			opt->flags |= CEPH_OPT_MYIP;
+			break;
+
+		case Opt_fsid:
+			err = parse_fsid(argstr[0].from, &opt->fsid);
+			if (err == 0)
+				opt->flags |= CEPH_OPT_FSID;
+			break;
+		case Opt_name:
+			opt->name = kstrndup(argstr[0].from,
+					      argstr[0].to-argstr[0].from,
+					      GFP_KERNEL);
+			break;
+		case Opt_secret:
+			opt->secret = kstrndup(argstr[0].from,
+						argstr[0].to-argstr[0].from,
+						GFP_KERNEL);
+			break;
+
+			/* misc */
+		case Opt_osdtimeout:
+			opt->osd_timeout = intval;
+			break;
+		case Opt_osdkeepalivetimeout:
+			opt->osd_keepalive_timeout = intval;
+			break;
+		case Opt_osd_idle_ttl:
+			opt->osd_idle_ttl = intval;
+			break;
+		case Opt_mount_timeout:
+			opt->mount_timeout = intval;
+			break;
+
+		case Opt_noshare:
+			opt->flags |= CEPH_OPT_NOSHARE;
+			break;
+
+		case Opt_nocrc:
+			opt->flags |= CEPH_OPT_NOCRC;
+			break;
+
+		default:
+			BUG_ON(token);
+		}
+	}
+
+	/* success */
+	*popt = opt;
+	return 0;
+
+out:
+	ceph_destroy_options(opt);
+	return err;
+}
+EXPORT_SYMBOL(ceph_parse_options);
+
+u64 ceph_client_id(struct ceph_client *client)
+{
+	return client->monc.auth->global_id;
+}
+EXPORT_SYMBOL(ceph_client_id);
+
+/*
+ * create a fresh client instance
+ */
+struct ceph_client *ceph_create_client(struct ceph_options *opt, void *private)
+{
+	struct ceph_client *client;
+	int err = -ENOMEM;
+
+	client = kzalloc(sizeof(*client), GFP_KERNEL);
+	if (client == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	client->private = private;
+	client->options = opt;
+
+	mutex_init(&client->mount_mutex);
+	init_waitqueue_head(&client->auth_wq);
+	client->auth_err = 0;
+
+	client->extra_mon_dispatch = NULL;
+	client->supported_features = CEPH_FEATURE_SUPPORTED_DEFAULT;
+	client->required_features = CEPH_FEATURE_REQUIRED_DEFAULT;
+
+	client->msgr = NULL;
+
+	/* subsystems */
+	err = ceph_monc_init(&client->monc, client);
+	if (err < 0)
+		goto fail;
+	err = ceph_osdc_init(&client->osdc, client);
+	if (err < 0)
+		goto fail_monc;
+
+	return client;
+
+fail_monc:
+	ceph_monc_stop(&client->monc);
+fail:
+	kfree(client);
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL(ceph_create_client);
+
+void ceph_destroy_client(struct ceph_client *client)
+{
+	dout("destroy_client %p\n", client);
+
+	/* unmount */
+	ceph_osdc_stop(&client->osdc);
+
+	/*
+	 * make sure mds and osd connections close out before destroying
+	 * the auth module, which is needed to free those connections'
+	 * ceph_authorizers.
+	 */
+	ceph_msgr_flush();
+
+	ceph_monc_stop(&client->monc);
+
+	ceph_debugfs_client_cleanup(client);
+
+	if (client->msgr)
+		ceph_messenger_destroy(client->msgr);
+
+	ceph_destroy_options(client->options);
+
+	kfree(client);
+	dout("destroy_client %p done\n", client);
+}
+EXPORT_SYMBOL(ceph_destroy_client);
+
+/*
+ * true if we have the mon map (and have thus joined the cluster)
+ */
+static int have_mon_and_osd_map(struct ceph_client *client)
+{
+	return client->monc.monmap && client->monc.monmap->epoch &&
+	       client->osdc.osdmap && client->osdc.osdmap->epoch;
+}
+
+/*
+ * mount: join the ceph cluster, and open root directory.
+ */
+int __ceph_open_session(struct ceph_client *client, unsigned long started)
+{
+	struct ceph_entity_addr *myaddr = NULL;
+	int err;
+	unsigned long timeout = client->options->mount_timeout * HZ;
+
+	/* initialize the messenger */
+	if (client->msgr == NULL) {
+		if (ceph_test_opt(client, MYIP))
+			myaddr = &client->options->my_addr;
+		client->msgr = ceph_messenger_create(myaddr,
+					client->supported_features,
+					client->required_features);
+		if (IS_ERR(client->msgr)) {
+			client->msgr = NULL;
+			return PTR_ERR(client->msgr);
+		}
+		client->msgr->nocrc = ceph_test_opt(client, NOCRC);
+	}
+
+	/* open session, and wait for mon and osd maps */
+	err = ceph_monc_open_session(&client->monc);
+	if (err < 0)
+		return err;
+
+	while (!have_mon_and_osd_map(client)) {
+		err = -EIO;
+		if (timeout && time_after_eq(jiffies, started + timeout))
+			return err;
+
+		/* wait */
+		dout("mount waiting for mon_map\n");
+		err = wait_event_interruptible_timeout(client->auth_wq,
+			have_mon_and_osd_map(client) || (client->auth_err < 0),
+			timeout);
+		if (err == -EINTR || err == -ERESTARTSYS)
+			return err;
+		if (client->auth_err < 0)
+			return client->auth_err;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(__ceph_open_session);
+
+
+int ceph_open_session(struct ceph_client *client)
+{
+	int ret;
+	unsigned long started = jiffies;  /* note the start time */
+
+	dout("open_session start\n");
+	mutex_lock(&client->mount_mutex);
+
+	ret = __ceph_open_session(client, started);
+
+	mutex_unlock(&client->mount_mutex);
+	return ret;
+}
+EXPORT_SYMBOL(ceph_open_session);
+
+
+static int __init init_ceph_lib(void)
+{
+	int ret = 0;
+
+	ret = ceph_debugfs_init();
+	if (ret < 0)
+		goto out;
+
+	ret = ceph_msgr_init();
+	if (ret < 0)
+		goto out_debugfs;
+
+	pr_info("loaded (mon/osd proto %d/%d, osdmap %d/%d %d/%d)\n",
+		CEPH_MONC_PROTOCOL, CEPH_OSDC_PROTOCOL,
+		CEPH_OSDMAP_VERSION, CEPH_OSDMAP_VERSION_EXT,
+		CEPH_OSDMAP_INC_VERSION, CEPH_OSDMAP_INC_VERSION_EXT);
+
+	return 0;
+
+out_debugfs:
+	ceph_debugfs_cleanup();
+out:
+	return ret;
+}
+
+static void __exit exit_ceph_lib(void)
+{
+	dout("exit_ceph_lib\n");
+	ceph_msgr_exit();
+	ceph_debugfs_cleanup();
+}
+
+module_init(init_ceph_lib);
+module_exit(exit_ceph_lib);
+
+MODULE_AUTHOR("Sage Weil <sage@newdream.net>");
+MODULE_AUTHOR("Yehuda Sadeh <yehuda@hq.newdream.net>");
+MODULE_AUTHOR("Patience Warnick <patience@newdream.net>");
+MODULE_DESCRIPTION("Ceph filesystem for Linux");
+MODULE_LICENSE("GPL");
diff --git a/net/ceph/ceph_fs.c b/net/ceph/ceph_fs.c
new file mode 100644
index 000000000000..a3a3a31d3c37
--- /dev/null
+++ b/net/ceph/ceph_fs.c
@@ -0,0 +1,75 @@
+/*
+ * Some non-inline ceph helpers
+ */
+#include <linux/module.h>
+#include <linux/ceph/types.h>
+
+/*
+ * return true if @layout appears to be valid
+ */
+int ceph_file_layout_is_valid(const struct ceph_file_layout *layout)
+{
+	__u32 su = le32_to_cpu(layout->fl_stripe_unit);
+	__u32 sc = le32_to_cpu(layout->fl_stripe_count);
+	__u32 os = le32_to_cpu(layout->fl_object_size);
+
+	/* stripe unit, object size must be non-zero, 64k increment */
+	if (!su || (su & (CEPH_MIN_STRIPE_UNIT-1)))
+		return 0;
+	if (!os || (os & (CEPH_MIN_STRIPE_UNIT-1)))
+		return 0;
+	/* object size must be a multiple of stripe unit */
+	if (os < su || os % su)
+		return 0;
+	/* stripe count must be non-zero */
+	if (!sc)
+		return 0;
+	return 1;
+}
+
+
+int ceph_flags_to_mode(int flags)
+{
+	int mode;
+
+#ifdef O_DIRECTORY  /* fixme */
+	if ((flags & O_DIRECTORY) == O_DIRECTORY)
+		return CEPH_FILE_MODE_PIN;
+#endif
+	if ((flags & O_APPEND) == O_APPEND)
+		flags |= O_WRONLY;
+
+	if ((flags & O_ACCMODE) == O_RDWR)
+		mode = CEPH_FILE_MODE_RDWR;
+	else if ((flags & O_ACCMODE) == O_WRONLY)
+		mode = CEPH_FILE_MODE_WR;
+	else
+		mode = CEPH_FILE_MODE_RD;
+
+#ifdef O_LAZY
+	if (flags & O_LAZY)
+		mode |= CEPH_FILE_MODE_LAZY;
+#endif
+
+	return mode;
+}
+EXPORT_SYMBOL(ceph_flags_to_mode);
+
+int ceph_caps_for_mode(int mode)
+{
+	int caps = CEPH_CAP_PIN;
+
+	if (mode & CEPH_FILE_MODE_RD)
+		caps |= CEPH_CAP_FILE_SHARED |
+			CEPH_CAP_FILE_RD | CEPH_CAP_FILE_CACHE;
+	if (mode & CEPH_FILE_MODE_WR)
+		caps |= CEPH_CAP_FILE_EXCL |
+			CEPH_CAP_FILE_WR | CEPH_CAP_FILE_BUFFER |
+			CEPH_CAP_AUTH_SHARED | CEPH_CAP_AUTH_EXCL |
+			CEPH_CAP_XATTR_SHARED | CEPH_CAP_XATTR_EXCL;
+	if (mode & CEPH_FILE_MODE_LAZY)
+		caps |= CEPH_CAP_FILE_LAZYIO;
+
+	return caps;
+}
+EXPORT_SYMBOL(ceph_caps_for_mode);
diff --git a/net/ceph/ceph_hash.c b/net/ceph/ceph_hash.c
new file mode 100644
index 000000000000..815ef8826796
--- /dev/null
+++ b/net/ceph/ceph_hash.c
@@ -0,0 +1,118 @@
+
+#include <linux/ceph/types.h>
+
+/*
+ * Robert Jenkin's hash function.
+ * http://burtleburtle.net/bob/hash/evahash.html
+ * This is in the public domain.
+ */
+#define mix(a, b, c)						\
+	do {							\
+		a = a - b;  a = a - c;  a = a ^ (c >> 13);	\
+		b = b - c;  b = b - a;  b = b ^ (a << 8);	\
+		c = c - a;  c = c - b;  c = c ^ (b >> 13);	\
+		a = a - b;  a = a - c;  a = a ^ (c >> 12);	\
+		b = b - c;  b = b - a;  b = b ^ (a << 16);	\
+		c = c - a;  c = c - b;  c = c ^ (b >> 5);	\
+		a = a - b;  a = a - c;  a = a ^ (c >> 3);	\
+		b = b - c;  b = b - a;  b = b ^ (a << 10);	\
+		c = c - a;  c = c - b;  c = c ^ (b >> 15);	\
+	} while (0)
+
+unsigned ceph_str_hash_rjenkins(const char *str, unsigned length)
+{
+	const unsigned char *k = (const unsigned char *)str;
+	__u32 a, b, c;  /* the internal state */
+	__u32 len;      /* how many key bytes still need mixing */
+
+	/* Set up the internal state */
+	len = length;
+	a = 0x9e3779b9;      /* the golden ratio; an arbitrary value */
+	b = a;
+	c = 0;               /* variable initialization of internal state */
+
+	/* handle most of the key */
+	while (len >= 12) {
+		a = a + (k[0] + ((__u32)k[1] << 8) + ((__u32)k[2] << 16) +
+			 ((__u32)k[3] << 24));
+		b = b + (k[4] + ((__u32)k[5] << 8) + ((__u32)k[6] << 16) +
+			 ((__u32)k[7] << 24));
+		c = c + (k[8] + ((__u32)k[9] << 8) + ((__u32)k[10] << 16) +
+			 ((__u32)k[11] << 24));
+		mix(a, b, c);
+		k = k + 12;
+		len = len - 12;
+	}
+
+	/* handle the last 11 bytes */
+	c = c + length;
+	switch (len) {            /* all the case statements fall through */
+	case 11:
+		c = c + ((__u32)k[10] << 24);
+	case 10:
+		c = c + ((__u32)k[9] << 16);
+	case 9:
+		c = c + ((__u32)k[8] << 8);
+		/* the first byte of c is reserved for the length */
+	case 8:
+		b = b + ((__u32)k[7] << 24);
+	case 7:
+		b = b + ((__u32)k[6] << 16);
+	case 6:
+		b = b + ((__u32)k[5] << 8);
+	case 5:
+		b = b + k[4];
+	case 4:
+		a = a + ((__u32)k[3] << 24);
+	case 3:
+		a = a + ((__u32)k[2] << 16);
+	case 2:
+		a = a + ((__u32)k[1] << 8);
+	case 1:
+		a = a + k[0];
+		/* case 0: nothing left to add */
+	}
+	mix(a, b, c);
+
+	return c;
+}
+
+/*
+ * linux dcache hash
+ */
+unsigned ceph_str_hash_linux(const char *str, unsigned length)
+{
+	unsigned long hash = 0;
+	unsigned char c;
+
+	while (length--) {
+		c = *str++;
+		hash = (hash + (c << 4) + (c >> 4)) * 11;
+	}
+	return hash;
+}
+
+
+unsigned ceph_str_hash(int type, const char *s, unsigned len)
+{
+	switch (type) {
+	case CEPH_STR_HASH_LINUX:
+		return ceph_str_hash_linux(s, len);
+	case CEPH_STR_HASH_RJENKINS:
+		return ceph_str_hash_rjenkins(s, len);
+	default:
+		return -1;
+	}
+}
+
+const char *ceph_str_hash_name(int type)
+{
+	switch (type) {
+	case CEPH_STR_HASH_LINUX:
+		return "linux";
+	case CEPH_STR_HASH_RJENKINS:
+		return "rjenkins";
+	default:
+		return "unknown";
+	}
+}
diff --git a/net/ceph/ceph_strings.c b/net/ceph/ceph_strings.c
new file mode 100644
index 000000000000..3fbda04de29c
--- /dev/null
+++ b/net/ceph/ceph_strings.c
@@ -0,0 +1,84 @@
+/*
+ * Ceph string constants
+ */
+#include <linux/module.h>
+#include <linux/ceph/types.h>
+
+const char *ceph_entity_type_name(int type)
+{
+	switch (type) {
+	case CEPH_ENTITY_TYPE_MDS: return "mds";
+	case CEPH_ENTITY_TYPE_OSD: return "osd";
+	case CEPH_ENTITY_TYPE_MON: return "mon";
+	case CEPH_ENTITY_TYPE_CLIENT: return "client";
+	case CEPH_ENTITY_TYPE_AUTH: return "auth";
+	default: return "unknown";
+	}
+}
+
+const char *ceph_osd_op_name(int op)
+{
+	switch (op) {
+	case CEPH_OSD_OP_READ: return "read";
+	case CEPH_OSD_OP_STAT: return "stat";
+
+	case CEPH_OSD_OP_MASKTRUNC: return "masktrunc";
+
+	case CEPH_OSD_OP_WRITE: return "write";
+	case CEPH_OSD_OP_DELETE: return "delete";
+	case CEPH_OSD_OP_TRUNCATE: return "truncate";
+	case CEPH_OSD_OP_ZERO: return "zero";
+	case CEPH_OSD_OP_WRITEFULL: return "writefull";
+	case CEPH_OSD_OP_ROLLBACK: return "rollback";
+
+	case CEPH_OSD_OP_APPEND: return "append";
+	case CEPH_OSD_OP_STARTSYNC: return "startsync";
+	case CEPH_OSD_OP_SETTRUNC: return "settrunc";
+	case CEPH_OSD_OP_TRIMTRUNC: return "trimtrunc";
+
+	case CEPH_OSD_OP_TMAPUP: return "tmapup";
+	case CEPH_OSD_OP_TMAPGET: return "tmapget";
+	case CEPH_OSD_OP_TMAPPUT: return "tmapput";
+
+	case CEPH_OSD_OP_GETXATTR: return "getxattr";
+	case CEPH_OSD_OP_GETXATTRS: return "getxattrs";
+	case CEPH_OSD_OP_SETXATTR: return "setxattr";
+	case CEPH_OSD_OP_SETXATTRS: return "setxattrs";
+	case CEPH_OSD_OP_RESETXATTRS: return "resetxattrs";
+	case CEPH_OSD_OP_RMXATTR: return "rmxattr";
+	case CEPH_OSD_OP_CMPXATTR: return "cmpxattr";
+
+	case CEPH_OSD_OP_PULL: return "pull";
+	case CEPH_OSD_OP_PUSH: return "push";
+	case CEPH_OSD_OP_BALANCEREADS: return "balance-reads";
+	case CEPH_OSD_OP_UNBALANCEREADS: return "unbalance-reads";
+	case CEPH_OSD_OP_SCRUB: return "scrub";
+
+	case CEPH_OSD_OP_WRLOCK: return "wrlock";
+	case CEPH_OSD_OP_WRUNLOCK: return "wrunlock";
+	case CEPH_OSD_OP_RDLOCK: return "rdlock";
+	case CEPH_OSD_OP_RDUNLOCK: return "rdunlock";
+	case CEPH_OSD_OP_UPLOCK: return "uplock";
+	case CEPH_OSD_OP_DNLOCK: return "dnlock";
+
+	case CEPH_OSD_OP_CALL: return "call";
+
+	case CEPH_OSD_OP_PGLS: return "pgls";
+	}
+	return "???";
+}
+
+
+const char *ceph_pool_op_name(int op)
+{
+	switch (op) {
+	case POOL_OP_CREATE: return "create";
+	case POOL_OP_DELETE: return "delete";
+	case POOL_OP_AUID_CHANGE: return "auid change";
+	case POOL_OP_CREATE_SNAP: return "create snap";
+	case POOL_OP_DELETE_SNAP: return "delete snap";
+	case POOL_OP_CREATE_UNMANAGED_SNAP: return "create unmanaged snap";
+	case POOL_OP_DELETE_UNMANAGED_SNAP: return "delete unmanaged snap";
+	}
+	return "???";
+}
diff --git a/net/ceph/crush/crush.c b/net/ceph/crush/crush.c
new file mode 100644
index 000000000000..d6ebb13a18a4
--- /dev/null
+++ b/net/ceph/crush/crush.c
@@ -0,0 +1,151 @@
+
+#ifdef __KERNEL__
+# include <linux/slab.h>
+#else
+# include <stdlib.h>
+# include <assert.h>
+# define kfree(x) do { if (x) free(x); } while (0)
+# define BUG_ON(x) assert(!(x))
+#endif
+
+#include <linux/crush/crush.h>
+
+const char *crush_bucket_alg_name(int alg)
+{
+	switch (alg) {
+	case CRUSH_BUCKET_UNIFORM: return "uniform";
+	case CRUSH_BUCKET_LIST: return "list";
+	case CRUSH_BUCKET_TREE: return "tree";
+	case CRUSH_BUCKET_STRAW: return "straw";
+	default: return "unknown";
+	}
+}
+
+/**
+ * crush_get_bucket_item_weight - Get weight of an item in given bucket
+ * @b: bucket pointer
+ * @p: item index in bucket
+ */
+int crush_get_bucket_item_weight(struct crush_bucket *b, int p)
+{
+	if (p >= b->size)
+		return 0;
+
+	switch (b->alg) {
+	case CRUSH_BUCKET_UNIFORM:
+		return ((struct crush_bucket_uniform *)b)->item_weight;
+	case CRUSH_BUCKET_LIST:
+		return ((struct crush_bucket_list *)b)->item_weights[p];
+	case CRUSH_BUCKET_TREE:
+		if (p & 1)
+			return ((struct crush_bucket_tree *)b)->node_weights[p];
+		return 0;
+	case CRUSH_BUCKET_STRAW:
+		return ((struct crush_bucket_straw *)b)->item_weights[p];
+	}
+	return 0;
+}
+
+/**
+ * crush_calc_parents - Calculate parent vectors for the given crush map.
+ * @map: crush_map pointer
+ */
+void crush_calc_parents(struct crush_map *map)
+{
+	int i, b, c;
+
+	for (b = 0; b < map->max_buckets; b++) {
+		if (map->buckets[b] == NULL)
+			continue;
+		for (i = 0; i < map->buckets[b]->size; i++) {
+			c = map->buckets[b]->items[i];
+			BUG_ON(c >= map->max_devices ||
+			       c < -map->max_buckets);
+			if (c >= 0)
+				map->device_parents[c] = map->buckets[b]->id;
+			else
+				map->bucket_parents[-1-c] = map->buckets[b]->id;
+		}
+	}
+}
+
+void crush_destroy_bucket_uniform(struct crush_bucket_uniform *b)
+{
+	kfree(b->h.perm);
+	kfree(b->h.items);
+	kfree(b);
+}
+
+void crush_destroy_bucket_list(struct crush_bucket_list *b)
+{
+	kfree(b->item_weights);
+	kfree(b->sum_weights);
+	kfree(b->h.perm);
+	kfree(b->h.items);
+	kfree(b);
+}
+
+void crush_destroy_bucket_tree(struct crush_bucket_tree *b)
+{
+	kfree(b->node_weights);
+	kfree(b);
+}
+
+void crush_destroy_bucket_straw(struct crush_bucket_straw *b)
+{
+	kfree(b->straws);
+	kfree(b->item_weights);
+	kfree(b->h.perm);
+	kfree(b->h.items);
+	kfree(b);
+}
+
+void crush_destroy_bucket(struct crush_bucket *b)
+{
+	switch (b->alg) {
+	case CRUSH_BUCKET_UNIFORM:
+		crush_destroy_bucket_uniform((struct crush_bucket_uniform *)b);
+		break;
+	case CRUSH_BUCKET_LIST:
+		crush_destroy_bucket_list((struct crush_bucket_list *)b);
+		break;
+	case CRUSH_BUCKET_TREE:
+		crush_destroy_bucket_tree((struct crush_bucket_tree *)b);
+		break;
+	case CRUSH_BUCKET_STRAW:
+		crush_destroy_bucket_straw((struct crush_bucket_straw *)b);
+		break;
+	}
+}
+
+/**
+ * crush_destroy - Destroy a crush_map
+ * @map: crush_map pointer
+ */
+void crush_destroy(struct crush_map *map)
+{
+	int b;
+
+	/* buckets */
+	if (map->buckets) {
+		for (b = 0; b < map->max_buckets; b++) {
+			if (map->buckets[b] == NULL)
+				continue;
+			crush_destroy_bucket(map->buckets[b]);
+		}
+		kfree(map->buckets);
+	}
+
+	/* rules */
+	if (map->rules) {
+		for (b = 0; b < map->max_rules; b++)
+			kfree(map->rules[b]);
+		kfree(map->rules);
+	}
+
+	kfree(map->bucket_parents);
+	kfree(map->device_parents);
+	kfree(map);
+}
+
+
diff --git a/net/ceph/crush/hash.c b/net/ceph/crush/hash.c
new file mode 100644
index 000000000000..5bb63e37a8a1
--- /dev/null
+++ b/net/ceph/crush/hash.c
@@ -0,0 +1,149 @@
+
+#include <linux/types.h>
+#include <linux/crush/hash.h>
+
+/*
+ * Robert Jenkins' function for mixing 32-bit values
+ * http://burtleburtle.net/bob/hash/evahash.html
+ * a, b = random bits, c = input and output
+ */
+#define crush_hashmix(a, b, c) do {			\
+		a = a-b;  a = a-c;  a = a^(c>>13);	\
+		b = b-c;  b = b-a;  b = b^(a<<8);	\
+		c = c-a;  c = c-b;  c = c^(b>>13);	\
+		a = a-b;  a = a-c;  a = a^(c>>12);	\
+		b = b-c;  b = b-a;  b = b^(a<<16);	\
+		c = c-a;  c = c-b;  c = c^(b>>5);	\
+		a = a-b;  a = a-c;  a = a^(c>>3);	\
+		b = b-c;  b = b-a;  b = b^(a<<10);	\
+		c = c-a;  c = c-b;  c = c^(b>>15);	\
+	} while (0)
+
+#define crush_hash_seed 1315423911
+
+static __u32 crush_hash32_rjenkins1(__u32 a)
+{
+	__u32 hash = crush_hash_seed ^ a;
+	__u32 b = a;
+	__u32 x = 231232;
+	__u32 y = 1232;
+	crush_hashmix(b, x, hash);
+	crush_hashmix(y, a, hash);
+	return hash;
+}
+
+static __u32 crush_hash32_rjenkins1_2(__u32 a, __u32 b)
+{
+	__u32 hash = crush_hash_seed ^ a ^ b;
+	__u32 x = 231232;
+	__u32 y = 1232;
+	crush_hashmix(a, b, hash);
+	crush_hashmix(x, a, hash);
+	crush_hashmix(b, y, hash);
+	return hash;
+}
+
+static __u32 crush_hash32_rjenkins1_3(__u32 a, __u32 b, __u32 c)
+{
+	__u32 hash = crush_hash_seed ^ a ^ b ^ c;
+	__u32 x = 231232;
+	__u32 y = 1232;
+	crush_hashmix(a, b, hash);
+	crush_hashmix(c, x, hash);
+	crush_hashmix(y, a, hash);
+	crush_hashmix(b, x, hash);
+	crush_hashmix(y, c, hash);
+	return hash;
+}
+
+static __u32 crush_hash32_rjenkins1_4(__u32 a, __u32 b, __u32 c, __u32 d)
+{
+	__u32 hash = crush_hash_seed ^ a ^ b ^ c ^ d;
+	__u32 x = 231232;
+	__u32 y = 1232;
+	crush_hashmix(a, b, hash);
+	crush_hashmix(c, d, hash);
+	crush_hashmix(a, x, hash);
+	crush_hashmix(y, b, hash);
+	crush_hashmix(c, x, hash);
+	crush_hashmix(y, d, hash);
+	return hash;
+}
+
+static __u32 crush_hash32_rjenkins1_5(__u32 a, __u32 b, __u32 c, __u32 d,
+				      __u32 e)
+{
+	__u32 hash = crush_hash_seed ^ a ^ b ^ c ^ d ^ e;
+	__u32 x = 231232;
+	__u32 y = 1232;
+	crush_hashmix(a, b, hash);
+	crush_hashmix(c, d, hash);
+	crush_hashmix(e, x, hash);
+	crush_hashmix(y, a, hash);
+	crush_hashmix(b, x, hash);
+	crush_hashmix(y, c, hash);
+	crush_hashmix(d, x, hash);
+	crush_hashmix(y, e, hash);
+	return hash;
+}
+
+
+__u32 crush_hash32(int type, __u32 a)
+{
+	switch (type) {
+	case CRUSH_HASH_RJENKINS1:
+		return crush_hash32_rjenkins1(a);
+	default:
+		return 0;
+	}
+}
+
+__u32 crush_hash32_2(int type, __u32 a, __u32 b)
+{
+	switch (type) {
+	case CRUSH_HASH_RJENKINS1:
+		return crush_hash32_rjenkins1_2(a, b);
+	default:
+		return 0;
+	}
+}
+
+__u32 crush_hash32_3(int type, __u32 a, __u32 b, __u32 c)
+{
+	switch (type) {
+	case CRUSH_HASH_RJENKINS1:
+		return crush_hash32_rjenkins1_3(a, b, c);
+	default:
+		return 0;
+	}
+}
+
+__u32 crush_hash32_4(int type, __u32 a, __u32 b, __u32 c, __u32 d)
+{
+	switch (type) {
+	case CRUSH_HASH_RJENKINS1:
+		return crush_hash32_rjenkins1_4(a, b, c, d);
+	default:
+		return 0;
+	}
+}
+
+__u32 crush_hash32_5(int type, __u32 a, __u32 b, __u32 c, __u32 d, __u32 e)
+{
+	switch (type) {
+	case CRUSH_HASH_RJENKINS1:
+		return crush_hash32_rjenkins1_5(a, b, c, d, e);
+	default:
+		return 0;
+	}
+}
+
+const char *crush_hash_name(int type)
+{
+	switch (type) {
+	case CRUSH_HASH_RJENKINS1:
+		return "rjenkins1";
+	default:
+		return "unknown";
+	}
+}
diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c
new file mode 100644
index 000000000000..42599e31dcad
--- /dev/null
+++ b/net/ceph/crush/mapper.c
@@ -0,0 +1,609 @@
+
+#ifdef __KERNEL__
+# include <linux/string.h>
+# include <linux/slab.h>
+# include <linux/bug.h>
+# include <linux/kernel.h>
+# ifndef dprintk
+#  define dprintk(args...)
+# endif
+#else
+# include <string.h>
+# include <stdio.h>
+# include <stdlib.h>
+# include <assert.h>
+# define BUG_ON(x) assert(!(x))
+# define dprintk(args...) /* printf(args) */
+# define kmalloc(x, f) malloc(x)
+# define kfree(x) free(x)
+#endif
+
+#include <linux/crush/crush.h>
+#include <linux/crush/hash.h>
+
+/*
+ * Implement the core CRUSH mapping algorithm.
+ */
+
+/**
+ * crush_find_rule - find a crush_rule id for a given ruleset, type, and size.
+ * @map: the crush_map
+ * @ruleset: the storage ruleset id (user defined)
+ * @type: storage ruleset type (user defined)
+ * @size: output set size
+ */
+int crush_find_rule(struct crush_map *map, int ruleset, int type, int size)
+{
+	int i;
+
+	for (i = 0; i < map->max_rules; i++) {
+		if (map->rules[i] &&
+		    map->rules[i]->mask.ruleset == ruleset &&
+		    map->rules[i]->mask.type == type &&
+		    map->rules[i]->mask.min_size <= size &&
+		    map->rules[i]->mask.max_size >= size)
+			return i;
+	}
+	return -1;
+}
+
+
+/*
+ * bucket choose methods
+ *
+ * For each bucket algorithm, we have a "choose" method that, given a
+ * crush input @x and replica position (usually, position in output set) @r,
+ * will produce an item in the bucket.
+ */
+
+/*
+ * Choose based on a random permutation of the bucket.
+ *
+ * We used to use some prime number arithmetic to do this, but it
+ * wasn't very random, and had some other bad behaviors.  Instead, we
+ * calculate an actual random permutation of the bucket members.
+ * Since this is expensive, we optimize for the r=0 case, which
+ * captures the vast majority of calls.
+ */
+static int bucket_perm_choose(struct crush_bucket *bucket,
+			      int x, int r)
+{
+	unsigned pr = r % bucket->size;
+	unsigned i, s;
+
+	/* start a new permutation if @x has changed */
+	if (bucket->perm_x != x || bucket->perm_n == 0) {
+		dprintk("bucket %d new x=%d\n", bucket->id, x);
+		bucket->perm_x = x;
+
+		/* optimize common r=0 case */
+		if (pr == 0) {
+			s = crush_hash32_3(bucket->hash, x, bucket->id, 0) %
+				bucket->size;
+			bucket->perm[0] = s;
+			bucket->perm_n = 0xffff;   /* magic value, see below */
+			goto out;
+		}
+
+		for (i = 0; i < bucket->size; i++)
+			bucket->perm[i] = i;
+		bucket->perm_n = 0;
+	} else if (bucket->perm_n == 0xffff) {
+		/* clean up after the r=0 case above */
+		for (i = 1; i < bucket->size; i++)
+			bucket->perm[i] = i;
+		bucket->perm[bucket->perm[0]] = 0;
+		bucket->perm_n = 1;
+	}
+
+	/* calculate permutation up to pr */
+	for (i = 0; i < bucket->perm_n; i++)
+		dprintk(" perm_choose have %d: %d\n", i, bucket->perm[i]);
+	while (bucket->perm_n <= pr) {
+		unsigned p = bucket->perm_n;
+		/* no point in swapping the final entry */
+		if (p < bucket->size - 1) {
+			i = crush_hash32_3(bucket->hash, x, bucket->id, p) %
+				(bucket->size - p);
+			if (i) {
+				unsigned t = bucket->perm[p + i];
+				bucket->perm[p + i] = bucket->perm[p];
+				bucket->perm[p] = t;
+			}
+			dprintk(" perm_choose swap %d with %d\n", p, p+i);
+		}
+		bucket->perm_n++;
+	}
+	for (i = 0; i < bucket->size; i++)
+		dprintk(" perm_choose  %d: %d\n", i, bucket->perm[i]);
+
+	s = bucket->perm[pr];
+out:
+	dprintk(" perm_choose %d sz=%d x=%d r=%d (%d) s=%d\n", bucket->id,
+		bucket->size, x, r, pr, s);
+	return bucket->items[s];
+}
+
+/* uniform */
+static int bucket_uniform_choose(struct crush_bucket_uniform *bucket,
+				 int x, int r)
+{
+	return bucket_perm_choose(&bucket->h, x, r);
+}
+
+/* list */
+static int bucket_list_choose(struct crush_bucket_list *bucket,
+			      int x, int r)
+{
+	int i;
+
+	for (i = bucket->h.size-1; i >= 0; i--) {
+		__u64 w = crush_hash32_4(bucket->h.hash,x, bucket->h.items[i],
+					 r, bucket->h.id);
+		w &= 0xffff;
+		dprintk("list_choose i=%d x=%d r=%d item %d weight %x "
+			"sw %x rand %llx",
+			i, x, r, bucket->h.items[i], bucket->item_weights[i],
+			bucket->sum_weights[i], w);
+		w *= bucket->sum_weights[i];
+		w = w >> 16;
+		/*dprintk(" scaled %llx\n", w);*/
+		if (w < bucket->item_weights[i])
+			return bucket->h.items[i];
+	}
+
+	BUG_ON(1);
+	return 0;
+}
+
+
+/* (binary) tree */
+static int height(int n)
+{
+	int h = 0;
+	while ((n & 1) == 0) {
+		h++;
+		n = n >> 1;
+	}
+	return h;
+}
+
+static int left(int x)
+{
+	int h = height(x);
+	return x - (1 << (h-1));
+}
+
+static int right(int x)
+{
+	int h = height(x);
+	return x + (1 << (h-1));
+}
+
+static int terminal(int x)
+{
+	return x & 1;
+}
+
+static int bucket_tree_choose(struct crush_bucket_tree *bucket,
+			      int x, int r)
+{
+	int n, l;
+	__u32 w;
+	__u64 t;
+
+	/* start at root */
+	n = bucket->num_nodes >> 1;
+
+	while (!terminal(n)) {
+		/* pick point in [0, w) */
+		w = bucket->node_weights[n];
+		t = (__u64)crush_hash32_4(bucket->h.hash, x, n, r,
+					  bucket->h.id) * (__u64)w;
+		t = t >> 32;
+
+		/* descend to the left or right? */
+		l = left(n);
+		if (t < bucket->node_weights[l])
+			n = l;
+		else
+			n = right(n);
+	}
+
+	return bucket->h.items[n >> 1];
+}
+
+
+/* straw */
+
+static int bucket_straw_choose(struct crush_bucket_straw *bucket,
+			       int x, int r)
+{
+	int i;
+	int high = 0;
+	__u64 high_draw = 0;
+	__u64 draw;
+
+	for (i = 0; i < bucket->h.size; i++) {
+		draw = crush_hash32_3(bucket->h.hash, x, bucket->h.items[i], r);
+		draw &= 0xffff;
+		draw *= bucket->straws[i];
+		if (i == 0 || draw > high_draw) {
+			high = i;
+			high_draw = draw;
+		}
+	}
+	return bucket->h.items[high];
+}
+
+static int crush_bucket_choose(struct crush_bucket *in, int x, int r)
+{
+	dprintk(" crush_bucket_choose %d x=%d r=%d\n", in->id, x, r);
+	switch (in->alg) {
+	case CRUSH_BUCKET_UNIFORM:
+		return bucket_uniform_choose((struct crush_bucket_uniform *)in,
+					  x, r);
+	case CRUSH_BUCKET_LIST:
+		return bucket_list_choose((struct crush_bucket_list *)in,
+					  x, r);
+	case CRUSH_BUCKET_TREE:
+		return bucket_tree_choose((struct crush_bucket_tree *)in,
+					  x, r);
+	case CRUSH_BUCKET_STRAW:
+		return bucket_straw_choose((struct crush_bucket_straw *)in,
+					   x, r);
+	default:
+		BUG_ON(1);
+		return in->items[0];
+	}
+}
+
+/*
+ * true if device is marked "out" (failed, fully offloaded)
+ * of the cluster
+ */
+static int is_out(struct crush_map *map, __u32 *weight, int item, int x)
+{
+	if (weight[item] >= 0x10000)
+		return 0;
+	if (weight[item] == 0)
+		return 1;
+	if ((crush_hash32_2(CRUSH_HASH_RJENKINS1, x, item) & 0xffff)
+	    < weight[item])
+		return 0;
+	return 1;
+}
+
+/**
+ * crush_choose - choose numrep distinct items of given type
+ * @map: the crush_map
+ * @bucket: the bucket we are choose an item from
+ * @x: crush input value
+ * @numrep: the number of items to choose
+ * @type: the type of item to choose
+ * @out: pointer to output vector
+ * @outpos: our position in that vector
+ * @firstn: true if choosing "first n" items, false if choosing "indep"
+ * @recurse_to_leaf: true if we want one device under each item of given type
+ * @out2: second output vector for leaf items (if @recurse_to_leaf)
+ */
+static int crush_choose(struct crush_map *map,
+			struct crush_bucket *bucket,
+			__u32 *weight,
+			int x, int numrep, int type,
+			int *out, int outpos,
+			int firstn, int recurse_to_leaf,
+			int *out2)
+{
+	int rep;
+	int ftotal, flocal;
+	int retry_descent, retry_bucket, skip_rep;
+	struct crush_bucket *in = bucket;
+	int r;
+	int i;
+	int item = 0;
+	int itemtype;
+	int collide, reject;
+	const int orig_tries = 5; /* attempts before we fall back to search */
+
+	dprintk("CHOOSE%s bucket %d x %d outpos %d numrep %d\n", recurse_to_leaf ? "_LEAF" : "",
+		bucket->id, x, outpos, numrep);
+
+	for (rep = outpos; rep < numrep; rep++) {
+		/* keep trying until we get a non-out, non-colliding item */
+		ftotal = 0;
+		skip_rep = 0;
+		do {
+			retry_descent = 0;
+			in = bucket;               /* initial bucket */
+
+			/* choose through intervening buckets */
+			flocal = 0;
+			do {
+				collide = 0;
+				retry_bucket = 0;
+				r = rep;
+				if (in->alg == CRUSH_BUCKET_UNIFORM) {
+					/* be careful */
+					if (firstn || numrep >= in->size)
+						/* r' = r + f_total */
+						r += ftotal;
+					else if (in->size % numrep == 0)
+						/* r'=r+(n+1)*f_local */
+						r += (numrep+1) *
+							(flocal+ftotal);
+					else
+						/* r' = r + n*f_local */
+						r += numrep * (flocal+ftotal);
+				} else {
+					if (firstn)
+						/* r' = r + f_total */
+						r += ftotal;
+					else
+						/* r' = r + n*f_local */
+						r += numrep * (flocal+ftotal);
+				}
+
+				/* bucket choose */
+				if (in->size == 0) {
+					reject = 1;
+					goto reject;
+				}
+				if (flocal >= (in->size>>1) &&
+				    flocal > orig_tries)
+					item = bucket_perm_choose(in, x, r);
+				else
+					item = crush_bucket_choose(in, x, r);
+				BUG_ON(item >= map->max_devices);
+
+				/* desired type? */
+				if (item < 0)
+					itemtype = map->buckets[-1-item]->type;
+				else
+					itemtype = 0;
+				dprintk("  item %d type %d\n", item, itemtype);
+
+				/* keep going? */
+				if (itemtype != type) {
+					BUG_ON(item >= 0 ||
+					       (-1-item) >= map->max_buckets);
+					in = map->buckets[-1-item];
+					retry_bucket = 1;
+					continue;
+				}
+
+				/* collision? */
+				for (i = 0; i < outpos; i++) {
+					if (out[i] == item) {
+						collide = 1;
+						break;
+					}
+				}
+
+				reject = 0;
+				if (recurse_to_leaf) {
+					if (item < 0) {
+						if (crush_choose(map,
+							 map->buckets[-1-item],
+							 weight,
+							 x, outpos+1, 0,
+							 out2, outpos,
+							 firstn, 0,
+							 NULL) <= outpos)
+							/* didn't get leaf */
+							reject = 1;
+					} else {
+						/* we already have a leaf! */
+						out2[outpos] = item;
+					}
+				}
+
+				if (!reject) {
+					/* out? */
+					if (itemtype == 0)
+						reject = is_out(map, weight,
+								item, x);
+					else
+						reject = 0;
+				}
+
+reject:
+				if (reject || collide) {
+					ftotal++;
+					flocal++;
+
+					if (collide && flocal < 3)
+						/* retry locally a few times */
+						retry_bucket = 1;
+					else if (flocal < in->size + orig_tries)
+						/* exhaustive bucket search */
+						retry_bucket = 1;
+					else if (ftotal < 20)
+						/* then retry descent */
+						retry_descent = 1;
+					else
+						/* else give up */
+						skip_rep = 1;
+					dprintk("  reject %d  collide %d  "
+						"ftotal %d  flocal %d\n",
+						reject, collide, ftotal,
+						flocal);
+				}
+			} while (retry_bucket);
+		} while (retry_descent);
+
+		if (skip_rep) {
+			dprintk("skip rep\n");
+			continue;
+		}
+
+		dprintk("CHOOSE got %d\n", item);
+		out[outpos] = item;
+		outpos++;
+	}
+
+	dprintk("CHOOSE returns %d\n", outpos);
+	return outpos;
+}
+
+
+/**
+ * crush_do_rule - calculate a mapping with the given input and rule
+ * @map: the crush_map
+ * @ruleno: the rule id
+ * @x: hash input
+ * @result: pointer to result vector
+ * @result_max: maximum result size
+ * @force: force initial replica choice; -1 for none
+ */
+int crush_do_rule(struct crush_map *map,
+		  int ruleno, int x, int *result, int result_max,
+		  int force, __u32 *weight)
+{
+	int result_len;
+	int force_context[CRUSH_MAX_DEPTH];
+	int force_pos = -1;
+	int a[CRUSH_MAX_SET];
+	int b[CRUSH_MAX_SET];
+	int c[CRUSH_MAX_SET];
+	int recurse_to_leaf;
+	int *w;
+	int wsize = 0;
+	int *o;
+	int osize;
+	int *tmp;
+	struct crush_rule *rule;
+	int step;
+	int i, j;
+	int numrep;
+	int firstn;
+	int rc = -1;
+
+	BUG_ON(ruleno >= map->max_rules);
+
+	rule = map->rules[ruleno];
+	result_len = 0;
+	w = a;
+	o = b;
+
+	/*
+	 * determine hierarchical context of force, if any.  note
+	 * that this may or may not correspond to the specific types
+	 * referenced by the crush rule.
+	 */
+	if (force >= 0) {
+		if (force >= map->max_devices ||
+		    map->device_parents[force] == 0) {
+			/*dprintk("CRUSH: forcefed device dne\n");*/
+			rc = -1;  /* force fed device dne */
+			goto out;
+		}
+		if (!is_out(map, weight, force, x)) {
+			while (1) {
+				force_context[++force_pos] = force;
+				if (force >= 0)
+					force = map->device_parents[force];
+				else
+					force = map->bucket_parents[-1-force];
+				if (force == 0)
+					break;
+			}
+		}
+	}
+
+	for (step = 0; step < rule->len; step++) {
+		firstn = 0;
+		switch (rule->steps[step].op) {
+		case CRUSH_RULE_TAKE:
+			w[0] = rule->steps[step].arg1;
+			if (force_pos >= 0) {
+				BUG_ON(force_context[force_pos] != w[0]);
+				force_pos--;
+			}
+			wsize = 1;
+			break;
+
+		case CRUSH_RULE_CHOOSE_LEAF_FIRSTN:
+		case CRUSH_RULE_CHOOSE_FIRSTN:
+			firstn = 1;
+		case CRUSH_RULE_CHOOSE_LEAF_INDEP:
+		case CRUSH_RULE_CHOOSE_INDEP:
+			BUG_ON(wsize == 0);
+
+			recurse_to_leaf =
+				rule->steps[step].op ==
+				 CRUSH_RULE_CHOOSE_LEAF_FIRSTN ||
+				rule->steps[step].op ==
+				CRUSH_RULE_CHOOSE_LEAF_INDEP;
+
+			/* reset output */
+			osize = 0;
+
+			for (i = 0; i < wsize; i++) {
+				/*
+				 * see CRUSH_N, CRUSH_N_MINUS macros.
+				 * basically, numrep <= 0 means relative to
+				 * the provided result_max
+				 */
+				numrep = rule->steps[step].arg1;
+				if (numrep <= 0) {
+					numrep += result_max;
+					if (numrep <= 0)
+						continue;
+				}
+				j = 0;
+				if (osize == 0 && force_pos >= 0) {
+					/* skip any intermediate types */
+					while (force_pos &&
+					       force_context[force_pos] < 0 &&
+					       rule->steps[step].arg2 !=
+					       map->buckets[-1 -
+					       force_context[force_pos]]->type)
+						force_pos--;
+					o[osize] = force_context[force_pos];
+					if (recurse_to_leaf)
+						c[osize] = force_context[0];
+					j++;
+					force_pos--;
+				}
+				osize += crush_choose(map,
+						      map->buckets[-1-w[i]],
+						      weight,
+						      x, numrep,
+						      rule->steps[step].arg2,
+						      o+osize, j,
+						      firstn,
+						      recurse_to_leaf, c+osize);
+			}
+
+			if (recurse_to_leaf)
+				/* copy final _leaf_ values to output set */
+				memcpy(o, c, osize*sizeof(*o));
+
+			/* swap t and w arrays */
+			tmp = o;
+			o = w;
+			w = tmp;
+			wsize = osize;
+			break;
+
+
+		case CRUSH_RULE_EMIT:
+			for (i = 0; i < wsize && result_len < result_max; i++) {
+				result[result_len] = w[i];
+				result_len++;
+			}
+			wsize = 0;
+			break;
+
+		default:
+			BUG_ON(1);
+		}
+	}
+	rc = result_len;
+
+out:
+	return rc;
+}
+
+
diff --git a/net/ceph/crypto.c b/net/ceph/crypto.c
new file mode 100644
index 000000000000..7b505b0c983f
--- /dev/null
+++ b/net/ceph/crypto.c
@@ -0,0 +1,412 @@
+
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/err.h>
+#include <linux/scatterlist.h>
+#include <linux/slab.h>
+#include <crypto/hash.h>
+
+#include <linux/ceph/decode.h>
+#include "crypto.h"
+
+int ceph_crypto_key_encode(struct ceph_crypto_key *key, void **p, void *end)
+{
+	if (*p + sizeof(u16) + sizeof(key->created) +
+	    sizeof(u16) + key->len > end)
+		return -ERANGE;
+	ceph_encode_16(p, key->type);
+	ceph_encode_copy(p, &key->created, sizeof(key->created));
+	ceph_encode_16(p, key->len);
+	ceph_encode_copy(p, key->key, key->len);
+	return 0;
+}
+
+int ceph_crypto_key_decode(struct ceph_crypto_key *key, void **p, void *end)
+{
+	ceph_decode_need(p, end, 2*sizeof(u16) + sizeof(key->created), bad);
+	key->type = ceph_decode_16(p);
+	ceph_decode_copy(p, &key->created, sizeof(key->created));
+	key->len = ceph_decode_16(p);
+	ceph_decode_need(p, end, key->len, bad);
+	key->key = kmalloc(key->len, GFP_NOFS);
+	if (!key->key)
+		return -ENOMEM;
+	ceph_decode_copy(p, key->key, key->len);
+	return 0;
+
+bad:
+	dout("failed to decode crypto key\n");
+	return -EINVAL;
+}
+
+int ceph_crypto_key_unarmor(struct ceph_crypto_key *key, const char *inkey)
+{
+	int inlen = strlen(inkey);
+	int blen = inlen * 3 / 4;
+	void *buf, *p;
+	int ret;
+
+	dout("crypto_key_unarmor %s\n", inkey);
+	buf = kmalloc(blen, GFP_NOFS);
+	if (!buf)
+		return -ENOMEM;
+	blen = ceph_unarmor(buf, inkey, inkey+inlen);
+	if (blen < 0) {
+		kfree(buf);
+		return blen;
+	}
+
+	p = buf;
+	ret = ceph_crypto_key_decode(key, &p, p + blen);
+	kfree(buf);
+	if (ret)
+		return ret;
+	dout("crypto_key_unarmor key %p type %d len %d\n", key,
+	     key->type, key->len);
+	return 0;
+}
+
+
+
+#define AES_KEY_SIZE 16
+
+static struct crypto_blkcipher *ceph_crypto_alloc_cipher(void)
+{
+	return crypto_alloc_blkcipher("cbc(aes)", 0, CRYPTO_ALG_ASYNC);
+}
+
+static const u8 *aes_iv = (u8 *)CEPH_AES_IV;
+
+static int ceph_aes_encrypt(const void *key, int key_len,
+			    void *dst, size_t *dst_len,
+			    const void *src, size_t src_len)
+{
+	struct scatterlist sg_in[2], sg_out[1];
+	struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher();
+	struct blkcipher_desc desc = { .tfm = tfm, .flags = 0 };
+	int ret;
+	void *iv;
+	int ivsize;
+	size_t zero_padding = (0x10 - (src_len & 0x0f));
+	char pad[16];
+
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
+
+	memset(pad, zero_padding, zero_padding);
+
+	*dst_len = src_len + zero_padding;
+
+	crypto_blkcipher_setkey((void *)tfm, key, key_len);
+	sg_init_table(sg_in, 2);
+	sg_set_buf(&sg_in[0], src, src_len);
+	sg_set_buf(&sg_in[1], pad, zero_padding);
+	sg_init_table(sg_out, 1);
+	sg_set_buf(sg_out, dst, *dst_len);
+	iv = crypto_blkcipher_crt(tfm)->iv;
+	ivsize = crypto_blkcipher_ivsize(tfm);
+
+	memcpy(iv, aes_iv, ivsize);
+	/*
+	print_hex_dump(KERN_ERR, "enc key: ", DUMP_PREFIX_NONE, 16, 1,
+		       key, key_len, 1);
+	print_hex_dump(KERN_ERR, "enc src: ", DUMP_PREFIX_NONE, 16, 1,
+			src, src_len, 1);
+	print_hex_dump(KERN_ERR, "enc pad: ", DUMP_PREFIX_NONE, 16, 1,
+			pad, zero_padding, 1);
+	*/
+	ret = crypto_blkcipher_encrypt(&desc, sg_out, sg_in,
+				     src_len + zero_padding);
+	crypto_free_blkcipher(tfm);
+	if (ret < 0)
+		pr_err("ceph_aes_crypt failed %d\n", ret);
+	/*
+	print_hex_dump(KERN_ERR, "enc out: ", DUMP_PREFIX_NONE, 16, 1,
+		       dst, *dst_len, 1);
+	*/
+	return 0;
+}
+
+static int ceph_aes_encrypt2(const void *key, int key_len, void *dst,
+			     size_t *dst_len,
+			     const void *src1, size_t src1_len,
+			     const void *src2, size_t src2_len)
+{
+	struct scatterlist sg_in[3], sg_out[1];
+	struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher();
+	struct blkcipher_desc desc = { .tfm = tfm, .flags = 0 };
+	int ret;
+	void *iv;
+	int ivsize;
+	size_t zero_padding = (0x10 - ((src1_len + src2_len) & 0x0f));
+	char pad[16];
+
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
+
+	memset(pad, zero_padding, zero_padding);
+
+	*dst_len = src1_len + src2_len + zero_padding;
+
+	crypto_blkcipher_setkey((void *)tfm, key, key_len);
+	sg_init_table(sg_in, 3);
+	sg_set_buf(&sg_in[0], src1, src1_len);
+	sg_set_buf(&sg_in[1], src2, src2_len);
+	sg_set_buf(&sg_in[2], pad, zero_padding);
+	sg_init_table(sg_out, 1);
+	sg_set_buf(sg_out, dst, *dst_len);
+	iv = crypto_blkcipher_crt(tfm)->iv;
+	ivsize = crypto_blkcipher_ivsize(tfm);
+
+	memcpy(iv, aes_iv, ivsize);
+	/*
+	print_hex_dump(KERN_ERR, "enc  key: ", DUMP_PREFIX_NONE, 16, 1,
+		       key, key_len, 1);
+	print_hex_dump(KERN_ERR, "enc src1: ", DUMP_PREFIX_NONE, 16, 1,
+			src1, src1_len, 1);
+	print_hex_dump(KERN_ERR, "enc src2: ", DUMP_PREFIX_NONE, 16, 1,
+			src2, src2_len, 1);
+	print_hex_dump(KERN_ERR, "enc  pad: ", DUMP_PREFIX_NONE, 16, 1,
+			pad, zero_padding, 1);
+	*/
+	ret = crypto_blkcipher_encrypt(&desc, sg_out, sg_in,
+				     src1_len + src2_len + zero_padding);
+	crypto_free_blkcipher(tfm);
+	if (ret < 0)
+		pr_err("ceph_aes_crypt2 failed %d\n", ret);
+	/*
+	print_hex_dump(KERN_ERR, "enc  out: ", DUMP_PREFIX_NONE, 16, 1,
+		       dst, *dst_len, 1);
+	*/
+	return 0;
+}
+
+static int ceph_aes_decrypt(const void *key, int key_len,
+			    void *dst, size_t *dst_len,
+			    const void *src, size_t src_len)
+{
+	struct scatterlist sg_in[1], sg_out[2];
+	struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher();
+	struct blkcipher_desc desc = { .tfm = tfm };
+	char pad[16];
+	void *iv;
+	int ivsize;
+	int ret;
+	int last_byte;
+
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
+
+	crypto_blkcipher_setkey((void *)tfm, key, key_len);
+	sg_init_table(sg_in, 1);
+	sg_init_table(sg_out, 2);
+	sg_set_buf(sg_in, src, src_len);
+	sg_set_buf(&sg_out[0], dst, *dst_len);
+	sg_set_buf(&sg_out[1], pad, sizeof(pad));
+
+	iv = crypto_blkcipher_crt(tfm)->iv;
+	ivsize = crypto_blkcipher_ivsize(tfm);
+
+	memcpy(iv, aes_iv, ivsize);
+
+	/*
+	print_hex_dump(KERN_ERR, "dec key: ", DUMP_PREFIX_NONE, 16, 1,
+		       key, key_len, 1);
+	print_hex_dump(KERN_ERR, "dec  in: ", DUMP_PREFIX_NONE, 16, 1,
+		       src, src_len, 1);
+	*/
+
+	ret = crypto_blkcipher_decrypt(&desc, sg_out, sg_in, src_len);
+	crypto_free_blkcipher(tfm);
+	if (ret < 0) {
+		pr_err("ceph_aes_decrypt failed %d\n", ret);
+		return ret;
+	}
+
+	if (src_len <= *dst_len)
+		last_byte = ((char *)dst)[src_len - 1];
+	else
+		last_byte = pad[src_len - *dst_len - 1];
+	if (last_byte <= 16 && src_len >= last_byte) {
+		*dst_len = src_len - last_byte;
+	} else {
+		pr_err("ceph_aes_decrypt got bad padding %d on src len %d\n",
+		       last_byte, (int)src_len);
+		return -EPERM;  /* bad padding */
+	}
+	/*
+	print_hex_dump(KERN_ERR, "dec out: ", DUMP_PREFIX_NONE, 16, 1,
+		       dst, *dst_len, 1);
+	*/
+	return 0;
+}
+
+static int ceph_aes_decrypt2(const void *key, int key_len,
+			     void *dst1, size_t *dst1_len,
+			     void *dst2, size_t *dst2_len,
+			     const void *src, size_t src_len)
+{
+	struct scatterlist sg_in[1], sg_out[3];
+	struct crypto_blkcipher *tfm = ceph_crypto_alloc_cipher();
+	struct blkcipher_desc desc = { .tfm = tfm };
+	char pad[16];
+	void *iv;
+	int ivsize;
+	int ret;
+	int last_byte;
+
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
+
+	sg_init_table(sg_in, 1);
+	sg_set_buf(sg_in, src, src_len);
+	sg_init_table(sg_out, 3);
+	sg_set_buf(&sg_out[0], dst1, *dst1_len);
+	sg_set_buf(&sg_out[1], dst2, *dst2_len);
+	sg_set_buf(&sg_out[2], pad, sizeof(pad));
+
+	crypto_blkcipher_setkey((void *)tfm, key, key_len);
+	iv = crypto_blkcipher_crt(tfm)->iv;
+	ivsize = crypto_blkcipher_ivsize(tfm);
+
+	memcpy(iv, aes_iv, ivsize);
+
+	/*
+	print_hex_dump(KERN_ERR, "dec  key: ", DUMP_PREFIX_NONE, 16, 1,
+		       key, key_len, 1);
+	print_hex_dump(KERN_ERR, "dec   in: ", DUMP_PREFIX_NONE, 16, 1,
+		       src, src_len, 1);
+	*/
+
+	ret = crypto_blkcipher_decrypt(&desc, sg_out, sg_in, src_len);
+	crypto_free_blkcipher(tfm);
+	if (ret < 0) {
+		pr_err("ceph_aes_decrypt failed %d\n", ret);
+		return ret;
+	}
+
+	if (src_len <= *dst1_len)
+		last_byte = ((char *)dst1)[src_len - 1];
+	else if (src_len <= *dst1_len + *dst2_len)
+		last_byte = ((char *)dst2)[src_len - *dst1_len - 1];
+	else
+		last_byte = pad[src_len - *dst1_len - *dst2_len - 1];
+	if (last_byte <= 16 && src_len >= last_byte) {
+		src_len -= last_byte;
+	} else {
+		pr_err("ceph_aes_decrypt got bad padding %d on src len %d\n",
+		       last_byte, (int)src_len);
+		return -EPERM;  /* bad padding */
+	}
+
+	if (src_len < *dst1_len) {
+		*dst1_len = src_len;
+		*dst2_len = 0;
+	} else {
+		*dst2_len = src_len - *dst1_len;
+	}
+	/*
+	print_hex_dump(KERN_ERR, "dec  out1: ", DUMP_PREFIX_NONE, 16, 1,
+		       dst1, *dst1_len, 1);
+	print_hex_dump(KERN_ERR, "dec  out2: ", DUMP_PREFIX_NONE, 16, 1,
+		       dst2, *dst2_len, 1);
+	*/
+
+	return 0;
+}
+
+
+int ceph_decrypt(struct ceph_crypto_key *secret, void *dst, size_t *dst_len,
+		 const void *src, size_t src_len)
+{
+	switch (secret->type) {
+	case CEPH_CRYPTO_NONE:
+		if (*dst_len < src_len)
+			return -ERANGE;
+		memcpy(dst, src, src_len);
+		*dst_len = src_len;
+		return 0;
+
+	case CEPH_CRYPTO_AES:
+		return ceph_aes_decrypt(secret->key, secret->len, dst,
+					dst_len, src, src_len);
+
+	default:
+		return -EINVAL;
+	}
+}
+
+int ceph_decrypt2(struct ceph_crypto_key *secret,
+			void *dst1, size_t *dst1_len,
+			void *dst2, size_t *dst2_len,
+			const void *src, size_t src_len)
+{
+	size_t t;
+
+	switch (secret->type) {
+	case CEPH_CRYPTO_NONE:
+		if (*dst1_len + *dst2_len < src_len)
+			return -ERANGE;
+		t = min(*dst1_len, src_len);
+		memcpy(dst1, src, t);
+		*dst1_len = t;
+		src += t;
+		src_len -= t;
+		if (src_len) {
+			t = min(*dst2_len, src_len);
+			memcpy(dst2, src, t);
+			*dst2_len = t;
+		}
+		return 0;
+
+	case CEPH_CRYPTO_AES:
+		return ceph_aes_decrypt2(secret->key, secret->len,
+					 dst1, dst1_len, dst2, dst2_len,
+					 src, src_len);
+
+	default:
+		return -EINVAL;
+	}
+}
+
+int ceph_encrypt(struct ceph_crypto_key *secret, void *dst, size_t *dst_len,
+		 const void *src, size_t src_len)
+{
+	switch (secret->type) {
+	case CEPH_CRYPTO_NONE:
+		if (*dst_len < src_len)
+			return -ERANGE;
+		memcpy(dst, src, src_len);
+		*dst_len = src_len;
+		return 0;
+
+	case CEPH_CRYPTO_AES:
+		return ceph_aes_encrypt(secret->key, secret->len, dst,
+					dst_len, src, src_len);
+
+	default:
+		return -EINVAL;
+	}
+}
+
+int ceph_encrypt2(struct ceph_crypto_key *secret, void *dst, size_t *dst_len,
+		  const void *src1, size_t src1_len,
+		  const void *src2, size_t src2_len)
+{
+	switch (secret->type) {
+	case CEPH_CRYPTO_NONE:
+		if (*dst_len < src1_len + src2_len)
+			return -ERANGE;
+		memcpy(dst, src1, src1_len);
+		memcpy(dst + src1_len, src2, src2_len);
+		*dst_len = src1_len + src2_len;
+		return 0;
+
+	case CEPH_CRYPTO_AES:
+		return ceph_aes_encrypt2(secret->key, secret->len, dst, dst_len,
+					 src1, src1_len, src2, src2_len);
+
+	default:
+		return -EINVAL;
+	}
+}
diff --git a/net/ceph/crypto.h b/net/ceph/crypto.h
new file mode 100644
index 000000000000..f9eccace592b
--- /dev/null
+++ b/net/ceph/crypto.h
@@ -0,0 +1,48 @@
+#ifndef _FS_CEPH_CRYPTO_H
+#define _FS_CEPH_CRYPTO_H
+
+#include <linux/ceph/types.h>
+#include <linux/ceph/buffer.h>
+
+/*
+ * cryptographic secret
+ */
+struct ceph_crypto_key {
+	int type;
+	struct ceph_timespec created;
+	int len;
+	void *key;
+};
+
+static inline void ceph_crypto_key_destroy(struct ceph_crypto_key *key)
+{
+	kfree(key->key);
+}
+
+extern int ceph_crypto_key_encode(struct ceph_crypto_key *key,
+				  void **p, void *end);
+extern int ceph_crypto_key_decode(struct ceph_crypto_key *key,
+				  void **p, void *end);
+extern int ceph_crypto_key_unarmor(struct ceph_crypto_key *key, const char *in);
+
+/* crypto.c */
+extern int ceph_decrypt(struct ceph_crypto_key *secret,
+			void *dst, size_t *dst_len,
+			const void *src, size_t src_len);
+extern int ceph_encrypt(struct ceph_crypto_key *secret,
+			void *dst, size_t *dst_len,
+			const void *src, size_t src_len);
+extern int ceph_decrypt2(struct ceph_crypto_key *secret,
+			void *dst1, size_t *dst1_len,
+			void *dst2, size_t *dst2_len,
+			const void *src, size_t src_len);
+extern int ceph_encrypt2(struct ceph_crypto_key *secret,
+			 void *dst, size_t *dst_len,
+			 const void *src1, size_t src1_len,
+			 const void *src2, size_t src2_len);
+
+/* armor.c */
+extern int ceph_armor(char *dst, const char *src, const char *end);
+extern int ceph_unarmor(char *dst, const char *src, const char *end);
+
+#endif
diff --git a/net/ceph/debugfs.c b/net/ceph/debugfs.c
new file mode 100644
index 000000000000..33d04999f4f2
--- /dev/null
+++ b/net/ceph/debugfs.c
@@ -0,0 +1,268 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/ctype.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
+
+#ifdef CONFIG_DEBUG_FS
+
+/*
+ * Implement /sys/kernel/debug/ceph fun
+ *
+ * /sys/kernel/debug/ceph/client*  - an instance of the ceph client
+ *      .../osdmap      - current osdmap
+ *      .../monmap      - current monmap
+ *      .../osdc        - active osd requests
+ *      .../monc        - mon client state
+ *      .../dentry_lru  - dump contents of dentry lru
+ *      .../caps        - expose cap (reservation) stats
+ *      .../bdi         - symlink to ../../bdi/something
+ */
+
+static struct dentry *ceph_debugfs_dir;
+
+static int monmap_show(struct seq_file *s, void *p)
+{
+	int i;
+	struct ceph_client *client = s->private;
+
+	if (client->monc.monmap == NULL)
+		return 0;
+
+	seq_printf(s, "epoch %d\n", client->monc.monmap->epoch);
+	for (i = 0; i < client->monc.monmap->num_mon; i++) {
+		struct ceph_entity_inst *inst =
+			&client->monc.monmap->mon_inst[i];
+
+		seq_printf(s, "\t%s%lld\t%s\n",
+			   ENTITY_NAME(inst->name),
+			   ceph_pr_addr(&inst->addr.in_addr));
+	}
+	return 0;
+}
+
+static int osdmap_show(struct seq_file *s, void *p)
+{
+	int i;
+	struct ceph_client *client = s->private;
+	struct rb_node *n;
+
+	if (client->osdc.osdmap == NULL)
+		return 0;
+	seq_printf(s, "epoch %d\n", client->osdc.osdmap->epoch);
+	seq_printf(s, "flags%s%s\n",
+		   (client->osdc.osdmap->flags & CEPH_OSDMAP_NEARFULL) ?
+		   " NEARFULL" : "",
+		   (client->osdc.osdmap->flags & CEPH_OSDMAP_FULL) ?
+		   " FULL" : "");
+	for (n = rb_first(&client->osdc.osdmap->pg_pools); n; n = rb_next(n)) {
+		struct ceph_pg_pool_info *pool =
+			rb_entry(n, struct ceph_pg_pool_info, node);
+		seq_printf(s, "pg_pool %d pg_num %d / %d, lpg_num %d / %d\n",
+			   pool->id, pool->v.pg_num, pool->pg_num_mask,
+			   pool->v.lpg_num, pool->lpg_num_mask);
+	}
+	for (i = 0; i < client->osdc.osdmap->max_osd; i++) {
+		struct ceph_entity_addr *addr =
+			&client->osdc.osdmap->osd_addr[i];
+		int state = client->osdc.osdmap->osd_state[i];
+		char sb[64];
+
+		seq_printf(s, "\tosd%d\t%s\t%3d%%\t(%s)\n",
+			   i, ceph_pr_addr(&addr->in_addr),
+			   ((client->osdc.osdmap->osd_weight[i]*100) >> 16),
+			   ceph_osdmap_state_str(sb, sizeof(sb), state));
+	}
+	return 0;
+}
+
+static int monc_show(struct seq_file *s, void *p)
+{
+	struct ceph_client *client = s->private;
+	struct ceph_mon_generic_request *req;
+	struct ceph_mon_client *monc = &client->monc;
+	struct rb_node *rp;
+
+	mutex_lock(&monc->mutex);
+
+	if (monc->have_mdsmap)
+		seq_printf(s, "have mdsmap %u\n", (unsigned)monc->have_mdsmap);
+	if (monc->have_osdmap)
+		seq_printf(s, "have osdmap %u\n", (unsigned)monc->have_osdmap);
+	if (monc->want_next_osdmap)
+		seq_printf(s, "want next osdmap\n");
+
+	for (rp = rb_first(&monc->generic_request_tree); rp; rp = rb_next(rp)) {
+		__u16 op;
+		req = rb_entry(rp, struct ceph_mon_generic_request, node);
+		op = le16_to_cpu(req->request->hdr.type);
+		if (op == CEPH_MSG_STATFS)
+			seq_printf(s, "%lld statfs\n", req->tid);
+		else
+			seq_printf(s, "%lld unknown\n", req->tid);
+	}
+
+	mutex_unlock(&monc->mutex);
+	return 0;
+}
+
+static int osdc_show(struct seq_file *s, void *pp)
+{
+	struct ceph_client *client = s->private;
+	struct ceph_osd_client *osdc = &client->osdc;
+	struct rb_node *p;
+
+	mutex_lock(&osdc->request_mutex);
+	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
+		struct ceph_osd_request *req;
+		struct ceph_osd_request_head *head;
+		struct ceph_osd_op *op;
+		int num_ops;
+		int opcode, olen;
+		int i;
+
+		req = rb_entry(p, struct ceph_osd_request, r_node);
+
+		seq_printf(s, "%lld\tosd%d\t%d.%x\t", req->r_tid,
+			   req->r_osd ? req->r_osd->o_osd : -1,
+			   le32_to_cpu(req->r_pgid.pool),
+			   le16_to_cpu(req->r_pgid.ps));
+
+		head = req->r_request->front.iov_base;
+		op = (void *)(head + 1);
+
+		num_ops = le16_to_cpu(head->num_ops);
+		olen = le32_to_cpu(head->object_len);
+		seq_printf(s, "%.*s", olen,
+			   (const char *)(head->ops + num_ops));
+
+		if (req->r_reassert_version.epoch)
+			seq_printf(s, "\t%u'%llu",
+			   (unsigned)le32_to_cpu(req->r_reassert_version.epoch),
+			   le64_to_cpu(req->r_reassert_version.version));
+		else
+			seq_printf(s, "\t");
+
+		for (i = 0; i < num_ops; i++) {
+			opcode = le16_to_cpu(op->op);
+			seq_printf(s, "\t%s", ceph_osd_op_name(opcode));
+			op++;
+		}
+
+		seq_printf(s, "\n");
+	}
+	mutex_unlock(&osdc->request_mutex);
+	return 0;
+}
+
+CEPH_DEFINE_SHOW_FUNC(monmap_show)
+CEPH_DEFINE_SHOW_FUNC(osdmap_show)
+CEPH_DEFINE_SHOW_FUNC(monc_show)
+CEPH_DEFINE_SHOW_FUNC(osdc_show)
+
+int ceph_debugfs_init(void)
+{
+	ceph_debugfs_dir = debugfs_create_dir("ceph", NULL);
+	if (!ceph_debugfs_dir)
+		return -ENOMEM;
+	return 0;
+}
+
+void ceph_debugfs_cleanup(void)
+{
+	debugfs_remove(ceph_debugfs_dir);
+}
+
+int ceph_debugfs_client_init(struct ceph_client *client)
+{
+	int ret = -ENOMEM;
+	char name[80];
+
+	snprintf(name, sizeof(name), "%pU.client%lld", &client->fsid,
+		 client->monc.auth->global_id);
+
+	client->debugfs_dir = debugfs_create_dir(name, ceph_debugfs_dir);
+	if (!client->debugfs_dir)
+		goto out;
+
+	client->monc.debugfs_file = debugfs_create_file("monc",
+						      0600,
+						      client->debugfs_dir,
+						      client,
+						      &monc_show_fops);
+	if (!client->monc.debugfs_file)
+		goto out;
+
+	client->osdc.debugfs_file = debugfs_create_file("osdc",
+						      0600,
+						      client->debugfs_dir,
+						      client,
+						      &osdc_show_fops);
+	if (!client->osdc.debugfs_file)
+		goto out;
+
+	client->debugfs_monmap = debugfs_create_file("monmap",
+					0600,
+					client->debugfs_dir,
+					client,
+					&monmap_show_fops);
+	if (!client->debugfs_monmap)
+		goto out;
+
+	client->debugfs_osdmap = debugfs_create_file("osdmap",
+					0600,
+					client->debugfs_dir,
+					client,
+					&osdmap_show_fops);
+	if (!client->debugfs_osdmap)
+		goto out;
+
+	return 0;
+
+out:
+	ceph_debugfs_client_cleanup(client);
+	return ret;
+}
+
+void ceph_debugfs_client_cleanup(struct ceph_client *client)
+{
+	debugfs_remove(client->debugfs_osdmap);
+	debugfs_remove(client->debugfs_monmap);
+	debugfs_remove(client->osdc.debugfs_file);
+	debugfs_remove(client->monc.debugfs_file);
+	debugfs_remove(client->debugfs_dir);
+}
+
+#else  /* CONFIG_DEBUG_FS */
+
+int ceph_debugfs_init(void)
+{
+	return 0;
+}
+
+void ceph_debugfs_cleanup(void)
+{
+}
+
+int ceph_debugfs_client_init(struct ceph_client *client,
+			     int (*module_debugfs_init)(struct ceph_client *))
+{
+	return 0;
+}
+
+void ceph_debugfs_client_cleanup(struct ceph_client *client)
+{
+}
+
+#endif  /* CONFIG_DEBUG_FS */
+
+EXPORT_SYMBOL(ceph_debugfs_init);
+EXPORT_SYMBOL(ceph_debugfs_cleanup);
diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c
new file mode 100644
index 000000000000..0e8157ee5d43
--- /dev/null
+++ b/net/ceph/messenger.c
@@ -0,0 +1,2453 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/crc32c.h>
+#include <linux/ctype.h>
+#include <linux/highmem.h>
+#include <linux/inet.h>
+#include <linux/kthread.h>
+#include <linux/net.h>
+#include <linux/slab.h>
+#include <linux/socket.h>
+#include <linux/string.h>
+#include <linux/bio.h>
+#include <linux/blkdev.h>
+#include <net/tcp.h>
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/pagelist.h>
+
+/*
+ * Ceph uses the messenger to exchange ceph_msg messages with other
+ * hosts in the system.  The messenger provides ordered and reliable
+ * delivery.  We tolerate TCP disconnects by reconnecting (with
+ * exponential backoff) in the case of a fault (disconnection, bad
+ * crc, protocol error).  Acks allow sent messages to be discarded by
+ * the sender.
+ */
+
+/* static tag bytes (protocol control messages) */
+static char tag_msg = CEPH_MSGR_TAG_MSG;
+static char tag_ack = CEPH_MSGR_TAG_ACK;
+static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE;
+
+#ifdef CONFIG_LOCKDEP
+static struct lock_class_key socket_class;
+#endif
+
+
+static void queue_con(struct ceph_connection *con);
+static void con_work(struct work_struct *);
+static void ceph_fault(struct ceph_connection *con);
+
+/*
+ * nicely render a sockaddr as a string.
+ */
+#define MAX_ADDR_STR 20
+#define MAX_ADDR_STR_LEN 60
+static char addr_str[MAX_ADDR_STR][MAX_ADDR_STR_LEN];
+static DEFINE_SPINLOCK(addr_str_lock);
+static int last_addr_str;
+
+const char *ceph_pr_addr(const struct sockaddr_storage *ss)
+{
+	int i;
+	char *s;
+	struct sockaddr_in *in4 = (void *)ss;
+	struct sockaddr_in6 *in6 = (void *)ss;
+
+	spin_lock(&addr_str_lock);
+	i = last_addr_str++;
+	if (last_addr_str == MAX_ADDR_STR)
+		last_addr_str = 0;
+	spin_unlock(&addr_str_lock);
+	s = addr_str[i];
+
+	switch (ss->ss_family) {
+	case AF_INET:
+		snprintf(s, MAX_ADDR_STR_LEN, "%pI4:%u", &in4->sin_addr,
+			 (unsigned int)ntohs(in4->sin_port));
+		break;
+
+	case AF_INET6:
+		snprintf(s, MAX_ADDR_STR_LEN, "[%pI6c]:%u", &in6->sin6_addr,
+			 (unsigned int)ntohs(in6->sin6_port));
+		break;
+
+	default:
+		sprintf(s, "(unknown sockaddr family %d)", (int)ss->ss_family);
+	}
+
+	return s;
+}
+EXPORT_SYMBOL(ceph_pr_addr);
+
+static void encode_my_addr(struct ceph_messenger *msgr)
+{
+	memcpy(&msgr->my_enc_addr, &msgr->inst.addr, sizeof(msgr->my_enc_addr));
+	ceph_encode_addr(&msgr->my_enc_addr);
+}
+
+/*
+ * work queue for all reading and writing to/from the socket.
+ */
+struct workqueue_struct *ceph_msgr_wq;
+
+int ceph_msgr_init(void)
+{
+	ceph_msgr_wq = create_workqueue("ceph-msgr");
+	if (IS_ERR(ceph_msgr_wq)) {
+		int ret = PTR_ERR(ceph_msgr_wq);
+		pr_err("msgr_init failed to create workqueue: %d\n", ret);
+		ceph_msgr_wq = NULL;
+		return ret;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_msgr_init);
+
+void ceph_msgr_exit(void)
+{
+	destroy_workqueue(ceph_msgr_wq);
+}
+EXPORT_SYMBOL(ceph_msgr_exit);
+
+void ceph_msgr_flush(void)
+{
+	flush_workqueue(ceph_msgr_wq);
+}
+EXPORT_SYMBOL(ceph_msgr_flush);
+
+
+/*
+ * socket callback functions
+ */
+
+/* data available on socket, or listen socket received a connect */
+static void ceph_data_ready(struct sock *sk, int count_unused)
+{
+	struct ceph_connection *con =
+		(struct ceph_connection *)sk->sk_user_data;
+	if (sk->sk_state != TCP_CLOSE_WAIT) {
+		dout("ceph_data_ready on %p state = %lu, queueing work\n",
+		     con, con->state);
+		queue_con(con);
+	}
+}
+
+/* socket has buffer space for writing */
+static void ceph_write_space(struct sock *sk)
+{
+	struct ceph_connection *con =
+		(struct ceph_connection *)sk->sk_user_data;
+
+	/* only queue to workqueue if there is data we want to write. */
+	if (test_bit(WRITE_PENDING, &con->state)) {
+		dout("ceph_write_space %p queueing write work\n", con);
+		queue_con(con);
+	} else {
+		dout("ceph_write_space %p nothing to write\n", con);
+	}
+
+	/* since we have our own write_space, clear the SOCK_NOSPACE flag */
+	clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+}
+
+/* socket's state has changed */
+static void ceph_state_change(struct sock *sk)
+{
+	struct ceph_connection *con =
+		(struct ceph_connection *)sk->sk_user_data;
+
+	dout("ceph_state_change %p state = %lu sk_state = %u\n",
+	     con, con->state, sk->sk_state);
+
+	if (test_bit(CLOSED, &con->state))
+		return;
+
+	switch (sk->sk_state) {
+	case TCP_CLOSE:
+		dout("ceph_state_change TCP_CLOSE\n");
+	case TCP_CLOSE_WAIT:
+		dout("ceph_state_change TCP_CLOSE_WAIT\n");
+		if (test_and_set_bit(SOCK_CLOSED, &con->state) == 0) {
+			if (test_bit(CONNECTING, &con->state))
+				con->error_msg = "connection failed";
+			else
+				con->error_msg = "socket closed";
+			queue_con(con);
+		}
+		break;
+	case TCP_ESTABLISHED:
+		dout("ceph_state_change TCP_ESTABLISHED\n");
+		queue_con(con);
+		break;
+	}
+}
+
+/*
+ * set up socket callbacks
+ */
+static void set_sock_callbacks(struct socket *sock,
+			       struct ceph_connection *con)
+{
+	struct sock *sk = sock->sk;
+	sk->sk_user_data = (void *)con;
+	sk->sk_data_ready = ceph_data_ready;
+	sk->sk_write_space = ceph_write_space;
+	sk->sk_state_change = ceph_state_change;
+}
+
+
+/*
+ * socket helpers
+ */
+
+/*
+ * initiate connection to a remote socket.
+ */
+static struct socket *ceph_tcp_connect(struct ceph_connection *con)
+{
+	struct sockaddr_storage *paddr = &con->peer_addr.in_addr;
+	struct socket *sock;
+	int ret;
+
+	BUG_ON(con->sock);
+	ret = sock_create_kern(con->peer_addr.in_addr.ss_family, SOCK_STREAM,
+			       IPPROTO_TCP, &sock);
+	if (ret)
+		return ERR_PTR(ret);
+	con->sock = sock;
+	sock->sk->sk_allocation = GFP_NOFS;
+
+#ifdef CONFIG_LOCKDEP
+	lockdep_set_class(&sock->sk->sk_lock, &socket_class);
+#endif
+
+	set_sock_callbacks(sock, con);
+
+	dout("connect %s\n", ceph_pr_addr(&con->peer_addr.in_addr));
+
+	ret = sock->ops->connect(sock, (struct sockaddr *)paddr, sizeof(*paddr),
+				 O_NONBLOCK);
+	if (ret == -EINPROGRESS) {
+		dout("connect %s EINPROGRESS sk_state = %u\n",
+		     ceph_pr_addr(&con->peer_addr.in_addr),
+		     sock->sk->sk_state);
+		ret = 0;
+	}
+	if (ret < 0) {
+		pr_err("connect %s error %d\n",
+		       ceph_pr_addr(&con->peer_addr.in_addr), ret);
+		sock_release(sock);
+		con->sock = NULL;
+		con->error_msg = "connect error";
+	}
+
+	if (ret < 0)
+		return ERR_PTR(ret);
+	return sock;
+}
+
+static int ceph_tcp_recvmsg(struct socket *sock, void *buf, size_t len)
+{
+	struct kvec iov = {buf, len};
+	struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL };
+
+	return kernel_recvmsg(sock, &msg, &iov, 1, len, msg.msg_flags);
+}
+
+/*
+ * write something.  @more is true if caller will be sending more data
+ * shortly.
+ */
+static int ceph_tcp_sendmsg(struct socket *sock, struct kvec *iov,
+		     size_t kvlen, size_t len, int more)
+{
+	struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL };
+
+	if (more)
+		msg.msg_flags |= MSG_MORE;
+	else
+		msg.msg_flags |= MSG_EOR;  /* superfluous, but what the hell */
+
+	return kernel_sendmsg(sock, &msg, iov, kvlen, len);
+}
+
+
+/*
+ * Shutdown/close the socket for the given connection.
+ */
+static int con_close_socket(struct ceph_connection *con)
+{
+	int rc;
+
+	dout("con_close_socket on %p sock %p\n", con, con->sock);
+	if (!con->sock)
+		return 0;
+	set_bit(SOCK_CLOSED, &con->state);
+	rc = con->sock->ops->shutdown(con->sock, SHUT_RDWR);
+	sock_release(con->sock);
+	con->sock = NULL;
+	clear_bit(SOCK_CLOSED, &con->state);
+	return rc;
+}
+
+/*
+ * Reset a connection.  Discard all incoming and outgoing messages
+ * and clear *_seq state.
+ */
+static void ceph_msg_remove(struct ceph_msg *msg)
+{
+	list_del_init(&msg->list_head);
+	ceph_msg_put(msg);
+}
+static void ceph_msg_remove_list(struct list_head *head)
+{
+	while (!list_empty(head)) {
+		struct ceph_msg *msg = list_first_entry(head, struct ceph_msg,
+							list_head);
+		ceph_msg_remove(msg);
+	}
+}
+
+static void reset_connection(struct ceph_connection *con)
+{
+	/* reset connection, out_queue, msg_ and connect_seq */
+	/* discard existing out_queue and msg_seq */
+	ceph_msg_remove_list(&con->out_queue);
+	ceph_msg_remove_list(&con->out_sent);
+
+	if (con->in_msg) {
+		ceph_msg_put(con->in_msg);
+		con->in_msg = NULL;
+	}
+
+	con->connect_seq = 0;
+	con->out_seq = 0;
+	if (con->out_msg) {
+		ceph_msg_put(con->out_msg);
+		con->out_msg = NULL;
+	}
+	con->out_keepalive_pending = false;
+	con->in_seq = 0;
+	con->in_seq_acked = 0;
+}
+
+/*
+ * mark a peer down.  drop any open connections.
+ */
+void ceph_con_close(struct ceph_connection *con)
+{
+	dout("con_close %p peer %s\n", con,
+	     ceph_pr_addr(&con->peer_addr.in_addr));
+	set_bit(CLOSED, &con->state);  /* in case there's queued work */
+	clear_bit(STANDBY, &con->state);  /* avoid connect_seq bump */
+	clear_bit(LOSSYTX, &con->state);  /* so we retry next connect */
+	clear_bit(KEEPALIVE_PENDING, &con->state);
+	clear_bit(WRITE_PENDING, &con->state);
+	mutex_lock(&con->mutex);
+	reset_connection(con);
+	con->peer_global_seq = 0;
+	cancel_delayed_work(&con->work);
+	mutex_unlock(&con->mutex);
+	queue_con(con);
+}
+EXPORT_SYMBOL(ceph_con_close);
+
+/*
+ * Reopen a closed connection, with a new peer address.
+ */
+void ceph_con_open(struct ceph_connection *con, struct ceph_entity_addr *addr)
+{
+	dout("con_open %p %s\n", con, ceph_pr_addr(&addr->in_addr));
+	set_bit(OPENING, &con->state);
+	clear_bit(CLOSED, &con->state);
+	memcpy(&con->peer_addr, addr, sizeof(*addr));
+	con->delay = 0;      /* reset backoff memory */
+	queue_con(con);
+}
+EXPORT_SYMBOL(ceph_con_open);
+
+/*
+ * return true if this connection ever successfully opened
+ */
+bool ceph_con_opened(struct ceph_connection *con)
+{
+	return con->connect_seq > 0;
+}
+
+/*
+ * generic get/put
+ */
+struct ceph_connection *ceph_con_get(struct ceph_connection *con)
+{
+	dout("con_get %p nref = %d -> %d\n", con,
+	     atomic_read(&con->nref), atomic_read(&con->nref) + 1);
+	if (atomic_inc_not_zero(&con->nref))
+		return con;
+	return NULL;
+}
+
+void ceph_con_put(struct ceph_connection *con)
+{
+	dout("con_put %p nref = %d -> %d\n", con,
+	     atomic_read(&con->nref), atomic_read(&con->nref) - 1);
+	BUG_ON(atomic_read(&con->nref) == 0);
+	if (atomic_dec_and_test(&con->nref)) {
+		BUG_ON(con->sock);
+		kfree(con);
+	}
+}
+
+/*
+ * initialize a new connection.
+ */
+void ceph_con_init(struct ceph_messenger *msgr, struct ceph_connection *con)
+{
+	dout("con_init %p\n", con);
+	memset(con, 0, sizeof(*con));
+	atomic_set(&con->nref, 1);
+	con->msgr = msgr;
+	mutex_init(&con->mutex);
+	INIT_LIST_HEAD(&con->out_queue);
+	INIT_LIST_HEAD(&con->out_sent);
+	INIT_DELAYED_WORK(&con->work, con_work);
+}
+EXPORT_SYMBOL(ceph_con_init);
+
+
+/*
+ * We maintain a global counter to order connection attempts.  Get
+ * a unique seq greater than @gt.
+ */
+static u32 get_global_seq(struct ceph_messenger *msgr, u32 gt)
+{
+	u32 ret;
+
+	spin_lock(&msgr->global_seq_lock);
+	if (msgr->global_seq < gt)
+		msgr->global_seq = gt;
+	ret = ++msgr->global_seq;
+	spin_unlock(&msgr->global_seq_lock);
+	return ret;
+}
+
+
+/*
+ * Prepare footer for currently outgoing message, and finish things
+ * off.  Assumes out_kvec* are already valid.. we just add on to the end.
+ */
+static void prepare_write_message_footer(struct ceph_connection *con, int v)
+{
+	struct ceph_msg *m = con->out_msg;
+
+	dout("prepare_write_message_footer %p\n", con);
+	con->out_kvec_is_msg = true;
+	con->out_kvec[v].iov_base = &m->footer;
+	con->out_kvec[v].iov_len = sizeof(m->footer);
+	con->out_kvec_bytes += sizeof(m->footer);
+	con->out_kvec_left++;
+	con->out_more = m->more_to_follow;
+	con->out_msg_done = true;
+}
+
+/*
+ * Prepare headers for the next outgoing message.
+ */
+static void prepare_write_message(struct ceph_connection *con)
+{
+	struct ceph_msg *m;
+	int v = 0;
+
+	con->out_kvec_bytes = 0;
+	con->out_kvec_is_msg = true;
+	con->out_msg_done = false;
+
+	/* Sneak an ack in there first?  If we can get it into the same
+	 * TCP packet that's a good thing. */
+	if (con->in_seq > con->in_seq_acked) {
+		con->in_seq_acked = con->in_seq;
+		con->out_kvec[v].iov_base = &tag_ack;
+		con->out_kvec[v++].iov_len = 1;
+		con->out_temp_ack = cpu_to_le64(con->in_seq_acked);
+		con->out_kvec[v].iov_base = &con->out_temp_ack;
+		con->out_kvec[v++].iov_len = sizeof(con->out_temp_ack);
+		con->out_kvec_bytes = 1 + sizeof(con->out_temp_ack);
+	}
+
+	m = list_first_entry(&con->out_queue,
+		       struct ceph_msg, list_head);
+	con->out_msg = m;
+	if (test_bit(LOSSYTX, &con->state)) {
+		list_del_init(&m->list_head);
+	} else {
+		/* put message on sent list */
+		ceph_msg_get(m);
+		list_move_tail(&m->list_head, &con->out_sent);
+	}
+
+	/*
+	 * only assign outgoing seq # if we haven't sent this message
+	 * yet.  if it is requeued, resend with it's original seq.
+	 */
+	if (m->needs_out_seq) {
+		m->hdr.seq = cpu_to_le64(++con->out_seq);
+		m->needs_out_seq = false;
+	}
+
+	dout("prepare_write_message %p seq %lld type %d len %d+%d+%d %d pgs\n",
+	     m, con->out_seq, le16_to_cpu(m->hdr.type),
+	     le32_to_cpu(m->hdr.front_len), le32_to_cpu(m->hdr.middle_len),
+	     le32_to_cpu(m->hdr.data_len),
+	     m->nr_pages);
+	BUG_ON(le32_to_cpu(m->hdr.front_len) != m->front.iov_len);
+
+	/* tag + hdr + front + middle */
+	con->out_kvec[v].iov_base = &tag_msg;
+	con->out_kvec[v++].iov_len = 1;
+	con->out_kvec[v].iov_base = &m->hdr;
+	con->out_kvec[v++].iov_len = sizeof(m->hdr);
+	con->out_kvec[v++] = m->front;
+	if (m->middle)
+		con->out_kvec[v++] = m->middle->vec;
+	con->out_kvec_left = v;
+	con->out_kvec_bytes += 1 + sizeof(m->hdr) + m->front.iov_len +
+		(m->middle ? m->middle->vec.iov_len : 0);
+	con->out_kvec_cur = con->out_kvec;
+
+	/* fill in crc (except data pages), footer */
+	con->out_msg->hdr.crc =
+		cpu_to_le32(crc32c(0, (void *)&m->hdr,
+				      sizeof(m->hdr) - sizeof(m->hdr.crc)));
+	con->out_msg->footer.flags = CEPH_MSG_FOOTER_COMPLETE;
+	con->out_msg->footer.front_crc =
+		cpu_to_le32(crc32c(0, m->front.iov_base, m->front.iov_len));
+	if (m->middle)
+		con->out_msg->footer.middle_crc =
+			cpu_to_le32(crc32c(0, m->middle->vec.iov_base,
+					   m->middle->vec.iov_len));
+	else
+		con->out_msg->footer.middle_crc = 0;
+	con->out_msg->footer.data_crc = 0;
+	dout("prepare_write_message front_crc %u data_crc %u\n",
+	     le32_to_cpu(con->out_msg->footer.front_crc),
+	     le32_to_cpu(con->out_msg->footer.middle_crc));
+
+	/* is there a data payload? */
+	if (le32_to_cpu(m->hdr.data_len) > 0) {
+		/* initialize page iterator */
+		con->out_msg_pos.page = 0;
+		if (m->pages)
+			con->out_msg_pos.page_pos =
+				le16_to_cpu(m->hdr.data_off) & ~PAGE_MASK;
+		else
+			con->out_msg_pos.page_pos = 0;
+		con->out_msg_pos.data_pos = 0;
+		con->out_msg_pos.did_page_crc = 0;
+		con->out_more = 1;  /* data + footer will follow */
+	} else {
+		/* no, queue up footer too and be done */
+		prepare_write_message_footer(con, v);
+	}
+
+	set_bit(WRITE_PENDING, &con->state);
+}
+
+/*
+ * Prepare an ack.
+ */
+static void prepare_write_ack(struct ceph_connection *con)
+{
+	dout("prepare_write_ack %p %llu -> %llu\n", con,
+	     con->in_seq_acked, con->in_seq);
+	con->in_seq_acked = con->in_seq;
+
+	con->out_kvec[0].iov_base = &tag_ack;
+	con->out_kvec[0].iov_len = 1;
+	con->out_temp_ack = cpu_to_le64(con->in_seq_acked);
+	con->out_kvec[1].iov_base = &con->out_temp_ack;
+	con->out_kvec[1].iov_len = sizeof(con->out_temp_ack);
+	con->out_kvec_left = 2;
+	con->out_kvec_bytes = 1 + sizeof(con->out_temp_ack);
+	con->out_kvec_cur = con->out_kvec;
+	con->out_more = 1;  /* more will follow.. eventually.. */
+	set_bit(WRITE_PENDING, &con->state);
+}
+
+/*
+ * Prepare to write keepalive byte.
+ */
+static void prepare_write_keepalive(struct ceph_connection *con)
+{
+	dout("prepare_write_keepalive %p\n", con);
+	con->out_kvec[0].iov_base = &tag_keepalive;
+	con->out_kvec[0].iov_len = 1;
+	con->out_kvec_left = 1;
+	con->out_kvec_bytes = 1;
+	con->out_kvec_cur = con->out_kvec;
+	set_bit(WRITE_PENDING, &con->state);
+}
+
+/*
+ * Connection negotiation.
+ */
+
+static void prepare_connect_authorizer(struct ceph_connection *con)
+{
+	void *auth_buf;
+	int auth_len = 0;
+	int auth_protocol = 0;
+
+	mutex_unlock(&con->mutex);
+	if (con->ops->get_authorizer)
+		con->ops->get_authorizer(con, &auth_buf, &auth_len,
+					 &auth_protocol, &con->auth_reply_buf,
+					 &con->auth_reply_buf_len,
+					 con->auth_retry);
+	mutex_lock(&con->mutex);
+
+	con->out_connect.authorizer_protocol = cpu_to_le32(auth_protocol);
+	con->out_connect.authorizer_len = cpu_to_le32(auth_len);
+
+	con->out_kvec[con->out_kvec_left].iov_base = auth_buf;
+	con->out_kvec[con->out_kvec_left].iov_len = auth_len;
+	con->out_kvec_left++;
+	con->out_kvec_bytes += auth_len;
+}
+
+/*
+ * We connected to a peer and are saying hello.
+ */
+static void prepare_write_banner(struct ceph_messenger *msgr,
+				 struct ceph_connection *con)
+{
+	int len = strlen(CEPH_BANNER);
+
+	con->out_kvec[0].iov_base = CEPH_BANNER;
+	con->out_kvec[0].iov_len = len;
+	con->out_kvec[1].iov_base = &msgr->my_enc_addr;
+	con->out_kvec[1].iov_len = sizeof(msgr->my_enc_addr);
+	con->out_kvec_left = 2;
+	con->out_kvec_bytes = len + sizeof(msgr->my_enc_addr);
+	con->out_kvec_cur = con->out_kvec;
+	con->out_more = 0;
+	set_bit(WRITE_PENDING, &con->state);
+}
+
+static void prepare_write_connect(struct ceph_messenger *msgr,
+				  struct ceph_connection *con,
+				  int after_banner)
+{
+	unsigned global_seq = get_global_seq(con->msgr, 0);
+	int proto;
+
+	switch (con->peer_name.type) {
+	case CEPH_ENTITY_TYPE_MON:
+		proto = CEPH_MONC_PROTOCOL;
+		break;
+	case CEPH_ENTITY_TYPE_OSD:
+		proto = CEPH_OSDC_PROTOCOL;
+		break;
+	case CEPH_ENTITY_TYPE_MDS:
+		proto = CEPH_MDSC_PROTOCOL;
+		break;
+	default:
+		BUG();
+	}
+
+	dout("prepare_write_connect %p cseq=%d gseq=%d proto=%d\n", con,
+	     con->connect_seq, global_seq, proto);
+
+	con->out_connect.features = cpu_to_le64(msgr->supported_features);
+	con->out_connect.host_type = cpu_to_le32(CEPH_ENTITY_TYPE_CLIENT);
+	con->out_connect.connect_seq = cpu_to_le32(con->connect_seq);
+	con->out_connect.global_seq = cpu_to_le32(global_seq);
+	con->out_connect.protocol_version = cpu_to_le32(proto);
+	con->out_connect.flags = 0;
+
+	if (!after_banner) {
+		con->out_kvec_left = 0;
+		con->out_kvec_bytes = 0;
+	}
+	con->out_kvec[con->out_kvec_left].iov_base = &con->out_connect;
+	con->out_kvec[con->out_kvec_left].iov_len = sizeof(con->out_connect);
+	con->out_kvec_left++;
+	con->out_kvec_bytes += sizeof(con->out_connect);
+	con->out_kvec_cur = con->out_kvec;
+	con->out_more = 0;
+	set_bit(WRITE_PENDING, &con->state);
+
+	prepare_connect_authorizer(con);
+}
+
+
+/*
+ * write as much of pending kvecs to the socket as we can.
+ *  1 -> done
+ *  0 -> socket full, but more to do
+ * <0 -> error
+ */
+static int write_partial_kvec(struct ceph_connection *con)
+{
+	int ret;
+
+	dout("write_partial_kvec %p %d left\n", con, con->out_kvec_bytes);
+	while (con->out_kvec_bytes > 0) {
+		ret = ceph_tcp_sendmsg(con->sock, con->out_kvec_cur,
+				       con->out_kvec_left, con->out_kvec_bytes,
+				       con->out_more);
+		if (ret <= 0)
+			goto out;
+		con->out_kvec_bytes -= ret;
+		if (con->out_kvec_bytes == 0)
+			break;            /* done */
+		while (ret > 0) {
+			if (ret >= con->out_kvec_cur->iov_len) {
+				ret -= con->out_kvec_cur->iov_len;
+				con->out_kvec_cur++;
+				con->out_kvec_left--;
+			} else {
+				con->out_kvec_cur->iov_len -= ret;
+				con->out_kvec_cur->iov_base += ret;
+				ret = 0;
+				break;
+			}
+		}
+	}
+	con->out_kvec_left = 0;
+	con->out_kvec_is_msg = false;
+	ret = 1;
+out:
+	dout("write_partial_kvec %p %d left in %d kvecs ret = %d\n", con,
+	     con->out_kvec_bytes, con->out_kvec_left, ret);
+	return ret;  /* done! */
+}
+
+#ifdef CONFIG_BLOCK
+static void init_bio_iter(struct bio *bio, struct bio **iter, int *seg)
+{
+	if (!bio) {
+		*iter = NULL;
+		*seg = 0;
+		return;
+	}
+	*iter = bio;
+	*seg = bio->bi_idx;
+}
+
+static void iter_bio_next(struct bio **bio_iter, int *seg)
+{
+	if (*bio_iter == NULL)
+		return;
+
+	BUG_ON(*seg >= (*bio_iter)->bi_vcnt);
+
+	(*seg)++;
+	if (*seg == (*bio_iter)->bi_vcnt)
+		init_bio_iter((*bio_iter)->bi_next, bio_iter, seg);
+}
+#endif
+
+/*
+ * Write as much message data payload as we can.  If we finish, queue
+ * up the footer.
+ *  1 -> done, footer is now queued in out_kvec[].
+ *  0 -> socket full, but more to do
+ * <0 -> error
+ */
+static int write_partial_msg_pages(struct ceph_connection *con)
+{
+	struct ceph_msg *msg = con->out_msg;
+	unsigned data_len = le32_to_cpu(msg->hdr.data_len);
+	size_t len;
+	int crc = con->msgr->nocrc;
+	int ret;
+	int total_max_write;
+	int in_trail = 0;
+	size_t trail_len = (msg->trail ? msg->trail->length : 0);
+
+	dout("write_partial_msg_pages %p msg %p page %d/%d offset %d\n",
+	     con, con->out_msg, con->out_msg_pos.page, con->out_msg->nr_pages,
+	     con->out_msg_pos.page_pos);
+
+#ifdef CONFIG_BLOCK
+	if (msg->bio && !msg->bio_iter)
+		init_bio_iter(msg->bio, &msg->bio_iter, &msg->bio_seg);
+#endif
+
+	while (data_len > con->out_msg_pos.data_pos) {
+		struct page *page = NULL;
+		void *kaddr = NULL;
+		int max_write = PAGE_SIZE;
+		int page_shift = 0;
+
+		total_max_write = data_len - trail_len -
+			con->out_msg_pos.data_pos;
+
+		/*
+		 * if we are calculating the data crc (the default), we need
+		 * to map the page.  if our pages[] has been revoked, use the
+		 * zero page.
+		 */
+
+		/* have we reached the trail part of the data? */
+		if (con->out_msg_pos.data_pos >= data_len - trail_len) {
+			in_trail = 1;
+
+			total_max_write = data_len - con->out_msg_pos.data_pos;
+
+			page = list_first_entry(&msg->trail->head,
+						struct page, lru);
+			if (crc)
+				kaddr = kmap(page);
+			max_write = PAGE_SIZE;
+		} else if (msg->pages) {
+			page = msg->pages[con->out_msg_pos.page];
+			if (crc)
+				kaddr = kmap(page);
+		} else if (msg->pagelist) {
+			page = list_first_entry(&msg->pagelist->head,
+						struct page, lru);
+			if (crc)
+				kaddr = kmap(page);
+#ifdef CONFIG_BLOCK
+		} else if (msg->bio) {
+			struct bio_vec *bv;
+
+			bv = bio_iovec_idx(msg->bio_iter, msg->bio_seg);
+			page = bv->bv_page;
+			page_shift = bv->bv_offset;
+			if (crc)
+				kaddr = kmap(page) + page_shift;
+			max_write = bv->bv_len;
+#endif
+		} else {
+			page = con->msgr->zero_page;
+			if (crc)
+				kaddr = page_address(con->msgr->zero_page);
+		}
+		len = min_t(int, max_write - con->out_msg_pos.page_pos,
+			    total_max_write);
+
+		if (crc && !con->out_msg_pos.did_page_crc) {
+			void *base = kaddr + con->out_msg_pos.page_pos;
+			u32 tmpcrc = le32_to_cpu(con->out_msg->footer.data_crc);
+
+			BUG_ON(kaddr == NULL);
+			con->out_msg->footer.data_crc =
+				cpu_to_le32(crc32c(tmpcrc, base, len));
+			con->out_msg_pos.did_page_crc = 1;
+		}
+		ret = kernel_sendpage(con->sock, page,
+				      con->out_msg_pos.page_pos + page_shift,
+				      len,
+				      MSG_DONTWAIT | MSG_NOSIGNAL |
+				      MSG_MORE);
+
+		if (crc &&
+		    (msg->pages || msg->pagelist || msg->bio || in_trail))
+			kunmap(page);
+
+		if (ret <= 0)
+			goto out;
+
+		con->out_msg_pos.data_pos += ret;
+		con->out_msg_pos.page_pos += ret;
+		if (ret == len) {
+			con->out_msg_pos.page_pos = 0;
+			con->out_msg_pos.page++;
+			con->out_msg_pos.did_page_crc = 0;
+			if (in_trail)
+				list_move_tail(&page->lru,
+					       &msg->trail->head);
+			else if (msg->pagelist)
+				list_move_tail(&page->lru,
+					       &msg->pagelist->head);
+#ifdef CONFIG_BLOCK
+			else if (msg->bio)
+				iter_bio_next(&msg->bio_iter, &msg->bio_seg);
+#endif
+		}
+	}
+
+	dout("write_partial_msg_pages %p msg %p done\n", con, msg);
+
+	/* prepare and queue up footer, too */
+	if (!crc)
+		con->out_msg->footer.flags |= CEPH_MSG_FOOTER_NOCRC;
+	con->out_kvec_bytes = 0;
+	con->out_kvec_left = 0;
+	con->out_kvec_cur = con->out_kvec;
+	prepare_write_message_footer(con, 0);
+	ret = 1;
+out:
+	return ret;
+}
+
+/*
+ * write some zeros
+ */
+static int write_partial_skip(struct ceph_connection *con)
+{
+	int ret;
+
+	while (con->out_skip > 0) {
+		struct kvec iov = {
+			.iov_base = page_address(con->msgr->zero_page),
+			.iov_len = min(con->out_skip, (int)PAGE_CACHE_SIZE)
+		};
+
+		ret = ceph_tcp_sendmsg(con->sock, &iov, 1, iov.iov_len, 1);
+		if (ret <= 0)
+			goto out;
+		con->out_skip -= ret;
+	}
+	ret = 1;
+out:
+	return ret;
+}
+
+/*
+ * Prepare to read connection handshake, or an ack.
+ */
+static void prepare_read_banner(struct ceph_connection *con)
+{
+	dout("prepare_read_banner %p\n", con);
+	con->in_base_pos = 0;
+}
+
+static void prepare_read_connect(struct ceph_connection *con)
+{
+	dout("prepare_read_connect %p\n", con);
+	con->in_base_pos = 0;
+}
+
+static void prepare_read_ack(struct ceph_connection *con)
+{
+	dout("prepare_read_ack %p\n", con);
+	con->in_base_pos = 0;
+}
+
+static void prepare_read_tag(struct ceph_connection *con)
+{
+	dout("prepare_read_tag %p\n", con);
+	con->in_base_pos = 0;
+	con->in_tag = CEPH_MSGR_TAG_READY;
+}
+
+/*
+ * Prepare to read a message.
+ */
+static int prepare_read_message(struct ceph_connection *con)
+{
+	dout("prepare_read_message %p\n", con);
+	BUG_ON(con->in_msg != NULL);
+	con->in_base_pos = 0;
+	con->in_front_crc = con->in_middle_crc = con->in_data_crc = 0;
+	return 0;
+}
+
+
+static int read_partial(struct ceph_connection *con,
+			int *to, int size, void *object)
+{
+	*to += size;
+	while (con->in_base_pos < *to) {
+		int left = *to - con->in_base_pos;
+		int have = size - left;
+		int ret = ceph_tcp_recvmsg(con->sock, object + have, left);
+		if (ret <= 0)
+			return ret;
+		con->in_base_pos += ret;
+	}
+	return 1;
+}
+
+
+/*
+ * Read all or part of the connect-side handshake on a new connection
+ */
+static int read_partial_banner(struct ceph_connection *con)
+{
+	int ret, to = 0;
+
+	dout("read_partial_banner %p at %d\n", con, con->in_base_pos);
+
+	/* peer's banner */
+	ret = read_partial(con, &to, strlen(CEPH_BANNER), con->in_banner);
+	if (ret <= 0)
+		goto out;
+	ret = read_partial(con, &to, sizeof(con->actual_peer_addr),
+			   &con->actual_peer_addr);
+	if (ret <= 0)
+		goto out;
+	ret = read_partial(con, &to, sizeof(con->peer_addr_for_me),
+			   &con->peer_addr_for_me);
+	if (ret <= 0)
+		goto out;
+out:
+	return ret;
+}
+
+static int read_partial_connect(struct ceph_connection *con)
+{
+	int ret, to = 0;
+
+	dout("read_partial_connect %p at %d\n", con, con->in_base_pos);
+
+	ret = read_partial(con, &to, sizeof(con->in_reply), &con->in_reply);
+	if (ret <= 0)
+		goto out;
+	ret = read_partial(con, &to, le32_to_cpu(con->in_reply.authorizer_len),
+			   con->auth_reply_buf);
+	if (ret <= 0)
+		goto out;
+
+	dout("read_partial_connect %p tag %d, con_seq = %u, g_seq = %u\n",
+	     con, (int)con->in_reply.tag,
+	     le32_to_cpu(con->in_reply.connect_seq),
+	     le32_to_cpu(con->in_reply.global_seq));
+out:
+	return ret;
+
+}
+
+/*
+ * Verify the hello banner looks okay.
+ */
+static int verify_hello(struct ceph_connection *con)
+{
+	if (memcmp(con->in_banner, CEPH_BANNER, strlen(CEPH_BANNER))) {
+		pr_err("connect to %s got bad banner\n",
+		       ceph_pr_addr(&con->peer_addr.in_addr));
+		con->error_msg = "protocol error, bad banner";
+		return -1;
+	}
+	return 0;
+}
+
+static bool addr_is_blank(struct sockaddr_storage *ss)
+{
+	switch (ss->ss_family) {
+	case AF_INET:
+		return ((struct sockaddr_in *)ss)->sin_addr.s_addr == 0;
+	case AF_INET6:
+		return
+		     ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[0] == 0 &&
+		     ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[1] == 0 &&
+		     ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[2] == 0 &&
+		     ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[3] == 0;
+	}
+	return false;
+}
+
+static int addr_port(struct sockaddr_storage *ss)
+{
+	switch (ss->ss_family) {
+	case AF_INET:
+		return ntohs(((struct sockaddr_in *)ss)->sin_port);
+	case AF_INET6:
+		return ntohs(((struct sockaddr_in6 *)ss)->sin6_port);
+	}
+	return 0;
+}
+
+static void addr_set_port(struct sockaddr_storage *ss, int p)
+{
+	switch (ss->ss_family) {
+	case AF_INET:
+		((struct sockaddr_in *)ss)->sin_port = htons(p);
+	case AF_INET6:
+		((struct sockaddr_in6 *)ss)->sin6_port = htons(p);
+	}
+}
+
+/*
+ * Parse an ip[:port] list into an addr array.  Use the default
+ * monitor port if a port isn't specified.
+ */
+int ceph_parse_ips(const char *c, const char *end,
+		   struct ceph_entity_addr *addr,
+		   int max_count, int *count)
+{
+	int i;
+	const char *p = c;
+
+	dout("parse_ips on '%.*s'\n", (int)(end-c), c);
+	for (i = 0; i < max_count; i++) {
+		const char *ipend;
+		struct sockaddr_storage *ss = &addr[i].in_addr;
+		struct sockaddr_in *in4 = (void *)ss;
+		struct sockaddr_in6 *in6 = (void *)ss;
+		int port;
+		char delim = ',';
+
+		if (*p == '[') {
+			delim = ']';
+			p++;
+		}
+
+		memset(ss, 0, sizeof(*ss));
+		if (in4_pton(p, end - p, (u8 *)&in4->sin_addr.s_addr,
+			     delim, &ipend))
+			ss->ss_family = AF_INET;
+		else if (in6_pton(p, end - p, (u8 *)&in6->sin6_addr.s6_addr,
+				  delim, &ipend))
+			ss->ss_family = AF_INET6;
+		else
+			goto bad;
+		p = ipend;
+
+		if (delim == ']') {
+			if (*p != ']') {
+				dout("missing matching ']'\n");
+				goto bad;
+			}
+			p++;
+		}
+
+		/* port? */
+		if (p < end && *p == ':') {
+			port = 0;
+			p++;
+			while (p < end && *p >= '0' && *p <= '9') {
+				port = (port * 10) + (*p - '0');
+				p++;
+			}
+			if (port > 65535 || port == 0)
+				goto bad;
+		} else {
+			port = CEPH_MON_PORT;
+		}
+
+		addr_set_port(ss, port);
+
+		dout("parse_ips got %s\n", ceph_pr_addr(ss));
+
+		if (p == end)
+			break;
+		if (*p != ',')
+			goto bad;
+		p++;
+	}
+
+	if (p != end)
+		goto bad;
+
+	if (count)
+		*count = i + 1;
+	return 0;
+
+bad:
+	pr_err("parse_ips bad ip '%.*s'\n", (int)(end - c), c);
+	return -EINVAL;
+}
+EXPORT_SYMBOL(ceph_parse_ips);
+
+static int process_banner(struct ceph_connection *con)
+{
+	dout("process_banner on %p\n", con);
+
+	if (verify_hello(con) < 0)
+		return -1;
+
+	ceph_decode_addr(&con->actual_peer_addr);
+	ceph_decode_addr(&con->peer_addr_for_me);
+
+	/*
+	 * Make sure the other end is who we wanted.  note that the other
+	 * end may not yet know their ip address, so if it's 0.0.0.0, give
+	 * them the benefit of the doubt.
+	 */
+	if (memcmp(&con->peer_addr, &con->actual_peer_addr,
+		   sizeof(con->peer_addr)) != 0 &&
+	    !(addr_is_blank(&con->actual_peer_addr.in_addr) &&
+	      con->actual_peer_addr.nonce == con->peer_addr.nonce)) {
+		pr_warning("wrong peer, want %s/%d, got %s/%d\n",
+			   ceph_pr_addr(&con->peer_addr.in_addr),
+			   (int)le32_to_cpu(con->peer_addr.nonce),
+			   ceph_pr_addr(&con->actual_peer_addr.in_addr),
+			   (int)le32_to_cpu(con->actual_peer_addr.nonce));
+		con->error_msg = "wrong peer at address";
+		return -1;
+	}
+
+	/*
+	 * did we learn our address?
+	 */
+	if (addr_is_blank(&con->msgr->inst.addr.in_addr)) {
+		int port = addr_port(&con->msgr->inst.addr.in_addr);
+
+		memcpy(&con->msgr->inst.addr.in_addr,
+		       &con->peer_addr_for_me.in_addr,
+		       sizeof(con->peer_addr_for_me.in_addr));
+		addr_set_port(&con->msgr->inst.addr.in_addr, port);
+		encode_my_addr(con->msgr);
+		dout("process_banner learned my addr is %s\n",
+		     ceph_pr_addr(&con->msgr->inst.addr.in_addr));
+	}
+
+	set_bit(NEGOTIATING, &con->state);
+	prepare_read_connect(con);
+	return 0;
+}
+
+static void fail_protocol(struct ceph_connection *con)
+{
+	reset_connection(con);
+	set_bit(CLOSED, &con->state);  /* in case there's queued work */
+
+	mutex_unlock(&con->mutex);
+	if (con->ops->bad_proto)
+		con->ops->bad_proto(con);
+	mutex_lock(&con->mutex);
+}
+
+static int process_connect(struct ceph_connection *con)
+{
+	u64 sup_feat = con->msgr->supported_features;
+	u64 req_feat = con->msgr->required_features;
+	u64 server_feat = le64_to_cpu(con->in_reply.features);
+
+	dout("process_connect on %p tag %d\n", con, (int)con->in_tag);
+
+	switch (con->in_reply.tag) {
+	case CEPH_MSGR_TAG_FEATURES:
+		pr_err("%s%lld %s feature set mismatch,"
+		       " my %llx < server's %llx, missing %llx\n",
+		       ENTITY_NAME(con->peer_name),
+		       ceph_pr_addr(&con->peer_addr.in_addr),
+		       sup_feat, server_feat, server_feat & ~sup_feat);
+		con->error_msg = "missing required protocol features";
+		fail_protocol(con);
+		return -1;
+
+	case CEPH_MSGR_TAG_BADPROTOVER:
+		pr_err("%s%lld %s protocol version mismatch,"
+		       " my %d != server's %d\n",
+		       ENTITY_NAME(con->peer_name),
+		       ceph_pr_addr(&con->peer_addr.in_addr),
+		       le32_to_cpu(con->out_connect.protocol_version),
+		       le32_to_cpu(con->in_reply.protocol_version));
+		con->error_msg = "protocol version mismatch";
+		fail_protocol(con);
+		return -1;
+
+	case CEPH_MSGR_TAG_BADAUTHORIZER:
+		con->auth_retry++;
+		dout("process_connect %p got BADAUTHORIZER attempt %d\n", con,
+		     con->auth_retry);
+		if (con->auth_retry == 2) {
+			con->error_msg = "connect authorization failure";
+			reset_connection(con);
+			set_bit(CLOSED, &con->state);
+			return -1;
+		}
+		con->auth_retry = 1;
+		prepare_write_connect(con->msgr, con, 0);
+		prepare_read_connect(con);
+		break;
+
+	case CEPH_MSGR_TAG_RESETSESSION:
+		/*
+		 * If we connected with a large connect_seq but the peer
+		 * has no record of a session with us (no connection, or
+		 * connect_seq == 0), they will send RESETSESION to indicate
+		 * that they must have reset their session, and may have
+		 * dropped messages.
+		 */
+		dout("process_connect got RESET peer seq %u\n",
+		     le32_to_cpu(con->in_connect.connect_seq));
+		pr_err("%s%lld %s connection reset\n",
+		       ENTITY_NAME(con->peer_name),
+		       ceph_pr_addr(&con->peer_addr.in_addr));
+		reset_connection(con);
+		prepare_write_connect(con->msgr, con, 0);
+		prepare_read_connect(con);
+
+		/* Tell ceph about it. */
+		mutex_unlock(&con->mutex);
+		pr_info("reset on %s%lld\n", ENTITY_NAME(con->peer_name));
+		if (con->ops->peer_reset)
+			con->ops->peer_reset(con);
+		mutex_lock(&con->mutex);
+		break;
+
+	case CEPH_MSGR_TAG_RETRY_SESSION:
+		/*
+		 * If we sent a smaller connect_seq than the peer has, try
+		 * again with a larger value.
+		 */
+		dout("process_connect got RETRY my seq = %u, peer_seq = %u\n",
+		     le32_to_cpu(con->out_connect.connect_seq),
+		     le32_to_cpu(con->in_connect.connect_seq));
+		con->connect_seq = le32_to_cpu(con->in_connect.connect_seq);
+		prepare_write_connect(con->msgr, con, 0);
+		prepare_read_connect(con);
+		break;
+
+	case CEPH_MSGR_TAG_RETRY_GLOBAL:
+		/*
+		 * If we sent a smaller global_seq than the peer has, try
+		 * again with a larger value.
+		 */
+		dout("process_connect got RETRY_GLOBAL my %u peer_gseq %u\n",
+		     con->peer_global_seq,
+		     le32_to_cpu(con->in_connect.global_seq));
+		get_global_seq(con->msgr,
+			       le32_to_cpu(con->in_connect.global_seq));
+		prepare_write_connect(con->msgr, con, 0);
+		prepare_read_connect(con);
+		break;
+
+	case CEPH_MSGR_TAG_READY:
+		if (req_feat & ~server_feat) {
+			pr_err("%s%lld %s protocol feature mismatch,"
+			       " my required %llx > server's %llx, need %llx\n",
+			       ENTITY_NAME(con->peer_name),
+			       ceph_pr_addr(&con->peer_addr.in_addr),
+			       req_feat, server_feat, req_feat & ~server_feat);
+			con->error_msg = "missing required protocol features";
+			fail_protocol(con);
+			return -1;
+		}
+		clear_bit(CONNECTING, &con->state);
+		con->peer_global_seq = le32_to_cpu(con->in_reply.global_seq);
+		con->connect_seq++;
+		con->peer_features = server_feat;
+		dout("process_connect got READY gseq %d cseq %d (%d)\n",
+		     con->peer_global_seq,
+		     le32_to_cpu(con->in_reply.connect_seq),
+		     con->connect_seq);
+		WARN_ON(con->connect_seq !=
+			le32_to_cpu(con->in_reply.connect_seq));
+
+		if (con->in_reply.flags & CEPH_MSG_CONNECT_LOSSY)
+			set_bit(LOSSYTX, &con->state);
+
+		prepare_read_tag(con);
+		break;
+
+	case CEPH_MSGR_TAG_WAIT:
+		/*
+		 * If there is a connection race (we are opening
+		 * connections to each other), one of us may just have
+		 * to WAIT.  This shouldn't happen if we are the
+		 * client.
+		 */
+		pr_err("process_connect peer connecting WAIT\n");
+
+	default:
+		pr_err("connect protocol error, will retry\n");
+		con->error_msg = "protocol error, garbage tag during connect";
+		return -1;
+	}
+	return 0;
+}
+
+
+/*
+ * read (part of) an ack
+ */
+static int read_partial_ack(struct ceph_connection *con)
+{
+	int to = 0;
+
+	return read_partial(con, &to, sizeof(con->in_temp_ack),
+			    &con->in_temp_ack);
+}
+
+
+/*
+ * We can finally discard anything that's been acked.
+ */
+static void process_ack(struct ceph_connection *con)
+{
+	struct ceph_msg *m;
+	u64 ack = le64_to_cpu(con->in_temp_ack);
+	u64 seq;
+
+	while (!list_empty(&con->out_sent)) {
+		m = list_first_entry(&con->out_sent, struct ceph_msg,
+				     list_head);
+		seq = le64_to_cpu(m->hdr.seq);
+		if (seq > ack)
+			break;
+		dout("got ack for seq %llu type %d at %p\n", seq,
+		     le16_to_cpu(m->hdr.type), m);
+		ceph_msg_remove(m);
+	}
+	prepare_read_tag(con);
+}
+
+
+
+
+static int read_partial_message_section(struct ceph_connection *con,
+					struct kvec *section,
+					unsigned int sec_len, u32 *crc)
+{
+	int ret, left;
+
+	BUG_ON(!section);
+
+	while (section->iov_len < sec_len) {
+		BUG_ON(section->iov_base == NULL);
+		left = sec_len - section->iov_len;
+		ret = ceph_tcp_recvmsg(con->sock, (char *)section->iov_base +
+				       section->iov_len, left);
+		if (ret <= 0)
+			return ret;
+		section->iov_len += ret;
+		if (section->iov_len == sec_len)
+			*crc = crc32c(0, section->iov_base,
+				      section->iov_len);
+	}
+
+	return 1;
+}
+
+static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con,
+				struct ceph_msg_header *hdr,
+				int *skip);
+
+
+static int read_partial_message_pages(struct ceph_connection *con,
+				      struct page **pages,
+				      unsigned data_len, int datacrc)
+{
+	void *p;
+	int ret;
+	int left;
+
+	left = min((int)(data_len - con->in_msg_pos.data_pos),
+		   (int)(PAGE_SIZE - con->in_msg_pos.page_pos));
+	/* (page) data */
+	BUG_ON(pages == NULL);
+	p = kmap(pages[con->in_msg_pos.page]);
+	ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
+			       left);
+	if (ret > 0 && datacrc)
+		con->in_data_crc =
+			crc32c(con->in_data_crc,
+				  p + con->in_msg_pos.page_pos, ret);
+	kunmap(pages[con->in_msg_pos.page]);
+	if (ret <= 0)
+		return ret;
+	con->in_msg_pos.data_pos += ret;
+	con->in_msg_pos.page_pos += ret;
+	if (con->in_msg_pos.page_pos == PAGE_SIZE) {
+		con->in_msg_pos.page_pos = 0;
+		con->in_msg_pos.page++;
+	}
+
+	return ret;
+}
+
+#ifdef CONFIG_BLOCK
+static int read_partial_message_bio(struct ceph_connection *con,
+				    struct bio **bio_iter, int *bio_seg,
+				    unsigned data_len, int datacrc)
+{
+	struct bio_vec *bv = bio_iovec_idx(*bio_iter, *bio_seg);
+	void *p;
+	int ret, left;
+
+	if (IS_ERR(bv))
+		return PTR_ERR(bv);
+
+	left = min((int)(data_len - con->in_msg_pos.data_pos),
+		   (int)(bv->bv_len - con->in_msg_pos.page_pos));
+
+	p = kmap(bv->bv_page) + bv->bv_offset;
+
+	ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
+			       left);
+	if (ret > 0 && datacrc)
+		con->in_data_crc =
+			crc32c(con->in_data_crc,
+				  p + con->in_msg_pos.page_pos, ret);
+	kunmap(bv->bv_page);
+	if (ret <= 0)
+		return ret;
+	con->in_msg_pos.data_pos += ret;
+	con->in_msg_pos.page_pos += ret;
+	if (con->in_msg_pos.page_pos == bv->bv_len) {
+		con->in_msg_pos.page_pos = 0;
+		iter_bio_next(bio_iter, bio_seg);
+	}
+
+	return ret;
+}
+#endif
+
+/*
+ * read (part of) a message.
+ */
+static int read_partial_message(struct ceph_connection *con)
+{
+	struct ceph_msg *m = con->in_msg;
+	int ret;
+	int to, left;
+	unsigned front_len, middle_len, data_len, data_off;
+	int datacrc = con->msgr->nocrc;
+	int skip;
+	u64 seq;
+
+	dout("read_partial_message con %p msg %p\n", con, m);
+
+	/* header */
+	while (con->in_base_pos < sizeof(con->in_hdr)) {
+		left = sizeof(con->in_hdr) - con->in_base_pos;
+		ret = ceph_tcp_recvmsg(con->sock,
+				       (char *)&con->in_hdr + con->in_base_pos,
+				       left);
+		if (ret <= 0)
+			return ret;
+		con->in_base_pos += ret;
+		if (con->in_base_pos == sizeof(con->in_hdr)) {
+			u32 crc = crc32c(0, (void *)&con->in_hdr,
+				 sizeof(con->in_hdr) - sizeof(con->in_hdr.crc));
+			if (crc != le32_to_cpu(con->in_hdr.crc)) {
+				pr_err("read_partial_message bad hdr "
+				       " crc %u != expected %u\n",
+				       crc, con->in_hdr.crc);
+				return -EBADMSG;
+			}
+		}
+	}
+	front_len = le32_to_cpu(con->in_hdr.front_len);
+	if (front_len > CEPH_MSG_MAX_FRONT_LEN)
+		return -EIO;
+	middle_len = le32_to_cpu(con->in_hdr.middle_len);
+	if (middle_len > CEPH_MSG_MAX_DATA_LEN)
+		return -EIO;
+	data_len = le32_to_cpu(con->in_hdr.data_len);
+	if (data_len > CEPH_MSG_MAX_DATA_LEN)
+		return -EIO;
+	data_off = le16_to_cpu(con->in_hdr.data_off);
+
+	/* verify seq# */
+	seq = le64_to_cpu(con->in_hdr.seq);
+	if ((s64)seq - (s64)con->in_seq < 1) {
+		pr_info("skipping %s%lld %s seq %lld, expected %lld\n",
+			ENTITY_NAME(con->peer_name),
+			ceph_pr_addr(&con->peer_addr.in_addr),
+			seq, con->in_seq + 1);
+		con->in_base_pos = -front_len - middle_len - data_len -
+			sizeof(m->footer);
+		con->in_tag = CEPH_MSGR_TAG_READY;
+		con->in_seq++;
+		return 0;
+	} else if ((s64)seq - (s64)con->in_seq > 1) {
+		pr_err("read_partial_message bad seq %lld expected %lld\n",
+		       seq, con->in_seq + 1);
+		con->error_msg = "bad message sequence # for incoming message";
+		return -EBADMSG;
+	}
+
+	/* allocate message? */
+	if (!con->in_msg) {
+		dout("got hdr type %d front %d data %d\n", con->in_hdr.type,
+		     con->in_hdr.front_len, con->in_hdr.data_len);
+		skip = 0;
+		con->in_msg = ceph_alloc_msg(con, &con->in_hdr, &skip);
+		if (skip) {
+			/* skip this message */
+			dout("alloc_msg said skip message\n");
+			BUG_ON(con->in_msg);
+			con->in_base_pos = -front_len - middle_len - data_len -
+				sizeof(m->footer);
+			con->in_tag = CEPH_MSGR_TAG_READY;
+			con->in_seq++;
+			return 0;
+		}
+		if (!con->in_msg) {
+			con->error_msg =
+				"error allocating memory for incoming message";
+			return -ENOMEM;
+		}
+		m = con->in_msg;
+		m->front.iov_len = 0;    /* haven't read it yet */
+		if (m->middle)
+			m->middle->vec.iov_len = 0;
+
+		con->in_msg_pos.page = 0;
+		if (m->pages)
+			con->in_msg_pos.page_pos = data_off & ~PAGE_MASK;
+		else
+			con->in_msg_pos.page_pos = 0;
+		con->in_msg_pos.data_pos = 0;
+	}
+
+	/* front */
+	ret = read_partial_message_section(con, &m->front, front_len,
+					   &con->in_front_crc);
+	if (ret <= 0)
+		return ret;
+
+	/* middle */
+	if (m->middle) {
+		ret = read_partial_message_section(con, &m->middle->vec,
+						   middle_len,
+						   &con->in_middle_crc);
+		if (ret <= 0)
+			return ret;
+	}
+#ifdef CONFIG_BLOCK
+	if (m->bio && !m->bio_iter)
+		init_bio_iter(m->bio, &m->bio_iter, &m->bio_seg);
+#endif
+
+	/* (page) data */
+	while (con->in_msg_pos.data_pos < data_len) {
+		if (m->pages) {
+			ret = read_partial_message_pages(con, m->pages,
+						 data_len, datacrc);
+			if (ret <= 0)
+				return ret;
+#ifdef CONFIG_BLOCK
+		} else if (m->bio) {
+
+			ret = read_partial_message_bio(con,
+						 &m->bio_iter, &m->bio_seg,
+						 data_len, datacrc);
+			if (ret <= 0)
+				return ret;
+#endif
+		} else {
+			BUG_ON(1);
+		}
+	}
+
+	/* footer */
+	to = sizeof(m->hdr) + sizeof(m->footer);
+	while (con->in_base_pos < to) {
+		left = to - con->in_base_pos;
+		ret = ceph_tcp_recvmsg(con->sock, (char *)&m->footer +
+				       (con->in_base_pos - sizeof(m->hdr)),
+				       left);
+		if (ret <= 0)
+			return ret;
+		con->in_base_pos += ret;
+	}
+	dout("read_partial_message got msg %p %d (%u) + %d (%u) + %d (%u)\n",
+	     m, front_len, m->footer.front_crc, middle_len,
+	     m->footer.middle_crc, data_len, m->footer.data_crc);
+
+	/* crc ok? */
+	if (con->in_front_crc != le32_to_cpu(m->footer.front_crc)) {
+		pr_err("read_partial_message %p front crc %u != exp. %u\n",
+		       m, con->in_front_crc, m->footer.front_crc);
+		return -EBADMSG;
+	}
+	if (con->in_middle_crc != le32_to_cpu(m->footer.middle_crc)) {
+		pr_err("read_partial_message %p middle crc %u != exp %u\n",
+		       m, con->in_middle_crc, m->footer.middle_crc);
+		return -EBADMSG;
+	}
+	if (datacrc &&
+	    (m->footer.flags & CEPH_MSG_FOOTER_NOCRC) == 0 &&
+	    con->in_data_crc != le32_to_cpu(m->footer.data_crc)) {
+		pr_err("read_partial_message %p data crc %u != exp. %u\n", m,
+		       con->in_data_crc, le32_to_cpu(m->footer.data_crc));
+		return -EBADMSG;
+	}
+
+	return 1; /* done! */
+}
+
+/*
+ * Process message.  This happens in the worker thread.  The callback should
+ * be careful not to do anything that waits on other incoming messages or it
+ * may deadlock.
+ */
+static void process_message(struct ceph_connection *con)
+{
+	struct ceph_msg *msg;
+
+	msg = con->in_msg;
+	con->in_msg = NULL;
+
+	/* if first message, set peer_name */
+	if (con->peer_name.type == 0)
+		con->peer_name = msg->hdr.src;
+
+	con->in_seq++;
+	mutex_unlock(&con->mutex);
+
+	dout("===== %p %llu from %s%lld %d=%s len %d+%d (%u %u %u) =====\n",
+	     msg, le64_to_cpu(msg->hdr.seq),
+	     ENTITY_NAME(msg->hdr.src),
+	     le16_to_cpu(msg->hdr.type),
+	     ceph_msg_type_name(le16_to_cpu(msg->hdr.type)),
+	     le32_to_cpu(msg->hdr.front_len),
+	     le32_to_cpu(msg->hdr.data_len),
+	     con->in_front_crc, con->in_middle_crc, con->in_data_crc);
+	con->ops->dispatch(con, msg);
+
+	mutex_lock(&con->mutex);
+	prepare_read_tag(con);
+}
+
+
+/*
+ * Write something to the socket.  Called in a worker thread when the
+ * socket appears to be writeable and we have something ready to send.
+ */
+static int try_write(struct ceph_connection *con)
+{
+	struct ceph_messenger *msgr = con->msgr;
+	int ret = 1;
+
+	dout("try_write start %p state %lu nref %d\n", con, con->state,
+	     atomic_read(&con->nref));
+
+more:
+	dout("try_write out_kvec_bytes %d\n", con->out_kvec_bytes);
+
+	/* open the socket first? */
+	if (con->sock == NULL) {
+		/*
+		 * if we were STANDBY and are reconnecting _this_
+		 * connection, bump connect_seq now.  Always bump
+		 * global_seq.
+		 */
+		if (test_and_clear_bit(STANDBY, &con->state))
+			con->connect_seq++;
+
+		prepare_write_banner(msgr, con);
+		prepare_write_connect(msgr, con, 1);
+		prepare_read_banner(con);
+		set_bit(CONNECTING, &con->state);
+		clear_bit(NEGOTIATING, &con->state);
+
+		BUG_ON(con->in_msg);
+		con->in_tag = CEPH_MSGR_TAG_READY;
+		dout("try_write initiating connect on %p new state %lu\n",
+		     con, con->state);
+		con->sock = ceph_tcp_connect(con);
+		if (IS_ERR(con->sock)) {
+			con->sock = NULL;
+			con->error_msg = "connect error";
+			ret = -1;
+			goto out;
+		}
+	}
+
+more_kvec:
+	/* kvec data queued? */
+	if (con->out_skip) {
+		ret = write_partial_skip(con);
+		if (ret <= 0)
+			goto done;
+		if (ret < 0) {
+			dout("try_write write_partial_skip err %d\n", ret);
+			goto done;
+		}
+	}
+	if (con->out_kvec_left) {
+		ret = write_partial_kvec(con);
+		if (ret <= 0)
+			goto done;
+	}
+
+	/* msg pages? */
+	if (con->out_msg) {
+		if (con->out_msg_done) {
+			ceph_msg_put(con->out_msg);
+			con->out_msg = NULL;   /* we're done with this one */
+			goto do_next;
+		}
+
+		ret = write_partial_msg_pages(con);
+		if (ret == 1)
+			goto more_kvec;  /* we need to send the footer, too! */
+		if (ret == 0)
+			goto done;
+		if (ret < 0) {
+			dout("try_write write_partial_msg_pages err %d\n",
+			     ret);
+			goto done;
+		}
+	}
+
+do_next:
+	if (!test_bit(CONNECTING, &con->state)) {
+		/* is anything else pending? */
+		if (!list_empty(&con->out_queue)) {
+			prepare_write_message(con);
+			goto more;
+		}
+		if (con->in_seq > con->in_seq_acked) {
+			prepare_write_ack(con);
+			goto more;
+		}
+		if (test_and_clear_bit(KEEPALIVE_PENDING, &con->state)) {
+			prepare_write_keepalive(con);
+			goto more;
+		}
+	}
+
+	/* Nothing to do! */
+	clear_bit(WRITE_PENDING, &con->state);
+	dout("try_write nothing else to write.\n");
+done:
+	ret = 0;
+out:
+	dout("try_write done on %p\n", con);
+	return ret;
+}
+
+
+
+/*
+ * Read what we can from the socket.
+ */
+static int try_read(struct ceph_connection *con)
+{
+	int ret = -1;
+
+	if (!con->sock)
+		return 0;
+
+	if (test_bit(STANDBY, &con->state))
+		return 0;
+
+	dout("try_read start on %p\n", con);
+
+more:
+	dout("try_read tag %d in_base_pos %d\n", (int)con->in_tag,
+	     con->in_base_pos);
+	if (test_bit(CONNECTING, &con->state)) {
+		if (!test_bit(NEGOTIATING, &con->state)) {
+			dout("try_read connecting\n");
+			ret = read_partial_banner(con);
+			if (ret <= 0)
+				goto done;
+			if (process_banner(con) < 0) {
+				ret = -1;
+				goto out;
+			}
+		}
+		ret = read_partial_connect(con);
+		if (ret <= 0)
+			goto done;
+		if (process_connect(con) < 0) {
+			ret = -1;
+			goto out;
+		}
+		goto more;
+	}
+
+	if (con->in_base_pos < 0) {
+		/*
+		 * skipping + discarding content.
+		 *
+		 * FIXME: there must be a better way to do this!
+		 */
+		static char buf[1024];
+		int skip = min(1024, -con->in_base_pos);
+		dout("skipping %d / %d bytes\n", skip, -con->in_base_pos);
+		ret = ceph_tcp_recvmsg(con->sock, buf, skip);
+		if (ret <= 0)
+			goto done;
+		con->in_base_pos += ret;
+		if (con->in_base_pos)
+			goto more;
+	}
+	if (con->in_tag == CEPH_MSGR_TAG_READY) {
+		/*
+		 * what's next?
+		 */
+		ret = ceph_tcp_recvmsg(con->sock, &con->in_tag, 1);
+		if (ret <= 0)
+			goto done;
+		dout("try_read got tag %d\n", (int)con->in_tag);
+		switch (con->in_tag) {
+		case CEPH_MSGR_TAG_MSG:
+			prepare_read_message(con);
+			break;
+		case CEPH_MSGR_TAG_ACK:
+			prepare_read_ack(con);
+			break;
+		case CEPH_MSGR_TAG_CLOSE:
+			set_bit(CLOSED, &con->state);   /* fixme */
+			goto done;
+		default:
+			goto bad_tag;
+		}
+	}
+	if (con->in_tag == CEPH_MSGR_TAG_MSG) {
+		ret = read_partial_message(con);
+		if (ret <= 0) {
+			switch (ret) {
+			case -EBADMSG:
+				con->error_msg = "bad crc";
+				ret = -EIO;
+				goto out;
+			case -EIO:
+				con->error_msg = "io error";
+				goto out;
+			default:
+				goto done;
+			}
+		}
+		if (con->in_tag == CEPH_MSGR_TAG_READY)
+			goto more;
+		process_message(con);
+		goto more;
+	}
+	if (con->in_tag == CEPH_MSGR_TAG_ACK) {
+		ret = read_partial_ack(con);
+		if (ret <= 0)
+			goto done;
+		process_ack(con);
+		goto more;
+	}
+
+done:
+	ret = 0;
+out:
+	dout("try_read done on %p\n", con);
+	return ret;
+
+bad_tag:
+	pr_err("try_read bad con->in_tag = %d\n", (int)con->in_tag);
+	con->error_msg = "protocol error, garbage tag";
+	ret = -1;
+	goto out;
+}
+
+
+/*
+ * Atomically queue work on a connection.  Bump @con reference to
+ * avoid races with connection teardown.
+ *
+ * There is some trickery going on with QUEUED and BUSY because we
+ * only want a _single_ thread operating on each connection at any
+ * point in time, but we want to use all available CPUs.
+ *
+ * The worker thread only proceeds if it can atomically set BUSY.  It
+ * clears QUEUED and does it's thing.  When it thinks it's done, it
+ * clears BUSY, then rechecks QUEUED.. if it's set again, it loops
+ * (tries again to set BUSY).
+ *
+ * To queue work, we first set QUEUED, _then_ if BUSY isn't set, we
+ * try to queue work.  If that fails (work is already queued, or BUSY)
+ * we give up (work also already being done or is queued) but leave QUEUED
+ * set so that the worker thread will loop if necessary.
+ */
+static void queue_con(struct ceph_connection *con)
+{
+	if (test_bit(DEAD, &con->state)) {
+		dout("queue_con %p ignoring: DEAD\n",
+		     con);
+		return;
+	}
+
+	if (!con->ops->get(con)) {
+		dout("queue_con %p ref count 0\n", con);
+		return;
+	}
+
+	set_bit(QUEUED, &con->state);
+	if (test_bit(BUSY, &con->state)) {
+		dout("queue_con %p - already BUSY\n", con);
+		con->ops->put(con);
+	} else if (!queue_work(ceph_msgr_wq, &con->work.work)) {
+		dout("queue_con %p - already queued\n", con);
+		con->ops->put(con);
+	} else {
+		dout("queue_con %p\n", con);
+	}
+}
+
+/*
+ * Do some work on a connection.  Drop a connection ref when we're done.
+ */
+static void con_work(struct work_struct *work)
+{
+	struct ceph_connection *con = container_of(work, struct ceph_connection,
+						   work.work);
+	int backoff = 0;
+
+more:
+	if (test_and_set_bit(BUSY, &con->state) != 0) {
+		dout("con_work %p BUSY already set\n", con);
+		goto out;
+	}
+	dout("con_work %p start, clearing QUEUED\n", con);
+	clear_bit(QUEUED, &con->state);
+
+	mutex_lock(&con->mutex);
+
+	if (test_bit(CLOSED, &con->state)) { /* e.g. if we are replaced */
+		dout("con_work CLOSED\n");
+		con_close_socket(con);
+		goto done;
+	}
+	if (test_and_clear_bit(OPENING, &con->state)) {
+		/* reopen w/ new peer */
+		dout("con_work OPENING\n");
+		con_close_socket(con);
+	}
+
+	if (test_and_clear_bit(SOCK_CLOSED, &con->state) ||
+	    try_read(con) < 0 ||
+	    try_write(con) < 0) {
+		mutex_unlock(&con->mutex);
+		backoff = 1;
+		ceph_fault(con);     /* error/fault path */
+		goto done_unlocked;
+	}
+
+done:
+	mutex_unlock(&con->mutex);
+
+done_unlocked:
+	clear_bit(BUSY, &con->state);
+	dout("con->state=%lu\n", con->state);
+	if (test_bit(QUEUED, &con->state)) {
+		if (!backoff || test_bit(OPENING, &con->state)) {
+			dout("con_work %p QUEUED reset, looping\n", con);
+			goto more;
+		}
+		dout("con_work %p QUEUED reset, but just faulted\n", con);
+		clear_bit(QUEUED, &con->state);
+	}
+	dout("con_work %p done\n", con);
+
+out:
+	con->ops->put(con);
+}
+
+
+/*
+ * Generic error/fault handler.  A retry mechanism is used with
+ * exponential backoff
+ */
+static void ceph_fault(struct ceph_connection *con)
+{
+	pr_err("%s%lld %s %s\n", ENTITY_NAME(con->peer_name),
+	       ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg);
+	dout("fault %p state %lu to peer %s\n",
+	     con, con->state, ceph_pr_addr(&con->peer_addr.in_addr));
+
+	if (test_bit(LOSSYTX, &con->state)) {
+		dout("fault on LOSSYTX channel\n");
+		goto out;
+	}
+
+	mutex_lock(&con->mutex);
+	if (test_bit(CLOSED, &con->state))
+		goto out_unlock;
+
+	con_close_socket(con);
+
+	if (con->in_msg) {
+		ceph_msg_put(con->in_msg);
+		con->in_msg = NULL;
+	}
+
+	/* Requeue anything that hasn't been acked */
+	list_splice_init(&con->out_sent, &con->out_queue);
+
+	/* If there are no messages in the queue, place the connection
+	 * in a STANDBY state (i.e., don't try to reconnect just yet). */
+	if (list_empty(&con->out_queue) && !con->out_keepalive_pending) {
+		dout("fault setting STANDBY\n");
+		set_bit(STANDBY, &con->state);
+	} else {
+		/* retry after a delay. */
+		if (con->delay == 0)
+			con->delay = BASE_DELAY_INTERVAL;
+		else if (con->delay < MAX_DELAY_INTERVAL)
+			con->delay *= 2;
+		dout("fault queueing %p delay %lu\n", con, con->delay);
+		con->ops->get(con);
+		if (queue_delayed_work(ceph_msgr_wq, &con->work,
+				       round_jiffies_relative(con->delay)) == 0)
+			con->ops->put(con);
+	}
+
+out_unlock:
+	mutex_unlock(&con->mutex);
+out:
+	/*
+	 * in case we faulted due to authentication, invalidate our
+	 * current tickets so that we can get new ones.
+	 */
+	if (con->auth_retry && con->ops->invalidate_authorizer) {
+		dout("calling invalidate_authorizer()\n");
+		con->ops->invalidate_authorizer(con);
+	}
+
+	if (con->ops->fault)
+		con->ops->fault(con);
+}
+
+
+
+/*
+ * create a new messenger instance
+ */
+struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr,
+					     u32 supported_features,
+					     u32 required_features)
+{
+	struct ceph_messenger *msgr;
+
+	msgr = kzalloc(sizeof(*msgr), GFP_KERNEL);
+	if (msgr == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	msgr->supported_features = supported_features;
+	msgr->required_features = required_features;
+
+	spin_lock_init(&msgr->global_seq_lock);
+
+	/* the zero page is needed if a request is "canceled" while the message
+	 * is being written over the socket */
+	msgr->zero_page = __page_cache_alloc(GFP_KERNEL | __GFP_ZERO);
+	if (!msgr->zero_page) {
+		kfree(msgr);
+		return ERR_PTR(-ENOMEM);
+	}
+	kmap(msgr->zero_page);
+
+	if (myaddr)
+		msgr->inst.addr = *myaddr;
+
+	/* select a random nonce */
+	msgr->inst.addr.type = 0;
+	get_random_bytes(&msgr->inst.addr.nonce, sizeof(msgr->inst.addr.nonce));
+	encode_my_addr(msgr);
+
+	dout("messenger_create %p\n", msgr);
+	return msgr;
+}
+EXPORT_SYMBOL(ceph_messenger_create);
+
+void ceph_messenger_destroy(struct ceph_messenger *msgr)
+{
+	dout("destroy %p\n", msgr);
+	kunmap(msgr->zero_page);
+	__free_page(msgr->zero_page);
+	kfree(msgr);
+	dout("destroyed messenger %p\n", msgr);
+}
+EXPORT_SYMBOL(ceph_messenger_destroy);
+
+/*
+ * Queue up an outgoing message on the given connection.
+ */
+void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
+{
+	if (test_bit(CLOSED, &con->state)) {
+		dout("con_send %p closed, dropping %p\n", con, msg);
+		ceph_msg_put(msg);
+		return;
+	}
+
+	/* set src+dst */
+	msg->hdr.src = con->msgr->inst.name;
+
+	BUG_ON(msg->front.iov_len != le32_to_cpu(msg->hdr.front_len));
+
+	msg->needs_out_seq = true;
+
+	/* queue */
+	mutex_lock(&con->mutex);
+	BUG_ON(!list_empty(&msg->list_head));
+	list_add_tail(&msg->list_head, &con->out_queue);
+	dout("----- %p to %s%lld %d=%s len %d+%d+%d -----\n", msg,
+	     ENTITY_NAME(con->peer_name), le16_to_cpu(msg->hdr.type),
+	     ceph_msg_type_name(le16_to_cpu(msg->hdr.type)),
+	     le32_to_cpu(msg->hdr.front_len),
+	     le32_to_cpu(msg->hdr.middle_len),
+	     le32_to_cpu(msg->hdr.data_len));
+	mutex_unlock(&con->mutex);
+
+	/* if there wasn't anything waiting to send before, queue
+	 * new work */
+	if (test_and_set_bit(WRITE_PENDING, &con->state) == 0)
+		queue_con(con);
+}
+EXPORT_SYMBOL(ceph_con_send);
+
+/*
+ * Revoke a message that was previously queued for send
+ */
+void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg)
+{
+	mutex_lock(&con->mutex);
+	if (!list_empty(&msg->list_head)) {
+		dout("con_revoke %p msg %p - was on queue\n", con, msg);
+		list_del_init(&msg->list_head);
+		ceph_msg_put(msg);
+		msg->hdr.seq = 0;
+	}
+	if (con->out_msg == msg) {
+		dout("con_revoke %p msg %p - was sending\n", con, msg);
+		con->out_msg = NULL;
+		if (con->out_kvec_is_msg) {
+			con->out_skip = con->out_kvec_bytes;
+			con->out_kvec_is_msg = false;
+		}
+		ceph_msg_put(msg);
+		msg->hdr.seq = 0;
+	}
+	mutex_unlock(&con->mutex);
+}
+
+/*
+ * Revoke a message that we may be reading data into
+ */
+void ceph_con_revoke_message(struct ceph_connection *con, struct ceph_msg *msg)
+{
+	mutex_lock(&con->mutex);
+	if (con->in_msg && con->in_msg == msg) {
+		unsigned front_len = le32_to_cpu(con->in_hdr.front_len);
+		unsigned middle_len = le32_to_cpu(con->in_hdr.middle_len);
+		unsigned data_len = le32_to_cpu(con->in_hdr.data_len);
+
+		/* skip rest of message */
+		dout("con_revoke_pages %p msg %p revoked\n", con, msg);
+			con->in_base_pos = con->in_base_pos -
+				sizeof(struct ceph_msg_header) -
+				front_len -
+				middle_len -
+				data_len -
+				sizeof(struct ceph_msg_footer);
+		ceph_msg_put(con->in_msg);
+		con->in_msg = NULL;
+		con->in_tag = CEPH_MSGR_TAG_READY;
+		con->in_seq++;
+	} else {
+		dout("con_revoke_pages %p msg %p pages %p no-op\n",
+		     con, con->in_msg, msg);
+	}
+	mutex_unlock(&con->mutex);
+}
+
+/*
+ * Queue a keepalive byte to ensure the tcp connection is alive.
+ */
+void ceph_con_keepalive(struct ceph_connection *con)
+{
+	if (test_and_set_bit(KEEPALIVE_PENDING, &con->state) == 0 &&
+	    test_and_set_bit(WRITE_PENDING, &con->state) == 0)
+		queue_con(con);
+}
+EXPORT_SYMBOL(ceph_con_keepalive);
+
+
+/*
+ * construct a new message with given type, size
+ * the new msg has a ref count of 1.
+ */
+struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags)
+{
+	struct ceph_msg *m;
+
+	m = kmalloc(sizeof(*m), flags);
+	if (m == NULL)
+		goto out;
+	kref_init(&m->kref);
+	INIT_LIST_HEAD(&m->list_head);
+
+	m->hdr.tid = 0;
+	m->hdr.type = cpu_to_le16(type);
