[001/111] ata_piix: fix locking around SIDPR access

[001/111] ata_piix: fix locking around SIDPR access

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Tejun Heo <tj@kernel.org>

commit 213373cf974fe69e78ec894b07f45ae2f5a3a078 upstream.

SIDPR window registers are shared across ports and as each access is
done in two steps, accesses to different ports under EH may race.
This primarily is caused by incorrect host locking in EH context and
should be fixed by defining locking requirements for each EH operation
which can be used during EH and enforcing them but for now work around
the problem by adding a dedicated SIDPR lock and grabbing it for each
SIDPR access.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Mark Knecht <markknecht@gmail.com>
Reported-by: Paul Check <paul@thechecks.ca>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/ata/ata_piix.c |    8 ++++++++
 1 file changed, 8 insertions(+)

--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -157,6 +157,7 @@ struct piix_map_db {
 struct piix_host_priv {
 	const int *map;
 	u32 saved_iocfg;
+	spinlock_t sidpr_lock;	/* FIXME: remove once locking in EH is fixed */
 	void __iomem *sidpr;
 };
 
@@ -948,12 +949,15 @@ static int piix_sidpr_scr_read(struct at
 			       unsigned int reg, u32 *val)
 {
 	struct piix_host_priv *hpriv = link->ap->host->private_data;
+	unsigned long flags;
 
 	if (reg >= ARRAY_SIZE(piix_sidx_map))
 		return -EINVAL;
 
+	spin_lock_irqsave(&hpriv->sidpr_lock, flags);
 	piix_sidpr_sel(link, reg);
 	*val = ioread32(hpriv->sidpr + PIIX_SIDPR_DATA);
+	spin_unlock_irqrestore(&hpriv->sidpr_lock, flags);
 	return 0;
 }
 
@@ -961,12 +965,15 @@ static int piix_sidpr_scr_write(struct a
 				unsigned int reg, u32 val)
 {
 	struct piix_host_priv *hpriv = link->ap->host->private_data;
+	unsigned long flags;
 
 	if (reg >= ARRAY_SIZE(piix_sidx_map))
 		return -EINVAL;
 
+	spin_lock_irqsave(&hpriv->sidpr_lock, flags);
 	piix_sidpr_sel(link, reg);
 	iowrite32(val, hpriv->sidpr + PIIX_SIDPR_DATA);
+	spin_unlock_irqrestore(&hpriv->sidpr_lock, flags);
 	return 0;
 }
 
@@ -1555,6 +1562,7 @@ static int __devinit piix_init_one(struc
 	hpriv = devm_kzalloc(dev, sizeof(*hpriv), GFP_KERNEL);
 	if (!hpriv)
 		return -ENOMEM;
+	spin_lock_init(&hpriv->sidpr_lock);
 
 	/* Save IOCFG, this will be used for cable detection, quirk
 	 * detection and restoration on detach.  This is necessary

[002/111] powerpc: fix build with make 3.82

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Sam Ravnborg <sam@ravnborg.org>

commit e32e78c5ee8aadef020fbaecbe6fb741ed9029fd upstream.

Thomas Backlund reported that the powerpc build broke with make 3.82.
It failed with the following message:

    arch/powerpc/Makefile:183: *** mixed implicit and normal rules.  Stop.

The fix is to avoid mixing non-wildcard and wildcard targets.

Reported-by: Thomas Backlund <tmb@mandriva.org>
Tested-by: Thomas Backlund <tmb@mandriva.org>
Cc: Michal Marek <mmarek@suse.cz>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/powerpc/Makefile |   16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

--- a/arch/powerpc/Makefile
+++ b/arch/powerpc/Makefile
@@ -158,9 +158,11 @@ drivers-$(CONFIG_OPROFILE)	+= arch/power
 # Default to zImage, override when needed
 all: zImage
 
-BOOT_TARGETS = zImage zImage.initrd uImage zImage% dtbImage% treeImage.% cuImage.% simpleImage.%
+# With make 3.82 we cannot mix normal and wildcard targets
+BOOT_TARGETS1 := zImage zImage.initrd uImaged
+BOOT_TARGETS2 := zImage% dtbImage% treeImage.% cuImage.% simpleImage.%
 
-PHONY += $(BOOT_TARGETS)
+PHONY += $(BOOT_TARGETS1) $(BOOT_TARGETS2)
 
 boot := arch/$(ARCH)/boot
 
@@ -175,10 +177,16 @@ relocs_check: arch/powerpc/relocs_check.
 zImage: relocs_check
 endif
 
-$(BOOT_TARGETS): vmlinux
+$(BOOT_TARGETS1): vmlinux
+	$(Q)$(MAKE) ARCH=ppc64 $(build)=$(boot) $(patsubst %,$(boot)/%,$@)
+$(BOOT_TARGETS2): vmlinux
+	$(Q)$(MAKE) ARCH=ppc64 $(build)=$(boot) $(patsubst %,$(boot)/%,$@)
+
+
+bootwrapper_install:
 	$(Q)$(MAKE) ARCH=ppc64 $(build)=$(boot) $(patsubst %,$(boot)/%,$@)
 
-bootwrapper_install %.dtb:
+%.dtb:
 	$(Q)$(MAKE) ARCH=ppc64 $(build)=$(boot) $(patsubst %,$(boot)/%,$@)
 
 define archhelp

[003/111] nvram: Fix write beyond end condition; prove to gcc copy is safe

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: H. Peter Anvin <hpa@zytor.com>

commit a01c7800420d2c294ca403988488a635d4087a6d upstream.

In nvram_write, first of all, correctly handle the case where the file
pointer is already beyond the end; we should return EOF in that case.

Second, make the logic a bit more explicit so that gcc can statically
prove that the copy_from_user() is safe.  Once the condition of the
beyond-end filepointer is eliminated, the copy is safe but gcc can't
prove it, causing build failures for i386 allyesconfig.

Third, eliminate the entirely superfluous variable "len", and just use
the passed-in variable "count" instead.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Wim Van Sebroeck <wim@iguana.be>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <tip-*@git.kernel.org>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/char/nvram.c |   14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

--- a/drivers/char/nvram.c
+++ b/drivers/char/nvram.c
@@ -265,10 +265,16 @@ static ssize_t nvram_write(struct file *
 	unsigned char contents[NVRAM_BYTES];
 	unsigned i = *ppos;
 	unsigned char *tmp;
-	int len;
 
-	len = (NVRAM_BYTES - i) < count ? (NVRAM_BYTES - i) : count;
-	if (copy_from_user(contents, buf, len))
+	if (i >= NVRAM_BYTES)
+		return 0;	/* Past EOF */
+
+	if (count > NVRAM_BYTES - i)
+		count = NVRAM_BYTES - i;
+	if (count > NVRAM_BYTES)
+		return -EFAULT;	/* Can't happen, but prove it to gcc */
+
+	if (copy_from_user(contents, buf, count))
 		return -EFAULT;
 
 	spin_lock_irq(&rtc_lock);
@@ -276,7 +282,7 @@ static ssize_t nvram_write(struct file *
 	if (!__nvram_check_checksum())
 		goto checksum_err;
 
-	for (tmp = contents; count-- > 0 && i < NVRAM_BYTES; ++i, ++tmp)
+	for (tmp = contents; count--; ++i, ++tmp)
 		__nvram_write_byte(*tmp, i);
 
 	__nvram_set_checksum();

[004/111] x86: Add memory modify constraints to xchg() and cmpxchg()

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: H. Peter Anvin <hpa@zytor.com>

commit 113fc5a6e8c2288619ff7e8187a6f556b7e0d372 upstream.

xchg() and cmpxchg() modify their memory operands, not merely read
them.  For some versions of gcc the "memory" clobber has apparently
dealt with the situation, but not for all.

Originally-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Glauber Costa <glommer@redhat.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Peter Palfrader <peter@palfrader.org>
Cc: Greg KH <gregkh@suse.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Zachary Amsden <zamsden@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
LKML-Reference: <4C4F7277.8050306@zytor.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/x86/include/asm/cmpxchg_32.h |   86 ++++++++++++++++++-------------------
 arch/x86/include/asm/cmpxchg_64.h |   88 ++++++++++++++++++++------------------
 2 files changed, 89 insertions(+), 85 deletions(-)

--- a/arch/x86/include/asm/cmpxchg_32.h
+++ b/arch/x86/include/asm/cmpxchg_32.h
@@ -34,12 +34,12 @@ static inline void __set_64bit(unsigned
 			       unsigned int low, unsigned int high)
 {
 	asm volatile("\n1:\t"
-		     "movl (%0), %%eax\n\t"
-		     "movl 4(%0), %%edx\n\t"
-		     LOCK_PREFIX "cmpxchg8b (%0)\n\t"
+		     "movl (%1), %%eax\n\t"
+		     "movl 4(%1), %%edx\n\t"
+		     LOCK_PREFIX "cmpxchg8b %0\n\t"
 		     "jnz 1b"
-		     : /* no outputs */
-		     : "D"(ptr),
+		     : "=m"(*ptr)
+		     : "D" (ptr),
 		       "b"(low),
 		       "c"(high)
 		     : "ax", "dx", "memory");
@@ -82,20 +82,20 @@ static inline unsigned long __xchg(unsig
 	switch (size) {
 	case 1:
 		asm volatile("xchgb %b0,%1"
-			     : "=q" (x)
-			     : "m" (*__xg(ptr)), "0" (x)
+			     : "=q" (x), "+m" (*__xg(ptr))
+			     : "0" (x)
 			     : "memory");
 		break;
 	case 2:
 		asm volatile("xchgw %w0,%1"
-			     : "=r" (x)
-			     : "m" (*__xg(ptr)), "0" (x)
+			     : "=r" (x), "+m" (*__xg(ptr))
+			     : "0" (x)
 			     : "memory");
 		break;
 	case 4:
 		asm volatile("xchgl %0,%1"
-			     : "=r" (x)
-			     : "m" (*__xg(ptr)), "0" (x)
+			     : "=r" (x), "+m" (*__xg(ptr))
+			     : "0" (x)
 			     : "memory");
 		break;
 	}
@@ -139,21 +139,21 @@ static inline unsigned long __cmpxchg(vo
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		asm volatile(LOCK_PREFIX "cmpxchgb %b1,%2"
-			     : "=a"(prev)
-			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile(LOCK_PREFIX "cmpxchgb %b2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "q"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 2:
-		asm volatile(LOCK_PREFIX "cmpxchgw %w1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile(LOCK_PREFIX "cmpxchgw %w2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 4:
-		asm volatile(LOCK_PREFIX "cmpxchgl %1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile(LOCK_PREFIX "cmpxchgl %2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	}
@@ -172,21 +172,21 @@ static inline unsigned long __sync_cmpxc
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		asm volatile("lock; cmpxchgb %b1,%2"
-			     : "=a"(prev)
-			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("lock; cmpxchgb %b2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "q"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 2:
-		asm volatile("lock; cmpxchgw %w1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("lock; cmpxchgw %w2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 4:
-		asm volatile("lock; cmpxchgl %1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("lock; cmpxchgl %2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	}
@@ -200,21 +200,21 @@ static inline unsigned long __cmpxchg_lo
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		asm volatile("cmpxchgb %b1,%2"
-			     : "=a"(prev)
-			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("cmpxchgb %b2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "q"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 2:
-		asm volatile("cmpxchgw %w1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("cmpxchgw %w2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 4:
-		asm volatile("cmpxchgl %1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("cmpxchgl %2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	}
@@ -226,11 +226,10 @@ static inline unsigned long long __cmpxc
 					     unsigned long long new)
 {
 	unsigned long long prev;
-	asm volatile(LOCK_PREFIX "cmpxchg8b %3"
-		     : "=A"(prev)
+	asm volatile(LOCK_PREFIX "cmpxchg8b %1"
+		     : "=A"(prev), "+m" (*__xg(ptr))
 		     : "b"((unsigned long)new),
 		       "c"((unsigned long)(new >> 32)),
-		       "m"(*__xg(ptr)),
 		       "0"(old)
 		     : "memory");
 	return prev;
@@ -241,11 +240,10 @@ static inline unsigned long long __cmpxc
 						   unsigned long long new)
 {
 	unsigned long long prev;
-	asm volatile("cmpxchg8b %3"
-		     : "=A"(prev)
+	asm volatile("cmpxchg8b %1"
+		     : "=A"(prev), "+m"(*__xg(ptr))
 		     : "b"((unsigned long)new),
 		       "c"((unsigned long)(new >> 32)),
-		       "m"(*__xg(ptr)),
 		       "0"(old)
 		     : "memory");
 	return prev;
--- a/arch/x86/include/asm/cmpxchg_64.h
+++ b/arch/x86/include/asm/cmpxchg_64.h
@@ -26,26 +26,26 @@ static inline unsigned long __xchg(unsig
 	switch (size) {
 	case 1:
 		asm volatile("xchgb %b0,%1"
-			     : "=q" (x)
-			     : "m" (*__xg(ptr)), "0" (x)
+			     : "=q" (x), "+m" (*__xg(ptr))
+			     : "0" (x)
 			     : "memory");
 		break;
 	case 2:
 		asm volatile("xchgw %w0,%1"
-			     : "=r" (x)
-			     : "m" (*__xg(ptr)), "0" (x)
+			     : "=r" (x), "+m" (*__xg(ptr))
+			     : "0" (x)
 			     : "memory");
 		break;
 	case 4:
 		asm volatile("xchgl %k0,%1"
-			     : "=r" (x)
-			     : "m" (*__xg(ptr)), "0" (x)
+			     : "=r" (x), "+m" (*__xg(ptr))
+			     : "0" (x)
 			     : "memory");
 		break;
 	case 8:
 		asm volatile("xchgq %0,%1"
-			     : "=r" (x)
-			     : "m" (*__xg(ptr)), "0" (x)
+			     : "=r" (x), "+m" (*__xg(ptr))
+			     : "0" (x)
 			     : "memory");
 		break;
 	}
@@ -66,27 +66,27 @@ static inline unsigned long __cmpxchg(vo
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		asm volatile(LOCK_PREFIX "cmpxchgb %b1,%2"
-			     : "=a"(prev)
-			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile(LOCK_PREFIX "cmpxchgb %b2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "q"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 2:
-		asm volatile(LOCK_PREFIX "cmpxchgw %w1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile(LOCK_PREFIX "cmpxchgw %w2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 4:
-		asm volatile(LOCK_PREFIX "cmpxchgl %k1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile(LOCK_PREFIX "cmpxchgl %k2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 8:
-		asm volatile(LOCK_PREFIX "cmpxchgq %1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile(LOCK_PREFIX "cmpxchgq %2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	}
@@ -105,21 +105,27 @@ static inline unsigned long __sync_cmpxc
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		asm volatile("lock; cmpxchgb %b1,%2"
-			     : "=a"(prev)
-			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("lock; cmpxchgb %b2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "q"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 2:
-		asm volatile("lock; cmpxchgw %w1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("lock; cmpxchgw %w2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 4:
-		asm volatile("lock; cmpxchgl %1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("lock; cmpxchgl %k2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
+			     : "memory");
+		return prev;
+	case 8:
+		asm volatile("lock; cmpxchgq %2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	}
@@ -133,27 +139,27 @@ static inline unsigned long __cmpxchg_lo
 	unsigned long prev;
 	switch (size) {
 	case 1:
-		asm volatile("cmpxchgb %b1,%2"
-			     : "=a"(prev)
-			     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("cmpxchgb %b2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "q"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 2:
-		asm volatile("cmpxchgw %w1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("cmpxchgw %w2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 4:
-		asm volatile("cmpxchgl %k1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("cmpxchgl %k2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	case 8:
-		asm volatile("cmpxchgq %1,%2"
-			     : "=a"(prev)
-			     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+		asm volatile("cmpxchgq %2,%1"
+			     : "=a"(prev), "+m"(*__xg(ptr))
+			     : "r"(new), "0"(old)
 			     : "memory");
 		return prev;
 	}

[005/111] x86, vmware: Preset lpj values when on VMware.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Alok Kataria <akataria@vmware.com>

commit 9f242dc10e0c3c1eb32d8c83c18650a35fd7f80d upstream.

When running on VMware's platform, we have seen situations where
the AP's try to calibrate the lpj values and fail to get good calibration
runs becasue of timing issues. As a result delays don't work correctly
on all cpus.

The solutions is to set preset_lpj value based on the current tsc frequency
value. This is similar to what KVM does as well.

Signed-off-by: Alok N Kataria <akataria@vmware.com>
LKML-Reference: <1280790637.14933.29.camel@ank32.eng.vmware.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/x86/kernel/cpu/vmware.c |   10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

--- a/arch/x86/kernel/cpu/vmware.c
+++ b/arch/x86/kernel/cpu/vmware.c
@@ -22,6 +22,7 @@
  */
 
 #include <linux/dmi.h>
+#include <linux/jiffies.h>
 #include <asm/div64.h>
 #include <asm/vmware.h>
 #include <asm/x86_init.h>
@@ -50,7 +51,7 @@ static inline int __vmware_platform(void
 
 static unsigned long vmware_get_tsc_khz(void)
 {
-	uint64_t tsc_hz;
+	uint64_t tsc_hz, lpj;
 	uint32_t eax, ebx, ecx, edx;
 
 	VMWARE_PORT(GETHZ, eax, ebx, ecx, edx);
@@ -61,6 +62,13 @@ static unsigned long vmware_get_tsc_khz(
 	printk(KERN_INFO "TSC freq read from hypervisor : %lu.%03lu MHz\n",
 			 (unsigned long) tsc_hz / 1000,
 			 (unsigned long) tsc_hz % 1000);
+
+	if (!preset_lpj) {
+		lpj = ((u64)tsc_hz * 1000);
+		do_div(lpj, HZ);
+		preset_lpj = lpj;
+	}
+
 	return tsc_hz;
 }
 

[006/111] Staging: line6: needs to select SND_PCM

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Randy Dunlap <randy.dunlap@oracle.com>

commit e928c077e5cdcd72ee762125b37232aec1ff49f3 upstream.

line6 uses snd_pcm*() functions, so it should select SND_PCM.

ERROR: "snd_pcm_period_elapsed" [drivers/staging/line6/line6usb.ko] undefined!
ERROR: "snd_pcm_set_ops" [drivers/staging/line6/line6usb.ko] undefined!
ERROR: "snd_pcm_lib_free_pages" [drivers/staging/line6/line6usb.ko] undefined!
ERROR: "snd_pcm_lib_ioctl" [drivers/staging/line6/line6usb.ko] undefined!
ERROR: "snd_pcm_lib_malloc_pages" [drivers/staging/line6/line6usb.ko] undefined!
ERROR: "snd_pcm_hw_constraint_ratdens" [drivers/staging/line6/line6usb.ko] undefined!
ERROR: "snd_pcm_format_physical_width" [drivers/staging/line6/line6usb.ko] undefined!
ERROR: "snd_pcm_lib_preallocate_pages_for_all" [drivers/staging/line6/line6usb.ko] undefined!
ERROR: "snd_pcm_new" [drivers/staging/line6/line6usb.ko] undefined!

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Markus Grabner <grabner@icg.tugraz.at>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/staging/line6/Kconfig |    1 +
 1 file changed, 1 insertion(+)

--- a/drivers/staging/line6/Kconfig
+++ b/drivers/staging/line6/Kconfig
@@ -2,6 +2,7 @@ config LINE6_USB
 	tristate "Line6 USB support"
 	depends on USB && SND
 	select SND_RAWMIDI
+	select SND_PCM
 	help
 	  This is a driver for the guitar amp, cab, and effects modeller
 	  PODxt Pro by Line6 (and similar devices), supporting the

[007/111] Staging: panel: Prevent double-calling of parport_release - fix oops.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Peter Huewe <peterhuewe@gmx.de>

commit 060132ae42cce3f9d2fd34d9a17b98362b44b9f9 upstream.

This patch prevents the code from calling parport_release and
parport_unregister_device twice with the same arguments - and thus fixes an oops.

Rationale:
After the first call the parport is already released and the
handle isn't valid anymore and calling parport_release and
parport_unregister_device twice isn't a good idea.

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Acked-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/staging/panel/panel.c |    2 ++
 1 file changed, 2 insertions(+)

--- a/drivers/staging/panel/panel.c
+++ b/drivers/staging/panel/panel.c
@@ -2181,6 +2181,7 @@ int panel_init(void)
 		if (pprt) {
 			parport_release(pprt);
 			parport_unregister_device(pprt);
+			pprt = NULL;
 		}
 		parport_unregister_driver(&panel_driver);
 		printk(KERN_ERR "Panel driver version " PANEL_VERSION
@@ -2230,6 +2231,7 @@ static void __exit panel_cleanup_module(
 		/* TODO: free all input signals */
 		parport_release(pprt);
 		parport_unregister_device(pprt);
+		pprt = NULL;
 	}
 	parport_unregister_driver(&panel_driver);
 }

[008/111] PCI: Do not run NVidia quirks related to MSI with MSI disabled

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Rafael J. Wysocki <rjw@sisk.pl>

commit 3d2a531804d16cd8df6dbbb0429c6f143e756049 upstream.

There is no reason to run NVidia-specific quirks related to HT MSI
mappings with MSI disabled via pci=nomsi, so make
__nv_msi_ht_cap_quirk() return immediately in that case.

This allows at least one machine to boot 100% of the time with
pci=nomsi (it still doesn't boot reliably without that).

Addresses https://bugzilla.kernel.org/show_bug.cgi?id=16443 .

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/pci/quirks.c |    3 +++
 1 file changed, 3 insertions(+)

--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -2356,6 +2356,9 @@ static void __devinit __nv_msi_ht_cap_qu
 	int pos;
 	int found;
 
+	if (!pci_msi_enabled())
+		return;
+
 	/* check if there is HT MSI cap or enabled on this device */
 	found = ht_check_msi_mapping(dev);
 

[009/111] PCI: disable MSI on VIA K8M800

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Tejun Heo <tj@kernel.org>

commit 549e15611b4ac1de51ef0e0a79c2704f50a638a2 upstream.

MSI delivery from on-board ahci controller doesn't work on K8M800.  At
this point, it's unclear whether the culprit is with the ahci
controller or the host bridge.  Given the track record and considering
the rather minimal impact of MSI, disabling it seems reasonable.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Rainer Hurtado Navarro <publio.escipion.el.africano@gmail.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/pci/quirks.c |    1 +
 1 file changed, 1 insertion(+)

--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -2084,6 +2084,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AT
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_VT3336, quirk_disable_all_msi);
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_VT3351, quirk_disable_all_msi);
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_VT3364, quirk_disable_all_msi);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8380_0, quirk_disable_all_msi);
 
 /* Disable MSI on chipsets that are known to not support it */
 static void __devinit quirk_disable_msi(struct pci_dev *dev)

[010/111] solos-pci: Fix race condition in tasklet RX handling

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: David Woodhouse <David.Woodhouse@intel.com>

commit 1f6ea6e511e5ec730d8e88651da1b7b6e8fd1333 upstream.

We were seeing faults in the solos-pci receive tasklet when packets
arrived for a VCC which was currently being closed:

[18842.727906] EIP: [<e082f490>] br2684_push+0x19/0x234 [br2684] SS:ESP 0068:dfb89d14

[18845.090712] [<c13ecff3>] ? do_page_fault+0x0/0x2e1
[18845.120042] [<e082f490>] ? br2684_push+0x19/0x234 [br2684]
[18845.153530] [<e084fa13>] solos_bh+0x28b/0x7c8 [solos_pci]
[18845.186488] [<e084f711>] ? solos_irq+0x2d/0x51 [solos_pci]
[18845.219960] [<c100387b>] ? handle_irq+0x3b/0x48
[18845.247732] [<c10265cb>] ? irq_exit+0x34/0x57
[18845.274437] [<c1025720>] tasklet_action+0x42/0x69
[18845.303247] [<c102643f>] __do_softirq+0x8e/0x129
[18845.331540] [<c10264ff>] do_softirq+0x25/0x2a
[18845.358274] [<c102664c>] _local_bh_enable_ip+0x5e/0x6a
[18845.389677] [<c102666d>] local_bh_enable+0xb/0xe
[18845.417944] [<e08490a8>] ppp_unregister_channel+0x32/0xbb [ppp_generic]
[18845.458193] [<e08731ad>] pppox_unbind_sock+0x18/0x1f [pppox]

This patch uses an RCU-inspired approach to fix it. In the RX tasklet's
find_vcc() function we first refuse to use a VCC which already has the
ATM_VF_READY bit cleared. And in the VCC close function, we synchronise
with the tasklet to ensure that it can't still be using the VCC before
we continue and allow the VCC to be destroyed.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Tested-by: Nathan Williams <nathan@traverse.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/atm/solos-pci.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

--- a/drivers/atm/solos-pci.c
+++ b/drivers/atm/solos-pci.c
@@ -774,7 +774,8 @@ static struct atm_vcc *find_vcc(struct a
 	sk_for_each(s, node, head) {
 		vcc = atm_sk(s);
 		if (vcc->dev == dev && vcc->vci == vci &&
-		    vcc->vpi == vpi && vcc->qos.rxtp.traffic_class != ATM_NONE)
+		    vcc->vpi == vpi && vcc->qos.rxtp.traffic_class != ATM_NONE &&
+		    test_bit(ATM_VF_READY, &vcc->flags))
 			goto out;
 	}
 	vcc = NULL;
@@ -900,6 +901,10 @@ static void pclose(struct atm_vcc *vcc)
 	clear_bit(ATM_VF_ADDR, &vcc->flags);
 	clear_bit(ATM_VF_READY, &vcc->flags);
 
+	/* Hold up vcc_destroy_socket() (our caller) until solos_bh() in the
+	   tasklet has finished processing any incoming packets (and, more to
+	   the point, using the vcc pointer). */
+	tasklet_unlock_wait(&card->tlet);
 	return;
 }
 

[011/111] splice: fix misuse of SPLICE_F_NONBLOCK

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Miklos Szeredi <miklos@szeredi.hu>

commit 6965031d331a642e31278fa1b5bd47f372ffdd5d upstream.

SPLICE_F_NONBLOCK is clearly documented to only affect blocking on the
pipe.  In __generic_file_splice_read(), however, it causes an EAGAIN
if the page is currently being read.

This makes it impossible to write an application that only wants
failure if the pipe is full.  For example if the same process is
handling both ends of a pipe and isn't otherwise able to determine
whether a splice to the pipe will fill it or not.

We could make the read non-blocking on O_NONBLOCK or some other splice
flag, but for now this is the simplest fix.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/splice.c |   12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)

--- a/fs/splice.c
+++ b/fs/splice.c
@@ -365,17 +365,7 @@ __generic_file_splice_read(struct file *
 		 * If the page isn't uptodate, we may need to start io on it
 		 */
 		if (!PageUptodate(page)) {
-			/*
-			 * If in nonblock mode then dont block on waiting
-			 * for an in-flight io page
-			 */
-			if (flags & SPLICE_F_NONBLOCK) {
-				if (!trylock_page(page)) {
-					error = -EAGAIN;
-					break;
-				}
-			} else
-				lock_page(page);
+			lock_page(page);
 
 			/*
 			 * Page was truncated, or invalidated by the

[012/111] drivers/video/w100fb.c: ignore void return value / fix build failure

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Peter Huewe <peterhuewe@gmx.de>

commit fa260c00c1aa5c657793a7221e40d2400df5afd8 upstream.

Fix a build failure "error: void value not ignored as it ought to be"
by removing an assignment of a void return value.  The functionality of
the code is not changed.

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Acked-by: Henrik Kretzschmar <henne@nachtwindheim.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/video/w100fb.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/drivers/video/w100fb.c
+++ b/drivers/video/w100fb.c
@@ -857,9 +857,9 @@ unsigned long w100fb_gpio_read(int port)
 void w100fb_gpio_write(int port, unsigned long value)
 {
 	if (port==W100_GPIO_PORT_A)
-		value = writel(value, remapped_regs + mmGPIO_DATA);
+		writel(value, remapped_regs + mmGPIO_DATA);
 	else
-		value = writel(value, remapped_regs + mmGPIO_DATA2);
+		writel(value, remapped_regs + mmGPIO_DATA2);
 }
 EXPORT_SYMBOL(w100fb_gpio_read);
 EXPORT_SYMBOL(w100fb_gpio_write);

[013/111] ide-cd: Do not access completed requests in the irq handler

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Borislav Petkov <bp@alien8.de>

commit 110712828365ccafcc61a7f4db44c31ed4cf8793 upstream.

ide_cd_error_cmd() can complete an erroneous request with leftover
buffers. Signal this with its return value so that the request is not
accessed after its completion in the irq handler and we oops.

Signed-off-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/ide/ide-cd.c |   14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -506,15 +506,22 @@ int ide_cd_queue_pc(ide_drive_t *drive,
 	return (flags & REQ_FAILED) ? -EIO : 0;
 }
 
-static void ide_cd_error_cmd(ide_drive_t *drive, struct ide_cmd *cmd)
+/*
+ * returns true if rq has been completed
+ */
+static bool ide_cd_error_cmd(ide_drive_t *drive, struct ide_cmd *cmd)
 {
 	unsigned int nr_bytes = cmd->nbytes - cmd->nleft;
 
 	if (cmd->tf_flags & IDE_TFLAG_WRITE)
 		nr_bytes -= cmd->last_xfer_len;
 
-	if (nr_bytes > 0)
+	if (nr_bytes > 0) {
 		ide_complete_rq(drive, 0, nr_bytes);
+		return true;
+	}
+
+	return false;
 }
 
 static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
@@ -679,7 +686,8 @@ out_end:
 		}
 
 		if (uptodate == 0 && rq->bio)
-			ide_cd_error_cmd(drive, cmd);
+			if (ide_cd_error_cmd(drive, cmd))
+				return ide_stopped;
 
 		/* make sure it's fully ended */
 		if (blk_fs_request(rq) == 0) {

[014/111] md/raid10: fix deadlock with unaligned read during resync

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: NeilBrown <neilb@suse.de>

commit 51e9ac77035a3dfcb6fc0a88a0d80b6f99b5edb1 upstream.

If the 'bio_split' path in raid10-read is used while
resync/recovery is happening it is possible to deadlock.
Fix this be elevating ->nr_waiting for the duration of both
parts of the split request.

This fixes a bug that has been present since 2.6.22
but has only started manifesting recently for unknown reasons.
It is suitable for and -stable since then.

Reported-by:  Justin Bronder <jsbronder@gentoo.org>
Tested-by:  Justin Bronder <jsbronder@gentoo.org>
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/md/raid10.c |   18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -824,11 +824,29 @@ static int make_request(struct request_q
 		 */
 		bp = bio_split(bio,
 			       chunk_sects - (bio->bi_sector & (chunk_sects - 1)) );
+
+		/* Each of these 'make_request' calls will call 'wait_barrier'.
+		 * If the first succeeds but the second blocks due to the resync
+		 * thread raising the barrier, we will deadlock because the
+		 * IO to the underlying device will be queued in generic_make_request
+		 * and will never complete, so will never reduce nr_pending.
+		 * So increment nr_waiting here so no new raise_barriers will
+		 * succeed, and so the second wait_barrier cannot block.
+		 */
+		spin_lock_irq(&conf->resync_lock);
+		conf->nr_waiting++;
+		spin_unlock_irq(&conf->resync_lock);
+
 		if (make_request(q, &bp->bio1))
 			generic_make_request(&bp->bio1);
 		if (make_request(q, &bp->bio2))
 			generic_make_request(&bp->bio2);
 
+		spin_lock_irq(&conf->resync_lock);
+		conf->nr_waiting--;
+		wake_up(&conf->wait_barrier);
+		spin_unlock_irq(&conf->resync_lock);
+
 		bio_pair_release(bp);
 		return 0;
 	bad_map:

[015/111] blkdev: cgroup whitelist permission fix

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Chris Wright <chrisw@sous-sol.org>

commit b7300b78d1a87625975a799a109a2f98d77757c8 upstream.

The cgroup device whitelist code gets confused when trying to grant
permission to a disk partition that is not currently open.  Part of
blkdev_open() includes __blkdev_get() on the whole disk.

Basically, the only ways to reliably allow a cgroup access to a partition
on a block device when using the whitelist are to 1) also give it access
to the whole block device or 2) make sure the partition is already open in
a different context.

The patch avoids the cgroup check for the whole disk case when opening a
partition.

Addresses https://bugzilla.redhat.com/show_bug.cgi?id=589662

Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Acked-by: Serge E. Hallyn <serue@us.ibm.com>
Tested-by: Serge E. Hallyn <serue@us.ibm.com>
Reported-by: Vivek Goyal <vgoyal@redhat.com>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: "Daniel P. Berrange" <berrange@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/block_dev.c |   10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1175,10 +1175,12 @@ static int __blkdev_get(struct block_dev
 	/*
 	 * hooks: /n/, see "layering violations".
 	 */
-	ret = devcgroup_inode_permission(bdev->bd_inode, perm);
-	if (ret != 0) {
-		bdput(bdev);
-		return ret;
+	if (!for_part) {
+		ret = devcgroup_inode_permission(bdev->bd_inode, perm);
+		if (ret != 0) {
+			bdput(bdev);
+			return ret;
+		}
 	}
 
 	lock_kernel();

[016/111] eCryptfs: Handle ioctl calls with unlocked and compat functions

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Tyler Hicks <tyhicks@linux.vnet.ibm.com>

commit c43f7b8fb03be8bcc579bfc4e6ab70eac887ab55 upstream.

Lower filesystems that only implemented unlocked_ioctl weren't being
passed ioctl calls because eCryptfs only checked for
lower_file->f_op->ioctl and returned -ENOTTY if it was NULL.

eCryptfs shouldn't implement ioctl(), since it doesn't require the BKL.
This patch introduces ecryptfs_unlocked_ioctl() and
ecryptfs_compat_ioctl(), which passes the calls on to the lower file
system.

https://bugs.launchpad.net/ecryptfs/+bug/469664

Reported-by: James Dupin <james.dupin@gmail.com>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ecryptfs/file.c |   56 +++++++++++++++++++++++++++++++++--------------------
 1 file changed, 35 insertions(+), 21 deletions(-)

--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -293,12 +293,40 @@ static int ecryptfs_fasync(int fd, struc
 	return rc;
 }
 
-static int ecryptfs_ioctl(struct inode *inode, struct file *file,
-			  unsigned int cmd, unsigned long arg);
+static long
+ecryptfs_unlocked_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct file *lower_file = NULL;
+	long rc = -ENOTTY;
+
+	if (ecryptfs_file_to_private(file))
+		lower_file = ecryptfs_file_to_lower(file);
+	if (lower_file && lower_file->f_op && lower_file->f_op->unlocked_ioctl)
+		rc = lower_file->f_op->unlocked_ioctl(lower_file, cmd, arg);
+	return rc;
+}
+
+#ifdef CONFIG_COMPAT
+static long
+ecryptfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct file *lower_file = NULL;
+	long rc = -ENOIOCTLCMD;
+
+	if (ecryptfs_file_to_private(file))
+		lower_file = ecryptfs_file_to_lower(file);
+	if (lower_file && lower_file->f_op && lower_file->f_op->compat_ioctl)
+		rc = lower_file->f_op->compat_ioctl(lower_file, cmd, arg);
+	return rc;
+}
+#endif
 
 const struct file_operations ecryptfs_dir_fops = {
 	.readdir = ecryptfs_readdir,
-	.ioctl = ecryptfs_ioctl,
+	.unlocked_ioctl = ecryptfs_unlocked_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = ecryptfs_compat_ioctl,
+#endif
 	.mmap = generic_file_mmap,
 	.open = ecryptfs_open,
 	.flush = ecryptfs_flush,
@@ -315,7 +343,10 @@ const struct file_operations ecryptfs_ma
 	.write = do_sync_write,
 	.aio_write = generic_file_aio_write,
 	.readdir = ecryptfs_readdir,
-	.ioctl = ecryptfs_ioctl,
+	.unlocked_ioctl = ecryptfs_unlocked_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = ecryptfs_compat_ioctl,
+#endif
 	.mmap = generic_file_mmap,
 	.open = ecryptfs_open,
 	.flush = ecryptfs_flush,
@@ -324,20 +355,3 @@ const struct file_operations ecryptfs_ma
 	.fasync = ecryptfs_fasync,
 	.splice_read = generic_file_splice_read,
 };
-
-static int
-ecryptfs_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
-	       unsigned long arg)
-{
-	int rc = 0;
-	struct file *lower_file = NULL;
-
-	if (ecryptfs_file_to_private(file))
-		lower_file = ecryptfs_file_to_lower(file);
-	if (lower_file && lower_file->f_op && lower_file->f_op->ioctl)
-		rc = lower_file->f_op->ioctl(ecryptfs_inode_to_lower(inode),
-					     lower_file, cmd, arg);
-	else
-		rc = -ENOTTY;
-	return rc;
-}

[017/111] ecryptfs: release reference to lower mount if interpose fails

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Lino Sanfilippo <LinoSanfilippo@gmx.de>

commit 31f73bee3e170b7cabb35db9e2f4bf7919b9d036 upstream.

In ecryptfs_lookup_and_interpose_lower() the lower mount is not decremented
if allocation of a dentry info struct failed. As a result the lower filesystem
cant be unmounted any more (since it is considered busy). This patch corrects
the reference counting.

Signed-off-by: Lino Sanfilippo <LinoSanfilippo@gmx.de>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ecryptfs/inode.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -272,7 +272,7 @@ int ecryptfs_lookup_and_interpose_lower(
 		printk(KERN_ERR "%s: Out of memory whilst attempting "
 		       "to allocate ecryptfs_dentry_info struct\n",
 			__func__);
-		goto out_dput;
+		goto out_put;
 	}
 	ecryptfs_set_dentry_lower(ecryptfs_dentry, lower_dentry);
 	ecryptfs_set_dentry_lower_mnt(ecryptfs_dentry, lower_mnt);
@@ -345,8 +345,9 @@ int ecryptfs_lookup_and_interpose_lower(
 out_free_kmem:
 	kmem_cache_free(ecryptfs_header_cache_2, page_virt);
 	goto out;
-out_dput:
+out_put:
 	dput(lower_dentry);
+	mntput(lower_mnt);
 	d_drop(ecryptfs_dentry);
 out:
 	return rc;

[018/111] fs/ecryptfs/file.c: introduce missing free

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Julia Lawall <julia@diku.dk>

commit ceeab92971e8af05c1e81a4ff2c271124b55bb9b upstream.

The comments in the code indicate that file_info should be released if the
function fails.  This releasing is done at the label out_free, not out.

The semantic match that finds this problem is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@r exists@
local idexpression x;
statement S;
expression E;
identifier f,f1,l;
position p1,p2;
expression *ptr != NULL;
@@

x@p1 = kmem_cache_zalloc(...);
...
if (x == NULL) S
<... when != x
     when != if (...) { <+...x...+> }
(
x->f1 = E
|
 (x->f1 == NULL || ...)
|
 f(...,x->f1,...)
)
...>
(
 return <+...x...+>;
|
 return@p2 ...;
)

@script:python@
p1 << r.p1;
p2 << r.p2;
@@

print "* file: %s kmem_cache_zalloc %s" % (p1[0].file,p1[0].line)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ecryptfs/file.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -198,7 +198,7 @@ static int ecryptfs_open(struct inode *i
 			       "the persistent file for the dentry with name "
 			       "[%s]; rc = [%d]\n", __func__,
 			       ecryptfs_dentry->d_name.name, rc);
-			goto out;
+			goto out_free;
 		}
 	}
 	if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_RDONLY)
@@ -206,7 +206,7 @@ static int ecryptfs_open(struct inode *i
 		rc = -EPERM;
 		printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs "
 		       "file must hence be opened RO\n", __func__);
-		goto out;
+		goto out_free;
 	}
 	ecryptfs_set_file_lower(
 		file, ecryptfs_inode_to_private(inode)->lower_file);

[019/111] bio, fs: update RWA_MASK, READA and SWRITE to match the corresponding BIO_RW_* bits

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Tejun Heo <tj@kernel.org>

commit aca27ba9618276dd2f777bcd5a1419589ccf1ca8 upstream.

Commit a82afdf (block: use the same failfast bits for bio and request)
moved BIO_RW_* bits around such that they match up with REQ_* bits.
Unfortunately, fs.h hard coded RW_MASK, RWA_MASK, READ, WRITE, READA
and SWRITE as 0, 1, 2 and 3, and expected them to match with BIO_RW_*
bits.  READ/WRITE didn't change but BIO_RW_AHEAD was moved to bit 4
instead of bit 1, breaking RWA_MASK, READA and SWRITE.

This patch updates RWA_MASK, READA and SWRITE such that they match the
BIO_RW_* bits again.  A follow up patch will update the definitions to
directly use BIO_RW_* bits so that this kind of breakage won't happen
again.

Neil also spotted missing RWA_MASK conversion.

Stable: The offending commit a82afdf was released with v2.6.32, so
this patch should be applied to all kernels since then but it must
_NOT_ be applied to kernels earlier than that.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-and-bisected-by: Vladislav Bolkhovitin <vst@vlnb.net>
Root-caused-by: Neil Brown <neilb@suse.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 include/linux/fs.h |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -145,11 +145,11 @@ struct inodes_stat_t {
  *
  */
 #define RW_MASK		1
-#define RWA_MASK	2
+#define RWA_MASK		16
 #define READ 0
 #define WRITE 1
-#define READA 2		/* read-ahead  - don't block if no resources */
-#define SWRITE 3	/* for ll_rw_block() - wait for buffer lock */
+#define READA			16 /* readahead - don't block if no resources */
+#define SWRITE			17 /* for ll_rw_block(), wait for buffer lock */
 #define READ_SYNC	(READ | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG))
 #define READ_META	(READ | (1 << BIO_RW_META))
 #define WRITE_SYNC_PLUG	(WRITE | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_NOIDLE))

[020/111] signalfd: fill in ssi_int for posix timers and message queues

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Nathan Lynch <ntl@pobox.com>

commit a2a20c412c86e0bb46a9ab0dd31bcfe6d201b913 upstream.

If signalfd is used to consume a signal generated by a POSIX interval
timer or POSIX message queue, the ssi_int field does not reflect the data
(sigevent->sigev_value) supplied to timer_create(2) or mq_notify(3).  (The
ssi_ptr field, however, is filled in.)

This behavior differs from signalfd's treatment of sigqueue-generated
signals -- see the default case in signalfd_copyinfo.  It also gives
results that differ from the case when a signal is handled conventionally
via a sigaction-registered handler.

So, set signalfd_siginfo->ssi_int in the remaining cases (__SI_TIMER,
__SI_MESGQ) where ssi_ptr is set.

akpm: a non-back-compatible change.  Merge into -stable to minimise the
number of kernels which are in the field and which miss this feature.

Signed-off-by: Nathan Lynch <ntl@pobox.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/signalfd.c |    2 ++
 1 file changed, 2 insertions(+)

--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -87,6 +87,7 @@ static int signalfd_copyinfo(struct sign
 		 err |= __put_user(kinfo->si_tid, &uinfo->ssi_tid);
 		 err |= __put_user(kinfo->si_overrun, &uinfo->ssi_overrun);
 		 err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr);
+		 err |= __put_user(kinfo->si_int, &uinfo->ssi_int);
 		break;
 	case __SI_POLL:
 		err |= __put_user(kinfo->si_band, &uinfo->ssi_band);
@@ -110,6 +111,7 @@ static int signalfd_copyinfo(struct sign
 		err |= __put_user(kinfo->si_pid, &uinfo->ssi_pid);
 		err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid);
 		err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr);
+		err |= __put_user(kinfo->si_int, &uinfo->ssi_int);
 		break;
 	default:
 		/*

[021/111] smsc911x: Add spinlocks around registers access

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Catalin Marinas <catalin.marinas@arm.com>

commit 492c5d943d6a04b124ba3a719dc746dc36b14cfb upstream.