+	m->hdr.priority = cpu_to_le16(CEPH_MSG_PRIO_DEFAULT);
+	m->hdr.version = 0;
+	m->hdr.front_len = cpu_to_le32(front_len);
+	m->hdr.middle_len = 0;
+	m->hdr.data_len = 0;
+	m->hdr.data_off = 0;
+	m->hdr.reserved = 0;
+	m->footer.front_crc = 0;
+	m->footer.middle_crc = 0;
+	m->footer.data_crc = 0;
+	m->footer.flags = 0;
+	m->front_max = front_len;
+	m->front_is_vmalloc = false;
+	m->more_to_follow = false;
+	m->pool = NULL;
+
+	/* front */
+	if (front_len) {
+		if (front_len > PAGE_CACHE_SIZE) {
+			m->front.iov_base = __vmalloc(front_len, flags,
+						      PAGE_KERNEL);
+			m->front_is_vmalloc = true;
+		} else {
+			m->front.iov_base = kmalloc(front_len, flags);
+		}
+		if (m->front.iov_base == NULL) {
+			pr_err("msg_new can't allocate %d bytes\n",
+			     front_len);
+			goto out2;
+		}
+	} else {
+		m->front.iov_base = NULL;
+	}
+	m->front.iov_len = front_len;
+
+	/* middle */
+	m->middle = NULL;
+
+	/* data */
+	m->nr_pages = 0;
+	m->pages = NULL;
+	m->pagelist = NULL;
+	m->bio = NULL;
+	m->bio_iter = NULL;
+	m->bio_seg = 0;
+	m->trail = NULL;
+
+	dout("ceph_msg_new %p front %d\n", m, front_len);
+	return m;
+
+out2:
+	ceph_msg_put(m);
+out:
+	pr_err("msg_new can't create type %d front %d\n", type, front_len);
+	return NULL;
+}
+EXPORT_SYMBOL(ceph_msg_new);
+
+/*
+ * Allocate "middle" portion of a message, if it is needed and wasn't
+ * allocated by alloc_msg.  This allows us to read a small fixed-size
+ * per-type header in the front and then gracefully fail (i.e.,
+ * propagate the error to the caller based on info in the front) when
+ * the middle is too large.
+ */
+static int ceph_alloc_middle(struct ceph_connection *con, struct ceph_msg *msg)
+{
+	int type = le16_to_cpu(msg->hdr.type);
+	int middle_len = le32_to_cpu(msg->hdr.middle_len);
+
+	dout("alloc_middle %p type %d %s middle_len %d\n", msg, type,
+	     ceph_msg_type_name(type), middle_len);
+	BUG_ON(!middle_len);
+	BUG_ON(msg->middle);
+
+	msg->middle = ceph_buffer_new(middle_len, GFP_NOFS);
+	if (!msg->middle)
+		return -ENOMEM;
+	return 0;
+}
+
+/*
+ * Generic message allocator, for incoming messages.
+ */
+static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con,
+				struct ceph_msg_header *hdr,
+				int *skip)
+{
+	int type = le16_to_cpu(hdr->type);
+	int front_len = le32_to_cpu(hdr->front_len);
+	int middle_len = le32_to_cpu(hdr->middle_len);
+	struct ceph_msg *msg = NULL;
+	int ret;
+
+	if (con->ops->alloc_msg) {
+		mutex_unlock(&con->mutex);
+		msg = con->ops->alloc_msg(con, hdr, skip);
+		mutex_lock(&con->mutex);
+		if (!msg || *skip)
+			return NULL;
+	}
+	if (!msg) {
+		*skip = 0;
+		msg = ceph_msg_new(type, front_len, GFP_NOFS);
+		if (!msg) {
+			pr_err("unable to allocate msg type %d len %d\n",
+			       type, front_len);
+			return NULL;
+		}
+	}
+	memcpy(&msg->hdr, &con->in_hdr, sizeof(con->in_hdr));
+
+	if (middle_len && !msg->middle) {
+		ret = ceph_alloc_middle(con, msg);
+		if (ret < 0) {
+			ceph_msg_put(msg);
+			return NULL;
+		}
+	}
+
+	return msg;
+}
+
+
+/*
+ * Free a generically kmalloc'd message.
+ */
+void ceph_msg_kfree(struct ceph_msg *m)
+{
+	dout("msg_kfree %p\n", m);
+	if (m->front_is_vmalloc)
+		vfree(m->front.iov_base);
+	else
+		kfree(m->front.iov_base);
+	kfree(m);
+}
+
+/*
+ * Drop a msg ref.  Destroy as needed.
+ */
+void ceph_msg_last_put(struct kref *kref)
+{
+	struct ceph_msg *m = container_of(kref, struct ceph_msg, kref);
+
+	dout("ceph_msg_put last one on %p\n", m);
+	WARN_ON(!list_empty(&m->list_head));
+
+	/* drop middle, data, if any */
+	if (m->middle) {
+		ceph_buffer_put(m->middle);
+		m->middle = NULL;
+	}
+	m->nr_pages = 0;
+	m->pages = NULL;
+
+	if (m->pagelist) {
+		ceph_pagelist_release(m->pagelist);
+		kfree(m->pagelist);
+		m->pagelist = NULL;
+	}
+
+	m->trail = NULL;
+
+	if (m->pool)
+		ceph_msgpool_put(m->pool, m);
+	else
+		ceph_msg_kfree(m);
+}
+EXPORT_SYMBOL(ceph_msg_last_put);
+
+void ceph_msg_dump(struct ceph_msg *msg)
+{
+	pr_debug("msg_dump %p (front_max %d nr_pages %d)\n", msg,
+		 msg->front_max, msg->nr_pages);
+	print_hex_dump(KERN_DEBUG, "header: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       &msg->hdr, sizeof(msg->hdr), true);
+	print_hex_dump(KERN_DEBUG, " front: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       msg->front.iov_base, msg->front.iov_len, true);
+	if (msg->middle)
+		print_hex_dump(KERN_DEBUG, "middle: ",
+			       DUMP_PREFIX_OFFSET, 16, 1,
+			       msg->middle->vec.iov_base,
+			       msg->middle->vec.iov_len, true);
+	print_hex_dump(KERN_DEBUG, "footer: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       &msg->footer, sizeof(msg->footer), true);
+}
+EXPORT_SYMBOL(ceph_msg_dump);
diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c
new file mode 100644
index 000000000000..8a079399174a
--- /dev/null
+++ b/net/ceph/mon_client.c
@@ -0,0 +1,1027 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/random.h>
+#include <linux/sched.h>
+
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/decode.h>
+
+#include <linux/ceph/auth.h>
+
+/*
+ * Interact with Ceph monitor cluster.  Handle requests for new map
+ * versions, and periodically resend as needed.  Also implement
+ * statfs() and umount().
+ *
+ * A small cluster of Ceph "monitors" are responsible for managing critical
+ * cluster configuration and state information.  An odd number (e.g., 3, 5)
+ * of cmon daemons use a modified version of the Paxos part-time parliament
+ * algorithm to manage the MDS map (mds cluster membership), OSD map, and
+ * list of clients who have mounted the file system.
+ *
+ * We maintain an open, active session with a monitor at all times in order to
+ * receive timely MDSMap updates.  We periodically send a keepalive byte on the
+ * TCP socket to ensure we detect a failure.  If the connection does break, we
+ * randomly hunt for a new monitor.  Once the connection is reestablished, we
+ * resend any outstanding requests.
+ */
+
+static const struct ceph_connection_operations mon_con_ops;
+
+static int __validate_auth(struct ceph_mon_client *monc);
+
+/*
+ * Decode a monmap blob (e.g., during mount).
+ */
+struct ceph_monmap *ceph_monmap_decode(void *p, void *end)
+{
+	struct ceph_monmap *m = NULL;
+	int i, err = -EINVAL;
+	struct ceph_fsid fsid;
+	u32 epoch, num_mon;
+	u16 version;
+	u32 len;
+
+	ceph_decode_32_safe(&p, end, len, bad);
+	ceph_decode_need(&p, end, len, bad);
+
+	dout("monmap_decode %p %p len %d\n", p, end, (int)(end-p));
+
+	ceph_decode_16_safe(&p, end, version, bad);
+
+	ceph_decode_need(&p, end, sizeof(fsid) + 2*sizeof(u32), bad);
+	ceph_decode_copy(&p, &fsid, sizeof(fsid));
+	epoch = ceph_decode_32(&p);
+
+	num_mon = ceph_decode_32(&p);
+	ceph_decode_need(&p, end, num_mon*sizeof(m->mon_inst[0]), bad);
+
+	if (num_mon >= CEPH_MAX_MON)
+		goto bad;
+	m = kmalloc(sizeof(*m) + sizeof(m->mon_inst[0])*num_mon, GFP_NOFS);
+	if (m == NULL)
+		return ERR_PTR(-ENOMEM);
+	m->fsid = fsid;
+	m->epoch = epoch;
+	m->num_mon = num_mon;
+	ceph_decode_copy(&p, m->mon_inst, num_mon*sizeof(m->mon_inst[0]));
+	for (i = 0; i < num_mon; i++)
+		ceph_decode_addr(&m->mon_inst[i].addr);
+
+	dout("monmap_decode epoch %d, num_mon %d\n", m->epoch,
+	     m->num_mon);
+	for (i = 0; i < m->num_mon; i++)
+		dout("monmap_decode  mon%d is %s\n", i,
+		     ceph_pr_addr(&m->mon_inst[i].addr.in_addr));
+	return m;
+
+bad:
+	dout("monmap_decode failed with %d\n", err);
+	kfree(m);
+	return ERR_PTR(err);
+}
+
+/*
+ * return true if *addr is included in the monmap.
+ */
+int ceph_monmap_contains(struct ceph_monmap *m, struct ceph_entity_addr *addr)
+{
+	int i;
+
+	for (i = 0; i < m->num_mon; i++)
+		if (memcmp(addr, &m->mon_inst[i].addr, sizeof(*addr)) == 0)
+			return 1;
+	return 0;
+}
+
+/*
+ * Send an auth request.
+ */
+static void __send_prepared_auth_request(struct ceph_mon_client *monc, int len)
+{
+	monc->pending_auth = 1;
+	monc->m_auth->front.iov_len = len;
+	monc->m_auth->hdr.front_len = cpu_to_le32(len);
+	ceph_con_revoke(monc->con, monc->m_auth);
+	ceph_msg_get(monc->m_auth);  /* keep our ref */
+	ceph_con_send(monc->con, monc->m_auth);
+}
+
+/*
+ * Close monitor session, if any.
+ */
+static void __close_session(struct ceph_mon_client *monc)
+{
+	if (monc->con) {
+		dout("__close_session closing mon%d\n", monc->cur_mon);
+		ceph_con_revoke(monc->con, monc->m_auth);
+		ceph_con_close(monc->con);
+		monc->cur_mon = -1;
+		monc->pending_auth = 0;
+		ceph_auth_reset(monc->auth);
+	}
+}
+
+/*
+ * Open a session with a (new) monitor.
+ */
+static int __open_session(struct ceph_mon_client *monc)
+{
+	char r;
+	int ret;
+
+	if (monc->cur_mon < 0) {
+		get_random_bytes(&r, 1);
+		monc->cur_mon = r % monc->monmap->num_mon;
+		dout("open_session num=%d r=%d -> mon%d\n",
+		     monc->monmap->num_mon, r, monc->cur_mon);
+		monc->sub_sent = 0;
+		monc->sub_renew_after = jiffies;  /* i.e., expired */
+		monc->want_next_osdmap = !!monc->want_next_osdmap;
+
+		dout("open_session mon%d opening\n", monc->cur_mon);
+		monc->con->peer_name.type = CEPH_ENTITY_TYPE_MON;
+		monc->con->peer_name.num = cpu_to_le64(monc->cur_mon);
+		ceph_con_open(monc->con,
+			      &monc->monmap->mon_inst[monc->cur_mon].addr);
+
+		/* initiatiate authentication handshake */
+		ret = ceph_auth_build_hello(monc->auth,
+					    monc->m_auth->front.iov_base,
+					    monc->m_auth->front_max);
+		__send_prepared_auth_request(monc, ret);
+	} else {
+		dout("open_session mon%d already open\n", monc->cur_mon);
+	}
+	return 0;
+}
+
+static bool __sub_expired(struct ceph_mon_client *monc)
+{
+	return time_after_eq(jiffies, monc->sub_renew_after);
+}
+
+/*
+ * Reschedule delayed work timer.
+ */
+static void __schedule_delayed(struct ceph_mon_client *monc)
+{
+	unsigned delay;
+
+	if (monc->cur_mon < 0 || __sub_expired(monc))
+		delay = 10 * HZ;
+	else
+		delay = 20 * HZ;
+	dout("__schedule_delayed after %u\n", delay);
+	schedule_delayed_work(&monc->delayed_work, delay);
+}
+
+/*
+ * Send subscribe request for mdsmap and/or osdmap.
+ */
+static void __send_subscribe(struct ceph_mon_client *monc)
+{
+	dout("__send_subscribe sub_sent=%u exp=%u want_osd=%d\n",
+	     (unsigned)monc->sub_sent, __sub_expired(monc),
+	     monc->want_next_osdmap);
+	if ((__sub_expired(monc) && !monc->sub_sent) ||
+	    monc->want_next_osdmap == 1) {
+		struct ceph_msg *msg = monc->m_subscribe;
+		struct ceph_mon_subscribe_item *i;
+		void *p, *end;
+		int num;
+
+		p = msg->front.iov_base;
+		end = p + msg->front_max;
+
+		num = 1 + !!monc->want_next_osdmap + !!monc->want_mdsmap;
+		ceph_encode_32(&p, num);
+
+		if (monc->want_next_osdmap) {
+			dout("__send_subscribe to 'osdmap' %u\n",
+			     (unsigned)monc->have_osdmap);
+			ceph_encode_string(&p, end, "osdmap", 6);
+			i = p;
+			i->have = cpu_to_le64(monc->have_osdmap);
+			i->onetime = 1;
+			p += sizeof(*i);
+			monc->want_next_osdmap = 2;  /* requested */
+		}
+		if (monc->want_mdsmap) {
+			dout("__send_subscribe to 'mdsmap' %u+\n",
+			     (unsigned)monc->have_mdsmap);
+			ceph_encode_string(&p, end, "mdsmap", 6);
+			i = p;
+			i->have = cpu_to_le64(monc->have_mdsmap);
+			i->onetime = 0;
+			p += sizeof(*i);
+		}
+		ceph_encode_string(&p, end, "monmap", 6);
+		i = p;
+		i->have = 0;
+		i->onetime = 0;
+		p += sizeof(*i);
+
+		msg->front.iov_len = p - msg->front.iov_base;
+		msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
+		ceph_con_revoke(monc->con, msg);
+		ceph_con_send(monc->con, ceph_msg_get(msg));
+
+		monc->sub_sent = jiffies | 1;  /* never 0 */
+	}
+}
+
+static void handle_subscribe_ack(struct ceph_mon_client *monc,
+				 struct ceph_msg *msg)
+{
+	unsigned seconds;
+	struct ceph_mon_subscribe_ack *h = msg->front.iov_base;
+
+	if (msg->front.iov_len < sizeof(*h))
+		goto bad;
+	seconds = le32_to_cpu(h->duration);
+
+	mutex_lock(&monc->mutex);
+	if (monc->hunting) {
+		pr_info("mon%d %s session established\n",
+			monc->cur_mon,
+			ceph_pr_addr(&monc->con->peer_addr.in_addr));
+		monc->hunting = false;
+	}
+	dout("handle_subscribe_ack after %d seconds\n", seconds);
+	monc->sub_renew_after = monc->sub_sent + (seconds >> 1)*HZ - 1;
+	monc->sub_sent = 0;
+	mutex_unlock(&monc->mutex);
+	return;
+bad:
+	pr_err("got corrupt subscribe-ack msg\n");
+	ceph_msg_dump(msg);
+}
+
+/*
+ * Keep track of which maps we have
+ */
+int ceph_monc_got_mdsmap(struct ceph_mon_client *monc, u32 got)
+{
+	mutex_lock(&monc->mutex);
+	monc->have_mdsmap = got;
+	mutex_unlock(&monc->mutex);
+	return 0;
+}
+EXPORT_SYMBOL(ceph_monc_got_mdsmap);
+
+int ceph_monc_got_osdmap(struct ceph_mon_client *monc, u32 got)
+{
+	mutex_lock(&monc->mutex);
+	monc->have_osdmap = got;
+	monc->want_next_osdmap = 0;
+	mutex_unlock(&monc->mutex);
+	return 0;
+}
+
+/*
+ * Register interest in the next osdmap
+ */
+void ceph_monc_request_next_osdmap(struct ceph_mon_client *monc)
+{
+	dout("request_next_osdmap have %u\n", monc->have_osdmap);
+	mutex_lock(&monc->mutex);
+	if (!monc->want_next_osdmap)
+		monc->want_next_osdmap = 1;
+	if (monc->want_next_osdmap < 2)
+		__send_subscribe(monc);
+	mutex_unlock(&monc->mutex);
+}
+
+/*
+ *
+ */
+int ceph_monc_open_session(struct ceph_mon_client *monc)
+{
+	if (!monc->con) {
+		monc->con = kmalloc(sizeof(*monc->con), GFP_KERNEL);
+		if (!monc->con)
+			return -ENOMEM;
+		ceph_con_init(monc->client->msgr, monc->con);
+		monc->con->private = monc;
+		monc->con->ops = &mon_con_ops;
+	}
+
+	mutex_lock(&monc->mutex);
+	__open_session(monc);
+	__schedule_delayed(monc);
+	mutex_unlock(&monc->mutex);
+	return 0;
+}
+EXPORT_SYMBOL(ceph_monc_open_session);
+
+/*
+ * The monitor responds with mount ack indicate mount success.  The
+ * included client ticket allows the client to talk to MDSs and OSDs.
+ */
+static void ceph_monc_handle_map(struct ceph_mon_client *monc,
+				 struct ceph_msg *msg)
+{
+	struct ceph_client *client = monc->client;
+	struct ceph_monmap *monmap = NULL, *old = monc->monmap;
+	void *p, *end;
+
+	mutex_lock(&monc->mutex);
+
+	dout("handle_monmap\n");
+	p = msg->front.iov_base;
+	end = p + msg->front.iov_len;
+
+	monmap = ceph_monmap_decode(p, end);
+	if (IS_ERR(monmap)) {
+		pr_err("problem decoding monmap, %d\n",
+		       (int)PTR_ERR(monmap));
+		goto out;
+	}
+
+	if (ceph_check_fsid(monc->client, &monmap->fsid) < 0) {
+		kfree(monmap);
+		goto out;
+	}
+
+	client->monc.monmap = monmap;
+	kfree(old);
+
+out:
+	mutex_unlock(&monc->mutex);
+	wake_up_all(&client->auth_wq);
+}
+
+/*
+ * generic requests (e.g., statfs, poolop)
+ */
+static struct ceph_mon_generic_request *__lookup_generic_req(
+	struct ceph_mon_client *monc, u64 tid)
+{
+	struct ceph_mon_generic_request *req;
+	struct rb_node *n = monc->generic_request_tree.rb_node;
+
+	while (n) {
+		req = rb_entry(n, struct ceph_mon_generic_request, node);
+		if (tid < req->tid)
+			n = n->rb_left;
+		else if (tid > req->tid)
+			n = n->rb_right;
+		else
+			return req;
+	}
+	return NULL;
+}
+
+static void __insert_generic_request(struct ceph_mon_client *monc,
+			    struct ceph_mon_generic_request *new)
+{
+	struct rb_node **p = &monc->generic_request_tree.rb_node;
+	struct rb_node *parent = NULL;
+	struct ceph_mon_generic_request *req = NULL;
+
+	while (*p) {
+		parent = *p;
+		req = rb_entry(parent, struct ceph_mon_generic_request, node);
+		if (new->tid < req->tid)
+			p = &(*p)->rb_left;
+		else if (new->tid > req->tid)
+			p = &(*p)->rb_right;
+		else
+			BUG();
+	}
+
+	rb_link_node(&new->node, parent, p);
+	rb_insert_color(&new->node, &monc->generic_request_tree);
+}
+
+static void release_generic_request(struct kref *kref)
+{
+	struct ceph_mon_generic_request *req =
+		container_of(kref, struct ceph_mon_generic_request, kref);
+
+	if (req->reply)
+		ceph_msg_put(req->reply);
+	if (req->request)
+		ceph_msg_put(req->request);
+
+	kfree(req);
+}
+
+static void put_generic_request(struct ceph_mon_generic_request *req)
+{
+	kref_put(&req->kref, release_generic_request);
+}
+
+static void get_generic_request(struct ceph_mon_generic_request *req)
+{
+	kref_get(&req->kref);
+}
+
+static struct ceph_msg *get_generic_reply(struct ceph_connection *con,
+					 struct ceph_msg_header *hdr,
+					 int *skip)
+{
+	struct ceph_mon_client *monc = con->private;
+	struct ceph_mon_generic_request *req;
+	u64 tid = le64_to_cpu(hdr->tid);
+	struct ceph_msg *m;
+
+	mutex_lock(&monc->mutex);
+	req = __lookup_generic_req(monc, tid);
+	if (!req) {
+		dout("get_generic_reply %lld dne\n", tid);
+		*skip = 1;
+		m = NULL;
+	} else {
+		dout("get_generic_reply %lld got %p\n", tid, req->reply);
+		m = ceph_msg_get(req->reply);
+		/*
+		 * we don't need to track the connection reading into
+		 * this reply because we only have one open connection
+		 * at a time, ever.
+		 */
+	}
+	mutex_unlock(&monc->mutex);
+	return m;
+}
+
+static int do_generic_request(struct ceph_mon_client *monc,
+			      struct ceph_mon_generic_request *req)
+{
+	int err;
+
+	/* register request */
+	mutex_lock(&monc->mutex);
+	req->tid = ++monc->last_tid;
+	req->request->hdr.tid = cpu_to_le64(req->tid);
+	__insert_generic_request(monc, req);
+	monc->num_generic_requests++;
+	ceph_con_send(monc->con, ceph_msg_get(req->request));
+	mutex_unlock(&monc->mutex);
+
+	err = wait_for_completion_interruptible(&req->completion);
+
+	mutex_lock(&monc->mutex);
+	rb_erase(&req->node, &monc->generic_request_tree);
+	monc->num_generic_requests--;
+	mutex_unlock(&monc->mutex);
+
+	if (!err)
+		err = req->result;
+	return err;
+}
+
+/*
+ * statfs
+ */
+static void handle_statfs_reply(struct ceph_mon_client *monc,
+				struct ceph_msg *msg)
+{
+	struct ceph_mon_generic_request *req;
+	struct ceph_mon_statfs_reply *reply = msg->front.iov_base;
+	u64 tid = le64_to_cpu(msg->hdr.tid);
+
+	if (msg->front.iov_len != sizeof(*reply))
+		goto bad;
+	dout("handle_statfs_reply %p tid %llu\n", msg, tid);
+
+	mutex_lock(&monc->mutex);
+	req = __lookup_generic_req(monc, tid);
+	if (req) {
+		*(struct ceph_statfs *)req->buf = reply->st;
+		req->result = 0;
+		get_generic_request(req);
+	}
+	mutex_unlock(&monc->mutex);
+	if (req) {
+		complete_all(&req->completion);
+		put_generic_request(req);
+	}
+	return;
+
+bad:
+	pr_err("corrupt generic reply, tid %llu\n", tid);
+	ceph_msg_dump(msg);
+}
+
+/*
+ * Do a synchronous statfs().
+ */
+int ceph_monc_do_statfs(struct ceph_mon_client *monc, struct ceph_statfs *buf)
+{
+	struct ceph_mon_generic_request *req;
+	struct ceph_mon_statfs *h;
+	int err;
+
+	req = kzalloc(sizeof(*req), GFP_NOFS);
+	if (!req)
+		return -ENOMEM;
+
+	kref_init(&req->kref);
+	req->buf = buf;
+	req->buf_len = sizeof(*buf);
+	init_completion(&req->completion);
+
+	err = -ENOMEM;
+	req->request = ceph_msg_new(CEPH_MSG_STATFS, sizeof(*h), GFP_NOFS);
+	if (!req->request)
+		goto out;
+	req->reply = ceph_msg_new(CEPH_MSG_STATFS_REPLY, 1024, GFP_NOFS);
+	if (!req->reply)
+		goto out;
+
+	/* fill out request */
+	h = req->request->front.iov_base;
+	h->monhdr.have_version = 0;
+	h->monhdr.session_mon = cpu_to_le16(-1);
+	h->monhdr.session_mon_tid = 0;
+	h->fsid = monc->monmap->fsid;
+
+	err = do_generic_request(monc, req);
+
+out:
+	kref_put(&req->kref, release_generic_request);
+	return err;
+}
+EXPORT_SYMBOL(ceph_monc_do_statfs);
+
+/*
+ * pool ops
+ */
+static int get_poolop_reply_buf(const char *src, size_t src_len,
+				char *dst, size_t dst_len)
+{
+	u32 buf_len;
+
+	if (src_len != sizeof(u32) + dst_len)
+		return -EINVAL;
+
+	buf_len = le32_to_cpu(*(u32 *)src);
+	if (buf_len != dst_len)
+		return -EINVAL;
+
+	memcpy(dst, src + sizeof(u32), dst_len);
+	return 0;
+}
+
+static void handle_poolop_reply(struct ceph_mon_client *monc,
+				struct ceph_msg *msg)
+{
+	struct ceph_mon_generic_request *req;
+	struct ceph_mon_poolop_reply *reply = msg->front.iov_base;
+	u64 tid = le64_to_cpu(msg->hdr.tid);
+
+	if (msg->front.iov_len < sizeof(*reply))
+		goto bad;
+	dout("handle_poolop_reply %p tid %llu\n", msg, tid);
+
+	mutex_lock(&monc->mutex);
+	req = __lookup_generic_req(monc, tid);
+	if (req) {
+		if (req->buf_len &&
+		    get_poolop_reply_buf(msg->front.iov_base + sizeof(*reply),
+				     msg->front.iov_len - sizeof(*reply),
+				     req->buf, req->buf_len) < 0) {
+			mutex_unlock(&monc->mutex);
+			goto bad;
+		}
+		req->result = le32_to_cpu(reply->reply_code);
+		get_generic_request(req);
+	}
+	mutex_unlock(&monc->mutex);
+	if (req) {
+		complete(&req->completion);
+		put_generic_request(req);
+	}
+	return;
+
+bad:
+	pr_err("corrupt generic reply, tid %llu\n", tid);
+	ceph_msg_dump(msg);
+}
+
+/*
+ * Do a synchronous pool op.
+ */
+int ceph_monc_do_poolop(struct ceph_mon_client *monc, u32 op,
+			u32 pool, u64 snapid,
+			char *buf, int len)
+{
+	struct ceph_mon_generic_request *req;
+	struct ceph_mon_poolop *h;
+	int err;
+
+	req = kzalloc(sizeof(*req), GFP_NOFS);
+	if (!req)
+		return -ENOMEM;
+
+	kref_init(&req->kref);
+	req->buf = buf;
+	req->buf_len = len;
+	init_completion(&req->completion);
+
+	err = -ENOMEM;
+	req->request = ceph_msg_new(CEPH_MSG_POOLOP, sizeof(*h), GFP_NOFS);
+	if (!req->request)
+		goto out;
+	req->reply = ceph_msg_new(CEPH_MSG_POOLOP_REPLY, 1024, GFP_NOFS);
+	if (!req->reply)
+		goto out;
+
+	/* fill out request */
+	req->request->hdr.version = cpu_to_le16(2);
+	h = req->request->front.iov_base;
+	h->monhdr.have_version = 0;
+	h->monhdr.session_mon = cpu_to_le16(-1);
+	h->monhdr.session_mon_tid = 0;
+	h->fsid = monc->monmap->fsid;
+	h->pool = cpu_to_le32(pool);
+	h->op = cpu_to_le32(op);
+	h->auid = 0;
+	h->snapid = cpu_to_le64(snapid);
+	h->name_len = 0;
+
+	err = do_generic_request(monc, req);
+
+out:
+	kref_put(&req->kref, release_generic_request);
+	return err;
+}
+
+int ceph_monc_create_snapid(struct ceph_mon_client *monc,
+			    u32 pool, u64 *snapid)
+{
+	return ceph_monc_do_poolop(monc,  POOL_OP_CREATE_UNMANAGED_SNAP,
+				   pool, 0, (char *)snapid, sizeof(*snapid));
+
+}
+EXPORT_SYMBOL(ceph_monc_create_snapid);
+
+int ceph_monc_delete_snapid(struct ceph_mon_client *monc,
+			    u32 pool, u64 snapid)
+{
+	return ceph_monc_do_poolop(monc,  POOL_OP_CREATE_UNMANAGED_SNAP,
+				   pool, snapid, 0, 0);
+
+}
+
+/*
+ * Resend pending generic requests.
+ */
+static void __resend_generic_request(struct ceph_mon_client *monc)
+{
+	struct ceph_mon_generic_request *req;
+	struct rb_node *p;
+
+	for (p = rb_first(&monc->generic_request_tree); p; p = rb_next(p)) {
+		req = rb_entry(p, struct ceph_mon_generic_request, node);
+		ceph_con_revoke(monc->con, req->request);
+		ceph_con_send(monc->con, ceph_msg_get(req->request));
+	}
+}
+
+/*
+ * Delayed work.  If we haven't mounted yet, retry.  Otherwise,
+ * renew/retry subscription as needed (in case it is timing out, or we
+ * got an ENOMEM).  And keep the monitor connection alive.
+ */
+static void delayed_work(struct work_struct *work)
+{
+	struct ceph_mon_client *monc =
+		container_of(work, struct ceph_mon_client, delayed_work.work);
+
+	dout("monc delayed_work\n");
+	mutex_lock(&monc->mutex);
+	if (monc->hunting) {
+		__close_session(monc);
+		__open_session(monc);  /* continue hunting */
+	} else {
+		ceph_con_keepalive(monc->con);
+
+		__validate_auth(monc);
+
+		if (monc->auth->ops->is_authenticated(monc->auth))
+			__send_subscribe(monc);
+	}
+	__schedule_delayed(monc);
+	mutex_unlock(&monc->mutex);
+}
+
+/*
+ * On startup, we build a temporary monmap populated with the IPs
+ * provided by mount(2).
+ */
+static int build_initial_monmap(struct ceph_mon_client *monc)
+{
+	struct ceph_options *opt = monc->client->options;
+	struct ceph_entity_addr *mon_addr = opt->mon_addr;
+	int num_mon = opt->num_mon;
+	int i;
+
+	/* build initial monmap */
+	monc->monmap = kzalloc(sizeof(*monc->monmap) +
+			       num_mon*sizeof(monc->monmap->mon_inst[0]),
+			       GFP_KERNEL);
+	if (!monc->monmap)
+		return -ENOMEM;
+	for (i = 0; i < num_mon; i++) {
+		monc->monmap->mon_inst[i].addr = mon_addr[i];
+		monc->monmap->mon_inst[i].addr.nonce = 0;
+		monc->monmap->mon_inst[i].name.type =
+			CEPH_ENTITY_TYPE_MON;
+		monc->monmap->mon_inst[i].name.num = cpu_to_le64(i);
+	}
+	monc->monmap->num_mon = num_mon;
+	monc->have_fsid = false;
+	return 0;
+}
+
+int ceph_monc_init(struct ceph_mon_client *monc, struct ceph_client *cl)
+{
+	int err = 0;
+
+	dout("init\n");
+	memset(monc, 0, sizeof(*monc));
+	monc->client = cl;
+	monc->monmap = NULL;
+	mutex_init(&monc->mutex);
+
+	err = build_initial_monmap(monc);
+	if (err)
+		goto out;
+
+	monc->con = NULL;
+
+	/* authentication */
+	monc->auth = ceph_auth_init(cl->options->name,
+				    cl->options->secret);
+	if (IS_ERR(monc->auth))
+		return PTR_ERR(monc->auth);
+	monc->auth->want_keys =
+		CEPH_ENTITY_TYPE_AUTH | CEPH_ENTITY_TYPE_MON |
+		CEPH_ENTITY_TYPE_OSD | CEPH_ENTITY_TYPE_MDS;
+
+	/* msgs */
+	err = -ENOMEM;
+	monc->m_subscribe_ack = ceph_msg_new(CEPH_MSG_MON_SUBSCRIBE_ACK,
+				     sizeof(struct ceph_mon_subscribe_ack),
+				     GFP_NOFS);
+	if (!monc->m_subscribe_ack)
+		goto out_monmap;
+
+	monc->m_subscribe = ceph_msg_new(CEPH_MSG_MON_SUBSCRIBE, 96, GFP_NOFS);
+	if (!monc->m_subscribe)
+		goto out_subscribe_ack;
+
+	monc->m_auth_reply = ceph_msg_new(CEPH_MSG_AUTH_REPLY, 4096, GFP_NOFS);
+	if (!monc->m_auth_reply)
+		goto out_subscribe;
+
+	monc->m_auth = ceph_msg_new(CEPH_MSG_AUTH, 4096, GFP_NOFS);
+	monc->pending_auth = 0;
+	if (!monc->m_auth)
+		goto out_auth_reply;
+
+	monc->cur_mon = -1;
+	monc->hunting = true;
+	monc->sub_renew_after = jiffies;
+	monc->sub_sent = 0;
+
+	INIT_DELAYED_WORK(&monc->delayed_work, delayed_work);
+	monc->generic_request_tree = RB_ROOT;
+	monc->num_generic_requests = 0;
+	monc->last_tid = 0;
+
+	monc->have_mdsmap = 0;
+	monc->have_osdmap = 0;
+	monc->want_next_osdmap = 1;
+	return 0;
+
+out_auth_reply:
+	ceph_msg_put(monc->m_auth_reply);
+out_subscribe:
+	ceph_msg_put(monc->m_subscribe);
+out_subscribe_ack:
+	ceph_msg_put(monc->m_subscribe_ack);
+out_monmap:
+	kfree(monc->monmap);
+out:
+	return err;
+}
+EXPORT_SYMBOL(ceph_monc_init);
+
+void ceph_monc_stop(struct ceph_mon_client *monc)
+{
+	dout("stop\n");
+	cancel_delayed_work_sync(&monc->delayed_work);
+
+	mutex_lock(&monc->mutex);
+	__close_session(monc);
+	if (monc->con) {
+		monc->con->private = NULL;
+		monc->con->ops->put(monc->con);
+		monc->con = NULL;
+	}
+	mutex_unlock(&monc->mutex);
+
+	ceph_auth_destroy(monc->auth);
+
+	ceph_msg_put(monc->m_auth);
+	ceph_msg_put(monc->m_auth_reply);
+	ceph_msg_put(monc->m_subscribe);
+	ceph_msg_put(monc->m_subscribe_ack);
+
+	kfree(monc->monmap);
+}
+EXPORT_SYMBOL(ceph_monc_stop);
+
+static void handle_auth_reply(struct ceph_mon_client *monc,
+			      struct ceph_msg *msg)
+{
+	int ret;
+	int was_auth = 0;
+
+	mutex_lock(&monc->mutex);
+	if (monc->auth->ops)
+		was_auth = monc->auth->ops->is_authenticated(monc->auth);
+	monc->pending_auth = 0;
+	ret = ceph_handle_auth_reply(monc->auth, msg->front.iov_base,
+				     msg->front.iov_len,
+				     monc->m_auth->front.iov_base,
+				     monc->m_auth->front_max);
+	if (ret < 0) {
+		monc->client->auth_err = ret;
+		wake_up_all(&monc->client->auth_wq);
+	} else if (ret > 0) {
+		__send_prepared_auth_request(monc, ret);
+	} else if (!was_auth && monc->auth->ops->is_authenticated(monc->auth)) {
+		dout("authenticated, starting session\n");
+
+		monc->client->msgr->inst.name.type = CEPH_ENTITY_TYPE_CLIENT;
+		monc->client->msgr->inst.name.num =
+					cpu_to_le64(monc->auth->global_id);
+
+		__send_subscribe(monc);
+		__resend_generic_request(monc);
+	}
+	mutex_unlock(&monc->mutex);
+}
+
+static int __validate_auth(struct ceph_mon_client *monc)
+{
+	int ret;
+
+	if (monc->pending_auth)
+		return 0;
+
+	ret = ceph_build_auth(monc->auth, monc->m_auth->front.iov_base,
+			      monc->m_auth->front_max);
+	if (ret <= 0)
+		return ret; /* either an error, or no need to authenticate */
+	__send_prepared_auth_request(monc, ret);
+	return 0;
+}
+
+int ceph_monc_validate_auth(struct ceph_mon_client *monc)
+{
+	int ret;
+
+	mutex_lock(&monc->mutex);
+	ret = __validate_auth(monc);
+	mutex_unlock(&monc->mutex);
+	return ret;
+}
+EXPORT_SYMBOL(ceph_monc_validate_auth);
+
+/*
+ * handle incoming message
+ */
+static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
+{
+	struct ceph_mon_client *monc = con->private;
+	int type = le16_to_cpu(msg->hdr.type);
+
+	if (!monc)
+		return;
+
+	switch (type) {
+	case CEPH_MSG_AUTH_REPLY:
+		handle_auth_reply(monc, msg);
+		break;
+
+	case CEPH_MSG_MON_SUBSCRIBE_ACK:
+		handle_subscribe_ack(monc, msg);
+		break;
+
+	case CEPH_MSG_STATFS_REPLY:
+		handle_statfs_reply(monc, msg);
+		break;
+
+	case CEPH_MSG_POOLOP_REPLY:
+		handle_poolop_reply(monc, msg);
+		break;
+
+	case CEPH_MSG_MON_MAP:
+		ceph_monc_handle_map(monc, msg);
+		break;
+
+	case CEPH_MSG_OSD_MAP:
+		ceph_osdc_handle_map(&monc->client->osdc, msg);
+		break;
+
+	default:
+		/* can the chained handler handle it? */
+		if (monc->client->extra_mon_dispatch &&
+		    monc->client->extra_mon_dispatch(monc->client, msg) == 0)
+			break;
+			
+		pr_err("received unknown message type %d %s\n", type,
+		       ceph_msg_type_name(type));
+	}
+	ceph_msg_put(msg);
+}
+
+/*
+ * Allocate memory for incoming message
+ */
+static struct ceph_msg *mon_alloc_msg(struct ceph_connection *con,
+				      struct ceph_msg_header *hdr,
+				      int *skip)
+{
+	struct ceph_mon_client *monc = con->private;
+	int type = le16_to_cpu(hdr->type);
+	int front_len = le32_to_cpu(hdr->front_len);
+	struct ceph_msg *m = NULL;
+
+	*skip = 0;
+
+	switch (type) {
+	case CEPH_MSG_MON_SUBSCRIBE_ACK:
+		m = ceph_msg_get(monc->m_subscribe_ack);
+		break;
+	case CEPH_MSG_POOLOP_REPLY:
+	case CEPH_MSG_STATFS_REPLY:
+		return get_generic_reply(con, hdr, skip);
+	case CEPH_MSG_AUTH_REPLY:
+		m = ceph_msg_get(monc->m_auth_reply);
+		break;
+	case CEPH_MSG_MON_MAP:
+	case CEPH_MSG_MDS_MAP:
+	case CEPH_MSG_OSD_MAP:
+		m = ceph_msg_new(type, front_len, GFP_NOFS);
+		break;
+	}
+
+	if (!m) {
+		pr_info("alloc_msg unknown type %d\n", type);
+		*skip = 1;
+	}
+	return m;
+}
+
+/*
+ * If the monitor connection resets, pick a new monitor and resubmit
+ * any pending requests.
+ */
+static void mon_fault(struct ceph_connection *con)
+{
+	struct ceph_mon_client *monc = con->private;
+
+	if (!monc)
+		return;
+
+	dout("mon_fault\n");
+	mutex_lock(&monc->mutex);
+	if (!con->private)
+		goto out;
+
+	if (monc->con && !monc->hunting)
+		pr_info("mon%d %s session lost, "
+			"hunting for new mon\n", monc->cur_mon,
+			ceph_pr_addr(&monc->con->peer_addr.in_addr));
+
+	__close_session(monc);
+	if (!monc->hunting) {
+		/* start hunting */
+		monc->hunting = true;
+		__open_session(monc);
+	} else {
+		/* already hunting, let's wait a bit */
+		__schedule_delayed(monc);
+	}
+out:
+	mutex_unlock(&monc->mutex);
+}
+
+static const struct ceph_connection_operations mon_con_ops = {
+	.get = ceph_con_get,
+	.put = ceph_con_put,
+	.dispatch = dispatch,
+	.fault = mon_fault,
+	.alloc_msg = mon_alloc_msg,
+};
diff --git a/net/ceph/msgpool.c b/net/ceph/msgpool.c
new file mode 100644
index 000000000000..d5f2d97ac05c
--- /dev/null
+++ b/net/ceph/msgpool.c
@@ -0,0 +1,64 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/err.h>
+#include <linux/sched.h>
+#include <linux/types.h>
+#include <linux/vmalloc.h>
+
+#include <linux/ceph/msgpool.h>
+
+static void *alloc_fn(gfp_t gfp_mask, void *arg)
+{
+	struct ceph_msgpool *pool = arg;
+	void *p;
+
+	p = ceph_msg_new(0, pool->front_len, gfp_mask);
+	if (!p)
+		pr_err("msgpool %s alloc failed\n", pool->name);
+	return p;
+}
+
+static void free_fn(void *element, void *arg)
+{
+	ceph_msg_put(element);
+}
+
+int ceph_msgpool_init(struct ceph_msgpool *pool,
+		      int front_len, int size, bool blocking, const char *name)
+{
+	pool->front_len = front_len;
+	pool->pool = mempool_create(size, alloc_fn, free_fn, pool);
+	if (!pool->pool)
+		return -ENOMEM;
+	pool->name = name;
+	return 0;
+}
+
+void ceph_msgpool_destroy(struct ceph_msgpool *pool)
+{
+	mempool_destroy(pool->pool);
+}
+
+struct ceph_msg *ceph_msgpool_get(struct ceph_msgpool *pool,
+				  int front_len)
+{
+	if (front_len > pool->front_len) {
+		pr_err("msgpool_get pool %s need front %d, pool size is %d\n",
+		       pool->name, front_len, pool->front_len);
+		WARN_ON(1);
+
+		/* try to alloc a fresh message */
+		return ceph_msg_new(0, front_len, GFP_NOFS);
+	}
+
+	return mempool_alloc(pool->pool, GFP_NOFS);
+}
+
+void ceph_msgpool_put(struct ceph_msgpool *pool, struct ceph_msg *msg)
+{
+	/* reset msg front_len; user may have changed it */
+	msg->front.iov_len = pool->front_len;
+	msg->hdr.front_len = cpu_to_le32(pool->front_len);
+
+	kref_init(&msg->kref);  /* retake single ref */
+}
diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
new file mode 100644
index 000000000000..79391994b3ed
--- /dev/null
+++ b/net/ceph/osd_client.c
@@ -0,0 +1,1773 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/highmem.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#ifdef CONFIG_BLOCK
+#include <linux/bio.h>
+#endif
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/osd_client.h>
+#include <linux/ceph/messenger.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/pagelist.h>
+
+#define OSD_OP_FRONT_LEN	4096
+#define OSD_OPREPLY_FRONT_LEN	512
+
+static const struct ceph_connection_operations osd_con_ops;
+static int __kick_requests(struct ceph_osd_client *osdc,
+			  struct ceph_osd *kickosd);
+
+static void kick_requests(struct ceph_osd_client *osdc, struct ceph_osd *osd);
+
+static int op_needs_trail(int op)
+{
+	switch (op) {
+	case CEPH_OSD_OP_GETXATTR:
+	case CEPH_OSD_OP_SETXATTR:
+	case CEPH_OSD_OP_CMPXATTR:
+	case CEPH_OSD_OP_CALL:
+		return 1;
+	default:
+		return 0;
+	}
+}
+
+static int op_has_extent(int op)
+{
+	return (op == CEPH_OSD_OP_READ ||
+		op == CEPH_OSD_OP_WRITE);
+}
+
+void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
+			struct ceph_file_layout *layout,
+			u64 snapid,
+			u64 off, u64 *plen, u64 *bno,
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op)
+{
+	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
+	u64 orig_len = *plen;
+	u64 objoff, objlen;    /* extent in object */
+
+	reqhead->snapid = cpu_to_le64(snapid);
+
+	/* object extent? */
+	ceph_calc_file_object_mapping(layout, off, plen, bno,
+				      &objoff, &objlen);
+	if (*plen < orig_len)
+		dout(" skipping last %llu, final file extent %llu~%llu\n",
+		     orig_len - *plen, off, *plen);
+
+	if (op_has_extent(op->op)) {
+		op->extent.offset = objoff;
+		op->extent.length = objlen;
+	}
+	req->r_num_pages = calc_pages_for(off, *plen);
+	if (op->op == CEPH_OSD_OP_WRITE)
+		op->payload_len = *plen;
+
+	dout("calc_layout bno=%llx %llu~%llu (%d pages)\n",
+	     *bno, objoff, objlen, req->r_num_pages);
+
+}
+EXPORT_SYMBOL(ceph_calc_raw_layout);
+
+/*
+ * Implement client access to distributed object storage cluster.
+ *
+ * All data objects are stored within a cluster/cloud of OSDs, or
+ * "object storage devices."  (Note that Ceph OSDs have _nothing_ to
+ * do with the T10 OSD extensions to SCSI.)  Ceph OSDs are simply
+ * remote daemons serving up and coordinating consistent and safe
+ * access to storage.
+ *
+ * Cluster membership and the mapping of data objects onto storage devices
+ * are described by the osd map.
+ *
+ * We keep track of pending OSD requests (read, write), resubmit
+ * requests to different OSDs when the cluster topology/data layout
+ * change, or retry the affected requests when the communications
+ * channel with an OSD is reset.
+ */
+
+/*
+ * calculate the mapping of a file extent onto an object, and fill out the
+ * request accordingly.  shorten extent as necessary if it crosses an
+ * object boundary.
+ *
+ * fill osd op in request message.
+ */
+static void calc_layout(struct ceph_osd_client *osdc,
+			struct ceph_vino vino,
+			struct ceph_file_layout *layout,
+			u64 off, u64 *plen,
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op)
+{
+	u64 bno;
+
+	ceph_calc_raw_layout(osdc, layout, vino.snap, off,
+			     plen, &bno, req, op);
+
+	sprintf(req->r_oid, "%llx.%08llx", vino.ino, bno);
+	req->r_oid_len = strlen(req->r_oid);
+}
+
+/*
+ * requests
+ */
+void ceph_osdc_release_request(struct kref *kref)
+{
+	struct ceph_osd_request *req = container_of(kref,
+						    struct ceph_osd_request,
+						    r_kref);
+
+	if (req->r_request)
+		ceph_msg_put(req->r_request);
+	if (req->r_reply)
+		ceph_msg_put(req->r_reply);
+	if (req->r_con_filling_msg) {
+		dout("release_request revoking pages %p from con %p\n",
+		     req->r_pages, req->r_con_filling_msg);
+		ceph_con_revoke_message(req->r_con_filling_msg,
+				      req->r_reply);
+		ceph_con_put(req->r_con_filling_msg);
+	}
+	if (req->r_own_pages)
+		ceph_release_page_vector(req->r_pages,
+					 req->r_num_pages);
+#ifdef CONFIG_BLOCK
+	if (req->r_bio)
+		bio_put(req->r_bio);
+#endif
+	ceph_put_snap_context(req->r_snapc);
+	if (req->r_trail) {
+		ceph_pagelist_release(req->r_trail);
+		kfree(req->r_trail);
+	}
+	if (req->r_mempool)
+		mempool_free(req, req->r_osdc->req_mempool);
+	else
+		kfree(req);
+}
+EXPORT_SYMBOL(ceph_osdc_release_request);
+
+static int get_num_ops(struct ceph_osd_req_op *ops, int *needs_trail)
+{
+	int i = 0;
+
+	if (needs_trail)
+		*needs_trail = 0;
+	while (ops[i].op) {
+		if (needs_trail && op_needs_trail(ops[i].op))
+			*needs_trail = 1;
+		i++;
+	}
+
+	return i;
+}
+
+struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
+					       int flags,
+					       struct ceph_snap_context *snapc,
+					       struct ceph_osd_req_op *ops,
+					       bool use_mempool,
+					       gfp_t gfp_flags,
+					       struct page **pages,
+					       struct bio *bio)
+{
+	struct ceph_osd_request *req;
+	struct ceph_msg *msg;
+	int needs_trail;
+	int num_op = get_num_ops(ops, &needs_trail);
+	size_t msg_size = sizeof(struct ceph_osd_request_head);
+
+	msg_size += num_op*sizeof(struct ceph_osd_op);
+
+	if (use_mempool) {
+		req = mempool_alloc(osdc->req_mempool, gfp_flags);
+		memset(req, 0, sizeof(*req));
+	} else {
+		req = kzalloc(sizeof(*req), gfp_flags);
+	}
+	if (req == NULL)
+		return NULL;
+
+	req->r_osdc = osdc;
+	req->r_mempool = use_mempool;
+
+	kref_init(&req->r_kref);
+	init_completion(&req->r_completion);
+	init_completion(&req->r_safe_completion);
+	INIT_LIST_HEAD(&req->r_unsafe_item);
+	req->r_flags = flags;
+
+	WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0);
+
+	/* create reply message */
+	if (use_mempool)
+		msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0);
+	else
+		msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY,
+				   OSD_OPREPLY_FRONT_LEN, gfp_flags);
+	if (!msg) {
+		ceph_osdc_put_request(req);
+		return NULL;
+	}
+	req->r_reply = msg;
+
+	/* allocate space for the trailing data */
+	if (needs_trail) {
+		req->r_trail = kmalloc(sizeof(struct ceph_pagelist), gfp_flags);
+		if (!req->r_trail) {
+			ceph_osdc_put_request(req);
+			return NULL;
+		}
+		ceph_pagelist_init(req->r_trail);
+	}
+	/* create request message; allow space for oid */
+	msg_size += 40;
+	if (snapc)
+		msg_size += sizeof(u64) * snapc->num_snaps;
+	if (use_mempool)
+		msg = ceph_msgpool_get(&osdc->msgpool_op, 0);
+	else
+		msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, gfp_flags);
+	if (!msg) {
+		ceph_osdc_put_request(req);
+		return NULL;
+	}
+
+	msg->hdr.type = cpu_to_le16(CEPH_MSG_OSD_OP);
+	memset(msg->front.iov_base, 0, msg->front.iov_len);
+
+	req->r_request = msg;
+	req->r_pages = pages;
+#ifdef CONFIG_BLOCK
+	if (bio) {
+		req->r_bio = bio;
+		bio_get(req->r_bio);
+	}
+#endif
+
+	return req;
+}
+EXPORT_SYMBOL(ceph_osdc_alloc_request);
+
+static void osd_req_encode_op(struct ceph_osd_request *req,
+			      struct ceph_osd_op *dst,
+			      struct ceph_osd_req_op *src)
+{
+	dst->op = cpu_to_le16(src->op);
+
+	switch (dst->op) {
+	case CEPH_OSD_OP_READ:
+	case CEPH_OSD_OP_WRITE:
+		dst->extent.offset =
+			cpu_to_le64(src->extent.offset);
+		dst->extent.length =
+			cpu_to_le64(src->extent.length);
+		dst->extent.truncate_size =
+			cpu_to_le64(src->extent.truncate_size);
+		dst->extent.truncate_seq =
+			cpu_to_le32(src->extent.truncate_seq);
+		break;
+
+	case CEPH_OSD_OP_GETXATTR:
+	case CEPH_OSD_OP_SETXATTR:
+	case CEPH_OSD_OP_CMPXATTR:
+		BUG_ON(!req->r_trail);
+
+		dst->xattr.name_len = cpu_to_le32(src->xattr.name_len);
+		dst->xattr.value_len = cpu_to_le32(src->xattr.value_len);
+		dst->xattr.cmp_op = src->xattr.cmp_op;
+		dst->xattr.cmp_mode = src->xattr.cmp_mode;
+		ceph_pagelist_append(req->r_trail, src->xattr.name,
+				     src->xattr.name_len);
+		ceph_pagelist_append(req->r_trail, src->xattr.val,
+				     src->xattr.value_len);
+		break;
+	case CEPH_OSD_OP_CALL:
+		BUG_ON(!req->r_trail);
+
+		dst->cls.class_len = src->cls.class_len;
+		dst->cls.method_len = src->cls.method_len;
+		dst->cls.indata_len = cpu_to_le32(src->cls.indata_len);
+
+		ceph_pagelist_append(req->r_trail, src->cls.class_name,
+				     src->cls.class_len);
+		ceph_pagelist_append(req->r_trail, src->cls.method_name,
+				     src->cls.method_len);
+		ceph_pagelist_append(req->r_trail, src->cls.indata,
+				     src->cls.indata_len);
+		break;
+	case CEPH_OSD_OP_ROLLBACK:
+		dst->snap.snapid = cpu_to_le64(src->snap.snapid);
+		break;
+	case CEPH_OSD_OP_STARTSYNC:
+		break;
+	default:
+		pr_err("unrecognized osd opcode %d\n", dst->op);
+		WARN_ON(1);
+		break;
+	}
+	dst->payload_len = cpu_to_le32(src->payload_len);
+}
+
+/*
+ * build new request AND message
+ *
+ */
+void ceph_osdc_build_request(struct ceph_osd_request *req,
+			     u64 off, u64 *plen,
+			     struct ceph_osd_req_op *src_ops,
+			     struct ceph_snap_context *snapc,
+			     struct timespec *mtime,
+			     const char *oid,
+			     int oid_len)
+{
+	struct ceph_msg *msg = req->r_request;
+	struct ceph_osd_request_head *head;
+	struct ceph_osd_req_op *src_op;
+	struct ceph_osd_op *op;
+	void *p;
+	int num_op = get_num_ops(src_ops, NULL);
+	size_t msg_size = sizeof(*head) + num_op*sizeof(*op);
+	int flags = req->r_flags;
+	u64 data_len = 0;
+	int i;
+
+	head = msg->front.iov_base;
+	op = (void *)(head + 1);
+	p = (void *)(op + num_op);
+
+	req->r_snapc = ceph_get_snap_context(snapc);
+
+	head->client_inc = cpu_to_le32(1); /* always, for now. */
+	head->flags = cpu_to_le32(flags);
+	if (flags & CEPH_OSD_FLAG_WRITE)
+		ceph_encode_timespec(&head->mtime, mtime);
+	head->num_ops = cpu_to_le16(num_op);
+
+
+	/* fill in oid */
+	head->object_len = cpu_to_le32(oid_len);
+	memcpy(p, oid, oid_len);
+	p += oid_len;
+
+	src_op = src_ops;
+	while (src_op->op) {
+		osd_req_encode_op(req, op, src_op);
+		src_op++;
+		op++;
+	}
+
+	if (req->r_trail)
+		data_len += req->r_trail->length;
+
+	if (snapc) {
+		head->snap_seq = cpu_to_le64(snapc->seq);
+		head->num_snaps = cpu_to_le32(snapc->num_snaps);
+		for (i = 0; i < snapc->num_snaps; i++) {
+			put_unaligned_le64(snapc->snaps[i], p);
+			p += sizeof(u64);
+		}
+	}
+
+	if (flags & CEPH_OSD_FLAG_WRITE) {
+		req->r_request->hdr.data_off = cpu_to_le16(off);
+		req->r_request->hdr.data_len = cpu_to_le32(*plen + data_len);
+	} else if (data_len) {
+		req->r_request->hdr.data_off = 0;
+		req->r_request->hdr.data_len = cpu_to_le32(data_len);
+	}
+
+	BUG_ON(p > msg->front.iov_base + msg->front.iov_len);
+	msg_size = p - msg->front.iov_base;
+	msg->front.iov_len = msg_size;
+	msg->hdr.front_len = cpu_to_le32(msg_size);
+	return;
+}
+EXPORT_SYMBOL(ceph_osdc_build_request);
+
+/*
+ * build new request AND message, calculate layout, and adjust file
+ * extent as needed.
+ *
+ * if the file was recently truncated, we include information about its
+ * old and new size so that the object can be updated appropriately.  (we
+ * avoid synchronously deleting truncated objects because it's slow.)
+ *
+ * if @do_sync, include a 'startsync' command so that the osd will flush
+ * data quickly.
+ */
+struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *osdc,
+					       struct ceph_file_layout *layout,
+					       struct ceph_vino vino,
+					       u64 off, u64 *plen,
+					       int opcode, int flags,
+					       struct ceph_snap_context *snapc,
+					       int do_sync,
+					       u32 truncate_seq,
+					       u64 truncate_size,
+					       struct timespec *mtime,
+					       bool use_mempool, int num_reply)
+{
+	struct ceph_osd_req_op ops[3];
+	struct ceph_osd_request *req;
+
+	ops[0].op = opcode;
+	ops[0].extent.truncate_seq = truncate_seq;
+	ops[0].extent.truncate_size = truncate_size;
+	ops[0].payload_len = 0;
+
+	if (do_sync) {
+		ops[1].op = CEPH_OSD_OP_STARTSYNC;
+		ops[1].payload_len = 0;
+		ops[2].op = 0;
+	} else
+		ops[1].op = 0;
+
+	req = ceph_osdc_alloc_request(osdc, flags,
+					 snapc, ops,
+					 use_mempool,
+					 GFP_NOFS, NULL, NULL);
+	if (IS_ERR(req))
+		return req;
+
+	/* calculate max write size */
+	calc_layout(osdc, vino, layout, off, plen, req, ops);
+	req->r_file_layout = *layout;  /* keep a copy */
+
+	ceph_osdc_build_request(req, off, plen, ops,
+				snapc,
+				mtime,
+				req->r_oid, req->r_oid_len);
+
+	return req;
+}
+EXPORT_SYMBOL(ceph_osdc_new_request);
+
+/*
+ * We keep osd requests in an rbtree, sorted by ->r_tid.
+ */
+static void __insert_request(struct ceph_osd_client *osdc,
+			     struct ceph_osd_request *new)
+{
+	struct rb_node **p = &osdc->requests.rb_node;
+	struct rb_node *parent = NULL;
+	struct ceph_osd_request *req = NULL;
+
+	while (*p) {
+		parent = *p;
+		req = rb_entry(parent, struct ceph_osd_request, r_node);
+		if (new->r_tid < req->r_tid)
+			p = &(*p)->rb_left;
+		else if (new->r_tid > req->r_tid)
+			p = &(*p)->rb_right;
+		else
+			BUG();
+	}
+
+	rb_link_node(&new->r_node, parent, p);
+	rb_insert_color(&new->r_node, &osdc->requests);
+}
+
+static struct ceph_osd_request *__lookup_request(struct ceph_osd_client *osdc,
+						 u64 tid)
+{
+	struct ceph_osd_request *req;
+	struct rb_node *n = osdc->requests.rb_node;
+
+	while (n) {
+		req = rb_entry(n, struct ceph_osd_request, r_node);
+		if (tid < req->r_tid)
+			n = n->rb_left;
+		else if (tid > req->r_tid)
+			n = n->rb_right;
+		else
+			return req;
+	}
+	return NULL;
+}
+
+static struct ceph_osd_request *
+__lookup_request_ge(struct ceph_osd_client *osdc,
+		    u64 tid)
+{
+	struct ceph_osd_request *req;
+	struct rb_node *n = osdc->requests.rb_node;
+
+	while (n) {
+		req = rb_entry(n, struct ceph_osd_request, r_node);
+		if (tid < req->r_tid) {
+			if (!n->rb_left)
+				return req;
+			n = n->rb_left;
+		} else if (tid > req->r_tid) {
+			n = n->rb_right;
+		} else {
+			return req;
+		}
+	}
+	return NULL;
+}
+
+
+/*
+ * If the osd connection drops, we need to resubmit all requests.
+ */
+static void osd_reset(struct ceph_connection *con)
+{
+	struct ceph_osd *osd = con->private;
+	struct ceph_osd_client *osdc;
+
+	if (!osd)
+		return;
+	dout("osd_reset osd%d\n", osd->o_osd);
+	osdc = osd->o_osdc;
+	down_read(&osdc->map_sem);
+	kick_requests(osdc, osd);
+	up_read(&osdc->map_sem);
+}
+
+/*
+ * Track open sessions with osds.
+ */
+static struct ceph_osd *create_osd(struct ceph_osd_client *osdc)
+{
+	struct ceph_osd *osd;
+
+	osd = kzalloc(sizeof(*osd), GFP_NOFS);
+	if (!osd)
+		return NULL;
+
+	atomic_set(&osd->o_ref, 1);
+	osd->o_osdc = osdc;
+	INIT_LIST_HEAD(&osd->o_requests);
+	INIT_LIST_HEAD(&osd->o_osd_lru);
+	osd->o_incarnation = 1;
+
+	ceph_con_init(osdc->client->msgr, &osd->o_con);
+	osd->o_con.private = osd;
+	osd->o_con.ops = &osd_con_ops;
+	osd->o_con.peer_name.type = CEPH_ENTITY_TYPE_OSD;
+
+	INIT_LIST_HEAD(&osd->o_keepalive_item);
+	return osd;
+}
+
+static struct ceph_osd *get_osd(struct ceph_osd *osd)
+{
+	if (atomic_inc_not_zero(&osd->o_ref)) {
+		dout("get_osd %p %d -> %d\n", osd, atomic_read(&osd->o_ref)-1,
+		     atomic_read(&osd->o_ref));
+		return osd;
+	} else {
+		dout("get_osd %p FAIL\n", osd);
+		return NULL;
+	}
+}
+
+static void put_osd(struct ceph_osd *osd)
+{
+	dout("put_osd %p %d -> %d\n", osd, atomic_read(&osd->o_ref),
+	     atomic_read(&osd->o_ref) - 1);
+	if (atomic_dec_and_test(&osd->o_ref)) {
+		struct ceph_auth_client *ac = osd->o_osdc->client->monc.auth;
+
+		if (osd->o_authorizer)
+			ac->ops->destroy_authorizer(ac, osd->o_authorizer);
+		kfree(osd);
+	}
+}
+
+/*
+ * remove an osd from our map
+ */
+static void __remove_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd)
+{
+	dout("__remove_osd %p\n", osd);
+	BUG_ON(!list_empty(&osd->o_requests));
+	rb_erase(&osd->o_node, &osdc->osds);
+	list_del_init(&osd->o_osd_lru);
+	ceph_con_close(&osd->o_con);
+	put_osd(osd);
+}
+
+static void __move_osd_to_lru(struct ceph_osd_client *osdc,
+			      struct ceph_osd *osd)
+{
+	dout("__move_osd_to_lru %p\n", osd);
+	BUG_ON(!list_empty(&osd->o_osd_lru));
+	list_add_tail(&osd->o_osd_lru, &osdc->osd_lru);
+	osd->lru_ttl = jiffies + osdc->client->options->osd_idle_ttl * HZ;
+}
+
+static void __remove_osd_from_lru(struct ceph_osd *osd)
+{
+	dout("__remove_osd_from_lru %p\n", osd);
+	if (!list_empty(&osd->o_osd_lru))
+		list_del_init(&osd->o_osd_lru);
+}
+
+static void remove_old_osds(struct ceph_osd_client *osdc, int remove_all)
+{
+	struct ceph_osd *osd, *nosd;
+
+	dout("__remove_old_osds %p\n", osdc);
+	mutex_lock(&osdc->request_mutex);
+	list_for_each_entry_safe(osd, nosd, &osdc->osd_lru, o_osd_lru) {
+		if (!remove_all && time_before(jiffies, osd->lru_ttl))
+			break;
+		__remove_osd(osdc, osd);
+	}
+	mutex_unlock(&osdc->request_mutex);
+}
+
+/*
+ * reset osd connect
+ */
+static int __reset_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd)
+{
+	struct ceph_osd_request *req;
+	int ret = 0;
+
+	dout("__reset_osd %p osd%d\n", osd, osd->o_osd);
+	if (list_empty(&osd->o_requests)) {
+		__remove_osd(osdc, osd);
+	} else if (memcmp(&osdc->osdmap->osd_addr[osd->o_osd],
+			  &osd->o_con.peer_addr,
+			  sizeof(osd->o_con.peer_addr)) == 0 &&
+		   !ceph_con_opened(&osd->o_con)) {
+		dout(" osd addr hasn't changed and connection never opened,"
+		     " letting msgr retry");
+		/* touch each r_stamp for handle_timeout()'s benfit */
+		list_for_each_entry(req, &osd->o_requests, r_osd_item)
+			req->r_stamp = jiffies;
+		ret = -EAGAIN;
+	} else {
+		ceph_con_close(&osd->o_con);
+		ceph_con_open(&osd->o_con, &osdc->osdmap->osd_addr[osd->o_osd]);
+		osd->o_incarnation++;
+	}
+	return ret;
+}
+
+static void __insert_osd(struct ceph_osd_client *osdc, struct ceph_osd *new)
+{
+	struct rb_node **p = &osdc->osds.rb_node;
+	struct rb_node *parent = NULL;
+	struct ceph_osd *osd = NULL;
+
+	while (*p) {
+		parent = *p;
+		osd = rb_entry(parent, struct ceph_osd, o_node);
+		if (new->o_osd < osd->o_osd)
+			p = &(*p)->rb_left;
+		else if (new->o_osd > osd->o_osd)
+			p = &(*p)->rb_right;
+		else
+			BUG();
+	}
+
+	rb_link_node(&new->o_node, parent, p);
+	rb_insert_color(&new->o_node, &osdc->osds);
+}
+
+static struct ceph_osd *__lookup_osd(struct ceph_osd_client *osdc, int o)
+{
+	struct ceph_osd *osd;
+	struct rb_node *n = osdc->osds.rb_node;
+
+	while (n) {
+		osd = rb_entry(n, struct ceph_osd, o_node);
+		if (o < osd->o_osd)
+			n = n->rb_left;
+		else if (o > osd->o_osd)
+			n = n->rb_right;
+		else
+			return osd;
+	}
+	return NULL;
+}
+
+static void __schedule_osd_timeout(struct ceph_osd_client *osdc)
+{
+	schedule_delayed_work(&osdc->timeout_work,
+			osdc->client->options->osd_keepalive_timeout * HZ);
+}
+
+static void __cancel_osd_timeout(struct ceph_osd_client *osdc)
+{
+	cancel_delayed_work(&osdc->timeout_work);
+}
+
+/*
+ * Register request, assign tid.  If this is the first request, set up
+ * the timeout event.
+ */
+static void register_request(struct ceph_osd_client *osdc,
+			     struct ceph_osd_request *req)
+{
+	mutex_lock(&osdc->request_mutex);
+	req->r_tid = ++osdc->last_tid;
+	req->r_request->hdr.tid = cpu_to_le64(req->r_tid);
+	INIT_LIST_HEAD(&req->r_req_lru_item);
+
+	dout("register_request %p tid %lld\n", req, req->r_tid);
+	__insert_request(osdc, req);
+	ceph_osdc_get_request(req);
+	osdc->num_requests++;
+
+	if (osdc->num_requests == 1) {
+		dout(" first request, scheduling timeout\n");
+		__schedule_osd_timeout(osdc);
+	}
+	mutex_unlock(&osdc->request_mutex);
+}
+
+/*
+ * called under osdc->request_mutex
+ */
+static void __unregister_request(struct ceph_osd_client *osdc,
+				 struct ceph_osd_request *req)
+{
+	dout("__unregister_request %p tid %lld\n", req, req->r_tid);
+	rb_erase(&req->r_node, &osdc->requests);
+	osdc->num_requests--;
+
+	if (req->r_osd) {
+		/* make sure the original request isn't in flight. */
+		ceph_con_revoke(&req->r_osd->o_con, req->r_request);
+
+		list_del_init(&req->r_osd_item);
+		if (list_empty(&req->r_osd->o_requests))
+			__move_osd_to_lru(osdc, req->r_osd);
+		req->r_osd = NULL;
+	}
+
+	ceph_osdc_put_request(req);
+
+	list_del_init(&req->r_req_lru_item);
+	if (osdc->num_requests == 0) {
+		dout(" no requests, canceling timeout\n");
+		__cancel_osd_timeout(osdc);
+	}
+}
+
+/*
+ * Cancel a previously queued request message
+ */
+static void __cancel_request(struct ceph_osd_request *req)
+{
+	if (req->r_sent && req->r_osd) {
+		ceph_con_revoke(&req->r_osd->o_con, req->r_request);
+		req->r_sent = 0;
+	}
+	list_del_init(&req->r_req_lru_item);
+}
+
+/*
+ * Pick an osd (the first 'up' osd in the pg), allocate the osd struct
+ * (as needed), and set the request r_osd appropriately.  If there is
+ * no up osd, set r_osd to NULL.
+ *
+ * Return 0 if unchanged, 1 if changed, or negative on error.
+ *
+ * Caller should hold map_sem for read and request_mutex.
+ */
+static int __map_osds(struct ceph_osd_client *osdc,
+		      struct ceph_osd_request *req)
+{
+	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
+	struct ceph_pg pgid;
+	int acting[CEPH_PG_MAX_SIZE];
+	int o = -1, num = 0;
+	int err;
+
+	dout("map_osds %p tid %lld\n", req, req->r_tid);
+	err = ceph_calc_object_layout(&reqhead->layout, req->r_oid,
+				      &req->r_file_layout, osdc->osdmap);
+	if (err)
+		return err;
+	pgid = reqhead->layout.ol_pgid;
+	req->r_pgid = pgid;
+
+	err = ceph_calc_pg_acting(osdc->osdmap, pgid, acting);
+	if (err > 0) {
+		o = acting[0];
+		num = err;
+	}
+
+	if ((req->r_osd && req->r_osd->o_osd == o &&
+	     req->r_sent >= req->r_osd->o_incarnation &&
+	     req->r_num_pg_osds == num &&
+	     memcmp(req->r_pg_osds, acting, sizeof(acting[0])*num) == 0) ||
+	    (req->r_osd == NULL && o == -1))
+		return 0;  /* no change */
+
+	dout("map_osds tid %llu pgid %d.%x osd%d (was osd%d)\n",
+	     req->r_tid, le32_to_cpu(pgid.pool), le16_to_cpu(pgid.ps), o,
+	     req->r_osd ? req->r_osd->o_osd : -1);
+
+	/* record full pg acting set */
+	memcpy(req->r_pg_osds, acting, sizeof(acting[0]) * num);
+	req->r_num_pg_osds = num;
+
+	if (req->r_osd) {
+		__cancel_request(req);
+		list_del_init(&req->r_osd_item);
+		req->r_osd = NULL;
+	}
+
+	req->r_osd = __lookup_osd(osdc, o);
+	if (!req->r_osd && o >= 0) {
+		err = -ENOMEM;
+		req->r_osd = create_osd(osdc);
+		if (!req->r_osd)
+			goto out;
+
+		dout("map_osds osd %p is osd%d\n", req->r_osd, o);
+		req->r_osd->o_osd = o;
+		req->r_osd->o_con.peer_name.num = cpu_to_le64(o);
+		__insert_osd(osdc, req->r_osd);
+
+		ceph_con_open(&req->r_osd->o_con, &osdc->osdmap->osd_addr[o]);
+	}
+
+	if (req->r_osd) {
+		__remove_osd_from_lru(req->r_osd);
+		list_add(&req->r_osd_item, &req->r_osd->o_requests);
+	}
+	err = 1;   /* osd or pg changed */
+
+out:
+	return err;
+}
+
+/*
+ * caller should hold map_sem (for read) and request_mutex
+ */
+static int __send_request(struct ceph_osd_client *osdc,
+			  struct ceph_osd_request *req)
+{
+	struct ceph_osd_request_head *reqhead;
+	int err;
+
+	err = __map_osds(osdc, req);
+	if (err < 0)
+		return err;
+	if (req->r_osd == NULL) {
+		dout("send_request %p no up osds in pg\n", req);
+		ceph_monc_request_next_osdmap(&osdc->client->monc);
+		return 0;
+	}
+
+	dout("send_request %p tid %llu to osd%d flags %d\n",
+	     req, req->r_tid, req->r_osd->o_osd, req->r_flags);
+
+	reqhead = req->r_request->front.iov_base;
+	reqhead->osdmap_epoch = cpu_to_le32(osdc->osdmap->epoch);
+	reqhead->flags |= cpu_to_le32(req->r_flags);  /* e.g., RETRY */
+	reqhead->reassert_version = req->r_reassert_version;
+
+	req->r_stamp = jiffies;
+	list_move_tail(&req->r_req_lru_item, &osdc->req_lru);
+
+	ceph_msg_get(req->r_request); /* send consumes a ref */
+	ceph_con_send(&req->r_osd->o_con, req->r_request);
+	req->r_sent = req->r_osd->o_incarnation;
+	return 0;
+}
+
+/*
+ * Timeout callback, called every N seconds when 1 or more osd
+ * requests has been active for more than N seconds.  When this
+ * happens, we ping all OSDs with requests who have timed out to
+ * ensure any communications channel reset is detected.  Reset the
+ * request timeouts another N seconds in the future as we go.
+ * Reschedule the timeout event another N seconds in future (unless
+ * there are no open requests).
+ */
+static void handle_timeout(struct work_struct *work)
+{
+	struct ceph_osd_client *osdc =
+		container_of(work, struct ceph_osd_client, timeout_work.work);
+	struct ceph_osd_request *req, *last_req = NULL;
+	struct ceph_osd *osd;
+	unsigned long timeout = osdc->client->options->osd_timeout * HZ;
+	unsigned long keepalive =
+		osdc->client->options->osd_keepalive_timeout * HZ;
+	unsigned long last_stamp = 0;
+	struct rb_node *p;
+	struct list_head slow_osds;
+
+	dout("timeout\n");
+	down_read(&osdc->map_sem);
+
+	ceph_monc_request_next_osdmap(&osdc->client->monc);
+
+	mutex_lock(&osdc->request_mutex);
+	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
+		req = rb_entry(p, struct ceph_osd_request, r_node);
+
+		if (req->r_resend) {
+			int err;
+
+			dout("osdc resending prev failed %lld\n", req->r_tid);
+			err = __send_request(osdc, req);
+			if (err)
+				dout("osdc failed again on %lld\n", req->r_tid);
+			else
+				req->r_resend = false;
+			continue;
+		}
+	}
+
+	/*
+	 * reset osds that appear to be _really_ unresponsive.  this
+	 * is a failsafe measure.. we really shouldn't be getting to
+	 * this point if the system is working properly.  the monitors
+	 * should mark the osd as failed and we should find out about
+	 * it from an updated osd map.
+	 */
+	while (timeout && !list_empty(&osdc->req_lru)) {
+		req = list_entry(osdc->req_lru.next, struct ceph_osd_request,
+				 r_req_lru_item);
+
+		if (time_before(jiffies, req->r_stamp + timeout))
+			break;
+
+		BUG_ON(req == last_req && req->r_stamp == last_stamp);
+		last_req = req;
+		last_stamp = req->r_stamp;
+
+		osd = req->r_osd;
+		BUG_ON(!osd);
+		pr_warning(" tid %llu timed out on osd%d, will reset osd\n",
+			   req->r_tid, osd->o_osd);
+		__kick_requests(osdc, osd);
+	}
+
+	/*
+	 * ping osds that are a bit slow.  this ensures that if there
+	 * is a break in the TCP connection we will notice, and reopen
+	 * a connection with that osd (from the fault callback).
+	 */
+	INIT_LIST_HEAD(&slow_osds);
+	list_for_each_entry(req, &osdc->req_lru, r_req_lru_item) {
+		if (time_before(jiffies, req->r_stamp + keepalive))
+			break;
+
+		osd = req->r_osd;
+		BUG_ON(!osd);
+		dout(" tid %llu is slow, will send keepalive on osd%d\n",
+		     req->r_tid, osd->o_osd);
+		list_move_tail(&osd->o_keepalive_item, &slow_osds);
+	}
+	while (!list_empty(&slow_osds)) {
+		osd = list_entry(slow_osds.next, struct ceph_osd,
+				 o_keepalive_item);
+		list_del_init(&osd->o_keepalive_item);
+		ceph_con_keepalive(&osd->o_con);
+	}
+
+	__schedule_osd_timeout(osdc);
+	mutex_unlock(&osdc->request_mutex);
+
+	up_read(&osdc->map_sem);
+}
+
+static void handle_osds_timeout(struct work_struct *work)
+{
+	struct ceph_osd_client *osdc =
+		container_of(work, struct ceph_osd_client,
+			     osds_timeout_work.work);
+	unsigned long delay =
+		osdc->client->options->osd_idle_ttl * HZ >> 2;
+
+	dout("osds timeout\n");
+	down_read(&osdc->map_sem);
+	remove_old_osds(osdc, 0);
+	up_read(&osdc->map_sem);
+
+	schedule_delayed_work(&osdc->osds_timeout_work,
+			      round_jiffies_relative(delay));
+}
+
+/*
+ * handle osd op reply.  either call the callback if it is specified,
+ * or do the completion to wake up the waiting thread.
+ */
+static void handle_reply(struct ceph_osd_client *osdc, struct ceph_msg *msg,
+			 struct ceph_connection *con)
+{
+	struct ceph_osd_reply_head *rhead = msg->front.iov_base;
+	struct ceph_osd_request *req;
+	u64 tid;
+	int numops, object_len, flags;
+	s32 result;
+
+	tid = le64_to_cpu(msg->hdr.tid);
+	if (msg->front.iov_len < sizeof(*rhead))
+		goto bad;
+	numops = le32_to_cpu(rhead->num_ops);
+	object_len = le32_to_cpu(rhead->object_len);
+	result = le32_to_cpu(rhead->result);
+	if (msg->front.iov_len != sizeof(*rhead) + object_len +
+	    numops * sizeof(struct ceph_osd_op))
+		goto bad;
+	dout("handle_reply %p tid %llu result %d\n", msg, tid, (int)result);
+
+	/* lookup */
+	mutex_lock(&osdc->request_mutex);
+	req = __lookup_request(osdc, tid);
+	if (req == NULL) {
+		dout("handle_reply tid %llu dne\n", tid);
+		mutex_unlock(&osdc->request_mutex);
+		return;
+	}
+	ceph_osdc_get_request(req);
+	flags = le32_to_cpu(rhead->flags);
+
+	/*
+	 * if this connection filled our message, drop our reference now, to
+	 * avoid a (safe but slower) revoke later.
+	 */
+	if (req->r_con_filling_msg == con && req->r_reply == msg) {
+		dout(" dropping con_filling_msg ref %p\n", con);
+		req->r_con_filling_msg = NULL;
+		ceph_con_put(con);
+	}
+
+	if (!req->r_got_reply) {
+		unsigned bytes;
+
+		req->r_result = le32_to_cpu(rhead->result);
+		bytes = le32_to_cpu(msg->hdr.data_len);
+		dout("handle_reply result %d bytes %d\n", req->r_result,
+		     bytes);
+		if (req->r_result == 0)
+			req->r_result = bytes;
+
+		/* in case this is a write and we need to replay, */
+		req->r_reassert_version = rhead->reassert_version;
+
+		req->r_got_reply = 1;
+	} else if ((flags & CEPH_OSD_FLAG_ONDISK) == 0) {
+		dout("handle_reply tid %llu dup ack\n", tid);
+		mutex_unlock(&osdc->request_mutex);
+		goto done;
+	}
+
+	dout("handle_reply tid %llu flags %d\n", tid, flags);
+
+	/* either this is a read, or we got the safe response */
+	if (result < 0 ||
+	    (flags & CEPH_OSD_FLAG_ONDISK) ||
+	    ((flags & CEPH_OSD_FLAG_WRITE) == 0))
+		__unregister_request(osdc, req);
+
+	mutex_unlock(&osdc->request_mutex);
+
+	if (req->r_callback)
+		req->r_callback(req, msg);
+	else
+		complete_all(&req->r_completion);
+
+	if (flags & CEPH_OSD_FLAG_ONDISK) {
+		if (req->r_safe_callback)
+			req->r_safe_callback(req, msg);
+		complete_all(&req->r_safe_completion);  /* fsync waiter */
+	}
+
+done:
+	ceph_osdc_put_request(req);
+	return;
+
+bad:
+	pr_err("corrupt osd_op_reply got %d %d expected %d\n",
+	       (int)msg->front.iov_len, le32_to_cpu(msg->hdr.front_len),
+	       (int)sizeof(*rhead));
+	ceph_msg_dump(msg);
+}
+
+
+static int __kick_requests(struct ceph_osd_client *osdc,
+			  struct ceph_osd *kickosd)
+{
+	struct ceph_osd_request *req;
+	struct rb_node *p, *n;
+	int needmap = 0;
+	int err;
+
+	dout("kick_requests osd%d\n", kickosd ? kickosd->o_osd : -1);
+	if (kickosd) {
+		err = __reset_osd(osdc, kickosd);
+		if (err == -EAGAIN)
+			return 1;
+	} else {
+		for (p = rb_first(&osdc->osds); p; p = n) {
+			struct ceph_osd *osd =
+				rb_entry(p, struct ceph_osd, o_node);
+
+			n = rb_next(p);
+			if (!ceph_osd_is_up(osdc->osdmap, osd->o_osd) ||
+			    memcmp(&osd->o_con.peer_addr,
+				   ceph_osd_addr(osdc->osdmap,
+						 osd->o_osd),
+				   sizeof(struct ceph_entity_addr)) != 0)
+				__reset_osd(osdc, osd);
+		}
+	}
+
+	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
+		req = rb_entry(p, struct ceph_osd_request, r_node);
+
+		if (req->r_resend) {
+			dout(" r_resend set on tid %llu\n", req->r_tid);
+			__cancel_request(req);
+			goto kick;
+		}
+		if (req->r_osd && kickosd == req->r_osd) {
+			__cancel_request(req);
+			goto kick;
+		}
+
+		err = __map_osds(osdc, req);
+		if (err == 0)
+			continue;  /* no change */
+		if (err < 0) {
+			/*
+			 * FIXME: really, we should set the request
+			 * error and fail if this isn't a 'nofail'
+			 * request, but that's a fair bit more
+			 * complicated to do.  So retry!
+			 */
+			dout(" setting r_resend on %llu\n", req->r_tid);
+			req->r_resend = true;
+			continue;
+		}
+		if (req->r_osd == NULL) {
+			dout("tid %llu maps to no valid osd\n", req->r_tid);
+			needmap++;  /* request a newer map */
+			continue;
+		}
+
+kick:
+		dout("kicking %p tid %llu osd%d\n", req, req->r_tid,
+		     req->r_osd ? req->r_osd->o_osd : -1);
+		req->r_flags |= CEPH_OSD_FLAG_RETRY;
+		err = __send_request(osdc, req);
+		if (err) {
+			dout(" setting r_resend on %llu\n", req->r_tid);
+			req->r_resend = true;
+		}
+	}
+
+	return needmap;
+}
+
+/*
+ * Resubmit osd requests whose osd or osd address has changed.  Request
+ * a new osd map if osds are down, or we are otherwise unable to determine
+ * how to direct a request.
+ *
+ * Close connections to down osds.
+ *
+ * If @who is specified, resubmit requests for that specific osd.
+ *
+ * Caller should hold map_sem for read and request_mutex.
+ */
+static void kick_requests(struct ceph_osd_client *osdc,
+			  struct ceph_osd *kickosd)
+{
+	int needmap;
+
+	mutex_lock(&osdc->request_mutex);
+	needmap = __kick_requests(osdc, kickosd);
+	mutex_unlock(&osdc->request_mutex);
+
+	if (needmap) {
+		dout("%d requests for down osds, need new map\n", needmap);
+		ceph_monc_request_next_osdmap(&osdc->client->monc);
+	}
+
+}
+/*
+ * Process updated osd map.
+ *
+ * The message contains any number of incremental and full maps, normally
+ * indicating some sort of topology change in the cluster.  Kick requests
+ * off to different OSDs as needed.
+ */
+void ceph_osdc_handle_map(struct ceph_osd_client *osdc, struct ceph_msg *msg)
+{
+	void *p, *end, *next;
+	u32 nr_maps, maplen;
+	u32 epoch;
+	struct ceph_osdmap *newmap = NULL, *oldmap;
+	int err;
+	struct ceph_fsid fsid;
+
+	dout("handle_map have %u\n", osdc->osdmap ? osdc->osdmap->epoch : 0);
+	p = msg->front.iov_base;
+	end = p + msg->front.iov_len;
+
+	/* verify fsid */
+	ceph_decode_need(&p, end, sizeof(fsid), bad);
+	ceph_decode_copy(&p, &fsid, sizeof(fsid));
+	if (ceph_check_fsid(osdc->client, &fsid) < 0)
+		return;
+
+	down_write(&osdc->map_sem);
+
+	/* incremental maps */
+	ceph_decode_32_safe(&p, end, nr_maps, bad);
+	dout(" %d inc maps\n", nr_maps);
+	while (nr_maps > 0) {
+		ceph_decode_need(&p, end, 2*sizeof(u32), bad);
+		epoch = ceph_decode_32(&p);
+		maplen = ceph_decode_32(&p);
+		ceph_decode_need(&p, end, maplen, bad);
+		next = p + maplen;
+		if (osdc->osdmap && osdc->osdmap->epoch+1 == epoch) {
+			dout("applying incremental map %u len %d\n",
+			     epoch, maplen);
+			newmap = osdmap_apply_incremental(&p, next,
+							  osdc->osdmap,
+							  osdc->client->msgr);
+			if (IS_ERR(newmap)) {
+				err = PTR_ERR(newmap);
+				goto bad;
+			}
+			BUG_ON(!newmap);
+			if (newmap != osdc->osdmap) {
+				ceph_osdmap_destroy(osdc->osdmap);
+				osdc->osdmap = newmap;
+			}
+		} else {
+			dout("ignoring incremental map %u len %d\n",
+			     epoch, maplen);
+		}
+		p = next;
+		nr_maps--;
+	}
+	if (newmap)
+		goto done;
+
+	/* full maps */
+	ceph_decode_32_safe(&p, end, nr_maps, bad);
+	dout(" %d full maps\n", nr_maps);
+	while (nr_maps) {
+		ceph_decode_need(&p, end, 2*sizeof(u32), bad);
+		epoch = ceph_decode_32(&p);
+		maplen = ceph_decode_32(&p);
+		ceph_decode_need(&p, end, maplen, bad);
+		if (nr_maps > 1) {
+			dout("skipping non-latest full map %u len %d\n",
+			     epoch, maplen);
+		} else if (osdc->osdmap && osdc->osdmap->epoch >= epoch) {
+			dout("skipping full map %u len %d, "
+			     "older than our %u\n", epoch, maplen,
+			     osdc->osdmap->epoch);
+		} else {
+			dout("taking full map %u len %d\n", epoch, maplen);
+			newmap = osdmap_decode(&p, p+maplen);
+			if (IS_ERR(newmap)) {
+				err = PTR_ERR(newmap);
+				goto bad;
+			}
+			BUG_ON(!newmap);
+			oldmap = osdc->osdmap;
+			osdc->osdmap = newmap;
+			if (oldmap)
+				ceph_osdmap_destroy(oldmap);
+		}
+		p += maplen;
+		nr_maps--;
+	}
+
+done:
+	downgrade_write(&osdc->map_sem);
+	ceph_monc_got_osdmap(&osdc->client->monc, osdc->osdmap->epoch);
+	if (newmap)
+		kick_requests(osdc, NULL);
+	up_read(&osdc->map_sem);
+	wake_up_all(&osdc->client->auth_wq);
+	return;
+
+bad:
+	pr_err("osdc handle_map corrupt msg\n");
+	ceph_msg_dump(msg);
+	up_write(&osdc->map_sem);
+	return;
+}
+
+/*
+ * Register request, send initial attempt.
+ */
+int ceph_osdc_start_request(struct ceph_osd_client *osdc,
+			    struct ceph_osd_request *req,
+			    bool nofail)
+{
+	int rc = 0;
+
+	req->r_request->pages = req->r_pages;
+	req->r_request->nr_pages = req->r_num_pages;
+#ifdef CONFIG_BLOCK
+	req->r_request->bio = req->r_bio;
+#endif
+	req->r_request->trail = req->r_trail;
+
+	register_request(osdc, req);
+
+	down_read(&osdc->map_sem);
+	mutex_lock(&osdc->request_mutex);
+	/*
+	 * a racing kick_requests() may have sent the message for us
+	 * while we dropped request_mutex above, so only send now if
+	 * the request still han't been touched yet.
+	 */
+	if (req->r_sent == 0) {
+		rc = __send_request(osdc, req);
+		if (rc) {
+			if (nofail) {
+				dout("osdc_start_request failed send, "
+				     " marking %lld\n", req->r_tid);
+				req->r_resend = true;
+				rc = 0;
+			} else {
+				__unregister_request(osdc, req);
+			}
+		}
+	}
+	mutex_unlock(&osdc->request_mutex);
+	up_read(&osdc->map_sem);
+	return rc;
+}
+EXPORT_SYMBOL(ceph_osdc_start_request);
+
+/*
+ * wait for a request to complete
+ */
+int ceph_osdc_wait_request(struct ceph_osd_client *osdc,
+			   struct ceph_osd_request *req)
+{
+	int rc;
+
+	rc = wait_for_completion_interruptible(&req->r_completion);
+	if (rc < 0) {
+		mutex_lock(&osdc->request_mutex);
+		__cancel_request(req);
+		__unregister_request(osdc, req);
+		mutex_unlock(&osdc->request_mutex);
+		dout("wait_request tid %llu canceled/timed out\n", req->r_tid);
+		return rc;
+	}
+
+	dout("wait_request tid %llu result %d\n", req->r_tid, req->r_result);
+	return req->r_result;
+}
+EXPORT_SYMBOL(ceph_osdc_wait_request);
+
+/*
+ * sync - wait for all in-flight requests to flush.  avoid starvation.
+ */
+void ceph_osdc_sync(struct ceph_osd_client *osdc)
+{
+	struct ceph_osd_request *req;
+	u64 last_tid, next_tid = 0;
+
+	mutex_lock(&osdc->request_mutex);
+	last_tid = osdc->last_tid;
+	while (1) {
+		req = __lookup_request_ge(osdc, next_tid);
+		if (!req)
+			break;
+		if (req->r_tid > last_tid)
+			break;
+
+		next_tid = req->r_tid + 1;
+		if ((req->r_flags & CEPH_OSD_FLAG_WRITE) == 0)
+			continue;
+
+		ceph_osdc_get_request(req);
+		mutex_unlock(&osdc->request_mutex);
+		dout("sync waiting on tid %llu (last is %llu)\n",
+		     req->r_tid, last_tid);
+		wait_for_completion(&req->r_safe_completion);
+		mutex_lock(&osdc->request_mutex);
+		ceph_osdc_put_request(req);
+	}
+	mutex_unlock(&osdc->request_mutex);
+	dout("sync done (thru tid %llu)\n", last_tid);
+}
+EXPORT_SYMBOL(ceph_osdc_sync);
+
+/*
+ * init, shutdown
+ */
+int ceph_osdc_init(struct ceph_osd_client *osdc, struct ceph_client *client)
+{
+	int err;
+
+	dout("init\n");
+	osdc->client = client;
+	osdc->osdmap = NULL;
+	init_rwsem(&osdc->map_sem);
+	init_completion(&osdc->map_waiters);
+	osdc->last_requested_map = 0;
+	mutex_init(&osdc->request_mutex);
+	osdc->last_tid = 0;
+	osdc->osds = RB_ROOT;
+	INIT_LIST_HEAD(&osdc->osd_lru);
+	osdc->requests = RB_ROOT;
+	INIT_LIST_HEAD(&osdc->req_lru);
+	osdc->num_requests = 0;
+	INIT_DELAYED_WORK(&osdc->timeout_work, handle_timeout);
+	INIT_DELAYED_WORK(&osdc->osds_timeout_work, handle_osds_timeout);
+
+	schedule_delayed_work(&osdc->osds_timeout_work,
+	   round_jiffies_relative(osdc->client->options->osd_idle_ttl * HZ));
+
+	err = -ENOMEM;
+	osdc->req_mempool = mempool_create_kmalloc_pool(10,
+					sizeof(struct ceph_osd_request));
+	if (!osdc->req_mempool)
+		goto out;
+
+	err = ceph_msgpool_init(&osdc->msgpool_op, OSD_OP_FRONT_LEN, 10, true,
+				"osd_op");
+	if (err < 0)
+		goto out_mempool;
+	err = ceph_msgpool_init(&osdc->msgpool_op_reply,
+				OSD_OPREPLY_FRONT_LEN, 10, true,
+				"osd_op_reply");
+	if (err < 0)
+		goto out_msgpool;
+	return 0;
+
+out_msgpool:
+	ceph_msgpool_destroy(&osdc->msgpool_op);
+out_mempool:
+	mempool_destroy(osdc->req_mempool);
+out:
+	return err;
+}
+EXPORT_SYMBOL(ceph_osdc_init);
+
+void ceph_osdc_stop(struct ceph_osd_client *osdc)
+{
+	cancel_delayed_work_sync(&osdc->timeout_work);
+	cancel_delayed_work_sync(&osdc->osds_timeout_work);
+	if (osdc->osdmap) {
+		ceph_osdmap_destroy(osdc->osdmap);
+		osdc->osdmap = NULL;
+	}
+	remove_old_osds(osdc, 1);
+	mempool_destroy(osdc->req_mempool);
+	ceph_msgpool_destroy(&osdc->msgpool_op);
+	ceph_msgpool_destroy(&osdc->msgpool_op_reply);
+}
+EXPORT_SYMBOL(ceph_osdc_stop);
+
+/*
+ * Read some contiguous pages.  If we cross a stripe boundary, shorten
+ * *plen.  Return number of bytes read, or error.
+ */
+int ceph_osdc_readpages(struct ceph_osd_client *osdc,
+			struct ceph_vino vino, struct ceph_file_layout *layout,
+			u64 off, u64 *plen,
+			u32 truncate_seq, u64 truncate_size,
+			struct page **pages, int num_pages)
+{
+	struct ceph_osd_request *req;
+	int rc = 0;
+
+	dout("readpages on ino %llx.%llx on %llu~%llu\n", vino.ino,
+	     vino.snap, off, *plen);
+	req = ceph_osdc_new_request(osdc, layout, vino, off, plen,
+				    CEPH_OSD_OP_READ, CEPH_OSD_FLAG_READ,
+				    NULL, 0, truncate_seq, truncate_size, NULL,
+				    false, 1);
+	if (!req)
+		return -ENOMEM;
+
+	/* it may be a short read due to an object boundary */
+	req->r_pages = pages;
+
+	dout("readpages  final extent is %llu~%llu (%d pages)\n",
+	     off, *plen, req->r_num_pages);
+
+	rc = ceph_osdc_start_request(osdc, req, false);
+	if (!rc)
+		rc = ceph_osdc_wait_request(osdc, req);
+
+	ceph_osdc_put_request(req);
+	dout("readpages result %d\n", rc);
+	return rc;
+}
+EXPORT_SYMBOL(ceph_osdc_readpages);
+
+/*
+ * do a synchronous write on N pages
+ */
+int ceph_osdc_writepages(struct ceph_osd_client *osdc, struct ceph_vino vino,
+			 struct ceph_file_layout *layout,
+			 struct ceph_snap_context *snapc,
+			 u64 off, u64 len,
+			 u32 truncate_seq, u64 truncate_size,
+			 struct timespec *mtime,
+			 struct page **pages, int num_pages,
+			 int flags, int do_sync, bool nofail)
+{
+	struct ceph_osd_request *req;
+	int rc = 0;
+
+	BUG_ON(vino.snap != CEPH_NOSNAP);
+	req = ceph_osdc_new_request(osdc, layout, vino, off, &len,
+				    CEPH_OSD_OP_WRITE,
+				    flags | CEPH_OSD_FLAG_ONDISK |
+					    CEPH_OSD_FLAG_WRITE,
+				    snapc, do_sync,
+				    truncate_seq, truncate_size, mtime,
+				    nofail, 1);
+	if (!req)
+		return -ENOMEM;
+
+	/* it may be a short write due to an object boundary */
+	req->r_pages = pages;
+	dout("writepages %llu~%llu (%d pages)\n", off, len,
+	     req->r_num_pages);
+
+	rc = ceph_osdc_start_request(osdc, req, nofail);
+	if (!rc)
+		rc = ceph_osdc_wait_request(osdc, req);
+
+	ceph_osdc_put_request(req);
+	if (rc == 0)
+		rc = len;
+	dout("writepages result %d\n", rc);
+	return rc;
+}
+EXPORT_SYMBOL(ceph_osdc_writepages);
+
+/*
+ * handle incoming message
+ */
+static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
+{
+	struct ceph_osd *osd = con->private;
+	struct ceph_osd_client *osdc;
+	int type = le16_to_cpu(msg->hdr.type);
+
+	if (!osd)
+		goto out;
+	osdc = osd->o_osdc;
+
+	switch (type) {
+	case CEPH_MSG_OSD_MAP:
+		ceph_osdc_handle_map(osdc, msg);
+		break;
+	case CEPH_MSG_OSD_OPREPLY:
+		handle_reply(osdc, msg, con);
+		break;
+
+	default:
+		pr_err("received unknown message type %d %s\n", type,
+		       ceph_msg_type_name(type));
+	}
+out:
+	ceph_msg_put(msg);
+}
+
+/*
+ * lookup and return message for incoming reply.  set up reply message
+ * pages.
+ */
+static struct ceph_msg *get_reply(struct ceph_connection *con,
+				  struct ceph_msg_header *hdr,
+				  int *skip)
+{
+	struct ceph_osd *osd = con->private;
+	struct ceph_osd_client *osdc = osd->o_osdc;
+	struct ceph_msg *m;
+	struct ceph_osd_request *req;
+	int front = le32_to_cpu(hdr->front_len);
+	int data_len = le32_to_cpu(hdr->data_len);
+	u64 tid;
+
+	tid = le64_to_cpu(hdr->tid);
+	mutex_lock(&osdc->request_mutex);
+	req = __lookup_request(osdc, tid);
+	if (!req) {
+		*skip = 1;
+		m = NULL;
+		pr_info("get_reply unknown tid %llu from osd%d\n", tid,
+			osd->o_osd);
+		goto out;
+	}
+
+	if (req->r_con_filling_msg) {
+		dout("get_reply revoking msg %p from old con %p\n",
+		     req->r_reply, req->r_con_filling_msg);
+		ceph_con_revoke_message(req->r_con_filling_msg, req->r_reply);
+		ceph_con_put(req->r_con_filling_msg);
+		req->r_con_filling_msg = NULL;
+	}
+
+	if (front > req->r_reply->front.iov_len) {
+		pr_warning("get_reply front %d > preallocated %d\n",
+			   front, (int)req->r_reply->front.iov_len);
+		m = ceph_msg_new(CEPH_MSG_OSD_OPREPLY, front, GFP_NOFS);
+		if (!m)
+			goto out;
+		ceph_msg_put(req->r_reply);
+		req->r_reply = m;
+	}
+	m = ceph_msg_get(req->r_reply);
+
+	if (data_len > 0) {
+		unsigned data_off = le16_to_cpu(hdr->data_off);
+		int want = calc_pages_for(data_off & ~PAGE_MASK, data_len);
+
+		if (unlikely(req->r_num_pages < want)) {
+			pr_warning("tid %lld reply %d > expected %d pages\n",
+				   tid, want, m->nr_pages);
+			*skip = 1;
+			ceph_msg_put(m);
+			m = NULL;
+			goto out;
+		}
+		m->pages = req->r_pages;
+		m->nr_pages = req->r_num_pages;
+#ifdef CONFIG_BLOCK
+		m->bio = req->r_bio;
+#endif
+	}
+	*skip = 0;
+	req->r_con_filling_msg = ceph_con_get(con);
+	dout("get_reply tid %lld %p\n", tid, m);
+
+out:
+	mutex_unlock(&osdc->request_mutex);
+	return m;
+
+}
+
+static struct ceph_msg *alloc_msg(struct ceph_connection *con,
+				  struct ceph_msg_header *hdr,
+				  int *skip)
+{
+	struct ceph_osd *osd = con->private;
+	int type = le16_to_cpu(hdr->type);
+	int front = le32_to_cpu(hdr->front_len);
+
+	switch (type) {
+	case CEPH_MSG_OSD_MAP:
+		return ceph_msg_new(type, front, GFP_NOFS);
+	case CEPH_MSG_OSD_OPREPLY:
+		return get_reply(con, hdr, skip);
+	default:
+		pr_info("alloc_msg unexpected msg type %d from osd%d\n", type,
+			osd->o_osd);
+		*skip = 1;
+		return NULL;
+	}
+}
+
+/*
+ * Wrappers to refcount containing ceph_osd struct
+ */
+static struct ceph_connection *get_osd_con(struct ceph_connection *con)
+{
+	struct ceph_osd *osd = con->private;
+	if (get_osd(osd))
+		return con;
+	return NULL;
+}
+
+static void put_osd_con(struct ceph_connection *con)
+{
+	struct ceph_osd *osd = con->private;
+	put_osd(osd);
+}
+
+/*
+ * authentication
+ */
+static int get_authorizer(struct ceph_connection *con,
+			  void **buf, int *len, int *proto,
+			  void **reply_buf, int *reply_len, int force_new)
+{
+	struct ceph_osd *o = con->private;
+	struct ceph_osd_client *osdc = o->o_osdc;
+	struct ceph_auth_client *ac = osdc->client->monc.auth;
+	int ret = 0;
+
+	if (force_new && o->o_authorizer) {
+		ac->ops->destroy_authorizer(ac, o->o_authorizer);
+		o->o_authorizer = NULL;
+	}
+	if (o->o_authorizer == NULL) {
+		ret = ac->ops->create_authorizer(
+			ac, CEPH_ENTITY_TYPE_OSD,
+			&o->o_authorizer,
+			&o->o_authorizer_buf,
+			&o->o_authorizer_buf_len,
+			&o->o_authorizer_reply_buf,
+			&o->o_authorizer_reply_buf_len);
+		if (ret)
+			return ret;
+	}
+
+	*proto = ac->protocol;
+	*buf = o->o_authorizer_buf;
+	*len = o->o_authorizer_buf_len;
+	*reply_buf = o->o_authorizer_reply_buf;
+	*reply_len = o->o_authorizer_reply_buf_len;
+	return 0;
+}
+
+
+static int verify_authorizer_reply(struct ceph_connection *con, int len)
+{
+	struct ceph_osd *o = con->private;
+	struct ceph_osd_client *osdc = o->o_osdc;
+	struct ceph_auth_client *ac = osdc->client->monc.auth;
+
+	return ac->ops->verify_authorizer_reply(ac, o->o_authorizer, len);
+}
+
+static int invalidate_authorizer(struct ceph_connection *con)
+{
+	struct ceph_osd *o = con->private;
+	struct ceph_osd_client *osdc = o->o_osdc;
+	struct ceph_auth_client *ac = osdc->client->monc.auth;
+
+	if (ac->ops->invalidate_authorizer)
+		ac->ops->invalidate_authorizer(ac, CEPH_ENTITY_TYPE_OSD);
+
+	return ceph_monc_validate_auth(&osdc->client->monc);
+}
+
+static const struct ceph_connection_operations osd_con_ops = {
+	.get = get_osd_con,
+	.put = put_osd_con,
+	.dispatch = dispatch,
+	.get_authorizer = get_authorizer,
+	.verify_authorizer_reply = verify_authorizer_reply,
+	.invalidate_authorizer = invalidate_authorizer,
+	.alloc_msg = alloc_msg,
+	.fault = osd_reset,
+};
diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c
new file mode 100644
index 000000000000..d73f3f6efa36
--- /dev/null
+++ b/net/ceph/osdmap.c
@@ -0,0 +1,1128 @@
+
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <asm/div64.h>
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/osdmap.h>
+#include <linux/ceph/decode.h>
+#include <linux/crush/hash.h>
+#include <linux/crush/mapper.h>
+
+char *ceph_osdmap_state_str(char *str, int len, int state)
+{
+	int flag = 0;
+
+	if (!len)
+		goto done;
+
+	*str = '\0';
+	if (state) {
+		if (state & CEPH_OSD_EXISTS) {
+			snprintf(str, len, "exists");
+			flag = 1;
+		}
+		if (state & CEPH_OSD_UP) {
+			snprintf(str, len, "%s%s%s", str, (flag ? ", " : ""),
+				 "up");
+			flag = 1;
+		}
+	} else {
+		snprintf(str, len, "doesn't exist");
+	}
+done:
+	return str;
+}
+
+/* maps */
+
+static int calc_bits_of(unsigned t)
+{
+	int b = 0;
+	while (t) {
+		t = t >> 1;
+		b++;
+	}
+	return b;
+}
+
+/*
+ * the foo_mask is the smallest value 2^n-1 that is >= foo.
+ */
+static void calc_pg_masks(struct ceph_pg_pool_info *pi)
+{
+	pi->pg_num_mask = (1 << calc_bits_of(le32_to_cpu(pi->v.pg_num)-1)) - 1;
+	pi->pgp_num_mask =
+		(1 << calc_bits_of(le32_to_cpu(pi->v.pgp_num)-1)) - 1;
+	pi->lpg_num_mask =
+		(1 << calc_bits_of(le32_to_cpu(pi->v.lpg_num)-1)) - 1;
+	pi->lpgp_num_mask =
+		(1 << calc_bits_of(le32_to_cpu(pi->v.lpgp_num)-1)) - 1;
+}
+
+/*
+ * decode crush map
+ */
+static int crush_decode_uniform_bucket(void **p, void *end,
+				       struct crush_bucket_uniform *b)
+{
+	dout("crush_decode_uniform_bucket %p to %p\n", *p, end);
+	ceph_decode_need(p, end, (1+b->h.size) * sizeof(u32), bad);
+	b->item_weight = ceph_decode_32(p);
+	return 0;
+bad:
+	return -EINVAL;
+}
+
+static int crush_decode_list_bucket(void **p, void *end,
+				    struct crush_bucket_list *b)
+{
+	int j;
+	dout("crush_decode_list_bucket %p to %p\n", *p, end);
+	b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
+	if (b->item_weights == NULL)
+		return -ENOMEM;
+	b->sum_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
+	if (b->sum_weights == NULL)
+		return -ENOMEM;
+	ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad);
+	for (j = 0; j < b->h.size; j++) {
+		b->item_weights[j] = ceph_decode_32(p);
+		b->sum_weights[j] = ceph_decode_32(p);
+	}
+	return 0;
+bad:
+	return -EINVAL;
+}
+
+static int crush_decode_tree_bucket(void **p, void *end,
+				    struct crush_bucket_tree *b)
+{
+	int j;
+	dout("crush_decode_tree_bucket %p to %p\n", *p, end);
+	ceph_decode_32_safe(p, end, b->num_nodes, bad);
+	b->node_weights = kcalloc(b->num_nodes, sizeof(u32), GFP_NOFS);
+	if (b->node_weights == NULL)
+		return -ENOMEM;
+	ceph_decode_need(p, end, b->num_nodes * sizeof(u32), bad);
+	for (j = 0; j < b->num_nodes; j++)
+		b->node_weights[j] = ceph_decode_32(p);
+	return 0;
+bad:
+	return -EINVAL;
+}
+
+static int crush_decode_straw_bucket(void **p, void *end,
+				     struct crush_bucket_straw *b)
+{
+	int j;
+	dout("crush_decode_straw_bucket %p to %p\n", *p, end);
+	b->item_weights = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
+	if (b->item_weights == NULL)
+		return -ENOMEM;
+	b->straws = kcalloc(b->h.size, sizeof(u32), GFP_NOFS);
+	if (b->straws == NULL)
+		return -ENOMEM;
+	ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad);
+	for (j = 0; j < b->h.size; j++) {
+		b->item_weights[j] = ceph_decode_32(p);
+		b->straws[j] = ceph_decode_32(p);
+	}
+	return 0;
+bad:
+	return -EINVAL;
+}
+
+static struct crush_map *crush_decode(void *pbyval, void *end)
+{
+	struct crush_map *c;
+	int err = -EINVAL;
+	int i, j;
+	void **p = &pbyval;
+	void *start = pbyval;
+	u32 magic;
+
+	dout("crush_decode %p to %p len %d\n", *p, end, (int)(end - *p));
+
+	c = kzalloc(sizeof(*c), GFP_NOFS);
+	if (c == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	ceph_decode_need(p, end, 4*sizeof(u32), bad);
+	magic = ceph_decode_32(p);
+	if (magic != CRUSH_MAGIC) {
+		pr_err("crush_decode magic %x != current %x\n",
+		       (unsigned)magic, (unsigned)CRUSH_MAGIC);
+		goto bad;
+	}
+	c->max_buckets = ceph_decode_32(p);
+	c->max_rules = ceph_decode_32(p);
+	c->max_devices = ceph_decode_32(p);
+
+	c->device_parents = kcalloc(c->max_devices, sizeof(u32), GFP_NOFS);
+	if (c->device_parents == NULL)
+		goto badmem;
+	c->bucket_parents = kcalloc(c->max_buckets, sizeof(u32), GFP_NOFS);
+	if (c->bucket_parents == NULL)
+		goto badmem;
+
+	c->buckets = kcalloc(c->max_buckets, sizeof(*c->buckets), GFP_NOFS);
+	if (c->buckets == NULL)
+		goto badmem;
+	c->rules = kcalloc(c->max_rules, sizeof(*c->rules), GFP_NOFS);
+	if (c->rules == NULL)
+		goto badmem;
+
+	/* buckets */
+	for (i = 0; i < c->max_buckets; i++) {
+		int size = 0;
+		u32 alg;
+		struct crush_bucket *b;
+
+		ceph_decode_32_safe(p, end, alg, bad);
+		if (alg == 0) {
+			c->buckets[i] = NULL;
+			continue;
+		}
+		dout("crush_decode bucket %d off %x %p to %p\n",
+		     i, (int)(*p-start), *p, end);
+
+		switch (alg) {
+		case CRUSH_BUCKET_UNIFORM:
+			size = sizeof(struct crush_bucket_uniform);
+			break;
+		case CRUSH_BUCKET_LIST:
+			size = sizeof(struct crush_bucket_list);
+			break;
+		case CRUSH_BUCKET_TREE:
+			size = sizeof(struct crush_bucket_tree);
+			break;
+		case CRUSH_BUCKET_STRAW:
+			size = sizeof(struct crush_bucket_straw);
+			break;
+		default:
+			err = -EINVAL;
+			goto bad;
+		}
+		BUG_ON(size == 0);
+		b = c->buckets[i] = kzalloc(size, GFP_NOFS);
+		if (b == NULL)
+			goto badmem;
+
+		ceph_decode_need(p, end, 4*sizeof(u32), bad);
+		b->id = ceph_decode_32(p);
+		b->type = ceph_decode_16(p);
+		b->alg = ceph_decode_8(p);
+		b->hash = ceph_decode_8(p);
+		b->weight = ceph_decode_32(p);
+		b->size = ceph_decode_32(p);
+
+		dout("crush_decode bucket size %d off %x %p to %p\n",
+		     b->size, (int)(*p-start), *p, end);
+
+		b->items = kcalloc(b->size, sizeof(__s32), GFP_NOFS);
+		if (b->items == NULL)
+			goto badmem;
+		b->perm = kcalloc(b->size, sizeof(u32), GFP_NOFS);
+		if (b->perm == NULL)
+			goto badmem;
+		b->perm_n = 0;
+
+		ceph_decode_need(p, end, b->size*sizeof(u32), bad);
+		for (j = 0; j < b->size; j++)
+			b->items[j] = ceph_decode_32(p);
+
+		switch (b->alg) {
+		case CRUSH_BUCKET_UNIFORM:
+			err = crush_decode_uniform_bucket(p, end,
+				  (struct crush_bucket_uniform *)b);
+			if (err < 0)
+				goto bad;
+			break;
+		case CRUSH_BUCKET_LIST:
+			err = crush_decode_list_bucket(p, end,
+			       (struct crush_bucket_list *)b);
+			if (err < 0)
+				goto bad;
+			break;
+		case CRUSH_BUCKET_TREE:
+			err = crush_decode_tree_bucket(p, end,
+				(struct crush_bucket_tree *)b);
+			if (err < 0)
+				goto bad;
+			break;
+		case CRUSH_BUCKET_STRAW:
+			err = crush_decode_straw_bucket(p, end,
+				(struct crush_bucket_straw *)b);
+			if (err < 0)
+				goto bad;
+			break;
+		}
+	}
+
+	/* rules */
+	dout("rule vec is %p\n", c->rules);
+	for (i = 0; i < c->max_rules; i++) {
+		u32 yes;
+		struct crush_rule *r;
+
+		ceph_decode_32_safe(p, end, yes, bad);
+		if (!yes) {
+			dout("crush_decode NO rule %d off %x %p to %p\n",
+			     i, (int)(*p-start), *p, end);
+			c->rules[i] = NULL;
+			continue;
+		}
+
+		dout("crush_decode rule %d off %x %p to %p\n",
+		     i, (int)(*p-start), *p, end);
+
+		/* len */
+		ceph_decode_32_safe(p, end, yes, bad);
+#if BITS_PER_LONG == 32
+		err = -EINVAL;
+		if (yes > ULONG_MAX / sizeof(struct crush_rule_step))
+			goto bad;
+#endif
+		r = c->rules[i] = kmalloc(sizeof(*r) +
+					  yes*sizeof(struct crush_rule_step),
+					  GFP_NOFS);
+		if (r == NULL)
+			goto badmem;
+		dout(" rule %d is at %p\n", i, r);
+		r->len = yes;
+		ceph_decode_copy_safe(p, end, &r->mask, 4, bad); /* 4 u8's */
+		ceph_decode_need(p, end, r->len*3*sizeof(u32), bad);
+		for (j = 0; j < r->len; j++) {
+			r->steps[j].op = ceph_decode_32(p);
+			r->steps[j].arg1 = ceph_decode_32(p);
+			r->steps[j].arg2 = ceph_decode_32(p);
+		}
+	}
+
+	/* ignore trailing name maps. */
+
+	dout("crush_decode success\n");
+	return c;
+
+badmem:
+	err = -ENOMEM;
+bad:
+	dout("crush_decode fail %d\n", err);
+	crush_destroy(c);
+	return ERR_PTR(err);
+}
+
+/*
+ * rbtree of pg_mapping for handling pg_temp (explicit mapping of pgid
+ * to a set of osds)
+ */
+static int pgid_cmp(struct ceph_pg l, struct ceph_pg r)
+{
+	u64 a = *(u64 *)&l;
+	u64 b = *(u64 *)&r;
+
+	if (a < b)
+		return -1;
+	if (a > b)
+		return 1;
+	return 0;
+}
+
+static int __insert_pg_mapping(struct ceph_pg_mapping *new,
+			       struct rb_root *root)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct ceph_pg_mapping *pg = NULL;
+	int c;
+
+	while (*p) {
+		parent = *p;
+		pg = rb_entry(parent, struct ceph_pg_mapping, node);
+		c = pgid_cmp(new->pgid, pg->pgid);
+		if (c < 0)
+			p = &(*p)->rb_left;
+		else if (c > 0)
+			p = &(*p)->rb_right;
+		else
+			return -EEXIST;
+	}
+
+	rb_link_node(&new->node, parent, p);
+	rb_insert_color(&new->node, root);
+	return 0;
+}
+
+static struct ceph_pg_mapping *__lookup_pg_mapping(struct rb_root *root,
+						   struct ceph_pg pgid)
+{
+	struct rb_node *n = root->rb_node;
+	struct ceph_pg_mapping *pg;
+	int c;
+
+	while (n) {
+		pg = rb_entry(n, struct ceph_pg_mapping, node);
+		c = pgid_cmp(pgid, pg->pgid);
+		if (c < 0)
+			n = n->rb_left;
+		else if (c > 0)
+			n = n->rb_right;
+		else
+			return pg;
+	}
+	return NULL;
+}
+
+/*
+ * rbtree of pg pool info
+ */
+static int __insert_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *new)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct ceph_pg_pool_info *pi = NULL;
+
+	while (*p) {
+		parent = *p;
+		pi = rb_entry(parent, struct ceph_pg_pool_info, node);
+		if (new->id < pi->id)
+			p = &(*p)->rb_left;
+		else if (new->id > pi->id)
+			p = &(*p)->rb_right;
+		else
+			return -EEXIST;
+	}
+
+	rb_link_node(&new->node, parent, p);
+	rb_insert_color(&new->node, root);
+	return 0;
+}
+
+static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id)
+{
+	struct ceph_pg_pool_info *pi;
+	struct rb_node *n = root->rb_node;
+
+	while (n) {
+		pi = rb_entry(n, struct ceph_pg_pool_info, node);
+		if (id < pi->id)
+			n = n->rb_left;
+		else if (id > pi->id)
+			n = n->rb_right;
+		else
+			return pi;
+	}
+	return NULL;
+}
+
+int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name)
+{
+	struct rb_node *rbp;
+
+	for (rbp = rb_first(&map->pg_pools); rbp; rbp = rb_next(rbp)) {
+		struct ceph_pg_pool_info *pi =
+			rb_entry(rbp, struct ceph_pg_pool_info, node);
+		if (pi->name && strcmp(pi->name, name) == 0)
+			return pi->id;
+	}
+	return -ENOENT;
+}
+EXPORT_SYMBOL(ceph_pg_poolid_by_name);
+
+static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi)
+{
+	rb_erase(&pi->node, root);
+	kfree(pi->name);
+	kfree(pi);
+}
+
+static int __decode_pool(void **p, void *end, struct ceph_pg_pool_info *pi)
+{
+	unsigned n, m;
+
+	ceph_decode_copy(p, &pi->v, sizeof(pi->v));
+	calc_pg_masks(pi);
+
+	/* num_snaps * snap_info_t */
+	n = le32_to_cpu(pi->v.num_snaps);
+	while (n--) {
+		ceph_decode_need(p, end, sizeof(u64) + 1 + sizeof(u64) +
+				 sizeof(struct ceph_timespec), bad);
+		*p += sizeof(u64) +       /* key */
+			1 + sizeof(u64) + /* u8, snapid */
+			sizeof(struct ceph_timespec);
+		m = ceph_decode_32(p);    /* snap name */
+		*p += m;
+	}
+
+	*p += le32_to_cpu(pi->v.num_removed_snap_intervals) * sizeof(u64) * 2;
+	return 0;
+
+bad:
+	return -EINVAL;
+}
+
+static int __decode_pool_names(void **p, void *end, struct ceph_osdmap *map)
+{
+	struct ceph_pg_pool_info *pi;
+	u32 num, len, pool;
+
+	ceph_decode_32_safe(p, end, num, bad);
+	dout(" %d pool names\n", num);
+	while (num--) {
+		ceph_decode_32_safe(p, end, pool, bad);
+		ceph_decode_32_safe(p, end, len, bad);
+		dout("  pool %d len %d\n", pool, len);
+		pi = __lookup_pg_pool(&map->pg_pools, pool);
+		if (pi) {
+			kfree(pi->name);
+			pi->name = kmalloc(len + 1, GFP_NOFS);
+			if (pi->name) {
+				memcpy(pi->name, *p, len);
+				pi->name[len] = '\0';
+				dout("  name is %s\n", pi->name);
+			}
+		}
+		*p += len;
+	}
+	return 0;
+
+bad:
+	return -EINVAL;
+}
+
+/*
+ * osd map
+ */
+void ceph_osdmap_destroy(struct ceph_osdmap *map)
+{
+	dout("osdmap_destroy %p\n", map);
+	if (map->crush)
+		crush_destroy(map->crush);
+	while (!RB_EMPTY_ROOT(&map->pg_temp)) {
+		struct ceph_pg_mapping *pg =
+			rb_entry(rb_first(&map->pg_temp),
+				 struct ceph_pg_mapping, node);
+		rb_erase(&pg->node, &map->pg_temp);
+		kfree(pg);
+	}
+	while (!RB_EMPTY_ROOT(&map->pg_pools)) {
+		struct ceph_pg_pool_info *pi =
+			rb_entry(rb_first(&map->pg_pools),
+				 struct ceph_pg_pool_info, node);
+		__remove_pg_pool(&map->pg_pools, pi);
+	}
+	kfree(map->osd_state);
+	kfree(map->osd_weight);
+	kfree(map->osd_addr);
+	kfree(map);
+}
+
+/*
+ * adjust max osd value.  reallocate arrays.
+ */
+static int osdmap_set_max_osd(struct ceph_osdmap *map, int max)
+{
+	u8 *state;
+	struct ceph_entity_addr *addr;
+	u32 *weight;
+
+	state = kcalloc(max, sizeof(*state), GFP_NOFS);
+	addr = kcalloc(max, sizeof(*addr), GFP_NOFS);
+	weight = kcalloc(max, sizeof(*weight), GFP_NOFS);
+	if (state == NULL || addr == NULL || weight == NULL) {
+		kfree(state);
+		kfree(addr);
+		kfree(weight);
+		return -ENOMEM;
+	}
+
+	/* copy old? */
+	if (map->osd_state) {
+		memcpy(state, map->osd_state, map->max_osd*sizeof(*state));
+		memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr));
+		memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight));
+		kfree(map->osd_state);
+		kfree(map->osd_addr);
+		kfree(map->osd_weight);
+	}
+
+	map->osd_state = state;
+	map->osd_weight = weight;
+	map->osd_addr = addr;
+	map->max_osd = max;
+	return 0;
+}
+
+/*
+ * decode a full map.
+ */
+struct ceph_osdmap *osdmap_decode(void **p, void *end)
+{
+	struct ceph_osdmap *map;
+	u16 version;
+	u32 len, max, i;
+	u8 ev;
+	int err = -EINVAL;
+	void *start = *p;
+	struct ceph_pg_pool_info *pi;
+
+	dout("osdmap_decode %p to %p len %d\n", *p, end, (int)(end - *p));
+
+	map = kzalloc(sizeof(*map), GFP_NOFS);
+	if (map == NULL)
+		return ERR_PTR(-ENOMEM);
+	map->pg_temp = RB_ROOT;
+
+	ceph_decode_16_safe(p, end, version, bad);
+	if (version > CEPH_OSDMAP_VERSION) {
+		pr_warning("got unknown v %d > %d of osdmap\n", version,
+			   CEPH_OSDMAP_VERSION);
+		goto bad;
+	}
+
+	ceph_decode_need(p, end, 2*sizeof(u64)+6*sizeof(u32), bad);
+	ceph_decode_copy(p, &map->fsid, sizeof(map->fsid));
+	map->epoch = ceph_decode_32(p);
+	ceph_decode_copy(p, &map->created, sizeof(map->created));
+	ceph_decode_copy(p, &map->modified, sizeof(map->modified));
+
+	ceph_decode_32_safe(p, end, max, bad);
+	while (max--) {
+		ceph_decode_need(p, end, 4 + 1 + sizeof(pi->v), bad);
+		pi = kzalloc(sizeof(*pi), GFP_NOFS);
+		if (!pi)
+			goto bad;
+		pi->id = ceph_decode_32(p);
+		ev = ceph_decode_8(p); /* encoding version */
+		if (ev > CEPH_PG_POOL_VERSION) {
+			pr_warning("got unknown v %d > %d of ceph_pg_pool\n",
+				   ev, CEPH_PG_POOL_VERSION);
+			kfree(pi);
+			goto bad;
+		}
+		err = __decode_pool(p, end, pi);
+		if (err < 0)
+			goto bad;
+		__insert_pg_pool(&map->pg_pools, pi);
+	}
+
+	if (version >= 5 && __decode_pool_names(p, end, map) < 0)
+		goto bad;
+
+	ceph_decode_32_safe(p, end, map->pool_max, bad);
+
+	ceph_decode_32_safe(p, end, map->flags, bad);
+
+	max = ceph_decode_32(p);
+
+	/* (re)alloc osd arrays */
+	err = osdmap_set_max_osd(map, max);
+	if (err < 0)
+		goto bad;
+	dout("osdmap_decode max_osd = %d\n", map->max_osd);
+
+	/* osds */
+	err = -EINVAL;
+	ceph_decode_need(p, end, 3*sizeof(u32) +
+			 map->max_osd*(1 + sizeof(*map->osd_weight) +
+				       sizeof(*map->osd_addr)), bad);
+	*p += 4; /* skip length field (should match max) */
+	ceph_decode_copy(p, map->osd_state, map->max_osd);
+
+	*p += 4; /* skip length field (should match max) */
+	for (i = 0; i < map->max_osd; i++)
+		map->osd_weight[i] = ceph_decode_32(p);
+
+	*p += 4; /* skip length field (should match max) */
+	ceph_decode_copy(p, map->osd_addr, map->max_osd*sizeof(*map->osd_addr));
+	for (i = 0; i < map->max_osd; i++)
+		ceph_decode_addr(&map->osd_addr[i]);
+
+	/* pg_temp */
+	ceph_decode_32_safe(p, end, len, bad);
+	for (i = 0; i < len; i++) {
+		int n, j;
+		struct ceph_pg pgid;
+		struct ceph_pg_mapping *pg;
+
+		ceph_decode_need(p, end, sizeof(u32) + sizeof(u64), bad);
+		ceph_decode_copy(p, &pgid, sizeof(pgid));
+		n = ceph_decode_32(p);
+		ceph_decode_need(p, end, n * sizeof(u32), bad);
+		err = -ENOMEM;
+		pg = kmalloc(sizeof(*pg) + n*sizeof(u32), GFP_NOFS);
+		if (!pg)
+			goto bad;
+		pg->pgid = pgid;
+		pg->len = n;
+		for (j = 0; j < n; j++)
+			pg->osds[j] = ceph_decode_32(p);
+
+		err = __insert_pg_mapping(pg, &map->pg_temp);
+		if (err)
+			goto bad;
+		dout(" added pg_temp %llx len %d\n", *(u64 *)&pgid, len);
+	}
+
+	/* crush */
+	ceph_decode_32_safe(p, end, len, bad);
+	dout("osdmap_decode crush len %d from off 0x%x\n", len,
+	     (int)(*p - start));
+	ceph_decode_need(p, end, len, bad);
+	map->crush = crush_decode(*p, end);
+	*p += len;
+	if (IS_ERR(map->crush)) {
+		err = PTR_ERR(map->crush);
+		map->crush = NULL;
+		goto bad;
+	}
+
+	/* ignore the rest of the map */
+	*p = end;
+
+	dout("osdmap_decode done %p %p\n", *p, end);
+	return map;
+
+bad:
+	dout("osdmap_decode fail\n");
+	ceph_osdmap_destroy(map);
+	return ERR_PTR(err);
+}
+
+/*
+ * decode and apply an incremental map update.
+ */
+struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
+					     struct ceph_osdmap *map,
+					     struct ceph_messenger *msgr)
+{
+	struct crush_map *newcrush = NULL;
+	struct ceph_fsid fsid;
+	u32 epoch = 0;
+	struct ceph_timespec modified;
+	u32 len, pool;
+	__s32 new_pool_max, new_flags, max;
+	void *start = *p;
+	int err = -EINVAL;
+	u16 version;
+	struct rb_node *rbp;
+
+	ceph_decode_16_safe(p, end, version, bad);
+	if (version > CEPH_OSDMAP_INC_VERSION) {
+		pr_warning("got unknown v %d > %d of inc osdmap\n", version,
+			   CEPH_OSDMAP_INC_VERSION);
+		goto bad;
+	}
+
+	ceph_decode_need(p, end, sizeof(fsid)+sizeof(modified)+2*sizeof(u32),
+			 bad);
+	ceph_decode_copy(p, &fsid, sizeof(fsid));
+	epoch = ceph_decode_32(p);
+	BUG_ON(epoch != map->epoch+1);
+	ceph_decode_copy(p, &modified, sizeof(modified));
+	new_pool_max = ceph_decode_32(p);
+	new_flags = ceph_decode_32(p);
+
+	/* full map? */
+	ceph_decode_32_safe(p, end, len, bad);
+	if (len > 0) {
+		dout("apply_incremental full map len %d, %p to %p\n",
+		     len, *p, end);
+		return osdmap_decode(p, min(*p+len, end));
+	}
+
+	/* new crush? */
+	ceph_decode_32_safe(p, end, len, bad);
+	if (len > 0) {
+		dout("apply_incremental new crush map len %d, %p to %p\n",
+		     len, *p, end);
+		newcrush = crush_decode(*p, min(*p+len, end));
+		if (IS_ERR(newcrush))
+			return ERR_CAST(newcrush);
+		*p += len;
+	}
+
+	/* new flags? */
+	if (new_flags >= 0)
+		map->flags = new_flags;
+	if (new_pool_max >= 0)
+		map->pool_max = new_pool_max;
+
+	ceph_decode_need(p, end, 5*sizeof(u32), bad);
+
+	/* new max? */
+	max = ceph_decode_32(p);
+	if (max >= 0) {
+		err = osdmap_set_max_osd(map, max);
+		if (err < 0)
+			goto bad;
+	}
+
+	map->epoch++;
+	map->modified = map->modified;
+	if (newcrush) {
+		if (map->crush)
+			crush_destroy(map->crush);
+		map->crush = newcrush;
+		newcrush = NULL;
+	}
+
+	/* new_pool */
+	ceph_decode_32_safe(p, end, len, bad);
+	while (len--) {
+		__u8 ev;
+		struct ceph_pg_pool_info *pi;
+
+		ceph_decode_32_safe(p, end, pool, bad);
+		ceph_decode_need(p, end, 1 + sizeof(pi->v), bad);
+		ev = ceph_decode_8(p);  /* encoding version */
+		if (ev > CEPH_PG_POOL_VERSION) {
+			pr_warning("got unknown v %d > %d of ceph_pg_pool\n",
+				   ev, CEPH_PG_POOL_VERSION);
+			goto bad;
+		}
+		pi = __lookup_pg_pool(&map->pg_pools, pool);
+		if (!pi) {
+			pi = kzalloc(sizeof(*pi), GFP_NOFS);
+			if (!pi) {
+				err = -ENOMEM;
+				goto bad;
+			}
+			pi->id = pool;
+			__insert_pg_pool(&map->pg_pools, pi);
+		}
+		err = __decode_pool(p, end, pi);
+		if (err < 0)
+			goto bad;
+	}
+	if (version >= 5 && __decode_pool_names(p, end, map) < 0)
+		goto bad;
+
+	/* old_pool */
+	ceph_decode_32_safe(p, end, len, bad);
+	while (len--) {
+		struct ceph_pg_pool_info *pi;
+
+		ceph_decode_32_safe(p, end, pool, bad);
+		pi = __lookup_pg_pool(&map->pg_pools, pool);
+		if (pi)
+			__remove_pg_pool(&map->pg_pools, pi);
+	}
+
+	/* new_up */
+	err = -EINVAL;
+	ceph_decode_32_safe(p, end, len, bad);
+	while (len--) {
+		u32 osd;
+		struct ceph_entity_addr addr;
+		ceph_decode_32_safe(p, end, osd, bad);
+		ceph_decode_copy_safe(p, end, &addr, sizeof(addr), bad);
+		ceph_decode_addr(&addr);
+		pr_info("osd%d up\n", osd);
+		BUG_ON(osd >= map->max_osd);
+		map->osd_state[osd] |= CEPH_OSD_UP;
+		map->osd_addr[osd] = addr;
+	}
+
+	/* new_down */
+	ceph_decode_32_safe(p, end, len, bad);
+	while (len--) {
+		u32 osd;
+		ceph_decode_32_safe(p, end, osd, bad);
+		(*p)++;  /* clean flag */
+		pr_info("osd%d down\n", osd);
+		if (osd < map->max_osd)
+			map->osd_state[osd] &= ~CEPH_OSD_UP;
+	}
+
+	/* new_weight */
+	ceph_decode_32_safe(p, end, len, bad);
+	while (len--) {
+		u32 osd, off;
+		ceph_decode_need(p, end, sizeof(u32)*2, bad);
+		osd = ceph_decode_32(p);
+		off = ceph_decode_32(p);
+		pr_info("osd%d weight 0x%x %s\n", osd, off,
+		     off == CEPH_OSD_IN ? "(in)" :
+		     (off == CEPH_OSD_OUT ? "(out)" : ""));
+		if (osd < map->max_osd)
+			map->osd_weight[osd] = off;
+	}
+
+	/* new_pg_temp */
+	rbp = rb_first(&map->pg_temp);
+	ceph_decode_32_safe(p, end, len, bad);
+	while (len--) {
+		struct ceph_pg_mapping *pg;
+		int j;
+		struct ceph_pg pgid;
+		u32 pglen;
+		ceph_decode_need(p, end, sizeof(u64) + sizeof(u32), bad);
+		ceph_decode_copy(p, &pgid, sizeof(pgid));
+		pglen = ceph_decode_32(p);
+
+		/* remove any? */
+		while (rbp && pgid_cmp(rb_entry(rbp, struct ceph_pg_mapping,
+						node)->pgid, pgid) <= 0) {
+			struct ceph_pg_mapping *cur =
+				rb_entry(rbp, struct ceph_pg_mapping, node);
+
+			rbp = rb_next(rbp);
+			dout(" removed pg_temp %llx\n", *(u64 *)&cur->pgid);
+			rb_erase(&cur->node, &map->pg_temp);
+			kfree(cur);
+		}
+
+		if (pglen) {
+			/* insert */
+			ceph_decode_need(p, end, pglen*sizeof(u32), bad);
+			pg = kmalloc(sizeof(*pg) + sizeof(u32)*pglen, GFP_NOFS);
+			if (!pg) {
+				err = -ENOMEM;
+				goto bad;
+			}
+			pg->pgid = pgid;
+			pg->len = pglen;
+			for (j = 0; j < pglen; j++)
+				pg->osds[j] = ceph_decode_32(p);
+			err = __insert_pg_mapping(pg, &map->pg_temp);
+			if (err) {
+				kfree(pg);
+				goto bad;
+			}
+			dout(" added pg_temp %llx len %d\n", *(u64 *)&pgid,
+			     pglen);
+		}
+	}
+	while (rbp) {
+		struct ceph_pg_mapping *cur =
+			rb_entry(rbp, struct ceph_pg_mapping, node);
+
+		rbp = rb_next(rbp);
+		dout(" removed pg_temp %llx\n", *(u64 *)&cur->pgid);
+		rb_erase(&cur->node, &map->pg_temp);
+		kfree(cur);
+	}
+
+	/* ignore the rest */
+	*p = end;
+	return map;
+
+bad:
+	pr_err("corrupt inc osdmap epoch %d off %d (%p of %p-%p)\n",
+	       epoch, (int)(*p - start), *p, start, end);
+	print_hex_dump(KERN_DEBUG, "osdmap: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       start, end - start, true);
+	if (newcrush)
+		crush_destroy(newcrush);
+	return ERR_PTR(err);
+}
+
+
+
+
+/*
+ * calculate file layout from given offset, length.
+ * fill in correct oid, logical length, and object extent
+ * offset, length.
+ *
+ * for now, we write only a single su, until we can
+ * pass a stride back to the caller.
+ */
+void ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
+				   u64 off, u64 *plen,
+				   u64 *ono,
+				   u64 *oxoff, u64 *oxlen)
+{
+	u32 osize = le32_to_cpu(layout->fl_object_size);
+	u32 su = le32_to_cpu(layout->fl_stripe_unit);
+	u32 sc = le32_to_cpu(layout->fl_stripe_count);
+	u32 bl, stripeno, stripepos, objsetno;
+	u32 su_per_object;
+	u64 t, su_offset;
+
+	dout("mapping %llu~%llu  osize %u fl_su %u\n", off, *plen,
+	     osize, su);
+	su_per_object = osize / su;
+	dout("osize %u / su %u = su_per_object %u\n", osize, su,
+	     su_per_object);
+
+	BUG_ON((su & ~PAGE_MASK) != 0);
+	/* bl = *off / su; */
+	t = off;
+	do_div(t, su);
+	bl = t;
+	dout("off %llu / su %u = bl %u\n", off, su, bl);
+
+	stripeno = bl / sc;
+	stripepos = bl % sc;
+	objsetno = stripeno / su_per_object;
+
+	*ono = objsetno * sc + stripepos;
+	dout("objset %u * sc %u = ono %u\n", objsetno, sc, (unsigned)*ono);
+
+	/* *oxoff = *off % layout->fl_stripe_unit;  # offset in su */
+	t = off;
+	su_offset = do_div(t, su);
+	*oxoff = su_offset + (stripeno % su_per_object) * su;
+
+	/*
+	 * Calculate the length of the extent being written to the selected
+	 * object. This is the minimum of the full length requested (plen) or
+	 * the remainder of the current stripe being written to.
+	 */
+	*oxlen = min_t(u64, *plen, su - su_offset);
+	*plen = *oxlen;
+
+	dout(" obj extent %llu~%llu\n", *oxoff, *oxlen);
+}
+EXPORT_SYMBOL(ceph_calc_file_object_mapping);
+
+/*
+ * calculate an object layout (i.e. pgid) from an oid,
+ * file_layout, and osdmap
+ */
+int ceph_calc_object_layout(struct ceph_object_layout *ol,
+			    const char *oid,
+			    struct ceph_file_layout *fl,
+			    struct ceph_osdmap *osdmap)
+{
+	unsigned num, num_mask;
+	struct ceph_pg pgid;
+	s32 preferred = (s32)le32_to_cpu(fl->fl_pg_preferred);
+	int poolid = le32_to_cpu(fl->fl_pg_pool);
+	struct ceph_pg_pool_info *pool;
+	unsigned ps;
+
+	BUG_ON(!osdmap);
+
+	pool = __lookup_pg_pool(&osdmap->pg_pools, poolid);
+	if (!pool)
+		return -EIO;
+	ps = ceph_str_hash(pool->v.object_hash, oid, strlen(oid));
+	if (preferred >= 0) {
+		ps += preferred;
+		num = le32_to_cpu(pool->v.lpg_num);
+		num_mask = pool->lpg_num_mask;
+	} else {
+		num = le32_to_cpu(pool->v.pg_num);
+		num_mask = pool->pg_num_mask;
+	}
+
+	pgid.ps = cpu_to_le16(ps);
+	pgid.preferred = cpu_to_le16(preferred);
+	pgid.pool = fl->fl_pg_pool;
+	if (preferred >= 0)
+		dout("calc_object_layout '%s' pgid %d.%xp%d\n", oid, poolid, ps,
+		     (int)preferred);
+	else
+		dout("calc_object_layout '%s' pgid %d.%x\n", oid, poolid, ps);
+
+	ol->ol_pgid = pgid;
+	ol->ol_stripe_unit = fl->fl_object_stripe_unit;
+	return 0;
+}
+EXPORT_SYMBOL(ceph_calc_object_layout);
+
+/*
+ * Calculate raw osd vector for the given pgid.  Return pointer to osd
+ * array, or NULL on failure.
+ */
+static int *calc_pg_raw(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
+			int *osds, int *num)
+{
+	struct ceph_pg_mapping *pg;
+	struct ceph_pg_pool_info *pool;
+	int ruleno;
+	unsigned poolid, ps, pps;
+	int preferred;
+
+	/* pg_temp? */
+	pg = __lookup_pg_mapping(&osdmap->pg_temp, pgid);
+	if (pg) {
+		*num = pg->len;
+		return pg->osds;
+	}
+
+	/* crush */
+	poolid = le32_to_cpu(pgid.pool);
+	ps = le16_to_cpu(pgid.ps);
+	preferred = (s16)le16_to_cpu(pgid.preferred);
+
+	/* don't forcefeed bad device ids to crush */
+	if (preferred >= osdmap->max_osd ||
+	    preferred >= osdmap->crush->max_devices)
+		preferred = -1;
+
+	pool = __lookup_pg_pool(&osdmap->pg_pools, poolid);
+	if (!pool)
+		return NULL;
+	ruleno = crush_find_rule(osdmap->crush, pool->v.crush_ruleset,
+				 pool->v.type, pool->v.size);
+	if (ruleno < 0) {
+		pr_err("no crush rule pool %d ruleset %d type %d size %d\n",
+		       poolid, pool->v.crush_ruleset, pool->v.type,
+		       pool->v.size);
+		return NULL;
+	}
+
+	if (preferred >= 0)
+		pps = ceph_stable_mod(ps,
+				      le32_to_cpu(pool->v.lpgp_num),
+				      pool->lpgp_num_mask);
+	else
+		pps = ceph_stable_mod(ps,
+				      le32_to_cpu(pool->v.pgp_num),
+				      pool->pgp_num_mask);
+	pps += poolid;
+	*num = crush_do_rule(osdmap->crush, ruleno, pps, osds,
+			     min_t(int, pool->v.size, *num),
+			     preferred, osdmap->osd_weight);
+	return osds;
+}
+
+/*
+ * Return acting set for given pgid.
+ */
+int ceph_calc_pg_acting(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
+			int *acting)
+{
+	int rawosds[CEPH_PG_MAX_SIZE], *osds;
+	int i, o, num = CEPH_PG_MAX_SIZE;
+
+	osds = calc_pg_raw(osdmap, pgid, rawosds, &num);
+	if (!osds)
+		return -1;
+
+	/* primary is first up osd */
+	o = 0;
+	for (i = 0; i < num; i++)
+		if (ceph_osd_is_up(osdmap, osds[i]))
+			acting[o++] = osds[i];
+	return o;
+}
+
+/*
+ * Return primary osd for given pgid, or -1 if none.
+ */
+int ceph_calc_pg_primary(struct ceph_osdmap *osdmap, struct ceph_pg pgid)
+{
+	int rawosds[CEPH_PG_MAX_SIZE], *osds;
+	int i, num = CEPH_PG_MAX_SIZE;
+
+	osds = calc_pg_raw(osdmap, pgid, rawosds, &num);
+	if (!osds)
+		return -1;
+
+	/* primary is first up osd */
+	for (i = 0; i < num; i++)
+		if (ceph_osd_is_up(osdmap, osds[i]))
+			return osds[i];
+	return -1;
+}
+EXPORT_SYMBOL(ceph_calc_pg_primary);
diff --git a/net/ceph/pagelist.c b/net/ceph/pagelist.c
new file mode 100644
index 000000000000..3b146cfac182
--- /dev/null
+++ b/net/ceph/pagelist.c
@@ -0,0 +1,65 @@
+
+#include <linux/module.h>
+#include <linux/gfp.h>
+#include <linux/pagemap.h>
+#include <linux/highmem.h>
+#include <linux/ceph/pagelist.h>
+
+static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl)
+{
+	struct page *page = list_entry(pl->head.prev, struct page,
+				       lru);
+	kunmap(page);
+}
+
+int ceph_pagelist_release(struct ceph_pagelist *pl)
+{
+	if (pl->mapped_tail)
+		ceph_pagelist_unmap_tail(pl);
+
+	while (!list_empty(&pl->head)) {
+		struct page *page = list_first_entry(&pl->head, struct page,
+						     lru);
+		list_del(&page->lru);
+		__free_page(page);
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_release);
+
+static int ceph_pagelist_addpage(struct ceph_pagelist *pl)
+{
+	struct page *page = __page_cache_alloc(GFP_NOFS);
+	if (!page)
+		return -ENOMEM;
+	pl->room += PAGE_SIZE;
+	list_add_tail(&page->lru, &pl->head);
+	if (pl->mapped_tail)
+		ceph_pagelist_unmap_tail(pl);
+	pl->mapped_tail = kmap(page);
+	return 0;
+}
+
+int ceph_pagelist_append(struct ceph_pagelist *pl, const void *buf, size_t len)
+{
+	while (pl->room < len) {
+		size_t bit = pl->room;
+		int ret;
+
+		memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK),
+		       buf, bit);
+		pl->length += bit;
+		pl->room -= bit;
+		buf += bit;
+		len -= bit;
+		ret = ceph_pagelist_addpage(pl);
+		if (ret)
+			return ret;
+	}
+
+	memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK), buf, len);
+	pl->length += len;
+	pl->room -= len;
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_append);
diff --git a/net/ceph/pagevec.c b/net/ceph/pagevec.c
new file mode 100644
index 000000000000..54caf0687155
--- /dev/null
+++ b/net/ceph/pagevec.c
@@ -0,0 +1,223 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/file.h>
+#include <linux/namei.h>
+#include <linux/writeback.h>
+
+#include <linux/ceph/libceph.h>
+
+/*
+ * build a vector of user pages
+ */
+struct page **ceph_get_direct_page_vector(const char __user *data,
+						 int num_pages,
+						 loff_t off, size_t len)
+{
+	struct page **pages;
+	int rc;
+
+	pages = kmalloc(sizeof(*pages) * num_pages, GFP_NOFS);
+	if (!pages)
+		return ERR_PTR(-ENOMEM);
+
+	down_read(&current->mm->mmap_sem);
+	rc = get_user_pages(current, current->mm, (unsigned long)data,
+			    num_pages, 0, 0, pages, NULL);
+	up_read(&current->mm->mmap_sem);
+	if (rc < 0)
+		goto fail;
+	return pages;
+
+fail:
+	kfree(pages);
+	return ERR_PTR(rc);
+}
+EXPORT_SYMBOL(ceph_get_direct_page_vector);
+
+void ceph_put_page_vector(struct page **pages, int num_pages)
+{
+	int i;
+
+	for (i = 0; i < num_pages; i++)
+		put_page(pages[i]);
+	kfree(pages);
+}
+EXPORT_SYMBOL(ceph_put_page_vector);
+
+void ceph_release_page_vector(struct page **pages, int num_pages)
+{
+	int i;
+
+	for (i = 0; i < num_pages; i++)
+		__free_pages(pages[i], 0);
+	kfree(pages);
+}
+EXPORT_SYMBOL(ceph_release_page_vector);
+
+/*
+ * allocate a vector new pages
+ */
+struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags)
+{
+	struct page **pages;
+	int i;
+
+	pages = kmalloc(sizeof(*pages) * num_pages, flags);
+	if (!pages)
+		return ERR_PTR(-ENOMEM);
+	for (i = 0; i < num_pages; i++) {
+		pages[i] = __page_cache_alloc(flags);
+		if (pages[i] == NULL) {
+			ceph_release_page_vector(pages, i);
+			return ERR_PTR(-ENOMEM);
+		}
+	}
+	return pages;
+}
+EXPORT_SYMBOL(ceph_alloc_page_vector);
+
+/*
+ * copy user data into a page vector
+ */
+int ceph_copy_user_to_page_vector(struct page **pages,
+					 const char __user *data,
+					 loff_t off, size_t len)
+{
+	int i = 0;
+	int po = off & ~PAGE_CACHE_MASK;
+	int left = len;
+	int l, bad;
+
+	while (left > 0) {
+		l = min_t(int, PAGE_CACHE_SIZE-po, left);
+		bad = copy_from_user(page_address(pages[i]) + po, data, l);
+		if (bad == l)
+			return -EFAULT;
+		data += l - bad;
+		left -= l - bad;
+		po += l - bad;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_user_to_page_vector);
+
+int ceph_copy_to_page_vector(struct page **pages,
+				    const char *data,
+				    loff_t off, size_t len)
+{
+	int i = 0;
+	size_t po = off & ~PAGE_CACHE_MASK;
+	size_t left = len;
+	size_t l;
+
+	while (left > 0) {
+		l = min_t(size_t, PAGE_CACHE_SIZE-po, left);
+		memcpy(page_address(pages[i]) + po, data, l);
+		data += l;
+		left -= l;
+		po += l;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_to_page_vector);
+
+int ceph_copy_from_page_vector(struct page **pages,
+				    char *data,
+				    loff_t off, size_t len)
+{
+	int i = 0;
+	size_t po = off & ~PAGE_CACHE_MASK;
+	size_t left = len;
+	size_t l;
+
+	while (left > 0) {
+		l = min_t(size_t, PAGE_CACHE_SIZE-po, left);
+		memcpy(data, page_address(pages[i]) + po, l);
+		data += l;
+		left -= l;
+		po += l;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_from_page_vector);
+
+/*
+ * copy user data from a page vector into a user pointer
+ */
+int ceph_copy_page_vector_to_user(struct page **pages,
+					 char __user *data,
+					 loff_t off, size_t len)
+{
+	int i = 0;
+	int po = off & ~PAGE_CACHE_MASK;
+	int left = len;
+	int l, bad;
+
+	while (left > 0) {
+		l = min_t(int, left, PAGE_CACHE_SIZE-po);
+		bad = copy_to_user(data, page_address(pages[i]) + po, l);
+		if (bad == l)
+			return -EFAULT;
+		data += l - bad;
+		left -= l - bad;
+		if (po) {
+			po += l - bad;
+			if (po == PAGE_CACHE_SIZE)
+				po = 0;
+		}
+		i++;
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_page_vector_to_user);
+
+/*
+ * Zero an extent within a page vector.  Offset is relative to the
+ * start of the first page.
+ */
+void ceph_zero_page_vector_range(int off, int len, struct page **pages)
+{
+	int i = off >> PAGE_CACHE_SHIFT;
+
+	off &= ~PAGE_CACHE_MASK;
+
+	dout("zero_page_vector_page %u~%u\n", off, len);
+
+	/* leading partial page? */
+	if (off) {
+		int end = min((int)PAGE_CACHE_SIZE, off + len);
+		dout("zeroing %d %p head from %d\n", i, pages[i],
+		     (int)off);
+		zero_user_segment(pages[i], off, end);
+		len -= (end - off);
+		i++;
+	}
+	while (len >= PAGE_CACHE_SIZE) {
+		dout("zeroing %d %p len=%d\n", i, pages[i], len);
+		zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE);
+		len -= PAGE_CACHE_SIZE;
+		i++;
+	}
+	/* trailing partial page? */
+	if (len) {
+		dout("zeroing %d %p tail to %d\n", i, pages[i], (int)len);
+		zero_user_segment(pages[i], 0, len);
+	}
+}
+EXPORT_SYMBOL(ceph_zero_page_vector_range);
+
-- 
cgit v1.2.3


From ac0b74d8a1ced8ea86147467daf06b15b130dd94 Mon Sep 17 00:00:00 2001
From: Greg Farnum <gregf@hq.newdream.net>
Date: Fri, 17 Sep 2010 10:10:55 -0700
Subject: ceph: add pagelist_reserve, pagelist_truncate, pagelist_set_cursor

These facilitate preallocation of pages so that we can encode into the pagelist
in an atomic context.

Signed-off-by: Greg Farnum <gregf@hq.newdream.net>
Signed-off-by: Sage Weil <sage@newdream.net>
---
 include/linux/ceph/pagelist.h |  21 +++++++++
 net/ceph/pagelist.c           | 106 ++++++++++++++++++++++++++++++++++++++----
 2 files changed, 118 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ceph/pagelist.h b/include/linux/ceph/pagelist.h
index cc9327aa1c98..9660d6b0a35d 100644
--- a/include/linux/ceph/pagelist.h
+++ b/include/linux/ceph/pagelist.h
@@ -8,6 +8,14 @@ struct ceph_pagelist {
 	void *mapped_tail;
 	size_t length;
 	size_t room;
+	struct list_head free_list;
+	size_t num_pages_free;
+};
+
+struct ceph_pagelist_cursor {
+	struct ceph_pagelist *pl;   /* pagelist, for error checking */
+	struct list_head *page_lru; /* page in list */
+	size_t room;		    /* room remaining to reset to */
 };
 
 static inline void ceph_pagelist_init(struct ceph_pagelist *pl)
@@ -16,11 +24,24 @@ static inline void ceph_pagelist_init(struct ceph_pagelist *pl)
 	pl->mapped_tail = NULL;
 	pl->length = 0;
 	pl->room = 0;
+	INIT_LIST_HEAD(&pl->free_list);
+	pl->num_pages_free = 0;
 }
+
 extern int ceph_pagelist_release(struct ceph_pagelist *pl);
 
 extern int ceph_pagelist_append(struct ceph_pagelist *pl, const void *d, size_t l);
 
+extern int ceph_pagelist_reserve(struct ceph_pagelist *pl, size_t space);
+
+extern int ceph_pagelist_free_reserve(struct ceph_pagelist *pl);
+
+extern void ceph_pagelist_set_cursor(struct ceph_pagelist *pl,
+				     struct ceph_pagelist_cursor *c);
+
+extern int ceph_pagelist_truncate(struct ceph_pagelist *pl,
+				  struct ceph_pagelist_cursor *c);
+
 static inline int ceph_pagelist_encode_64(struct ceph_pagelist *pl, u64 v)
 {
 	__le64 ev = cpu_to_le64(v);
diff --git a/net/ceph/pagelist.c b/net/ceph/pagelist.c
index 3b146cfac182..b8cbc456d0bb 100644
--- a/net/ceph/pagelist.c
+++ b/net/ceph/pagelist.c
@@ -7,35 +7,42 @@
 
 static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl)
 {
-	struct page *page = list_entry(pl->head.prev, struct page,
-				       lru);
-	kunmap(page);
+	if (pl->mapped_tail) {
+		struct page *page = list_entry(pl->head.prev, struct page, lru);
+		kunmap(page);
+		pl->mapped_tail = NULL;
+	}
 }
 
 int ceph_pagelist_release(struct ceph_pagelist *pl)
 {
-	if (pl->mapped_tail)
-		ceph_pagelist_unmap_tail(pl);
-
+	ceph_pagelist_unmap_tail(pl);
 	while (!list_empty(&pl->head)) {
 		struct page *page = list_first_entry(&pl->head, struct page,
 						     lru);
 		list_del(&page->lru);
 		__free_page(page);
 	}
+	ceph_pagelist_free_reserve(pl);
 	return 0;
 }
 EXPORT_SYMBOL(ceph_pagelist_release);
 
 static int ceph_pagelist_addpage(struct ceph_pagelist *pl)
 {
-	struct page *page = __page_cache_alloc(GFP_NOFS);
+	struct page *page;
+
+	if (!pl->num_pages_free) {
+		page = __page_cache_alloc(GFP_NOFS);
+	} else {
+		page = list_first_entry(&pl->free_list, struct page, lru);
+		list_del(&page->lru);
+	}
 	if (!page)
 		return -ENOMEM;
 	pl->room += PAGE_SIZE;
+	ceph_pagelist_unmap_tail(pl);
 	list_add_tail(&page->lru, &pl->head);
-	if (pl->mapped_tail)
-		ceph_pagelist_unmap_tail(pl);
 	pl->mapped_tail = kmap(page);
 	return 0;
 }
@@ -63,3 +70,84 @@ int ceph_pagelist_append(struct ceph_pagelist *pl, const void *buf, size_t len)
 	return 0;
 }
 EXPORT_SYMBOL(ceph_pagelist_append);
+
+/**
+ * Allocate enough pages for a pagelist to append the given amount
+ * of data without without allocating.
+ * Returns: 0 on success, -ENOMEM on error.
+ */
+int ceph_pagelist_reserve(struct ceph_pagelist *pl, size_t space)
+{
+	if (space <= pl->room)
+		return 0;
+	space -= pl->room;
+	space = (space + PAGE_SIZE - 1) >> PAGE_SHIFT;   /* conv to num pages */
+
+	while (space > pl->num_pages_free) {
+		struct page *page = __page_cache_alloc(GFP_NOFS);
+		if (!page)
+			return -ENOMEM;
+		list_add_tail(&page->lru, &pl->free_list);
+		++pl->num_pages_free;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_reserve);
+
+/**
+ * Free any pages that have been preallocated.
+ */
+int ceph_pagelist_free_reserve(struct ceph_pagelist *pl)
+{
+	while (!list_empty(&pl->free_list)) {
+		struct page *page = list_first_entry(&pl->free_list,
+						     struct page, lru);
+		list_del(&page->lru);
+		__free_page(page);
+		--pl->num_pages_free;
+	}
+	BUG_ON(pl->num_pages_free);
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_free_reserve);
+
+/**
+ * Create a truncation point.
+ */
+void ceph_pagelist_set_cursor(struct ceph_pagelist *pl,
+			      struct ceph_pagelist_cursor *c)
+{
+	c->pl = pl;
+	c->page_lru = pl->head.prev;
+	c->room = pl->room;
+}
+EXPORT_SYMBOL(ceph_pagelist_set_cursor);
+
+/**
+ * Truncate a pagelist to the given point. Move extra pages to reserve.
+ * This won't sleep.
+ * Returns: 0 on success,
+ *          -EINVAL if the pagelist doesn't match the trunc point pagelist
+ */
+int ceph_pagelist_truncate(struct ceph_pagelist *pl,
+			   struct ceph_pagelist_cursor *c)
+{
+	struct page *page;
+
+	if (pl != c->pl)
+		return -EINVAL;
+	ceph_pagelist_unmap_tail(pl);
+	while (pl->head.prev != c->page_lru) {
+		page = list_entry(pl->head.prev, struct page, lru);
+		list_del(&page->lru);                /* remove from pagelist */
+		list_add_tail(&page->lru, &pl->free_list); /* add to reserve */
+		++pl->num_pages_free;
+	}
+	pl->room = c->room;
+	if (!list_empty(&pl->head)) {
+		page = list_entry(pl->head.prev, struct page, lru);
+		pl->mapped_tail = kmap(page);
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_truncate);
-- 
cgit v1.2.3


From 571dba52a34015a5a7aa5d480a86936878444a6f Mon Sep 17 00:00:00 2001
From: Greg Farnum <gregf@hq.newdream.net>
Date: Fri, 24 Sep 2010 14:56:40 -0700
Subject: ceph: add CEPH_MDS_OP_SETDIRLAYOUT and associated ioctl.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 fs/ceph/ioctl.c              | 66 ++++++++++++++++++++++++++++++++++++++++++++
 fs/ceph/ioctl.h              |  4 ++-
 include/linux/ceph/ceph_fs.h |  1 +
 3 files changed, 70 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 899578b0c46b..8888c9ba68db 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -91,6 +91,68 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg)
 	return err;
 }
 
+/*
+ * Set a layout policy on a directory inode. All items in the tree
+ * rooted at this inode will inherit this layout on creation,
+ * (It doesn't apply retroactively )
+ * unless a subdirectory has its own layout policy.
+ */
+static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct ceph_mds_request *req;
+	struct ceph_ioctl_layout l;
+	int err, i;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
+
+	/* copy and validate */
+	if (copy_from_user(&l, arg, sizeof(l)))
+		return -EFAULT;
+
+	if ((l.object_size & ~PAGE_MASK) ||
+	    (l.stripe_unit & ~PAGE_MASK) ||
+	    !l.stripe_unit ||
+	    (l.object_size &&
+	        (unsigned)l.object_size % (unsigned)l.stripe_unit))
+		return -EINVAL;
+
+	/* make sure it's a valid data pool */
+	if (l.data_pool > 0) {
+		mutex_lock(&mdsc->mutex);
+		err = -EINVAL;
+		for (i = 0; i < mdsc->mdsmap->m_num_data_pg_pools; i++)
+			if (mdsc->mdsmap->m_data_pg_pools[i] == l.data_pool) {
+				err = 0;
+				break;
+			}
+		mutex_unlock(&mdsc->mutex);
+		if (err)
+			return err;
+	}
+
+	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETDIRLAYOUT,
+				       USE_AUTH_MDS);
+
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+	req->r_inode = igrab(inode);
+
+	req->r_args.setlayout.layout.fl_stripe_unit =
+			cpu_to_le32(l.stripe_unit);
+	req->r_args.setlayout.layout.fl_stripe_count =
+			cpu_to_le32(l.stripe_count);
+	req->r_args.setlayout.layout.fl_object_size =
+			cpu_to_le32(l.object_size);
+	req->r_args.setlayout.layout.fl_pg_pool =
+			cpu_to_le32(l.data_pool);
+	req->r_args.setlayout.layout.fl_pg_preferred =
+			cpu_to_le32(l.preferred_osd);
+
+	err = ceph_mdsc_do_request(mdsc, inode, req);
+	ceph_mdsc_put_request(req);
+	return err;
+}
+
 /*
  * Return object name, size/offset information, and location (OSD
  * number, network address) for a given file offset.
@@ -177,11 +239,15 @@ long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case CEPH_IOC_SET_LAYOUT:
 		return ceph_ioctl_set_layout(file, (void __user *)arg);
 
+	case CEPH_IOC_SET_LAYOUT_POLICY:
+		return ceph_ioctl_set_layout_policy(file, (void __user *)arg);
+
 	case CEPH_IOC_GET_DATALOC:
 		return ceph_ioctl_get_dataloc(file, (void __user *)arg);
 
 	case CEPH_IOC_LAZYIO:
 		return ceph_ioctl_lazyio(file);
 	}
+
 	return -ENOTTY;
 }
diff --git a/fs/ceph/ioctl.h b/fs/ceph/ioctl.h
index 88451a3b6857..a6ce54e94eb5 100644
--- a/fs/ceph/ioctl.h
+++ b/fs/ceph/ioctl.h
@@ -4,7 +4,7 @@
 #include <linux/ioctl.h>
 #include <linux/types.h>
 
-#define CEPH_IOCTL_MAGIC 0x97
+#define CEPH_IOCTL_MAGIC 0x98
 
 /* just use u64 to align sanely on all archs */
 struct ceph_ioctl_layout {
@@ -17,6 +17,8 @@ struct ceph_ioctl_layout {
 				   struct ceph_ioctl_layout)
 #define CEPH_IOC_SET_LAYOUT _IOW(CEPH_IOCTL_MAGIC, 2,		\
 				   struct ceph_ioctl_layout)
+#define CEPH_IOC_SET_LAYOUT_POLICY _IOW(CEPH_IOCTL_MAGIC, 5,	\
+				   struct ceph_ioctl_layout)
 
 /*
  * Extract identity, address of the OSD and object storing a given
diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h
index d5619ac86711..c3c74aef289d 100644
--- a/include/linux/ceph/ceph_fs.h
+++ b/include/linux/ceph/ceph_fs.h
@@ -299,6 +299,7 @@ enum {
 	CEPH_MDS_OP_SETATTR    = 0x01108,
 	CEPH_MDS_OP_SETFILELOCK= 0x01109,
 	CEPH_MDS_OP_GETFILELOCK= 0x00110,
+	CEPH_MDS_OP_SETDIRLAYOUT=0x0110a,
 
 	CEPH_MDS_OP_MKNOD      = 0x01201,
 	CEPH_MDS_OP_LINK       = 0x01202,
-- 
cgit v1.2.3


From b0ae19811375031ae3b3fecc65b702a9c6e5cc28 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Fri, 15 Oct 2010 04:21:18 +0900
Subject: security: remove unused parameter from security_task_setscheduler()

All security modules shouldn't change sched_param parameter of
security_task_setscheduler().  This is not only meaningless, but also
make a harmful result if caller pass a static variable.

This patch remove policy and sched_param parameter from
security_task_setscheduler() becuase none of security module is
using it.

Cc: James Morris <jmorris@namei.org>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 arch/mips/kernel/mips-mt-fpaff.c |  2 +-
 include/linux/security.h         | 14 +++++---------
 kernel/cpuset.c                  |  4 ++--
 kernel/sched.c                   |  4 ++--
 security/commoncap.c             |  5 +----
 security/security.c              |  5 ++---
 security/selinux/hooks.c         |  4 ++--
 security/smack/smack_lsm.c       |  5 ++---
 8 files changed, 17 insertions(+), 26 deletions(-)

(limited to 'include/linux')

diff --git a/arch/mips/kernel/mips-mt-fpaff.c b/arch/mips/kernel/mips-mt-fpaff.c
index 2340f11dc29c..9a526ba6f257 100644
--- a/arch/mips/kernel/mips-mt-fpaff.c
+++ b/arch/mips/kernel/mips-mt-fpaff.c
@@ -103,7 +103,7 @@ asmlinkage long mipsmt_sys_sched_setaffinity(pid_t pid, unsigned int len,
 	if (!check_same_owner(p) && !capable(CAP_SYS_NICE))
 		goto out_unlock;
 
-	retval = security_task_setscheduler(p, 0, NULL);
+	retval = security_task_setscheduler(p)
 	if (retval)
 		goto out_unlock;
 
diff --git a/include/linux/security.h b/include/linux/security.h
index a22219afff09..294a0b228123 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -74,7 +74,7 @@ extern int cap_file_mmap(struct file *file, unsigned long reqprot,
 extern int cap_task_fix_setuid(struct cred *new, const struct cred *old, int flags);
 extern int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3,
 			  unsigned long arg4, unsigned long arg5);
-extern int cap_task_setscheduler(struct task_struct *p, int policy, struct sched_param *lp);
+extern int cap_task_setscheduler(struct task_struct *p);
 extern int cap_task_setioprio(struct task_struct *p, int ioprio);
 extern int cap_task_setnice(struct task_struct *p, int nice);
 extern int cap_syslog(int type, bool from_file);
@@ -1501,8 +1501,7 @@ struct security_operations {
 	int (*task_getioprio) (struct task_struct *p);
 	int (*task_setrlimit) (struct task_struct *p, unsigned int resource,
 			struct rlimit *new_rlim);
-	int (*task_setscheduler) (struct task_struct *p, int policy,
-				  struct sched_param *lp);
+	int (*task_setscheduler) (struct task_struct *p);
 	int (*task_getscheduler) (struct task_struct *p);
 	int (*task_movememory) (struct task_struct *p);
 	int (*task_kill) (struct task_struct *p,
@@ -1752,8 +1751,7 @@ int security_task_setioprio(struct task_struct *p, int ioprio);
 int security_task_getioprio(struct task_struct *p);
 int security_task_setrlimit(struct task_struct *p, unsigned int resource,
 		struct rlimit *new_rlim);
-int security_task_setscheduler(struct task_struct *p,
-				int policy, struct sched_param *lp);
+int security_task_setscheduler(struct task_struct *p);
 int security_task_getscheduler(struct task_struct *p);
 int security_task_movememory(struct task_struct *p);
 int security_task_kill(struct task_struct *p, struct siginfo *info,
@@ -2320,11 +2318,9 @@ static inline int security_task_setrlimit(struct task_struct *p,
 	return 0;
 }
 
-static inline int security_task_setscheduler(struct task_struct *p,
-					     int policy,
-					     struct sched_param *lp)
+static inline int security_task_setscheduler(struct task_struct *p)
 {
-	return cap_task_setscheduler(p, policy, lp);
+	return cap_task_setscheduler(p);
 }
 
 static inline int security_task_getscheduler(struct task_struct *p)
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index b23c0979bbe7..51b143e2a07a 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -1397,7 +1397,7 @@ static int cpuset_can_attach(struct cgroup_subsys *ss, struct cgroup *cont,
 	if (tsk->flags & PF_THREAD_BOUND)
 		return -EINVAL;
 
-	ret = security_task_setscheduler(tsk, 0, NULL);
+	ret = security_task_setscheduler(tsk);
 	if (ret)
 		return ret;
 	if (threadgroup) {
@@ -1405,7 +1405,7 @@ static int cpuset_can_attach(struct cgroup_subsys *ss, struct cgroup *cont,
 
 		rcu_read_lock();
 		list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) {
-			ret = security_task_setscheduler(c, 0, NULL);
+			ret = security_task_setscheduler(c);
 			if (ret) {
 				rcu_read_unlock();
 				return ret;
diff --git a/kernel/sched.c b/kernel/sched.c
index dc85ceb90832..df6579d9b4df 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4645,7 +4645,7 @@ recheck:
 	}
 
 	if (user) {
-		retval = security_task_setscheduler(p, policy, param);
+		retval = security_task_setscheduler(p);
 		if (retval)
 			return retval;
 	}
@@ -4887,7 +4887,7 @@ long sched_setaffinity(pid_t pid, const struct cpumask *in_mask)
 	if (!check_same_owner(p) && !capable(CAP_SYS_NICE))
 		goto out_unlock;
 
-	retval = security_task_setscheduler(p, 0, NULL);
+	retval = security_task_setscheduler(p);
 	if (retval)
 		goto out_unlock;
 
diff --git a/security/commoncap.c b/security/commoncap.c
index 9d172e6e330c..5e632b4857e4 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -719,14 +719,11 @@ static int cap_safe_nice(struct task_struct *p)
 /**
  * cap_task_setscheduler - Detemine if scheduler policy change is permitted
  * @p: The task to affect
- * @policy: The policy to effect
- * @lp: The parameters to the scheduling policy
  *
  * Detemine if the requested scheduler policy change is permitted for the
  * specified task, returning 0 if permission is granted, -ve if denied.
  */
-int cap_task_setscheduler(struct task_struct *p, int policy,
-			   struct sched_param *lp)
+int cap_task_setscheduler(struct task_struct *p)
 {
 	return cap_safe_nice(p);
 }
diff --git a/security/security.c b/security/security.c
index 43b6463ebbfb..1cbcdfa4b015 100644
--- a/security/security.c
+++ b/security/security.c
@@ -778,10 +778,9 @@ int security_task_setrlimit(struct task_struct *p, unsigned int resource,
 	return security_ops->task_setrlimit(p, resource, new_rlim);
 }
 
-int security_task_setscheduler(struct task_struct *p,
-				int policy, struct sched_param *lp)
+int security_task_setscheduler(struct task_struct *p)
 {
-	return security_ops->task_setscheduler(p, policy, lp);
+	return security_ops->task_setscheduler(p);
 }
 
 int security_task_getscheduler(struct task_struct *p)
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 4796ddd4e721..db2b331de89a 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3354,11 +3354,11 @@ static int selinux_task_setrlimit(struct task_struct *p, unsigned int resource,
 	return 0;
 }
 
-static int selinux_task_setscheduler(struct task_struct *p, int policy, struct sched_param *lp)
+static int selinux_task_setscheduler(struct task_struct *p)
 {
 	int rc;
 
-	rc = cap_task_setscheduler(p, policy, lp);
+	rc = cap_task_setscheduler(p);
 	if (rc)
 		return rc;
 
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index c448d57ae2b7..174aec44bfac 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -1281,12 +1281,11 @@ static int smack_task_getioprio(struct task_struct *p)
  *
  * Return 0 if read access is permitted
  */
-static int smack_task_setscheduler(struct task_struct *p, int policy,
-				   struct sched_param *lp)
+static int smack_task_setscheduler(struct task_struct *p)
 {
 	int rc;
 
-	rc = cap_task_setscheduler(p, policy, lp);
+	rc = cap_task_setscheduler(p);
 	if (rc == 0)
 		rc = smk_curacc_on_task(p, MAY_WRITE);
 	return rc;
-- 
cgit v1.2.3


From 2606fd1fa5710205b23ee859563502aa18362447 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Wed, 13 Oct 2010 16:24:41 -0400
Subject: secmark: make secmark object handling generic

Right now secmark has lots of direct selinux calls.  Use all LSM calls and
remove all SELinux specific knowledge.  The only SELinux specific knowledge
we leave is the mode.  The only point is to make sure that other LSMs at
least test this generic code before they assume it works.  (They may also
have to make changes if they do not represent labels as strings)

Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Paul Moore <paul.moore@hp.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/netfilter/xt_SECMARK.h | 12 ++-----
 include/linux/security.h             | 25 ++++++++++++++
 include/linux/selinux.h              | 63 ------------------------------------
 net/netfilter/xt_CT.c                |  1 -
 net/netfilter/xt_SECMARK.c           | 35 ++++++++++----------
 security/capability.c                | 17 +++++++++-
 security/security.c                  | 18 +++++++++++
 security/selinux/exports.c           | 49 ----------------------------
 security/selinux/hooks.c             | 24 ++++++++++++++
 security/selinux/include/security.h  |  1 +
 10 files changed, 104 insertions(+), 141 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/xt_SECMARK.h b/include/linux/netfilter/xt_SECMARK.h
index 6fcd3448b186..989092bd6274 100644
--- a/include/linux/netfilter/xt_SECMARK.h
+++ b/include/linux/netfilter/xt_SECMARK.h
@@ -11,18 +11,12 @@
  * packets are being marked for.
  */
 #define SECMARK_MODE_SEL	0x01		/* SELinux */
-#define SECMARK_SELCTX_MAX	256
-
-struct xt_secmark_target_selinux_info {
-	__u32 selsid;
-	char selctx[SECMARK_SELCTX_MAX];
-};
+#define SECMARK_SECCTX_MAX	256
 
 struct xt_secmark_target_info {
 	__u8 mode;
-	union {
-		struct xt_secmark_target_selinux_info sel;
-	} u;
+	__u32 secid;
+	char secctx[SECMARK_SECCTX_MAX];
 };
 
 #endif /*_XT_SECMARK_H_target */
diff --git a/include/linux/security.h b/include/linux/security.h
index 294a0b228123..d70adc394f62 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -959,6 +959,12 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	Sets the new child socket's sid to the openreq sid.
  * @inet_conn_established:
  *	Sets the connection's peersid to the secmark on skb.
+ * @secmark_relabel_packet:
+ *	check if the process should be allowed to relabel packets to the given secid
+ * @security_secmark_refcount_inc
+ *	tells the LSM to increment the number of secmark labeling rules loaded
+ * @security_secmark_refcount_dec
+ *	tells the LSM to decrement the number of secmark labeling rules loaded
  * @req_classify_flow:
  *	Sets the flow's sid to the openreq sid.
  * @tun_dev_create:
@@ -1593,6 +1599,9 @@ struct security_operations {
 				  struct request_sock *req);
 	void (*inet_csk_clone) (struct sock *newsk, const struct request_sock *req);
 	void (*inet_conn_established) (struct sock *sk, struct sk_buff *skb);
+	int (*secmark_relabel_packet) (u32 secid);
+	void (*secmark_refcount_inc) (void);
+	void (*secmark_refcount_dec) (void);
 	void (*req_classify_flow) (const struct request_sock *req, struct flowi *fl);
 	int (*tun_dev_create)(void);
 	void (*tun_dev_post_create)(struct sock *sk);
@@ -2547,6 +2556,9 @@ void security_inet_csk_clone(struct sock *newsk,
 			const struct request_sock *req);
 void security_inet_conn_established(struct sock *sk,
 			struct sk_buff *skb);
+int security_secmark_relabel_packet(u32 secid);
+void security_secmark_refcount_inc(void);
+void security_secmark_refcount_dec(void);
 int security_tun_dev_create(void);
 void security_tun_dev_post_create(struct sock *sk);
 int security_tun_dev_attach(struct sock *sk);
@@ -2701,6 +2713,19 @@ static inline void security_inet_conn_established(struct sock *sk,
 {
 }
 
+static inline int security_secmark_relabel_packet(u32 secid)
+{
+	return 0;
+}
+
+static inline void security_secmark_refcount_inc(void)
+{
+}
+
+static inline void security_secmark_refcount_dec(void)
+{
+}
+
 static inline int security_tun_dev_create(void)
 {
 	return 0;
diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index 82e0f26a1299..44f459612690 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -20,75 +20,12 @@ struct kern_ipc_perm;
 
 #ifdef CONFIG_SECURITY_SELINUX
 
-/**
- *     selinux_string_to_sid - map a security context string to a security ID
- *     @str: the security context string to be mapped
- *     @sid: ID value returned via this.
- *
- *     Returns 0 if successful, with the SID stored in sid.  A value
- *     of zero for sid indicates no SID could be determined (but no error
- *     occurred).
- */
-int selinux_string_to_sid(char *str, u32 *sid);
-
-/**
- *     selinux_secmark_relabel_packet_permission - secmark permission check
- *     @sid: SECMARK ID value to be applied to network packet
- *
- *     Returns 0 if the current task is allowed to set the SECMARK label of
- *     packets with the supplied security ID.  Note that it is implicit that
- *     the packet is always being relabeled from the default unlabeled value,
- *     and that the access control decision is made in the AVC.
- */
-int selinux_secmark_relabel_packet_permission(u32 sid);
-
-/**
- *     selinux_secmark_refcount_inc - increments the secmark use counter
- *
- *     SELinux keeps track of the current SECMARK targets in use so it knows
- *     when to apply SECMARK label access checks to network packets.  This
- *     function incements this reference count to indicate that a new SECMARK
- *     target has been configured.
- */
-void selinux_secmark_refcount_inc(void);
-
-/**
- *     selinux_secmark_refcount_dec - decrements the secmark use counter
- *
- *     SELinux keeps track of the current SECMARK targets in use so it knows
- *     when to apply SECMARK label access checks to network packets.  This
- *     function decements this reference count to indicate that one of the
- *     existing SECMARK targets has been removed/flushed.
- */
-void selinux_secmark_refcount_dec(void);
-
 /**
  * selinux_is_enabled - is SELinux enabled?
  */
 bool selinux_is_enabled(void);
 #else
 
-static inline int selinux_string_to_sid(const char *str, u32 *sid)
-{
-       *sid = 0;
-       return 0;
-}
-
-static inline int selinux_secmark_relabel_packet_permission(u32 sid)
-{
-	return 0;
-}
-
-static inline void selinux_secmark_refcount_inc(void)
-{
-	return;
-}
-
-static inline void selinux_secmark_refcount_dec(void)
-{
-	return;
-}
-
 static inline bool selinux_is_enabled(void)
 {
 	return false;
diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c
index 0cb6053f02fd..782e51986a6f 100644
--- a/net/netfilter/xt_CT.c
+++ b/net/netfilter/xt_CT.c
@@ -9,7 +9,6 @@
 #include <linux/module.h>
 #include <linux/gfp.h>
 #include <linux/skbuff.h>
-#include <linux/selinux.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv6/ip6_tables.h>
 #include <linux/netfilter/x_tables.h>
diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c
index 364ad1600129..9faf5e050b79 100644
--- a/net/netfilter/xt_SECMARK.c
+++ b/net/netfilter/xt_SECMARK.c
@@ -14,8 +14,8 @@
  */
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 #include <linux/module.h>
+#include <linux/security.h>
 #include <linux/skbuff.h>
-#include <linux/selinux.h>
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/xt_SECMARK.h>
 
@@ -39,9 +39,8 @@ secmark_tg(struct sk_buff *skb, const struct xt_action_param *par)
 
 	switch (mode) {
 	case SECMARK_MODE_SEL:
-		secmark = info->u.sel.selsid;
+		secmark = info->secid;
 		break;
-
 	default:
 		BUG();
 	}
@@ -50,33 +49,33 @@ secmark_tg(struct sk_buff *skb, const struct xt_action_param *par)
 	return XT_CONTINUE;
 }
 
-static int checkentry_selinux(struct xt_secmark_target_info *info)
+static int checkentry_lsm(struct xt_secmark_target_info *info)
 {
 	int err;
-	struct xt_secmark_target_selinux_info *sel = &info->u.sel;
 
-	sel->selctx[SECMARK_SELCTX_MAX - 1] = '\0';
+	info->secctx[SECMARK_SECCTX_MAX - 1] = '\0';
+	info->secid = 0;
 
-	err = selinux_string_to_sid(sel->selctx, &sel->selsid);
+	err = security_secctx_to_secid(info->secctx, strlen(info->secctx),
+				       &info->secid);
 	if (err) {
 		if (err == -EINVAL)
-			pr_info("invalid SELinux context \'%s\'\n",
-				sel->selctx);
+			pr_info("invalid security context \'%s\'\n", info->secctx);
 		return err;
 	}
 
-	if (!sel->selsid) {
-		pr_info("unable to map SELinux context \'%s\'\n", sel->selctx);
+	if (!info->secid) {
+		pr_info("unable to map security context \'%s\'\n", info->secctx);
 		return -ENOENT;
 	}
 
-	err = selinux_secmark_relabel_packet_permission(sel->selsid);
+	err = security_secmark_relabel_packet(info->secid);
 	if (err) {
 		pr_info("unable to obtain relabeling permission\n");
 		return err;
 	}
 
-	selinux_secmark_refcount_inc();
+	security_secmark_refcount_inc();
 	return 0;
 }
 
@@ -100,16 +99,16 @@ static int secmark_tg_check(const struct xt_tgchk_param *par)
 
 	switch (info->mode) {
 	case SECMARK_MODE_SEL:
-		err = checkentry_selinux(info);
-		if (err)
-			return err;
 		break;
-
 	default:
 		pr_info("invalid mode: %hu\n", info->mode);
 		return -EINVAL;
 	}
 
+	err = checkentry_lsm(info);
+	if (err)
+		return err;
+
 	if (!mode)
 		mode = info->mode;
 	return 0;
@@ -119,7 +118,7 @@ static void secmark_tg_destroy(const struct xt_tgdtor_param *par)
 {
 	switch (mode) {
 	case SECMARK_MODE_SEL:
-		selinux_secmark_refcount_dec();
+		security_secmark_refcount_dec();
 	}
 }
 
diff --git a/security/capability.c b/security/capability.c
index 95a6599a37bb..30ae00fbecd5 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -677,7 +677,18 @@ static void cap_inet_conn_established(struct sock *sk, struct sk_buff *skb)
 {
 }
 
+static int cap_secmark_relabel_packet(u32 secid)
+{
+	return 0;
+}
 
+static void cap_secmark_refcount_inc(void)
+{
+}
+
+static void cap_secmark_refcount_dec(void)
+{
+}
 
 static void cap_req_classify_flow(const struct request_sock *req,
 				  struct flowi *fl)
@@ -777,7 +788,8 @@ static int cap_secid_to_secctx(u32 secid, char **secdata, u32 *seclen)
 
 static int cap_secctx_to_secid(const char *secdata, u32 seclen, u32 *secid)
 {
-	return -EOPNOTSUPP;
+	*secid = 0;
+	return 0;
 }
 
 static void cap_release_secctx(char *secdata, u32 seclen)
@@ -1018,6 +1030,9 @@ void __init security_fixup_ops(struct security_operations *ops)
 	set_to_cap_if_null(ops, inet_conn_request);
 	set_to_cap_if_null(ops, inet_csk_clone);
 	set_to_cap_if_null(ops, inet_conn_established);
+	set_to_cap_if_null(ops, secmark_relabel_packet);
+	set_to_cap_if_null(ops, secmark_refcount_inc);
+	set_to_cap_if_null(ops, secmark_refcount_dec);
 	set_to_cap_if_null(ops, req_classify_flow);
 	set_to_cap_if_null(ops, tun_dev_create);
 	set_to_cap_if_null(ops, tun_dev_post_create);
diff --git a/security/security.c b/security/security.c
index 1cbcdfa4b015..b50f472061a4 100644
--- a/security/security.c
+++ b/security/security.c
@@ -1136,6 +1136,24 @@ void security_inet_conn_established(struct sock *sk,
 	security_ops->inet_conn_established(sk, skb);
 }
 
+int security_secmark_relabel_packet(u32 secid)
+{
+	return security_ops->secmark_relabel_packet(secid);
+}
+EXPORT_SYMBOL(security_secmark_relabel_packet);
+
+void security_secmark_refcount_inc(void)
+{
+	security_ops->secmark_refcount_inc();
+}
+EXPORT_SYMBOL(security_secmark_refcount_inc);
+
+void security_secmark_refcount_dec(void)
+{
+	security_ops->secmark_refcount_dec();
+}
+EXPORT_SYMBOL(security_secmark_refcount_dec);
+
 int security_tun_dev_create(void)
 {
 	return security_ops->tun_dev_create();
diff --git a/security/selinux/exports.c b/security/selinux/exports.c
index c0a454aee1e0..90664385dead 100644
--- a/security/selinux/exports.c
+++ b/security/selinux/exports.c
@@ -11,58 +11,9 @@
  * it under the terms of the GNU General Public License version 2,
  * as published by the Free Software Foundation.
  */
-#include <linux/types.h>
-#include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/selinux.h>
-#include <linux/fs.h>
-#include <linux/ipc.h>
-#include <asm/atomic.h>
 
 #include "security.h"
-#include "objsec.h"
-
-/* SECMARK reference count */
-extern atomic_t selinux_secmark_refcount;
-
-int selinux_string_to_sid(char *str, u32 *sid)
-{
-	if (selinux_enabled)
-		return security_context_to_sid(str, strlen(str), sid);
-	else {
-		*sid = 0;
-		return 0;
-	}
-}
-EXPORT_SYMBOL_GPL(selinux_string_to_sid);
-
-int selinux_secmark_relabel_packet_permission(u32 sid)
-{
-	if (selinux_enabled) {
-		const struct task_security_struct *__tsec;
-		u32 tsid;
-
-		__tsec = current_security();
-		tsid = __tsec->sid;
-
-		return avc_has_perm(tsid, sid, SECCLASS_PACKET,
-				    PACKET__RELABELTO, NULL);
-	}
-	return 0;
-}
-EXPORT_SYMBOL_GPL(selinux_secmark_relabel_packet_permission);
-
-void selinux_secmark_refcount_inc(void)
-{
-	atomic_inc(&selinux_secmark_refcount);
-}
-EXPORT_SYMBOL_GPL(selinux_secmark_refcount_inc);
-
-void selinux_secmark_refcount_dec(void)
-{
-	atomic_dec(&selinux_secmark_refcount);
-}
-EXPORT_SYMBOL_GPL(selinux_secmark_refcount_dec);
 
 bool selinux_is_enabled(void)
 {
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index db2b331de89a..d9154cf90ae1 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -4279,6 +4279,27 @@ static void selinux_inet_conn_established(struct sock *sk, struct sk_buff *skb)
 	selinux_skb_peerlbl_sid(skb, family, &sksec->peer_sid);
 }
 
+static int selinux_secmark_relabel_packet(u32 sid)
+{
+	const struct task_security_struct *__tsec;
+	u32 tsid;
+
+	__tsec = current_security();
+	tsid = __tsec->sid;
+
+	return avc_has_perm(tsid, sid, SECCLASS_PACKET, PACKET__RELABELTO, NULL);
+}
+
+static void selinux_secmark_refcount_inc(void)
+{
+	atomic_inc(&selinux_secmark_refcount);
+}
+
+static void selinux_secmark_refcount_dec(void)
+{
+	atomic_dec(&selinux_secmark_refcount);
+}
+
 static void selinux_req_classify_flow(const struct request_sock *req,
 				      struct flowi *fl)
 {
@@ -5533,6 +5554,9 @@ static struct security_operations selinux_ops = {
 	.inet_conn_request =		selinux_inet_conn_request,
 	.inet_csk_clone =		selinux_inet_csk_clone,
 	.inet_conn_established =	selinux_inet_conn_established,
+	.secmark_relabel_packet =	selinux_secmark_relabel_packet,
+	.secmark_refcount_inc =		selinux_secmark_refcount_inc,
+	.secmark_refcount_dec =		selinux_secmark_refcount_dec,
 	.req_classify_flow =		selinux_req_classify_flow,
 	.tun_dev_create =		selinux_tun_dev_create,
 	.tun_dev_post_create = 		selinux_tun_dev_post_create,
diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h
index 4b66f19bb1f3..611a526afae7 100644
--- a/security/selinux/include/security.h
+++ b/security/selinux/include/security.h
@@ -9,6 +9,7 @@
 #define _SELINUX_SECURITY_H_
 
 #include <linux/magic.h>
+#include <linux/types.h>
 #include "flask.h"
 
 #define SECSID_NULL			0x00000000 /* unspecified SID */
-- 
cgit v1.2.3


From d5630b9d276bd389299ffea620b7c340ab19bcf5 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Wed, 13 Oct 2010 16:24:48 -0400
Subject: security: secid_to_secctx returns len when data is NULL

With the (long ago) interface change to have the secid_to_secctx functions
do the string allocation instead of having the caller do the allocation we
lost the ability to query the security server for the length of the
upcoming string.  The SECMARK code would like to allocate a netlink skb
with enough length to hold the string but it is just too unclean to do the
string allocation twice or to do the allocation the first time and hold
onto the string and slen.  This patch adds the ability to call
security_secid_to_secctx() with a NULL data pointer and it will just set
the slen pointer.

Signed-off-by: Eric Paris <eparis@redhat.com>
Reviewed-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/security.h       |  6 +++++-
 security/selinux/ss/services.c | 11 +++++++++--
 security/smack/smack_lsm.c     |  3 ++-
 3 files changed, 16 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/security.h b/include/linux/security.h
index d70adc394f62..b8246a8df7d2 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1285,9 +1285,13 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	Return 0 if permission is granted.
  *
  * @secid_to_secctx:
- *	Convert secid to security context.
+ *	Convert secid to security context.  If secdata is NULL the length of
+ *	the result will be returned in seclen, but no secdata will be returned.
+ *	This does mean that the length could change between calls to check the
+ *	length and the next call which actually allocates and returns the secdata.
  *	@secid contains the security ID.
  *	@secdata contains the pointer that stores the converted security context.
+ *	@seclen pointer which contains the length of the data
  * @secctx_to_secid:
  *	Convert security context to secid.
  *	@secid contains the pointer to the generated security ID.
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 494ff527c174..60964d79e5eb 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -991,7 +991,8 @@ static int context_struct_to_string(struct context *context, char **scontext, u3
 {
 	char *scontextp;
 
-	*scontext = NULL;
+	if (scontext)
+		*scontext = NULL;
 	*scontext_len = 0;
 
 	if (context->len) {
@@ -1008,6 +1009,9 @@ static int context_struct_to_string(struct context *context, char **scontext, u3
 	*scontext_len += strlen(policydb.p_type_val_to_name[context->type - 1]) + 1;
 	*scontext_len += mls_compute_context_len(context);
 
+	if (!scontext)
+		return 0;
+
 	/* Allocate space for the context; caller must free this space. */
 	scontextp = kmalloc(*scontext_len, GFP_ATOMIC);
 	if (!scontextp)
@@ -1047,7 +1051,8 @@ static int security_sid_to_context_core(u32 sid, char **scontext,
 	struct context *context;
 	int rc = 0;
 
-	*scontext = NULL;
+	if (scontext)
+		*scontext = NULL;
 	*scontext_len  = 0;
 
 	if (!ss_initialized) {
@@ -1055,6 +1060,8 @@ static int security_sid_to_context_core(u32 sid, char **scontext,
 			char *scontextp;
 
 			*scontext_len = strlen(initial_sid_to_string[sid]) + 1;
+			if (!scontext)
+				goto out;
 			scontextp = kmalloc(*scontext_len, GFP_ATOMIC);
 			if (!scontextp) {
 				rc = -ENOMEM;
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 174aec44bfac..bc39f4067af6 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -3004,7 +3004,8 @@ static int smack_secid_to_secctx(u32 secid, char **secdata, u32 *seclen)
 {
 	char *sp = smack_from_secid(secid);
 
-	*secdata = sp;
+	if (secdata)
+		*secdata = sp;
 	*seclen = strlen(sp);
 	return 0;
 }
-- 
cgit v1.2.3


From 1cc63249adfa957b34ca51effdee90ff8261d63f Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Wed, 13 Oct 2010 16:24:54 -0400
Subject: conntrack: export lsm context rather than internal secid via netlink

The conntrack code can export the internal secid to userspace.  These are
dynamic, can change on lsm changes, and have no meaning in userspace.  We
should instead be sending lsm contexts to userspace instead.  This patch sends
the secctx (rather than secid) to userspace over the netlink socket.  We use a
new field CTA_SECCTX and stop using the the old CTA_SECMARK field since it did
not send particularly useful information.

Signed-off-by: Eric Paris <eparis@redhat.com>
Reviewed-by: Paul Moore <paul.moore@hp.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/netfilter/nfnetlink_conntrack.h | 10 +++++-
 net/netfilter/nf_conntrack_netlink.c          | 46 +++++++++++++++++++++------
 2 files changed, 45 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index 9ed534c991b9..70cd0603911c 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -39,8 +39,9 @@ enum ctattr_type {
 	CTA_TUPLE_MASTER,
 	CTA_NAT_SEQ_ADJ_ORIG,
 	CTA_NAT_SEQ_ADJ_REPLY,
-	CTA_SECMARK,
+	CTA_SECMARK,		/* obsolete */
 	CTA_ZONE,
+	CTA_SECCTX,
 	__CTA_MAX
 };
 #define CTA_MAX (__CTA_MAX - 1)
@@ -172,4 +173,11 @@ enum ctattr_help {
 };
 #define CTA_HELP_MAX (__CTA_HELP_MAX - 1)
 
+enum ctattr_secctx {
+	CTA_SECCTX_UNSPEC,
+	CTA_SECCTX_NAME,
+	__CTA_SECCTX_MAX
+};
+#define CTA_SECCTX_MAX (__CTA_SECCTX_MAX - 1)
+
 #endif /* _IPCONNTRACK_NETLINK_H */
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 5bae1cd15eea..b3c628555cf3 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -22,6 +22,7 @@
 #include <linux/rculist_nulls.h>
 #include <linux/types.h>
 #include <linux/timer.h>
+#include <linux/security.h>
 #include <linux/skbuff.h>
 #include <linux/errno.h>
 #include <linux/netlink.h>
@@ -245,16 +246,31 @@ nla_put_failure:
 
 #ifdef CONFIG_NF_CONNTRACK_SECMARK
 static inline int
-ctnetlink_dump_secmark(struct sk_buff *skb, const struct nf_conn *ct)
+ctnetlink_dump_secctx(struct sk_buff *skb, const struct nf_conn *ct)
 {
-	NLA_PUT_BE32(skb, CTA_SECMARK, htonl(ct->secmark));
-	return 0;
+	struct nlattr *nest_secctx;
+	int len, ret;
+	char *secctx;
+
+	ret = security_secid_to_secctx(ct->secmark, &secctx, &len);
+	if (ret)
+		return ret;
+
+	ret = -1;
+	nest_secctx = nla_nest_start(skb, CTA_SECCTX | NLA_F_NESTED);
+	if (!nest_secctx)
+		goto nla_put_failure;
 
+	NLA_PUT_STRING(skb, CTA_SECCTX_NAME, secctx);
+	nla_nest_end(skb, nest_secctx);
+
+	ret = 0;
 nla_put_failure:
-	return -1;
+	security_release_secctx(secctx, len);
+	return ret;
 }
 #else
-#define ctnetlink_dump_secmark(a, b) (0)
+#define ctnetlink_dump_secctx(a, b) (0)
 #endif
 
 #define master_tuple(ct) &(ct->master->tuplehash[IP_CT_DIR_ORIGINAL].tuple)
@@ -391,7 +407,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq,
 	    ctnetlink_dump_protoinfo(skb, ct) < 0 ||
 	    ctnetlink_dump_helpinfo(skb, ct) < 0 ||
 	    ctnetlink_dump_mark(skb, ct) < 0 ||
-	    ctnetlink_dump_secmark(skb, ct) < 0 ||
+	    ctnetlink_dump_secctx(skb, ct) < 0 ||
 	    ctnetlink_dump_id(skb, ct) < 0 ||
 	    ctnetlink_dump_use(skb, ct) < 0 ||
 	    ctnetlink_dump_master(skb, ct) < 0 ||
@@ -437,6 +453,17 @@ ctnetlink_counters_size(const struct nf_conn *ct)
 	       ;
 }
 
+#ifdef CONFIG_NF_CONNTRACK_SECMARK
+static int ctnetlink_nlmsg_secctx_size(const struct nf_conn *ct)
+{
+	int len;
+
+	security_secid_to_secctx(ct->secmark, NULL, &len);
+
+	return sizeof(char) * len;
+}
+#endif
+
 static inline size_t
 ctnetlink_nlmsg_size(const struct nf_conn *ct)
 {
@@ -453,7 +480,8 @@ ctnetlink_nlmsg_size(const struct nf_conn *ct)
 	       + nla_total_size(0) /* CTA_HELP */
 	       + nla_total_size(NF_CT_HELPER_NAME_LEN) /* CTA_HELP_NAME */
 #ifdef CONFIG_NF_CONNTRACK_SECMARK
-	       + nla_total_size(sizeof(u_int32_t)) /* CTA_SECMARK */
+	       + nla_total_size(0) /* CTA_SECCTX */
+	       + nla_total_size(ctnetlink_nlmsg_secctx_size(ct)) /* CTA_SECCTX_NAME */
 #endif
 #ifdef CONFIG_NF_NAT_NEEDED
 	       + 2 * nla_total_size(0) /* CTA_NAT_SEQ_ADJ_ORIG|REPL */
@@ -554,11 +582,9 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item)
 		    && ctnetlink_dump_helpinfo(skb, ct) < 0)
 			goto nla_put_failure;
 
-#ifdef CONFIG_NF_CONNTRACK_SECMARK
 		if ((events & (1 << IPCT_SECMARK) || ct->secmark)
-		    && ctnetlink_dump_secmark(skb, ct) < 0)
+		    && ctnetlink_dump_secctx(skb, ct) < 0)
 			goto nla_put_failure;
-#endif
 
 		if (events & (1 << IPCT_RELATED) &&
 		    ctnetlink_dump_master(skb, ct) < 0)
-- 
cgit v1.2.3


From 686a0f3d71203bbfcc186900bbb8ac2cfc3d803c Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Wed, 13 Oct 2010 17:50:02 -0400
Subject: kernel: rounddown helper function

The roundup() helper function will round a given value up to a multiple of
another given value.  aka  roundup(11, 7) would give 14 = 7 * 2.  This new
function does the opposite.  It will round a given number down to the
nearest multiple of the second number: rounddown(11, 7) would give 7.

I need this in some future SELinux code and can carry the macro myself, but
figured I would put it in the core kernel so others might find and use it
if need be.

Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/kernel.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 2b0a35e6bc69..6d6eea7f7b1e 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -59,6 +59,12 @@ extern const char linux_proc_banner[];
 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
 #define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
 #define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y))
+#define rounddown(x, y) (				\
+{							\
+	typeof(x) __x = (x);				\
+	__x - (__x % (y));				\
+}							\
+)
 #define DIV_ROUND_CLOSEST(x, divisor)(			\
 {							\
 	typeof(divisor) __divisor = divisor;		\
-- 
cgit v1.2.3


From b28efd54d9d5c8005a29cd8782335beb9daaa32d Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Wed, 13 Oct 2010 17:50:08 -0400
Subject: kernel: roundup should only reference arguments once

Currently the roundup macro references it's arguments more than one time.
This patch changes it so it will only use its arguments once.

Suggested-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 include/linux/kernel.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 6d6eea7f7b1e..1759ba5adce8 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -58,7 +58,12 @@ extern const char linux_proc_banner[];
 
 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
 #define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
-#define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y))
+#define roundup(x, y) (					\
+{							\
+	typeof(y) __y = y;				\
+	(((x) + (__y - 1)) / __y) * __y;		\
+}							\
+)
 #define rounddown(x, y) (				\
 {							\
 	typeof(x) __x = (x);				\
-- 
cgit v1.2.3


From b738127dfb469bb9f595cdace30e7f881e8146b2 Mon Sep 17 00:00:00 2001
From: Jesse Gross <jesse@nicira.com>
Date: Wed, 20 Oct 2010 13:56:02 +0000
Subject: vlan: Rename VLAN_GROUP_ARRAY_LEN to VLAN_N_VID.

VLAN_GROUP_ARRAY_LEN is simply the number of possible vlan VIDs.
Since vlan groups will soon be more of an implementation detail
for vlan devices, rename the constant to be descriptive of its
actual purpose.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/benet/be.h             |  2 +-
 drivers/net/benet/be_main.c        |  2 +-
 drivers/net/e1000/e1000_main.c     |  2 +-
 drivers/net/e1000e/netdev.c        |  2 +-
 drivers/net/igb/igb_main.c         |  2 +-
 drivers/net/igbvf/netdev.c         |  2 +-
 drivers/net/ixgb/ixgb_main.c       |  2 +-
 drivers/net/ixgbe/ixgbe_main.c     |  2 +-
 drivers/net/ixgbevf/ixgbevf_main.c |  2 +-
 drivers/net/qlcnic/qlcnic_main.c   |  2 +-
 drivers/net/vmxnet3/vmxnet3_drv.c  |  2 +-
 drivers/net/vxge/vxge-main.c       |  2 +-
 drivers/s390/net/qeth_l3_main.c    |  6 +++---
 include/linux/if_vlan.h            |  4 ++--
 net/8021q/vlan.c                   | 16 ++++++++--------
 net/bridge/netfilter/ebt_vlan.c    |  4 ++--
 16 files changed, 27 insertions(+), 27 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h
index 1afabb1e6620..59a17b569b7f 100644
--- a/drivers/net/benet/be.h
+++ b/drivers/net/benet/be.h
@@ -263,7 +263,7 @@ struct be_adapter {
 	struct vlan_group *vlan_grp;
 	u16 vlans_added;
 	u16 max_vlans;	/* Number of vlans supported */
-	u8 vlan_tag[VLAN_GROUP_ARRAY_LEN];
+	u8 vlan_tag[VLAN_N_VID];
 	struct be_dma_mem mc_cmd_mem;
 
 	struct be_dma_mem stats_cmd;
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index 9a1cd28b426d..4b59e53890b2 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -626,7 +626,7 @@ static int be_vid_config(struct be_adapter *adapter, bool vf, u32 vf_num)
 
 	if (adapter->vlans_added <= adapter->max_vlans)  {
 		/* Construct VLAN Table to give to HW */
-		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+		for (i = 0; i < VLAN_N_VID; i++) {
 			if (adapter->vlan_tag[i]) {
 				vtag[ntags] = cpu_to_le16(i);
 				ntags++;
diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index cb3f84b81793..232ac2dcb497 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -4541,7 +4541,7 @@ static void e1000_restore_vlan(struct e1000_adapter *adapter)
 
 	if (adapter->vlgrp) {
 		u16 vid;
-		for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+		for (vid = 0; vid < VLAN_N_VID; vid++) {
 			if (!vlan_group_get_device(adapter->vlgrp, vid))
 				continue;
 			e1000_vlan_rx_add_vid(adapter->netdev, vid);
diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index 992b622fe205..5d6cdea0eb14 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -2545,7 +2545,7 @@ static void e1000_restore_vlan(struct e1000_adapter *adapter)
 	if (!adapter->vlgrp)
 		return;
 
-	for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+	for (vid = 0; vid < VLAN_N_VID; vid++) {
 		if (!vlan_group_get_device(adapter->vlgrp, vid))
 			continue;
 		e1000_vlan_rx_add_vid(adapter->netdev, vid);
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index b8dccc0ac089..0f0939caf091 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -6153,7 +6153,7 @@ static void igb_restore_vlan(struct igb_adapter *adapter)
 
 	if (adapter->vlgrp) {
 		u16 vid;
-		for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+		for (vid = 0; vid < VLAN_N_VID; vid++) {
 			if (!vlan_group_get_device(adapter->vlgrp, vid))
 				continue;
 			igb_vlan_rx_add_vid(adapter->netdev, vid);
diff --git a/drivers/net/igbvf/netdev.c b/drivers/net/igbvf/netdev.c
index 6693323a6cf5..ebfaa68ee630 100644
--- a/drivers/net/igbvf/netdev.c
+++ b/drivers/net/igbvf/netdev.c
@@ -1254,7 +1254,7 @@ static void igbvf_restore_vlan(struct igbvf_adapter *adapter)
 	if (!adapter->vlgrp)
 		return;
 
-	for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+	for (vid = 0; vid < VLAN_N_VID; vid++) {
 		if (!vlan_group_get_device(adapter->vlgrp, vid))
 			continue;
 		igbvf_vlan_rx_add_vid(adapter->netdev, vid);
diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c
index 80e62578ffa0..666207a9c039 100644
--- a/drivers/net/ixgb/ixgb_main.c
+++ b/drivers/net/ixgb/ixgb_main.c
@@ -2223,7 +2223,7 @@ ixgb_restore_vlan(struct ixgb_adapter *adapter)
 
 	if (adapter->vlgrp) {
 		u16 vid;
-		for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+		for (vid = 0; vid < VLAN_N_VID; vid++) {
 			if (!vlan_group_get_device(adapter->vlgrp, vid))
 				continue;
 			ixgb_vlan_rx_add_vid(adapter->netdev, vid);
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 790a0dae1247..1d424428079f 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -3185,7 +3185,7 @@ static void ixgbe_restore_vlan(struct ixgbe_adapter *adapter)
 
 	if (adapter->vlgrp) {
 		u16 vid;
-		for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+		for (vid = 0; vid < VLAN_N_VID; vid++) {
 			if (!vlan_group_get_device(adapter->vlgrp, vid))
 				continue;
 			ixgbe_vlan_rx_add_vid(adapter->netdev, vid);
diff --git a/drivers/net/ixgbevf/ixgbevf_main.c b/drivers/net/ixgbevf/ixgbevf_main.c
index 0866a1cf4d7b..78bfbe42ca9b 100644
--- a/drivers/net/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ixgbevf/ixgbevf_main.c
@@ -1495,7 +1495,7 @@ static void ixgbevf_restore_vlan(struct ixgbevf_adapter *adapter)
 
 	if (adapter->vlgrp) {
 		u16 vid;
-		for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+		for (vid = 0; vid < VLAN_N_VID; vid++) {
 			if (!vlan_group_get_device(adapter->vlgrp, vid))
 				continue;
 			ixgbevf_vlan_rx_add_vid(adapter->netdev, vid);
diff --git a/drivers/net/qlcnic/qlcnic_main.c b/drivers/net/qlcnic/qlcnic_main.c
index 4aada0b8ceb1..f047c7c48314 100644
--- a/drivers/net/qlcnic/qlcnic_main.c
+++ b/drivers/net/qlcnic/qlcnic_main.c
@@ -4093,7 +4093,7 @@ qlcnic_restore_indev_addr(struct net_device *netdev, unsigned long event)
 	if (!adapter->vlgrp)
 		return;
 
-	for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+	for (vid = 0; vid < VLAN_N_VID; vid++) {
 		dev = vlan_group_get_device(adapter->vlgrp, vid);
 		if (!dev)
 			continue;
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index 198ce92af0c3..b1de73b1bf1a 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -1634,7 +1634,7 @@ vmxnet3_restore_vlan(struct vmxnet3_adapter *adapter)
 		u32 *vfTable = adapter->shared->devRead.rxFilterConf.vfTable;
 		bool activeVlan = false;
 
-		for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+		for (vid = 0; vid < VLAN_N_VID; vid++) {
 			if (vlan_group_get_device(adapter->vlan_grp, vid)) {
 				VMXNET3_SET_VFTABLE_ENTRY(vfTable, vid);
 				activeVlan = true;
diff --git a/drivers/net/vxge/vxge-main.c b/drivers/net/vxge/vxge-main.c
index 5378b849f54f..0bda7fe05d4b 100644
--- a/drivers/net/vxge/vxge-main.c
+++ b/drivers/net/vxge/vxge-main.c
@@ -1862,7 +1862,7 @@ enum vxge_hw_status vxge_restore_vpath_vid_table(struct vxge_vpath *vpath)
 
 	if (vdev->vlgrp && vpath->is_open) {
 
-		for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) {
+		for (vid = 0; vid < VLAN_N_VID; vid++) {
 			if (!vlan_group_get_device(vdev->vlgrp, vid))
 				continue;
 			/* Add these vlan to the vid table */
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index c094707fcbff..74d1401a5d5e 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1820,7 +1820,7 @@ static void qeth_l3_add_vlan_mc(struct qeth_card *card)
 		return;
 
 	vg = card->vlangrp;
-	for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+	for (i = 0; i < VLAN_N_VID; i++) {
 		struct net_device *netdev = vlan_group_get_device(vg, i);
 		if (netdev == NULL ||
 		    !(netdev->flags & IFF_UP))
@@ -1883,7 +1883,7 @@ static void qeth_l3_add_vlan_mc6(struct qeth_card *card)
 		return;
 
 	vg = card->vlangrp;
-	for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+	for (i = 0; i < VLAN_N_VID; i++) {
 		struct net_device *netdev = vlan_group_get_device(vg, i);
 		if (netdev == NULL ||
 		    !(netdev->flags & IFF_UP))
@@ -2247,7 +2247,7 @@ static int qeth_l3_verify_vlan_dev(struct net_device *dev,
 	if (!vg)
 		return rc;
 
-	for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+	for (i = 0; i < VLAN_N_VID; i++) {
 		if (vlan_group_get_device(vg, i) == dev) {
 			rc = QETH_VLAN_CARD;
 			break;
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index a52320751bfc..494cce866564 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -68,6 +68,7 @@ static inline struct vlan_ethhdr *vlan_eth_hdr(const struct sk_buff *skb)
 #define VLAN_CFI_MASK		0x1000 /* Canonical Format Indicator */
 #define VLAN_TAG_PRESENT	VLAN_CFI_MASK
 #define VLAN_VID_MASK		0x0fff /* VLAN Identifier */
+#define VLAN_N_VID		4096
 
 /* found in socket.c */
 extern void vlan_ioctl_set(int (*hook)(struct net *, void __user *));
@@ -76,9 +77,8 @@ extern void vlan_ioctl_set(int (*hook)(struct net *, void __user *));
  * depends on completely exhausting the VLAN identifier space.  Thus
  * it gives constant time look-up, but in many cases it wastes memory.
  */
-#define VLAN_GROUP_ARRAY_LEN          4096
 #define VLAN_GROUP_ARRAY_SPLIT_PARTS  8
-#define VLAN_GROUP_ARRAY_PART_LEN     (VLAN_GROUP_ARRAY_LEN/VLAN_GROUP_ARRAY_SPLIT_PARTS)
+#define VLAN_GROUP_ARRAY_PART_LEN     (VLAN_N_VID/VLAN_GROUP_ARRAY_SPLIT_PARTS)
 
 struct vlan_group {
 	struct net_device	*real_dev; /* The ethernet(like) device
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 25c21332e9c3..54f22d820ef5 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -439,7 +439,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 	switch (event) {
 	case NETDEV_CHANGE:
 		/* Propagate real device state to vlan devices */
-		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+		for (i = 0; i < VLAN_N_VID; i++) {
 			vlandev = vlan_group_get_device(grp, i);
 			if (!vlandev)
 				continue;
@@ -450,7 +450,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 
 	case NETDEV_CHANGEADDR:
 		/* Adjust unicast filters on underlying device */
-		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+		for (i = 0; i < VLAN_N_VID; i++) {
 			vlandev = vlan_group_get_device(grp, i);
 			if (!vlandev)
 				continue;
@@ -464,7 +464,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 		break;
 
 	case NETDEV_CHANGEMTU:
-		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+		for (i = 0; i < VLAN_N_VID; i++) {
 			vlandev = vlan_group_get_device(grp, i);
 			if (!vlandev)
 				continue;
@@ -478,7 +478,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 
 	case NETDEV_FEAT_CHANGE:
 		/* Propagate device features to underlying device */
-		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+		for (i = 0; i < VLAN_N_VID; i++) {
 			vlandev = vlan_group_get_device(grp, i);
 			if (!vlandev)
 				continue;
@@ -490,7 +490,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 
 	case NETDEV_DOWN:
 		/* Put all VLANs for this dev in the down state too.  */
-		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+		for (i = 0; i < VLAN_N_VID; i++) {
 			vlandev = vlan_group_get_device(grp, i);
 			if (!vlandev)
 				continue;
@@ -508,7 +508,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 
 	case NETDEV_UP:
 		/* Put all VLANs for this dev in the up state too.  */
-		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+		for (i = 0; i < VLAN_N_VID; i++) {
 			vlandev = vlan_group_get_device(grp, i);
 			if (!vlandev)
 				continue;
@@ -532,7 +532,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 		/* Delete all VLANs for this dev. */
 		grp->killall = 1;
 
-		for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) {
+		for (i = 0; i < VLAN_N_VID; i++) {
 			vlandev = vlan_group_get_device(grp, i);
 			if (!vlandev)
 				continue;
@@ -540,7 +540,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 			/* unregistration of last vlan destroys group, abort
 			 * afterwards */
 			if (grp->nr_vlans == 1)
-				i = VLAN_GROUP_ARRAY_LEN;
+				i = VLAN_N_VID;
 
 			unregister_vlan_dev(vlandev, &list);
 		}
diff --git a/net/bridge/netfilter/ebt_vlan.c b/net/bridge/netfilter/ebt_vlan.c
index cc11d6b8e507..eae67bf0446c 100644
--- a/net/bridge/netfilter/ebt_vlan.c
+++ b/net/bridge/netfilter/ebt_vlan.c
@@ -118,10 +118,10 @@ static int ebt_vlan_mt_check(const struct xt_mtchk_param *par)
 	 * 0 - The null VLAN ID.
 	 * 1 - The default Port VID (PVID)
 	 * 0x0FFF - Reserved for implementation use.
-	 * if_vlan.h: VLAN_GROUP_ARRAY_LEN 4096. */
+	 * if_vlan.h: VLAN_N_VID 4096. */
 	if (GET_BITMASK(EBT_VLAN_ID)) {
 		if (!!info->id) { /* if id!=0 => check vid range */
-			if (info->id > VLAN_GROUP_ARRAY_LEN) {
+			if (info->id > VLAN_N_VID) {
 				pr_debug("id %d is out of range (1-4096)\n",
 					 info->id);
 				return -EINVAL;
-- 
cgit v1.2.3


From 7b9c60903714bf0a19d746b228864bad3497284e Mon Sep 17 00:00:00 2001
From: Jesse Gross <jesse@nicira.com>
Date: Wed, 20 Oct 2010 13:56:04 +0000
Subject: vlan: Enable software emulation for vlan accleration.

Currently users of hardware vlan accleration need to know whether
the device supports it before generating packets.  However, vlan
acceleration will soon be available in a more flexible manner so
knowing ahead of time becomes much more difficult.  This adds
a software fallback path for vlan packets on devices without the
necessary offloading support, similar to other types of hardware
accleration.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 14 +++++++++++---
 net/core/dev.c            | 36 +++++++++++++++++++++++++++++++++---
 2 files changed, 44 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 880d56565828..2861565a27d9 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2248,9 +2248,17 @@ static inline int skb_gso_ok(struct sk_buff *skb, int features)
 
 static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb)
 {
-	return skb_is_gso(skb) &&
-	       (!skb_gso_ok(skb, dev->features) ||
-		unlikely(skb->ip_summed != CHECKSUM_PARTIAL));
+	if (skb_is_gso(skb)) {
+		int features = dev->features;
+
+		if (skb->protocol == htons(ETH_P_8021Q) || skb->vlan_tci)
+			features &= dev->vlan_features;
+
+		return (!skb_gso_ok(skb, features) ||
+			unlikely(skb->ip_summed != CHECKSUM_PARTIAL));
+	}
+
+	return 0;
 }
 
 static inline void netif_set_gso_max_size(struct net_device *dev,
diff --git a/net/core/dev.c b/net/core/dev.c
index 4c3ac53e4b16..1bfd96b1fbd4 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1694,7 +1694,12 @@ static bool can_checksum_protocol(unsigned long features, __be16 protocol)
 
 static bool dev_can_checksum(struct net_device *dev, struct sk_buff *skb)
 {
-	if (can_checksum_protocol(dev->features, skb->protocol))
+	int features = dev->features;
+
+	if (vlan_tx_tag_present(skb))
+		features &= dev->vlan_features;
+
+	if (can_checksum_protocol(features, skb->protocol))
 		return true;
 
 	if (skb->protocol == htons(ETH_P_8021Q)) {
@@ -1793,6 +1798,16 @@ struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features)
 	__be16 type = skb->protocol;
 	int err;
 
+	if (type == htons(ETH_P_8021Q)) {
+		struct vlan_ethhdr *veh;
+
+		if (unlikely(!pskb_may_pull(skb, VLAN_ETH_HLEN)))
+			return ERR_PTR(-EINVAL);
+
+		veh = (struct vlan_ethhdr *)skb->data;
+		type = veh->h_vlan_encapsulated_proto;
+	}
+
 	skb_reset_mac_header(skb);
 	skb->mac_len = skb->network_header - skb->mac_header;
 	__skb_pull(skb, skb->mac_len);
@@ -1964,9 +1979,14 @@ static inline void skb_orphan_try(struct sk_buff *skb)
 static inline int skb_needs_linearize(struct sk_buff *skb,
 				      struct net_device *dev)
 {
+	int features = dev->features;
+
+	if (skb->protocol == htons(ETH_P_8021Q) || vlan_tx_tag_present(skb))
+		features &= dev->vlan_features;
+
 	return skb_is_nonlinear(skb) &&
-	       ((skb_has_frag_list(skb) && !(dev->features & NETIF_F_FRAGLIST)) ||
-	        (skb_shinfo(skb)->nr_frags && (!(dev->features & NETIF_F_SG) ||
+	       ((skb_has_frag_list(skb) && !(features & NETIF_F_FRAGLIST)) ||
+		(skb_shinfo(skb)->nr_frags && (!(features & NETIF_F_SG) ||
 					      illegal_highdma(dev, skb))));
 }
 
@@ -1989,6 +2009,15 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
 
 		skb_orphan_try(skb);
 
+		if (vlan_tx_tag_present(skb) &&
+		    !(dev->features & NETIF_F_HW_VLAN_TX)) {
+			skb = __vlan_put_tag(skb, vlan_tx_tag_get(skb));
+			if (unlikely(!skb))
+				goto out;
+
+			skb->vlan_tci = 0;
+		}
+
 		if (netif_needs_gso(dev, skb)) {
 			if (unlikely(dev_gso_segment(skb)))
 				goto out_kfree_skb;
@@ -2050,6 +2079,7 @@ out_kfree_gso_skb:
 		skb->destructor = DEV_GSO_CB(skb)->destructor;
 out_kfree_skb:
 	kfree_skb(skb);
+out:
 	return rc;
 }
 
-- 
cgit v1.2.3


From 65ac6a5fa658b90f1be700c55e7cd72e4611015d Mon Sep 17 00:00:00 2001
From: Jesse Gross <jesse@nicira.com>
Date: Wed, 20 Oct 2010 13:56:05 +0000
Subject: vlan: Avoid hash table lookup to find group.

A struct net_device always maps to zero or one vlan groups and we
always know the device when we are looking up a group.  We currently
do a hash table lookup on the device to find the group but it is
much simpler to just store a pointer.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h   | 19 ++++++++++++++
 include/linux/netdevice.h |  5 +++-
 net/8021q/vlan.c          | 64 ++++++++---------------------------------------
 net/8021q/vlan.h          | 17 -------------
 net/8021q/vlan_dev.c      |  2 +-
 5 files changed, 34 insertions(+), 73 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 494cce866564..4047781da727 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -16,6 +16,7 @@
 #ifdef __KERNEL__
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
 
 #define VLAN_HLEN	4		/* The additional bytes (on top of the Ethernet header)
 					 * that VLAN requires.
@@ -114,6 +115,18 @@ static inline void vlan_group_set_device(struct vlan_group *vg,
 #define vlan_tx_tag_get(__skb)		((__skb)->vlan_tci & ~VLAN_TAG_PRESENT)
 
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
+/* Must be invoked with rcu_read_lock or with RTNL. */
+static inline struct net_device *vlan_find_dev(struct net_device *real_dev,
+					       u16 vlan_id)
+{
+	struct vlan_group *grp = rcu_dereference_rtnl(real_dev->vlgrp);
+
+	if (grp)
+		return vlan_group_get_device(grp, vlan_id);
+
+	return NULL;
+}
+
 extern struct net_device *vlan_dev_real_dev(const struct net_device *dev);
 extern u16 vlan_dev_vlan_id(const struct net_device *dev);
 
@@ -128,6 +141,12 @@ vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp,
 	       unsigned int vlan_tci);
 
 #else
+static inline struct net_device *vlan_find_dev(struct net_device *real_dev,
+					       u16 vlan_id)
+{
+	return NULL;
+}
+
 static inline struct net_device *vlan_dev_real_dev(const struct net_device *dev)
 {
 	BUG();
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 2861565a27d9..9c78312ce142 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -942,7 +942,10 @@ struct net_device {
 
 
 	/* Protocol specific pointers */
-	
+
+#if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
+	struct vlan_group	*vlgrp;		/* VLAN group */
+#endif
 #ifdef CONFIG_NET_DSA
 	void			*dsa_ptr;	/* dsa specific data */
 #endif
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 54f22d820ef5..f862dccf6bb0 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -44,9 +44,6 @@
 
 int vlan_net_id __read_mostly;
 
-/* Our listing of VLAN group(s) */
-static struct hlist_head vlan_group_hash[VLAN_GRP_HASH_SIZE];
-
 const char vlan_fullname[] = "802.1Q VLAN Support";
 const char vlan_version[] = DRV_VERSION;
 static const char vlan_copyright[] = "Ben Greear <greearb@candelatech.com>";
@@ -59,40 +56,6 @@ static struct packet_type vlan_packet_type __read_mostly = {
 
 /* End of global variables definitions. */
 
-static inline unsigned int vlan_grp_hashfn(unsigned int idx)
-{
-	return ((idx >> VLAN_GRP_HASH_SHIFT) ^ idx) & VLAN_GRP_HASH_MASK;
-}
-
-/* Must be invoked with RCU read lock (no preempt) */
-static struct vlan_group *__vlan_find_group(struct net_device *real_dev)
-{
-	struct vlan_group *grp;
-	struct hlist_node *n;
-	int hash = vlan_grp_hashfn(real_dev->ifindex);
-
-	hlist_for_each_entry_rcu(grp, n, &vlan_group_hash[hash], hlist) {
-		if (grp->real_dev == real_dev)
-			return grp;
-	}
-
-	return NULL;
-}
-
-/*  Find the protocol handler.  Assumes VID < VLAN_VID_MASK.
- *
- * Must be invoked with RCU read lock (no preempt)
- */
-struct net_device *__find_vlan_dev(struct net_device *real_dev, u16 vlan_id)
-{
-	struct vlan_group *grp = __vlan_find_group(real_dev);
-
-	if (grp)
-		return vlan_group_get_device(grp, vlan_id);
-
-	return NULL;
-}
-
 static void vlan_group_free(struct vlan_group *grp)
 {
 	int i;
@@ -111,8 +74,6 @@ static struct vlan_group *vlan_group_alloc(struct net_device *real_dev)
 		return NULL;
 
 	grp->real_dev = real_dev;
-	hlist_add_head_rcu(&grp->hlist,
-			&vlan_group_hash[vlan_grp_hashfn(real_dev->ifindex)]);
 	return grp;
 }
 
@@ -151,7 +112,7 @@ void unregister_vlan_dev(struct net_device *dev, struct list_head *head)
 
 	ASSERT_RTNL();
 
-	grp = __vlan_find_group(real_dev);
+	grp = real_dev->vlgrp;
 	BUG_ON(!grp);
 
 	/* Take it out of our own structures, but be sure to interlock with
@@ -173,11 +134,10 @@ void unregister_vlan_dev(struct net_device *dev, struct list_head *head)
 	if (grp->nr_vlans == 0) {
 		vlan_gvrp_uninit_applicant(real_dev);
 
+		rcu_assign_pointer(real_dev->vlgrp, NULL);
 		if (real_dev->features & NETIF_F_HW_VLAN_RX)
 			ops->ndo_vlan_rx_register(real_dev, NULL);
 
-		hlist_del_rcu(&grp->hlist);
-
 		/* Free the group, after all cpu's are done. */
 		call_rcu(&grp->rcu, vlan_rcu_free);
 	}
@@ -207,7 +167,7 @@ int vlan_check_real_dev(struct net_device *real_dev, u16 vlan_id)
 		return -EOPNOTSUPP;
 	}
 
-	if (__find_vlan_dev(real_dev, vlan_id) != NULL)
+	if (vlan_find_dev(real_dev, vlan_id) != NULL)
 		return -EEXIST;
 
 	return 0;
@@ -222,7 +182,7 @@ int register_vlan_dev(struct net_device *dev)
 	struct vlan_group *grp, *ngrp = NULL;
 	int err;
 
-	grp = __vlan_find_group(real_dev);
+	grp = real_dev->vlgrp;
 	if (!grp) {
 		ngrp = grp = vlan_group_alloc(real_dev);
 		if (!grp)
@@ -252,8 +212,11 @@ int register_vlan_dev(struct net_device *dev)
 	vlan_group_set_device(grp, vlan_id, dev);
 	grp->nr_vlans++;
 
-	if (ngrp && real_dev->features & NETIF_F_HW_VLAN_RX)
-		ops->ndo_vlan_rx_register(real_dev, ngrp);
+	if (ngrp) {
+		if (real_dev->features & NETIF_F_HW_VLAN_RX)
+			ops->ndo_vlan_rx_register(real_dev, ngrp);
+		rcu_assign_pointer(real_dev->vlgrp, ngrp);
+	}
 	if (real_dev->features & NETIF_F_HW_VLAN_FILTER)
 		ops->ndo_vlan_rx_add_vid(real_dev, vlan_id);
 
@@ -264,7 +227,6 @@ out_uninit_applicant:
 		vlan_gvrp_uninit_applicant(real_dev);
 out_free_group:
 	if (ngrp) {
-		hlist_del_rcu(&ngrp->hlist);
 		/* Free the group, after all cpu's are done. */
 		call_rcu(&ngrp->rcu, vlan_rcu_free);
 	}
@@ -428,7 +390,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 		dev->netdev_ops->ndo_vlan_rx_add_vid(dev, 0);
 	}
 
-	grp = __vlan_find_group(dev);
+	grp = dev->vlgrp;
 	if (!grp)
 		goto out;
 
@@ -746,8 +708,6 @@ err0:
 
 static void __exit vlan_cleanup_module(void)
 {
-	unsigned int i;
-
 	vlan_ioctl_set(NULL);
 	vlan_netlink_fini();
 
@@ -755,10 +715,6 @@ static void __exit vlan_cleanup_module(void)
 
 	dev_remove_pack(&vlan_packet_type);
 
-	/* This table must be empty if there are no module references left. */
-	for (i = 0; i < VLAN_GRP_HASH_SIZE; i++)
-		BUG_ON(!hlist_empty(&vlan_group_hash[i]));
-
 	unregister_pernet_subsys(&vlan_net_ops);
 	rcu_barrier(); /* Wait for completion of call_rcu()'s */
 
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 8d9503ad01da..db01b3181fdc 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -72,23 +72,6 @@ static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
 	return netdev_priv(dev);
 }
 
-#define VLAN_GRP_HASH_SHIFT	5
-#define VLAN_GRP_HASH_SIZE	(1 << VLAN_GRP_HASH_SHIFT)
-#define VLAN_GRP_HASH_MASK	(VLAN_GRP_HASH_SIZE - 1)
-
-/*  Find a VLAN device by the MAC address of its Ethernet device, and
- *  it's VLAN ID.  The default configuration is to have VLAN's scope
- *  to be box-wide, so the MAC will be ignored.  The mac will only be
- *  looked at if we are configured to have a separate set of VLANs per
- *  each MAC addressable interface.  Note that this latter option does
- *  NOT follow the spec for VLANs, but may be useful for doing very
- *  large quantities of VLAN MUX/DEMUX onto FrameRelay or ATM PVCs.
- *
- *  Must be invoked with rcu_read_lock (ie preempt disabled)
- *  or with RTNL.
- */
-struct net_device *__find_vlan_dev(struct net_device *real_dev, u16 vlan_id);
-
 /* found in vlan_dev.c */
 int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 		  struct packet_type *ptype, struct net_device *orig_dev);
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index f54251edd40d..14e3d1fa07a0 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -158,7 +158,7 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 	vlan_id = vlan_tci & VLAN_VID_MASK;
 
 	rcu_read_lock();
-	vlan_dev = __find_vlan_dev(dev, vlan_id);
+	vlan_dev = vlan_find_dev(dev, vlan_id);
 
 	/* If the VLAN device is defined, we use it.
 	 * If not, and the VID is 0, it is a 802.1p packet (not
-- 
cgit v1.2.3


From 3701e51382a026cba10c60b03efabe534fba4ca4 Mon Sep 17 00:00:00 2001
From: Jesse Gross <jesse@nicira.com>
Date: Wed, 20 Oct 2010 13:56:06 +0000
Subject: vlan: Centralize handling of hardware acceleration.

Currently each driver that is capable of vlan hardware acceleration
must be aware of the vlan groups that are configured and then pass
the stripped tag to a specialized receive function.  This is

different from other types of hardware offload in that it places a
significant amount of knowledge in the driver itself rather keeping
it in the networking core.

This makes vlan offloading function more similarly to other forms
of offloading (such as checksum offloading or TSO) by doing the
following:
* On receive, stripped vlans are passed directly to the network
core, without attempting to check for vlan groups or reconstructing
the header if no group
* vlans are made less special by folding the logic into the main
receive routines
* On transmit, the device layer will add the vlan header in software
if the hardware doesn't support it, instead of spreading that logic
out in upper layers, such as bonding.

There are a number of advantages to this:
* Fixes all bugs with drivers incorrectly dropping vlan headers at once.
* Avoids having to disable VLAN acceleration when in promiscuous mode
(good for bridging since it always puts devices in promiscuous mode).
* Keeps VLAN tag separate until given to ultimate consumer, which
avoids needing to do header reconstruction as in tg3 unless absolutely
necessary.
* Consolidates common code in core networking.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h   |   6 ++-
 include/linux/netdevice.h |   1 -
 net/8021q/vlan.c          |   9 +---
 net/8021q/vlan_core.c     | 125 ++++++++++------------------------------------
 net/core/dev.c            |  47 ++++++-----------
 5 files changed, 48 insertions(+), 140 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 4047781da727..a0d9786c202d 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -132,7 +132,7 @@ extern u16 vlan_dev_vlan_id(const struct net_device *dev);
 
 extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 			     u16 vlan_tci, int polling);
-extern void vlan_hwaccel_do_receive(struct sk_buff *skb);
+extern bool vlan_hwaccel_do_receive(struct sk_buff **skb);
 extern gro_result_t
 vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp,
 		 unsigned int vlan_tci, struct sk_buff *skb);
@@ -166,8 +166,10 @@ static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 	return NET_XMIT_SUCCESS;
 }
 
-static inline void vlan_hwaccel_do_receive(struct sk_buff *skb)
+static inline bool vlan_hwaccel_do_receive(struct sk_buff **skb)
 {
+	BUG();
+	return false;
 }
 
 static inline gro_result_t
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9c78312ce142..ed7db7eebbf3 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1768,7 +1768,6 @@ extern int netdev_rx_handler_register(struct net_device *dev,
 				      void *rx_handler_data);
 extern void netdev_rx_handler_unregister(struct net_device *dev);
 
-extern void		netif_nit_deliver(struct sk_buff *skb);
 extern int		dev_valid_name(const char *name);
 extern int		dev_ioctl(struct net *net, unsigned int cmd, void __user *);
 extern int		dev_ethtool(struct net *net, struct ifreq *);
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index f862dccf6bb0..05b867e43757 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -135,7 +135,7 @@ void unregister_vlan_dev(struct net_device *dev, struct list_head *head)
 		vlan_gvrp_uninit_applicant(real_dev);
 
 		rcu_assign_pointer(real_dev->vlgrp, NULL);
-		if (real_dev->features & NETIF_F_HW_VLAN_RX)
+		if (ops->ndo_vlan_rx_register)
 			ops->ndo_vlan_rx_register(real_dev, NULL);
 
 		/* Free the group, after all cpu's are done. */
@@ -156,11 +156,6 @@ int vlan_check_real_dev(struct net_device *real_dev, u16 vlan_id)
 		return -EOPNOTSUPP;
 	}
 
-	if ((real_dev->features & NETIF_F_HW_VLAN_RX) && !ops->ndo_vlan_rx_register) {
-		pr_info("8021q: device %s has buggy VLAN hw accel\n", name);
-		return -EOPNOTSUPP;
-	}
-
 	if ((real_dev->features & NETIF_F_HW_VLAN_FILTER) &&
 	    (!ops->ndo_vlan_rx_add_vid || !ops->ndo_vlan_rx_kill_vid)) {
 		pr_info("8021q: Device %s has buggy VLAN hw accel\n", name);
@@ -213,7 +208,7 @@ int register_vlan_dev(struct net_device *dev)
 	grp->nr_vlans++;
 
 	if (ngrp) {
-		if (real_dev->features & NETIF_F_HW_VLAN_RX)
+		if (ops->ndo_vlan_rx_register)
 			ops->ndo_vlan_rx_register(real_dev, ngrp);
 		rcu_assign_pointer(real_dev->vlgrp, ngrp);
 	}
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index dee727ce0291..69b2f79800a5 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -4,54 +4,29 @@
 #include <linux/netpoll.h>
 #include "vlan.h"
 
-/* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
-int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
-		      u16 vlan_tci, int polling)
+bool vlan_hwaccel_do_receive(struct sk_buff **skbp)
 {
+	struct sk_buff *skb = *skbp;
+	u16 vlan_id = skb->vlan_tci & VLAN_VID_MASK;
 	struct net_device *vlan_dev;
-	u16 vlan_id;
-
-	if (netpoll_rx(skb))
-		return NET_RX_DROP;
-
-	if (skb_bond_should_drop(skb, ACCESS_ONCE(skb->dev->master)))
-		skb->deliver_no_wcard = 1;
+	struct vlan_rx_stats *rx_stats;
 
-	skb->skb_iif = skb->dev->ifindex;
-	__vlan_hwaccel_put_tag(skb, vlan_tci);
-	vlan_id = vlan_tci & VLAN_VID_MASK;
-	vlan_dev = vlan_group_get_device(grp, vlan_id);
-
-	if (vlan_dev)
-		skb->dev = vlan_dev;
-	else if (vlan_id) {
-		if (!(skb->dev->flags & IFF_PROMISC))
-			goto drop;
-		skb->pkt_type = PACKET_OTHERHOST;
+	vlan_dev = vlan_find_dev(skb->dev, vlan_id);
+	if (!vlan_dev) {
+		if (vlan_id)
+			skb->pkt_type = PACKET_OTHERHOST;
+		return false;
 	}
 
-	return polling ? netif_receive_skb(skb) : netif_rx(skb);
+	skb = *skbp = skb_share_check(skb, GFP_ATOMIC);
+	if (unlikely(!skb))
+		return false;
 
-drop:
-	atomic_long_inc(&skb->dev->rx_dropped);
-	dev_kfree_skb_any(skb);
-	return NET_RX_DROP;
-}
-EXPORT_SYMBOL(__vlan_hwaccel_rx);
-
-void vlan_hwaccel_do_receive(struct sk_buff *skb)
-{
-	struct net_device *dev = skb->dev;
-	struct vlan_rx_stats     *rx_stats;
-
-	skb->dev = vlan_dev_real_dev(dev);
-	netif_nit_deliver(skb);
-
-	skb->dev = dev;
-	skb->priority = vlan_get_ingress_priority(dev, skb->vlan_tci);
+	skb->dev = vlan_dev;
+	skb->priority = vlan_get_ingress_priority(vlan_dev, skb->vlan_tci);
 	skb->vlan_tci = 0;
 
-	rx_stats = this_cpu_ptr(vlan_dev_info(dev)->vlan_rx_stats);
+	rx_stats = this_cpu_ptr(vlan_dev_info(vlan_dev)->vlan_rx_stats);
 
 	u64_stats_update_begin(&rx_stats->syncp);
 	rx_stats->rx_packets++;
@@ -68,11 +43,13 @@ void vlan_hwaccel_do_receive(struct sk_buff *skb)
 		 * This allows the VLAN to have a different MAC than the
 		 * underlying device, and still route correctly. */
 		if (!compare_ether_addr(eth_hdr(skb)->h_dest,
-					dev->dev_addr))
+					vlan_dev->dev_addr))
 			skb->pkt_type = PACKET_HOST;
 		break;
 	}
 	u64_stats_update_end(&rx_stats->syncp);
+
+	return true;
 }
 
 struct net_device *vlan_dev_real_dev(const struct net_device *dev)
@@ -87,75 +64,27 @@ u16 vlan_dev_vlan_id(const struct net_device *dev)
 }
 EXPORT_SYMBOL(vlan_dev_vlan_id);
 
-static gro_result_t
-vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp,
-		unsigned int vlan_tci, struct sk_buff *skb)
+/* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
+int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
+		      u16 vlan_tci, int polling)
 {
-	struct sk_buff *p;
-	struct net_device *vlan_dev;
-	u16 vlan_id;
-
-	if (skb_bond_should_drop(skb, ACCESS_ONCE(skb->dev->master)))
-		skb->deliver_no_wcard = 1;
-
-	skb->skb_iif = skb->dev->ifindex;
 	__vlan_hwaccel_put_tag(skb, vlan_tci);
-	vlan_id = vlan_tci & VLAN_VID_MASK;
-	vlan_dev = vlan_group_get_device(grp, vlan_id);
-
-	if (vlan_dev)
-		skb->dev = vlan_dev;
-	else if (vlan_id) {
-		if (!(skb->dev->flags & IFF_PROMISC))
-			goto drop;
-		skb->pkt_type = PACKET_OTHERHOST;
-	}
-
-	for (p = napi->gro_list; p; p = p->next) {
-		unsigned long diffs;
-
-		diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev;
-		diffs |= compare_ether_header(skb_mac_header(p),
-					      skb_gro_mac_header(skb));
-		NAPI_GRO_CB(p)->same_flow = !diffs;
-		NAPI_GRO_CB(p)->flush = 0;
-	}
-
-	return dev_gro_receive(napi, skb);
-
-drop:
-	atomic_long_inc(&skb->dev->rx_dropped);
-	return GRO_DROP;
+	return polling ? netif_receive_skb(skb) : netif_rx(skb);
 }
+EXPORT_SYMBOL(__vlan_hwaccel_rx);
 
 gro_result_t vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp,
 			      unsigned int vlan_tci, struct sk_buff *skb)
 {
-	if (netpoll_rx_on(skb))
-		return vlan_hwaccel_receive_skb(skb, grp, vlan_tci)
-			? GRO_DROP : GRO_NORMAL;
-
-	skb_gro_reset_offset(skb);
-
-	return napi_skb_finish(vlan_gro_common(napi, grp, vlan_tci, skb), skb);
+	__vlan_hwaccel_put_tag(skb, vlan_tci);
+	return napi_gro_receive(napi, skb);
 }
 EXPORT_SYMBOL(vlan_gro_receive);
 
 gro_result_t vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp,
 			    unsigned int vlan_tci)
 {
-	struct sk_buff *skb = napi_frags_skb(napi);
-
-	if (!skb)
-		return GRO_DROP;
-
-	if (netpoll_rx_on(skb)) {
-		skb->protocol = eth_type_trans(skb, skb->dev);
-		return vlan_hwaccel_receive_skb(skb, grp, vlan_tci)
-			? GRO_DROP : GRO_NORMAL;
-	}
-
-	return napi_frags_finish(napi, skb,
-				 vlan_gro_common(napi, grp, vlan_tci, skb));
+	__vlan_hwaccel_put_tag(napi->skb, vlan_tci);
+	return napi_gro_frags(napi);
 }
 EXPORT_SYMBOL(vlan_gro_frags);
diff --git a/net/core/dev.c b/net/core/dev.c
index 1bfd96b1fbd4..97fd6bc2004c 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2789,33 +2789,6 @@ out:
 }
 #endif
 
-/*
- * 	netif_nit_deliver - deliver received packets to network taps
- * 	@skb: buffer
- *
- * 	This function is used to deliver incoming packets to network
- * 	taps. It should be used when the normal netif_receive_skb path
- * 	is bypassed, for example because of VLAN acceleration.
- */
-void netif_nit_deliver(struct sk_buff *skb)
-{
-	struct packet_type *ptype;
-
-	if (list_empty(&ptype_all))
-		return;
-
-	skb_reset_network_header(skb);
-	skb_reset_transport_header(skb);
-	skb->mac_len = skb->network_header - skb->mac_header;
-
-	rcu_read_lock();
-	list_for_each_entry_rcu(ptype, &ptype_all, list) {
-		if (!ptype->dev || ptype->dev == skb->dev)
-			deliver_skb(skb, ptype, skb->dev);
-	}
-	rcu_read_unlock();
-}
-
 /**
  *	netdev_rx_handler_register - register receive handler
  *	@dev: device to register a handler for
@@ -2925,9 +2898,6 @@ static int __netif_receive_skb(struct sk_buff *skb)
 	if (!netdev_tstamp_prequeue)
 		net_timestamp_check(skb);
 
-	if (vlan_tx_tag_present(skb))
-		vlan_hwaccel_do_receive(skb);
-
 	/* if we've gotten here through NAPI, check netpoll */
 	if (netpoll_receive_skb(skb))
 		return NET_RX_DROP;
@@ -2940,8 +2910,7 @@ static int __netif_receive_skb(struct sk_buff *skb)
 	 * be delivered to pkt handlers that are exact matches.  Also
 	 * the deliver_no_wcard flag will be set.  If packet handlers
 	 * are sensitive to duplicate packets these skbs will need to
-	 * be dropped at the handler.  The vlan accel path may have
-	 * already set the deliver_no_wcard flag.
+	 * be dropped at the handler.
 	 */
 	null_or_orig = NULL;
 	orig_dev = skb->dev;
@@ -3000,6 +2969,18 @@ ncls:
 			goto out;
 	}
 
+	if (vlan_tx_tag_present(skb)) {
+		if (pt_prev) {
+			ret = deliver_skb(skb, pt_prev, orig_dev);
+			pt_prev = NULL;
+		}
+		if (vlan_hwaccel_do_receive(&skb)) {
+			ret = __netif_receive_skb(skb);
+			goto out;
+		} else if (unlikely(!skb))
+			goto out;
+	}
+
 	/*
 	 * Make sure frames received on VLAN interfaces stacked on
 	 * bonding interfaces still make their way to any base bonding
@@ -3264,6 +3245,7 @@ __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 		unsigned long diffs;
 
 		diffs = (unsigned long)p->dev ^ (unsigned long)skb->dev;
+		diffs |= p->vlan_tci ^ skb->vlan_tci;
 		diffs |= compare_ether_header(skb_mac_header(p),
 					      skb_gro_mac_header(skb));
 		NAPI_GRO_CB(p)->same_flow = !diffs;
@@ -3323,6 +3305,7 @@ void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb)
 {
 	__skb_pull(skb, skb_headlen(skb));
 	skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb));
+	skb->vlan_tci = 0;
 
 	napi->skb = skb;
 }
-- 
cgit v1.2.3


From d5dbda23804156ae6f35025ade5307a49d1db6d7 Mon Sep 17 00:00:00 2001
From: Jesse Gross <jesse@nicira.com>
Date: Wed, 20 Oct 2010 13:56:07 +0000
Subject: ethtool: Add support for vlan accleration.

Now that vlan acceleration is handled consistently regardless of usage,
it is possible to enable and disable it at will.  This adds support for
Ethtool operations that change the offloading status for debugging
purposes, similar to other forms of hardware acceleration.

Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool.h | 2 ++
 net/core/ethtool.c      | 3 ++-
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 8a3338ceb438..6628a507fd3b 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -309,6 +309,8 @@ struct ethtool_perm_addr {
  * flag differs from the read-only value.
  */
 enum ethtool_flags {
+	ETH_FLAG_TXVLAN		= (1 << 7),	/* TX VLAN offload enabled */
+	ETH_FLAG_RXVLAN		= (1 << 8),	/* RX VLAN offload enabled */
 	ETH_FLAG_LRO		= (1 << 15),	/* LRO is enabled */
 	ETH_FLAG_NTUPLE		= (1 << 27),	/* N-tuple filters enabled */
 	ETH_FLAG_RXHASH		= (1 << 28),
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 685c7005e87f..956a9f4971cb 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -132,7 +132,8 @@ EXPORT_SYMBOL(ethtool_op_set_ufo);
  * NETIF_F_xxx values in include/linux/netdevice.h
  */
 static const u32 flags_dup_features =
-	(ETH_FLAG_LRO | ETH_FLAG_NTUPLE | ETH_FLAG_RXHASH);
+	(ETH_FLAG_LRO | ETH_FLAG_RXVLAN | ETH_FLAG_TXVLAN | ETH_FLAG_NTUPLE |
+	 ETH_FLAG_RXHASH);
 
 u32 ethtool_op_get_flags(struct net_device *dev)
 {
-- 
cgit v1.2.3


From 11a691bea48887c27425cc40bf291e74c922df25 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Thu, 21 Oct 2010 10:32:29 +0200
Subject: block: Turn bvec_k{un,}map_irq() into static inline functions

Convert bvec_k{un,}map_irq() from macros to static inline functions if
!CONFIG_HIGHMEM, so we can easier detect mistakes like the one fixed in
93055c31045a2d5599ec613a0c6cdcefc481a460 ("ps3disk: passing wrong variable =
to
bvec_kunmap_irq()")

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
---
 include/linux/bio.h | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/bio.h b/include/linux/bio.h
index 2c3fd7421607..ba679992d39b 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -346,8 +346,15 @@ static inline void bvec_kunmap_irq(char *buffer, unsigned long *flags)
 }
 
 #else
-#define bvec_kmap_irq(bvec, flags)	(page_address((bvec)->bv_page) + (bvec)->bv_offset)
-#define bvec_kunmap_irq(buf, flags)	do { *(flags) = 0; } while (0)
+static inline char *bvec_kmap_irq(struct bio_vec *bvec, unsigned long *flags)
+{
+	return page_address(bvec->bv_page) + bvec->bv_offset;
+}
+
+static inline void bvec_kunmap_irq(char *buffer, unsigned long *flags)
+{
+	*flags = 0;
+}
 #endif
 
 static inline char *__bio_kmap_irq(struct bio *bio, unsigned short idx,
-- 
cgit v1.2.3


From 11165f1457181e4499e5eada442434a07827ffd8 Mon Sep 17 00:00:00 2001
From: stephen hemminger <shemminger@vyatta.com>
Date: Mon, 18 Oct 2010 14:27:29 +0000
Subject: socket: localize functions

A couple of functions in socket.c are only used there and
should be localized.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/socket.h | 1 -
 net/socket.c           | 7 ++++---
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/socket.h b/include/linux/socket.h
index a8f56e1ec760..5146b50202ce 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -326,7 +326,6 @@ extern long verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *a
 extern int memcpy_toiovec(struct iovec *v, unsigned char *kdata, int len);
 extern int memcpy_toiovecend(const struct iovec *v, unsigned char *kdata,
 			     int offset, int len);
-extern int move_addr_to_user(struct sockaddr *kaddr, int klen, void __user *uaddr, int __user *ulen);
 extern int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr *kaddr);
 extern int put_cmsg(struct msghdr*, int level, int type, int len, void *data);
 
diff --git a/net/socket.c b/net/socket.c
index 717a5f1c8792..72da57d6ab7b 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -209,8 +209,8 @@ int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr *kaddr)
  *	specified. Zero is returned for a success.
  */
 
-int move_addr_to_user(struct sockaddr *kaddr, int klen, void __user *uaddr,
-		      int __user *ulen)
+static int move_addr_to_user(struct sockaddr *kaddr, int klen,
+			     void __user *uaddr, int __user *ulen)
 {
 	int err;
 	int len;
@@ -661,7 +661,8 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk,
 }
 EXPORT_SYMBOL_GPL(__sock_recv_timestamp);
 
-inline void sock_recv_drops(struct msghdr *msg, struct sock *sk, struct sk_buff *skb)
+static inline void sock_recv_drops(struct msghdr *msg, struct sock *sk,
+				   struct sk_buff *skb)
 {
 	if (sock_flag(sk, SOCK_RXQ_OVFL) && skb && skb->dropcount)
 		put_cmsg(msg, SOL_SOCKET, SO_RXQ_OVFL,
-- 
cgit v1.2.3


From 8c974438085d2c81b006daeaab8801eedbd19758 Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@tuxdriver.com>
Date: Thu, 21 Oct 2010 01:06:15 +0000
Subject: Revert c6537d6742985da1fbf12ae26cde6a096fd35b5c

Backout the tipc changes to the flags int he subscription message.  These
changees, while reasonable on the surface, interefere with user space ABI
compatibility which is a no-no.  This was part of the changes to fix the
endianess issues in the TIPC protocol, which would be really nice to do but we
need to do so in a way that is backwards compatible with user space.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tipc.h | 30 ++++++++++++++++++------------
 net/tipc/subscr.c    | 15 +++++----------
 2 files changed, 23 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/tipc.h b/include/linux/tipc.h
index 181c8d0e6f73..d10614b29d59 100644
--- a/include/linux/tipc.h
+++ b/include/linux/tipc.h
@@ -127,17 +127,23 @@ static inline unsigned int tipc_node(__u32 addr)
  * TIPC topology subscription service definitions
  */
 
-#define TIPC_SUB_SERVICE     	0x00  	/* Filter for service availability    */
-#define TIPC_SUB_PORTS     	0x01  	/* Filter for port availability  */
-#define TIPC_SUB_CANCEL         0x04    /* Cancel a subscription         */
+#define TIPC_SUB_PORTS     	0x01  	/* filter for port availability */
+#define TIPC_SUB_SERVICE     	0x02  	/* filter for service availability */
+#define TIPC_SUB_CANCEL         0x04    /* cancel a subscription */
+#if 0
+/* The following filter options are not currently implemented */
+#define TIPC_SUB_NO_BIND_EVTS	0x04	/* filter out "publish" events */
+#define TIPC_SUB_NO_UNBIND_EVTS	0x08	/* filter out "withdraw" events */
+#define TIPC_SUB_SINGLE_EVT	0x10	/* expire after first event */
+#endif
 
 #define TIPC_WAIT_FOREVER	~0	/* timeout for permanent subscription */
 
 struct tipc_subscr {
-	struct tipc_name_seq seq;	/* NBO. Name sequence of interest */
-	__u32 timeout;			/* NBO. Subscription duration (in ms) */
-        __u32 filter;   		/* NBO. Bitmask of filter options */
-	char usr_handle[8];		/* Opaque. Available for subscriber use */
+	struct tipc_name_seq seq;	/* name sequence of interest */
+	__u32 timeout;			/* subscription duration (in ms) */
+        __u32 filter;   		/* bitmask of filter options */
+	char usr_handle[8];		/* available for subscriber use */
 };
 
 #define TIPC_PUBLISHED		1	/* publication event */
@@ -145,11 +151,11 @@ struct tipc_subscr {
 #define TIPC_SUBSCR_TIMEOUT	3	/* subscription timeout event */
 
 struct tipc_event {
-	__u32 event;			/* NBO. Event type, as defined above */
-	__u32 found_lower;		/* NBO. Matching name seq instances  */
-	__u32 found_upper;		/*  "      "       "   "    "        */
-	struct tipc_portid port;	/* NBO. Associated port              */
-	struct tipc_subscr s;		/* Original, associated subscription */
+	__u32 event;			/* event type */
+	__u32 found_lower;		/* matching name seq instances */
+	__u32 found_upper;		/*    "      "    "     "      */
+	struct tipc_portid port;	/* associated port */
+	struct tipc_subscr s;		/* associated subscription */
 };
 
 /*
diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index ab6eab4c45e2..ff123e56114a 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -274,7 +274,7 @@ static void subscr_cancel(struct tipc_subscr *s,
 {
 	struct subscription *sub;
 	struct subscription *sub_temp;
-	__u32 type, lower, upper, timeout, filter;
+	__u32 type, lower, upper;
 	int found = 0;
 
 	/* Find first matching subscription, exit if not found */
@@ -282,18 +282,12 @@ static void subscr_cancel(struct tipc_subscr *s,
 	type = ntohl(s->seq.type);
 	lower = ntohl(s->seq.lower);
 	upper = ntohl(s->seq.upper);
-	timeout = ntohl(s->timeout);
-	filter = ntohl(s->filter) & ~TIPC_SUB_CANCEL;
 
 	list_for_each_entry_safe(sub, sub_temp, &subscriber->subscription_list,
 				 subscription_list) {
 			if ((type == sub->seq.type) &&
 			    (lower == sub->seq.lower) &&
-			    (upper == sub->seq.upper) &&
-			    (timeout == sub->timeout) &&
-                            (filter == sub->filter) &&
-                             !memcmp(s->usr_handle,sub->evt.s.usr_handle,
-				     sizeof(s->usr_handle)) ){
+			    (upper == sub->seq.upper)) {
 				found = 1;
 				break;
 			}
@@ -310,7 +304,7 @@ static void subscr_cancel(struct tipc_subscr *s,
 		k_term_timer(&sub->timer);
 		spin_lock_bh(subscriber->lock);
 	}
-	dbg("Cancel: removing sub %u,%u,%u from subscriber %p list\n",
+	dbg("Cancel: removing sub %u,%u,%u from subscriber %x list\n",
 	    sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber);
 	subscr_del(sub);
 }
@@ -358,7 +352,8 @@ static struct subscription *subscr_subscribe(struct tipc_subscr *s,
 	sub->seq.upper = ntohl(s->seq.upper);
 	sub->timeout = ntohl(s->timeout);
 	sub->filter = ntohl(s->filter);
-	if ((sub->filter && (sub->filter != TIPC_SUB_PORTS)) ||
+	if ((!(sub->filter & TIPC_SUB_PORTS) ==
+	     !(sub->filter & TIPC_SUB_SERVICE)) ||
 	    (sub->seq.lower > sub->seq.upper)) {
 		warn("Subscription rejected, illegal request\n");
 		kfree(sub);
-- 
cgit v1.2.3


From d0c2b0d265a0f1f92922a99a31def9da582197ac Mon Sep 17 00:00:00 2001
From: stephen hemminger <shemminger@vyatta.com>
Date: Tue, 19 Oct 2010 07:12:10 +0000
Subject: napi: unexport napi_reuse_skb

The function napi_reuse_skb is only used inside core.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 2 --
 net/core/dev.c            | 3 +--
 2 files changed, 1 insertion(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ed7db7eebbf3..fcd3dda86322 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1748,8 +1748,6 @@ extern gro_result_t	napi_skb_finish(gro_result_t ret, struct sk_buff *skb);
 extern gro_result_t	napi_gro_receive(struct napi_struct *napi,
 					 struct sk_buff *skb);
 extern void		napi_gro_flush(struct napi_struct *napi);
-extern void		napi_reuse_skb(struct napi_struct *napi,
-				       struct sk_buff *skb);
 extern struct sk_buff *	napi_get_frags(struct napi_struct *napi);
 extern gro_result_t	napi_frags_finish(struct napi_struct *napi,
 					  struct sk_buff *skb,
diff --git a/net/core/dev.c b/net/core/dev.c
index 97fd6bc2004c..b2269ac04cb8 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3301,7 +3301,7 @@ gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(napi_gro_receive);
 
-void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb)
+static void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb)
 {
 	__skb_pull(skb, skb_headlen(skb));
 	skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb));
@@ -3309,7 +3309,6 @@ void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb)
 
 	napi->skb = skb;
 }
-EXPORT_SYMBOL(napi_reuse_skb);
 
 struct sk_buff *napi_get_frags(struct napi_struct *napi)
 {
-- 
cgit v1.2.3


From 30e1f7a8122145f44f45c95366e27b6bb0b08428 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <borislav.petkov@amd.com>
Date: Thu, 2 Sep 2010 17:26:48 +0200
Subject: EDAC: Export edac sysfs class to users.