On SMP systems, the SMSC911x registers may be accessed by multiple CPUs
and this seems to put the chip in an inconsistent state. The patch adds
spinlocks to the smsc911x_reg_read, smsc911x_reg_write,
smsc911x_rx_readfifo and smsc911x_tx_writefifo functions.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/net/smsc911x.c |   92 +++++++++++++++++++++++++++----------------------
 1 file changed, 52 insertions(+), 40 deletions(-)

--- a/drivers/net/smsc911x.c
+++ b/drivers/net/smsc911x.c
@@ -85,8 +85,7 @@ struct smsc911x_data {
 	 */
 	spinlock_t mac_lock;
 
-	/* spinlock to ensure 16-bit accesses are serialised.
-	 * unused with a 32-bit bus */
+	/* spinlock to ensure register accesses are serialised */
 	spinlock_t dev_lock;
 
 	struct phy_device *phy_dev;
@@ -119,37 +118,33 @@ struct smsc911x_data {
 	unsigned int hashlo;
 };
 
-/* The 16-bit access functions are significantly slower, due to the locking
- * necessary.  If your bus hardware can be configured to do this for you
- * (in response to a single 32-bit operation from software), you should use
- * the 32-bit access functions instead. */
-
-static inline u32 smsc911x_reg_read(struct smsc911x_data *pdata, u32 reg)
+static inline u32 __smsc911x_reg_read(struct smsc911x_data *pdata, u32 reg)
 {
 	if (pdata->config.flags & SMSC911X_USE_32BIT)
 		return readl(pdata->ioaddr + reg);
 
-	if (pdata->config.flags & SMSC911X_USE_16BIT) {
-		u32 data;
-		unsigned long flags;
-
-		/* these two 16-bit reads must be performed consecutively, so
-		 * must not be interrupted by our own ISR (which would start
-		 * another read operation) */
-		spin_lock_irqsave(&pdata->dev_lock, flags);
-		data = ((readw(pdata->ioaddr + reg) & 0xFFFF) |
+	if (pdata->config.flags & SMSC911X_USE_16BIT)
+		return ((readw(pdata->ioaddr + reg) & 0xFFFF) |
 			((readw(pdata->ioaddr + reg + 2) & 0xFFFF) << 16));
-		spin_unlock_irqrestore(&pdata->dev_lock, flags);
-
-		return data;
-	}
 
 	BUG();
 	return 0;
 }
 
-static inline void smsc911x_reg_write(struct smsc911x_data *pdata, u32 reg,
-				      u32 val)
+static inline u32 smsc911x_reg_read(struct smsc911x_data *pdata, u32 reg)
+{
+	u32 data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&pdata->dev_lock, flags);
+	data = __smsc911x_reg_read(pdata, reg);
+	spin_unlock_irqrestore(&pdata->dev_lock, flags);
+
+	return data;
+}
+
+static inline void __smsc911x_reg_write(struct smsc911x_data *pdata, u32 reg,
+					u32 val)
 {
 	if (pdata->config.flags & SMSC911X_USE_32BIT) {
 		writel(val, pdata->ioaddr + reg);
@@ -157,44 +152,54 @@ static inline void smsc911x_reg_write(st
 	}
 
 	if (pdata->config.flags & SMSC911X_USE_16BIT) {
-		unsigned long flags;
-
-		/* these two 16-bit writes must be performed consecutively, so
-		 * must not be interrupted by our own ISR (which would start
-		 * another read operation) */
-		spin_lock_irqsave(&pdata->dev_lock, flags);
 		writew(val & 0xFFFF, pdata->ioaddr + reg);
 		writew((val >> 16) & 0xFFFF, pdata->ioaddr + reg + 2);
-		spin_unlock_irqrestore(&pdata->dev_lock, flags);
 		return;
 	}
 
 	BUG();
 }
 
+static inline void smsc911x_reg_write(struct smsc911x_data *pdata, u32 reg,
+				      u32 val)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&pdata->dev_lock, flags);
+	__smsc911x_reg_write(pdata, reg, val);
+	spin_unlock_irqrestore(&pdata->dev_lock, flags);
+}
+
 /* Writes a packet to the TX_DATA_FIFO */
 static inline void
 smsc911x_tx_writefifo(struct smsc911x_data *pdata, unsigned int *buf,
 		      unsigned int wordcount)
 {
+	unsigned long flags;
+
+	spin_lock_irqsave(&pdata->dev_lock, flags);
+
 	if (pdata->config.flags & SMSC911X_SWAP_FIFO) {
 		while (wordcount--)
-			smsc911x_reg_write(pdata, TX_DATA_FIFO, swab32(*buf++));
-		return;
+			__smsc911x_reg_write(pdata, TX_DATA_FIFO,
+					     swab32(*buf++));
+		goto out;
 	}
 
 	if (pdata->config.flags & SMSC911X_USE_32BIT) {
 		writesl(pdata->ioaddr + TX_DATA_FIFO, buf, wordcount);
-		return;
+		goto out;
 	}
 
 	if (pdata->config.flags & SMSC911X_USE_16BIT) {
 		while (wordcount--)
-			smsc911x_reg_write(pdata, TX_DATA_FIFO, *buf++);
-		return;
+			__smsc911x_reg_write(pdata, TX_DATA_FIFO, *buf++);
+		goto out;
 	}
 
 	BUG();
+out:
+	spin_unlock_irqrestore(&pdata->dev_lock, flags);
 }
 
 /* Reads a packet out of the RX_DATA_FIFO */
@@ -202,24 +207,31 @@ static inline void
 smsc911x_rx_readfifo(struct smsc911x_data *pdata, unsigned int *buf,
 		     unsigned int wordcount)
 {
+	unsigned long flags;
+
+	spin_lock_irqsave(&pdata->dev_lock, flags);
+
 	if (pdata->config.flags & SMSC911X_SWAP_FIFO) {
 		while (wordcount--)
-			*buf++ = swab32(smsc911x_reg_read(pdata, RX_DATA_FIFO));
-		return;
+			*buf++ = swab32(__smsc911x_reg_read(pdata,
+							    RX_DATA_FIFO));
+		goto out;
 	}
 
 	if (pdata->config.flags & SMSC911X_USE_32BIT) {
 		readsl(pdata->ioaddr + RX_DATA_FIFO, buf, wordcount);
-		return;
+		goto out;
 	}
 
 	if (pdata->config.flags & SMSC911X_USE_16BIT) {
 		while (wordcount--)
-			*buf++ = smsc911x_reg_read(pdata, RX_DATA_FIFO);
-		return;
+			*buf++ = __smsc911x_reg_read(pdata, RX_DATA_FIFO);
+		goto out;
 	}
 
 	BUG();
+out:
+	spin_unlock_irqrestore(&pdata->dev_lock, flags);
 }
 
 /* waits for MAC not busy, with timeout.  Only called by smsc911x_mac_read

[022/111] ARM: 6299/1: errata: TLBIASIDIS and TLBIMVAIS operations can broadcast a faulty ASID

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Will Deacon <will.deacon@arm.com>

commit cdf357f1e13a08a11261edacb3083746f65c1ed9 upstream.

On versions of the Cortex-A9 prior to r2p0, performing TLB invalidations by
ASID match can result in the incorrect ASID being broadcast to other CPUs.
As a consequence of this, the targetted TLB entries are not invalidated
across the system.

This workaround changes the TLB flushing routines to invalidate entries
regardless of the ASID.

Tested-by: Rob Clark <rob@ti.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/arm/Kconfig                |   12 ++++++++++++
 arch/arm/include/asm/tlbflush.h |    8 ++++++++
 2 files changed, 20 insertions(+)

--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -887,6 +887,18 @@ config ARM_ERRATA_460075
 	  ACTLR register. Note that setting specific bits in the ACTLR register
 	  may not be available in non-secure mode.
 
+config ARM_ERRATA_720789
+	bool "ARM errata: TLBIASIDIS and TLBIMVAIS operations can broadcast a faulty ASID"
+	depends on CPU_V7 && SMP
+	help
+	  This option enables the workaround for the 720789 Cortex-A9 (prior to
+	  r2p0) erratum. A faulty ASID can be sent to the other CPUs for the
+	  broadcasted CP15 TLB maintenance operations TLBIASIDIS and TLBIMVAIS.
+	  As a consequence of this erratum, some TLB entries which should be
+	  invalidated are not, resulting in an incoherency in the system page
+	  tables. The workaround changes the TLB flushing routines to invalidate
+	  entries regardless of the ASID.
+
 endmenu
 
 source "arch/arm/common/Kconfig"
--- a/arch/arm/include/asm/tlbflush.h
+++ b/arch/arm/include/asm/tlbflush.h
@@ -369,7 +369,11 @@ static inline void local_flush_tlb_mm(st
 	if (tlb_flag(TLB_V6_I_ASID))
 		asm("mcr p15, 0, %0, c8, c5, 2" : : "r" (asid) : "cc");
 	if (tlb_flag(TLB_V7_UIS_ASID))
+#ifdef CONFIG_ARM_ERRATA_720789
+		asm("mcr p15, 0, %0, c8, c3, 0" : : "r" (zero) : "cc");
+#else
 		asm("mcr p15, 0, %0, c8, c3, 2" : : "r" (asid) : "cc");
+#endif
 
 	if (tlb_flag(TLB_BTB)) {
 		/* flush the branch target cache */
@@ -409,7 +413,11 @@ local_flush_tlb_page(struct vm_area_stru
 	if (tlb_flag(TLB_V6_I_PAGE))
 		asm("mcr p15, 0, %0, c8, c5, 1" : : "r" (uaddr) : "cc");
 	if (tlb_flag(TLB_V7_UIS_PAGE))
+#ifdef CONFIG_ARM_ERRATA_720789
+		asm("mcr p15, 0, %0, c8, c3, 3" : : "r" (uaddr & PAGE_MASK) : "cc");
+#else
 		asm("mcr p15, 0, %0, c8, c3, 1" : : "r" (uaddr) : "cc");
+#endif
 
 	if (tlb_flag(TLB_BTB)) {
 		/* flush the branch target cache */

[023/111] ARM: 6280/1: imx: Fix build failure when including <mach/gpio.h> without <linux/spinlock.h>

[Greg KH]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1097 bytes --]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

commit 868003ca7ad17ac6c1606dc36101f10a7825b399 upstream.

This is a follow up to

	14cb0de (arm/imx/gpio: add spinlock protection)

and fixes the following build failure:

	  CC      arch/arm/mach-imx/pcm970-baseboard.o
	In file included from arch/arm/include/asm/gpio.h:6,
			 from include/linux/gpio.h:8,
			 from arch/arm/mach-imx/pcm970-baseboard.c:20:
	arch/arm/plat-mxc/include/mach/gpio.h:40: error: expected specifier-qualifier-list before 'spinlock_t'

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

---
 arch/arm/plat-mxc/include/mach/gpio.h |    1 +
 1 file changed, 1 insertion(+)

--- a/arch/arm/plat-mxc/include/mach/gpio.h
+++ b/arch/arm/plat-mxc/include/mach/gpio.h
@@ -19,6 +19,7 @@
 #ifndef __ASM_ARCH_MXC_GPIO_H__
 #define __ASM_ARCH_MXC_GPIO_H__
 
+#include <linux/spinlock.h>
 #include <mach/hardware.h>
 #include <asm-generic/gpio.h>
 

[024/111] USB: resizing usbmon binary interface buffer causes protection faults

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Steven Robertson <steven@strobe.cc>

commit 33d973ad88ceb83ed1449592b7574b5b5bb33ac6 upstream.

Enlarging the buffer size via the MON_IOCT_RING_SIZE ioctl causes
general protection faults. It appears the culprit is an incorrect
argument to mon_free_buff: instead of passing the size of the current
buffer being freed, the size of the new buffer is passed.

Use the correct size argument to mon_free_buff when changing the size of
the buffer.

Signed-off-by: Steven Robertson <steven@strobe.cc>
Acked-by: Pete Zaitcev <zaitcev@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/mon/mon_bin.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/usb/mon/mon_bin.c
+++ b/drivers/usb/mon/mon_bin.c
@@ -971,7 +971,7 @@ static int mon_bin_ioctl(struct inode *i
 
 		mutex_lock(&rp->fetch_lock);
 		spin_lock_irqsave(&rp->b_lock, flags);
-		mon_free_buff(rp->b_vec, size/CHUNK_SIZE);
+		mon_free_buff(rp->b_vec, rp->b_size/CHUNK_SIZE);
 		kfree(rp->b_vec);
 		rp->b_vec  = vec;
 		rp->b_size = size;

[025/111] USB delay init quirk for logitech Harmony 700-series devices

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Phil Dibowitz <phil@ipom.com>

commit 93362a875fc69881ae69299efaf19a55a1f57db0 upstream.

The Logitech Harmony 700 series needs an extra delay during
initialization.  This patch adds a USB quirk which enables such a delay
and adds the device to the quirks list.

Signed-off-by: Phil Dibowitz <phil@ipom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/core/hub.c     |    6 +++++-
 drivers/usb/core/quirks.c  |    3 +++
 include/linux/usb/quirks.h |    4 ++++
 3 files changed, 12 insertions(+), 1 deletion(-)

--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -22,6 +22,7 @@
 #include <linux/kthread.h>
 #include <linux/mutex.h>
 #include <linux/freezer.h>
+#include <linux/usb/quirks.h>
 
 #include <asm/uaccess.h>
 #include <asm/byteorder.h>
@@ -1768,7 +1769,6 @@ int usb_new_device(struct usb_device *ud
 	if (udev->parent)
 		usb_autoresume_device(udev->parent);
 
-	usb_detect_quirks(udev);
 	err = usb_enumerate_device(udev);	/* Read descriptors */
 	if (err < 0)
 		goto fail;
@@ -3063,6 +3063,10 @@ static void hub_port_connect_change(stru
 		if (status < 0)
 			goto loop;
 
+		usb_detect_quirks(udev);
+		if (udev->quirks & USB_QUIRK_DELAY_INIT)
+			msleep(1000);
+
 		/* consecutive bus-powered hubs aren't reliable; they can
 		 * violate the voltage drop budget.  if the new child has
 		 * a "powered" LED, users should notice we didn't enable it
--- a/drivers/usb/core/quirks.c
+++ b/drivers/usb/core/quirks.c
@@ -38,6 +38,9 @@ static const struct usb_device_id usb_qu
 	/* Creative SB Audigy 2 NX */
 	{ USB_DEVICE(0x041e, 0x3020), .driver_info = USB_QUIRK_RESET_RESUME },
 
+	/* Logitech Harmony 700-series */
+	{ USB_DEVICE(0x046d, 0xc122), .driver_info = USB_QUIRK_DELAY_INIT },
+
 	/* Philips PSC805 audio device */
 	{ USB_DEVICE(0x0471, 0x0155), .driver_info = USB_QUIRK_RESET_RESUME },
 
--- a/include/linux/usb/quirks.h
+++ b/include/linux/usb/quirks.h
@@ -19,4 +19,8 @@
 /* device can't handle its Configuration or Interface strings */
 #define USB_QUIRK_CONFIG_INTF_STRINGS	0x00000008
 
+/* device needs a pause during initialization, after we read the device
+   descriptor */
+#define USB_QUIRK_DELAY_INIT		0x00000040
+
 #endif /* __LINUX_USB_QUIRKS_H */

[026/111] USB: serial: enabling support for Segway RMP in ftdi_sio

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: John G. Rogers <jgrogers@gmail.com>

commit afad19648f70c6493193e0a774bd754b7790b4a0 upstream.

I have added the ProductID=0xe729 VendorID=FTDI_VID=0x0403 which will
enable support for the Segway Robotic Mobility Platform (RMP200) in the
ftdi_sio kernel module.  Currently, users of the Segway RMP200 must use
a RUN+="/sbin/modprobe -q ftdi-sio product=0xe729 vendor=0x0403 in a
udev rule to get the ftdi_sio module to handle the usb interface and
mount it on /dev/ttyXXX.  This is not a good solution because some users
will have multiple USB to Serial converters which will use the ftdi_sio
module.

Signed-off-by: John Rogers <jgrogers@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/serial/ftdi_sio.c     |    1 +
 drivers/usb/serial/ftdi_sio_ids.h |    5 +++++
 2 files changed, 6 insertions(+)

--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -752,6 +752,7 @@ static struct usb_device_id id_table_com
 		.driver_info = (kernel_ulong_t)&ftdi_jtag_quirk },
 	{ USB_DEVICE(FTDI_VID, XVERVE_SIGNALYZER_SH4_PID),
 		.driver_info = (kernel_ulong_t)&ftdi_jtag_quirk },
+	{ USB_DEVICE(FTDI_VID, SEGWAY_RMP200_PID) },
 	{ },					/* Optional parameter entry */
 	{ }					/* Terminating entry */
 };
--- a/drivers/usb/serial/ftdi_sio_ids.h
+++ b/drivers/usb/serial/ftdi_sio_ids.h
@@ -1039,3 +1039,8 @@
 #define XVERVE_SIGNALYZER_SH2_PID	0xBCA2
 #define XVERVE_SIGNALYZER_SH4_PID	0xBCA4
 
+/*
+ * Segway Robotic Mobility Platform USB interface (using VID 0x0403)
+ * Submitted by John G. Rogers
+ */
+#define SEGWAY_RMP200_PID	0xe729

[027/111] USB: option: Huawei ETS 1220 support added

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Pavel Kazlou <p.i.kazlou@gmail.com>

commit b972302b0a13aaddc9e90da2b4b52722e5d0e776 upstream.

The patch adds Huawei ETS 1220 product id into the list of supported
devices in 'option' usb serial driver.

Signed-off-by: Pavel Kazlou <p.i.kazlou@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/serial/option.c |    2 ++
 1 file changed, 2 insertions(+)

--- a/drivers/usb/serial/option.c
+++ b/drivers/usb/serial/option.c
@@ -166,6 +166,7 @@ static int  option_resume(struct usb_ser
 #define HUAWEI_PRODUCT_E143E			0x143E
 #define HUAWEI_PRODUCT_E143F			0x143F
 #define HUAWEI_PRODUCT_E14AC			0x14AC
+#define HUAWEI_PRODUCT_ETS1220			0x1803
 
 #define QUANTA_VENDOR_ID			0x0408
 #define QUANTA_PRODUCT_Q101			0xEA02
@@ -469,6 +470,7 @@ static struct usb_device_id option_ids[]
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143D, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143E, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143F, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_ETS1220, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E14AC) },
 	{ USB_DEVICE(AMOI_VENDOR_ID, AMOI_PRODUCT_9508) },
 	{ USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V640) }, /* Novatel Merlin V640/XV620 */

[028/111] USB: option: add huawei k3765 k4505 devices to work properly

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Andrew Bird <ajb@spheresystems.co.uk>

commit 0372a754be9aa43e19fd86c9bc04796d43b55e38 upstream.

This patch adds the product IDs of Huawei's K3765 and K4505 mobile
broadband usb modems to option.c. It also adds a quirk to the option
probe function so that binding to the device's network interface(class
0xff) is avoided. This is necessary to allow another driver to bind to
that, and to avoid programs like wvdial opening a nonfunctioning tty
during modem discovery.

Signed-off-by: Andrew Bird <ajb@spheresystems.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/serial/option.c |   11 +++++++++++
 1 file changed, 11 insertions(+)

--- a/drivers/usb/serial/option.c
+++ b/drivers/usb/serial/option.c
@@ -165,6 +165,8 @@ static int  option_resume(struct usb_ser
 #define HUAWEI_PRODUCT_E143D			0x143D
 #define HUAWEI_PRODUCT_E143E			0x143E
 #define HUAWEI_PRODUCT_E143F			0x143F
+#define HUAWEI_PRODUCT_K4505			0x1464
+#define HUAWEI_PRODUCT_K3765			0x1465
 #define HUAWEI_PRODUCT_E14AC			0x14AC
 #define HUAWEI_PRODUCT_ETS1220			0x1803
 
@@ -470,6 +472,8 @@ static struct usb_device_id option_ids[]
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143D, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143E, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143F, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_K4505, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_K3765, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_ETS1220, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E14AC) },
 	{ USB_DEVICE(AMOI_VENDOR_ID, AMOI_PRODUCT_9508) },
@@ -1009,6 +1013,13 @@ static int option_probe(struct usb_seria
 		serial->interface->cur_altsetting->desc.bInterfaceClass != 0xff)
 		return -ENODEV;
 
+	/* Don't bind network interfaces on Huawei K3765 & K4505 */
+	if (serial->dev->descriptor.idVendor == HUAWEI_VENDOR_ID &&
+		(serial->dev->descriptor.idProduct == HUAWEI_PRODUCT_K3765 ||
+			serial->dev->descriptor.idProduct == HUAWEI_PRODUCT_K4505) &&
+		serial->interface->cur_altsetting->desc.bInterfaceNumber == 1)
+		return -ENODEV;
+
 	data = serial->private = kzalloc(sizeof(struct option_intf_private), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;

[029/111] USB: ftdi_sio: device id for Navitator

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: dranch@trinnet.net <dranch@trinnet.net>

commit b6180ef7c99574c3350bbffa2a3a9d675321543d upstream.

This patch is to add a US Interface, Inc. "Navigator" USB device.
Specifically, it's a HAM Radio USB sound modem that also
incorporates three pairs of unique FTDI serial ports.  The standard
Linux FTDI serial driver will only recognize the first two serial
ports of an unknown FDTI derived device and this patch adds in
recognition to these specific new IDs.

Signed-off-by: David A. Ranch <dranch@trinnet.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/serial/ftdi_sio.c     |    3 +++
 drivers/usb/serial/ftdi_sio_ids.h |    5 +++++
 2 files changed, 8 insertions(+)

--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -162,6 +162,9 @@ static struct usb_device_id id_table_com
 	{ USB_DEVICE(FTDI_VID, FTDI_SCS_DEVICE_5_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_SCS_DEVICE_6_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_SCS_DEVICE_7_PID) },
+	{ USB_DEVICE(FTDI_VID, FTDI_USINT_CAT_PID) },
+	{ USB_DEVICE(FTDI_VID, FTDI_USINT_WKEY_PID) },
+	{ USB_DEVICE(FTDI_VID, FTDI_USINT_RS232_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_ACTZWAVE_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_IRTRANS_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_IPLUS_PID) },
--- a/drivers/usb/serial/ftdi_sio_ids.h
+++ b/drivers/usb/serial/ftdi_sio_ids.h
@@ -40,6 +40,11 @@
 
 #define FTDI_NXTCAM_PID		0xABB8 /* NXTCam for Mindstorms NXT */
 
+/* US Interface Navigator (http://www.usinterface.com/) */
+#define FTDI_USINT_CAT_PID	0xb810	/* Navigator CAT and 2nd PTT lines */
+#define FTDI_USINT_WKEY_PID	0xb811	/* Navigator WKEY and FSK lines */
+#define FTDI_USINT_RS232_PID	0xb812	/* Navigator RS232 and CONFIG lines */
+
 /* OOCDlink by Joern Kaipf <joernk@web.de>
  * (http://www.joernonline.de/dw/doku.php?id=start&idx=projects:oocdlink) */
 #define FTDI_OOCDLINK_PID	0xbaf8	/* Amontec JTAGkey */

[030/111] USB: cp210x: Add four new device IDs

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Alessio Igor Bogani <abogani@texware.it>

commit 356c5a4834a74c621715f7a7f16ded914eecbd3c upstream.

Signed-off-by: Alessio Igor Bogani <abogani@texware.it>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/serial/cp210x.c |    4 ++++
 1 file changed, 4 insertions(+)

--- a/drivers/usb/serial/cp210x.c
+++ b/drivers/usb/serial/cp210x.c
@@ -128,6 +128,10 @@ static struct usb_device_id id_table []
 	{ USB_DEVICE(0x1843, 0x0200) }, /* Vaisala USB Instrument Cable */
 	{ USB_DEVICE(0x18EF, 0xE00F) }, /* ELV USB-I2C-Interface */
 	{ USB_DEVICE(0x413C, 0x9500) }, /* DW700 GPS USB interface */
+	{ USB_DEVICE(0x16DC, 0x0010) }, /* W-IE-NE-R Plein & Baus GmbH PL512 Power Supply */
+	{ USB_DEVICE(0x16DC, 0x0011) }, /* W-IE-NE-R Plein & Baus GmbH RCM Remote Control for MARATON Power Supply */
+	{ USB_DEVICE(0x16DC, 0x0012) }, /* W-IE-NE-R Plein & Baus GmbH MPOD Multi Channel Power Supply */
+	{ USB_DEVICE(0x16DC, 0x0015) }, /* W-IE-NE-R Plein & Baus GmbH CML Control, Monitoring and Data Logger */
 	{ } /* Terminating Entry */
 };
 

[031/111] USB: usbtest: avoid to free coherent buffer in atomic context

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Ming Lei <tom.leiming@gmail.com>

commit e10e1bec8e6654de4591ef45ddd6a6d1e5b2591c upstream.

This patch fixes the warning below:
[30753.755998] ------------[ cut here ]------------
[30753.755998] WARNING: at /home/tom/git/linux-2.6/linux-2.6-next/arch/x86/include/asm/dma-mapping.h:155 hcd_buffer_free+0xb1/0xd4 [usbcore]()
[30753.755998] Hardware name: 6475EK2
[30753.755998] Modules linked in: uvcvideo ehci_hcd usbtest cdc_ether usbnet vfat fat usb_storage nfsd lockd nfs_acl auth_rpcgss exportfs mii tun videodev v4l1_compat v4l2_compat_ioctl32 fuse bridge stp llc sunrpc ipv6 cpufreq_ondemand acpi_cpufreq freq_table mperf kvm_intel kvm arc4 ecb ath5k usbhid mac80211 snd_hda_codec_conexant ch341 usbserial ath cfg80211 thinkpad_acpi snd_hda_intel pcspkr wmi hwmon yenta_socket iTCO_wdt iTCO_vendor_support i2c_i801 e1000e snd_hda_codec snd_hwdep snd_pcm snd_timer snd soundcore snd_page_alloc pata_acpi uhci_hcd ohci_hcd usbcore i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: uvcvideo]
[30753.755998] Pid: 0, comm: swapper Tainted: G        W   2.6.35-rc6-gkh-wl+ #49
[30753.755998] Call Trace:
[30753.755998]  <IRQ>  [<ffffffff8104478a>] warn_slowpath_common+0x80/0x98
[30753.755998]  [<ffffffff810447b7>] warn_slowpath_null+0x15/0x17
[30753.755998]  [<ffffffffa00ce02d>] hcd_buffer_free+0xb1/0xd4 [usbcore]
[30753.755998]  [<ffffffffa00c1345>] usb_free_coherent+0x1c/0x1e [usbcore]
[30753.755998]  [<ffffffffa00b13e4>] simple_free_urb+0x23/0x2f [usbtest]
[30753.755998]  [<ffffffffa00b210b>] iso_callback+0xbb/0x10f [usbtest]
[30753.755998]  [<ffffffffa00c7390>] usb_hcd_giveback_urb+0x8c/0xc0 [usbcore]
[30753.755998]  [<ffffffffa0449b35>] ehci_urb_done+0x84/0x95 [ehci_hcd]
[30753.755998]  [<ffffffffa044b5a5>] ehci_work+0x41a/0x7dd [ehci_hcd]
[30753.755998]  [<ffffffffa044e298>] ehci_irq+0x33b/0x370 [ehci_hcd]
[30753.755998]  [<ffffffff8100fb05>] ? sched_clock+0x9/0xd
[30753.755998]  [<ffffffff8105e641>] ? sched_clock_local+0x1c/0x82
[30753.755998]  [<ffffffff8105e76a>] ? sched_clock_cpu+0xc3/0xce
[30753.755998]  [<ffffffff81067c7e>] ? trace_hardirqs_off+0xd/0xf
[30753.755998]  [<ffffffff8105e7b8>] ? cpu_clock+0x43/0x5e
[30753.755998]  [<ffffffffa00c6999>] usb_hcd_irq+0x45/0xa1 [usbcore]
[30753.755998]  [<ffffffff81092e02>] handle_IRQ_event+0x20/0xa5
[30753.755998]  [<ffffffff81094cea>] handle_fasteoi_irq+0x92/0xd2
[30753.755998]  [<ffffffff8100c0ed>] handle_irq+0x1f/0x2a
[30753.755998]  [<ffffffff8100b75d>] do_IRQ+0x57/0xbe
[30753.755998]  [<ffffffff8136a693>] ret_from_intr+0x0/0x16
[30753.755998]  <EOI>  [<ffffffff81223baa>] ? acpi_idle_enter_bm+0x231/0x269
[30753.755998]  [<ffffffff81223ba3>] ? acpi_idle_enter_bm+0x22a/0x269
[30753.755998]  [<ffffffff812c4b6b>] cpuidle_idle_call+0x99/0xce
[30753.755998]  [<ffffffff81008dd5>] cpu_idle+0x61/0xaa
[30753.755998]  [<ffffffff8136374b>] start_secondary+0x1c2/0x1c6
[30753.755998] ---[ end trace 904cfaf7ab4cb1a2 ]---

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/misc/usbtest.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

--- a/drivers/usb/misc/usbtest.c
+++ b/drivers/usb/misc/usbtest.c
@@ -1382,7 +1382,6 @@ static void iso_callback (struct urb *ur
 			break;
 		}
 	}
-	simple_free_urb (urb);
 
 	ctx->pending--;
 	if (ctx->pending == 0) {
@@ -1499,6 +1498,7 @@ test_iso_queue (struct usbtest_dev *dev,
 			}
 
 			simple_free_urb (urbs [i]);
+			urbs[i] = NULL;
 			context.pending--;
 			context.submit_error = 1;
 			break;
@@ -1508,6 +1508,10 @@ test_iso_queue (struct usbtest_dev *dev,
 
 	wait_for_completion (&context.done);
 
+	for (i = 0; i < param->sglen; i++) {
+		if (urbs[i])
+			simple_free_urb(urbs[i]);
+	}
 	/*
 	 * Isochronous transfers are expected to fail sometimes.  As an
 	 * arbitrary limit, we will report an error if any submissions

[032/111] USB: fix thread-unsafe anchor utiliy routines

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Christian Lamparter <chunkeey@googlemail.com>

commit b3e670443b7fb8a2d29831b62b44a039c283e351 upstream.

This patch fixes a race condition in two utility routines
related to the removal/unlinking of urbs from an anchor.

If two threads are concurrently accessing the same anchor,
both could end up with the same urb - thinking they are
the exclusive owner.

Alan Stern pointed out a related issue in
usb_unlink_anchored_urbs:

"The URB isn't removed from the anchor until it completes
 (as a by-product of completion, in fact), which might not
 be for quite some time after the unlink call returns.
 In the meantime, the subroutine will keep trying to unlink
 it, over and over again."

Cc: Oliver Neukum <oneukum@suse.de>
Cc: Greg Kroah-Hartman <greg@kroah.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/usb/core/urb.c |   50 ++++++++++++++++++++-----------------------------
 1 file changed, 21 insertions(+), 29 deletions(-)

--- a/drivers/usb/core/urb.c
+++ b/drivers/usb/core/urb.c
@@ -137,6 +137,16 @@ void usb_anchor_urb(struct urb *urb, str
 }
 EXPORT_SYMBOL_GPL(usb_anchor_urb);
 
+/* Callers must hold anchor->lock */
+static void __usb_unanchor_urb(struct urb *urb, struct usb_anchor *anchor)
+{
+	urb->anchor = NULL;
+	list_del(&urb->anchor_list);
+	usb_put_urb(urb);
+	if (list_empty(&anchor->urb_list))
+		wake_up(&anchor->wait);
+}
+
 /**
  * usb_unanchor_urb - unanchors an URB
  * @urb: pointer to the urb to anchor
@@ -156,17 +166,14 @@ void usb_unanchor_urb(struct urb *urb)
 		return;
 
 	spin_lock_irqsave(&anchor->lock, flags);
-	if (unlikely(anchor != urb->anchor)) {
-		/* we've lost the race to another thread */
-		spin_unlock_irqrestore(&anchor->lock, flags);
-		return;
-	}
-	urb->anchor = NULL;
-	list_del(&urb->anchor_list);
+	/*
+	 * At this point, we could be competing with another thread which
+	 * has the same intention. To protect the urb from being unanchored
+	 * twice, only the winner of the race gets the job.
+	 */
+	if (likely(anchor == urb->anchor))
+		__usb_unanchor_urb(urb, anchor);
 	spin_unlock_irqrestore(&anchor->lock, flags);
-	usb_put_urb(urb);
-	if (list_empty(&anchor->urb_list))
-		wake_up(&anchor->wait);
 }
 EXPORT_SYMBOL_GPL(usb_unanchor_urb);
 
@@ -725,20 +732,11 @@ EXPORT_SYMBOL_GPL(usb_unpoison_anchored_
 void usb_unlink_anchored_urbs(struct usb_anchor *anchor)
 {
 	struct urb *victim;
-	unsigned long flags;
 
-	spin_lock_irqsave(&anchor->lock, flags);
-	while (!list_empty(&anchor->urb_list)) {
-		victim = list_entry(anchor->urb_list.prev, struct urb,
-				    anchor_list);
-		usb_get_urb(victim);
-		spin_unlock_irqrestore(&anchor->lock, flags);
-		/* this will unanchor the URB */
+	while ((victim = usb_get_from_anchor(anchor)) != NULL) {
 		usb_unlink_urb(victim);
 		usb_put_urb(victim);
-		spin_lock_irqsave(&anchor->lock, flags);
 	}
-	spin_unlock_irqrestore(&anchor->lock, flags);
 }
 EXPORT_SYMBOL_GPL(usb_unlink_anchored_urbs);
 
@@ -775,12 +773,11 @@ struct urb *usb_get_from_anchor(struct u
 		victim = list_entry(anchor->urb_list.next, struct urb,
 				    anchor_list);
 		usb_get_urb(victim);
-		spin_unlock_irqrestore(&anchor->lock, flags);
-		usb_unanchor_urb(victim);
+		__usb_unanchor_urb(victim, anchor);
 	} else {
-		spin_unlock_irqrestore(&anchor->lock, flags);
 		victim = NULL;
 	}
+	spin_unlock_irqrestore(&anchor->lock, flags);
 
 	return victim;
 }
@@ -802,12 +799,7 @@ void usb_scuttle_anchored_urbs(struct us
 	while (!list_empty(&anchor->urb_list)) {
 		victim = list_entry(anchor->urb_list.prev, struct urb,
 				    anchor_list);
-		usb_get_urb(victim);
-		spin_unlock_irqrestore(&anchor->lock, flags);
-		/* this may free the URB */
-		usb_unanchor_urb(victim);
-		usb_put_urb(victim);
-		spin_lock_irqsave(&anchor->lock, flags);
+		__usb_unanchor_urb(victim, anchor);
 	}
 	spin_unlock_irqrestore(&anchor->lock, flags);
 }

[033/111] drm/edid: Fix the HDTV hack sync adjustment

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Adam Jackson <ajax@redhat.com>

commit a4967de6cbb260ad0f6612a1d2035e119ef1578f upstream.

We're adjusting horizontal timings only here, moving vsync was just a
slavish translation of a typo in the X server.

Signed-off-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/gpu/drm/drm_edid.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -565,8 +565,8 @@ struct drm_display_mode *drm_mode_std(st
 		mode = drm_cvt_mode(dev, hsize, vsize, vrefresh_rate, 0, 0,
 				    false);
 		mode->hdisplay = 1366;
-		mode->vsync_start = mode->vsync_start - 1;
-		mode->vsync_end = mode->vsync_end - 1;
+		mode->hsync_start = mode->hsync_start - 1;
+		mode->hsync_end = mode->hsync_end - 1;
 		return mode;
 	}
 	mode = NULL;

[034/111] Bluetooth: Added support for controller shipped with iMac i5

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Cyril Lacoux <clacoux@ifeelgood.org>

commit 0a79f67445de50ca0a8dc1d34f3cc406d89c28b2 upstream.

Device class is ff(vend.) instead of e0(wlcon).

Output from command `usb-devices`:
T:  Bus=01 Lev=03 Prnt=03 Port=00 Cnt=01 Dev#=  6 Spd=12  MxCh= 0
D:  Ver= 2.00 Cls=ff(vend.) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=05ac ProdID=8215 Rev=01.82
S:  Manufacturer=Apple Inc.
S:  Product=Bluetooth USB Host Controller
S:  SerialNumber=7C6D62936607
C:  #Ifs= 4 Cfg#= 1 Atr=e0 MxPwr=0mA
I:  If#= 0 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=01 Prot=01 Driver=btusb
I:  If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
I:  If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
I:  If#= 3 Alt= 0 #EPs= 0 Cls=fe(app. ) Sub=01 Prot=00 Driver=(none)

Signed-off-by: Cyril Lacoux <clacoux@ifeelgood.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/bluetooth/btusb.c |    3 +++
 1 file changed, 3 insertions(+)

--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -59,6 +59,9 @@ static struct usb_device_id btusb_table[
 	/* Generic Bluetooth USB device */
 	{ USB_DEVICE_INFO(0xe0, 0x01, 0x01) },
 
+	/* Apple iMac11,1 */
+	{ USB_DEVICE(0x05ac, 0x8215) },
+
 	/* AVM BlueFRITZ! USB v2.0 */
 	{ USB_DEVICE(0x057c, 0x3800) },
 

[035/111] jfs: dont allow os2 xattr namespace overlap with others

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Dave Kleikamp <shaggy@linux.vnet.ibm.com>

commit aca0fa34bdaba39bfddddba8ca70dba4782e8fe6 upstream.

It's currently possible to bypass xattr namespace access rules by
prefixing valid xattr names with "os2.", since the os2 namespace stores
extended attributes in a legacy format with no prefix.

This patch adds checking to deny access to any valid namespace prefix
following "os2.".

Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Reported-by: Sergey Vlasov <vsu@altlinux.ru>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/jfs/xattr.c |   87 ++++++++++++++++++++++++---------------------------------
 1 file changed, 38 insertions(+), 49 deletions(-)

--- a/fs/jfs/xattr.c
+++ b/fs/jfs/xattr.c
@@ -85,46 +85,25 @@ struct ea_buffer {
 #define EA_MALLOC	0x0008
 
 
+static int is_known_namespace(const char *name)
+{
+	if (strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN) &&
+	    strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN) &&
+	    strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN) &&
+	    strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN))
+		return false;
+
+	return true;
+}
+
 /*
  * These three routines are used to recognize on-disk extended attributes
  * that are in a recognized namespace.  If the attribute is not recognized,
  * "os2." is prepended to the name
  */
-static inline int is_os2_xattr(struct jfs_ea *ea)
+static int is_os2_xattr(struct jfs_ea *ea)
 {
-	/*
-	 * Check for "system."
-	 */
-	if ((ea->namelen >= XATTR_SYSTEM_PREFIX_LEN) &&
-	    !strncmp(ea->name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
-		return false;
-	/*
-	 * Check for "user."
-	 */
-	if ((ea->namelen >= XATTR_USER_PREFIX_LEN) &&
-	    !strncmp(ea->name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN))
-		return false;
-	/*
-	 * Check for "security."
-	 */
-	if ((ea->namelen >= XATTR_SECURITY_PREFIX_LEN) &&
-	    !strncmp(ea->name, XATTR_SECURITY_PREFIX,
-		     XATTR_SECURITY_PREFIX_LEN))
-		return false;
-	/*
-	 * Check for "trusted."
-	 */
-	if ((ea->namelen >= XATTR_TRUSTED_PREFIX_LEN) &&
-	    !strncmp(ea->name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN))
-		return false;
-	/*
-	 * Add any other valid namespace prefixes here
-	 */
-
-	/*
-	 * We assume it's OS/2's flat namespace
-	 */
-	return true;
+	return !is_known_namespace(ea->name);
 }
 
 static inline int name_size(struct jfs_ea *ea)
@@ -762,13 +741,23 @@ static int can_set_xattr(struct inode *i
 	if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
 		return can_set_system_xattr(inode, name, value, value_len);
 
+	if (!strncmp(name, XATTR_OS2_PREFIX, XATTR_OS2_PREFIX_LEN)) {
+		/*
+		 * This makes sure that we aren't trying to set an
+		 * attribute in a different namespace by prefixing it
+		 * with "os2."
+		 */
+		if (is_known_namespace(name + XATTR_OS2_PREFIX_LEN))
+				return -EOPNOTSUPP;
+		return 0;
+	}
+
 	/*
 	 * Don't allow setting an attribute in an unknown namespace.
 	 */
 	if (strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) &&
 	    strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN) &&
-	    strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN) &&
-	    strncmp(name, XATTR_OS2_PREFIX, XATTR_OS2_PREFIX_LEN))
+	    strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN))
 		return -EOPNOTSUPP;
 
 	return 0;
@@ -950,19 +939,8 @@ ssize_t __jfs_getxattr(struct inode *ino
 	int xattr_size;
 	ssize_t size;
 	int namelen = strlen(name);
-	char *os2name = NULL;
 	char *value;
 
-	if (strncmp(name, XATTR_OS2_PREFIX, XATTR_OS2_PREFIX_LEN) == 0) {
-		os2name = kmalloc(namelen - XATTR_OS2_PREFIX_LEN + 1,
-				  GFP_KERNEL);
-		if (!os2name)
-			return -ENOMEM;
-		strcpy(os2name, name + XATTR_OS2_PREFIX_LEN);
-		name = os2name;
-		namelen -= XATTR_OS2_PREFIX_LEN;
-	}
-
 	down_read(&JFS_IP(inode)->xattr_sem);
 
 	xattr_size = ea_get(inode, &ea_buf, 0);
@@ -1000,8 +978,6 @@ ssize_t __jfs_getxattr(struct inode *ino
       out:
 	up_read(&JFS_IP(inode)->xattr_sem);
 
-	kfree(os2name);
-
 	return size;
 }
 
@@ -1010,6 +986,19 @@ ssize_t jfs_getxattr(struct dentry *dent
 {
 	int err;
 
+	if (strncmp(name, XATTR_OS2_PREFIX, XATTR_OS2_PREFIX_LEN) == 0) {
+		/*
+		 * skip past "os2." prefix
+		 */
+		name += XATTR_OS2_PREFIX_LEN;
+		/*
+		 * Don't allow retrieving properly prefixed attributes
+		 * by prepending them with "os2."
+		 */
+		if (is_known_namespace(name))
+			return -EOPNOTSUPP;
+	}
+
 	err = __jfs_getxattr(dentry->d_inode, name, data, buf_size);
 
 	return err;

[036/111] arp_notify: allow drivers to explicitly request a notification event.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Ian Campbell <Ian.Campbell@citrix.com>

commit 06c4648d46d1b757d6b9591a86810be79818b60c upstream.

Currently such notifications are only generated when the device comes up or the
address changes. However one use case for these notifications is to enable
faster network recovery after a virtual machine migration (by causing switches
to relearn their MAC tables). A migration appears to the network stack as a
temporary loss of carrier and therefore does not trigger either of the current
conditions. Rather than adding carrier up as a trigger (which can cause issues
when interfaces a flapping) simply add an interface which the driver can use
to explicitly trigger the notification.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Stephen Hemminger <shemminger@linux-foundation.org>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 include/linux/netdevice.h |    2 ++
 include/linux/notifier.h  |    1 +
 net/ipv4/devinet.c        |    1 +
 net/sched/sch_generic.c   |   18 ++++++++++++++++++
 4 files changed, 22 insertions(+)

--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1560,6 +1560,8 @@ extern void netif_carrier_on(struct net_
 
 extern void netif_carrier_off(struct net_device *dev);
 
+extern void netif_notify_peers(struct net_device *dev);
+
 /**
  *	netif_dormant_on - mark device as dormant.
  *	@dev: network device
--- a/include/linux/notifier.h
+++ b/include/linux/notifier.h
@@ -201,6 +201,7 @@ static inline int notifier_to_errno(int
 #define NETDEV_PRE_UP		0x000D
 #define NETDEV_BONDING_OLDTYPE  0x000E
 #define NETDEV_BONDING_NEWTYPE  0x000F
+#define NETDEV_NOTIFY_PEERS	0x0012
 
 #define SYS_DOWN	0x0001	/* Notify of system down */
 #define SYS_RESTART	SYS_DOWN
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1076,6 +1076,7 @@ static int inetdev_event(struct notifier
 		}
 		ip_mc_up(in_dev);
 		/* fall through */
+	case NETDEV_NOTIFY_PEERS:
 	case NETDEV_CHANGEADDR:
 		/* Send gratuitous ARP to notify of link change */
 		if (IN_DEV_ARP_NOTIFY(in_dev)) {
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -330,6 +330,24 @@ void netif_carrier_off(struct net_device
 }
 EXPORT_SYMBOL(netif_carrier_off);
 
+/**
+ * 	netif_notify_peers - notify network peers about existence of @dev
+ * 	@dev: network device
+ *
+ * Generate traffic such that interested network peers are aware of
+ * @dev, such as by generating a gratuitous ARP. This may be used when
+ * a device wants to inform the rest of the network about some sort of
+ * reconfiguration such as a failover event or virtual machine
+ * migration.
+ */
+void netif_notify_peers(struct net_device *dev)
+{
+	rtnl_lock();
+	call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, dev);
+	rtnl_unlock();
+}
+EXPORT_SYMBOL(netif_notify_peers);
+
 /* "NOOP" scheduler: the best scheduler, recommended for all interfaces
    under all circumstances. It is difficult to invent anything faster or
    cheaper.

[037/111] xen: netfront: explicitly generate arp_notify event after migration.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Ian Campbell <Ian.Campbell@citrix.com>

commit 592970675c9522bde588b945388c7995c8b51328 upstream.

Use newly introduced netif_notify_peers() method to ensure a gratuitous ARP is
generated after a migration.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Stephen Hemminger <shemminger@linux-foundation.org>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: xen-devel@lists.xensource.com
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/net/xen-netfront.c |    1 +
 1 file changed, 1 insertion(+)

--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -1619,6 +1619,7 @@ static void backend_changed(struct xenbu
 		if (xennet_connect(netdev) != 0)
 			break;
 		xenbus_switch_state(dev, XenbusStateConnected);
+		netif_notify_peers(netdev);
 		break;
 
 	case XenbusStateClosing:

[038/111] net: Fix NETDEV_NOTIFY_PEERS to not conflict with NETDEV_BONDING_DESLAVE.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: David S. Miller <davem@davemloft.net>

commit 38117d1495e587fbb10d6e55733139a27893cef5 upstream.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 include/linux/notifier.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/include/linux/notifier.h
+++ b/include/linux/notifier.h
@@ -201,7 +201,7 @@ static inline int notifier_to_errno(int
 #define NETDEV_PRE_UP		0x000D
 #define NETDEV_BONDING_OLDTYPE  0x000E
 #define NETDEV_BONDING_NEWTYPE  0x000F
-#define NETDEV_NOTIFY_PEERS	0x0012
+#define NETDEV_NOTIFY_PEERS	0x0013
 
 #define SYS_DOWN	0x0001	/* Notify of system down */
 #define SYS_RESTART	SYS_DOWN

[039/111] irq: Add new IRQ flag IRQF_NO_SUSPEND

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Ian Campbell <ian.campbell@citrix.com>

commit 685fd0b4ea3f0f1d5385610b0d5b57775a8d5842 upstream.

A small number of users of IRQF_TIMER are using it for the implied no
suspend behaviour on interrupts which are not timer interrupts.

Therefore add a new IRQF_NO_SUSPEND flag, rename IRQF_TIMER to
__IRQF_TIMER and redefine IRQF_TIMER in terms of these new flags.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: xen-devel@lists.xensource.com
Cc: linux-input@vger.kernel.org
Cc: linuxppc-dev@ozlabs.org
Cc: devicetree-discuss@lists.ozlabs.org
LKML-Reference: <1280398595-29708-1-git-send-email-ian.campbell@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 include/linux/interrupt.h |    7 ++++++-
 kernel/irq/manage.c       |    2 +-
 2 files changed, 7 insertions(+), 2 deletions(-)

--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -52,16 +52,21 @@
  * IRQF_ONESHOT - Interrupt is not reenabled after the hardirq handler finished.
  *                Used by threaded interrupts which need to keep the
  *                irq line disabled until the threaded handler has been run.
+ * IRQF_NO_SUSPEND - Do not disable this IRQ during suspend
+ *
  */
 #define IRQF_DISABLED		0x00000020
 #define IRQF_SAMPLE_RANDOM	0x00000040
 #define IRQF_SHARED		0x00000080
 #define IRQF_PROBE_SHARED	0x00000100
-#define IRQF_TIMER		0x00000200
+#define __IRQF_TIMER		0x00000200
 #define IRQF_PERCPU		0x00000400
 #define IRQF_NOBALANCING	0x00000800
 #define IRQF_IRQPOLL		0x00001000
 #define IRQF_ONESHOT		0x00002000
+#define IRQF_NO_SUSPEND		0x00004000
+
+#define IRQF_TIMER		(__IRQF_TIMER | IRQF_NO_SUSPEND)
 
 /*
  * Bits used by threaded handlers:
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -200,7 +200,7 @@ static inline int setup_affinity(unsigne
 void __disable_irq(struct irq_desc *desc, unsigned int irq, bool suspend)
 {
 	if (suspend) {
-		if (!desc->action || (desc->action->flags & IRQF_TIMER))
+		if (!desc->action || (desc->action->flags & IRQF_NO_SUSPEND))
 			return;
 		desc->status |= IRQ_SUSPENDED;
 	}

[040/111] xen: Do not suspend IPI IRQs.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Ian Campbell <ian.campbell@citrix.com>

commit 4877c737283813bdb4bebfa3168c1585f6e3a8ca upstream.

In general the semantics of IPIs are that they are are expected to
continue functioning after dpm_suspend_noirq().

Specifically I have seen a deadlock between the callfunc IPI and the
stop machine used by xen's do_suspend() routine. If one CPU has already
called dpm_suspend_noirq() then there is a window where it can be sent
a callfunc IPI before all the other CPUs have entered stop_cpu().

If this happens then the first CPU ends up spinning in stop_cpu()
waiting for the other to rendezvous in state STOPMACHINE_PREPARE while
the other is spinning in csd_lock_wait().

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: xen-devel@lists.xensource.com
LKML-Reference: <1280398595-29708-4-git-send-email-ian.campbell@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/xen/events.c |    1 +
 1 file changed, 1 insertion(+)

--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -535,6 +535,7 @@ int bind_ipi_to_irqhandler(enum ipi_vect
 	if (irq < 0)
 		return irq;
 
+	irqflags |= IRQF_NO_SUSPEND;
 	retval = request_irq(irq, handler, irqflags, devname, dev_id);
 	if (retval != 0) {
 		unbind_from_irq(irq);

[041/111] ext4: fix freeze deadlock under IO

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Eric Sandeen <sandeen@sandeen.net>

commit 437f88cc031ffe7f37f3e705367f4fe1f4be8b0f upstream.

Commit 6b0310fbf087ad6 caused a regression resulting in deadlocks
when freezing a filesystem which had active IO; the vfs_check_frozen
level (SB_FREEZE_WRITE) did not let the freeze-related IO syncing
through.  Duh.

Changing the test to FREEZE_TRANS should let the normal freeze
syncing get through the fs, but still block any transactions from
starting once the fs is completely frozen.

I tested this by running fsstress in the background while periodically
snapshotting the fs and running fsck on the result.  I ran into
occasional deadlocks, but different ones.  I think this is a
fine fix for the problem at hand, and the other deadlocky things
will need more investigation.

Reported-by: Phillip Susi <psusi@cfl.rr.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ext4/super.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -227,7 +227,7 @@ handle_t *ext4_journal_start_sb(struct s
 	if (sb->s_flags & MS_RDONLY)
 		return ERR_PTR(-EROFS);
 
-	vfs_check_frozen(sb, SB_FREEZE_WRITE);
+	vfs_check_frozen(sb, SB_FREEZE_TRANS);
 	/* Special case here: if the journal has aborted behind our
 	 * backs (eg. EIO in the commit thread), then we still need to
 	 * take the FS itself readonly cleanly. */
@@ -3390,7 +3390,7 @@ int ext4_force_commit(struct super_block
 
 	journal = EXT4_SB(sb)->s_journal;
 	if (journal) {
-		vfs_check_frozen(sb, SB_FREEZE_WRITE);
+		vfs_check_frozen(sb, SB_FREEZE_TRANS);
 		ret = ext4_journal_force_commit(journal);
 	}
 

[042/111] drm/i915: Use RSEN instead of HTPLG for tfp410 monitor detection.

[Greg KH]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 835 bytes --]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Dave Müller <dave.mueller@gmx.ch>

commit f458823b864c6def488f951a79986fa205aba4f1 upstream.

Presence detection of a digital monitor seems not to be reliable using
the HTPLG bit.

Dave Müller <dave.mueller@gmx.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/gpu/drm/i915/dvo_tfp410.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/gpu/drm/i915/dvo_tfp410.c
+++ b/drivers/gpu/drm/i915/dvo_tfp410.c
@@ -214,7 +214,7 @@ static enum drm_connector_status tfp410_
 	uint8_t ctl2;
 
 	if (tfp410_readb(dvo, TFP410_CTL_2, &ctl2)) {
-		if (ctl2 & TFP410_CTL_2_HTPLG)
+		if (ctl2 & TFP410_CTL_2_RSEN)
 			ret = connector_status_connected;
 		else
 			ret = connector_status_disconnected;

[043/111] Btrfs: Avoid superfluous tree-log writeout

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 8cef4e160d74920ad1725f58c89fd75ec4c4ac38 upstream.

We allow two log transactions at a time, but use same flag
to mark dirty tree-log btree blocks. So we may flush dirty
blocks belonging to newer log transaction when committing a
log transaction. This patch fixes the issue by using two
flags to mark dirty tree-log btree blocks.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/disk-io.c     |    6 +++---
 fs/btrfs/extent-tree.c |   12 ++++++++++--
 fs/btrfs/transaction.c |   21 +++++++++++----------
 fs/btrfs/transaction.h |    6 +++---
 fs/btrfs/tree-log.c    |   33 ++++++++++++++++++++-------------
 5 files changed, 47 insertions(+), 31 deletions(-)

--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -980,12 +980,12 @@ int btrfs_free_log_root_tree(struct btrf
 
 	while (1) {
 		ret = find_first_extent_bit(&log_root_tree->dirty_log_pages,
-				    0, &start, &end, EXTENT_DIRTY);
+				0, &start, &end, EXTENT_DIRTY | EXTENT_NEW);
 		if (ret)
 			break;
 
-		clear_extent_dirty(&log_root_tree->dirty_log_pages,
-				   start, end, GFP_NOFS);
+		clear_extent_bits(&log_root_tree->dirty_log_pages, start, end,
+				  EXTENT_DIRTY | EXTENT_NEW, GFP_NOFS);
 	}
 	eb = fs_info->log_root_tree->node;
 
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -4919,8 +4919,16 @@ struct extent_buffer *btrfs_init_new_buf
 	btrfs_set_buffer_uptodate(buf);
 
 	if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
-		set_extent_dirty(&root->dirty_log_pages, buf->start,
-			 buf->start + buf->len - 1, GFP_NOFS);
+		/*
+		 * we allow two log transactions at a time, use different
+		 * EXENT bit to differentiate dirty pages.
+		 */
+		if (root->log_transid % 2 == 0)
+			set_extent_dirty(&root->dirty_log_pages, buf->start,
+					buf->start + buf->len - 1, GFP_NOFS);
+		else
+			set_extent_new(&root->dirty_log_pages, buf->start,
+					buf->start + buf->len - 1, GFP_NOFS);
 	} else {
 		set_extent_dirty(&trans->transaction->dirty_pages, buf->start,
 			 buf->start + buf->len - 1, GFP_NOFS);
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -354,7 +354,7 @@ int btrfs_end_transaction_throttle(struc
  * those extents are sent to disk but does not wait on them
  */
 int btrfs_write_marked_extents(struct btrfs_root *root,
-			       struct extent_io_tree *dirty_pages)
+			       struct extent_io_tree *dirty_pages, int mark)
 {
 	int ret;
 	int err = 0;
@@ -367,7 +367,7 @@ int btrfs_write_marked_extents(struct bt
 
 	while (1) {
 		ret = find_first_extent_bit(dirty_pages, start, &start, &end,
-					    EXTENT_DIRTY);
+					    mark);
 		if (ret)
 			break;
 		while (start <= end) {
@@ -413,7 +413,7 @@ int btrfs_write_marked_extents(struct bt
  * on all the pages and clear them from the dirty pages state tree
  */
 int btrfs_wait_marked_extents(struct btrfs_root *root,
-			      struct extent_io_tree *dirty_pages)
+			      struct extent_io_tree *dirty_pages, int mark)
 {
 	int ret;
 	int err = 0;
@@ -425,12 +425,12 @@ int btrfs_wait_marked_extents(struct btr
 	unsigned long index;
 
 	while (1) {
-		ret = find_first_extent_bit(dirty_pages, 0, &start, &end,
-					    EXTENT_DIRTY);
+		ret = find_first_extent_bit(dirty_pages, start, &start, &end,
+					    mark);
 		if (ret)
 			break;
 
-		clear_extent_dirty(dirty_pages, start, end, GFP_NOFS);
+		clear_extent_bits(dirty_pages, start, end, mark, GFP_NOFS);
 		while (start <= end) {
 			index = start >> PAGE_CACHE_SHIFT;
 			start = (u64)(index + 1) << PAGE_CACHE_SHIFT;
@@ -460,13 +460,13 @@ int btrfs_wait_marked_extents(struct btr
  * those extents are on disk for transaction or log commit
  */
 int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
-					struct extent_io_tree *dirty_pages)
+				struct extent_io_tree *dirty_pages, int mark)
 {
 	int ret;
 	int ret2;
 
-	ret = btrfs_write_marked_extents(root, dirty_pages);
-	ret2 = btrfs_wait_marked_extents(root, dirty_pages);
+	ret = btrfs_write_marked_extents(root, dirty_pages, mark);
+	ret2 = btrfs_wait_marked_extents(root, dirty_pages, mark);
 	return ret || ret2;
 }
 
@@ -479,7 +479,8 @@ int btrfs_write_and_wait_transaction(str
 		return filemap_write_and_wait(btree_inode->i_mapping);
 	}
 	return btrfs_write_and_wait_marked_extents(root,
-					   &trans->transaction->dirty_pages);
+					   &trans->transaction->dirty_pages,
+					   EXTENT_DIRTY);
 }
 
 /*
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -107,10 +107,10 @@ void btrfs_throttle(struct btrfs_root *r
 int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root);
 int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
-					struct extent_io_tree *dirty_pages);
+				struct extent_io_tree *dirty_pages, int mark);
 int btrfs_write_marked_extents(struct btrfs_root *root,
-					struct extent_io_tree *dirty_pages);
+				struct extent_io_tree *dirty_pages, int mark);
 int btrfs_wait_marked_extents(struct btrfs_root *root,
-					struct extent_io_tree *dirty_pages);
+				struct extent_io_tree *dirty_pages, int mark);
 int btrfs_transaction_in_commit(struct btrfs_fs_info *info);
 #endif
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -1977,10 +1977,11 @@ int btrfs_sync_log(struct btrfs_trans_ha
 {
 	int index1;
 	int index2;
+	int mark;
 	int ret;
 	struct btrfs_root *log = root->log_root;
 	struct btrfs_root *log_root_tree = root->fs_info->log_root_tree;
-	u64 log_transid = 0;
+	unsigned long log_transid = 0;
 
 	mutex_lock(&root->log_mutex);
 	index1 = root->log_transid % 2;
@@ -2014,24 +2015,29 @@ int btrfs_sync_log(struct btrfs_trans_ha
 		goto out;
 	}
 
+	log_transid = root->log_transid;
+	if (log_transid % 2 == 0)
+		mark = EXTENT_DIRTY;
+	else
+		mark = EXTENT_NEW;
+
 	/* we start IO on  all the marked extents here, but we don't actually
 	 * wait for them until later.
 	 */
-	ret = btrfs_write_marked_extents(log, &log->dirty_log_pages);
+	ret = btrfs_write_marked_extents(log, &log->dirty_log_pages, mark);
 	BUG_ON(ret);
 
 	btrfs_set_root_node(&log->root_item, log->node);
 
 	root->log_batch = 0;
-	log_transid = root->log_transid;
 	root->log_transid++;
 	log->log_transid = root->log_transid;
 	root->log_start_pid = 0;
 	smp_mb();
 	/*
-	 * log tree has been flushed to disk, new modifications of
-	 * the log will be written to new positions. so it's safe to
-	 * allow log writers to go in.
+	 * IO has been started, blocks of the log tree have WRITTEN flag set
+	 * in their headers. new modifications of the log will be written to
+	 * new positions. so it's safe to allow log writers to go in.
 	 */
 	mutex_unlock(&root->log_mutex);
 
@@ -2052,7 +2058,7 @@ int btrfs_sync_log(struct btrfs_trans_ha
 
 	index2 = log_root_tree->log_transid % 2;
 	if (atomic_read(&log_root_tree->log_commit[index2])) {
-		btrfs_wait_marked_extents(log, &log->dirty_log_pages);
+		btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
 		wait_log_commit(trans, log_root_tree,
 				log_root_tree->log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
@@ -2072,16 +2078,17 @@ int btrfs_sync_log(struct btrfs_trans_ha
 	 * check the full commit flag again
 	 */
 	if (root->fs_info->last_trans_log_full_commit == trans->transid) {
-		btrfs_wait_marked_extents(log, &log->dirty_log_pages);
+		btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
 		mutex_unlock(&log_root_tree->log_mutex);
 		ret = -EAGAIN;
 		goto out_wake_log_root;
 	}
 
 	ret = btrfs_write_and_wait_marked_extents(log_root_tree,
-				&log_root_tree->dirty_log_pages);
+				&log_root_tree->dirty_log_pages,
+				EXTENT_DIRTY | EXTENT_NEW);
 	BUG_ON(ret);
-	btrfs_wait_marked_extents(log, &log->dirty_log_pages);
+	btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
 
 	btrfs_set_super_log_root(&root->fs_info->super_for_commit,
 				log_root_tree->node->start);
@@ -2147,12 +2154,12 @@ int btrfs_free_log(struct btrfs_trans_ha
 
 	while (1) {
 		ret = find_first_extent_bit(&log->dirty_log_pages,
-				    0, &start, &end, EXTENT_DIRTY);
+				0, &start, &end, EXTENT_DIRTY | EXTENT_NEW);
 		if (ret)
 			break;
 
-		clear_extent_dirty(&log->dirty_log_pages,
-				   start, end, GFP_NOFS);
+		clear_extent_bits(&log->dirty_log_pages, start, end,
+				  EXTENT_DIRTY | EXTENT_NEW, GFP_NOFS);
 	}
 
 	if (log->log_transid > 0) {

[044/111] Btrfs: Add btrfs_duplicate_item

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit ad48fd754676bfae4139be1a897b1ea58f9aaf21 upstream.

btrfs_duplicate_item duplicates item with new key, guaranteeing
the source item and the new items are in the same tree leaf and
contiguous. It allows us to split file extent in place, without
using lock_extent to prevent bookend extent race.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/ctree.c |  198 ++++++++++++++++++++++++++++++++++++++-----------------
 fs/btrfs/ctree.h |    4 +
 2 files changed, 143 insertions(+), 59 deletions(-)

--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -37,6 +37,11 @@ static int balance_node_right(struct btr
 			      struct extent_buffer *src_buf);
 static int del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		   struct btrfs_path *path, int level, int slot);
+static int setup_items_for_insert(struct btrfs_trans_handle *trans,
+			struct btrfs_root *root, struct btrfs_path *path,
+			struct btrfs_key *cpu_key, u32 *data_size,
+			u32 total_data, u32 total_size, int nr);
+
 
 struct btrfs_path *btrfs_alloc_path(void)
 {
@@ -2997,75 +3002,85 @@ again:
 	return ret;
 }
 
-/*
- * This function splits a single item into two items,
- * giving 'new_key' to the new item and splitting the
- * old one at split_offset (from the start of the item).
- *
- * The path may be released by this operation.  After
- * the split, the path is pointing to the old item.  The
- * new item is going to be in the same node as the old one.
- *
- * Note, the item being split must be smaller enough to live alone on
- * a tree block with room for one extra struct btrfs_item
- *
- * This allows us to split the item in place, keeping a lock on the
- * leaf the entire time.
- */
-int btrfs_split_item(struct btrfs_trans_handle *trans,
-		     struct btrfs_root *root,
-		     struct btrfs_path *path,
-		     struct btrfs_key *new_key,
-		     unsigned long split_offset)
+static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
+					 struct btrfs_root *root,
+					 struct btrfs_path *path, int ins_len)
 {
-	u32 item_size;
+	struct btrfs_key key;
 	struct extent_buffer *leaf;
-	struct btrfs_key orig_key;
-	struct btrfs_item *item;
-	struct btrfs_item *new_item;
-	int ret = 0;
-	int slot;
-	u32 nritems;
-	u32 orig_offset;
-	struct btrfs_disk_key disk_key;
-	char *buf;
+	struct btrfs_file_extent_item *fi;
+	u64 extent_len = 0;
+	u32 item_size;
+	int ret;
 
 	leaf = path->nodes[0];
-	btrfs_item_key_to_cpu(leaf, &orig_key, path->slots[0]);
-	if (btrfs_leaf_free_space(root, leaf) >= sizeof(struct btrfs_item))
-		goto split;
+	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
+
+	BUG_ON(key.type != BTRFS_EXTENT_DATA_KEY &&
+	       key.type != BTRFS_EXTENT_CSUM_KEY);
+
+	if (btrfs_leaf_free_space(root, leaf) >= ins_len)
+		return 0;
 
 	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	if (key.type == BTRFS_EXTENT_DATA_KEY) {
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+				    struct btrfs_file_extent_item);
+		extent_len = btrfs_file_extent_num_bytes(leaf, fi);
+	}
 	btrfs_release_path(root, path);
 
-	path->search_for_split = 1;
 	path->keep_locks = 1;
-
-	ret = btrfs_search_slot(trans, root, &orig_key, path, 0, 1);
+	path->search_for_split = 1;
+	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
 	path->search_for_split = 0;
+	if (ret < 0)
+		goto err;
 
+	ret = -EAGAIN;
+	leaf = path->nodes[0];
 	/* if our item isn't there or got smaller, return now */
-	if (ret != 0 || item_size != btrfs_item_size_nr(path->nodes[0],
-							path->slots[0])) {
-		path->keep_locks = 0;
-		return -EAGAIN;
+	if (ret > 0 || item_size != btrfs_item_size_nr(leaf, path->slots[0]))
+		goto err;
+
+	if (key.type == BTRFS_EXTENT_DATA_KEY) {
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+				    struct btrfs_file_extent_item);
+		if (extent_len != btrfs_file_extent_num_bytes(leaf, fi))
+			goto err;
 	}
 
 	btrfs_set_path_blocking(path);
-	ret = split_leaf(trans, root, &orig_key, path,
-			 sizeof(struct btrfs_item), 1);
-	path->keep_locks = 0;
+	ret = split_leaf(trans, root, &key, path, ins_len, 1);
 	BUG_ON(ret);
 
+	path->keep_locks = 0;
 	btrfs_unlock_up_safe(path, 1);
+	return 0;
+err:
+	path->keep_locks = 0;
+	return ret;
+}
+
+static noinline int split_item(struct btrfs_trans_handle *trans,
+			       struct btrfs_root *root,
+			       struct btrfs_path *path,
+			       struct btrfs_key *new_key,
+			       unsigned long split_offset)
+{
+	struct extent_buffer *leaf;
+	struct btrfs_item *item;
+	struct btrfs_item *new_item;
+	int slot;
+	char *buf;
+	u32 nritems;
+	u32 item_size;
+	u32 orig_offset;
+	struct btrfs_disk_key disk_key;
+
 	leaf = path->nodes[0];
 	BUG_ON(btrfs_leaf_free_space(root, leaf) < sizeof(struct btrfs_item));
 
-split:
-	/*
-	 * make sure any changes to the path from split_leaf leave it
-	 * in a blocking state
-	 */
 	btrfs_set_path_blocking(path);
 
 	item = btrfs_item_nr(leaf, path->slots[0]);
@@ -3073,19 +3088,19 @@ split:
 	item_size = btrfs_item_size(leaf, item);
 
 	buf = kmalloc(item_size, GFP_NOFS);
+	if (!buf)
+		return -ENOMEM;
+
 	read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf,
 			    path->slots[0]), item_size);
-	slot = path->slots[0] + 1;
-	leaf = path->nodes[0];
 
+	slot = path->slots[0] + 1;
 	nritems = btrfs_header_nritems(leaf);
-
 	if (slot != nritems) {
 		/* shift the items */
 		memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + 1),
-			      btrfs_item_nr_offset(slot),
-			      (nritems - slot) * sizeof(struct btrfs_item));
-
+				btrfs_item_nr_offset(slot),
+				(nritems - slot) * sizeof(struct btrfs_item));
 	}
 
 	btrfs_cpu_key_to_disk(&disk_key, new_key);
@@ -3113,16 +3128,81 @@ split:
 			    item_size - split_offset);
 	btrfs_mark_buffer_dirty(leaf);
 
-	ret = 0;
-	if (btrfs_leaf_free_space(root, leaf) < 0) {
-		btrfs_print_leaf(root, leaf);
-		BUG();
-	}
+	BUG_ON(btrfs_leaf_free_space(root, leaf) < 0);
 	kfree(buf);
+	return 0;
+}
+
+/*
+ * This function splits a single item into two items,
+ * giving 'new_key' to the new item and splitting the
+ * old one at split_offset (from the start of the item).
+ *
+ * The path may be released by this operation.  After
+ * the split, the path is pointing to the old item.  The
+ * new item is going to be in the same node as the old one.
+ *
+ * Note, the item being split must be smaller enough to live alone on
+ * a tree block with room for one extra struct btrfs_item
+ *
+ * This allows us to split the item in place, keeping a lock on the
+ * leaf the entire time.
+ */
+int btrfs_split_item(struct btrfs_trans_handle *trans,
+		     struct btrfs_root *root,
+		     struct btrfs_path *path,
+		     struct btrfs_key *new_key,
+		     unsigned long split_offset)
+{
+	int ret;
+	ret = setup_leaf_for_split(trans, root, path,
+				   sizeof(struct btrfs_item));
+	if (ret)
+		return ret;
+
+	ret = split_item(trans, root, path, new_key, split_offset);
 	return ret;
 }
 
 /*
+ * This function duplicate a item, giving 'new_key' to the new item.
+ * It guarantees both items live in the same tree leaf and the new item
+ * is contiguous with the original item.
+ *
+ * This allows us to split file extent in place, keeping a lock on the
+ * leaf the entire time.
+ */
+int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
+			 struct btrfs_root *root,
+			 struct btrfs_path *path,
+			 struct btrfs_key *new_key)
+{
+	struct extent_buffer *leaf;
+	int ret;
+	u32 item_size;
+
+	leaf = path->nodes[0];
+	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	ret = setup_leaf_for_split(trans, root, path,
+				   item_size + sizeof(struct btrfs_item));
+	if (ret)
+		return ret;
+
+	path->slots[0]++;
+	ret = setup_items_for_insert(trans, root, path, new_key, &item_size,
+				     item_size, item_size +
+				     sizeof(struct btrfs_item), 1);
+	BUG_ON(ret);
+
+	leaf = path->nodes[0];
+	memcpy_extent_buffer(leaf,
+			     btrfs_item_ptr_offset(leaf, path->slots[0]),
+			     btrfs_item_ptr_offset(leaf, path->slots[0] - 1),
+			     item_size);
+	return 0;
+}
+
+/*
  * make the item pointed to by the path smaller.  new_size indicates
  * how small to make it, and from_end tells us if we just chop bytes
  * off the end of the item or if we shift the item to chop bytes off
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2089,6 +2089,10 @@ int btrfs_split_item(struct btrfs_trans_
 		     struct btrfs_path *path,
 		     struct btrfs_key *new_key,
 		     unsigned long split_offset);
+int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
+			 struct btrfs_root *root,
+			 struct btrfs_path *path,
+			 struct btrfs_key *new_key);
 int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
 		      *root, struct btrfs_key *key, struct btrfs_path *p, int
 		      ins_len, int cow);

[045/111] Btrfs: Rewrite btrfs_drop_extents

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 920bbbfb05c9fce22e088d20eb9dcb8f96342de9 upstream.

Rewrite btrfs_drop_extents by using btrfs_duplicate_item, so we can
avoid calling lock_extent within transaction.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/ctree.h    |    7 
 fs/btrfs/file.c     |  661 ++++++++++++++++++++--------------------------------
 fs/btrfs/inode.c    |   27 --
 fs/btrfs/ioctl.c    |    3 
 fs/btrfs/tree-log.c |    4 
 5 files changed, 278 insertions(+), 424 deletions(-)

--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2349,12 +2349,9 @@ int btrfs_drop_extent_cache(struct inode
 			    int skip_pinned);
 int btrfs_check_file(struct btrfs_root *root, struct inode *inode);
 extern const struct file_operations btrfs_file_operations;
-int btrfs_drop_extents(struct btrfs_trans_handle *trans,
-		       struct btrfs_root *root, struct inode *inode,
-		       u64 start, u64 end, u64 locked_end,
-		       u64 inline_limit, u64 *hint_block, int drop_cache);
+int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode,
+		       u64 start, u64 end, u64 *hint_byte, int drop_cache);
 int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
-			      struct btrfs_root *root,
 			      struct inode *inode, u64 start, u64 end);
 int btrfs_release_file(struct inode *inode, struct file *file);
 
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -265,319 +265,247 @@ int btrfs_drop_extent_cache(struct inode
  * If an extent intersects the range but is not entirely inside the range
  * it is either truncated or split.  Anything entirely inside the range
  * is deleted from the tree.
- *
- * inline_limit is used to tell this code which offsets in the file to keep
- * if they contain inline extents.
  */
-noinline int btrfs_drop_extents(struct btrfs_trans_handle *trans,
-		       struct btrfs_root *root, struct inode *inode,
-		       u64 start, u64 end, u64 locked_end,
-		       u64 inline_limit, u64 *hint_byte, int drop_cache)
+int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode,
+		       u64 start, u64 end, u64 *hint_byte, int drop_cache)
 {
-	u64 extent_end = 0;
-	u64 search_start = start;
-	u64 ram_bytes = 0;
-	u64 disk_bytenr = 0;
-	u64 orig_locked_end = locked_end;
-	u8 compression;
-	u8 encryption;
-	u16 other_encoding = 0;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_buffer *leaf;
-	struct btrfs_file_extent_item *extent;
+	struct btrfs_file_extent_item *fi;
 	struct btrfs_path *path;
 	struct btrfs_key key;
-	struct btrfs_file_extent_item old;
-	int keep;
-	int slot;
-	int bookend;
-	int found_type = 0;
-	int found_extent;
-	int found_inline;
+	struct btrfs_key new_key;
+	u64 search_start = start;
+	u64 disk_bytenr = 0;
+	u64 num_bytes = 0;
+	u64 extent_offset = 0;
+	u64 extent_end = 0;
+	int del_nr = 0;
+	int del_slot = 0;
+	int extent_type;
 	int recow;
 	int ret;
 
-	inline_limit = 0;
 	if (drop_cache)
 		btrfs_drop_extent_cache(inode, start, end - 1, 0);
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
+
 	while (1) {
 		recow = 0;
-		btrfs_release_path(root, path);
 		ret = btrfs_lookup_file_extent(trans, root, path, inode->i_ino,
 					       search_start, -1);
 		if (ret < 0)
-			goto out;
-		if (ret > 0) {
-			if (path->slots[0] == 0) {
-				ret = 0;
-				goto out;
-			}
-			path->slots[0]--;
+			break;
+		if (ret > 0 && path->slots[0] > 0 && search_start == start) {
+			leaf = path->nodes[0];
+			btrfs_item_key_to_cpu(leaf, &key, path->slots[0] - 1);
+			if (key.objectid == inode->i_ino &&
+			    key.type == BTRFS_EXTENT_DATA_KEY)
+				path->slots[0]--;
 		}
+		ret = 0;
 next_slot:
-		keep = 0;
-		bookend = 0;
-		found_extent = 0;
-		found_inline = 0;
-		compression = 0;
-		encryption = 0;
-		extent = NULL;
 		leaf = path->nodes[0];
-		slot = path->slots[0];
-		ret = 0;
-		btrfs_item_key_to_cpu(leaf, &key, slot);
-		if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY &&
-		    key.offset >= end) {
-			goto out;
-		}
-		if (btrfs_key_type(&key) > BTRFS_EXTENT_DATA_KEY ||
-		    key.objectid != inode->i_ino) {
-			goto out;
-		}
-		if (recow) {
-			search_start = max(key.offset, start);
-			continue;
-		}
-		if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY) {
-			extent = btrfs_item_ptr(leaf, slot,
-						struct btrfs_file_extent_item);
-			found_type = btrfs_file_extent_type(leaf, extent);
-			compression = btrfs_file_extent_compression(leaf,
-								    extent);
-			encryption = btrfs_file_extent_encryption(leaf,
-								  extent);
-			other_encoding = btrfs_file_extent_other_encoding(leaf,
-								  extent);
-			if (found_type == BTRFS_FILE_EXTENT_REG ||
-			    found_type == BTRFS_FILE_EXTENT_PREALLOC) {
-				extent_end =
-				     btrfs_file_extent_disk_bytenr(leaf,
-								   extent);
-				if (extent_end)
-					*hint_byte = extent_end;
-
-				extent_end = key.offset +
-				     btrfs_file_extent_num_bytes(leaf, extent);
-				ram_bytes = btrfs_file_extent_ram_bytes(leaf,
-								extent);
-				found_extent = 1;
-			} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
-				found_inline = 1;
-				extent_end = key.offset +
-				     btrfs_file_extent_inline_len(leaf, extent);
+		if (path->slots[0] >= btrfs_header_nritems(leaf)) {
+			BUG_ON(del_nr > 0);
+			ret = btrfs_next_leaf(root, path);
+			if (ret < 0)
+				break;
+			if (ret > 0) {
+				ret = 0;
+				break;
 			}
+			leaf = path->nodes[0];
+			recow = 1;
+		}
+
+		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
+		if (key.objectid > inode->i_ino ||
+		    key.type > BTRFS_EXTENT_DATA_KEY || key.offset >= end)
+			break;
+
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+				    struct btrfs_file_extent_item);
+		extent_type = btrfs_file_extent_type(leaf, fi);
+
+		if (extent_type == BTRFS_FILE_EXTENT_REG ||
+		    extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
+			disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
+			num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
+			extent_offset = btrfs_file_extent_offset(leaf, fi);
+			extent_end = key.offset +
+				btrfs_file_extent_num_bytes(leaf, fi);
+		} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
+			extent_end = key.offset +
+				btrfs_file_extent_inline_len(leaf, fi);
 		} else {
+			WARN_ON(1);
 			extent_end = search_start;
 		}
 
-		/* we found nothing we can drop */
-		if ((!found_extent && !found_inline) ||
-		    search_start >= extent_end) {
-			int nextret;
-			u32 nritems;
-			nritems = btrfs_header_nritems(leaf);
-			if (slot >= nritems - 1) {
-				nextret = btrfs_next_leaf(root, path);
-				if (nextret)
-					goto out;
-				recow = 1;
-			} else {
-				path->slots[0]++;
-			}
+		if (extent_end <= search_start) {
+			path->slots[0]++;
 			goto next_slot;
 		}
 
-		if (end <= extent_end && start >= key.offset && found_inline)
-			*hint_byte = EXTENT_MAP_INLINE;
+		search_start = max(key.offset, start);
+		if (recow) {
+			btrfs_release_path(root, path);
+			continue;
+		}
 
-		if (found_extent) {
-			read_extent_buffer(leaf, &old, (unsigned long)extent,
-					   sizeof(old));
-		}
-
-		if (end < extent_end && end >= key.offset) {
-			bookend = 1;
-			if (found_inline && start <= key.offset)
-				keep = 1;
-		}
-
-		if (bookend && found_extent) {
-			if (locked_end < extent_end) {
-				ret = try_lock_extent(&BTRFS_I(inode)->io_tree,
-						locked_end, extent_end - 1,
-						GFP_NOFS);
-				if (!ret) {
-					btrfs_release_path(root, path);
-					lock_extent(&BTRFS_I(inode)->io_tree,
-						locked_end, extent_end - 1,
-						GFP_NOFS);
-					locked_end = extent_end;
-					continue;
-				}
-				locked_end = extent_end;
+		/*
+		 *     | - range to drop - |
+		 *  | -------- extent -------- |
+		 */
+		if (start > key.offset && end < extent_end) {
+			BUG_ON(del_nr > 0);
+			BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
+
+			memcpy(&new_key, &key, sizeof(new_key));
+			new_key.offset = start;
+			ret = btrfs_duplicate_item(trans, root, path,
+						   &new_key);
+			if (ret == -EAGAIN) {
+				btrfs_release_path(root, path);
+				continue;
 			}
-			disk_bytenr = le64_to_cpu(old.disk_bytenr);
-			if (disk_bytenr != 0) {
+			if (ret < 0)
+				break;
+
+			leaf = path->nodes[0];
+			fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							start - key.offset);
+
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+
+			extent_offset += start - key.offset;
+			btrfs_set_file_extent_offset(leaf, fi, extent_offset);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							extent_end - start);
+			btrfs_mark_buffer_dirty(leaf);
+
+			if (disk_bytenr > 0) {
 				ret = btrfs_inc_extent_ref(trans, root,
-					   disk_bytenr,
-					   le64_to_cpu(old.disk_num_bytes), 0,
-					   root->root_key.objectid,
-					   key.objectid, key.offset -
-					   le64_to_cpu(old.offset));
+						disk_bytenr, num_bytes, 0,
+						root->root_key.objectid,
+						new_key.objectid,
+						start - extent_offset);
 				BUG_ON(ret);
+				*hint_byte = disk_bytenr;
 			}
+			key.offset = start;
 		}
+		/*
+		 *  | ---- range to drop ----- |
+		 *      | -------- extent -------- |
+		 */
+		if (start <= key.offset && end < extent_end) {
+			BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
+
+			memcpy(&new_key, &key, sizeof(new_key));
+			new_key.offset = end;
+			btrfs_set_item_key_safe(trans, root, path, &new_key);
 
-		if (found_inline) {
-			u64 mask = root->sectorsize - 1;
-			search_start = (extent_end + mask) & ~mask;
-		} else
-			search_start = extent_end;
-
-		/* truncate existing extent */
-		if (start > key.offset) {
-			u64 new_num;
-			u64 old_num;
-			keep = 1;
-			WARN_ON(start & (root->sectorsize - 1));
-			if (found_extent) {
-				new_num = start - key.offset;
-				old_num = btrfs_file_extent_num_bytes(leaf,
-								      extent);
-				*hint_byte =
-					btrfs_file_extent_disk_bytenr(leaf,
-								      extent);
-				if (btrfs_file_extent_disk_bytenr(leaf,
-								  extent)) {
-					inode_sub_bytes(inode, old_num -
-							new_num);
-				}
-				btrfs_set_file_extent_num_bytes(leaf,
-							extent, new_num);
-				btrfs_mark_buffer_dirty(leaf);
-			} else if (key.offset < inline_limit &&
-				   (end > extent_end) &&
-				   (inline_limit < extent_end)) {
-				u32 new_size;
-				new_size = btrfs_file_extent_calc_inline_size(
-						   inline_limit - key.offset);
-				inode_sub_bytes(inode, extent_end -
-						inline_limit);
-				btrfs_set_file_extent_ram_bytes(leaf, extent,
-							new_size);
-				if (!compression && !encryption) {
-					btrfs_truncate_item(trans, root, path,
-							    new_size, 1);
-				}
+			extent_offset += end - key.offset;
+			btrfs_set_file_extent_offset(leaf, fi, extent_offset);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							extent_end - end);
+			btrfs_mark_buffer_dirty(leaf);
+			if (disk_bytenr > 0) {
+				inode_sub_bytes(inode, end - key.offset);
+				*hint_byte = disk_bytenr;
 			}
+			break;
 		}
-		/* delete the entire extent */
-		if (!keep) {
-			if (found_inline)
-				inode_sub_bytes(inode, extent_end -
-						key.offset);
-			ret = btrfs_del_item(trans, root, path);
-			/* TODO update progress marker and return */
-			BUG_ON(ret);
-			extent = NULL;
-			btrfs_release_path(root, path);
-			/* the extent will be freed later */
-		}
-		if (bookend && found_inline && start <= key.offset) {
-			u32 new_size;
-			new_size = btrfs_file_extent_calc_inline_size(
-						   extent_end - end);
-			inode_sub_bytes(inode, end - key.offset);
-			btrfs_set_file_extent_ram_bytes(leaf, extent,
-							new_size);
-			if (!compression && !encryption)
-				ret = btrfs_truncate_item(trans, root, path,
-							  new_size, 0);
-			BUG_ON(ret);
-		}
-		/* create bookend, splitting the extent in two */
-		if (bookend && found_extent) {
-			struct btrfs_key ins;
-			ins.objectid = inode->i_ino;
-			ins.offset = end;
-			btrfs_set_key_type(&ins, BTRFS_EXTENT_DATA_KEY);
-
-			btrfs_release_path(root, path);
-			path->leave_spinning = 1;
-			ret = btrfs_insert_empty_item(trans, root, path, &ins,
-						      sizeof(*extent));
-			BUG_ON(ret);
 
-			leaf = path->nodes[0];
-			extent = btrfs_item_ptr(leaf, path->slots[0],
-						struct btrfs_file_extent_item);
-			write_extent_buffer(leaf, &old,
-					    (unsigned long)extent, sizeof(old));
-
-			btrfs_set_file_extent_compression(leaf, extent,
-							  compression);
-			btrfs_set_file_extent_encryption(leaf, extent,
-							 encryption);
-			btrfs_set_file_extent_other_encoding(leaf, extent,
-							     other_encoding);
-			btrfs_set_file_extent_offset(leaf, extent,
-				    le64_to_cpu(old.offset) + end - key.offset);
-			WARN_ON(le64_to_cpu(old.num_bytes) <
-				(extent_end - end));
-			btrfs_set_file_extent_num_bytes(leaf, extent,
-							extent_end - end);
+		search_start = extent_end;
+		/*
+		 *       | ---- range to drop ----- |
+		 *  | -------- extent -------- |
+		 */
+		if (start > key.offset && end >= extent_end) {
+			BUG_ON(del_nr > 0);
+			BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
 
-			/*
-			 * set the ram bytes to the size of the full extent
-			 * before splitting.  This is a worst case flag,
-			 * but its the best we can do because we don't know
-			 * how splitting affects compression
-			 */
-			btrfs_set_file_extent_ram_bytes(leaf, extent,
-							ram_bytes);
-			btrfs_set_file_extent_type(leaf, extent, found_type);
-
-			btrfs_unlock_up_safe(path, 1);
-			btrfs_mark_buffer_dirty(path->nodes[0]);
-			btrfs_set_lock_blocking(path->nodes[0]);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							start - key.offset);
+			btrfs_mark_buffer_dirty(leaf);
+			if (disk_bytenr > 0) {
+				inode_sub_bytes(inode, extent_end - start);
+				*hint_byte = disk_bytenr;
+			}
+			if (end == extent_end)
+				break;
 
-			path->leave_spinning = 0;
-			btrfs_release_path(root, path);
-			if (disk_bytenr != 0)
-				inode_add_bytes(inode, extent_end - end);
+			path->slots[0]++;
+			goto next_slot;
 		}
 
-		if (found_extent && !keep) {
-			u64 old_disk_bytenr = le64_to_cpu(old.disk_bytenr);
+		/*
+		 *  | ---- range to drop ----- |
+		 *    | ------ extent ------ |
+		 */
+		if (start <= key.offset && end >= extent_end) {
+			if (del_nr == 0) {
+				del_slot = path->slots[0];
+				del_nr = 1;
+			} else {
+				BUG_ON(del_slot + del_nr != path->slots[0]);
+				del_nr++;
+			}
 
-			if (old_disk_bytenr != 0) {
+			if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
 				inode_sub_bytes(inode,
-						le64_to_cpu(old.num_bytes));
+						extent_end - key.offset);
+				extent_end = ALIGN(extent_end,
+						   root->sectorsize);
+			} else if (disk_bytenr > 0) {
 				ret = btrfs_free_extent(trans, root,
-						old_disk_bytenr,
-						le64_to_cpu(old.disk_num_bytes),
-						0, root->root_key.objectid,
+						disk_bytenr, num_bytes, 0,
+						root->root_key.objectid,
 						key.objectid, key.offset -
-						le64_to_cpu(old.offset));
+						extent_offset);
 				BUG_ON(ret);
-				*hint_byte = old_disk_bytenr;
+				inode_sub_bytes(inode,
+						extent_end - key.offset);
+				*hint_byte = disk_bytenr;
 			}
-		}
 
-		if (search_start >= end) {
-			ret = 0;
-			goto out;
+			if (end == extent_end)
+				break;
+
+			if (path->slots[0] + 1 < btrfs_header_nritems(leaf)) {
+				path->slots[0]++;
+				goto next_slot;
+			}
+
+			ret = btrfs_del_items(trans, root, path, del_slot,
+					      del_nr);
+			BUG_ON(ret);
+
+			del_nr = 0;
+			del_slot = 0;
+
+			btrfs_release_path(root, path);
+			continue;
 		}
+
+		BUG_ON(1);
 	}
-out:
-	btrfs_free_path(path);
-	if (locked_end > orig_locked_end) {
-		unlock_extent(&BTRFS_I(inode)->io_tree, orig_locked_end,
-			      locked_end - 1, GFP_NOFS);
+
+	if (del_nr > 0) {
+		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
+		BUG_ON(ret);
 	}
+
+	btrfs_free_path(path);
 	return ret;
 }
 
@@ -620,23 +548,23 @@ static int extent_mergeable(struct exten
  * two or three.
  */
 int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
-			      struct btrfs_root *root,
 			      struct inode *inode, u64 start, u64 end)
 {
+	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_buffer *leaf;
 	struct btrfs_path *path;
 	struct btrfs_file_extent_item *fi;
 	struct btrfs_key key;
+	struct btrfs_key new_key;
 	u64 bytenr;
 	u64 num_bytes;
 	u64 extent_end;
 	u64 orig_offset;
 	u64 other_start;
 	u64 other_end;
-	u64 split = start;
-	u64 locked_end = end;
-	int extent_type;
-	int split_end = 1;
+	u64 split;
+	int del_nr = 0;
+	int del_slot = 0;
 	int ret;
 
 	btrfs_drop_extent_cache(inode, start, end - 1, 0);
@@ -644,12 +572,10 @@ int btrfs_mark_extent_written(struct btr
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 again:
+	split = start;
 	key.objectid = inode->i_ino;
 	key.type = BTRFS_EXTENT_DATA_KEY;
-	if (split == start)
-		key.offset = split;
-	else
-		key.offset = split - 1;
+	key.offset = split;
 
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 	if (ret > 0 && path->slots[0] > 0)
@@ -661,8 +587,8 @@ again:
 	       key.type != BTRFS_EXTENT_DATA_KEY);
 	fi = btrfs_item_ptr(leaf, path->slots[0],
 			    struct btrfs_file_extent_item);
-	extent_type = btrfs_file_extent_type(leaf, fi);
-	BUG_ON(extent_type != BTRFS_FILE_EXTENT_PREALLOC);
+	BUG_ON(btrfs_file_extent_type(leaf, fi) !=
+	       BTRFS_FILE_EXTENT_PREALLOC);
 	extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
 	BUG_ON(key.offset > start || extent_end < end);
 
@@ -670,150 +596,91 @@ again:
 	num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
 	orig_offset = key.offset - btrfs_file_extent_offset(leaf, fi);
 
-	if (key.offset == start)
-		split = end;
-
-	if (key.offset == start && extent_end == end) {
-		int del_nr = 0;
-		int del_slot = 0;
-		other_start = end;
-		other_end = 0;
-		if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
-				     bytenr, &other_start, &other_end)) {
-			extent_end = other_end;
-			del_slot = path->slots[0] + 1;
-			del_nr++;
-			ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
-						0, root->root_key.objectid,
-						inode->i_ino, orig_offset);
-			BUG_ON(ret);
-		}
-		other_start = 0;
-		other_end = start;
-		if (extent_mergeable(leaf, path->slots[0] - 1, inode->i_ino,
-				     bytenr, &other_start, &other_end)) {
-			key.offset = other_start;
-			del_slot = path->slots[0];
-			del_nr++;
-			ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
-						0, root->root_key.objectid,
-						inode->i_ino, orig_offset);
-			BUG_ON(ret);
-		}
-		split_end = 0;
-		if (del_nr == 0) {
-			btrfs_set_file_extent_type(leaf, fi,
-						   BTRFS_FILE_EXTENT_REG);
-			goto done;
+	while (start > key.offset || end < extent_end) {
+		if (key.offset == start)
+			split = end;
+
+		memcpy(&new_key, &key, sizeof(new_key));
+		new_key.offset = split;
+		ret = btrfs_duplicate_item(trans, root, path, &new_key);
+		if (ret == -EAGAIN) {
+			btrfs_release_path(root, path);
+			goto again;
 		}
+		BUG_ON(ret < 0);
 
-		fi = btrfs_item_ptr(leaf, del_slot - 1,
+		leaf = path->nodes[0];
+		fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
 				    struct btrfs_file_extent_item);
-		btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG);
 		btrfs_set_file_extent_num_bytes(leaf, fi,
-						extent_end - key.offset);
+						split - key.offset);
+
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+				    struct btrfs_file_extent_item);
+
+		btrfs_set_file_extent_offset(leaf, fi, split - orig_offset);
+		btrfs_set_file_extent_num_bytes(leaf, fi,
+						extent_end - split);
 		btrfs_mark_buffer_dirty(leaf);
 
-		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
+		ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, 0,
+					   root->root_key.objectid,
+					   inode->i_ino, orig_offset);
 		BUG_ON(ret);
-		goto release;
-	} else if (split == start) {
-		if (locked_end < extent_end) {
-			ret = try_lock_extent(&BTRFS_I(inode)->io_tree,
-					locked_end, extent_end - 1, GFP_NOFS);
-			if (!ret) {
-				btrfs_release_path(root, path);
-				lock_extent(&BTRFS_I(inode)->io_tree,
-					locked_end, extent_end - 1, GFP_NOFS);
-				locked_end = extent_end;
-				goto again;
-			}
-			locked_end = extent_end;
-		}
-		btrfs_set_file_extent_num_bytes(leaf, fi, split - key.offset);
-	} else  {
-		BUG_ON(key.offset != start);
-		key.offset = split;
-		btrfs_set_file_extent_offset(leaf, fi, key.offset -
-					     orig_offset);
-		btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - split);
-		btrfs_set_item_key_safe(trans, root, path, &key);
-		extent_end = split;
-	}
 
-	if (extent_end == end) {
-		split_end = 0;
-		extent_type = BTRFS_FILE_EXTENT_REG;
-	}
-	if (extent_end == end && split == start) {
-		other_start = end;
-		other_end = 0;
-		if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
-				     bytenr, &other_start, &other_end)) {
-			path->slots[0]++;
-			fi = btrfs_item_ptr(leaf, path->slots[0],
-					    struct btrfs_file_extent_item);
-			key.offset = split;
-			btrfs_set_item_key_safe(trans, root, path, &key);
-			btrfs_set_file_extent_offset(leaf, fi, key.offset -
-						     orig_offset);
-			btrfs_set_file_extent_num_bytes(leaf, fi,
-							other_end - split);
-			goto done;
-		}
-	}
-	if (extent_end == end && split == end) {
-		other_start = 0;
-		other_end = start;
-		if (extent_mergeable(leaf, path->slots[0] - 1 , inode->i_ino,
-				     bytenr, &other_start, &other_end)) {
+		if (split == start) {
+			key.offset = start;
+		} else {
+			BUG_ON(start != key.offset);
 			path->slots[0]--;
-			fi = btrfs_item_ptr(leaf, path->slots[0],
-					    struct btrfs_file_extent_item);
-			btrfs_set_file_extent_num_bytes(leaf, fi, extent_end -
-							other_start);
-			goto done;
+			extent_end = end;
 		}
 	}
 