Move toplevel sysfs class to the stub and make it available to
non-modularized code too. Add proper refcounting of its users and move
the registration functionality into the reference counting routines.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 drivers/edac/edac_device_sysfs.c | 16 +++++---
 drivers/edac/edac_mc_sysfs.c     |  9 +++--
 drivers/edac/edac_module.c       | 79 +---------------------------------------
 drivers/edac/edac_module.h       |  1 -
 drivers/edac/edac_pci_sysfs.c    | 10 +++--
 drivers/edac/edac_stub.c         | 51 ++++++++++++++++++++++++--
 include/linux/edac.h             |  4 ++
 7 files changed, 75 insertions(+), 95 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/edac/edac_device_sysfs.c b/drivers/edac/edac_device_sysfs.c
index 070968178a24..2941dca91aae 100644
--- a/drivers/edac/edac_device_sysfs.c
+++ b/drivers/edac/edac_device_sysfs.c
@@ -13,6 +13,7 @@
 #include <linux/ctype.h>
 #include <linux/module.h>
 #include <linux/slab.h>
+#include <linux/edac.h>
 
 #include "edac_core.h"
 #include "edac_module.h"
@@ -235,7 +236,7 @@ int edac_device_register_sysfs_main_kobj(struct edac_device_ctl_info *edac_dev)
 	debugf1("%s()\n", __func__);
 
 	/* get the /sys/devices/system/edac reference */
-	edac_class = edac_get_edac_class();
+	edac_class = edac_get_sysfs_class();
 	if (edac_class == NULL) {
 		debugf1("%s() no edac_class error\n", __func__);
 		err = -ENODEV;
@@ -255,7 +256,7 @@ int edac_device_register_sysfs_main_kobj(struct edac_device_ctl_info *edac_dev)
 
 	if (!try_module_get(edac_dev->owner)) {
 		err = -ENODEV;
-		goto err_out;
+		goto err_mod_get;
 	}
 
 	/* register */
@@ -282,6 +283,9 @@ int edac_device_register_sysfs_main_kobj(struct edac_device_ctl_info *edac_dev)
 err_kobj_reg:
 	module_put(edac_dev->owner);
 
+err_mod_get:
+	edac_put_sysfs_class();
+
 err_out:
 	return err;
 }
@@ -290,12 +294,11 @@ err_out:
  * edac_device_unregister_sysfs_main_kobj:
  *	the '..../edac/<name>' kobject
  */
-void edac_device_unregister_sysfs_main_kobj(
-					struct edac_device_ctl_info *edac_dev)
+void edac_device_unregister_sysfs_main_kobj(struct edac_device_ctl_info *dev)
 {
 	debugf0("%s()\n", __func__);
 	debugf4("%s() name of kobject is: %s\n",
-		__func__, kobject_name(&edac_dev->kobj));
+		__func__, kobject_name(&dev->kobj));
 
 	/*
 	 * Unregister the edac device's kobject and
@@ -304,7 +307,8 @@ void edac_device_unregister_sysfs_main_kobj(
 	 *   a) module_put() this module
 	 *   b) 'kfree' the memory
 	 */
-	kobject_put(&edac_dev->kobj);
+	kobject_put(&dev->kobj);
+	edac_put_sysfs_class();
 }
 
 /* edac_dev -> instance information */
diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index aa93ad82ee07..a4135860149b 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -11,6 +11,7 @@
 
 #include <linux/ctype.h>
 #include <linux/slab.h>
+#include <linux/edac.h>
 #include <linux/bug.h>
 
 #include "edac_core.h"
@@ -1017,7 +1018,7 @@ int edac_sysfs_setup_mc_kset(void)
 	debugf1("%s()\n", __func__);
 
 	/* get the /sys/devices/system/edac class reference */
-	edac_class = edac_get_edac_class();
+	edac_class = edac_get_sysfs_class();
 	if (edac_class == NULL) {
 		debugf1("%s() no edac_class error=%d\n", __func__, err);
 		goto fail_out;
@@ -1028,15 +1029,16 @@ int edac_sysfs_setup_mc_kset(void)
 	if (!mc_kset) {
 		err = -ENOMEM;
 		debugf1("%s() Failed to register '.../edac/mc'\n", __func__);
-		goto fail_out;
+		goto fail_kset;
 	}
 
 	debugf1("%s() Registered '.../edac/mc' kobject\n", __func__);
 
 	return 0;
 
+fail_kset:
+	edac_put_sysfs_class();
 
-	/* error unwind stack */
 fail_out:
 	return err;
 }
@@ -1049,5 +1051,6 @@ fail_out:
 void edac_sysfs_teardown_mc_kset(void)
 {
 	kset_unregister(mc_kset);
+	edac_put_sysfs_class();
 }
 
diff --git a/drivers/edac/edac_module.c b/drivers/edac/edac_module.c
index 7e1374afd967..be4b075c3098 100644
--- a/drivers/edac/edac_module.c
+++ b/drivers/edac/edac_module.c
@@ -26,15 +26,6 @@ EXPORT_SYMBOL_GPL(edac_debug_level);
 /* scope is to module level only */
 struct workqueue_struct *edac_workqueue;
 
-/*
- * sysfs object: /sys/devices/system/edac
- *	need to export to other files in this modules
- */
-static struct sysdev_class edac_class = {
-	.name = "edac",
-};
-static int edac_class_valid;
-
 /*
  * edac_op_state_to_string()
  */
@@ -54,60 +45,6 @@ char *edac_op_state_to_string(int opstate)
 	return "UNKNOWN";
 }
 
-/*
- * edac_get_edac_class()
- *
- *	return pointer to the edac class of 'edac'
- */
-struct sysdev_class *edac_get_edac_class(void)
-{
-	struct sysdev_class *classptr = NULL;
-
-	if (edac_class_valid)
-		classptr = &edac_class;
-
-	return classptr;
-}
-
-/*
- * edac_register_sysfs_edac_name()
- *
- *	register the 'edac' into /sys/devices/system
- *
- * return:
- *	0  success
- *	!0 error
- */
-static int edac_register_sysfs_edac_name(void)
-{
-	int err;
-
-	/* create the /sys/devices/system/edac directory */
-	err = sysdev_class_register(&edac_class);
-
-	if (err) {
-		debugf1("%s() error=%d\n", __func__, err);
-		return err;
-	}
-
-	edac_class_valid = 1;
-	return 0;
-}
-
-/*
- * sysdev_class_unregister()
- *
- *	unregister the 'edac' from /sys/devices/system
- */
-static void edac_unregister_sysfs_edac_name(void)
-{
-	/* only if currently registered, then unregister it */
-	if (edac_class_valid)
-		sysdev_class_unregister(&edac_class);
-
-	edac_class_valid = 0;
-}
-
 /*
  * edac_workqueue_setup
  *	initialize the edac work queue for polling operations
@@ -153,22 +90,12 @@ static int __init edac_init(void)
 	 */
 	edac_pci_clear_parity_errors();
 
-	/*
-	 * perform the registration of the /sys/devices/system/edac class object
-	 */
-	if (edac_register_sysfs_edac_name()) {
-		edac_printk(KERN_ERR, EDAC_MC,
-			"Error initializing 'edac' kobject\n");
-		err = -ENODEV;
-		goto error;
-	}
-
 	/*
 	 * now set up the mc_kset under the edac class object
 	 */
 	err = edac_sysfs_setup_mc_kset();
 	if (err)
-		goto sysfs_setup_fail;
+		goto error;
 
 	/* Setup/Initialize the workq for this core */
 	err = edac_workqueue_setup();
@@ -183,9 +110,6 @@ static int __init edac_init(void)
 workq_fail:
 	edac_sysfs_teardown_mc_kset();
 
-sysfs_setup_fail:
-	edac_unregister_sysfs_edac_name();
-
 error:
 	return err;
 }
@@ -201,7 +125,6 @@ static void __exit edac_exit(void)
 	/* tear down the various subsystems */
 	edac_workqueue_teardown();
 	edac_sysfs_teardown_mc_kset();
-	edac_unregister_sysfs_edac_name();
 }
 
 /*
diff --git a/drivers/edac/edac_module.h b/drivers/edac/edac_module.h
index 233d4798c3aa..17aabb7b90ec 100644
--- a/drivers/edac/edac_module.h
+++ b/drivers/edac/edac_module.h
@@ -42,7 +42,6 @@ extern void edac_device_unregister_sysfs_main_kobj(
 				struct edac_device_ctl_info *edac_dev);
 extern int edac_device_create_sysfs(struct edac_device_ctl_info *edac_dev);
 extern void edac_device_remove_sysfs(struct edac_device_ctl_info *edac_dev);
-extern struct sysdev_class *edac_get_edac_class(void);
 
 /* edac core workqueue: single CPU mode */
 extern struct workqueue_struct *edac_workqueue;
diff --git a/drivers/edac/edac_pci_sysfs.c b/drivers/edac/edac_pci_sysfs.c
index c39697df9cb4..023b01cb5175 100644
--- a/drivers/edac/edac_pci_sysfs.c
+++ b/drivers/edac/edac_pci_sysfs.c
@@ -7,7 +7,7 @@
  *
  */
 #include <linux/module.h>
-#include <linux/sysdev.h>
+#include <linux/edac.h>
 #include <linux/slab.h>
 #include <linux/ctype.h>
 
@@ -354,7 +354,7 @@ static int edac_pci_main_kobj_setup(void)
 	/* First time, so create the main kobject and its
 	 * controls and atributes
 	 */
-	edac_class = edac_get_edac_class();
+	edac_class = edac_get_sysfs_class();
 	if (edac_class == NULL) {
 		debugf1("%s() no edac_class\n", __func__);
 		err = -ENODEV;
@@ -368,7 +368,7 @@ static int edac_pci_main_kobj_setup(void)
 	if (!try_module_get(THIS_MODULE)) {
 		debugf1("%s() try_module_get() failed\n", __func__);
 		err = -ENODEV;
-		goto decrement_count_fail;
+		goto mod_get_fail;
 	}
 
 	edac_pci_top_main_kobj = kzalloc(sizeof(struct kobject), GFP_KERNEL);
@@ -403,6 +403,9 @@ kobject_init_and_add_fail:
 kzalloc_fail:
 	module_put(THIS_MODULE);
 
+mod_get_fail:
+	edac_put_sysfs_class();
+
 decrement_count_fail:
 	/* if are on this error exit, nothing to tear down */
 	atomic_dec(&edac_pci_sysfs_refcount);
@@ -429,6 +432,7 @@ static void edac_pci_main_kobj_teardown(void)
 			__func__);
 		kobject_put(edac_pci_top_main_kobj);
 	}
+	edac_put_sysfs_class();
 }
 
 /*
diff --git a/drivers/edac/edac_stub.c b/drivers/edac/edac_stub.c
index 20b428aa155e..aab970760b75 100644
--- a/drivers/edac/edac_stub.c
+++ b/drivers/edac/edac_stub.c
@@ -3,10 +3,13 @@
  *
  * Author: Dave Jiang <djiang@mvista.com>
  *
- * 2007 (c) MontaVista Software, Inc. This file is licensed under
- * the terms of the GNU General Public License version 2. This program
- * is licensed "as is" without any warranty of any kind, whether express
- * or implied.
+ * 2007 (c) MontaVista Software, Inc.
+ * 2010 (c) Advanced Micro Devices Inc.
+ *	    Borislav Petkov <borislav.petkov@amd.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
  *
  */
 #include <linux/module.h>
@@ -23,6 +26,8 @@ EXPORT_SYMBOL_GPL(edac_handlers);
 int edac_err_assert = 0;
 EXPORT_SYMBOL_GPL(edac_err_assert);
 
+static atomic_t edac_class_valid = ATOMIC_INIT(0);
+
 /*
  * called to determine if there is an EDAC driver interested in
  * knowing an event (such as NMI) occurred
@@ -44,3 +49,41 @@ void edac_atomic_assert_error(void)
 	edac_err_assert++;
 }
 EXPORT_SYMBOL_GPL(edac_atomic_assert_error);
+
+/*
+ * sysfs object: /sys/devices/system/edac
+ *	need to export to other files
+ */
+struct sysdev_class edac_class = {
+	.name = "edac",
+};
+EXPORT_SYMBOL_GPL(edac_class);
+
+/* return pointer to the 'edac' node in sysfs */
+struct sysdev_class *edac_get_sysfs_class(void)
+{
+	int err = 0;
+
+	if (atomic_read(&edac_class_valid))
+		goto out;
+
+	/* create the /sys/devices/system/edac directory */
+	err = sysdev_class_register(&edac_class);
+	if (err) {
+		printk(KERN_ERR "Error registering toplevel EDAC sysfs dir\n");
+		return NULL;
+	}
+
+out:
+	atomic_inc(&edac_class_valid);
+	return &edac_class;
+}
+EXPORT_SYMBOL_GPL(edac_get_sysfs_class);
+
+void edac_put_sysfs_class(void)
+{
+	/* last user unregisters it */
+	if (atomic_dec_and_test(&edac_class_valid))
+		sysdev_class_unregister(&edac_class);
+}
+EXPORT_SYMBOL_GPL(edac_put_sysfs_class);
diff --git a/include/linux/edac.h b/include/linux/edac.h
index 7cf92e8a4196..36c66443bdfd 100644
--- a/include/linux/edac.h
+++ b/include/linux/edac.h
@@ -13,6 +13,7 @@
 #define _LINUX_EDAC_H_
 
 #include <asm/atomic.h>
+#include <linux/sysdev.h>
 
 #define EDAC_OPSTATE_INVAL	-1
 #define EDAC_OPSTATE_POLL	0
@@ -22,9 +23,12 @@
 extern int edac_op_state;
 extern int edac_err_assert;
 extern atomic_t edac_handlers;
+extern struct sysdev_class edac_class;
 
 extern int edac_handler_set(void);
 extern void edac_atomic_assert_error(void);
+extern struct sysdev_class *edac_get_sysfs_class(void);
+extern void edac_put_sysfs_class(void);
 
 static inline void opstate_init(void)
 {
-- 
cgit v1.2.3


From 6de5bd128d381ad88ac6d419a5e597048eb468cf Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sat, 11 Sep 2010 18:00:57 +0200
Subject: BKL: introduce CONFIG_BKL.

With all the patches we have queued in the BKL removal tree, only a
few dozen modules are left that actually rely on the BKL, and even
there are lots of low-hanging fruit. We need to decide what to do
about them, this patch illustrates one of the options:

Every user of the BKL is marked as 'depends on BKL' in Kconfig,
and the CONFIG_BKL becomes a user-visible option. If it gets
disabled, no BKL using module can be built any more and the BKL
code itself is compiled out.

The one exception is file locking, which is practically always
enabled and does a 'select BKL' instead. This effectively forces
CONFIG_BKL to be enabled until we have solved the fs/lockd
mess and can apply the patch that removes the BKL from fs/locks.c.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 drivers/gpu/drm/Kconfig         | 5 ++++-
 drivers/media/Kconfig           | 1 +
 drivers/net/appletalk/Kconfig   | 1 +
 drivers/staging/cx25821/Kconfig | 1 +
 drivers/staging/easycap/Kconfig | 1 +
 drivers/staging/go7007/Kconfig  | 1 +
 drivers/staging/usbip/Kconfig   | 2 +-
 fs/Kconfig                      | 1 +
 fs/adfs/Kconfig                 | 1 +
 fs/autofs/Kconfig               | 1 +
 fs/hpfs/Kconfig                 | 1 +
 fs/nfs/Kconfig                  | 1 +
 fs/nfsd/Kconfig                 | 1 +
 fs/smbfs/Kconfig                | 1 +
 fs/udf/Kconfig                  | 1 +
 fs/ufs/Kconfig                  | 1 +
 include/linux/smp_lock.h        | 7 +++++--
 init/Kconfig                    | 2 +-
 lib/Kconfig.debug               | 9 +++++++++
 net/ipx/Kconfig                 | 1 +
 net/x25/Kconfig                 | 1 +
 21 files changed, 36 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 4cab0c6397e3..7af443672626 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -73,7 +73,8 @@ source "drivers/gpu/drm/radeon/Kconfig"
 
 config DRM_I810
 	tristate "Intel I810"
-	depends on DRM && AGP && AGP_INTEL
+	# BKL usage in order to avoid AB-BA deadlocks, may become BROKEN_ON_SMP
+	depends on DRM && AGP && AGP_INTEL && BKL
 	help
 	  Choose this option if you have an Intel I810 graphics card.  If M is
 	  selected, the module will be called i810.  AGP support is required
@@ -86,6 +87,8 @@ choice
 
 config DRM_I830
 	tristate "i830 driver"
+	# BKL usage in order to avoid AB-BA deadlocks, i830 may get removed
+	depends on BKL
 	help
 	  Choose this option if you have a system that has Intel 830M, 845G,
 	  852GM, 855GM or 865G integrated graphics.  If M is selected, the
diff --git a/drivers/media/Kconfig b/drivers/media/Kconfig
index a28541b2b1a2..bad2cedb8d96 100644
--- a/drivers/media/Kconfig
+++ b/drivers/media/Kconfig
@@ -19,6 +19,7 @@ comment "Multimedia core support"
 
 config VIDEO_DEV
 	tristate "Video For Linux"
+	depends on BKL # used in many drivers for ioctl handling, need to kill
 	---help---
 	  V4L core support for video capture and overlay devices, webcams and
 	  AM/FM radio cards.
diff --git a/drivers/net/appletalk/Kconfig b/drivers/net/appletalk/Kconfig
index 0a0e0cd81a23..20f97e7017ce 100644
--- a/drivers/net/appletalk/Kconfig
+++ b/drivers/net/appletalk/Kconfig
@@ -3,6 +3,7 @@
 #
 config ATALK
 	tristate "Appletalk protocol support"
+	depends on BKL # waiting to be removed from net/appletalk/ddp.c
 	select LLC
 	---help---
 	  AppleTalk is the protocol that Apple computers can use to communicate
diff --git a/drivers/staging/cx25821/Kconfig b/drivers/staging/cx25821/Kconfig
index df7756a95fad..813cb355ac01 100644
--- a/drivers/staging/cx25821/Kconfig
+++ b/drivers/staging/cx25821/Kconfig
@@ -1,6 +1,7 @@
 config VIDEO_CX25821
 	tristate "Conexant cx25821 support"
 	depends on DVB_CORE && VIDEO_DEV && PCI && I2C && INPUT
+	depends on BKL # please fix
 	select I2C_ALGOBIT
 	select VIDEO_BTCX
 	select VIDEO_TVEEPROM
diff --git a/drivers/staging/easycap/Kconfig b/drivers/staging/easycap/Kconfig
index bd96f39f2735..9d5fe4ddc30a 100644
--- a/drivers/staging/easycap/Kconfig
+++ b/drivers/staging/easycap/Kconfig
@@ -1,6 +1,7 @@
 config EASYCAP
 	tristate "EasyCAP USB ID 05e1:0408 support"
 	depends on USB && VIDEO_DEV
+	depends on BKL # please fix
 
 	---help---
 	  This is an integrated audio/video driver for EasyCAP cards with
diff --git a/drivers/staging/go7007/Kconfig b/drivers/staging/go7007/Kconfig
index e47f683a323e..75fa46805527 100644
--- a/drivers/staging/go7007/Kconfig
+++ b/drivers/staging/go7007/Kconfig
@@ -1,6 +1,7 @@
 config VIDEO_GO7007
 	tristate "WIS GO7007 MPEG encoder support"
 	depends on VIDEO_DEV && PCI && I2C && INPUT
+	depends on BKL # please fix
 	depends on SND
 	select VIDEOBUF_DMA_SG
 	select VIDEO_IR
diff --git a/drivers/staging/usbip/Kconfig b/drivers/staging/usbip/Kconfig
index 2c1d10acb8b5..b11ec379b5c2 100644
--- a/drivers/staging/usbip/Kconfig
+++ b/drivers/staging/usbip/Kconfig
@@ -1,6 +1,6 @@
 config USB_IP_COMMON
 	tristate "USB IP support (EXPERIMENTAL)"
-	depends on USB && NET && EXPERIMENTAL
+	depends on USB && NET && EXPERIMENTAL && BKL
 	default N
 	---help---
 	  This enables pushing USB packets over IP to allow remote
diff --git a/fs/Kconfig b/fs/Kconfig
index 3d185308ec88..65781de44fc0 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -50,6 +50,7 @@ endif # BLOCK
 config FILE_LOCKING
 	bool "Enable POSIX file locking API" if EMBEDDED
 	default y
+	select BKL # while lockd still uses it.
 	help
 	  This option enables standard file locking support, required
           for filesystems like NFS and for the flock() system
diff --git a/fs/adfs/Kconfig b/fs/adfs/Kconfig
index e55182a74605..1dd5f34b3cf2 100644
--- a/fs/adfs/Kconfig
+++ b/fs/adfs/Kconfig
@@ -1,6 +1,7 @@
 config ADFS_FS
 	tristate "ADFS file system support (EXPERIMENTAL)"
 	depends on BLOCK && EXPERIMENTAL
+	depends on BKL # need to fix
 	help
 	  The Acorn Disc Filing System is the standard file system of the
 	  RiscOS operating system which runs on Acorn's ARM-based Risc PC
diff --git a/fs/autofs/Kconfig b/fs/autofs/Kconfig
index 5f3bea90911e..480e210c83ab 100644
--- a/fs/autofs/Kconfig
+++ b/fs/autofs/Kconfig
@@ -1,5 +1,6 @@
 config AUTOFS_FS
 	tristate "Kernel automounter support"
+	depends on BKL # unfixable, just use autofs4
 	help
 	  The automounter is a tool to automatically mount remote file systems
 	  on demand. This implementation is partially kernel-based to reduce
diff --git a/fs/hpfs/Kconfig b/fs/hpfs/Kconfig
index 56bd15c5bf6c..63b6f5632318 100644
--- a/fs/hpfs/Kconfig
+++ b/fs/hpfs/Kconfig
@@ -1,6 +1,7 @@
 config HPFS_FS
 	tristate "OS/2 HPFS file system support"
 	depends on BLOCK
+	depends on BKL # nontrivial to fix
 	help
 	  OS/2 is IBM's operating system for PC's, the same as Warp, and HPFS
 	  is the file system used for organizing files on OS/2 hard disk
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index 6c2aad49d731..10b9524bb908 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -1,6 +1,7 @@
 config NFS_FS
 	tristate "NFS client support"
 	depends on INET && FILE_LOCKING
+	depends on BKL # fix as soon as lockd is done
 	select LOCKD
 	select SUNRPC
 	select NFS_ACL_SUPPORT if NFS_V3_ACL
diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
index 95932f523aef..429d4a142276 100644
--- a/fs/nfsd/Kconfig
+++ b/fs/nfsd/Kconfig
@@ -2,6 +2,7 @@ config NFSD
 	tristate "NFS server support"
 	depends on INET
 	depends on FILE_LOCKING
+	depends on BKL # fix as soon as lockd is done
 	select LOCKD
 	select SUNRPC
 	select EXPORTFS
diff --git a/fs/smbfs/Kconfig b/fs/smbfs/Kconfig
index e668127c8b2e..2bc24a8c4039 100644
--- a/fs/smbfs/Kconfig
+++ b/fs/smbfs/Kconfig
@@ -1,5 +1,6 @@
 config SMB_FS
 	tristate "SMB file system support (OBSOLETE, please use CIFS)"
+	depends on BKL # probably unfixable
 	depends on INET
 	select NLS
 	help
diff --git a/fs/udf/Kconfig b/fs/udf/Kconfig
index 0e0e99bd6bce..f8def3c8ea4c 100644
--- a/fs/udf/Kconfig
+++ b/fs/udf/Kconfig
@@ -1,5 +1,6 @@
 config UDF_FS
 	tristate "UDF file system support"
+	depends on BKL # needs serious work to remove
 	select CRC_ITU_T
 	help
 	  This is the new file system used on some CD-ROMs and DVDs. Say Y if
diff --git a/fs/ufs/Kconfig b/fs/ufs/Kconfig
index e4f10a40768a..30c8f223253d 100644
--- a/fs/ufs/Kconfig
+++ b/fs/ufs/Kconfig
@@ -1,6 +1,7 @@
 config UFS_FS
 	tristate "UFS file system support (read only)"
 	depends on BLOCK
+	depends on BKL # probably fixable
 	help
 	  BSD and derivate versions of Unix (such as SunOS, FreeBSD, NetBSD,
 	  OpenBSD and NeXTstep) use a file system called UFS. Some System V
diff --git a/include/linux/smp_lock.h b/include/linux/smp_lock.h
index 2ea1dd1ba21c..291f721144c2 100644
--- a/include/linux/smp_lock.h
+++ b/include/linux/smp_lock.h
@@ -54,12 +54,15 @@ static inline void cycle_kernel_lock(void)
 
 #else
 
+#ifdef CONFIG_BKL /* provoke build bug if not set */
 #define lock_kernel()
 #define unlock_kernel()
-#define release_kernel_lock(task)		do { } while(0)
 #define cycle_kernel_lock()			do { } while(0)
-#define reacquire_kernel_lock(task)		0
 #define kernel_locked()				1
+#endif /* CONFIG_BKL */
+
+#define release_kernel_lock(task)		do { } while(0)
+#define reacquire_kernel_lock(task)		0
 
 #endif /* CONFIG_LOCK_KERNEL */
 #endif /* __LINUX_SMPLOCK_H */
diff --git a/init/Kconfig b/init/Kconfig
index 2de5b1cbadd9..2005a1d49928 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -64,7 +64,7 @@ config BROKEN_ON_SMP
 
 config LOCK_KERNEL
 	bool
-	depends on SMP || PREEMPT
+	depends on (SMP || PREEMPT) && BKL
 	default y
 
 config INIT_ENV_ARG_LIMIT
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 1b4afd2e6ca0..088eea1c2bef 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -461,6 +461,15 @@ config DEBUG_MUTEXES
 	 This feature allows mutex semantics violations to be detected and
 	 reported.
 
+config BKL
+	bool "Big Kernel Lock" if (SMP || PREEMPT)
+	default y
+	help
+	  This is the traditional lock that is used in old code instead
+	  of proper locking. All drivers that use the BKL should depend
+	  on this symbol.
+	  Say Y here unless you are working on removing the BKL.
+
 config DEBUG_LOCK_ALLOC
 	bool "Lock debugging: detect incorrect freeing of live locks"
 	depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
diff --git a/net/ipx/Kconfig b/net/ipx/Kconfig
index e9ad0062fbb6..02549cb2c328 100644
--- a/net/ipx/Kconfig
+++ b/net/ipx/Kconfig
@@ -3,6 +3,7 @@
 #
 config IPX
 	tristate "The IPX protocol"
+	depends on BKL # should be fixable
 	select LLC
 	---help---
 	  This is support for the Novell networking protocol, IPX, commonly
diff --git a/net/x25/Kconfig b/net/x25/Kconfig
index e6759c9660bb..2196e55e4f61 100644
--- a/net/x25/Kconfig
+++ b/net/x25/Kconfig
@@ -5,6 +5,7 @@
 config X25
 	tristate "CCITT X.25 Packet Layer (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
+	depends on BKL # should be fixable
 	---help---
 	  X.25 is a set of standardized network protocols, similar in scope to
 	  frame relay; the one physical line from your box to the X.25 network
-- 
cgit v1.2.3


From 6c46862280c5f55eda7750391bc65cd7e08c7535 Mon Sep 17 00:00:00 2001
From: Balazs Scheidler <bazsi@balabit.hu>
Date: Thu, 21 Oct 2010 16:08:28 +0200
Subject: tproxy: added tproxy sockopt interface in the IPV6 layer

Support for IPV6_RECVORIGDSTADDR sockopt for UDP sockets were contributed by
Harry Mason.

Signed-off-by: Balazs Scheidler <bazsi@balabit.hu>
Signed-off-by: KOVACS Krisztian <hidden@balabit.hu>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/in6.h      |  4 ++++
 include/linux/ipv6.h     |  4 +++-
 net/ipv6/datagram.c      | 19 +++++++++++++++++++
 net/ipv6/ipv6_sockglue.c | 23 +++++++++++++++++++++++
 4 files changed, 49 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/in6.h b/include/linux/in6.h
index c4bf46f764bf..097a34b55560 100644
--- a/include/linux/in6.h
+++ b/include/linux/in6.h
@@ -268,6 +268,10 @@ struct in6_flowlabel_req {
 /* RFC5082: Generalized Ttl Security Mechanism */
 #define IPV6_MINHOPCOUNT		73
 
+#define IPV6_ORIGDSTADDR        74
+#define IPV6_RECVORIGDSTADDR    IPV6_ORIGDSTADDR
+#define IPV6_TRANSPARENT        75
+
 /*
  * Multicast Routing:
  * see include/linux/mroute6.h.
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index e62683ba88e6..8e429d0e0405 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -341,7 +341,9 @@ struct ipv6_pinfo {
 				odstopts:1,
                                 rxflow:1,
 				rxtclass:1,
-				rxpmtu:1;
+				rxpmtu:1,
+				rxorigdstaddr:1;
+				/* 2 bits hole */
 		} bits;
 		__u16		all;
 	} rxopt;
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index ef371aa01ac5..320bdb877eed 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -577,6 +577,25 @@ int datagram_recv_ctl(struct sock *sk, struct msghdr *msg, struct sk_buff *skb)
 		u8 *ptr = nh + opt->dst1;
 		put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, (ptr[1]+1)<<3, ptr);
 	}
+	if (np->rxopt.bits.rxorigdstaddr) {
+		struct sockaddr_in6 sin6;
+		u16 *ports = (u16 *) skb_transport_header(skb);
+
+		if (skb_transport_offset(skb) + 4 <= skb->len) {
+			/* All current transport protocols have the port numbers in the
+			 * first four bytes of the transport header and this function is
+			 * written with this assumption in mind.
+			 */
+
+			sin6.sin6_family = AF_INET6;
+			ipv6_addr_copy(&sin6.sin6_addr, &ipv6_hdr(skb)->daddr);
+			sin6.sin6_port = ports[1];
+			sin6.sin6_flowinfo = 0;
+			sin6.sin6_scope_id = 0;
+
+			put_cmsg(msg, SOL_IPV6, IPV6_ORIGDSTADDR, sizeof(sin6), &sin6);
+		}
+	}
 	return 0;
 }
 
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index a7f66bc8f0b0..0553867a317f 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -342,6 +342,21 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 		retv = 0;
 		break;
 
+	case IPV6_TRANSPARENT:
+		if (optlen < sizeof(int))
+			goto e_inval;
+		/* we don't have a separate transparent bit for IPV6 we use the one in the IPv4 socket */
+		inet_sk(sk)->transparent = valbool;
+		retv = 0;
+		break;
+
+	case IPV6_RECVORIGDSTADDR:
+		if (optlen < sizeof(int))
+			goto e_inval;
+		np->rxopt.bits.rxorigdstaddr = valbool;
+		retv = 0;
+		break;
+
 	case IPV6_HOPOPTS:
 	case IPV6_RTHDRDSTOPTS:
 	case IPV6_RTHDR:
@@ -1104,6 +1119,14 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname,
 		break;
 	}
 
+	case IPV6_TRANSPARENT:
+		val = inet_sk(sk)->transparent;
+		break;
+
+	case IPV6_RECVORIGDSTADDR:
+		val = np->rxopt.bits.rxorigdstaddr;
+		break;
+
 	case IPV6_UNICAST_HOPS:
 	case IPV6_MULTICAST_HOPS:
 	{
-- 
cgit v1.2.3


From 6ad7889327a5ee6ab4220bd34e4428c7d0de0f32 Mon Sep 17 00:00:00 2001
From: Balazs Scheidler <bazsi@balabit.hu>
Date: Thu, 21 Oct 2010 16:17:26 +0200
Subject: tproxy: added IPv6 support to the TPROXY target

This requires a new revision as the old target structure was
IPv4 specific.

Signed-off-by: Balazs Scheidler <bazsi@balabit.hu>
Signed-off-by: KOVACS Krisztian <hidden@balabit.hu>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/xt_TPROXY.h |  13 +-
 net/netfilter/xt_TPROXY.c           | 262 +++++++++++++++++++++++++++++++-----
 2 files changed, 235 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/xt_TPROXY.h b/include/linux/netfilter/xt_TPROXY.h
index 152e8f97132b..3f3d69361289 100644
--- a/include/linux/netfilter/xt_TPROXY.h
+++ b/include/linux/netfilter/xt_TPROXY.h
@@ -1,5 +1,5 @@
-#ifndef _XT_TPROXY_H_target
-#define _XT_TPROXY_H_target
+#ifndef _XT_TPROXY_H
+#define _XT_TPROXY_H
 
 /* TPROXY target is capable of marking the packet to perform
  * redirection. We can get rid of that whenever we get support for
@@ -11,4 +11,11 @@ struct xt_tproxy_target_info {
 	__be16 lport;
 };
 
-#endif /* _XT_TPROXY_H_target */
+struct xt_tproxy_target_info_v1 {
+	u_int32_t mark_mask;
+	u_int32_t mark_value;
+	union nf_inet_addr laddr;
+	__be16 lport;
+};
+
+#endif /* _XT_TPROXY_H */
diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c
index e0b6900a92c1..d5f97e2302b8 100644
--- a/net/netfilter/xt_TPROXY.c
+++ b/net/netfilter/xt_TPROXY.c
@@ -1,7 +1,7 @@
 /*
  * Transparent proxy support for Linux/iptables
  *
- * Copyright (c) 2006-2007 BalaBit IT Ltd.
+ * Copyright (c) 2006-2010 BalaBit IT Ltd.
  * Author: Balazs Scheidler, Krisztian Kovacs
  *
  * This program is free software; you can redistribute it and/or modify
@@ -19,15 +19,18 @@
 
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter_ipv4/ip_tables.h>
+#include <linux/netfilter_ipv6/ip6_tables.h>
 #include <linux/netfilter/xt_TPROXY.h>
 
 #include <net/netfilter/ipv4/nf_defrag_ipv4.h>
+#include <net/netfilter/ipv6/nf_defrag_ipv6.h>
 #include <net/netfilter/nf_tproxy_core.h>
 
 /**
- * tproxy_handle_time_wait() - handle TCP TIME_WAIT reopen redirections
+ * tproxy_handle_time_wait4() - handle IPv4 TCP TIME_WAIT reopen redirections
  * @skb:	The skb being processed.
- * @par:	Iptables target parameters.
+ * @laddr:	IPv4 address to redirect to or zero.
+ * @lport:	TCP port to redirect to or zero.
  * @sk:		The TIME_WAIT TCP socket found by the lookup.
  *
  * We have to handle SYN packets arriving to TIME_WAIT sockets
@@ -35,16 +38,16 @@
  * redirect the new connection to the proxy if there's a listener
  * socket present.
  *
- * tproxy_handle_time_wait() consumes the socket reference passed in.
+ * tproxy_handle_time_wait4() consumes the socket reference passed in.
  *
  * Returns the listener socket if there's one, the TIME_WAIT socket if
  * no such listener is found, or NULL if the TCP header is incomplete.
  */
 static struct sock *
-tproxy_handle_time_wait(struct sk_buff *skb, const struct xt_action_param *par, struct sock *sk)
+tproxy_handle_time_wait4(struct sk_buff *skb, __be32 laddr, __be16 lport,
+			struct sock *sk)
 {
 	const struct iphdr *iph = ip_hdr(skb);
-	const struct xt_tproxy_target_info *tgi = par->targinfo;
 	struct tcphdr _hdr, *hp;
 
 	hp = skb_header_pointer(skb, ip_hdrlen(skb), sizeof(_hdr), &_hdr);
@@ -59,13 +62,64 @@ tproxy_handle_time_wait(struct sk_buff *skb, const struct xt_action_param *par,
 		struct sock *sk2;
 
 		sk2 = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
-					    iph->saddr, tgi->laddr ? tgi->laddr : iph->daddr,
-					    hp->source, tgi->lport ? tgi->lport : hp->dest,
-					    par->in, NFT_LOOKUP_LISTENER);
+					    iph->saddr, laddr ? laddr : iph->daddr,
+					    hp->source, lport ? lport : hp->dest,
+					    skb->dev, NFT_LOOKUP_LISTENER);
+		if (sk2) {
+			inet_twsk_deschedule(inet_twsk(sk), &tcp_death_row);
+			inet_twsk_put(inet_twsk(sk));
+			sk = sk2;
+		}
+	}
+
+	return sk;
+}
+
+/**
+ * tproxy_handle_time_wait6() - handle IPv6 TCP TIME_WAIT reopen redirections
+ * @skb:	The skb being processed.
+ * @tproto:	Transport protocol.
+ * @thoff:	Transport protocol header offset.
+ * @par:	Iptables target parameters.
+ * @sk:		The TIME_WAIT TCP socket found by the lookup.
+ *
+ * We have to handle SYN packets arriving to TIME_WAIT sockets
+ * differently: instead of reopening the connection we should rather
+ * redirect the new connection to the proxy if there's a listener
+ * socket present.
+ *
+ * tproxy_handle_time_wait6() consumes the socket reference passed in.
+ *
+ * Returns the listener socket if there's one, the TIME_WAIT socket if
+ * no such listener is found, or NULL if the TCP header is incomplete.
+ */
+static struct sock *
+tproxy_handle_time_wait6(struct sk_buff *skb, int tproto, int thoff,
+			 const struct xt_action_param *par,
+			 struct sock *sk)
+{
+	const struct ipv6hdr *iph = ipv6_hdr(skb);
+	struct tcphdr _hdr, *hp;
+	const struct xt_tproxy_target_info_v1 *tgi = par->targinfo;
+
+	hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr);
+	if (hp == NULL) {
+		inet_twsk_put(inet_twsk(sk));
+		return NULL;
+	}
+
+	if (hp->syn && !hp->rst && !hp->ack && !hp->fin) {
+		/* SYN to a TIME_WAIT socket, we'd rather redirect it
+		 * to a listener socket if there's one */
+		struct sock *sk2;
+
+		sk2 = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+					    &iph->saddr,
+					    !ipv6_addr_any(&tgi->laddr.in6) ? &tgi->laddr.in6 : &iph->daddr,
+					    hp->source,
+					    tgi->lport ? tgi->lport : hp->dest,
+					    skb->dev, NFT_LOOKUP_LISTENER);
 		if (sk2) {
-			/* yeah, there's one, let's kill the TIME_WAIT
-			 * socket and redirect to the listener
-			 */
 			inet_twsk_deschedule(inet_twsk(sk), &tcp_death_row);
 			inet_twsk_put(inet_twsk(sk));
 			sk = sk2;
@@ -76,10 +130,10 @@ tproxy_handle_time_wait(struct sk_buff *skb, const struct xt_action_param *par,
 }
 
 static unsigned int
-tproxy_tg(struct sk_buff *skb, const struct xt_action_param *par)
+tproxy_tg4(struct sk_buff *skb, __be32 laddr, __be16 lport,
+	   u_int32_t mark_mask, u_int32_t mark_value)
 {
 	const struct iphdr *iph = ip_hdr(skb);
-	const struct xt_tproxy_target_info *tgi = par->targinfo;
 	struct udphdr _hdr, *hp;
 	struct sock *sk;
 
@@ -87,18 +141,105 @@ tproxy_tg(struct sk_buff *skb, const struct xt_action_param *par)
 	if (hp == NULL)
 		return NF_DROP;
 
+	/* check if there's an ongoing connection on the packet
+	 * addresses, this happens if the redirect already happened
+	 * and the current packet belongs to an already established
+	 * connection */
 	sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
 				   iph->saddr, iph->daddr,
 				   hp->source, hp->dest,
-				   par->in, NFT_LOOKUP_ESTABLISHED);
+				   skb->dev, NFT_LOOKUP_ESTABLISHED);
 
 	/* UDP has no TCP_TIME_WAIT state, so we never enter here */
 	if (sk && sk->sk_state == TCP_TIME_WAIT)
-		sk = tproxy_handle_time_wait(skb, par, sk);
+		/* reopening a TIME_WAIT connection needs special handling */
+		sk = tproxy_handle_time_wait4(skb, laddr, lport, sk);
 	else if (!sk)
+		/* no, there's no established connection, check if
+		 * there's a listener on the redirected addr/port */
 		sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol,
-					   iph->saddr, tgi->laddr ? tgi->laddr : iph->daddr,
-					   hp->source, tgi->lport ? tgi->lport : hp->dest,
+					   iph->saddr, laddr ? laddr : iph->daddr,
+					   hp->source, lport ? lport : hp->dest,
+					   skb->dev, NFT_LOOKUP_LISTENER);
+
+	/* NOTE: assign_sock consumes our sk reference */
+	if (sk && nf_tproxy_assign_sock(skb, sk)) {
+		/* This should be in a separate target, but we don't do multiple
+		   targets on the same rule yet */
+		skb->mark = (skb->mark & ~mark_mask) ^ mark_value;
+
+		pr_debug("redirecting: proto %hhu %pI4:%hu -> %pI4:%hu, mark: %x\n",
+			 iph->protocol, &iph->daddr, ntohs(hp->dest),
+			 &laddr, ntohs(lport), skb->mark);
+		return NF_ACCEPT;
+	}
+
+	pr_debug("no socket, dropping: proto %hhu %08x:%hu -> %08x:%hu, mark: %x\n",
+		 iph->protocol, ntohl(iph->daddr), ntohs(hp->dest),
+		 ntohl(laddr), ntohs(lport), skb->mark);
+	return NF_DROP;
+}
+
+static unsigned int
+tproxy_tg4_v0(struct sk_buff *skb, const struct xt_action_param *par)
+{
+	const struct xt_tproxy_target_info *tgi = par->targinfo;
+
+	return tproxy_tg4(skb, tgi->laddr, tgi->lport, tgi->mark_mask, tgi->mark_value);
+}
+
+static unsigned int
+tproxy_tg4_v1(struct sk_buff *skb, const struct xt_action_param *par)
+{
+	const struct xt_tproxy_target_info_v1 *tgi = par->targinfo;
+
+	return tproxy_tg4(skb, tgi->laddr.ip, tgi->lport, tgi->mark_mask, tgi->mark_value);
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+static unsigned int
+tproxy_tg6_v1(struct sk_buff *skb, const struct xt_action_param *par)
+{
+	const struct ipv6hdr *iph = ipv6_hdr(skb);
+	const struct xt_tproxy_target_info_v1 *tgi = par->targinfo;
+	struct udphdr _hdr, *hp;
+	struct sock *sk;
+	int thoff;
+	int tproto;
+
+	tproto = ipv6_find_hdr(skb, &thoff, -1, NULL);
+	if (tproto < 0) {
+		pr_debug("unable to find transport header in IPv6 packet, dropping\n");
+		return NF_DROP;
+	}
+
+	hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr);
+	if (hp == NULL) {
+		pr_debug("unable to grab transport header contents in IPv6 packet, dropping\n");
+		return NF_DROP;
+	}
+
+	/* check if there's an ongoing connection on the packet
+	 * addresses, this happens if the redirect already happened
+	 * and the current packet belongs to an already established
+	 * connection */
+	sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+				   &iph->saddr, &iph->daddr,
+				   hp->source, hp->dest,
+				   par->in, NFT_LOOKUP_ESTABLISHED);
+
+	/* UDP has no TCP_TIME_WAIT state, so we never enter here */
+	if (sk && sk->sk_state == TCP_TIME_WAIT)
+		/* reopening a TIME_WAIT connection needs special handling */
+		sk = tproxy_handle_time_wait6(skb, tproto, thoff, par, sk);
+	else if (!sk)
+		/* no there's no established connection, check if
+		 * there's a listener on the redirected addr/port */
+		sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto,
+					   &iph->saddr,
+					   !ipv6_addr_any(&tgi->laddr.in6) ? &tgi->laddr.in6 : &iph->daddr,
+					   hp->source,
+					   tgi->lport ? tgi->lport : hp->dest,
 					   par->in, NFT_LOOKUP_LISTENER);
 
 	/* NOTE: assign_sock consumes our sk reference */
@@ -107,19 +248,33 @@ tproxy_tg(struct sk_buff *skb, const struct xt_action_param *par)
 		   targets on the same rule yet */
 		skb->mark = (skb->mark & ~tgi->mark_mask) ^ tgi->mark_value;
 
-		pr_debug("redirecting: proto %u %08x:%u -> %08x:%u, mark: %x\n",
-			 iph->protocol, ntohl(iph->daddr), ntohs(hp->dest),
-			 ntohl(tgi->laddr), ntohs(tgi->lport), skb->mark);
+		pr_debug("redirecting: proto %hhu %pI6:%hu -> %pI6:%hu, mark: %x\n",
+			 tproto, &iph->saddr, ntohs(hp->dest),
+			 &tgi->laddr.in6, ntohs(tgi->lport), skb->mark);
 		return NF_ACCEPT;
 	}
 
-	pr_debug("no socket, dropping: proto %u %08x:%u -> %08x:%u, mark: %x\n",
-		 iph->protocol, ntohl(iph->daddr), ntohs(hp->dest),
-		 ntohl(tgi->laddr), ntohs(tgi->lport), skb->mark);
+	pr_debug("no socket, dropping: proto %hhu %pI6:%hu -> %pI6:%hu, mark: %x\n",
+		 tproto, &iph->saddr, ntohs(hp->dest),
+		 &tgi->laddr.in6, ntohs(tgi->lport), skb->mark);
 	return NF_DROP;
 }
 
-static int tproxy_tg_check(const struct xt_tgchk_param *par)
+static int tproxy_tg6_check(const struct xt_tgchk_param *par)
+{
+	const struct ip6t_ip6 *i = par->entryinfo;
+
+	if ((i->proto == IPPROTO_TCP || i->proto == IPPROTO_UDP)
+	    && !(i->flags & IP6T_INV_PROTO))
+		return 0;
+
+	pr_info("Can be used only in combination with "
+		"either -p tcp or -p udp\n");
+	return -EINVAL;
+}
+#endif
+
+static int tproxy_tg4_check(const struct xt_tgchk_param *par)
 {
 	const struct ipt_ip *i = par->entryinfo;
 
@@ -132,31 +287,64 @@ static int tproxy_tg_check(const struct xt_tgchk_param *par)
 	return -EINVAL;
 }
 
-static struct xt_target tproxy_tg_reg __read_mostly = {
-	.name		= "TPROXY",
-	.family		= NFPROTO_IPV4,
-	.table		= "mangle",
-	.target		= tproxy_tg,
-	.targetsize	= sizeof(struct xt_tproxy_target_info),
-	.checkentry	= tproxy_tg_check,
-	.hooks		= 1 << NF_INET_PRE_ROUTING,
-	.me		= THIS_MODULE,
+static struct xt_target tproxy_tg_reg[] __read_mostly = {
+	{
+		.name		= "TPROXY",
+		.family		= NFPROTO_IPV4,
+		.table		= "mangle",
+		.target		= tproxy_tg4_v0,
+		.revision	= 0,
+		.targetsize	= sizeof(struct xt_tproxy_target_info),
+		.checkentry	= tproxy_tg4_check,
+		.hooks		= 1 << NF_INET_PRE_ROUTING,
+		.me		= THIS_MODULE,
+	},
+	{
+		.name		= "TPROXY",
+		.family		= NFPROTO_IPV4,
+		.table		= "mangle",
+		.target		= tproxy_tg4_v1,
+		.revision	= 1,
+		.targetsize	= sizeof(struct xt_tproxy_target_info_v1),
+		.checkentry	= tproxy_tg4_check,
+		.hooks		= 1 << NF_INET_PRE_ROUTING,
+		.me		= THIS_MODULE,
+	},
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	{
+		.name		= "TPROXY",
+		.family		= NFPROTO_IPV6,
+		.table		= "mangle",
+		.target		= tproxy_tg6_v1,
+		.revision	= 1,
+		.targetsize	= sizeof(struct xt_tproxy_target_info_v1),
+		.checkentry	= tproxy_tg6_check,
+		.hooks		= 1 << NF_INET_PRE_ROUTING,
+		.me		= THIS_MODULE,
+	},
+#endif
+
 };
 
 static int __init tproxy_tg_init(void)
 {
 	nf_defrag_ipv4_enable();
-	return xt_register_target(&tproxy_tg_reg);
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	nf_defrag_ipv6_enable();
+#endif
+
+	return xt_register_targets(tproxy_tg_reg, ARRAY_SIZE(tproxy_tg_reg));
 }
 
 static void __exit tproxy_tg_exit(void)
 {
-	xt_unregister_target(&tproxy_tg_reg);
+	xt_unregister_targets(tproxy_tg_reg, ARRAY_SIZE(tproxy_tg_reg));
 }
 
 module_init(tproxy_tg_init);
 module_exit(tproxy_tg_exit);
 MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Krisztian Kovacs");
+MODULE_AUTHOR("Balazs Scheidler, Krisztian Kovacs");
 MODULE_DESCRIPTION("Netfilter transparent proxy (TPROXY) target module.");
 MODULE_ALIAS("ipt_TPROXY");
+MODULE_ALIAS("ip6t_TPROXY");
-- 
cgit v1.2.3


From d9027470b88631d0956ac37cdadfdeb9cdcf2c99 Mon Sep 17 00:00:00 2001
From: Gwendal Grignou <gwendal@google.com>
Date: Tue, 25 May 2010 12:31:38 -0700
Subject: [libata] Add ATA transport class

This is a scheleton for libata transport class.
All information is read only, exporting information from libata:
- ata_port class: one per ATA port
- ata_link class: one per ATA port or 15 for SATA Port Multiplier
- ata_device class: up to 2 for PATA link, usually one for SATA.

Signed-off-by: Gwendal Grignou <gwendal@google.com>
Reviewed-by: Grant Grundler <grundler@google.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 Documentation/ABI/testing/sysfs-ata |  99 +++++
 drivers/ata/Makefile                |   2 +-
 drivers/ata/libata-core.c           |  49 ++-
 drivers/ata/libata-eh.c             |  35 +-
 drivers/ata/libata-pmp.c            |  18 +-
 drivers/ata/libata-scsi.c           |  21 +-
 drivers/ata/libata-transport.c      | 773 ++++++++++++++++++++++++++++++++++++
 drivers/ata/libata-transport.h      |  18 +
 drivers/ata/libata.h                |   7 +
 include/linux/libata.h              |   5 +
 10 files changed, 987 insertions(+), 40 deletions(-)
 create mode 100644 Documentation/ABI/testing/sysfs-ata
 create mode 100644 drivers/ata/libata-transport.c
 create mode 100644 drivers/ata/libata-transport.h

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-ata b/Documentation/ABI/testing/sysfs-ata
new file mode 100644
index 000000000000..0a932155cbba
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-ata
@@ -0,0 +1,99 @@
+What:		/sys/class/ata_...
+Date:		August 2008
+Contact:	Gwendal Grignou<gwendal@google.com>
+Description:
+
+Provide a place in sysfs for storing the ATA topology of the system.  This allows
+retrieving various information about ATA objects.
+
+Files under /sys/class/ata_port
+-------------------------------
+
+	For each port, a directory ataX is created where X is the ata_port_id of
+	the port. The device parent is the ata host device.
+
+idle_irq (read)
+
+	Number of IRQ received by the port while idle [some ata HBA only].
+
+nr_pmp_links (read)
+
+	If a SATA Port Multiplier (PM) is connected, number of link behind it.
+
+Files under /sys/class/ata_link
+-------------------------------
+
+	Behind each port, there is a ata_link. If there is a SATA PM in the
+	topology, 15 ata_link objects are created.
+
+	If a link is behind a port, the directory name is linkX, where X is
+	ata_port_id of the port.
+	If a link is behind a PM, its name is linkX.Y where X is ata_port_id
+	of the parent port and Y the PM port.
+
+hw_sata_spd_limit
+
+	Maximum speed supported by the connected SATA device.
+
+sata_spd_limit
+
+	Maximum speed imposed by libata.
+
+sata_spd
+
+	Current speed of the link [1.5, 3Gps,...].
+
+Files under /sys/class/ata_device
+---------------------------------
+
+	Behind each link, up to two ata device are created.
+	The name of the directory is devX[.Y].Z where:
+	- X is ata_port_id of the port where the device is connected,
+	- Y the port of the PM if any, and
+	- Z the device id: for PATA, there is usually 2 devices [0,1],
+	only 1 for SATA.
+
+class
+	Device class. Can be "ata" for disk, "atapi" for packet device,
+	"pmp" for PM, or "none" if no device was found behind the link.
+
+dma_mode
+
+	Transfer modes supported by the device when in DMA mode.
+	Mostly used by PATA device.
+
+pio_mode
+
+	Transfer modes supported by the device when in PIO mode.
+	Mostly used by PATA device.
+
+xfer_mode
+
+	Current transfer mode.
+
+id
+
+	Cached result of IDENTIFY command, as described in ATA8 7.16 and 7.17.
+	Only valid if the device is not a PM.
+
+gscr
+
+	Cached result of the dump of PM GSCR register.
+	Valid registers are:
+	0: 	SATA_PMP_GSCR_PROD_ID,
+	1: 	SATA_PMP_GSCR_REV,
+	2: 	SATA_PMP_GSCR_PORT_INFO,
+	32:	SATA_PMP_GSCR_ERROR,
+	33:	SATA_PMP_GSCR_ERROR_EN,
+	64:	SATA_PMP_GSCR_FEAT,
+	96:	SATA_PMP_GSCR_FEAT_EN,
+	130:	SATA_PMP_GSCR_SII_GPIO
+	Only valid if the device is a PM.
+
+spdn_cnt
+
+	Number of time libata decided to lower the speed of link due to errors.
+
+ering
+
+	Formatted output of the error ring of the device.
diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile
index d5df04a395ca..ccd461bb5ee4 100644
--- a/drivers/ata/Makefile
+++ b/drivers/ata/Makefile
@@ -99,7 +99,7 @@ obj-$(CONFIG_ATA_GENERIC)	+= ata_generic.o
 # Should be last libata driver
 obj-$(CONFIG_PATA_LEGACY)	+= pata_legacy.o
 
-libata-objs	:= libata-core.o libata-scsi.o libata-eh.o
+libata-objs	:= libata-core.o libata-scsi.o libata-eh.o libata-transport.o
 libata-$(CONFIG_ATA_SFF)	+= libata-sff.o
 libata-$(CONFIG_SATA_PMP)	+= libata-pmp.o
 libata-$(CONFIG_ATA_ACPI)	+= libata-acpi.o
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 932eaee50245..4012b33e8b8a 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -68,7 +68,7 @@
 #include <linux/ratelimit.h>
 
 #include "libata.h"
-
+#include "libata-transport.h"
 
 /* debounce timing parameters in msecs { interval, duration, timeout } */
 const unsigned long sata_deb_timing_normal[]		= {   5,  100, 2000 };
@@ -1017,7 +1017,7 @@ const char *ata_mode_string(unsigned long xfer_mask)
 	return "<n/a>";
 }
 
-static const char *sata_spd_string(unsigned int spd)
+const char *sata_spd_string(unsigned int spd)
 {
 	static const char * const spd_str[] = {
 		"1.5 Gbps",
@@ -5517,7 +5517,8 @@ void ata_link_init(struct ata_port *ap, struct ata_link *link, int pmp)
 	int i;
 
 	/* clear everything except for devices */
-	memset(link, 0, offsetof(struct ata_link, device[0]));
+	memset((void *)link + ATA_LINK_CLEAR_BEGIN, 0,
+	       ATA_LINK_CLEAR_END - ATA_LINK_CLEAR_BEGIN);
 
 	link->ap = ap;
 	link->pmp = pmp;
@@ -5591,7 +5592,7 @@ struct ata_port *ata_port_alloc(struct ata_host *host)
 	ap = kzalloc(sizeof(*ap), GFP_KERNEL);
 	if (!ap)
 		return NULL;
-
+	
 	ap->pflags |= ATA_PFLAG_INITIALIZING;
 	ap->lock = &host->lock;
 	ap->print_id = -1;
@@ -6093,9 +6094,18 @@ int ata_host_register(struct ata_host *host, struct scsi_host_template *sht)
 	for (i = 0; i < host->n_ports; i++)
 		host->ports[i]->print_id = ata_print_id++;
 
+	
+	/* Create associated sysfs transport objects  */
+	for (i = 0; i < host->n_ports; i++) {
+		rc = ata_tport_add(host->dev,host->ports[i]);
+		if (rc) {
+			goto err_tadd;
+		}
+	}
+
 	rc = ata_scsi_add_hosts(host, sht);
 	if (rc)
-		return rc;
+		goto err_tadd;
 
 	/* associate with ACPI nodes */
 	ata_acpi_associate(host);
@@ -6136,6 +6146,13 @@ int ata_host_register(struct ata_host *host, struct scsi_host_template *sht)
 	}
 
 	return 0;
+
+ err_tadd:
+	while (--i >= 0) {
+		ata_tport_delete(host->ports[i]);
+	}
+	return rc;
+
 }
 
 /**
@@ -6226,6 +6243,13 @@ static void ata_port_detach(struct ata_port *ap)
 	cancel_rearming_delayed_work(&ap->hotplug_task);
 
  skip_eh:
+	if (ap->pmp_link) {
+		int i;
+		for (i = 0; i < SATA_PMP_MAX_PORTS; i++)
+			ata_tlink_delete(&ap->pmp_link[i]);
+	}
+	ata_tport_delete(ap);
+
 	/* remove the associated SCSI host */
 	scsi_remove_host(ap->scsi_host);
 }
@@ -6542,7 +6566,7 @@ static void __init ata_parse_force_param(void)
 
 static int __init ata_init(void)
 {
-	int rc = -ENOMEM;
+	int rc;
 
 	ata_parse_force_param();
 
@@ -6552,12 +6576,25 @@ static int __init ata_init(void)
 		return rc;
 	}
 
+	libata_transport_init();
+	ata_scsi_transport_template = ata_attach_transport();
+	if (!ata_scsi_transport_template) {
+		ata_sff_exit();
+		rc = -ENOMEM;
+		goto err_out;
+	}		
+
 	printk(KERN_DEBUG "libata version " DRV_VERSION " loaded.\n");
 	return 0;
+
+err_out:
+	return rc;
 }
 
 static void __exit ata_exit(void)
 {
+	ata_release_transport(ata_scsi_transport_template);
+	libata_transport_exit();
 	ata_sff_exit();
 	kfree(ata_force_tbl);
 }
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index e48302eae55f..95838b382231 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -57,6 +57,7 @@ enum {
 	/* error flags */
 	ATA_EFLAG_IS_IO			= (1 << 0),
 	ATA_EFLAG_DUBIOUS_XFER		= (1 << 1),
+	ATA_EFLAG_OLD_ER                = (1 << 31),
 
 	/* error categories */
 	ATA_ECAT_NONE			= 0,
@@ -396,14 +397,9 @@ static struct ata_ering_entry *ata_ering_top(struct ata_ering *ering)
 	return NULL;
 }
 
-static void ata_ering_clear(struct ata_ering *ering)
-{
-	memset(ering, 0, sizeof(*ering));
-}
-
-static int ata_ering_map(struct ata_ering *ering,
-			 int (*map_fn)(struct ata_ering_entry *, void *),
-			 void *arg)
+int ata_ering_map(struct ata_ering *ering,
+		  int (*map_fn)(struct ata_ering_entry *, void *),
+		  void *arg)
 {
 	int idx, rc = 0;
 	struct ata_ering_entry *ent;
@@ -422,6 +418,17 @@ static int ata_ering_map(struct ata_ering *ering,
 	return rc;
 }
 
+int ata_ering_clear_cb(struct ata_ering_entry *ent, void *void_arg)
+{
+	ent->eflags |= ATA_EFLAG_OLD_ER;
+	return 0;
+}
+
+static void ata_ering_clear(struct ata_ering *ering)
+{
+	ata_ering_map(ering, ata_ering_clear_cb, NULL);
+}
+
 static unsigned int ata_eh_dev_action(struct ata_device *dev)
 {
 	struct ata_eh_context *ehc = &dev->link->eh_context;
@@ -572,19 +579,19 @@ void ata_scsi_error(struct Scsi_Host *host)
 		int nr_timedout = 0;
 
 		spin_lock_irqsave(ap->lock, flags);
-		
+
 		/* This must occur under the ap->lock as we don't want
 		   a polled recovery to race the real interrupt handler
-		   
+
 		   The lost_interrupt handler checks for any completed but
 		   non-notified command and completes much like an IRQ handler.
-		   
+
 		   We then fall into the error recovery code which will treat
 		   this as if normal completion won the race */
 
 		if (ap->ops->lost_interrupt)
 			ap->ops->lost_interrupt(ap);
-			
+
 		list_for_each_entry_safe(scmd, tmp, &host->eh_cmd_q, eh_entry) {
 			struct ata_queued_cmd *qc;
 
@@ -628,7 +635,7 @@ void ata_scsi_error(struct Scsi_Host *host)
 		ap->eh_tries = ATA_EH_MAX_TRIES;
 	} else
 		spin_unlock_wait(ap->lock);
-		
+
 	/* If we timed raced normal completion and there is nothing to
 	   recover nr_timedout == 0 why exactly are we doing error recovery ? */
 
@@ -1755,7 +1762,7 @@ static int speed_down_verdict_cb(struct ata_ering_entry *ent, void *void_arg)
 	struct speed_down_verdict_arg *arg = void_arg;
 	int cat;
 
-	if (ent->timestamp < arg->since)
+	if ((ent->eflags & ATA_EFLAG_OLD_ER) || (ent->timestamp < arg->since))
 		return -1;
 
 	cat = ata_eh_categorize_error(ent->eflags, ent->err_mask,
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 224faabd7b7e..505470237d79 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -11,6 +11,7 @@
 #include <linux/libata.h>
 #include <linux/slab.h>
 #include "libata.h"
+#include "libata-transport.h"
 
 const struct ata_port_operations sata_pmp_port_ops = {
 	.inherits		= &sata_port_ops,
@@ -312,10 +313,10 @@ static int sata_pmp_configure(struct ata_device *dev, int print_info)
 	return rc;
 }
 
-static int sata_pmp_init_links(struct ata_port *ap, int nr_ports)
+static int sata_pmp_init_links (struct ata_port *ap, int nr_ports)
 {
 	struct ata_link *pmp_link = ap->pmp_link;
-	int i;
+	int i, err;
 
 	if (!pmp_link) {
 		pmp_link = kzalloc(sizeof(pmp_link[0]) * SATA_PMP_MAX_PORTS,
@@ -327,6 +328,13 @@ static int sata_pmp_init_links(struct ata_port *ap, int nr_ports)
 			ata_link_init(ap, &pmp_link[i], i);
 
 		ap->pmp_link = pmp_link;
+
+		for (i = 0; i < SATA_PMP_MAX_PORTS; i++) {
+			err = ata_tlink_add(&pmp_link[i]);
+			if (err) {
+				goto err_tlink;
+			}
+		}
 	}
 
 	for (i = 0; i < nr_ports; i++) {
@@ -339,6 +347,12 @@ static int sata_pmp_init_links(struct ata_port *ap, int nr_ports)
 	}
 
 	return 0;
+  err_tlink:
+	while (--i >= 0)
+		ata_tlink_delete(&pmp_link[i]);
+	kfree(pmp_link);
+	ap->pmp_link = NULL;
+	return err;
 }
 
 static void sata_pmp_quirks(struct ata_port *ap)
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index a89172c100f5..c16f5c151735 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -51,6 +51,7 @@
 #include <asm/unaligned.h>
 
 #include "libata.h"
+#include "libata-transport.h"
 
 #define SECTOR_SIZE		512
 #define ATA_SCSI_RBUF_SIZE	4096
@@ -64,9 +65,6 @@ static struct ata_device *__ata_scsi_find_dev(struct ata_port *ap,
 					const struct scsi_device *scsidev);
 static struct ata_device *ata_scsi_find_dev(struct ata_port *ap,
 					    const struct scsi_device *scsidev);
-static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
-			      unsigned int id, unsigned int lun);
-
 
 #define RW_RECOVERY_MPAGE 0x1
 #define RW_RECOVERY_MPAGE_LEN 12
@@ -106,17 +104,6 @@ static const u8 def_control_mpage[CONTROL_MPAGE_LEN] = {
 	0, 30	/* extended self test time, see 05-359r1 */
 };
 
-/*
- * libata transport template.  libata doesn't do real transport stuff.
- * It just needs the eh_timed_out hook.
- */
-static struct scsi_transport_template ata_scsi_transport_template = {
-	.eh_strategy_handler	= ata_scsi_error,
-	.eh_timed_out		= ata_scsi_timed_out,
-	.user_scan		= ata_scsi_user_scan,
-};
-
-
 static const struct {
 	enum link_pm	value;
 	const char	*name;
@@ -3334,7 +3321,7 @@ int ata_scsi_add_hosts(struct ata_host *host, struct scsi_host_template *sht)
 		*(struct ata_port **)&shost->hostdata[0] = ap;
 		ap->scsi_host = shost;
 
-		shost->transportt = &ata_scsi_transport_template;
+		shost->transportt = ata_scsi_transport_template;
 		shost->unique_id = ap->print_id;
 		shost->max_id = 16;
 		shost->max_lun = 1;
@@ -3616,8 +3603,8 @@ void ata_scsi_hotplug(struct work_struct *work)
  *	RETURNS:
  *	Zero.
  */
-static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
-			      unsigned int id, unsigned int lun)
+int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
+		       unsigned int id, unsigned int lun)
 {
 	struct ata_port *ap = ata_shost_to_port(shost);
 	unsigned long flags;
diff --git a/drivers/ata/libata-transport.c b/drivers/ata/libata-transport.c
new file mode 100644
index 000000000000..a80860f537e4
--- /dev/null
+++ b/drivers/ata/libata-transport.c
@@ -0,0 +1,773 @@
+/*
+ *  Copyright 2008 ioogle, Inc.  All rights reserved.
+ *	Released under GPL v2.
+ *
+ * Libata transport class.
+ *
+ * The ATA transport class contains common code to deal with ATA HBAs,
+ * an approximated representation of ATA topologies in the driver model,
+ * and various sysfs attributes to expose these topologies and management
+ * interfaces to user-space.
+ *
+ * There are 3 objects defined in in this class:
+ * - ata_port
+ * - ata_link
+ * - ata_device
+ * Each port has a link object. Each link can have up to two devices for PATA
+ * and generally one for SATA.
+ * If there is SATA port multiplier [PMP], 15 additional ata_link object are
+ * created.
+ *
+ * These objects are created when the ata host is initialized and when a PMP is
+ * found. They are removed only when the HBA is removed, cleaned before the
+ * error handler runs.
+ */
+
+
+#include <linux/kernel.h>
+#include <linux/blkdev.h>
+#include <linux/spinlock.h>
+#include <scsi/scsi_transport.h>
+#include <linux/libata.h>
+#include <linux/hdreg.h>
+#include <linux/uaccess.h>
+
+#include "libata.h"
+#include "libata-transport.h"
+
+#define ATA_PORT_ATTRS		2
+#define ATA_LINK_ATTRS		3
+#define ATA_DEV_ATTRS		9
+
+struct scsi_transport_template;
+struct scsi_transport_template *ata_scsi_transport_template;
+
+struct ata_internal {
+	struct scsi_transport_template t;
+
+	struct device_attribute private_port_attrs[ATA_PORT_ATTRS];
+	struct device_attribute private_link_attrs[ATA_LINK_ATTRS];
+	struct device_attribute private_dev_attrs[ATA_DEV_ATTRS];
+
+	struct transport_container link_attr_cont;
+	struct transport_container dev_attr_cont;
+
+	/*
+	 * The array of null terminated pointers to attributes
+	 * needed by scsi_sysfs.c
+	 */
+	struct device_attribute *link_attrs[ATA_LINK_ATTRS + 1];
+	struct device_attribute *port_attrs[ATA_PORT_ATTRS + 1];
+	struct device_attribute *dev_attrs[ATA_DEV_ATTRS + 1];
+};
+#define to_ata_internal(tmpl)	container_of(tmpl, struct ata_internal, t)
+
+
+#define tdev_to_device(d)					\
+	container_of((d), struct ata_device, tdev)
+#define transport_class_to_dev(dev)				\
+	tdev_to_device((dev)->parent)
+
+#define tdev_to_link(d)						\
+	container_of((d), struct ata_link, tdev)
+#define transport_class_to_link(dev)				\
+	tdev_to_link((dev)->parent)
+
+#define tdev_to_port(d)						\
+	container_of((d), struct ata_port, tdev)
+#define transport_class_to_port(dev)				\
+	tdev_to_port((dev)->parent)
+
+
+/* Device objects are always created whit link objects */
+static int ata_tdev_add(struct ata_device *dev);
+static void ata_tdev_delete(struct ata_device *dev);
+
+
+/*
+ * Hack to allow attributes of the same name in different objects.
+ */
+#define ATA_DEVICE_ATTR(_prefix,_name,_mode,_show,_store) \
+	struct device_attribute device_attr_##_prefix##_##_name = \
+	__ATTR(_name,_mode,_show,_store)
+
+#define ata_bitfield_name_match(title, table)			\
+static ssize_t							\
+get_ata_##title##_names(u32 table_key, char *buf)		\
+{								\
+	char *prefix = "";					\
+	ssize_t len = 0;					\
+	int i;							\
+								\
+	for (i = 0; i < ARRAY_SIZE(table); i++) {		\
+		if (table[i].value & table_key) {		\
+			len += sprintf(buf + len, "%s%s",	\
+				prefix, table[i].name);		\
+			prefix = ", ";				\
+		}						\
+	}							\
+	len += sprintf(buf + len, "\n");			\
+	return len;						\
+}
+
+#define ata_bitfield_name_search(title, table)			\
+static ssize_t							\
+get_ata_##title##_names(u32 table_key, char *buf)		\
+{								\
+	ssize_t len = 0;					\
+	int i;							\
+								\
+	for (i = 0; i < ARRAY_SIZE(table); i++) {		\
+		if (table[i].value == table_key) {		\
+			len += sprintf(buf + len, "%s",		\
+				table[i].name);			\
+			break;					\
+		}						\
+	}							\
+	len += sprintf(buf + len, "\n");			\
+	return len;						\
+}
+
+static struct {
+	u32		value;
+	char		*name;
+} ata_class_names[] = {
+	{ ATA_DEV_UNKNOWN,		"unknown" },
+	{ ATA_DEV_ATA,			"ata" },
+	{ ATA_DEV_ATA_UNSUP,		"ata" },
+	{ ATA_DEV_ATAPI,		"atapi" },
+	{ ATA_DEV_ATAPI_UNSUP,		"atapi" },
+	{ ATA_DEV_PMP,			"pmp" },
+	{ ATA_DEV_PMP_UNSUP,		"pmp" },
+	{ ATA_DEV_SEMB,			"semb" },
+	{ ATA_DEV_SEMB_UNSUP,		"semb" },
+	{ ATA_DEV_NONE,			"none" }
+};
+ata_bitfield_name_search(class, ata_class_names)
+
+
+static struct {
+	u32		value;
+	char		*name;
+} ata_err_names[] = {
+	{ AC_ERR_DEV,			"DeviceError" },
+	{ AC_ERR_HSM,			"HostStateMachineError" },
+	{ AC_ERR_TIMEOUT,		"Timeout" },
+	{ AC_ERR_MEDIA,			"MediaError" },
+	{ AC_ERR_ATA_BUS,		"BusError" },
+	{ AC_ERR_HOST_BUS,		"HostBusError" },
+	{ AC_ERR_SYSTEM,		"SystemError" },
+	{ AC_ERR_INVALID,		"InvalidArg" },
+	{ AC_ERR_OTHER,			"Unknown" },
+	{ AC_ERR_NODEV_HINT,		"NoDeviceHint" },
+	{ AC_ERR_NCQ,		 	"NCQError" }
+};
+ata_bitfield_name_match(err, ata_err_names)
+
+static struct {
+	u32		value;
+	char		*name;
+} ata_xfer_names[] = {
+	{ XFER_UDMA_7,			"XFER_UDMA_7" },
+	{ XFER_UDMA_6,			"XFER_UDMA_6" },
+	{ XFER_UDMA_5,			"XFER_UDMA_5" },
+	{ XFER_UDMA_4,			"XFER_UDMA_4" },
+	{ XFER_UDMA_3,			"XFER_UDMA_3" },
+	{ XFER_UDMA_2,			"XFER_UDMA_2" },
+	{ XFER_UDMA_1,			"XFER_UDMA_1" },
+	{ XFER_UDMA_0,			"XFER_UDMA_0" },
+	{ XFER_MW_DMA_4,		"XFER_MW_DMA_4" },
+	{ XFER_MW_DMA_3,		"XFER_MW_DMA_3" },
+	{ XFER_MW_DMA_2,		"XFER_MW_DMA_2" },
+	{ XFER_MW_DMA_1,		"XFER_MW_DMA_1" },
+	{ XFER_MW_DMA_0,		"XFER_MW_DMA_0" },
+	{ XFER_SW_DMA_2,		"XFER_SW_DMA_2" },
+	{ XFER_SW_DMA_1,		"XFER_SW_DMA_1" },
+	{ XFER_SW_DMA_0,		"XFER_SW_DMA_0" },
+	{ XFER_PIO_6,			"XFER_PIO_6" },
+	{ XFER_PIO_5,			"XFER_PIO_5" },
+	{ XFER_PIO_4,			"XFER_PIO_4" },
+	{ XFER_PIO_3,			"XFER_PIO_3" },
+	{ XFER_PIO_2,			"XFER_PIO_2" },
+	{ XFER_PIO_1,			"XFER_PIO_1" },
+	{ XFER_PIO_0,			"XFER_PIO_0" },
+	{ XFER_PIO_SLOW,		"XFER_PIO_SLOW" }
+};
+ata_bitfield_name_match(xfer,ata_xfer_names)
+
+/*
+ * ATA Port attributes
+ */
+#define ata_port_show_simple(field, name, format_string, cast)		\
+static ssize_t								\
+show_ata_port_##name(struct device *dev,				\
+		     struct device_attribute *attr, char *buf)		\
+{									\
+	struct ata_port *ap = transport_class_to_port(dev);		\
+									\
+	return snprintf(buf, 20, format_string, cast ap->field);	\
+}
+
+#define ata_port_simple_attr(field, name, format_string, type)		\
+	ata_port_show_simple(field, name, format_string, (type))	\
+static DEVICE_ATTR(name, S_IRUGO, show_ata_port_##name, NULL)
+
+ata_port_simple_attr(nr_pmp_links, nr_pmp_links, "%d\n", int);
+ata_port_simple_attr(stats.idle_irq, idle_irq, "%ld\n", unsigned long);
+
+static DECLARE_TRANSPORT_CLASS(ata_port_class,
+			       "ata_port", NULL, NULL, NULL);
+
+static void ata_tport_release(struct device *dev)
+{
+	put_device(dev->parent);
+}
+
+/**
+ * ata_is_port --  check if a struct device represents a ATA port
+ * @dev:	device to check
+ *
+ * Returns:
+ *	%1 if the device represents a ATA Port, %0 else
+ */
+int ata_is_port(const struct device *dev)
+{
+	return dev->release == ata_tport_release;
+}
+
+static int ata_tport_match(struct attribute_container *cont,
+			   struct device *dev)
+{
+	if (!ata_is_port(dev))
+		return 0;
+	return &ata_scsi_transport_template->host_attrs.ac == cont;
+}
+
+/**
+ * ata_tport_delete  --  remove ATA PORT
+ * @port:	ATA PORT to remove
+ *
+ * Removes the specified ATA PORT.  Remove the associated link as well.
+ */
+void ata_tport_delete(struct ata_port *ap)
+{
+	struct device *dev = &ap->tdev;
+
+	ata_tlink_delete(&ap->link);
+
+	transport_remove_device(dev);
+	device_del(dev);
+	transport_destroy_device(dev);
+	put_device(dev);
+}
+
+/** ata_tport_add - initialize a transport ATA port structure
+ *
+ * @parent:	parent device
+ * @ap:		existing ata_port structure
+ *
+ * Initialize a ATA port structure for sysfs.  It will be added to the device
+ * tree below the device specified by @parent which could be a PCI device.
+ *
+ * Returns %0 on success
+ */
+int ata_tport_add(struct device *parent,
+		  struct ata_port *ap)
+{
+	int error;
+	struct device *dev = &ap->tdev;
+
+	device_initialize(dev);
+
+	dev->parent = get_device(parent);
+	dev->release = ata_tport_release;
+	dev_set_name(dev, "ata%d", ap->print_id);
+	transport_setup_device(dev);
+	error = device_add(dev);
+	if (error) {
+		goto tport_err;
+	}
+
+	transport_add_device(dev);
+	transport_configure_device(dev);
+
+	error = ata_tlink_add(&ap->link);
+	if (error) {
+		goto tport_link_err;
+	}
+	return 0;
+
+ tport_link_err:
+	transport_remove_device(dev);
+	device_del(dev);
+
+ tport_err:
+	transport_destroy_device(dev);
+	put_device(dev);
+	return error;
+}
+
+
+/*
+ * ATA link attributes
+ */
+
+
+#define ata_link_show_linkspeed(field)					\
+static ssize_t								\
+show_ata_link_##field(struct device *dev,				\
+		      struct device_attribute *attr, char *buf)		\
+{									\
+	struct ata_link *link = transport_class_to_link(dev);		\
+									\
+	return sprintf(buf,"%s\n", sata_spd_string(fls(link->field)));	\
+}
+
+#define ata_link_linkspeed_attr(field)					\
+	ata_link_show_linkspeed(field)					\
+static DEVICE_ATTR(field, S_IRUGO, show_ata_link_##field, NULL)
+
+ata_link_linkspeed_attr(hw_sata_spd_limit);
+ata_link_linkspeed_attr(sata_spd_limit);
+ata_link_linkspeed_attr(sata_spd);
+
+
+static DECLARE_TRANSPORT_CLASS(ata_link_class,
+		"ata_link", NULL, NULL, NULL);
+
+static void ata_tlink_release(struct device *dev)
+{
+	put_device(dev->parent);
+}
+
+/**
+ * ata_is_link --  check if a struct device represents a ATA link
+ * @dev:	device to check
+ *
+ * Returns:
+ *	%1 if the device represents a ATA link, %0 else
+ */
+int ata_is_link(const struct device *dev)
+{
+	return dev->release == ata_tlink_release;
+}
+
+static int ata_tlink_match(struct attribute_container *cont,
+			   struct device *dev)
+{
+	struct ata_internal* i = to_ata_internal(ata_scsi_transport_template);
+	if (!ata_is_link(dev))
+		return 0;
+	return &i->link_attr_cont.ac == cont;
+}
+
+/**
+ * ata_tlink_delete  --  remove ATA LINK
+ * @port:	ATA LINK to remove
+ *
+ * Removes the specified ATA LINK.  remove associated ATA device(s) as well.
+ */
+void ata_tlink_delete(struct ata_link *link)
+{
+	struct device *dev = &link->tdev;
+	struct ata_device *ata_dev;
+
+	ata_for_each_dev(ata_dev, link, ALL) {
+		ata_tdev_delete(ata_dev);
+	}
+
+	transport_remove_device(dev);
+	device_del(dev);
+	transport_destroy_device(dev);
+	put_device(dev);
+}
+
+/**
+ * ata_tlink_add  --  initialize a transport ATA link structure
+ * @link:	allocated ata_link structure.
+ *
+ * Initialize an ATA LINK structure for sysfs.  It will be added in the
+ * device tree below the ATA PORT it belongs to.
+ *
+ * Returns %0 on success
+ */
+int ata_tlink_add(struct ata_link *link)
+{
+	struct device *dev = &link->tdev;
+	struct ata_port *ap = link->ap;
+	struct ata_device *ata_dev;
+	int error;
+
+	device_initialize(dev);
+	dev->parent = get_device(&ap->tdev);
+	dev->release = ata_tlink_release;
+	if (ata_is_host_link(link))
+		dev_set_name(dev, "link%d", ap->print_id);
+        else
+		dev_set_name(dev, "link%d.%d", ap->print_id, link->pmp);
+
+	transport_setup_device(dev);
+
+	error = device_add(dev);
+	if (error) {
+		goto tlink_err;
+	}
+
+	transport_add_device(dev);
+	transport_configure_device(dev);
+
+	ata_for_each_dev(ata_dev, link, ALL) {
+		error = ata_tdev_add(ata_dev);
+		if (error) {
+			goto tlink_dev_err;
+		}
+	}
+	return 0;
+  tlink_dev_err:
+	while (--ata_dev >= link->device) {
+		ata_tdev_delete(ata_dev);
+	}
+	transport_remove_device(dev);
+	device_del(dev);
+  tlink_err:
+	transport_destroy_device(dev);
+	put_device(dev);
+	return error;
+}
+
+/*
+ * ATA device attributes
+ */
+
+#define ata_dev_show_class(title, field)				\
+static ssize_t								\
+show_ata_dev_##field(struct device *dev,				\
+		     struct device_attribute *attr, char *buf)		\
+{									\
+	struct ata_device *ata_dev = transport_class_to_dev(dev);	\
+									\
+	return get_ata_##title##_names(ata_dev->field, buf);		\
+}
+
+#define ata_dev_attr(title, field)					\
+	ata_dev_show_class(title, field)				\
+static DEVICE_ATTR(field, S_IRUGO, show_ata_dev_##field, NULL)
+
+ata_dev_attr(class, class);
+ata_dev_attr(xfer, pio_mode);
+ata_dev_attr(xfer, dma_mode);
+ata_dev_attr(xfer, xfer_mode);
+
+
+#define ata_dev_show_simple(field, format_string, cast)		\
+static ssize_t								\
+show_ata_dev_##field(struct device *dev,				\
+		     struct device_attribute *attr, char *buf)		\
+{									\
+	struct ata_device *ata_dev = transport_class_to_dev(dev);	\
+									\
+	return snprintf(buf, 20, format_string, cast ata_dev->field);	\
+}
+
+#define ata_dev_simple_attr(field, format_string, type)	\
+	ata_dev_show_simple(field, format_string, (type))	\
+static DEVICE_ATTR(field, S_IRUGO, 			\
+		   show_ata_dev_##field, NULL)
+
+ata_dev_simple_attr(spdn_cnt, "%d\n", int);
+
+struct ata_show_ering_arg {
+	char* buf;
+	int written;
+};
+
+static int ata_show_ering(struct ata_ering_entry *ent, void *void_arg)
+{
+	struct ata_show_ering_arg* arg = void_arg;
+	struct timespec time;
+
+	jiffies_to_timespec(ent->timestamp,&time);
+	arg->written += sprintf(arg->buf + arg->written,
+			       "[%5lu.%06lu]",
+			       time.tv_sec, time.tv_nsec);
+	arg->written += get_ata_err_names(ent->err_mask,
+					  arg->buf + arg->written);
+	return 0;
+}
+
+static ssize_t
+show_ata_dev_ering(struct device *dev,
+		   struct device_attribute *attr, char *buf)
+{
+	struct ata_device *ata_dev = transport_class_to_dev(dev);
+	struct ata_show_ering_arg arg = { buf, 0 };
+
+	ata_ering_map(&ata_dev->ering, ata_show_ering, &arg);
+	return arg.written;
+}
+
+
+static DEVICE_ATTR(ering, S_IRUGO, show_ata_dev_ering, NULL);
+
+static ssize_t
+show_ata_dev_id(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct ata_device *ata_dev = transport_class_to_dev(dev);
+	int written = 0, i = 0;
+
+	if (ata_dev->class == ATA_DEV_PMP)
+		return 0;
+	for(i=0;i<ATA_ID_WORDS;i++)  {
+		written += snprintf(buf+written, 20, "%04x%c",
+				    ata_dev->id[i],
+				    ((i+1) & 7) ? ' ' : '\n');
+	}
+	return written;
+}
+
+static DEVICE_ATTR(id, S_IRUGO, show_ata_dev_id, NULL);
+
+static ssize_t
+show_ata_dev_gscr(struct device *dev,
+		  struct device_attribute *attr, char *buf)
+{
+	struct ata_device *ata_dev = transport_class_to_dev(dev);
+	int written = 0, i = 0;
+
+	if (ata_dev->class != ATA_DEV_PMP)
+		return 0;
+	for(i=0;i<SATA_PMP_GSCR_DWORDS;i++)  {
+		written += snprintf(buf+written, 20, "%08x%c",
+				    ata_dev->gscr[i],
+				    ((i+1) & 3) ? ' ' : '\n');
+	}
+	if (SATA_PMP_GSCR_DWORDS & 3)
+		buf[written-1] = '\n';
+	return written;
+}
+
+static DEVICE_ATTR(gscr, S_IRUGO, show_ata_dev_gscr, NULL);
+
+static DECLARE_TRANSPORT_CLASS(ata_dev_class,
+			       "ata_device", NULL, NULL, NULL);
+
+static void ata_tdev_release(struct device *dev)
+{
+	put_device(dev->parent);
+}
+
+/**
+ * ata_is_ata_dev  --  check if a struct device represents a ATA device
+ * @dev:	device to check
+ *
+ * Returns:
+ *	%1 if the device represents a ATA device, %0 else
+ */
+int ata_is_ata_dev(const struct device *dev)
+{
+	return dev->release == ata_tdev_release;
+}
+
+static int ata_tdev_match(struct attribute_container *cont,
+			  struct device *dev)
+{
+	struct ata_internal* i = to_ata_internal(ata_scsi_transport_template);
+	if (!ata_is_ata_dev(dev))
+		return 0;
+	return &i->dev_attr_cont.ac == cont;
+}
+
+/**
+ * ata_tdev_free  --  free a ATA LINK
+ * @dev:	ATA PHY to free
+ *
+ * Frees the specified ATA PHY.
+ *
+ * Note:
+ *   This function must only be called on a PHY that has not
+ *   successfully been added using ata_tdev_add().
+ */
+static void ata_tdev_free(struct ata_device *dev)
+{
+	transport_destroy_device(&dev->tdev);
+	put_device(&dev->tdev);
+}
+
+/**
+ * ata_tdev_delete  --  remove ATA device
+ * @port:	ATA PORT to remove
+ *
+ * Removes the specified ATA device.
+ */
+static void ata_tdev_delete(struct ata_device *ata_dev)
+{
+	struct device *dev = &ata_dev->tdev;
+
+	transport_remove_device(dev);
+	device_del(dev);
+	ata_tdev_free(ata_dev);
+}
+
+
+/**
+ * ata_tdev_add  --  initialize a transport ATA device structure.
+ * @ata_dev:	ata_dev structure.
+ *
+ * Initialize an ATA device structure for sysfs.  It will be added in the
+ * device tree below the ATA LINK device it belongs to.
+ *
+ * Returns %0 on success
+ */
+static int ata_tdev_add(struct ata_device *ata_dev)
+{
+	struct device *dev = &ata_dev->tdev;
+	struct ata_link *link = ata_dev->link;
+	struct ata_port *ap = link->ap;
+	int error;
+
+	device_initialize(dev);
+	dev->parent = get_device(&link->tdev);
+	dev->release = ata_tdev_release;
+	if (ata_is_host_link(link))
+		dev_set_name(dev, "dev%d.%d", ap->print_id,ata_dev->devno);
+        else
+		dev_set_name(dev, "dev%d.%d.0", ap->print_id, link->pmp);
+
+	transport_setup_device(dev);
+	error = device_add(dev);
+	if (error) {
+		ata_tdev_free(ata_dev);
+		return error;
+	}
+
+	transport_add_device(dev);
+	transport_configure_device(dev);
+	return 0;
+}
+
+
+/*
+ * Setup / Teardown code
+ */
+
+#define SETUP_TEMPLATE(attrb, field, perm, test)			\
+	i->private_##attrb[count] = dev_attr_##field;		       	\
+	i->private_##attrb[count].attr.mode = perm;			\
+	i->attrb[count] = &i->private_##attrb[count];			\
+	if (test)							\
+		count++
+
+#define SETUP_LINK_ATTRIBUTE(field)					\
+	SETUP_TEMPLATE(link_attrs, field, S_IRUGO, 1)
+
+#define SETUP_PORT_ATTRIBUTE(field)					\
+	SETUP_TEMPLATE(port_attrs, field, S_IRUGO, 1)
+
+#define SETUP_DEV_ATTRIBUTE(field)					\
+	SETUP_TEMPLATE(dev_attrs, field, S_IRUGO, 1)
+
+/**
+ * ata_attach_transport  --  instantiate ATA transport template
+ */
+struct scsi_transport_template *ata_attach_transport(void)
+{
+	struct ata_internal *i;
+	int count;
+
+	i = kzalloc(sizeof(struct ata_internal), GFP_KERNEL);
+	if (!i)
+		return NULL;
+
+	i->t.eh_strategy_handler	= ata_scsi_error;
+	i->t.eh_timed_out		= ata_scsi_timed_out;
+	i->t.user_scan			= ata_scsi_user_scan;
+
+	i->t.host_attrs.ac.attrs = &i->port_attrs[0];
+	i->t.host_attrs.ac.class = &ata_port_class.class;
+	i->t.host_attrs.ac.match = ata_tport_match;
+	transport_container_register(&i->t.host_attrs);
+
+	i->link_attr_cont.ac.class = &ata_link_class.class;
+	i->link_attr_cont.ac.attrs = &i->link_attrs[0];
+	i->link_attr_cont.ac.match = ata_tlink_match;
+	transport_container_register(&i->link_attr_cont);
+
+	i->dev_attr_cont.ac.class = &ata_dev_class.class;
+	i->dev_attr_cont.ac.attrs = &i->dev_attrs[0];
+	i->dev_attr_cont.ac.match = ata_tdev_match;
+	transport_container_register(&i->dev_attr_cont);
+
+	count = 0;
+	SETUP_PORT_ATTRIBUTE(nr_pmp_links);
+	SETUP_PORT_ATTRIBUTE(idle_irq);
+	BUG_ON(count > ATA_PORT_ATTRS);
+	i->port_attrs[count] = NULL;
+
+	count = 0;
+	SETUP_LINK_ATTRIBUTE(hw_sata_spd_limit);
+	SETUP_LINK_ATTRIBUTE(sata_spd_limit);
+	SETUP_LINK_ATTRIBUTE(sata_spd);
+	BUG_ON(count > ATA_LINK_ATTRS);
+	i->link_attrs[count] = NULL;
+
+	count = 0;
+	SETUP_DEV_ATTRIBUTE(class);
+	SETUP_DEV_ATTRIBUTE(pio_mode);
+	SETUP_DEV_ATTRIBUTE(dma_mode);
+	SETUP_DEV_ATTRIBUTE(xfer_mode);
+	SETUP_DEV_ATTRIBUTE(spdn_cnt);
+	SETUP_DEV_ATTRIBUTE(ering);
+	SETUP_DEV_ATTRIBUTE(id);
+	SETUP_DEV_ATTRIBUTE(gscr);
+	BUG_ON(count > ATA_DEV_ATTRS);
+	i->dev_attrs[count] = NULL;
+
+	return &i->t;
+}
+
+/**
+ * ata_release_transport  --  release ATA transport template instance
+ * @t:		transport template instance
+ */
+void ata_release_transport(struct scsi_transport_template *t)
+{
+	struct ata_internal *i = to_ata_internal(t);
+
+	transport_container_unregister(&i->t.host_attrs);
+	transport_container_unregister(&i->link_attr_cont);
+	transport_container_unregister(&i->dev_attr_cont);
+
+	kfree(i);
+}
+
+__init int libata_transport_init(void)
+{
+	int error;
+
+	error = transport_class_register(&ata_link_class);
+	if (error)
+		goto out_unregister_transport;
+	error = transport_class_register(&ata_port_class);
+	if (error)
+		goto out_unregister_link;
+	error = transport_class_register(&ata_dev_class);
+	if (error)
+		goto out_unregister_port;
+	return 0;
+
+ out_unregister_port:
+	transport_class_unregister(&ata_port_class);
+ out_unregister_link:
+	transport_class_unregister(&ata_link_class);
+ out_unregister_transport:
+	return error;
+
+}
+
+void __exit libata_transport_exit(void)
+{
+	transport_class_unregister(&ata_link_class);
+	transport_class_unregister(&ata_port_class);
+	transport_class_unregister(&ata_dev_class);
+}
diff --git a/drivers/ata/libata-transport.h b/drivers/ata/libata-transport.h
new file mode 100644
index 000000000000..2820cf864f11
--- /dev/null
+++ b/drivers/ata/libata-transport.h
@@ -0,0 +1,18 @@
+#ifndef _LIBATA_TRANSPORT_H
+#define _LIBATA_TRANSPORT_H
+
+
+extern struct scsi_transport_template *ata_scsi_transport_template;
+
+int ata_tlink_add(struct ata_link *link);
+void ata_tlink_delete(struct ata_link *link);
+
+int ata_tport_add(struct device *parent, struct ata_port *ap);
+void ata_tport_delete(struct ata_port *ap);
+
+struct scsi_transport_template *ata_attach_transport(void);
+void ata_release_transport(struct scsi_transport_template *t);
+
+__init int libata_transport_init(void);
+void __exit libata_transport_exit(void);
+#endif
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 9ce1ecc63e39..142102b94df5 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -102,6 +102,7 @@ extern int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg);
 extern struct ata_port *ata_port_alloc(struct ata_host *host);
 extern void ata_dev_enable_pm(struct ata_device *dev, enum link_pm policy);
 extern void ata_lpm_schedule(struct ata_port *ap, enum link_pm);
+extern const char *sata_spd_string(unsigned int spd);
 
 /* libata-acpi.c */
 #ifdef CONFIG_ATA_ACPI
@@ -137,6 +138,9 @@ extern void ata_scsi_hotplug(struct work_struct *work);
 extern void ata_schedule_scsi_eh(struct Scsi_Host *shost);
 extern void ata_scsi_dev_rescan(struct work_struct *work);
 extern int ata_bus_probe(struct ata_port *ap);
+extern int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
+			      unsigned int id, unsigned int lun);
+
 
 /* libata-eh.c */
 extern unsigned long ata_internal_cmd_timeout(struct ata_device *dev, u8 cmd);
@@ -164,6 +168,9 @@ extern int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 			  ata_postreset_fn_t postreset,
 			  struct ata_link **r_failed_disk);
 extern void ata_eh_finish(struct ata_port *ap);
+extern int ata_ering_map(struct ata_ering *ering,
+			 int (*map_fn)(struct ata_ering_entry *, void *),
+		  	 void *arg);
 
 /* libata-pmp.c */
 #ifdef CONFIG_SATA_PMP
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 45fb2967b66d..c50f66d4382c 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -604,6 +604,7 @@ struct ata_device {
 	union acpi_object	*gtf_cache;
 	unsigned int		gtf_filter;
 #endif
+	struct device		tdev;
 	/* n_sector is CLEAR_BEGIN, read comment above CLEAR_BEGIN */
 	u64			n_sectors;	/* size of device, if ATA */
 	u64			n_native_sectors; /* native size, if ATA */
@@ -690,6 +691,7 @@ struct ata_link {
 	struct ata_port		*ap;
 	int			pmp;		/* port multiplier port # */
 
+	struct device		tdev;
 	unsigned int		active_tag;	/* active tag on this link */
 	u32			sactive;	/* active NCQ commands */
 
@@ -707,6 +709,8 @@ struct ata_link {
 
 	struct ata_device	device[ATA_MAX_DEVICES];
 };
+#define ATA_LINK_CLEAR_BEGIN		offsetof(struct ata_link, active_tag)
+#define ATA_LINK_CLEAR_END		offsetof(struct ata_link, device[0])
 
 struct ata_port {
 	struct Scsi_Host	*scsi_host; /* our co-allocated scsi host */
@@ -752,6 +756,7 @@ struct ata_port {
 	struct ata_port_stats	stats;
 	struct ata_host		*host;
 	struct device 		*dev;
+	struct device		tdev;
 
 	struct mutex		scsi_scan_mutex;
 	struct delayed_work	hotplug_task;
-- 
cgit v1.2.3


From 295124dce4ddfd40b1f12d3ffd2779673e87c701 Mon Sep 17 00:00:00 2001
From: Grant Grundler <grundler@google.com>
Date: Tue, 17 Aug 2010 10:56:53 -0700
Subject: [libata] support for > 512 byte sectors (e.g. 4K Native)

This change enables my x86 machine to recognize and talk to a
"Native 4K" SATA device.

When I started working on this, I didn't know Matthew Wilcox had
posted a similar patch 2 years ago:
  http://git.kernel.org/?p=linux/kernel/git/willy/ata.git;a=shortlog;h=refs/heads/ata-large-sectors

Gwendal Grignou pointed me at the the above code and small portions of
this patch include Matthew's work. That's why Mathew is first on the
"Signed-off-by:". I've NOT included his use of a bitmap to determine
512 vs Native for ATA command block size - just used a simple table.
And bugs are almost certainly mine.

Lastly, the patch has been tested with a native 4K 'Engineering
Sample' drive provided by Hitachi GST.

Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Signed-off-by: Grant Grundler <grundler@google.com>
Reviewed-by: Gwendal Grignou <gwendal@google.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-scsi.c | 96 ++++++++++++++++++++++++++++++++---------------
 include/linux/ata.h       | 46 ++++++++++++++++++++---
 2 files changed, 105 insertions(+), 37 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index c16f5c151735..f1c0118c6d4b 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -53,7 +53,6 @@
 #include "libata.h"
 #include "libata-transport.h"
 
-#define SECTOR_SIZE		512
 #define ATA_SCSI_RBUF_SIZE	4096
 
 static DEFINE_SPINLOCK(ata_scsi_rbuf_lock);
@@ -503,7 +502,7 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg)
 	memset(scsi_cmd, 0, sizeof(scsi_cmd));
 
 	if (args[3]) {
-		argsize = SECTOR_SIZE * args[3];
+		argsize = ATA_SECT_SIZE * args[3];
 		argbuf = kmalloc(argsize, GFP_KERNEL);
 		if (argbuf == NULL) {
 			rc = -ENOMEM;
@@ -1137,8 +1136,9 @@ static int ata_scsi_dev_config(struct scsi_device *sdev,
 		blk_queue_dma_drain(q, atapi_drain_needed, buf, ATAPI_MAX_DRAIN);
 	} else {
 		/* ATA devices must be sector aligned */
+		sdev->sector_size = ata_id_logical_sector_size(dev->id);
 		blk_queue_update_dma_alignment(sdev->request_queue,
-					       ATA_SECT_SIZE - 1);
+					       sdev->sector_size - 1);
 		sdev->manage_start_stop = 1;
 	}
 
@@ -1153,6 +1153,7 @@ static int ata_scsi_dev_config(struct scsi_device *sdev,
 		scsi_adjust_queue_depth(sdev, MSG_SIMPLE_TAG, depth);
 	}
 
+	dev->sdev = sdev;
 	return 0;
 }
 
@@ -1683,7 +1684,7 @@ static unsigned int ata_scsi_rw_xlat(struct ata_queued_cmd *qc)
 		goto nothing_to_do;
 
 	qc->flags |= ATA_QCFLAG_IO;
-	qc->nbytes = n_block * ATA_SECT_SIZE;
+	qc->nbytes = n_block * scmd->device->sector_size;
 
 	rc = ata_build_rw_tf(&qc->tf, qc->dev, block, n_block, tf_flags,
 			     qc->tag);
@@ -2110,7 +2111,7 @@ static unsigned int ata_scsiop_inq_89(struct ata_scsi_args *args, u8 *rbuf)
 
 static unsigned int ata_scsiop_inq_b0(struct ata_scsi_args *args, u8 *rbuf)
 {
-	u32 min_io_sectors;
+	u16 min_io_sectors;
 
 	rbuf[1] = 0xb0;
 	rbuf[3] = 0x3c;		/* required VPD size with unmap support */
@@ -2122,10 +2123,7 @@ static unsigned int ata_scsiop_inq_b0(struct ata_scsi_args *args, u8 *rbuf)
 	 * logical than physical sector size we need to figure out what the
 	 * latter is.
 	 */
-	if (ata_id_has_large_logical_sectors(args->id))
-		min_io_sectors = ata_id_logical_per_physical_sectors(args->id);
-	else
-		min_io_sectors = 1;
+	min_io_sectors = 1 << ata_id_log2_per_physical_sector(args->id);
 	put_unaligned_be16(min_io_sectors, &rbuf[6]);
 
 	/*
@@ -2384,21 +2382,13 @@ static unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf)
 {
 	struct ata_device *dev = args->dev;
 	u64 last_lba = dev->n_sectors - 1; /* LBA of the last block */
-	u8 log_per_phys = 0;
-	u16 lowest_aligned = 0;
-	u16 word_106 = dev->id[106];
-	u16 word_209 = dev->id[209];
-
-	if ((word_106 & 0xc000) == 0x4000) {
-		/* Number and offset of logical sectors per physical sector */
-		if (word_106 & (1 << 13))
-			log_per_phys = word_106 & 0xf;
-		if ((word_209 & 0xc000) == 0x4000) {
-			u16 first = dev->id[209] & 0x3fff;
-			if (first > 0)
-				lowest_aligned = (1 << log_per_phys) - first;
-		}
-	}
+	u32 sector_size; /* physical sector size in bytes */
+	u8 log2_per_phys;
+	u16 lowest_aligned;
+
+	sector_size = ata_id_logical_sector_size(dev->id);
+	log2_per_phys = ata_id_log2_per_physical_sector(dev->id);
+	lowest_aligned = ata_id_logical_sector_offset(dev->id, log2_per_phys);
 
 	VPRINTK("ENTER\n");
 
@@ -2413,8 +2403,10 @@ static unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf)
 		rbuf[3] = last_lba;
 
 		/* sector size */
-		rbuf[6] = ATA_SECT_SIZE >> 8;
-		rbuf[7] = ATA_SECT_SIZE & 0xff;
+		rbuf[4] = sector_size >> (8 * 3);
+		rbuf[5] = sector_size >> (8 * 2);
+		rbuf[6] = sector_size >> (8 * 1);
+		rbuf[7] = sector_size;
 	} else {
 		/* sector count, 64-bit */
 		rbuf[0] = last_lba >> (8 * 7);
@@ -2427,11 +2419,13 @@ static unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf)
 		rbuf[7] = last_lba;
 
 		/* sector size */
-		rbuf[10] = ATA_SECT_SIZE >> 8;
-		rbuf[11] = ATA_SECT_SIZE & 0xff;
+		rbuf[ 8] = sector_size >> (8 * 3);
+		rbuf[ 9] = sector_size >> (8 * 2);
+		rbuf[10] = sector_size >> (8 * 1);
+		rbuf[11] = sector_size;
 
 		rbuf[12] = 0;
-		rbuf[13] = log_per_phys;
+		rbuf[13] = log2_per_phys;
 		rbuf[14] = (lowest_aligned >> 8) & 0x3f;
 		rbuf[15] = lowest_aligned;
 
@@ -2875,9 +2869,8 @@ static unsigned int ata_scsi_pass_thru(struct ata_queued_cmd *qc)
 	tf->device = dev->devno ?
 		tf->device | ATA_DEV1 : tf->device & ~ATA_DEV1;
 
-	/* READ/WRITE LONG use a non-standard sect_size */
-	qc->sect_size = ATA_SECT_SIZE;
 	switch (tf->command) {
+	/* READ/WRITE LONG use a non-standard sect_size */
 	case ATA_CMD_READ_LONG:
 	case ATA_CMD_READ_LONG_ONCE:
 	case ATA_CMD_WRITE_LONG:
@@ -2885,6 +2878,45 @@ static unsigned int ata_scsi_pass_thru(struct ata_queued_cmd *qc)
 		if (tf->protocol != ATA_PROT_PIO || tf->nsect != 1)
 			goto invalid_fld;
 		qc->sect_size = scsi_bufflen(scmd);
+		break;
+
+	/* commands using reported Logical Block size (e.g. 512 or 4K) */
+	case ATA_CMD_CFA_WRITE_NE:
+	case ATA_CMD_CFA_TRANS_SECT:
+	case ATA_CMD_CFA_WRITE_MULT_NE:
+	/* XXX: case ATA_CMD_CFA_WRITE_SECTORS_WITHOUT_ERASE: */
+	case ATA_CMD_READ:
+	case ATA_CMD_READ_EXT:
+	case ATA_CMD_READ_QUEUED:
+	/* XXX: case ATA_CMD_READ_QUEUED_EXT: */
+	case ATA_CMD_FPDMA_READ:
+	case ATA_CMD_READ_MULTI:
+	case ATA_CMD_READ_MULTI_EXT:
+	case ATA_CMD_PIO_READ:
+	case ATA_CMD_PIO_READ_EXT:
+	case ATA_CMD_READ_STREAM_DMA_EXT:
+	case ATA_CMD_READ_STREAM_EXT:
+	case ATA_CMD_VERIFY:
+	case ATA_CMD_VERIFY_EXT:
+	case ATA_CMD_WRITE:
+	case ATA_CMD_WRITE_EXT:
+	case ATA_CMD_WRITE_FUA_EXT:
+	case ATA_CMD_WRITE_QUEUED:
+	case ATA_CMD_WRITE_QUEUED_FUA_EXT:
+	case ATA_CMD_FPDMA_WRITE:
+	case ATA_CMD_WRITE_MULTI:
+	case ATA_CMD_WRITE_MULTI_EXT:
+	case ATA_CMD_WRITE_MULTI_FUA_EXT:
+	case ATA_CMD_PIO_WRITE:
+	case ATA_CMD_PIO_WRITE_EXT:
+	case ATA_CMD_WRITE_STREAM_DMA_EXT:
+	case ATA_CMD_WRITE_STREAM_EXT:
+		qc->sect_size = scmd->device->sector_size;
+		break;
+
+	/* Everything else uses 512 byte "sectors" */
+	default:
+		qc->sect_size = ATA_SECT_SIZE;
 	}
 
 	/*
@@ -3380,6 +3412,8 @@ void ata_scsi_scan_host(struct ata_port *ap, int sync)
 			if (!IS_ERR(sdev)) {
 				dev->sdev = sdev;
 				scsi_device_put(sdev);
+			} else {
+				dev->sdev = NULL;
 			}
 		}
 	}
diff --git a/include/linux/ata.h b/include/linux/ata.h
index fe6e681a9d74..0c4929fa34d3 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -89,6 +89,7 @@ enum {
 	ATA_ID_SPG		= 98,
 	ATA_ID_LBA_CAPACITY_2	= 100,
 	ATA_ID_SECTOR_SIZE	= 106,
+	ATA_ID_LOGICAL_SECTOR_SIZE	= 117,	/* and 118 */
 	ATA_ID_LAST_LUN		= 126,
 	ATA_ID_DLF		= 128,
 	ATA_ID_CSFO		= 129,
@@ -640,16 +641,49 @@ static inline int ata_id_flush_ext_enabled(const u16 *id)
 	return (id[ATA_ID_CFS_ENABLE_2] & 0x2400) == 0x2400;
 }
 
-static inline int ata_id_has_large_logical_sectors(const u16 *id)
+static inline u32 ata_id_logical_sector_size(const u16 *id)
 {
-	if ((id[ATA_ID_SECTOR_SIZE] & 0xc000) != 0x4000)
-		return 0;
-	return id[ATA_ID_SECTOR_SIZE] & (1 << 13);
+	/* T13/1699-D Revision 6a, Sep 6, 2008. Page 128.
+	 * IDENTIFY DEVICE data, word 117-118.
+	 * 0xd000 ignores bit 13 (logical:physical > 1)
+	 */
+	if ((id[ATA_ID_SECTOR_SIZE] & 0xd000) == 0x5000)
+		return (((id[ATA_ID_LOGICAL_SECTOR_SIZE+1] << 16)
+			 + id[ATA_ID_LOGICAL_SECTOR_SIZE]) * sizeof(u16)) ;
+	return ATA_SECT_SIZE;
+}
+
+static inline u8 ata_id_log2_per_physical_sector(const u16 *id)
+{
+	/* T13/1699-D Revision 6a, Sep 6, 2008. Page 128.
+	 * IDENTIFY DEVICE data, word 106.
+	 * 0xe000 ignores bit 12 (logical sector > 512 bytes)
+	 */
+	if ((id[ATA_ID_SECTOR_SIZE] & 0xe000) == 0x6000)
+		return (id[ATA_ID_SECTOR_SIZE] & 0xf);
+	return 0;
 }
 
-static inline u16 ata_id_logical_per_physical_sectors(const u16 *id)
+/* Offset of logical sectors relative to physical sectors.
+ *
+ * If device has more than one logical sector per physical sector
+ * (aka 512 byte emulation), vendors might offset the "sector 0" address
+ * so sector 63 is "naturally aligned" - e.g. FAT partition table.
+ * This avoids Read/Mod/Write penalties when using FAT partition table
+ * and updating "well aligned" (FS perspective) physical sectors on every
+ * transaction.
+ */
+static inline u16 ata_id_logical_sector_offset(const u16 *id,
+	 u8 log2_per_phys)
 {
-	return 1 << (id[ATA_ID_SECTOR_SIZE] & 0xf);
+	u16 word_209 = id[209];
+
+	if ((log2_per_phys > 1) && (word_209 & 0xc000) == 0x4000) {
+		u16 first = word_209 & 0x3fff;
+		if (first > 0)
+			return (1 << log2_per_phys) - first;
+	}
+	return 0;
 }
 
 static inline int ata_id_has_lba48(const u16 *id)
-- 
cgit v1.2.3


From c93b263e0d4fa8ce5fec0142a98196d1a127e845 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 1 Sep 2010 17:50:04 +0200
Subject: libata: clean up lpm related symbols and sysfs show/store functions

Link power management related symbols are in confusing state w/ mixed
usages of lpm, ipm and pm.  This patch cleans up lpm related symbols
and sysfs show/store functions as follows.

* lpm states - NOT_AVAILABLE, MIN_POWER, MAX_PERFORMANCE and
  MEDIUM_POWER are renamed to ATA_LPM_UNKNOWN and
  ATA_LPM_{MIN|MAX|MED}_POWER.

* Pre/postfixes are unified to lpm.

* sysfs show/store functions for link_power_management_policy were
  curiously named get/put and unnecessarily complex.  Renamed to
  show/store and simplified.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/ahci.c          |  2 +-
 drivers/ata/ahci.h          |  2 +-
 drivers/ata/ahci_platform.c |  2 +-
 drivers/ata/libahci.c       | 20 +++++++-------
 drivers/ata/libata-core.c   | 50 +++++++++++++++++-----------------
 drivers/ata/libata-eh.c     |  2 +-
 drivers/ata/libata-scsi.c   | 66 +++++++++++++++------------------------------
 drivers/ata/libata.h        |  6 +++--
 include/linux/libata.h      | 29 ++++++++++----------
 9 files changed, 79 insertions(+), 100 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 99d0e5a51148..68f7b4216501 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1209,7 +1209,7 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 				   0x100 + ap->port_no * 0x80, "port");
 
 		/* set initial link pm policy */
-		ap->pm_policy = NOT_AVAILABLE;
+		ap->lpm_policy = ATA_LPM_UNKNOWN;
 
 		/* set enclosure management message type */
 		if (ap->flags & ATA_FLAG_EM)
diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h
index e5fdeebf9ef0..93867d3f8dd3 100644
--- a/drivers/ata/ahci.h
+++ b/drivers/ata/ahci.h
@@ -216,7 +216,7 @@ enum {
 	AHCI_FLAG_COMMON		= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
 					  ATA_FLAG_MMIO | ATA_FLAG_PIO_DMA |
 					  ATA_FLAG_ACPI_SATA | ATA_FLAG_AN |
-					  ATA_FLAG_IPM,
+					  ATA_FLAG_LPM,
 
 	ICH_MAP				= 0x90, /* ICH MAP register */
 
diff --git a/drivers/ata/ahci_platform.c b/drivers/ata/ahci_platform.c
index 84b643270e7a..07556853c0d6 100644
--- a/drivers/ata/ahci_platform.c
+++ b/drivers/ata/ahci_platform.c
@@ -130,7 +130,7 @@ static int __init ahci_probe(struct platform_device *pdev)
 		ata_port_desc(ap, "port 0x%x", 0x100 + ap->port_no * 0x80);
 
 		/* set initial link pm policy */
-		ap->pm_policy = NOT_AVAILABLE;
+		ap->lpm_policy = ATA_LPM_UNKNOWN;
 
 		/* set enclosure management message type */
 		if (ap->flags & ATA_FLAG_EM)
diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index 8eea309ea212..ed7803f2b4f1 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -56,8 +56,7 @@ MODULE_PARM_DESC(skip_host_reset, "skip global host reset (0=don't skip, 1=skip)
 module_param_named(ignore_sss, ahci_ignore_sss, int, 0444);
 MODULE_PARM_DESC(ignore_sss, "Ignore staggered spinup flag (0=don't ignore, 1=ignore)");
 
-static int ahci_enable_alpm(struct ata_port *ap,
-		enum link_pm policy);
+static int ahci_enable_alpm(struct ata_port *ap, enum ata_lpm_policy policy);
 static void ahci_disable_alpm(struct ata_port *ap);
 static ssize_t ahci_led_show(struct ata_port *ap, char *buf);
 static ssize_t ahci_led_store(struct ata_port *ap, const char *buf,
@@ -649,7 +648,7 @@ static void ahci_disable_alpm(struct ata_port *ap)
 	u32 cmd;
 	struct ahci_port_priv *pp = ap->private_data;
 
-	/* IPM bits should be disabled by libata-core */
+	/* LPM bits should be disabled by libata-core */
 	/* get the existing command bits */
 	cmd = readl(port_mmio + PORT_CMD);
 
@@ -691,8 +690,7 @@ static void ahci_disable_alpm(struct ata_port *ap)
 	 */
 }
 
-static int ahci_enable_alpm(struct ata_port *ap,
-	enum link_pm policy)
+static int ahci_enable_alpm(struct ata_port *ap, enum ata_lpm_policy policy)
 {
 	struct ahci_host_priv *hpriv = ap->host->private_data;
 	void __iomem *port_mmio = ahci_port_base(ap);
@@ -705,21 +703,21 @@ static int ahci_enable_alpm(struct ata_port *ap,
 		return -EINVAL;
 
 	switch (policy) {
-	case MAX_PERFORMANCE:
-	case NOT_AVAILABLE:
+	case ATA_LPM_MAX_POWER:
+	case ATA_LPM_UNKNOWN:
 		/*
-		 * if we came here with NOT_AVAILABLE,
+		 * if we came here with ATA_LPM_UNKNOWN,
 		 * it just means this is the first time we
 		 * have tried to enable - default to max performance,
 		 * and let the user go to lower power modes on request.
 		 */
 		ahci_disable_alpm(ap);
 		return 0;
-	case MIN_POWER:
+	case ATA_LPM_MIN_POWER:
 		/* configure HBA to enter SLUMBER */
 		asp = PORT_CMD_ASP;
 		break;
-	case MEDIUM_POWER:
+	case ATA_LPM_MED_POWER:
 		/* configure HBA to enter PARTIAL */
 		asp = 0;
 		break;
@@ -762,7 +760,7 @@ static int ahci_enable_alpm(struct ata_port *ap,
 	writel(cmd, port_mmio + PORT_CMD);
 	cmd = readl(port_mmio + PORT_CMD);
 
-	/* IPM bits should be set by libata-core */
+	/* LPM bits should be set by libata-core */
 	return 0;
 }
 
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 92cd5f375b8f..380ceb000aad 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -1030,7 +1030,7 @@ const char *sata_spd_string(unsigned int spd)
 	return spd_str[spd - 1];
 }
 
-static int ata_dev_set_dipm(struct ata_device *dev, enum link_pm policy)
+static int ata_dev_set_dipm(struct ata_device *dev, enum ata_lpm_policy policy)
 {
 	struct ata_link *link = dev->link;
 	struct ata_port *ap = link->ap;
@@ -1040,14 +1040,14 @@ static int ata_dev_set_dipm(struct ata_device *dev, enum link_pm policy)
 
 	/*
 	 * disallow DIPM for drivers which haven't set
-	 * ATA_FLAG_IPM.  This is because when DIPM is enabled,
+	 * ATA_FLAG_LPM.  This is because when DIPM is enabled,
 	 * phy ready will be set in the interrupt status on
 	 * state changes, which will cause some drivers to
 	 * think there are errors - additionally drivers will
 	 * need to disable hot plug.
 	 */
-	if (!(ap->flags & ATA_FLAG_IPM) || !ata_dev_enabled(dev)) {
-		ap->pm_policy = NOT_AVAILABLE;
+	if (!(ap->flags & ATA_FLAG_LPM) || !ata_dev_enabled(dev)) {
+		ap->lpm_policy = ATA_LPM_UNKNOWN;
 		return -EINVAL;
 	}
 
@@ -1066,8 +1066,8 @@ static int ata_dev_set_dipm(struct ata_device *dev, enum link_pm policy)
 		return rc;
 
 	switch (policy) {
-	case MIN_POWER:
-		/* no restrictions on IPM transitions */
+	case ATA_LPM_MIN_POWER:
+		/* no restrictions on LPM transitions */
 		scontrol &= ~(0x3 << 8);
 		rc = sata_scr_write(link, SCR_CONTROL, scontrol);
 		if (rc)
@@ -1078,8 +1078,8 @@ static int ata_dev_set_dipm(struct ata_device *dev, enum link_pm policy)
 			err_mask = ata_dev_set_feature(dev,
 					SETFEATURES_SATA_ENABLE, SATA_DIPM);
 		break;
-	case MEDIUM_POWER:
-		/* allow IPM to PARTIAL */
+	case ATA_LPM_MED_POWER:
+		/* allow LPM to PARTIAL */
 		scontrol &= ~(0x1 << 8);
 		scontrol |= (0x2 << 8);
 		rc = sata_scr_write(link, SCR_CONTROL, scontrol);
@@ -1087,21 +1087,21 @@ static int ata_dev_set_dipm(struct ata_device *dev, enum link_pm policy)
 			return rc;
 
 		/*
-		 * we don't have to disable DIPM since IPM flags
+		 * we don't have to disable DIPM since LPM flags
 		 * disallow transitions to SLUMBER, which effectively
 		 * disable DIPM if it does not support PARTIAL
 		 */
 		break;
-	case NOT_AVAILABLE:
-	case MAX_PERFORMANCE:
-		/* disable all IPM transitions */
+	case ATA_LPM_UNKNOWN:
+	case ATA_LPM_MAX_POWER:
+		/* disable all LPM transitions */
 		scontrol |= (0x3 << 8);
 		rc = sata_scr_write(link, SCR_CONTROL, scontrol);
 		if (rc)
 			return rc;
 
 		/*
-		 * we don't have to disable DIPM since IPM flags
+		 * we don't have to disable DIPM since LPM flags
 		 * disallow all transitions which effectively
 		 * disable DIPM anyway.
 		 */
@@ -1125,9 +1125,9 @@ static int ata_dev_set_dipm(struct ata_device *dev, enum link_pm policy)
  *	enabling Host Initiated Power management.
  *
  *	Locking: Caller.
- *	Returns: -EINVAL if IPM is not supported, 0 otherwise.
+ *	Returns: -EINVAL if LPM is not supported, 0 otherwise.
  */
-void ata_dev_enable_pm(struct ata_device *dev, enum link_pm policy)
+void ata_dev_enable_pm(struct ata_device *dev, enum ata_lpm_policy policy)
 {
 	int rc = 0;
 	struct ata_port *ap = dev->link->ap;
@@ -1141,9 +1141,9 @@ void ata_dev_enable_pm(struct ata_device *dev, enum link_pm policy)
 
 enable_pm_out:
 	if (rc)
-		ap->pm_policy = MAX_PERFORMANCE;
+		ap->lpm_policy = ATA_LPM_MAX_POWER;
 	else
-		ap->pm_policy = policy;
+		ap->lpm_policy = policy;
 	return /* rc */;	/* hopefully we can use 'rc' eventually */
 }
 
@@ -1164,15 +1164,15 @@ static void ata_dev_disable_pm(struct ata_device *dev)
 {
 	struct ata_port *ap = dev->link->ap;
 
-	ata_dev_set_dipm(dev, MAX_PERFORMANCE);
+	ata_dev_set_dipm(dev, ATA_LPM_MAX_POWER);
 	if (ap->ops->disable_pm)
 		ap->ops->disable_pm(ap);
 }
 #endif	/* CONFIG_PM */
 
-void ata_lpm_schedule(struct ata_port *ap, enum link_pm policy)
+void ata_lpm_schedule(struct ata_port *ap, enum ata_lpm_policy policy)
 {
-	ap->pm_policy = policy;
+	ap->lpm_policy = policy;
 	ap->link.eh_info.action |= ATA_EH_LPM;
 	ap->link.eh_info.flags |= ATA_EHI_NO_AUTOPSY;
 	ata_port_schedule_eh(ap);
@@ -1201,7 +1201,7 @@ static void ata_lpm_disable(struct ata_host *host)
 
 	for (i = 0; i < host->n_ports; i++) {
 		struct ata_port *ap = host->ports[i];
-		ata_lpm_schedule(ap, ap->pm_policy);
+		ata_lpm_schedule(ap, ap->lpm_policy);
 	}
 }
 #endif	/* CONFIG_PM */
@@ -2564,7 +2564,7 @@ int ata_dev_configure(struct ata_device *dev)
 	if (dev->flags & ATA_DFLAG_LBA48)
 		dev->max_sectors = ATA_MAX_SECTORS_LBA48;
 
-	if (!(dev->horkage & ATA_HORKAGE_IPM)) {
+	if (!(dev->horkage & ATA_HORKAGE_LPM)) {
 		if (ata_id_has_hipm(dev->id))
 			dev->flags |= ATA_DFLAG_HIPM;
 		if (ata_id_has_dipm(dev->id))
@@ -2591,11 +2591,11 @@ int ata_dev_configure(struct ata_device *dev)
 		dev->max_sectors = min_t(unsigned int, ATA_MAX_SECTORS_128,
 					 dev->max_sectors);
 
-	if (ata_dev_blacklisted(dev) & ATA_HORKAGE_IPM) {
-		dev->horkage |= ATA_HORKAGE_IPM;
+	if (ata_dev_blacklisted(dev) & ATA_HORKAGE_LPM) {
+		dev->horkage |= ATA_HORKAGE_LPM;
 
 		/* reset link pm_policy for this port to no pm */
-		ap->pm_policy = MAX_PERFORMANCE;
+		ap->lpm_policy = ATA_LPM_MAX_POWER;
 	}
 
 	if (ap->ops->dev_config)
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 95838b382231..96aa75ddf87f 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -3570,7 +3570,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 		/* configure link power saving */
 		if (ehc->i.action & ATA_EH_LPM)
 			ata_for_each_dev(dev, link, ALL)
-				ata_dev_enable_pm(dev, ap->pm_policy);
+				ata_dev_enable_pm(dev, ap->lpm_policy);
 
 		/* this link is okay now */
 		ehc->i.flags = 0;
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index f1c0118c6d4b..aa56681f68db 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -103,72 +103,50 @@ static const u8 def_control_mpage[CONTROL_MPAGE_LEN] = {
 	0, 30	/* extended self test time, see 05-359r1 */
 };
 
-static const struct {
-	enum link_pm	value;
-	const char	*name;
-} link_pm_policy[] = {
-	{ NOT_AVAILABLE, "max_performance" },
-	{ MIN_POWER, "min_power" },
-	{ MAX_PERFORMANCE, "max_performance" },
-	{ MEDIUM_POWER, "medium_power" },
+static const char *ata_lpm_policy_names[] = {
+	[ATA_LPM_UNKNOWN]	= "max_performance",
+	[ATA_LPM_MAX_POWER]	= "max_performance",
+	[ATA_LPM_MED_POWER]	= "medium_power",
+	[ATA_LPM_MIN_POWER]	= "min_power",
 };
 
-static const char *ata_scsi_lpm_get(enum link_pm policy)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(link_pm_policy); i++)
-		if (link_pm_policy[i].value == policy)
-			return link_pm_policy[i].name;
-
-	return NULL;
-}
-
-static ssize_t ata_scsi_lpm_put(struct device *dev,
-				struct device_attribute *attr,
-				const char *buf, size_t count)
+static ssize_t ata_scsi_lpm_store(struct device *dev,
+				  struct device_attribute *attr,
+				  const char *buf, size_t count)
 {
 	struct Scsi_Host *shost = class_to_shost(dev);
 	struct ata_port *ap = ata_shost_to_port(shost);
-	enum link_pm policy = 0;
-	int i;
+	enum ata_lpm_policy policy;
 
-	/*
-	 * we are skipping array location 0 on purpose - this
-	 * is because a value of NOT_AVAILABLE is displayed
-	 * to the user as max_performance, but when the user
-	 * writes "max_performance", they actually want the
-	 * value to match MAX_PERFORMANCE.
-	 */
-	for (i = 1; i < ARRAY_SIZE(link_pm_policy); i++) {
-		const int len = strlen(link_pm_policy[i].name);
-		if (strncmp(link_pm_policy[i].name, buf, len) == 0) {
-			policy = link_pm_policy[i].value;
+	/* UNKNOWN is internal state, iterate from MAX_POWER */
+	for (policy = ATA_LPM_MAX_POWER;
+	     policy < ARRAY_SIZE(ata_lpm_policy_names); policy++) {
+		const char *name = ata_lpm_policy_names[policy];
+
+		if (strncmp(name, buf, strlen(name)) == 0)
 			break;
-		}
 	}
-	if (!policy)
+	if (policy == ARRAY_SIZE(ata_lpm_policy_names))
 		return -EINVAL;
 
 	ata_lpm_schedule(ap, policy);
 	return count;
 }
 
-static ssize_t
-ata_scsi_lpm_show(struct device *dev, struct device_attribute *attr, char *buf)
+static ssize_t ata_scsi_lpm_show(struct device *dev,
+				 struct device_attribute *attr, char *buf)
 {
 	struct Scsi_Host *shost = class_to_shost(dev);
 	struct ata_port *ap = ata_shost_to_port(shost);
-	const char *policy =
-		ata_scsi_lpm_get(ap->pm_policy);
 
-	if (!policy)
+	if (ap->lpm_policy >= ARRAY_SIZE(ata_lpm_policy_names))
 		return -EINVAL;
 
-	return snprintf(buf, 23, "%s\n", policy);
+	return snprintf(buf, PAGE_SIZE, "%s\n",
+			ata_lpm_policy_names[ap->lpm_policy]);
 }
 DEVICE_ATTR(link_power_management_policy, S_IRUGO | S_IWUSR,
-		ata_scsi_lpm_show, ata_scsi_lpm_put);
+	    ata_scsi_lpm_show, ata_scsi_lpm_store);
 EXPORT_SYMBOL_GPL(dev_attr_link_power_management_policy);
 
 static ssize_t ata_scsi_park_show(struct device *device,
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 142102b94df5..1471462e3e3c 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -100,8 +100,10 @@ extern int sata_link_init_spd(struct ata_link *link);
 extern int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg);
 extern int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg);
 extern struct ata_port *ata_port_alloc(struct ata_host *host);
-extern void ata_dev_enable_pm(struct ata_device *dev, enum link_pm policy);
-extern void ata_lpm_schedule(struct ata_port *ap, enum link_pm);
+extern void ata_dev_enable_pm(struct ata_device *dev,
+			      enum ata_lpm_policy policy);
+extern void ata_lpm_schedule(struct ata_port *ap,
+			     enum ata_lpm_policy policy);
 extern const char *sata_spd_string(unsigned int spd);
 
 /* libata-acpi.c */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index c50f66d4382c..c5bdc90fd319 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -196,7 +196,7 @@ enum {
 	ATA_FLAG_ACPI_SATA	= (1 << 17), /* need native SATA ACPI layout */
 	ATA_FLAG_AN		= (1 << 18), /* controller supports AN */
 	ATA_FLAG_PMP		= (1 << 19), /* controller supports PMP */
-	ATA_FLAG_IPM		= (1 << 20), /* driver can handle IPM */
+	ATA_FLAG_LPM		= (1 << 20), /* driver can handle LPM */
 	ATA_FLAG_EM		= (1 << 21), /* driver supports enclosure
 					      * management */
 	ATA_FLAG_SW_ACTIVITY	= (1 << 22), /* driver supports sw activity
@@ -377,7 +377,7 @@ enum {
 	ATA_HORKAGE_BROKEN_HPA	= (1 << 4),	/* Broken HPA */
 	ATA_HORKAGE_DISABLE	= (1 << 5),	/* Disable it */
 	ATA_HORKAGE_HPA_SIZE	= (1 << 6),	/* native size off by one */
-	ATA_HORKAGE_IPM		= (1 << 7),	/* Link PM problems */
+	ATA_HORKAGE_LPM		= (1 << 7),	/* Link PM problems */
 	ATA_HORKAGE_IVB		= (1 << 8),	/* cbl det validity bit bugs */
 	ATA_HORKAGE_STUCK_ERR	= (1 << 9),	/* stuck ERR on next PACKET */
 	ATA_HORKAGE_BRIDGE_OK	= (1 << 10),	/* no bridge limits */
@@ -464,6 +464,17 @@ enum ata_completion_errors {
 	AC_ERR_NCQ		= (1 << 10), /* marker for offending NCQ qc */
 };
 
+/*
+ * Link power management policy: If you alter this, you also need to
+ * alter libata-scsi.c (for the ascii descriptions)
+ */
+enum ata_lpm_policy {
+	ATA_LPM_UNKNOWN,
+	ATA_LPM_MAX_POWER,
+	ATA_LPM_MED_POWER,
+	ATA_LPM_MIN_POWER,
+};
+
 /* forward declarations */
 struct scsi_device;
 struct ata_port_operations;
@@ -478,16 +489,6 @@ typedef int (*ata_reset_fn_t)(struct ata_link *link, unsigned int *classes,
 			      unsigned long deadline);
 typedef void (*ata_postreset_fn_t)(struct ata_link *link, unsigned int *classes);
 
-/*
- * host pm policy: If you alter this, you also need to alter libata-scsi.c
- * (for the ascii descriptions)
- */
-enum link_pm {
-	NOT_AVAILABLE,
-	MIN_POWER,
-	MAX_PERFORMANCE,
-	MEDIUM_POWER,
-};
 extern struct device_attribute dev_attr_link_power_management_policy;
 extern struct device_attribute dev_attr_unload_heads;
 extern struct device_attribute dev_attr_em_message_type;
@@ -772,7 +773,7 @@ struct ata_port {
 
 	pm_message_t		pm_mesg;
 	int			*pm_result;
-	enum link_pm		pm_policy;
+	enum ata_lpm_policy	lpm_policy;
 
 	struct timer_list	fastdrain_timer;
 	unsigned long		fastdrain_cnt;
@@ -838,7 +839,7 @@ struct ata_port_operations {
 	int  (*scr_write)(struct ata_link *link, unsigned int sc_reg, u32 val);
 	void (*pmp_attach)(struct ata_port *ap);
 	void (*pmp_detach)(struct ata_port *ap);
-	int  (*enable_pm)(struct ata_port *ap, enum link_pm policy);
+	int  (*enable_pm)(struct ata_port *ap, enum ata_lpm_policy policy);
 	void (*disable_pm)(struct ata_port *ap);
 
 	/*
-- 
cgit v1.2.3


From 1152b2617a6e1943b6b82e07c962950e56f1000c Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 1 Sep 2010 17:50:05 +0200
Subject: libata: implement sata_link_scr_lpm() and make ata_dev_set_feature()
 global

Link power management is about to be reimplemented.  Prepare for it.