-	btrfs_mark_buffer_dirty(leaf);
-
-	ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, 0,
-				   root->root_key.objectid,
-				   inode->i_ino, orig_offset);
-	BUG_ON(ret);
-	btrfs_release_path(root, path);
-
-	key.offset = start;
-	ret = btrfs_insert_empty_item(trans, root, path, &key, sizeof(*fi));
-	BUG_ON(ret);
-
-	leaf = path->nodes[0];
 	fi = btrfs_item_ptr(leaf, path->slots[0],
 			    struct btrfs_file_extent_item);
-	btrfs_set_file_extent_generation(leaf, fi, trans->transid);
-	btrfs_set_file_extent_type(leaf, fi, extent_type);
-	btrfs_set_file_extent_disk_bytenr(leaf, fi, bytenr);
-	btrfs_set_file_extent_disk_num_bytes(leaf, fi, num_bytes);
-	btrfs_set_file_extent_offset(leaf, fi, key.offset - orig_offset);
-	btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - key.offset);
-	btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes);
-	btrfs_set_file_extent_compression(leaf, fi, 0);
-	btrfs_set_file_extent_encryption(leaf, fi, 0);
-	btrfs_set_file_extent_other_encoding(leaf, fi, 0);
-done:
-	btrfs_mark_buffer_dirty(leaf);
 
-release:
-	btrfs_release_path(root, path);
-	if (split_end && split == start) {
-		split = end;
-		goto again;
+	other_start = end;
+	other_end = 0;
+	if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
+			     bytenr, &other_start, &other_end)) {
+		extent_end = other_end;
+		del_slot = path->slots[0] + 1;
+		del_nr++;
+		ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
+					0, root->root_key.objectid,
+					inode->i_ino, orig_offset);
+		BUG_ON(ret);
+	}
+	other_start = 0;
+	other_end = start;
+	if (extent_mergeable(leaf, path->slots[0] - 1, inode->i_ino,
+			     bytenr, &other_start, &other_end)) {
+		key.offset = other_start;
+		del_slot = path->slots[0];
+		del_nr++;
+		ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
+					0, root->root_key.objectid,
+					inode->i_ino, orig_offset);
+		BUG_ON(ret);
 	}
-	if (locked_end > end) {
-		unlock_extent(&BTRFS_I(inode)->io_tree, end, locked_end - 1,
-			      GFP_NOFS);
+	if (del_nr == 0) {
+		btrfs_set_file_extent_type(leaf, fi,
+					   BTRFS_FILE_EXTENT_REG);
+		btrfs_mark_buffer_dirty(leaf);
+		goto out;
 	}
+
+	fi = btrfs_item_ptr(leaf, del_slot - 1,
+			    struct btrfs_file_extent_item);
+	btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG);
+	btrfs_set_file_extent_num_bytes(leaf, fi,
+					extent_end - key.offset);
+	btrfs_mark_buffer_dirty(leaf);
+
+	ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
+	BUG_ON(ret);
+out:
 	btrfs_free_path(path);
 	return 0;
 }
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -230,8 +230,7 @@ static noinline int cow_file_range_inlin
 		return 1;
 	}
 
-	ret = btrfs_drop_extents(trans, root, inode, start,
-				 aligned_end, aligned_end, start,
+	ret = btrfs_drop_extents(trans, inode, start, aligned_end,
 				 &hint_byte, 1);
 	BUG_ON(ret);
 
@@ -1596,7 +1595,6 @@ static int insert_reserved_file_extent(s
 				       struct inode *inode, u64 file_pos,
 				       u64 disk_bytenr, u64 disk_num_bytes,
 				       u64 num_bytes, u64 ram_bytes,
-				       u64 locked_end,
 				       u8 compression, u8 encryption,
 				       u16 other_encoding, int extent_type)
 {
@@ -1622,9 +1620,8 @@ static int insert_reserved_file_extent(s
 	 * the caller is expected to unpin it and allow it to be merged
 	 * with the others.
 	 */
-	ret = btrfs_drop_extents(trans, root, inode, file_pos,
-				 file_pos + num_bytes, locked_end,
-				 file_pos, &hint, 0);
+	ret = btrfs_drop_extents(trans, inode, file_pos, file_pos + num_bytes,
+				 &hint, 0);
 	BUG_ON(ret);
 
 	ins.objectid = inode->i_ino;
@@ -1746,7 +1743,7 @@ static int btrfs_finish_ordered_io(struc
 		compressed = 1;
 	if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
 		BUG_ON(compressed);
-		ret = btrfs_mark_extent_written(trans, root, inode,
+		ret = btrfs_mark_extent_written(trans, inode,
 						ordered_extent->file_offset,
 						ordered_extent->file_offset +
 						ordered_extent->len);
@@ -1758,8 +1755,6 @@ static int btrfs_finish_ordered_io(struc
 						ordered_extent->disk_len,
 						ordered_extent->len,
 						ordered_extent->len,
-						ordered_extent->file_offset +
-						ordered_extent->len,
 						compressed, 0, 0,
 						BTRFS_FILE_EXTENT_REG);
 		unpin_extent_cache(&BTRFS_I(inode)->extent_tree,
@@ -3209,11 +3204,9 @@ int btrfs_cont_expand(struct inode *inod
 		if (test_bit(EXTENT_FLAG_VACANCY, &em->flags)) {
 			u64 hint_byte = 0;
 			hole_size = last_byte - cur_offset;
-			err = btrfs_drop_extents(trans, root, inode,
-						 cur_offset,
+			err = btrfs_drop_extents(trans, inode, cur_offset,
 						 cur_offset + hole_size,
-						 block_end,
-						 cur_offset, &hint_byte, 1);
+						 &hint_byte, 1);
 			if (err)
 				break;
 
@@ -5643,7 +5636,7 @@ out_fail:
 
 static int prealloc_file_range(struct btrfs_trans_handle *trans,
 			       struct inode *inode, u64 start, u64 end,
-			       u64 locked_end, u64 alloc_hint, int mode)
+			       u64 alloc_hint, int mode)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_key ins;
@@ -5669,8 +5662,7 @@ static int prealloc_file_range(struct bt
 		ret = insert_reserved_file_extent(trans, inode,
 						  cur_offset, ins.objectid,
 						  ins.offset, ins.offset,
-						  ins.offset, locked_end,
-						  0, 0, 0,
+						  ins.offset, 0, 0, 0,
 						  BTRFS_FILE_EXTENT_PREALLOC);
 		BUG_ON(ret);
 		btrfs_drop_extent_cache(inode, cur_offset,
@@ -5779,8 +5771,7 @@ static long btrfs_fallocate(struct inode
 		last_byte = (last_byte + mask) & ~mask;
 		if (em->block_start == EXTENT_MAP_HOLE) {
 			ret = prealloc_file_range(trans, inode, cur_offset,
-					last_byte, locked_end + 1,
-					alloc_hint, mode);
+						last_byte, alloc_hint, mode);
 			if (ret < 0) {
 				free_extent_map(em);
 				break;
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1032,8 +1032,7 @@ static noinline long btrfs_ioctl_clone(s
 	BUG_ON(!trans);
 
 	/* punch hole in destination first */
-	btrfs_drop_extents(trans, root, inode, off, off + len,
-			   off + len, 0, &hint_byte, 1);
+	btrfs_drop_extents(trans, inode, off, off + len, &hint_byte, 1);
 
 	/* clone data */
 	key.objectid = src->i_ino;
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -542,8 +542,8 @@ static noinline int replay_one_extent(st
 
 	saved_nbytes = inode_get_bytes(inode);
 	/* drop any overlapping extents */
-	ret = btrfs_drop_extents(trans, root, inode,
-			 start, extent_end, extent_end, start, &alloc_hint, 1);
+	ret = btrfs_drop_extents(trans, inode, start, extent_end,
+				 &alloc_hint, 1);
 	BUG_ON(ret);
 
 	if (found_type == BTRFS_FILE_EXTENT_REG ||

[046/111] Btrfs: Fix disk_i_size update corner case

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit c216775458a2ee345d9412a2770c2916acfb5d30 upstream.

There are some cases file extents are inserted without involving
ordered struct. In these cases, we update disk_i_size directly,
without checking pending ordered extent and DELALLOC bit. This
patch extends btrfs_ordered_update_i_size() to handle these cases.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/btrfs_inode.h  |    5 --
 fs/btrfs/inode.c        |   71 ++++++++++++++++++++------------
 fs/btrfs/ordered-data.c |  105 +++++++++++++++++++++++++++++++++++++-----------
 fs/btrfs/ordered-data.h |    2 
 4 files changed, 127 insertions(+), 56 deletions(-)

--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -44,9 +44,6 @@ struct btrfs_inode {
 	 */
 	struct extent_io_tree io_failure_tree;
 
-	/* held while inesrting or deleting extents from files */
-	struct mutex extent_mutex;
-
 	/* held while logging the inode in tree-log.c */
 	struct mutex log_mutex;
 
@@ -166,7 +163,7 @@ static inline struct btrfs_inode *BTRFS_
 
 static inline void btrfs_i_size_write(struct inode *inode, u64 size)
 {
-	inode->i_size = size;
+	i_size_write(inode, size);
 	BTRFS_I(inode)->disk_i_size = size;
 }
 
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -188,8 +188,18 @@ static noinline int insert_inline_extent
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_free_path(path);
 
+	/*
+	 * we're an inline extent, so nobody can
+	 * extend the file past i_size without locking
+	 * a page we already have locked.
+	 *
+	 * We must do any isize and inode updates
+	 * before we unlock the pages.  Otherwise we
+	 * could end up racing with unlink.
+	 */
 	BTRFS_I(inode)->disk_i_size = inode->i_size;
 	btrfs_update_inode(trans, root, inode);
+
 	return 0;
 fail:
 	btrfs_free_path(path);
@@ -415,7 +425,6 @@ again:
 						    start, end,
 						    total_compressed, pages);
 		}
-		btrfs_end_transaction(trans, root);
 		if (ret == 0) {
 			/*
 			 * inline extent creation worked, we don't need
@@ -429,9 +438,11 @@ again:
 			     EXTENT_CLEAR_DELALLOC |
 			     EXTENT_CLEAR_ACCOUNTING |
 			     EXTENT_SET_WRITEBACK | EXTENT_END_WRITEBACK);
-			ret = 0;
+
+			btrfs_end_transaction(trans, root);
 			goto free_pages_out;
 		}
+		btrfs_end_transaction(trans, root);
 	}
 
 	if (will_compress) {
@@ -542,7 +553,6 @@ static noinline int submit_compressed_ex
 	if (list_empty(&async_cow->extents))
 		return 0;
 
-	trans = btrfs_join_transaction(root, 1);
 
 	while (!list_empty(&async_cow->extents)) {
 		async_extent = list_entry(async_cow->extents.next,
@@ -589,19 +599,15 @@ retry:
 		lock_extent(io_tree, async_extent->start,
 			    async_extent->start + async_extent->ram_size - 1,
 			    GFP_NOFS);
-		/*
-		 * here we're doing allocation and writeback of the
-		 * compressed pages
-		 */
-		btrfs_drop_extent_cache(inode, async_extent->start,
-					async_extent->start +
-					async_extent->ram_size - 1, 0);
 
+		trans = btrfs_join_transaction(root, 1);
 		ret = btrfs_reserve_extent(trans, root,
 					   async_extent->compressed_size,
 					   async_extent->compressed_size,
 					   0, alloc_hint,
 					   (u64)-1, &ins, 1);
+		btrfs_end_transaction(trans, root);
+
 		if (ret) {
 			int i;
 			for (i = 0; i < async_extent->nr_pages; i++) {
@@ -617,6 +623,14 @@ retry:
 			goto retry;
 		}
 
+		/*
+		 * here we're doing allocation and writeback of the
+		 * compressed pages
+		 */
+		btrfs_drop_extent_cache(inode, async_extent->start,
+					async_extent->start +
+					async_extent->ram_size - 1, 0);
+
 		em = alloc_extent_map(GFP_NOFS);
 		em->start = async_extent->start;
 		em->len = async_extent->ram_size;
@@ -648,8 +662,6 @@ retry:
 					       BTRFS_ORDERED_COMPRESSED);
 		BUG_ON(ret);
 
-		btrfs_end_transaction(trans, root);
-
 		/*
 		 * clear dirty, set writeback and unlock the pages.
 		 */
@@ -671,13 +683,11 @@ retry:
 				    async_extent->nr_pages);
 
 		BUG_ON(ret);
-		trans = btrfs_join_transaction(root, 1);
 		alloc_hint = ins.objectid + ins.offset;
 		kfree(async_extent);
 		cond_resched();
 	}
 
-	btrfs_end_transaction(trans, root);
 	return 0;
 }
 
@@ -741,6 +751,7 @@ static noinline int cow_file_range(struc
 				     EXTENT_CLEAR_DIRTY |
 				     EXTENT_SET_WRITEBACK |
 				     EXTENT_END_WRITEBACK);
+
 			*nr_written = *nr_written +
 			     (end - start + PAGE_CACHE_SIZE) / PAGE_CACHE_SIZE;
 			*page_started = 1;
@@ -1727,18 +1738,27 @@ static int btrfs_finish_ordered_io(struc
 		}
 	}
 
-	trans = btrfs_join_transaction(root, 1);
-
 	if (!ordered_extent)
 		ordered_extent = btrfs_lookup_ordered_extent(inode, start);
 	BUG_ON(!ordered_extent);
-	if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags))
-		goto nocow;
+	if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
+		BUG_ON(!list_empty(&ordered_extent->list));
+		ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent);
+		if (!ret) {
+			trans = btrfs_join_transaction(root, 1);
+			ret = btrfs_update_inode(trans, root, inode);
+			BUG_ON(ret);
+			btrfs_end_transaction(trans, root);
+		}
+		goto out;
+	}
 
 	lock_extent(io_tree, ordered_extent->file_offset,
 		    ordered_extent->file_offset + ordered_extent->len - 1,
 		    GFP_NOFS);
 
+	trans = btrfs_join_transaction(root, 1);
+
 	if (test_bit(BTRFS_ORDERED_COMPRESSED, &ordered_extent->flags))
 		compressed = 1;
 	if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
@@ -1765,22 +1785,20 @@ static int btrfs_finish_ordered_io(struc
 	unlock_extent(io_tree, ordered_extent->file_offset,
 		    ordered_extent->file_offset + ordered_extent->len - 1,
 		    GFP_NOFS);
-nocow:
 	add_pending_csums(trans, inode, ordered_extent->file_offset,
 			  &ordered_extent->list);
 
-	mutex_lock(&BTRFS_I(inode)->extent_mutex);
-	btrfs_ordered_update_i_size(inode, ordered_extent);
-	btrfs_update_inode(trans, root, inode);
-	btrfs_remove_ordered_extent(inode, ordered_extent);
-	mutex_unlock(&BTRFS_I(inode)->extent_mutex);
-
+	/* this also removes the ordered extent from the tree */
+	btrfs_ordered_update_i_size(inode, 0, ordered_extent);
+	ret = btrfs_update_inode(trans, root, inode);
+	BUG_ON(ret);
+	btrfs_end_transaction(trans, root);
+out:
 	/* once for us */
 	btrfs_put_ordered_extent(ordered_extent);
 	/* once for the tree */
 	btrfs_put_ordered_extent(ordered_extent);
 
-	btrfs_end_transaction(trans, root);
 	return 0;
 }
 
@@ -3562,7 +3580,6 @@ static noinline void init_btrfs_i(struct
 	INIT_LIST_HEAD(&BTRFS_I(inode)->ordered_operations);
 	RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node);
 	btrfs_ordered_inode_tree_init(&BTRFS_I(inode)->ordered_tree);
-	mutex_init(&BTRFS_I(inode)->extent_mutex);
 	mutex_init(&BTRFS_I(inode)->log_mutex);
 }
 
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -291,16 +291,16 @@ int btrfs_put_ordered_extent(struct btrf
 
 /*
  * remove an ordered extent from the tree.  No references are dropped
- * but, anyone waiting on this extent is woken up.
+ * and you must wake_up entry->wait.  You must hold the tree mutex
+ * while you call this function.
  */
-int btrfs_remove_ordered_extent(struct inode *inode,
+static int __btrfs_remove_ordered_extent(struct inode *inode,
 				struct btrfs_ordered_extent *entry)
 {
 	struct btrfs_ordered_inode_tree *tree;
 	struct rb_node *node;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
-	mutex_lock(&tree->mutex);
 	node = &entry->rb_node;
 	rb_erase(node, &tree->tree);
 	tree->last = NULL;
@@ -326,9 +326,26 @@ int btrfs_remove_ordered_extent(struct i
 	}
 	spin_unlock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
 
+	return 0;
+}
+
+/*
+ * remove an ordered extent from the tree.  No references are dropped
+ * but any waiters are woken.
+ */
+int btrfs_remove_ordered_extent(struct inode *inode,
+				struct btrfs_ordered_extent *entry)
+{
+	struct btrfs_ordered_inode_tree *tree;
+	int ret;
+
+	tree = &BTRFS_I(inode)->ordered_tree;
+	mutex_lock(&tree->mutex);
+	ret = __btrfs_remove_ordered_extent(inode, entry);
 	mutex_unlock(&tree->mutex);
 	wake_up(&entry->wait);
-	return 0;
+
+	return ret;
 }
 
 /*
@@ -589,7 +606,7 @@ out:
  * After an extent is done, call this to conditionally update the on disk
  * i_size.  i_size is updated to cover any fully written part of the file.
  */
-int btrfs_ordered_update_i_size(struct inode *inode,
+int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 				struct btrfs_ordered_extent *ordered)
 {
 	struct btrfs_ordered_inode_tree *tree = &BTRFS_I(inode)->ordered_tree;
@@ -597,18 +614,30 @@ int btrfs_ordered_update_i_size(struct i
 	u64 disk_i_size;
 	u64 new_i_size;
 	u64 i_size_test;
+	u64 i_size = i_size_read(inode);
 	struct rb_node *node;
+	struct rb_node *prev = NULL;
 	struct btrfs_ordered_extent *test;
+	int ret = 1;
+
+	if (ordered)
+		offset = entry_end(ordered);
 
 	mutex_lock(&tree->mutex);
 	disk_i_size = BTRFS_I(inode)->disk_i_size;
 
+	/* truncate file */
+	if (disk_i_size > i_size) {
+		BTRFS_I(inode)->disk_i_size = i_size;
+		ret = 0;
+		goto out;
+	}
+
 	/*
 	 * if the disk i_size is already at the inode->i_size, or
 	 * this ordered extent is inside the disk i_size, we're done
 	 */
-	if (disk_i_size >= inode->i_size ||
-	    ordered->file_offset + ordered->len <= disk_i_size) {
+	if (disk_i_size == i_size || offset <= disk_i_size) {
 		goto out;
 	}
 
@@ -616,8 +645,7 @@ int btrfs_ordered_update_i_size(struct i
 	 * we can't update the disk_isize if there are delalloc bytes
 	 * between disk_i_size and  this ordered extent
 	 */
-	if (test_range_bit(io_tree, disk_i_size,
-			   ordered->file_offset + ordered->len - 1,
+	if (test_range_bit(io_tree, disk_i_size, offset - 1,
 			   EXTENT_DELALLOC, 0, NULL)) {
 		goto out;
 	}
@@ -626,20 +654,32 @@ int btrfs_ordered_update_i_size(struct i
 	 * if we find an ordered extent then we can't update disk i_size
 	 * yet
 	 */
-	node = &ordered->rb_node;
-	while (1) {
-		node = rb_prev(node);
-		if (!node)
-			break;
+	if (ordered) {
+		node = rb_prev(&ordered->rb_node);
+	} else {
+		prev = tree_search(tree, offset);
+		/*
+		 * we insert file extents without involving ordered struct,
+		 * so there should be no ordered struct cover this offset
+		 */
+		if (prev) {
+			test = rb_entry(prev, struct btrfs_ordered_extent,
+					rb_node);
+			BUG_ON(offset_in_entry(test, offset));
+		}
+		node = prev;
+	}
+	while (node) {
 		test = rb_entry(node, struct btrfs_ordered_extent, rb_node);
 		if (test->file_offset + test->len <= disk_i_size)
 			break;
-		if (test->file_offset >= inode->i_size)
+		if (test->file_offset >= i_size)
 			break;
 		if (test->file_offset >= disk_i_size)
 			goto out;
+		node = rb_prev(node);
 	}
-	new_i_size = min_t(u64, entry_end(ordered), i_size_read(inode));
+	new_i_size = min_t(u64, offset, i_size);
 
 	/*
 	 * at this point, we know we can safely update i_size to at least
@@ -647,7 +687,14 @@ int btrfs_ordered_update_i_size(struct i
 	 * walk forward and see if ios from higher up in the file have
 	 * finished.
 	 */
-	node = rb_next(&ordered->rb_node);
+	if (ordered) {
+		node = rb_next(&ordered->rb_node);
+	} else {
+		if (prev)
+			node = rb_next(prev);
+		else
+			node = rb_first(&tree->tree);
+	}
 	i_size_test = 0;
 	if (node) {
 		/*
@@ -655,10 +702,10 @@ int btrfs_ordered_update_i_size(struct i
 		 * between our ordered extent and the next one.
 		 */
 		test = rb_entry(node, struct btrfs_ordered_extent, rb_node);
-		if (test->file_offset > entry_end(ordered))
+		if (test->file_offset > offset)
 			i_size_test = test->file_offset;
 	} else {
-		i_size_test = i_size_read(inode);
+		i_size_test = i_size;
 	}
 
 	/*
@@ -667,15 +714,25 @@ int btrfs_ordered_update_i_size(struct i
 	 * are no delalloc bytes in this area, it is safe to update
 	 * disk_i_size to the end of the region.
 	 */
-	if (i_size_test > entry_end(ordered) &&
-	    !test_range_bit(io_tree, entry_end(ordered), i_size_test - 1,
-			   EXTENT_DELALLOC, 0, NULL)) {
-		new_i_size = min_t(u64, i_size_test, i_size_read(inode));
+	if (i_size_test > offset &&
+	    !test_range_bit(io_tree, offset, i_size_test - 1,
+			    EXTENT_DELALLOC, 0, NULL)) {
+		new_i_size = min_t(u64, i_size_test, i_size);
 	}
 	BTRFS_I(inode)->disk_i_size = new_i_size;
+	ret = 0;
 out:
+	/*
+	 * we need to remove the ordered extent with the tree lock held
+	 * so that other people calling this function don't find our fully
+	 * processed ordered entry and skip updating the i_size
+	 */
+	if (ordered)
+		__btrfs_remove_ordered_extent(inode, ordered);
 	mutex_unlock(&tree->mutex);
-	return 0;
+	if (ordered)
+		wake_up(&ordered->wait);
+	return ret;
 }
 
 /*
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -150,7 +150,7 @@ void btrfs_start_ordered_extent(struct i
 int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len);
 struct btrfs_ordered_extent *
 btrfs_lookup_first_ordered_extent(struct inode * inode, u64 file_offset);
-int btrfs_ordered_update_i_size(struct inode *inode,
+int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 				struct btrfs_ordered_extent *ordered);
 int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr, u32 *sum);
 int btrfs_wait_ordered_extents(struct btrfs_root *root, int nocow_only);

[047/111] Btrfs: Avoid orphan inodes cleanup while replaying log

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit c71bf099abddf3e0fdc27f251ba76fca1461d49a upstream.

We do log replay in a single transaction, so it's not good to do unbound
operations. This patch cleans up orphan inodes cleanup after replaying
the log. It also avoids doing other unbound operations such as truncating
a file during replaying log. These unbound operations are postponed to
the orphan inode cleanup stage.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/ctree.h      |    5 +++--
 fs/btrfs/disk-io.c    |   17 +++++++++++------
 fs/btrfs/inode.c      |   19 ++++++++++++++++---
 fs/btrfs/relocation.c |    1 +
 fs/btrfs/tree-log.c   |   49 ++++++++++++++++++++++++-------------------------
 5 files changed, 55 insertions(+), 36 deletions(-)

--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -859,8 +859,9 @@ struct btrfs_fs_info {
 	struct mutex ordered_operations_mutex;
 	struct rw_semaphore extent_commit_sem;
 
-	struct rw_semaphore subvol_sem;
+	struct rw_semaphore cleanup_work_sem;
 
+	struct rw_semaphore subvol_sem;
 	struct srcu_struct subvol_srcu;
 
 	struct list_head trans_list;
@@ -1034,12 +1035,12 @@ struct btrfs_root {
 	int ref_cows;
 	int track_dirty;
 	int in_radix;
+	int clean_orphans;
 
 	u64 defrag_trans_start;
 	struct btrfs_key defrag_progress;
 	struct btrfs_key defrag_max;
 	int defrag_running;
-	int defrag_level;
 	char *name;
 	int in_sysfs;
 
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -892,6 +892,8 @@ static int __setup_root(u32 nodesize, u3
 	root->stripesize = stripesize;
 	root->ref_cows = 0;
 	root->track_dirty = 0;
+	root->in_radix = 0;
+	root->clean_orphans = 0;
 
 	root->fs_info = fs_info;
 	root->objectid = objectid;
@@ -928,7 +930,6 @@ static int __setup_root(u32 nodesize, u3
 	root->defrag_trans_start = fs_info->generation;
 	init_completion(&root->kobj_unregister);
 	root->defrag_running = 0;
-	root->defrag_level = 0;
 	root->root_key.objectid = objectid;
 	root->anon_super.s_root = NULL;
 	root->anon_super.s_dev = 0;
@@ -1210,8 +1211,10 @@ again:
 	ret = radix_tree_insert(&fs_info->fs_roots_radix,
 				(unsigned long)root->root_key.objectid,
 				root);
-	if (ret == 0)
+	if (ret == 0) {
 		root->in_radix = 1;
+		root->clean_orphans = 1;
+	}
 	spin_unlock(&fs_info->fs_roots_radix_lock);
 	radix_tree_preload_end();
 	if (ret) {
@@ -1225,10 +1228,6 @@ again:
 	ret = btrfs_find_dead_roots(fs_info->tree_root,
 				    root->root_key.objectid);
 	WARN_ON(ret);
-
-	if (!(fs_info->sb->s_flags & MS_RDONLY))
-		btrfs_orphan_cleanup(root);
-
 	return root;
 fail:
 	free_fs_root(root);
@@ -1689,6 +1688,7 @@ struct btrfs_root *open_ctree(struct sup
 	mutex_init(&fs_info->cleaner_mutex);
 	mutex_init(&fs_info->volume_mutex);
 	init_rwsem(&fs_info->extent_commit_sem);
+	init_rwsem(&fs_info->cleanup_work_sem);
 	init_rwsem(&fs_info->subvol_sem);
 
 	btrfs_init_free_cluster(&fs_info->meta_alloc_cluster);
@@ -2388,6 +2388,11 @@ int btrfs_commit_super(struct btrfs_root
 	mutex_lock(&root->fs_info->cleaner_mutex);
 	btrfs_clean_old_snapshots(root);
 	mutex_unlock(&root->fs_info->cleaner_mutex);
+
+	/* wait until ongoing cleanup work done */
+	down_write(&root->fs_info->cleanup_work_sem);
+	up_write(&root->fs_info->cleanup_work_sem);
+
 	trans = btrfs_start_transaction(root, 1);
 	ret = btrfs_commit_transaction(trans, root);
 	BUG_ON(ret);
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2093,16 +2093,17 @@ void btrfs_orphan_cleanup(struct btrfs_r
 	struct inode *inode;
 	int ret = 0, nr_unlink = 0, nr_truncate = 0;
 
-	path = btrfs_alloc_path();
-	if (!path)
+	if (!xchg(&root->clean_orphans, 0))
 		return;
+
+	path = btrfs_alloc_path();
+	BUG_ON(!path);
 	path->reada = -1;
 
 	key.objectid = BTRFS_ORPHAN_OBJECTID;
 	btrfs_set_key_type(&key, BTRFS_ORPHAN_ITEM_KEY);
 	key.offset = (u64)-1;
 
-
 	while (1) {
 		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 		if (ret < 0) {
@@ -3298,6 +3299,11 @@ void btrfs_delete_inode(struct inode *in
 	}
 	btrfs_wait_ordered_range(inode, 0, (u64)-1);
 
+	if (root->fs_info->log_root_recovering) {
+		BUG_ON(!list_empty(&BTRFS_I(inode)->i_orphan));
+		goto no_delete;
+	}
+
 	if (inode->i_nlink > 0) {
 		BUG_ON(btrfs_root_refs(&root->root_item) != 0);
 		goto no_delete;
@@ -3705,6 +3711,13 @@ struct inode *btrfs_lookup_dentry(struct
 	}
 	srcu_read_unlock(&root->fs_info->subvol_srcu, index);
 
+	if (root != sub_root) {
+		down_read(&root->fs_info->cleanup_work_sem);
+		if (!(inode->i_sb->s_flags & MS_RDONLY))
+			btrfs_orphan_cleanup(sub_root);
+		up_read(&root->fs_info->cleanup_work_sem);
+	}
+
 	return inode;
 }
 
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3755,6 +3755,7 @@ out:
 				       BTRFS_DATA_RELOC_TREE_OBJECTID);
 		if (IS_ERR(fs_root))
 			err = PTR_ERR(fs_root);
+		btrfs_orphan_cleanup(fs_root);
 	}
 	return err;
 }
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -930,6 +930,17 @@ out_nowrite:
 	return 0;
 }
 
+static int insert_orphan_item(struct btrfs_trans_handle *trans,
+			      struct btrfs_root *root, u64 offset)
+{
+	int ret;
+	ret = btrfs_find_orphan_item(root, offset);
+	if (ret > 0)
+		ret = btrfs_insert_orphan_item(trans, root, offset);
+	return ret;
+}
+
+
 /*
  * There are a few corners where the link count of the file can't
  * be properly maintained during replay.  So, instead of adding
@@ -997,9 +1008,13 @@ static noinline int fixup_inode_link_cou
 	}
 	BTRFS_I(inode)->index_cnt = (u64)-1;
 
-	if (inode->i_nlink == 0 && S_ISDIR(inode->i_mode)) {
-		ret = replay_dir_deletes(trans, root, NULL, path,
-					 inode->i_ino, 1);
+	if (inode->i_nlink == 0) {
+		if (S_ISDIR(inode->i_mode)) {
+			ret = replay_dir_deletes(trans, root, NULL, path,
+						 inode->i_ino, 1);
+			BUG_ON(ret);
+		}
+		ret = insert_orphan_item(trans, root, inode->i_ino);
 		BUG_ON(ret);
 	}
 	btrfs_free_path(path);
@@ -1587,7 +1602,6 @@ static int replay_one_buffer(struct btrf
 		/* inode keys are done during the first stage */
 		if (key.type == BTRFS_INODE_ITEM_KEY &&
 		    wc->stage == LOG_WALK_REPLAY_INODES) {
-			struct inode *inode;
 			struct btrfs_inode_item *inode_item;
 			u32 mode;
 
@@ -1603,31 +1617,16 @@ static int replay_one_buffer(struct btrf
 					     eb, i, &key);
 			BUG_ON(ret);
 
-			/* for regular files, truncate away
-			 * extents past the new EOF
+			/* for regular files, make sure corresponding
+			 * orhpan item exist. extents past the new EOF
+			 * will be truncated later by orphan cleanup.
 			 */
 			if (S_ISREG(mode)) {
-				inode = read_one_inode(root,
-						       key.objectid);
-				BUG_ON(!inode);
-
-				ret = btrfs_truncate_inode_items(wc->trans,
-					root, inode, inode->i_size,
-					BTRFS_EXTENT_DATA_KEY);
+				ret = insert_orphan_item(wc->trans, root,
+							 key.objectid);
 				BUG_ON(ret);
-
-				/* if the nlink count is zero here, the iput
-				 * will free the inode.  We bump it to make
-				 * sure it doesn't get freed until the link
-				 * count fixup is done
-				 */
-				if (inode->i_nlink == 0) {
-					btrfs_inc_nlink(inode);
-					btrfs_update_inode(wc->trans,
-							   root, inode);
-				}
-				iput(inode);
 			}
+
 			ret = link_to_fixup_dir(wc->trans, root,
 						path, key.objectid);
 			BUG_ON(ret);

[048/111] Btrfs: Avoid orphan inodes cleanup during committing transaction

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 2e4bfab97055aa6acdd0637913bd705c2d6506d6 upstream.

btrfs_lookup_dentry may trigger orphan cleanup, so it's not good
to call it while committing a transaction.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/ioctl.c       |   29 +++++++++++++++++------------
 fs/btrfs/transaction.c |    4 ----
 2 files changed, 17 insertions(+), 16 deletions(-)

--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -237,7 +237,6 @@ static noinline int create_subvol(struct
 	u64 objectid;
 	u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
 	u64 index = 0;
-	unsigned long nr = 1;
 
 	/*
 	 * 1 - inode item
@@ -342,24 +341,21 @@ static noinline int create_subvol(struct
 
 	d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
 fail:
-	nr = trans->blocks_used;
 	err = btrfs_commit_transaction(trans, root);
 	if (err && !ret)
 		ret = err;
 
 	btrfs_unreserve_metadata_space(root, 6);
-	btrfs_btree_balance_dirty(root, nr);
 	return ret;
 }
 
 static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 			   char *name, int namelen)
 {
+	struct inode *inode;
 	struct btrfs_pending_snapshot *pending_snapshot;
 	struct btrfs_trans_handle *trans;
-	int ret = 0;
-	int err;
-	unsigned long nr = 0;
+	int ret;
 
 	if (!root->ref_cows)
 		return -EINVAL;
@@ -372,20 +368,20 @@ static int create_snapshot(struct btrfs_
 	 */
 	ret = btrfs_reserve_metadata_space(root, 6);
 	if (ret)
-		goto fail_unlock;
+		goto fail;
 
 	pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
 	if (!pending_snapshot) {
 		ret = -ENOMEM;
 		btrfs_unreserve_metadata_space(root, 6);
-		goto fail_unlock;
+		goto fail;
 	}
 	pending_snapshot->name = kmalloc(namelen + 1, GFP_NOFS);
 	if (!pending_snapshot->name) {
 		ret = -ENOMEM;
 		kfree(pending_snapshot);
 		btrfs_unreserve_metadata_space(root, 6);
-		goto fail_unlock;
+		goto fail;
 	}
 	memcpy(pending_snapshot->name, name, namelen);
 	pending_snapshot->name[namelen] = '\0';
@@ -395,10 +391,19 @@ static int create_snapshot(struct btrfs_
 	pending_snapshot->root = root;
 	list_add(&pending_snapshot->list,
 		 &trans->transaction->pending_snapshots);
-	err = btrfs_commit_transaction(trans, root);
+	ret = btrfs_commit_transaction(trans, root);
+	BUG_ON(ret);
+	btrfs_unreserve_metadata_space(root, 6);
 
-fail_unlock:
-	btrfs_btree_balance_dirty(root, nr);
+	inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
+	if (IS_ERR(inode)) {
+		ret = PTR_ERR(inode);
+		goto fail;
+	}
+	BUG_ON(!inode);
+	d_instantiate(dentry, inode);
+	ret = 0;
+fail:
 	return ret;
 }
 
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -796,7 +796,6 @@ static noinline int create_pending_snaps
 	memcpy(&pending->root_key, &key, sizeof(key));
 fail:
 	kfree(new_root_item);
-	btrfs_unreserve_metadata_space(root, 6);
 	return ret;
 }
 
@@ -808,7 +807,6 @@ static noinline int finish_pending_snaps
 	u64 index = 0;
 	struct btrfs_trans_handle *trans;
 	struct inode *parent_inode;
-	struct inode *inode;
 	struct btrfs_root *parent_root;
 
 	parent_inode = pending->dentry->d_parent->d_inode;
@@ -840,8 +838,6 @@ static noinline int finish_pending_snaps
 
 	BUG_ON(ret);
 
-	inode = btrfs_lookup_dentry(parent_inode, pending->dentry);
-	d_instantiate(pending->dentry, inode);
 fail:
 	btrfs_end_transaction(trans, fs_info->fs_root);
 	return ret;

[049/111] Btrfs: Make fallocate(2) more ENOSPC friendly

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 5a303d5d4b8055d2e5a03e92d04745bfc5881a22 upstream.

fallocate(2) may allocate large number of file extents, so it's not
good to do it in a single transaction. This patch make fallocate(2)
start a new transaction for each file extents it allocates.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/inode.c |   65 +++++++++++++++++++++++++++----------------------------
 1 file changed, 32 insertions(+), 33 deletions(-)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5664,10 +5664,10 @@ out_fail:
 	return err;
 }
 
-static int prealloc_file_range(struct btrfs_trans_handle *trans,
-			       struct inode *inode, u64 start, u64 end,
+static int prealloc_file_range(struct inode *inode, u64 start, u64 end,
 			       u64 alloc_hint, int mode)
 {
+	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_key ins;
 	u64 alloc_size;
@@ -5678,17 +5678,23 @@ static int prealloc_file_range(struct bt
 	while (num_bytes > 0) {
 		alloc_size = min(num_bytes, root->fs_info->max_extent);
 
-		ret = btrfs_reserve_metadata_space(root, 1);
-		if (ret)
-			goto out;
-
 		ret = btrfs_reserve_extent(trans, root, alloc_size,
 					   root->sectorsize, 0, alloc_hint,
 					   (u64)-1, &ins, 1);
 		if (ret) {
 			WARN_ON(1);
-			goto out;
+			break;
 		}
+
+		ret = btrfs_reserve_metadata_space(root, 3);
+		if (ret) {
+			btrfs_free_reserved_extent(root, ins.objectid,
+						   ins.offset);
+			break;
+		}
+
+		trans = btrfs_start_transaction(root, 1);
+
 		ret = insert_reserved_file_extent(trans, inode,
 						  cur_offset, ins.objectid,
 						  ins.offset, ins.offset,
@@ -5697,22 +5703,25 @@ static int prealloc_file_range(struct bt
 		BUG_ON(ret);
 		btrfs_drop_extent_cache(inode, cur_offset,
 					cur_offset + ins.offset -1, 0);
+
 		num_bytes -= ins.offset;
 		cur_offset += ins.offset;
 		alloc_hint = ins.objectid + ins.offset;
-		btrfs_unreserve_metadata_space(root, 1);
-	}
-out:
-	if (cur_offset > start) {
+
 		inode->i_ctime = CURRENT_TIME;
 		BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC;
 		if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-		    cur_offset > i_size_read(inode))
-			btrfs_i_size_write(inode, cur_offset);
+		    cur_offset > inode->i_size) {
+			i_size_write(inode, cur_offset);
+			btrfs_ordered_update_i_size(inode, cur_offset, NULL);
+		}
+
 		ret = btrfs_update_inode(trans, root, inode);
 		BUG_ON(ret);
-	}
 
+		btrfs_end_transaction(trans, root);
+		btrfs_unreserve_metadata_space(root, 3);
+	}
 	return ret;
 }
 
@@ -5727,8 +5736,6 @@ static long btrfs_fallocate(struct inode
 	u64 locked_end;
 	u64 mask = BTRFS_I(inode)->root->sectorsize - 1;
 	struct extent_map *em;
-	struct btrfs_trans_handle *trans;
-	struct btrfs_root *root;
 	int ret;
 
 	alloc_start = offset & ~mask;
@@ -5747,9 +5754,7 @@ static long btrfs_fallocate(struct inode
 			goto out;
 	}
 
-	root = BTRFS_I(inode)->root;
-
-	ret = btrfs_check_data_free_space(root, inode,
+	ret = btrfs_check_data_free_space(BTRFS_I(inode)->root, inode,
 					  alloc_end - alloc_start);
 	if (ret)
 		goto out;
@@ -5758,12 +5763,6 @@ static long btrfs_fallocate(struct inode
 	while (1) {
 		struct btrfs_ordered_extent *ordered;
 
-		trans = btrfs_start_transaction(BTRFS_I(inode)->root, 1);
-		if (!trans) {
-			ret = -EIO;
-			goto out_free;
-		}
-
 		/* the extent lock is ordered inside the running
 		 * transaction
 		 */
@@ -5777,8 +5776,6 @@ static long btrfs_fallocate(struct inode
 			btrfs_put_ordered_extent(ordered);
 			unlock_extent(&BTRFS_I(inode)->io_tree,
 				      alloc_start, locked_end, GFP_NOFS);
-			btrfs_end_transaction(trans, BTRFS_I(inode)->root);
-
 			/*
 			 * we can't wait on the range with the transaction
 			 * running or with the extent lock held
@@ -5799,9 +5796,12 @@ static long btrfs_fallocate(struct inode
 		BUG_ON(IS_ERR(em) || !em);
 		last_byte = min(extent_map_end(em), alloc_end);
 		last_byte = (last_byte + mask) & ~mask;
-		if (em->block_start == EXTENT_MAP_HOLE) {
-			ret = prealloc_file_range(trans, inode, cur_offset,
-						last_byte, alloc_hint, mode);
+		if (em->block_start == EXTENT_MAP_HOLE ||
+		    (cur_offset >= inode->i_size &&
+		     !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) {
+			ret = prealloc_file_range(inode,
+						  cur_offset, last_byte,
+						  alloc_hint, mode);
 			if (ret < 0) {
 				free_extent_map(em);
 				break;
@@ -5820,9 +5820,8 @@ static long btrfs_fallocate(struct inode
 	unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
 		      GFP_NOFS);
 
-	btrfs_end_transaction(trans, BTRFS_I(inode)->root);
-out_free:
-	btrfs_free_reserved_data_space(root, inode, alloc_end - alloc_start);
+	btrfs_free_reserved_data_space(BTRFS_I(inode)->root, inode,
+				       alloc_end - alloc_start);
 out:
 	mutex_unlock(&inode->i_mutex);
 	return ret;

[050/111] Btrfs: Make truncate(2) more ENOSPC friendly

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 8082510e7124cc50d728f1b875639cb4e22312cc upstream.

truncating and deleting regular files are unbound operations,
so it's not good to do them in a single transaction. This
patch makes btrfs_truncate and btrfs_delete_inode start a
new transaction after all items in a tree leaf are deleted.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/inode.c      |  316 ++++++++++++++++++++++++++++++--------------------
 fs/btrfs/relocation.c |   33 +++--
 2 files changed, 212 insertions(+), 137 deletions(-)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2848,37 +2848,40 @@ out:
  * min_type is the minimum key type to truncate down to.  If set to 0, this
  * will kill all the items on this inode, including the INODE_ITEM_KEY.
  */
-noinline int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
-					struct btrfs_root *root,
-					struct inode *inode,
-					u64 new_size, u32 min_type)
+int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
+			       struct btrfs_root *root,
+			       struct inode *inode,
+			       u64 new_size, u32 min_type)
 {
-	int ret;
 	struct btrfs_path *path;
-	struct btrfs_key key;
-	struct btrfs_key found_key;
-	u32 found_type = (u8)-1;
 	struct extent_buffer *leaf;
 	struct btrfs_file_extent_item *fi;
+	struct btrfs_key key;
+	struct btrfs_key found_key;
 	u64 extent_start = 0;
 	u64 extent_num_bytes = 0;
 	u64 extent_offset = 0;
 	u64 item_end = 0;
+	u64 mask = root->sectorsize - 1;
+	u32 found_type = (u8)-1;
 	int found_extent;
 	int del_item;
 	int pending_del_nr = 0;
 	int pending_del_slot = 0;
 	int extent_type = -1;
 	int encoding;
-	u64 mask = root->sectorsize - 1;
+	int ret;
+	int err = 0;
+
+	BUG_ON(new_size > 0 && min_type != BTRFS_EXTENT_DATA_KEY);
 
 	if (root->ref_cows)
 		btrfs_drop_extent_cache(inode, new_size & (~mask), (u64)-1, 0);
+
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 	path->reada = -1;
 
-	/* FIXME, add redo link to tree so we don't leak on crash */
 	key.objectid = inode->i_ino;
 	key.offset = (u64)-1;
 	key.type = (u8)-1;
@@ -2886,17 +2889,17 @@ noinline int btrfs_truncate_inode_items(
 search_again:
 	path->leave_spinning = 1;
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
-	if (ret < 0)
-		goto error;
+	if (ret < 0) {
+		err = ret;
+		goto out;
+	}
 
 	if (ret > 0) {
 		/* there are no items in the tree for us to truncate, we're
 		 * done
 		 */
-		if (path->slots[0] == 0) {
-			ret = 0;
-			goto error;
-		}
+		if (path->slots[0] == 0)
+			goto out;
 		path->slots[0]--;
 	}
 
@@ -2931,28 +2934,17 @@ search_again:
 			}
 			item_end--;
 		}
-		if (item_end < new_size) {
-			if (found_type == BTRFS_DIR_ITEM_KEY)
-				found_type = BTRFS_INODE_ITEM_KEY;
-			else if (found_type == BTRFS_EXTENT_ITEM_KEY)
-				found_type = BTRFS_EXTENT_DATA_KEY;
-			else if (found_type == BTRFS_EXTENT_DATA_KEY)
-				found_type = BTRFS_XATTR_ITEM_KEY;
-			else if (found_type == BTRFS_XATTR_ITEM_KEY)
-				found_type = BTRFS_INODE_REF_KEY;
-			else if (found_type)
-				found_type--;
-			else
+		if (found_type > min_type) {
+			del_item = 1;
+		} else {
+			if (item_end < new_size)
 				break;
-			btrfs_set_key_type(&key, found_type);
-			goto next;
+			if (found_key.offset >= new_size)
+				del_item = 1;
+			else
+				del_item = 0;
 		}
-		if (found_key.offset >= new_size)
-			del_item = 1;
-		else
-			del_item = 0;
 		found_extent = 0;
-
 		/* FIXME, shrink the extent if the ref count is only 1 */
 		if (found_type != BTRFS_EXTENT_DATA_KEY)
 			goto delete;
@@ -3039,42 +3031,36 @@ delete:
 						inode->i_ino, extent_offset);
 			BUG_ON(ret);
 		}
-next:
-		if (path->slots[0] == 0) {
-			if (pending_del_nr)
-				goto del_pending;
-			btrfs_release_path(root, path);
-			if (found_type == BTRFS_INODE_ITEM_KEY)
-				break;
-			goto search_again;
-		}
 
-		path->slots[0]--;
-		if (pending_del_nr &&
-		    path->slots[0] + 1 != pending_del_slot) {
-			struct btrfs_key debug;
-del_pending:
-			btrfs_item_key_to_cpu(path->nodes[0], &debug,
-					      pending_del_slot);
-			ret = btrfs_del_items(trans, root, path,
-					      pending_del_slot,
-					      pending_del_nr);
-			BUG_ON(ret);
-			pending_del_nr = 0;
+		if (found_type == BTRFS_INODE_ITEM_KEY)
+			break;
+
+		if (path->slots[0] == 0 ||
+		    path->slots[0] != pending_del_slot) {
+			if (root->ref_cows) {
+				err = -EAGAIN;
+				goto out;
+			}
+			if (pending_del_nr) {
+				ret = btrfs_del_items(trans, root, path,
+						pending_del_slot,
+						pending_del_nr);
+				BUG_ON(ret);
+				pending_del_nr = 0;
+			}
 			btrfs_release_path(root, path);
-			if (found_type == BTRFS_INODE_ITEM_KEY)
-				break;
 			goto search_again;
+		} else {
+			path->slots[0]--;
 		}
 	}
-	ret = 0;
-error:
+out:
 	if (pending_del_nr) {
 		ret = btrfs_del_items(trans, root, path, pending_del_slot,
 				      pending_del_nr);
 	}
 	btrfs_free_path(path);
-	return ret;
+	return err;
 }
 
 /*
@@ -3194,10 +3180,6 @@ int btrfs_cont_expand(struct inode *inod
 	if (size <= hole_start)
 		return 0;
 
-	err = btrfs_truncate_page(inode->i_mapping, inode->i_size);
-	if (err)
-		return err;
-
 	while (1) {
 		struct btrfs_ordered_extent *ordered;
 		btrfs_wait_ordered_range(inode, hole_start,
@@ -3210,9 +3192,6 @@ int btrfs_cont_expand(struct inode *inod
 		btrfs_put_ordered_extent(ordered);
 	}
 
-	trans = btrfs_start_transaction(root, 1);
-	btrfs_set_trans_block_group(trans, inode);
-
 	cur_offset = hole_start;
 	while (1) {
 		em = btrfs_get_extent(inode, NULL, 0, cur_offset,
@@ -3220,38 +3199,120 @@ int btrfs_cont_expand(struct inode *inod
 		BUG_ON(IS_ERR(em) || !em);
 		last_byte = min(extent_map_end(em), block_end);
 		last_byte = (last_byte + mask) & ~mask;
-		if (test_bit(EXTENT_FLAG_VACANCY, &em->flags)) {
+		if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) {
 			u64 hint_byte = 0;
 			hole_size = last_byte - cur_offset;
-			err = btrfs_drop_extents(trans, inode, cur_offset,
-						 cur_offset + hole_size,
-						 &hint_byte, 1);
-			if (err)
-				break;
 
-			err = btrfs_reserve_metadata_space(root, 1);
+			err = btrfs_reserve_metadata_space(root, 2);
 			if (err)
 				break;
 
+			trans = btrfs_start_transaction(root, 1);
+			btrfs_set_trans_block_group(trans, inode);
+
+			err = btrfs_drop_extents(trans, inode, cur_offset,
+						 cur_offset + hole_size,
+						 &hint_byte, 1);
+			BUG_ON(err);
+
 			err = btrfs_insert_file_extent(trans, root,
 					inode->i_ino, cur_offset, 0,
 					0, hole_size, 0, hole_size,
 					0, 0, 0);
+			BUG_ON(err);
+
 			btrfs_drop_extent_cache(inode, hole_start,
 					last_byte - 1, 0);
-			btrfs_unreserve_metadata_space(root, 1);
+
+			btrfs_end_transaction(trans, root);
+			btrfs_unreserve_metadata_space(root, 2);
 		}
 		free_extent_map(em);
 		cur_offset = last_byte;
-		if (err || cur_offset >= block_end)
+		if (cur_offset >= block_end)
 			break;
 	}
 
-	btrfs_end_transaction(trans, root);
 	unlock_extent(io_tree, hole_start, block_end - 1, GFP_NOFS);
 	return err;
 }
 
+static int btrfs_setattr_size(struct inode *inode, struct iattr *attr)
+{
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_trans_handle *trans;
+	unsigned long nr;
+	int ret;
+
+	if (attr->ia_size == inode->i_size)
+		return 0;
+
+	if (attr->ia_size > inode->i_size) {
+		unsigned long limit;
+		limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur;
+		if (attr->ia_size > inode->i_sb->s_maxbytes)
+			return -EFBIG;
+		if (limit != RLIM_INFINITY && attr->ia_size > limit) {
+			send_sig(SIGXFSZ, current, 0);
+			return -EFBIG;
+		}
+	}
+
+	ret = btrfs_reserve_metadata_space(root, 1);
+	if (ret)
+		return ret;
+
+	trans = btrfs_start_transaction(root, 1);
+	btrfs_set_trans_block_group(trans, inode);
+
+	ret = btrfs_orphan_add(trans, inode);
+	BUG_ON(ret);
+
+	nr = trans->blocks_used;
+	btrfs_end_transaction(trans, root);
+	btrfs_unreserve_metadata_space(root, 1);
+	btrfs_btree_balance_dirty(root, nr);
+
+	if (attr->ia_size > inode->i_size) {
+		ret = btrfs_cont_expand(inode, attr->ia_size);
+		if (ret) {
+			btrfs_truncate(inode);
+			return ret;
+		}
+
+		i_size_write(inode, attr->ia_size);
+		btrfs_ordered_update_i_size(inode, inode->i_size, NULL);
+
+		trans = btrfs_start_transaction(root, 1);
+		btrfs_set_trans_block_group(trans, inode);
+
+		ret = btrfs_update_inode(trans, root, inode);
+		BUG_ON(ret);
+		if (inode->i_nlink > 0) {
+			ret = btrfs_orphan_del(trans, inode);
+			BUG_ON(ret);
+		}
+		nr = trans->blocks_used;
+		btrfs_end_transaction(trans, root);
+		btrfs_btree_balance_dirty(root, nr);
+		return 0;
+	}
+
+	/*
+	 * We're truncating a file that used to have good data down to
+	 * zero. Make sure it gets into the ordered flush list so that
+	 * any new writes get down to disk quickly.
+	 */
+	if (attr->ia_size == 0)
+		BTRFS_I(inode)->ordered_data_close = 1;
+
+	/* we don't support swapfiles, so vmtruncate shouldn't fail */
+	ret = vmtruncate(inode, attr->ia_size);
+	BUG_ON(ret);
+
+	return 0;
+}
+
 static int btrfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = dentry->d_inode;
@@ -3262,23 +3323,14 @@ static int btrfs_setattr(struct dentry *
 		return err;
 
 	if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE)) {
-		if (attr->ia_size > inode->i_size) {
-			err = btrfs_cont_expand(inode, attr->ia_size);
-			if (err)
-				return err;
-		} else if (inode->i_size > 0 &&
-			   attr->ia_size == 0) {
-
-			/* we're truncating a file that used to have good
-			 * data down to zero.  Make sure it gets into
-			 * the ordered flush list so that any new writes
-			 * get down to disk quickly.
-			 */
-			BTRFS_I(inode)->ordered_data_close = 1;
-		}
+		err = btrfs_setattr_size(inode, attr);
+		if (err)
+			return err;
 	}
+	attr->ia_valid &= ~ATTR_SIZE;
 
-	err = inode_setattr(inode, attr);
+	if (attr->ia_valid)
+		err = inode_setattr(inode, attr);
 
 	if (!err && ((attr->ia_valid & ATTR_MODE)))
 		err = btrfs_acl_chmod(inode);
@@ -3310,30 +3362,32 @@ void btrfs_delete_inode(struct inode *in
 	}
 
 	btrfs_i_size_write(inode, 0);
-	trans = btrfs_join_transaction(root, 1);
 
-	btrfs_set_trans_block_group(trans, inode);
-	ret = btrfs_truncate_inode_items(trans, root, inode, inode->i_size, 0);
-	if (ret) {
-		btrfs_orphan_del(NULL, inode);
-		goto no_delete_lock;
-	}
+	while (1) {
+		trans = btrfs_start_transaction(root, 1);
+		btrfs_set_trans_block_group(trans, inode);
+		ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0);
 
-	btrfs_orphan_del(trans, inode);
+		if (ret != -EAGAIN)
+			break;
 
-	nr = trans->blocks_used;
-	clear_inode(inode);
+		nr = trans->blocks_used;
+		btrfs_end_transaction(trans, root);
+		trans = NULL;
+		btrfs_btree_balance_dirty(root, nr);
+	}
 
-	btrfs_end_transaction(trans, root);
-	btrfs_btree_balance_dirty(root, nr);
-	return;
+	if (ret == 0) {
+		ret = btrfs_orphan_del(trans, inode);
+		BUG_ON(ret);
+	}
 
-no_delete_lock:
 	nr = trans->blocks_used;
 	btrfs_end_transaction(trans, root);
 	btrfs_btree_balance_dirty(root, nr);
 no_delete:
 	clear_inode(inode);
+	return;
 }
 
 /*
@@ -5097,17 +5151,20 @@ static void btrfs_truncate(struct inode
 	unsigned long nr;
 	u64 mask = root->sectorsize - 1;
 
-	if (!S_ISREG(inode->i_mode))
-		return;
-	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
+	if (!S_ISREG(inode->i_mode)) {
+		WARN_ON(1);
 		return;
+	}
 
 	ret = btrfs_truncate_page(inode->i_mapping, inode->i_size);
 	if (ret)
 		return;
+
 	btrfs_wait_ordered_range(inode, inode->i_size & (~mask), (u64)-1);
+	btrfs_ordered_update_i_size(inode, inode->i_size, NULL);
 
 	trans = btrfs_start_transaction(root, 1);
+	btrfs_set_trans_block_group(trans, inode);
 
 	/*
 	 * setattr is responsible for setting the ordered_data_close flag,
@@ -5129,21 +5186,32 @@ static void btrfs_truncate(struct inode
 	if (inode->i_size == 0 && BTRFS_I(inode)->ordered_data_close)
 		btrfs_add_ordered_operation(trans, root, inode);
 
-	btrfs_set_trans_block_group(trans, inode);
-	btrfs_i_size_write(inode, inode->i_size);
+	while (1) {
+		ret = btrfs_truncate_inode_items(trans, root, inode,
+						 inode->i_size,
+						 BTRFS_EXTENT_DATA_KEY);
+		if (ret != -EAGAIN)
+			break;
 
-	ret = btrfs_orphan_add(trans, inode);
-	if (ret)
-		goto out;
-	/* FIXME, add redo link to tree so we don't leak on crash */
-	ret = btrfs_truncate_inode_items(trans, root, inode, inode->i_size,
-				      BTRFS_EXTENT_DATA_KEY);
-	btrfs_update_inode(trans, root, inode);
+		ret = btrfs_update_inode(trans, root, inode);
+		BUG_ON(ret);
+
+		nr = trans->blocks_used;
+		btrfs_end_transaction(trans, root);
+		btrfs_btree_balance_dirty(root, nr);
+
+		trans = btrfs_start_transaction(root, 1);
+		btrfs_set_trans_block_group(trans, inode);
+	}
 
-	ret = btrfs_orphan_del(trans, inode);
+	if (ret == 0 && inode->i_nlink > 0) {
+		ret = btrfs_orphan_del(trans, inode);
+		BUG_ON(ret);
+	}
+
+	ret = btrfs_update_inode(trans, root, inode);
 	BUG_ON(ret);
 
-out:
 	nr = trans->blocks_used;
 	ret = btrfs_end_transaction_throttle(trans, root);
 	BUG_ON(ret);
@@ -5240,9 +5308,9 @@ void btrfs_destroy_inode(struct inode *i
 
 	spin_lock(&root->list_lock);
 	if (!list_empty(&BTRFS_I(inode)->i_orphan)) {
-		printk(KERN_ERR "BTRFS: inode %lu: inode still on the orphan"
-		       " list\n", inode->i_ino);
-		dump_stack();
+		printk(KERN_INFO "BTRFS: inode %lu still on the orphan list\n",
+		       inode->i_ino);
+		list_del_init(&BTRFS_I(inode)->i_orphan);
 	}
 	spin_unlock(&root->list_lock);
 
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1561,6 +1561,20 @@ static int invalidate_extent_cache(struc
 	return 0;
 }
 
+static void put_inodes(struct list_head *list)
+{
+	struct inodevec *ivec;
+	while (!list_empty(list)) {
+		ivec = list_entry(list->next, struct inodevec, list);
+		list_del(&ivec->list);
+		while (ivec->nr > 0) {
+			ivec->nr--;
+			iput(ivec->inode[ivec->nr]);
+		}
+		kfree(ivec);
+	}
+}
+
 static int find_next_key(struct btrfs_path *path, int level,
 			 struct btrfs_key *key)
 
@@ -1723,6 +1737,11 @@ static noinline_for_stack int merge_relo
 
 		btrfs_btree_balance_dirty(root, nr);
 
+		/*
+		 * put inodes outside transaction, otherwise we may deadlock.
+		 */
+		put_inodes(&inode_list);
+
 		if (replaced && rc->stage == UPDATE_DATA_PTRS)
 			invalidate_extent_cache(root, &key, &next_key);
 	}
@@ -1752,19 +1771,7 @@ out:
 
 	btrfs_btree_balance_dirty(root, nr);
 
-	/*
-	 * put inodes while we aren't holding the tree locks
-	 */
-	while (!list_empty(&inode_list)) {
-		struct inodevec *ivec;
-		ivec = list_entry(inode_list.next, struct inodevec, list);
-		list_del(&ivec->list);
-		while (ivec->nr > 0) {
-			ivec->nr--;
-			iput(ivec->inode[ivec->nr]);
-		}
-		kfree(ivec);
-	}
+	put_inodes(&inode_list);
 
 	if (replaced && rc->stage == UPDATE_DATA_PTRS)
 		invalidate_extent_cache(root, &key, &next_key);

[051/111] Btrfs: Pass transaction handle to security and ACL initialization functions

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit f34f57a3ab4e73304d78c125682f1a53cd3975f2 upstream.

Pass transaction handle down to security and ACL initialization
functions, so we can avoid starting nested transactions

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/acl.c      |   23 ++++++++------
 fs/btrfs/ctree.h    |   13 +++++---
 fs/btrfs/dir-item.c |   19 ++++--------
 fs/btrfs/inode.c    |   15 +++++----
 fs/btrfs/xattr.c    |   80 +++++++++++++++++++++++++++++++++++-----------------
 fs/btrfs/xattr.h    |    9 +++--
 6 files changed, 96 insertions(+), 63 deletions(-)

--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -94,7 +94,8 @@ static int btrfs_xattr_get_acl(struct in
 /*
  * Needs to be called with fs_mutex held
  */
-static int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
+static int btrfs_set_acl(struct btrfs_trans_handle *trans,
+			 struct inode *inode, struct posix_acl *acl, int type)
 {
 	int ret, size = 0;
 	const char *name;
@@ -140,8 +141,7 @@ static int btrfs_set_acl(struct inode *i
 			goto out;
 	}
 
-	ret = __btrfs_setxattr(inode, name, value, size, 0);
-
+	ret = __btrfs_setxattr(trans, inode, name, value, size, 0);
 out:
 	kfree(value);
 
@@ -154,7 +154,7 @@ out:
 static int btrfs_xattr_set_acl(struct inode *inode, int type,
 			       const void *value, size_t size)
 {
-	int ret = 0;
+	int ret;
 	struct posix_acl *acl = NULL;
 
 	if (!is_owner_or_cap(inode))
@@ -170,7 +170,7 @@ static int btrfs_xattr_set_acl(struct in
 		}
 	}
 
-	ret = btrfs_set_acl(inode, acl, type);
+	ret = btrfs_set_acl(NULL, inode, acl, type);
 
 	posix_acl_release(acl);
 
@@ -224,7 +224,8 @@ int btrfs_check_acl(struct inode *inode,
  * stuff has been fixed to work with that.  If the locking stuff changes, we
  * need to re-evaluate the acl locking stuff.
  */
-int btrfs_init_acl(struct inode *inode, struct inode *dir)
+int btrfs_init_acl(struct btrfs_trans_handle *trans,
+		   struct inode *inode, struct inode *dir)
 {
 	struct posix_acl *acl = NULL;
 	int ret = 0;
@@ -249,7 +250,8 @@ int btrfs_init_acl(struct inode *inode,
 		mode_t mode;
 
 		if (S_ISDIR(inode->i_mode)) {
-			ret = btrfs_set_acl(inode, acl, ACL_TYPE_DEFAULT);
+			ret = btrfs_set_acl(trans, inode, acl,
+					    ACL_TYPE_DEFAULT);
 			if (ret)
 				goto failed;
 		}
@@ -264,7 +266,7 @@ int btrfs_init_acl(struct inode *inode,
 			inode->i_mode = mode;
 			if (ret > 0) {
 				/* we need an acl */
-				ret = btrfs_set_acl(inode, clone,
+				ret = btrfs_set_acl(trans, inode, clone,
 						    ACL_TYPE_ACCESS);
 			}
 		}
@@ -297,7 +299,7 @@ int btrfs_acl_chmod(struct inode *inode)
 
 	ret = posix_acl_chmod_masq(clone, inode->i_mode);
 	if (!ret)
-		ret = btrfs_set_acl(inode, clone, ACL_TYPE_ACCESS);
+		ret = btrfs_set_acl(NULL, inode, clone, ACL_TYPE_ACCESS);
 
 	posix_acl_release(clone);
 
@@ -323,7 +325,8 @@ int btrfs_acl_chmod(struct inode *inode)
 	return 0;
 }
 
-int btrfs_init_acl(struct inode *inode, struct inode *dir)
+int btrfs_init_acl(struct btrfs_trans_handle *trans,
+		   struct inode *inode, struct inode *dir)
 {
 	return 0;
 }
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -310,6 +310,9 @@ struct btrfs_header {
 #define BTRFS_MAX_INLINE_DATA_SIZE(r) (BTRFS_LEAF_DATA_SIZE(r) - \
 					sizeof(struct btrfs_item) - \
 					sizeof(struct btrfs_file_extent_item))
+#define BTRFS_MAX_XATTR_SIZE(r)	(BTRFS_LEAF_DATA_SIZE(r) - \
+				 sizeof(struct btrfs_item) -\
+				 sizeof(struct btrfs_dir_item))
 
 
 /*
@@ -2201,9 +2204,10 @@ int btrfs_delete_one_dir_name(struct btr
 			      struct btrfs_path *path,
 			      struct btrfs_dir_item *di);
 int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
-			    struct btrfs_root *root, const char *name,
-			    u16 name_len, const void *data, u16 data_len,
-			    u64 dir);
+			    struct btrfs_root *root,
+			    struct btrfs_path *path, u64 objectid,
+			    const char *name, u16 name_len,
+			    const void *data, u16 data_len);
 struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans,
 					  struct btrfs_root *root,
 					  struct btrfs_path *path, u64 dir,
@@ -2382,7 +2386,8 @@ int btrfs_check_acl(struct inode *inode,
 #else
 #define btrfs_check_acl NULL
 #endif
-int btrfs_init_acl(struct inode *inode, struct inode *dir);
+int btrfs_init_acl(struct btrfs_trans_handle *trans,
+		   struct inode *inode, struct inode *dir);
 int btrfs_acl_chmod(struct inode *inode);
 
 /* relocation.c */
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -68,12 +68,12 @@ static struct btrfs_dir_item *insert_wit
  * into the tree
  */
 int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
-			    struct btrfs_root *root, const char *name,
-			    u16 name_len, const void *data, u16 data_len,
-			    u64 dir)
+			    struct btrfs_root *root,
+			    struct btrfs_path *path, u64 objectid,
+			    const char *name, u16 name_len,
+			    const void *data, u16 data_len)
 {
 	int ret = 0;
-	struct btrfs_path *path;
 	struct btrfs_dir_item *dir_item;
 	unsigned long name_ptr, data_ptr;
 	struct btrfs_key key, location;
@@ -81,15 +81,11 @@ int btrfs_insert_xattr_item(struct btrfs
 	struct extent_buffer *leaf;
 	u32 data_size;
 
-	key.objectid = dir;
+	BUG_ON(name_len + data_len > BTRFS_MAX_XATTR_SIZE(root));
+
+	key.objectid = objectid;
 	btrfs_set_key_type(&key, BTRFS_XATTR_ITEM_KEY);
 	key.offset = btrfs_name_hash(name, name_len);
-	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
-	if (name_len + data_len + sizeof(struct btrfs_dir_item) >
-	    BTRFS_LEAF_DATA_SIZE(root) - sizeof(struct btrfs_item))
-		return -ENOSPC;
 
 	data_size = sizeof(*dir_item) + name_len + data_len;
 	dir_item = insert_with_overflow(trans, root, path, &key, data_size,
@@ -117,7 +113,6 @@ int btrfs_insert_xattr_item(struct btrfs
 	write_extent_buffer(leaf, data, data_ptr, data_len);
 	btrfs_mark_buffer_dirty(path->nodes[0]);
 
-	btrfs_free_path(path);
 	return ret;
 }
 
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -88,13 +88,14 @@ static noinline int cow_file_range(struc
 				   u64 start, u64 end, int *page_started,
 				   unsigned long *nr_written, int unlock);
 
-static int btrfs_init_inode_security(struct inode *inode,  struct inode *dir)
+static int btrfs_init_inode_security(struct btrfs_trans_handle *trans,
+				     struct inode *inode,  struct inode *dir)
 {
 	int err;
 
-	err = btrfs_init_acl(inode, dir);
+	err = btrfs_init_acl(trans, inode, dir);
 	if (!err)
-		err = btrfs_xattr_security_init(inode, dir);
+		err = btrfs_xattr_security_init(trans, inode, dir);
 	return err;
 }
 
@@ -4296,7 +4297,7 @@ static int btrfs_mknod(struct inode *dir
 	if (IS_ERR(inode))
 		goto out_unlock;
 
-	err = btrfs_init_inode_security(inode, dir);
+	err = btrfs_init_inode_security(trans, inode, dir);
 	if (err) {
 		drop_inode = 1;
 		goto out_unlock;
@@ -4367,7 +4368,7 @@ static int btrfs_create(struct inode *di
 	if (IS_ERR(inode))
 		goto out_unlock;
 
-	err = btrfs_init_inode_security(inode, dir);
+	err = btrfs_init_inode_security(trans, inode, dir);
 	if (err) {
 		drop_inode = 1;
 		goto out_unlock;
@@ -4500,7 +4501,7 @@ static int btrfs_mkdir(struct inode *dir
 
 	drop_on_err = 1;
 
-	err = btrfs_init_inode_security(inode, dir);
+	err = btrfs_init_inode_security(trans, inode, dir);
 	if (err)
 		goto out_fail;
 
@@ -5660,7 +5661,7 @@ static int btrfs_symlink(struct inode *d
 	if (IS_ERR(inode))
 		goto out_unlock;
 
-	err = btrfs_init_inode_security(inode, dir);
+	err = btrfs_init_inode_security(trans, inode, dir);
 	if (err) {
 		drop_inode = 1;
 		goto out_unlock;
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -85,22 +85,23 @@ out:
 	return ret;
 }
 
-int __btrfs_setxattr(struct inode *inode, const char *name,
-			    const void *value, size_t size, int flags)
+static int do_setxattr(struct btrfs_trans_handle *trans,
+		       struct inode *inode, const char *name,
+		       const void *value, size_t size, int flags)
 {
 	struct btrfs_dir_item *di;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_trans_handle *trans;
 	struct btrfs_path *path;
-	int ret = 0, mod = 0;
+	size_t name_len = strlen(name);
+	int ret = 0;
+
+	if (name_len + size > BTRFS_MAX_XATTR_SIZE(root))
+		return -ENOSPC;
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
 
-	trans = btrfs_join_transaction(root, 1);
-	btrfs_set_trans_block_group(trans, inode);
-
 	/* first lets see if we already have this xattr */
 	di = btrfs_lookup_xattr(trans, root, path, inode->i_ino, name,
 				strlen(name), -1);
@@ -118,15 +119,12 @@ int __btrfs_setxattr(struct inode *inode
 		}
 
 		ret = btrfs_delete_one_dir_name(trans, root, path, di);
-		if (ret)
-			goto out;
+		BUG_ON(ret);
 		btrfs_release_path(root, path);
 
 		/* if we don't have a value then we are removing the xattr */
-		if (!value) {
-			mod = 1;
+		if (!value)
 			goto out;
-		}
 	} else {
 		btrfs_release_path(root, path);
 
@@ -138,20 +136,45 @@ int __btrfs_setxattr(struct inode *inode
 	}
 
 	/* ok we have to create a completely new xattr */
-	ret = btrfs_insert_xattr_item(trans, root, name, strlen(name),
-				      value, size, inode->i_ino);
+	ret = btrfs_insert_xattr_item(trans, root, path, inode->i_ino,
+				      name, name_len, value, size);
+	BUG_ON(ret);
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
+int __btrfs_setxattr(struct btrfs_trans_handle *trans,
+		     struct inode *inode, const char *name,
+		     const void *value, size_t size, int flags)
+{
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	int ret;
+
+	if (trans)
+		return do_setxattr(trans, inode, name, value, size, flags);
+
+	ret = btrfs_reserve_metadata_space(root, 2);
 	if (ret)
-		goto out;
-	mod = 1;
+		return ret;
 
-out:
-	if (mod) {
-		inode->i_ctime = CURRENT_TIME;
-		ret = btrfs_update_inode(trans, root, inode);
+	trans = btrfs_start_transaction(root, 1);
+	if (!trans) {
+		ret = -ENOMEM;
+		goto out;
 	}
+	btrfs_set_trans_block_group(trans, inode);
 
-	btrfs_end_transaction(trans, root);
-	btrfs_free_path(path);
+	ret = do_setxattr(trans, inode, name, value, size, flags);
+	if (ret)
+		goto out;
+
+	inode->i_ctime = CURRENT_TIME;
+	ret = btrfs_update_inode(trans, root, inode);
+	BUG_ON(ret);
+out:
+	btrfs_end_transaction_throttle(trans, root);
+	btrfs_unreserve_metadata_space(root, 2);
 	return ret;
 }
 
@@ -314,7 +337,9 @@ int btrfs_setxattr(struct dentry *dentry
 
 	if (size == 0)
 		value = "";  /* empty EA, do not remove */
-	return __btrfs_setxattr(dentry->d_inode, name, value, size, flags);
+
+	return __btrfs_setxattr(NULL, dentry->d_inode, name, value, size,
+				flags);
 }
 
 int btrfs_removexattr(struct dentry *dentry, const char *name)
@@ -329,10 +354,13 @@ int btrfs_removexattr(struct dentry *den
 
 	if (!btrfs_is_valid_xattr(name))
 		return -EOPNOTSUPP;
-	return __btrfs_setxattr(dentry->d_inode, name, NULL, 0, XATTR_REPLACE);
+
+	return __btrfs_setxattr(NULL, dentry->d_inode, name, NULL, 0,
+				XATTR_REPLACE);
 }
 
-int btrfs_xattr_security_init(struct inode *inode, struct inode *dir)
+int btrfs_xattr_security_init(struct btrfs_trans_handle *trans,
+			      struct inode *inode, struct inode *dir)
 {
 	int err;
 	size_t len;
@@ -354,7 +382,7 @@ int btrfs_xattr_security_init(struct ino
 	} else {
 		strcpy(name, XATTR_SECURITY_PREFIX);
 		strcpy(name + XATTR_SECURITY_PREFIX_LEN, suffix);
-		err = __btrfs_setxattr(inode, name, value, len, 0);
+		err = __btrfs_setxattr(trans, inode, name, value, len, 0);
 		kfree(name);
 	}
 
--- a/fs/btrfs/xattr.h
+++ b/fs/btrfs/xattr.h
@@ -27,15 +27,16 @@ extern struct xattr_handler *btrfs_xattr
 
 extern ssize_t __btrfs_getxattr(struct inode *inode, const char *name,
 		void *buffer, size_t size);
-extern int __btrfs_setxattr(struct inode *inode, const char *name,
-		const void *value, size_t size, int flags);
-
+extern int __btrfs_setxattr(struct btrfs_trans_handle *trans,
+			    struct inode *inode, const char *name,
+			    const void *value, size_t size, int flags);
 extern ssize_t btrfs_getxattr(struct dentry *dentry, const char *name,
 		void *buffer, size_t size);
 extern int btrfs_setxattr(struct dentry *dentry, const char *name,
 		const void *value, size_t size, int flags);
 extern int btrfs_removexattr(struct dentry *dentry, const char *name);
 
-extern int btrfs_xattr_security_init(struct inode *inode, struct inode *dir);
+extern int btrfs_xattr_security_init(struct btrfs_trans_handle *trans,
+				     struct inode *inode, struct inode *dir);
 
 #endif /* __XATTR__ */

[052/111] Btrfs: Add delayed iput

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 24bbcf0442ee04660a5a030efdbb6d03f1c275cb upstream.

iput() can trigger new transactions if we are dropping the
final reference, so calling it in btrfs_commit_transaction
may end up deadlock. This patch adds delayed iput to avoid
the issue.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/ctree.h        |    7 +++++-
 fs/btrfs/disk-io.c      |    4 +++
 fs/btrfs/extent-tree.c  |    8 +++---
 fs/btrfs/inode.c        |   55 ++++++++++++++++++++++++++++++++++++++++++++++--
 fs/btrfs/ordered-data.c |   10 ++++++--
 fs/btrfs/ordered-data.h |    3 +-
 fs/btrfs/relocation.c   |    4 +--
 fs/btrfs/super.c        |    4 +--
 fs/btrfs/transaction.c  |   13 ++++++++---
 9 files changed, 90 insertions(+), 18 deletions(-)

--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -872,6 +872,9 @@ struct btrfs_fs_info {
 	struct list_head dead_roots;
 	struct list_head caching_block_groups;
 
+	spinlock_t delayed_iput_lock;
+	struct list_head delayed_iputs;
+
 	atomic_t nr_async_submits;
 	atomic_t async_submit_draining;
 	atomic_t nr_async_bios;
@@ -2301,7 +2304,7 @@ int btrfs_truncate_inode_items(struct bt
 			       struct inode *inode, u64 new_size,
 			       u32 min_type);
 
-int btrfs_start_delalloc_inodes(struct btrfs_root *root);
+int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput);
 int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end);
 int btrfs_writepages(struct address_space *mapping,
 		     struct writeback_control *wbc);
@@ -2341,6 +2344,8 @@ int btrfs_orphan_del(struct btrfs_trans_
 void btrfs_orphan_cleanup(struct btrfs_root *root);
 int btrfs_cont_expand(struct inode *inode, loff_t size);
 int btrfs_invalidate_inodes(struct btrfs_root *root);
+void btrfs_add_delayed_iput(struct inode *inode);
+void btrfs_run_delayed_iputs(struct btrfs_root *root);
 extern const struct dentry_operations btrfs_dentry_operations;
 
 /* ioctl.c */
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1476,6 +1476,7 @@ static int cleaner_kthread(void *arg)
 
 		if (!(root->fs_info->sb->s_flags & MS_RDONLY) &&
 		    mutex_trylock(&root->fs_info->cleaner_mutex)) {
+			btrfs_run_delayed_iputs(root);
 			btrfs_clean_old_snapshots(root);
 			mutex_unlock(&root->fs_info->cleaner_mutex);
 		}
@@ -1605,6 +1606,7 @@ struct btrfs_root *open_ctree(struct sup
 	INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
 	INIT_LIST_HEAD(&fs_info->trans_list);
 	INIT_LIST_HEAD(&fs_info->dead_roots);
+	INIT_LIST_HEAD(&fs_info->delayed_iputs);
 	INIT_LIST_HEAD(&fs_info->hashers);
 	INIT_LIST_HEAD(&fs_info->delalloc_inodes);
 	INIT_LIST_HEAD(&fs_info->ordered_operations);
@@ -1613,6 +1615,7 @@ struct btrfs_root *open_ctree(struct sup
 	spin_lock_init(&fs_info->new_trans_lock);
 	spin_lock_init(&fs_info->ref_cache_lock);
 	spin_lock_init(&fs_info->fs_roots_radix_lock);
+	spin_lock_init(&fs_info->delayed_iput_lock);
 
 	init_completion(&fs_info->kobj_unregister);
 	fs_info->tree_root = tree_root;
@@ -2386,6 +2389,7 @@ int btrfs_commit_super(struct btrfs_root
 	int ret;
 
 	mutex_lock(&root->fs_info->cleaner_mutex);
+	btrfs_run_delayed_iputs(root);
 	btrfs_clean_old_snapshots(root);
 	mutex_unlock(&root->fs_info->cleaner_mutex);
 
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -2880,9 +2880,9 @@ static noinline void flush_delalloc_asyn
 	root = async->root;
 	info = async->info;
 
-	btrfs_start_delalloc_inodes(root);
+	btrfs_start_delalloc_inodes(root, 0);
 	wake_up(&info->flush_wait);
-	btrfs_wait_ordered_extents(root, 0);
+	btrfs_wait_ordered_extents(root, 0, 0);
 
 	spin_lock(&info->lock);
 	info->flushing = 0;
@@ -2956,8 +2956,8 @@ static void flush_delalloc(struct btrfs_
 	return;
 
 flush:
-	btrfs_start_delalloc_inodes(root);
-	btrfs_wait_ordered_extents(root, 0);
+	btrfs_start_delalloc_inodes(root, 0);
+	btrfs_wait_ordered_extents(root, 0, 0);
 
 	spin_lock(&info->lock);
 	info->flushing = 0;
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2022,6 +2022,54 @@ zeroit:
 	return -EIO;
 }
 
+struct delayed_iput {
+	struct list_head list;
+	struct inode *inode;
+};
+
+void btrfs_add_delayed_iput(struct inode *inode)
+{
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+	struct delayed_iput *delayed;
+
+	if (atomic_add_unless(&inode->i_count, -1, 1))
+		return;
+
+	delayed = kmalloc(sizeof(*delayed), GFP_NOFS | __GFP_NOFAIL);
+	delayed->inode = inode;
+
+	spin_lock(&fs_info->delayed_iput_lock);
+	list_add_tail(&delayed->list, &fs_info->delayed_iputs);
+	spin_unlock(&fs_info->delayed_iput_lock);
+}
+
+void btrfs_run_delayed_iputs(struct btrfs_root *root)
+{
+	LIST_HEAD(list);
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct delayed_iput *delayed;
+	int empty;
+
+	spin_lock(&fs_info->delayed_iput_lock);
+	empty = list_empty(&fs_info->delayed_iputs);
+	spin_unlock(&fs_info->delayed_iput_lock);
+	if (empty)
+		return;
+
+	down_read(&root->fs_info->cleanup_work_sem);
+	spin_lock(&fs_info->delayed_iput_lock);
+	list_splice_init(&fs_info->delayed_iputs, &list);
+	spin_unlock(&fs_info->delayed_iput_lock);
+
+	while (!list_empty(&list)) {
+		delayed = list_entry(list.next, struct delayed_iput, list);
+		list_del(&delayed->list);
+		iput(delayed->inode);
+		kfree(delayed);
+	}
+	up_read(&root->fs_info->cleanup_work_sem);
+}
+
 /*
  * This creates an orphan entry for the given inode in case something goes
  * wrong in the middle of an unlink/truncate.
@@ -5568,7 +5616,7 @@ out_fail:
  * some fairly slow code that needs optimization. This walks the list
  * of all the inodes with pending delalloc and forces them to disk.
  */
-int btrfs_start_delalloc_inodes(struct btrfs_root *root)
+int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
 {
 	struct list_head *head = &root->fs_info->delalloc_inodes;
 	struct btrfs_inode *binode;
@@ -5587,7 +5635,10 @@ int btrfs_start_delalloc_inodes(struct b
 		spin_unlock(&root->fs_info->delalloc_lock);
 		if (inode) {
 			filemap_flush(inode->i_mapping);
-			iput(inode);
+			if (delay_iput)
+				btrfs_add_delayed_iput(inode);
+			else
+				iput(inode);
 		}
 		cond_resched();
 		spin_lock(&root->fs_info->delalloc_lock);
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -352,7 +352,8 @@ int btrfs_remove_ordered_extent(struct i
  * wait for all the ordered extents in a root.  This is done when balancing
  * space between drives.
  */
-int btrfs_wait_ordered_extents(struct btrfs_root *root, int nocow_only)
+int btrfs_wait_ordered_extents(struct btrfs_root *root,
+			       int nocow_only, int delay_iput)
 {
 	struct list_head splice;
 	struct list_head *cur;
@@ -389,7 +390,10 @@ int btrfs_wait_ordered_extents(struct bt
 		if (inode) {
 			btrfs_start_ordered_extent(inode, ordered, 1);
 			btrfs_put_ordered_extent(ordered);
-			iput(inode);
+			if (delay_iput)
+				btrfs_add_delayed_iput(inode);
+			else
+				iput(inode);
 		} else {
 			btrfs_put_ordered_extent(ordered);
 		}
@@ -447,7 +451,7 @@ again:
 				btrfs_wait_ordered_range(inode, 0, (u64)-1);
 			else
 				filemap_flush(inode->i_mapping);
-			iput(inode);
+			btrfs_add_delayed_iput(inode);
 		}
 
 		cond_resched();
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -153,9 +153,10 @@ btrfs_lookup_first_ordered_extent(struct
 int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 				struct btrfs_ordered_extent *ordered);
 int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr, u32 *sum);
-int btrfs_wait_ordered_extents(struct btrfs_root *root, int nocow_only);
 int btrfs_run_ordered_operations(struct btrfs_root *root, int wait);
 int btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root,
 				struct inode *inode);
+int btrfs_wait_ordered_extents(struct btrfs_root *root,
+			       int nocow_only, int delay_iput);
 #endif
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3541,8 +3541,8 @@ int btrfs_relocate_block_group(struct bt
 	       (unsigned long long)rc->block_group->key.objectid,
 	       (unsigned long long)rc->block_group->flags);
 
-	btrfs_start_delalloc_inodes(fs_info->tree_root);
-	btrfs_wait_ordered_extents(fs_info->tree_root, 0);
+	btrfs_start_delalloc_inodes(fs_info->tree_root, 0);
+	btrfs_wait_ordered_extents(fs_info->tree_root, 0, 0);
 
 	while (1) {
 		rc->extents_found = 0;
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -405,8 +405,8 @@ int btrfs_sync_fs(struct super_block *sb
 		return 0;
 	}
 
-	btrfs_start_delalloc_inodes(root);
-	btrfs_wait_ordered_extents(root, 0);
+	btrfs_start_delalloc_inodes(root, 0);
+	btrfs_wait_ordered_extents(root, 0, 0);
 
 	trans = btrfs_start_transaction(root, 1);
 	ret = btrfs_commit_transaction(trans, root);
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -333,6 +333,9 @@ static int __btrfs_end_transaction(struc
 	memset(trans, 0, sizeof(*trans));
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
 
+	if (throttle)
+		btrfs_run_delayed_iputs(root);
+
 	return 0;
 }
 
@@ -991,11 +994,11 @@ int btrfs_commit_transaction(struct btrf
 		mutex_unlock(&root->fs_info->trans_mutex);
 
 		if (flush_on_commit) {
-			btrfs_start_delalloc_inodes(root);
-			ret = btrfs_wait_ordered_extents(root, 0);
+			btrfs_start_delalloc_inodes(root, 1);
+			ret = btrfs_wait_ordered_extents(root, 0, 1);
 			BUG_ON(ret);
 		} else if (snap_pending) {
-			ret = btrfs_wait_ordered_extents(root, 1);
+			ret = btrfs_wait_ordered_extents(root, 0, 1);
 			BUG_ON(ret);
 		}
 
@@ -1113,6 +1116,10 @@ int btrfs_commit_transaction(struct btrf
 		current->journal_info = NULL;
 
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
+
+	if (current != root->fs_info->transaction_kthread)
+		btrfs_run_delayed_iputs(root);
+
 	return ret;
 }
 

[053/111] Btrfs: Fix btrfs_drop_extent_cache for skip pinned case

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 55ef68990029fcd8d04d42fc184aa7fb18cf309e upstream.

The check for skip pinned case is wrong, it may breaks the
while loop too soon.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/file.c |   10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -179,18 +179,14 @@ int btrfs_drop_extent_cache(struct inode
 		}
 		flags = em->flags;
 		if (skip_pinned && test_bit(EXTENT_FLAG_PINNED, &em->flags)) {
-			if (em->start <= start &&
-			    (!testend || em->start + em->len >= start + len)) {
+			if (testend && em->start + em->len >= start + len) {
 				free_extent_map(em);
 				write_unlock(&em_tree->lock);
 				break;
 			}
-			if (start < em->start) {
-				len = em->start - start;
-			} else {
+			start = em->start + em->len;
+			if (testend)
 				len = start + len - (em->start + em->len);
-				start = em->start + em->len;
-			}
 			free_extent_map(em);
 			write_unlock(&em_tree->lock);
 			continue;

[054/111] Btrfs: Fix per root used space accounting

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 86b9f2eca5e0984145e3c7698a7cd6dd65c2a93f upstream.

The bytes_used field in root item was originally planned to
trace the amount of used data and tree blocks. But it never
worked right since we can't trace freeing of data accurately.
This patch changes it to only trace the amount of tree blocks.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/ctree.c       |   31 ++++++++++++++-----------------
 fs/btrfs/ctree.h       |    4 ++++
 fs/btrfs/extent-tree.c |   31 +++++++++++++++++++++++--------
 fs/btrfs/ioctl.c       |    2 +-
 fs/btrfs/transaction.c |    6 +++++-
 5 files changed, 47 insertions(+), 27 deletions(-)

--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -456,9 +456,8 @@ static noinline int __btrfs_cow_block(st
 		extent_buffer_get(cow);
 		spin_unlock(&root->node_lock);
 
-		btrfs_free_extent(trans, root, buf->start, buf->len,
-				  parent_start, root->root_key.objectid,
-				  level, 0);
+		btrfs_free_tree_block(trans, root, buf->start, buf->len,
+				parent_start, root->root_key.objectid, level);
 		free_extent_buffer(buf);
 		add_root_to_dirty_list(root);
 	} else {
@@ -473,9 +472,8 @@ static noinline int __btrfs_cow_block(st
 		btrfs_set_node_ptr_generation(parent, parent_slot,
 					      trans->transid);
 		btrfs_mark_buffer_dirty(parent);
-		btrfs_free_extent(trans, root, buf->start, buf->len,
-				  parent_start, root->root_key.objectid,
-				  level, 0);
+		btrfs_free_tree_block(trans, root, buf->start, buf->len,
+				parent_start, root->root_key.objectid, level);
 	}
 	if (unlock_orig)
 		btrfs_tree_unlock(buf);
@@ -1035,8 +1033,8 @@ static noinline int balance_level(struct
 		btrfs_tree_unlock(mid);
 		/* once for the path */
 		free_extent_buffer(mid);
-		ret = btrfs_free_extent(trans, root, mid->start, mid->len,
-					0, root->root_key.objectid, level, 1);
+		ret = btrfs_free_tree_block(trans, root, mid->start, mid->len,
+					    0, root->root_key.objectid, level);
 		/* once for the root ptr */
 		free_extent_buffer(mid);
 		return ret;
@@ -1100,10 +1098,10 @@ static noinline int balance_level(struct
 				       1);
 			if (wret)
 				ret = wret;
-			wret = btrfs_free_extent(trans, root, bytenr,
-						 blocksize, 0,
-						 root->root_key.objectid,
-						 level, 0);
+			wret = btrfs_free_tree_block(trans, root,
+						     bytenr, blocksize, 0,
+						     root->root_key.objectid,
+						     level);
 			if (wret)
 				ret = wret;
 		} else {
@@ -1148,9 +1146,8 @@ static noinline int balance_level(struct
 		wret = del_ptr(trans, root, path, level + 1, pslot);
 		if (wret)
 			ret = wret;
-		wret = btrfs_free_extent(trans, root, bytenr, blocksize,
-					 0, root->root_key.objectid,
-					 level, 0);
+		wret = btrfs_free_tree_block(trans, root, bytenr, blocksize,
+					 0, root->root_key.objectid, level);
 		if (wret)
 			ret = wret;
 	} else {
@@ -3794,8 +3791,8 @@ static noinline int btrfs_del_leaf(struc
 	 */
 	btrfs_unlock_up_safe(path, 0);
 
-	ret = btrfs_free_extent(trans, root, leaf->start, leaf->len,
-				0, root->root_key.objectid, 0, 0);
+	ret = btrfs_free_tree_block(trans, root, leaf->start, leaf->len,
+				    0, root->root_key.objectid, 0);
 	return ret;
 }
 /*
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1982,6 +1982,10 @@ struct extent_buffer *btrfs_alloc_free_b
 					u64 parent, u64 root_objectid,
 					struct btrfs_disk_key *key, int level,
 					u64 hint, u64 empty_size);
+int btrfs_free_tree_block(struct btrfs_trans_handle *trans,
+			  struct btrfs_root *root,
+			  u64 bytenr, u32 blocksize,
+			  u64 parent, u64 root_objectid, int level);
 struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
 					    struct btrfs_root *root,
 					    u64 bytenr, u32 blocksize,
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3454,14 +3454,6 @@ static int update_block_group(struct btr
 	else
 		old_val -= num_bytes;
 	btrfs_set_super_bytes_used(&info->super_copy, old_val);
-
-	/* block accounting for root item */
-	old_val = btrfs_root_used(&root->root_item);
-	if (alloc)
-		old_val += num_bytes;
-	else
-		old_val -= num_bytes;
-	btrfs_set_root_used(&root->root_item, old_val);
 	spin_unlock(&info->delalloc_lock);
 
 	while (total) {
@@ -4049,6 +4041,21 @@ int btrfs_free_extent(struct btrfs_trans
 	return ret;
 }
 
+int btrfs_free_tree_block(struct btrfs_trans_handle *trans,
+			  struct btrfs_root *root,
+			  u64 bytenr, u32 blocksize,
+			  u64 parent, u64 root_objectid, int level)
+{
+	u64 used;
+	spin_lock(&root->node_lock);
+	used = btrfs_root_used(&root->root_item) - blocksize;
+	btrfs_set_root_used(&root->root_item, used);
+	spin_unlock(&root->node_lock);
+
+	return btrfs_free_extent(trans, root, bytenr, blocksize,
+				 parent, root_objectid, level, 0);
+}
+
 static u64 stripe_align(struct btrfs_root *root, u64 val)
 {
 	u64 mask = ((u64)root->stripesize - 1);
@@ -4897,6 +4904,14 @@ static int alloc_tree_block(struct btrfs
 					extent_op);
 		BUG_ON(ret);
 	}
+
+	if (root_objectid == root->root_key.objectid) {
+		u64 used;
+		spin_lock(&root->node_lock);
+		used = btrfs_root_used(&root->root_item) + num_bytes;
+		btrfs_set_root_used(&root->root_item, used);
+		spin_unlock(&root->node_lock);
+	}
 	return ret;
 }
 
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -289,7 +289,7 @@ static noinline int create_subvol(struct
 	btrfs_set_root_generation(&root_item, trans->transid);
 	btrfs_set_root_level(&root_item, 0);
 	btrfs_set_root_refs(&root_item, 1);
-	btrfs_set_root_used(&root_item, 0);
+	btrfs_set_root_used(&root_item, leaf->len);
 	btrfs_set_root_last_snapshot(&root_item, 0);
 
 	memset(&root_item.drop_progress, 0, sizeof(root_item.drop_progress));
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -501,13 +501,16 @@ static int update_cowonly_root(struct bt
 {
 	int ret;
 	u64 old_root_bytenr;
+	u64 old_root_used;
 	struct btrfs_root *tree_root = root->fs_info->tree_root;
 
+	old_root_used = btrfs_root_used(&root->root_item);
 	btrfs_write_dirty_block_groups(trans, root);
 
 	while (1) {
 		old_root_bytenr = btrfs_root_bytenr(&root->root_item);
-		if (old_root_bytenr == root->node->start)
+		if (old_root_bytenr == root->node->start &&
+		    old_root_used == btrfs_root_used(&root->root_item))
 			break;
 
 		btrfs_set_root_node(&root->root_item, root->node);
@@ -516,6 +519,7 @@ static int update_cowonly_root(struct bt
 					&root->root_item);
 		BUG_ON(ret);
 
+		old_root_used = btrfs_root_used(&root->root_item);
 		ret = btrfs_write_dirty_block_groups(trans, root);
 		BUG_ON(ret);
 	}

[055/111] Btrfs: dont add extent 0 to the free space cache v2

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 06b2331f8333ec6edf41662757ce8882cc1747d5 upstream.