* Implement sata_link_scr_lpm().

* Drop static from ata_dev_set_feature() and make it available to
  other libata files.

* Trivial whitespace adjustments.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 75 +++++++++++++++++++++++++++++++++++++++++++----
 drivers/ata/libata.h      |  2 ++
 include/linux/libata.h    |  2 ++
 3 files changed, 74 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 380ceb000aad..b8024451234c 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -91,8 +91,6 @@ const struct ata_port_operations sata_port_ops = {
 static unsigned int ata_dev_init_params(struct ata_device *dev,
 					u16 heads, u16 sectors);
 static unsigned int ata_dev_set_xfermode(struct ata_device *dev);
-static unsigned int ata_dev_set_feature(struct ata_device *dev,
-					u8 enable, u8 feature);
 static void ata_dev_xfermask(struct ata_device *dev);
 static unsigned long ata_dev_blacklisted(const struct ata_device *dev);
 
@@ -3628,7 +3626,7 @@ int ata_wait_after_reset(struct ata_link *link, unsigned long deadline,
  *	@params: timing parameters { interval, duratinon, timeout } in msec
  *	@deadline: deadline jiffies for the operation
  *
-*	Make sure SStatus of @link reaches stable state, determined by
+ *	Make sure SStatus of @link reaches stable state, determined by
  *	holding the same value where DET is not 1 for @duration polled
  *	every @interval, before @timeout.  Timeout constraints the
  *	beginning of the stable state.  Because DET gets stuck at 1 on
@@ -3759,6 +3757,72 @@ int sata_link_resume(struct ata_link *link, const unsigned long *params,
 	return rc != -EINVAL ? rc : 0;
 }
 
+/**
+ *	sata_link_scr_lpm - manipulate SControl IPM and SPM fields
+ *	@link: ATA link to manipulate SControl for
+ *	@policy: LPM policy to configure
+ *	@spm_wakeup: initiate LPM transition to active state
+ *
+ *	Manipulate the IPM field of the SControl register of @link
+ *	according to @policy.  If @policy is ATA_LPM_MAX_POWER and
+ *	@spm_wakeup is %true, the SPM field is manipulated to wake up
+ *	the link.  This function also clears PHYRDY_CHG before
+ *	returning.
+ *
+ *	LOCKING:
+ *	EH context.
+ *
+ *	RETURNS:
+ *	0 on succes, -errno otherwise.
+ */
+int sata_link_scr_lpm(struct ata_link *link, enum ata_lpm_policy policy,
+		      bool spm_wakeup)
+{
+	struct ata_eh_context *ehc = &link->eh_context;
+	bool woken_up = false;
+	u32 scontrol;
+	int rc;
+
+	rc = sata_scr_read(link, SCR_CONTROL, &scontrol);
+	if (rc)
+		return rc;
+
+	switch (policy) {
+	case ATA_LPM_MAX_POWER:
+		/* disable all LPM transitions */
+		scontrol |= (0x3 << 8);
+		/* initiate transition to active state */
+		if (spm_wakeup) {
+			scontrol |= (0x4 << 12);
+			woken_up = true;
+		}
+		break;
+	case ATA_LPM_MED_POWER:
+		/* allow LPM to PARTIAL */
+		scontrol &= ~(0x1 << 8);
+		scontrol |= (0x2 << 8);
+		break;
+	case ATA_LPM_MIN_POWER:
+		/* no restrictions on LPM transitions */
+		scontrol &= ~(0x3 << 8);
+		break;
+	default:
+		WARN_ON(1);
+	}
+
+	rc = sata_scr_write(link, SCR_CONTROL, scontrol);
+	if (rc)
+		return rc;
+
+	/* give the link time to transit out of LPM state */
+	if (woken_up)
+		msleep(10);
+
+	/* clear PHYRDY_CHG from SError */
+	ehc->i.serror &= ~SERR_PHYRDY_CHG;
+	return sata_scr_write(link, SCR_ERROR, SERR_PHYRDY_CHG);
+}
+
 /**
  *	ata_std_prereset - prepare for reset
  *	@link: ATA link to be reset
@@ -4551,6 +4615,7 @@ static unsigned int ata_dev_set_xfermode(struct ata_device *dev)
 	DPRINTK("EXIT, err_mask=%x\n", err_mask);
 	return err_mask;
 }
+
 /**
  *	ata_dev_set_feature - Issue SET FEATURES - SATA FEATURES
  *	@dev: Device to which command will be sent
@@ -4566,8 +4631,7 @@ static unsigned int ata_dev_set_xfermode(struct ata_device *dev)
  *	RETURNS:
  *	0 on success, AC_ERR_* mask otherwise.
  */
-static unsigned int ata_dev_set_feature(struct ata_device *dev, u8 enable,
-					u8 feature)
+unsigned int ata_dev_set_feature(struct ata_device *dev, u8 enable, u8 feature)
 {
 	struct ata_taskfile tf;
 	unsigned int err_mask;
@@ -6732,6 +6796,7 @@ EXPORT_SYMBOL_GPL(sata_set_spd);
 EXPORT_SYMBOL_GPL(ata_wait_after_reset);
 EXPORT_SYMBOL_GPL(sata_link_debounce);
 EXPORT_SYMBOL_GPL(sata_link_resume);
+EXPORT_SYMBOL_GPL(sata_link_scr_lpm);
 EXPORT_SYMBOL_GPL(ata_std_prereset);
 EXPORT_SYMBOL_GPL(sata_link_hardreset);
 EXPORT_SYMBOL_GPL(sata_std_hardreset);
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 1471462e3e3c..65b6a73c3ac7 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -86,6 +86,8 @@ extern int ata_dev_revalidate(struct ata_device *dev, unsigned int new_class,
 extern int ata_dev_configure(struct ata_device *dev);
 extern int sata_down_spd_limit(struct ata_link *link, u32 spd_limit);
 extern int ata_down_xfermask_limit(struct ata_device *dev, unsigned int sel);
+extern unsigned int ata_dev_set_feature(struct ata_device *dev,
+					u8 enable, u8 feature);
 extern void ata_sg_clean(struct ata_queued_cmd *qc);
 extern void ata_qc_free(struct ata_queued_cmd *qc);
 extern void ata_qc_issue(struct ata_queued_cmd *qc);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index c5bdc90fd319..7770eeb21039 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -952,6 +952,8 @@ extern int sata_link_debounce(struct ata_link *link,
 			const unsigned long *params, unsigned long deadline);
 extern int sata_link_resume(struct ata_link *link, const unsigned long *params,
 			    unsigned long deadline);
+extern int sata_link_scr_lpm(struct ata_link *link, enum ata_lpm_policy policy,
+			     bool spm_wakeup);
 extern int sata_link_hardreset(struct ata_link *link,
 			const unsigned long *timing, unsigned long deadline,
 			bool *online, int (*check_ready)(struct ata_link *));
-- 
cgit v1.2.3


From 6b7ae9545ad9875a289f4191c0216b473e313cb9 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 1 Sep 2010 17:50:06 +0200
Subject: libata: reimplement link power management

The current LPM implementation has the following issues.

* Operation order isn't well thought-out.  e.g. HIPM should be
  configured after IPM in SControl is properly configured.  Not the
  other way around.

* Suspend/resume paths call ata_lpm_enable/disable() which must only
  be called from EH context directly.  Also, ata_lpm_enable/disable()
  were called whether LPM was in use or not.

* Implementation is per-port when it should be per-link.  As a result,
  it can't be used for controllers with slave links or PMP.

* LPM state isn't managed consistently.  After a link reset for
  whatever reason including suspend/resume the actual LPM state would
  be reset leaving ap->lpm_policy inconsistent.

* Generic/driver-specific logic boundary isn't clear.  Currently,
  libahci has to mangle stuff which libata EH proper should be
  handling.  This makes the implementation unnecessarily complex and
  fragile.

* Tied to ALPM.  Doesn't consider DIPM only cases and doesn't check
  whether the device allows HIPM.

* Error handling isn't implemented.

Given the extent of mismatch with the rest of libata, I don't think
trying to fix it piecewise makes much sense.  This patch reimplements
LPM support.

* The new implementation is per-link.  The target policy is still
  port-wide (ap->target_lpm_policy) but all the mechanisms and states
  are per-link and integrate well with the rest of link abstraction
  and can work with slave and PMP links.

* Core EH has proper control of LPM state.  LPM state is reconfigured
  when and only when reconfiguration is necessary.  It makes sure that
  LPM state is reset when probing for new device on the link.
  Controller agnostic logic is now implemented in libata EH proper and
  driver implementation only has to deal with controller specifics.

* Proper error handling.  LPM config failure is attributed to the
  device on the link and LPM is disabled for the link if it fails
  repeatedly.

* ops->enable/disable_pm() are replaced with single ops->set_lpm()
  which takes @policy and @hints.  This simplifies driver specific
  implementation.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/ahci.c          |   3 -
 drivers/ata/ahci.h          |   1 -
 drivers/ata/ahci_platform.c |   3 -
 drivers/ata/libahci.c       | 158 +++++++++-------------------------
 drivers/ata/libata-core.c   | 201 +-------------------------------------------
 drivers/ata/libata-eh.c     | 164 +++++++++++++++++++++++++++++++-----
 drivers/ata/libata-scsi.c   |  11 ++-
 drivers/ata/libata.h        |   4 -
 include/linux/libata.h      |  17 ++--
 9 files changed, 206 insertions(+), 356 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 68f7b4216501..328826381a2d 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1208,9 +1208,6 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		ata_port_pbar_desc(ap, AHCI_PCI_BAR,
 				   0x100 + ap->port_no * 0x80, "port");
 
-		/* set initial link pm policy */
-		ap->lpm_policy = ATA_LPM_UNKNOWN;
-
 		/* set enclosure management message type */
 		if (ap->flags & ATA_FLAG_EM)
 			ap->em_message_type = hpriv->em_msg_type;
diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h
index 93867d3f8dd3..1a2aacfdebc5 100644
--- a/drivers/ata/ahci.h
+++ b/drivers/ata/ahci.h
@@ -201,7 +201,6 @@ enum {
 	AHCI_HFLAG_MV_PATA		= (1 << 4), /* PATA port */
 	AHCI_HFLAG_NO_MSI		= (1 << 5), /* no PCI MSI */
 	AHCI_HFLAG_NO_PMP		= (1 << 6), /* no PMP */
-	AHCI_HFLAG_NO_HOTPLUG		= (1 << 7), /* ignore PxSERR.DIAG.N */
 	AHCI_HFLAG_SECT255		= (1 << 8), /* max 255 sectors */
 	AHCI_HFLAG_YES_NCQ		= (1 << 9), /* force NCQ cap on */
 	AHCI_HFLAG_NO_SUSPEND		= (1 << 10), /* don't suspend */
diff --git a/drivers/ata/ahci_platform.c b/drivers/ata/ahci_platform.c
index 07556853c0d6..6fef1fa75c54 100644
--- a/drivers/ata/ahci_platform.c
+++ b/drivers/ata/ahci_platform.c
@@ -129,9 +129,6 @@ static int __init ahci_probe(struct platform_device *pdev)
 		ata_port_desc(ap, "mmio %pR", mem);
 		ata_port_desc(ap, "port 0x%x", 0x100 + ap->port_no * 0x80);
 
-		/* set initial link pm policy */
-		ap->lpm_policy = ATA_LPM_UNKNOWN;
-
 		/* set enclosure management message type */
 		if (ap->flags & ATA_FLAG_EM)
 			ap->em_message_type = hpriv->em_msg_type;
diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index ed7803f2b4f1..437f92597788 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -56,8 +56,8 @@ MODULE_PARM_DESC(skip_host_reset, "skip global host reset (0=don't skip, 1=skip)
 module_param_named(ignore_sss, ahci_ignore_sss, int, 0444);
 MODULE_PARM_DESC(ignore_sss, "Ignore staggered spinup flag (0=don't ignore, 1=ignore)");
 
-static int ahci_enable_alpm(struct ata_port *ap, enum ata_lpm_policy policy);
-static void ahci_disable_alpm(struct ata_port *ap);
+static int ahci_set_lpm(struct ata_link *link, enum ata_lpm_policy policy,
+			unsigned hints);
 static ssize_t ahci_led_show(struct ata_port *ap, char *buf);
 static ssize_t ahci_led_store(struct ata_port *ap, const char *buf,
 			      size_t size);
@@ -163,8 +163,7 @@ struct ata_port_operations ahci_ops = {
 	.pmp_attach		= ahci_pmp_attach,
 	.pmp_detach		= ahci_pmp_detach,
 
-	.enable_pm		= ahci_enable_alpm,
-	.disable_pm		= ahci_disable_alpm,
+	.set_lpm		= ahci_set_lpm,
 	.em_show		= ahci_led_show,
 	.em_store		= ahci_led_store,
 	.sw_activity_show	= ahci_activity_show,
@@ -641,126 +640,56 @@ static void ahci_power_up(struct ata_port *ap)
 	writel(cmd | PORT_CMD_ICC_ACTIVE, port_mmio + PORT_CMD);
 }
 
-static void ahci_disable_alpm(struct ata_port *ap)
+static int ahci_set_lpm(struct ata_link *link, enum ata_lpm_policy policy,
+			unsigned int hints)
 {
+	struct ata_port *ap = link->ap;
 	struct ahci_host_priv *hpriv = ap->host->private_data;
-	void __iomem *port_mmio = ahci_port_base(ap);
-	u32 cmd;
 	struct ahci_port_priv *pp = ap->private_data;
-
-	/* LPM bits should be disabled by libata-core */
-	/* get the existing command bits */
-	cmd = readl(port_mmio + PORT_CMD);
-
-	/* disable ALPM and ASP */
-	cmd &= ~PORT_CMD_ASP;
-	cmd &= ~PORT_CMD_ALPE;
-
-	/* force the interface back to active */
-	cmd |= PORT_CMD_ICC_ACTIVE;
-
-	/* write out new cmd value */
-	writel(cmd, port_mmio + PORT_CMD);
-	cmd = readl(port_mmio + PORT_CMD);
-
-	/* wait 10ms to be sure we've come out of any low power state */
-	msleep(10);
-
-	/* clear out any PhyRdy stuff from interrupt status */
-	writel(PORT_IRQ_PHYRDY, port_mmio + PORT_IRQ_STAT);
-
-	/* go ahead and clean out PhyRdy Change from Serror too */
-	ahci_scr_write(&ap->link, SCR_ERROR, ((1 << 16) | (1 << 18)));
-
-	/*
-	 * Clear flag to indicate that we should ignore all PhyRdy
-	 * state changes
-	 */
-	hpriv->flags &= ~AHCI_HFLAG_NO_HOTPLUG;
-
-	/*
-	 * Enable interrupts on Phy Ready.
-	 */
-	pp->intr_mask |= PORT_IRQ_PHYRDY;
-	writel(pp->intr_mask, port_mmio + PORT_IRQ_MASK);
-
-	/*
-	 * don't change the link pm policy - we can be called
-	 * just to turn of link pm temporarily
-	 */
-}
-
-static int ahci_enable_alpm(struct ata_port *ap, enum ata_lpm_policy policy)
-{
-	struct ahci_host_priv *hpriv = ap->host->private_data;
 	void __iomem *port_mmio = ahci_port_base(ap);
-	u32 cmd;
-	struct ahci_port_priv *pp = ap->private_data;
-	u32 asp;
-
-	/* Make sure the host is capable of link power management */
-	if (!(hpriv->cap & HOST_CAP_ALPM))
-		return -EINVAL;
 
-	switch (policy) {
-	case ATA_LPM_MAX_POWER:
-	case ATA_LPM_UNKNOWN:
+	if (policy != ATA_LPM_MAX_POWER) {
 		/*
-		 * if we came here with ATA_LPM_UNKNOWN,
-		 * it just means this is the first time we
-		 * have tried to enable - default to max performance,
-		 * and let the user go to lower power modes on request.
+		 * Disable interrupts on Phy Ready. This keeps us from
+		 * getting woken up due to spurious phy ready
+		 * interrupts.
 		 */
-		ahci_disable_alpm(ap);
-		return 0;
-	case ATA_LPM_MIN_POWER:
-		/* configure HBA to enter SLUMBER */
-		asp = PORT_CMD_ASP;
-		break;
-	case ATA_LPM_MED_POWER:
-		/* configure HBA to enter PARTIAL */
-		asp = 0;
-		break;
-	default:
-		return -EINVAL;
+		pp->intr_mask &= ~PORT_IRQ_PHYRDY;
+		writel(pp->intr_mask, port_mmio + PORT_IRQ_MASK);
+
+		sata_link_scr_lpm(link, policy, false);
 	}
 
-	/*
-	 * Disable interrupts on Phy Ready. This keeps us from
-	 * getting woken up due to spurious phy ready interrupts
-	 * TBD - Hot plug should be done via polling now, is
-	 * that even supported?
-	 */
-	pp->intr_mask &= ~PORT_IRQ_PHYRDY;
-	writel(pp->intr_mask, port_mmio + PORT_IRQ_MASK);
+	if (hpriv->cap & HOST_CAP_ALPM) {
+		u32 cmd = readl(port_mmio + PORT_CMD);
 
-	/*
-	 * Set a flag to indicate that we should ignore all PhyRdy
-	 * state changes since these can happen now whenever we
-	 * change link state
-	 */
-	hpriv->flags |= AHCI_HFLAG_NO_HOTPLUG;
+		if (policy == ATA_LPM_MAX_POWER || !(hints & ATA_LPM_HIPM)) {
+			cmd &= ~(PORT_CMD_ASP | PORT_CMD_ALPE);
+			cmd |= PORT_CMD_ICC_ACTIVE;
 
-	/* get the existing command bits */
-	cmd = readl(port_mmio + PORT_CMD);
+			writel(cmd, port_mmio + PORT_CMD);
+			readl(port_mmio + PORT_CMD);
 
-	/*
-	 * Set ASP based on Policy
-	 */
-	cmd |= asp;
+			/* wait 10ms to be sure we've come out of LPM state */
+			msleep(10);
+		} else {
+			cmd |= PORT_CMD_ALPE;
+			if (policy == ATA_LPM_MIN_POWER)
+				cmd |= PORT_CMD_ASP;
 
-	/*
-	 * Setting this bit will instruct the HBA to aggressively
-	 * enter a lower power link state when it's appropriate and
-	 * based on the value set above for ASP
-	 */
-	cmd |= PORT_CMD_ALPE;
+			/* write out new cmd value */
+			writel(cmd, port_mmio + PORT_CMD);
+		}
+	}
 
-	/* write out new cmd value */
-	writel(cmd, port_mmio + PORT_CMD);
-	cmd = readl(port_mmio + PORT_CMD);
+	if (policy == ATA_LPM_MAX_POWER) {
+		sata_link_scr_lpm(link, policy, false);
+
+		/* turn PHYRDY IRQ back on */
+		pp->intr_mask |= PORT_IRQ_PHYRDY;
+		writel(pp->intr_mask, port_mmio + PORT_IRQ_MASK);
+	}
 
-	/* LPM bits should be set by libata-core */
 	return 0;
 }
 
@@ -1658,15 +1587,10 @@ static void ahci_port_intr(struct ata_port *ap)
 	if (unlikely(resetting))
 		status &= ~PORT_IRQ_BAD_PMP;
 
-	/* If we are getting PhyRdy, this is
-	 * just a power state change, we should
-	 * clear out this, plus the PhyRdy/Comm
-	 * Wake bits from Serror
-	 */
-	if ((hpriv->flags & AHCI_HFLAG_NO_HOTPLUG) &&
-		(status & PORT_IRQ_PHYRDY)) {
+	/* if LPM is enabled, PHYRDY doesn't mean anything */
+	if (ap->link.lpm_policy > ATA_LPM_MAX_POWER) {
 		status &= ~PORT_IRQ_PHYRDY;
-		ahci_scr_write(&ap->link, SCR_ERROR, ((1 << 16) | (1 << 18)));
+		ahci_scr_write(&ap->link, SCR_ERROR, SERR_PHYRDY_CHG);
 	}
 
 	if (unlikely(status & PORT_IRQ_ERROR)) {
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index b8024451234c..7c5538b9fa3b 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -1028,182 +1028,6 @@ const char *sata_spd_string(unsigned int spd)
 	return spd_str[spd - 1];
 }
 
-static int ata_dev_set_dipm(struct ata_device *dev, enum ata_lpm_policy policy)
-{
-	struct ata_link *link = dev->link;
-	struct ata_port *ap = link->ap;
-	u32 scontrol;
-	unsigned int err_mask;
-	int rc;
-
-	/*
-	 * disallow DIPM for drivers which haven't set
-	 * ATA_FLAG_LPM.  This is because when DIPM is enabled,
-	 * phy ready will be set in the interrupt status on
-	 * state changes, which will cause some drivers to
-	 * think there are errors - additionally drivers will
-	 * need to disable hot plug.
-	 */
-	if (!(ap->flags & ATA_FLAG_LPM) || !ata_dev_enabled(dev)) {
-		ap->lpm_policy = ATA_LPM_UNKNOWN;
-		return -EINVAL;
-	}
-
-	/*
-	 * For DIPM, we will only enable it for the
-	 * min_power setting.
-	 *
-	 * Why?  Because Disks are too stupid to know that
-	 * If the host rejects a request to go to SLUMBER
-	 * they should retry at PARTIAL, and instead it
-	 * just would give up.  So, for medium_power to
-	 * work at all, we need to only allow HIPM.
-	 */
-	rc = sata_scr_read(link, SCR_CONTROL, &scontrol);
-	if (rc)
-		return rc;
-
-	switch (policy) {
-	case ATA_LPM_MIN_POWER:
-		/* no restrictions on LPM transitions */
-		scontrol &= ~(0x3 << 8);
-		rc = sata_scr_write(link, SCR_CONTROL, scontrol);
-		if (rc)
-			return rc;
-
-		/* enable DIPM */
-		if (dev->flags & ATA_DFLAG_DIPM)
-			err_mask = ata_dev_set_feature(dev,
-					SETFEATURES_SATA_ENABLE, SATA_DIPM);
-		break;
-	case ATA_LPM_MED_POWER:
-		/* allow LPM to PARTIAL */
-		scontrol &= ~(0x1 << 8);
-		scontrol |= (0x2 << 8);
-		rc = sata_scr_write(link, SCR_CONTROL, scontrol);
-		if (rc)
-			return rc;
-
-		/*
-		 * we don't have to disable DIPM since LPM flags
-		 * disallow transitions to SLUMBER, which effectively
-		 * disable DIPM if it does not support PARTIAL
-		 */
-		break;
-	case ATA_LPM_UNKNOWN:
-	case ATA_LPM_MAX_POWER:
-		/* disable all LPM transitions */
-		scontrol |= (0x3 << 8);
-		rc = sata_scr_write(link, SCR_CONTROL, scontrol);
-		if (rc)
-			return rc;
-
-		/*
-		 * we don't have to disable DIPM since LPM flags
-		 * disallow all transitions which effectively
-		 * disable DIPM anyway.
-		 */
-		break;
-	}
-
-	/* FIXME: handle SET FEATURES failure */
-	(void) err_mask;
-
-	return 0;
-}
-
-/**
- *	ata_dev_enable_pm - enable SATA interface power management
- *	@dev:  device to enable power management
- *	@policy: the link power management policy
- *
- *	Enable SATA Interface power management.  This will enable
- *	Device Interface Power Management (DIPM) for min_power
- * 	policy, and then call driver specific callbacks for
- *	enabling Host Initiated Power management.
- *
- *	Locking: Caller.
- *	Returns: -EINVAL if LPM is not supported, 0 otherwise.
- */
-void ata_dev_enable_pm(struct ata_device *dev, enum ata_lpm_policy policy)
-{
-	int rc = 0;
-	struct ata_port *ap = dev->link->ap;
-
-	/* set HIPM first, then DIPM */
-	if (ap->ops->enable_pm)
-		rc = ap->ops->enable_pm(ap, policy);
-	if (rc)
-		goto enable_pm_out;
-	rc = ata_dev_set_dipm(dev, policy);
-
-enable_pm_out:
-	if (rc)
-		ap->lpm_policy = ATA_LPM_MAX_POWER;
-	else
-		ap->lpm_policy = policy;
-	return /* rc */;	/* hopefully we can use 'rc' eventually */
-}
-
-#ifdef CONFIG_PM
-/**
- *	ata_dev_disable_pm - disable SATA interface power management
- *	@dev: device to disable power management
- *
- *	Disable SATA Interface power management.  This will disable
- *	Device Interface Power Management (DIPM) without changing
- * 	policy,  call driver specific callbacks for disabling Host
- * 	Initiated Power management.
- *
- *	Locking: Caller.
- *	Returns: void
- */
-static void ata_dev_disable_pm(struct ata_device *dev)
-{
-	struct ata_port *ap = dev->link->ap;
-
-	ata_dev_set_dipm(dev, ATA_LPM_MAX_POWER);
-	if (ap->ops->disable_pm)
-		ap->ops->disable_pm(ap);
-}
-#endif	/* CONFIG_PM */
-
-void ata_lpm_schedule(struct ata_port *ap, enum ata_lpm_policy policy)
-{
-	ap->lpm_policy = policy;
-	ap->link.eh_info.action |= ATA_EH_LPM;
-	ap->link.eh_info.flags |= ATA_EHI_NO_AUTOPSY;
-	ata_port_schedule_eh(ap);
-}
-
-#ifdef CONFIG_PM
-static void ata_lpm_enable(struct ata_host *host)
-{
-	struct ata_link *link;
-	struct ata_port *ap;
-	struct ata_device *dev;
-	int i;
-
-	for (i = 0; i < host->n_ports; i++) {
-		ap = host->ports[i];
-		ata_for_each_link(link, ap, EDGE) {
-			ata_for_each_dev(dev, link, ALL)
-				ata_dev_disable_pm(dev);
-		}
-	}
-}
-
-static void ata_lpm_disable(struct ata_host *host)
-{
-	int i;
-
-	for (i = 0; i < host->n_ports; i++) {
-		struct ata_port *ap = host->ports[i];
-		ata_lpm_schedule(ap, ap->lpm_policy);
-	}
-}
-#endif	/* CONFIG_PM */
-
 /**
  *	ata_dev_classify - determine device type based on ATA-spec signature
  *	@tf: ATA taskfile register set for device to be identified
@@ -2562,13 +2386,6 @@ int ata_dev_configure(struct ata_device *dev)
 	if (dev->flags & ATA_DFLAG_LBA48)
 		dev->max_sectors = ATA_MAX_SECTORS_LBA48;
 
-	if (!(dev->horkage & ATA_HORKAGE_LPM)) {
-		if (ata_id_has_hipm(dev->id))
-			dev->flags |= ATA_DFLAG_HIPM;
-		if (ata_id_has_dipm(dev->id))
-			dev->flags |= ATA_DFLAG_DIPM;
-	}
-
 	/* Limit PATA drive on SATA cable bridge transfers to udma5,
 	   200 sectors */
 	if (ata_dev_knobble(dev)) {
@@ -2589,13 +2406,6 @@ int ata_dev_configure(struct ata_device *dev)
 		dev->max_sectors = min_t(unsigned int, ATA_MAX_SECTORS_128,
 					 dev->max_sectors);
 
-	if (ata_dev_blacklisted(dev) & ATA_HORKAGE_LPM) {
-		dev->horkage |= ATA_HORKAGE_LPM;
-
-		/* reset link pm_policy for this port to no pm */
-		ap->lpm_policy = ATA_LPM_MAX_POWER;
-	}
-
 	if (ap->ops->dev_config)
 		ap->ops->dev_config(dev);
 
@@ -5494,12 +5304,6 @@ int ata_host_suspend(struct ata_host *host, pm_message_t mesg)
 	unsigned int ehi_flags = ATA_EHI_QUIET;
 	int rc;
 
-	/*
-	 * disable link pm on all ports before requesting
-	 * any pm activity
-	 */
-	ata_lpm_enable(host);
-
 	/*
 	 * On some hardware, device fails to respond after spun down
 	 * for suspend.  As the device won't be used before being
@@ -5533,9 +5337,6 @@ void ata_host_resume(struct ata_host *host)
 	ata_host_request_pm(host, PMSG_ON, ATA_EH_RESET,
 			    ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, 0);
 	host->dev->power.power_state = PMSG_ON;
-
-	/* reenable link pm */
-	ata_lpm_disable(host);
 }
 #endif
 
@@ -6096,7 +5897,7 @@ static void async_port_probe(void *data, async_cookie_t cookie)
 		spin_lock_irqsave(ap->lock, flags);
 
 		ehi->probe_mask |= ATA_ALL_DEVICES;
-		ehi->action |= ATA_EH_RESET | ATA_EH_LPM;
+		ehi->action |= ATA_EH_RESET;
 		ehi->flags |= ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET;
 
 		ap->pflags &= ~ATA_PFLAG_INITIALIZING;
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 96aa75ddf87f..a645cd3ab163 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1580,9 +1580,9 @@ static void ata_eh_analyze_serror(struct ata_link *link)
 	 * host links.  For disabled PMP links, only N bit is
 	 * considered as X bit is left at 1 for link plugging.
 	 */
-	hotplug_mask = 0;
-
-	if (!(link->flags & ATA_LFLAG_DISABLED) || ata_is_host_link(link))
+	if (link->lpm_policy != ATA_LPM_MAX_POWER)
+		hotplug_mask = 0;	/* hotplug doesn't work w/ LPM */
+	else if (!(link->flags & ATA_LFLAG_DISABLED) || ata_is_host_link(link))
 		hotplug_mask = SERR_PHYRDY_CHG | SERR_DEV_XCHG;
 	else
 		hotplug_mask = SERR_PHYRDY_CHG;
@@ -2784,8 +2784,9 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	ata_eh_done(link, NULL, ATA_EH_RESET);
 	if (slave)
 		ata_eh_done(slave, NULL, ATA_EH_RESET);
-	ehc->last_reset = jiffies;	/* update to completion time */
+	ehc->last_reset = jiffies;		/* update to completion time */
 	ehc->i.action |= ATA_EH_REVALIDATE;
+	link->lpm_policy = ATA_LPM_UNKNOWN;	/* reset LPM state */
 
 	rc = 0;
  out:
@@ -3211,6 +3212,121 @@ static int ata_eh_maybe_retry_flush(struct ata_device *dev)
 	return rc;
 }
 
+/**
+ *	ata_eh_set_lpm - configure SATA interface power management
+ *	@link: link to configure power management
+ *	@policy: the link power management policy
+ *	@r_failed_dev: out parameter for failed device
+ *
+ *	Enable SATA Interface power management.  This will enable
+ *	Device Interface Power Management (DIPM) for min_power
+ * 	policy, and then call driver specific callbacks for
+ *	enabling Host Initiated Power management.
+ *
+ *	LOCKING:
+ *	EH context.
+ *
+ *	RETURNS:
+ *	0 on success, -errno on failure.
+ */
+static int ata_eh_set_lpm(struct ata_link *link, enum ata_lpm_policy policy,
+			  struct ata_device **r_failed_dev)
+{
+	struct ata_port *ap = link->ap;
+	struct ata_eh_context *ehc = &link->eh_context;
+	struct ata_device *dev, *link_dev = NULL, *lpm_dev = NULL;
+	unsigned int hints = ATA_LPM_EMPTY | ATA_LPM_HIPM;
+	unsigned int err_mask;
+	int rc;
+
+	/* if the link or host doesn't do LPM, noop */
+	if ((link->flags & ATA_LFLAG_NO_LPM) || (ap && !ap->ops->set_lpm))
+		return 0;
+
+	/*
+	 * DIPM is enabled only for MIN_POWER as some devices
+	 * misbehave when the host NACKs transition to SLUMBER.  Order
+	 * device and link configurations such that the host always
+	 * allows DIPM requests.
+	 */
+	ata_for_each_dev(dev, link, ENABLED) {
+		bool hipm = ata_id_has_hipm(dev->id);
+		bool dipm = ata_id_has_dipm(dev->id);
+
+		/* find the first enabled and LPM enabled devices */
+		if (!link_dev)
+			link_dev = dev;
+
+		if (!lpm_dev && (hipm || dipm))
+			lpm_dev = dev;
+
+		hints &= ~ATA_LPM_EMPTY;
+		if (!hipm)
+			hints &= ~ATA_LPM_HIPM;
+
+		/* disable DIPM before changing link config */
+		if (policy != ATA_LPM_MIN_POWER && dipm) {
+			err_mask = ata_dev_set_feature(dev,
+					SETFEATURES_SATA_DISABLE, SATA_DIPM);
+			if (err_mask && err_mask != AC_ERR_DEV) {
+				ata_dev_printk(dev, KERN_WARNING,
+					"failed to disable DIPM, Emask 0x%x\n",
+					err_mask);
+				rc = -EIO;
+				goto fail;
+			}
+		}
+	}
+
+	rc = ap->ops->set_lpm(link, policy, hints);
+	if (!rc && ap->slave_link)
+		rc = ap->ops->set_lpm(ap->slave_link, policy, hints);
+
+	/*
+	 * Attribute link config failure to the first (LPM) enabled
+	 * device on the link.
+	 */
+	if (rc) {
+		if (rc == -EOPNOTSUPP) {
+			link->flags |= ATA_LFLAG_NO_LPM;
+			return 0;
+		}
+		dev = lpm_dev ? lpm_dev : link_dev;
+		goto fail;
+	}
+
+	/* host config updated, enable DIPM if transitioning to MIN_POWER */
+	ata_for_each_dev(dev, link, ENABLED) {
+		if (policy == ATA_LPM_MIN_POWER && ata_id_has_dipm(dev->id)) {
+			err_mask = ata_dev_set_feature(dev,
+					SETFEATURES_SATA_ENABLE, SATA_DIPM);
+			if (err_mask && err_mask != AC_ERR_DEV) {
+				ata_dev_printk(dev, KERN_WARNING,
+					"failed to enable DIPM, Emask 0x%x\n",
+					err_mask);
+				rc = -EIO;
+				goto fail;
+			}
+		}
+	}
+
+	link->lpm_policy = policy;
+	if (ap && ap->slave_link)
+		ap->slave_link->lpm_policy = policy;
+	return 0;
+
+fail:
+	/* if no device or only one more chance is left, disable LPM */
+	if (!dev || ehc->tries[dev->devno] <= 2) {
+		ata_link_printk(link, KERN_WARNING,
+				"disabling LPM on the link\n");
+		link->flags |= ATA_LFLAG_NO_LPM;
+	}
+	if (r_failed_dev)
+		*r_failed_dev = dev;
+	return rc;
+}
+
 static int ata_link_nr_enabled(struct ata_link *link)
 {
 	struct ata_device *dev;
@@ -3295,6 +3411,10 @@ static int ata_eh_schedule_probe(struct ata_device *dev)
 	ehc->saved_xfer_mode[dev->devno] = 0;
 	ehc->saved_ncq_enabled &= ~(1 << dev->devno);
 
+	/* the link maybe in a deep sleep, wake it up */
+	if (link->lpm_policy > ATA_LPM_MAX_POWER)
+		link->ap->ops->set_lpm(link, ATA_LPM_MAX_POWER, ATA_LPM_EMPTY);
+
 	/* Record and count probe trials on the ering.  The specific
 	 * error mask used is irrelevant.  Because a successful device
 	 * detection clears the ering, this count accumulates only if
@@ -3396,8 +3516,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 {
 	struct ata_link *link;
 	struct ata_device *dev;
-	int nr_failed_devs;
-	int rc;
+	int rc, nr_fails;
 	unsigned long flags, deadline;
 
 	DPRINTK("ENTER\n");
@@ -3438,7 +3557,6 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 
  retry:
 	rc = 0;
-	nr_failed_devs = 0;
 
 	/* if UNLOADING, finish immediately */
 	if (ap->pflags & ATA_PFLAG_UNLOADING)
@@ -3523,13 +3641,17 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 	}
 
 	/* the rest */
-	ata_for_each_link(link, ap, EDGE) {
+	nr_fails = 0;
+	ata_for_each_link(link, ap, PMP_FIRST) {
 		struct ata_eh_context *ehc = &link->eh_context;
 
+		if (sata_pmp_attached(ap) && ata_is_host_link(link))
+			goto config_lpm;
+
 		/* revalidate existing devices and attach new ones */
 		rc = ata_eh_revalidate_and_attach(link, &dev);
 		if (rc)
-			goto dev_fail;
+			goto rest_fail;
 
 		/* if PMP got attached, return, pmp EH will take care of it */
 		if (link->device->class == ATA_DEV_PMP) {
@@ -3541,7 +3663,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 		if (ehc->i.flags & ATA_EHI_SETMODE) {
 			rc = ata_set_mode(link, &dev);
 			if (rc)
-				goto dev_fail;
+				goto rest_fail;
 			ehc->i.flags &= ~ATA_EHI_SETMODE;
 		}
 
@@ -3554,7 +3676,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 					continue;
 				rc = atapi_eh_clear_ua(dev);
 				if (rc)
-					goto dev_fail;
+					goto rest_fail;
 			}
 		}
 
@@ -3564,21 +3686,25 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 				continue;
 			rc = ata_eh_maybe_retry_flush(dev);
 			if (rc)
-				goto dev_fail;
+				goto rest_fail;
 		}
 
+	config_lpm:
 		/* configure link power saving */
-		if (ehc->i.action & ATA_EH_LPM)
-			ata_for_each_dev(dev, link, ALL)
-				ata_dev_enable_pm(dev, ap->lpm_policy);
+		if (link->lpm_policy != ap->target_lpm_policy) {
+			rc = ata_eh_set_lpm(link, ap->target_lpm_policy, &dev);
+			if (rc)
+				goto rest_fail;
+		}
 
 		/* this link is okay now */
 		ehc->i.flags = 0;
 		continue;
 
-dev_fail:
-		nr_failed_devs++;
-		ata_eh_handle_dev_fail(dev, rc);
+	rest_fail:
+		nr_fails++;
+		if (dev)
+			ata_eh_handle_dev_fail(dev, rc);
 
 		if (ap->pflags & ATA_PFLAG_FROZEN) {
 			/* PMP reset requires working host port.
@@ -3590,7 +3716,7 @@ dev_fail:
 		}
 	}
 
-	if (nr_failed_devs)
+	if (nr_fails)
 		goto retry;
 
  out:
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index aa56681f68db..56f6224fd6e6 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -117,6 +117,7 @@ static ssize_t ata_scsi_lpm_store(struct device *dev,
 	struct Scsi_Host *shost = class_to_shost(dev);
 	struct ata_port *ap = ata_shost_to_port(shost);
 	enum ata_lpm_policy policy;
+	unsigned long flags;
 
 	/* UNKNOWN is internal state, iterate from MAX_POWER */
 	for (policy = ATA_LPM_MAX_POWER;
@@ -129,7 +130,11 @@ static ssize_t ata_scsi_lpm_store(struct device *dev,
 	if (policy == ARRAY_SIZE(ata_lpm_policy_names))
 		return -EINVAL;
 
-	ata_lpm_schedule(ap, policy);
+	spin_lock_irqsave(ap->lock, flags);
+	ap->target_lpm_policy = policy;
+	ata_port_schedule_eh(ap);
+	spin_unlock_irqrestore(ap->lock, flags);
+
 	return count;
 }
 
@@ -139,11 +144,11 @@ static ssize_t ata_scsi_lpm_show(struct device *dev,
 	struct Scsi_Host *shost = class_to_shost(dev);
 	struct ata_port *ap = ata_shost_to_port(shost);
 
-	if (ap->lpm_policy >= ARRAY_SIZE(ata_lpm_policy_names))
+	if (ap->target_lpm_policy >= ARRAY_SIZE(ata_lpm_policy_names))
 		return -EINVAL;
 
 	return snprintf(buf, PAGE_SIZE, "%s\n",
-			ata_lpm_policy_names[ap->lpm_policy]);
+			ata_lpm_policy_names[ap->target_lpm_policy]);
 }
 DEVICE_ATTR(link_power_management_policy, S_IRUGO | S_IWUSR,
 	    ata_scsi_lpm_show, ata_scsi_lpm_store);
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 65b6a73c3ac7..55a6f413a550 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -102,10 +102,6 @@ extern int sata_link_init_spd(struct ata_link *link);
 extern int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg);
 extern int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg);
 extern struct ata_port *ata_port_alloc(struct ata_host *host);
-extern void ata_dev_enable_pm(struct ata_device *dev,
-			      enum ata_lpm_policy policy);
-extern void ata_lpm_schedule(struct ata_port *ap,
-			     enum ata_lpm_policy policy);
 extern const char *sata_spd_string(unsigned int spd);
 
 /* libata-acpi.c */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 7770eeb21039..bc4ee218b185 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -172,6 +172,7 @@ enum {
 	ATA_LFLAG_NO_RETRY	= (1 << 5), /* don't retry this link */
 	ATA_LFLAG_DISABLED	= (1 << 6), /* link is disabled */
 	ATA_LFLAG_SW_ACTIVITY	= (1 << 7), /* keep activity stats */
+	ATA_LFLAG_NO_LPM	= (1 << 8), /* disable LPM on this link */
 
 	/* struct ata_port flags */
 	ATA_FLAG_SLAVE_POSS	= (1 << 0), /* host supports slave dev */
@@ -324,12 +325,11 @@ enum {
 	ATA_EH_HARDRESET	= (1 << 2), /* meaningful only in ->prereset */
 	ATA_EH_RESET		= ATA_EH_SOFTRESET | ATA_EH_HARDRESET,
 	ATA_EH_ENABLE_LINK	= (1 << 3),
-	ATA_EH_LPM		= (1 << 4),  /* link power management action */
 	ATA_EH_PARK		= (1 << 5), /* unload heads and stop I/O */
 
 	ATA_EH_PERDEV_MASK	= ATA_EH_REVALIDATE | ATA_EH_PARK,
 	ATA_EH_ALL_ACTIONS	= ATA_EH_REVALIDATE | ATA_EH_RESET |
-				  ATA_EH_ENABLE_LINK | ATA_EH_LPM,
+				  ATA_EH_ENABLE_LINK,
 
 	/* ata_eh_info->flags */
 	ATA_EHI_HOTPLUGGED	= (1 << 0),  /* could have been hotplugged */
@@ -377,7 +377,6 @@ enum {
 	ATA_HORKAGE_BROKEN_HPA	= (1 << 4),	/* Broken HPA */
 	ATA_HORKAGE_DISABLE	= (1 << 5),	/* Disable it */
 	ATA_HORKAGE_HPA_SIZE	= (1 << 6),	/* native size off by one */
-	ATA_HORKAGE_LPM		= (1 << 7),	/* Link PM problems */
 	ATA_HORKAGE_IVB		= (1 << 8),	/* cbl det validity bit bugs */
 	ATA_HORKAGE_STUCK_ERR	= (1 << 9),	/* stuck ERR on next PACKET */
 	ATA_HORKAGE_BRIDGE_OK	= (1 << 10),	/* no bridge limits */
@@ -475,6 +474,11 @@ enum ata_lpm_policy {
 	ATA_LPM_MIN_POWER,
 };
 
+enum ata_lpm_hints {
+	ATA_LPM_EMPTY		= (1 << 0), /* port empty/probing */
+	ATA_LPM_HIPM		= (1 << 1), /* may use HIPM */
+};
+
 /* forward declarations */
 struct scsi_device;
 struct ata_port_operations;
@@ -702,6 +706,7 @@ struct ata_link {
 	unsigned int		hw_sata_spd_limit;
 	unsigned int		sata_spd_limit;
 	unsigned int		sata_spd;	/* current SATA PHY speed */
+	enum ata_lpm_policy	lpm_policy;
 
 	/* record runtime error info, protected by host_set lock */
 	struct ata_eh_info	eh_info;
@@ -773,7 +778,7 @@ struct ata_port {
 
 	pm_message_t		pm_mesg;
 	int			*pm_result;
-	enum ata_lpm_policy	lpm_policy;
+	enum ata_lpm_policy	target_lpm_policy;
 
 	struct timer_list	fastdrain_timer;
 	unsigned long		fastdrain_cnt;
@@ -839,8 +844,8 @@ struct ata_port_operations {
 	int  (*scr_write)(struct ata_link *link, unsigned int sc_reg, u32 val);
 	void (*pmp_attach)(struct ata_port *ap);
 	void (*pmp_detach)(struct ata_port *ap);
-	int  (*enable_pm)(struct ata_port *ap, enum ata_lpm_policy policy);
-	void (*disable_pm)(struct ata_port *ap);
+	int  (*set_lpm)(struct ata_link *link, enum ata_lpm_policy policy,
+			unsigned hints);
 
 	/*
 	 * Start, stop, suspend and resume
-- 
cgit v1.2.3


From 97750cebb3000a9cc08f8ce8dc8c7143be7d7201 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Mon, 6 Sep 2010 17:56:29 +0200
Subject: libata: add @ap to ata_wait_register() and introduce ata_msleep()

Add optional @ap argument to ata_wait_register() and replace msleep()
calls with ata_msleep() which take optional @ap in addition to the
duration.  These will be used to implement EH exclusion.

This patch doesn't cause any behavior difference.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libahci.c         | 18 +++++++++---------
 drivers/ata/libata-core.c     | 21 ++++++++++++++-------
 drivers/ata/libata-eh.c       |  2 +-
 drivers/ata/libata-sff.c      | 12 ++++++------
 drivers/ata/pata_bf54x.c      |  4 ++--
 drivers/ata/pata_samsung_cf.c |  2 +-
 drivers/ata/pata_scc.c        |  4 ++--
 drivers/ata/sata_fsl.c        | 19 ++++++++++---------
 drivers/ata/sata_inic162x.c   |  2 +-
 drivers/ata/sata_sil24.c      | 14 +++++++-------
 drivers/ata/sata_via.c        |  2 +-
 include/linux/libata.h        |  5 +++--
 12 files changed, 57 insertions(+), 48 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index 437f92597788..524dbe8be163 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -567,7 +567,7 @@ int ahci_stop_engine(struct ata_port *ap)
 	writel(tmp, port_mmio + PORT_CMD);
 
 	/* wait for engine to stop. This could be as long as 500 msec */
-	tmp = ata_wait_register(port_mmio + PORT_CMD,
+	tmp = ata_wait_register(ap, port_mmio + PORT_CMD,
 				PORT_CMD_LIST_ON, PORT_CMD_LIST_ON, 1, 500);
 	if (tmp & PORT_CMD_LIST_ON)
 		return -EIO;
@@ -614,7 +614,7 @@ static int ahci_stop_fis_rx(struct ata_port *ap)
 	writel(tmp, port_mmio + PORT_CMD);
 
 	/* wait for completion, spec says 500ms, give it 1000 */
-	tmp = ata_wait_register(port_mmio + PORT_CMD, PORT_CMD_FIS_ON,
+	tmp = ata_wait_register(ap, port_mmio + PORT_CMD, PORT_CMD_FIS_ON,
 				PORT_CMD_FIS_ON, 10, 1000);
 	if (tmp & PORT_CMD_FIS_ON)
 		return -EBUSY;
@@ -671,7 +671,7 @@ static int ahci_set_lpm(struct ata_link *link, enum ata_lpm_policy policy,
 			readl(port_mmio + PORT_CMD);
 
 			/* wait 10ms to be sure we've come out of LPM state */
-			msleep(10);
+			ata_msleep(ap, 10);
 		} else {
 			cmd |= PORT_CMD_ALPE;
 			if (policy == ATA_LPM_MIN_POWER)
@@ -740,7 +740,7 @@ static void ahci_start_port(struct ata_port *ap)
 							       emp->led_state,
 							       4);
 				if (rc == -EBUSY)
-					msleep(1);
+					ata_msleep(ap, 1);
 				else
 					break;
 			}
@@ -799,7 +799,7 @@ int ahci_reset_controller(struct ata_host *host)
 		 * reset must complete within 1 second, or
 		 * the hardware should be considered fried.
 		 */
-		tmp = ata_wait_register(mmio + HOST_CTL, HOST_RESET,
+		tmp = ata_wait_register(NULL, mmio + HOST_CTL, HOST_RESET,
 					HOST_RESET, 10, 1000);
 
 		if (tmp & HOST_RESET) {
@@ -1179,7 +1179,7 @@ int ahci_kick_engine(struct ata_port *ap)
 	writel(tmp, port_mmio + PORT_CMD);
 
 	rc = 0;
-	tmp = ata_wait_register(port_mmio + PORT_CMD,
+	tmp = ata_wait_register(ap, port_mmio + PORT_CMD,
 				PORT_CMD_CLO, PORT_CMD_CLO, 1, 500);
 	if (tmp & PORT_CMD_CLO)
 		rc = -EIO;
@@ -1209,8 +1209,8 @@ static int ahci_exec_polled_cmd(struct ata_port *ap, int pmp,
 	writel(1, port_mmio + PORT_CMD_ISSUE);
 
 	if (timeout_msec) {
-		tmp = ata_wait_register(port_mmio + PORT_CMD_ISSUE, 0x1, 0x1,
-					1, timeout_msec);
+		tmp = ata_wait_register(ap, port_mmio + PORT_CMD_ISSUE,
+					0x1, 0x1, 1, timeout_msec);
 		if (tmp & 0x1) {
 			ahci_kick_engine(ap);
 			return -EBUSY;
@@ -1257,7 +1257,7 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class,
 	}
 
 	/* spec says at least 5us, but be generous and sleep for 1ms */
-	msleep(1);
+	ata_msleep(ap, 1);
 
 	/* issue the second D2H Register FIS */
 	tf.ctl &= ~ATA_SRST;
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 7c5538b9fa3b..42d9ce29f50d 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -3404,7 +3404,7 @@ int ata_wait_ready(struct ata_link *link, unsigned long deadline,
 			warned = 1;
 		}
 
-		msleep(50);
+		ata_msleep(link->ap, 50);
 	}
 }
 
@@ -3425,7 +3425,7 @@ int ata_wait_ready(struct ata_link *link, unsigned long deadline,
 int ata_wait_after_reset(struct ata_link *link, unsigned long deadline,
 				int (*check_ready)(struct ata_link *link))
 {
-	msleep(ATA_WAIT_AFTER_RESET);
+	ata_msleep(link->ap, ATA_WAIT_AFTER_RESET);
 
 	return ata_wait_ready(link, deadline, check_ready);
 }
@@ -3473,7 +3473,7 @@ int sata_link_debounce(struct ata_link *link, const unsigned long *params,
 	last_jiffies = jiffies;
 
 	while (1) {
-		msleep(interval);
+		ata_msleep(link->ap, interval);
 		if ((rc = sata_scr_read(link, SCR_STATUS, &cur)))
 			return rc;
 		cur &= 0xf;
@@ -3538,7 +3538,7 @@ int sata_link_resume(struct ata_link *link, const unsigned long *params,
 		 * immediately after resuming.  Delay 200ms before
 		 * debouncing.
 		 */
-		msleep(200);
+		ata_msleep(link->ap, 200);
 
 		/* is SControl restored correctly? */
 		if ((rc = sata_scr_read(link, SCR_CONTROL, &scontrol)))
@@ -3742,7 +3742,7 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 	/* Couldn't find anything in SATA I/II specs, but AHCI-1.1
 	 * 10.4.2 says at least 1 ms.
 	 */
-	msleep(1);
+	ata_msleep(link->ap, 1);
 
 	/* bring link back */
 	rc = sata_link_resume(link, timing, deadline);
@@ -6483,8 +6483,14 @@ int ata_ratelimit(void)
 	return __ratelimit(&ratelimit);
 }
 
+void ata_msleep(struct ata_port *ap, unsigned int msecs)
+{
+	msleep(msecs);
+}
+
 /**
  *	ata_wait_register - wait until register value changes
+ *	@ap: ATA port to wait register for, can be NULL
  *	@reg: IO-mapped register
  *	@mask: Mask to apply to read register value
  *	@val: Wait condition
@@ -6506,7 +6512,7 @@ int ata_ratelimit(void)
  *	RETURNS:
  *	The final register value.
  */
-u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
+u32 ata_wait_register(struct ata_port *ap, void __iomem *reg, u32 mask, u32 val,
 		      unsigned long interval, unsigned long timeout)
 {
 	unsigned long deadline;
@@ -6521,7 +6527,7 @@ u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
 	deadline = ata_deadline(jiffies, timeout);
 
 	while ((tmp & mask) == val && time_before(jiffies, deadline)) {
-		msleep(interval);
+		ata_msleep(ap, interval);
 		tmp = ioread32(reg);
 	}
 
@@ -6605,6 +6611,7 @@ EXPORT_SYMBOL_GPL(ata_std_postreset);
 EXPORT_SYMBOL_GPL(ata_dev_classify);
 EXPORT_SYMBOL_GPL(ata_dev_pair);
 EXPORT_SYMBOL_GPL(ata_ratelimit);
+EXPORT_SYMBOL_GPL(ata_msleep);
 EXPORT_SYMBOL_GPL(ata_wait_register);
 EXPORT_SYMBOL_GPL(ata_scsi_queuecmd);
 EXPORT_SYMBOL_GPL(ata_scsi_slave_config);
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 06a4db1ec10e..6780f4d16e81 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -779,7 +779,7 @@ void ata_port_wait_eh(struct ata_port *ap)
 
 	/* make sure SCSI EH is complete */
 	if (scsi_host_in_recovery(ap->scsi_host)) {
-		msleep(10);
+		ata_msleep(ap, 10);
 		goto retry;
 	}
 }
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 193f242eec05..14d18bf81255 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -222,7 +222,7 @@ int ata_sff_busy_sleep(struct ata_port *ap,
 	timeout = ata_deadline(timer_start, tmout_pat);
 	while (status != 0xff && (status & ATA_BUSY) &&
 	       time_before(jiffies, timeout)) {
-		msleep(50);
+		ata_msleep(ap, 50);
 		status = ata_sff_busy_wait(ap, ATA_BUSY, 3);
 	}
 
@@ -234,7 +234,7 @@ int ata_sff_busy_sleep(struct ata_port *ap,
 	timeout = ata_deadline(timer_start, tmout);
 	while (status != 0xff && (status & ATA_BUSY) &&
 	       time_before(jiffies, timeout)) {
-		msleep(50);
+		ata_msleep(ap, 50);
 		status = ap->ops->sff_check_status(ap);
 	}
 
@@ -360,7 +360,7 @@ static void ata_dev_select(struct ata_port *ap, unsigned int device,
 
 	if (wait) {
 		if (can_sleep && ap->link.device[device].class == ATA_DEV_ATAPI)
-			msleep(150);
+			ata_msleep(ap, 150);
 		ata_wait_idle(ap);
 	}
 }
@@ -1356,7 +1356,7 @@ fsm_start:
 	 */
 	status = ata_sff_busy_wait(ap, ATA_BUSY, 5);
 	if (status & ATA_BUSY) {
-		msleep(2);
+		ata_msleep(ap, 2);
 		status = ata_sff_busy_wait(ap, ATA_BUSY, 10);
 		if (status & ATA_BUSY) {
 			ata_sff_queue_pio_task(link, ATA_SHORT_PAUSE);
@@ -1937,7 +1937,7 @@ int ata_sff_wait_after_reset(struct ata_link *link, unsigned int devmask,
 	unsigned int dev1 = devmask & (1 << 1);
 	int rc, ret = 0;
 
-	msleep(ATA_WAIT_AFTER_RESET);
+	ata_msleep(ap, ATA_WAIT_AFTER_RESET);
 
 	/* always check readiness of the master device */
 	rc = ata_sff_wait_ready(link, deadline);
@@ -1966,7 +1966,7 @@ int ata_sff_wait_after_reset(struct ata_link *link, unsigned int devmask,
 			lbal = ioread8(ioaddr->lbal_addr);
 			if ((nsect == 1) && (lbal == 1))
 				break;
-			msleep(50);	/* give drive a breather */
+			ata_msleep(ap, 50);	/* give drive a breather */
 		}
 
 		rc = ata_sff_wait_ready(link, deadline);
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index 9cae65de750e..e1423cd2531f 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1046,7 +1046,7 @@ static void bfin_bus_post_reset(struct ata_port *ap, unsigned int devmask)
 			dev1 = 0;
 			break;
 		}
-		msleep(50);	/* give drive a breather */
+		ata_msleep(ap, 50);	/* give drive a breather */
 	}
 	if (dev1)
 		ata_sff_busy_sleep(ap, ATA_TMOUT_BOOT_QUICK, ATA_TMOUT_BOOT);
@@ -1087,7 +1087,7 @@ static unsigned int bfin_bus_softreset(struct ata_port *ap,
 	 *
 	 * Old drivers/ide uses the 2mS rule and then waits for ready
 	 */
-	msleep(150);
+	ata_msleep(ap, 150);
 
 	/* Before we perform post reset processing we want to see if
 	 * the bus shows 0xFF because the odd clown forgets the D7
diff --git a/drivers/ata/pata_samsung_cf.c b/drivers/ata/pata_samsung_cf.c
index 6f9cfb24b751..8a51d673e5b2 100644
--- a/drivers/ata/pata_samsung_cf.c
+++ b/drivers/ata/pata_samsung_cf.c
@@ -322,7 +322,7 @@ static int pata_s3c_wait_after_reset(struct ata_link *link,
 {
 	int rc;
 
-	msleep(ATA_WAIT_AFTER_RESET);
+	ata_msleep(link->ap, ATA_WAIT_AFTER_RESET);
 
 	/* always check readiness of the master device */
 	rc = ata_sff_wait_ready(link, deadline);
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index fe36966f7e34..093715c3273a 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -530,7 +530,7 @@ static int scc_wait_after_reset(struct ata_link *link, unsigned int devmask,
 	 *
 	 * Old drivers/ide uses the 2mS rule and then waits for ready.
 	 */
-	msleep(150);
+	ata_msleep(ap, 150);
 
 	/* always check readiness of the master device */
 	rc = ata_sff_wait_ready(link, deadline);
@@ -559,7 +559,7 @@ static int scc_wait_after_reset(struct ata_link *link, unsigned int devmask,
 			lbal = in_be32(ioaddr->lbal_addr);
 			if ((nsect == 1) && (lbal == 1))
 				break;
-			msleep(50);	/* give drive a breather */
+			ata_msleep(ap, 50);	/* give drive a breather */
 		}
 
 		rc = ata_sff_wait_ready(link, deadline);
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index 1440dc0af242..b0214d00d50b 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -678,7 +678,7 @@ static void sata_fsl_port_stop(struct ata_port *ap)
 	iowrite32(temp, hcr_base + HCONTROL);
 
 	/* Poll for controller to go offline - should happen immediately */
-	ata_wait_register(hcr_base + HSTATUS, ONLINE, ONLINE, 1, 1);
+	ata_wait_register(ap, hcr_base + HSTATUS, ONLINE, ONLINE, 1, 1);
 
 	ap->private_data = NULL;
 	dma_free_coherent(dev, SATA_FSL_PORT_PRIV_DMA_SZ,
@@ -729,7 +729,8 @@ try_offline_again:
 	iowrite32(temp, hcr_base + HCONTROL);
 
 	/* Poll for controller to go offline */
-	temp = ata_wait_register(hcr_base + HSTATUS, ONLINE, ONLINE, 1, 500);
+	temp = ata_wait_register(ap, hcr_base + HSTATUS, ONLINE, ONLINE,
+				 1, 500);
 
 	if (temp & ONLINE) {
 		ata_port_printk(ap, KERN_ERR,
@@ -752,7 +753,7 @@ try_offline_again:
 	/*
 	 * PHY reset should remain asserted for atleast 1ms
 	 */
-	msleep(1);
+	ata_msleep(ap, 1);
 
 	/*
 	 * Now, bring the host controller online again, this can take time
@@ -766,7 +767,7 @@ try_offline_again:
 	temp |= HCONTROL_PMP_ATTACHED;
 	iowrite32(temp, hcr_base + HCONTROL);
 
-	temp = ata_wait_register(hcr_base + HSTATUS, ONLINE, 0, 1, 500);
+	temp = ata_wait_register(ap, hcr_base + HSTATUS, ONLINE, 0, 1, 500);
 
 	if (!(temp & ONLINE)) {
 		ata_port_printk(ap, KERN_ERR,
@@ -784,7 +785,7 @@ try_offline_again:
 	 * presence
 	 */
 
-	temp = ata_wait_register(hcr_base + HSTATUS, 0xFF, 0, 1, 500);
+	temp = ata_wait_register(ap, hcr_base + HSTATUS, 0xFF, 0, 1, 500);
 	if ((!(temp & 0x10)) || ata_link_offline(link)) {
 		ata_port_printk(ap, KERN_WARNING,
 				"No Device OR PHYRDY change,Hstatus = 0x%x\n",
@@ -797,7 +798,7 @@ try_offline_again:
 	 * Wait for the first D2H from device,i.e,signature update notification
 	 */
 	start_jiffies = jiffies;
-	temp = ata_wait_register(hcr_base + HSTATUS, 0xFF, 0x10,
+	temp = ata_wait_register(ap, hcr_base + HSTATUS, 0xFF, 0x10,
 			500, jiffies_to_msecs(deadline - start_jiffies));
 
 	if ((temp & 0xFF) != 0x18) {
@@ -880,7 +881,7 @@ static int sata_fsl_softreset(struct ata_link *link, unsigned int *class,
 		iowrite32(pmp, CQPMP + hcr_base);
 	iowrite32(1, CQ + hcr_base);
 
-	temp = ata_wait_register(CQ + hcr_base, 0x1, 0x1, 1, 5000);
+	temp = ata_wait_register(ap, CQ + hcr_base, 0x1, 0x1, 1, 5000);
 	if (temp & 0x1) {
 		ata_port_printk(ap, KERN_WARNING, "ATA_SRST issue failed\n");
 
@@ -896,7 +897,7 @@ static int sata_fsl_softreset(struct ata_link *link, unsigned int *class,
 		goto err;
 	}
 
-	msleep(1);
+	ata_msleep(ap, 1);
 
 	/*
 	 * SATA device enters reset state after receving a Control register
@@ -915,7 +916,7 @@ static int sata_fsl_softreset(struct ata_link *link, unsigned int *class,
 	if (pmp != SATA_PMP_CTRL_PORT)
 		iowrite32(pmp, CQPMP + hcr_base);
 	iowrite32(1, CQ + hcr_base);
-	msleep(150);		/* ?? */
+	ata_msleep(ap, 150);		/* ?? */
 
 	/*
 	 * The above command would have signalled an interrupt on command
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index a36149ebf4a2..83a44471b189 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -614,7 +614,7 @@ static int inic_hardreset(struct ata_link *link, unsigned int *class,
 
 	writew(IDMA_CTL_RST_ATA, idma_ctl);
 	readw(idma_ctl);	/* flush */
-	msleep(1);
+	ata_msleep(ap, 1);
 	writew(0, idma_ctl);
 
 	rc = sata_link_resume(link, timing, deadline);
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index be7726d7686d..af41c6fd1254 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -589,9 +589,9 @@ static int sil24_init_port(struct ata_port *ap)
 		sil24_clear_pmp(ap);
 
 	writel(PORT_CS_INIT, port + PORT_CTRL_STAT);
-	ata_wait_register(port + PORT_CTRL_STAT,
+	ata_wait_register(ap, port + PORT_CTRL_STAT,
 			  PORT_CS_INIT, PORT_CS_INIT, 10, 100);
-	tmp = ata_wait_register(port + PORT_CTRL_STAT,
+	tmp = ata_wait_register(ap, port + PORT_CTRL_STAT,
 				PORT_CS_RDY, 0, 10, 100);
 
 	if ((tmp & (PORT_CS_INIT | PORT_CS_RDY)) != PORT_CS_RDY) {
@@ -631,7 +631,7 @@ static int sil24_exec_polled_cmd(struct ata_port *ap, int pmp,
 	writel((u64)paddr >> 32, port + PORT_CMD_ACTIVATE + 4);
 
 	irq_mask = (PORT_IRQ_COMPLETE | PORT_IRQ_ERROR) << PORT_IRQ_RAW_SHIFT;
-	irq_stat = ata_wait_register(port + PORT_IRQ_STAT, irq_mask, 0x0,
+	irq_stat = ata_wait_register(ap, port + PORT_IRQ_STAT, irq_mask, 0x0,
 				     10, timeout_msec);
 
 	writel(irq_mask, port + PORT_IRQ_STAT); /* clear IRQs */
@@ -719,9 +719,9 @@ static int sil24_hardreset(struct ata_link *link, unsigned int *class,
 				"state, performing PORT_RST\n");
 
 		writel(PORT_CS_PORT_RST, port + PORT_CTRL_STAT);
-		msleep(10);
+		ata_msleep(ap, 10);
 		writel(PORT_CS_PORT_RST, port + PORT_CTRL_CLR);
-		ata_wait_register(port + PORT_CTRL_STAT, PORT_CS_RDY, 0,
+		ata_wait_register(ap, port + PORT_CTRL_STAT, PORT_CS_RDY, 0,
 				  10, 5000);
 
 		/* restore port configuration */
@@ -740,7 +740,7 @@ static int sil24_hardreset(struct ata_link *link, unsigned int *class,
 		tout_msec = 5000;
 
 	writel(PORT_CS_DEV_RST, port + PORT_CTRL_STAT);
-	tmp = ata_wait_register(port + PORT_CTRL_STAT,
+	tmp = ata_wait_register(ap, port + PORT_CTRL_STAT,
 				PORT_CS_DEV_RST, PORT_CS_DEV_RST, 10,
 				tout_msec);
 
@@ -1253,7 +1253,7 @@ static void sil24_init_controller(struct ata_host *host)
 		tmp = readl(port + PORT_CTRL_STAT);
 		if (tmp & PORT_CS_PORT_RST) {
 			writel(PORT_CS_PORT_RST, port + PORT_CTRL_CLR);
-			tmp = ata_wait_register(port + PORT_CTRL_STAT,
+			tmp = ata_wait_register(NULL, port + PORT_CTRL_STAT,
 						PORT_CS_PORT_RST,
 						PORT_CS_PORT_RST, 10, 100);
 			if (tmp & PORT_CS_PORT_RST)
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 4730c42a5ee5..c21589986c69 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -349,7 +349,7 @@ static int vt6420_prereset(struct ata_link *link, unsigned long deadline)
 
 	/* wait for phy to become ready, if necessary */
 	do {
-		msleep(200);
+		ata_msleep(link->ap, 200);
 		svia_scr_read(link, SCR_STATUS, &sstatus);
 		if ((sstatus & 0xf) != 1)
 			break;
diff --git a/include/linux/libata.h b/include/linux/libata.h
index bc4ee218b185..2fbd22bd68ce 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1004,8 +1004,9 @@ extern int ata_host_suspend(struct ata_host *host, pm_message_t mesg);
 extern void ata_host_resume(struct ata_host *host);
 #endif
 extern int ata_ratelimit(void);
-extern u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
-			     unsigned long interval, unsigned long timeout);
+extern void ata_msleep(struct ata_port *ap, unsigned int msecs);
+extern u32 ata_wait_register(struct ata_port *ap, void __iomem *reg, u32 mask,
+			u32 val, unsigned long interval, unsigned long timeout);
 extern int atapi_cmd_type(u8 opcode);
 extern void ata_tf_to_fis(const struct ata_taskfile *tf,
 			  u8 pmp, int is_cmd, u8 *fis);
-- 
cgit v1.2.3


From c0c362b60e259e3480a36ef70280d545818844f0 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 6 Sep 2010 17:57:14 +0200
Subject: libata: implement cross-port EH exclusion

In libata, the non-EH code paths should always take and release
ap->lock explicitly when accessing hardware or shared data structures.
However, once EH is active, it's assumed that the port is owned by EH
and EH methods don't explicitly take ap->lock unless race from irq
handler or other code paths are expected.  However, libata EH didn't
guarantee exclusion among EHs for ports of the same host.  IOW,
multiple EHs may execute in parallel on multiple ports of the same
controller.

In many cases, especially in SATA, the ports are completely
independent of each other and this doesn't cause problems; however,
there are cases where different ports share the same resource, which
lead to obscure timing related bugs such as the one fixed by commit
213373cf (ata_piix: fix locking around SIDPR access).

This patch implements exclusion among EHs of the same host.  When EH
begins, it acquires per-host EH ownership by calling ata_eh_acquire().
When EH finishes, the ownership is released by calling
ata_eh_release().  EH ownership is also released whenever the EH
thread goes to sleep from ata_msleep() or explicitly and reacquired
after waking up.

This ensures that while EH is actively accessing the hardware, it has
exclusive access to it while allowing EHs to interleave and progress
in parallel as they hit waiting stages, which dominate the time spent
in EH.  This achieves cross-port EH exclusion without pervasive and
fragile changes while still allowing parallel EH for the most part.

This was first reported by yuanding02@gmail.com more than three years
ago in the following bugzilla.  :-)

  https://bugzilla.kernel.org/show_bug.cgi?id=8223

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Reported-by: yuanding02@gmail.com
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 30 ++++++++++++++++++++++++++++++
 drivers/ata/libata-eh.c   | 44 +++++++++++++++++++++++++++++++++++++++++++-
 drivers/ata/libata.h      |  2 ++
 include/linux/libata.h    |  5 +++++
 4 files changed, 80 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 42d9ce29f50d..7f77c67d267c 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -1628,8 +1628,14 @@ unsigned ata_exec_internal_sg(struct ata_device *dev,
 		}
 	}
 
+	if (ap->ops->error_handler)
+		ata_eh_release(ap);
+
 	rc = wait_for_completion_timeout(&wait, msecs_to_jiffies(timeout));
 
+	if (ap->ops->error_handler)
+		ata_eh_acquire(ap);
+
 	ata_sff_flush_pio_task(ap);
 
 	if (!rc) {
@@ -5570,6 +5576,7 @@ struct ata_host *ata_host_alloc(struct device *dev, int max_ports)
 	dev_set_drvdata(dev, host);
 
 	spin_lock_init(&host->lock);
+	mutex_init(&host->eh_mutex);
 	host->dev = dev;
 	host->n_ports = max_ports;
 
@@ -5867,6 +5874,7 @@ void ata_host_init(struct ata_host *host, struct device *dev,
 		   unsigned long flags, struct ata_port_operations *ops)
 {
 	spin_lock_init(&host->lock);
+	mutex_init(&host->eh_mutex);
 	host->dev = dev;
 	host->flags = flags;
 	host->ops = ops;
@@ -6483,9 +6491,31 @@ int ata_ratelimit(void)
 	return __ratelimit(&ratelimit);
 }
 
+/**
+ *	ata_msleep - ATA EH owner aware msleep
+ *	@ap: ATA port to attribute the sleep to
+ *	@msecs: duration to sleep in milliseconds
+ *
+ *	Sleeps @msecs.  If the current task is owner of @ap's EH, the
+ *	ownership is released before going to sleep and reacquired
+ *	after the sleep is complete.  IOW, other ports sharing the
+ *	@ap->host will be allowed to own the EH while this task is
+ *	sleeping.
+ *
+ *	LOCKING:
+ *	Might sleep.
+ */
 void ata_msleep(struct ata_port *ap, unsigned int msecs)
 {
+	bool owns_eh = ap && ap->host->eh_owner == current;
+
+	if (owns_eh)
+		ata_eh_release(ap);
+
 	msleep(msecs);
+
+	if (owns_eh)
+		ata_eh_acquire(ap);
 }
 
 /**
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 6780f4d16e81..5e590504f3aa 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -462,6 +462,41 @@ static void ata_eh_clear_action(struct ata_link *link, struct ata_device *dev,
 	}
 }
 
+/**
+ *	ata_eh_acquire - acquire EH ownership
+ *	@ap: ATA port to acquire EH ownership for
+ *
+ *	Acquire EH ownership for @ap.  This is the basic exclusion
+ *	mechanism for ports sharing a host.  Only one port hanging off
+ *	the same host can claim the ownership of EH.
+ *
+ *	LOCKING:
+ *	EH context.
+ */
+void ata_eh_acquire(struct ata_port *ap)
+{
+	mutex_lock(&ap->host->eh_mutex);
+	WARN_ON_ONCE(ap->host->eh_owner);
+	ap->host->eh_owner = current;
+}
+
+/**
+ *	ata_eh_release - release EH ownership
+ *	@ap: ATA port to release EH ownership for
+ *
+ *	Release EH ownership for @ap if the caller.  The caller must
+ *	have acquired EH ownership using ata_eh_acquire() previously.
+ *
+ *	LOCKING:
+ *	EH context.
+ */
+void ata_eh_release(struct ata_port *ap)
+{
+	WARN_ON_ONCE(ap->host->eh_owner != current);
+	ap->host->eh_owner = NULL;
+	mutex_unlock(&ap->host->eh_mutex);
+}
+
 /**
  *	ata_scsi_timed_out - SCSI layer time out callback
  *	@cmd: timed out SCSI command
@@ -639,11 +674,13 @@ void ata_scsi_error(struct Scsi_Host *host)
 	/* If we timed raced normal completion and there is nothing to
 	   recover nr_timedout == 0 why exactly are we doing error recovery ? */
 
- repeat:
 	/* invoke error handler */
 	if (ap->ops->error_handler) {
 		struct ata_link *link;
 
+		/* acquire EH ownership */
+		ata_eh_acquire(ap);
+ repeat:
 		/* kill fast drain timer */
 		del_timer_sync(&ap->fastdrain_timer);
 
@@ -718,6 +755,7 @@ void ata_scsi_error(struct Scsi_Host *host)
 		host->host_eh_scheduled = 0;
 
 		spin_unlock_irqrestore(ap->lock, flags);
+		ata_eh_release(ap);
 	} else {
 		WARN_ON(ata_qc_from_tag(ap, ap->link.active_tag) == NULL);
 		ap->ops->eng_timeout(ap);
@@ -2818,8 +2856,10 @@ int ata_eh_reset(struct ata_link *link, int classify,
 			"reset failed (errno=%d), retrying in %u secs\n",
 			rc, DIV_ROUND_UP(jiffies_to_msecs(delta), 1000));
 
+		ata_eh_release(ap);
 		while (delta)
 			delta = schedule_timeout_uninterruptible(delta);
+		ata_eh_acquire(ap);
 	}
 
 	if (try == max_tries - 1) {
@@ -3635,8 +3675,10 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 		if (time_before_eq(deadline, now))
 			break;
 
+		ata_eh_release(ap);
 		deadline = wait_for_completion_timeout(&ap->park_req_pending,
 						       deadline - now);
+		ata_eh_acquire(ap);
 	} while (deadline);
 	ata_for_each_link(link, ap, EDGE) {
 		ata_for_each_dev(dev, link, ALL) {
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 7c070a4b1c08..a9be110dbf51 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -145,6 +145,8 @@ extern int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel,
 /* libata-eh.c */
 extern unsigned long ata_internal_cmd_timeout(struct ata_device *dev, u8 cmd);
 extern void ata_internal_cmd_timed_out(struct ata_device *dev, u8 cmd);
+extern void ata_eh_acquire(struct ata_port *ap);
+extern void ata_eh_release(struct ata_port *ap);
 extern enum blk_eh_timer_return ata_scsi_timed_out(struct scsi_cmnd *cmd);
 extern void ata_scsi_error(struct Scsi_Host *host);
 extern void ata_port_wait_eh(struct ata_port *ap);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 2fbd22bd68ce..52112d39d71e 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -37,6 +37,7 @@
 #include <scsi/scsi_host.h>
 #include <linux/acpi.h>
 #include <linux/cdrom.h>
+#include <linux/sched.h>
 
 /*
  * Define if arch has non-standard setup.  This is a _PCI_ standard
@@ -535,6 +536,10 @@ struct ata_host {
 	void			*private_data;
 	struct ata_port_operations *ops;
 	unsigned long		flags;
+
+	struct mutex		eh_mutex;
+	struct task_struct	*eh_owner;
+
 #ifdef CONFIG_ATA_ACPI
 	acpi_handle		acpi_handle;
 #endif
-- 
cgit v1.2.3


From b34e90429ce8a23546b6b927d4e151df4c113644 Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Fri, 10 Sep 2010 12:19:43 +0100
Subject: libata: reorder ata_queued_cmd to remove alignment padding on 64 bit
 builds

Reorder structure ata_queued_cmd to remove 8 bytes of alignment padding
on 64 bit builds & therefore reduce the size of structure ata_port by
256 bytes.

Overall this will have little impact, other than reducing the amount of
memory that is cleared when allocating ata_ports.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/libata.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/libata.h b/include/linux/libata.h
index 52112d39d71e..15efec05df67 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -570,13 +570,13 @@ struct ata_queued_cmd {
 	unsigned int		extrabytes;
 	unsigned int		curbytes;
 
-	struct scatterlist	*cursg;
-	unsigned int		cursg_ofs;
-
 	struct scatterlist	sgent;
 
 	struct scatterlist	*sg;
 
+	struct scatterlist	*cursg;
+	unsigned int		cursg_ofs;
+
 	unsigned int		err_mask;
 	struct ata_taskfile	result_tf;
 	ata_qc_cb_t		complete_fn;
-- 
cgit v1.2.3


From 89692c03226a066a017048cf7fbacbaa645f0e79 Mon Sep 17 00:00:00 2001
From: Andrea Gelmini <andrea.gelmini@gelma.net>
Date: Sat, 16 Oct 2010 15:19:18 +0200
Subject: include/linux/libata.h: fix typo

Signed-off-by: Andrea Gelmini <andrea.gelmini@gelma.net>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/libata.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/libata.h b/include/linux/libata.h
index 15efec05df67..15b77b8dc7e1 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -342,7 +342,7 @@ enum {
 	ATA_EHI_DID_HARDRESET	= (1 << 17), /* already soft-reset this port */
 	ATA_EHI_PRINTINFO	= (1 << 18), /* print configuration info */
 	ATA_EHI_SETMODE		= (1 << 19), /* configure transfer mode */
-	ATA_EHI_POST_SETMODE	= (1 << 20), /* revaildating after setmode */
+	ATA_EHI_POST_SETMODE	= (1 << 20), /* revalidating after setmode */
 
 	ATA_EHI_DID_RESET	= ATA_EHI_DID_SOFTRESET | ATA_EHI_DID_HARDRESET,
 
-- 
cgit v1.2.3


From 0988c4c7fb5881377ec20a6452f739a722e97c6b Mon Sep 17 00:00:00 2001
From: Jesse Gross <jesse@nicira.com>
Date: Thu, 21 Oct 2010 11:30:42 +0000
Subject: vlan: Calling vlan_hwaccel_do_receive() is always valid.

It is now acceptable to receive vlan tagged packets at any time,
even if CONFIG_VLAN_8021Q is not set.  This means that calling
vlan_hwaccel_do_receive() should not result in BUG() but rather just
behave as if there were no vlan devices configured.

Reported-by: Vladislav Zolotarov <vladz@broadcom.com>
Signed-off-by: Jesse Gross <jesse@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index a0d9786c202d..c2f3a72712ce 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -168,7 +168,8 @@ static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 
 static inline bool vlan_hwaccel_do_receive(struct sk_buff **skb)
 {
-	BUG();
+	if ((*skb)->vlan_tci & VLAN_VID_MASK)
+		(*skb)->pkt_type = PACKET_OTHERHOST;
 	return false;
 }
 
-- 
cgit v1.2.3


From c64a0926710153b9d44c979d2942f4a8648fd74e Mon Sep 17 00:00:00 2001
From: Kevin Hilman <khilman@ti.com>
Date: Wed, 25 Aug 2010 12:50:00 -0700
Subject: driver core: platform_bus: allow runtime override of dev_pm_ops

Currently, the platform_bus allows customization of several of the
busses dev_pm_ops methods by using weak symbols so that platform code
can override them.  The weak-symbol approach is not scalable when
wanting to support multiple platforms in a single kernel binary.

Instead, provide __init methods for platform code to customize the
dev_pm_ops methods at runtime.

NOTE: after these dynamic methods are merged, the weak symbols should
      be removed from drivers/base/platform.c.  AFAIK, this will only
      affect SH and sh-mobile which should be converted to use this
      runtime approach instead of the weak symbols.  After SH &
      sh-mobile are converted, the weak symobols could be removed.

Tested on OMAP3.

Cc: Magnus Damm <magnus.damm@gmail.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/platform.c         | 35 +++++++++++++++++++++++++++++++++++
 include/linux/platform_device.h |  3 +++
 2 files changed, 38 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index 579906f88b09..a01abf9ebf7b 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -976,6 +976,41 @@ struct bus_type platform_bus_type = {
 };
 EXPORT_SYMBOL_GPL(platform_bus_type);
 
+/**
+ * platform_bus_get_pm_ops() - return pointer to busses dev_pm_ops
+ *
+ * This function can be used by platform code to get the current
+ * set of dev_pm_ops functions used by the platform_bus_type.
+ */
+const struct dev_pm_ops * __init platform_bus_get_pm_ops(void)
+{
+	return platform_bus_type.pm;
+}
+
+/**
+ * platform_bus_set_pm_ops() - update dev_pm_ops for the platform_bus_type
+ *
+ * @pm: pointer to new dev_pm_ops struct to be used for platform_bus_type
+ *
+ * Platform code can override the dev_pm_ops methods of
+ * platform_bus_type by using this function.  It is expected that
+ * platform code will first do a platform_bus_get_pm_ops(), then
+ * kmemdup it, then customize selected methods and pass a pointer to
+ * the new struct dev_pm_ops to this function.
+ *
+ * Since platform-specific code is customizing methods for *all*
+ * devices (not just platform-specific devices) it is expected that
+ * any custom overrides of these functions will keep existing behavior
+ * and simply extend it.  For example, any customization of the
+ * runtime PM methods should continue to call the pm_generic_*
+ * functions as the default ones do in addition to the
+ * platform-specific behavior.
+ */
+void __init platform_bus_set_pm_ops(const struct dev_pm_ops *pm)
+{
+	platform_bus_type.pm = pm;
+}
+
 int __init platform_bus_init(void)
 {
 	int error;
diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h
index d7ecad0093bb..2e700ec0601f 100644
--- a/include/linux/platform_device.h
+++ b/include/linux/platform_device.h
@@ -138,6 +138,9 @@ extern struct platform_device *platform_create_bundle(struct platform_driver *dr
 					struct resource *res, unsigned int n_res,
 					const void *data, size_t size);
 
+extern const struct dev_pm_ops * platform_bus_get_pm_ops(void);
+extern void platform_bus_set_pm_ops(const struct dev_pm_ops *pm);
+
 /* early platform driver interface */
 struct early_platform_driver {
 	const char *class_str;
-- 
cgit v1.2.3


From e52eec13cd6b7f30ab19081b387813e03e592ae5 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@linux.intel.com>
Date: Wed, 8 Sep 2010 16:54:17 +0200
Subject: SYSFS: Allow boot time switching between deprecated and modern sysfs
 layout

I have some systems which need legacy sysfs due to old tools that are
making assumptions that a directory can never be a symlink to another
directory, and it's a big hazzle to compile separate kernels for them.

This patch turns CONFIG_SYSFS_DEPRECATED into a run time option
that can be switched on/off the kernel command line. This way
the same binary can be used in both cases with just a option
on the command line.

The old CONFIG_SYSFS_DEPRECATED_V2 option is still there to set
the default. I kept the weird name to not break existing
config files.

Also the compat code can be still completely disabled by undefining
CONFIG_SYSFS_DEPRECATED_SWITCH -- just the optimizer takes
care of this now instead of lots of ifdefs. This makes the code
look nicer.

v2: This is an updated version on top of Kay's patch to only
handle the block devices. I tested it on my old systems
and that seems to work.

Cc: axboe@kernel.dk
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/kernel-parameters.txt |  9 +++++++++
 block/genhd.c                       |  7 ++-----
 drivers/base/class.c                |  4 ++--
 drivers/base/core.c                 | 26 +++++++++++++++++---------
 fs/partitions/check.c               | 19 +++++++++----------
 include/linux/device.h              |  7 +++++++
 init/Kconfig                        | 26 ++++++++++++++++++++++----
 7 files changed, 68 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 3d854c9ae53a..67fa3fdc128f 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -2365,6 +2365,15 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	switches=	[HW,M68k]
 
+	sysfs.deprecated=0|1 [KNL]
+			Enable/disable old style sysfs layout for old udev
+			on older distributions. When this option is enabled
+			very new udev will not work anymore. When this option
+			is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
+			in older udev will not work anymore.
+			Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
+			the kernel configuration.
+
 	sysrq_always_enabled
 			[KNL]
 			Ignore sysrq setting - this boot parameter will
diff --git a/block/genhd.c b/block/genhd.c
index 59a2db6fecef..4e28a840bde0 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -22,9 +22,7 @@
 #include "blk.h"
 
 static DEFINE_MUTEX(block_class_lock);
-#ifndef CONFIG_SYSFS_DEPRECATED
 struct kobject *block_depr;
-#endif
 
 /* for extended dynamic devt allocation, currently only one major is used */
 #define MAX_EXT_DEVT		(1 << MINORBITS)
@@ -803,10 +801,9 @@ static int __init genhd_device_init(void)
 
 	register_blkdev(BLOCK_EXT_MAJOR, "blkext");
 
-#ifndef CONFIG_SYSFS_DEPRECATED
 	/* create top-level block dir */
-	block_depr = kobject_create_and_add("block", NULL);
-#endif
+	if (!sysfs_deprecated)
+		block_depr = kobject_create_and_add("block", NULL);
 	return 0;
 }
 
diff --git a/drivers/base/class.c b/drivers/base/class.c
index 1078969889fd..9c63a5687d69 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -184,9 +184,9 @@ int __class_register(struct class *cls, struct lock_class_key *key)
 	if (!cls->dev_kobj)
 		cls->dev_kobj = sysfs_dev_char_kobj;
 
-#if defined(CONFIG_SYSFS_DEPRECATED) && defined(CONFIG_BLOCK)
+#if defined(CONFIG_BLOCK)
 	/* let the block class directory show up in the root of sysfs */
-	if (cls != &block_class)
+	if (!sysfs_deprecated || cls != &block_class)
 		cp->class_subsys.kobj.kset = class_kset;
 #else
 	cp->class_subsys.kobj.kset = class_kset;
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 6cf9069f3150..f7f906f0a2f2 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -26,6 +26,19 @@
 #include "base.h"
 #include "power/power.h"
 
+#ifdef CONFIG_SYSFS_DEPRECATED
+#ifdef CONFIG_SYSFS_DEPRECATED_V2
+long sysfs_deprecated = 1;
+#else
+long sysfs_deprecated = 0;
+#endif
+static __init int sysfs_deprecated_setup(char *arg)
+{
+	return strict_strtol(arg, 10, &sysfs_deprecated);
+}
+early_param("sysfs.deprecated", sysfs_deprecated_setup);
+#endif
+
 int (*platform_notify)(struct device *dev) = NULL;
 int (*platform_notify_remove)(struct device *dev) = NULL;
 static struct kobject *dev_kobj;
@@ -617,14 +630,13 @@ static struct kobject *get_device_parent(struct device *dev,
 		struct kobject *parent_kobj;
 		struct kobject *k;
 
-#ifdef CONFIG_SYSFS_DEPRECATED
 		/* block disks show up in /sys/block */
-		if (dev->class == &block_class) {
+		if (sysfs_deprecated && dev->class == &block_class) {
 			if (parent && parent->class == &block_class)
 				return &parent->kobj;
 			return &block_class.p->class_subsys.kobj;
 		}
-#endif
+
 		/*
 		 * If we have no parent, we live in "virtual".
 		 * Class-devices with a non class-device as parent, live
@@ -707,11 +719,9 @@ static int device_add_class_symlinks(struct device *dev)
 			goto out_subsys;
 	}
 
-#ifdef CONFIG_SYSFS_DEPRECATED
 	/* /sys/block has directories and does not need symlinks */
-	if (dev->class == &block_class)
+	if (sysfs_deprecated && dev->class == &block_class)
 		return 0;
-#endif
 
 	/* link in the class directory pointing to the device */
 	error = sysfs_create_link(&dev->class->p->class_subsys.kobj,
@@ -738,10 +748,8 @@ static void device_remove_class_symlinks(struct device *dev)
 	if (dev->parent && device_is_not_partition(dev))
 		sysfs_remove_link(&dev->kobj, "device");
 	sysfs_remove_link(&dev->kobj, "subsystem");
-#ifdef CONFIG_SYSFS_DEPRECATED
-	if (dev->class == &block_class)
+	if (sysfs_deprecated && dev->class == &block_class)
 		return;
-#endif
 	sysfs_delete_link(&dev->class->p->class_subsys.kobj, &dev->kobj, dev_name(dev));
 }
 
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 79fbf3f390f0..137bf9787853 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -513,14 +513,14 @@ void register_disk(struct gendisk *disk)
 
 	if (device_add(ddev))
 		return;
-#ifndef CONFIG_SYSFS_DEPRECATED
-	err = sysfs_create_link(block_depr, &ddev->kobj,
-				kobject_name(&ddev->kobj));
-	if (err) {
-		device_del(ddev);
-		return;
+	if (!sysfs_deprecated) {
+		err = sysfs_create_link(block_depr, &ddev->kobj,
+					kobject_name(&ddev->kobj));
+		if (err) {
+			device_del(ddev);
+			return;
+		}
 	}
-#endif
 	disk->part0.holder_dir = kobject_create_and_add("holders", &ddev->kobj);
 	disk->slave_dir = kobject_create_and_add("slaves", &ddev->kobj);
 
@@ -737,8 +737,7 @@ void del_gendisk(struct gendisk *disk)
 	kobject_put(disk->part0.holder_dir);
 	kobject_put(disk->slave_dir);
 	disk->driverfs_dev = NULL;
-#ifndef CONFIG_SYSFS_DEPRECATED
-	sysfs_remove_link(block_depr, dev_name(disk_to_dev(disk)));
-#endif
+	if (!sysfs_deprecated)
+		sysfs_remove_link(block_depr, dev_name(disk_to_dev(disk)));
 	device_del(disk_to_dev(disk));
 }
diff --git a/include/linux/device.h b/include/linux/device.h
index 516fecacf27b..dd4895313468 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -751,4 +751,11 @@ do {						     \
 	MODULE_ALIAS("char-major-" __stringify(major) "-" __stringify(minor))
 #define MODULE_ALIAS_CHARDEV_MAJOR(major) \
 	MODULE_ALIAS("char-major-" __stringify(major) "-*")
+
+#ifdef CONFIG_SYSFS_DEPRECATED
+extern long sysfs_deprecated;
+#else
+#define sysfs_deprecated 0
+#endif
+
 #endif /* _DEVICE_H_ */
diff --git a/init/Kconfig b/init/Kconfig
index 137609f33ebc..d742b6fca8d2 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -660,8 +660,12 @@ config SYSFS_DEPRECATED
 	depends on SYSFS
 	default n
 	help
-	  This option switches the layout of the "block" class devices, to not
-	  show up in /sys/class/block/, but only in /sys/block/.
+	  This option adds code that switches the layout of the "block" class
+	  devices, to not show up in /sys/class/block/, but only in
+	  /sys/block/.
+
+	  This switch is only active when the sysfs.deprecated=1 boot option is
+	  passed or the SYSFS_DEPRECATED_V2 option is set.
 
 	  This option allows new kernels to run on old distributions and tools,
 	  which might get confused by /sys/class/block/. Since 2007/2008 all
@@ -672,8 +676,22 @@ config SYSFS_DEPRECATED
 	  option enabled.
 
 	  Only if you are using a new kernel on an old distribution, you might
-	  need to say Y here. Never say Y, if the original kernel, that came
-	  with your distribution, has not set this option.
+	  need to say Y here.
+
+config SYSFS_DEPRECATED_V2
+	bool "enabled deprecated sysfs features by default"
+	default n
+	depends on SYSFS
+	depends on SYSFS_DEPRECATED
+	help
+	  Enable deprecated sysfs by default.
+
+	  See the CONFIG_SYSFS_DEPRECATED option for more details about this
+	  option.
+
+	  Only if you are using a new kernel on an old distribution, you might
+	  need to say Y here. Even then, odds are you would not need it
+	  enabled, you can always pass the boot option if absolutely necessary.
 
 config RELAY
 	bool "Kernel->user space relay support (formerly relayfs)"
-- 
cgit v1.2.3


From 6427a7655afd7f07dfa83736defd1d94656c83e5 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@aristanetworks.com>
Date: Tue, 14 Sep 2010 11:37:36 -0700
Subject: uio: Cleanup irq handling.

Change the value of UIO_IRQ_NONE -2 to 0.  0 is well defined in the rest
of the kernel as the value to indicate an irq has not been assigned.

Update the calls to request_irq and free_irq to only ignore UIO_IRQ_NONE
and UIO_IRQ_CUSTOM allowing the rest of the kernel's possible irq
numbers to be used.

Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Hans J. Koch <hjk@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/uio/uio.c          | 14 +++++++-------
 include/linux/uio_driver.h |  2 +-
 2 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index 8132288920b2..0fd2cda74244 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -512,7 +512,7 @@ static unsigned int uio_poll(struct file *filep, poll_table *wait)
 	struct uio_listener *listener = filep->private_data;
 	struct uio_device *idev = listener->dev;
 
-	if (idev->info->irq == UIO_IRQ_NONE)
+	if (!idev->info->irq)
 		return -EIO;
 
 	poll_wait(filep, &idev->wait, wait);
@@ -530,7 +530,7 @@ static ssize_t uio_read(struct file *filep, char __user *buf,
 	ssize_t retval;
 	s32 event_count;
 
-	if (idev->info->irq == UIO_IRQ_NONE)
+	if (!idev->info->irq)
 		return -EIO;
 
 	if (count != sizeof(s32))
@@ -578,7 +578,7 @@ static ssize_t uio_write(struct file *filep, const char __user *buf,
 	ssize_t retval;
 	s32 irq_on;
 
-	if (idev->info->irq == UIO_IRQ_NONE)
+	if (!idev->info->irq)
 		return -EIO;
 
 	if (count != sizeof(s32))
@@ -825,9 +825,9 @@ int __uio_register_device(struct module *owner,
 
 	info->uio_dev = idev;
 
-	if (idev->info->irq >= 0) {
-		ret = request_irq(idev->info->irq, uio_interrupt,
-				  idev->info->irq_flags, idev->info->name, idev);
+	if (info->irq && (info->irq != UIO_IRQ_CUSTOM)) {
+		ret = request_irq(info->irq, uio_interrupt,
+				  info->irq_flags, info->name, idev);
 		if (ret)
 			goto err_request_irq;
 	}
@@ -863,7 +863,7 @@ void uio_unregister_device(struct uio_info *info)
 
 	uio_free_minor(idev);
 
-	if (info->irq >= 0)
+	if (info->irq && (info->irq != UIO_IRQ_CUSTOM))
 		free_irq(info->irq, idev);
 
 	uio_dev_del_attributes(idev);
diff --git a/include/linux/uio_driver.h b/include/linux/uio_driver.h
index 5dcc9ff72f69..d6188e5a52df 100644
--- a/include/linux/uio_driver.h
+++ b/include/linux/uio_driver.h
@@ -108,7 +108,7 @@ extern void uio_event_notify(struct uio_info *info);
 
 /* defines for uio_info->irq */
 #define UIO_IRQ_CUSTOM	-1
-#define UIO_IRQ_NONE	-2
+#define UIO_IRQ_NONE	0
 
 /* defines for uio_mem->memtype */
 #define UIO_MEM_NONE	0
-- 
cgit v1.2.3


From c25d1dfbd403209025df41a737f82ce8f43d93f5 Mon Sep 17 00:00:00 2001
From: Robin Holt <holt@sgi.com>
Date: Wed, 29 Sep 2010 14:00:54 -0500
Subject: kobject: Introduce kset_find_obj_hinted.

One call chain getting to kset_find_obj is:
  link_mem_sections()
    find_mem_section()
      kset_find_obj()

This is done during boot.  The memory sections were added in a linearly
increasing order and link_mem_sections tends to utilize them in that
same linear order.

Introduce a kset_find_obj_hinted which is passed the result of the
previous kset_find_obj which it uses for a quick "is the next object
our desired object" check before falling back to the old behavior.

Signed-off-by: Robin Holt <holt@sgi.com>
To: Robert P. J. Day <rpjday@crashcourse.ca>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h |  2 ++
 lib/kobject.c           | 39 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 7950a37a7146..8f6d12151048 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -191,6 +191,8 @@ static inline struct kobj_type *get_ktype(struct kobject *kobj)
 }
 
 extern struct kobject *kset_find_obj(struct kset *, const char *);
+extern struct kobject *kset_find_obj_hinted(struct kset *, const char *,
+						struct kobject *);
 
 /* The global /sys/kernel/ kobject for people to chain off of */
 extern struct kobject *kernel_kobj;
diff --git a/lib/kobject.c b/lib/kobject.c
index f07c57252e82..82dc34c095c2 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -745,18 +745,57 @@ void kset_unregister(struct kset *k)
  * take a reference and return the object.
  */
 struct kobject *kset_find_obj(struct kset *kset, const char *name)
+{
+	return kset_find_obj_hinted(kset, name, NULL);
+}
+
+/**
+ * kset_find_obj_hinted - search for object in kset given a predecessor hint.
+ * @kset: kset we're looking in.
+ * @name: object's name.
+ * @hint: hint to possible object's predecessor.
+ *
+ * Check the hint's next object and if it is a match return it directly,
+ * otherwise, fall back to the behavior of kset_find_obj().  Either way
+ * a reference for the returned object is held and the reference on the
+ * hinted object is released.
+ */
+struct kobject *kset_find_obj_hinted(struct kset *kset, const char *name,
+				     struct kobject *hint)
 {
 	struct kobject *k;
 	struct kobject *ret = NULL;
 
 	spin_lock(&kset->list_lock);
+
+	if (!hint)
+		goto slow_search;
+
+	/* end of list detection */
+	if (hint->entry.next == kset->list.next)
+		goto slow_search;
+
+	k = container_of(hint->entry.next, struct kobject, entry);
+	if (!kobject_name(k) || strcmp(kobject_name(k), name))
+		goto slow_search;
+
+	ret = kobject_get(k);
+	goto unlock_exit;
+
+slow_search:
 	list_for_each_entry(k, &kset->list, entry) {
 		if (kobject_name(k) && !strcmp(kobject_name(k), name)) {
 			ret = kobject_get(k);
 			break;
 		}
 	}
+
+unlock_exit:
 	spin_unlock(&kset->list_lock);
+
+	if (hint)
+		kobject_put(hint);
+
 	return ret;
 }
 
-- 
cgit v1.2.3


From 98383031ed77c6eb49ab612166fef9c0efe1604a Mon Sep 17 00:00:00 2001
From: Robin Holt <holt@sgi.com>
Date: Wed, 29 Sep 2010 14:00:55 -0500
Subject: driver core: Introduce find_memory_block_hinted which utilizes
 kset_find_obj_hinted.

Introduce a find_memory_block_hinted() which utilizes the
recently added kset_find_obj_hinted().

Signed-off-by: Robin Holt <holt@sgi.com>
To: Dave Hansen <haveblue@us.ibm.com>
To: Matt Tolentino <matthew.e.tolentino@intel.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/memory.c  | 28 ++++++++++++++++++----------
 include/linux/memory.h |  2 ++
 2 files changed, 20 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 933442f40321..a7994409b9a5 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -468,28 +468,23 @@ static int add_memory_block(int nid, struct mem_section *section,
 	return ret;
 }
 
-/*
- * For now, we have a linear search to go find the appropriate
- * memory_block corresponding to a particular phys_index. If
- * this gets to be a real problem, we can always use a radix
- * tree or something here.
- *
- * This could be made generic for all sysdev classes.
- */
-struct memory_block *find_memory_block(struct mem_section *section)
+struct memory_block *find_memory_block_hinted(struct mem_section *section,
+					      struct memory_block *hint)
 {
 	struct kobject *kobj;
 	struct sys_device *sysdev;
 	struct memory_block *mem;
 	char name[sizeof(MEMORY_CLASS_NAME) + 9 + 1];
 
+	kobj = hint ? &hint->sysdev.kobj : NULL;
+
 	/*
 	 * This only works because we know that section == sysdev->id
 	 * slightly redundant with sysdev_register()
 	 */
 	sprintf(&name[0], "%s%d", MEMORY_CLASS_NAME, __section_nr(section));
 
-	kobj = kset_find_obj(&memory_sysdev_class.kset, name);
+	kobj = kset_find_obj_hinted(&memory_sysdev_class.kset, name, kobj);
 	if (!kobj)
 		return NULL;
 
@@ -499,6 +494,19 @@ struct memory_block *find_memory_block(struct mem_section *section)
 	return mem;
 }
 
+/*
+ * For now, we have a linear search to go find the appropriate
+ * memory_block corresponding to a particular phys_index. If
+ * this gets to be a real problem, we can always use a radix
+ * tree or something here.
+ *
+ * This could be made generic for all sysdev classes.
+ */
+struct memory_block *find_memory_block(struct mem_section *section)
+{
+	return find_memory_block_hinted(section, NULL);
+}
+
 int remove_memory_block(unsigned long node_id, struct mem_section *section,
 		int phys_device)
 {
diff --git a/include/linux/memory.h b/include/linux/memory.h
index 85582e1bcee9..c4f3127dbd48 100644
--- a/include/linux/memory.h
+++ b/include/linux/memory.h
@@ -113,6 +113,8 @@ extern int memory_dev_init(void);
 extern int remove_memory_block(unsigned long, struct mem_section *, int);
 extern int memory_notify(unsigned long val, void *v);
 extern int memory_isolate_notify(unsigned long val, void *v);
+extern struct memory_block *find_memory_block_hinted(struct mem_section *,
+							struct memory_block *);
 extern struct memory_block *find_memory_block(struct mem_section *);
 #define CONFIG_MEM_BLOCK_SIZE	(PAGES_PER_SECTION<<PAGE_SHIFT)
 enum mem_add_context { BOOT, HOTPLUG };
-- 
cgit v1.2.3


From 07681215975e05a1454b0afdeef07deb0db626ee Mon Sep 17 00:00:00 2001
From: Nathan Fontenot <nfont@austin.ibm.com>
Date: Tue, 19 Oct 2010 12:46:19 -0500
Subject: Driver core: Add section count to memory_block struct

Add a section count property to the memory_block struct to track the number
of memory sections that have been added/removed from a memory block. This
allows us to know when the last memory section of a memory block has been
removed so we can remove the memory block.

Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Reviewed-by: Robin Holt <holt@sgi.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/memory.c  | 17 +++++++++++------
 include/linux/memory.h |  2 ++
 2 files changed, 13 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 5185bcff2de9..cafeaaf0428f 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -490,6 +490,7 @@ static int add_memory_block(int nid, struct mem_section *section,
 
 	mem->phys_index = __section_nr(section);
 	mem->state = state;
+	mem->section_count++;
 	mutex_init(&mem->state_mutex);
 	start_pfn = section_nr_to_pfn(mem->phys_index);
 	mem->phys_device = arch_get_memory_phys_device(start_pfn);
@@ -519,12 +520,16 @@ int remove_memory_block(unsigned long node_id, struct mem_section *section,
 
 	mutex_lock(&mem_sysfs_mutex);
 	mem = find_memory_block(section);
-	unregister_mem_sect_under_nodes(mem);
-	mem_remove_simple_file(mem, phys_index);
-	mem_remove_simple_file(mem, state);
-	mem_remove_simple_file(mem, phys_device);
-	mem_remove_simple_file(mem, removable);
-	unregister_memory(mem, section);
+
+	mem->section_count--;
+	if (mem->section_count == 0) {
+		unregister_mem_sect_under_nodes(mem);
+		mem_remove_simple_file(mem, phys_index);
+		mem_remove_simple_file(mem, state);
+		mem_remove_simple_file(mem, phys_device);
+		mem_remove_simple_file(mem, removable);
+		unregister_memory(mem, section);
+	}
 
 	mutex_unlock(&mem_sysfs_mutex);
 	return 0;
diff --git a/include/linux/memory.h b/include/linux/memory.h
index c4f3127dbd48..06c1fa0a5c7b 100644
--- a/include/linux/memory.h
+++ b/include/linux/memory.h
@@ -23,6 +23,8 @@
 struct memory_block {
 	unsigned long phys_index;
 	unsigned long state;
+	int section_count;
+
 	/*
 	 * This serializes all state change requests.  It isn't
 	 * held during creation because the control files are
-- 
cgit v1.2.3


From 30004ac9c090dcdcca99556b4587b3bad828731a Mon Sep 17 00:00:00 2001
From: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Date: Mon, 9 Aug 2010 18:22:49 +0400
Subject: tty: add tty_struct->dev pointer to corresponding device instance

Some device drivers (mostly tty line disciplines) would like to have way
know a struct device instance corresponding to passed tty_struct. Add
a struct device pointer to struct tty_struct and populate it during
initialize_tty_struct().

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Acked-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/char/tty_io.c | 17 +++++++++++++++++
 include/linux/tty.h   |  1 +
 2 files changed, 18 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 613c852ee0fe..dc184d4b5638 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -183,6 +183,8 @@ struct tty_struct *alloc_tty_struct(void)
 
 void free_tty_struct(struct tty_struct *tty)
 {
+	if (tty->dev)
+		put_device(tty->dev);
 	kfree(tty->write_buf);
 	tty_buffer_free_all(tty);
 	kfree(tty);
@@ -2783,6 +2785,20 @@ void do_SAK(struct tty_struct *tty)
 
 EXPORT_SYMBOL(do_SAK);
 
+static int dev_match_devt(struct device *dev, void *data)
+{
+	dev_t *devt = data;
+	return dev->devt == *devt;
+}
+
+/* Must put_device() after it's unused! */
+static struct device *tty_get_device(struct tty_struct *tty)
+{
+	dev_t devt = tty_devnum(tty);
+	return class_find_device(tty_class, NULL, &devt, dev_match_devt);
+}
+
+
 /**
  *	initialize_tty_struct
  *	@tty: tty to initialize
@@ -2823,6 +2839,7 @@ void initialize_tty_struct(struct tty_struct *tty,
 	tty->ops = driver->ops;
 	tty->index = idx;
 	tty_line_name(driver, idx, tty->name);
+	tty->dev = tty_get_device(tty);
 }
 
 /**
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 67d64e6efe7a..d94eb86266c4 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -256,6 +256,7 @@ struct tty_operations;
 struct tty_struct {
 	int	magic;
 	struct kref kref;
+	struct device *dev;
 	struct tty_driver *driver;
 	const struct tty_operations *ops;
 	int index;
-- 
cgit v1.2.3


From f573bd1764f0f3f47754ca1ae7b2eb2909798a60 Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@kernel.org>
Date: Tue, 24 Aug 2010 07:48:34 +0300
Subject: tty: Remove __GFP_NOFAIL from tty_add_file()

This patch removes __GFP_NOFAIL use from tty_add_file() and adds proper error
handling to the call-sites of the function.

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/char/pty.c    |  4 +++-
 drivers/char/tty_io.c | 15 +++++++++++----
 include/linux/tty.h   |  2 +-
 3 files changed, 15 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/pty.c b/drivers/char/pty.c
index c350d01716bd..923a48585501 100644
--- a/drivers/char/pty.c
+++ b/drivers/char/pty.c
@@ -676,7 +676,9 @@ static int ptmx_open(struct inode *inode, struct file *filp)
 
 	set_bit(TTY_PTY_LOCK, &tty->flags); /* LOCK THE SLAVE */
 
-	tty_add_file(tty, filp);
+	retval = tty_add_file(tty, filp);
+	if (retval)
+		goto out;
 
 	retval = devpts_pty_new(inode, tty->link);
 	if (retval)
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index dc184d4b5638..d6c659f2f659 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -196,12 +196,13 @@ static inline struct tty_struct *file_tty(struct file *file)
 }
 
 /* Associate a new file with the tty structure */
-void tty_add_file(struct tty_struct *tty, struct file *file)
+int tty_add_file(struct tty_struct *tty, struct file *file)
 {
 	struct tty_file_private *priv;
 
-	/* XXX: must implement proper error handling in callers */
-	priv = kmalloc(sizeof(*priv), GFP_KERNEL|__GFP_NOFAIL);
+	priv = kmalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
 
 	priv->tty = tty;
 	priv->file = file;
@@ -210,6 +211,8 @@ void tty_add_file(struct tty_struct *tty, struct file *file)
 	spin_lock(&tty_files_lock);
 	list_add(&priv->list, &tty->tty_files);
 	spin_unlock(&tty_files_lock);
+
+	return 0;
 }
 
 /* Delete file from its tty */
@@ -1877,7 +1880,11 @@ got_driver:
 		return PTR_ERR(tty);
 	}
 
-	tty_add_file(tty, filp);
+	retval = tty_add_file(tty, filp);
+	if (retval) {
+		tty_unlock();
+		return retval;
+	}
 
 	check_tty_count(tty, "tty_open");
 	if (tty->driver->type == TTY_DRIVER_TYPE_PTY &&
diff --git a/include/linux/tty.h b/include/linux/tty.h
index d94eb86266c4..86be0cdeb11b 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -466,7 +466,7 @@ extern void proc_clear_tty(struct task_struct *p);
 extern struct tty_struct *get_current_tty(void);
 extern void tty_default_fops(struct file_operations *fops);
 extern struct tty_struct *alloc_tty_struct(void);
-extern void tty_add_file(struct tty_struct *tty, struct file *file);
+extern int tty_add_file(struct tty_struct *tty, struct file *file);
 extern void free_tty_struct(struct tty_struct *tty);
 extern void initialize_tty_struct(struct tty_struct *tty,
 		struct tty_driver *driver, int idx);
-- 
cgit v1.2.3


From d281da7ff6f70efca0553c288bb883e8605b3862 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Thu, 16 Sep 2010 18:21:24 +0100
Subject: tty: Make tiocgicount a handler

Dan Rosenberg noted that various drivers return the struct with uncleared
fields. Instead of spending forever trying to stomp all the drivers that
get it wrong (and every new driver) do the job in one place.

This first patch adds the needed operations and hooks them up, including
the needed USB midlayer and serial core plumbing.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/char/tty_io.c           | 21 +++++++++++++++++++++
 drivers/serial/serial_core.c    | 35 ++++++++++++++++-------------------
 drivers/usb/serial/usb-serial.c | 13 +++++++++++++
 include/linux/tty_driver.h      |  9 +++++++++
 include/linux/usb/serial.h      |  2 ++
 5 files changed, 61 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index e185db36f8ad..c05c5af5aa04 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -96,6 +96,7 @@
 #include <linux/bitops.h>
 #include <linux/delay.h>
 #include <linux/seq_file.h>
+#include <linux/serial.h>
 
 #include <linux/uaccess.h>
 #include <asm/system.h>
@@ -2511,6 +2512,20 @@ static int tty_tiocmset(struct tty_struct *tty, struct file *file, unsigned int
 	return tty->ops->tiocmset(tty, file, set, clear);
 }
 
+static int tty_tiocgicount(struct tty_struct *tty, void __user *arg)
+{
+	int retval = -EINVAL;
+	struct serial_icounter_struct icount;
+	memset(&icount, 0, sizeof(icount));
+	if (tty->ops->get_icount)
+		retval = tty->ops->get_icount(tty, &icount);
+	if (retval != 0)
+		return retval;
+	if (copy_to_user(arg, &icount, sizeof(icount)))
+		return -EFAULT;
+	return 0;
+}
+
 struct tty_struct *tty_pair_get_tty(struct tty_struct *tty)
 {
 	if (tty->driver->type == TTY_DRIVER_TYPE_PTY &&
@@ -2631,6 +2646,12 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case TIOCMBIC:
 	case TIOCMBIS:
 		return tty_tiocmset(tty, file, cmd, p);
+	case TIOCGICOUNT:
+		retval = tty_tiocgicount(tty, p);
+		/* For the moment allow fall through to the old method */
+        	if (retval != -EINVAL)
+			return retval;
+		break;
 	case TCFLSH:
 		switch (arg) {
 		case TCIFLUSH:
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index bc6cddd10294..c4ea14670d44 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -1074,10 +1074,10 @@ uart_wait_modem_status(struct uart_state *state, unsigned long arg)
  * NB: both 1->0 and 0->1 transitions are counted except for
  *     RI where only 0->1 is counted.
  */
-static int uart_get_count(struct uart_state *state,
-			  struct serial_icounter_struct __user *icnt)
+static int uart_get_icount(struct tty_struct *tty,
+			  struct serial_icounter_struct *icount)
 {
-	struct serial_icounter_struct icount;
+	struct uart_state *state = tty->driver_data;
 	struct uart_icount cnow;
 	struct uart_port *uport = state->uart_port;
 
@@ -1085,19 +1085,19 @@ static int uart_get_count(struct uart_state *state,
 	memcpy(&cnow, &uport->icount, sizeof(struct uart_icount));
 	spin_unlock_irq(&uport->lock);
 
-	icount.cts         = cnow.cts;
-	icount.dsr         = cnow.dsr;
-	icount.rng         = cnow.rng;
-	icount.dcd         = cnow.dcd;
-	icount.rx          = cnow.rx;
-	icount.tx          = cnow.tx;
-	icount.frame       = cnow.frame;
-	icount.overrun     = cnow.overrun;
-	icount.parity      = cnow.parity;
-	icount.brk         = cnow.brk;
-	icount.buf_overrun = cnow.buf_overrun;
+	icount->cts         = cnow.cts;
+	icount->dsr         = cnow.dsr;
+	icount->rng         = cnow.rng;
+	icount->dcd         = cnow.dcd;
+	icount->rx          = cnow.rx;
+	icount->tx          = cnow.tx;
+	icount->frame       = cnow.frame;
+	icount->overrun     = cnow.overrun;
+	icount->parity      = cnow.parity;
+	icount->brk         = cnow.brk;
+	icount->buf_overrun = cnow.buf_overrun;
 
-	return copy_to_user(icnt, &icount, sizeof(icount)) ? -EFAULT : 0;
+	return 0;
 }
 
 /*
@@ -1150,10 +1150,6 @@ uart_ioctl(struct tty_struct *tty, struct file *filp, unsigned int cmd,
 	case TIOCMIWAIT:
 		ret = uart_wait_modem_status(state, arg);
 		break;
-
-	case TIOCGICOUNT:
-		ret = uart_get_count(state, uarg);
-		break;
 	}
 
 	if (ret != -ENOIOCTLCMD)
@@ -2295,6 +2291,7 @@ static const struct tty_operations uart_ops = {
 #endif
 	.tiocmget	= uart_tiocmget,
 	.tiocmset	= uart_tiocmset,
+	.get_icount	= uart_get_icount,
 #ifdef CONFIG_CONSOLE_POLL
 	.poll_init	= uart_poll_init,
 	.poll_get_char	= uart_poll_get_char,
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 7a2177c79bde..e64da74bdcc5 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -519,6 +519,18 @@ static int serial_tiocmset(struct tty_struct *tty, struct file *file,
 	return -EINVAL;
 }
 
+static int serial_get_icount(struct tty_struct *tty,
+				struct serial_icounter_struct *icount)
+{
+	struct usb_serial_port *port = tty->driver_data;
+
+	dbg("%s - port %d", __func__, port->number);
+
+	if (port->serial->type->get_icount)
+		return port->serial->type->get_icount(tty, icount);
+	return -EINVAL;
+}
+
 /*
  * We would be calling tty_wakeup here, but unfortunately some line
  * disciplines have an annoying habit of calling tty->write from
@@ -1195,6 +1207,7 @@ static const struct tty_operations serial_ops = {
 	.chars_in_buffer =	serial_chars_in_buffer,
 	.tiocmget =		serial_tiocmget,
 	.tiocmset =		serial_tiocmset,
+	.get_icount = 		serial_get_icount,
 	.cleanup = 		serial_cleanup,
 	.install = 		serial_install,
 	.proc_fops =		&serial_proc_fops,
diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h
index b08677982525..db2d227694da 100644
--- a/include/linux/tty_driver.h
+++ b/include/linux/tty_driver.h
@@ -224,6 +224,12 @@
  *	unless the tty also has a valid tty->termiox pointer.
  *
  *	Optional: Called under the termios lock
+ *
+ * int (*get_icount)(struct tty_struct *tty, struct serial_icounter *icount);
+ *
+ *	Called when the device receives a TIOCGICOUNT ioctl. Passed a kernel
+ *	structure to complete. This method is optional and will only be called
+ *	if provided (otherwise EINVAL will be returned).
  */
 
 #include <linux/fs.h>
@@ -232,6 +238,7 @@
 
 struct tty_struct;
 struct tty_driver;
+struct serial_icounter_struct;
 
 struct tty_operations {
 	struct tty_struct * (*lookup)(struct tty_driver *driver,
@@ -268,6 +275,8 @@ struct tty_operations {
 			unsigned int set, unsigned int clear);
 	int (*resize)(struct tty_struct *tty, struct winsize *ws);
 	int (*set_termiox)(struct tty_struct *tty, struct termiox *tnew);
+	int (*get_icount)(struct tty_struct *tty,
+				struct serial_icounter_struct *icount);
 #ifdef CONFIG_CONSOLE_POLL
 	int (*poll_init)(struct tty_driver *driver, int line, char *options);
 	int (*poll_get_char)(struct tty_driver *driver, int line);
diff --git a/include/linux/usb/serial.h b/include/linux/usb/serial.h
index 55675b1efb28..16d682f4f7c3 100644
--- a/include/linux/usb/serial.h
+++ b/include/linux/usb/serial.h
@@ -271,6 +271,8 @@ struct usb_serial_driver {
 	int  (*tiocmget)(struct tty_struct *tty, struct file *file);
 	int  (*tiocmset)(struct tty_struct *tty, struct file *file,
 			 unsigned int set, unsigned int clear);
+	int  (*get_icount)(struct tty_struct *tty,
+			struct serial_icounter_struct *icount);
 	/* Called by the tty layer for port level work. There may or may not
 	   be an attached tty at this point */
 	void (*dtr_rts)(struct usb_serial_port *port, int on);
-- 
cgit v1.2.3


From 432c9ed22aff641039ccd400cdabf983fabc285e Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@fluxnic.net>
Date: Fri, 1 Oct 2010 00:10:44 -0400
Subject: vcs: invoke the vt update callback when /dev/vcs* is written to

A notifier chain is called whenever the vt code modifies a terminal
content, except for one case which is when the modification comes
through writes to /dev/vcs* devices.  Let's add the missing notifier
invocation at the end of vcs_write() for that case too.

Signed-off-by: Nicolas Pitre <nicolas.pitre@canonical.com>
Cc: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/char/vc_screen.c  | 2 ++
 drivers/char/vt.c         | 5 +++++
 include/linux/selection.h | 1 +
 3 files changed, 8 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/char/vc_screen.c b/drivers/char/vc_screen.c
index 6f7054e1a516..273ab44cc91d 100644
--- a/drivers/char/vc_screen.c
+++ b/drivers/char/vc_screen.c
@@ -538,6 +538,8 @@ vcs_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 	}
 	*ppos += written;
 	ret = written;
+	if (written)
+		vcs_scr_updated(vc);
 
 unlock_out:
 	release_console_sem();
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 281aada7b4a1..a8ec48ed14d9 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -4182,6 +4182,11 @@ void vcs_scr_writew(struct vc_data *vc, u16 val, u16 *org)
 	}
 }
 
+void vcs_scr_updated(struct vc_data *vc)
+{
+	notify_update(vc);
+}
+
 /*
  *	Visible symbols for modules
  */
diff --git a/include/linux/selection.h b/include/linux/selection.h
index 8cdaa1151d2e..85193aa8c1e3 100644
--- a/include/linux/selection.h
+++ b/include/linux/selection.h
@@ -39,5 +39,6 @@ extern void putconsxy(struct vc_data *vc, unsigned char *p);
 
 extern u16 vcs_scr_readw(struct vc_data *vc, const u16 *org);
 extern void vcs_scr_writew(struct vc_data *vc, u16 val, u16 *org);
+extern void vcs_scr_updated(struct vc_data *vc);
 
 #endif
-- 
cgit v1.2.3


From 54381067ed7873e6173d6fe32818a585ad667723 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <cbouatmailru@gmail.com>
Date: Fri, 1 Oct 2010 17:21:25 +0400
Subject: serial: Factor out uart_poll_timeout() from 8250 driver

Soon we will use that handy function in the altera_uart driver.

Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
Cc: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/serial/8250.c       | 14 ++++----------
 include/linux/serial_core.h |  8 ++++++++
 2 files changed, 12 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 31b8cca179cd..b586406f04ca 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -1722,12 +1722,6 @@ static void serial_unlink_irq_chain(struct uart_8250_port *up)
 	mutex_unlock(&hash_mutex);
 }
 
-/* Base timer interval for polling */
-static inline int poll_timeout(int timeout)
-{
-	return timeout > 6 ? (timeout / 2 - 2) : 1;
-}
-
 /*
  * This function is used to handle ports that do not have an
  * interrupt.  This doesn't work very well for 16450's, but gives
@@ -1742,7 +1736,7 @@ static void serial8250_timeout(unsigned long data)
 	iir = serial_in(up, UART_IIR);
 	if (!(iir & UART_IIR_NO_INT))
 		serial8250_handle_port(up);
-	mod_timer(&up->timer, jiffies + poll_timeout(up->port.timeout));
+	mod_timer(&up->timer, jiffies + uart_poll_timeout(&up->port));
 }
 
 static void serial8250_backup_timeout(unsigned long data)
@@ -1787,7 +1781,7 @@ static void serial8250_backup_timeout(unsigned long data)
 
 	/* Standard timer interval plus 0.2s to keep the port running */
 	mod_timer(&up->timer,
-		jiffies + poll_timeout(up->port.timeout) + HZ / 5);
+		jiffies + uart_poll_timeout(&up->port) + HZ / 5);
 }
 
 static unsigned int serial8250_tx_empty(struct uart_port *port)
@@ -2071,7 +2065,7 @@ static int serial8250_startup(struct uart_port *port)
 		up->timer.function = serial8250_backup_timeout;
 		up->timer.data = (unsigned long)up;
 		mod_timer(&up->timer, jiffies +
-			  poll_timeout(up->port.timeout) + HZ / 5);
+			uart_poll_timeout(port) + HZ / 5);
 	}
 
 	/*
@@ -2081,7 +2075,7 @@ static int serial8250_startup(struct uart_port *port)
 	 */
 	if (!is_real_interrupt(up->port.irq)) {
 		up->timer.data = (unsigned long)up;
-		mod_timer(&up->timer, jiffies + poll_timeout(up->port.timeout));
+		mod_timer(&up->timer, jiffies + uart_poll_timeout(port));
 	} else {
 		retval = serial_link_irq_chain(up);
 		if (retval)
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 563e23400913..ac48082f3559 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -411,6 +411,14 @@ unsigned int uart_get_baud_rate(struct uart_port *port, struct ktermios *termios
 				unsigned int max);
 unsigned int uart_get_divisor(struct uart_port *port, unsigned int baud);
 
+/* Base timer interval for polling */
+static inline int uart_poll_timeout(struct uart_port *port)
+{
+	int timeout = port->timeout;
+
+	return timeout > 6 ? (timeout / 2 - 2) : 1;
+}
+
 /*
  * Console helpers.
  */
-- 
cgit v1.2.3


From 0d426eda7c94d864ead913f7099c623521368443 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <cbouatmailru@gmail.com>
Date: Fri, 1 Oct 2010 17:21:54 +0400
Subject: altera_uart: Add support for different address strides

Some controllers implement registers with a stride, to support
those we must implement the proper IO accessors.

Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
Acked-by: Tobias Klauser <tklauser@distanz.ch>
Cc: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/serial/altera_uart.c | 58 ++++++++++++++++++++++++++++----------------
 include/linux/altera_uart.h  |  1 +
 2 files changed, 38 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/serial/altera_uart.c b/drivers/serial/altera_uart.c
index 1dfeffae5231..f1985aaf2cbc 100644
--- a/drivers/serial/altera_uart.c
+++ b/drivers/serial/altera_uart.c
@@ -82,9 +82,23 @@ struct altera_uart {
 	unsigned short imr;	/* Local IMR mirror */
 };
 
+static u32 altera_uart_readl(struct uart_port *port, int reg)
+{
+	struct altera_uart_platform_uart *platp = port->private_data;
+
+	return readl(port->membase + (reg << platp->bus_shift));
+}
+
+static void altera_uart_writel(struct uart_port *port, u32 dat, int reg)
+{
+	struct altera_uart_platform_uart *platp = port->private_data;
+
+	writel(dat, port->membase + (reg << platp->bus_shift));
+}
+
 static unsigned int altera_uart_tx_empty(struct uart_port *port)
 {
-	return (readl(port->membase + ALTERA_UART_STATUS_REG) &
+	return (altera_uart_readl(port, ALTERA_UART_STATUS_REG) &
 		ALTERA_UART_STATUS_TMT_MSK) ? TIOCSER_TEMT : 0;
 }
 
@@ -93,8 +107,7 @@ static unsigned int altera_uart_get_mctrl(struct uart_port *port)
 	struct altera_uart *pp = container_of(port, struct altera_uart, port);
 	unsigned int sigs;
 
-	sigs =
-	    (readl(port->membase + ALTERA_UART_STATUS_REG) &
+	sigs = (altera_uart_readl(port, ALTERA_UART_STATUS_REG) &
 	     ALTERA_UART_STATUS_CTS_MSK) ? TIOCM_CTS : 0;
 	sigs |= (pp->sigs & TIOCM_RTS);
 
@@ -110,7 +123,7 @@ static void altera_uart_set_mctrl(struct uart_port *port, unsigned int sigs)
 		pp->imr |= ALTERA_UART_CONTROL_RTS_MSK;
 	else
 		pp->imr &= ~ALTERA_UART_CONTROL_RTS_MSK;
-	writel(pp->imr, port->membase + ALTERA_UART_CONTROL_REG);
+	altera_uart_writel(port, pp->imr, ALTERA_UART_CONTROL_REG);
 }
 
 static void altera_uart_start_tx(struct uart_port *port)
@@ -118,7 +131,7 @@ static void altera_uart_start_tx(struct uart_port *port)
 	struct altera_uart *pp = container_of(port, struct altera_uart, port);
 
 	pp->imr |= ALTERA_UART_CONTROL_TRDY_MSK;
-	writel(pp->imr, port->membase + ALTERA_UART_CONTROL_REG);
+	altera_uart_writel(port, pp->imr, ALTERA_UART_CONTROL_REG);
 }
 
 static void altera_uart_stop_tx(struct uart_port *port)
@@ -126,7 +139,7 @@ static void altera_uart_stop_tx(struct uart_port *port)
 	struct altera_uart *pp = container_of(port, struct altera_uart, port);
 
 	pp->imr &= ~ALTERA_UART_CONTROL_TRDY_MSK;
-	writel(pp->imr, port->membase + ALTERA_UART_CONTROL_REG);
+	altera_uart_writel(port, pp->imr, ALTERA_UART_CONTROL_REG);
 }
 
 static void altera_uart_stop_rx(struct uart_port *port)
@@ -134,7 +147,7 @@ static void altera_uart_stop_rx(struct uart_port *port)
 	struct altera_uart *pp = container_of(port, struct altera_uart, port);
 
 	pp->imr &= ~ALTERA_UART_CONTROL_RRDY_MSK;
-	writel(pp->imr, port->membase + ALTERA_UART_CONTROL_REG);
+	altera_uart_writel(port, pp->imr, ALTERA_UART_CONTROL_REG);
 }
 
 static void altera_uart_break_ctl(struct uart_port *port, int break_state)
@@ -147,7 +160,7 @@ static void altera_uart_break_ctl(struct uart_port *port, int break_state)
 		pp->imr |= ALTERA_UART_CONTROL_TRBK_MSK;
 	else
 		pp->imr &= ~ALTERA_UART_CONTROL_TRBK_MSK;
-	writel(pp->imr, port->membase + ALTERA_UART_CONTROL_REG);
+	altera_uart_writel(port, pp->imr, ALTERA_UART_CONTROL_REG);
 	spin_unlock_irqrestore(&port->lock, flags);
 }
 
@@ -171,7 +184,7 @@ static void altera_uart_set_termios(struct uart_port *port,
 
 	spin_lock_irqsave(&port->lock, flags);
 	uart_update_timeout(port, termios->c_cflag, baud);
-	writel(baudclk, port->membase + ALTERA_UART_DIVISOR_REG);
+	altera_uart_writel(port, baudclk, ALTERA_UART_DIVISOR_REG);
 	spin_unlock_irqrestore(&port->lock, flags);
 }
 
@@ -181,14 +194,15 @@ static void altera_uart_rx_chars(struct altera_uart *pp)
 	unsigned char ch, flag;
 	unsigned short status;
 
-	while ((status = readl(port->membase + ALTERA_UART_STATUS_REG)) &
+	while ((status = altera_uart_readl(port, ALTERA_UART_STATUS_REG)) &
 	       ALTERA_UART_STATUS_RRDY_MSK) {
-		ch = readl(port->membase + ALTERA_UART_RXDATA_REG);
+		ch = altera_uart_readl(port, ALTERA_UART_RXDATA_REG);
 		flag = TTY_NORMAL;
 		port->icount.rx++;
 
 		if (status & ALTERA_UART_STATUS_E_MSK) {
-			writel(status, port->membase + ALTERA_UART_STATUS_REG);
+			altera_uart_writel(port, status,
+					   ALTERA_UART_STATUS_REG);
 
 			if (status & ALTERA_UART_STATUS_BRK_MSK) {
 				port->icount.brk++;
@@ -228,18 +242,18 @@ static void altera_uart_tx_chars(struct altera_uart *pp)
 
 	if (port->x_char) {
 		/* Send special char - probably flow control */
-		writel(port->x_char, port->membase + ALTERA_UART_TXDATA_REG);
+		altera_uart_writel(port, port->x_char, ALTERA_UART_TXDATA_REG);
 		port->x_char = 0;
 		port->icount.tx++;
 		return;
 	}
 
-	while (readl(port->membase + ALTERA_UART_STATUS_REG) &
+	while (altera_uart_readl(port, ALTERA_UART_STATUS_REG) &
 	       ALTERA_UART_STATUS_TRDY_MSK) {
 		if (xmit->head == xmit->tail)
 			break;
-		writel(xmit->buf[xmit->tail],
-		       port->membase + ALTERA_UART_TXDATA_REG);
+		altera_uart_writel(port, xmit->buf[xmit->tail],
+		       ALTERA_UART_TXDATA_REG);
 		xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
 		port->icount.tx++;
 	}
@@ -249,7 +263,7 @@ static void altera_uart_tx_chars(struct altera_uart *pp)
 
 	if (xmit->head == xmit->tail) {
 		pp->imr &= ~ALTERA_UART_CONTROL_TRDY_MSK;
-		writel(pp->imr, port->membase + ALTERA_UART_CONTROL_REG);
+		altera_uart_writel(port, pp->imr, ALTERA_UART_CONTROL_REG);
 	}
 }
 
@@ -259,7 +273,7 @@ static irqreturn_t altera_uart_interrupt(int irq, void *data)
 	struct altera_uart *pp = container_of(port, struct altera_uart, port);
 	unsigned int isr;
 
-	isr = readl(port->membase + ALTERA_UART_STATUS_REG) & pp->imr;
+	isr = altera_uart_readl(port, ALTERA_UART_STATUS_REG) & pp->imr;
 
 	spin_lock(&port->lock);
 	if (isr & ALTERA_UART_STATUS_RRDY_MSK)
@@ -285,9 +299,9 @@ static void altera_uart_config_port(struct uart_port *port, int flags)
 	port->type = PORT_ALTERA_UART;
 
 	/* Clear mask, so no surprise interrupts. */
-	writel(0, port->membase + ALTERA_UART_CONTROL_REG);
+	altera_uart_writel(port, 0, ALTERA_UART_CONTROL_REG);
 	/* Clear status register */
-	writel(0, port->membase + ALTERA_UART_STATUS_REG);
+	altera_uart_writel(port, 0, ALTERA_UART_STATUS_REG);
 }
 
 static int altera_uart_startup(struct uart_port *port)
@@ -407,6 +421,7 @@ int __init early_altera_uart_setup(struct altera_uart_platform_uart *platp)
 		port->uartclk = platp[i].uartclk;
 		port->flags = ASYNC_BOOT_AUTOCONF;
 		port->ops = &altera_uart_ops;
+		port->private_data = platp;
 	}
 
 	return 0;
@@ -414,7 +429,7 @@ int __init early_altera_uart_setup(struct altera_uart_platform_uart *platp)
 
 static void altera_uart_console_putc(struct uart_port *port, const char c)
 {
-	while (!(readl(port->membase + ALTERA_UART_STATUS_REG) &
+	while (!(altera_uart_readl(port, ALTERA_UART_STATUS_REG) &
 		 ALTERA_UART_STATUS_TRDY_MSK))
 		cpu_relax();
 
@@ -535,6 +550,7 @@ static int __devinit altera_uart_probe(struct platform_device *pdev)
 	port->uartclk = platp->uartclk;
 	port->ops = &altera_uart_ops;
 	port->flags = ASYNC_BOOT_AUTOCONF;
+	port->private_data = platp;
 
 	uart_add_one_port(&altera_uart_driver, port);
 
diff --git a/include/linux/altera_uart.h b/include/linux/altera_uart.h
index 8d441064a30d..c022c82db7ca 100644
--- a/include/linux/altera_uart.h
+++ b/include/linux/altera_uart.h
@@ -9,6 +9,7 @@ struct altera_uart_platform_uart {
 	unsigned long mapbase;	/* Physical address base */
 	unsigned int irq;	/* Interrupt vector */
 	unsigned int uartclk;	/* UART clock rate */
+	unsigned int bus_shift;	/* Bus shift (address stride) */
 };
 
 #endif /* __ALTUART_H */
-- 
cgit v1.2.3


From 5d89a48acfbaae02e7ecf97d4d8cc570a31964c5 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <cbouatmailru@gmail.com>
Date: Fri, 1 Oct 2010 17:22:55 +0400
Subject: altera_uart: Fix missing prototype for registering an early console

Simply add an early_altera_uart_setup() prototype declaration, otherwise
platform code have to do it in .c files, which is ugly.

Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
Acked-by: Tobias Klauser <tklauser@distanz.ch>
Cc: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/altera_uart.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/altera_uart.h b/include/linux/altera_uart.h
index c022c82db7ca..a10a90791976 100644
--- a/include/linux/altera_uart.h
+++ b/include/linux/altera_uart.h
@@ -5,6 +5,8 @@
 #ifndef	__ALTUART_H
 #define	__ALTUART_H
 
+#include <linux/init.h>
+
 struct altera_uart_platform_uart {
 	unsigned long mapbase;	/* Physical address base */
 	unsigned int irq;	/* Interrupt vector */
@@ -12,4 +14,6 @@ struct altera_uart_platform_uart {
 	unsigned int bus_shift;	/* Bus shift (address stride) */
 };
 
+int __init early_altera_uart_setup(struct altera_uart_platform_uart *platp);
+
 #endif /* __ALTUART_H */
-- 
cgit v1.2.3


From c161afe9759ddcc174d08e7c4f683d08ac9ba86f Mon Sep 17 00:00:00 2001
From: Manuel Lauss <manuel.lauss@googlemail.com>
Date: Sat, 25 Sep 2010 15:13:45 +0200
Subject: 8250: allow platforms to override PM hook.

Add a hook for platforms to specify custom pm methods.

Signed-off-by: Manuel Lauss <manuel.lauss@googlemail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/serial/8250.c       | 27 ++++++++++++++++-----------
 include/linux/serial_8250.h |  4 ++++
 include/linux/serial_core.h |  2 ++
 3 files changed, 22 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index b586406f04ca..6994afb5571a 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -154,12 +154,6 @@ struct uart_8250_port {
 	unsigned char		lsr_saved_flags;
 #define MSR_SAVE_FLAGS UART_MSR_ANY_DELTA
 	unsigned char		msr_saved_flags;
-
-	/*
-	 * We provide a per-port pm hook.
-	 */
-	void			(*pm)(struct uart_port *port,
-				      unsigned int state, unsigned int old);
 };
 
 struct irq_info {
@@ -2436,16 +2430,24 @@ serial8250_set_ldisc(struct uart_port *port, int new)
 		port->flags &= ~UPF_HARDPPS_CD;
 }
 
-static void
-serial8250_pm(struct uart_port *port, unsigned int state,
-	      unsigned int oldstate)
+
+void serial8250_do_pm(struct uart_port *port, unsigned int state,
+		      unsigned int oldstate)
 {
 	struct uart_8250_port *p = (struct uart_8250_port *)port;
 
 	serial8250_set_sleep(p, state != 0);
+}
+EXPORT_SYMBOL(serial8250_do_pm);
 
-	if (p->pm)
-		p->pm(port, state, oldstate);
+static void
+serial8250_pm(struct uart_port *port, unsigned int state,
+	      unsigned int oldstate)
+{
+	if (port->pm)
+		port->pm(port, state, oldstate);
+	else
+		serial8250_do_pm(port, state, oldstate);
 }
 
 static unsigned int serial8250_port_size(struct uart_8250_port *pt)
@@ -3006,6 +3008,7 @@ static int __devinit serial8250_probe(struct platform_device *dev)
 		port.serial_in		= p->serial_in;
 		port.serial_out		= p->serial_out;
 		port.set_termios	= p->set_termios;
+		port.pm			= p->pm;
 		port.dev		= &dev->dev;
 		port.irqflags		|= irqflag;
 		ret = serial8250_register_port(&port);
@@ -3172,6 +3175,8 @@ int serial8250_register_port(struct uart_port *port)
 		/*  Possibly override set_termios call */
 		if (port->set_termios)
 			uart->port.set_termios = port->set_termios;
+		if (port->pm)
+			uart->port.pm = port->pm;
 
 		ret = uart_add_one_port(&serial8250_reg, &uart->port);
 		if (ret == 0)
diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h
index 7638deaaba65..bf9c2bdb2e05 100644
--- a/include/linux/serial_8250.h
+++ b/include/linux/serial_8250.h
@@ -35,6 +35,8 @@ struct plat_serial8250_port {
 	void		(*set_termios)(struct uart_port *,
 			               struct ktermios *new,
 			               struct ktermios *old);
+	void		(*pm)(struct uart_port *, unsigned int state,
+			      unsigned old);
 };
 
 /*
@@ -76,5 +78,7 @@ extern int serial8250_find_port_for_earlycon(void);
 extern int setup_early_serial8250_console(char *cmdline);
 extern void serial8250_do_set_termios(struct uart_port *port,
 		struct ktermios *termios, struct ktermios *old);
+extern void serial8250_do_pm(struct uart_port *port, unsigned int state,
+			     unsigned int oldstate);
 
 #endif
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index ac48082f3559..99e5994e6f84 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -289,6 +289,8 @@ struct uart_port {
 	void			(*set_termios)(struct uart_port *,
 				               struct ktermios *new,
 				               struct ktermios *old);
+	void			(*pm)(struct uart_port *, unsigned int state,
+				      unsigned int old);
 	unsigned int		irq;			/* irq number */
 	unsigned long		irqflags;		/* irq flags  */
 	unsigned int		uartclk;		/* base uart clock */
-- 
cgit v1.2.3


From af7f3743567e3d5b40e2f9c21541b7f40b99c103 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Mon, 18 Oct 2010 11:38:02 -0700
Subject: serial: abstraction for 8250 legacy ports

Not every platform that has generic legacy 8250 ports manages to have them
clocked the right way or without errata. Provide a generic interface to
allow platforms to override the default behaviour in a manner that dumps
the complexity in *their* code not the 8250 driver.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dirk Brandewie <dirk.brandewie@gmail.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/serial/8250.c       | 17 +++++++++++++++++
 include/linux/serial_8250.h |  4 ++++
 2 files changed, 21 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 6994afb5571a..37f19a678c97 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -2672,6 +2672,16 @@ static struct uart_ops serial8250_pops = {
 
 static struct uart_8250_port serial8250_ports[UART_NR];
 
+static void (*serial8250_isa_config)(int port, struct uart_port *up,
+	unsigned short *capabilities);
+
+void serial8250_set_isa_configurator(
+	void (*v)(int port, struct uart_port *up, unsigned short *capabilities))
+{
+	serial8250_isa_config = v;
+}
+EXPORT_SYMBOL(serial8250_set_isa_configurator);
+
 static void __init serial8250_isa_init_ports(void)
 {
 	struct uart_8250_port *up;
@@ -2717,6 +2727,9 @@ static void __init serial8250_isa_init_ports(void)
 		up->port.regshift = old_serial_port[i].iomem_reg_shift;
 		set_io_from_upio(&up->port);
 		up->port.irqflags |= irqflag;
+		if (serial8250_isa_config != NULL)
+			serial8250_isa_config(i, &up->port, &up->capabilities);
+
 	}
 }
 
@@ -3178,6 +3191,10 @@ int serial8250_register_port(struct uart_port *port)
 		if (port->pm)
 			uart->port.pm = port->pm;
 
+		if (serial8250_isa_config != NULL)
+			serial8250_isa_config(0, &uart->port,
+					&uart->capabilities);
+
 		ret = uart_add_one_port(&serial8250_reg, &uart->port);
 		if (ret == 0)
 			ret = uart->port.line;
diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h
index bf9c2bdb2e05..97f5b45bbc07 100644
--- a/include/linux/serial_8250.h
+++ b/include/linux/serial_8250.h
@@ -81,4 +81,8 @@ extern void serial8250_do_set_termios(struct uart_port *port,
 extern void serial8250_do_pm(struct uart_port *port, unsigned int state,
 			     unsigned int oldstate);
 
+extern void serial8250_set_isa_configurator(void (*v)
+					(int port, struct uart_port *up,
+						unsigned short *capabilities));
+
 #endif
-- 
cgit v1.2.3


From f0ae849df1cd6b597130d890f2107ee31bf02c19 Mon Sep 17 00:00:00 2001
From: Hao Wu <hao.wu@intel.com>
Date: Thu, 5 Aug 2010 14:17:28 +0100
Subject: usb: Add Intel Langwell USB OTG Transceiver Driver

This adds support for the USB transceiver driver in the Langwell chipset used
on the Intel MID platforms. It folds up the original patch set which includes
basic support for the device, PHY low power mode (Please notice that there is
a limitation, after we drive VBus down, 2ms delay is required from SCU FW to
sync up OTGSC register with USBCFG register), software timers (the hardware
timers do not work in low power mode), HNP, SRP.


Signed-off-by: Hao Wu <hao.wu@intel.com>
Signed-off-by: Alek Du <alek.du@intel.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/otg/Kconfig          |   14 +
 drivers/usb/otg/Makefile         |    1 +
 drivers/usb/otg/langwell_otg.c   | 2408 ++++++++++++++++++++++++++++++++++++++
 include/linux/usb/langwell_otg.h |  139 +++
 4 files changed, 2562 insertions(+)
 create mode 100644 drivers/usb/otg/langwell_otg.c
 create mode 100644 include/linux/usb/langwell_otg.h

(limited to 'include/linux')

diff --git a/drivers/usb/otg/Kconfig b/drivers/usb/otg/Kconfig
index 3b1289572d72..299dfd2510cb 100644
--- a/drivers/usb/otg/Kconfig
+++ b/drivers/usb/otg/Kconfig
@@ -67,4 +67,18 @@ config NOP_USB_XCEIV
 	 built-in with usb ip or which are autonomous and doesn't require any
 	 phy programming such as ISP1x04 etc.
 
+config USB_LANGWELL_OTG
+	tristate "Intel Langwell USB OTG dual-role support"
+	depends on USB && X86_MRST
+	select USB_OTG
+	select USB_OTG_UTILS
+	help
+	  Say Y here if you want to build Intel Langwell USB OTG
+	  transciever driver in kernel. This driver implements role
+	  switch between EHCI host driver and Langwell USB OTG
+	  client driver.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called langwell_otg.
+
 endif # USB || OTG
diff --git a/drivers/usb/otg/Makefile b/drivers/usb/otg/Makefile
index aeb49a8ec412..b6609db3a849 100644
--- a/drivers/usb/otg/Makefile
+++ b/drivers/usb/otg/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_USB_OTG_UTILS)	+= otg.o
 obj-$(CONFIG_USB_GPIO_VBUS)	+= gpio_vbus.o
 obj-$(CONFIG_ISP1301_OMAP)	+= isp1301_omap.o
 obj-$(CONFIG_TWL4030_USB)	+= twl4030-usb.o
+obj-$(CONFIG_USB_LANGWELL_OTG)	+= langwell_otg.o
 obj-$(CONFIG_NOP_USB_XCEIV)	+= nop-usb-xceiv.o
 obj-$(CONFIG_USB_ULPI)		+= ulpi.o
 
diff --git a/drivers/usb/otg/langwell_otg.c b/drivers/usb/otg/langwell_otg.c
new file mode 100644
index 000000000000..879188086daf
--- /dev/null
+++ b/drivers/usb/otg/langwell_otg.c
@@ -0,0 +1,2408 @@
+/*
+ * Intel Langwell USB OTG transceiver driver
+ * Copyright (C) 2008 - 2009, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope 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.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ */
+/* This driver helps to switch Langwell OTG controller function between host
+ * and peripheral. It works with EHCI driver and Langwell client controller
+ * driver together.
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/moduleparam.h>
+#include <linux/usb/ch9.h>
+#include <linux/usb/gadget.h>
+#include <linux/usb.h>
+#include <linux/usb/otg.h>
+#include <linux/usb/hcd.h>
+#include <linux/notifier.h>
+#include <linux/delay.h>
+#include <asm/intel_scu_ipc.h>
+
+#include <linux/usb/langwell_otg.h>
+
+#define	DRIVER_DESC		"Intel Langwell USB OTG transceiver driver"
+#define	DRIVER_VERSION		"3.0.0.32L.0003"
+
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_AUTHOR("Henry Yuan <hang.yuan@intel.com>, Hao Wu <hao.wu@intel.com>");
+MODULE_VERSION(DRIVER_VERSION);
+MODULE_LICENSE("GPL");
+
+static const char driver_name[] = "langwell_otg";
+
+static int langwell_otg_probe(struct pci_dev *pdev,
+			const struct pci_device_id *id);
+static void langwell_otg_remove(struct pci_dev *pdev);
+static int langwell_otg_suspend(struct pci_dev *pdev, pm_message_t message);
+static int langwell_otg_resume(struct pci_dev *pdev);
+
+static int langwell_otg_set_host(struct otg_transceiver *otg,
+				struct usb_bus *host);
+static int langwell_otg_set_peripheral(struct otg_transceiver *otg,
+				struct usb_gadget *gadget);
+static int langwell_otg_start_srp(struct otg_transceiver *otg);
+
+static const struct pci_device_id pci_ids[] = {{
+	.class =        ((PCI_CLASS_SERIAL_USB << 8) | 0xfe),
+	.class_mask =   ~0,
+	.vendor =	0x8086,
+	.device =	0x0811,
+	.subvendor =	PCI_ANY_ID,
+	.subdevice =	PCI_ANY_ID,
+}, { /* end: all zeroes */ }
+};
+
+static struct pci_driver otg_pci_driver = {
+	.name =		(char *) driver_name,
+	.id_table =	pci_ids,
+
+	.probe =	langwell_otg_probe,
+	.remove =	langwell_otg_remove,
+
+	.suspend =	langwell_otg_suspend,
+	.resume =	langwell_otg_resume,
+};
+
+static const char *state_string(enum usb_otg_state state)
+{
+	switch (state) {
+	case OTG_STATE_A_IDLE:
+		return "a_idle";
+	case OTG_STATE_A_WAIT_VRISE:
+		return "a_wait_vrise";
+	case OTG_STATE_A_WAIT_BCON:
+		return "a_wait_bcon";
+	case OTG_STATE_A_HOST:
+		return "a_host";
+	case OTG_STATE_A_SUSPEND:
+		return "a_suspend";
+	case OTG_STATE_A_PERIPHERAL:
+		return "a_peripheral";
+	case OTG_STATE_A_WAIT_VFALL:
+		return "a_wait_vfall";
+	case OTG_STATE_A_VBUS_ERR:
+		return "a_vbus_err";
+	case OTG_STATE_B_IDLE:
+		return "b_idle";
+	case OTG_STATE_B_SRP_INIT:
+		return "b_srp_init";
+	case OTG_STATE_B_PERIPHERAL:
+		return "b_peripheral";
+	case OTG_STATE_B_WAIT_ACON:
+		return "b_wait_acon";
+	case OTG_STATE_B_HOST:
+		return "b_host";
+	default:
+		return "UNDEFINED";
+	}
+}
+
+/* HSM timers */
+static inline struct langwell_otg_timer *otg_timer_initializer
+(void (*function)(unsigned long), unsigned long expires, unsigned long data)
+{
+	struct langwell_otg_timer *timer;
+	timer = kmalloc(sizeof(struct langwell_otg_timer), GFP_KERNEL);
+	if (timer == NULL)
+		return timer;
+
+	timer->function = function;
+	timer->expires = expires;
+	timer->data = data;
+	return timer;
+}
+
+static struct langwell_otg_timer *a_wait_vrise_tmr, *a_aidl_bdis_tmr,
+	*b_se0_srp_tmr, *b_srp_init_tmr;
+
+static struct list_head active_timers;
+
+static struct langwell_otg *the_transceiver;
+
+/* host/client notify transceiver when event affects HNP state */
+void langwell_update_transceiver(void)
+{
+	struct langwell_otg *lnw = the_transceiver;
+
+	dev_dbg(lnw->dev, "transceiver is updated\n");
+
+	if (!lnw->qwork)
+		return ;
+
+	queue_work(lnw->qwork, &lnw->work);
+}
+EXPORT_SYMBOL(langwell_update_transceiver);
+
+static int langwell_otg_set_host(struct otg_transceiver *otg,
+					struct usb_bus *host)
+{
+	otg->host = host;
+
+	return 0;
+}
+
+static int langwell_otg_set_peripheral(struct otg_transceiver *otg,
+					struct usb_gadget *gadget)
+{
+	otg->gadget = gadget;
+
+	return 0;
+}
+
+static int langwell_otg_set_power(struct otg_transceiver *otg,
+				unsigned mA)
+{
+	return 0;
+}
+
+/* A-device drives vbus, controlled through PMIC CHRGCNTL register*/
+static int langwell_otg_set_vbus(struct otg_transceiver *otg, bool enabled)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	u8 r;
+
+	dev_dbg(lnw->dev, "%s <--- %s\n", __func__, enabled ? "on" : "off");
+
+	/* FIXME: surely we should cache this on the first read. If not use
+	   readv to avoid two transactions */
+	if (intel_scu_ipc_ioread8(0x00, &r) < 0) {
+		dev_dbg(lnw->dev, "Failed to read PMIC register 0xD2");
+		return -EBUSY;
+	}
+	if ((r & 0x03) != 0x02) {
+		dev_dbg(lnw->dev, "not NEC PMIC attached\n");
+		return -EBUSY;
+	}
+
+	if (intel_scu_ipc_ioread8(0x20, &r) < 0) {
+		dev_dbg(lnw->dev, "Failed to read PMIC register 0xD2");
+		return -EBUSY;
+	}
+
+	if ((r & 0x20) == 0) {
+		dev_dbg(lnw->dev, "no battery attached\n");
+		return -EBUSY;
+	}
+
+	/* Workaround for battery attachment issue */
+	if (r == 0x34) {
+		dev_dbg(lnw->dev, "no battery attached on SH\n");
+		return -EBUSY;
+	}
+
+	dev_dbg(lnw->dev, "battery attached. 2 reg = %x\n", r);
+
+	/* workaround: FW detect writing 0x20/0xc0 to d4 event.
+	 * this is only for NEC PMIC.
+	 */
+
+	if (intel_scu_ipc_iowrite8(0xD4, enabled ? 0x20 : 0xC0))
+		dev_dbg(lnw->dev, "Failed to write PMIC.\n");
+
+	dev_dbg(lnw->dev, "%s --->\n", __func__);
+
+	return 0;
+}
+
+/* charge vbus or discharge vbus through a resistor to ground */
+static void langwell_otg_chrg_vbus(int on)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	u32	val;
+
+	val = readl(lnw->iotg.base + CI_OTGSC);
+
+	if (on)
+		writel((val & ~OTGSC_INTSTS_MASK) | OTGSC_VC,
+				lnw->iotg.base + CI_OTGSC);
+	else
+		writel((val & ~OTGSC_INTSTS_MASK) | OTGSC_VD,
+				lnw->iotg.base + CI_OTGSC);
+}
+
+/* Start SRP */
+static int langwell_otg_start_srp(struct otg_transceiver *otg)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	u32				val;
+
+	dev_dbg(lnw->dev, "%s --->\n", __func__);
+
+	val = readl(iotg->base + CI_OTGSC);
+
+	writel((val & ~OTGSC_INTSTS_MASK) | OTGSC_HADP,
+				iotg->base + CI_OTGSC);
+
+	/* Check if the data plus is finished or not */
+	msleep(8);
+	val = readl(iotg->base + CI_OTGSC);
+	if (val & (OTGSC_HADP | OTGSC_DP))
+		dev_dbg(lnw->dev, "DataLine SRP Error\n");
+
+	/* Disable interrupt - b_sess_vld */
+	val = readl(iotg->base + CI_OTGSC);
+	val &= (~(OTGSC_BSVIE | OTGSC_BSEIE));
+	writel(val, iotg->base + CI_OTGSC);
+
+	/* Start VBus SRP, drive vbus to generate VBus pulse */
+	iotg->otg.set_vbus(&iotg->otg, true);
+	msleep(15);
+	iotg->otg.set_vbus(&iotg->otg, false);
+
+	/* Enable interrupt - b_sess_vld*/
+	val = readl(iotg->base + CI_OTGSC);
+	dev_dbg(lnw->dev, "after VBUS pulse otgsc = %x\n", val);
+
+	val |= (OTGSC_BSVIE | OTGSC_BSEIE);
+	writel(val, iotg->base + CI_OTGSC);
+
+	/* If Vbus is valid, then update the hsm */
+	if (val & OTGSC_BSV) {
+		dev_dbg(lnw->dev, "no b_sess_vld interrupt\n");
+
+		lnw->iotg.hsm.b_sess_vld = 1;
+		langwell_update_transceiver();
+	}
+
+	dev_dbg(lnw->dev, "%s <---\n", __func__);
+	return 0;
+}
+
+/* stop SOF via bus_suspend */
+static void langwell_otg_loc_sof(int on)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	struct usb_hcd		*hcd;
+	int			err;
+
+	dev_dbg(lnw->dev, "%s ---> %s\n", __func__, on ? "suspend" : "resume");
+
+	hcd = bus_to_hcd(lnw->iotg.otg.host);
+	if (on)
+		err = hcd->driver->bus_resume(hcd);
+	else
+		err = hcd->driver->bus_suspend(hcd);
+
+	if (err)
+		dev_dbg(lnw->dev, "Fail to resume/suspend USB bus - %d\n", err);
+
+	dev_dbg(lnw->dev, "%s <---\n", __func__);
+}
+
+static int langwell_otg_check_otgsc(void)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	u32				otgsc, usbcfg;
+
+	dev_dbg(lnw->dev, "check sync OTGSC and USBCFG registers\n");
+
+	otgsc = readl(lnw->iotg.base + CI_OTGSC);
+	usbcfg = readl(lnw->usbcfg);
+
+	dev_dbg(lnw->dev, "OTGSC = %08x, USBCFG = %08x\n",
+					otgsc, usbcfg);
+	dev_dbg(lnw->dev, "OTGSC_AVV = %d\n", !!(otgsc & OTGSC_AVV));
+	dev_dbg(lnw->dev, "USBCFG.VBUSVAL = %d\n",
+					!!(usbcfg & USBCFG_VBUSVAL));
+	dev_dbg(lnw->dev, "OTGSC_ASV = %d\n", !!(otgsc & OTGSC_ASV));
+	dev_dbg(lnw->dev, "USBCFG.AVALID = %d\n",
+					!!(usbcfg & USBCFG_AVALID));
+	dev_dbg(lnw->dev, "OTGSC_BSV = %d\n", !!(otgsc & OTGSC_BSV));
+	dev_dbg(lnw->dev, "USBCFG.BVALID = %d\n",
+					!!(usbcfg & USBCFG_BVALID));
+	dev_dbg(lnw->dev, "OTGSC_BSE = %d\n", !!(otgsc & OTGSC_BSE));
+	dev_dbg(lnw->dev, "USBCFG.SESEND = %d\n",
+					!!(usbcfg & USBCFG_SESEND));
+
+	/* Check USBCFG VBusValid/AValid/BValid/SessEnd */
+	if (!!(otgsc & OTGSC_AVV) ^ !!(usbcfg & USBCFG_VBUSVAL)) {
+		dev_dbg(lnw->dev, "OTGSC.AVV != USBCFG.VBUSVAL\n");
+		goto err;
+	}
+	if (!!(otgsc & OTGSC_ASV) ^ !!(usbcfg & USBCFG_AVALID)) {
+		dev_dbg(lnw->dev, "OTGSC.ASV != USBCFG.AVALID\n");
+		goto err;
+	}
+	if (!!(otgsc & OTGSC_BSV) ^ !!(usbcfg & USBCFG_BVALID)) {
+		dev_dbg(lnw->dev, "OTGSC.BSV != USBCFG.BVALID\n");
+		goto err;
+	}
+	if (!!(otgsc & OTGSC_BSE) ^ !!(usbcfg & USBCFG_SESEND)) {
+		dev_dbg(lnw->dev, "OTGSC.BSE != USBCFG.SESSEN\n");
+		goto err;
+	}
+
+	dev_dbg(lnw->dev, "OTGSC and USBCFG are synced\n");
+
+	return 0;
+
+err:
+	dev_warn(lnw->dev, "OTGSC isn't equal to USBCFG\n");
+	return -EPIPE;
+}
+
+
+static void langwell_otg_phy_low_power(int on)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	u8				val, phcd;
+	int				retval;
+
+	dev_dbg(lnw->dev, "%s ---> %s mode\n",
+			__func__, on ? "Low power" : "Normal");
+
+	phcd = 0x40;
+
+	val = readb(iotg->base + CI_HOSTPC1 + 2);
+
+	if (on) {
+		/* Due to hardware issue, after set PHCD, sync will failed
+		 * between USBCFG and OTGSC, so before set PHCD, check if
+		 * sync is in process now. If the answer is "yes", then do
+		 * not touch PHCD bit */
+		retval = langwell_otg_check_otgsc();
+		if (retval) {
+			dev_dbg(lnw->dev, "Skip PHCD programming..\n");
+			return ;
+		}
+
+		writeb(val | phcd, iotg->base + CI_HOSTPC1 + 2);
+	} else
+		writeb(val & ~phcd, iotg->base + CI_HOSTPC1 + 2);
+
+	dev_dbg(lnw->dev, "%s <--- done\n", __func__);
+}
+
+/* After drv vbus, add 2 ms delay to set PHCD */
+static void langwell_otg_phy_low_power_wait(int on)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+
+	dev_dbg(lnw->dev, "add 2ms delay before programing PHCD\n");
+
+	mdelay(2);
+	langwell_otg_phy_low_power(on);
+}
+
+/* Enable/Disable OTG interrupt */
+static void langwell_otg_intr(int on)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	u32				val;
+
+	dev_dbg(lnw->dev, "%s ---> %s\n", __func__, on ? "on" : "off");
+
+	val = readl(iotg->base + CI_OTGSC);
+
+	/* OTGSC_INT_MASK doesn't contains 1msInt */
+	if (on) {
+		val = val | (OTGSC_INT_MASK);
+		writel(val, iotg->base + CI_OTGSC);
+	} else {
+		val = val & ~(OTGSC_INT_MASK);
+		writel(val, iotg->base + CI_OTGSC);
+	}
+
+	dev_dbg(lnw->dev, "%s <---\n", __func__);
+}
+
+/* set HAAR: Hardware Assist Auto-Reset */
+static void langwell_otg_HAAR(int on)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	u32				val;
+
+	dev_dbg(lnw->dev, "%s ---> %s\n", __func__, on ? "on" : "off");
+
+	val = readl(iotg->base + CI_OTGSC);
+	if (on)
+		writel((val & ~OTGSC_INTSTS_MASK) | OTGSC_HAAR,
+					iotg->base + CI_OTGSC);
+	else
+		writel((val & ~OTGSC_INTSTS_MASK) & ~OTGSC_HAAR,
+					iotg->base + CI_OTGSC);
+
+	dev_dbg(lnw->dev, "%s <---\n", __func__);
+}
+
+/* set HABA: Hardware Assist B-Disconnect to A-Connect */
+static void langwell_otg_HABA(int on)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	u32				val;
+
+	dev_dbg(lnw->dev, "%s ---> %s\n", __func__, on ? "on" : "off");
+
+	val = readl(iotg->base + CI_OTGSC);
+	if (on)
+		writel((val & ~OTGSC_INTSTS_MASK) | OTGSC_HABA,
+					iotg->base + CI_OTGSC);
+	else
+		writel((val & ~OTGSC_INTSTS_MASK) & ~OTGSC_HABA,
+					iotg->base + CI_OTGSC);
+
+	dev_dbg(lnw->dev, "%s <---\n", __func__);
+}
+
+static int langwell_otg_check_se0_srp(int on)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	int			delay_time = TB_SE0_SRP * 10;
+	u32			val;
+
+	dev_dbg(lnw->dev, "%s --->\n", __func__);
+
+	do {
+		udelay(100);
+		if (!delay_time--)
+			break;
+		val = readl(lnw->iotg.base + CI_PORTSC1);
+		val &= PORTSC_LS;
+	} while (!val);
+
+	dev_dbg(lnw->dev, "%s <---\n", __func__);
+	return val;
+}
+
+/* The timeout callback function to set time out bit */
+static void set_tmout(unsigned long indicator)
+{
+	*(int *)indicator = 1;
+}
+
+void langwell_otg_nsf_msg(unsigned long indicator)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+
+	switch (indicator) {
+	case 2:
+	case 4:
+	case 6:
+	case 7:
+		dev_warn(lnw->dev,
+			"OTG:NSF-%lu - deivce not responding\n", indicator);
+		break;
+	case 3:
+		dev_warn(lnw->dev,
+			"OTG:NSF-%lu - deivce not supported\n", indicator);
+		break;
+	default:
+		dev_warn(lnw->dev, "Do not have this kind of NSF\n");
+		break;
+	}
+}
+
+/* Initialize timers */
+static int langwell_otg_init_timers(struct otg_hsm *hsm)
+{
+	/* HSM used timers */
+	a_wait_vrise_tmr = otg_timer_initializer(&set_tmout, TA_WAIT_VRISE,
+				(unsigned long)&hsm->a_wait_vrise_tmout);
+	if (a_wait_vrise_tmr == NULL)
+		return -ENOMEM;
+	a_aidl_bdis_tmr = otg_timer_initializer(&set_tmout, TA_AIDL_BDIS,
+				(unsigned long)&hsm->a_aidl_bdis_tmout);
+	if (a_aidl_bdis_tmr == NULL)
+		return -ENOMEM;
+	b_se0_srp_tmr = otg_timer_initializer(&set_tmout, TB_SE0_SRP,
+				(unsigned long)&hsm->b_se0_srp);
+	if (b_se0_srp_tmr == NULL)
+		return -ENOMEM;
+	b_srp_init_tmr = otg_timer_initializer(&set_tmout, TB_SRP_INIT,
+				(unsigned long)&hsm->b_srp_init_tmout);
+	if (b_srp_init_tmr == NULL)
+		return -ENOMEM;
+
+	return 0;
+}
+
+/* Free timers */
+static void langwell_otg_free_timers(void)
+{
+	kfree(a_wait_vrise_tmr);
+	kfree(a_aidl_bdis_tmr);
+	kfree(b_se0_srp_tmr);
+	kfree(b_srp_init_tmr);
+}
+
+/* The timeout callback function to set time out bit */
+static void langwell_otg_timer_fn(unsigned long indicator)
+{
+	struct langwell_otg *lnw = the_transceiver;
+
+	*(int *)indicator = 1;
+
+	dev_dbg(lnw->dev, "kernel timer - timeout\n");
+
+	langwell_update_transceiver();
+}
+
+/* kernel timer used instead of HW based interrupt */
+static void langwell_otg_add_ktimer(enum langwell_otg_timer_type timers)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	unsigned long		j = jiffies;
+	unsigned long		data, time;
+
+	switch (timers) {
+	case TA_WAIT_VRISE_TMR:
+		iotg->hsm.a_wait_vrise_tmout = 0;
+		data = (unsigned long)&iotg->hsm.a_wait_vrise_tmout;
+		time = TA_WAIT_VRISE;
+		break;
+	case TA_WAIT_BCON_TMR:
+		iotg->hsm.a_wait_bcon_tmout = 0;
+		data = (unsigned long)&iotg->hsm.a_wait_bcon_tmout;
+		time = TA_WAIT_BCON;
+		break;
+	case TA_AIDL_BDIS_TMR:
+		iotg->hsm.a_aidl_bdis_tmout = 0;
+		data = (unsigned long)&iotg->hsm.a_aidl_bdis_tmout;
+		time = TA_AIDL_BDIS;
+		break;
+	case TB_ASE0_BRST_TMR:
+		iotg->hsm.b_ase0_brst_tmout = 0;
+		data = (unsigned long)&iotg->hsm.b_ase0_brst_tmout;
+		time = TB_ASE0_BRST;
+		break;
+	case TB_SRP_INIT_TMR:
+		iotg->hsm.b_srp_init_tmout = 0;
+		data = (unsigned long)&iotg->hsm.b_srp_init_tmout;
+		time = TB_SRP_INIT;
+		break;
+	case TB_SRP_FAIL_TMR:
+		iotg->hsm.b_srp_fail_tmout = 0;
+		data = (unsigned long)&iotg->hsm.b_srp_fail_tmout;
+		time = TB_SRP_FAIL;
+		break;
+	case TB_BUS_SUSPEND_TMR:
+		iotg->hsm.b_bus_suspend_tmout = 0;
+		data = (unsigned long)&iotg->hsm.b_bus_suspend_tmout;
+		time = TB_BUS_SUSPEND;
+		break;
+	default:
+		dev_dbg(lnw->dev, "unkown timer, cannot enable it\n");
+		return;
+	}
+
+	lnw->hsm_timer.data = data;
+	lnw->hsm_timer.function = langwell_otg_timer_fn;
+	lnw->hsm_timer.expires = j + time * HZ / 1000; /* milliseconds */
+
+	add_timer(&lnw->hsm_timer);
+
+	dev_dbg(lnw->dev, "add timer successfully\n");
+}
+
+/* Add timer to timer list */
+static void langwell_otg_add_timer(void *gtimer)
+{
+	struct langwell_otg_timer *timer = (struct langwell_otg_timer *)gtimer;
+	struct langwell_otg_timer *tmp_timer;
+	struct intel_mid_otg_xceiv *iotg = &the_transceiver->iotg;
+	u32	val32;
+
+	/* Check if the timer is already in the active list,
+	 * if so update timer count
+	 */
+	list_for_each_entry(tmp_timer, &active_timers, list)
+		if (tmp_timer == timer) {
+			timer->count = timer->expires;
+			return;
+		}
+	timer->count = timer->expires;
+
+	if (list_empty(&active_timers)) {
+		val32 = readl(iotg->base + CI_OTGSC);
+		writel(val32 | OTGSC_1MSE, iotg->base + CI_OTGSC);
+	}
+
+	list_add_tail(&timer->list, &active_timers);
+}
+
+/* Remove timer from the timer list; clear timeout status */
+static void langwell_otg_del_timer(void *gtimer)
+{
+	struct langwell_otg *lnw = the_transceiver;
+	struct langwell_otg_timer *timer = (struct langwell_otg_timer *)gtimer;
+	struct langwell_otg_timer *tmp_timer, *del_tmp;
+	u32 val32;
+
+	list_for_each_entry_safe(tmp_timer, del_tmp, &active_timers, list)
+		if (tmp_timer == timer)
+			list_del(&timer->list);
+
+	if (list_empty(&active_timers)) {
+		val32 = readl(lnw->iotg.base + CI_OTGSC);
+		writel(val32 & ~OTGSC_1MSE, lnw->iotg.base + CI_OTGSC);
+	}
+}
+
+/* Reduce timer count by 1, and find timeout conditions.*/
+static int langwell_otg_tick_timer(u32 *int_sts)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	struct langwell_otg_timer *tmp_timer, *del_tmp;
+	int expired = 0;
+
+	list_for_each_entry_safe(tmp_timer, del_tmp, &active_timers, list) {
+		tmp_timer->count--;
+		/* check if timer expires */
+		if (!tmp_timer->count) {
+			list_del(&tmp_timer->list);
+			tmp_timer->function(tmp_timer->data);
+			expired = 1;
+		}
+	}
+
+	if (list_empty(&active_timers)) {
+		dev_dbg(lnw->dev, "tick timer: disable 1ms int\n");
+		*int_sts = *int_sts & ~OTGSC_1MSE;
+	}
+	return expired;
+}
+
+static void reset_otg(void)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	int			delay_time = 1000;
+	u32			val;
+
+	dev_dbg(lnw->dev, "reseting OTG controller ...\n");
+	val = readl(lnw->iotg.base + CI_USBCMD);
+	writel(val | USBCMD_RST, lnw->iotg.base + CI_USBCMD);
+	do {
+		udelay(100);
+		if (!delay_time--)
+			dev_dbg(lnw->dev, "reset timeout\n");
+		val = readl(lnw->iotg.base + CI_USBCMD);
+		val &= USBCMD_RST;
+	} while (val != 0);
+	dev_dbg(lnw->dev, "reset done.\n");
+}
+
+static void set_host_mode(void)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	u32			val;
+
+	reset_otg();
+	val = readl(lnw->iotg.base + CI_USBMODE);
+	val = (val & (~USBMODE_CM)) | USBMODE_HOST;
+	writel(val, lnw->iotg.base + CI_USBMODE);
+}
+
+static void set_client_mode(void)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	u32			val;
+
+	reset_otg();
+	val = readl(lnw->iotg.base + CI_USBMODE);
+	val = (val & (~USBMODE_CM)) | USBMODE_DEVICE;
+	writel(val, lnw->iotg.base + CI_USBMODE);
+}
+
+static void init_hsm(void)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	u32				val32;
+
+	/* read OTGSC after reset */
+	val32 = readl(lnw->iotg.base + CI_OTGSC);
+	dev_dbg(lnw->dev, "%s: OTGSC init value = 0x%x\n", __func__, val32);
+
+	/* set init state */
+	if (val32 & OTGSC_ID) {
+		iotg->hsm.id = 1;
+		iotg->otg.default_a = 0;
+		set_client_mode();
+		iotg->otg.state = OTG_STATE_B_IDLE;
+	} else {
+		iotg->hsm.id = 0;
+		iotg->otg.default_a = 1;
+		set_host_mode();
+		iotg->otg.state = OTG_STATE_A_IDLE;
+	}
+
+	/* set session indicator */
+	if (val32 & OTGSC_BSE)
+		iotg->hsm.b_sess_end = 1;
+	if (val32 & OTGSC_BSV)
+		iotg->hsm.b_sess_vld = 1;
+	if (val32 & OTGSC_ASV)
+		iotg->hsm.a_sess_vld = 1;
+	if (val32 & OTGSC_AVV)
+		iotg->hsm.a_vbus_vld = 1;
+
+	/* defautly power the bus */
+	iotg->hsm.a_bus_req = 1;
+	iotg->hsm.a_bus_drop = 0;
+	/* defautly don't request bus as B device */
+	iotg->hsm.b_bus_req = 0;
+	/* no system error */
+	iotg->hsm.a_clr_err = 0;
+
+	langwell_otg_phy_low_power_wait(1);
+}
+
+static void update_hsm(void)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	u32				val32;
+
+	/* read OTGSC */
+	val32 = readl(lnw->iotg.base + CI_OTGSC);
+	dev_dbg(lnw->dev, "%s: OTGSC value = 0x%x\n", __func__, val32);
+
+	iotg->hsm.id = !!(val32 & OTGSC_ID);
+	iotg->hsm.b_sess_end = !!(val32 & OTGSC_BSE);
+	iotg->hsm.b_sess_vld = !!(val32 & OTGSC_BSV);
+	iotg->hsm.a_sess_vld = !!(val32 & OTGSC_ASV);
+	iotg->hsm.a_vbus_vld = !!(val32 & OTGSC_AVV);
+}
+
+static irqreturn_t otg_dummy_irq(int irq, void *_dev)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	void __iomem		*reg_base = _dev;
+	u32			val;
+	u32			int_mask = 0;
+
+	val = readl(reg_base + CI_USBMODE);
+	if ((val & USBMODE_CM) != USBMODE_DEVICE)
+		return IRQ_NONE;
+
+	val = readl(reg_base + CI_USBSTS);
+	int_mask = val & INTR_DUMMY_MASK;
+
+	if (int_mask == 0)
+		return IRQ_NONE;
+
+	/* clear hsm.b_conn here since host driver can't detect it
+	*  otg_dummy_irq called means B-disconnect happened.
+	*/
+	if (lnw->iotg.hsm.b_conn) {
+		lnw->iotg.hsm.b_conn = 0;
+		if (spin_trylock(&lnw->wq_lock)) {
+			langwell_update_transceiver();
+			spin_unlock(&lnw->wq_lock);
+		}
+	}
+
+	/* Clear interrupts */
+	writel(int_mask, reg_base + CI_USBSTS);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t otg_irq(int irq, void *_dev)
+{
+	struct langwell_otg		*lnw = _dev;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	u32				int_sts, int_en;
+	u32				int_mask = 0;
+	int				flag = 0;
+
+	int_sts = readl(lnw->iotg.base + CI_OTGSC);
+	int_en = (int_sts & OTGSC_INTEN_MASK) >> 8;
+	int_mask = int_sts & int_en;
+	if (int_mask == 0)
+		return IRQ_NONE;
+
+	if (int_mask & OTGSC_IDIS) {
+		dev_dbg(lnw->dev, "%s: id change int\n", __func__);
+		iotg->hsm.id = (int_sts & OTGSC_ID) ? 1 : 0;
+		dev_dbg(lnw->dev, "id = %d\n", iotg->hsm.id);
+		flag = 1;
+	}
+	if (int_mask & OTGSC_DPIS) {
+		dev_dbg(lnw->dev, "%s: data pulse int\n", __func__);
+		iotg->hsm.a_srp_det = (int_sts & OTGSC_DPS) ? 1 : 0;
+		dev_dbg(lnw->dev, "data pulse = %d\n", iotg->hsm.a_srp_det);
+		flag = 1;
+	}
+	if (int_mask & OTGSC_BSEIS) {
+		dev_dbg(lnw->dev, "%s: b session end int\n", __func__);
+		iotg->hsm.b_sess_end = (int_sts & OTGSC_BSE) ? 1 : 0;
+		dev_dbg(lnw->dev, "b_sess_end = %d\n", iotg->hsm.b_sess_end);
+		flag = 1;
+	}
+	if (int_mask & OTGSC_BSVIS) {
+		dev_dbg(lnw->dev, "%s: b session valid int\n", __func__);
+		iotg->hsm.b_sess_vld = (int_sts & OTGSC_BSV) ? 1 : 0;
+		dev_dbg(lnw->dev, "b_sess_vld = %d\n", iotg->hsm.b_sess_end);
+		flag = 1;
+	}
+	if (int_mask & OTGSC_ASVIS) {
+		dev_dbg(lnw->dev, "%s: a session valid int\n", __func__);
+		iotg->hsm.a_sess_vld = (int_sts & OTGSC_ASV) ? 1 : 0;
+		dev_dbg(lnw->dev, "a_sess_vld = %d\n", iotg->hsm.a_sess_vld);
+		flag = 1;
+	}
+	if (int_mask & OTGSC_AVVIS) {
+		dev_dbg(lnw->dev, "%s: a vbus valid int\n", __func__);
+		iotg->hsm.a_vbus_vld = (int_sts & OTGSC_AVV) ? 1 : 0;
+		dev_dbg(lnw->dev, "a_vbus_vld = %d\n", iotg->hsm.a_vbus_vld);
+		flag = 1;
+	}
+
+	if (int_mask & OTGSC_1MSS) {
+		/* need to schedule otg_work if any timer is expired */
+		if (langwell_otg_tick_timer(&int_sts))
+			flag = 1;
+	}
+
+	writel((int_sts & ~OTGSC_INTSTS_MASK) | int_mask,
+					lnw->iotg.base + CI_OTGSC);
+	if (flag)
+		langwell_update_transceiver();
+
+	return IRQ_HANDLED;
+}
+
+static int langwell_otg_iotg_notify(struct notifier_block *nb,
+				unsigned long action, void *data)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = data;
+	int				flag = 0;
+
+	if (iotg == NULL)
+		return NOTIFY_BAD;
+
+	if (lnw == NULL)
+		return NOTIFY_BAD;
+
+	switch (action) {
+	case MID_OTG_NOTIFY_CONNECT:
+		dev_dbg(lnw->dev, "Lnw OTG Notify Connect Event\n");
+		if (iotg->otg.default_a == 1)
+			iotg->hsm.b_conn = 1;
+		else
+			iotg->hsm.a_conn = 1;
+		flag = 1;
+		break;
+	case MID_OTG_NOTIFY_DISCONN:
+		dev_dbg(lnw->dev, "Lnw OTG Notify Disconnect Event\n");
+		if (iotg->otg.default_a == 1)
+			iotg->hsm.b_conn = 0;
+		else
+			iotg->hsm.a_conn = 0;
+		flag = 1;
+		break;
+	case MID_OTG_NOTIFY_HSUSPEND:
+		dev_dbg(lnw->dev, "Lnw OTG Notify Host Bus suspend Event\n");
+		if (iotg->otg.default_a == 1)
+			iotg->hsm.a_suspend_req = 1;
+		else
+			iotg->hsm.b_bus_req = 0;
+		flag = 1;
+		break;
+	case MID_OTG_NOTIFY_HRESUME:
+		dev_dbg(lnw->dev, "Lnw OTG Notify Host Bus resume Event\n");
+		if (iotg->otg.default_a == 1)
+			iotg->hsm.b_bus_resume = 1;
+		flag = 1;
+		break;
+	case MID_OTG_NOTIFY_CSUSPEND:
+		dev_dbg(lnw->dev, "Lnw OTG Notify Client Bus suspend Event\n");
+		if (iotg->otg.default_a == 1) {
+			if (iotg->hsm.b_bus_suspend_vld == 2) {
+				iotg->hsm.b_bus_suspend = 1;
+				iotg->hsm.b_bus_suspend_vld = 0;
+				flag = 1;
+			} else {
+				iotg->hsm.b_bus_suspend_vld++;
+				flag = 0;
+			}
+		} else {
+			if (iotg->hsm.a_bus_suspend == 0) {
+				iotg->hsm.a_bus_suspend = 1;
+				flag = 1;
+			}
+		}
+		break;
+	case MID_OTG_NOTIFY_CRESUME:
+		dev_dbg(lnw->dev, "Lnw OTG Notify Client Bus resume Event\n");
+		if (iotg->otg.default_a == 0)
+			iotg->hsm.a_bus_suspend = 0;
+		flag = 0;
+		break;
+	case MID_OTG_NOTIFY_HOSTADD:
+		dev_dbg(lnw->dev, "Lnw OTG Nofity Host Driver Add\n");
+		flag = 1;
+		break;
+	case MID_OTG_NOTIFY_HOSTREMOVE:
+		dev_dbg(lnw->dev, "Lnw OTG Nofity Host Driver remove\n");
+		flag = 1;
+		break;
+	case MID_OTG_NOTIFY_CLIENTADD:
+		dev_dbg(lnw->dev, "Lnw OTG Nofity Client Driver Add\n");
+		flag = 1;
+		break;
+	case MID_OTG_NOTIFY_CLIENTREMOVE:
+		dev_dbg(lnw->dev, "Lnw OTG Nofity Client Driver remove\n");
+		flag = 1;
+		break;
+	default:
+		dev_dbg(lnw->dev, "Lnw OTG Nofity unknown notify message\n");
+		return NOTIFY_DONE;
+	}
+
+	if (flag)
+		langwell_update_transceiver();
+
+	return NOTIFY_OK;
+}
+
+static void langwell_otg_work(struct work_struct *work)
+{
+	struct langwell_otg		*lnw;
+	struct intel_mid_otg_xceiv	*iotg;
+	int				retval;
+	struct pci_dev			*pdev;
+
+	lnw = container_of(work, struct langwell_otg, work);
+	iotg = &lnw->iotg;
+	pdev = to_pci_dev(lnw->dev);
+
+	dev_dbg(lnw->dev, "%s: old state = %s\n", __func__,
+			state_string(iotg->otg.state));
+
+	switch (iotg->otg.state) {
+	case OTG_STATE_UNDEFINED:
+	case OTG_STATE_B_IDLE:
+		if (!iotg->hsm.id) {
+			langwell_otg_del_timer(b_srp_init_tmr);
+			del_timer_sync(&lnw->hsm_timer);
+
+			iotg->otg.default_a = 1;
+			iotg->hsm.a_srp_det = 0;
+
+			langwell_otg_chrg_vbus(0);
+			set_host_mode();
+			langwell_otg_phy_low_power(1);
+
+			iotg->otg.state = OTG_STATE_A_IDLE;
+			langwell_update_transceiver();
+		} else if (iotg->hsm.b_sess_vld) {
+			langwell_otg_del_timer(b_srp_init_tmr);
+			del_timer_sync(&lnw->hsm_timer);
+			iotg->hsm.b_sess_end = 0;
+			iotg->hsm.a_bus_suspend = 0;
+			langwell_otg_chrg_vbus(0);
+
+			if (lnw->iotg.start_peripheral) {
+				lnw->iotg.start_peripheral(&lnw->iotg);
+				iotg->otg.state = OTG_STATE_B_PERIPHERAL;
+			} else
+				dev_dbg(lnw->dev, "client driver not loaded\n");
+
+		} else if (iotg->hsm.b_srp_init_tmout) {
+			iotg->hsm.b_srp_init_tmout = 0;
+			dev_warn(lnw->dev, "SRP init timeout\n");
+		} else if (iotg->hsm.b_srp_fail_tmout) {
+			iotg->hsm.b_srp_fail_tmout = 0;
+			iotg->hsm.b_bus_req = 0;
+
+			/* No silence failure */
+			langwell_otg_nsf_msg(6);
+		} else if (iotg->hsm.b_bus_req && iotg->hsm.b_sess_end) {
+			del_timer_sync(&lnw->hsm_timer);
+			/* workaround for b_se0_srp detection */
+			retval = langwell_otg_check_se0_srp(0);
+			if (retval) {
+				iotg->hsm.b_bus_req = 0;
+				dev_dbg(lnw->dev, "LS isn't SE0, try later\n");
+			} else {
+				/* clear the PHCD before start srp */
+				langwell_otg_phy_low_power(0);
+
+				/* Start SRP */
+				langwell_otg_add_timer(b_srp_init_tmr);
+				iotg->otg.start_srp(&iotg->otg);
+				langwell_otg_del_timer(b_srp_init_tmr);
+				langwell_otg_add_ktimer(TB_SRP_FAIL_TMR);
+
+				/* reset PHY low power mode here */
+				langwell_otg_phy_low_power_wait(1);
+			}
+		}
+		break;
+	case OTG_STATE_B_SRP_INIT:
+		if (!iotg->hsm.id) {
+			iotg->otg.default_a = 1;
+			iotg->hsm.a_srp_det = 0;
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			langwell_otg_chrg_vbus(0);
+			set_host_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_A_IDLE;
+			langwell_update_transceiver();
+		} else if (iotg->hsm.b_sess_vld) {
+			langwell_otg_chrg_vbus(0);
+			if (lnw->iotg.start_peripheral) {
+				lnw->iotg.start_peripheral(&lnw->iotg);
+				iotg->otg.state = OTG_STATE_B_PERIPHERAL;
+			} else
+				dev_dbg(lnw->dev, "client driver not loaded\n");
+		}
+		break;
+	case OTG_STATE_B_PERIPHERAL:
+		if (!iotg->hsm.id) {
+			iotg->otg.default_a = 1;
+			iotg->hsm.a_srp_det = 0;
+
+			langwell_otg_chrg_vbus(0);
+
+			if (lnw->iotg.stop_peripheral)
+				lnw->iotg.stop_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver has been removed.\n");
+
+			set_host_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_A_IDLE;
+			langwell_update_transceiver();
+		} else if (!iotg->hsm.b_sess_vld) {
+			iotg->hsm.b_hnp_enable = 0;
+
+			if (lnw->iotg.stop_peripheral)
+				lnw->iotg.stop_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver has been removed.\n");
+
+			iotg->otg.state = OTG_STATE_B_IDLE;
+		} else if (iotg->hsm.b_bus_req && iotg->otg.gadget &&
+					iotg->otg.gadget->b_hnp_enable &&
+					iotg->hsm.a_bus_suspend) {
+
+			if (lnw->iotg.stop_peripheral)
+				lnw->iotg.stop_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver has been removed.\n");
+
+			langwell_otg_HAAR(1);
+			iotg->hsm.a_conn = 0;
+
+			if (lnw->iotg.start_host) {
+				lnw->iotg.start_host(&lnw->iotg);
+				iotg->otg.state = OTG_STATE_B_WAIT_ACON;
+			} else
+				dev_dbg(lnw->dev,
+						"host driver not loaded.\n");
+
+			iotg->hsm.a_bus_resume = 0;
+			langwell_otg_add_ktimer(TB_ASE0_BRST_TMR);
+		}
+		break;
+
+	case OTG_STATE_B_WAIT_ACON:
+		if (!iotg->hsm.id) {
+			/* delete hsm timer for b_ase0_brst_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			iotg->otg.default_a = 1;
+			iotg->hsm.a_srp_det = 0;
+
+			langwell_otg_chrg_vbus(0);
+
+			langwell_otg_HAAR(0);
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			set_host_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_A_IDLE;
+			langwell_update_transceiver();
+		} else if (!iotg->hsm.b_sess_vld) {
+			/* delete hsm timer for b_ase0_brst_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			iotg->hsm.b_hnp_enable = 0;
+			iotg->hsm.b_bus_req = 0;
+
+			langwell_otg_chrg_vbus(0);
+			langwell_otg_HAAR(0);
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			set_client_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+		} else if (iotg->hsm.a_conn) {
+			/* delete hsm timer for b_ase0_brst_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			langwell_otg_HAAR(0);
+			iotg->otg.state = OTG_STATE_B_HOST;
+			langwell_update_transceiver();
+		} else if (iotg->hsm.a_bus_resume ||
+				iotg->hsm.b_ase0_brst_tmout) {
+			/* delete hsm timer for b_ase0_brst_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			langwell_otg_HAAR(0);
+			langwell_otg_nsf_msg(7);
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			iotg->hsm.a_bus_suspend = 0;
+			iotg->hsm.b_bus_req = 0;
+
+			if (lnw->iotg.start_peripheral)
+				lnw->iotg.start_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver not loaded.\n");
+
+			iotg->otg.state = OTG_STATE_B_PERIPHERAL;
+		}
+		break;
+
+	case OTG_STATE_B_HOST:
+		if (!iotg->hsm.id) {
+			iotg->otg.default_a = 1;
+			iotg->hsm.a_srp_det = 0;
+
+			langwell_otg_chrg_vbus(0);
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			set_host_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_A_IDLE;
+			langwell_update_transceiver();
+		} else if (!iotg->hsm.b_sess_vld) {
+			iotg->hsm.b_hnp_enable = 0;
+			iotg->hsm.b_bus_req = 0;
+
+			langwell_otg_chrg_vbus(0);
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			set_client_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+		} else if ((!iotg->hsm.b_bus_req) ||
+				(!iotg->hsm.a_conn)) {
+			iotg->hsm.b_bus_req = 0;
+			langwell_otg_loc_sof(0);
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			iotg->hsm.a_bus_suspend = 0;
+
+			if (lnw->iotg.start_peripheral)
+				lnw->iotg.start_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+						"client driver not loaded.\n");
+
+			iotg->otg.state = OTG_STATE_B_PERIPHERAL;
+		}
+		break;
+
+	case OTG_STATE_A_IDLE:
+		iotg->otg.default_a = 1;
+		if (iotg->hsm.id) {
+			iotg->otg.default_a = 0;
+			iotg->hsm.b_bus_req = 0;
+			iotg->hsm.vbus_srp_up = 0;
+
+			langwell_otg_chrg_vbus(0);
+			set_client_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+			langwell_update_transceiver();
+		} else if (!iotg->hsm.a_bus_drop &&
+			(iotg->hsm.a_srp_det || iotg->hsm.a_bus_req)) {
+			langwell_otg_phy_low_power(0);
+
+			/* Turn on VBus */
+			iotg->otg.set_vbus(&iotg->otg, true);
+
+			iotg->hsm.vbus_srp_up = 0;
+			iotg->hsm.a_wait_vrise_tmout = 0;
+			langwell_otg_add_timer(a_wait_vrise_tmr);
+			iotg->otg.state = OTG_STATE_A_WAIT_VRISE;
+			langwell_update_transceiver();
+		} else if (!iotg->hsm.a_bus_drop && iotg->hsm.a_sess_vld) {
+			iotg->hsm.vbus_srp_up = 1;
+		} else if (!iotg->hsm.a_sess_vld && iotg->hsm.vbus_srp_up) {
+			msleep(10);
+			langwell_otg_phy_low_power(0);
+
+			/* Turn on VBus */
+			iotg->otg.set_vbus(&iotg->otg, true);
+			iotg->hsm.a_srp_det = 1;
+			iotg->hsm.vbus_srp_up = 0;
+			iotg->hsm.a_wait_vrise_tmout = 0;
+			langwell_otg_add_timer(a_wait_vrise_tmr);
+			iotg->otg.state = OTG_STATE_A_WAIT_VRISE;
+			langwell_update_transceiver();
+		} else if (!iotg->hsm.a_sess_vld &&
+				!iotg->hsm.vbus_srp_up) {
+			langwell_otg_phy_low_power(1);
+		}
+		break;
+	case OTG_STATE_A_WAIT_VRISE:
+		if (iotg->hsm.id) {
+			langwell_otg_del_timer(a_wait_vrise_tmr);
+			iotg->hsm.b_bus_req = 0;
+			iotg->otg.default_a = 0;
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			set_client_mode();
+			langwell_otg_phy_low_power_wait(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+		} else if (iotg->hsm.a_vbus_vld) {
+			langwell_otg_del_timer(a_wait_vrise_tmr);
+			iotg->hsm.b_conn = 0;
+			if (lnw->iotg.start_host)
+				lnw->iotg.start_host(&lnw->iotg);
+			else {
+				dev_dbg(lnw->dev, "host driver not loaded.\n");
+				break;
+			}
+
+			langwell_otg_add_ktimer(TA_WAIT_BCON_TMR);
+			iotg->otg.state = OTG_STATE_A_WAIT_BCON;
+		} else if (iotg->hsm.a_wait_vrise_tmout) {
+			iotg->hsm.b_conn = 0;
+			if (iotg->hsm.a_vbus_vld) {
+				if (lnw->iotg.start_host)
+					lnw->iotg.start_host(&lnw->iotg);
+				else {
+					dev_dbg(lnw->dev,
+						"host driver not loaded.\n");
+					break;
+				}
+				langwell_otg_add_ktimer(TA_WAIT_BCON_TMR);
+				iotg->otg.state = OTG_STATE_A_WAIT_BCON;
+			} else {
+
+				/* Turn off VBus */
+				iotg->otg.set_vbus(&iotg->otg, false);
+				langwell_otg_phy_low_power_wait(1);
+				iotg->otg.state = OTG_STATE_A_VBUS_ERR;
+			}
+		}
+		break;
+	case OTG_STATE_A_WAIT_BCON:
+		if (iotg->hsm.id) {
+			/* delete hsm timer for a_wait_bcon_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			iotg->otg.default_a = 0;
+			iotg->hsm.b_bus_req = 0;
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			set_client_mode();
+			langwell_otg_phy_low_power_wait(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+			langwell_update_transceiver();
+		} else if (!iotg->hsm.a_vbus_vld) {
+			/* delete hsm timer for a_wait_bcon_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			langwell_otg_phy_low_power_wait(1);
+			iotg->otg.state = OTG_STATE_A_VBUS_ERR;
+		} else if (iotg->hsm.a_bus_drop ||
+				(iotg->hsm.a_wait_bcon_tmout &&
+				!iotg->hsm.a_bus_req)) {
+			/* delete hsm timer for a_wait_bcon_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			iotg->otg.state = OTG_STATE_A_WAIT_VFALL;
+		} else if (iotg->hsm.b_conn) {
+			/* delete hsm timer for a_wait_bcon_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			iotg->hsm.a_suspend_req = 0;
+			iotg->otg.state = OTG_STATE_A_HOST;
+			if (iotg->hsm.a_srp_det && iotg->otg.host &&
+					!iotg->otg.host->b_hnp_enable) {
+				/* SRP capable peripheral-only device */
+				iotg->hsm.a_bus_req = 1;
+				iotg->hsm.a_srp_det = 0;
+			} else if (!iotg->hsm.a_bus_req && iotg->otg.host &&
+					iotg->otg.host->b_hnp_enable) {
+				/* It is not safe enough to do a fast
+				 * transistion from A_WAIT_BCON to
+				 * A_SUSPEND */
+				msleep(10000);
+				if (iotg->hsm.a_bus_req)
+					break;
+
+				if (request_irq(pdev->irq,
+					otg_dummy_irq, IRQF_SHARED,
+					driver_name, iotg->base) != 0) {
+					dev_dbg(lnw->dev,
+						"request interrupt %d fail\n",
+						pdev->irq);
+				}
+
+				langwell_otg_HABA(1);
+				iotg->hsm.b_bus_resume = 0;
+				iotg->hsm.a_aidl_bdis_tmout = 0;
+
+				langwell_otg_loc_sof(0);
+				/* clear PHCD to enable HW timer */
+				langwell_otg_phy_low_power(0);
+				langwell_otg_add_timer(a_aidl_bdis_tmr);
+				iotg->otg.state = OTG_STATE_A_SUSPEND;
+			} else if (!iotg->hsm.a_bus_req && iotg->otg.host &&
+				!iotg->otg.host->b_hnp_enable) {
+				if (lnw->iotg.stop_host)
+					lnw->iotg.stop_host(&lnw->iotg);
+				else
+					dev_dbg(lnw->dev,
+						"host driver removed.\n");
+
+				/* Turn off VBus */
+				iotg->otg.set_vbus(&iotg->otg, false);
+				iotg->otg.state = OTG_STATE_A_WAIT_VFALL;
+			}
+		}
+		break;
+	case OTG_STATE_A_HOST:
+		if (iotg->hsm.id) {
+			iotg->otg.default_a = 0;
+			iotg->hsm.b_bus_req = 0;
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			set_client_mode();
+			langwell_otg_phy_low_power_wait(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+			langwell_update_transceiver();
+		} else if (iotg->hsm.a_bus_drop ||
+				(iotg->otg.host &&
+				!iotg->otg.host->b_hnp_enable &&
+					!iotg->hsm.a_bus_req)) {
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			iotg->otg.state = OTG_STATE_A_WAIT_VFALL;
+		} else if (!iotg->hsm.a_vbus_vld) {
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			langwell_otg_phy_low_power_wait(1);
+			iotg->otg.state = OTG_STATE_A_VBUS_ERR;
+		} else if (iotg->otg.host &&
+				iotg->otg.host->b_hnp_enable &&
+				!iotg->hsm.a_bus_req) {
+			/* Set HABA to enable hardware assistance to signal
+			 *  A-connect after receiver B-disconnect. Hardware
+			 *  will then set client mode and enable URE, SLE and
+			 *  PCE after the assistance. otg_dummy_irq is used to
+			 *  clean these ints when client driver is not resumed.
+			 */
+			if (request_irq(pdev->irq, otg_dummy_irq, IRQF_SHARED,
+					driver_name, iotg->base) != 0) {
+				dev_dbg(lnw->dev,
+					"request interrupt %d failed\n",
+						pdev->irq);
+			}
+
+			/* set HABA */
+			langwell_otg_HABA(1);
+			iotg->hsm.b_bus_resume = 0;
+			iotg->hsm.a_aidl_bdis_tmout = 0;
+			langwell_otg_loc_sof(0);
+			/* clear PHCD to enable HW timer */
+			langwell_otg_phy_low_power(0);
+			langwell_otg_add_timer(a_aidl_bdis_tmr);
+			iotg->otg.state = OTG_STATE_A_SUSPEND;
+		} else if (!iotg->hsm.b_conn || !iotg->hsm.a_bus_req) {
+			langwell_otg_add_ktimer(TA_WAIT_BCON_TMR);
+			iotg->otg.state = OTG_STATE_A_WAIT_BCON;
+		}
+		break;
+	case OTG_STATE_A_SUSPEND:
+		if (iotg->hsm.id) {
+			langwell_otg_del_timer(a_aidl_bdis_tmr);
+			langwell_otg_HABA(0);
+			free_irq(pdev->irq, iotg->base);
+			iotg->otg.default_a = 0;
+			iotg->hsm.b_bus_req = 0;
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			set_client_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+			langwell_update_transceiver();
+		} else if (iotg->hsm.a_bus_req ||
+				iotg->hsm.b_bus_resume) {
+			langwell_otg_del_timer(a_aidl_bdis_tmr);
+			langwell_otg_HABA(0);
+			free_irq(pdev->irq, iotg->base);
+			iotg->hsm.a_suspend_req = 0;
+			langwell_otg_loc_sof(1);
+			iotg->otg.state = OTG_STATE_A_HOST;
+		} else if (iotg->hsm.a_aidl_bdis_tmout ||
+				iotg->hsm.a_bus_drop) {
+			langwell_otg_del_timer(a_aidl_bdis_tmr);
+			langwell_otg_HABA(0);
+			free_irq(pdev->irq, iotg->base);
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			iotg->otg.state = OTG_STATE_A_WAIT_VFALL;
+		} else if (!iotg->hsm.b_conn && iotg->otg.host &&
+				iotg->otg.host->b_hnp_enable) {
+			langwell_otg_del_timer(a_aidl_bdis_tmr);
+			langwell_otg_HABA(0);
+			free_irq(pdev->irq, iotg->base);
+
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			iotg->hsm.b_bus_suspend = 0;
+			iotg->hsm.b_bus_suspend_vld = 0;
+
+			/* msleep(200); */
+			if (lnw->iotg.start_peripheral)
+				lnw->iotg.start_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver not loaded.\n");
+
+			langwell_otg_add_ktimer(TB_BUS_SUSPEND_TMR);
+			iotg->otg.state = OTG_STATE_A_PERIPHERAL;
+			break;
+		} else if (!iotg->hsm.a_vbus_vld) {
+			langwell_otg_del_timer(a_aidl_bdis_tmr);
+			langwell_otg_HABA(0);
+			free_irq(pdev->irq, iotg->base);
+			if (lnw->iotg.stop_host)
+				lnw->iotg.stop_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"host driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			langwell_otg_phy_low_power_wait(1);
+			iotg->otg.state = OTG_STATE_A_VBUS_ERR;
+		}
+		break;
+	case OTG_STATE_A_PERIPHERAL:
+		if (iotg->hsm.id) {
+			/* delete hsm timer for b_bus_suspend_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+			iotg->otg.default_a = 0;
+			iotg->hsm.b_bus_req = 0;
+			if (lnw->iotg.stop_peripheral)
+				lnw->iotg.stop_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			set_client_mode();
+			langwell_otg_phy_low_power_wait(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+			langwell_update_transceiver();
+		} else if (!iotg->hsm.a_vbus_vld) {
+			/* delete hsm timer for b_bus_suspend_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			if (lnw->iotg.stop_peripheral)
+				lnw->iotg.stop_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			langwell_otg_phy_low_power_wait(1);
+			iotg->otg.state = OTG_STATE_A_VBUS_ERR;
+		} else if (iotg->hsm.a_bus_drop) {
+			/* delete hsm timer for b_bus_suspend_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			if (lnw->iotg.stop_peripheral)
+				lnw->iotg.stop_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver has been removed.\n");
+
+			/* Turn off VBus */
+			iotg->otg.set_vbus(&iotg->otg, false);
+			iotg->otg.state = OTG_STATE_A_WAIT_VFALL;
+		} else if (iotg->hsm.b_bus_suspend) {
+			/* delete hsm timer for b_bus_suspend_tmr */
+			del_timer_sync(&lnw->hsm_timer);
+
+			if (lnw->iotg.stop_peripheral)
+				lnw->iotg.stop_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver has been removed.\n");
+
+			if (lnw->iotg.start_host)
+				lnw->iotg.start_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+						"host driver not loaded.\n");
+			langwell_otg_add_ktimer(TA_WAIT_BCON_TMR);
+			iotg->otg.state = OTG_STATE_A_WAIT_BCON;
+		} else if (iotg->hsm.b_bus_suspend_tmout) {
+			u32	val;
+			val = readl(lnw->iotg.base + CI_PORTSC1);
+			if (!(val & PORTSC_SUSP))
+				break;
+
+			if (lnw->iotg.stop_peripheral)
+				lnw->iotg.stop_peripheral(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+					"client driver has been removed.\n");
+
+			if (lnw->iotg.start_host)
+				lnw->iotg.start_host(&lnw->iotg);
+			else
+				dev_dbg(lnw->dev,
+						"host driver not loaded.\n");
+			langwell_otg_add_ktimer(TA_WAIT_BCON_TMR);
+			iotg->otg.state = OTG_STATE_A_WAIT_BCON;
+		}
+		break;
+	case OTG_STATE_A_VBUS_ERR:
+		if (iotg->hsm.id) {
+			iotg->otg.default_a = 0;
+			iotg->hsm.a_clr_err = 0;
+			iotg->hsm.a_srp_det = 0;
+			set_client_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+			langwell_update_transceiver();
+		} else if (iotg->hsm.a_clr_err) {
+			iotg->hsm.a_clr_err = 0;
+			iotg->hsm.a_srp_det = 0;
+			reset_otg();
+			init_hsm();
+			if (iotg->otg.state == OTG_STATE_A_IDLE)
+				langwell_update_transceiver();
+		} else {
+			/* FW will clear PHCD bit when any VBus
+			 * event detected. Reset PHCD to 1 again */
+			langwell_otg_phy_low_power(1);
+		}
+		break;
+	case OTG_STATE_A_WAIT_VFALL:
+		if (iotg->hsm.id) {
+			iotg->otg.default_a = 0;
+			set_client_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_B_IDLE;
+			langwell_update_transceiver();
+		} else if (iotg->hsm.a_bus_req) {
+
+			/* Turn on VBus */
+			iotg->otg.set_vbus(&iotg->otg, true);
+			iotg->hsm.a_wait_vrise_tmout = 0;
+			langwell_otg_add_timer(a_wait_vrise_tmr);
+			iotg->otg.state = OTG_STATE_A_WAIT_VRISE;
+		} else if (!iotg->hsm.a_sess_vld) {
+			iotg->hsm.a_srp_det = 0;
+			set_host_mode();
+			langwell_otg_phy_low_power(1);
+			iotg->otg.state = OTG_STATE_A_IDLE;
+		}
+		break;
+	default:
+		;
+	}
+
+	dev_dbg(lnw->dev, "%s: new state = %s\n", __func__,
+			state_string(iotg->otg.state));
+}
+
+static ssize_t
+show_registers(struct device *_dev, struct device_attribute *attr, char *buf)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	char			*next;
+	unsigned		size, t;
+
+	next = buf;
+	size = PAGE_SIZE;
+
+	t = scnprintf(next, size,
+		"\n"
+		"USBCMD = 0x%08x\n"
+		"USBSTS = 0x%08x\n"
+		"USBINTR = 0x%08x\n"
+		"ASYNCLISTADDR = 0x%08x\n"
+		"PORTSC1 = 0x%08x\n"
+		"HOSTPC1 = 0x%08x\n"
+		"OTGSC = 0x%08x\n"
+		"USBMODE = 0x%08x\n",
+		readl(lnw->iotg.base + 0x30),
+		readl(lnw->iotg.base + 0x34),
+		readl(lnw->iotg.base + 0x38),
+		readl(lnw->iotg.base + 0x48),
+		readl(lnw->iotg.base + 0x74),
+		readl(lnw->iotg.base + 0xb4),
+		readl(lnw->iotg.base + 0xf4),
+		readl(lnw->iotg.base + 0xf8)
+	     );
+	size -= t;
+	next += t;
+
+	return PAGE_SIZE - size;
+}
+static DEVICE_ATTR(registers, S_IRUGO, show_registers, NULL);
+
+static ssize_t
+show_hsm(struct device *_dev, struct device_attribute *attr, char *buf)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	char				*next;
+	unsigned			size, t;
+
+	next = buf;
+	size = PAGE_SIZE;
+
+	if (iotg->otg.host)
+		iotg->hsm.a_set_b_hnp_en = iotg->otg.host->b_hnp_enable;
+
+	if (iotg->otg.gadget)
+		iotg->hsm.b_hnp_enable = iotg->otg.gadget->b_hnp_enable;
+
+	t = scnprintf(next, size,
+		"\n"
+		"current state = %s\n"
+		"a_bus_resume = \t%d\n"
+		"a_bus_suspend = \t%d\n"
+		"a_conn = \t%d\n"
+		"a_sess_vld = \t%d\n"
+		"a_srp_det = \t%d\n"
+		"a_vbus_vld = \t%d\n"
+		"b_bus_resume = \t%d\n"
+		"b_bus_suspend = \t%d\n"
+		"b_conn = \t%d\n"
+		"b_se0_srp = \t%d\n"
+		"b_sess_end = \t%d\n"
+		"b_sess_vld = \t%d\n"
+		"id = \t%d\n"
+		"a_set_b_hnp_en = \t%d\n"
+		"b_srp_done = \t%d\n"
+		"b_hnp_enable = \t%d\n"
+		"a_wait_vrise_tmout = \t%d\n"
+		"a_wait_bcon_tmout = \t%d\n"
+		"a_aidl_bdis_tmout = \t%d\n"
+		"b_ase0_brst_tmout = \t%d\n"
+		"a_bus_drop = \t%d\n"
+		"a_bus_req = \t%d\n"
+		"a_clr_err = \t%d\n"
+		"a_suspend_req = \t%d\n"
+		"b_bus_req = \t%d\n"
+		"b_bus_suspend_tmout = \t%d\n"
+		"b_bus_suspend_vld = \t%d\n",
+		state_string(iotg->otg.state),
+		iotg->hsm.a_bus_resume,
+		iotg->hsm.a_bus_suspend,
+		iotg->hsm.a_conn,
+		iotg->hsm.a_sess_vld,
+		iotg->hsm.a_srp_det,
+		iotg->hsm.a_vbus_vld,
+		iotg->hsm.b_bus_resume,
+		iotg->hsm.b_bus_suspend,
+		iotg->hsm.b_conn,
+		iotg->hsm.b_se0_srp,
+		iotg->hsm.b_sess_end,
+		iotg->hsm.b_sess_vld,
+		iotg->hsm.id,
+		iotg->hsm.a_set_b_hnp_en,
+		iotg->hsm.b_srp_done,
+		iotg->hsm.b_hnp_enable,
+		iotg->hsm.a_wait_vrise_tmout,
+		iotg->hsm.a_wait_bcon_tmout,
+		iotg->hsm.a_aidl_bdis_tmout,
+		iotg->hsm.b_ase0_brst_tmout,
+		iotg->hsm.a_bus_drop,
+		iotg->hsm.a_bus_req,
+		iotg->hsm.a_clr_err,
+		iotg->hsm.a_suspend_req,
+		iotg->hsm.b_bus_req,
+		iotg->hsm.b_bus_suspend_tmout,
+		iotg->hsm.b_bus_suspend_vld
+		);
+	size -= t;
+	next += t;
+
+	return PAGE_SIZE - size;
+}
+static DEVICE_ATTR(hsm, S_IRUGO, show_hsm, NULL);
+
+static ssize_t
+get_a_bus_req(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	char			*next;
+	unsigned		size, t;
+
+	next = buf;
+	size = PAGE_SIZE;
+
+	t = scnprintf(next, size, "%d", lnw->iotg.hsm.a_bus_req);
+	size -= t;
+	next += t;
+
+	return PAGE_SIZE - size;
+}
+
+static ssize_t
+set_a_bus_req(struct device *dev, struct device_attribute *attr,
+		const char *buf, size_t count)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+
+	if (!iotg->otg.default_a)
+		return -1;
+	if (count > 2)
+		return -1;
+
+	if (buf[0] == '0') {
+		iotg->hsm.a_bus_req = 0;
+		dev_dbg(lnw->dev, "User request: a_bus_req = 0\n");
+	} else if (buf[0] == '1') {
+		/* If a_bus_drop is TRUE, a_bus_req can't be set */
+		if (iotg->hsm.a_bus_drop)
+			return -1;
+		iotg->hsm.a_bus_req = 1;
+		dev_dbg(lnw->dev, "User request: a_bus_req = 1\n");
+	}
+	if (spin_trylock(&lnw->wq_lock)) {
+		langwell_update_transceiver();
+		spin_unlock(&lnw->wq_lock);
+	}
+	return count;
+}
+static DEVICE_ATTR(a_bus_req, S_IRUGO | S_IWUGO, get_a_bus_req, set_a_bus_req);
+
+static ssize_t
+get_a_bus_drop(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	char			*next;
+	unsigned		size, t;
+
+	next = buf;
+	size = PAGE_SIZE;
+
+	t = scnprintf(next, size, "%d", lnw->iotg.hsm.a_bus_drop);
+	size -= t;
+	next += t;
+
+	return PAGE_SIZE - size;
+}
+
+static ssize_t
+set_a_bus_drop(struct device *dev, struct device_attribute *attr,
+		const char *buf, size_t count)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+
+	if (!iotg->otg.default_a)
+		return -1;
+	if (count > 2)
+		return -1;
+
+	if (buf[0] == '0') {
+		iotg->hsm.a_bus_drop = 0;
+		dev_dbg(lnw->dev, "User request: a_bus_drop = 0\n");
+	} else if (buf[0] == '1') {
+		iotg->hsm.a_bus_drop = 1;
+		iotg->hsm.a_bus_req = 0;
+		dev_dbg(lnw->dev, "User request: a_bus_drop = 1\n");
+		dev_dbg(lnw->dev, "User request: and a_bus_req = 0\n");
+	}
+	if (spin_trylock(&lnw->wq_lock)) {
+		langwell_update_transceiver();
+		spin_unlock(&lnw->wq_lock);
+	}
+	return count;
+}
+static DEVICE_ATTR(a_bus_drop, S_IRUGO | S_IWUGO,
+	get_a_bus_drop, set_a_bus_drop);
+
+static ssize_t
+get_b_bus_req(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	char			*next;
+	unsigned		size, t;
+
+	next = buf;
+	size = PAGE_SIZE;
+
+	t = scnprintf(next, size, "%d", lnw->iotg.hsm.b_bus_req);
+	size -= t;
+	next += t;
+
+	return PAGE_SIZE - size;
+}
+
+static ssize_t
+set_b_bus_req(struct device *dev, struct device_attribute *attr,
+		const char *buf, size_t count)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+
+	if (iotg->otg.default_a)
+		return -1;
+
+	if (count > 2)
+		return -1;
+
+	if (buf[0] == '0') {
+		iotg->hsm.b_bus_req = 0;
+		dev_dbg(lnw->dev, "User request: b_bus_req = 0\n");
+	} else if (buf[0] == '1') {
+		iotg->hsm.b_bus_req = 1;
+		dev_dbg(lnw->dev, "User request: b_bus_req = 1\n");
+	}
+	if (spin_trylock(&lnw->wq_lock)) {
+		langwell_update_transceiver();
+		spin_unlock(&lnw->wq_lock);
+	}
+	return count;
+}
+static DEVICE_ATTR(b_bus_req, S_IRUGO | S_IWUGO, get_b_bus_req, set_b_bus_req);
+
+static ssize_t
+set_a_clr_err(struct device *dev, struct device_attribute *attr,
+		const char *buf, size_t count)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+
+	if (!iotg->otg.default_a)
+		return -1;
+	if (count > 2)
+		return -1;
+
+	if (buf[0] == '1') {
+		iotg->hsm.a_clr_err = 1;
+		dev_dbg(lnw->dev, "User request: a_clr_err = 1\n");
+	}
+	if (spin_trylock(&lnw->wq_lock)) {
+		langwell_update_transceiver();
+		spin_unlock(&lnw->wq_lock);
+	}
+	return count;
+}
+static DEVICE_ATTR(a_clr_err, S_IWUGO, NULL, set_a_clr_err);
+
+static struct attribute *inputs_attrs[] = {
+	&dev_attr_a_bus_req.attr,
+	&dev_attr_a_bus_drop.attr,
+	&dev_attr_b_bus_req.attr,
+	&dev_attr_a_clr_err.attr,
+	NULL,
+};
+
+static struct attribute_group debug_dev_attr_group = {
+	.name = "inputs",
+	.attrs = inputs_attrs,
+};
+
+static int langwell_otg_probe(struct pci_dev *pdev,
+		const struct pci_device_id *id)
+{
+	unsigned long		resource, len;
+	void __iomem		*base = NULL;
+	int			retval;
+	u32			val32;
+	struct langwell_otg	*lnw;
+	char			qname[] = "langwell_otg_queue";
+
+	retval = 0;
+	dev_dbg(&pdev->dev, "\notg controller is detected.\n");
+	if (pci_enable_device(pdev) < 0) {
+		retval = -ENODEV;
+		goto done;
+	}
+
+	lnw = kzalloc(sizeof *lnw, GFP_KERNEL);
+	if (lnw == NULL) {
+		retval = -ENOMEM;
+		goto done;
+	}
+	the_transceiver = lnw;
+
+	/* control register: BAR 0 */
+	resource = pci_resource_start(pdev, 0);
+	len = pci_resource_len(pdev, 0);
+	if (!request_mem_region(resource, len, driver_name)) {
+		retval = -EBUSY;
+		goto err;
+	}
+	lnw->region = 1;
+
+	base = ioremap_nocache(resource, len);
+	if (base == NULL) {
+		retval = -EFAULT;
+		goto err;
+	}
+	lnw->iotg.base = base;
+
+	if (!request_mem_region(USBCFG_ADDR, USBCFG_LEN, driver_name)) {
+		retval = -EBUSY;
+		goto err;
+	}
+	lnw->cfg_region = 1;
+
+	/* For the SCCB.USBCFG register */
+	base = ioremap_nocache(USBCFG_ADDR, USBCFG_LEN);
+	if (base == NULL) {
+		retval = -EFAULT;
+		goto err;
+	}
+	lnw->usbcfg = base;
+
+	if (!pdev->irq) {
+		dev_dbg(&pdev->dev, "No IRQ.\n");
+		retval = -ENODEV;
+		goto err;
+	}
+
+	lnw->qwork = create_singlethread_workqueue(qname);
+	if (!lnw->qwork) {
+		dev_dbg(&pdev->dev, "cannot create workqueue %s\n", qname);
+		retval = -ENOMEM;
+		goto err;
+	}
+	INIT_WORK(&lnw->work, langwell_otg_work);
+
+	/* OTG common part */
+	lnw->dev = &pdev->dev;
+	lnw->iotg.otg.dev = lnw->dev;
+	lnw->iotg.otg.label = driver_name;
+	lnw->iotg.otg.set_host = langwell_otg_set_host;
+	lnw->iotg.otg.set_peripheral = langwell_otg_set_peripheral;
+	lnw->iotg.otg.set_power = langwell_otg_set_power;
+	lnw->iotg.otg.set_vbus = langwell_otg_set_vbus;
+	lnw->iotg.otg.start_srp = langwell_otg_start_srp;
+	lnw->iotg.otg.state = OTG_STATE_UNDEFINED;
+
+	if (otg_set_transceiver(&lnw->iotg.otg)) {
+		dev_dbg(lnw->dev, "can't set transceiver\n");
+		retval = -EBUSY;
+		goto err;
+	}
+
+	reset_otg();
+	init_hsm();
+
+	spin_lock_init(&lnw->lock);
+	spin_lock_init(&lnw->wq_lock);
+	INIT_LIST_HEAD(&active_timers);
+	retval = langwell_otg_init_timers(&lnw->iotg.hsm);
+	if (retval) {
+		dev_dbg(&pdev->dev, "Failed to init timers\n");
+		goto err;
+	}
+
+	init_timer(&lnw->hsm_timer);
+	ATOMIC_INIT_NOTIFIER_HEAD(&lnw->iotg.iotg_notifier);
+
+	lnw->iotg_notifier.notifier_call = langwell_otg_iotg_notify;
+
+	retval = intel_mid_otg_register_notifier(&lnw->iotg,
+						&lnw->iotg_notifier);
+	if (retval) {
+		dev_dbg(lnw->dev, "Failed to register notifier\n");
+		goto err;
+	}
+
+	if (request_irq(pdev->irq, otg_irq, IRQF_SHARED,
+				driver_name, lnw) != 0) {
+		dev_dbg(lnw->dev, "request interrupt %d failed\n", pdev->irq);
+		retval = -EBUSY;
+		goto err;
+	}
+
+	/* enable OTGSC int */
+	val32 = OTGSC_DPIE | OTGSC_BSEIE | OTGSC_BSVIE |
+		OTGSC_ASVIE | OTGSC_AVVIE | OTGSC_IDIE | OTGSC_IDPU;
+	writel(val32, lnw->iotg.base + CI_OTGSC);
+
+	retval = device_create_file(&pdev->dev, &dev_attr_registers);
+	if (retval < 0) {
+		dev_dbg(lnw->dev,
+			"Can't register sysfs attribute: %d\n", retval);
+		goto err;
+	}
+
+	retval = device_create_file(&pdev->dev, &dev_attr_hsm);
+	if (retval < 0) {
+		dev_dbg(lnw->dev, "Can't hsm sysfs attribute: %d\n", retval);
+		goto err;
+	}
+
+	retval = sysfs_create_group(&pdev->dev.kobj, &debug_dev_attr_group);
+	if (retval < 0) {
+		dev_dbg(lnw->dev,
+			"Can't register sysfs attr group: %d\n", retval);
+		goto err;
+	}
+
+	if (lnw->iotg.otg.state == OTG_STATE_A_IDLE)
+		langwell_update_transceiver();
+
+	return 0;
+
+err:
+	if (the_transceiver)
+		langwell_otg_remove(pdev);
+done:
+	return retval;
+}
+
+static void langwell_otg_remove(struct pci_dev *pdev)
+{
+	struct langwell_otg *lnw = the_transceiver;
+
+	if (lnw->qwork) {
+		flush_workqueue(lnw->qwork);
+		destroy_workqueue(lnw->qwork);
+	}
+	intel_mid_otg_unregister_notifier(&lnw->iotg, &lnw->iotg_notifier);
+	langwell_otg_free_timers();
+
+	/* disable OTGSC interrupt as OTGSC doesn't change in reset */
+	writel(0, lnw->iotg.base + CI_OTGSC);
+
+	if (pdev->irq)
+		free_irq(pdev->irq, lnw);
+	if (lnw->usbcfg)
+		iounmap(lnw->usbcfg);
+	if (lnw->cfg_region)
+		release_mem_region(USBCFG_ADDR, USBCFG_LEN);
+	if (lnw->iotg.base)
+		iounmap(lnw->iotg.base);
+	if (lnw->region)
+		release_mem_region(pci_resource_start(pdev, 0),
+				pci_resource_len(pdev, 0));
+
+	otg_set_transceiver(NULL);
+	pci_disable_device(pdev);
+	sysfs_remove_group(&pdev->dev.kobj, &debug_dev_attr_group);
+	device_remove_file(&pdev->dev, &dev_attr_hsm);
+	device_remove_file(&pdev->dev, &dev_attr_registers);
+	kfree(lnw);
+	lnw = NULL;
+}
+
+static void transceiver_suspend(struct pci_dev *pdev)
+{
+	pci_save_state(pdev);
+	pci_set_power_state(pdev, PCI_D3hot);
+	langwell_otg_phy_low_power(1);
+}
+
+static int langwell_otg_suspend(struct pci_dev *pdev, pm_message_t message)
+{
+	struct langwell_otg		*lnw = the_transceiver;
+	struct intel_mid_otg_xceiv	*iotg = &lnw->iotg;
+	int				ret = 0;
+
+	/* Disbale OTG interrupts */
+	langwell_otg_intr(0);
+
+	if (pdev->irq)
+		free_irq(pdev->irq, lnw);
+
+	/* Prevent more otg_work */
+	flush_workqueue(lnw->qwork);
+	destroy_workqueue(lnw->qwork);
+	lnw->qwork = NULL;
+
+	/* start actions */
+	switch (iotg->otg.state) {
+	case OTG_STATE_A_WAIT_VFALL:
+		iotg->otg.state = OTG_STATE_A_IDLE;
+	case OTG_STATE_A_IDLE:
+	case OTG_STATE_B_IDLE:
+	case OTG_STATE_A_VBUS_ERR:
+		transceiver_suspend(pdev);
+		break;
+	case OTG_STATE_A_WAIT_VRISE:
+		langwell_otg_del_timer(a_wait_vrise_tmr);
+		iotg->hsm.a_srp_det = 0;
+
+		/* Turn off VBus */
+		iotg->otg.set_vbus(&iotg->otg, false);
+		iotg->otg.state = OTG_STATE_A_IDLE;
+		transceiver_suspend(pdev);
+		break;
+	case OTG_STATE_A_WAIT_BCON:
+		del_timer_sync(&lnw->hsm_timer);
+		if (lnw->iotg.stop_host)
+			lnw->iotg.stop_host(&lnw->iotg);
+		else
+			dev_dbg(&pdev->dev, "host driver has been removed.\n");
+
+		iotg->hsm.a_srp_det = 0;
+
+		/* Turn off VBus */
+		iotg->otg.set_vbus(&iotg->otg, false);
+		iotg->otg.state = OTG_STATE_A_IDLE;
+		transceiver_suspend(pdev);
+		break;
+	case OTG_STATE_A_HOST:
+		if (lnw->iotg.stop_host)
+			lnw->iotg.stop_host(&lnw->iotg);
+		else
+			dev_dbg(&pdev->dev, "host driver has been removed.\n");
+
+		iotg->hsm.a_srp_det = 0;
+
+		/* Turn off VBus */
+		iotg->otg.set_vbus(&iotg->otg, false);
+
+		iotg->otg.state = OTG_STATE_A_IDLE;
+		transceiver_suspend(pdev);
+		break;
+	case OTG_STATE_A_SUSPEND:
+		langwell_otg_del_timer(a_aidl_bdis_tmr);
+		langwell_otg_HABA(0);
+		if (lnw->iotg.stop_host)
+			lnw->iotg.stop_host(&lnw->iotg);
+		else
+			dev_dbg(lnw->dev, "host driver has been removed.\n");
+		iotg->hsm.a_srp_det = 0;
+
+		/* Turn off VBus */
+		iotg->otg.set_vbus(&iotg->otg, false);
+		iotg->otg.state = OTG_STATE_A_IDLE;
+		transceiver_suspend(pdev);
+		break;
+	case OTG_STATE_A_PERIPHERAL:
+		del_timer_sync(&lnw->hsm_timer);
+
+		if (lnw->iotg.stop_peripheral)
+			lnw->iotg.stop_peripheral(&lnw->iotg);
+		else
+			dev_dbg(&pdev->dev,
+				"client driver has been removed.\n");
+		iotg->hsm.a_srp_det = 0;
+
+		/* Turn off VBus */
+		iotg->otg.set_vbus(&iotg->otg, false);
+		iotg->otg.state = OTG_STATE_A_IDLE;
+		transceiver_suspend(pdev);
+		break;
+	case OTG_STATE_B_HOST:
+		if (lnw->iotg.stop_host)
+			lnw->iotg.stop_host(&lnw->iotg);
+		else
+			dev_dbg(&pdev->dev, "host driver has been removed.\n");
+		iotg->hsm.b_bus_req = 0;
+		iotg->otg.state = OTG_STATE_B_IDLE;
+		transceiver_suspend(pdev);
+		break;
+	case OTG_STATE_B_PERIPHERAL:
+		if (lnw->iotg.stop_peripheral)
+			lnw->iotg.stop_peripheral(&lnw->iotg);
+		else
+			dev_dbg(&pdev->dev,
+				"client driver has been removed.\n");
+		iotg->otg.state = OTG_STATE_B_IDLE;
+		transceiver_suspend(pdev);
+		break;
+	case OTG_STATE_B_WAIT_ACON:
+		/* delete hsm timer for b_ase0_brst_tmr */
+		del_timer_sync(&lnw->hsm_timer);
+
+		langwell_otg_HAAR(0);
+
+		if (lnw->iotg.stop_host)
+			lnw->iotg.stop_host(&lnw->iotg);
+		else
+			dev_dbg(&pdev->dev, "host driver has been removed.\n");
+		iotg->hsm.b_bus_req = 0;
+		iotg->otg.state = OTG_STATE_B_IDLE;
+		transceiver_suspend(pdev);
+		break;
+	default:
+		dev_dbg(lnw->dev, "error state before suspend\n");
+		break;
+	}
+
+	return ret;
+}
+
+static void transceiver_resume(struct pci_dev *pdev)
+{
+	pci_restore_state(pdev);
+	pci_set_power_state(pdev, PCI_D0);
+}
+
+static int langwell_otg_resume(struct pci_dev *pdev)
+{
+	struct langwell_otg	*lnw = the_transceiver;
+	int			ret = 0;
+
+	transceiver_resume(pdev);
+
+	lnw->qwork = create_singlethread_workqueue("langwell_otg_queue");
+	if (!lnw->qwork) {
+		dev_dbg(&pdev->dev, "cannot create langwell otg workqueuen");
+		ret = -ENOMEM;
+		goto error;
+	}
+
+	if (request_irq(pdev->irq, otg_irq, IRQF_SHARED,
+				driver_name, lnw) != 0) {
+		dev_dbg(&pdev->dev, "request interrupt %d failed\n", pdev->irq);
+		ret = -EBUSY;
+		goto error;
+	}
+
+	/* enable OTG interrupts */
+	langwell_otg_intr(1);
+
+	update_hsm();
+
+	langwell_update_transceiver();
+
+	return ret;
+error:
+	langwell_otg_intr(0);
+	transceiver_suspend(pdev);
+	return ret;
+}
+
+static int __init langwell_otg_init(void)
+{
+	return pci_register_driver(&otg_pci_driver);
+}
+module_init(langwell_otg_init);
+
+static void __exit langwell_otg_cleanup(void)
+{
+	pci_unregister_driver(&otg_pci_driver);
+}
+module_exit(langwell_otg_cleanup);
diff --git a/include/linux/usb/langwell_otg.h b/include/linux/usb/langwell_otg.h
new file mode 100644
index 000000000000..a6562f1d4e2b
--- /dev/null
+++ b/include/linux/usb/langwell_otg.h
@@ -0,0 +1,139 @@
+/*
+ * Intel Langwell USB OTG transceiver driver
+ * Copyright (C) 2008, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope 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.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ */
+
+#ifndef __LANGWELL_OTG_H
+#define __LANGWELL_OTG_H
+
+#include <linux/usb/intel_mid_otg.h>
+
+#define CI_USBCMD		0x30
+#	define USBCMD_RST		BIT(1)
+#	define USBCMD_RS		BIT(0)
+#define CI_USBSTS		0x34
+#	define USBSTS_SLI		BIT(8)
+#	define USBSTS_URI		BIT(6)
+#	define USBSTS_PCI		BIT(2)
+#define CI_PORTSC1		0x74
+#	define PORTSC_PP		BIT(12)
+#	define PORTSC_LS		(BIT(11) | BIT(10))
+#	define PORTSC_SUSP		BIT(7)
+#	define PORTSC_CCS		BIT(0)
+#define CI_HOSTPC1		0xb4
+#	define HOSTPC1_PHCD		BIT(22)
+#define CI_OTGSC		0xf4
+#	define OTGSC_DPIE		BIT(30)
+#	define OTGSC_1MSE		BIT(29)
+#	define OTGSC_BSEIE		BIT(28)
+#	define OTGSC_BSVIE		BIT(27)
+#	define OTGSC_ASVIE		BIT(26)
+#	define OTGSC_AVVIE		BIT(25)
+#	define OTGSC_IDIE		BIT(24)
+#	define OTGSC_DPIS		BIT(22)
+#	define OTGSC_1MSS		BIT(21)
+#	define OTGSC_BSEIS		BIT(20)
+#	define OTGSC_BSVIS		BIT(19)
+#	define OTGSC_ASVIS		BIT(18)
+#	define OTGSC_AVVIS		BIT(17)
+#	define OTGSC_IDIS		BIT(16)
+#	define OTGSC_DPS		BIT(14)
+#	define OTGSC_1MST		BIT(13)
+#	define OTGSC_BSE		BIT(12)
+#	define OTGSC_BSV		BIT(11)
+#	define OTGSC_ASV		BIT(10)
+#	define OTGSC_AVV		BIT(9)
+#	define OTGSC_ID			BIT(8)
+#	define OTGSC_HABA		BIT(7)
+#	define OTGSC_HADP		BIT(6)
+#	define OTGSC_IDPU		BIT(5)
+#	define OTGSC_DP			BIT(4)
+#	define OTGSC_OT			BIT(3)
+#	define OTGSC_HAAR		BIT(2)
+#	define OTGSC_VC			BIT(1)
+#	define OTGSC_VD			BIT(0)
+#	define OTGSC_INTEN_MASK		(0x7f << 24)
+#	define OTGSC_INT_MASK		(0x5f << 24)
+#	define OTGSC_INTSTS_MASK	(0x7f << 16)
+#define CI_USBMODE		0xf8
+#	define USBMODE_CM		(BIT(1) | BIT(0))
+#	define USBMODE_IDLE		0
+#	define USBMODE_DEVICE		0x2
+#	define USBMODE_HOST		0x3
+#define USBCFG_ADDR			0xff10801c
+#define USBCFG_LEN			4
+#	define USBCFG_VBUSVAL		BIT(14)
+#	define USBCFG_AVALID		BIT(13)
+#	define USBCFG_BVALID		BIT(12)
+#	define USBCFG_SESEND		BIT(11)
+
+#define INTR_DUMMY_MASK (USBSTS_SLI | USBSTS_URI | USBSTS_PCI)
+
+enum langwell_otg_timer_type {
+	TA_WAIT_VRISE_TMR,
+	TA_WAIT_BCON_TMR,
+	TA_AIDL_BDIS_TMR,
+	TB_ASE0_BRST_TMR,
+	TB_SE0_SRP_TMR,
+	TB_SRP_INIT_TMR,
+	TB_SRP_FAIL_TMR,
+	TB_BUS_SUSPEND_TMR
+};
+
+#define TA_WAIT_VRISE	100
+#define TA_WAIT_BCON	30000
+#define TA_AIDL_BDIS	15000
+#define TB_ASE0_BRST	5000
+#define TB_SE0_SRP	2
+#define TB_SRP_INIT	100
+#define TB_SRP_FAIL	5500
+#define TB_BUS_SUSPEND	500
+
+struct langwell_otg_timer {
+	unsigned long expires;	/* Number of count increase to timeout */
+	unsigned long count;	/* Tick counter */
+	void (*function)(unsigned long);	/* Timeout function */
+	unsigned long data;	/* Data passed to function */
+	struct list_head list;
+};
+
+struct langwell_otg {
+	struct intel_mid_otg_xceiv	iotg;
+	struct device			*dev;
+
+	void __iomem			*usbcfg;	/* SCCBUSB config Reg */
+
+	unsigned			region;
+	unsigned			cfg_region;
+
+	struct work_struct		work;
+	struct workqueue_struct		*qwork;
+	struct timer_list		hsm_timer;
+
+	spinlock_t			lock;
+	spinlock_t			wq_lock;
+
+	struct notifier_block		iotg_notifier;
+};
+
+static inline
+struct langwell_otg *mid_xceiv_to_lnw(struct intel_mid_otg_xceiv *iotg)
+{
+	return container_of(iotg, struct langwell_otg, iotg);
+}
+
+#endif /* __LANGWELL_OTG_H__ */
-- 
cgit v1.2.3


From 37b5801e16d2e192fe2b20f4af33aa8c6e8786f3 Mon Sep 17 00:00:00 2001
From: Parirajan Muthalagu <parirajan.muthalagu@stericsson.com>
Date: Wed, 25 Aug 2010 16:33:26 +0530
Subject: USB Gadget: Verify VBUS current before setting the device
 self-powered bit

Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Praveena Nadahally <praveen.nadahally@stericsson.com>
Signed-off-by: Parirajan Muthalagu <parirajan.muthalagu@stericsson.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/composite.c |  8 +++++++-
 include/linux/usb/ch9.h        | 10 ++++++++++
 2 files changed, 17 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c
index 1160c55de7f2..eaa9a599df63 100644
--- a/drivers/usb/gadget/composite.c
+++ b/drivers/usb/gadget/composite.c
@@ -1074,7 +1074,13 @@ static int composite_bind(struct usb_gadget *gadget)
 	cdev->bufsiz = USB_BUFSIZ;
 	cdev->driver = composite;
 
-	usb_gadget_set_selfpowered(gadget);
+	/*
+	 * As per USB compliance update, a device that is actively drawing
+	 * more than 100mA from USB must report itself as bus-powered in
+	 * the GetStatus(DEVICE) call.
+	 */
+	if (CONFIG_USB_GADGET_VBUS_DRAW <= USB_SELF_POWER_VBUS_MAX_DRAW)
+		usb_gadget_set_selfpowered(gadget);
 
 	/* interface and string IDs start at zero via kzalloc.
 	 * we force endpoints to start unassigned; few controller
diff --git a/include/linux/usb/ch9.h b/include/linux/usb/ch9.h
index da2ed77d3e8d..b0f7e9f57176 100644
--- a/include/linux/usb/ch9.h
+++ b/include/linux/usb/ch9.h
@@ -808,4 +808,14 @@ enum usb_device_state {
 	 */
 };
 
+/*-------------------------------------------------------------------------*/
+
+/*
+ * As per USB compliance update, a device that is actively drawing
+ * more than 100mA from USB must report itself as bus-powered in
+ * the GetStatus(DEVICE) call.
+ * http://compliance.usb.org/index.asp?UpdateFile=Electrical&Format=Standard#34
+ */
+#define USB_SELF_POWER_VBUS_MAX_DRAW		100
+
 #endif /* __LINUX_USB_CH9_H */
-- 
cgit v1.2.3


From ad1a8102f957f4d25fc58cdc10736c5ade7557e1 Mon Sep 17 00:00:00 2001
From: Michal Nazarewicz <m.nazarewicz@samsung.com>
Date: Thu, 12 Aug 2010 17:43:46 +0200
Subject: USB: gadget: composite: Better string override handling

The iManufatcurer, iProduct and iSerialNumber composite module
parameters were only used when the gadget driver registers
strings for manufacturer, product and serial number.  If the
gadget never bothered to set corresponding fields in USB device
descriptors those module parameters are ignored.

This commit makes the parameters work even if the strings ID
have not been assigned.  It also changes the way IDs are
overridden -- what IDs are overridden is now saved in
usb_composite_dev structure -- which makes it unnecessary to
modify the string tables the way previous code did.

The commit also adds a iProduct and iManufatcurer fields to the
usb_composite_device structure.  If they are set, appropriate
strings are reserved and added to device descriptor.  This makes
it unnecessary for gadget drivers to maintain code for setting
those.  If iProduct is not set it defaults to
usb_composite_device::name; if iManufatcurer is not set
a default "<system> <release> with <gadget-name>" is used.

The last thing is that if needs_serial field of
usb_composite_device is set and user failed to provided
iSerialNumber parameter a warning is issued.

Signed-off-by: Michal Nazarewicz <m.nazarewicz@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/composite.c | 96 ++++++++++++++++++++++++++++--------------
 include/linux/usb/composite.h  | 13 ++++++
 2 files changed, 77 insertions(+), 32 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c
index eaa9a599df63..717de39627c7 100644
--- a/drivers/usb/gadget/composite.c
+++ b/drivers/usb/gadget/composite.c
@@ -24,6 +24,7 @@
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/device.h>
+#include <linux/utsname.h>
 
 #include <linux/usb/composite.h>
 
@@ -69,6 +70,8 @@ static char *iSerialNumber;
 module_param(iSerialNumber, charp, 0);
 MODULE_PARM_DESC(iSerialNumber, "SerialNumber string");
 
+static char composite_manufacturer[50];
+
 /*-------------------------------------------------------------------------*/
 
 /**
@@ -599,6 +602,7 @@ static int get_string(struct usb_composite_dev *cdev,
 	struct usb_configuration	*c;
 	struct usb_function		*f;
 	int				len;
+	const char			*str;
 
 	/* Yes, not only is USB's I18N support probably more than most
 	 * folk will ever care about ... also, it's all supported here.
@@ -638,9 +642,29 @@ static int get_string(struct usb_composite_dev *cdev,
 		return s->bLength;
 	}
 
-	/* Otherwise, look up and return a specified string.  String IDs
-	 * are device-scoped, so we look up each string table we're told
-	 * about.  These lookups are infrequent; simpler-is-better here.
+	/* Otherwise, look up and return a specified string.  First
+	 * check if the string has not been overridden.
+	 */
+	if (cdev->manufacturer_override == id)
+		str = iManufacturer ?: composite->iManufacturer ?:
+			composite_manufacturer;
+	else if (cdev->product_override == id)
+		str = iProduct ?: composite->iProduct;
+	else if (cdev->serial_override == id)
+		str = iSerialNumber;
+	else
+		str = NULL;
+	if (str) {
+		struct usb_gadget_strings strings = {
+			.language = language,
+			.strings  = &(struct usb_string) { 0xff, str }
+		};
+		return usb_gadget_get_string(&strings, 0xff, buf);
+	}
+
+	/* String IDs are device-scoped, so we look up each string
+	 * table we're told about.  These lookups are infrequent;
+	 * simpler-is-better here.
 	 */
 	if (composite->strings) {
 		len = lookup_string(composite->strings, buf, language, id);
@@ -1025,26 +1049,17 @@ composite_unbind(struct usb_gadget *gadget)
 	composite = NULL;
 }
 
-static void
-string_override_one(struct usb_gadget_strings *tab, u8 id, const char *s)
+static u8 override_id(struct usb_composite_dev *cdev, u8 *desc)
 {
-	struct usb_string		*str = tab->strings;
-
-	for (str = tab->strings; str->s; str++) {
-		if (str->id == id) {
-			str->s = s;
-			return;
-		}
+	if (!*desc) {
+		int ret = usb_string_id(cdev);
+		if (unlikely(ret < 0))
+			WARNING(cdev, "failed to override string ID\n");
+		else
+			*desc = ret;
 	}
-}
 
-static void
-string_override(struct usb_gadget_strings **tab, u8 id, const char *s)
-{
-	while (*tab) {
-		string_override_one(*tab, id, s);
-		tab++;
-	}
+	return *desc;
 }
 
 static int composite_bind(struct usb_gadget *gadget)
@@ -1107,19 +1122,34 @@ static int composite_bind(struct usb_gadget *gadget)
 	cdev->desc = *composite->dev;
 	cdev->desc.bMaxPacketSize0 = gadget->ep0->maxpacket;
 
-	/* strings can't be assigned before bind() allocates the
-	 * releavnt identifiers
-	 */
-	if (cdev->desc.iManufacturer && iManufacturer)
-		string_override(composite->strings,
-			cdev->desc.iManufacturer, iManufacturer);
-	if (cdev->desc.iProduct && iProduct)
-		string_override(composite->strings,
-			cdev->desc.iProduct, iProduct);
-	if (cdev->desc.iSerialNumber && iSerialNumber)
-		string_override(composite->strings,
-			cdev->desc.iSerialNumber, iSerialNumber);
+	/* stirng overrides */
+	if (iManufacturer || !cdev->desc.iManufacturer) {
+		if (!iManufacturer && !composite->iManufacturer &&
+		    !*composite_manufacturer)
+			snprintf(composite_manufacturer,
+				 sizeof composite_manufacturer,
+				 "%s %s with %s",
+				 init_utsname()->sysname,
+				 init_utsname()->release,
+				 gadget->name);
+
+		cdev->manufacturer_override =
+			override_id(cdev, &cdev->desc.iManufacturer);
+	}
+
+	if (iProduct || (!cdev->desc.iProduct && composite->iProduct))
+		cdev->product_override =
+			override_id(cdev, &cdev->desc.iProduct);
+
+	if (iSerialNumber)
+		cdev->serial_override =
+			override_id(cdev, &cdev->desc.iSerialNumber);
+
+	/* has userspace failed to provide a serial number? */
+	if (composite->needs_serial && !cdev->desc.iSerialNumber)
+		WARNING(cdev, "userspace failed to provide iSerialNumber\n");
 
+	/* finish up */
 	status = device_create_file(&gadget->dev, &dev_attr_suspended);
 	if (status)
 		goto fail;
@@ -1217,6 +1247,8 @@ int usb_composite_register(struct usb_composite_driver *driver)
 	if (!driver || !driver->dev || !driver->bind || composite)
 		return -EINVAL;
 
+	if (!driver->iProduct)
+		driver->iProduct = driver->name;
 	if (!driver->name)
 		driver->name = "composite";
 	composite_driver.function =  (char *) driver->name;
diff --git a/include/linux/usb/composite.h b/include/linux/usb/composite.h
index 617068134ae8..a78e813d27e4 100644
--- a/include/linux/usb/composite.h
+++ b/include/linux/usb/composite.h
@@ -237,10 +237,17 @@ int usb_add_config(struct usb_composite_dev *,
 /**
  * struct usb_composite_driver - groups configurations into a gadget
  * @name: For diagnostics, identifies the driver.
+ * @iProduct: Used as iProduct override if @dev->iProduct is not set.
+ *	If NULL value of @name is taken.
+ * @iManufacturer: Used as iManufacturer override if @dev->iManufacturer is
+ *	not set. If NULL a default "<system> <release> with <udc>" value
+ *	will be used.
  * @dev: Template descriptor for the device, including default device
  *	identifiers.
  * @strings: tables of strings, keyed by identifiers assigned during bind()
  *	and language IDs provided in control requests
+ * @needs_serial: set to 1 if the gadget needs userspace to provide
+ * 	a serial number.  If one is not provided, warning will be printed.
  * @bind: (REQUIRED) Used to allocate resources that are shared across the
  *	whole device, such as string IDs, and add its configurations using
  *	@usb_add_config().  This may fail by returning a negative errno
@@ -266,8 +273,11 @@ int usb_add_config(struct usb_composite_dev *,
  */
 struct usb_composite_driver {
 	const char				*name;
+	const char				*iProduct;
+	const char				*iManufacturer;
 	const struct usb_device_descriptor	*dev;
 	struct usb_gadget_strings		**strings;
+	unsigned		needs_serial:1;
 
 	/* REVISIT:  bind() functions can be marked __init, which
 	 * makes trouble for section mismatch analysis.  See if
@@ -334,6 +344,9 @@ struct usb_composite_dev {
 	struct list_head		configs;
 	struct usb_composite_driver	*driver;
 	u8				next_string_id;
+	u8				manufacturer_override;
+	u8				product_override;
+	u8				serial_override;
 
 	/* the gadget driver won't enable the data pullup
 	 * while the deactivation count is nonzero.
-- 
cgit v1.2.3


From b0fca50f5a94a268ed02cfddf44448051ed9343f Mon Sep 17 00:00:00 2001
From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Date: Thu, 12 Aug 2010 17:43:53 +0200
Subject: usb gadget: don't save bind callback in struct usb_gadget_driver
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To accomplish this the function to register a gadget driver takes the bind
function as a second argument.  To make things clearer rename the function
to resemble platform_driver_probe.