If block group 0 is completely free, btrfs_read_block_groups will
add extent [0, BTRFS_SUPER_INFO_OFFSET) to the free space cache.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/extent-tree.c |   10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -195,6 +195,14 @@ static int exclude_super_stripes(struct
 	int stripe_len;
 	int i, nr, ret;
 
+	if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
+		stripe_len = BTRFS_SUPER_INFO_OFFSET - cache->key.objectid;
+		cache->bytes_super += stripe_len;
+		ret = add_excluded_extent(root, cache->key.objectid,
+					  stripe_len);
+		BUG_ON(ret);
+	}
+
 	for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
 		bytenr = btrfs_sb_offset(i);
 		ret = btrfs_rmap_block(&root->fs_info->mapping_tree,
@@ -255,7 +263,7 @@ static u64 add_new_free_space(struct btr
 		if (ret)
 			break;
 
-		if (extent_start == start) {
+		if (extent_start <= start) {
 			start = extent_end + 1;
 		} else if (extent_start > start && extent_start < end) {
 			size = extent_start - start;

[056/111] Btrfs: fail mount on bad mount options

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Sage Weil <sage@newdream.net>

commit a7a3f7cadd9bdee569243f7ead9550aa16b60e07 upstream.

We shouldn't silently ignore unrecognized options.

Signed-off-by: Sage Weil <sage@newdream.net>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/super.c |    9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -128,6 +128,7 @@ int btrfs_parse_options(struct btrfs_roo
 	substring_t args[MAX_OPT_ARGS];
 	char *p, *num;
 	int intarg;
+	int ret = 0;
 
 	if (!options)
 		return 0;
@@ -262,12 +263,18 @@ int btrfs_parse_options(struct btrfs_roo
 		case Opt_discard:
 			btrfs_set_opt(info->mount_opt, DISCARD);
 			break;
+		case Opt_err:
+			printk(KERN_INFO "btrfs: unrecognized mount option "
+			       "'%s'\n", p);
+			ret = -EINVAL;
+			goto out;
 		default:
 			break;
 		}
 	}
+out:
 	kfree(options);
-	return 0;
+	return ret;
 }
 
 /*

[057/111] Btrfs: deny sys_link across subvolumes.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: TARUISI Hiroaki <taruishi.hiroak@jp.fujitsu.com>

commit 4a8be425a8fb8fbb5d881eb55fa6634c3463b9c9 upstream.

I rebased Christian Parpart's patch to deny hard link across
subvolumes. Original patch modifies also btrfs_rename, but
I excluded it because we can move across subvolumes now and
it make no problem.
-----------------

Hard link across subvolumes should not allowed in Btrfs.
btrfs_link checks root of 'to' directory is same as root
of 'from' file. If not same, btrfs_link returns -EPERM.

Signed-off-by: TARUISI Hiroaki <taruishi.hiroak@jp.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/inode.c |    4 ++++
 1 file changed, 4 insertions(+)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4462,6 +4462,10 @@ static int btrfs_link(struct dentry *old
 	if (inode->i_nlink == 0)
 		return -ENOENT;
 
+	/* do not allow sys_link's with other subvols of the same device */
+	if (root->objectid != BTRFS_I(inode)->root->objectid)
+		return -EPERM;
+
 	/*
 	 * 1 item for inode ref
 	 * 2 items for dir items

[058/111] Btrfs: Show discard option in /proc/mounts

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Matthew Wilcox <matthew@wil.cx>

commit 20a5239a5d0f340e29827a6a2d28a138001c44b8 upstream.

Christoph's patch e244a0aeb6a599c19a7c802cda6e2d67c847b154 doesn't display
the discard option in /proc/mounts, leading to some confusion for me.
Here's the missing bit.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/super.c |    2 ++
 1 file changed, 2 insertions(+)

--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -457,6 +457,8 @@ static int btrfs_show_options(struct seq
 		seq_puts(seq, ",notreelog");
 	if (btrfs_test_opt(root, FLUSHONCOMMIT))
 		seq_puts(seq, ",flushoncommit");
+	if (btrfs_test_opt(root, DISCARD))
+		seq_puts(seq, ",discard");
 	if (!(root->fs_info->sb->s_flags & MS_POSIXACL))
 		seq_puts(seq, ",noacl");
 	return 0;

[059/111] Btrfs: make metadata chunks smaller

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit 83d3c9696fed237a3d96fce18299e2fcf112109f upstream.

This patch makes us a bit less zealous about making sure we have enough free
metadata space by pearing down the size of new metadata chunks to 256mb instead
of 1gb.  Also, we used to try an allocate metadata chunks when allocating data,
but that sort of thing is done elsewhere now so we can just remove it.  With my
-ENOSPC test I used to have 3gb reserved for metadata out of 75gb, now I have
1.7gb.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/extent-tree.c |   11 +----------
 fs/btrfs/volumes.c     |    2 +-
 2 files changed, 2 insertions(+), 11 deletions(-)

--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -4593,7 +4593,6 @@ int btrfs_reserve_extent(struct btrfs_tr
 {
 	int ret;
 	u64 search_start = 0;
-	struct btrfs_fs_info *info = root->fs_info;
 
 	data = btrfs_get_alloc_profile(root, data);
 again:
@@ -4601,17 +4600,9 @@ again:
 	 * the only place that sets empty_size is btrfs_realloc_node, which
 	 * is not called recursively on allocations
 	 */
-	if (empty_size || root->ref_cows) {
-		if (!(data & BTRFS_BLOCK_GROUP_METADATA)) {
-			ret = do_chunk_alloc(trans, root->fs_info->extent_root,
-				     2 * 1024 * 1024,
-				     BTRFS_BLOCK_GROUP_METADATA |
-				     (info->metadata_alloc_profile &
-				      info->avail_metadata_alloc_bits), 0);
-		}
+	if (empty_size || root->ref_cows)
 		ret = do_chunk_alloc(trans, root->fs_info->extent_root,
 				     num_bytes + 2 * 1024 * 1024, data, 0);
-	}
 
 	WARN_ON(num_bytes < root->sectorsize);
 	ret = find_free_extent(trans, root, num_bytes, empty_size,
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -2209,7 +2209,7 @@ static int __btrfs_alloc_chunk(struct bt
 		max_chunk_size = 10 * calc_size;
 		min_stripe_size = 64 * 1024 * 1024;
 	} else if (type & BTRFS_BLOCK_GROUP_METADATA) {
-		max_chunk_size = 4 * calc_size;
+		max_chunk_size = 256 * 1024 * 1024;
 		min_stripe_size = 32 * 1024 * 1024;
 	} else if (type & BTRFS_BLOCK_GROUP_SYSTEM) {
 		calc_size = 8 * 1024 * 1024;

[060/111] Btrfs: make sure fallocate properly starts a transaction

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Chris Mason <chris.mason@oracle.com>

commit 3a1abec9f6880cf406593c392636199ea1c6c917 upstream.

The recent patch to make fallocate enospc friendly would send
down a NULL trans handle to the allocator.  This moves the
transaction start to properly fix things.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/inode.c |   13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5802,23 +5802,23 @@ static int prealloc_file_range(struct in
 	while (num_bytes > 0) {
 		alloc_size = min(num_bytes, root->fs_info->max_extent);
 
+		trans = btrfs_start_transaction(root, 1);
+
 		ret = btrfs_reserve_extent(trans, root, alloc_size,
 					   root->sectorsize, 0, alloc_hint,
 					   (u64)-1, &ins, 1);
 		if (ret) {
 			WARN_ON(1);
-			break;
+			goto stop_trans;
 		}
 
 		ret = btrfs_reserve_metadata_space(root, 3);
 		if (ret) {
 			btrfs_free_reserved_extent(root, ins.objectid,
 						   ins.offset);
-			break;
+			goto stop_trans;
 		}
 
-		trans = btrfs_start_transaction(root, 1);
-
 		ret = insert_reserved_file_extent(trans, inode,
 						  cur_offset, ins.objectid,
 						  ins.offset, ins.offset,
@@ -5847,6 +5847,11 @@ static int prealloc_file_range(struct in
 		btrfs_unreserve_metadata_space(root, 3);
 	}
 	return ret;
+
+stop_trans:
+	btrfs_end_transaction(trans, root);
+	return ret;
+
 }
 
 static long btrfs_fallocate(struct inode *inode, int mode,

[061/111] btrfs: fix missing last-entry in readdir(3)

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Jan Engelhardt <jengelh@medozas.de>

commit 406266ab9ac8ed8b085c58aacd9e3161480dc5d5 upstream.

parent 49313cdac7b34c9f7ecbb1780cfc648b1c082cd7 (v2.6.32-1-g49313cd)
commit ff48c08e1c05c67e8348ab6f8a24de8034e0e34d
Author: Jan Engelhardt <jengelh@medozas.de>
Date:   Wed Dec 9 22:57:36 2009 +0100

Btrfs: fix missing last-entry in readdir(3)

When one does a 32-bit readdir(3), the last entry of a directory is
missing. This is however not due to passing a large value to filldir,
but it seems to have to do with glibc doing telldir or something
quirky. In any case, this patch fixes it in practice.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/inode.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3995,7 +3995,11 @@ skip:
 
 	/* Reached end of directory/root. Bump pos past the last item. */
 	if (key_type == BTRFS_DIR_INDEX_KEY)
-		filp->f_pos = INT_LIMIT(off_t);
+		/*
+		 * 32-bit glibc will use getdents64, but then strtol -
+		 * so the last number we can serve is this.
+		 */
+		filp->f_pos = 0x7fffffff;
 	else
 		filp->f_pos++;
 nopos:

[062/111] Btrfs: align offsets for btrfs_ordered_update_i_size

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <yanzheng@21cn.com>

commit a038fab0cb873c75d6675e2bcffce8a3935bdce7 upstream.

Some callers of btrfs_ordered_update_i_size can now pass in
a NULL for the ordered extent to update against.  This makes
sure we properly align the offset they pass in when deciding
how much to bump the on disk i_size.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/ordered-data.c |    2 ++
 1 file changed, 2 insertions(+)

--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -626,6 +626,8 @@ int btrfs_ordered_update_i_size(struct i
 
 	if (ordered)
 		offset = entry_end(ordered);
+	else
+		offset = ALIGN(offset, BTRFS_I(inode)->root->sectorsize);
 
 	mutex_lock(&tree->mutex);
 	disk_i_size = BTRFS_I(inode)->disk_i_size;

[063/111] Btrfs, fix memory leaks in error paths

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Jiri Slaby <jslaby@suse.cz>

commit 2423fdfb96e3f9ff3baeb6c4c78d74145547891d upstream.

Stanse found 2 memory leaks in relocate_block_group and
__btrfs_map_block. cluster and multi are not freed/assigned on all
paths. Fix that.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: linux-btrfs@vger.kernel.org
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/relocation.c |    4 +++-
 fs/btrfs/volumes.c    |    4 +++-
 2 files changed, 6 insertions(+), 2 deletions(-)

--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3281,8 +3281,10 @@ static noinline_for_stack int relocate_b
 		return -ENOMEM;
 
 	path = btrfs_alloc_path();
-	if (!path)
+	if (!path) {
+		kfree(cluster);
 		return -ENOMEM;
+	}
 
 	rc->extents_found = 0;
 	rc->extents_skipped = 0;
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -2649,8 +2649,10 @@ again:
 	em = lookup_extent_mapping(em_tree, logical, *length);
 	read_unlock(&em_tree->lock);
 
-	if (!em && unplug_page)
+	if (!em && unplug_page) {
+		kfree(multi);
 		return 0;
+	}
 
 	if (!em) {
 		printk(KERN_CRIT "unable to find logical %llu len %llu\n",

[064/111] Btrfs: Fix race in btrfs_mark_extent_written

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 6c7d54ac87f338c479d9729e8392eca3f76e11e1 upstream.

Fix bug reported by Johannes Hirte. The reason of that bug
is btrfs_del_items is called after btrfs_duplicate_item and
btrfs_del_items triggers tree balance. The fix is check that
case and call btrfs_search_slot when needed.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/file.c |  100 ++++++++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 80 insertions(+), 20 deletions(-)

--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -506,7 +506,8 @@ next_slot:
 }
 
 static int extent_mergeable(struct extent_buffer *leaf, int slot,
-			    u64 objectid, u64 bytenr, u64 *start, u64 *end)
+			    u64 objectid, u64 bytenr, u64 orig_offset,
+			    u64 *start, u64 *end)
 {
 	struct btrfs_file_extent_item *fi;
 	struct btrfs_key key;
@@ -522,6 +523,7 @@ static int extent_mergeable(struct exten
 	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
 	if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG ||
 	    btrfs_file_extent_disk_bytenr(leaf, fi) != bytenr ||
+	    btrfs_file_extent_offset(leaf, fi) != key.offset - orig_offset ||
 	    btrfs_file_extent_compression(leaf, fi) ||
 	    btrfs_file_extent_encryption(leaf, fi) ||
 	    btrfs_file_extent_other_encoding(leaf, fi))
@@ -561,6 +563,7 @@ int btrfs_mark_extent_written(struct btr
 	u64 split;
 	int del_nr = 0;
 	int del_slot = 0;
+	int recow;
 	int ret;
 
 	btrfs_drop_extent_cache(inode, start, end - 1, 0);
@@ -568,6 +571,7 @@ int btrfs_mark_extent_written(struct btr
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 again:
+	recow = 0;
 	split = start;
 	key.objectid = inode->i_ino;
 	key.type = BTRFS_EXTENT_DATA_KEY;
@@ -591,12 +595,60 @@ again:
 	bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
 	num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
 	orig_offset = key.offset - btrfs_file_extent_offset(leaf, fi);
+	memcpy(&new_key, &key, sizeof(new_key));
+
+	if (start == key.offset && end < extent_end) {
+		other_start = 0;
+		other_end = start;
+		if (extent_mergeable(leaf, path->slots[0] - 1,
+				     inode->i_ino, bytenr, orig_offset,
+				     &other_start, &other_end)) {
+			new_key.offset = end;
+			btrfs_set_item_key_safe(trans, root, path, &new_key);
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							extent_end - end);
+			btrfs_set_file_extent_offset(leaf, fi,
+						     end - orig_offset);
+			fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							end - other_start);
+			btrfs_mark_buffer_dirty(leaf);
+			goto out;
+		}
+	}
+
+	if (start > key.offset && end == extent_end) {
+		other_start = end;
+		other_end = 0;
+		if (extent_mergeable(leaf, path->slots[0] + 1,
+				     inode->i_ino, bytenr, orig_offset,
+				     &other_start, &other_end)) {
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							start - key.offset);
+			path->slots[0]++;
+			new_key.offset = start;
+			btrfs_set_item_key_safe(trans, root, path, &new_key);
+
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							other_end - start);
+			btrfs_set_file_extent_offset(leaf, fi,
+						     start - orig_offset);
+			btrfs_mark_buffer_dirty(leaf);
+			goto out;
+		}
+	}
 
 	while (start > key.offset || end < extent_end) {
 		if (key.offset == start)
 			split = end;
 
-		memcpy(&new_key, &key, sizeof(new_key));
 		new_key.offset = split;
 		ret = btrfs_duplicate_item(trans, root, path, &new_key);
 		if (ret == -EAGAIN) {
@@ -631,15 +683,18 @@ again:
 			path->slots[0]--;
 			extent_end = end;
 		}
+		recow = 1;
 	}
 
-	fi = btrfs_item_ptr(leaf, path->slots[0],
-			    struct btrfs_file_extent_item);
-
 	other_start = end;
 	other_end = 0;
-	if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
-			     bytenr, &other_start, &other_end)) {
+	if (extent_mergeable(leaf, path->slots[0] + 1,
+			     inode->i_ino, bytenr, orig_offset,
+			     &other_start, &other_end)) {
+		if (recow) {
+			btrfs_release_path(root, path);
+			goto again;
+		}
 		extent_end = other_end;
 		del_slot = path->slots[0] + 1;
 		del_nr++;
@@ -650,8 +705,13 @@ again:
 	}
 	other_start = 0;
 	other_end = start;
-	if (extent_mergeable(leaf, path->slots[0] - 1, inode->i_ino,
-			     bytenr, &other_start, &other_end)) {
+	if (extent_mergeable(leaf, path->slots[0] - 1,
+			     inode->i_ino, bytenr, orig_offset,
+			     &other_start, &other_end)) {
+		if (recow) {
+			btrfs_release_path(root, path);
+			goto again;
+		}
 		key.offset = other_start;
 		del_slot = path->slots[0];
 		del_nr++;
@@ -660,22 +720,22 @@ again:
 					inode->i_ino, orig_offset);
 		BUG_ON(ret);
 	}
+	fi = btrfs_item_ptr(leaf, path->slots[0],
+			   struct btrfs_file_extent_item);
 	if (del_nr == 0) {
 		btrfs_set_file_extent_type(leaf, fi,
 					   BTRFS_FILE_EXTENT_REG);
 		btrfs_mark_buffer_dirty(leaf);
-		goto out;
-	}
-
-	fi = btrfs_item_ptr(leaf, del_slot - 1,
-			    struct btrfs_file_extent_item);
-	btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG);
-	btrfs_set_file_extent_num_bytes(leaf, fi,
-					extent_end - key.offset);
-	btrfs_mark_buffer_dirty(leaf);
+	} else {
+		btrfs_set_file_extent_type(leaf, fi,
+					   BTRFS_FILE_EXTENT_REG);
+		btrfs_set_file_extent_num_bytes(leaf, fi,
+						extent_end - key.offset);
+		btrfs_mark_buffer_dirty(leaf);
 
-	ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
-	BUG_ON(ret);
+		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
+		BUG_ON(ret);
+	}
 out:
 	btrfs_free_path(path);
 	return 0;

[065/111] Btrfs: fix regression in orphan cleanup

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit 6c090a11e1c403b727a6a8eff0b97d5fb9e95cb5 upstream.

Currently orphan cleanup only ever gets triggered if we cross subvolumes during
a lookup, which means that if we just mount a plain jane fs that has orphans in
it, they will never get cleaned up.  This results in panic's like these

http://www.kerneloops.org/oops.php?number=1109085

where adding an orphan entry results in -EEXIST being returned and we panic.  In
order to fix this, we check to see on lookup if our root has had the orphan
cleanup done, and if not go ahead and do it.  This is easily reproduceable by
running this testcase

#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <string.h>
#include <unistd.h>
#include <stdio.h>

int main(int argc, char **argv)
{
	char data[4096];
	char newdata[4096];
	int fd1, fd2;

	memset(data, 'a', 4096);
	memset(newdata, 'b', 4096);

	while (1) {
		int i;

		fd1 = creat("file1", 0666);
		if (fd1 < 0)
			break;

		for (i = 0; i < 512; i++)
			write(fd1, data, 4096);

		fsync(fd1);
		close(fd1);

		fd2 = creat("file2", 0666);
		if (fd2 < 0)
			break;

		ftruncate(fd2, 4096 * 512);

		for (i = 0; i < 512; i++)
			write(fd2, newdata, 4096);
		close(fd2);

		i = rename("file2", "file1");
		unlink("file1");
	}

	return 0;
}

and then pulling the power on the box, and then trying to run that test again
when the box comes back up.  I've tested this locally and it fixes the problem.
Thanks to Tomas Carnecky for helping me track this down initially.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/inode.c |    6 ++++++
 1 file changed, 6 insertions(+)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3796,6 +3796,12 @@ struct inode *btrfs_lookup_dentry(struct
 
 	if (location.type == BTRFS_INODE_ITEM_KEY) {
 		inode = btrfs_iget(dir->i_sb, &location, root);
+		if (unlikely(root->clean_orphans) &&
+		    !(inode->i_sb->s_flags & MS_RDONLY)) {
+			down_read(&root->fs_info->cleanup_work_sem);
+			btrfs_orphan_cleanup(root);
+			up_read(&root->fs_info->cleanup_work_sem);
+		}
 		return inode;
 	}
 

[066/111] Btrfs: deal with NULL acl sent to btrfs_set_acl

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Chris Mason <chris.mason@oracle.com>

commit a9cc71a60c29a09174bee2fcef8f924c529fd4b7 upstream.

It is legal for btrfs_set_acl to be sent a NULL acl.  This
makes sure we don't dereference it.  A similar patch was sent by
Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de>

Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/acl.c |   12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -112,12 +112,14 @@ static int btrfs_set_acl(struct btrfs_tr
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		mode = inode->i_mode;
-		ret = posix_acl_equiv_mode(acl, &mode);
-		if (ret < 0)
-			return ret;
-		ret = 0;
-		inode->i_mode = mode;
 		name = POSIX_ACL_XATTR_ACCESS;
+		if (acl) {
+			ret = posix_acl_equiv_mode(acl, &mode);
+			if (ret < 0)
+				return ret;
+			inode->i_mode = mode;
+		}
+		ret = 0;
 		break;
 	case ACL_TYPE_DEFAULT:
 		if (!S_ISDIR(inode->i_mode))

[067/111] Btrfs: fix possible panic on unmount

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit 11dfe35a0108097f2df1f042c485fa7f758c2cdf upstream.

We can race with the unmount of an fs and the stopping of a kthread where we
will free the block group before we're done using it.  The reason for this is
because we do not hold a reference on the block group while its caching, since
the allocator drops its reference once it exits or moves on to the next block
group.  This patch fixes the problem by taking a reference to the block group
before we start caching and dropping it when we're done to make sure all
accesses to the block group are safe.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/extent-tree.c |   32 +++++++++++++++++++-------------
 1 file changed, 19 insertions(+), 13 deletions(-)

--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -83,6 +83,17 @@ static int block_group_bits(struct btrfs
 	return (cache->flags & bits) == bits;
 }
 
+void btrfs_get_block_group(struct btrfs_block_group_cache *cache)
+{
+	atomic_inc(&cache->count);
+}
+
+void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
+{
+	if (atomic_dec_and_test(&cache->count))
+		kfree(cache);
+}
+
 /*
  * this adds the block group to the fs_info rb tree for the block group
  * cache
@@ -156,7 +167,7 @@ block_group_cache_tree_search(struct btr
 		}
 	}
 	if (ret)
-		atomic_inc(&ret->count);
+		btrfs_get_block_group(ret);
 	spin_unlock(&info->block_group_cache_lock);
 
 	return ret;
@@ -407,6 +418,8 @@ err:
 
 	put_caching_control(caching_ctl);
 	atomic_dec(&block_group->space_info->caching_threads);
+	btrfs_put_block_group(block_group);
+
 	return 0;
 }
 
@@ -447,6 +460,7 @@ static int cache_block_group(struct btrf
 	up_write(&fs_info->extent_commit_sem);
 
 	atomic_inc(&cache->space_info->caching_threads);
+	btrfs_get_block_group(cache);
 
 	tsk = kthread_run(caching_kthread, cache, "btrfs-cache-%llu\n",
 			  cache->key.objectid);
@@ -486,12 +500,6 @@ struct btrfs_block_group_cache *btrfs_lo
 	return cache;
 }
 
-void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
-{
-	if (atomic_dec_and_test(&cache->count))
-		kfree(cache);
-}
-
 static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
 						  u64 flags)
 {
@@ -2582,7 +2590,7 @@ next_block_group(struct btrfs_root *root
 	if (node) {
 		cache = rb_entry(node, struct btrfs_block_group_cache,
 				 cache_node);
-		atomic_inc(&cache->count);
+		btrfs_get_block_group(cache);
 	} else
 		cache = NULL;
 	spin_unlock(&root->fs_info->block_group_cache_lock);
@@ -4227,7 +4235,7 @@ search:
 		u64 offset;
 		int cached;
 
-		atomic_inc(&block_group->count);
+		btrfs_get_block_group(block_group);
 		search_start = block_group->key.objectid;
 
 have_block_group:
@@ -4315,7 +4323,7 @@ have_block_group:
 
 				btrfs_put_block_group(block_group);
 				block_group = last_ptr->block_group;
-				atomic_inc(&block_group->count);
+				btrfs_get_block_group(block_group);
 				spin_unlock(&last_ptr->lock);
 				spin_unlock(&last_ptr->refill_lock);
 
@@ -7395,9 +7403,7 @@ int btrfs_free_block_groups(struct btrfs
 			wait_block_group_cache_done(block_group);
 
 		btrfs_remove_free_space_cache(block_group);
-
-		WARN_ON(atomic_read(&block_group->count) != 1);
-		kfree(block_group);
+		btrfs_put_block_group(block_group);
 
 		spin_lock(&info->block_group_cache_lock);
 	}

[068/111] Btrfs: Use correct values when updating inode i_size on fallocate

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>

commit d1ea6a61454e7d7ff0873d0ad1ae27d5807da0d3 upstream.

commit f2bc9dd07e3424c4ec5f3949961fe053d47bc825
Author: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Date:   Wed Jan 20 12:57:53 2010 +0530

    Btrfs: Use correct values when updating inode i_size on fallocate

    Even though we allocate more, we should be updating inode i_size
    as per the arguments passed

    Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>

Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/inode.c |   13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5799,7 +5799,7 @@ out_fail:
 }
 
 static int prealloc_file_range(struct inode *inode, u64 start, u64 end,
-			       u64 alloc_hint, int mode)
+			u64 alloc_hint, int mode, loff_t actual_len)
 {
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -5808,6 +5808,7 @@ static int prealloc_file_range(struct in
 	u64 cur_offset = start;
 	u64 num_bytes = end - start;
 	int ret = 0;
+	u64 i_size;
 
 	while (num_bytes > 0) {
 		alloc_size = min(num_bytes, root->fs_info->max_extent);
@@ -5846,8 +5847,12 @@ static int prealloc_file_range(struct in
 		BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC;
 		if (!(mode & FALLOC_FL_KEEP_SIZE) &&
 		    cur_offset > inode->i_size) {
-			i_size_write(inode, cur_offset);
-			btrfs_ordered_update_i_size(inode, cur_offset, NULL);
+			if (cur_offset > actual_len)
+				i_size  = actual_len;
+			else
+				i_size = cur_offset;
+			i_size_write(inode, i_size);
+			btrfs_ordered_update_i_size(inode, i_size, NULL);
 		}
 
 		ret = btrfs_update_inode(trans, root, inode);
@@ -5940,7 +5945,7 @@ static long btrfs_fallocate(struct inode
 		     !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) {
 			ret = prealloc_file_range(inode,
 						  cur_offset, last_byte,
-						  alloc_hint, mode);
+						alloc_hint, mode, offset+len);
 			if (ret < 0) {
 				free_extent_map(em);
 				break;

[069/111] Btrfs: fix a memory leak in btrfs_init_acl

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yang Hongyang <yanghy@cn.fujitsu.com>

commit f858153c367a397235d3e81136741e40e44faf1d upstream.

In btrfs_init_acl() cloned acl is not released

Signed-off-by: Yang Hongyang <yanghy@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/acl.c |    1 +
 1 file changed, 1 insertion(+)

--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -272,6 +272,7 @@ int btrfs_init_acl(struct btrfs_trans_ha
 						    ACL_TYPE_ACCESS);
 			}
 		}
+		posix_acl_release(clone);
 	}
 failed:
 	posix_acl_release(acl);

[070/111] Btrfs: run orphan cleanup on default fs root

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit e3acc2a6850efff647f1c5458524eb3a8bcba20a upstream.

This patch revert's commit

6c090a11e1c403b727a6a8eff0b97d5fb9e95cb5

Since it introduces this problem where we can run orphan cleanup on a
volume that can have orphan entries re-added.  Instead of my original
fix, Yan Zheng pointed out that we can just revert my original fix and
then run the orphan cleanup in open_ctree after we look up the fs_root.
I have tested this with all the tests that gave me problems and this
patch fixes both problems.  Thanks,

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/disk-io.c |    6 ++++++
 fs/btrfs/inode.c   |    6 ------
 2 files changed, 6 insertions(+), 6 deletions(-)

--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1993,6 +1993,12 @@ struct btrfs_root *open_ctree(struct sup
 	if (!fs_info->fs_root)
 		goto fail_trans_kthread;
 
+	if (!(sb->s_flags & MS_RDONLY)) {
+		down_read(&fs_info->cleanup_work_sem);
+		btrfs_orphan_cleanup(fs_info->fs_root);
+		up_read(&fs_info->cleanup_work_sem);
+	}
+
 	return tree_root;
 
 fail_trans_kthread:
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3796,12 +3796,6 @@ struct inode *btrfs_lookup_dentry(struct
 
 	if (location.type == BTRFS_INODE_ITEM_KEY) {
 		inode = btrfs_iget(dir->i_sb, &location, root);
-		if (unlikely(root->clean_orphans) &&
-		    !(inode->i_sb->s_flags & MS_RDONLY)) {
-			down_read(&root->fs_info->cleanup_work_sem);
-			btrfs_orphan_cleanup(root);
-			up_read(&root->fs_info->cleanup_work_sem);
-		}
 		return inode;
 	}
 

[071/111] Btrfs: do not mark the chunk as readonly if in degraded mode

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit f48b90756bd834dda852ff514f2690d3175b1f44 upstream.

If a RAID setup has chunks that span multiple disks, and one of those
disks has failed, btrfs_chunk_readonly will return 1 since one of the
disks in that chunk's stripes is dead and therefore not writeable.  So
instead if we are in degraded mode, return 0 so we can go ahead and
allocate stuff.  Without this patch all of the block groups in a RAID1
setup will end up read-only, which will mean we can't add new disks to
the array since we won't be able to make allocations.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/volumes.c |    5 +++++
 1 file changed, 5 insertions(+)

--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -2538,6 +2538,11 @@ int btrfs_chunk_readonly(struct btrfs_ro
 	if (!em)
 		return 1;
 
+	if (btrfs_test_opt(root, DEGRADED)) {
+		free_extent_map(em);
+		return 0;
+	}
+
 	map = (struct map_lookup *)em->bdev;
 	for (i = 0; i < map->num_stripes; i++) {
 		if (!map->stripes[i].dev->writeable) {

[072/111] Btrfs: check return value of open_bdev_exclusive properly

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit 7f59203abeaf18bf3497b308891f95a4489810ad upstream.

Hit this problem while testing RAID1 failure stuff.  open_bdev_exclusive
returns ERR_PTR(), not NULL.  So change the return value properly.  This
is important if you accidently specify a device that doesn't exist when
trying to add a new device to an array, you will panic the box
dereferencing bdev.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/volumes.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1434,8 +1434,8 @@ int btrfs_init_new_device(struct btrfs_r
 		return -EINVAL;
 
 	bdev = open_bdev_exclusive(device_path, 0, root->fs_info->bdev_holder);
-	if (!bdev)
-		return -EIO;
+	if (IS_ERR(bdev))
+		return PTR_ERR(bdev);
 
 	if (root->fs_info->fs_devices->seeding) {
 		seeding_dev = 1;

[073/111] Btrfs: check total number of devices when removing missing

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit 035fe03a7ad56982b30ab3a522b7b08d58feccd0 upstream.

If you have a disk failure in RAID1 and then add a new disk to the
array, and then try to remove the missing volume, it will fail.  The
reason is the sanity check only looks at the total number of rw devices,
which is just 2 because we have 2 good disks and 1 bad one.  Instead
check the total number of devices in the array to make sure we can
actually remove the device.  Tested this with a failed disk setup and
with this test we can now run

btrfs-vol -r missing /mount/point

and it works fine.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/volumes.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1135,7 +1135,7 @@ int btrfs_rm_device(struct btrfs_root *r
 		root->fs_info->avail_metadata_alloc_bits;
 
 	if ((all_avail & BTRFS_BLOCK_GROUP_RAID10) &&
-	    root->fs_info->fs_devices->rw_devices <= 4) {
+	    root->fs_info->fs_devices->num_devices <= 4) {
 		printk(KERN_ERR "btrfs: unable to go below four devices "
 		       "on raid10\n");
 		ret = -EINVAL;
@@ -1143,7 +1143,7 @@ int btrfs_rm_device(struct btrfs_root *r
 	}
 
 	if ((all_avail & BTRFS_BLOCK_GROUP_RAID1) &&
-	    root->fs_info->fs_devices->rw_devices <= 2) {
+	    root->fs_info->fs_devices->num_devices <= 2) {
 		printk(KERN_ERR "btrfs: unable to go below two "
 		       "devices on raid1\n");
 		ret = -EINVAL;

[074/111] Btrfs: fix race between allocate and release extent buffer.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit f044ba7835b84e69c68b620ca8fa27e5ef67759d upstream.

Increase extent buffer's reference count while holding the lock.
Otherwise it can race with try_release_extent_buffer.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/btrfs/extent_io.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -3165,10 +3165,9 @@ struct extent_buffer *alloc_extent_buffe
 		spin_unlock(&tree->buffer_lock);
 		goto free_eb;
 	}
-	spin_unlock(&tree->buffer_lock);
-
 	/* add one reference for the tree */
 	atomic_inc(&eb->refs);
+	spin_unlock(&tree->buffer_lock);
 	return eb;
 
 free_eb:

[075/111] Btrfs: make error return negative in btrfs_sync_file()

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Roel Kluin <roel.kluin@gmail.com>

commit 014e4ac4f7d9c981750491fa40ea35efadc9ed49 upstream.

It appears the error return should be negative

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/file.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1133,7 +1133,7 @@ int btrfs_sync_file(struct file *file, s
 	}
 	mutex_lock(&dentry->d_inode->i_mutex);
 out:
-	return ret > 0 ? EIO : ret;
+	return ret > 0 ? -EIO : ret;
 }
 
 static const struct vm_operations_struct btrfs_file_vm_ops = {

[076/111] Btrfs: remove BUG_ON() due to mounting bad filesystem

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Miao Xie <miaox@cn.fujitsu.com>

commit d7ce5843bb28ada6845ab2ae8510ba3f12d33154 upstream.

Mounting a bad filesystem caused a BUG_ON(). The following is steps to
reproduce it.
 # mkfs.btrfs /dev/sda2
 # mount /dev/sda2 /mnt
 # mkfs.btrfs /dev/sda1 /dev/sda2
 (the program says that /dev/sda2 was mounted, and then exits. )
 # umount /mnt
 # mount /dev/sda1 /mnt

At the third step, mkfs.btrfs exited in the way of make filesystem. So the
initialization of the filesystem didn't finish. So the filesystem was bad, and
it caused BUG_ON() when mounting it. But BUG_ON() should be called by the wrong
code, not user's operation, so I think it is a bug of btrfs.

This patch fixes it.

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/disk-io.c    |    7 ++++++-
 fs/btrfs/relocation.c |    3 ++-
 2 files changed, 8 insertions(+), 2 deletions(-)

--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1982,7 +1982,12 @@ struct btrfs_root *open_ctree(struct sup
 
 	if (!(sb->s_flags & MS_RDONLY)) {
 		ret = btrfs_recover_relocation(tree_root);
-		BUG_ON(ret);
+		if (ret < 0) {
+			printk(KERN_WARNING
+			       "btrfs: failed to recover relocation\n");
+			err = -EINVAL;
+			goto fail_trans_kthread;
+		}
 	}
 
 	location.objectid = BTRFS_FS_TREE_OBJECTID;
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3764,7 +3764,8 @@ out:
 				       BTRFS_DATA_RELOC_TREE_OBJECTID);
 		if (IS_ERR(fs_root))
 			err = PTR_ERR(fs_root);
-		btrfs_orphan_cleanup(fs_root);
+		else
+			btrfs_orphan_cleanup(fs_root);
 	}
 	return err;
 }

[077/111] Btrfs: Fix oopsen when dropping empty tree.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Yan, Zheng <zheng.yan@oracle.com>

commit 7a7965f83e89f0be506a96769938a721e4e5ae50 upstream.

When dropping a empty tree, walk_down_tree() skips checking
extent information for the tree root. This will triggers a
BUG_ON in walk_up_proc().

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/extent-tree.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -5402,10 +5402,6 @@ static noinline int walk_down_tree(struc
 	int ret;
 
 	while (level >= 0) {
-		if (path->slots[level] >=
-		    btrfs_header_nritems(path->nodes[level]))
-			break;
-
 		ret = walk_down_proc(trans, root, path, wc, lookup_info);
 		if (ret > 0)
 			break;
@@ -5413,6 +5409,10 @@ static noinline int walk_down_tree(struc
 		if (level == 0)
 			break;
 
+		if (path->slots[level] >=
+		    btrfs_header_nritems(path->nodes[level]))
+			break;
+
 		ret = do_walk_down(trans, root, path, wc, &lookup_info);
 		if (ret > 0) {
 			path->slots[level]++;

[078/111] Btrfs: do not try and lookup the file extent when finishing ordered io

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit efd049fb26a162c3830fd3cb1001fdc09b147f3b upstream.

When running the following fio job

[torrent]
filename=torrent-test
rw=randwrite
size=4g
filesize=4g
bs=4k
ioengine=sync

you would see long stalls where no work was being done.  That is because we were
doing all this extra work to read in the file extent outside of the transaction,
however in the random io case this ends up hurting us because the file extents
are not there to begin with.  So axe this logic, since we end up reading in the
file extent when we go to update it anyway.  This took the fio job from 11 mb/s
with several ~10 second stalls to 24 mb/s to a couple of 1-2 second stalls.

Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/inode.c |   46 ++--------------------------------------------
 1 file changed, 2 insertions(+), 44 deletions(-)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1680,24 +1680,6 @@ static int insert_reserved_file_extent(s
  * before we start the transaction.  It limits the amount of btree
  * reads required while inside the transaction.
  */
-static noinline void reada_csum(struct btrfs_root *root,
-				struct btrfs_path *path,
-				struct btrfs_ordered_extent *ordered_extent)
-{
-	struct btrfs_ordered_sum *sum;
-	u64 bytenr;
-
-	sum = list_entry(ordered_extent->list.next, struct btrfs_ordered_sum,
-			 list);
-	bytenr = sum->sums[0].bytenr;
-
-	/*
-	 * we don't care about the results, the point of this search is
-	 * just to get the btree leaves into ram
-	 */
-	btrfs_lookup_csum(NULL, root->fs_info->csum_root, path, bytenr, 0);
-}
-
 /* as ordered data IO finishes, this gets called so we can finish
  * an ordered extent if the range of bytes in the file it covers are
  * fully written.
@@ -1708,7 +1690,6 @@ static int btrfs_finish_ordered_io(struc
 	struct btrfs_trans_handle *trans;
 	struct btrfs_ordered_extent *ordered_extent = NULL;
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
-	struct btrfs_path *path;
 	int compressed = 0;
 	int ret;
 
@@ -1716,32 +1697,9 @@ static int btrfs_finish_ordered_io(struc
 	if (!ret)
 		return 0;
 
-	/*
-	 * before we join the transaction, try to do some of our IO.
-	 * This will limit the amount of IO that we have to do with
-	 * the transaction running.  We're unlikely to need to do any
-	 * IO if the file extents are new, the disk_i_size checks
-	 * covers the most common case.
-	 */
-	if (start < BTRFS_I(inode)->disk_i_size) {
-		path = btrfs_alloc_path();
-		if (path) {
-			ret = btrfs_lookup_file_extent(NULL, root, path,
-						       inode->i_ino,
-						       start, 0);
-			ordered_extent = btrfs_lookup_ordered_extent(inode,
-								     start);
-			if (!list_empty(&ordered_extent->list)) {
-				btrfs_release_path(root, path);
-				reada_csum(root, path, ordered_extent);
-			}
-			btrfs_free_path(path);
-		}
-	}
-
-	if (!ordered_extent)
-		ordered_extent = btrfs_lookup_ordered_extent(inode, start);
+	ordered_extent = btrfs_lookup_ordered_extent(inode, start);
 	BUG_ON(!ordered_extent);
+
 	if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
 		BUG_ON(!list_empty(&ordered_extent->list));
 		ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent);

[079/111] Btrfs: apply updated fallocate i_size fix

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>

commit 23b5c50945f2294add0137799400329c0ebba290 upstream.

This version of the i_size fix for fallocate makes sure we only update
the i_size when the current fallocate is really operating outside of
i_size.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/inode.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5798,7 +5798,9 @@ static int prealloc_file_range(struct in
 		inode->i_ctime = CURRENT_TIME;
 		BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC;
 		if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-		    cur_offset > inode->i_size) {
+			(actual_len > inode->i_size) &&
+			(cur_offset > inode->i_size)) {
+
 			if (cur_offset > actual_len)
 				i_size  = actual_len;
 			else

[080/111] Btrfs: btrfs_mark_extent_written uses the wrong slot

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Shaohua Li <shaohua.li@intel.com>

commit 3f6fae9559225741c91f1320090b285da1413290 upstream.

My test do: fallocate a big file and do write. The file is 512M, but
after file write is done btrfs-debug-tree shows:
item 6 key (257 EXTENT_DATA 0) itemoff 3516 itemsize 53
                extent data disk byte 1103101952 nr 536870912
                extent data offset 0 nr 399634432 ram 536870912
                extent compression 0
Looks like a regression introducted by
6c7d54ac87f338c479d9729e8392eca3f76e11e1, where we set wrong slot.

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Acked-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/file.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -720,13 +720,15 @@ again:
 					inode->i_ino, orig_offset);
 		BUG_ON(ret);
 	}
-	fi = btrfs_item_ptr(leaf, path->slots[0],
-			   struct btrfs_file_extent_item);
 	if (del_nr == 0) {
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+			   struct btrfs_file_extent_item);
 		btrfs_set_file_extent_type(leaf, fi,
 					   BTRFS_FILE_EXTENT_REG);
 		btrfs_mark_buffer_dirty(leaf);
 	} else {
+		fi = btrfs_item_ptr(leaf, del_slot - 1,
+			   struct btrfs_file_extent_item);
 		btrfs_set_file_extent_type(leaf, fi,
 					   BTRFS_FILE_EXTENT_REG);
 		btrfs_set_file_extent_num_bytes(leaf, fi,

[081/111] Btrfs: kfree correct pointer during mount option parsing

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Josef Bacik <josef@redhat.com>

commit da495ecc0fb096b383754952a1c152147bc95b52 upstream.

We kstrdup the options string, but then strsep screws with the pointer,
so when we kfree() it, we're not giving it the right pointer.

Tested-by: Andy Lutomirski <luto@mit.edu>

Signed-off-by: Chris Mason <chris.mason@oracle.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/btrfs/super.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -126,7 +126,7 @@ int btrfs_parse_options(struct btrfs_roo
 {
 	struct btrfs_fs_info *info = root->fs_info;
 	substring_t args[MAX_OPT_ARGS];
-	char *p, *num;
+	char *p, *num, *orig;
 	int intarg;
 	int ret = 0;
 
@@ -141,6 +141,7 @@ int btrfs_parse_options(struct btrfs_roo
 	if (!options)
 		return -ENOMEM;
 
+	orig = options;
 
 	while ((p = strsep(&options, ",")) != NULL) {
 		int token;
@@ -273,7 +274,7 @@ int btrfs_parse_options(struct btrfs_roo
 		}
 	}
 out:
-	kfree(options);
+	kfree(orig);
 	return ret;
 }
 

[082/111] nohz: Introduce arch_needs_cpu

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

commit 3c5d92a0cfb5103c0d5ab74d4ae6373d3af38148 upstream.

Allow the architecture to request a normal jiffy tick when the system
goes idle and tick_nohz_stop_sched_tick is called . On s390 the hook is
used to prevent the system going fully idle if there has been an
interrupt other than a clock comparator interrupt since the last wakeup.

On s390 the HiperSockets response time for 1 connection ping-pong goes
down from 42 to 34 microseconds. The CPU cost decreases by 27%.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
LKML-Reference: <20090929122533.402715150@de.ibm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Jolly <jjolly@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/s390/include/asm/cputime.h |    8 ++++++++
 arch/s390/kernel/s390_ext.c     |    2 ++
 arch/s390/kernel/vtime.c        |    2 ++
 drivers/s390/cio/cio.c          |    1 +
 include/linux/tick.h            |    3 +++
 kernel/time/tick-sched.c        |   13 ++++++++-----
 6 files changed, 24 insertions(+), 5 deletions(-)

--- a/arch/s390/include/asm/cputime.h
+++ b/arch/s390/include/asm/cputime.h
@@ -183,6 +183,7 @@ struct s390_idle_data {
 	unsigned long long idle_count;
 	unsigned long long idle_enter;
 	unsigned long long idle_time;
+	int nohz_delay;
 };
 
 DECLARE_PER_CPU(struct s390_idle_data, s390_idle);
@@ -198,4 +199,11 @@ static inline void s390_idle_check(void)
 		vtime_start_cpu();
 }
 
+static inline int s390_nohz_delay(int cpu)
+{
+	return per_cpu(s390_idle, cpu).nohz_delay != 0;
+}
+
+#define arch_needs_cpu(cpu) s390_nohz_delay(cpu)
+
 #endif /* _S390_CPUTIME_H */
--- a/arch/s390/kernel/s390_ext.c
+++ b/arch/s390/kernel/s390_ext.c
@@ -126,6 +126,8 @@ void __irq_entry do_extint(struct pt_reg
 		/* Serve timer interrupts first. */
 		clock_comparator_work();
 	kstat_cpu(smp_processor_id()).irqs[EXTERNAL_INTERRUPT]++;
+	if (code != 0x1004)
+		__get_cpu_var(s390_idle).nohz_delay = 1;
         index = ext_hash(code);
 	for (p = ext_int_hash[index]; p; p = p->next) {
 		if (likely(p->code == code))
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -167,6 +167,8 @@ void vtime_stop_cpu(void)
 	/* Wait for external, I/O or machine check interrupt. */
 	psw.mask = psw_kernel_bits | PSW_MASK_WAIT | PSW_MASK_IO | PSW_MASK_EXT;
 
+	idle->nohz_delay = 0;
+
 	/* Check if the CPU timer needs to be reprogrammed. */
 	if (vq->do_spt) {
 		__u64 vmax = VTIMER_MAX_SLICE;
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -618,6 +618,7 @@ void __irq_entry do_IRQ(struct pt_regs *
 	old_regs = set_irq_regs(regs);
 	s390_idle_check();
 	irq_enter();
+	__get_cpu_var(s390_idle).nohz_delay = 1;
 	if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator)
 		/* Serve timer interrupts first. */
 		clock_comparator_work();
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -98,6 +98,9 @@ extern int tick_check_oneshot_change(int
 extern struct tick_sched *tick_get_tick_sched(int cpu);
 extern void tick_check_idle(int cpu);
 extern int tick_oneshot_mode_active(void);
+#  ifndef arch_needs_cpu
+#   define arch_needs_cpu(cpu) (0)
+#  endif
 # else
 static inline void tick_clock_notify(void) { }
 static inline int tick_check_oneshot_change(int allow_nohz) { return 0; }
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -289,12 +289,15 @@ void tick_nohz_stop_sched_tick(int inidl
 			time_delta = KTIME_MAX;
 	} while (read_seqretry(&xtime_lock, seq));
 
-	/* Get the next timer wheel timer */
-	next_jiffies = get_next_timer_interrupt(last_jiffies);
-	delta_jiffies = next_jiffies - last_jiffies;
-
-	if (rcu_needs_cpu(cpu) || printk_needs_cpu(cpu))
+	if (rcu_needs_cpu(cpu) || printk_needs_cpu(cpu) ||
+	    arch_needs_cpu(cpu)) {
+		next_jiffies = last_jiffies + 1;
 		delta_jiffies = 1;
+	} else {
+		/* Get the next timer wheel timer */
+		next_jiffies = get_next_timer_interrupt(last_jiffies);
+		delta_jiffies = next_jiffies - last_jiffies;
+	}
 	/*
 	 * Do not stop the tick, if we are only one off
 	 * or if the cpu is required for rcu

[083/111] nohz: Reuse ktime in sub-functions of tick_check_idle.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

commit eed3b9cf3fe3fcc7a50238dfcab63a63914e8f42 upstream.