This fixes many section mismatches like

	WARNING: drivers/usb/gadget/g_printer.o(.data+0xc): Section mismatch in
	reference from the variable printer_driver to the function
	.init.text:printer_bind()
	The variable printer_driver references
	the function __init printer_bind()

All callers are fixed.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
[m.nazarewicz@samsung.com: added dbgp]
Signed-off-by: Michał Nazarewicz <m.nazarewicz@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/amd5536udc.c     |  9 +++++----
 drivers/usb/gadget/at91_udc.c       | 11 ++++++-----
 drivers/usb/gadget/atmel_usba_udc.c |  7 ++++---
 drivers/usb/gadget/ci13xxx_udc.c    | 18 ++++++++++--------
 drivers/usb/gadget/composite.c      |  3 +--
 drivers/usb/gadget/dbgp.c           |  3 +--
 drivers/usb/gadget/dummy_hcd.c      | 10 +++++-----
 drivers/usb/gadget/file_storage.c   |  3 +--
 drivers/usb/gadget/fsl_qe_udc.c     | 12 ++++++------
 drivers/usb/gadget/fsl_udc_core.c   | 10 +++++-----
 drivers/usb/gadget/gmidi.c          |  3 +--
 drivers/usb/gadget/goku_udc.c       |  9 +++++----
 drivers/usb/gadget/imx_udc.c        |  9 +++++----
 drivers/usb/gadget/inode.c          |  6 ++----
 drivers/usb/gadget/langwell_udc.c   |  9 +++++----
 drivers/usb/gadget/lh7a40x_udc.c    | 10 +++++-----
 drivers/usb/gadget/m66592-udc.c     |  9 +++++----
 drivers/usb/gadget/net2280.c        | 10 +++++-----
 drivers/usb/gadget/omap_udc.c       | 10 +++++-----
 drivers/usb/gadget/printer.c        |  5 ++---
 drivers/usb/gadget/pxa25x_udc.c     |  9 +++++----
 drivers/usb/gadget/pxa27x_udc.c     | 12 +++++++-----
 drivers/usb/gadget/r8a66597-udc.c   |  9 +++++----
 drivers/usb/gadget/s3c-hsotg.c      |  9 +++++----
 drivers/usb/gadget/s3c2410_udc.c    | 17 ++++++++---------
 drivers/usb/musb/musb_gadget.c      | 11 ++++++-----
 include/linux/usb/gadget.h          | 20 ++++++++------------
 27 files changed, 128 insertions(+), 125 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/gadget/amd5536udc.c b/drivers/usb/gadget/amd5536udc.c
index 731150d4b1d9..fadebfd53b47 100644
--- a/drivers/usb/gadget/amd5536udc.c
+++ b/drivers/usb/gadget/amd5536udc.c
@@ -1954,13 +1954,14 @@ static int setup_ep0(struct udc *dev)
 }
 
 /* Called by gadget driver to register itself */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct udc		*dev = udc;
 	int			retval;
 	u32 tmp;
 
-	if (!driver || !driver->bind || !driver->setup
+	if (!driver || !bind || !driver->setup
 			|| driver->speed != USB_SPEED_HIGH)
 		return -EINVAL;
 	if (!dev)
@@ -1972,7 +1973,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	dev->driver = driver;
 	dev->gadget.dev.driver = &driver->driver;
 
-	retval = driver->bind(&dev->gadget);
+	retval = bind(&dev->gadget);
 
 	/* Some gadget drivers use both ep0 directions.
 	 * NOTE: to gadget driver, ep0 is just one endpoint...
@@ -2000,7 +2001,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	return 0;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 /* shutdown requests and disconnect from gadget */
 static void
diff --git a/drivers/usb/gadget/at91_udc.c b/drivers/usb/gadget/at91_udc.c
index 93ead19507b6..387e503b9d14 100644
--- a/drivers/usb/gadget/at91_udc.c
+++ b/drivers/usb/gadget/at91_udc.c
@@ -1628,7 +1628,8 @@ static void at91_vbus_timer(unsigned long data)
 		schedule_work(&udc->vbus_timer_work);
 }
 
-int usb_gadget_register_driver (struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct at91_udc	*udc = &controller;
 	int		retval;
@@ -1636,7 +1637,7 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 
 	if (!driver
 			|| driver->speed < USB_SPEED_FULL
-			|| !driver->bind
+			|| !bind
 			|| !driver->setup) {
 		DBG("bad parameter.\n");
 		return -EINVAL;
@@ -1653,9 +1654,9 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	udc->enabled = 1;
 	udc->selfpowered = 1;
 
-	retval = driver->bind(&udc->gadget);
+	retval = bind(&udc->gadget);
 	if (retval) {
-		DBG("driver->bind() returned %d\n", retval);
+		DBG("bind() returned %d\n", retval);
 		udc->driver = NULL;
 		udc->gadget.dev.driver = NULL;
 		dev_set_drvdata(&udc->gadget.dev, NULL);
@@ -1671,7 +1672,7 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	DBG("bound to %s\n", driver->driver.name);
 	return 0;
 }
-EXPORT_SYMBOL (usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int usb_gadget_unregister_driver (struct usb_gadget_driver *driver)
 {
diff --git a/drivers/usb/gadget/atmel_usba_udc.c b/drivers/usb/gadget/atmel_usba_udc.c
index d623c7bda1f6..e4810c6a0b1f 100644
--- a/drivers/usb/gadget/atmel_usba_udc.c
+++ b/drivers/usb/gadget/atmel_usba_udc.c
@@ -1789,7 +1789,8 @@ out:
 	return IRQ_HANDLED;
 }
 
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct usba_udc *udc = &the_udc;
 	unsigned long flags;
@@ -1812,7 +1813,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	clk_enable(udc->pclk);
 	clk_enable(udc->hclk);
 
-	ret = driver->bind(&udc->gadget);
+	ret = bind(&udc->gadget);
 	if (ret) {
 		DBG(DBG_ERR, "Could not bind to driver %s: error %d\n",
 			driver->driver.name, ret);
@@ -1841,7 +1842,7 @@ err_driver_bind:
 	udc->gadget.dev.driver = NULL;
 	return ret;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
 {
diff --git a/drivers/usb/gadget/ci13xxx_udc.c b/drivers/usb/gadget/ci13xxx_udc.c
index 699695128e33..98b36fc88c77 100644
--- a/drivers/usb/gadget/ci13xxx_udc.c
+++ b/drivers/usb/gadget/ci13xxx_udc.c
@@ -2340,12 +2340,15 @@ static const struct usb_ep_ops usb_ep_ops = {
 static const struct usb_gadget_ops usb_gadget_ops;
 
 /**
- * usb_gadget_register_driver: register a gadget driver
+ * usb_gadget_probe_driver: register a gadget driver
+ * @driver: the driver being registered
+ * @bind: the driver's bind callback
  *
- * Check usb_gadget_register_driver() at "usb_gadget.h" for details
- * Interrupts are enabled here
+ * Check usb_gadget_probe_driver() at <linux/usb/gadget.h> for details.
+ * Interrupts are enabled here.
  */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct ci13xxx *udc = _udc;
 	unsigned long i, k, flags;
@@ -2354,7 +2357,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	trace("%p", driver);
 
 	if (driver             == NULL ||
-	    driver->bind       == NULL ||
+	    bind               == NULL ||
 	    driver->unbind     == NULL ||
 	    driver->setup      == NULL ||
 	    driver->disconnect == NULL ||
@@ -2430,7 +2433,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	udc->gadget.dev.driver = &driver->driver;
 
 	spin_unlock_irqrestore(udc->lock, flags);
-	retval = driver->bind(&udc->gadget);                /* MAY SLEEP */
+	retval = bind(&udc->gadget);                /* MAY SLEEP */
 	spin_lock_irqsave(udc->lock, flags);
 
 	if (retval) {
@@ -2447,7 +2450,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 		usb_gadget_unregister_driver(driver);
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 /**
  * usb_gadget_unregister_driver: unregister a gadget driver
@@ -2462,7 +2465,6 @@ int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
 	trace("%p", driver);
 
 	if (driver             == NULL ||
-	    driver->bind       == NULL ||
 	    driver->unbind     == NULL ||
 	    driver->setup      == NULL ||
 	    driver->disconnect == NULL ||
diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c
index 717de39627c7..a3009bf01229 100644
--- a/drivers/usb/gadget/composite.c
+++ b/drivers/usb/gadget/composite.c
@@ -1213,7 +1213,6 @@ composite_resume(struct usb_gadget *gadget)
 static struct usb_gadget_driver composite_driver = {
 	.speed		= USB_SPEED_HIGH,
 
-	.bind		= composite_bind,
 	.unbind		= composite_unbind,
 
 	.setup		= composite_setup,
@@ -1255,7 +1254,7 @@ int usb_composite_register(struct usb_composite_driver *driver)
 	composite_driver.driver.name = driver->name;
 	composite = driver;
 
-	return usb_gadget_register_driver(&composite_driver);
+	return usb_gadget_probe_driver(&composite_driver, composite_bind);
 }
 
 /**
diff --git a/drivers/usb/gadget/dbgp.c b/drivers/usb/gadget/dbgp.c
index abe4a2ec5625..e5ac8a316fec 100644
--- a/drivers/usb/gadget/dbgp.c
+++ b/drivers/usb/gadget/dbgp.c
@@ -403,7 +403,6 @@ fail:
 static struct usb_gadget_driver dbgp_driver = {
 	.function = "dbgp",
 	.speed = USB_SPEED_HIGH,
-	.bind = dbgp_bind,
 	.unbind = dbgp_unbind,
 	.setup = dbgp_setup,
 	.disconnect = dbgp_disconnect,
@@ -415,7 +414,7 @@ static struct usb_gadget_driver dbgp_driver = {
 
 static int __init dbgp_init(void)
 {
-	return usb_gadget_register_driver(&dbgp_driver);
+	return usb_gadget_probe_driver(&dbgp_driver, dbgp_bind);
 }
 
 static void __exit dbgp_exit(void)
diff --git a/drivers/usb/gadget/dummy_hcd.c b/drivers/usb/gadget/dummy_hcd.c
index dc6546248ed9..7bb9d78aac27 100644
--- a/drivers/usb/gadget/dummy_hcd.c
+++ b/drivers/usb/gadget/dummy_hcd.c
@@ -748,7 +748,8 @@ static DEVICE_ATTR (function, S_IRUGO, show_function, NULL);
  */
 
 int
-usb_gadget_register_driver (struct usb_gadget_driver *driver)
+usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct dummy	*dum = the_controller;
 	int		retval, i;
@@ -757,8 +758,7 @@ usb_gadget_register_driver (struct usb_gadget_driver *driver)
 		return -EINVAL;
 	if (dum->driver)
 		return -EBUSY;
-	if (!driver->bind || !driver->setup
-			|| driver->speed == USB_SPEED_UNKNOWN)
+	if (!bind || !driver->setup || driver->speed == USB_SPEED_UNKNOWN)
 		return -EINVAL;
 
 	/*
@@ -796,7 +796,7 @@ usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	dum->gadget.dev.driver = &driver->driver;
 	dev_dbg (udc_dev(dum), "binding gadget driver '%s'\n",
 			driver->driver.name);
-	retval = driver->bind(&dum->gadget);
+	retval = bind(&dum->gadget);
 	if (retval) {
 		dum->driver = NULL;
 		dum->gadget.dev.driver = NULL;
@@ -812,7 +812,7 @@ usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	usb_hcd_poll_rh_status (dummy_to_hcd (dum));
 	return 0;
 }
-EXPORT_SYMBOL (usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int
 usb_gadget_unregister_driver (struct usb_gadget_driver *driver)
diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c
index aae04c20b0ea..132a1c0877bd 100644
--- a/drivers/usb/gadget/file_storage.c
+++ b/drivers/usb/gadget/file_storage.c
@@ -3608,7 +3608,6 @@ static struct usb_gadget_driver		fsg_driver = {
 	.speed		= USB_SPEED_FULL,
 #endif
 	.function	= (char *) fsg_string_product,
-	.bind		= fsg_bind,
 	.unbind		= fsg_unbind,
 	.disconnect	= fsg_disconnect,
 	.setup		= fsg_setup,
@@ -3650,7 +3649,7 @@ static int __init fsg_init(void)
 	if ((rc = fsg_alloc()) != 0)
 		return rc;
 	fsg = the_fsg;
-	if ((rc = usb_gadget_register_driver(&fsg_driver)) != 0)
+	if ((rc = usb_gadget_probe_driver(&fsg_driver, fsg_bind)) != 0)
 		kref_put(&fsg->ref, fsg_release);
 	return rc;
 }
diff --git a/drivers/usb/gadget/fsl_qe_udc.c b/drivers/usb/gadget/fsl_qe_udc.c
index a5ea2c1d8c93..792d5ef40137 100644
--- a/drivers/usb/gadget/fsl_qe_udc.c
+++ b/drivers/usb/gadget/fsl_qe_udc.c
@@ -2302,9 +2302,10 @@ static irqreturn_t qe_udc_irq(int irq, void *_udc)
 }
 
 /*-------------------------------------------------------------------------
-	Gadget driver register and unregister.
+	Gadget driver probe and unregister.
  --------------------------------------------------------------------------*/
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	int retval;
 	unsigned long flags = 0;
@@ -2315,8 +2316,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	if (!driver || (driver->speed != USB_SPEED_FULL
 			&& driver->speed != USB_SPEED_HIGH)
-			|| !driver->bind || !driver->disconnect
-			|| !driver->setup)
+			|| !bind || !driver->disconnect || !driver->setup)
 		return -EINVAL;
 
 	if (udc_controller->driver)
@@ -2332,7 +2332,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	udc_controller->gadget.speed = (enum usb_device_speed)(driver->speed);
 	spin_unlock_irqrestore(&udc_controller->lock, flags);
 
-	retval = driver->bind(&udc_controller->gadget);
+	retval = bind(&udc_controller->gadget);
 	if (retval) {
 		dev_err(udc_controller->dev, "bind to %s --> %d",
 				driver->driver.name, retval);
@@ -2353,7 +2353,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 		udc_controller->gadget.name, driver->driver.name);
 	return 0;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
 {
diff --git a/drivers/usb/gadget/fsl_udc_core.c b/drivers/usb/gadget/fsl_udc_core.c
index 08a9a62a39e3..c16b402a876b 100644
--- a/drivers/usb/gadget/fsl_udc_core.c
+++ b/drivers/usb/gadget/fsl_udc_core.c
@@ -1765,7 +1765,8 @@ static irqreturn_t fsl_udc_irq(int irq, void *_udc)
  * Hook to gadget drivers
  * Called by initialization code of gadget drivers
 *----------------------------------------------------------------*/
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	int retval = -ENODEV;
 	unsigned long flags = 0;
@@ -1775,8 +1776,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	if (!driver || (driver->speed != USB_SPEED_FULL
 				&& driver->speed != USB_SPEED_HIGH)
-			|| !driver->bind || !driver->disconnect
-			|| !driver->setup)
+			|| !bind || !driver->disconnect || !driver->setup)
 		return -EINVAL;
 
 	if (udc_controller->driver)
@@ -1792,7 +1792,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	spin_unlock_irqrestore(&udc_controller->lock, flags);
 
 	/* bind udc driver to gadget driver */
-	retval = driver->bind(&udc_controller->gadget);
+	retval = bind(&udc_controller->gadget);
 	if (retval) {
 		VDBG("bind to %s --> %d", driver->driver.name, retval);
 		udc_controller->gadget.dev.driver = NULL;
@@ -1814,7 +1814,7 @@ out:
 		       retval);
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 /* Disconnect from gadget driver */
 int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
diff --git a/drivers/usb/gadget/gmidi.c b/drivers/usb/gadget/gmidi.c
index b7bf88019b06..0ab7e141d494 100644
--- a/drivers/usb/gadget/gmidi.c
+++ b/drivers/usb/gadget/gmidi.c
@@ -1292,7 +1292,6 @@ static void gmidi_resume(struct usb_gadget *gadget)
 static struct usb_gadget_driver gmidi_driver = {
 	.speed		= USB_SPEED_FULL,
 	.function	= (char *)longname,
-	.bind		= gmidi_bind,
 	.unbind		= gmidi_unbind,
 
 	.setup		= gmidi_setup,
@@ -1309,7 +1308,7 @@ static struct usb_gadget_driver gmidi_driver = {
 
 static int __init gmidi_init(void)
 {
-	return usb_gadget_register_driver(&gmidi_driver);
+	return usb_gadget_probe_driver(&gmidi_driver, gmidi_bind);
 }
 module_init(gmidi_init);
 
diff --git a/drivers/usb/gadget/goku_udc.c b/drivers/usb/gadget/goku_udc.c
index 1088d08c7ed8..49fbd4dbeb94 100644
--- a/drivers/usb/gadget/goku_udc.c
+++ b/drivers/usb/gadget/goku_udc.c
@@ -1343,14 +1343,15 @@ static struct goku_udc	*the_controller;
  * disconnect is reported.  then a host may connect again, or
  * the driver might get unbound.
  */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct goku_udc	*dev = the_controller;
 	int			retval;
 
 	if (!driver
 			|| driver->speed < USB_SPEED_FULL
-			|| !driver->bind
+			|| !bind
 			|| !driver->disconnect
 			|| !driver->setup)
 		return -EINVAL;
@@ -1363,7 +1364,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	driver->driver.bus = NULL;
 	dev->driver = driver;
 	dev->gadget.dev.driver = &driver->driver;
-	retval = driver->bind(&dev->gadget);
+	retval = bind(&dev->gadget);
 	if (retval) {
 		DBG(dev, "bind to driver %s --> error %d\n",
 				driver->driver.name, retval);
@@ -1380,7 +1381,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	DBG(dev, "registered gadget driver '%s'\n", driver->driver.name);
 	return 0;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 static void
 stop_activity(struct goku_udc *dev, struct usb_gadget_driver *driver)
diff --git a/drivers/usb/gadget/imx_udc.c b/drivers/usb/gadget/imx_udc.c
index e743122fcd93..ed0266462c57 100644
--- a/drivers/usb/gadget/imx_udc.c
+++ b/drivers/usb/gadget/imx_udc.c
@@ -1319,14 +1319,15 @@ static struct imx_udc_struct controller = {
  * USB gadged driver functions
  *******************************************************************************
  */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct imx_udc_struct *imx_usb = &controller;
 	int retval;
 
 	if (!driver
 		|| driver->speed < USB_SPEED_FULL
-		|| !driver->bind
+		|| !bind
 		|| !driver->disconnect
 		|| !driver->setup)
 			return -EINVAL;
@@ -1342,7 +1343,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	retval = device_add(&imx_usb->gadget.dev);
 	if (retval)
 		goto fail;
-	retval = driver->bind(&imx_usb->gadget);
+	retval = bind(&imx_usb->gadget);
 	if (retval) {
 		D_ERR(imx_usb->dev, "<%s> bind to driver %s --> error %d\n",
 			__func__, driver->driver.name, retval);
@@ -1362,7 +1363,7 @@ fail:
 	imx_usb->gadget.dev.driver = NULL;
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
 {
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index fc35406fc80c..e2e8eda83f0d 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -1775,7 +1775,6 @@ static struct usb_gadget_driver gadgetfs_driver = {
 	.speed		= USB_SPEED_FULL,
 #endif
 	.function	= (char *) driver_desc,
-	.bind		= gadgetfs_bind,
 	.unbind		= gadgetfs_unbind,
 	.setup		= gadgetfs_setup,
 	.disconnect	= gadgetfs_disconnect,
@@ -1798,7 +1797,6 @@ static int gadgetfs_probe (struct usb_gadget *gadget)
 
 static struct usb_gadget_driver probe_driver = {
 	.speed		= USB_SPEED_HIGH,
-	.bind		= gadgetfs_probe,
 	.unbind		= gadgetfs_nop,
 	.setup		= (void *)gadgetfs_nop,
 	.disconnect	= gadgetfs_nop,
@@ -1908,7 +1906,7 @@ dev_config (struct file *fd, const char __user *buf, size_t len, loff_t *ptr)
 
 	/* triggers gadgetfs_bind(); then we can enumerate. */
 	spin_unlock_irq (&dev->lock);
-	value = usb_gadget_register_driver (&gadgetfs_driver);
+	value = usb_gadget_probe_driver(&gadgetfs_driver, gadgetfs_bind);
 	if (value != 0) {
 		kfree (dev->buf);
 		dev->buf = NULL;
@@ -2047,7 +2045,7 @@ gadgetfs_fill_super (struct super_block *sb, void *opts, int silent)
 		return -ESRCH;
 
 	/* fake probe to determine $CHIP */
-	(void) usb_gadget_register_driver (&probe_driver);
+	(void) usb_gadget_probe_driver(&probe_driver, gadgetfs_probe);
 	if (!CHIP)
 		return -ENODEV;
 
diff --git a/drivers/usb/gadget/langwell_udc.c b/drivers/usb/gadget/langwell_udc.c
index ccc9c070a30d..1ef17a6dcb51 100644
--- a/drivers/usb/gadget/langwell_udc.c
+++ b/drivers/usb/gadget/langwell_udc.c
@@ -1855,7 +1855,8 @@ static DEVICE_ATTR(remote_wakeup, S_IWUSR, NULL, store_remote_wakeup);
  * the driver might get unbound.
  */
 
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct langwell_udc	*dev = the_controller;
 	unsigned long		flags;
@@ -1878,7 +1879,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	spin_unlock_irqrestore(&dev->lock, flags);
 
-	retval = driver->bind(&dev->gadget);
+	retval = bind(&dev->gadget);
 	if (retval) {
 		dev_dbg(&dev->pdev->dev, "bind to driver %s --> %d\n",
 				driver->driver.name, retval);
@@ -1916,7 +1917,7 @@ err_unbind:
 	dev_dbg(&dev->pdev->dev, "<--- %s()\n", __func__);
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 
 /* unregister gadget driver */
@@ -1930,7 +1931,7 @@ int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
 
 	dev_dbg(&dev->pdev->dev, "---> %s()\n", __func__);
 
-	if (unlikely(!driver || !driver->bind || !driver->unbind))
+	if (unlikely(!driver || !driver->unbind))
 		return -EINVAL;
 
 	/* exit PHY low power suspend */
diff --git a/drivers/usb/gadget/lh7a40x_udc.c b/drivers/usb/gadget/lh7a40x_udc.c
index fded3fca793b..6b58bd8ce623 100644
--- a/drivers/usb/gadget/lh7a40x_udc.c
+++ b/drivers/usb/gadget/lh7a40x_udc.c
@@ -408,7 +408,8 @@ static void udc_enable(struct lh7a40x_udc *dev)
 /*
   Register entry point for the peripheral controller driver.
 */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct lh7a40x_udc *dev = the_controller;
 	int retval;
@@ -417,7 +418,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	if (!driver
 			|| driver->speed != USB_SPEED_FULL
-			|| !driver->bind
+			|| !bind
 			|| !driver->disconnect
 			|| !driver->setup)
 		return -EINVAL;
@@ -431,7 +432,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	dev->gadget.dev.driver = &driver->driver;
 
 	device_add(&dev->gadget.dev);
-	retval = driver->bind(&dev->gadget);
+	retval = bind(&dev->gadget);
 	if (retval) {
 		printk(KERN_WARNING "%s: bind to driver %s --> error %d\n",
 		       dev->gadget.name, driver->driver.name, retval);
@@ -453,8 +454,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	return 0;
 }
-
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 /*
   Unregister entry point for the peripheral controller driver.
diff --git a/drivers/usb/gadget/m66592-udc.c b/drivers/usb/gadget/m66592-udc.c
index e03058fe23cb..51b19f3027e7 100644
--- a/drivers/usb/gadget/m66592-udc.c
+++ b/drivers/usb/gadget/m66592-udc.c
@@ -1454,14 +1454,15 @@ static struct usb_ep_ops m66592_ep_ops = {
 /*-------------------------------------------------------------------------*/
 static struct m66592 *the_controller;
 
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct m66592 *m66592 = the_controller;
 	int retval;
 
 	if (!driver
 			|| driver->speed != USB_SPEED_HIGH
-			|| !driver->bind
+			|| !bind
 			|| !driver->setup)
 		return -EINVAL;
 	if (!m66592)
@@ -1480,7 +1481,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 		goto error;
 	}
 
-	retval = driver->bind (&m66592->gadget);
+	retval = bind(&m66592->gadget);
 	if (retval) {
 		pr_err("bind to driver error (%d)\n", retval);
 		device_del(&m66592->gadget.dev);
@@ -1505,7 +1506,7 @@ error:
 
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
 {
diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c
index 9498be87a724..d09155b25d73 100644
--- a/drivers/usb/gadget/net2280.c
+++ b/drivers/usb/gadget/net2280.c
@@ -1929,7 +1929,8 @@ static void ep0_start (struct net2280 *dev)
  * disconnect is reported.  then a host may connect again, or
  * the driver might get unbound.
  */
-int usb_gadget_register_driver (struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct net2280		*dev = the_controller;
 	int			retval;
@@ -1941,8 +1942,7 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	 */
 	if (!driver
 			|| driver->speed != USB_SPEED_HIGH
-			|| !driver->bind
-			|| !driver->setup)
+			|| !bind || !driver->setup)
 		return -EINVAL;
 	if (!dev)
 		return -ENODEV;
@@ -1957,7 +1957,7 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	driver->driver.bus = NULL;
 	dev->driver = driver;
 	dev->gadget.dev.driver = &driver->driver;
-	retval = driver->bind (&dev->gadget);
+	retval = bind(&dev->gadget);
 	if (retval) {
 		DEBUG (dev, "bind to driver %s --> %d\n",
 				driver->driver.name, retval);
@@ -1993,7 +1993,7 @@ err_unbind:
 	dev->driver = NULL;
 	return retval;
 }
-EXPORT_SYMBOL (usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 static void
 stop_activity (struct net2280 *dev, struct usb_gadget_driver *driver)
diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c
index f81e4f025f23..61d3ca6619bb 100644
--- a/drivers/usb/gadget/omap_udc.c
+++ b/drivers/usb/gadget/omap_udc.c
@@ -2102,7 +2102,8 @@ static inline int machine_without_vbus_sense(void)
 		);
 }
 
-int usb_gadget_register_driver (struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	int		status = -ENODEV;
 	struct omap_ep	*ep;
@@ -2114,8 +2115,7 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	if (!driver
 			// FIXME if otg, check:  driver->is_otg
 			|| driver->speed < USB_SPEED_FULL
-			|| !driver->bind
-			|| !driver->setup)
+			|| !bind || !driver->setup)
 		return -EINVAL;
 
 	spin_lock_irqsave(&udc->lock, flags);
@@ -2145,7 +2145,7 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver)
 	if (udc->dc_clk != NULL)
 		omap_udc_enable_clock(1);
 
-	status = driver->bind (&udc->gadget);
+	status = bind(&udc->gadget);
 	if (status) {
 		DBG("bind to %s --> %d\n", driver->driver.name, status);
 		udc->gadget.dev.driver = NULL;
@@ -2186,7 +2186,7 @@ done:
 		omap_udc_enable_clock(0);
 	return status;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int usb_gadget_unregister_driver (struct usb_gadget_driver *driver)
 {
diff --git a/drivers/usb/gadget/printer.c b/drivers/usb/gadget/printer.c
index 04c7bbf121af..ded080a1c8ce 100644
--- a/drivers/usb/gadget/printer.c
+++ b/drivers/usb/gadget/printer.c
@@ -1543,7 +1543,6 @@ static struct usb_gadget_driver printer_driver = {
 	.speed		= DEVSPEED,
 
 	.function	= (char *) driver_desc,
-	.bind		= printer_bind,
 	.unbind		= printer_unbind,
 
 	.setup		= printer_setup,
@@ -1579,11 +1578,11 @@ init(void)
 		return status;
 	}
 
-	status = usb_gadget_register_driver(&printer_driver);
+	status = usb_gadget_probe_driver(&printer_driver, printer_bind);
 	if (status) {
 		class_destroy(usb_gadget_class);
 		unregister_chrdev_region(g_printer_devno, 1);
-		DBG(dev, "usb_gadget_register_driver %x\n", status);
+		DBG(dev, "usb_gadget_probe_driver %x\n", status);
 	}
 
 	return status;
diff --git a/drivers/usb/gadget/pxa25x_udc.c b/drivers/usb/gadget/pxa25x_udc.c
index be5fb34d9602..b37f92cb71bc 100644
--- a/drivers/usb/gadget/pxa25x_udc.c
+++ b/drivers/usb/gadget/pxa25x_udc.c
@@ -1280,14 +1280,15 @@ static void udc_enable (struct pxa25x_udc *dev)
  * disconnect is reported.  then a host may connect again, or
  * the driver might get unbound.
  */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct pxa25x_udc	*dev = the_controller;
 	int			retval;
 
 	if (!driver
 			|| driver->speed < USB_SPEED_FULL
-			|| !driver->bind
+			|| !bind
 			|| !driver->disconnect
 			|| !driver->setup)
 		return -EINVAL;
@@ -1308,7 +1309,7 @@ fail:
 		dev->gadget.dev.driver = NULL;
 		return retval;
 	}
-	retval = driver->bind(&dev->gadget);
+	retval = bind(&dev->gadget);
 	if (retval) {
 		DMSG("bind to driver %s --> error %d\n",
 				driver->driver.name, retval);
@@ -1338,7 +1339,7 @@ fail:
 bind_fail:
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 static void
 stop_activity(struct pxa25x_udc *dev, struct usb_gadget_driver *driver)
diff --git a/drivers/usb/gadget/pxa27x_udc.c b/drivers/usb/gadget/pxa27x_udc.c
index 980762453a9c..027d66f81620 100644
--- a/drivers/usb/gadget/pxa27x_udc.c
+++ b/drivers/usb/gadget/pxa27x_udc.c
@@ -1792,8 +1792,9 @@ static void udc_enable(struct pxa_udc *udc)
 }
 
 /**
- * usb_gadget_register_driver - Register gadget driver
+ * usb_gadget_probe_driver - Register gadget driver
  * @driver: gadget driver
+ * @bind: bind function
  *
  * When a driver is successfully registered, it will receive control requests
  * including set_configuration(), which enables non-control requests.  Then
@@ -1805,12 +1806,13 @@ static void udc_enable(struct pxa_udc *udc)
  *
  * Returns 0 if no error, -EINVAL, -ENODEV, -EBUSY otherwise
  */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct pxa_udc *udc = the_controller;
 	int retval;
 
-	if (!driver || driver->speed < USB_SPEED_FULL || !driver->bind
+	if (!driver || driver->speed < USB_SPEED_FULL || !bind
 			|| !driver->disconnect || !driver->setup)
 		return -EINVAL;
 	if (!udc)
@@ -1828,7 +1830,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 		dev_err(udc->dev, "device_add error %d\n", retval);
 		goto add_fail;
 	}
-	retval = driver->bind(&udc->gadget);
+	retval = bind(&udc->gadget);
 	if (retval) {
 		dev_err(udc->dev, "bind to driver %s --> error %d\n",
 			driver->driver.name, retval);
@@ -1859,7 +1861,7 @@ add_fail:
 	udc->gadget.dev.driver = NULL;
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 
 /**
diff --git a/drivers/usb/gadget/r8a66597-udc.c b/drivers/usb/gadget/r8a66597-udc.c
index 2456ccd9965e..95092151f901 100644
--- a/drivers/usb/gadget/r8a66597-udc.c
+++ b/drivers/usb/gadget/r8a66597-udc.c
@@ -1405,14 +1405,15 @@ static struct usb_ep_ops r8a66597_ep_ops = {
 /*-------------------------------------------------------------------------*/
 static struct r8a66597 *the_controller;
 
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct r8a66597 *r8a66597 = the_controller;
 	int retval;
 
 	if (!driver
 			|| driver->speed != USB_SPEED_HIGH
-			|| !driver->bind
+			|| !bind
 			|| !driver->setup)
 		return -EINVAL;
 	if (!r8a66597)
@@ -1431,7 +1432,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 		goto error;
 	}
 
-	retval = driver->bind(&r8a66597->gadget);
+	retval = bind(&r8a66597->gadget);
 	if (retval) {
 		printk(KERN_ERR "bind to driver error (%d)\n", retval);
 		device_del(&r8a66597->gadget.dev);
@@ -1456,7 +1457,7 @@ error:
 
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
 {
diff --git a/drivers/usb/gadget/s3c-hsotg.c b/drivers/usb/gadget/s3c-hsotg.c
index a229744a8c7d..ef825c3baed9 100644
--- a/drivers/usb/gadget/s3c-hsotg.c
+++ b/drivers/usb/gadget/s3c-hsotg.c
@@ -2523,7 +2523,8 @@ static int s3c_hsotg_corereset(struct s3c_hsotg *hsotg)
 	return 0;
 }
 
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct s3c_hsotg *hsotg = our_hsotg;
 	int ret;
@@ -2543,7 +2544,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 		dev_err(hsotg->dev, "%s: bad speed\n", __func__);
 	}
 
-	if (!driver->bind || !driver->setup) {
+	if (!bind || !driver->setup) {
 		dev_err(hsotg->dev, "%s: missing entry points\n", __func__);
 		return -EINVAL;
 	}
@@ -2562,7 +2563,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 		goto err;
 	}
 
-	ret = driver->bind(&hsotg->gadget);
+	ret = bind(&hsotg->gadget);
 	if (ret) {
 		dev_err(hsotg->dev, "failed bind %s\n", driver->driver.name);
 
@@ -2687,7 +2688,7 @@ err:
 	hsotg->gadget.dev.driver = NULL;
 	return ret;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 int usb_gadget_unregister_driver(struct usb_gadget_driver *driver)
 {
diff --git a/drivers/usb/gadget/s3c2410_udc.c b/drivers/usb/gadget/s3c2410_udc.c
index ea2b3c7ebee5..c2448950a8d8 100644
--- a/drivers/usb/gadget/s3c2410_udc.c
+++ b/drivers/usb/gadget/s3c2410_udc.c
@@ -1632,15 +1632,15 @@ static void s3c2410_udc_enable(struct s3c2410_udc *dev)
 }
 
 /*
- *	usb_gadget_register_driver
+ *	usb_gadget_probe_driver
  */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	struct s3c2410_udc *udc = the_controller;
 	int		retval;
 
-	dprintk(DEBUG_NORMAL, "usb_gadget_register_driver() '%s'\n",
-		driver->driver.name);
+	dprintk(DEBUG_NORMAL, "%s() '%s'\n", __func__, driver->driver.name);
 
 	/* Sanity checks */
 	if (!udc)
@@ -1649,10 +1649,9 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	if (udc->driver)
 		return -EBUSY;
 
-	if (!driver->bind || !driver->setup
-			|| driver->speed < USB_SPEED_FULL) {
+	if (!bind || !driver->setup || driver->speed < USB_SPEED_FULL) {
 		printk(KERN_ERR "Invalid driver: bind %p setup %p speed %d\n",
-			driver->bind, driver->setup, driver->speed);
+			bind, driver->setup, driver->speed);
 		return -EINVAL;
 	}
 #if defined(MODULE)
@@ -1675,7 +1674,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	dprintk(DEBUG_NORMAL, "binding gadget driver '%s'\n",
 		driver->driver.name);
 
-	if ((retval = driver->bind (&udc->gadget)) != 0) {
+	if ((retval = bind(&udc->gadget)) != 0) {
 		device_del(&udc->gadget.dev);
 		goto register_error;
 	}
@@ -1690,6 +1689,7 @@ register_error:
 	udc->gadget.dev.driver = NULL;
 	return retval;
 }
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 /*
  *	usb_gadget_unregister_driver
@@ -2049,7 +2049,6 @@ static void __exit udc_exit(void)
 }
 
 EXPORT_SYMBOL(usb_gadget_unregister_driver);
-EXPORT_SYMBOL(usb_gadget_register_driver);
 
 module_init(udc_init);
 module_exit(udc_exit);
diff --git a/drivers/usb/musb/musb_gadget.c b/drivers/usb/musb/musb_gadget.c
index d065e23f123e..ecd5f8cffcbf 100644
--- a/drivers/usb/musb/musb_gadget.c
+++ b/drivers/usb/musb/musb_gadget.c
@@ -1699,9 +1699,11 @@ void musb_gadget_cleanup(struct musb *musb)
  * -ENOMEM no memeory to perform the operation
  *
  * @param driver the gadget driver
+ * @param bind the driver's bind function
  * @return <0 if error, 0 if everything is fine
  */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver)
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *))
 {
 	int retval;
 	unsigned long flags;
@@ -1709,8 +1711,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	if (!driver
 			|| driver->speed != USB_SPEED_HIGH
-			|| !driver->bind
-			|| !driver->setup)
+			|| !bind || !driver->setup)
 		return -EINVAL;
 
 	/* driver must be initialized to support peripheral mode */
@@ -1738,7 +1739,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 	spin_unlock_irqrestore(&musb->lock, flags);
 
 	if (retval == 0) {
-		retval = driver->bind(&musb->g);
+		retval = bind(&musb->g);
 		if (retval != 0) {
 			DBG(3, "bind to driver %s failed --> %d\n",
 					driver->driver.name, retval);
@@ -1786,7 +1787,7 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	return retval;
 }
-EXPORT_SYMBOL(usb_gadget_register_driver);
+EXPORT_SYMBOL(usb_gadget_probe_driver);
 
 static void stop_activity(struct musb *musb, struct usb_gadget_driver *driver)
 {
diff --git a/include/linux/usb/gadget.h b/include/linux/usb/gadget.h
index d3ef42d7d2f0..006412ce2303 100644
--- a/include/linux/usb/gadget.h
+++ b/include/linux/usb/gadget.h
@@ -705,11 +705,6 @@ static inline int usb_gadget_disconnect(struct usb_gadget *gadget)
  * struct usb_gadget_driver - driver for usb 'slave' devices
  * @function: String describing the gadget's function
  * @speed: Highest speed the driver handles.
- * @bind: Invoked when the driver is bound to a gadget, usually
- *	after registering the driver.
- *	At that point, ep0 is fully initialized, and ep_list holds
- *	the currently-available endpoints.
- *	Called in a context that permits sleeping.
  * @setup: Invoked for ep0 control requests that aren't handled by
  *	the hardware level driver. Most calls must be handled by
  *	the gadget driver, including descriptor and configuration
@@ -774,7 +769,6 @@ static inline int usb_gadget_disconnect(struct usb_gadget *gadget)
 struct usb_gadget_driver {
 	char			*function;
 	enum usb_device_speed	speed;
-	int			(*bind)(struct usb_gadget *);
 	void			(*unbind)(struct usb_gadget *);
 	int			(*setup)(struct usb_gadget *,
 					const struct usb_ctrlrequest *);
@@ -798,17 +792,19 @@ struct usb_gadget_driver {
  */
 
 /**
- * usb_gadget_register_driver - register a gadget driver
- * @driver:the driver being registered
+ * usb_gadget_probe_driver - probe a gadget driver
+ * @driver: the driver being registered
+ * @bind: the driver's bind callback
  * Context: can sleep
  *
  * Call this in your gadget driver's module initialization function,
  * to tell the underlying usb controller driver about your driver.
- * The driver's bind() function will be called to bind it to a
- * gadget before this registration call returns.  It's expected that
- * the bind() functions will be in init sections.
+ * The @bind() function will be called to bind it to a gadget before this
+ * registration call returns.  It's expected that the @bind() function will
+ * be in init sections.
  */
-int usb_gadget_register_driver(struct usb_gadget_driver *driver);
+int usb_gadget_probe_driver(struct usb_gadget_driver *driver,
+		int (*bind)(struct usb_gadget *));
 
 /**
  * usb_gadget_unregister_driver - unregister a gadget driver
-- 
cgit v1.2.3


From 07a18bd716ed5dea336429404b132568cfaaef95 Mon Sep 17 00:00:00 2001
From: Michal Nazarewicz <m.nazarewicz@samsung.com>
Date: Thu, 12 Aug 2010 17:43:54 +0200
Subject: usb gadget: don't save bind callback in struct usb_composite_driver
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The bind function is most of the time only called at init time so there
is no need to save a pointer to it in the composite driver structure.

This fixes many section mismatches reported by modpost.

Signed-off-by: Michał Nazarewicz <m.nazarewicz@samsung.com>
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/audio.c        |  3 +--
 drivers/usb/gadget/cdc2.c         |  3 +--
 drivers/usb/gadget/composite.c    | 15 +++++++++++----
 drivers/usb/gadget/ether.c        |  3 +--
 drivers/usb/gadget/g_ffs.c        |  3 +--
 drivers/usb/gadget/hid.c          |  3 +--
 drivers/usb/gadget/mass_storage.c |  3 +--
 drivers/usb/gadget/multi.c        |  3 +--
 drivers/usb/gadget/nokia.c        |  3 +--
 drivers/usb/gadget/serial.c       |  3 +--
 drivers/usb/gadget/webcam.c       |  3 +--
 drivers/usb/gadget/zero.c         |  3 +--
 include/linux/usb/composite.h     | 19 +++++--------------
 13 files changed, 27 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/gadget/audio.c b/drivers/usb/gadget/audio.c
index a62af7b59094..5a65fbb4e20b 100644
--- a/drivers/usb/gadget/audio.c
+++ b/drivers/usb/gadget/audio.c
@@ -166,13 +166,12 @@ static struct usb_composite_driver audio_driver = {
 	.name		= "g_audio",
 	.dev		= &device_desc,
 	.strings	= audio_strings,
-	.bind		= audio_bind,
 	.unbind		= __exit_p(audio_unbind),
 };
 
 static int __init init(void)
 {
-	return usb_composite_register(&audio_driver);
+	return usb_composite_probe(&audio_driver, audio_bind);
 }
 module_init(init);
 
diff --git a/drivers/usb/gadget/cdc2.c b/drivers/usb/gadget/cdc2.c
index 928137d3dbdc..1f2a9b1e4f2d 100644
--- a/drivers/usb/gadget/cdc2.c
+++ b/drivers/usb/gadget/cdc2.c
@@ -245,7 +245,6 @@ static struct usb_composite_driver cdc_driver = {
 	.name		= "g_cdc",
 	.dev		= &device_desc,
 	.strings	= dev_strings,
-	.bind		= cdc_bind,
 	.unbind		= __exit_p(cdc_unbind),
 };
 
@@ -255,7 +254,7 @@ MODULE_LICENSE("GPL");
 
 static int __init init(void)
 {
-	return usb_composite_register(&cdc_driver);
+	return usb_composite_probe(&cdc_driver, cdc_bind);
 }
 module_init(init);
 
diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c
index a3009bf01229..c531a7e05f1e 100644
--- a/drivers/usb/gadget/composite.c
+++ b/drivers/usb/gadget/composite.c
@@ -40,6 +40,7 @@
 #define USB_BUFSIZ	1024
 
 static struct usb_composite_driver *composite;
+static int (*composite_gadget_bind)(struct usb_composite_dev *cdev);
 
 /* Some systems will need runtime overrides for the  product identifers
  * published in the device descriptor, either numbers or strings or both.
@@ -1115,7 +1116,7 @@ static int composite_bind(struct usb_gadget *gadget)
 	 * serial number), register function drivers, potentially update
 	 * power state and consumption, etc
 	 */
-	status = composite->bind(cdev);
+	status = composite_gadget_bind(cdev);
 	if (status < 0)
 		goto fail;
 
@@ -1227,8 +1228,12 @@ static struct usb_gadget_driver composite_driver = {
 };
 
 /**
- * usb_composite_register() - register a composite driver
+ * usb_composite_probe() - register a composite driver
  * @driver: the driver to register
+ * @bind: the callback used to allocate resources that are shared across the
+ *	whole device, such as string IDs, and add its configurations using
+ *	@usb_add_config().  This may fail by returning a negative errno
+ *	value; it should return zero on successful initialization.
  * Context: single threaded during gadget setup
  *
  * This function is used to register drivers using the composite driver
@@ -1241,9 +1246,10 @@ static struct usb_gadget_driver composite_driver = {
  * while it was binding.  That would usually be done in order to wait for
  * some userspace participation.
  */
-int usb_composite_register(struct usb_composite_driver *driver)
+extern int usb_composite_probe(struct usb_composite_driver *driver,
+			       int (*bind)(struct usb_composite_dev *cdev))
 {
-	if (!driver || !driver->dev || !driver->bind || composite)
+	if (!driver || !driver->dev || !bind || composite)
 		return -EINVAL;
 
 	if (!driver->iProduct)
@@ -1253,6 +1259,7 @@ int usb_composite_register(struct usb_composite_driver *driver)
 	composite_driver.function =  (char *) driver->name;
 	composite_driver.driver.name = driver->name;
 	composite = driver;
+	composite_gadget_bind = bind;
 
 	return usb_gadget_probe_driver(&composite_driver, composite_bind);
 }
diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c
index 400f80372d93..33076bca9936 100644
--- a/drivers/usb/gadget/ether.c
+++ b/drivers/usb/gadget/ether.c
@@ -402,7 +402,6 @@ static struct usb_composite_driver eth_driver = {
 	.name		= "g_ether",
 	.dev		= &device_desc,
 	.strings	= dev_strings,
-	.bind		= eth_bind,
 	.unbind		= __exit_p(eth_unbind),
 };
 
@@ -412,7 +411,7 @@ MODULE_LICENSE("GPL");
 
 static int __init init(void)
 {
-	return usb_composite_register(&eth_driver);
+	return usb_composite_probe(&eth_driver, eth_bind);
 }
 module_init(init);
 
diff --git a/drivers/usb/gadget/g_ffs.c b/drivers/usb/gadget/g_ffs.c
index 52fd3fa0d130..9fcb15879506 100644
--- a/drivers/usb/gadget/g_ffs.c
+++ b/drivers/usb/gadget/g_ffs.c
@@ -147,7 +147,6 @@ static struct usb_composite_driver gfs_driver = {
 	.name		= DRIVER_NAME,
 	.dev		= &gfs_dev_desc,
 	.strings	= gfs_dev_strings,
-	.bind		= gfs_bind,
 	.unbind		= gfs_unbind,
 	.iProduct	= DRIVER_DESC,
 };
@@ -187,7 +186,7 @@ static int functionfs_ready_callback(struct ffs_data *ffs)
 		return -EBUSY;
 
 	gfs_ffs_data = ffs;
-	ret = usb_composite_register(&gfs_driver);
+	ret = usb_composite_probe(&gfs_driver, gfs_bind);
 	if (unlikely(ret < 0))
 		clear_bit(0, &gfs_registered);
 	return ret;
diff --git a/drivers/usb/gadget/hid.c b/drivers/usb/gadget/hid.c
index 775722686ed8..77f495212fb9 100644
--- a/drivers/usb/gadget/hid.c
+++ b/drivers/usb/gadget/hid.c
@@ -256,7 +256,6 @@ static struct usb_composite_driver hidg_driver = {
 	.name		= "g_hid",
 	.dev		= &device_desc,
 	.strings	= dev_strings,
-	.bind		= hid_bind,
 	.unbind		= __exit_p(hid_unbind),
 };
 
@@ -282,7 +281,7 @@ static int __init hidg_init(void)
 	if (status < 0)
 		return status;
 
-	status = usb_composite_register(&hidg_driver);
+	status = usb_composite_probe(&hidg_driver, hid_bind);
 	if (status < 0)
 		platform_driver_unregister(&hidg_plat_driver);
 
diff --git a/drivers/usb/gadget/mass_storage.c b/drivers/usb/gadget/mass_storage.c
index 05e9bd330348..a5e4a777d922 100644
--- a/drivers/usb/gadget/mass_storage.c
+++ b/drivers/usb/gadget/mass_storage.c
@@ -169,7 +169,6 @@ static int __init msg_bind(struct usb_composite_dev *cdev)
 static struct usb_composite_driver msg_driver = {
 	.name		= "g_mass_storage",
 	.dev		= &msg_device_desc,
-	.bind		= msg_bind,
 	.iProduct	= DRIVER_DESC,
 	.needs_serial	= 1,
 };
@@ -180,7 +179,7 @@ MODULE_LICENSE("GPL");
 
 static int __init msg_init(void)
 {
-	return usb_composite_register(&msg_driver);
+	return usb_composite_probe(&msg_driver, msg_bind);
 }
 module_init(msg_init);
 
diff --git a/drivers/usb/gadget/multi.c b/drivers/usb/gadget/multi.c
index ca51661d71db..91170a02a9a3 100644
--- a/drivers/usb/gadget/multi.c
+++ b/drivers/usb/gadget/multi.c
@@ -353,7 +353,6 @@ static struct usb_composite_driver multi_driver = {
 	.name		= "g_multi",
 	.dev		= &device_desc,
 	.strings	= dev_strings,
-	.bind		= multi_bind,
 	.unbind		= __exit_p(multi_unbind),
 	.iProduct	= DRIVER_DESC,
 	.needs_serial	= 1,
@@ -362,7 +361,7 @@ static struct usb_composite_driver multi_driver = {
 
 static int __init multi_init(void)
 {
-	return usb_composite_register(&multi_driver);
+	return usb_composite_probe(&multi_driver, multi_bind);
 }
 module_init(multi_init);
 
diff --git a/drivers/usb/gadget/nokia.c b/drivers/usb/gadget/nokia.c
index 7d6b66a85724..8aec728882a0 100644
--- a/drivers/usb/gadget/nokia.c
+++ b/drivers/usb/gadget/nokia.c
@@ -241,13 +241,12 @@ static struct usb_composite_driver nokia_driver = {
 	.name		= "g_nokia",
 	.dev		= &device_desc,
 	.strings	= dev_strings,
-	.bind		= nokia_bind,
 	.unbind		= __exit_p(nokia_unbind),
 };
 
 static int __init nokia_init(void)
 {
-	return usb_composite_register(&nokia_driver);
+	return usb_composite_probe(&nokia_driver, nokia_bind);
 }
 module_init(nokia_init);
 
diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c
index f46a60962dab..0b81d7b741d1 100644
--- a/drivers/usb/gadget/serial.c
+++ b/drivers/usb/gadget/serial.c
@@ -242,7 +242,6 @@ static struct usb_composite_driver gserial_driver = {
 	.name		= "g_serial",
 	.dev		= &device_desc,
 	.strings	= dev_strings,
-	.bind		= gs_bind,
 };
 
 static int __init init(void)
@@ -271,7 +270,7 @@ static int __init init(void)
 	}
 	strings_dev[STRING_DESCRIPTION_IDX].s = serial_config_driver.label;
 
-	return usb_composite_register(&gserial_driver);
+	return usb_composite_probe(&gserial_driver, gs_bind);
 }
 module_init(init);
 
diff --git a/drivers/usb/gadget/webcam.c b/drivers/usb/gadget/webcam.c
index 288d21155abe..de65b8078e05 100644
--- a/drivers/usb/gadget/webcam.c
+++ b/drivers/usb/gadget/webcam.c
@@ -373,14 +373,13 @@ static struct usb_composite_driver webcam_driver = {
 	.name		= "g_webcam",
 	.dev		= &webcam_device_descriptor,
 	.strings	= webcam_device_strings,
-	.bind		= webcam_bind,
 	.unbind		= webcam_unbind,
 };
 
 static int __init
 webcam_init(void)
 {
-	return usb_composite_register(&webcam_driver);
+	return usb_composite_probe(&webcam_driver, webcam_bind);
 }
 
 static void __exit
diff --git a/drivers/usb/gadget/zero.c b/drivers/usb/gadget/zero.c
index 807280d069f9..6d16db9d9d2d 100644
--- a/drivers/usb/gadget/zero.c
+++ b/drivers/usb/gadget/zero.c
@@ -340,7 +340,6 @@ static struct usb_composite_driver zero_driver = {
 	.name		= "zero",
 	.dev		= &device_desc,
 	.strings	= dev_strings,
-	.bind		= zero_bind,
 	.unbind		= zero_unbind,
 	.suspend	= zero_suspend,
 	.resume		= zero_resume,
@@ -351,7 +350,7 @@ MODULE_LICENSE("GPL");
 
 static int __init init(void)
 {
-	return usb_composite_register(&zero_driver);
+	return usb_composite_probe(&zero_driver, zero_bind);
 }
 module_init(init);
 
diff --git a/include/linux/usb/composite.h b/include/linux/usb/composite.h
index a78e813d27e4..e28b6626802c 100644
--- a/include/linux/usb/composite.h
+++ b/include/linux/usb/composite.h
@@ -248,11 +248,7 @@ int usb_add_config(struct usb_composite_dev *,
  *	and language IDs provided in control requests
  * @needs_serial: set to 1 if the gadget needs userspace to provide
  * 	a serial number.  If one is not provided, warning will be printed.
- * @bind: (REQUIRED) Used to allocate resources that are shared across the
- *	whole device, such as string IDs, and add its configurations using
- *	@usb_add_config().  This may fail by returning a negative errno
- *	value; it should return zero on successful initialization.
- * @unbind: Reverses @bind(); called as a side effect of unregistering
+ * @unbind: Reverses bind; called as a side effect of unregistering
  *	this driver.
  * @disconnect: optional driver disconnect method
  * @suspend: Notifies when the host stops sending USB traffic,
@@ -263,7 +259,7 @@ int usb_add_config(struct usb_composite_dev *,
  * Devices default to reporting self powered operation.  Devices which rely
  * on bus powered operation should report this in their @bind() method.
  *
- * Before returning from @bind, various fields in the template descriptor
+ * Before returning from bind, various fields in the template descriptor
  * may be overridden.  These include the idVendor/idProduct/bcdDevice values
  * normally to bind the appropriate host side driver, and the three strings
  * (iManufacturer, iProduct, iSerialNumber) normally used to provide user
@@ -279,12 +275,6 @@ struct usb_composite_driver {
 	struct usb_gadget_strings		**strings;
 	unsigned		needs_serial:1;
 
-	/* REVISIT:  bind() functions can be marked __init, which
-	 * makes trouble for section mismatch analysis.  See if
-	 * we can't restructure things to avoid mismatching...
-	 */
-
-	int			(*bind)(struct usb_composite_dev *);
 	int			(*unbind)(struct usb_composite_dev *);
 
 	void			(*disconnect)(struct usb_composite_dev *);
@@ -294,8 +284,9 @@ struct usb_composite_driver {
 	void			(*resume)(struct usb_composite_dev *);
 };
 
-extern int usb_composite_register(struct usb_composite_driver *);
-extern void usb_composite_unregister(struct usb_composite_driver *);
+extern int usb_composite_probe(struct usb_composite_driver *driver,
+			       int (*bind)(struct usb_composite_dev *cdev));
+extern void usb_composite_unregister(struct usb_composite_driver *driver);
 
 
 /**
-- 
cgit v1.2.3


From c9bfff9c98671ad50e4abbfe1ab606a9957f7539 Mon Sep 17 00:00:00 2001
From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Date: Thu, 12 Aug 2010 17:43:55 +0200
Subject: usb gadget: don't save bind callback in struct usb_configuration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The bind function is most of the time only called at init time so there
is no need to save a pointer to it in the configuration structure.

This fixes many section mismatches reported by modpost.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
[m.nazarewicz@samsung.com: updated for -next]
Signed-off-by: Michał Nazarewicz <m.nazarewicz@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/audio.c        |  3 +--
 drivers/usb/gadget/cdc2.c         |  3 +--
 drivers/usb/gadget/composite.c    | 14 ++++++++------
 drivers/usb/gadget/ether.c        |  7 +++----
 drivers/usb/gadget/f_loopback.c   |  3 +--
 drivers/usb/gadget/f_sourcesink.c |  3 +--
 drivers/usb/gadget/g_ffs.c        |  3 +--
 drivers/usb/gadget/hid.c          |  3 +--
 drivers/usb/gadget/mass_storage.c |  3 +--
 drivers/usb/gadget/multi.c        | 10 ++++------
 drivers/usb/gadget/nokia.c        |  8 ++++----
 drivers/usb/gadget/serial.c       |  4 ++--
 drivers/usb/gadget/webcam.c       |  4 ++--
 include/linux/usb/composite.h     |  8 +++-----
 14 files changed, 33 insertions(+), 43 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/gadget/audio.c b/drivers/usb/gadget/audio.c
index 5a65fbb4e20b..93b999e49ef3 100644
--- a/drivers/usb/gadget/audio.c
+++ b/drivers/usb/gadget/audio.c
@@ -105,7 +105,6 @@ static int __init audio_do_config(struct usb_configuration *c)
 
 static struct usb_configuration audio_config_driver = {
 	.label			= DRIVER_DESC,
-	.bind			= audio_do_config,
 	.bConfigurationValue	= 1,
 	/* .iConfiguration = DYNAMIC */
 	.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
@@ -145,7 +144,7 @@ static int __init audio_bind(struct usb_composite_dev *cdev)
 	strings_dev[STRING_PRODUCT_IDX].id = status;
 	device_desc.iProduct = status;
 
-	status = usb_add_config(cdev, &audio_config_driver);
+	status = usb_add_config(cdev, &audio_config_driver, audio_do_config);
 	if (status < 0)
 		goto fail;
 
diff --git a/drivers/usb/gadget/cdc2.c b/drivers/usb/gadget/cdc2.c
index 1f2a9b1e4f2d..2720ab07ef1a 100644
--- a/drivers/usb/gadget/cdc2.c
+++ b/drivers/usb/gadget/cdc2.c
@@ -151,7 +151,6 @@ static int __init cdc_do_config(struct usb_configuration *c)
 
 static struct usb_configuration cdc_config_driver = {
 	.label			= "CDC Composite (ECM + ACM)",
-	.bind			= cdc_do_config,
 	.bConfigurationValue	= 1,
 	/* .iConfiguration = DYNAMIC */
 	.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
@@ -218,7 +217,7 @@ static int __init cdc_bind(struct usb_composite_dev *cdev)
 	device_desc.iProduct = status;
 
 	/* register our configuration */
-	status = usb_add_config(cdev, &cdc_config_driver);
+	status = usb_add_config(cdev, &cdc_config_driver, cdc_do_config);
 	if (status < 0)
 		goto fail1;
 
diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c
index c531a7e05f1e..5e2bd7428424 100644
--- a/drivers/usb/gadget/composite.c
+++ b/drivers/usb/gadget/composite.c
@@ -474,18 +474,20 @@ done:
  * usb_add_config() - add a configuration to a device.
  * @cdev: wraps the USB gadget
  * @config: the configuration, with bConfigurationValue assigned
+ * @bind: the configuration's bind function
  * Context: single threaded during gadget setup
  *
- * One of the main tasks of a composite driver's bind() routine is to
+ * One of the main tasks of a composite @bind() routine is to
  * add each of the configurations it supports, using this routine.
  *
- * This function returns the value of the configuration's bind(), which
+ * This function returns the value of the configuration's @bind(), which
  * is zero for success else a negative errno value.  Binding configurations
  * assigns global resources including string IDs, and per-configuration
  * resources such as interface IDs and endpoints.
  */
 int usb_add_config(struct usb_composite_dev *cdev,
-		struct usb_configuration *config)
+		struct usb_configuration *config,
+		int (*bind)(struct usb_configuration *))
 {
 	int				status = -EINVAL;
 	struct usb_configuration	*c;
@@ -494,7 +496,7 @@ int usb_add_config(struct usb_composite_dev *cdev,
 			config->bConfigurationValue,
 			config->label, config);
 
-	if (!config->bConfigurationValue || !config->bind)
+	if (!config->bConfigurationValue || !bind)
 		goto done;
 
 	/* Prevent duplicate configuration identifiers */
@@ -511,7 +513,7 @@ int usb_add_config(struct usb_composite_dev *cdev,
 	INIT_LIST_HEAD(&config->functions);
 	config->next_interface_id = 0;
 
-	status = config->bind(config);
+	status = bind(config);
 	if (status < 0) {
 		list_del(&config->list);
 		config->cdev = NULL;
@@ -537,7 +539,7 @@ int usb_add_config(struct usb_composite_dev *cdev,
 		}
 	}
 
-	/* set_alt(), or next config->bind(), sets up
+	/* set_alt(), or next bind(), sets up
 	 * ep->driver_data as needed.
 	 */
 	usb_ep_autoconfig_reset(cdev->gadget);
diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c
index 33076bca9936..1690c9d68256 100644
--- a/drivers/usb/gadget/ether.c
+++ b/drivers/usb/gadget/ether.c
@@ -251,7 +251,6 @@ static int __init rndis_do_config(struct usb_configuration *c)
 
 static struct usb_configuration rndis_config_driver = {
 	.label			= "RNDIS",
-	.bind			= rndis_do_config,
 	.bConfigurationValue	= 2,
 	/* .iConfiguration = DYNAMIC */
 	.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
@@ -289,7 +288,6 @@ static int __init eth_do_config(struct usb_configuration *c)
 
 static struct usb_configuration eth_config_driver = {
 	/* .label = f(hardware) */
-	.bind			= eth_do_config,
 	.bConfigurationValue	= 1,
 	/* .iConfiguration = DYNAMIC */
 	.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
@@ -373,12 +371,13 @@ static int __init eth_bind(struct usb_composite_dev *cdev)
 
 	/* register our configuration(s); RNDIS first, if it's used */
 	if (has_rndis()) {
-		status = usb_add_config(cdev, &rndis_config_driver);
+		status = usb_add_config(cdev, &rndis_config_driver,
+				rndis_do_config);
 		if (status < 0)
 			goto fail;
 	}
 
-	status = usb_add_config(cdev, &eth_config_driver);
+	status = usb_add_config(cdev, &eth_config_driver, eth_do_config);
 	if (status < 0)
 		goto fail;
 
diff --git a/drivers/usb/gadget/f_loopback.c b/drivers/usb/gadget/f_loopback.c
index e91d1b16d9be..b37960f9e753 100644
--- a/drivers/usb/gadget/f_loopback.c
+++ b/drivers/usb/gadget/f_loopback.c
@@ -349,7 +349,6 @@ static int __init loopback_bind_config(struct usb_configuration *c)
 static struct usb_configuration loopback_driver = {
 	.label		= "loopback",
 	.strings	= loopback_strings,
-	.bind		= loopback_bind_config,
 	.bConfigurationValue = 2,
 	.bmAttributes	= USB_CONFIG_ATT_SELFPOWER,
 	/* .iConfiguration = DYNAMIC */
@@ -382,5 +381,5 @@ int __init loopback_add(struct usb_composite_dev *cdev, bool autoresume)
 		loopback_driver.bmAttributes |= USB_CONFIG_ATT_WAKEUP;
 	}
 
-	return usb_add_config(cdev, &loopback_driver);
+	return usb_add_config(cdev, &loopback_driver, loopback_bind_config);
 }
diff --git a/drivers/usb/gadget/f_sourcesink.c b/drivers/usb/gadget/f_sourcesink.c
index 6d3cc443d914..e403a534dd55 100644
--- a/drivers/usb/gadget/f_sourcesink.c
+++ b/drivers/usb/gadget/f_sourcesink.c
@@ -498,7 +498,6 @@ unknown:
 static struct usb_configuration sourcesink_driver = {
 	.label		= "source/sink",
 	.strings	= sourcesink_strings,
-	.bind		= sourcesink_bind_config,
 	.setup		= sourcesink_setup,
 	.bConfigurationValue = 3,
 	.bmAttributes	= USB_CONFIG_ATT_SELFPOWER,
@@ -532,5 +531,5 @@ int __init sourcesink_add(struct usb_composite_dev *cdev, bool autoresume)
 		sourcesink_driver.bmAttributes |= USB_CONFIG_ATT_WAKEUP;
 	}
 
-	return usb_add_config(cdev, &sourcesink_driver);
+	return usb_add_config(cdev, &sourcesink_driver, sourcesink_bind_config);
 }
diff --git a/drivers/usb/gadget/g_ffs.c b/drivers/usb/gadget/g_ffs.c
index 9fcb15879506..af75e3620849 100644
--- a/drivers/usb/gadget/g_ffs.c
+++ b/drivers/usb/gadget/g_ffs.c
@@ -234,11 +234,10 @@ static int gfs_bind(struct usb_composite_dev *cdev)
 
 		c->c.label			= gfs_strings[i].s;
 		c->c.iConfiguration		= gfs_strings[i].id;
-		c->c.bind			= gfs_do_config;
 		c->c.bConfigurationValue	= 1 + i;
 		c->c.bmAttributes		= USB_CONFIG_ATT_SELFPOWER;
 
-		ret = usb_add_config(cdev, &c->c);
+		ret = usb_add_config(cdev, &c->c, gfs_do_config);
 		if (unlikely(ret < 0))
 			goto error_unbind;
 	}
diff --git a/drivers/usb/gadget/hid.c b/drivers/usb/gadget/hid.c
index 77f495212fb9..2523e54097bd 100644
--- a/drivers/usb/gadget/hid.c
+++ b/drivers/usb/gadget/hid.c
@@ -148,7 +148,6 @@ static int __init do_config(struct usb_configuration *c)
 
 static struct usb_configuration config_driver = {
 	.label			= "HID Gadget",
-	.bind			= do_config,
 	.bConfigurationValue	= 1,
 	/* .iConfiguration = DYNAMIC */
 	.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
@@ -201,7 +200,7 @@ static int __init hid_bind(struct usb_composite_dev *cdev)
 	device_desc.iProduct = status;
 
 	/* register our configuration */
-	status = usb_add_config(cdev, &config_driver);
+	status = usb_add_config(cdev, &config_driver, do_config);
 	if (status < 0)
 		return status;
 
diff --git a/drivers/usb/gadget/mass_storage.c b/drivers/usb/gadget/mass_storage.c
index a5e4a777d922..0769179dbdb0 100644
--- a/drivers/usb/gadget/mass_storage.c
+++ b/drivers/usb/gadget/mass_storage.c
@@ -141,7 +141,6 @@ static int __init msg_do_config(struct usb_configuration *c)
 
 static struct usb_configuration msg_config_driver = {
 	.label			= "Linux File-Backed Storage",
-	.bind			= msg_do_config,
 	.bConfigurationValue	= 1,
 	.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
 };
@@ -153,7 +152,7 @@ static int __init msg_bind(struct usb_composite_dev *cdev)
 {
 	int status;
 
-	status = usb_add_config(cdev, &msg_config_driver);
+	status = usb_add_config(cdev, &msg_config_driver, msg_do_config);
 	if (status < 0)
 		return status;
 
diff --git a/drivers/usb/gadget/multi.c b/drivers/usb/gadget/multi.c
index 91170a02a9a3..d9feced348e3 100644
--- a/drivers/usb/gadget/multi.c
+++ b/drivers/usb/gadget/multi.c
@@ -164,7 +164,7 @@ static u8 hostaddr[ETH_ALEN];
 
 #ifdef USB_ETH_RNDIS
 
-static __ref int rndis_do_config(struct usb_configuration *c)
+static __init int rndis_do_config(struct usb_configuration *c)
 {
 	int ret;
 
@@ -191,7 +191,6 @@ static __ref int rndis_do_config(struct usb_configuration *c)
 static int rndis_config_register(struct usb_composite_dev *cdev)
 {
 	static struct usb_configuration config = {
-		.bind			= rndis_do_config,
 		.bConfigurationValue	= MULTI_RNDIS_CONFIG_NUM,
 		.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
 	};
@@ -199,7 +198,7 @@ static int rndis_config_register(struct usb_composite_dev *cdev)
 	config.label          = strings_dev[MULTI_STRING_RNDIS_CONFIG_IDX].s;
 	config.iConfiguration = strings_dev[MULTI_STRING_RNDIS_CONFIG_IDX].id;
 
-	return usb_add_config(cdev, &config);
+	return usb_add_config(cdev, &config, rndis_do_config);
 }
 
 #else
@@ -216,7 +215,7 @@ static int rndis_config_register(struct usb_composite_dev *cdev)
 
 #ifdef CONFIG_USB_G_MULTI_CDC
 
-static __ref int cdc_do_config(struct usb_configuration *c)
+static __init int cdc_do_config(struct usb_configuration *c)
 {
 	int ret;
 
@@ -243,7 +242,6 @@ static __ref int cdc_do_config(struct usb_configuration *c)
 static int cdc_config_register(struct usb_composite_dev *cdev)
 {
 	static struct usb_configuration config = {
-		.bind			= cdc_do_config,
 		.bConfigurationValue	= MULTI_CDC_CONFIG_NUM,
 		.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
 	};
@@ -251,7 +249,7 @@ static int cdc_config_register(struct usb_composite_dev *cdev)
 	config.label          = strings_dev[MULTI_STRING_CDC_CONFIG_IDX].s;
 	config.iConfiguration = strings_dev[MULTI_STRING_CDC_CONFIG_IDX].id;
 
-	return usb_add_config(cdev, &config);
+	return usb_add_config(cdev, &config, cdc_do_config);
 }
 
 #else
diff --git a/drivers/usb/gadget/nokia.c b/drivers/usb/gadget/nokia.c
index 8aec728882a0..b5364f9d7cd2 100644
--- a/drivers/usb/gadget/nokia.c
+++ b/drivers/usb/gadget/nokia.c
@@ -135,7 +135,6 @@ static int __init nokia_bind_config(struct usb_configuration *c)
 
 static struct usb_configuration nokia_config_500ma_driver = {
 	.label		= "Bus Powered",
-	.bind		= nokia_bind_config,
 	.bConfigurationValue = 1,
 	/* .iConfiguration = DYNAMIC */
 	.bmAttributes	= USB_CONFIG_ATT_ONE,
@@ -144,7 +143,6 @@ static struct usb_configuration nokia_config_500ma_driver = {
 
 static struct usb_configuration nokia_config_100ma_driver = {
 	.label		= "Self Powered",
-	.bind		= nokia_bind_config,
 	.bConfigurationValue = 2,
 	/* .iConfiguration = DYNAMIC */
 	.bmAttributes	= USB_CONFIG_ATT_ONE | USB_CONFIG_ATT_SELFPOWER,
@@ -206,11 +204,13 @@ static int __init nokia_bind(struct usb_composite_dev *cdev)
 	}
 
 	/* finaly register the configuration */
-	status = usb_add_config(cdev, &nokia_config_500ma_driver);
+	status = usb_add_config(cdev, &nokia_config_500ma_driver,
+			nokia_bind_config);
 	if (status < 0)
 		goto err_usb;
 
-	status = usb_add_config(cdev, &nokia_config_100ma_driver);
+	status = usb_add_config(cdev, &nokia_config_100ma_driver,
+			nokia_bind_config);
 	if (status < 0)
 		goto err_usb;
 
diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c
index 0b81d7b741d1..1ac57a973aa9 100644
--- a/drivers/usb/gadget/serial.c
+++ b/drivers/usb/gadget/serial.c
@@ -155,7 +155,6 @@ static int __init serial_bind_config(struct usb_configuration *c)
 
 static struct usb_configuration serial_config_driver = {
 	/* .label = f(use_acm) */
-	.bind		= serial_bind_config,
 	/* .bConfigurationValue = f(use_acm) */
 	/* .iConfiguration = DYNAMIC */
 	.bmAttributes	= USB_CONFIG_ATT_SELFPOWER,
@@ -225,7 +224,8 @@ static int __init gs_bind(struct usb_composite_dev *cdev)
 	}
 
 	/* register our configuration */
-	status = usb_add_config(cdev, &serial_config_driver);
+	status = usb_add_config(cdev, &serial_config_driver,
+			serial_bind_config);
 	if (status < 0)
 		goto fail;
 
diff --git a/drivers/usb/gadget/webcam.c b/drivers/usb/gadget/webcam.c
index de65b8078e05..a5a0fdb808c7 100644
--- a/drivers/usb/gadget/webcam.c
+++ b/drivers/usb/gadget/webcam.c
@@ -317,7 +317,6 @@ webcam_config_bind(struct usb_configuration *c)
 
 static struct usb_configuration webcam_config_driver = {
 	.label			= webcam_config_label,
-	.bind			= webcam_config_bind,
 	.bConfigurationValue	= 1,
 	.iConfiguration		= 0, /* dynamic */
 	.bmAttributes		= USB_CONFIG_ATT_SELFPOWER,
@@ -354,7 +353,8 @@ webcam_bind(struct usb_composite_dev *cdev)
 	webcam_config_driver.iConfiguration = ret;
 
 	/* Register our configuration. */
-	if ((ret = usb_add_config(cdev, &webcam_config_driver)) < 0)
+	if ((ret = usb_add_config(cdev, &webcam_config_driver,
+					webcam_config_bind)) < 0)
 		goto error;
 
 	INFO(cdev, "Webcam Video Gadget\n");
diff --git a/include/linux/usb/composite.h b/include/linux/usb/composite.h
index e28b6626802c..3d29a7dcac2d 100644
--- a/include/linux/usb/composite.h
+++ b/include/linux/usb/composite.h
@@ -161,8 +161,6 @@ ep_choose(struct usb_gadget *g, struct usb_endpoint_descriptor *hs,
  *	and by language IDs provided in control requests.
  * @descriptors: Table of descriptors preceding all function descriptors.
  *	Examples include OTG and vendor-specific descriptors.
- * @bind: Called from @usb_add_config() to allocate resources unique to this
- *	configuration and to call @usb_add_function() for each function used.
  * @unbind: Reverses @bind; called as a side effect of unregistering the
  *	driver which added this configuration.
  * @setup: Used to delegate control requests that aren't handled by standard
@@ -207,8 +205,7 @@ struct usb_configuration {
 	 * we can't restructure things to avoid mismatching...
 	 */
 
-	/* configuration management:  bind/unbind */
-	int			(*bind)(struct usb_configuration *);
+	/* configuration management: unbind/setup */
 	void			(*unbind)(struct usb_configuration *);
 	int			(*setup)(struct usb_configuration *,
 					const struct usb_ctrlrequest *);
@@ -232,7 +229,8 @@ struct usb_configuration {
 };
 
 int usb_add_config(struct usb_composite_dev *,
-		struct usb_configuration *);
+		struct usb_configuration *,
+		int (*)(struct usb_configuration *));
 
 /**
  * struct usb_composite_driver - groups configurations into a gadget
-- 
cgit v1.2.3


From 5ea081785dde6041eb2f4acc2369abbb9099a981 Mon Sep 17 00:00:00 2001
From: Michal Nazarewicz <m.nazarewicz@samsung.com>
Date: Thu, 12 Aug 2010 17:43:56 +0200
Subject: init.h: add some more documentation to __ref* tags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The __ref* tags may have been confusing for new kernel
developers (I was confused by them for sure) so adding a few
more sentences to comment to clear things up for people who
see those for the first time.

Signed-off-by: Michal Nazarewicz <m.nazarewicz@samsung.com>
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/init.h | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/init.h b/include/linux/init.h
index de994304e0bb..577671c55153 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -46,16 +46,23 @@
 #define __exitdata	__section(.exit.data)
 #define __exit_call	__used __section(.exitcall.exit)
 
-/* modpost check for section mismatches during the kernel build.
+/*
+ * modpost check for section mismatches during the kernel build.
  * A section mismatch happens when there are references from a
  * code or data section to an init section (both code or data).
  * The init sections are (for most archs) discarded by the kernel
  * when early init has completed so all such references are potential bugs.
  * For exit sections the same issue exists.
+ *
  * The following markers are used for the cases where the reference to
  * the *init / *exit section (code or data) is valid and will teach
- * modpost not to issue a warning.
- * The markers follow same syntax rules as __init / __initdata. */
+ * modpost not to issue a warning.  Intended semantics is that a code or
+ * data tagged __ref* can reference code or data from init section without
+ * producing a warning (of course, no warning does not mean code is
+ * correct, so optimally document why the __ref is needed and why it's OK).
+ *
+ * The markers follow same syntax rules as __init / __initdata.
+ */
 #define __ref            __section(.ref.text) noinline
 #define __refdata        __section(.ref.data)
 #define __refconst       __section(.ref.rodata)
-- 
cgit v1.2.3


From d39a0edad60dc65cf4774ee732aa7a84cf35c27a Mon Sep 17 00:00:00 2001
From: Hao Wu <hao.wu@intel.com>
Date: Thu, 9 Sep 2010 22:35:39 +0100
Subject: USB OTG: Add common data structure for Intel MID Platform
 (Langwell/Penwell)

This patch adds one new header file for the common data structure used in
Intel Penwell/Langwell MID Platform OTG Transceiver drivers. After switched
to the common data structure, Langwell/Penwell OTG Transceiver driver will
provide an unified interface to host/client driver.

Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Hao Wu <hao.wu@intel.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/intel_mid_otg.h | 180 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 180 insertions(+)
 create mode 100644 include/linux/usb/intel_mid_otg.h

(limited to 'include/linux')

diff --git a/include/linux/usb/intel_mid_otg.h b/include/linux/usb/intel_mid_otg.h
new file mode 100644
index 000000000000..a0ccf795f362
--- /dev/null
+++ b/include/linux/usb/intel_mid_otg.h
@@ -0,0 +1,180 @@
+/*
+ * Intel MID (Langwell/Penwell) USB OTG Transceiver driver
+ * Copyright (C) 2008 - 2010, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope 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.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ */
+
+#ifndef __INTEL_MID_OTG_H
+#define __INTEL_MID_OTG_H
+
+#include <linux/pm.h>
+#include <linux/usb/otg.h>
+#include <linux/notifier.h>
+
+struct intel_mid_otg_xceiv;
+
+/* This is a common data structure for Intel MID platform to
+ * save values of the OTG state machine */
+struct otg_hsm {
+	/* Input */
+	int a_bus_resume;
+	int a_bus_suspend;
+	int a_conn;
+	int a_sess_vld;
+	int a_srp_det;
+	int a_vbus_vld;
+	int b_bus_resume;
+	int b_bus_suspend;
+	int b_conn;
+	int b_se0_srp;
+	int b_ssend_srp;
+	int b_sess_end;
+	int b_sess_vld;
+	int id;
+/* id values */
+#define ID_B		0x05
+#define ID_A		0x04
+#define ID_ACA_C	0x03
+#define ID_ACA_B	0x02
+#define ID_ACA_A	0x01
+	int power_up;
+	int adp_change;
+	int test_device;
+
+	/* Internal variables */
+	int a_set_b_hnp_en;
+	int b_srp_done;
+	int b_hnp_enable;
+	int hnp_poll_enable;
+
+	/* Timeout indicator for timers */
+	int a_wait_vrise_tmout;
+	int a_wait_bcon_tmout;
+	int a_aidl_bdis_tmout;
+	int a_bidl_adis_tmout;
+	int a_bidl_adis_tmr;
+	int a_wait_vfall_tmout;
+	int b_ase0_brst_tmout;
+	int b_bus_suspend_tmout;
+	int b_srp_init_tmout;
+	int b_srp_fail_tmout;
+	int b_srp_fail_tmr;
+	int b_adp_sense_tmout;
+
+	/* Informative variables */
+	int a_bus_drop;
+	int a_bus_req;
+	int a_clr_err;
+	int b_bus_req;
+	int a_suspend_req;
+	int b_bus_suspend_vld;
+
+	/* Output */
+	int drv_vbus;
+	int loc_conn;
+	int loc_sof;
+
+	/* Others */
+	int vbus_srp_up;
+};
+
+/* must provide ULPI access function to read/write registers implemented in
+ * ULPI address space */
+struct iotg_ulpi_access_ops {
+	int	(*read)(struct intel_mid_otg_xceiv *iotg, u8 reg, u8 *val);
+	int	(*write)(struct intel_mid_otg_xceiv *iotg, u8 reg, u8 val);
+};
+
+#define OTG_A_DEVICE	0x0
+#define OTG_B_DEVICE	0x1
+
+/*
+ * the Intel MID (Langwell/Penwell) otg transceiver driver needs to interact
+ * with device and host drivers to implement the USB OTG related feature. More
+ * function members are added based on otg_transceiver data structure for this
+ * purpose.
+ */
+struct intel_mid_otg_xceiv {
+	struct otg_transceiver	otg;
+	struct otg_hsm		hsm;
+
+	/* base address */
+	void __iomem		*base;
+
+	/* ops to access ulpi */
+	struct iotg_ulpi_access_ops	ulpi_ops;
+
+	/* atomic notifier for interrupt context */
+	struct atomic_notifier_head	iotg_notifier;
+
+	/* start/stop USB Host function */
+	int	(*start_host)(struct intel_mid_otg_xceiv *iotg);
+	int	(*stop_host)(struct intel_mid_otg_xceiv *iotg);
+
+	/* start/stop USB Peripheral function */
+	int	(*start_peripheral)(struct intel_mid_otg_xceiv *iotg);
+	int	(*stop_peripheral)(struct intel_mid_otg_xceiv *iotg);
+
+	/* start/stop ADP sense/probe function */
+	int	(*set_adp_probe)(struct intel_mid_otg_xceiv *iotg,
+					bool enabled, int dev);
+	int	(*set_adp_sense)(struct intel_mid_otg_xceiv *iotg,
+					bool enabled);
+
+#ifdef CONFIG_PM
+	/* suspend/resume USB host function */
+	int	(*suspend_host)(struct intel_mid_otg_xceiv *iotg,
+					pm_message_t message);
+	int	(*resume_host)(struct intel_mid_otg_xceiv *iotg);
+
+	int	(*suspend_peripheral)(struct intel_mid_otg_xceiv *iotg,
+					pm_message_t message);
+	int	(*resume_peripheral)(struct intel_mid_otg_xceiv *iotg);
+#endif
+
+};
+static inline
+struct intel_mid_otg_xceiv *otg_to_mid_xceiv(struct otg_transceiver *otg)
+{
+	return container_of(otg, struct intel_mid_otg_xceiv, otg);
+}
+
+#define MID_OTG_NOTIFY_CONNECT		0x0001
+#define MID_OTG_NOTIFY_DISCONN		0x0002
+#define MID_OTG_NOTIFY_HSUSPEND		0x0003
+#define MID_OTG_NOTIFY_HRESUME		0x0004
+#define MID_OTG_NOTIFY_CSUSPEND		0x0005
+#define MID_OTG_NOTIFY_CRESUME		0x0006
+#define MID_OTG_NOTIFY_HOSTADD		0x0007
+#define MID_OTG_NOTIFY_HOSTREMOVE	0x0008
+#define MID_OTG_NOTIFY_CLIENTADD	0x0009
+#define MID_OTG_NOTIFY_CLIENTREMOVE	0x000a
+
+static inline int
+intel_mid_otg_register_notifier(struct intel_mid_otg_xceiv *iotg,
+				struct notifier_block *nb)
+{
+	return atomic_notifier_chain_register(&iotg->iotg_notifier, nb);
+}
+
+static inline void
+intel_mid_otg_unregister_notifier(struct intel_mid_otg_xceiv *iotg,
+				struct notifier_block *nb)
+{
+	atomic_notifier_chain_unregister(&iotg->iotg_notifier, nb);
+}
+
+#endif /* __INTEL_MID_OTG_H */
-- 
cgit v1.2.3


From 56e9406ca22968e3c9dc27d6dc0f1825e13bfff9 Mon Sep 17 00:00:00 2001
From: Hao Wu <hao.wu@intel.com>
Date: Thu, 9 Sep 2010 22:35:54 +0100
Subject: USB OTG Langwell: Update OTG Kconfig and driver version.

This patch updated Kconfig for langwell otg transceiver driver.
Add ipc driver(INTEL_SCU_IPC) as a dependency. Driver version is
updated too.

Signed-off-by: Hao Wu <hao.wu@intel.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/otg/Kconfig          | 2 +-
 drivers/usb/otg/langwell_otg.c   | 4 ++--
 include/linux/usb/langwell_otg.h | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/otg/Kconfig b/drivers/usb/otg/Kconfig
index 299dfd2510cb..5ce07528cd0c 100644
--- a/drivers/usb/otg/Kconfig
+++ b/drivers/usb/otg/Kconfig
@@ -69,7 +69,7 @@ config NOP_USB_XCEIV
 
 config USB_LANGWELL_OTG
 	tristate "Intel Langwell USB OTG dual-role support"
-	depends on USB && X86_MRST
+	depends on USB && PCI && INTEL_SCU_IPC
 	select USB_OTG
 	select USB_OTG_UTILS
 	help
diff --git a/drivers/usb/otg/langwell_otg.c b/drivers/usb/otg/langwell_otg.c
index 879188086daf..bdc3ea66be69 100644
--- a/drivers/usb/otg/langwell_otg.c
+++ b/drivers/usb/otg/langwell_otg.c
@@ -1,6 +1,6 @@
 /*
  * Intel Langwell USB OTG transceiver driver
- * Copyright (C) 2008 - 2009, Intel Corporation.
+ * Copyright (C) 2008 - 2010, Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -40,7 +40,7 @@
 #include <linux/usb/langwell_otg.h>
 
 #define	DRIVER_DESC		"Intel Langwell USB OTG transceiver driver"
-#define	DRIVER_VERSION		"3.0.0.32L.0003"
+#define	DRIVER_VERSION		"July 10, 2010"
 
 MODULE_DESCRIPTION(DRIVER_DESC);
 MODULE_AUTHOR("Henry Yuan <hang.yuan@intel.com>, Hao Wu <hao.wu@intel.com>");
diff --git a/include/linux/usb/langwell_otg.h b/include/linux/usb/langwell_otg.h
index a6562f1d4e2b..51f17b16d312 100644
--- a/include/linux/usb/langwell_otg.h
+++ b/include/linux/usb/langwell_otg.h
@@ -1,6 +1,6 @@
 /*
  * Intel Langwell USB OTG transceiver driver
- * Copyright (C) 2008, Intel Corporation.
+ * Copyright (C) 2008 - 2010, Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
-- 
cgit v1.2.3


From 1f53c0e9bbf654ed93f63deee2bf5c9a1816966e Mon Sep 17 00:00:00 2001
From: Yauheni Kaliuta <yauheni.kaliuta@nokia.com>
Date: Mon, 20 Sep 2010 15:40:26 +0300
Subject: USB: cdc.h: ncm: typo and style fixes

Some typos were in the initial commit, make the spelling
according to the spec.

Add some more comments.

Also change constant names according to the style of the rest
of the file

Signed-off-by: Yauheni Kaliuta <yauheni.kaliuta@nokia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/cdc.h | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/usb/cdc.h b/include/linux/usb/cdc.h
index c117a68d04a7..583264abca0c 100644
--- a/include/linux/usb/cdc.h
+++ b/include/linux/usb/cdc.h
@@ -274,13 +274,13 @@ struct usb_cdc_notification {
 /*
  * Class Specific structures and constants
  *
- * CDC NCM parameter structure, CDC NCM subclass 6.2.1
+ * CDC NCM NTB parameters structure, CDC NCM subclass 6.2.1
  *
  */
 
-struct usb_cdc_ncm_ntb_parameter {
+struct usb_cdc_ncm_ntb_parameters {
 	__le16	wLength;
-	__le16	bmNtbFormatSupported;
+	__le16	bmNtbFormatsSupported;
 	__le32	dwNtbInMaxSize;
 	__le16	wNdpInDivisor;
 	__le16	wNdpInPayloadRemainder;
@@ -297,8 +297,8 @@ struct usb_cdc_ncm_ntb_parameter {
  * CDC NCM transfer headers, CDC NCM subclass 3.2
  */
 
-#define NCM_NTH16_SIGN		0x484D434E /* NCMH */
-#define NCM_NTH32_SIGN		0x686D636E /* ncmh */
+#define USB_CDC_NCM_NTH16_SIGN		0x484D434E /* NCMH */
+#define USB_CDC_NCM_NTH32_SIGN		0x686D636E /* ncmh */
 
 struct usb_cdc_ncm_nth16 {
 	__le32	dwSignature;
@@ -320,11 +320,12 @@ struct usb_cdc_ncm_nth32 {
  * CDC NCM datagram pointers, CDC NCM subclass 3.3
  */
 
-#define NCM_NDP16_CRC_SIGN	0x314D434E /* NCM1 */
-#define NCM_NDP16_NOCRC_SIGN	0x304D434E /* NCM0 */
-#define NCM_NDP32_CRC_SIGN	0x316D636E /* ncm1 */
-#define NCM_NDP32_NOCRC_SIGN	0x306D636E /* ncm0 */
+#define USB_CDC_NCM_NDP16_CRC_SIGN	0x314D434E /* NCM1 */
+#define USB_CDC_NCM_NDP16_NOCRC_SIGN	0x304D434E /* NCM0 */
+#define USB_CDC_NCM_NDP32_CRC_SIGN	0x316D636E /* ncm1 */
+#define USB_CDC_NCM_NDP32_NOCRC_SIGN	0x306D636E /* ncm0 */
 
+/* 16-bit NCM Datagram Pointer Table */
 struct usb_cdc_ncm_ndp16 {
 	__le32	dwSignature;
 	__le16	wLength;
@@ -332,11 +333,12 @@ struct usb_cdc_ncm_ndp16 {
 	__u8	data[0];
 } __attribute__ ((packed));
 
+/* 32-bit NCM Datagram Pointer Table */
 struct usb_cdc_ncm_ndp32 {
 	__le32	dwSignature;
 	__le16	wLength;
 	__le16	wReserved6;
-	__le32	dwNextFpIndex;
+	__le32	dwNextNdpIndex;
 	__le32	dwReserved12;
 	__u8	data[0];
 } __attribute__ ((packed));
-- 
cgit v1.2.3


From 7fc09170cedc329ad274433b4f1a653e603600b5 Mon Sep 17 00:00:00 2001
From: Yauheni Kaliuta <yauheni.kaliuta@nokia.com>
Date: Mon, 20 Sep 2010 15:40:27 +0300
Subject: Revert "USB: ncm: added ncm.h with auxiliary definitions"

This reverts commit 65e0b499105ec8ff3bc4ab7680873dec20127f9d.

Since the host and gadget implementations are different, there is
no common code for the file, remove for now.

Signed-off-by: Yauheni Kaliuta <yauheni.kaliuta@nokia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/ncm.h | 114 ------------------------------------------------
 1 file changed, 114 deletions(-)
 delete mode 100644 include/linux/usb/ncm.h

(limited to 'include/linux')

diff --git a/include/linux/usb/ncm.h b/include/linux/usb/ncm.h
deleted file mode 100644
index 006d1064c8b2..000000000000
--- a/include/linux/usb/ncm.h
+++ /dev/null
@@ -1,114 +0,0 @@
-/*
- * USB CDC NCM auxiliary definitions
- */
-
-#ifndef __LINUX_USB_NCM_H
-#define __LINUX_USB_NCM_H
-
-#include <linux/types.h>
-#include <linux/usb/cdc.h>
-#include <asm/unaligned.h>
-
-#define NCM_NTB_MIN_IN_SIZE		2048
-#define NCM_NTB_MIN_OUT_SIZE		2048
-
-#define NCM_CONTROL_TIMEOUT		(5 * 1000)
-
-/* bmNetworkCapabilities */
-
-#define NCM_NCAP_ETH_FILTER	(1 << 0)
-#define NCM_NCAP_NET_ADDRESS	(1 << 1)
-#define NCM_NCAP_ENCAP_COMM	(1 << 2)
-#define NCM_NCAP_MAX_DGRAM	(1 << 3)
-#define NCM_NCAP_CRC_MODE	(1 << 4)
-
-/*
- * Here are options for NCM Datagram Pointer table (NDP) parser.
- * There are 2 different formats: NDP16 and NDP32 in the spec (ch. 3),
- * in NDP16 offsets and sizes fields are 1 16bit word wide,
- * in NDP32 -- 2 16bit words wide. Also signatures are different.
- * To make the parser code the same, put the differences in the structure,
- * and switch pointers to the structures when the format is changed.
- */
-
-struct ndp_parser_opts {
-	u32		nth_sign;
-	u32		ndp_sign;
-	unsigned	nth_size;
-	unsigned	ndp_size;
-	unsigned	ndplen_align;
-	/* sizes in u16 units */
-	unsigned	dgram_item_len; /* index or length */
-	unsigned	block_length;
-	unsigned	fp_index;
-	unsigned	reserved1;
-	unsigned	reserved2;
-	unsigned	next_fp_index;
-};
-
-#define INIT_NDP16_OPTS {					\
-		.nth_sign = NCM_NTH16_SIGN,			\
-		.ndp_sign = NCM_NDP16_NOCRC_SIGN,		\
-		.nth_size = sizeof(struct usb_cdc_ncm_nth16),	\
-		.ndp_size = sizeof(struct usb_cdc_ncm_ndp16),	\
-		.ndplen_align = 4,				\
-		.dgram_item_len = 1,				\
-		.block_length = 1,				\
-		.fp_index = 1,					\
-		.reserved1 = 0,					\
-		.reserved2 = 0,					\
-		.next_fp_index = 1,				\
-	}
-
-
-#define INIT_NDP32_OPTS {					\
-		.nth_sign = NCM_NTH32_SIGN,			\
-		.ndp_sign = NCM_NDP32_NOCRC_SIGN,		\
-		.nth_size = sizeof(struct usb_cdc_ncm_nth32),	\
-		.ndp_size = sizeof(struct usb_cdc_ncm_ndp32),	\
-		.ndplen_align = 8,				\
-		.dgram_item_len = 2,				\
-		.block_length = 2,				\
-		.fp_index = 2,					\
-		.reserved1 = 1,					\
-		.reserved2 = 2,					\
-		.next_fp_index = 2,				\
-	}
-
-static inline void put_ncm(__le16 **p, unsigned size, unsigned val)
-{
-	switch (size) {
-	case 1:
-		put_unaligned_le16((u16)val, *p);
-		break;
-	case 2:
-		put_unaligned_le32((u32)val, *p);
-
-		break;
-	default:
-		BUG();
-	}
-
-	*p += size;
-}
-
-static inline unsigned get_ncm(__le16 **p, unsigned size)
-{
-	unsigned tmp;
-
-	switch (size) {
-	case 1:
-		tmp = get_unaligned_le16(*p);
-		break;
-	case 2:
-		tmp = get_unaligned_le32(*p);
-		break;
-	default:
-		BUG();
-	}
-
-	*p += size;
-	return tmp;
-}
-
-#endif /* __LINUX_USB_NCM_H */
-- 
cgit v1.2.3


From e5dcd531ac7a040f1b4d35f58914a36ad6174e84 Mon Sep 17 00:00:00 2001
From: Yauheni Kaliuta <yauheni.kaliuta@nokia.com>
Date: Mon, 20 Sep 2010 15:40:28 +0300
Subject: USB: cdc.h: ncm: add missed constants and structures

Make a dedicated structure for datagram pointer entry. There is no
explicit declaration in the spec, but it's used by the host
implementation and makes the structure more clear.

Add some missed constants from the spec

Signed-off-by: Yauheni Kaliuta <yauheni.kaliuta@nokia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/cdc.h | 57 +++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 55 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/usb/cdc.h b/include/linux/usb/cdc.h
index 583264abca0c..2d5b6f296aa3 100644
--- a/include/linux/usb/cdc.h
+++ b/include/linux/usb/cdc.h
@@ -32,6 +32,8 @@
 
 #define USB_CDC_PROTO_EEM			7
 
+#define USB_CDC_NCM_PROTO_NTB			1
+
 /*-------------------------------------------------------------------------*/
 
 /*
@@ -325,14 +327,26 @@ struct usb_cdc_ncm_nth32 {
 #define USB_CDC_NCM_NDP32_CRC_SIGN	0x316D636E /* ncm1 */
 #define USB_CDC_NCM_NDP32_NOCRC_SIGN	0x306D636E /* ncm0 */
 
+/* 16-bit NCM Datagram Pointer Entry */
+struct usb_cdc_ncm_dpe16 {
+	__le16	wDatagramIndex;
+	__le16	wDatagramLength;
+} __attribute__((__packed__));
+
 /* 16-bit NCM Datagram Pointer Table */
 struct usb_cdc_ncm_ndp16 {
 	__le32	dwSignature;
 	__le16	wLength;
 	__le16	wNextFpIndex;
-	__u8	data[0];
+	struct	usb_cdc_ncm_dpe16 dpe16[0];
 } __attribute__ ((packed));
 
+/* 32-bit NCM Datagram Pointer Entry */
+struct usb_cdc_ncm_dpe32 {
+	__le32	wDatagramIndex;
+	__le32	wDatagramLength;
+} __attribute__((__packed__));
+
 /* 32-bit NCM Datagram Pointer Table */
 struct usb_cdc_ncm_ndp32 {
 	__le32	dwSignature;
@@ -340,7 +354,46 @@ struct usb_cdc_ncm_ndp32 {
 	__le16	wReserved6;
 	__le32	dwNextNdpIndex;
 	__le32	dwReserved12;
-	__u8	data[0];
+	struct	usb_cdc_ncm_dpe32 dpe32[0];
 } __attribute__ ((packed));
 
+/* CDC NCM subclass 3.2.1 and 3.2.2 */
+#define USB_CDC_NCM_NDP16_INDEX_MIN			0x000C
+#define USB_CDC_NCM_NDP32_INDEX_MIN			0x0010
+
+/* CDC NCM subclass 3.3.3 Datagram Formatting */
+#define USB_CDC_NCM_DATAGRAM_FORMAT_CRC			0x30
+#define USB_CDC_NCM_DATAGRAM_FORMAT_NOCRC		0X31
+
+/* CDC NCM subclass 4.2 NCM Communications Interface Protocol Code */
+#define USB_CDC_NCM_PROTO_CODE_NO_ENCAP_COMMANDS	0x00
+#define USB_CDC_NCM_PROTO_CODE_EXTERN_PROTO		0xFE
+
+/* CDC NCM subclass 5.2.1 NCM Functional Descriptor, bmNetworkCapabilities */
+#define USB_CDC_NCM_NCAP_ETH_FILTER			(1 << 0)
+#define USB_CDC_NCM_NCAP_NET_ADDRESS			(1 << 1)
+#define USB_CDC_NCM_NCAP_ENCAP_COMMAND			(1 << 2)
+#define USB_CDC_NCM_NCAP_MAX_DATAGRAM_SIZE		(1 << 3)
+#define USB_CDC_NCM_NCAP_CRC_MODE			(1 << 4)
+
+/* CDC NCM subclass Table 6-3: NTB Parameter Structure */
+#define USB_CDC_NCM_NTB16_SUPPORTED			(1 << 0)
+#define USB_CDC_NCM_NTB32_SUPPORTED			(1 << 1)
+
+/* CDC NCM subclass Table 6-3: NTB Parameter Structure */
+#define USB_CDC_NCM_NDP_ALIGN_MIN_SIZE			0x04
+#define USB_CDC_NCM_NTB_MAX_LENGTH			0x1C
+
+/* CDC NCM subclass 6.2.5 SetNtbFormat */
+#define USB_CDC_NCM_NTB16_FORMAT			0x00
+#define USB_CDC_NCM_NTB32_FORMAT			0x01
+
+/* CDC NCM subclass 6.2.7 SetNtbInputSize */
+#define USB_CDC_NCM_NTB_MIN_IN_SIZE			2048
+#define USB_CDC_NCM_NTB_MIN_OUT_SIZE			2048
+
+/* CDC NCM subclass 6.2.11 SetCrcMode */
+#define USB_CDC_NCM_CRC_NOT_APPENDED			0x00
+#define USB_CDC_NCM_CRC_APPENDED			0x01
+
 #endif /* __LINUX_USB_CDC_H */
-- 
cgit v1.2.3


From 6195e3c6aa84dbbf80a60731168118824bd58bba Mon Sep 17 00:00:00 2001
From: Yauheni Kaliuta <yauheni.kaliuta@nokia.com>
Date: Fri, 24 Sep 2010 09:43:27 +0300
Subject: USB: cdc.h: ncm: fix one more typo

In usb_cdc_ncm_dpe32 the fields are 32 bit long and according
to usb style (hungarian notation) should be called dwDatagramIndex
and dwDatagramLength (see CDC NCM subclass spec, 3.3.2). Actually,
they were called wDatagramIndex, wDatagramLength.

Signed-off-by: Yauheni Kaliuta <yauheni.kaliuta@nokia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/cdc.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/usb/cdc.h b/include/linux/usb/cdc.h
index 2d5b6f296aa3..5e86dc771da4 100644
--- a/include/linux/usb/cdc.h
+++ b/include/linux/usb/cdc.h
@@ -343,8 +343,8 @@ struct usb_cdc_ncm_ndp16 {
 
 /* 32-bit NCM Datagram Pointer Entry */
 struct usb_cdc_ncm_dpe32 {
-	__le32	wDatagramIndex;
-	__le32	wDatagramLength;
+	__le32	dwDatagramIndex;
+	__le32	dwDatagramLength;
 } __attribute__((__packed__));
 
 /* 32-bit NCM Datagram Pointer Table */
-- 
cgit v1.2.3


From 8fa7fd74ef398370383df276ca41082ba35aafd8 Mon Sep 17 00:00:00 2001
From: Michal Nazarewicz <mina86@mina86.com>
Date: Thu, 7 Oct 2010 13:05:21 +0200
Subject: USB: storage: Use USB_ prefix instead of US_ prefix

This commit changes prefix for some of the USB mass storage
class related macros (ie. USB_SC_ for subclass and USB_PR_
for class).

Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/block/ub.c                     |   2 +-
 drivers/usb/storage/scsiglue.c         |   8 +-
 drivers/usb/storage/sddr09.c           |   2 +-
 drivers/usb/storage/transport.c        |  10 +-
 drivers/usb/storage/unusual_alauda.h   |   4 +-
 drivers/usb/storage/unusual_cypress.h  |   4 +-
 drivers/usb/storage/unusual_datafab.h  |  20 +-
 drivers/usb/storage/unusual_devs.h     | 572 ++++++++++++++++-----------------
 drivers/usb/storage/unusual_freecom.h  |   2 +-
 drivers/usb/storage/unusual_isd200.h   |  12 +-
 drivers/usb/storage/unusual_jumpshot.h |   2 +-
 drivers/usb/storage/unusual_karma.h    |   2 +-
 drivers/usb/storage/unusual_onetouch.h |   4 +-
 drivers/usb/storage/unusual_sddr09.h   |  12 +-
 drivers/usb/storage/unusual_sddr55.h   |   8 +-
 drivers/usb/storage/unusual_usbat.h    |   8 +-
 drivers/usb/storage/usb.c              |  30 +-
 include/linux/usb_usual.h              |  59 ++--
 18 files changed, 381 insertions(+), 380 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/ub.c b/drivers/block/ub.c
index c48e14878582..f3fd454e28f9 100644
--- a/drivers/block/ub.c
+++ b/drivers/block/ub.c
@@ -396,7 +396,7 @@ static int ub_probe_lun(struct ub_dev *sc, int lnum);
 #else
 
 static const struct usb_device_id ub_usb_ids[] = {
-	{ USB_INTERFACE_INFO(USB_CLASS_MASS_STORAGE, US_SC_SCSI, US_PR_BULK) },
+	{ USB_INTERFACE_INFO(USB_CLASS_MASS_STORAGE, USB_SC_SCSI, USB_PR_BULK) },
 	{ }
 };
 
diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
index d8d98cfecada..e80362d148f3 100644
--- a/drivers/usb/storage/scsiglue.c
+++ b/drivers/usb/storage/scsiglue.c
@@ -113,7 +113,7 @@ static int slave_alloc (struct scsi_device *sdev)
 	 * Let the scanning code know if this target merely sets
 	 * Peripheral Device Type to 0x1f to indicate no LUN.
 	 */
-	if (us->subclass == US_SC_UFI)
+	if (us->subclass == USB_SC_UFI)
 		sdev->sdev_target->pdt_1f_for_no_lun = 1;
 
 	return 0;
@@ -176,7 +176,7 @@ static int slave_configure(struct scsi_device *sdev)
 		/* Disk-type devices use MODE SENSE(6) if the protocol
 		 * (SubClass) is Transparent SCSI, otherwise they use
 		 * MODE SENSE(10). */
-		if (us->subclass != US_SC_SCSI && us->subclass != US_SC_CYP_ATACB)
+		if (us->subclass != USB_SC_SCSI && us->subclass != USB_SC_CYP_ATACB)
 			sdev->use_10_for_ms = 1;
 
 		/* Many disks only accept MODE SENSE transfer lengths of
@@ -245,7 +245,7 @@ static int slave_configure(struct scsi_device *sdev)
 		 * capacity will be decremented or is correct. */
 		if (!(us->fflags & (US_FL_FIX_CAPACITY | US_FL_CAPACITY_OK |
 					US_FL_SCM_MULT_TARG)) &&
-				us->protocol == US_PR_BULK)
+				us->protocol == USB_PR_BULK)
 			us->use_last_sector_hacks = 1;
 	} else {
 
@@ -261,7 +261,7 @@ static int slave_configure(struct scsi_device *sdev)
 	 * scsi_level == 0 (UNKNOWN).  Hence such devices must necessarily
 	 * be single-LUN.
 	 */
-	if ((us->protocol == US_PR_CB || us->protocol == US_PR_CBI) &&
+	if ((us->protocol == USB_PR_CB || us->protocol == USB_PR_CBI) &&
 			sdev->scsi_level == SCSI_UNKNOWN)
 		us->max_lun = 0;
 
diff --git a/drivers/usb/storage/sddr09.c b/drivers/usb/storage/sddr09.c
index ab5f9f37575a..bcb9a709d349 100644
--- a/drivers/usb/storage/sddr09.c
+++ b/drivers/usb/storage/sddr09.c
@@ -1760,7 +1760,7 @@ static int sddr09_probe(struct usb_interface *intf,
 	if (result)
 		return result;
 
-	if (us->protocol == US_PR_DPCM_USB) {
+	if (us->protocol == USB_PR_DPCM_USB) {
 		us->transport_name = "Control/Bulk-EUSB/SDDR09";
 		us->transport = dpcm_transport;
 		us->transport_reset = usb_stor_CB_reset;
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 64ec073e89de..00418995d8e9 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -642,7 +642,7 @@ void usb_stor_invoke_transport(struct scsi_cmnd *srb, struct us_data *us)
 	 * unless the operation involved a data-in transfer.  Devices
 	 * can signal most data-in errors by stalling the bulk-in pipe.
 	 */
-	if ((us->protocol == US_PR_CB || us->protocol == US_PR_DPCM_USB) &&
+	if ((us->protocol == USB_PR_CB || us->protocol == USB_PR_DPCM_USB) &&
 			srb->sc_data_direction != DMA_FROM_DEVICE) {
 		US_DEBUGP("-- CB transport device requiring auto-sense\n");
 		need_auto_sense = 1;
@@ -701,8 +701,8 @@ Retry_Sense:
 		scsi_eh_prep_cmnd(srb, &ses, NULL, 0, sense_size);
 
 		/* FIXME: we must do the protocol translation here */
-		if (us->subclass == US_SC_RBC || us->subclass == US_SC_SCSI ||
-				us->subclass == US_SC_CYP_ATACB)
+		if (us->subclass == USB_SC_RBC || us->subclass == USB_SC_SCSI ||
+				us->subclass == USB_SC_CYP_ATACB)
 			srb->cmd_len = 6;
 		else
 			srb->cmd_len = 12;
@@ -926,7 +926,7 @@ int usb_stor_CB_transport(struct scsi_cmnd *srb, struct us_data *us)
 	/* NOTE: CB does not have a status stage.  Silly, I know.  So
 	 * we have to catch this at a higher level.
 	 */
-	if (us->protocol != US_PR_CBI)
+	if (us->protocol != USB_PR_CBI)
 		return USB_STOR_TRANSPORT_GOOD;
 
 	result = usb_stor_intr_transfer(us, us->iobuf, 2);
@@ -942,7 +942,7 @@ int usb_stor_CB_transport(struct scsi_cmnd *srb, struct us_data *us)
 	 * that this means we could be ignoring a real error on these
 	 * commands, but that can't be helped.
 	 */
-	if (us->subclass == US_SC_UFI) {
+	if (us->subclass == USB_SC_UFI) {
 		if (srb->cmnd[0] == REQUEST_SENSE ||
 		    srb->cmnd[0] == INQUIRY)
 			return USB_STOR_TRANSPORT_GOOD;
diff --git a/drivers/usb/storage/unusual_alauda.h b/drivers/usb/storage/unusual_alauda.h
index 8c412f885dd2..fa3e9edaa2cf 100644
--- a/drivers/usb/storage/unusual_alauda.h
+++ b/drivers/usb/storage/unusual_alauda.h
@@ -21,11 +21,11 @@
 UNUSUAL_DEV(  0x0584, 0x0008, 0x0102, 0x0102,
 		"Fujifilm",
 		"DPC-R1 (Alauda)",
-		US_SC_SCSI, US_PR_ALAUDA, init_alauda, 0),
+		USB_SC_SCSI, USB_PR_ALAUDA, init_alauda, 0),
 
 UNUSUAL_DEV(  0x07b4, 0x010a, 0x0102, 0x0102,
 		"Olympus",
 		"MAUSB-10 (Alauda)",
-		US_SC_SCSI, US_PR_ALAUDA, init_alauda, 0),
+		USB_SC_SCSI, USB_PR_ALAUDA, init_alauda, 0),
 
 #endif /* defined(CONFIG_USB_STORAGE_ALAUDA) || ... */
diff --git a/drivers/usb/storage/unusual_cypress.h b/drivers/usb/storage/unusual_cypress.h
index 44be6d75dab6..c854fdebe0ae 100644
--- a/drivers/usb/storage/unusual_cypress.h
+++ b/drivers/usb/storage/unusual_cypress.h
@@ -23,12 +23,12 @@
 UNUSUAL_DEV(  0x04b4, 0x6830, 0x0000, 0x9999,
 		"Cypress",
 		"Cypress AT2LP",
-		US_SC_CYP_ATACB, US_PR_DEVICE, NULL, 0),
+		USB_SC_CYP_ATACB, USB_PR_DEVICE, NULL, 0),
 
 /* CY7C68310 : support atacb and atacb2 */
 UNUSUAL_DEV(  0x04b4, 0x6831, 0x0000, 0x9999,
 		"Cypress",
 		"Cypress ISD-300LP",
-		US_SC_CYP_ATACB, US_PR_DEVICE, NULL, 0),
+		USB_SC_CYP_ATACB, USB_PR_DEVICE, NULL, 0),
 
 #endif /* defined(CONFIG_USB_STORAGE_CYPRESS_ATACB) || ... */
diff --git a/drivers/usb/storage/unusual_datafab.h b/drivers/usb/storage/unusual_datafab.h
index c9298ce9f223..582a603c78be 100644
--- a/drivers/usb/storage/unusual_datafab.h
+++ b/drivers/usb/storage/unusual_datafab.h
@@ -21,7 +21,7 @@
 UNUSUAL_DEV(  0x07c4, 0xa000, 0x0000, 0x0015,
 		"Datafab",
 		"MDCFE-B USB CF Reader",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		0),
 
 /*
@@ -38,45 +38,45 @@ UNUSUAL_DEV(  0x07c4, 0xa000, 0x0000, 0x0015,
 UNUSUAL_DEV( 0x07c4, 0xa001, 0x0000, 0xffff,
 		"SIIG/Datafab",
 		"SIIG/Datafab Memory Stick+CF Reader/Writer",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		0),
 
 /* Reported by Josef Reisinger <josef.reisinger@netcologne.de> */
 UNUSUAL_DEV( 0x07c4, 0xa002, 0x0000, 0xffff,
 		"Datafab/Unknown",
 		"MD2/MD3 Disk enclosure",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		US_FL_SINGLE_LUN),
 
 UNUSUAL_DEV( 0x07c4, 0xa003, 0x0000, 0xffff,
 		"Datafab/Unknown",
 		"Datafab-based Reader",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		0),
 
 UNUSUAL_DEV( 0x07c4, 0xa004, 0x0000, 0xffff,
 		"Datafab/Unknown",
 		"Datafab-based Reader",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		0),
 
 UNUSUAL_DEV( 0x07c4, 0xa005, 0x0000, 0xffff,
 		"PNY/Datafab",
 		"PNY/Datafab CF+SM Reader",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		0),
 
 UNUSUAL_DEV( 0x07c4, 0xa006, 0x0000, 0xffff,
 		"Simple Tech/Datafab",
 		"Simple Tech/Datafab CF+SM Reader",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		0),
 
 /* Submitted by Olaf Hering <olh@suse.de> */
 UNUSUAL_DEV(  0x07c4, 0xa109, 0x0000, 0xffff,
 		"Datafab Systems, Inc.",
 		"USB to CF + SM Combo (LC1)",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		0),
 
 /* Reported by Felix Moeller <felix@derklecks.de>
@@ -86,13 +86,13 @@ UNUSUAL_DEV(  0x07c4, 0xa109, 0x0000, 0xffff,
 UNUSUAL_DEV(  0x07c4, 0xa10b, 0x0000, 0xffff,
 		"DataFab Systems Inc.",
 		"USB CF+MS",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		0),
 
 UNUSUAL_DEV( 0x0c0b, 0xa109, 0x0000, 0xffff,
 		"Acomdata",
 		"CF",
-		US_SC_SCSI, US_PR_DATAFAB, NULL,
+		USB_SC_SCSI, USB_PR_DATAFAB, NULL,
 		US_FL_SINGLE_LUN),
 
 #endif /* defined(CONFIG_USB_STORAGE_DATAFAB) || ... */
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 2c897eefadde..f43314a22ba4 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -63,26 +63,26 @@
 UNUSUAL_DEV(  0x03eb, 0x2002, 0x0100, 0x0100,
 		"ATMEL",
 		"SND1 Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE),
 
 /* Reported by Rodolfo Quesada <rquesada@roqz.net> */
 UNUSUAL_DEV(  0x03ee, 0x6906, 0x0003, 0x0003,
 		"VIA Technologies Inc.",
 		"Mitsumi multi cardreader",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 UNUSUAL_DEV(  0x03f0, 0x0107, 0x0200, 0x0200,
 		"HP",
 		"CD-Writer+",
-		US_SC_8070, US_PR_CB, NULL, 0),
+		USB_SC_8070, USB_PR_CB, NULL, 0),
 
 /* Reported by Ben Efros <ben@pc-doctor.com> */
 UNUSUAL_DEV(  0x03f0, 0x070c, 0x0000, 0x0000,
 		"HP",
 		"Personal Media Drive",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SANE_SENSE ),
 
 /* Reported by Grant Grundler <grundler@parisc-linux.org>
@@ -91,7 +91,7 @@ UNUSUAL_DEV(  0x03f0, 0x070c, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x03f0, 0x4002, 0x0001, 0x0001,
 		"HP",
 		"PhotoSmart R707",
-		US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_CAPACITY),
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_FIX_CAPACITY),
 
 /* Reported by Sebastian Kapfer <sebastian_kapfer@gmx.net>
  * and Olaf Hering <olh@suse.de> (different bcd's, same vendor/product)
@@ -100,14 +100,14 @@ UNUSUAL_DEV(  0x03f0, 0x4002, 0x0001, 0x0001,
 UNUSUAL_DEV(  0x0409, 0x0040, 0x0000, 0x9999,
 		"NEC",
 		"NEC USB UF000x",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Patch submitted by Mihnea-Costin Grigore <mihnea@zulu.ro> */
 UNUSUAL_DEV(  0x040d, 0x6205, 0x0003, 0x0003,
 		"VIA Technologies Inc.",
 		"USB 2.0 Card Reader",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Deduced by Jonathan Woithe <jwoithe@physics.adelaide.edu.au>
@@ -117,40 +117,40 @@ UNUSUAL_DEV(  0x040d, 0x6205, 0x0003, 0x0003,
 UNUSUAL_DEV(  0x0411, 0x001c, 0x0113, 0x0113,
 		"Buffalo",
 		"DUB-P40G HDD",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* Submitted by Ernestas Vaiciukevicius <ernisv@gmail.com> */
 UNUSUAL_DEV(  0x0419, 0x0100, 0x0100, 0x0100,
 		"Samsung Info. Systems America, Inc.",
 		"MP3 Player",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Orgad Shaneh <orgads@gmail.com> */
 UNUSUAL_DEV(  0x0419, 0xaace, 0x0100, 0x0100,
 		"Samsung", "MP3 Player",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Christian Leber <christian@leber.de> */
 UNUSUAL_DEV(  0x0419, 0xaaf5, 0x0100, 0x0100,
 		"TrekStor",
 		"i.Beat 115 2.0",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_NOT_LOCKABLE ),
 
 /* Reported by Stefan Werner <dustbln@gmx.de> */
 UNUSUAL_DEV(  0x0419, 0xaaf6, 0x0100, 0x0100,
 		"TrekStor",
 		"i.Beat Joy 2.0",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Pete Zaitcev <zaitcev@redhat.com>, bz#176584 */
 UNUSUAL_DEV(  0x0420, 0x0001, 0x0100, 0x0100,
 		"GENERIC", "MP3 PLAYER", /* MyMusix PD-205 on the outside. */
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Andrew Nayenko <relan@bk.ru>
@@ -158,28 +158,28 @@ UNUSUAL_DEV(  0x0420, 0x0001, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x0421, 0x0019, 0x0592, 0x0610,
 		"Nokia",
 		"Nokia 6288",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64 ),
 
 /* Reported by Mario Rettig <mariorettig@web.de> */
 UNUSUAL_DEV(  0x0421, 0x042e, 0x0100, 0x0100,
 		"Nokia",
 		"Nokia 3250",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_FIX_CAPACITY ),
 
 /* Reported by <honkkis@gmail.com> */
 UNUSUAL_DEV(  0x0421, 0x0433, 0x0100, 0x0100,
 		"Nokia",
 		"E70",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_FIX_CAPACITY ),
 
 /* Reported by Jon Hart <Jon.Hart@web.de> */
 UNUSUAL_DEV(  0x0421, 0x0434, 0x0100, 0x0100,
 		"Nokia",
 		"E60",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Sumedha Swamy <sumedhaswamy@gmail.com> and
@@ -187,7 +187,7 @@ UNUSUAL_DEV(  0x0421, 0x0434, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x0421, 0x0444, 0x0100, 0x0100,
 		"Nokia",
 		"N91",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_FIX_CAPACITY ),
 
 /* Reported by Jiri Slaby <jirislaby@gmail.com> and
@@ -195,42 +195,42 @@ UNUSUAL_DEV(  0x0421, 0x0444, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x0421, 0x0446, 0x0100, 0x0100,
 		"Nokia",
 		"N80",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_FIX_CAPACITY ),
 
 /* Reported by Matthew Bloch <matthew@bytemark.co.uk> */
 UNUSUAL_DEV(  0x0421, 0x044e, 0x0100, 0x0100,
 		"Nokia",
 		"E61",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_FIX_CAPACITY ),
 
 /* Reported by Bardur Arantsson <bardur@scientician.net> */
 UNUSUAL_DEV(  0x0421, 0x047c, 0x0370, 0x0610,
 		"Nokia",
 		"6131",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64 ),
 
 /* Reported by Manuel Osdoba <manuel.osdoba@tu-ilmenau.de> */
 UNUSUAL_DEV( 0x0421, 0x0492, 0x0452, 0x9999,
 		"Nokia",
 		"Nokia 6233",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64 ),
 
 /* Reported by Alex Corcoles <alex@corcoles.net> */
 UNUSUAL_DEV(  0x0421, 0x0495, 0x0370, 0x0370,
 		"Nokia",
 		"6234",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64 ),
 
 #ifdef NO_SDDR09
 UNUSUAL_DEV(  0x0436, 0x0005, 0x0100, 0x0100,
 		"Microtech",
 		"CameraMate",
-		US_SC_SCSI, US_PR_CB, NULL,
+		USB_SC_SCSI, USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN ),
 #endif
 
@@ -239,7 +239,7 @@ UNUSUAL_DEV(  0x0436, 0x0005, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x0451, 0x5416, 0x0100, 0x0100,
 		"Neuros Audio",
 		"USB 2.0 HD 2.5",
-		US_SC_DEVICE, US_PR_BULK, NULL,
+		USB_SC_DEVICE, USB_PR_BULK, NULL,
 		US_FL_NEED_OVERRIDE ),
 
 /*
@@ -250,7 +250,7 @@ UNUSUAL_DEV(  0x0451, 0x5416, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x0457, 0x0150, 0x0100, 0x0100,
 		"USBest Technology",	/* sold by Transcend */
 		"USB Mass Storage Device",
-		US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_NOT_LOCKABLE ),
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_NOT_LOCKABLE ),
 
 /*
 * Bohdan Linda <bohdan.linda@gmail.com>
@@ -260,7 +260,7 @@ UNUSUAL_DEV(  0x0457, 0x0150, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x0457, 0x0151, 0x0100, 0x0100,
 		"USB 2.0",
 		"Flash Disk",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NOT_LOCKABLE ),
 
 /* Reported by Tamas Kerecsen <kerecsen@bigfoot.com>
@@ -272,7 +272,7 @@ UNUSUAL_DEV(  0x0457, 0x0151, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x045e, 0xffff, 0x0000, 0x0000,
 		"Mitac",
 		"GPS",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64 ),
 
 /*
@@ -284,32 +284,32 @@ UNUSUAL_DEV(  0x045e, 0xffff, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x046b, 0xff40, 0x0100, 0x0100,
 		"AMI",
 		"Virtual Floppy",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NO_WP_DETECT),
 
 /* Patch submitted by Philipp Friedrich <philipp@void.at> */
 UNUSUAL_DEV(  0x0482, 0x0100, 0x0100, 0x0100,
 		"Kyocera",
 		"Finecam S3x",
-		US_SC_8070, US_PR_CB, NULL, US_FL_FIX_INQUIRY),
+		USB_SC_8070, USB_PR_CB, NULL, US_FL_FIX_INQUIRY),
 
 /* Patch submitted by Philipp Friedrich <philipp@void.at> */
 UNUSUAL_DEV(  0x0482, 0x0101, 0x0100, 0x0100,
 		"Kyocera",
 		"Finecam S4",
-		US_SC_8070, US_PR_CB, NULL, US_FL_FIX_INQUIRY),
+		USB_SC_8070, USB_PR_CB, NULL, US_FL_FIX_INQUIRY),
 
 /* Patch submitted by Stephane Galles <stephane.galles@free.fr> */
 UNUSUAL_DEV(  0x0482, 0x0103, 0x0100, 0x0100,
 		"Kyocera",
 		"Finecam S5",
-		US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_INQUIRY),
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_FIX_INQUIRY),
 
 /* Patch submitted by Jens Taprogge <jens.taprogge@taprogge.org> */
 UNUSUAL_DEV(  0x0482, 0x0107, 0x0100, 0x0100,
 		"Kyocera",
 		"CONTAX SL300R T*",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_NOT_LOCKABLE),
 
 /* Reported by Paul Stewart <stewart@wetlogic.net>
@@ -317,7 +317,7 @@ UNUSUAL_DEV(  0x0482, 0x0107, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x04a4, 0x0004, 0x0001, 0x0001,
 		"Hitachi",
 		"DVD-CAM DZ-MV100A Camcorder",
-		US_SC_SCSI, US_PR_CB, NULL, US_FL_SINGLE_LUN),
+		USB_SC_SCSI, USB_PR_CB, NULL, US_FL_SINGLE_LUN),
 
 /* BENQ DC5330
  * Reported by Manuel Fombuena <mfombuena@ya.com> and
@@ -325,7 +325,7 @@ UNUSUAL_DEV(  0x04a4, 0x0004, 0x0001, 0x0001,
 UNUSUAL_DEV(  0x04a5, 0x3010, 0x0100, 0x0100,
 		"Tekom Technologies, Inc",
 		"300_CAMERA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Patch for Nikon coolpix 2000
@@ -333,14 +333,14 @@ UNUSUAL_DEV(  0x04a5, 0x3010, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x04b0, 0x0301, 0x0010, 0x0010,
 		"NIKON",
 		"NIKON DSC E2000",
-		US_SC_DEVICE, US_PR_DEVICE,NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE,NULL,
 		US_FL_NOT_LOCKABLE ),
 
 /* Reported by Doug Maxey (dwm@austin.ibm.com) */
 UNUSUAL_DEV(  0x04b3, 0x4001, 0x0110, 0x0110,
 		"IBM",
 		"IBM RSA2",
-		US_SC_DEVICE, US_PR_CB, NULL,
+		USB_SC_DEVICE, USB_PR_CB, NULL,
 		US_FL_MAX_SECTORS_MIN),
 
 /* Reported by Simon Levitt <simon@whattf.com>
@@ -348,14 +348,14 @@ UNUSUAL_DEV(  0x04b3, 0x4001, 0x0110, 0x0110,
 UNUSUAL_DEV(  0x04b8, 0x0601, 0x0100, 0x0100,
 		"Epson",
 		"875DC Storage",
-		US_SC_SCSI, US_PR_CB, NULL, US_FL_FIX_INQUIRY),
+		USB_SC_SCSI, USB_PR_CB, NULL, US_FL_FIX_INQUIRY),
 
 /* Reported by Khalid Aziz <khalid@gonehiking.org>
  * This entry is needed because the device reports Sub=ff */
 UNUSUAL_DEV(  0x04b8, 0x0602, 0x0110, 0x0110,
 		"Epson",
 		"785EPX Storage",
-		US_SC_SCSI, US_PR_BULK, NULL, US_FL_SINGLE_LUN),
+		USB_SC_SCSI, USB_PR_BULK, NULL, US_FL_SINGLE_LUN),
 
 /* Not sure who reported this originally but
  * Pavel Machek <pavel@ucw.cz> reported that the extra US_FL_SINGLE_LUN
@@ -363,7 +363,7 @@ UNUSUAL_DEV(  0x04b8, 0x0602, 0x0110, 0x0110,
 UNUSUAL_DEV(  0x04cb, 0x0100, 0x0000, 0x2210,
 		"Fujifilm",
 		"FinePix 1400Zoom",
-		US_SC_UFI, US_PR_DEVICE, NULL, US_FL_FIX_INQUIRY | US_FL_SINGLE_LUN),
+		USB_SC_UFI, USB_PR_DEVICE, NULL, US_FL_FIX_INQUIRY | US_FL_SINGLE_LUN),
 
 /* Reported by Ondrej Zary <linux@rainbow-software.org>
  * The device reports one sector more and breaks when that sector is accessed
@@ -371,7 +371,7 @@ UNUSUAL_DEV(  0x04cb, 0x0100, 0x0000, 0x2210,
 UNUSUAL_DEV(  0x04ce, 0x0002, 0x026c, 0x026c,
 		"ScanLogic",
 		"SL11R-IDE",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY),
 
 /* Reported by Kriston Fincher <kriston@airmail.net>
@@ -382,27 +382,27 @@ UNUSUAL_DEV(  0x04ce, 0x0002, 0x026c, 0x026c,
 UNUSUAL_DEV(  0x04da, 0x0901, 0x0100, 0x0200,
 		"Panasonic",
 		"LS-120 Camera",
-		US_SC_UFI, US_PR_DEVICE, NULL, 0),
+		USB_SC_UFI, USB_PR_DEVICE, NULL, 0),
 
 /* From Yukihiro Nakai, via zaitcev@yahoo.com.
  * This is needed for CB instead of CBI */
 UNUSUAL_DEV(  0x04da, 0x0d05, 0x0000, 0x0000,
 		"Sharp CE-CW05",
 		"CD-R/RW Drive",
-		US_SC_8070, US_PR_CB, NULL, 0),
+		USB_SC_8070, USB_PR_CB, NULL, 0),
 
 /* Reported by Adriaan Penning <a.penning@luon.net> */
 UNUSUAL_DEV(  0x04da, 0x2372, 0x0000, 0x9999,
 		"Panasonic",
 		"DMC-LCx Camera",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_NOT_LOCKABLE ),
 
 /* Reported by Simeon Simeonov <simeonov_2000@yahoo.com> */
 UNUSUAL_DEV(  0x04da, 0x2373, 0x0000, 0x9999,
 		"LEICA",
 		"D-LUX Camera",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_NOT_LOCKABLE ),
 
 /* Most of the following entries were developed with the help of
@@ -411,19 +411,19 @@ UNUSUAL_DEV(  0x04da, 0x2373, 0x0000, 0x9999,
 UNUSUAL_DEV(  0x04e6, 0x0001, 0x0200, 0x0200,
 		"Matshita",
 		"LS-120",
-		US_SC_8020, US_PR_CB, NULL, 0),
+		USB_SC_8020, USB_PR_CB, NULL, 0),
 
 UNUSUAL_DEV(  0x04e6, 0x0002, 0x0100, 0x0100,
 		"Shuttle",
 		"eUSCSI Bridge",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_euscsi_init, 
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_euscsi_init,
 		US_FL_SCM_MULT_TARG ),
 
 #ifdef NO_SDDR09
 UNUSUAL_DEV(  0x04e6, 0x0005, 0x0100, 0x0208,
 		"SCM Microsystems",
 		"eUSB CompactFlash Adapter",
-		US_SC_SCSI, US_PR_CB, NULL,
+		USB_SC_SCSI, USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN),
 #endif
 
@@ -431,54 +431,54 @@ UNUSUAL_DEV(  0x04e6, 0x0005, 0x0100, 0x0208,
 UNUSUAL_DEV(  0x04e6, 0x0006, 0x0100, 0x0100,
 		"SCM Microsystems Inc.",
 		"eUSB MMC Adapter",
-		US_SC_SCSI, US_PR_CB, NULL,
+		USB_SC_SCSI, USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN),
 
 /* Reported by Daniel Nouri <dpunktnpunkt@web.de> */
 UNUSUAL_DEV(  0x04e6, 0x0006, 0x0205, 0x0205,
 		"Shuttle",
 		"eUSB MMC Adapter",
-		US_SC_SCSI, US_PR_DEVICE, NULL,
+		USB_SC_SCSI, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN),
 
 UNUSUAL_DEV(  0x04e6, 0x0007, 0x0100, 0x0200,
 		"Sony",
 		"Hifd",
-		US_SC_SCSI, US_PR_CB, NULL,
+		USB_SC_SCSI, USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN),
 
 UNUSUAL_DEV(  0x04e6, 0x0009, 0x0200, 0x0200,
 		"Shuttle",
 		"eUSB ATA/ATAPI Adapter",
-		US_SC_8020, US_PR_CB, NULL, 0),
+		USB_SC_8020, USB_PR_CB, NULL, 0),
 
 UNUSUAL_DEV(  0x04e6, 0x000a, 0x0200, 0x0200,
 		"Shuttle",
 		"eUSB CompactFlash Adapter",
-		US_SC_8020, US_PR_CB, NULL, 0),
+		USB_SC_8020, USB_PR_CB, NULL, 0),
 
 UNUSUAL_DEV(  0x04e6, 0x000B, 0x0100, 0x0100,
 		"Shuttle",
 		"eUSCSI Bridge",
-		US_SC_SCSI, US_PR_BULK, usb_stor_euscsi_init,
+		USB_SC_SCSI, USB_PR_BULK, usb_stor_euscsi_init,
 		US_FL_SCM_MULT_TARG ), 
 
 UNUSUAL_DEV(  0x04e6, 0x000C, 0x0100, 0x0100,
 		"Shuttle",
 		"eUSCSI Bridge",
-		US_SC_SCSI, US_PR_BULK, usb_stor_euscsi_init,
+		USB_SC_SCSI, USB_PR_BULK, usb_stor_euscsi_init,
 		US_FL_SCM_MULT_TARG ),
 
 UNUSUAL_DEV(  0x04e6, 0x0101, 0x0200, 0x0200,
 		"Shuttle",
 		"CD-RW Device",
-		US_SC_8020, US_PR_CB, NULL, 0),
+		USB_SC_8020, USB_PR_CB, NULL, 0),
 
 /* Reported by Dmitry Khlystov <adminimus@gmail.com> */
 UNUSUAL_DEV(  0x04e8, 0x507c, 0x0220, 0x0220,
 		"Samsung",
 		"YP-U3",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64),
 
 /* Entry and supporting patch by Theodore Kilgore <kilgota@auburn.edu>.
@@ -488,14 +488,14 @@ UNUSUAL_DEV(  0x04e8, 0x507c, 0x0220, 0x0220,
 UNUSUAL_DEV(  0x04fc, 0x80c2, 0x0100, 0x0100,
 		"Kobian Mercury",
 		"Binocam DCB-132",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_BULK32),
 
 /* Reported by Bob Sass <rls@vectordb.com> -- only rev 1.33 tested */
 UNUSUAL_DEV(  0x050d, 0x0115, 0x0133, 0x0133,
 		"Belkin",
 		"USB SCSI Adaptor",
-		US_SC_SCSI, US_PR_BULK, usb_stor_euscsi_init,
+		USB_SC_SCSI, USB_PR_BULK, usb_stor_euscsi_init,
 		US_FL_SCM_MULT_TARG ),
 
 /* Iomega Clik! Drive 
@@ -505,14 +505,14 @@ UNUSUAL_DEV(  0x050d, 0x0115, 0x0133, 0x0133,
 UNUSUAL_DEV(  0x0525, 0xa140, 0x0100, 0x0100,
 		"Iomega",
 		"USB Clik! 40",
-		US_SC_8070, US_PR_DEVICE, NULL,
+		USB_SC_8070, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* Added by Alan Stern <stern@rowland.harvard.edu> */
 COMPLIANT_DEV(0x0525, 0xa4a5, 0x0000, 0x9999,
 		"Linux",
 		"File-backed Storage Gadget",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_CAPACITY_OK ),
 
 /* Yakumo Mega Image 37
@@ -520,7 +520,7 @@ COMPLIANT_DEV(0x0525, 0xa4a5, 0x0000, 0x9999,
 UNUSUAL_DEV(  0x052b, 0x1801, 0x0100, 0x0100,
 		"Tekom Technologies, Inc",
 		"300_CAMERA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Another Yakumo camera.
@@ -528,14 +528,14 @@ UNUSUAL_DEV(  0x052b, 0x1801, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x052b, 0x1804, 0x0100, 0x0100,
 		"Tekom Technologies, Inc",
 		"300_CAMERA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Iacopo Spalletti <avvisi@spalletti.it> */
 UNUSUAL_DEV(  0x052b, 0x1807, 0x0100, 0x0100,
 		"Tekom Technologies, Inc",
 		"300_CAMERA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Yakumo Mega Image 47
@@ -543,7 +543,7 @@ UNUSUAL_DEV(  0x052b, 0x1807, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x052b, 0x1905, 0x0100, 0x0100,
 		"Tekom Technologies, Inc",
 		"400_CAMERA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Paul Ortyl <ortylp@3miasto.net>
@@ -551,13 +551,13 @@ UNUSUAL_DEV(  0x052b, 0x1905, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x052b, 0x1911, 0x0100, 0x0100,
 		"Tekom Technologies, Inc",
 		"400_CAMERA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 UNUSUAL_DEV(  0x054c, 0x0010, 0x0106, 0x0450,
 		"Sony",
 		"DSC-S30/S70/S75/505V/F505/F707/F717/P8",
-		US_SC_SCSI, US_PR_DEVICE, NULL,
+		USB_SC_SCSI, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN | US_FL_NOT_LOCKABLE | US_FL_NO_WP_DETECT ),
 
 /* Submitted by Lars Jacob <jacob.lars@googlemail.com>
@@ -565,7 +565,7 @@ UNUSUAL_DEV(  0x054c, 0x0010, 0x0106, 0x0450,
 UNUSUAL_DEV(  0x054c, 0x0010, 0x0500, 0x0610,
 		"Sony",
 		"DSC-T1/T5/H5",
-		US_SC_8070, US_PR_DEVICE, NULL,
+		USB_SC_8070, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 
@@ -573,88 +573,88 @@ UNUSUAL_DEV(  0x054c, 0x0010, 0x0500, 0x0610,
 UNUSUAL_DEV(  0x054c, 0x0025, 0x0100, 0x0100,
 		"Sony",
 		"Memorystick NW-MS7",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Submitted by Olaf Hering, <olh@suse.de> SuSE Bugzilla #49049 */
 UNUSUAL_DEV(  0x054c, 0x002c, 0x0501, 0x2000,
 		"Sony",
 		"USB Floppy Drive",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 UNUSUAL_DEV(  0x054c, 0x002d, 0x0100, 0x0100,
 		"Sony",
 		"Memorystick MSAC-US1",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Submitted by Klaus Mueller <k.mueller@intershop.de> */
 UNUSUAL_DEV(  0x054c, 0x002e, 0x0106, 0x0310,
 		"Sony",
 		"Handycam",
-		US_SC_SCSI, US_PR_DEVICE, NULL,
+		USB_SC_SCSI, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Submitted by Rajesh Kumble Nayak <nayak@obs-nice.fr> */
 UNUSUAL_DEV(  0x054c, 0x002e, 0x0500, 0x0500,
 		"Sony",
 		"Handycam HC-85",
-		US_SC_UFI, US_PR_DEVICE, NULL,
+		USB_SC_UFI, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 UNUSUAL_DEV(  0x054c, 0x0032, 0x0000, 0x9999,
 		"Sony",
 		"Memorystick MSC-U01N",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Submitted by Michal Mlotek <mlotek@foobar.pl> */
 UNUSUAL_DEV(  0x054c, 0x0058, 0x0000, 0x9999,
 		"Sony",
 		"PEG N760c Memorystick",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 		
 UNUSUAL_DEV(  0x054c, 0x0069, 0x0000, 0x9999,
 		"Sony",
 		"Memorystick MSC-U03",
-		US_SC_UFI, US_PR_CB, NULL,
+		USB_SC_UFI, USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Submitted by Nathan Babb <nathan@lexi.com> */
 UNUSUAL_DEV(  0x054c, 0x006d, 0x0000, 0x9999,
 		"Sony",
 		"PEG Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* Submitted by Frank Engel <frankie@cse.unsw.edu.au> */
 UNUSUAL_DEV(  0x054c, 0x0099, 0x0000, 0x9999,
 		"Sony",
 		"PEG Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* Submitted by Mike Alborn <malborn@deandra.homeip.net> */
 UNUSUAL_DEV(  0x054c, 0x016a, 0x0000, 0x9999,
 		"Sony",
 		"PEG Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* floppy reports multiple luns */
 UNUSUAL_DEV(  0x055d, 0x2020, 0x0000, 0x0210,
 		"SAMSUNG",
 		"SFD-321U [FW 0C]",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* We keep this entry to force the transport; firmware 3.00 and later is ok. */
 UNUSUAL_DEV(  0x057b, 0x0000, 0x0000, 0x0299,
 		"Y-E Data",
 		"Flashbuster-U",
-		US_SC_DEVICE,  US_PR_CB, NULL,
+		USB_SC_DEVICE,  USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN),
 
 /* Reported by Johann Cardon <johann.cardon@free.fr>
@@ -664,20 +664,20 @@ UNUSUAL_DEV(  0x057b, 0x0000, 0x0000, 0x0299,
 UNUSUAL_DEV(  0x057b, 0x0022, 0x0000, 0x9999,
 		"Y-E Data",
 		"Silicon Media R/W",
-		US_SC_DEVICE, US_PR_DEVICE, NULL, 0),
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL, 0),
 
 /* Reported by RTE <raszilki@yandex.ru> */
 UNUSUAL_DEV(  0x058f, 0x6387, 0x0141, 0x0141,
 		"JetFlash",
 		"TS1GJF2A/120",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64 ),
 
 /* Fabrizio Fellini <fello@libero.it> */
 UNUSUAL_DEV(  0x0595, 0x4343, 0x0000, 0x2210,
 		"Fujifilm",
 		"Digital Camera EX-20 DSC",
-		US_SC_8070, US_PR_DEVICE, NULL, 0 ),
+		USB_SC_8070, USB_PR_DEVICE, NULL, 0 ),
 
 /* Reported by Andre Welter <a.r.welter@gmx.de>
  * This antique device predates the release of the Bulk-only Transport
@@ -688,14 +688,14 @@ UNUSUAL_DEV(  0x0595, 0x4343, 0x0000, 0x2210,
 UNUSUAL_DEV(  0x059b, 0x0001, 0x0100, 0x0100,
 		"Iomega",
 		"ZIP 100",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Reported by <Hendryk.Pfeiffer@gmx.de> */
 UNUSUAL_DEV(  0x059f, 0x0643, 0x0000, 0x0000,
 		"LaCie",
 		"DVD+-RW",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_GO_SLOW ),
 
 /* Submitted by Joel Bourquard <numlock@freesurf.ch>
@@ -705,7 +705,7 @@ UNUSUAL_DEV(  0x059f, 0x0643, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x05ab, 0x0060, 0x1104, 0x1110,
 		"In-System",
 		"PyroGate External CD-ROM Enclosure (FCD-523)",
-		US_SC_SCSI, US_PR_BULK, NULL,
+		USB_SC_SCSI, USB_PR_BULK, NULL,
 		US_FL_NEED_OVERRIDE ),
 
 /* Submitted by Sven Anderson <sven-linux@anderson.de>
@@ -717,26 +717,26 @@ UNUSUAL_DEV(  0x05ab, 0x0060, 0x1104, 0x1110,
 UNUSUAL_DEV( 0x05ac, 0x1202, 0x0000, 0x9999,
 		"Apple",
 		"iPod",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY ),
 
 /* Reported by Avi Kivity <avi@argo.co.il> */
 UNUSUAL_DEV( 0x05ac, 0x1203, 0x0000, 0x9999,
 		"Apple",
 		"iPod",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY ),
 
 UNUSUAL_DEV( 0x05ac, 0x1204, 0x0000, 0x9999,
 		"Apple",
 		"iPod",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_NOT_LOCKABLE ),
 
 UNUSUAL_DEV( 0x05ac, 0x1205, 0x0000, 0x9999,
 		"Apple",
 		"iPod",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY ),
 
 /*
@@ -746,7 +746,7 @@ UNUSUAL_DEV( 0x05ac, 0x1205, 0x0000, 0x9999,
 UNUSUAL_DEV( 0x05ac, 0x120a, 0x0000, 0x9999,
 		"Apple",
 		"iPod",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY ),
 
 /* Reported by Dan Williams <dcbw@redhat.com>
@@ -758,14 +758,14 @@ UNUSUAL_DEV( 0x05ac, 0x120a, 0x0000, 0x9999,
 UNUSUAL_DEV(  0x05c6, 0x1000, 0x0000, 0x9999,
 		"Option N.V.",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, option_ms_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, option_ms_init,
 		0),
 
 /* Reported by Blake Matheny <bmatheny@purdue.edu> */
 UNUSUAL_DEV(  0x05dc, 0xb002, 0x0000, 0x0113,
 		"Lexar",
 		"USB CF Reader",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* The following two entries are for a Genesys USB to IDE
@@ -782,20 +782,20 @@ UNUSUAL_DEV(  0x05dc, 0xb002, 0x0000, 0x0113,
 UNUSUAL_DEV(  0x05e3, 0x0701, 0x0000, 0xffff,
 		"Genesys Logic",
 		"USB to IDE Optical",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_GO_SLOW | US_FL_MAX_SECTORS_64 | US_FL_IGNORE_RESIDUE ),
 
 UNUSUAL_DEV(  0x05e3, 0x0702, 0x0000, 0xffff,
 		"Genesys Logic",
 		"USB to IDE Disk",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_GO_SLOW | US_FL_MAX_SECTORS_64 | US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Ben Efros <ben@pc-doctor.com> */
 UNUSUAL_DEV(  0x05e3, 0x0723, 0x9451, 0x9451,
 		"Genesys Logic",
 		"USB to SATA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SANE_SENSE ),
 
 /* Reported by Hanno Boeck <hanno@gmx.de>
@@ -803,33 +803,33 @@ UNUSUAL_DEV(  0x05e3, 0x0723, 0x9451, 0x9451,
 UNUSUAL_DEV(  0x0636, 0x0003, 0x0000, 0x9999,
 		"Vivitar",
 		"Vivicam 35Xx",
-		US_SC_SCSI, US_PR_BULK, NULL,
+		USB_SC_SCSI, USB_PR_BULK, NULL,
 		US_FL_FIX_INQUIRY ),
 
 UNUSUAL_DEV(  0x0644, 0x0000, 0x0100, 0x0100,
 		"TEAC",
 		"Floppy Drive",
-		US_SC_UFI, US_PR_CB, NULL, 0 ),
+		USB_SC_UFI, USB_PR_CB, NULL, 0 ),
 
 /* Reported by Darsen Lu <darsen@micro.ee.nthu.edu.tw> */
 UNUSUAL_DEV( 0x066f, 0x8000, 0x0001, 0x0001,
 		"SigmaTel",
 		"USBMSC Audio Player",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY ),
 
 /* Reported by Daniel Kukula <daniel.kuku@gmail.com> */
 UNUSUAL_DEV( 0x067b, 0x1063, 0x0100, 0x0100,
 		"Prolific Technology, Inc.",
 		"Prolific Storage Gadget",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_BAD_SENSE ),
 
 /* Reported by Rogerio Brito <rbrito@ime.usp.br> */
 UNUSUAL_DEV( 0x067b, 0x2317, 0x0001, 0x001,
 		"Prolific Technology, Inc.",
 		"Mass Storage Device",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NOT_LOCKABLE ),
 
 /* Reported by Richard -=[]=- <micro_flyer@hotmail.com> */
@@ -838,45 +838,45 @@ UNUSUAL_DEV( 0x067b, 0x2317, 0x0001, 0x001,
 UNUSUAL_DEV( 0x067b, 0x2507, 0x0001, 0x0100,
 		"Prolific Technology Inc.",
 		"Mass Storage Device",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_GO_SLOW ),
 
 /* Reported by Alex Butcher <alex.butcher@assursys.co.uk> */
 UNUSUAL_DEV( 0x067b, 0x3507, 0x0001, 0x0101,
 		"Prolific Technology Inc.",
 		"ATAPI-6 Bridge Controller",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_GO_SLOW ),
 
 /* Submitted by Benny Sjostrand <benny@hostmobility.com> */
 UNUSUAL_DEV( 0x0686, 0x4011, 0x0001, 0x0001,
 		"Minolta",
 		"Dimage F300",
-		US_SC_SCSI, US_PR_BULK, NULL, 0 ),
+		USB_SC_SCSI, USB_PR_BULK, NULL, 0 ),
 
 /* Reported by Miguel A. Fosas <amn3s1a@ono.com> */
 UNUSUAL_DEV(  0x0686, 0x4017, 0x0001, 0x0001,
 		"Minolta",
 		"DIMAGE E223",
-		US_SC_SCSI, US_PR_DEVICE, NULL, 0 ),
+		USB_SC_SCSI, USB_PR_DEVICE, NULL, 0 ),
 
 UNUSUAL_DEV(  0x0693, 0x0005, 0x0100, 0x0100,
 		"Hagiwara",
 		"Flashgate",
-		US_SC_SCSI, US_PR_BULK, NULL, 0 ),
+		USB_SC_SCSI, USB_PR_BULK, NULL, 0 ),
 
 /* Reported by David Hamilton <niftimusmaximus@lycos.com> */
 UNUSUAL_DEV(  0x069b, 0x3004, 0x0001, 0x0001,
 		"Thomson Multimedia Inc.",
 		"RCA RD1080 MP3 Player",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY ),
 
 /* Reported by Adrian Pilchowiec <adi1981@epf.pl> */
 UNUSUAL_DEV(  0x071b, 0x3203, 0x0000, 0x0000,
 		"RockChip",
 		"MP3",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NO_WP_DETECT | US_FL_MAX_SECTORS_64),
 
 /* Reported by Jean-Baptiste Onofre <jb@nanthrax.net>
@@ -886,7 +886,7 @@ UNUSUAL_DEV(  0x071b, 0x3203, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x071b, 0x32bb, 0x0000, 0x0000,
 		"RockChip",
 		"MTP",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NO_WP_DETECT | US_FL_MAX_SECTORS_64),
 
 /* Reported by Massimiliano Ghilardi <massimiliano.ghilardi@gmail.com>
@@ -902,59 +902,59 @@ UNUSUAL_DEV(  0x071b, 0x32bb, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x071b, 0x3203, 0x0100, 0x0100,
 		"RockChip",
 		"ROCK MP3",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_MAX_SECTORS_64),
 
 /* Reported by Olivier Blondeau <zeitoun@gmail.com> */
 UNUSUAL_DEV(  0x0727, 0x0306, 0x0100, 0x0100,
 		"ATMEL",
 		"SND1 Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE),
 
 /* Submitted by Roman Hodek <roman@hodek.net> */
 UNUSUAL_DEV(  0x0781, 0x0001, 0x0200, 0x0200,
 		"Sandisk",
 		"ImageMate SDDR-05a",
-		US_SC_SCSI, US_PR_CB, NULL,
+		USB_SC_SCSI, USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN ),
 
 UNUSUAL_DEV(  0x0781, 0x0002, 0x0009, 0x0009,
 		"SanDisk Corporation",
 		"ImageMate CompactFlash USB",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY ),
 
 UNUSUAL_DEV(  0x0781, 0x0100, 0x0100, 0x0100,
 		"Sandisk",
 		"ImageMate SDDR-12",
-		US_SC_SCSI, US_PR_CB, NULL,
+		USB_SC_SCSI, USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Reported by Eero Volotinen <eero@ping-viini.org> */
 UNUSUAL_DEV(  0x07ab, 0xfccd, 0x0000, 0x9999,
 		"Freecom Technologies",
 		"FHD-Classic",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY),
 
 UNUSUAL_DEV(  0x07af, 0x0004, 0x0100, 0x0133,
 		"Microtech",
 		"USB-SCSI-DB25",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_euscsi_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_euscsi_init,
 		US_FL_SCM_MULT_TARG ), 
 
 UNUSUAL_DEV(  0x07af, 0x0005, 0x0100, 0x0100,
 		"Microtech",
 		"USB-SCSI-HD50",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_euscsi_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_euscsi_init,
 		US_FL_SCM_MULT_TARG ),
 
 #ifdef NO_SDDR09
 UNUSUAL_DEV(  0x07af, 0x0006, 0x0100, 0x0100,
 		"Microtech",
 		"CameraMate",
-		US_SC_SCSI, US_PR_CB, NULL,
+		USB_SC_SCSI, USB_PR_CB, NULL,
 		US_FL_SINGLE_LUN ),
 #endif
 
@@ -967,7 +967,7 @@ UNUSUAL_DEV(  0x07af, 0x0006, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x07c4, 0xa400, 0x0000, 0xffff,
 		"Datafab",
 		"KECF-USB",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY | US_FL_FIX_CAPACITY ),
 
 /* Reported by Rauch Wolke <rauchwolke@gmx.net>
@@ -976,7 +976,7 @@ UNUSUAL_DEV(  0x07c4, 0xa400, 0x0000, 0xffff,
 UNUSUAL_DEV(  0x07c4, 0xa4a5, 0x0000, 0xffff,
 		"Simple Tech/Datafab",
 		"CF+SM Reader",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_MAX_SECTORS_64 ),
 
 /* Casio QV 2x00/3x00/4000/8000 digital still cameras are not conformant
@@ -986,42 +986,42 @@ UNUSUAL_DEV(  0x07c4, 0xa4a5, 0x0000, 0xffff,
  * - They don't like the INQUIRY command. So we must handle this command
  *   of the SCSI layer ourselves.
  * - Some cameras with idProduct=0x1001 and bcdDevice=0x1000 have
- *   bInterfaceProtocol=0x00 (US_PR_CBI) while others have 0x01 (US_PR_CB).
- *   So don't remove the US_PR_CB override!
- * - Cameras with bcdDevice=0x9009 require the US_SC_8070 override.
+ *   bInterfaceProtocol=0x00 (USB_PR_CBI) while others have 0x01 (USB_PR_CB).
+ *   So don't remove the USB_PR_CB override!
+ * - Cameras with bcdDevice=0x9009 require the USB_SC_8070 override.
  */
 UNUSUAL_DEV( 0x07cf, 0x1001, 0x1000, 0x9999,
 		"Casio",
 		"QV DigitalCamera",
-		US_SC_8070, US_PR_CB, NULL,
+		USB_SC_8070, USB_PR_CB, NULL,
 		US_FL_NEED_OVERRIDE | US_FL_FIX_INQUIRY ),
 
 /* Submitted by Hartmut Wahl <hwahl@hwahl.de>*/
 UNUSUAL_DEV( 0x0839, 0x000a, 0x0001, 0x0001,
 		"Samsung",
 		"Digimax 410",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY),
 
 /* Reported by Luciano Rocha <luciano@eurotux.com> */
 UNUSUAL_DEV( 0x0840, 0x0082, 0x0001, 0x0001,
 		"Argosy",
 		"Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY),
 
 /* Reported and patched by Nguyen Anh Quynh <aquynh@gmail.com> */
 UNUSUAL_DEV( 0x0840, 0x0084, 0x0001, 0x0001,
 		"Argosy",
 		"Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY),
 
 /* Reported by Martijn Hijdra <martijn.hijdra@gmail.com> */
 UNUSUAL_DEV( 0x0840, 0x0085, 0x0001, 0x0001,
 		"Argosy",
 		"Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY),
 
 /* Entry and supporting patch by Theodore Kilgore <kilgota@auburn.edu>.
@@ -1033,7 +1033,7 @@ UNUSUAL_DEV( 0x0840, 0x0085, 0x0001, 0x0001,
 UNUSUAL_DEV(  0x084d, 0x0011, 0x0110, 0x0110,
 		"Grandtech",
 		"DC2MEGA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_BULK32),
 
 /* Andrew Lunn <andrew@lunn.ch>
@@ -1044,14 +1044,14 @@ UNUSUAL_DEV(  0x084d, 0x0011, 0x0110, 0x0110,
 UNUSUAL_DEV(  0x0851, 0x1543, 0x0200, 0x0200,
 		"PanDigital",
 		"Photo Frame",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NOT_LOCKABLE),
 
 /* Submitted by Jan De Luyck <lkml@kcore.org> */
 UNUSUAL_DEV(  0x08bd, 0x1100, 0x0000, 0x0000,
 		"CITIZEN",
 		"X1DE-USB",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN),
 
 /* Submitted by Dylan Taft <d13f00l@gmail.com>
@@ -1060,7 +1060,7 @@ UNUSUAL_DEV(  0x08bd, 0x1100, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x08ca, 0x3103, 0x0100, 0x0100,
 		"AIPTEK",
 		"Aiptek USB Keychain MP3 Player",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE),
 
 /* Entry needed for flags. Moreover, all devices with this ID use
@@ -1071,7 +1071,7 @@ UNUSUAL_DEV(  0x08ca, 0x3103, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x090a, 0x1001, 0x0100, 0x0100,
 		"Trumpion",
 		"t33520 USB Flash Card Controller",
-		US_SC_DEVICE, US_PR_BULK, NULL,
+		USB_SC_DEVICE, USB_PR_BULK, NULL,
 		US_FL_NEED_OVERRIDE ),
 
 /* Reported by Filippo Bardelli <filibard@libero.it>
@@ -1080,21 +1080,21 @@ UNUSUAL_DEV(  0x090a, 0x1001, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x090a, 0x1050, 0x0100, 0x0100,
 		"Trumpion Microelectronics, Inc.",
 		"33520 USB Digital Voice Recorder",
-		US_SC_UFI, US_PR_DEVICE, NULL,
+		USB_SC_UFI, USB_PR_DEVICE, NULL,
 		0),
 
 /* Trumpion Microelectronics MP3 player (felipe_alfaro@linuxmail.org) */
 UNUSUAL_DEV( 0x090a, 0x1200, 0x0000, 0x9999,
 		"Trumpion",
 		"MP3 player",
-		US_SC_RBC, US_PR_BULK, NULL,
+		USB_SC_RBC, USB_PR_BULK, NULL,
 		0 ),
 
 /* aeb */
 UNUSUAL_DEV( 0x090c, 0x1132, 0x0000, 0xffff,
 		"Feiya",
 		"5-in-1 Card Reader",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY ),
 
 /* This Pentax still camera is not conformant
@@ -1107,7 +1107,7 @@ UNUSUAL_DEV( 0x090c, 0x1132, 0x0000, 0xffff,
 UNUSUAL_DEV( 0x0a17, 0x0004, 0x1000, 0x1000,
 		"Pentax",
 		"Optio 2/3/400",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* These are virtual windows driver CDs, which the zd1211rw driver
@@ -1115,13 +1115,13 @@ UNUSUAL_DEV( 0x0a17, 0x0004, 0x1000, 0x1000,
 UNUSUAL_DEV( 0x0ace, 0x2011, 0x0101, 0x0101,
 		"ZyXEL",
 		"G-220F USB-WLAN Install",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_DEVICE ),
 
 UNUSUAL_DEV( 0x0ace, 0x20ff, 0x0101, 0x0101,
 		"SiteCom",
 		"WL-117 USB-WLAN Install",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_DEVICE ),
 
 /* Reported by Dan Williams <dcbw@redhat.com>
@@ -1133,7 +1133,7 @@ UNUSUAL_DEV( 0x0ace, 0x20ff, 0x0101, 0x0101,
 UNUSUAL_DEV(  0x0af0, 0x6971, 0x0000, 0x9999,
 		"Option N.V.",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, option_ms_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, option_ms_init,
 		0),
 
 /* Reported by F. Aben <f.aben@option.com>
@@ -1143,7 +1143,7 @@ UNUSUAL_DEV(  0x0af0, 0x6971, 0x0000, 0x9999,
 UNUSUAL_DEV( 0x0af0, 0x7401, 0x0000, 0x0000,
 		"Option",
 		"GI 0401 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 /* Reported by Jan Dumon <j.dumon@option.com>
@@ -1153,104 +1153,104 @@ UNUSUAL_DEV( 0x0af0, 0x7401, 0x0000, 0x0000,
 UNUSUAL_DEV( 0x0af0, 0x7501, 0x0000, 0x0000,
 		"Option",
 		"GI 0431 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0x7701, 0x0000, 0x0000,
 		"Option",
 		"GI 0451 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0x7706, 0x0000, 0x0000,
 		"Option",
 		"GI 0451 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0x7901, 0x0000, 0x0000,
 		"Option",
 		"GI 0452 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0x7A01, 0x0000, 0x0000,
 		"Option",
 		"GI 0461 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0x7A05, 0x0000, 0x0000,
 		"Option",
 		"GI 0461 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0x8300, 0x0000, 0x0000,
 		"Option",
 		"GI 033x SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0x8302, 0x0000, 0x0000,
 		"Option",
 		"GI 033x SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0x8304, 0x0000, 0x0000,
 		"Option",
 		"GI 033x SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0xc100, 0x0000, 0x0000,
 		"Option",
 		"GI 070x SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0xd057, 0x0000, 0x0000,
 		"Option",
 		"GI 1505 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0xd058, 0x0000, 0x0000,
 		"Option",
 		"GI 1509 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0xd157, 0x0000, 0x0000,
 		"Option",
 		"GI 1515 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0xd257, 0x0000, 0x0000,
 		"Option",
 		"GI 1215 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 UNUSUAL_DEV( 0x0af0, 0xd357, 0x0000, 0x0000,
 		"Option",
 		"GI 1505 SD-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 /* Reported by Ben Efros <ben@pc-doctor.com> */
 UNUSUAL_DEV( 0x0bc2, 0x3010, 0x0000, 0x0000,
 		"Seagate",
 		"FreeAgent Pro",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SANE_SENSE ),
 
 UNUSUAL_DEV(  0x0d49, 0x7310, 0x0000, 0x9999,
 		"Maxtor",
 		"USB to SATA",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SANE_SENSE),
 
 /*
@@ -1260,14 +1260,14 @@ UNUSUAL_DEV(  0x0d49, 0x7310, 0x0000, 0x9999,
 UNUSUAL_DEV( 0x0c45, 0x1060, 0x0100, 0x0100,
 		"Unknown",
 		"Unknown",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SINGLE_LUN ),
 
 /* Submitted by Joris Struyve <joris@struyve.be> */
 UNUSUAL_DEV( 0x0d96, 0x410a, 0x0001, 0xffff,
 		"Medion",
 		"MD 7425",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY),
 
 /*
@@ -1278,13 +1278,13 @@ UNUSUAL_DEV( 0x0d96, 0x410a, 0x0001, 0xffff,
 UNUSUAL_DEV(  0x0d96, 0x5200, 0x0001, 0x0200,
 		"Jenoptik",
 		"JD 5200 z3",
-		US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_INQUIRY),
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_FIX_INQUIRY),
 
 /* Reported by  Jason Johnston <killean@shaw.ca> */
 UNUSUAL_DEV(  0x0dc4, 0x0073, 0x0000, 0x0000,
 		"Macpower Technology Co.LTD.",
 		"USB 2.0 3.5\" DEVICE",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY),
 
 /* Reported by Lubomir Blaha <tritol@trilogic.cz>
@@ -1295,7 +1295,7 @@ UNUSUAL_DEV(  0x0dc4, 0x0073, 0x0000, 0x0000,
 UNUSUAL_DEV( 0x0dd8, 0x1060, 0x0000, 0xffff,
 		"Netac",
 		"USB-CF-Card",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* Reported by Edward Chapman (taken from linux-usb mailing list)
@@ -1303,7 +1303,7 @@ UNUSUAL_DEV( 0x0dd8, 0x1060, 0x0000, 0xffff,
 UNUSUAL_DEV( 0x0dd8, 0xd202, 0x0000, 0x9999,
 		"Netac",
 		"USB Flash Disk",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 
@@ -1312,28 +1312,28 @@ UNUSUAL_DEV( 0x0dd8, 0xd202, 0x0000, 0x9999,
 UNUSUAL_DEV( 0x0dda, 0x0001, 0x0012, 0x0012,
 		"WINWARD",
 		"Music Disk",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Ian McConnell <ian at emit.demon.co.uk> */
 UNUSUAL_DEV(  0x0dda, 0x0301, 0x0012, 0x0012,
 		"PNP_MP3",
 		"PNP_MP3 PLAYER",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Jim McCloskey <mcclosk@ucsc.edu> */
 UNUSUAL_DEV( 0x0e21, 0x0520, 0x0100, 0x0100,
 		"Cowon Systems",
 		"iAUDIO M5",
-		US_SC_DEVICE, US_PR_BULK, NULL,
+		USB_SC_DEVICE, USB_PR_BULK, NULL,
 		US_FL_NEED_OVERRIDE ),
 
 /* Submitted by Antoine Mairesse <antoine.mairesse@free.fr> */
 UNUSUAL_DEV( 0x0ed1, 0x6660, 0x0100, 0x0300,
 		"USB",
 		"Solid state disk",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
 /* Submitted by Daniel Drake <dsd@gentoo.org>
@@ -1341,14 +1341,14 @@ UNUSUAL_DEV( 0x0ed1, 0x6660, 0x0100, 0x0300,
 UNUSUAL_DEV(  0x0ea0, 0x2168, 0x0110, 0x0110,
 		"Ours Technology",
 		"Flash Disk",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Rastislav Stanik <rs_kernel@yahoo.com> */
 UNUSUAL_DEV(  0x0ea0, 0x6828, 0x0110, 0x0110,
 		"USB",
 		"Flash Disk",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Benjamin Schiller <sbenni@gmx.de>
@@ -1356,7 +1356,7 @@ UNUSUAL_DEV(  0x0ea0, 0x6828, 0x0110, 0x0110,
 UNUSUAL_DEV(  0x0ed1, 0x7636, 0x0103, 0x0103,
 		"Typhoon",
 		"My DJ 1820",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_GO_SLOW | US_FL_MAX_SECTORS_64),
 
 /* Patch by Leonid Petrov mail at lpetrov.net
@@ -1367,7 +1367,7 @@ UNUSUAL_DEV(  0x0ed1, 0x7636, 0x0103, 0x0103,
 UNUSUAL_DEV(  0x0f19, 0x0103, 0x0100, 0x0100,
 		"Oracom Co., Ltd",
 		"ORC-200M",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* David Kuehling <dvdkhlng@gmx.de>:
@@ -1377,21 +1377,21 @@ UNUSUAL_DEV(  0x0f19, 0x0103, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x0f19, 0x0105, 0x0100, 0x0100,
 		"C-MEX",
 		"A-VOX",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Michael Stattmann <michael@stattmann.com> */
 UNUSUAL_DEV(  0x0fce, 0xd008, 0x0000, 0x0000,
 		"Sony Ericsson",
 		"V800-Vodafone 802",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NO_WP_DETECT ),
 
 /* Reported by The Solutor <thesolutor@gmail.com> */
 UNUSUAL_DEV(  0x0fce, 0xd0e1, 0x0000, 0x0000,
 		"Sony Ericsson",
 		"MD400",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_DEVICE),
 
 /* Reported by Jan Mate <mate@fiit.stuba.sk>
@@ -1399,21 +1399,21 @@ UNUSUAL_DEV(  0x0fce, 0xd0e1, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x0fce, 0xe030, 0x0000, 0x0000,
 		"Sony Ericsson",
 		"P990i",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Emmanuel Vasilakis <evas@forthnet.gr> */
 UNUSUAL_DEV(  0x0fce, 0xe031, 0x0000, 0x0000,
 		"Sony Ericsson",
 		"M600i",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_FIX_CAPACITY ),
 
 /* Reported by Ricardo Barberis <ricardo@dattatec.com> */
 UNUSUAL_DEV(  0x0fce, 0xe092, 0x0000, 0x0000,
 		"Sony Ericsson",
 		"P1i",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Kevin Cernekee <kpc-usbdev@gelato.uiuc.edu>
@@ -1425,13 +1425,13 @@ UNUSUAL_DEV(  0x0fce, 0xe092, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x1019, 0x0c55, 0x0000, 0x0110,
 		"Desknote",
 		"UCR-61S2B",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_ucr61s2b_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_ucr61s2b_init,
 		0 ),
 
 UNUSUAL_DEV(  0x1058, 0x0704, 0x0000, 0x9999,
 		"Western Digital",
 		"External HDD",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_SANE_SENSE),
 
 /* Reported by Fabio Venturi <f.venturi@tdnet.it>
@@ -1440,7 +1440,7 @@ UNUSUAL_DEV(  0x1058, 0x0704, 0x0000, 0x9999,
 UNUSUAL_DEV(  0x10d6, 0x2200, 0x0100, 0x0100,
 		"Actions Semiconductor",
 		"Mtp device",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0),
 
 /* Reported by Pascal Terjan <pterjan@mandriva.com>
@@ -1449,7 +1449,7 @@ UNUSUAL_DEV(  0x10d6, 0x2200, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x1186, 0x3e04, 0x0000, 0x0000,
            "D-Link",
            "USB Mass Storage",
-           US_SC_DEVICE, US_PR_DEVICE, option_ms_init, US_FL_IGNORE_DEVICE),
+           USB_SC_DEVICE, USB_PR_DEVICE, option_ms_init, US_FL_IGNORE_DEVICE),
 
 /* Reported by Kevin Lloyd <linux@sierrawireless.com>
  * Entry is needed for the initializer function override,
@@ -1459,7 +1459,7 @@ UNUSUAL_DEV(  0x1186, 0x3e04, 0x0000, 0x0000,
 UNUSUAL_DEV(  0x1199, 0x0fff, 0x0000, 0x9999,
 		"Sierra Wireless",
 		"USB MMC Storage",
-		US_SC_DEVICE, US_PR_DEVICE, sierra_ms_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, sierra_ms_init,
 		0),
 
 /* Reported by Jaco Kroon <jaco@kroon.co.za>
@@ -1469,7 +1469,7 @@ UNUSUAL_DEV(  0x1199, 0x0fff, 0x0000, 0x9999,
 UNUSUAL_DEV(  0x1210, 0x0003, 0x0100, 0x0100,
 		"Digitech HMG",
 		"DigiTech Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by fangxiaozhi <huananhu@huawei.com>
@@ -1478,353 +1478,353 @@ UNUSUAL_DEV(  0x1210, 0x0003, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x12d1, 0x1001, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1003, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1004, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1401, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1402, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1403, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1404, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1405, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1406, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1407, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1408, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1409, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x140A, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x140B, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x140C, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x140D, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x140E, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x140F, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1410, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1411, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1412, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1413, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1414, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1415, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1416, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1417, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1418, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1419, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x141A, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x141B, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x141C, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x141D, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x141E, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x141F, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1420, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1421, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1422, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1423, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1424, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1425, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1426, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1427, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1428, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1429, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x142A, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x142B, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x142C, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x142D, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x142E, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x142F, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1430, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1431, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1432, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1433, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1434, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1435, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1436, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1437, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1438, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x1439, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x143A, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x143B, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x143C, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x143D, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x143E, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 UNUSUAL_DEV(  0x12d1, 0x143F, 0x0000, 0x0000,
 		"HUAWEI MOBILE",
 		"Mass Storage",
-		US_SC_DEVICE, US_PR_DEVICE, usb_stor_huawei_e220_init,
+		USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
 		0),
 
 /* Reported by Vilius Bilinkevicius <vilisas AT xxx DOT lt) */
 UNUSUAL_DEV(  0x132b, 0x000b, 0x0001, 0x0001,
 		"Minolta",
 		"Dimage Z10",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		0 ),
 
 /* Reported by Kotrla Vitezslav <kotrla@ceb.cz> */
 UNUSUAL_DEV(  0x1370, 0x6828, 0x0110, 0x0110,
 		"SWISSBIT",
 		"Black Silver",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Francesco Foresti <frafore@tiscali.it> */
 UNUSUAL_DEV(  0x14cd, 0x6600, 0x0201, 0x0201,
 		"Super Top",
 		"IDE DEVICE",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Alexandre Oliva <oliva@lsd.ic.unicamp.br>
@@ -1833,7 +1833,7 @@ UNUSUAL_DEV(  0x14cd, 0x6600, 0x0201, 0x0201,
 UNUSUAL_DEV(  0x152d, 0x2329, 0x0100, 0x0100,
 		"JMicron",
 		"USB to ATA/ATAPI Bridge",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE | US_FL_SANE_SENSE ),
 
 /* Reported by Robert Schedel <r.schedel@yahoo.de>
@@ -1841,7 +1841,7 @@ UNUSUAL_DEV(  0x152d, 0x2329, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x1652, 0x6600, 0x0201, 0x0201,
 		"Teac",
 		"HD-35PUK-B",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Hans de Goede <hdegoede@redhat.com>
@@ -1851,18 +1851,18 @@ UNUSUAL_DEV(  0x1652, 0x6600, 0x0201, 0x0201,
 UNUSUAL_DEV( 0x1908, 0x1315, 0x0000, 0x0000,
 		"BUILDWIN",
 		"Photo Frame",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_BAD_SENSE ),
 UNUSUAL_DEV( 0x1908, 0x1320, 0x0000, 0x0000,
 		"BUILDWIN",
 		"Photo Frame",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_BAD_SENSE ),
 
 UNUSUAL_DEV( 0x2116, 0x0320, 0x0001, 0x0001,
 		"ST",
 		"2A",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY),
 
 /* patch submitted by Davide Perini <perini.davide@dpsoftware.org>
@@ -1871,7 +1871,7 @@ UNUSUAL_DEV( 0x2116, 0x0320, 0x0001, 0x0001,
 UNUSUAL_DEV(  0x22b8, 0x3010, 0x0001, 0x0001,
 		"Motorola",
 		"RAZR V3x",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_CAPACITY | US_FL_IGNORE_RESIDUE ),
 
 /*
@@ -1882,14 +1882,14 @@ UNUSUAL_DEV(  0x22b8, 0x3010, 0x0001, 0x0001,
 UNUSUAL_DEV(  0x22b8, 0x6426, 0x0101, 0x0101,
 		"Motorola",
 		"MSnc.",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY | US_FL_FIX_CAPACITY | US_FL_BULK_IGNORE_TAG),
 
 /* Reported by Radovan Garabik <garabik@kassiopeia.juls.savba.sk> */
 UNUSUAL_DEV(  0x2735, 0x100b, 0x0000, 0x9999,
 		"MPIO",
 		"HS200",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_GO_SLOW ),
 
 /* Reported by Frederic Marchal <frederic.marchal@wowcompany.com>
@@ -1898,21 +1898,21 @@ UNUSUAL_DEV(  0x2735, 0x100b, 0x0000, 0x9999,
 UNUSUAL_DEV(  0x3340, 0xffff, 0x0000, 0x0000,
 		"Mitac",
 		"Mio DigiWalker USB Sync",
-		US_SC_DEVICE,US_PR_DEVICE,NULL,
+		USB_SC_DEVICE,USB_PR_DEVICE,NULL,
 		US_FL_MAX_SECTORS_64 ),
 
 /* Reported by Andrey Rahmatullin <wrar@altlinux.org> */
 UNUSUAL_DEV(  0x4102, 0x1020, 0x0100,  0x0100,
 		"iRiver",
 		"MP3 T10",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
 /* Reported by Sergey Pinaev <dfo@antex.ru> */
 UNUSUAL_DEV(  0x4102, 0x1059, 0x0000,  0x0000,
                "iRiver",
                "P7K",
-               US_SC_DEVICE, US_PR_DEVICE, NULL,
+               USB_SC_DEVICE, USB_PR_DEVICE, NULL,
                US_FL_MAX_SECTORS_64 ),
 
 /*
@@ -1922,41 +1922,41 @@ UNUSUAL_DEV(  0x4102, 0x1059, 0x0000,  0x0000,
 UNUSUAL_DEV(  0x4146, 0xba01, 0x0100, 0x0100,
 		"Iomega",
 		"Micro Mini 1GB",
-		US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_NOT_LOCKABLE ),
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_NOT_LOCKABLE ),
 
 /* Reported by Andrew Simmons <andrew.simmons@gmail.com> */
 UNUSUAL_DEV(  0xed06, 0x4500, 0x0001, 0x0001,
 		"DataStor",
 		"USB4500 FW1.04",
-		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_CAPACITY_HEURISTICS),
 
 /* Reported by Alessio Treglia <quadrispro@ubuntu.com> */
 UNUSUAL_DEV( 0xed10, 0x7636, 0x0001, 0x0001,
 		"TGE",
 		"Digital MP3 Audio Player",
-		US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_NOT_LOCKABLE ),
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_NOT_LOCKABLE ),
 
 /* Control/Bulk transport for all SubClass values */
-USUAL_DEV(US_SC_RBC, US_PR_CB, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_8020, US_PR_CB, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_QIC, US_PR_CB, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_UFI, US_PR_CB, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_8070, US_PR_CB, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_SCSI, US_PR_CB, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_RBC, USB_PR_CB, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_8020, USB_PR_CB, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_QIC, USB_PR_CB, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_UFI, USB_PR_CB, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_8070, USB_PR_CB, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_SCSI, USB_PR_CB, USB_US_TYPE_STOR),
 
 /* Control/Bulk/Interrupt transport for all SubClass values */
-USUAL_DEV(US_SC_RBC, US_PR_CBI, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_8020, US_PR_CBI, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_QIC, US_PR_CBI, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_UFI, US_PR_CBI, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_8070, US_PR_CBI, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_SCSI, US_PR_CBI, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_RBC, USB_PR_CBI, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_8020, USB_PR_CBI, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_QIC, USB_PR_CBI, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_UFI, USB_PR_CBI, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_8070, USB_PR_CBI, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_SCSI, USB_PR_CBI, USB_US_TYPE_STOR),
 
 /* Bulk-only transport for all SubClass values */
-USUAL_DEV(US_SC_RBC, US_PR_BULK, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_8020, US_PR_BULK, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_QIC, US_PR_BULK, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_UFI, US_PR_BULK, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_8070, US_PR_BULK, USB_US_TYPE_STOR),
-USUAL_DEV(US_SC_SCSI, US_PR_BULK, 0),
+USUAL_DEV(USB_SC_RBC, USB_PR_BULK, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_8020, USB_PR_BULK, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_QIC, USB_PR_BULK, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_UFI, USB_PR_BULK, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_8070, USB_PR_BULK, USB_US_TYPE_STOR),
+USUAL_DEV(USB_SC_SCSI, USB_PR_BULK, 0),
diff --git a/drivers/usb/storage/unusual_freecom.h b/drivers/usb/storage/unusual_freecom.h
index 375867942391..59a261155b98 100644
--- a/drivers/usb/storage/unusual_freecom.h
+++ b/drivers/usb/storage/unusual_freecom.h
@@ -21,6 +21,6 @@
 UNUSUAL_DEV(  0x07ab, 0xfc01, 0x0000, 0x9999,
 		"Freecom",
 		"USB-IDE",
-		US_SC_QIC, US_PR_FREECOM, init_freecom, 0),
+		USB_SC_QIC, USB_PR_FREECOM, init_freecom, 0),
 
 #endif /* defined(CONFIG_USB_STORAGE_FREECOM) || ... */
diff --git a/drivers/usb/storage/unusual_isd200.h b/drivers/usb/storage/unusual_isd200.h
index 0d99dde3382a..14cca0c48302 100644
--- a/drivers/usb/storage/unusual_isd200.h
+++ b/drivers/usb/storage/unusual_isd200.h
@@ -21,37 +21,37 @@
 UNUSUAL_DEV(  0x054c, 0x002b, 0x0100, 0x0110,
 		"Sony",
 		"Portable USB Harddrive V2",
-		US_SC_ISD200, US_PR_BULK, isd200_Initialization,
+		USB_SC_ISD200, USB_PR_BULK, isd200_Initialization,
 		0),
 
 UNUSUAL_DEV(  0x05ab, 0x0031, 0x0100, 0x0110,
 		"In-System",
 		"USB/IDE Bridge (ATA/ATAPI)",
-		US_SC_ISD200, US_PR_BULK, isd200_Initialization,
+		USB_SC_ISD200, USB_PR_BULK, isd200_Initialization,
 		0),
 
 UNUSUAL_DEV(  0x05ab, 0x0301, 0x0100, 0x0110,
 		"In-System",
 		"Portable USB Harddrive V2",
-		US_SC_ISD200, US_PR_BULK, isd200_Initialization,
+		USB_SC_ISD200, USB_PR_BULK, isd200_Initialization,
 		0),
 
 UNUSUAL_DEV(  0x05ab, 0x0351, 0x0100, 0x0110,
 		"In-System",
 		"Portable USB Harddrive V2",
-		US_SC_ISD200, US_PR_BULK, isd200_Initialization,
+		USB_SC_ISD200, USB_PR_BULK, isd200_Initialization,
 		0),
 
 UNUSUAL_DEV(  0x05ab, 0x5701, 0x0100, 0x0110,
 		"In-System",
 		"USB Storage Adapter V2",
-		US_SC_ISD200, US_PR_BULK, isd200_Initialization,
+		USB_SC_ISD200, USB_PR_BULK, isd200_Initialization,
 		0),
 
 UNUSUAL_DEV(  0x0bf6, 0xa001, 0x0100, 0x0110,
 		"ATI",
 		"USB Cable 205",
-		US_SC_ISD200, US_PR_BULK, isd200_Initialization,
+		USB_SC_ISD200, USB_PR_BULK, isd200_Initialization,
 		0),
 
 #endif /* defined(CONFIG_USB_STORAGE_ISD200) || ... */
diff --git a/drivers/usb/storage/unusual_jumpshot.h b/drivers/usb/storage/unusual_jumpshot.h
index 2e549b1c2c62..54be78b5d643 100644
--- a/drivers/usb/storage/unusual_jumpshot.h
+++ b/drivers/usb/storage/unusual_jumpshot.h
@@ -21,7 +21,7 @@
 UNUSUAL_DEV(  0x05dc, 0x0001, 0x0000, 0x0001,
 		"Lexar",
 		"Jumpshot USB CF Reader",
-		US_SC_SCSI, US_PR_JUMPSHOT, NULL,
+		USB_SC_SCSI, USB_PR_JUMPSHOT, NULL,
 		US_FL_NEED_OVERRIDE),
 
 #endif /* defined(CONFIG_USB_STORAGE_JUMPSHOT) || ... */
diff --git a/drivers/usb/storage/unusual_karma.h b/drivers/usb/storage/unusual_karma.h
index 12ae3a03e802..6df03972a22c 100644
--- a/drivers/usb/storage/unusual_karma.h
+++ b/drivers/usb/storage/unusual_karma.h
@@ -21,6 +21,6 @@
 UNUSUAL_DEV(  0x045a, 0x5210, 0x0101, 0x0101,
 		"Rio",
 		"Rio Karma",
-		US_SC_SCSI, US_PR_KARMA, rio_karma_init, 0),
+		USB_SC_SCSI, USB_PR_KARMA, rio_karma_init, 0),
 
 #endif /* defined(CONFIG_USB_STORAGE_KARMA) || ... */
diff --git a/drivers/usb/storage/unusual_onetouch.h b/drivers/usb/storage/unusual_onetouch.h
index bd9306b637df..0abb819c7405 100644
--- a/drivers/usb/storage/unusual_onetouch.h
+++ b/drivers/usb/storage/unusual_onetouch.h
@@ -24,13 +24,13 @@
 UNUSUAL_DEV(  0x0d49, 0x7000, 0x0000, 0x9999,
 		"Maxtor",
 		"OneTouch External Harddrive",
-		US_SC_DEVICE, US_PR_DEVICE, onetouch_connect_input,
+		USB_SC_DEVICE, USB_PR_DEVICE, onetouch_connect_input,
 		0),
 
 UNUSUAL_DEV(  0x0d49, 0x7010, 0x0000, 0x9999,
 		"Maxtor",
 		"OneTouch External Harddrive",
-		US_SC_DEVICE, US_PR_DEVICE, onetouch_connect_input,
+		USB_SC_DEVICE, USB_PR_DEVICE, onetouch_connect_input,
 		0),
 
 #endif /* defined(CONFIG_USB_STORAGE_ONETOUCH) || ... */
diff --git a/drivers/usb/storage/unusual_sddr09.h b/drivers/usb/storage/unusual_sddr09.h
index 50cab511a4d7..59a7e37b6c11 100644
--- a/drivers/usb/storage/unusual_sddr09.h
+++ b/drivers/usb/storage/unusual_sddr09.h
@@ -21,36 +21,36 @@
 UNUSUAL_DEV(  0x0436, 0x0005, 0x0100, 0x0100,
 		"Microtech",
 		"CameraMate (DPCM_USB)",
-		US_SC_SCSI, US_PR_DPCM_USB, NULL, 0),
+		USB_SC_SCSI, USB_PR_DPCM_USB, NULL, 0),
 
 UNUSUAL_DEV(  0x04e6, 0x0003, 0x0000, 0x9999,
 		"Sandisk",
 		"ImageMate SDDR09",
-		US_SC_SCSI, US_PR_EUSB_SDDR09, usb_stor_sddr09_init,
+		USB_SC_SCSI, USB_PR_EUSB_SDDR09, usb_stor_sddr09_init,
 		0),
 
 /* This entry is from Andries.Brouwer@cwi.nl */
 UNUSUAL_DEV(  0x04e6, 0x0005, 0x0100, 0x0208,
 		"SCM Microsystems",
 		"eUSB SmartMedia / CompactFlash Adapter",
-		US_SC_SCSI, US_PR_DPCM_USB, usb_stor_sddr09_dpcm_init,
+		USB_SC_SCSI, USB_PR_DPCM_USB, usb_stor_sddr09_dpcm_init,
 		0),
 
 UNUSUAL_DEV(  0x066b, 0x0105, 0x0100, 0x0100,
 		"Olympus",
 		"Camedia MAUSB-2",
-		US_SC_SCSI, US_PR_EUSB_SDDR09, usb_stor_sddr09_init,
+		USB_SC_SCSI, USB_PR_EUSB_SDDR09, usb_stor_sddr09_init,
 		0),
 
 UNUSUAL_DEV(  0x0781, 0x0200, 0x0000, 0x9999,
 		"Sandisk",
 		"ImageMate SDDR-09",
-		US_SC_SCSI, US_PR_EUSB_SDDR09, usb_stor_sddr09_init,
+		USB_SC_SCSI, USB_PR_EUSB_SDDR09, usb_stor_sddr09_init,
 		0),
 
 UNUSUAL_DEV(  0x07af, 0x0006, 0x0100, 0x0100,
 		"Microtech",
 		"CameraMate (DPCM_USB)",
-		US_SC_SCSI, US_PR_DPCM_USB, NULL, 0),
+		USB_SC_SCSI, USB_PR_DPCM_USB, NULL, 0),
 
 #endif /* defined(CONFIG_USB_STORAGE_SDDR09) || ... */
diff --git a/drivers/usb/storage/unusual_sddr55.h b/drivers/usb/storage/unusual_sddr55.h
index ae81ef7a1cfd..fcb7e12c598f 100644
--- a/drivers/usb/storage/unusual_sddr55.h
+++ b/drivers/usb/storage/unusual_sddr55.h
@@ -22,23 +22,23 @@
 UNUSUAL_DEV( 0x07c4, 0xa103, 0x0000, 0x9999,
 		"Datafab",
 		"MDSM-B reader",
-		US_SC_SCSI, US_PR_SDDR55, NULL,
+		USB_SC_SCSI, USB_PR_SDDR55, NULL,
 		US_FL_FIX_INQUIRY),
 
 /* SM part - aeb <Andries.Brouwer@cwi.nl> */
 UNUSUAL_DEV(  0x07c4, 0xa109, 0x0000, 0xffff,
 		"Datafab Systems, Inc.",
 		"USB to CF + SM Combo (LC1)",
-		US_SC_SCSI, US_PR_SDDR55, NULL, 0),
+		USB_SC_SCSI, USB_PR_SDDR55, NULL, 0),
 
 UNUSUAL_DEV( 0x0c0b, 0xa109, 0x0000, 0xffff,
 		"Acomdata",
 		"SM",
-		US_SC_SCSI, US_PR_SDDR55, NULL, 0),
+		USB_SC_SCSI, USB_PR_SDDR55, NULL, 0),
 
 UNUSUAL_DEV(  0x55aa, 0xa103, 0x0000, 0x9999,
 		"Sandisk",
 		"ImageMate SDDR55",
-		US_SC_SCSI, US_PR_SDDR55, NULL, 0),
+		USB_SC_SCSI, USB_PR_SDDR55, NULL, 0),
 
 #endif /* defined(CONFIG_USB_STORAGE_SDDR55) || ... */
diff --git a/drivers/usb/storage/unusual_usbat.h b/drivers/usb/storage/unusual_usbat.h
index 80e869f10180..38e79c4e6d6a 100644
--- a/drivers/usb/storage/unusual_usbat.h
+++ b/drivers/usb/storage/unusual_usbat.h
@@ -21,23 +21,23 @@
 UNUSUAL_DEV(  0x03f0, 0x0207, 0x0001, 0x0001,
 		"HP",
 		"CD-Writer+ 8200e",
-		US_SC_8070, US_PR_USBAT, init_usbat_cd, 0),
+		USB_SC_8070, USB_PR_USBAT, init_usbat_cd, 0),
 
 UNUSUAL_DEV(  0x03f0, 0x0307, 0x0001, 0x0001,
 		"HP",
 		"CD-Writer+ CD-4e",
-		US_SC_8070, US_PR_USBAT, init_usbat_cd, 0),
+		USB_SC_8070, USB_PR_USBAT, init_usbat_cd, 0),
 
 UNUSUAL_DEV(  0x04e6, 0x1010, 0x0000, 0x9999,
 		"Shuttle/SCM",
 		"USBAT-02",
-		US_SC_SCSI, US_PR_USBAT, init_usbat_flash,
+		USB_SC_SCSI, USB_PR_USBAT, init_usbat_flash,
 		US_FL_SINGLE_LUN),
 
 UNUSUAL_DEV(  0x0781, 0x0005, 0x0005, 0x0005,
 		"Sandisk",
 		"ImageMate SDDR-05b",
-		US_SC_SCSI, US_PR_USBAT, init_usbat_flash,
+		USB_SC_SCSI, USB_PR_USBAT, init_usbat_flash,
 		US_FL_SINGLE_LUN),
 
 #endif /* defined(CONFIG_USB_STORAGE_USBAT) || ... */
diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c
index 90bb0175a152..4219c197cb08 100644
--- a/drivers/usb/storage/usb.c
+++ b/drivers/usb/storage/usb.c
@@ -512,10 +512,10 @@ static int get_device_info(struct us_data *us, const struct usb_device_id *id,
 
 	/* Store the entries */
 	us->unusual_dev = unusual_dev;
-	us->subclass = (unusual_dev->useProtocol == US_SC_DEVICE) ?
+	us->subclass = (unusual_dev->useProtocol == USB_SC_DEVICE) ?
 			idesc->bInterfaceSubClass :
 			unusual_dev->useProtocol;
-	us->protocol = (unusual_dev->useTransport == US_PR_DEVICE) ?
+	us->protocol = (unusual_dev->useTransport == USB_PR_DEVICE) ?
 			idesc->bInterfaceProtocol :
 			unusual_dev->useTransport;
 	us->fflags = USB_US_ORIG_FLAGS(id->driver_info);
@@ -552,10 +552,10 @@ static int get_device_info(struct us_data *us, const struct usb_device_id *id,
 		struct usb_device_descriptor *ddesc = &dev->descriptor;
 		int msg = -1;
 
-		if (unusual_dev->useProtocol != US_SC_DEVICE &&
+		if (unusual_dev->useProtocol != USB_SC_DEVICE &&
 			us->subclass == idesc->bInterfaceSubClass)
 			msg += 1;
-		if (unusual_dev->useTransport != US_PR_DEVICE &&
+		if (unusual_dev->useTransport != USB_PR_DEVICE &&
 			us->protocol == idesc->bInterfaceProtocol)
 			msg += 2;
 		if (msg >= 0 && !(us->fflags & US_FL_NEED_OVERRIDE))
@@ -582,21 +582,21 @@ static int get_device_info(struct us_data *us, const struct usb_device_id *id,
 static void get_transport(struct us_data *us)
 {
 	switch (us->protocol) {
-	case US_PR_CB:
+	case USB_PR_CB:
 		us->transport_name = "Control/Bulk";
 		us->transport = usb_stor_CB_transport;
 		us->transport_reset = usb_stor_CB_reset;
 		us->max_lun = 7;
 		break;
 
-	case US_PR_CBI:
+	case USB_PR_CBI:
 		us->transport_name = "Control/Bulk/Interrupt";
 		us->transport = usb_stor_CB_transport;
 		us->transport_reset = usb_stor_CB_reset;
 		us->max_lun = 7;
 		break;
 
-	case US_PR_BULK:
+	case USB_PR_BULK:
 		us->transport_name = "Bulk";
 		us->transport = usb_stor_Bulk_transport;
 		us->transport_reset = usb_stor_Bulk_reset;
@@ -608,35 +608,35 @@ static void get_transport(struct us_data *us)
 static void get_protocol(struct us_data *us)
 {
 	switch (us->subclass) {
-	case US_SC_RBC:
+	case USB_SC_RBC:
 		us->protocol_name = "Reduced Block Commands (RBC)";
 		us->proto_handler = usb_stor_transparent_scsi_command;
 		break;
 
-	case US_SC_8020:
+	case USB_SC_8020:
 		us->protocol_name = "8020i";
 		us->proto_handler = usb_stor_pad12_command;
 		us->max_lun = 0;
 		break;
 
-	case US_SC_QIC:
+	case USB_SC_QIC:
 		us->protocol_name = "QIC-157";
 		us->proto_handler = usb_stor_pad12_command;
 		us->max_lun = 0;
 		break;
 
-	case US_SC_8070:
+	case USB_SC_8070:
 		us->protocol_name = "8070i";
 		us->proto_handler = usb_stor_pad12_command;
 		us->max_lun = 0;
 		break;
 
-	case US_SC_SCSI:
+	case USB_SC_SCSI:
 		us->protocol_name = "Transparent SCSI";
 		us->proto_handler = usb_stor_transparent_scsi_command;
 		break;
 
-	case US_SC_UFI:
+	case USB_SC_UFI:
 		us->protocol_name = "Uniform Floppy Interface (UFI)";
 		us->proto_handler = usb_stor_ufi_command;
 		break;
@@ -679,7 +679,7 @@ static int get_pipes(struct us_data *us)
 		}
 	}
 
-	if (!ep_in || !ep_out || (us->protocol == US_PR_CBI && !ep_int)) {
+	if (!ep_in || !ep_out || (us->protocol == USB_PR_CBI && !ep_int)) {
 		US_DEBUGP("Endpoint sanity check failed! Rejecting dev.\n");
 		return -EIO;
 	}
@@ -834,7 +834,7 @@ static int usb_stor_scan_thread(void * __us)
 	if (!test_bit(US_FLIDX_DONT_SCAN, &us->dflags)) {
 
 		/* For bulk-only devices, determine the max LUN value */
-		if (us->protocol == US_PR_BULK &&
+		if (us->protocol == USB_PR_BULK &&
 				!(us->fflags & US_FL_SINGLE_LUN)) {
 			mutex_lock(&us->dev_mutex);
 			us->max_lun = usb_stor_Bulk_max_lun(us);
diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h
index a4b947e470a5..f387c436042e 100644
--- a/include/linux/usb_usual.h
+++ b/include/linux/usb_usual.h
@@ -81,35 +81,36 @@ enum { US_DO_ALL_FLAGS };
 
 /* Sub Classes */
 
-#define US_SC_RBC	0x01		/* Typically, flash devices */
-#define US_SC_8020	0x02		/* CD-ROM */
-#define US_SC_QIC	0x03		/* QIC-157 Tapes */
-#define US_SC_UFI	0x04		/* Floppy */
-#define US_SC_8070	0x05		/* Removable media */
-#define US_SC_SCSI	0x06		/* Transparent */
-#define US_SC_LOCKABLE	0x07		/* Password-protected */
-
-#define US_SC_ISD200    0xf0		/* ISD200 ATA */
-#define US_SC_CYP_ATACB 0xf1		/* Cypress ATACB */
-#define US_SC_DEVICE	0xff		/* Use device's value */
-
-/* Protocols */
-
-#define US_PR_CBI	0x00		/* Control/Bulk/Interrupt */
-#define US_PR_CB	0x01		/* Control/Bulk w/o interrupt */
-#define US_PR_BULK	0x50		/* bulk only */
-
-#define US_PR_USBAT	0x80		/* SCM-ATAPI bridge */
-#define US_PR_EUSB_SDDR09	0x81	/* SCM-SCSI bridge for SDDR-09 */
-#define US_PR_SDDR55	0x82		/* SDDR-55 (made up) */
-#define US_PR_DPCM_USB  0xf0		/* Combination CB/SDDR09 */
-#define US_PR_FREECOM   0xf1		/* Freecom */
-#define US_PR_DATAFAB   0xf2		/* Datafab chipsets */
-#define US_PR_JUMPSHOT  0xf3		/* Lexar Jumpshot */
-#define US_PR_ALAUDA    0xf4		/* Alauda chipsets */
-#define US_PR_KARMA     0xf5		/* Rio Karma */
-
-#define US_PR_DEVICE	0xff		/* Use device's value */
+#define USB_SC_RBC	0x01		/* Typically, flash devices */
+#define USB_SC_8020	0x02		/* CD-ROM */
+#define USB_SC_QIC	0x03		/* QIC-157 Tapes */
+#define USB_SC_UFI	0x04		/* Floppy */
+#define USB_SC_8070	0x05		/* Removable media */
+#define USB_SC_SCSI	0x06		/* Transparent */
+#define USB_SC_LOCKABLE	0x07		/* Password-protected */
+
+#define USB_SC_ISD200	0xf0		/* ISD200 ATA */
+#define USB_SC_CYP_ATACB	0xf1	/* Cypress ATACB */
+#define USB_SC_DEVICE	0xff		/* Use device's value */
+
+/* Storage protocol codes */
+
+#define USB_PR_CBI	0x00		/* Control/Bulk/Interrupt */
+#define USB_PR_CB	0x01		/* Control/Bulk w/o interrupt */
+#define USB_PR_BULK	0x50		/* bulk only */
+#define USB_PR_UAS	0x62		/* USB Attached SCSI */
+
+#define USB_PR_USBAT	0x80		/* SCM-ATAPI bridge */
+#define USB_PR_EUSB_SDDR09	0x81	/* SCM-SCSI bridge for SDDR-09 */
+#define USB_PR_SDDR55	0x82		/* SDDR-55 (made up) */
+#define USB_PR_DPCM_USB	0xf0		/* Combination CB/SDDR09 */
+#define USB_PR_FREECOM	0xf1		/* Freecom */
+#define USB_PR_DATAFAB	0xf2		/* Datafab chipsets */
+#define USB_PR_JUMPSHOT	0xf3		/* Lexar Jumpshot */
+#define USB_PR_ALAUDA	0xf4		/* Alauda chipsets */
+#define USB_PR_KARMA	0xf5		/* Rio Karma */
+
+#define USB_PR_DEVICE	0xff		/* Use device's value */
 
 /*
  */
-- 
cgit v1.2.3


From ae6d22fe1812ce8d40add3eb74ede9cfd2eae44f Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew.r.wilcox@intel.com>
Date: Thu, 7 Oct 2010 13:05:22 +0200
Subject: USB: Move USB Storage definitions to their own header file

The libusual header file is hard to use from code that isn't part
of libusual.  As the comment suggests, these definitions are moved to
their own header file, paralleling other USB classes.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
[mina86@mina86.com: updated to use USB_ prefix and added #include guard]
Signed-off-by: Michal Nazarewicz <mina86@mina86.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

index 0000000..d7fc910
---
 include/linux/usb/storage.h | 48 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb_usual.h   | 38 +----------------------------------
 2 files changed, 49 insertions(+), 37 deletions(-)
 create mode 100644 include/linux/usb/storage.h

(limited to 'include/linux')

diff --git a/include/linux/usb/storage.h b/include/linux/usb/storage.h
new file mode 100644
index 000000000000..d7fc910f1dc4
--- /dev/null
+++ b/include/linux/usb/storage.h
@@ -0,0 +1,48 @@
+#ifndef __LINUX_USB_STORAGE_H
+#define __LINUX_USB_STORAGE_H
+
+/*
+ * linux/usb/storage.h
+ *
+ * Copyright Matthew Wilcox for Intel Corp, 2010
+ *
+ * This file contains definitions taken from the
+ * USB Mass Storage Class Specification Overview
+ *
+ * Distributed under the terms of the GNU GPL, version two.
+ */
+
+/* Storage subclass codes */
+
+#define USB_SC_RBC	0x01		/* Typically, flash devices */
+#define USB_SC_8020	0x02		/* CD-ROM */
+#define USB_SC_QIC	0x03		/* QIC-157 Tapes */
+#define USB_SC_UFI	0x04		/* Floppy */
+#define USB_SC_8070	0x05		/* Removable media */
+#define USB_SC_SCSI	0x06		/* Transparent */
+#define USB_SC_LOCKABLE	0x07		/* Password-protected */
+
+#define USB_SC_ISD200	0xf0		/* ISD200 ATA */
+#define USB_SC_CYP_ATACB	0xf1	/* Cypress ATACB */
+#define USB_SC_DEVICE	0xff		/* Use device's value */
+
+/* Storage protocol codes */
+
+#define USB_PR_CBI	0x00		/* Control/Bulk/Interrupt */
+#define USB_PR_CB	0x01		/* Control/Bulk w/o interrupt */
+#define USB_PR_BULK	0x50		/* bulk only */
+#define USB_PR_UAS	0x62		/* USB Attached SCSI */
+
+#define USB_PR_USBAT	0x80		/* SCM-ATAPI bridge */
+#define USB_PR_EUSB_SDDR09	0x81	/* SCM-SCSI bridge for SDDR-09 */
+#define USB_PR_SDDR55	0x82		/* SDDR-55 (made up) */
+#define USB_PR_DPCM_USB	0xf0		/* Combination CB/SDDR09 */
+#define USB_PR_FREECOM	0xf1		/* Freecom */
+#define USB_PR_DATAFAB	0xf2		/* Datafab chipsets */
+#define USB_PR_JUMPSHOT	0xf3		/* Lexar Jumpshot */
+#define USB_PR_ALAUDA	0xf4		/* Alauda chipsets */
+#define USB_PR_KARMA	0xf5		/* Rio Karma */
+
+#define USB_PR_DEVICE	0xff		/* Use device's value */
+
+#endif
diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h
index f387c436042e..f091dc6e5a00 100644
--- a/include/linux/usb_usual.h
+++ b/include/linux/usb_usual.h
@@ -74,43 +74,7 @@ enum { US_DO_ALL_FLAGS };
 #define USB_US_TYPE(flags) 		(((flags) >> 24) & 0xFF)
 #define USB_US_ORIG_FLAGS(flags)	((flags) & 0x00FFFFFF)
 
-/*
- * This is probably not the best place to keep these constants, conceptually.
- * But it's the only header included into all places which need them.
- */
-
-/* Sub Classes */
-
-#define USB_SC_RBC	0x01		/* Typically, flash devices */
-#define USB_SC_8020	0x02		/* CD-ROM */
-#define USB_SC_QIC	0x03		/* QIC-157 Tapes */
-#define USB_SC_UFI	0x04		/* Floppy */
-#define USB_SC_8070	0x05		/* Removable media */
-#define USB_SC_SCSI	0x06		/* Transparent */
-#define USB_SC_LOCKABLE	0x07		/* Password-protected */
-
-#define USB_SC_ISD200	0xf0		/* ISD200 ATA */
-#define USB_SC_CYP_ATACB	0xf1	/* Cypress ATACB */
-#define USB_SC_DEVICE	0xff		/* Use device's value */
-
-/* Storage protocol codes */
-
-#define USB_PR_CBI	0x00		/* Control/Bulk/Interrupt */
-#define USB_PR_CB	0x01		/* Control/Bulk w/o interrupt */
-#define USB_PR_BULK	0x50		/* bulk only */
-#define USB_PR_UAS	0x62		/* USB Attached SCSI */
-
-#define USB_PR_USBAT	0x80		/* SCM-ATAPI bridge */
-#define USB_PR_EUSB_SDDR09	0x81	/* SCM-SCSI bridge for SDDR-09 */
-#define USB_PR_SDDR55	0x82		/* SDDR-55 (made up) */
-#define USB_PR_DPCM_USB	0xf0		/* Combination CB/SDDR09 */
-#define USB_PR_FREECOM	0xf1		/* Freecom */
-#define USB_PR_DATAFAB	0xf2		/* Datafab chipsets */
-#define USB_PR_JUMPSHOT	0xf3		/* Lexar Jumpshot */
-#define USB_PR_ALAUDA	0xf4		/* Alauda chipsets */
-#define USB_PR_KARMA	0xf5		/* Rio Karma */
-
-#define USB_PR_DEVICE	0xff		/* Use device's value */
+#include <linux/usb/storage.h>
 
 /*
  */
-- 
cgit v1.2.3


From 496dda704bca1208e08773ba39b29a69536f5381 Mon Sep 17 00:00:00 2001
From: Maulik Mankad <x0082077@ti.com>
Date: Fri, 24 Sep 2010 13:44:06 +0300
Subject: usb: musb: host: unmap the buffer for PIO data transfers

The USB stack maps the buffer for DMA if the controller supports DMA.
MUSB controller can perform DMA as well as PIO transfers.
The buffer needs to be unmapped before CPU can perform
PIO data transfers.

Export unmap_urb_for_dma() so that drivers can perform
the DMA unmapping in a sane way.

Signed-off-by: Maulik Mankad <x0082077@ti.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c       | 3 ++-
 drivers/usb/musb/musb_host.c | 5 +++++
 include/linux/usb/hcd.h      | 1 +
 3 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 5cca00a6d09d..cb2d894321da 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1263,7 +1263,7 @@ static void hcd_free_coherent(struct usb_bus *bus, dma_addr_t *dma_handle,
 	*dma_handle = 0;
 }
 
-static void unmap_urb_for_dma(struct usb_hcd *hcd, struct urb *urb)
+void unmap_urb_for_dma(struct usb_hcd *hcd, struct urb *urb)
 {
 	enum dma_data_direction dir;
 
@@ -1307,6 +1307,7 @@ static void unmap_urb_for_dma(struct usb_hcd *hcd, struct urb *urb)
 			URB_DMA_MAP_SG | URB_DMA_MAP_PAGE |
 			URB_DMA_MAP_SINGLE | URB_MAP_LOCAL);
 }
+EXPORT_SYMBOL_GPL(unmap_urb_for_dma);
 
 static int map_urb_for_dma(struct usb_hcd *hcd, struct urb *urb,
 			   gfp_t mem_flags)
diff --git a/drivers/usb/musb/musb_host.c b/drivers/usb/musb/musb_host.c
index 9e65c47cc98b..62e39fc57211 100644
--- a/drivers/usb/musb/musb_host.c
+++ b/drivers/usb/musb/musb_host.c
@@ -41,6 +41,7 @@
 #include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/list.h>
+#include <linux/dma-mapping.h>
 
 #include "musb_core.h"
 #include "musb_host.h"
@@ -1332,6 +1333,8 @@ void musb_host_tx(struct musb *musb, u8 epnum)
 	 */
 	if (length > qh->maxpacket)
 		length = qh->maxpacket;
+	/* Unmap the buffer so that CPU can use it */
+	unmap_urb_for_dma(musb_to_hcd(musb), urb);
 	musb_write_fifo(hw_ep, length, urb->transfer_buffer + offset);
 	qh->segsize = length;
 
@@ -1752,6 +1755,8 @@ void musb_host_rx(struct musb *musb, u8 epnum)
 #endif	/* Mentor DMA */
 
 		if (!dma) {
+			/* Unmap the buffer so that CPU can use it */
+			unmap_urb_for_dma(musb_to_hcd(musb), urb);
 			done = musb_host_packet_rx(musb, urb,
 					epnum, iso_err);
 			DBG(6, "read %spacket\n", done ? "last " : "");
diff --git a/include/linux/usb/hcd.h b/include/linux/usb/hcd.h
index 3b571f1ffbb3..fe89f7c298aa 100644
--- a/include/linux/usb/hcd.h
+++ b/include/linux/usb/hcd.h
@@ -329,6 +329,7 @@ extern int usb_hcd_submit_urb(struct urb *urb, gfp_t mem_flags);
 extern int usb_hcd_unlink_urb(struct urb *urb, int status);
 extern void usb_hcd_giveback_urb(struct usb_hcd *hcd, struct urb *urb,
 		int status);
+extern void unmap_urb_for_dma(struct usb_hcd *, struct urb *);
 extern void usb_hcd_flush_endpoint(struct usb_device *udev,
 		struct usb_host_endpoint *ep);
 extern void usb_hcd_disable_endpoint(struct usb_device *udev,
-- 
cgit v1.2.3


From 748eee0986f0d51c7bc39f194d515a8d8248ebdd Mon Sep 17 00:00:00 2001
From: Grazvydas Ignotas <notasas@gmail.com>
Date: Mon, 27 Sep 2010 15:17:18 +0300
Subject: USB: Add more empty functions in otg.h

Add empty functions for get/put transceiver functions too, so that
drivers that optionally use them can call them without worrying that
they might not exist, eliminating ifdefs.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Acked-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/otg.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/usb/otg.h b/include/linux/usb/otg.h
index 545cba73ccaf..0a5b3711e502 100644
--- a/include/linux/usb/otg.h
+++ b/include/linux/usb/otg.h
@@ -164,8 +164,19 @@ otg_shutdown(struct otg_transceiver *otg)
 }
 
 /* for usb host and peripheral controller drivers */
+#ifdef CONFIG_USB_OTG_UTILS
 extern struct otg_transceiver *otg_get_transceiver(void);
 extern void otg_put_transceiver(struct otg_transceiver *);
+#else
+static inline struct otg_transceiver *otg_get_transceiver(void)
+{
+	return NULL;
+}
+
+static inline void otg_put_transceiver(struct otg_transceiver *x)
+{
+}
+#endif
 
 /* Context: can sleep */
 static inline int
-- 
cgit v1.2.3


From 230f7ede6c2f0e403f29e03e0251a470aa9350dd Mon Sep 17 00:00:00 2001
From: Anatolij Gustschin <agust@denx.de>
Date: Tue, 28 Sep 2010 20:55:21 +0200
Subject: USB: add USB EHCI support for MPC5121 SoC

Extends FSL EHCI platform driver glue layer to support
MPC5121 USB controllers. MPC5121 Rev 2.0 silicon EHCI
registers are in big endian format. The appropriate flags
are set using the information in the platform data structure.
MPC83xx system interface registers are not available on
MPC512x, so the access to these registers is isolated in
MPC512x case. Furthermore the USB controller clocks
must be enabled before 512x register accesses which is
done by providing platform specific init callback.

The MPC512x internal USB PHY doesn't provide supply voltage.
For boards using different power switches allow specifying
DRVVBUS and PWR_FAULT signal polarity of the MPC5121 internal
PHY using "fsl,invert-drvvbus" and "fsl,invert-pwr-fault"
properties in the device tree USB nodes. Adds documentation
for this new device tree bindings.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/powerpc/dts-bindings/fsl/usb.txt | 22 ++++++
 drivers/usb/Kconfig                            |  1 +
 drivers/usb/host/Kconfig                       |  6 +-
 drivers/usb/host/ehci-fsl.c                    | 99 ++++++++++++++++++--------
 drivers/usb/host/ehci-fsl.h                    | 13 +++-
 drivers/usb/host/ehci-mem.c                    |  2 +-
 drivers/usb/host/fsl-mph-dr-of.c               | 89 +++++++++++++++++++++++
 include/linux/fsl_devices.h                    | 15 ++++
 8 files changed, 215 insertions(+), 32 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/powerpc/dts-bindings/fsl/usb.txt b/Documentation/powerpc/dts-bindings/fsl/usb.txt
index b00152402694..bd5723f0b67e 100644
--- a/Documentation/powerpc/dts-bindings/fsl/usb.txt
+++ b/Documentation/powerpc/dts-bindings/fsl/usb.txt
@@ -8,6 +8,7 @@ and additions :
 Required properties :
  - compatible : Should be "fsl-usb2-mph" for multi port host USB
    controllers, or "fsl-usb2-dr" for dual role USB controllers
+   or "fsl,mpc5121-usb2-dr" for dual role USB controllers of MPC5121
  - phy_type : For multi port host USB controllers, should be one of
    "ulpi", or "serial". For dual role USB controllers, should be
    one of "ulpi", "utmi", "utmi_wide", or "serial".
@@ -33,6 +34,12 @@ Recommended properties :
  - interrupt-parent : the phandle for the interrupt controller that
    services interrupts for this device.
 
+Optional properties :
+ - fsl,invert-drvvbus : boolean; for MPC5121 USB0 only. Indicates the
+   port power polarity of internal PHY signal DRVVBUS is inverted.
+ - fsl,invert-pwr-fault : boolean; for MPC5121 USB0 only. Indicates
+   the PWR_FAULT signal polarity is inverted.
+
 Example multi port host USB controller device node :
 	usb@22000 {
 		compatible = "fsl-usb2-mph";
@@ -57,3 +64,18 @@ Example dual role USB controller device node :
 		dr_mode = "otg";
 		phy = "ulpi";
 	};
+
+Example dual role USB controller device node for MPC5121ADS:
+
+	usb@4000 {
+		compatible = "fsl,mpc5121-usb2-dr";
+		reg = <0x4000 0x1000>;
+		#address-cells = <1>;
+		#size-cells = <0>;
+		interrupt-parent = < &ipic >;
+		interrupts = <44 0x8>;
+		dr_mode = "otg";
+		phy_type = "utmi_wide";
+		fsl,invert-drvvbus;
+		fsl,invert-pwr-fault;
+	};
diff --git a/drivers/usb/Kconfig b/drivers/usb/Kconfig
index 4aa00e6e57ad..67eb3770868f 100644
--- a/drivers/usb/Kconfig
+++ b/drivers/usb/Kconfig
@@ -59,6 +59,7 @@ config USB_ARCH_HAS_OHCI
 config USB_ARCH_HAS_EHCI
 	boolean
 	default y if PPC_83xx
+	default y if PPC_MPC512x
 	default y if SOC_AU1200
 	default y if ARCH_IXP4XX
 	default y if ARCH_W90X900
diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
index f3a90b0fc422..bf2e7d234533 100644
--- a/drivers/usb/host/Kconfig
+++ b/drivers/usb/host/Kconfig
@@ -93,12 +93,14 @@ config USB_EHCI_TT_NEWSCHED
 
 config USB_EHCI_BIG_ENDIAN_MMIO
 	bool
-	depends on USB_EHCI_HCD && (PPC_CELLEB || PPC_PS3 || 440EPX || ARCH_IXP4XX || XPS_USB_HCD_XILINX)
+	depends on USB_EHCI_HCD && (PPC_CELLEB || PPC_PS3 || 440EPX || ARCH_IXP4XX || \
+				    XPS_USB_HCD_XILINX || PPC_MPC512x)
 	default y
 
 config USB_EHCI_BIG_ENDIAN_DESC
 	bool
-	depends on USB_EHCI_HCD && (440EPX || ARCH_IXP4XX || XPS_USB_HCD_XILINX)
+	depends on USB_EHCI_HCD && (440EPX || ARCH_IXP4XX || XPS_USB_HCD_XILINX || \
+				    PPC_MPC512x)
 	default y
 
 config XPS_USB_HCD_XILINX
diff --git a/drivers/usb/host/ehci-fsl.c b/drivers/usb/host/ehci-fsl.c
index 8600317bd60b..86e42892016d 100644
--- a/drivers/usb/host/ehci-fsl.c
+++ b/drivers/usb/host/ehci-fsl.c
@@ -116,13 +116,33 @@ static int usb_hcd_fsl_probe(const struct hc_driver *driver,
 		goto err3;
 	}
 
-	/* Enable USB controller */
-	temp = in_be32(hcd->regs + 0x500);
-	out_be32(hcd->regs + 0x500, temp | 0x4);
+	pdata->regs = hcd->regs;
 
-	/* Set to Host mode */
-	temp = in_le32(hcd->regs + 0x1a8);
-	out_le32(hcd->regs + 0x1a8, temp | 0x3);
+	/*
+	 * do platform specific init: check the clock, grab/config pins, etc.
+	 */
+	if (pdata->init && pdata->init(pdev)) {
+		retval = -ENODEV;
+		goto err3;
+	}
+
+	/*
+	 * Check if it is MPC5121 SoC, otherwise set pdata->have_sysif_regs
+	 * flag for 83xx or 8536 system interface registers.
+	 */
+	if (pdata->big_endian_mmio)
+		temp = in_be32(hcd->regs + FSL_SOC_USB_ID);
+	else
+		temp = in_le32(hcd->regs + FSL_SOC_USB_ID);
+
+	if ((temp & ID_MSK) != (~((temp & NID_MSK) >> 8) & ID_MSK))
+		pdata->have_sysif_regs = 1;
+
+	/* Enable USB controller, 83xx or 8536 */
+	if (pdata->have_sysif_regs)
+		setbits32(hcd->regs + FSL_SOC_USB_CTRL, 0x4);
+
+	/* Don't need to set host mode here. It will be done by tdi_reset() */
 
 	retval = usb_add_hcd(hcd, irq, IRQF_DISABLED | IRQF_SHARED);
 	if (retval != 0)
@@ -137,6 +157,8 @@ static int usb_hcd_fsl_probe(const struct hc_driver *driver,
 	usb_put_hcd(hcd);
       err1:
 	dev_err(&pdev->dev, "init %s fail, %d\n", dev_name(&pdev->dev), retval);
+	if (pdata->exit)
+		pdata->exit(pdev);
 	return retval;
 }
 
@@ -154,17 +176,30 @@ static int usb_hcd_fsl_probe(const struct hc_driver *driver,
 static void usb_hcd_fsl_remove(struct usb_hcd *hcd,
 			       struct platform_device *pdev)
 {
+	struct fsl_usb2_platform_data *pdata = pdev->dev.platform_data;
+
 	usb_remove_hcd(hcd);
+
+	/*
+	 * do platform specific un-initialization:
+	 * release iomux pins, disable clock, etc.
+	 */
+	if (pdata->exit)
+		pdata->exit(pdev);
 	iounmap(hcd->regs);
 	release_mem_region(hcd->rsrc_start, hcd->rsrc_len);
 	usb_put_hcd(hcd);
 }
 
-static void mpc83xx_setup_phy(struct ehci_hcd *ehci,
-			      enum fsl_usb2_phy_modes phy_mode,
-			      unsigned int port_offset)
+static void ehci_fsl_setup_phy(struct ehci_hcd *ehci,
+			       enum fsl_usb2_phy_modes phy_mode,
+			       unsigned int port_offset)
 {
-	u32 portsc = 0;
+	u32 portsc;
+
+	portsc = ehci_readl(ehci, &ehci->regs->port_status[port_offset]);
+	portsc &= ~(PORT_PTS_MSK | PORT_PTS_PTW);
+
 	switch (phy_mode) {
 	case FSL_USB2_PHY_ULPI:
 		portsc |= PORT_PTS_ULPI;
@@ -184,20 +219,21 @@ static void mpc83xx_setup_phy(struct ehci_hcd *ehci,
 	ehci_writel(ehci, portsc, &ehci->regs->port_status[port_offset]);
 }
 
-static void mpc83xx_usb_setup(struct usb_hcd *hcd)
+static void ehci_fsl_usb_setup(struct ehci_hcd *ehci)
 {
-	struct ehci_hcd *ehci = hcd_to_ehci(hcd);
+	struct usb_hcd *hcd = ehci_to_hcd(ehci);
 	struct fsl_usb2_platform_data *pdata;
 	void __iomem *non_ehci = hcd->regs;
 	u32 temp;
 
-	pdata =
-	    (struct fsl_usb2_platform_data *)hcd->self.controller->
-	    platform_data;
+	pdata = hcd->self.controller->platform_data;
+
 	/* Enable PHY interface in the control reg. */
-	temp = in_be32(non_ehci + FSL_SOC_USB_CTRL);
-	out_be32(non_ehci + FSL_SOC_USB_CTRL, temp | 0x00000004);
-	out_be32(non_ehci + FSL_SOC_USB_SNOOP1, 0x0000001b);
+	if (pdata->have_sysif_regs) {
+		temp = in_be32(non_ehci + FSL_SOC_USB_CTRL);
+		out_be32(non_ehci + FSL_SOC_USB_CTRL, temp | 0x00000004);
+		out_be32(non_ehci + FSL_SOC_USB_SNOOP1, 0x0000001b);
+	}
 
 #if defined(CONFIG_PPC32) && !defined(CONFIG_NOT_COHERENT_CACHE)
 	/*
@@ -214,7 +250,7 @@ static void mpc83xx_usb_setup(struct usb_hcd *hcd)
 
 	if ((pdata->operating_mode == FSL_USB2_DR_HOST) ||
 			(pdata->operating_mode == FSL_USB2_DR_OTG))
-		mpc83xx_setup_phy(ehci, pdata->phy_mode, 0);
+		ehci_fsl_setup_phy(ehci, pdata->phy_mode, 0);
 
 	if (pdata->operating_mode == FSL_USB2_MPH_HOST) {
 		unsigned int chip, rev, svr;
@@ -228,25 +264,27 @@ static void mpc83xx_usb_setup(struct usb_hcd *hcd)
 			ehci->has_fsl_port_bug = 1;
 
 		if (pdata->port_enables & FSL_USB2_PORT0_ENABLED)
-			mpc83xx_setup_phy(ehci, pdata->phy_mode, 0);
+			ehci_fsl_setup_phy(ehci, pdata->phy_mode, 0);
 		if (pdata->port_enables & FSL_USB2_PORT1_ENABLED)
-			mpc83xx_setup_phy(ehci, pdata->phy_mode, 1);
+			ehci_fsl_setup_phy(ehci, pdata->phy_mode, 1);
 	}
 
+	if (pdata->have_sysif_regs) {
 #ifdef CONFIG_PPC_85xx
-	out_be32(non_ehci + FSL_SOC_USB_PRICTRL, 0x00000008);
-	out_be32(non_ehci + FSL_SOC_USB_AGECNTTHRSH, 0x00000080);
+		out_be32(non_ehci + FSL_SOC_USB_PRICTRL, 0x00000008);
+		out_be32(non_ehci + FSL_SOC_USB_AGECNTTHRSH, 0x00000080);
 #else
-	out_be32(non_ehci + FSL_SOC_USB_PRICTRL, 0x0000000c);
-	out_be32(non_ehci + FSL_SOC_USB_AGECNTTHRSH, 0x00000040);
+		out_be32(non_ehci + FSL_SOC_USB_PRICTRL, 0x0000000c);
+		out_be32(non_ehci + FSL_SOC_USB_AGECNTTHRSH, 0x00000040);
 #endif
-	out_be32(non_ehci + FSL_SOC_USB_SICTRL, 0x00000001);
+		out_be32(non_ehci + FSL_SOC_USB_SICTRL, 0x00000001);
+	}
 }
 
 /* called after powerup, by probe or system-pm "wakeup" */
 static int ehci_fsl_reinit(struct ehci_hcd *ehci)
 {
-	mpc83xx_usb_setup(ehci_to_hcd(ehci));
+	ehci_fsl_usb_setup(ehci);
 	ehci_port_power(ehci, 0);
 
 	return 0;
@@ -257,6 +295,11 @@ static int ehci_fsl_setup(struct usb_hcd *hcd)
 {
 	struct ehci_hcd *ehci = hcd_to_ehci(hcd);
 	int retval;
+	struct fsl_usb2_platform_data *pdata;
+
+	pdata = hcd->self.controller->platform_data;
+	ehci->big_endian_desc = pdata->big_endian_desc;
+	ehci->big_endian_mmio = pdata->big_endian_mmio;
 
 	/* EHCI registers start at offset 0x100 */
 	ehci->caps = hcd->regs + 0x100;
@@ -370,7 +413,7 @@ static const struct hc_driver ehci_fsl_hc_driver = {
 	 * generic hardware linkage
 	 */
 	.irq = ehci_irq,
-	.flags = HCD_USB2,
+	.flags = HCD_USB2 | HCD_MEMORY,
 
 	/*
 	 * basic lifecycle operations
diff --git a/drivers/usb/host/ehci-fsl.h b/drivers/usb/host/ehci-fsl.h
index eb537aa54610..2c8353795226 100644
--- a/drivers/usb/host/ehci-fsl.h
+++ b/drivers/usb/host/ehci-fsl.h
@@ -1,4 +1,4 @@
-/* Copyright (c) 2005 freescale semiconductor
+/* Copyright (C) 2005-2010 Freescale Semiconductor, Inc.
  * Copyright (c) 2005 MontaVista Software
  *
  * This program is free software; you can redistribute  it and/or modify it
@@ -19,6 +19,9 @@
 #define _EHCI_FSL_H
 
 /* offsets for the non-ehci registers in the FSL SOC USB controller */
+#define FSL_SOC_USB_ID		0x0
+#define ID_MSK			0x3f
+#define NID_MSK			0x3f00
 #define FSL_SOC_USB_ULPIVP	0x170
 #define FSL_SOC_USB_PORTSC1	0x184
 #define PORT_PTS_MSK		(3<<30)
@@ -27,6 +30,14 @@
 #define	PORT_PTS_SERIAL		(3<<30)
 #define PORT_PTS_PTW		(1<<28)
 #define FSL_SOC_USB_PORTSC2	0x188
+
+#define FSL_SOC_USB_USBGENCTRL	0x200
+#define USBGENCTRL_PPP		(1 << 3)
+#define USBGENCTRL_PFP		(1 << 2)
+#define FSL_SOC_USB_ISIPHYCTRL	0x204
+#define ISIPHYCTRL_PXE		(1)
+#define ISIPHYCTRL_PHYE		(1 << 4)
+
 #define FSL_SOC_USB_SNOOP1	0x400	/* NOTE: big-endian */
 #define FSL_SOC_USB_SNOOP2	0x404	/* NOTE: big-endian */
 #define FSL_SOC_USB_AGECNTTHRSH	0x408	/* NOTE: big-endian */
diff --git a/drivers/usb/host/ehci-mem.c b/drivers/usb/host/ehci-mem.c
index 1f3f01eacaf0..d36e4e75e08d 100644
--- a/drivers/usb/host/ehci-mem.c
+++ b/drivers/usb/host/ehci-mem.c
@@ -40,7 +40,7 @@ static inline void ehci_qtd_init(struct ehci_hcd *ehci, struct ehci_qtd *qtd,
 {
 	memset (qtd, 0, sizeof *qtd);
 	qtd->qtd_dma = dma;
-	qtd->hw_token = cpu_to_le32 (QTD_STS_HALT);
+	qtd->hw_token = cpu_to_hc32(ehci, QTD_STS_HALT);
 	qtd->hw_next = EHCI_LIST_END(ehci);
 	qtd->hw_alt_next = EHCI_LIST_END(ehci);
 	INIT_LIST_HEAD (&qtd->qtd_list);
diff --git a/drivers/usb/host/fsl-mph-dr-of.c b/drivers/usb/host/fsl-mph-dr-of.c
index 12db5d5cb0bc..574b99ea0700 100644
--- a/drivers/usb/host/fsl-mph-dr-of.c
+++ b/drivers/usb/host/fsl-mph-dr-of.c
@@ -15,6 +15,7 @@
 #include <linux/err.h>
 #include <linux/io.h>
 #include <linux/of_platform.h>
+#include <linux/clk.h>
 
 struct fsl_usb2_dev_data {
 	char *dr_mode;		/* controller mode */
@@ -153,6 +154,12 @@ static int __devinit fsl_usb2_mph_dr_of_probe(struct platform_device *ofdev)
 
 		pdata->operating_mode = FSL_USB2_MPH_HOST;
 	} else {
+		if (of_get_property(np, "fsl,invert-drvvbus", NULL))
+			pdata->invert_drvvbus = 1;
+
+		if (of_get_property(np, "fsl,invert-pwr-fault", NULL))
+			pdata->invert_pwr_fault = 1;
+
 		/* setup mode selected in the device tree */
 		pdata->operating_mode = dev_data->op_mode;
 	}
@@ -186,9 +193,91 @@ static int __devexit fsl_usb2_mph_dr_of_remove(struct platform_device *ofdev)
 	return 0;
 }
 
+#ifdef CONFIG_PPC_MPC512x
+
+#define USBGENCTRL		0x200		/* NOTE: big endian */
+#define GC_WU_INT_CLR		(1 << 5)	/* Wakeup int clear */
+#define GC_ULPI_SEL		(1 << 4)	/* ULPI i/f select (usb0 only)*/
+#define GC_PPP			(1 << 3)	/* Inv. Port Power Polarity */
+#define GC_PFP			(1 << 2)	/* Inv. Power Fault Polarity */
+#define GC_WU_ULPI_EN		(1 << 1)	/* Wakeup on ULPI event */
+#define GC_WU_IE		(1 << 1)	/* Wakeup interrupt enable */
+
+#define ISIPHYCTRL		0x204		/* NOTE: big endian */
+#define PHYCTRL_PHYE		(1 << 4)	/* On-chip UTMI PHY enable */
+#define PHYCTRL_BSENH		(1 << 3)	/* Bit Stuff Enable High */
+#define PHYCTRL_BSEN		(1 << 2)	/* Bit Stuff Enable */
+#define PHYCTRL_LSFE		(1 << 1)	/* Line State Filter Enable */
+#define PHYCTRL_PXE		(1 << 0)	/* PHY oscillator enable */
+
+int fsl_usb2_mpc5121_init(struct platform_device *pdev)
+{
+	struct fsl_usb2_platform_data *pdata = pdev->dev.platform_data;
+	struct clk *clk;
+	char clk_name[10];
+	int base, clk_num;
+
+	base = pdev->resource->start & 0xf000;
+	if (base == 0x3000)
+		clk_num = 1;
+	else if (base == 0x4000)
+		clk_num = 2;
+	else
+		return -ENODEV;
+
+	snprintf(clk_name, sizeof(clk_name), "usb%d_clk", clk_num);
+	clk = clk_get(&pdev->dev, clk_name);
+	if (IS_ERR(clk)) {
+		dev_err(&pdev->dev, "failed to get clk\n");
+		return PTR_ERR(clk);
+	}
+
+	clk_enable(clk);
+	pdata->clk = clk;
+
+	if (pdata->phy_mode == FSL_USB2_PHY_UTMI_WIDE) {
+		u32 reg = 0;
+
+		if (pdata->invert_drvvbus)
+			reg |= GC_PPP;
+
+		if (pdata->invert_pwr_fault)
+			reg |= GC_PFP;
+
+		out_be32(pdata->regs + ISIPHYCTRL, PHYCTRL_PHYE | PHYCTRL_PXE);
+		out_be32(pdata->regs + USBGENCTRL, reg);
+	}
+	return 0;
+}
+
+static void fsl_usb2_mpc5121_exit(struct platform_device *pdev)
+{
+	struct fsl_usb2_platform_data *pdata = pdev->dev.platform_data;
+
+	pdata->regs = NULL;
+
+	if (pdata->clk) {
+		clk_disable(pdata->clk);
+		clk_put(pdata->clk);
+	}
+}
+
+struct fsl_usb2_platform_data fsl_usb2_mpc5121_pd = {
+	.big_endian_desc = 1,
+	.big_endian_mmio = 1,
+	.es = 1,
+	.le_setup_buf = 1,
+	.init = fsl_usb2_mpc5121_init,
+	.exit = fsl_usb2_mpc5121_exit,
+};
+#endif /* CONFIG_PPC_MPC512x */
+
 static const struct of_device_id fsl_usb2_mph_dr_of_match[] = {
 	{ .compatible = "fsl-usb2-mph", },
 	{ .compatible = "fsl-usb2-dr", },
+#ifdef CONFIG_PPC_MPC512x
+	{ .compatible = "fsl,mpc5121-usb2-dr", .data = &fsl_usb2_mpc5121_pd, },
+#endif
 	{},
 };
 
diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h
index 28e33fea5107..d5f9a7431bd0 100644
--- a/include/linux/fsl_devices.h
+++ b/include/linux/fsl_devices.h
@@ -58,11 +58,26 @@ enum fsl_usb2_phy_modes {
 	FSL_USB2_PHY_SERIAL,
 };
 
+struct clk;
+struct platform_device;
+
 struct fsl_usb2_platform_data {
 	/* board specific information */
 	enum fsl_usb2_operating_modes	operating_mode;
 	enum fsl_usb2_phy_modes		phy_mode;
 	unsigned int			port_enables;
+
+	int		(*init)(struct platform_device *);
+	void		(*exit)(struct platform_device *);
+	void __iomem	*regs;		/* ioremap'd register base */
+	struct clk	*clk;
+	unsigned	big_endian_mmio:1;
+	unsigned	big_endian_desc:1;
+	unsigned	es:1;		/* need USBMODE:ES */
+	unsigned	le_setup_buf:1;
+	unsigned	have_sysif_regs:1;
+	unsigned	invert_drvvbus:1;
+	unsigned	invert_pwr_fault:1;
 };
 
 /* Flags in fsl_usb2_mph_platform_data */
-- 
cgit v1.2.3


From 1dae423dd9b247b048eda00cb598c755e5933213 Mon Sep 17 00:00:00 2001
From: Martin Fuzzey <mfuzzey@gmail.com>
Date: Fri, 1 Oct 2010 00:21:55 +0200
Subject: USB: introduce unmap_urb_setup_for_dma()

Split unmap_urb_for_dma() to allow just the setup buffer
to be unmapped. This allows HCDs to use PIO for the setup
buffer if it is not suitable for DMA.

Signed-off-by: Martin Fuzzey <mfuzzey@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c  | 18 +++++++++++++-----
 include/linux/usb/hcd.h |  1 +
 2 files changed, 14 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index cb2d894321da..61800f77dac8 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1263,10 +1263,8 @@ static void hcd_free_coherent(struct usb_bus *bus, dma_addr_t *dma_handle,
 	*dma_handle = 0;
 }
 
-void unmap_urb_for_dma(struct usb_hcd *hcd, struct urb *urb)
+void unmap_urb_setup_for_dma(struct usb_hcd *hcd, struct urb *urb)
 {
-	enum dma_data_direction dir;
-
 	if (urb->transfer_flags & URB_SETUP_MAP_SINGLE)
 		dma_unmap_single(hcd->self.controller,
 				urb->setup_dma,
@@ -1279,6 +1277,17 @@ void unmap_urb_for_dma(struct usb_hcd *hcd, struct urb *urb)
 				sizeof(struct usb_ctrlrequest),
 				DMA_TO_DEVICE);
 
+	/* Make it safe to call this routine more than once */
+	urb->transfer_flags &= ~(URB_SETUP_MAP_SINGLE | URB_SETUP_MAP_LOCAL);
+}
+EXPORT_SYMBOL_GPL(unmap_urb_setup_for_dma);
+
+void unmap_urb_for_dma(struct usb_hcd *hcd, struct urb *urb)
+{
+	enum dma_data_direction dir;
+
+	unmap_urb_setup_for_dma(hcd, urb);
+
 	dir = usb_urb_dir_in(urb) ? DMA_FROM_DEVICE : DMA_TO_DEVICE;
 	if (urb->transfer_flags & URB_DMA_MAP_SG)
 		dma_unmap_sg(hcd->self.controller,
@@ -1303,8 +1312,7 @@ void unmap_urb_for_dma(struct usb_hcd *hcd, struct urb *urb)
 				dir);
 
 	/* Make it safe to call this routine more than once */
-	urb->transfer_flags &= ~(URB_SETUP_MAP_SINGLE | URB_SETUP_MAP_LOCAL |
-			URB_DMA_MAP_SG | URB_DMA_MAP_PAGE |
+	urb->transfer_flags &= ~(URB_DMA_MAP_SG | URB_DMA_MAP_PAGE |
 			URB_DMA_MAP_SINGLE | URB_MAP_LOCAL);
 }
 EXPORT_SYMBOL_GPL(unmap_urb_for_dma);
diff --git a/include/linux/usb/hcd.h b/include/linux/usb/hcd.h
index fe89f7c298aa..0b6e751ea0b1 100644
--- a/include/linux/usb/hcd.h
+++ b/include/linux/usb/hcd.h
@@ -329,6 +329,7 @@ extern int usb_hcd_submit_urb(struct urb *urb, gfp_t mem_flags);
 extern int usb_hcd_unlink_urb(struct urb *urb, int status);
 extern void usb_hcd_giveback_urb(struct usb_hcd *hcd, struct urb *urb,
 		int status);
+extern void unmap_urb_setup_for_dma(struct usb_hcd *, struct urb *);
 extern void unmap_urb_for_dma(struct usb_hcd *, struct urb *);
 extern void usb_hcd_flush_endpoint(struct usb_device *udev,
 		struct usb_host_endpoint *ep);
-- 
cgit v1.2.3


From ae38c78a03e1b77ad45248fcf097e4568e740209 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Fri, 1 Oct 2010 14:20:10 -0700
Subject: usb-storage: add new no_read_disc_info quirk

Appotech ax3003 (the larger brother of the ax203) based devices are even
more buggy then the ax203.  They will go of into lala land when ever they
see a READ_DISC_INFO scsi command.  So add a new US_FL which tells the
scsi sr driver to not issue any READ_DISC_INFO scsi commands.

[akpm@linux-foundation.org: fix build]
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/scsiglue.c     | 4 ++++
 drivers/usb/storage/unusual_devs.h | 5 +++++
 include/linux/usb_usual.h          | 4 +++-
 3 files changed, 12 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
index e80362d148f3..a1128ff5cc2c 100644
--- a/drivers/usb/storage/scsiglue.c
+++ b/drivers/usb/storage/scsiglue.c
@@ -253,6 +253,10 @@ static int slave_configure(struct scsi_device *sdev)
 		 * or to force 192-byte transfer lengths for MODE SENSE.
 		 * But they do need to use MODE SENSE(10). */
 		sdev->use_10_for_ms = 1;
+
+		/* Some (fake) usb cdrom devices don't like READ_DISC_INFO */
+		if (us->fflags & US_FL_NO_READ_DISC_INFO)
+			sdev->no_read_disc_info = 1;
 	}
 
 	/* The CB and CBI transports have no way to pass LUN values
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index f43314a22ba4..c8264ff5457e 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -1858,6 +1858,11 @@ UNUSUAL_DEV( 0x1908, 0x1320, 0x0000, 0x0000,
 		"Photo Frame",
 		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_BAD_SENSE ),
+UNUSUAL_DEV( 0x1908, 0x3335, 0x0200, 0x0200,
+		"BUILDWIN",
+		"Photo Frame",
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
+		US_FL_NO_READ_DISC_INFO ),
 
 UNUSUAL_DEV( 0x2116, 0x0320, 0x0001, 0x0001,
 		"ST",
diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h
index f091dc6e5a00..e62e9fe08883 100644
--- a/include/linux/usb_usual.h
+++ b/include/linux/usb_usual.h
@@ -58,7 +58,9 @@
 	US_FLAG(CAPACITY_OK,	0x00010000)			\
 		/* READ CAPACITY response is correct */		\
 	US_FLAG(BAD_SENSE,	0x00020000)			\
-		/* Bad Sense (never more than 18 bytes) */
+		/* Bad Sense (never more than 18 bytes) */	\
+	US_FLAG(NO_READ_DISC_INFO,	0x00040000)		\
+		/* cannot handle READ_DISC_INFO */
 
 #define US_FLAG(name, value)	US_FL_##name = value ,
 enum { US_DO_ALL_FLAGS };
-- 
cgit v1.2.3


From 00914025cc4e783d4703b4db1d47b41f389e50c8 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Fri, 1 Oct 2010 14:20:11 -0700
Subject: usb-storage: add new no_read_capacity_16 quirk

Some Rockbox based mp4 players will crash when ever they see a
read_capacity_16 scsi command.  So add a new US_FL which tells the scsi sd
driver to not issue any read_capacity_16 scsi commands.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/scsiglue.c     | 4 ++++
 drivers/usb/storage/unusual_devs.h | 3 ++-
 include/linux/usb_usual.h          | 4 +++-
 3 files changed, 9 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
index a1128ff5cc2c..a688b1e686ea 100644
--- a/drivers/usb/storage/scsiglue.c
+++ b/drivers/usb/storage/scsiglue.c
@@ -209,6 +209,10 @@ static int slave_configure(struct scsi_device *sdev)
 		if (us->fflags & US_FL_CAPACITY_HEURISTICS)
 			sdev->guess_capacity = 1;
 
+		/* Some devices cannot handle READ_CAPACITY_16 */
+		if (us->fflags & US_FL_NO_READ_CAPACITY_16)
+			sdev->no_read_capacity_16 = 1;
+
 		/* assume SPC3 or latter devices support sense size > 18 */
 		if (sdev->scsi_level > SCSI_SPC_2)
 			us->fflags |= US_FL_SANE_SENSE;
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index c8264ff5457e..6ccdd3dd5259 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -877,7 +877,8 @@ UNUSUAL_DEV(  0x071b, 0x3203, 0x0000, 0x0000,
 		"RockChip",
 		"MP3",
 		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
-		US_FL_NO_WP_DETECT | US_FL_MAX_SECTORS_64),
+		US_FL_NO_WP_DETECT | US_FL_MAX_SECTORS_64 |
+		US_FL_NO_READ_CAPACITY_16),
 
 /* Reported by Jean-Baptiste Onofre <jb@nanthrax.net>
  * Support the following product :
diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h
index e62e9fe08883..71693d4a4fe1 100644
--- a/include/linux/usb_usual.h
+++ b/include/linux/usb_usual.h
@@ -60,7 +60,9 @@
 	US_FLAG(BAD_SENSE,	0x00020000)			\
 		/* Bad Sense (never more than 18 bytes) */	\
 	US_FLAG(NO_READ_DISC_INFO,	0x00040000)		\
-		/* cannot handle READ_DISC_INFO */
+		/* cannot handle READ_DISC_INFO */		\
+	US_FLAG(NO_READ_CAPACITY_16,	0x00080000)		\
+		/* cannot handle READ_CAPACITY_16 */
 
 #define US_FLAG(name, value)	US_FL_##name = value ,
 enum { US_DO_ALL_FLAGS };
-- 
cgit v1.2.3


From 562e7c71c6708353bfe7b615576bcbcf7afd522e Mon Sep 17 00:00:00 2001
From: Tatyana Brokhman <tlinder@codeaurora.org>
Date: Sat, 9 Oct 2010 16:46:12 +0200
Subject: usb: usb3.0 ch9 definitions

Adding SuperSpeed usb definitions as defined by ch9 of the USB3.0 spec.
This patch is a preparation for adding SuperSpeed support to the gadget
framework.

Signed-off-by: Tatyana Brokhman <tlinder@codeaurora.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/ch9.h | 58 ++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 57 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/usb/ch9.h b/include/linux/usb/ch9.h
index b0f7e9f57176..f917bbbc8901 100644
--- a/include/linux/usb/ch9.h
+++ b/include/linux/usb/ch9.h
@@ -123,8 +123,23 @@
 #define USB_DEVICE_A_ALT_HNP_SUPPORT	5	/* (otg) other RH port does */
 #define USB_DEVICE_DEBUG_MODE		6	/* (special devices only) */
 
+/*
+ * New Feature Selectors as added by USB 3.0
+ * See USB 3.0 spec Table 9-6
+ */
+#define USB_DEVICE_U1_ENABLE	48	/* dev may initiate U1 transition */
+#define USB_DEVICE_U2_ENABLE	49	/* dev may initiate U2 transition */
+#define USB_DEVICE_LTM_ENABLE	50	/* dev may send LTM */
+#define USB_INTRF_FUNC_SUSPEND	0	/* function suspend */
+
+#define USB_INTR_FUNC_SUSPEND_OPT_MASK	0xFF00
+
 #define USB_ENDPOINT_HALT		0	/* IN/OUT will STALL */
 
+/* Bit array elements as returned by the USB_REQ_GET_STATUS request. */
+#define USB_DEV_STAT_U1_ENABLED		2	/* transition into U1 state */
+#define USB_DEV_STAT_U2_ENABLED		3	/* transition into U2 state */
+#define USB_DEV_STAT_LTM_ENABLED	4	/* Latency tolerance messages */
 
 /**
  * struct usb_ctrlrequest - SETUP data for a USB device control request
@@ -675,6 +690,7 @@ struct usb_bos_descriptor {
 	__u8  bNumDeviceCaps;
 } __attribute__((packed));
 
+#define USB_DT_BOS_SIZE		5
 /*-------------------------------------------------------------------------*/
 
 /* USB_DT_DEVICE_CAPABILITY:  grouped with BOS */
@@ -712,16 +728,56 @@ struct usb_wireless_cap_descriptor {	/* Ultra Wide Band */
 	__u8  bReserved;
 } __attribute__((packed));
 
+/* USB 2.0 Extension descriptor */
 #define	USB_CAP_TYPE_EXT		2
 
 struct usb_ext_cap_descriptor {		/* Link Power Management */
 	__u8  bLength;
 	__u8  bDescriptorType;
 	__u8  bDevCapabilityType;
-	__u8  bmAttributes;
+	__le32 bmAttributes;
 #define USB_LPM_SUPPORT			(1 << 1)	/* supports LPM */
 } __attribute__((packed));
 
+#define USB_DT_USB_EXT_CAP_SIZE	7
+
+/*
+ * SuperSpeed USB Capability descriptor: Defines the set of SuperSpeed USB
+ * specific device level capabilities
+ */
+#define		USB_SS_CAP_TYPE		3
+struct usb_ss_cap_descriptor {		/* Link Power Management */
+	__u8  bLength;
+	__u8  bDescriptorType;
+	__u8  bDevCapabilityType;
+	__u8  bmAttributes;
+#define USB_LTM_SUPPORT			(1 << 1) /* supports LTM */
+	__le16 wSpeedSupported;
+#define USB_LOW_SPEED_OPERATION		(1)	 /* Low speed operation */
+#define USB_FULL_SPEED_OPERATION	(1 << 1) /* Full speed operation */
+#define USB_HIGH_SPEED_OPERATION	(1 << 2) /* High speed operation */
+#define USB_5GBPS_OPERATION		(1 << 3) /* Operation at 5Gbps */
+	__u8  bFunctionalitySupport;
+	__u8  bU1devExitLat;
+	__le16 bU2DevExitLat;
+} __attribute__((packed));
+
+#define USB_DT_USB_SS_CAP_SIZE	10
+
+/*
+ * Container ID Capability descriptor: Defines the instance unique ID used to
+ * identify the instance across all operating modes
+ */
+#define	CONTAINER_ID_TYPE	4
+struct usb_ss_container_id_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;
+	__u8  bDevCapabilityType;
+	__u8  bReserved;
+	__u8  ContainerID[16]; /* 128-bit number */
+} __attribute__((packed));
+
+#define USB_DT_USB_SS_CONTN_ID_SIZE	20
 /*-------------------------------------------------------------------------*/
 
 /* USB_DT_WIRELESS_ENDPOINT_COMP:  companion descriptor associated with
-- 
cgit v1.2.3


From 69cb1ec4ce4da4bc4c07bb09c4c98b3e25d99fb1 Mon Sep 17 00:00:00 2001
From: Eric Bénard <eric@eukrea.com>
Date: Fri, 15 Oct 2010 14:30:58 +0200
Subject: mxc_udc: add workaround for ENGcm09152 for i.MX35
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

this patch gives the possibility to workaround bug ENGcm09152
on i.MX35 when the hardware workaround is also implemented on
the board.
It covers the workaround described on page 25 of the following Errata :
http://cache.freescale.com/files/dsp/doc/errata/IMX35CE.pdf

Signed-off-by: Eric Bénard <eric@eukrea.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/mach-mx3/mach-cpuimx35.c |  1 +
 drivers/usb/gadget/fsl_mxc_udc.c  | 15 +++++++++++++++
 include/linux/fsl_devices.h       |  3 +++
 3 files changed, 19 insertions(+)

(limited to 'include/linux')

diff --git a/arch/arm/mach-mx3/mach-cpuimx35.c b/arch/arm/mach-mx3/mach-cpuimx35.c
index 2a4f8b781ba4..4d161b3fca65 100644
--- a/arch/arm/mach-mx3/mach-cpuimx35.c
+++ b/arch/arm/mach-mx3/mach-cpuimx35.c
@@ -155,6 +155,7 @@ static struct mxc_usbh_platform_data usbh1_pdata = {
 static struct fsl_usb2_platform_data otg_device_pdata = {
 	.operating_mode	= FSL_USB2_DR_DEVICE,
 	.phy_mode	= FSL_USB2_PHY_UTMI,
+	.workaround	= FLS_USB2_WORKAROUND_ENGCM09152,
 };
 
 static int otg_mode_host;
diff --git a/drivers/usb/gadget/fsl_mxc_udc.c b/drivers/usb/gadget/fsl_mxc_udc.c
index eafa6d2c5ed7..5bdbfe619853 100644
--- a/drivers/usb/gadget/fsl_mxc_udc.c
+++ b/drivers/usb/gadget/fsl_mxc_udc.c
@@ -22,6 +22,10 @@
 static struct clk *mxc_ahb_clk;
 static struct clk *mxc_usb_clk;
 
+/* workaround ENGcm09152 for i.MX35 */
+#define USBPHYCTRL_OTGBASE_OFFSET	0x608
+#define USBPHYCTRL_EVDO			(1 << 23)
+
 int fsl_udc_clk_init(struct platform_device *pdev)
 {
 	struct fsl_usb2_platform_data *pdata;
@@ -84,6 +88,17 @@ eenahb:
 void fsl_udc_clk_finalize(struct platform_device *pdev)
 {
 	struct fsl_usb2_platform_data *pdata = pdev->dev.platform_data;
+#if defined(CONFIG_ARCH_MX35)
+	unsigned int v;
+
+	/* workaround ENGcm09152 for i.MX35 */
+	if (pdata->workaround & FLS_USB2_WORKAROUND_ENGCM09152) {
+		v = readl(MX35_IO_ADDRESS(MX35_OTG_BASE_ADDR +
+				USBPHYCTRL_OTGBASE_OFFSET));
+		writel(v | USBPHYCTRL_EVDO, MX35_IO_ADDRESS(MX35_OTG_BASE_ADDR +
+				USBPHYCTRL_OTGBASE_OFFSET));
+	}
+#endif
 
 	/* ULPI transceivers don't need usbpll */
 	if (pdata->phy_mode == FSL_USB2_PHY_ULPI) {
diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h
index d5f9a7431bd0..4eb56ed75fbc 100644
--- a/include/linux/fsl_devices.h
+++ b/include/linux/fsl_devices.h
@@ -66,6 +66,7 @@ struct fsl_usb2_platform_data {
 	enum fsl_usb2_operating_modes	operating_mode;
 	enum fsl_usb2_phy_modes		phy_mode;
 	unsigned int			port_enables;
+	unsigned int			workaround;
 
 	int		(*init)(struct platform_device *);
 	void		(*exit)(struct platform_device *);
@@ -84,6 +85,8 @@ struct fsl_usb2_platform_data {
 #define FSL_USB2_PORT0_ENABLED	0x00000001
 #define FSL_USB2_PORT1_ENABLED	0x00000002
 
+#define FLS_USB2_WORKAROUND_ENGCM09152	(1 << 0)
+
 struct spi_device;
 
 struct fsl_spi_platform_data {
-- 
cgit v1.2.3


From f7030bbc446430ecd12c9ad02cf0ea94934e5f91 Mon Sep 17 00:00:00 2001
From: Jason Wessel <jason.wessel@windriver.com>
Date: Mon, 11 Oct 2010 10:20:14 -0500
Subject: kdb: Allow kernel loadable modules to add kdb shell functions

In order to allow kernel modules to dynamically add a command to the
kdb shell the kdb_register, kdb_register_repeat, kdb_unregister, and
kdb_printf need to be exported as GPL symbols.

Any kernel module that adds a dynamic kdb shell function should only
need to include linux/kdb.h.

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
---
 include/linux/kdb.h            | 43 ++++++++++++++++++++++++++++++++++++++++++
 kernel/debug/kdb/kdb_io.c      |  2 +-
 kernel/debug/kdb/kdb_main.c    |  4 ++++
 kernel/debug/kdb/kdb_private.h | 39 --------------------------------------
 4 files changed, 48 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kdb.h b/include/linux/kdb.h
index ea6e5244ed3f..deda197ced62 100644
--- a/include/linux/kdb.h
+++ b/include/linux/kdb.h
@@ -28,6 +28,41 @@ extern int kdb_poll_idx;
 extern int kdb_initial_cpu;
 extern atomic_t kdb_event;
 
+/* Types and messages used for dynamically added kdb shell commands */
+
+#define KDB_MAXARGS    16 /* Maximum number of arguments to a function  */
+
+typedef enum {
+	KDB_REPEAT_NONE = 0,	/* Do not repeat this command */
+	KDB_REPEAT_NO_ARGS,	/* Repeat the command without arguments */
+	KDB_REPEAT_WITH_ARGS,	/* Repeat the command including its arguments */
+} kdb_repeat_t;
+
+typedef int (*kdb_func_t)(int, const char **);
+
+/* KDB return codes from a command or internal kdb function */
+#define KDB_NOTFOUND	(-1)
+#define KDB_ARGCOUNT	(-2)
+#define KDB_BADWIDTH	(-3)
+#define KDB_BADRADIX	(-4)
+#define KDB_NOTENV	(-5)
+#define KDB_NOENVVALUE	(-6)
+#define KDB_NOTIMP	(-7)
+#define KDB_ENVFULL	(-8)
+#define KDB_ENVBUFFULL	(-9)
+#define KDB_TOOMANYBPT	(-10)
+#define KDB_TOOMANYDBREGS (-11)
+#define KDB_DUPBPT	(-12)
+#define KDB_BPTNOTFOUND	(-13)
+#define KDB_BADMODE	(-14)
+#define KDB_BADINT	(-15)
+#define KDB_INVADDRFMT  (-16)
+#define KDB_BADREG      (-17)
+#define KDB_BADCPUNUM   (-18)
+#define KDB_BADLENGTH	(-19)
+#define KDB_NOBP	(-20)
+#define KDB_BADADDR	(-21)
+
 /*
  * kdb_diemsg
  *
@@ -105,9 +140,17 @@ int kdb_process_cpu(const struct task_struct *p)
 /* kdb access to register set for stack dumping */
 extern struct pt_regs *kdb_current_regs;
 
+/* Dynamic kdb shell command registration */
+extern int kdb_register(char *, kdb_func_t, char *, char *, short);
+extern int kdb_register_repeat(char *, kdb_func_t, char *, char *,
+			       short, kdb_repeat_t);
+extern int kdb_unregister(char *);
 #else /* ! CONFIG_KGDB_KDB */
 #define kdb_printf(...)
 #define kdb_init(x)
+#define kdb_register(...)
+#define kdb_register_repeat(...)
+#define kdb_uregister(x)
 #endif	/* CONFIG_KGDB_KDB */
 enum {
 	KDB_NOT_INITIALIZED,
diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c
index c9b7f4f90bba..96fdaac46a80 100644
--- a/kernel/debug/kdb/kdb_io.c
+++ b/kernel/debug/kdb/kdb_io.c
@@ -823,4 +823,4 @@ int kdb_printf(const char *fmt, ...)
 
 	return r;
 }
-
+EXPORT_SYMBOL_GPL(kdb_printf);
diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c
index caf057a3de0e..5448990a299e 100644
--- a/kernel/debug/kdb/kdb_main.c
+++ b/kernel/debug/kdb/kdb_main.c
@@ -2783,6 +2783,8 @@ int kdb_register_repeat(char *cmd,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(kdb_register_repeat);
+
 
 /*
  * kdb_register - Compatibility register function for commands that do
@@ -2805,6 +2807,7 @@ int kdb_register(char *cmd,
 	return kdb_register_repeat(cmd, func, usage, help, minlen,
 				   KDB_REPEAT_NONE);
 }
+EXPORT_SYMBOL_GPL(kdb_register);
 
 /*
  * kdb_unregister - This function is used to unregister a kernel
@@ -2833,6 +2836,7 @@ int kdb_unregister(char *cmd)
 	/* Couldn't find it.  */
 	return 1;
 }
+EXPORT_SYMBOL_GPL(kdb_unregister);
 
 /* Initialize the kdb command table. */
 static void __init kdb_inittab(void)
diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h
index be775f7e81e0..1921e6e4c0bc 100644
--- a/kernel/debug/kdb/kdb_private.h
+++ b/kernel/debug/kdb/kdb_private.h
@@ -15,29 +15,6 @@
 #include <linux/kgdb.h>
 #include "../debug_core.h"
 
-/* Kernel Debugger Error codes.  Must not overlap with command codes. */
-#define KDB_NOTFOUND	(-1)
-#define KDB_ARGCOUNT	(-2)
-#define KDB_BADWIDTH	(-3)
-#define KDB_BADRADIX	(-4)
-#define KDB_NOTENV	(-5)
-#define KDB_NOENVVALUE	(-6)
-#define KDB_NOTIMP	(-7)
-#define KDB_ENVFULL	(-8)
-#define KDB_ENVBUFFULL	(-9)
-#define KDB_TOOMANYBPT	(-10)
-#define KDB_TOOMANYDBREGS (-11)
-#define KDB_DUPBPT	(-12)
-#define KDB_BPTNOTFOUND	(-13)
-#define KDB_BADMODE	(-14)
-#define KDB_BADINT	(-15)
-#define KDB_INVADDRFMT  (-16)
-#define KDB_BADREG      (-17)
-#define KDB_BADCPUNUM   (-18)
-#define KDB_BADLENGTH	(-19)
-#define KDB_NOBP	(-20)
-#define KDB_BADADDR	(-21)
-
 /* Kernel Debugger Command codes.  Must not overlap with error codes. */
 #define KDB_CMD_GO	(-1001)
 #define KDB_CMD_CPU	(-1002)
@@ -93,17 +70,6 @@
  */
 #define KDB_MAXBPT	16
 
-/* Maximum number of arguments to a function  */
-#define KDB_MAXARGS    16
-
-typedef enum {
-	KDB_REPEAT_NONE = 0,	/* Do not repeat this command */
-	KDB_REPEAT_NO_ARGS,	/* Repeat the command without arguments */
-	KDB_REPEAT_WITH_ARGS,	/* Repeat the command including its arguments */
-} kdb_repeat_t;
-
-typedef int (*kdb_func_t)(int, const char **);
-
 /* Symbol table format returned by kallsyms. */
 typedef struct __ksymtab {
 		unsigned long value;	/* Address of symbol */
@@ -123,11 +89,6 @@ extern int kallsyms_symbol_next(char *prefix_name, int flag);
 extern int kallsyms_symbol_complete(char *prefix_name, int max_len);
 
 /* Exported Symbols for kernel loadable modules to use. */
-extern int kdb_register(char *, kdb_func_t, char *, char *, short);
-extern int kdb_register_repeat(char *, kdb_func_t, char *, char *,
-			       short, kdb_repeat_t);
-extern int kdb_unregister(char *);
-
 extern int kdb_getarea_size(void *, unsigned long, size_t);
 extern int kdb_putarea_size(unsigned long, void *, size_t);
 
-- 
cgit v1.2.3


From 91b152aa85bbcf076e269565394c31964f940371 Mon Sep 17 00:00:00 2001
From: Jason Wessel <jason.wessel@windriver.com>
Date: Mon, 23 Aug 2010 09:20:14 -0500
Subject: kdb,kgdb: fix sparse fixups

Fix the following sparse warnings:

kdb_main.c:328:5: warning: symbol 'kdbgetu64arg' was not declared. Should it be static?
kgdboc.c:246:12: warning: symbol 'kgdboc_early_init' was not declared. Should it be static?
kgdb.c:652:26: warning: incorrect type in argument 1 (different address spaces)
kgdb.c:652:26:    expected void const *ptr
kgdb.c:652:26:    got struct perf_event *[noderef] <asn:3>*pev

The one in kgdb.c required the (void * __force) because of the return
code from register_wide_hw_breakpoint looking like:

        return (void __percpu __force *)ERR_PTR(err);

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
---
 arch/x86/kernel/kgdb.c         | 2 +-
 drivers/serial/kgdboc.c        | 2 +-
 include/linux/kdb.h            | 8 ++++++++
 kernel/debug/kdb/kdb_private.h | 9 +--------
 4 files changed, 11 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
index 497f97386412..101bf22cf164 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -649,7 +649,7 @@ void kgdb_arch_late(void)
 		if (breakinfo[i].pev)
 			continue;
 		breakinfo[i].pev = register_wide_hw_breakpoint(&attr, NULL);
-		if (IS_ERR(breakinfo[i].pev)) {
+		if (IS_ERR((void * __force)breakinfo[i].pev)) {
 			printk(KERN_ERR "kgdb: Could not allocate hw"
 			       "breakpoints\nDisabling the kernel debugger\n");
 			breakinfo[i].pev = NULL;
diff --git a/drivers/serial/kgdboc.c b/drivers/serial/kgdboc.c
index 39f9a1adaa75..d4b711c9a416 100644
--- a/drivers/serial/kgdboc.c
+++ b/drivers/serial/kgdboc.c
@@ -243,7 +243,7 @@ static struct kgdb_io kgdboc_io_ops = {
 
 #ifdef CONFIG_KGDB_SERIAL_CONSOLE
 /* This is only available if kgdboc is a built in for early debugging */
-int __init kgdboc_early_init(char *opt)
+static int __init kgdboc_early_init(char *opt)
 {
 	/* save the first character of the config string because the
 	 * init routine can destroy it.
diff --git a/include/linux/kdb.h b/include/linux/kdb.h
index deda197ced62..aadff7cc2b84 100644
--- a/include/linux/kdb.h
+++ b/include/linux/kdb.h
@@ -139,6 +139,14 @@ int kdb_process_cpu(const struct task_struct *p)
 
 /* kdb access to register set for stack dumping */
 extern struct pt_regs *kdb_current_regs;
+#ifdef CONFIG_KALLSYMS
+extern const char *kdb_walk_kallsyms(loff_t *pos);
+#else /* ! CONFIG_KALLSYMS */
+static inline const char *kdb_walk_kallsyms(loff_t *pos)
+{
+	return NULL;
+}
+#endif /* ! CONFIG_KALLSYMS */
 
 /* Dynamic kdb shell command registration */
 extern int kdb_register(char *, kdb_func_t, char *, char *, short);
diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h
index 1921e6e4c0bc..35d69ed1dfb5 100644
--- a/kernel/debug/kdb/kdb_private.h
+++ b/kernel/debug/kdb/kdb_private.h
@@ -105,6 +105,7 @@ extern int kdb_getword(unsigned long *, unsigned long, size_t);
 extern int kdb_putword(unsigned long, unsigned long, size_t);
 
 extern int kdbgetularg(const char *, unsigned long *);
+extern int kdbgetu64arg(const char *, u64 *);
 extern char *kdbgetenv(const char *);
 extern int kdbgetaddrarg(int, const char **, int*, unsigned long *,
 			 long *, char **);
@@ -216,14 +217,6 @@ extern void kdb_ps1(const struct task_struct *p);
 extern void kdb_print_nameval(const char *name, unsigned long val);
 extern void kdb_send_sig_info(struct task_struct *p, struct siginfo *info);
 extern void kdb_meminfo_proc_show(void);
-#ifdef CONFIG_KALLSYMS
-extern const char *kdb_walk_kallsyms(loff_t *pos);
-#else /* ! CONFIG_KALLSYMS */
-static inline const char *kdb_walk_kallsyms(loff_t *pos)
-{
-	return NULL;
-}
-#endif /* ! CONFIG_KALLSYMS */
 extern char *kdb_getstr(char *, size_t, char *);
 
 /* Defines for kdb_symbol_print */
-- 
cgit v1.2.3


From 9566a7a851eb7201e3207eab53ee81efd0850fee Mon Sep 17 00:00:00 2001
From: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Date: Tue, 10 Aug 2010 00:58:41 +0900
Subject: nilfs2: accept future revisions

Compatibility of nilfs partitions is now managed with three feature
sets.  This changes old compatibility check with revision number so
that it can accept future revisions.

Note that we can stop support of experimental versions of nilfs that
doesn't know the feature sets by incrementing NILFS_CURRENT_REV.  We
don't have to do it soon, but it would be a possible option whenever
the need arises.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
---
 fs/nilfs2/the_nilfs.c     | 4 ++--
 include/linux/nilfs2_fs.h | 1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c
index ba7c10c917fc..461b7211e14f 100644
--- a/fs/nilfs2/the_nilfs.c
+++ b/fs/nilfs2/the_nilfs.c
@@ -468,8 +468,8 @@ static unsigned long long nilfs_max_size(unsigned int blkbits)
 static int nilfs_store_disk_layout(struct the_nilfs *nilfs,
 				   struct nilfs_super_block *sbp)
 {
-	if (le32_to_cpu(sbp->s_rev_level) != NILFS_CURRENT_REV) {
-		printk(KERN_ERR "NILFS: revision mismatch "
+	if (le32_to_cpu(sbp->s_rev_level) < NILFS_MIN_SUPP_REV) {
+		printk(KERN_ERR "NILFS: unsupported revision "
 		       "(superblock rev.=%d.%d, current rev.=%d.%d). "
 		       "Please check the version of mkfs.nilfs.\n",
 		       le32_to_cpu(sbp->s_rev_level),
diff --git a/include/linux/nilfs2_fs.h b/include/linux/nilfs2_fs.h
index f5487b6f91ed..b07f5cdff5e2 100644
--- a/include/linux/nilfs2_fs.h
+++ b/include/linux/nilfs2_fs.h
@@ -229,6 +229,7 @@ struct nilfs_super_block {
  */
 #define NILFS_CURRENT_REV	2	/* current major revision */
 #define NILFS_MINOR_REV		0	/* minor revision */
+#define NILFS_MIN_SUPP_REV	2	/* minimum supported revision */
 
 /*
  * Feature set definitions
-- 
cgit v1.2.3


From 6c43f41000312fefa482c3bfdd97e7f81d6be0ec Mon Sep 17 00:00:00 2001
From: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Date: Fri, 20 Aug 2010 20:10:38 +0900
Subject: nilfs2: keep zero value in i_cno except for gc-inodes

On-memory inode structures of nilfs have a member "i_cno" which stores
a checkpoint number related to the inode.  For gc-inodes, this field
indicates version of data each gc-inode caches for GC.  Log writer
temporarily uses "i_cno" to transfer the latest checkpoint number.

This stops the latter use and lets only gc-inodes use it.

The purpose of this patch is to allow the successive change use
"i_cno" for inode lookup.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
---
 fs/nilfs2/segment.c       | 29 ++++++++++++++---------------
 fs/nilfs2/segment.h       |  3 ++-
 include/linux/nilfs2_fs.h |  8 ++++++++
 3 files changed, 24 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c
index 9fd051a33c4f..eee4b223c293 100644
--- a/fs/nilfs2/segment.c
+++ b/fs/nilfs2/segment.c
@@ -366,8 +366,7 @@ static int nilfs_segctor_reset_segment_buffer(struct nilfs_sc_info *sci)
 
 	if (nilfs_doing_gc())
 		flags = NILFS_SS_GC;
-	err = nilfs_segbuf_reset(segbuf, flags, sci->sc_seg_ctime,
-				 sci->sc_sbi->s_nilfs->ns_cno);
+	err = nilfs_segbuf_reset(segbuf, flags, sci->sc_seg_ctime, sci->sc_cno);
 	if (unlikely(err))
 		return err;
 
@@ -440,17 +439,26 @@ static void nilfs_segctor_end_finfo(struct nilfs_sc_info *sci,
 	struct nilfs_finfo *finfo;
 	struct nilfs_inode_info *ii;
 	struct nilfs_segment_buffer *segbuf;
+	__u64 cno;
 
 	if (sci->sc_blk_cnt == 0)
 		return;
 
 	ii = NILFS_I(inode);
+
+	if (test_bit(NILFS_I_GCINODE, &ii->i_state))
+		cno = ii->i_cno;
+	else if (NILFS_ROOT_METADATA_FILE(inode->i_ino))
+		cno = 0;
+	else
+		cno = sci->sc_cno;
+
 	finfo = nilfs_segctor_map_segsum_entry(sci, &sci->sc_finfo_ptr,
 						 sizeof(*finfo));
 	finfo->fi_ino = cpu_to_le64(inode->i_ino);
 	finfo->fi_nblocks = cpu_to_le32(sci->sc_blk_cnt);
 	finfo->fi_ndatablk = cpu_to_le32(sci->sc_datablk_cnt);
-	finfo->fi_cno = cpu_to_le64(ii->i_cno);
+	finfo->fi_cno = cpu_to_le64(cno);
 
 	segbuf = sci->sc_curseg;
 	segbuf->sb_sum.sumbytes = sci->sc_binfo_ptr.offset +
@@ -1976,7 +1984,6 @@ static int nilfs_segctor_check_in_files(struct nilfs_sc_info *sci,
 					struct nilfs_sb_info *sbi)
 {
 	struct nilfs_inode_info *ii, *n;
-	__u64 cno = sbi->s_nilfs->ns_cno;
 
 	spin_lock(&sbi->s_inode_lock);
  retry:
@@ -2002,7 +2009,6 @@ static int nilfs_segctor_check_in_files(struct nilfs_sc_info *sci,
 				brelse(ibh);
 			goto retry;
 		}
-		ii->i_cno = cno;
 
 		clear_bit(NILFS_I_QUEUED, &ii->i_state);
 		set_bit(NILFS_I_BUSY, &ii->i_state);
@@ -2011,8 +2017,6 @@ static int nilfs_segctor_check_in_files(struct nilfs_sc_info *sci,
 	}
 	spin_unlock(&sbi->s_inode_lock);
 
-	NILFS_I(sbi->s_ifile)->i_cno = cno;
-
 	return 0;
 }
 
@@ -2021,19 +2025,13 @@ static void nilfs_segctor_check_out_files(struct nilfs_sc_info *sci,
 {
 	struct nilfs_transaction_info *ti = current->journal_info;
 	struct nilfs_inode_info *ii, *n;
-	__u64 cno = sbi->s_nilfs->ns_cno;
 
 	spin_lock(&sbi->s_inode_lock);
 	list_for_each_entry_safe(ii, n, &sci->sc_dirty_files, i_dirty) {
 		if (!test_and_clear_bit(NILFS_I_UPDATED, &ii->i_state) ||
-		    test_bit(NILFS_I_DIRTY, &ii->i_state)) {
-			/* The current checkpoint number (=nilfs->ns_cno) is
-			   changed between check-in and check-out only if the
-			   super root is written out.  So, we can update i_cno
-			   for the inodes that remain in the dirty list. */
-			ii->i_cno = cno;
+		    test_bit(NILFS_I_DIRTY, &ii->i_state))
 			continue;
-		}
+
 		clear_bit(NILFS_I_BUSY, &ii->i_state);
 		brelse(ii->i_bh);
 		ii->i_bh = NULL;
@@ -2054,6 +2052,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 	int err;
 
 	sci->sc_stage.scnt = NILFS_ST_INIT;
+	sci->sc_cno = nilfs->ns_cno;
 
 	err = nilfs_segctor_check_in_files(sci, sbi);
 	if (unlikely(err))
diff --git a/fs/nilfs2/segment.h b/fs/nilfs2/segment.h
index 17c487bd8152..675d932148a4 100644
--- a/fs/nilfs2/segment.h
+++ b/fs/nilfs2/segment.h
@@ -107,6 +107,7 @@ struct nilfs_segsum_pointer {
  * @sc_datablk_cnt: Data block count of a file
  * @sc_nblk_this_inc: Number of blocks included in the current logical segment
  * @sc_seg_ctime: Creation time
+ * @sc_cno: checkpoint number of current log
  * @sc_flags: Internal flags
  * @sc_state_lock: spinlock for sc_state and so on
  * @sc_state: Segctord state flags
@@ -156,7 +157,7 @@ struct nilfs_sc_info {
 	unsigned long		sc_datablk_cnt;
 	unsigned long		sc_nblk_this_inc;
 	time_t			sc_seg_ctime;
-
+	__u64			sc_cno;
 	unsigned long		sc_flags;
 
 	spinlock_t		sc_state_lock;
diff --git a/include/linux/nilfs2_fs.h b/include/linux/nilfs2_fs.h
index b07f5cdff5e2..bcdb34c68d08 100644
--- a/include/linux/nilfs2_fs.h
+++ b/include/linux/nilfs2_fs.h
@@ -270,6 +270,14 @@ struct nilfs_super_block {
 #define NILFS_MIN_NRSVSEGS	8	/* Minimum number of reserved
 					   segments */
 
+/*
+ * We call DAT, cpfile, and sufile root metadata files.  Inodes of
+ * these files are written in super root block instead of ifile, and
+ * garbage collector doesn't keep any past versions of these files.
+ */
+#define NILFS_ROOT_METADATA_FILE(ino) \
+	((ino) >= NILFS_DAT_INO && (ino) <= NILFS_SUFILE_INO)
+
 /*
  * bytes offset of secondary super block
  */
-- 
cgit v1.2.3


From 8e656fd518784b49453f60c5f78b78703bc85cb2 Mon Sep 17 00:00:00 2001
From: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Date: Fri, 27 Aug 2010 00:23:02 +0900
Subject: nilfs2: make snapshots in checkpoint tree exportable

The previous export operations cannot handle multiple versions of
a filesystem if they belong to the same sb instance.

This adds a new type of file handle and extends export operations so
that they can get the inode specified by a checkpoint number as well
as an inode number and a generation number.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
---
 fs/nilfs2/export.h       |  17 ++++++
 fs/nilfs2/namei.c        | 137 +++++++++++++++++++++++++++++++++++++++++------
 fs/nilfs2/nilfs.h        |   3 --
 fs/nilfs2/super.c        |  52 +-----------------
 include/linux/exportfs.h |  13 +++++
 5 files changed, 151 insertions(+), 71 deletions(-)
 create mode 100644 fs/nilfs2/export.h

(limited to 'include/linux')

diff --git a/fs/nilfs2/export.h b/fs/nilfs2/export.h
new file mode 100644
index 000000000000..a71cc412b651
--- /dev/null
+++ b/fs/nilfs2/export.h
@@ -0,0 +1,17 @@
+#ifndef NILFS_EXPORT_H
+#define NILFS_EXPORT_H
+
+#include <linux/exportfs.h>
+
+extern const struct export_operations nilfs_export_ops;
+
+struct nilfs_fid {
+	u64 cno;
+	u64 ino;
+	u32 gen;
+
+	u32 parent_gen;
+	u64 parent_ino;
+} __attribute__ ((packed));
+
+#endif
diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c
index 1110d56a23f5..a65f46560fbe 100644
--- a/fs/nilfs2/namei.c
+++ b/fs/nilfs2/namei.c
@@ -40,7 +40,11 @@
 
 #include <linux/pagemap.h>
 #include "nilfs.h"
+#include "export.h"
 
+#define NILFS_FID_SIZE_NON_CONNECTABLE \
+	(offsetof(struct nilfs_fid, parent_gen) / 4)
+#define NILFS_FID_SIZE_CONNECTABLE	(sizeof(struct nilfs_fid) / 4)
 
 static inline int nilfs_add_nondir(struct dentry *dentry, struct inode *inode)
 {
@@ -77,23 +81,6 @@ nilfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 	return d_splice_alias(inode, dentry);
 }
 
-struct dentry *nilfs_get_parent(struct dentry *child)
-{
-	unsigned long ino;
-	struct inode *inode;
-	struct qstr dotdot = {.name = "..", .len = 2};
-
-	ino = nilfs_inode_by_name(child->d_inode, &dotdot);
-	if (!ino)
-		return ERR_PTR(-ENOENT);
-
-	inode = nilfs_iget(child->d_inode->i_sb,
-			   NILFS_I(child->d_inode)->i_root, ino);
-	if (IS_ERR(inode))
-		return ERR_CAST(inode);
-	return d_obtain_alias(inode);
-}
-
 /*
  * By the time this is called, we already have created
  * the directory cache entry for the new file, but it
@@ -469,6 +456,115 @@ out:
 	return err;
 }
 
+/*
+ * Export operations
+ */
+static struct dentry *nilfs_get_parent(struct dentry *child)
+{
+	unsigned long ino;
+	struct inode *inode;
+	struct qstr dotdot = {.name = "..", .len = 2};
+	struct nilfs_root *root;
+
+	ino = nilfs_inode_by_name(child->d_inode, &dotdot);
+	if (!ino)
+		return ERR_PTR(-ENOENT);
+
+	root = NILFS_I(child->d_inode)->i_root;
+
+	inode = nilfs_iget(child->d_inode->i_sb, root, ino);
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
+
+	return d_obtain_alias(inode);
+}
+
+static struct dentry *nilfs_get_dentry(struct super_block *sb, u64 cno,
+				       u64 ino, u32 gen)
+{
+	struct nilfs_root *root;
+	struct inode *inode;
+
+	if (ino < NILFS_FIRST_INO(sb) && ino != NILFS_ROOT_INO)
+		return ERR_PTR(-ESTALE);
+
+	root = nilfs_lookup_root(NILFS_SB(sb)->s_nilfs, cno);
+	if (!root)
+		return ERR_PTR(-ESTALE);
+
+	inode = nilfs_iget(sb, root, ino);
+	nilfs_put_root(root);
+
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
+	if (gen && inode->i_generation != gen) {
+		iput(inode);
+		return ERR_PTR(-ESTALE);
+	}
+	return d_obtain_alias(inode);
+}
+
+static struct dentry *nilfs_fh_to_dentry(struct super_block *sb, struct fid *fh,
+					 int fh_len, int fh_type)
+{
+	struct nilfs_fid *fid = (struct nilfs_fid *)fh;
+
+	if ((fh_len != NILFS_FID_SIZE_NON_CONNECTABLE &&
+	     fh_len != NILFS_FID_SIZE_CONNECTABLE) ||
+	    (fh_type != FILEID_NILFS_WITH_PARENT &&
+	     fh_type != FILEID_NILFS_WITHOUT_PARENT))
+		return NULL;
+
+	return nilfs_get_dentry(sb, fid->cno, fid->ino, fid->gen);
+}
+
+static struct dentry *nilfs_fh_to_parent(struct super_block *sb, struct fid *fh,
+					 int fh_len, int fh_type)
+{
+	struct nilfs_fid *fid = (struct nilfs_fid *)fh;
+
+	if (fh_len != NILFS_FID_SIZE_CONNECTABLE ||
+	    fh_type != FILEID_NILFS_WITH_PARENT)
+		return NULL;
+
+	return nilfs_get_dentry(sb, fid->cno, fid->parent_ino, fid->parent_gen);
+}
+
+static int nilfs_encode_fh(struct dentry *dentry, __u32 *fh, int *lenp,
+			   int connectable)
+{
+	struct nilfs_fid *fid = (struct nilfs_fid *)fh;
+	struct inode *inode = dentry->d_inode;
+	struct nilfs_root *root = NILFS_I(inode)->i_root;
+	int type;
+
+	if (*lenp < NILFS_FID_SIZE_NON_CONNECTABLE ||
+	    (connectable && *lenp < NILFS_FID_SIZE_CONNECTABLE))
+		return 255;
+
+	fid->cno = root->cno;
+	fid->ino = inode->i_ino;
+	fid->gen = inode->i_generation;
+
+	if (connectable && !S_ISDIR(inode->i_mode)) {
+		struct inode *parent;
+
+		spin_lock(&dentry->d_lock);
+		parent = dentry->d_parent->d_inode;
+		fid->parent_ino = parent->i_ino;
+		fid->parent_gen = parent->i_generation;
+		spin_unlock(&dentry->d_lock);
+
+		type = FILEID_NILFS_WITH_PARENT;
+		*lenp = NILFS_FID_SIZE_CONNECTABLE;
+	} else {
+		type = FILEID_NILFS_WITHOUT_PARENT;
+		*lenp = NILFS_FID_SIZE_NON_CONNECTABLE;
+	}
+
+	return type;
+}
+
 const struct inode_operations nilfs_dir_inode_operations = {
 	.create		= nilfs_create,
 	.lookup		= nilfs_lookup,
@@ -493,3 +589,10 @@ const struct inode_operations nilfs_symlink_inode_operations = {
 	.follow_link	= page_follow_link_light,
 	.put_link	= page_put_link,
 };
+
+const struct export_operations nilfs_export_ops = {
+	.encode_fh = nilfs_encode_fh,
+	.fh_to_dentry = nilfs_fh_to_dentry,
+	.fh_to_parent = nilfs_fh_to_parent,
+	.get_parent = nilfs_get_parent,
+};
diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h
index 21d90c4b4e2d..aa7940f7ecf1 100644
--- a/fs/nilfs2/nilfs.h
+++ b/fs/nilfs2/nilfs.h
@@ -264,9 +264,6 @@ extern int nilfs_set_file_dirty(struct nilfs_sb_info *, struct inode *,
 extern int nilfs_mark_inode_dirty(struct inode *);
 extern void nilfs_dirty_inode(struct inode *);
 
-/* namei.c */
-extern struct dentry *nilfs_get_parent(struct dentry *);
-
 /* super.c */
 extern struct inode *nilfs_alloc_inode_common(struct the_nilfs *);
 extern struct inode *nilfs_alloc_inode(struct super_block *);
diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c
index a1c0e38a7706..adbf5826b837 100644
--- a/fs/nilfs2/super.c
+++ b/fs/nilfs2/super.c
@@ -48,10 +48,10 @@
 #include <linux/vfs.h>
 #include <linux/writeback.h>
 #include <linux/kobject.h>
-#include <linux/exportfs.h>
 #include <linux/seq_file.h>
 #include <linux/mount.h>
 #include "nilfs.h"
+#include "export.h"
 #include "mdt.h"
 #include "alloc.h"
 #include "btree.h"
@@ -556,56 +556,6 @@ static const struct super_operations nilfs_sops = {
 	.show_options = nilfs_show_options
 };
 
-static struct inode *
-nilfs_nfs_get_inode(struct super_block *sb, u64 ino, u32 generation)
-{
-	struct inode *inode;
-	struct nilfs_root *root;
-
-	if (ino < NILFS_FIRST_INO(sb) && ino != NILFS_ROOT_INO &&
-	    ino != NILFS_SKETCH_INO)
-		return ERR_PTR(-ESTALE);
-
-	root = nilfs_lookup_root(NILFS_SB(sb)->s_nilfs,
-				 NILFS_CPTREE_CURRENT_CNO);
-	if (!root)
-		return ERR_PTR(-ESTALE);
-
-	/* new file handle type is required to export snapshots */
-	inode = nilfs_iget(sb, root, ino);
-	nilfs_put_root(root);
-	if (IS_ERR(inode))
-		return ERR_CAST(inode);
-	if (generation && inode->i_generation != generation) {
-		iput(inode);
-		return ERR_PTR(-ESTALE);
-	}
-
-	return inode;
-}
-
-static struct dentry *
-nilfs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len,
-		   int fh_type)
-{
-	return generic_fh_to_dentry(sb, fid, fh_len, fh_type,
-				    nilfs_nfs_get_inode);
-}
-
-static struct dentry *
-nilfs_fh_to_parent(struct super_block *sb, struct fid *fid, int fh_len,
-		   int fh_type)
-{
-	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
-				    nilfs_nfs_get_inode);
-}
-
-static const struct export_operations nilfs_export_ops = {
-	.fh_to_dentry = nilfs_fh_to_dentry,
-	.fh_to_parent = nilfs_fh_to_parent,
-	.get_parent = nilfs_get_parent,
-};
-
 enum {
 	Opt_err_cont, Opt_err_panic, Opt_err_ro,
 	Opt_barrier, Opt_nobarrier, Opt_snapshot, Opt_order, Opt_norecovery,
diff --git a/include/linux/exportfs.h b/include/linux/exportfs.h
index a9cd507f8cd2..28028988c862 100644
--- a/include/linux/exportfs.h
+++ b/include/linux/exportfs.h
@@ -67,6 +67,19 @@ enum fid_type {
 	 * 32 bit parent block number, 32 bit parent generation number
 	 */
 	FILEID_UDF_WITH_PARENT = 0x52,
+
+	/*
+	 * 64 bit checkpoint number, 64 bit inode number,
+	 * 32 bit generation number.
+	 */
+	FILEID_NILFS_WITHOUT_PARENT = 0x61,
+
+	/*
+	 * 64 bit checkpoint number, 64 bit inode number,
+	 * 32 bit generation number, 32 bit parent generation.
+	 * 64 bit parent inode number.
+	 */
+	FILEID_NILFS_WITH_PARENT = 0x62,
 };
 
 struct fid {
-- 
cgit v1.2.3


From b453c95eb8d6a3b2348e9c7bc28a7d223cb640e3 Mon Sep 17 00:00:00 2001
From: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Date: Wed, 25 Aug 2010 23:52:46 +0900
Subject: nilfs2: get rid of snapshot mount flag

This flag is a fake used to distinguish type of super block instance.
And, it got obsolete by the unification of sb.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
---
 include/linux/nilfs2_fs.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/nilfs2_fs.h b/include/linux/nilfs2_fs.h
index bcdb34c68d08..46604671ccd5 100644
--- a/include/linux/nilfs2_fs.h
+++ b/include/linux/nilfs2_fs.h
@@ -147,7 +147,6 @@ struct nilfs_super_root {
 #define NILFS_MOUNT_ERRORS_CONT		0x0010  /* Continue on errors */
 #define NILFS_MOUNT_ERRORS_RO		0x0020  /* Remount fs ro on errors */
 #define NILFS_MOUNT_ERRORS_PANIC	0x0040  /* Panic on errors */
-#define NILFS_MOUNT_SNAPSHOT		0x0080  /* Snapshot flag */
 #define NILFS_MOUNT_BARRIER		0x1000  /* Use block barriers */
 #define NILFS_MOUNT_STRICT_ORDER	0x2000  /* Apply strict in-order
 						   semantics also for data */
-- 
cgit v1.2.3


From c486f3895d6dc751f7c0f04f0fa67390ce4d168e Mon Sep 17 00:00:00 2001
From: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Date: Sun, 3 Oct 2010 17:44:03 +0900
Subject: nilfs2: change license of exported header file

This allows other projects to carry copies of the header file related
to ABI and disk format (i.e. "nilfs2_fs.h") without it or distributors
having to worry about effects on the project's overall license terms.
It's also desired for switching the license of nilfs library to LGPL.

Jiro SEKIBA pointed out these license issues (Message-ID:
<87tylo7msw.wl%jir@sekiba.com>), and he suggested switching license of
the library and nilfs2_fs.h to GNU Lesser General Public License.  We
take in his suggestion to avoid the license issues.

Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Cc: Jiro SEKIBA <jir@unicus.jp>
Cc: linux-nilfs <linux-nilfs@vger.kernel.org>
---
 include/linux/nilfs2_fs.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nilfs2_fs.h b/include/linux/nilfs2_fs.h
index 46604671ccd5..227e49dd5720 100644
--- a/include/linux/nilfs2_fs.h
+++ b/include/linux/nilfs2_fs.h
@@ -4,16 +4,16 @@
  * Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation.
  *
  * 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, or
+ * it under the terms of the GNU Lesser General Public License as published
+ * by the Free Software Foundation; either version 2.1 of the License, or
  * (at your option) any later version.
  *
  * 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.
+ * GNU Lesser General Public License for more details.
  *
- * You should have received a copy of the GNU General Public License
+ * You should have received a copy of the GNU Lesser General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  *
-- 
cgit v1.2.3