On a system with NOHZ=y tick_check_idle calls tick_nohz_stop_idle and
tick_nohz_update_jiffies. Given the right conditions (ts->idle_active
and/or ts->tick_stopped) both function get a time stamp with ktime_get.
The same time stamp can be reused if both function require one.

On s390 this change has the additional benefit that gcc inlines the
tick_nohz_stop_idle function into tick_check_idle. The number of
instructions to execute tick_check_idle drops from 225 to 144
(without the ktime_get optimization it is 367 vs 215 instructions).

before:

 0)               |  tick_check_idle() {
 0)               |    tick_nohz_stop_idle() {
 0)               |      ktime_get() {
 0)               |        read_tod_clock() {
 0)   0.601 us    |        }
 0)   1.765 us    |      }
 0)   3.047 us    |    }
 0)               |    ktime_get() {
 0)               |      read_tod_clock() {
 0)   0.570 us    |      }
 0)   1.727 us    |    }
 0)               |    tick_do_update_jiffies64() {
 0)   0.609 us    |    }
 0)   8.055 us    |  }

after:

 0)               |  tick_check_idle() {
 0)               |    ktime_get() {
 0)               |      read_tod_clock() {
 0)   0.617 us    |      }
 0)   1.773 us    |    }
 0)               |    tick_do_update_jiffies64() {
 0)   0.593 us    |    }
 0)   4.477 us    |  }

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: john stultz <johnstul@us.ibm.com>
LKML-Reference: <20090929122533.206589318@de.ibm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Jolly <jjolly@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 kernel/time/tick-sched.c |   62 +++++++++++++++++++++++++----------------------
 1 file changed, 33 insertions(+), 29 deletions(-)

--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -134,18 +134,13 @@ __setup("nohz=", setup_tick_nohz);
  * value. We do this unconditionally on any cpu, as we don't know whether the
  * cpu, which has the update task assigned is in a long sleep.
  */
-static void tick_nohz_update_jiffies(void)
+static void tick_nohz_update_jiffies(ktime_t now)
 {
 	int cpu = smp_processor_id();
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
 	unsigned long flags;
-	ktime_t now;
-
-	if (!ts->tick_stopped)
-		return;
 
 	cpumask_clear_cpu(cpu, nohz_cpu_mask);
-	now = ktime_get();
 	ts->idle_waketime = now;
 
 	local_irq_save(flags);
@@ -155,20 +150,17 @@ static void tick_nohz_update_jiffies(voi
 	touch_softlockup_watchdog();
 }
 
-static void tick_nohz_stop_idle(int cpu)
+static void tick_nohz_stop_idle(int cpu, ktime_t now)
 {
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+	ktime_t delta;
 
-	if (ts->idle_active) {
-		ktime_t now, delta;
-		now = ktime_get();
-		delta = ktime_sub(now, ts->idle_entrytime);
-		ts->idle_lastupdate = now;
-		ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
-		ts->idle_active = 0;
+	delta = ktime_sub(now, ts->idle_entrytime);
+	ts->idle_lastupdate = now;
+	ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
+	ts->idle_active = 0;
 
-		sched_clock_idle_wakeup_event(0);
-	}
+	sched_clock_idle_wakeup_event(0);
 }
 
 static ktime_t tick_nohz_start_idle(struct tick_sched *ts)
@@ -463,7 +455,11 @@ void tick_nohz_restart_sched_tick(void)
 	ktime_t now;
 
 	local_irq_disable();
-	tick_nohz_stop_idle(cpu);
+	if (ts->idle_active || (ts->inidle && ts->tick_stopped))
+		now = ktime_get();
+
+	if (ts->idle_active)
+		tick_nohz_stop_idle(cpu, now);
 
 	if (!ts->inidle || !ts->tick_stopped) {
 		ts->inidle = 0;
@@ -477,7 +473,6 @@ void tick_nohz_restart_sched_tick(void)
 
 	/* Update jiffies first */
 	select_nohz_load_balancer(0);
-	now = ktime_get();
 	tick_do_update_jiffies64(now);
 	cpumask_clear_cpu(cpu, nohz_cpu_mask);
 
@@ -611,22 +606,18 @@ static void tick_nohz_switch_to_nohz(voi
  * timer and do not touch the other magic bits which need to be done
  * when idle is left.
  */
-static void tick_nohz_kick_tick(int cpu)
+static void tick_nohz_kick_tick(int cpu, ktime_t now)
 {
 #if 0
 	/* Switch back to 2.6.27 behaviour */
 
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
-	ktime_t delta, now;
-
-	if (!ts->tick_stopped)
-		return;
+	ktime_t delta;
 
 	/*
 	 * Do not touch the tick device, when the next expiry is either
 	 * already reached or less/equal than the tick period.
 	 */
-	now = ktime_get();
 	delta =	ktime_sub(hrtimer_get_expires(&ts->sched_timer), now);
 	if (delta.tv64 <= tick_period.tv64)
 		return;
@@ -635,9 +626,26 @@ static void tick_nohz_kick_tick(int cpu)
 #endif
 }
 
+static inline void tick_check_nohz(int cpu)
+{
+	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+	ktime_t now;
+
+	if (!ts->idle_active && !ts->tick_stopped)
+		return;
+	now = ktime_get();
+	if (ts->idle_active)
+		tick_nohz_stop_idle(cpu, now);
+	if (ts->tick_stopped) {
+		tick_nohz_update_jiffies(now);
+		tick_nohz_kick_tick(cpu, now);
+	}
+}
+
 #else
 
 static inline void tick_nohz_switch_to_nohz(void) { }
+static inline void tick_check_nohz(int cpu) { }
 
 #endif /* NO_HZ */
 
@@ -647,11 +655,7 @@ static inline void tick_nohz_switch_to_n
 void tick_check_idle(int cpu)
 {
 	tick_check_oneshot_broadcast(cpu);
-#ifdef CONFIG_NO_HZ
-	tick_nohz_stop_idle(cpu);
-	tick_nohz_update_jiffies();
-	tick_nohz_kick_tick(cpu);
-#endif
+	tick_check_nohz(cpu);
 }
 
 /*

[084/111] timekeeping: Fix clock_gettime vsyscall time warp

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Lin Ming <ming.m.lin@intel.com>

commit 0696b711e4be45fa104c12329f617beb29c03f78 upstream.

Since commit 0a544198 "timekeeping: Move NTP adjusted clock multiplier
to struct timekeeper" the clock multiplier of vsyscall is updated with
the unmodified clock multiplier of the clock source and not with the
NTP adjusted multiplier of the timekeeper.

This causes user space observerable time warps:
new CLOCK-warp maximum: 120 nsecs,  00000025c337c537 -> 00000025c337c4bf

Add a new argument "mult" to update_vsyscall() and hand in the
timekeeping internal NTP adjusted multiplier.

Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Cc: "Zhang Yanmin" <yanmin_zhang@linux.intel.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Tony Luck <tony.luck@intel.com>
LKML-Reference: <1258436990.17765.83.camel@minggr.sh.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Kurt Garloff <garloff@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/ia64/kernel/time.c       |    4 ++--
 arch/powerpc/kernel/time.c    |    5 +++--
 arch/s390/kernel/time.c       |    3 ++-
 arch/x86/kernel/vsyscall_64.c |    5 +++--
 include/linux/clocksource.h   |    6 ++++--
 kernel/time/timekeeping.c     |    6 +++---
 6 files changed, 17 insertions(+), 12 deletions(-)

--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -473,7 +473,7 @@ void update_vsyscall_tz(void)
 {
 }
 
-void update_vsyscall(struct timespec *wall, struct clocksource *c)
+void update_vsyscall(struct timespec *wall, struct clocksource *c, u32 mult)
 {
         unsigned long flags;
 
@@ -481,7 +481,7 @@ void update_vsyscall(struct timespec *wa
 
         /* copy fsyscall clock data */
         fsyscall_gtod_data.clk_mask = c->mask;
-        fsyscall_gtod_data.clk_mult = c->mult;
+        fsyscall_gtod_data.clk_mult = mult;
         fsyscall_gtod_data.clk_shift = c->shift;
         fsyscall_gtod_data.clk_fsys_mmio = c->fsys_mmio;
         fsyscall_gtod_data.clk_cycle_last = c->cycle_last;
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -864,7 +864,8 @@ static cycle_t timebase_read(struct cloc
 	return (cycle_t)get_tb();
 }
 
-void update_vsyscall(struct timespec *wall_time, struct clocksource *clock)
+void update_vsyscall(struct timespec *wall_time, struct clocksource *clock,
+		     u32 mult)
 {
 	u64 t2x, stamp_xsec;
 
@@ -877,7 +878,7 @@ void update_vsyscall(struct timespec *wa
 
 	/* XXX this assumes clock->shift == 22 */
 	/* 4611686018 ~= 2^(20+64-22) / 1e9 */
-	t2x = (u64) clock->mult * 4611686018ULL;
+	t2x = (u64) mult * 4611686018ULL;
 	stamp_xsec = (u64) xtime.tv_nsec * XSEC_PER_SEC;
 	do_div(stamp_xsec, 1000000000);
 	stamp_xsec += (u64) xtime.tv_sec * XSEC_PER_SEC;
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -214,7 +214,8 @@ struct clocksource * __init clocksource_
 	return &clocksource_tod;
 }
 
-void update_vsyscall(struct timespec *wall_time, struct clocksource *clock)
+void update_vsyscall(struct timespec *wall_time, struct clocksource *clock,
+		     u32 mult)
 {
 	if (clock != &clocksource_tod)
 		return;
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -73,7 +73,8 @@ void update_vsyscall_tz(void)
 	write_sequnlock_irqrestore(&vsyscall_gtod_data.lock, flags);
 }
 
-void update_vsyscall(struct timespec *wall_time, struct clocksource *clock)
+void update_vsyscall(struct timespec *wall_time, struct clocksource *clock,
+		     u32 mult)
 {
 	unsigned long flags;
 
@@ -82,7 +83,7 @@ void update_vsyscall(struct timespec *wa
 	vsyscall_gtod_data.clock.vread = clock->vread;
 	vsyscall_gtod_data.clock.cycle_last = clock->cycle_last;
 	vsyscall_gtod_data.clock.mask = clock->mask;
-	vsyscall_gtod_data.clock.mult = clock->mult;
+	vsyscall_gtod_data.clock.mult = mult;
 	vsyscall_gtod_data.clock.shift = clock->shift;
 	vsyscall_gtod_data.wall_time_sec = wall_time->tv_sec;
 	vsyscall_gtod_data.wall_time_nsec = wall_time->tv_nsec;
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -282,10 +282,12 @@ extern struct clocksource * __init __wea
 extern void clocksource_mark_unstable(struct clocksource *cs);
 
 #ifdef CONFIG_GENERIC_TIME_VSYSCALL
-extern void update_vsyscall(struct timespec *ts, struct clocksource *c);
+extern void
+update_vsyscall(struct timespec *ts, struct clocksource *c, u32 mult);
 extern void update_vsyscall_tz(void);
 #else
-static inline void update_vsyscall(struct timespec *ts, struct clocksource *c)
+static inline void
+update_vsyscall(struct timespec *ts, struct clocksource *c, u32 mult)
 {
 }
 
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -177,7 +177,7 @@ void timekeeping_leap_insert(int leapsec
 {
 	xtime.tv_sec += leapsecond;
 	wall_to_monotonic.tv_sec -= leapsecond;
-	update_vsyscall(&xtime, timekeeper.clock);
+	update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult);
 }
 
 #ifdef CONFIG_GENERIC_TIME
@@ -337,7 +337,7 @@ int do_settimeofday(struct timespec *tv)
 	timekeeper.ntp_error = 0;
 	ntp_clear();
 
-	update_vsyscall(&xtime, timekeeper.clock);
+	update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult);
 
 	write_sequnlock_irqrestore(&xtime_lock, flags);
 
@@ -822,7 +822,7 @@ void update_wall_time(void)
 	update_xtime_cache(nsecs);
 
 	/* check to see if there is a new clocksource to use */
-	update_vsyscall(&xtime, timekeeper.clock);
+	update_vsyscall(&xtime, timekeeper.clock, timekeeper.mult);
 }
 
 /**

[085/111] sched: Fix granularity of task_u/stime()

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>

commit 761b1d26df542fd5eb348837351e4d2f3bc7bffe upstream.

Originally task_s/utime() were designed to return clock_t but
later changed to return cputime_t by following commit:

  commit efe567fc8281661524ffa75477a7c4ca9b466c63
  Author: Christian Borntraeger <borntraeger@de.ibm.com>
  Date:   Thu Aug 23 15:18:02 2007 +0200

It only changed the type of return value, but not the
implementation. As the result the granularity of task_s/utime()
is still that of clock_t, not that of cputime_t.

So using task_s/utime() in __exit_signal() makes values
accumulated to the signal struct to be rounded and coarse
grained.

This patch removes casts to clock_t in task_u/stime(), to keep
granularity of cputime_t over the calculation.

v2:
  Use div_u64() to avoid error "undefined reference to `__udivdi3`"
  on some 32bit systems.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: xiyou.wangcong@gmail.com
Cc: Spencer Candland <spencer@bluehost.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Stanislaw Gruszka <sgruszka@redhat.com>
LKML-Reference: <4AFB9029.9000208@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 kernel/sched.c |   22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5216,41 +5216,45 @@ cputime_t task_stime(struct task_struct
 	return p->stime;
 }
 #else
+
+#ifndef nsecs_to_cputime
+# define nsecs_to_cputime(__nsecs) \
+	msecs_to_cputime(div_u64((__nsecs), NSEC_PER_MSEC))
+#endif
+
 cputime_t task_utime(struct task_struct *p)
 {
-	clock_t utime = cputime_to_clock_t(p->utime),
-		total = utime + cputime_to_clock_t(p->stime);
+	cputime_t utime = p->utime, total = utime + p->stime;
 	u64 temp;
 
 	/*
 	 * Use CFS's precise accounting:
 	 */
-	temp = (u64)nsec_to_clock_t(p->se.sum_exec_runtime);
+	temp = (u64)nsecs_to_cputime(p->se.sum_exec_runtime);
 
 	if (total) {
 		temp *= utime;
 		do_div(temp, total);
 	}
-	utime = (clock_t)temp;
+	utime = (cputime_t)temp;
 
-	p->prev_utime = max(p->prev_utime, clock_t_to_cputime(utime));
+	p->prev_utime = max(p->prev_utime, utime);
 	return p->prev_utime;
 }
 
 cputime_t task_stime(struct task_struct *p)
 {
-	clock_t stime;
+	cputime_t stime;
 
 	/*
 	 * Use CFS's precise accounting. (we subtract utime from
 	 * the total, to make sure the total observed by userspace
 	 * grows monotonically - apps rely on that):
 	 */
-	stime = nsec_to_clock_t(p->se.sum_exec_runtime) -
-			cputime_to_clock_t(task_utime(p));
+	stime = nsecs_to_cputime(p->se.sum_exec_runtime) - task_utime(p);
 
 	if (stime >= 0)
-		p->prev_stime = max(p->prev_stime, clock_t_to_cputime(stime));
+		p->prev_stime = max(p->prev_stime, stime);
 
 	return p->prev_stime;
 }

[086/111] sched, cputime: Introduce thread_group_times()

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>

commit 0cf55e1ec08bb5a22e068309e2d8ba1180ab4239 upstream.

This is a real fix for problem of utime/stime values decreasing
described in the thread:

   http://lkml.org/lkml/2009/11/3/522

Now cputime is accounted in the following way:

 - {u,s}time in task_struct are increased every time when the thread
   is interrupted by a tick (timer interrupt).

 - When a thread exits, its {u,s}time are added to signal->{u,s}time,
   after adjusted by task_times().

 - When all threads in a thread_group exits, accumulated {u,s}time
   (and also c{u,s}time) in signal struct are added to c{u,s}time
   in signal struct of the group's parent.

So {u,s}time in task struct are "raw" tick count, while
{u,s}time and c{u,s}time in signal struct are "adjusted" values.

And accounted values are used by:

 - task_times(), to get cputime of a thread:
   This function returns adjusted values that originates from raw
   {u,s}time and scaled by sum_exec_runtime that accounted by CFS.

 - thread_group_cputime(), to get cputime of a thread group:
   This function returns sum of all {u,s}time of living threads in
   the group, plus {u,s}time in the signal struct that is sum of
   adjusted cputimes of all exited threads belonged to the group.

The problem is the return value of thread_group_cputime(),
because it is mixed sum of "raw" value and "adjusted" value:

  group's {u,s}time = foreach(thread){{u,s}time} + exited({u,s}time)

This misbehavior can break {u,s}time monotonicity.
Assume that if there is a thread that have raw values greater
than adjusted values (e.g. interrupted by 1000Hz ticks 50 times
but only runs 45ms) and if it exits, cputime will decrease (e.g.
-5ms).

To fix this, we could do:

  group's {u,s}time = foreach(t){task_times(t)} + exited({u,s}time)

But task_times() contains hard divisions, so applying it for
every thread should be avoided.

This patch fixes the above problem in the following way:

 - Modify thread's exit (= __exit_signal()) not to use task_times().
   It means {u,s}time in signal struct accumulates raw values instead
   of adjusted values.  As the result it makes thread_group_cputime()
   to return pure sum of "raw" values.

 - Introduce a new function thread_group_times(*task, *utime, *stime)
   that converts "raw" values of thread_group_cputime() to "adjusted"
   values, in same calculation procedure as task_times().

 - Modify group's exit (= wait_task_zombie()) to use this introduced
   thread_group_times().  It make c{u,s}time in signal struct to
   have adjusted values like before this patch.

 - Replace some thread_group_cputime() by thread_group_times().
   This replacements are only applied where conveys the "adjusted"
   cputime to users, and where already uses task_times() near by it.
   (i.e. sys_times(), getrusage(), and /proc/<PID>/stat.)

This patch have a positive side effect:

 - Before this patch, if a group contains many short-life threads
   (e.g. runs 0.9ms and not interrupted by ticks), the group's
   cputime could be invisible since thread's cputime was accumulated
   after adjusted: imagine adjustment function as adj(ticks, runtime),
     {adj(0, 0.9) + adj(0, 0.9) + ....} = {0 + 0 + ....} = 0.
   After this patch it will not happen because the adjustment is
   applied after accumulated.

v2:
 - remove if()s, put new variables into signal_struct.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Spencer Candland <spencer@bluehost.com>
Cc: Americo Wang <xiyou.wangcong@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Stanislaw Gruszka <sgruszka@redhat.com>
LKML-Reference: <4B162517.8040909@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 fs/proc/array.c       |    5 +----
 include/linux/sched.h |    4 ++++
 kernel/exit.c         |   20 ++++++++++++--------
 kernel/fork.c         |    3 +++
 kernel/sched.c        |   41 +++++++++++++++++++++++++++++++++++++++++
 kernel/sys.c          |   18 ++++++++----------
 6 files changed, 69 insertions(+), 22 deletions(-)

--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -405,7 +405,6 @@ static int do_task_stat(struct seq_file
 
 		/* add up live thread stats at the group level */
 		if (whole) {
-			struct task_cputime cputime;
 			struct task_struct *t = task;
 			do {
 				min_flt += t->min_flt;
@@ -416,9 +415,7 @@ static int do_task_stat(struct seq_file
 
 			min_flt += sig->min_flt;
 			maj_flt += sig->maj_flt;
-			thread_group_cputime(task, &cputime);
-			utime = cputime.utime;
-			stime = cputime.stime;
+			thread_group_times(task, &utime, &stime);
 			gtime = cputime_add(gtime, sig->gtime);
 		}
 
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -628,6 +628,9 @@ struct signal_struct {
 	cputime_t utime, stime, cutime, cstime;
 	cputime_t gtime;
 	cputime_t cgtime;
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
+	cputime_t prev_utime, prev_stime;
+#endif
 	unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw;
 	unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt;
 	unsigned long inblock, oublock, cinblock, coublock;
@@ -1725,6 +1728,7 @@ static inline void put_task_struct(struc
 extern cputime_t task_utime(struct task_struct *p);
 extern cputime_t task_stime(struct task_struct *p);
 extern cputime_t task_gtime(struct task_struct *p);
+extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st);
 
 /*
  * Per process flags
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -110,8 +110,8 @@ static void __exit_signal(struct task_st
 		 * We won't ever get here for the group leader, since it
 		 * will have been the last reference on the signal_struct.
 		 */
-		sig->utime = cputime_add(sig->utime, task_utime(tsk));
-		sig->stime = cputime_add(sig->stime, task_stime(tsk));
+		sig->utime = cputime_add(sig->utime, tsk->utime);
+		sig->stime = cputime_add(sig->stime, tsk->stime);
 		sig->gtime = cputime_add(sig->gtime, task_gtime(tsk));
 		sig->min_flt += tsk->min_flt;
 		sig->maj_flt += tsk->maj_flt;
@@ -1205,6 +1205,7 @@ static int wait_task_zombie(struct wait_
 		struct signal_struct *psig;
 		struct signal_struct *sig;
 		unsigned long maxrss;
+		cputime_t tgutime, tgstime;
 
 		/*
 		 * The resource counters for the group leader are in its
@@ -1220,20 +1221,23 @@ static int wait_task_zombie(struct wait_
 		 * need to protect the access to parent->signal fields,
 		 * as other threads in the parent group can be right
 		 * here reaping other children at the same time.
+		 *
+		 * We use thread_group_times() to get times for the thread
+		 * group, which consolidates times for all threads in the
+		 * group including the group leader.
 		 */
+		thread_group_times(p, &tgutime, &tgstime);
 		spin_lock_irq(&p->real_parent->sighand->siglock);
 		psig = p->real_parent->signal;
 		sig = p->signal;
 		psig->cutime =
 			cputime_add(psig->cutime,
-			cputime_add(p->utime,
-			cputime_add(sig->utime,
-				    sig->cutime)));
+			cputime_add(tgutime,
+				    sig->cutime));
 		psig->cstime =
 			cputime_add(psig->cstime,
-			cputime_add(p->stime,
-			cputime_add(sig->stime,
-				    sig->cstime)));
+			cputime_add(tgstime,
+				    sig->cstime));
 		psig->cgtime =
 			cputime_add(psig->cgtime,
 			cputime_add(p->gtime,
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -884,6 +884,9 @@ static int copy_signal(unsigned long clo
 	sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero;
 	sig->gtime = cputime_zero;
 	sig->cgtime = cputime_zero;
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
+	sig->prev_utime = sig->prev_stime = cputime_zero;
+#endif
 	sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0;
 	sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0;
 	sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0;
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5215,6 +5215,16 @@ cputime_t task_stime(struct task_struct
 {
 	return p->stime;
 }
+
+void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
+{
+	struct task_cputime cputime;
+
+	thread_group_cputime(p, &cputime);
+
+	*ut = cputime.utime;
+	*st = cputime.stime;
+}
 #else
 
 #ifndef nsecs_to_cputime
@@ -5258,6 +5268,37 @@ cputime_t task_stime(struct task_struct
 
 	return p->prev_stime;
 }
+
+/*
+ * Must be called with siglock held.
+ */
+void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
+{
+	struct signal_struct *sig = p->signal;
+	struct task_cputime cputime;
+	cputime_t rtime, utime, total;
+
+	thread_group_cputime(p, &cputime);
+
+	total = cputime_add(cputime.utime, cputime.stime);
+	rtime = nsecs_to_cputime(cputime.sum_exec_runtime);
+
+	if (total) {
+		u64 temp;
+
+		temp = (u64)(rtime * cputime.utime);
+		do_div(temp, total);
+		utime = (cputime_t)temp;
+	} else
+		utime = rtime;
+
+	sig->prev_utime = max(sig->prev_utime, utime);
+	sig->prev_stime = max(sig->prev_stime,
+			      cputime_sub(rtime, sig->prev_utime));
+
+	*ut = sig->prev_utime;
+	*st = sig->prev_stime;
+}
 #endif
 
 inline cputime_t task_gtime(struct task_struct *p)
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -911,16 +911,15 @@ change_okay:
 
 void do_sys_times(struct tms *tms)
 {
-	struct task_cputime cputime;
-	cputime_t cutime, cstime;
+	cputime_t tgutime, tgstime, cutime, cstime;
 
-	thread_group_cputime(current, &cputime);
 	spin_lock_irq(&current->sighand->siglock);
+	thread_group_times(current, &tgutime, &tgstime);
 	cutime = current->signal->cutime;
 	cstime = current->signal->cstime;
 	spin_unlock_irq(&current->sighand->siglock);
-	tms->tms_utime = cputime_to_clock_t(cputime.utime);
-	tms->tms_stime = cputime_to_clock_t(cputime.stime);
+	tms->tms_utime = cputime_to_clock_t(tgutime);
+	tms->tms_stime = cputime_to_clock_t(tgstime);
 	tms->tms_cutime = cputime_to_clock_t(cutime);
 	tms->tms_cstime = cputime_to_clock_t(cstime);
 }
@@ -1338,8 +1337,7 @@ static void k_getrusage(struct task_stru
 {
 	struct task_struct *t;
 	unsigned long flags;
-	cputime_t utime, stime;
-	struct task_cputime cputime;
+	cputime_t tgutime, tgstime, utime, stime;
 	unsigned long maxrss = 0;
 
 	memset((char *) r, 0, sizeof *r);
@@ -1373,9 +1371,9 @@ static void k_getrusage(struct task_stru
 				break;
 
 		case RUSAGE_SELF:
-			thread_group_cputime(p, &cputime);
-			utime = cputime_add(utime, cputime.utime);
-			stime = cputime_add(stime, cputime.stime);
+			thread_group_times(p, &tgutime, &tgstime);
+			utime = cputime_add(utime, tgutime);
+			stime = cputime_add(stime, tgstime);
 			r->ru_nvcsw += p->signal->nvcsw;
 			r->ru_nivcsw += p->signal->nivcsw;
 			r->ru_minflt += p->signal->min_flt;

[087/111] mutex: Dont spin when the owner CPU is offline or other weird cases

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Benjamin Herrenschmidt <benh@kernel.crashing.org>

commit 4b402210486c6414fe5fbfd85934a0a22da56b04 upstream.

Due to recent load-balancer changes that delay the task migration to
the next wakeup, the adaptive mutex spinning ends up in a live lock
when the owner's CPU gets offlined because the cpu_online() check
lives before the owner running check.

This patch changes mutex_spin_on_owner() to return 0 (don't spin) in
any case where we aren't sure about the owner struct validity or CPU
number, and if the said CPU is offline. There is no point going back &
re-evaluate spinning in corner cases like that, let's just go to
sleep.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1271212509.13059.135.camel@pasglop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 kernel/sched.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5590,7 +5590,7 @@ int mutex_spin_on_owner(struct mutex *lo
 	 * the mutex owner just released it and exited.
 	 */
 	if (probe_kernel_address(&owner->cpu, cpu))
-		goto out;
+		return 0;
 #else
 	cpu = owner->cpu;
 #endif
@@ -5600,14 +5600,14 @@ int mutex_spin_on_owner(struct mutex *lo
 	 * the cpu field may no longer be valid.
 	 */
 	if (cpu >= nr_cpumask_bits)
-		goto out;
+		return 0;
 
 	/*
 	 * We need to validate that we can do a
 	 * get_cpu() and that we have the percpu area.
 	 */
 	if (!cpu_online(cpu))
-		goto out;
+		return 0;
 
 	rq = cpu_rq(cpu);
 
@@ -5626,7 +5626,7 @@ int mutex_spin_on_owner(struct mutex *lo
 
 		cpu_relax();
 	}
-out:
+
 	return 1;
 }
 #endif

[088/111] [IA64] fix SBA IOMMU to handle allocation failure properly

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>

commit e2a465675dc089e9a56ba2fa2a5fbd9bd8844d18 upstream.

It's possible that SBA IOMMU might fail to find I/O space under heavy
I/Os.  SBA IOMMU panics on allocation failure but it shouldn't; drivers
can handle the failure.  The majority of other IOMMU drivers don't panic
on allocation failure.

This patch fixes SBA IOMMU path to handle allocation failure properly.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Acked-by: Leonardo Chiquitto <lchiquitto@novell.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/ia64/hp/common/sba_iommu.c |   38 +++++++++++++++++++++++++++++---------
 1 file changed, 29 insertions(+), 9 deletions(-)

--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -677,12 +677,19 @@ sba_alloc_range(struct ioc *ioc, struct
 			spin_unlock_irqrestore(&ioc->saved_lock, flags);
 
 			pide = sba_search_bitmap(ioc, dev, pages_needed, 0);
-			if (unlikely(pide >= (ioc->res_size << 3)))
-				panic(__FILE__ ": I/O MMU @ %p is out of mapping resources\n",
-				      ioc->ioc_hpa);
+			if (unlikely(pide >= (ioc->res_size << 3))) {
+				printk(KERN_WARNING "%s: I/O MMU @ %p is"
+				       "out of mapping resources, %u %u %lx\n",
+				       __func__, ioc->ioc_hpa, ioc->res_size,
+				       pages_needed, dma_get_seg_boundary(dev));
+				return -1;
+			}
 #else
-			panic(__FILE__ ": I/O MMU @ %p is out of mapping resources\n",
-			      ioc->ioc_hpa);
+			printk(KERN_WARNING "%s: I/O MMU @ %p is"
+			       "out of mapping resources, %u %u %lx\n",
+			       __func__, ioc->ioc_hpa, ioc->res_size,
+			       pages_needed, dma_get_seg_boundary(dev));
+			return -1;
 #endif
 		}
 	}
@@ -965,6 +972,8 @@ static dma_addr_t sba_map_page(struct de
 #endif
 
 	pide = sba_alloc_range(ioc, dev, size);
+	if (pide < 0)
+		return 0;
 
 	iovp = (dma_addr_t) pide << iovp_shift;
 
@@ -1320,6 +1329,7 @@ sba_coalesce_chunks(struct ioc *ioc, str
 	unsigned long dma_offset, dma_len; /* start/len of DMA stream */
 	int n_mappings = 0;
 	unsigned int max_seg_size = dma_get_max_seg_size(dev);
+	int idx;
 
 	while (nents > 0) {
 		unsigned long vaddr = (unsigned long) sba_sg_address(startsg);
@@ -1418,16 +1428,22 @@ sba_coalesce_chunks(struct ioc *ioc, str
 		vcontig_sg->dma_length = vcontig_len;
 		dma_len = (dma_len + dma_offset + ~iovp_mask) & iovp_mask;
 		ASSERT(dma_len <= DMA_CHUNK_SIZE);
-		dma_sg->dma_address = (dma_addr_t) (PIDE_FLAG
-			| (sba_alloc_range(ioc, dev, dma_len) << iovp_shift)
-			| dma_offset);
+		idx = sba_alloc_range(ioc, dev, dma_len);
+		if (idx < 0) {
+			dma_sg->dma_length = 0;
+			return -1;
+		}
+		dma_sg->dma_address = (dma_addr_t)(PIDE_FLAG | (idx << iovp_shift)
+						   | dma_offset);
 		n_mappings++;
 	}
 
 	return n_mappings;
 }
 
-
+static void sba_unmap_sg_attrs(struct device *dev, struct scatterlist *sglist,
+			       int nents, enum dma_data_direction dir,
+			       struct dma_attrs *attrs);
 /**
  * sba_map_sg - map Scatter/Gather list
  * @dev: instance of PCI owned by the driver that's asking.
@@ -1493,6 +1509,10 @@ static int sba_map_sg_attrs(struct devic
 	** Access to the virtual address is what forces a two pass algorithm.
 	*/
 	coalesced = sba_coalesce_chunks(ioc, dev, sglist, nents);
+	if (coalesced < 0) {
+		sba_unmap_sg_attrs(dev, sglist, nents, dir, attrs);
+		return 0;
+	}
 
 	/*
 	** Program the I/O Pdir

[089/111] crypto: testmgr - Fix complain about lack test for internal used algorithm

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Song Youquan <youquan.song@intel.com>

commit 863b557a88f8c033f7419fabafef4712a5055f85 upstream.

When load aesni-intel and ghash_clmulni-intel driver,kernel will complain no
 test for some internal used algorithm.
The strange information as following:

alg: No test for __aes-aesni (__driver-aes-aesni)
alg: No test for __ecb-aes-aesni (__driver-ecb-aes-aesni)
alg: No test for __cbc-aes-aesni (__driver-cbc-aes-aesni)
alg: No test for __ecb-aes-aesni (cryptd(__driver-ecb-aes-aesni)
alg: No test for __ghash (__ghash-pclmulqdqni)
alg: No test for __ghash (cryptd(__ghash-pclmulqdqni))

This patch add NULL test entries for these algorithm and driver.

Signed-off-by: Song Youquan <youquan.song@intel.com>
Signed-off-by: Hang Ying <ying.huang@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 crypto/testmgr.c |   84 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 84 insertions(+)

--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -1477,9 +1477,54 @@ static int alg_test_cprng(const struct a
 	return err;
 }
 
+static int alg_test_null(const struct alg_test_desc *desc,
+			     const char *driver, u32 type, u32 mask)
+{
+	return 0;
+}
+
 /* Please keep this list sorted by algorithm name. */
 static const struct alg_test_desc alg_test_descs[] = {
 	{
+		.alg = "__driver-cbc-aes-aesni",
+		.test = alg_test_null,
+		.suite = {
+			.cipher = {
+				.enc = {
+					.vecs = NULL,
+					.count = 0
+				},
+				.dec = {
+					.vecs = NULL,
+					.count = 0
+				}
+			}
+		}
+	}, {
+		.alg = "__driver-ecb-aes-aesni",
+		.test = alg_test_null,
+		.suite = {
+			.cipher = {
+				.enc = {
+					.vecs = NULL,
+					.count = 0
+				},
+				.dec = {
+					.vecs = NULL,
+					.count = 0
+				}
+			}
+		}
+	}, {
+		.alg = "__ghash-pclmulqdqni",
+		.test = alg_test_null,
+		.suite = {
+			.hash = {
+				.vecs = NULL,
+				.count = 0
+			}
+		}
+	}, {
 		.alg = "ansi_cprng",
 		.test = alg_test_cprng,
 		.fips_allowed = 1,
@@ -1623,6 +1668,30 @@ static const struct alg_test_desc alg_te
 			}
 		}
 	}, {
+		.alg = "cryptd(__driver-ecb-aes-aesni)",
+		.test = alg_test_null,
+		.suite = {
+			.cipher = {
+				.enc = {
+					.vecs = NULL,
+					.count = 0
+				},
+				.dec = {
+					.vecs = NULL,
+					.count = 0
+				}
+			}
+		}
+	}, {
+		.alg = "cryptd(__ghash-pclmulqdqni)",
+		.test = alg_test_null,
+		.suite = {
+			.hash = {
+				.vecs = NULL,
+				.count = 0
+			}
+		}
+	}, {
 		.alg = "ctr(aes)",
 		.test = alg_test_skcipher,
 		.fips_allowed = 1,
@@ -1668,6 +1737,21 @@ static const struct alg_test_desc alg_te
 				}
 			}
 		}
+	}, {
+		.alg = "ecb(__aes-aesni)",
+		.test = alg_test_null,
+		.suite = {
+			.cipher = {
+				.enc = {
+					.vecs = NULL,
+					.count = 0
+				},
+				.dec = {
+					.vecs = NULL,
+					.count = 0
+				}
+			}
+		}
 	}, {
 		.alg = "ecb(aes)",
 		.test = alg_test_skcipher,

[090/111] memory hotplug: fix a bug on /dev/mem for 64-bit kernels

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Shaohui Zheng <shaohui.zheng@intel.com>

commit ea0854170c95245a258b386c7a9314399c949fe0 upstream.

Newly added memory can not be accessed via /dev/mem, because we do not
update the variables high_memory, max_pfn and max_low_pfn.

Add a function update_end_of_memory_vars() to update these variables for
64-bit kernels.

[akpm@linux-foundation.org: simplify comment]
Signed-off-by: Shaohui Zheng <shaohui.zheng@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Li Haicheng <haicheng.li@intel.com>
Reviewed-by: Wu Fengguang <fengguang.wu@intel.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/x86/mm/init_64.c |   19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -49,6 +49,7 @@
 #include <asm/numa.h>
 #include <asm/cacheflush.h>
 #include <asm/init.h>
+#include <linux/bootmem.h>
 
 static unsigned long dma_reserve __initdata;
 
@@ -615,6 +616,21 @@ void __init paging_init(void)
  */
 #ifdef CONFIG_MEMORY_HOTPLUG
 /*
+ * After memory hotplug the variables max_pfn, max_low_pfn and high_memory need
+ * updating.
+ */
+static void  update_end_of_memory_vars(u64 start, u64 size)
+{
+	unsigned long end_pfn = PFN_UP(start + size);
+
+	if (end_pfn > max_pfn) {
+		max_pfn = end_pfn;
+		max_low_pfn = end_pfn;
+		high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1;
+	}
+}
+
+/*
  * Memory is added always to NORMAL zone. This means you will never get
  * additional DMA/DMA32 memory.
  */
@@ -633,6 +649,9 @@ int arch_add_memory(int nid, u64 start,
 	ret = __add_pages(nid, zone, start_pfn, nr_pages);
 	WARN_ON_ONCE(ret);
 
+	/* update max_pfn, max_low_pfn and high_memory */
+	update_end_of_memory_vars(start, size);
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(arch_add_memory);

[091/111] x86: Fix out of order of gsi

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Eric W. Biederman <ebiederm@xmission.com>

commit fad539956c9e69749a03f7817d22d1bab87657bf upstream.

Iranna D Ankad reported that IBM x3950 systems have boot
problems after this commit:

 |
 | commit b9c61b70075c87a8612624736faf4a2de5b1ed30
 |
 |    x86/pci: update pirq_enable_irq() to setup io apic routing
 |

The problem is that with the patch, the machine freezes when
console=ttyS0,... kernel serial parameter is passed.

It seem to freeze at DVD initialization and the whole problem
seem to be DVD/pata related, but somehow exposed through the
serial parameter.

Such apic problems can expose really weird behavior:

  ACPI: IOAPIC (id[0x10] address[0xfecff000] gsi_base[0])
  IOAPIC[0]: apic_id 16, version 0, address 0xfecff000, GSI 0-2
  ACPI: IOAPIC (id[0x0f] address[0xfec00000] gsi_base[3])
  IOAPIC[1]: apic_id 15, version 0, address 0xfec00000, GSI 3-38
  ACPI: IOAPIC (id[0x0e] address[0xfec01000] gsi_base[39])
  IOAPIC[2]: apic_id 14, version 0, address 0xfec01000, GSI 39-74
  ACPI: INT_SRC_OVR (bus 0 bus_irq 1 global_irq 4 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 5 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 3 global_irq 6 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 4 global_irq 7 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 6 global_irq 9 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 7 global_irq 10 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 8 global_irq 11 low edge)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 12 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 12 global_irq 15 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 13 global_irq 16 dfl dfl)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 14 global_irq 17 low edge)
  ACPI: INT_SRC_OVR (bus 0 bus_irq 15 global_irq 18 dfl dfl)

It turns out that the system has three io apic controllers, but
boot ioapic routing is in the second one, and that gsi_base is
not 0 - it is using a bunch of INT_SRC_OVR...

So these recent changes:

 1. one set routing for first io apic controller
 2. assume irq = gsi

... will break that system.

So try to remap those gsis, need to seperate boot_ioapic_idx
detection out of enable_IO_APIC() and call them early.

So introduce boot_ioapic_idx, and remap_ioapic_gsi()...

 -v2: shift gsi with delta instead of gsi_base of boot_ioapic_idx

 -v3: double check with find_isa_irq_apic(0, mp_INT) to get right
      boot_ioapic_idx

 -v4: nr_legacy_irqs

 -v5: add print out for boot_ioapic_idx, and also make it could be
      applied for current kernel and previous kernel

 -v6: add bus_irq, in acpi_sci_ioapic_setup, so can get overwride
      for sci right mapping...

 -v7: looks like pnpacpi get irq instead of gsi, so need to revert
      them back...

 -v8: split into two patches

 -v9: according to Eric, use fixed 16 for shifting instead of remap

 -v10: still need to touch rsparser.c

 -v11: just revert back to way Eric suggest...
      anyway the ioapic in first ioapic is blocked by second...

 -v12: two patches, this one will add more loop but check apic_id and irq > 16

Reported-by: Iranna D Ankad <iranna.ankad@in.ibm.com>
Bisected-by: Iranna D Ankad <iranna.ankad@in.ibm.com>
Tested-by: Gary Hade <garyhade@us.ibm.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Thomas Renninger <trenn@suse.de>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: len.brown@intel.com
LKML-Reference: <4B8A321A.1000008@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/x86/kernel/apic/io_apic.c |   28 ++++++++++------------------
 1 file changed, 10 insertions(+), 18 deletions(-)

--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -1484,7 +1484,7 @@ static struct {
 
 static void __init setup_IO_APIC_irqs(void)
 {
-	int apic_id = 0, pin, idx, irq;
+	int apic_id, pin, idx, irq;
 	int notcon = 0;
 	struct irq_desc *desc;
 	struct irq_cfg *cfg;
@@ -1492,14 +1492,7 @@ static void __init setup_IO_APIC_irqs(vo
 
 	apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
 
-#ifdef CONFIG_ACPI
-	if (!acpi_disabled && acpi_ioapic) {
-		apic_id = mp_find_ioapic(0);
-		if (apic_id < 0)
-			apic_id = 0;
-	}
-#endif
-
+	for (apic_id = 0; apic_id < nr_ioapics; apic_id++)
 	for (pin = 0; pin < nr_ioapic_registers[apic_id]; pin++) {
 		idx = find_irq_entry(apic_id, pin, mp_INT);
 		if (idx == -1) {
@@ -1521,6 +1514,9 @@ static void __init setup_IO_APIC_irqs(vo
 
 		irq = pin_2_irq(idx, apic_id, pin);
 
+		if ((apic_id > 0) && (irq > 16))
+			continue;
+
 		/*
 		 * Skip the timer IRQ if there's a quirk handler
 		 * installed and if it returns 1:
@@ -4083,27 +4079,23 @@ int acpi_get_override_irq(int bus_irq, i
 #ifdef CONFIG_SMP
 void __init setup_ioapic_dest(void)
 {
-	int pin, ioapic = 0, irq, irq_entry;
+	int pin, ioapic, irq, irq_entry;
 	struct irq_desc *desc;
 	const struct cpumask *mask;
 
 	if (skip_ioapic_setup == 1)
 		return;
 
-#ifdef CONFIG_ACPI
-	if (!acpi_disabled && acpi_ioapic) {
-		ioapic = mp_find_ioapic(0);
-		if (ioapic < 0)
-			ioapic = 0;
-	}
-#endif
-
+	for (ioapic = 0; ioapic < nr_ioapics; ioapic++)
 	for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) {
 		irq_entry = find_irq_entry(ioapic, pin, mp_INT);
 		if (irq_entry == -1)
 			continue;
 		irq = pin_2_irq(irq_entry, ioapic, pin);
 
+		if ((ioapic > 0) && (irq > 16))
+			continue;
+
 		desc = irq_to_desc(irq);
 
 		/*

[092/111] HWPOISON: remove the anonymous entry

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Wu Fengguang <fengguang.wu@intel.com>

commit 9b9a29ecd75e310f75a9243e1c3538ad34598fcb upstream.

(PG_swapbacked && !PG_lru) pages should not happen.
Better to treat them as unknown pages.

Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 mm/memory-failure.c |    1 -
 1 file changed, 1 deletion(-)

--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -589,7 +589,6 @@ static struct page_state {
 
 	{ lru|dirty,	lru|dirty,	"LRU",		me_pagecache_dirty },
 	{ lru|dirty,	lru,		"clean LRU",	me_pagecache_clean },
-	{ swapbacked,	swapbacked,	"anonymous",	me_pagecache_clean },
 
 	/*
 	 * Catchall entry: must be at end.

[093/111] HWPOISON: abort on failed unmap

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Wu Fengguang <fengguang.wu@intel.com>

commit 1668bfd5be9d8a52536c4865000fbbe065a3613b upstream.

Don't try to isolate a still mapped page. Otherwise we will hit the
BUG_ON(page_mapped(page)) in __remove_from_page_cache().

Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 mm/memory-failure.c |   20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -637,7 +637,7 @@ static int page_action(struct page_state
  * Do all that is necessary to remove user space mappings. Unmap
  * the pages and send SIGBUS to the processes if the data was dirty.
  */
-static void hwpoison_user_mappings(struct page *p, unsigned long pfn,
+static int hwpoison_user_mappings(struct page *p, unsigned long pfn,
 				  int trapno)
 {
 	enum ttu_flags ttu = TTU_UNMAP | TTU_IGNORE_MLOCK | TTU_IGNORE_ACCESS;
@@ -647,15 +647,18 @@ static void hwpoison_user_mappings(struc
 	int i;
 	int kill = 1;
 
-	if (PageReserved(p) || PageCompound(p) || PageSlab(p) || PageKsm(p))
-		return;
+	if (PageReserved(p) || PageSlab(p))
+		return SWAP_SUCCESS;
 
 	/*
 	 * This check implies we don't kill processes if their pages
 	 * are in the swap cache early. Those are always late kills.
 	 */
 	if (!page_mapped(p))
-		return;
+		return SWAP_SUCCESS;
+
+	if (PageCompound(p) || PageKsm(p))
+		return SWAP_FAIL;
 
 	if (PageSwapCache(p)) {
 		printk(KERN_ERR
@@ -717,6 +720,8 @@ static void hwpoison_user_mappings(struc
 	 */
 	kill_procs_ao(&tokill, !!PageDirty(p), trapno,
 		      ret != SWAP_SUCCESS, pfn);
+
+	return ret;
 }
 
 int __memory_failure(unsigned long pfn, int trapno, int ref)
@@ -786,8 +791,13 @@ int __memory_failure(unsigned long pfn,
 
 	/*
 	 * Now take care of user space mappings.
+	 * Abort on fail: __remove_from_page_cache() assumes unmapped page.
 	 */
-	hwpoison_user_mappings(p, pfn, trapno);
+	if (hwpoison_user_mappings(p, pfn, trapno) != SWAP_SUCCESS) {
+		printk(KERN_ERR "MCE %#lx: cannot unmap page, give up\n", pfn);
+		res = -EBUSY;
+		goto out;
+	}
 
 	/*
 	 * Torn down by someone else?

[094/111] powerpc/eeh: Fix a bug when pci structure is null

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Breno Leitao <leitao@linux.vnet.ibm.com>

commit 8d3d50bf1913561ef3b1f5b53115c5a481ba9b1e upstream.

During a EEH recover, the pci_dev structure can be null, mainly if an
eeh event is detected during cpi config operation. In this case, the
pci_dev will not be known (and will be null) the kernel will crash
with the following message:

Unable to handle kernel paging request for data at address 0x000000a0
Faulting instruction address: 0xc00000000006b8b4
Oops: Kernel access of bad area, sig: 11 [#1]

NIP [c00000000006b8b4] .eeh_event_handler+0x10c/0x1a0
LR [c00000000006b8a8] .eeh_event_handler+0x100/0x1a0
Call Trace:
[c0000003a80dff00] [c00000000006b8a8] .eeh_event_handler+0x100/0x1a0
[c0000003a80dff90] [c000000000031f1c] .kernel_thread+0x54/0x70

The bug occurs because pci_name() tries to access a null pointer.
This patch just guarantee that pci_name() is not called on Null pointers.

Signed-off-by: Breno Leitao <leitao@linux.vnet.ibm.com>
Signed-off-by: Linas Vepstas <linasvepstas@gmail.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 arch/powerpc/include/asm/ppc-pci.h          |    5 +++++
 arch/powerpc/platforms/pseries/eeh.c        |    4 ++--
 arch/powerpc/platforms/pseries/eeh_driver.c |    4 ++--
 arch/powerpc/platforms/pseries/eeh_event.c  |    2 +-
 4 files changed, 10 insertions(+), 5 deletions(-)

--- a/arch/powerpc/include/asm/ppc-pci.h
+++ b/arch/powerpc/include/asm/ppc-pci.h
@@ -137,6 +137,11 @@ struct device_node * find_device_pe(stru
 void eeh_sysfs_add_device(struct pci_dev *pdev);
 void eeh_sysfs_remove_device(struct pci_dev *pdev);
 
+static inline const char *eeh_pci_name(struct pci_dev *pdev)
+{
+	return pdev ? pci_name(pdev) : "<null>";
+}
+
 #endif /* CONFIG_EEH */
 
 #else /* CONFIG_PCI */
--- a/arch/powerpc/platforms/pseries/eeh.c
+++ b/arch/powerpc/platforms/pseries/eeh.c
@@ -491,7 +491,7 @@ int eeh_dn_check_failure(struct device_n
 	    pdn->eeh_mode & EEH_MODE_NOCHECK) {
 		ignored_check++;
 		pr_debug("EEH: Ignored check (%x) for %s %s\n",
-			 pdn->eeh_mode, pci_name (dev), dn->full_name);
+			 pdn->eeh_mode, eeh_pci_name(dev), dn->full_name);
 		return 0;
 	}
 
@@ -515,7 +515,7 @@ int eeh_dn_check_failure(struct device_n
 			printk (KERN_ERR "EEH: %d reads ignored for recovering device at "
 				"location=%s driver=%s pci addr=%s\n",
 				pdn->eeh_check_count, location,
-				dev->driver->name, pci_name(dev));
+				dev->driver->name, eeh_pci_name(dev));
 			printk (KERN_ERR "EEH: Might be infinite loop in %s driver\n",
 				dev->driver->name);
 			dump_stack();
--- a/arch/powerpc/platforms/pseries/eeh_driver.c
+++ b/arch/powerpc/platforms/pseries/eeh_driver.c
@@ -353,7 +353,7 @@ struct pci_dn * handle_eeh_events (struc
 		location = location ? location : "unknown";
 		printk(KERN_ERR "EEH: Error: Cannot find partition endpoint "
 		                "for location=%s pci addr=%s\n",
-		        location, pci_name(event->dev));
+		        location, eeh_pci_name(event->dev));
 		return NULL;
 	}
 
@@ -384,7 +384,7 @@ struct pci_dn * handle_eeh_events (struc
 		pci_str = pci_name (frozen_pdn->pcidev);
 		drv_str = pcid_name (frozen_pdn->pcidev);
 	} else {
-		pci_str = pci_name (event->dev);
+		pci_str = eeh_pci_name(event->dev);
 		drv_str = pcid_name (event->dev);
 	}
 	
--- a/arch/powerpc/platforms/pseries/eeh_event.c
+++ b/arch/powerpc/platforms/pseries/eeh_event.c
@@ -80,7 +80,7 @@ static int eeh_event_handler(void * dumm
 	eeh_mark_slot(event->dn, EEH_MODE_RECOVERING);
 
 	printk(KERN_INFO "EEH: Detected PCI bus error on device %s\n",
-	       pci_name(event->dev));
+	       eeh_pci_name(event->dev));
 
 	pdn = handle_eeh_events(event);
 

[095/111] ACPI: Fix regression where _PPC is not read at boot even when ignore_ppc=0

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Darrick J. Wong <djwong@us.ibm.com>

commit 455c0d71d46e86b0b7ff2c9dcfc19bc162302ee9 upstream.

Earlier, Ingo Molnar posted a patch to make it so that the kernel would avoid
reading _PPC on his broken T60.  Unfortunately, it seems that with Thomas
Renninger's patch last July to eliminate _PPC evaluations when the processor
driver loads, the kernel never actually reads _PPC at all!  This is problematic
if you happen to boot your non-T60 computer in a state where the BIOS _wants_
_PPC to be something other than zero.

So, put the _PPC evaluation back into acpi_processor_get_performance_info if
ignore_ppc isn't 1.

Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/acpi/processor_perflib.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

--- a/drivers/acpi/processor_perflib.c
+++ b/drivers/acpi/processor_perflib.c
@@ -356,7 +356,11 @@ static int acpi_processor_get_performanc
 	if (result)
 		goto update_bios;
 
-	return 0;
+	/* We need to call _PPC once when cpufreq starts */
+	if (ignore_ppc != 1)
+		result = acpi_processor_get_platform_limit(pr);
+
+	return result;
 
 	/*
 	 * Having _PPC but missing frequencies (_PSS, _PCT) is a very good hint that

[096/111] ext4: Make sure the MOVE_EXT ioctl cant overwrite append-only files

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Theodore Ts'o <tytso@mit.edu>

commit 1f5a81e41f8b1a782c68d3843e9ec1bfaadf7d72 upstream.

Dan Roseberg has reported a problem with the MOVE_EXT ioctl.  If the
donor file is an append-only file, we should not allow the operation
to proceed, lest we end up overwriting the contents of an append-only
file.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: Dan Rosenberg <dan.j.rosenberg@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ext4/move_extent.c |    3 +++
 1 file changed, 3 insertions(+)

--- a/fs/ext4/move_extent.c
+++ b/fs/ext4/move_extent.c
@@ -958,6 +958,9 @@ mext_check_arguments(struct inode *orig_
 		return -EINVAL;
 	}
 
+	if (IS_IMMUTABLE(donor_inode) || IS_APPEND(donor_inode))
+		return -EPERM;
+
 	/* Ext4 move extent does not support swapfile */
 	if (IS_SWAPFILE(orig_inode) || IS_SWAPFILE(donor_inode)) {
 		ext4_debug("ext4 move extent: The argument files should "

[097/111] ext4: Fix optional-arg mount options

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Eric Sandeen <sandeen@redhat.com>

commit 15121c18a22ae483279f76dc9e554334b800d0f7 upstream.

We have 2 mount options, "barrier" and "auto_da_alloc" which may or
may not take a 1/0 argument.  This causes the ext4 superblock mount
code to subtract uninitialized pointers and pass the result to
kmalloc, which results in very noisy failures.

Per Ted's suggestion, initialize the args struct so that
we know whether match_token() found an argument for the
option, and skip match_int() if not.

Also, return error (0) from parse_options if we thought
we found an argument, but match_int() Fails.

Reported-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Acked-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ext4/super.c |   23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1218,6 +1218,11 @@ static int parse_options(char *options,
 		if (!*p)
 			continue;
 
+		/*
+		 * Initialize args struct so we know whether arg was
+		 * found; some options take optional arguments.
+		 */
+		args[0].to = args[0].from = 0;
 		token = match_token(p, tokens, args);
 		switch (token) {
 		case Opt_bsd_df:
@@ -1503,10 +1508,11 @@ set_qf_format:
 			clear_opt(sbi->s_mount_opt, BARRIER);
 			break;
 		case Opt_barrier:
-			if (match_int(&args[0], &option)) {
-				set_opt(sbi->s_mount_opt, BARRIER);
-				break;
-			}
+			if (args[0].from) {
+				if (match_int(&args[0], &option))
+					return 0;
+			} else
+				option = 1;	/* No argument, default to 1 */
 			if (option)
 				set_opt(sbi->s_mount_opt, BARRIER);
 			else
@@ -1579,10 +1585,11 @@ set_qf_format:
 			set_opt(sbi->s_mount_opt,NO_AUTO_DA_ALLOC);
 			break;
 		case Opt_auto_da_alloc:
-			if (match_int(&args[0], &option)) {
-				clear_opt(sbi->s_mount_opt, NO_AUTO_DA_ALLOC);
-				break;
-			}
+			if (args[0].from) {
+				if (match_int(&args[0], &option))
+					return 0;
+			} else
+				option = 1;	/* No argument, default to 1 */
 			if (option)
 				clear_opt(sbi->s_mount_opt, NO_AUTO_DA_ALLOC);
 			else

[098/111] reiserfs: properly honor read-only devices

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Jeff Mahoney <jeffm@suse.com>

commit 3f8b5ee33293d43ca360771b535dfae8c57259dc upstream.

The reiserfs journal behaves inconsistently when determining whether to
allow a mount of a read-only device.

This is due to the use of the continue_replay variable to short circuit
the journal scanning.  If it's set, it's assumed that there are
transactions to replay, but there may not be.  If it's unset, it's assumed
that there aren't any, and that may not be the case either.

I've observed two failure cases:
1) Where a clean file system on a read-only device refuses to mount
2) Where a clean file system on a read-only device passes the
   optimization and then tries writing the journal header to update
   the latest mount id.

The former is easily observable by using a freshly created file system on
a read-only loopback device.

This patch moves the check into journal_read_transaction, where it can
bail out before it's about to replay a transaction.  That way it can go
through and skip transactions where appropriate, yet still refuse to mount
a file system with outstanding transactions.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/reiserfs/journal.c |   15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -2184,6 +2184,15 @@ static int journal_read_transaction(stru
 		brelse(d_bh);
 		return 1;
 	}
+
+	if (bdev_read_only(sb->s_bdev)) {
+		reiserfs_warning(sb, "clm-2076",
+				 "device is readonly, unable to replay log");
+		brelse(c_bh);
+		brelse(d_bh);
+		return -EROFS;
+	}
+
 	trans_id = get_desc_trans_id(desc);
 	/* now we know we've got a good transaction, and it was inside the valid time ranges */
 	log_blocks = kmalloc(get_desc_trans_len(desc) *
@@ -2422,12 +2431,6 @@ static int journal_read(struct super_blo
 		goto start_log_replay;
 	}
 
-	if (continue_replay && bdev_read_only(sb->s_bdev)) {
-		reiserfs_warning(sb, "clm-2076",
-				 "device is readonly, unable to replay log");
-		return -1;
-	}
-
 	/* ok, there are transactions that need to be replayed.  start with the first log block, find
 	 ** all the valid transactions, and pick out the oldest.
 	 */

[099/111] reiserfs: fix oops while creating privroot with selinux enabled

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Jeff Mahoney <jeffm@suse.com>

commit 6cb4aff0a77cc0e6bae9475d62205319e3ebbf3f upstream.

Commit 57fe60df ("reiserfs: add atomic addition of selinux attributes
during inode creation") contains a bug that will cause it to oops when
mounting a file system that didn't previously contain extended attributes
on a system using security.* xattrs.

The issue is that while creating the privroot during mount
reiserfs_security_init calls reiserfs_xattr_jcreate_nblocks which
dereferences the xattr root.  The xattr root doesn't exist, so we get an
oops.

Addresses http://bugzilla.kernel.org/show_bug.cgi?id=15309

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/reiserfs/xattr_security.c   |    2 +-
 include/linux/reiserfs_xattr.h |    5 +++++
 2 files changed, 6 insertions(+), 1 deletion(-)

--- a/fs/reiserfs/xattr_security.c
+++ b/fs/reiserfs/xattr_security.c
@@ -75,7 +75,7 @@ int reiserfs_security_init(struct inode
 		return error;
 	}
 
-	if (sec->length) {
+	if (sec->length && reiserfs_xattrs_initialized(inode->i_sb)) {
 		blocks = reiserfs_xattr_jcreate_nblocks(inode) +
 			 reiserfs_xattr_nblocks(inode, sec->length);
 		/* We don't want to count the directories twice if we have
--- a/include/linux/reiserfs_xattr.h
+++ b/include/linux/reiserfs_xattr.h
@@ -70,6 +70,11 @@ int reiserfs_security_write(struct reise
 void reiserfs_security_free(struct reiserfs_security_handle *sec);
 #endif
 
+static inline int reiserfs_xattrs_initialized(struct super_block *sb)
+{
+	return REISERFS_SB(sb)->priv_root != NULL;
+}
+
 #define xattr_size(size) ((size) + sizeof(struct reiserfs_xattr_header))
 static inline loff_t reiserfs_xattr_nblocks(struct inode *inode, loff_t size)
 {

[100/111] dlm: always use GFP_NOFS

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: David Teigland <teigland@redhat.com>

commit 573c24c4af6664ffcd9aa7ba617a35fde2b95534 upstream.

Replace all GFP_KERNEL and ls_allocation with GFP_NOFS.
ls_allocation would be GFP_KERNEL for userland lockspaces
and GFP_NOFS for file system lockspaces.

It was discovered that any lockspaces on the system can
affect all others by triggering memory reclaim in the
file system which could in turn call back into the dlm
to acquire locks, deadlocking dlm threads that were
shared by all lockspaces, like dlm_recv.

Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/dlm/config.c       |   24 ++++++++++++------------
 fs/dlm/debug_fs.c     |    2 +-
 fs/dlm/dir.c          |    7 +++----
 fs/dlm/dlm_internal.h |    1 -
 fs/dlm/lock.c         |    6 +++---
 fs/dlm/lockspace.c    |   15 +++++----------
 fs/dlm/lowcomms.c     |    6 +++---
 fs/dlm/member.c       |    8 ++++----
 fs/dlm/memory.c       |    6 +++---
 fs/dlm/netlink.c      |    2 +-
 fs/dlm/plock.c        |    6 +++---
 fs/dlm/rcom.c         |    2 +-
 fs/dlm/requestqueue.c |    2 +-
 fs/dlm/user.c         |   12 ++++++------
 14 files changed, 46 insertions(+), 53 deletions(-)

--- a/fs/dlm/config.c
+++ b/fs/dlm/config.c
@@ -410,10 +410,10 @@ static struct config_group *make_cluster
 	struct dlm_comms *cms = NULL;
 	void *gps = NULL;
 
-	cl = kzalloc(sizeof(struct dlm_cluster), GFP_KERNEL);
-	gps = kcalloc(3, sizeof(struct config_group *), GFP_KERNEL);
-	sps = kzalloc(sizeof(struct dlm_spaces), GFP_KERNEL);
-	cms = kzalloc(sizeof(struct dlm_comms), GFP_KERNEL);
+	cl = kzalloc(sizeof(struct dlm_cluster), GFP_NOFS);
+	gps = kcalloc(3, sizeof(struct config_group *), GFP_NOFS);
+	sps = kzalloc(sizeof(struct dlm_spaces), GFP_NOFS);
+	cms = kzalloc(sizeof(struct dlm_comms), GFP_NOFS);
 
 	if (!cl || !gps || !sps || !cms)
 		goto fail;
@@ -482,9 +482,9 @@ static struct config_group *make_space(s
 	struct dlm_nodes *nds = NULL;
 	void *gps = NULL;
 
-	sp = kzalloc(sizeof(struct dlm_space), GFP_KERNEL);
-	gps = kcalloc(2, sizeof(struct config_group *), GFP_KERNEL);
-	nds = kzalloc(sizeof(struct dlm_nodes), GFP_KERNEL);
+	sp = kzalloc(sizeof(struct dlm_space), GFP_NOFS);
+	gps = kcalloc(2, sizeof(struct config_group *), GFP_NOFS);
+	nds = kzalloc(sizeof(struct dlm_nodes), GFP_NOFS);
 
 	if (!sp || !gps || !nds)
 		goto fail;
@@ -536,7 +536,7 @@ static struct config_item *make_comm(str
 {
 	struct dlm_comm *cm;
 
-	cm = kzalloc(sizeof(struct dlm_comm), GFP_KERNEL);
+	cm = kzalloc(sizeof(struct dlm_comm), GFP_NOFS);
 	if (!cm)
 		return ERR_PTR(-ENOMEM);
 
@@ -569,7 +569,7 @@ static struct config_item *make_node(str
 	struct dlm_space *sp = config_item_to_space(g->cg_item.ci_parent);
 	struct dlm_node *nd;
 
-	nd = kzalloc(sizeof(struct dlm_node), GFP_KERNEL);
+	nd = kzalloc(sizeof(struct dlm_node), GFP_NOFS);
 	if (!nd)
 		return ERR_PTR(-ENOMEM);
 
@@ -705,7 +705,7 @@ static ssize_t comm_addr_write(struct dl
 	if (cm->addr_count >= DLM_MAX_ADDR_COUNT)
 		return -ENOSPC;
 
-	addr = kzalloc(sizeof(*addr), GFP_KERNEL);
+	addr = kzalloc(sizeof(*addr), GFP_NOFS);
 	if (!addr)
 		return -ENOMEM;
 
@@ -868,7 +868,7 @@ int dlm_nodeid_list(char *lsname, int **
 
 	ids_count = sp->members_count;
 
-	ids = kcalloc(ids_count, sizeof(int), GFP_KERNEL);
+	ids = kcalloc(ids_count, sizeof(int), GFP_NOFS);
 	if (!ids) {
 		rv = -ENOMEM;
 		goto out;
@@ -886,7 +886,7 @@ int dlm_nodeid_list(char *lsname, int **
 	if (!new_count)
 		goto out_ids;
 
-	new = kcalloc(new_count, sizeof(int), GFP_KERNEL);
+	new = kcalloc(new_count, sizeof(int), GFP_NOFS);
 	if (!new) {
 		kfree(ids);
 		rv = -ENOMEM;
--- a/fs/dlm/debug_fs.c
+++ b/fs/dlm/debug_fs.c
@@ -404,7 +404,7 @@ static void *table_seq_start(struct seq_
 	if (bucket >= ls->ls_rsbtbl_size)
 		return NULL;
 
-	ri = kzalloc(sizeof(struct rsbtbl_iter), GFP_KERNEL);
+	ri = kzalloc(sizeof(struct rsbtbl_iter), GFP_NOFS);
 	if (!ri)
 		return NULL;
 	if (n == 0)
--- a/fs/dlm/dir.c
+++ b/fs/dlm/dir.c
@@ -49,8 +49,7 @@ static struct dlm_direntry *get_free_de(
 	spin_unlock(&ls->ls_recover_list_lock);
 
 	if (!found)
-		de = kzalloc(sizeof(struct dlm_direntry) + len,
-			     ls->ls_allocation);
+		de = kzalloc(sizeof(struct dlm_direntry) + len, GFP_NOFS);
 	return de;
 }
 
@@ -212,7 +211,7 @@ int dlm_recover_directory(struct dlm_ls
 
 	dlm_dir_clear(ls);
 
-	last_name = kmalloc(DLM_RESNAME_MAXLEN, ls->ls_allocation);
+	last_name = kmalloc(DLM_RESNAME_MAXLEN, GFP_NOFS);
 	if (!last_name)
 		goto out;
 
@@ -323,7 +322,7 @@ static int get_entry(struct dlm_ls *ls,
 	if (namelen > DLM_RESNAME_MAXLEN)
 		return -EINVAL;
 
-	de = kzalloc(sizeof(struct dlm_direntry) + namelen, ls->ls_allocation);
+	de = kzalloc(sizeof(struct dlm_direntry) + namelen, GFP_NOFS);
 	if (!de)
 		return -ENOMEM;
 
--- a/fs/dlm/dlm_internal.h
+++ b/fs/dlm/dlm_internal.h
@@ -473,7 +473,6 @@ struct dlm_ls {
 	int			ls_low_nodeid;
 	int			ls_total_weight;
 	int			*ls_node_array;
-	gfp_t			ls_allocation;
 
 	struct dlm_rsb		ls_stub_rsb;	/* for returning errors */
 	struct dlm_lkb		ls_stub_lkb;	/* for returning errors */
--- a/fs/dlm/lock.c
+++ b/fs/dlm/lock.c
@@ -2689,7 +2689,7 @@ static int _create_message(struct dlm_ls
 	   pass into lowcomms_commit and a message buffer (mb) that we
 	   write our data into */
 
-	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, ls->ls_allocation, &mb);
+	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, GFP_NOFS, &mb);
 	if (!mh)
 		return -ENOBUFS;
 
@@ -4512,7 +4512,7 @@ int dlm_user_request(struct dlm_ls *ls,
 	}
 
 	if (flags & DLM_LKF_VALBLK) {
-		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_KERNEL);
+		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_NOFS);
 		if (!ua->lksb.sb_lvbptr) {
 			kfree(ua);
 			__put_lkb(ls, lkb);
@@ -4582,7 +4582,7 @@ int dlm_user_convert(struct dlm_ls *ls,
 	ua = lkb->lkb_ua;
 
 	if (flags & DLM_LKF_VALBLK && !ua->lksb.sb_lvbptr) {
-		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_KERNEL);
+		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_NOFS);
 		if (!ua->lksb.sb_lvbptr) {
 			error = -ENOMEM;
 			goto out_put;
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -430,7 +430,7 @@ static int new_lockspace(const char *nam
 
 	error = -ENOMEM;
 
-	ls = kzalloc(sizeof(struct dlm_ls) + namelen, GFP_KERNEL);
+	ls = kzalloc(sizeof(struct dlm_ls) + namelen, GFP_NOFS);
 	if (!ls)
 		goto out;
 	memcpy(ls->ls_name, name, namelen);
@@ -443,11 +443,6 @@ static int new_lockspace(const char *nam
 	if (flags & DLM_LSFL_TIMEWARN)
 		set_bit(LSFL_TIMEWARN, &ls->ls_flags);
 
-	if (flags & DLM_LSFL_FS)
-		ls->ls_allocation = GFP_NOFS;
-	else
-		ls->ls_allocation = GFP_KERNEL;
-
 	/* ls_exflags are forced to match among nodes, and we don't
 	   need to require all nodes to have some flags set */
 	ls->ls_exflags = (flags & ~(DLM_LSFL_TIMEWARN | DLM_LSFL_FS |
@@ -456,7 +451,7 @@ static int new_lockspace(const char *nam
 	size = dlm_config.ci_rsbtbl_size;
 	ls->ls_rsbtbl_size = size;
 
-	ls->ls_rsbtbl = kmalloc(sizeof(struct dlm_rsbtable) * size, GFP_KERNEL);
+	ls->ls_rsbtbl = kmalloc(sizeof(struct dlm_rsbtable) * size, GFP_NOFS);
 	if (!ls->ls_rsbtbl)
 		goto out_lsfree;
 	for (i = 0; i < size; i++) {
@@ -468,7 +463,7 @@ static int new_lockspace(const char *nam
 	size = dlm_config.ci_lkbtbl_size;
 	ls->ls_lkbtbl_size = size;
 
-	ls->ls_lkbtbl = kmalloc(sizeof(struct dlm_lkbtable) * size, GFP_KERNEL);
+	ls->ls_lkbtbl = kmalloc(sizeof(struct dlm_lkbtable) * size, GFP_NOFS);
 	if (!ls->ls_lkbtbl)
 		goto out_rsbfree;
 	for (i = 0; i < size; i++) {
@@ -480,7 +475,7 @@ static int new_lockspace(const char *nam
 	size = dlm_config.ci_dirtbl_size;
 	ls->ls_dirtbl_size = size;
 
-	ls->ls_dirtbl = kmalloc(sizeof(struct dlm_dirtable) * size, GFP_KERNEL);
+	ls->ls_dirtbl = kmalloc(sizeof(struct dlm_dirtable) * size, GFP_NOFS);
 	if (!ls->ls_dirtbl)
 		goto out_lkbfree;
 	for (i = 0; i < size; i++) {
@@ -527,7 +522,7 @@ static int new_lockspace(const char *nam
 	mutex_init(&ls->ls_requestqueue_mutex);
 	mutex_init(&ls->ls_clear_proc_locks);
 
-	ls->ls_recover_buf = kmalloc(dlm_config.ci_buffer_size, GFP_KERNEL);
+	ls->ls_recover_buf = kmalloc(dlm_config.ci_buffer_size, GFP_NOFS);
 	if (!ls->ls_recover_buf)
 		goto out_dirfree;
 
--- a/fs/dlm/lowcomms.c
+++ b/fs/dlm/lowcomms.c
@@ -1060,7 +1060,7 @@ static void init_local(void)
 		if (dlm_our_addr(&sas, i))
 			break;
 
-		addr = kmalloc(sizeof(*addr), GFP_KERNEL);
+		addr = kmalloc(sizeof(*addr), GFP_NOFS);
 		if (!addr)
 			break;
 		memcpy(addr, &sas, sizeof(*addr));
@@ -1099,7 +1099,7 @@ static int sctp_listen_for_all(void)
 	struct sockaddr_storage localaddr;
 	struct sctp_event_subscribe subscribe;
 	int result = -EINVAL, num = 1, i, addr_len;
-	struct connection *con = nodeid2con(0, GFP_KERNEL);
+	struct connection *con = nodeid2con(0, GFP_NOFS);
 	int bufsize = NEEDED_RMEM;
 
 	if (!con)
@@ -1171,7 +1171,7 @@ out:
 static int tcp_listen_for_all(void)
 {
 	struct socket *sock = NULL;
-	struct connection *con = nodeid2con(0, GFP_KERNEL);
+	struct connection *con = nodeid2con(0, GFP_NOFS);
 	int result = -EINVAL;
 
 	if (!con)
--- a/fs/dlm/member.c
+++ b/fs/dlm/member.c
@@ -48,7 +48,7 @@ static int dlm_add_member(struct dlm_ls
 	struct dlm_member *memb;
 	int w, error;
 
-	memb = kzalloc(sizeof(struct dlm_member), ls->ls_allocation);
+	memb = kzalloc(sizeof(struct dlm_member), GFP_NOFS);
 	if (!memb)
 		return -ENOMEM;
 
@@ -143,7 +143,7 @@ static void make_member_array(struct dlm
 
 	ls->ls_total_weight = total;
 
-	array = kmalloc(sizeof(int) * total, ls->ls_allocation);
+	array = kmalloc(sizeof(int) * total, GFP_NOFS);
 	if (!array)
 		return;
 
@@ -226,7 +226,7 @@ int dlm_recover_members(struct dlm_ls *l
 			continue;
 		log_debug(ls, "new nodeid %d is a re-added member", rv->new[i]);
 
-		memb = kzalloc(sizeof(struct dlm_member), ls->ls_allocation);
+		memb = kzalloc(sizeof(struct dlm_member), GFP_NOFS);
 		if (!memb)
 			return -ENOMEM;
 		memb->nodeid = rv->new[i];
@@ -341,7 +341,7 @@ int dlm_ls_start(struct dlm_ls *ls)
 	int *ids = NULL, *new = NULL;
 	int error, ids_count = 0, new_count = 0;
 
-	rv = kzalloc(sizeof(struct dlm_recover), ls->ls_allocation);
+	rv = kzalloc(sizeof(struct dlm_recover), GFP_NOFS);
 	if (!rv)
 		return -ENOMEM;
 
--- a/fs/dlm/memory.c
+++ b/fs/dlm/memory.c
@@ -39,7 +39,7 @@ char *dlm_allocate_lvb(struct dlm_ls *ls
 {
 	char *p;
 
-	p = kzalloc(ls->ls_lvblen, ls->ls_allocation);
+	p = kzalloc(ls->ls_lvblen, GFP_NOFS);
 	return p;
 }
 
@@ -57,7 +57,7 @@ struct dlm_rsb *dlm_allocate_rsb(struct
 
 	DLM_ASSERT(namelen <= DLM_RESNAME_MAXLEN,);
 
-	r = kzalloc(sizeof(*r) + namelen, ls->ls_allocation);
+	r = kzalloc(sizeof(*r) + namelen, GFP_NOFS);
 	return r;
 }
 
@@ -72,7 +72,7 @@ struct dlm_lkb *dlm_allocate_lkb(struct
 {
 	struct dlm_lkb *lkb;
 
-	lkb = kmem_cache_zalloc(lkb_cache, ls->ls_allocation);
+	lkb = kmem_cache_zalloc(lkb_cache, GFP_NOFS);
 	return lkb;
 }
 
--- a/fs/dlm/netlink.c
+++ b/fs/dlm/netlink.c
@@ -26,7 +26,7 @@ static int prepare_data(u8 cmd, struct s
 	struct sk_buff *skb;
 	void *data;
 
-	skb = genlmsg_new(size, GFP_KERNEL);
+	skb = genlmsg_new(size, GFP_NOFS);
 	if (!skb)
 		return -ENOMEM;
 
--- a/fs/dlm/plock.c
+++ b/fs/dlm/plock.c
@@ -82,7 +82,7 @@ int dlm_posix_lock(dlm_lockspace_t *lock
 	if (!ls)
 		return -EINVAL;
 
-	xop = kzalloc(sizeof(*xop), GFP_KERNEL);
+	xop = kzalloc(sizeof(*xop), GFP_NOFS);
 	if (!xop) {
 		rv = -ENOMEM;
 		goto out;
@@ -211,7 +211,7 @@ int dlm_posix_unlock(dlm_lockspace_t *lo
 	if (!ls)
 		return -EINVAL;
 
-	op = kzalloc(sizeof(*op), GFP_KERNEL);
+	op = kzalloc(sizeof(*op), GFP_NOFS);
 	if (!op) {
 		rv = -ENOMEM;
 		goto out;
@@ -266,7 +266,7 @@ int dlm_posix_get(dlm_lockspace_t *locks
 	if (!ls)
 		return -EINVAL;
 
-	op = kzalloc(sizeof(*op), GFP_KERNEL);
+	op = kzalloc(sizeof(*op), GFP_NOFS);
 	if (!op) {
 		rv = -ENOMEM;
 		goto out;
--- a/fs/dlm/rcom.c
+++ b/fs/dlm/rcom.c
@@ -38,7 +38,7 @@ static int create_rcom(struct dlm_ls *ls
 	char *mb;
 	int mb_len = sizeof(struct dlm_rcom) + len;
 
-	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, ls->ls_allocation, &mb);
+	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, GFP_NOFS, &mb);
 	if (!mh) {
 		log_print("create_rcom to %d type %d len %d ENOBUFS",
 			  to_nodeid, type, len);
--- a/fs/dlm/requestqueue.c
+++ b/fs/dlm/requestqueue.c
@@ -35,7 +35,7 @@ void dlm_add_requestqueue(struct dlm_ls
 	struct rq_entry *e;
 	int length = ms->m_header.h_length - sizeof(struct dlm_message);
 
-	e = kmalloc(sizeof(struct rq_entry) + length, ls->ls_allocation);
+	e = kmalloc(sizeof(struct rq_entry) + length, GFP_NOFS);
 	if (!e) {
 		log_print("dlm_add_requestqueue: out of memory len %d", length);
 		return;
--- a/fs/dlm/user.c
+++ b/fs/dlm/user.c
@@ -267,7 +267,7 @@ static int device_user_lock(struct dlm_u
 		goto out;
 	}
 
-	ua = kzalloc(sizeof(struct dlm_user_args), GFP_KERNEL);
+	ua = kzalloc(sizeof(struct dlm_user_args), GFP_NOFS);
 	if (!ua)
 		goto out;
 	ua->proc = proc;
@@ -307,7 +307,7 @@ static int device_user_unlock(struct dlm
 	if (!ls)
 		return -ENOENT;
 
-	ua = kzalloc(sizeof(struct dlm_user_args), GFP_KERNEL);
+	ua = kzalloc(sizeof(struct dlm_user_args), GFP_NOFS);
 	if (!ua)
 		goto out;
 	ua->proc = proc;
@@ -352,7 +352,7 @@ static int dlm_device_register(struct dl
 
 	error = -ENOMEM;
 	len = strlen(name) + strlen(name_prefix) + 2;
-	ls->ls_device.name = kzalloc(len, GFP_KERNEL);
+	ls->ls_device.name = kzalloc(len, GFP_NOFS);
 	if (!ls->ls_device.name)
 		goto fail;
 
@@ -520,7 +520,7 @@ static ssize_t device_write(struct file
 #endif
 		return -EINVAL;
 
-	kbuf = kzalloc(count + 1, GFP_KERNEL);
+	kbuf = kzalloc(count + 1, GFP_NOFS);
 	if (!kbuf)
 		return -ENOMEM;
 
@@ -546,7 +546,7 @@ static ssize_t device_write(struct file
 
 		/* add 1 after namelen so that the name string is terminated */
 		kbuf = kzalloc(sizeof(struct dlm_write_request) + namelen + 1,
-			       GFP_KERNEL);
+			       GFP_NOFS);
 		if (!kbuf) {
 			kfree(k32buf);
 			return -ENOMEM;
@@ -648,7 +648,7 @@ static int device_open(struct inode *ino
 	if (!ls)
 		return -ENOENT;
 
-	proc = kzalloc(sizeof(struct dlm_user_proc), GFP_KERNEL);
+	proc = kzalloc(sizeof(struct dlm_user_proc), GFP_NOFS);
 	if (!proc) {
 		dlm_put_lockspace(ls);
 		return -ENOMEM;

[101/111] dlm: fix ordering of bast and cast

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: David Teigland <teigland@redhat.com>

commit 7fe2b3190b8b299409f13cf3a6f85c2bd371f8bb upstream.

When both blocking and completion callbacks are queued for lock,
the dlm would always deliver the completion callback (cast) first.
In some cases the blocking callback (bast) is queued before the
cast, though, and should be delivered first.  This patch keeps
track of the order in which they were queued and delivers them
in that order.

This patch also keeps track of the granted mode in the last cast
and eliminates the following bast if the bast mode is compatible
with the preceding cast mode.  This happens when a remotely mastered
lock is demoted, e.g. EX->NL, in which case the local node queues
a cast immediately after sending the demote message.  In this way
a cast can be queued for a mode, e.g. NL, that makes an in-transit
bast extraneous.

Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/dlm/ast.c          |   72 +++++++++++++++++++++++++++++++++++++++-----------
 fs/dlm/ast.h          |    4 +-
 fs/dlm/dlm_internal.h |   10 +++++-
 fs/dlm/lock.c         |    4 +-
 fs/dlm/user.c         |   10 ++++--
 fs/dlm/user.h         |    4 +-
 6 files changed, 77 insertions(+), 27 deletions(-)

--- a/fs/dlm/ast.c
+++ b/fs/dlm/ast.c
@@ -2,7 +2,7 @@
 *******************************************************************************
 **
 **  Copyright (C) Sistina Software, Inc.  1997-2003  All rights reserved.
-**  Copyright (C) 2004-2008 Red Hat, Inc.  All rights reserved.
+**  Copyright (C) 2004-2010 Red Hat, Inc.  All rights reserved.
 **
 **  This copyrighted material is made available to anyone wishing to use,
 **  modify, copy, or redistribute it subject to the terms and conditions
@@ -33,10 +33,10 @@ void dlm_del_ast(struct dlm_lkb *lkb)
 	spin_unlock(&ast_queue_lock);
 }
 
-void dlm_add_ast(struct dlm_lkb *lkb, int type, int bastmode)
+void dlm_add_ast(struct dlm_lkb *lkb, int type, int mode)
 {
 	if (lkb->lkb_flags & DLM_IFL_USER) {
-		dlm_user_add_ast(lkb, type, bastmode);
+		dlm_user_add_ast(lkb, type, mode);
 		return;
 	}
 
@@ -44,10 +44,21 @@ void dlm_add_ast(struct dlm_lkb *lkb, in
 	if (!(lkb->lkb_ast_type & (AST_COMP | AST_BAST))) {
 		kref_get(&lkb->lkb_ref);
 		list_add_tail(&lkb->lkb_astqueue, &ast_queue);
+		lkb->lkb_ast_first = type;
 	}
+
+	/* sanity check, this should not happen */
+
+	if ((type == AST_COMP) && (lkb->lkb_ast_type & AST_COMP))
+		log_print("repeat cast %d castmode %d lock %x %s",
+			  mode, lkb->lkb_castmode,
+			  lkb->lkb_id, lkb->lkb_resource->res_name);
+
 	lkb->lkb_ast_type |= type;
-	if (bastmode)
-		lkb->lkb_bastmode = bastmode;
+	if (type == AST_BAST)
+		lkb->lkb_bastmode = mode;
+	else
+		lkb->lkb_castmode = mode;
 	spin_unlock(&ast_queue_lock);
 
 	set_bit(WAKE_ASTS, &astd_wakeflags);
@@ -59,9 +70,9 @@ static void process_asts(void)
 	struct dlm_ls *ls = NULL;
 	struct dlm_rsb *r = NULL;
 	struct dlm_lkb *lkb;
-	void (*cast) (void *astparam);
-	void (*bast) (void *astparam, int mode);
-	int type = 0, bastmode;
+	void (*castfn) (void *astparam);
+	void (*bastfn) (void *astparam, int mode);
+	int type, first, bastmode, castmode, do_bast, do_cast, last_castmode;
 
 repeat:
 	spin_lock(&ast_queue_lock);
@@ -75,17 +86,48 @@ repeat:
 		list_del(&lkb->lkb_astqueue);
 		type = lkb->lkb_ast_type;
 		lkb->lkb_ast_type = 0;
+		first = lkb->lkb_ast_first;
+		lkb->lkb_ast_first = 0;
 		bastmode = lkb->lkb_bastmode;
-
+		castmode = lkb->lkb_castmode;
+		castfn = lkb->lkb_astfn;
+		bastfn = lkb->lkb_bastfn;
 		spin_unlock(&ast_queue_lock);
-		cast = lkb->lkb_astfn;
-		bast = lkb->lkb_bastfn;
 
-		if ((type & AST_COMP) && cast)
-			cast(lkb->lkb_astparam);
+		do_cast = (type & AST_COMP) && castfn;
+		do_bast = (type & AST_BAST) && bastfn;
+
+		/* Skip a bast if its blocking mode is compatible with the
+		   granted mode of the preceding cast. */
 
-		if ((type & AST_BAST) && bast)
-			bast(lkb->lkb_astparam, bastmode);
+		if (do_bast) {
+			if (first == AST_COMP)
+				last_castmode = castmode;
+			else
+				last_castmode = lkb->lkb_castmode_done;
+			if (dlm_modes_compat(bastmode, last_castmode))
+				do_bast = 0;
+		}
+
+		if (first == AST_COMP) {
+			if (do_cast)
+				castfn(lkb->lkb_astparam);
+			if (do_bast)
+				bastfn(lkb->lkb_astparam, bastmode);
+		} else if (first == AST_BAST) {
+			if (do_bast)
+				bastfn(lkb->lkb_astparam, bastmode);
+			if (do_cast)
+				castfn(lkb->lkb_astparam);
+		} else {
+			log_error(ls, "bad ast_first %d ast_type %d",
+				  first, type);
+		}
+
+		if (do_cast)
+			lkb->lkb_castmode_done = castmode;
+		if (do_bast)
+			lkb->lkb_bastmode_done = bastmode;
 
 		/* this removes the reference added by dlm_add_ast
 		   and may result in the lkb being freed */
--- a/fs/dlm/ast.h
+++ b/fs/dlm/ast.h
@@ -1,7 +1,7 @@
 /******************************************************************************
 *******************************************************************************
 **
-**  Copyright (C) 2005-2008 Red Hat, Inc.  All rights reserved.
+**  Copyright (C) 2005-2010 Red Hat, Inc.  All rights reserved.
 **
 **  This copyrighted material is made available to anyone wishing to use,
 **  modify, copy, or redistribute it subject to the terms and conditions
@@ -13,7 +13,7 @@
 #ifndef __ASTD_DOT_H__
 #define __ASTD_DOT_H__
 
-void dlm_add_ast(struct dlm_lkb *lkb, int type, int bastmode);
+void dlm_add_ast(struct dlm_lkb *lkb, int type, int mode);
 void dlm_del_ast(struct dlm_lkb *lkb);
 
 void dlm_astd_wake(void);
--- a/fs/dlm/dlm_internal.h
+++ b/fs/dlm/dlm_internal.h
@@ -2,7 +2,7 @@
 *******************************************************************************
 **
 **  Copyright (C) Sistina Software, Inc.  1997-2003  All rights reserved.
-**  Copyright (C) 2004-2008 Red Hat, Inc.  All rights reserved.
+**  Copyright (C) 2004-2010 Red Hat, Inc.  All rights reserved.
 **
 **  This copyrighted material is made available to anyone wishing to use,
 **  modify, copy, or redistribute it subject to the terms and conditions
@@ -232,11 +232,17 @@ struct dlm_lkb {
 	int8_t			lkb_status;     /* granted, waiting, convert */
 	int8_t			lkb_rqmode;	/* requested lock mode */
 	int8_t			lkb_grmode;	/* granted lock mode */
-	int8_t			lkb_bastmode;	/* requested mode */
 	int8_t			lkb_highbast;	/* highest mode bast sent for */
+
 	int8_t			lkb_wait_type;	/* type of reply waiting for */
 	int8_t			lkb_wait_count;
 	int8_t			lkb_ast_type;	/* type of ast queued for */
+	int8_t			lkb_ast_first;	/* type of first ast queued */
+
+	int8_t			lkb_bastmode;	/* req mode of queued bast */
+	int8_t			lkb_castmode;	/* gr mode of queued cast */
+	int8_t			lkb_bastmode_done; /* last delivered bastmode */
+	int8_t			lkb_castmode_done; /* last delivered castmode */
 
 	struct list_head	lkb_idtbl_list;	/* lockspace lkbtbl */
 	struct list_head	lkb_statequeue;	/* rsb g/c/w list */
--- a/fs/dlm/lock.c
+++ b/fs/dlm/lock.c
@@ -1,7 +1,7 @@
 /******************************************************************************
 *******************************************************************************
 **
-**  Copyright (C) 2005-2008 Red Hat, Inc.  All rights reserved.
+**  Copyright (C) 2005-2010 Red Hat, Inc.  All rights reserved.
 **
 **  This copyrighted material is made available to anyone wishing to use,
 **  modify, copy, or redistribute it subject to the terms and conditions
@@ -307,7 +307,7 @@ static void queue_cast(struct dlm_rsb *r
 	lkb->lkb_lksb->sb_status = rv;
 	lkb->lkb_lksb->sb_flags = lkb->lkb_sbflags;
 
-	dlm_add_ast(lkb, AST_COMP, 0);
+	dlm_add_ast(lkb, AST_COMP, lkb->lkb_grmode);
 }
 
 static inline void queue_cast_overlap(struct dlm_rsb *r, struct dlm_lkb *lkb)
--- a/fs/dlm/user.c
+++ b/fs/dlm/user.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2006-2009 Red Hat, Inc.  All rights reserved.
+ * Copyright (C) 2006-2010 Red Hat, Inc.  All rights reserved.
  *
  * This copyrighted material is made available to anyone wishing to use,
  * modify, copy, or redistribute it subject to the terms and conditions
@@ -173,7 +173,7 @@ static int lkb_is_endoflife(struct dlm_l
 /* we could possibly check if the cancel of an orphan has resulted in the lkb
    being removed and then remove that lkb from the orphans list and free it */
 
-void dlm_user_add_ast(struct dlm_lkb *lkb, int type, int bastmode)
+void dlm_user_add_ast(struct dlm_lkb *lkb, int type, int mode)
 {
 	struct dlm_ls *ls;
 	struct dlm_user_args *ua;
@@ -206,8 +206,10 @@ void dlm_user_add_ast(struct dlm_lkb *lk
 
 	ast_type = lkb->lkb_ast_type;
 	lkb->lkb_ast_type |= type;
-	if (bastmode)
-		lkb->lkb_bastmode = bastmode;
+	if (type == AST_BAST)
+		lkb->lkb_bastmode = mode;
+	else
+		lkb->lkb_castmode = mode;
 
 	if (!ast_type) {
 		kref_get(&lkb->lkb_ref);
--- a/fs/dlm/user.h
+++ b/fs/dlm/user.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2006-2008 Red Hat, Inc.  All rights reserved.
+ * Copyright (C) 2006-2010 Red Hat, Inc.  All rights reserved.
  *
  * This copyrighted material is made available to anyone wishing to use,
  * modify, copy, or redistribute it subject to the terms and conditions
@@ -9,7 +9,7 @@
 #ifndef __USER_DOT_H__
 #define __USER_DOT_H__
 
-void dlm_user_add_ast(struct dlm_lkb *lkb, int type, int bastmode);
+void dlm_user_add_ast(struct dlm_lkb *lkb, int type, int mode);
 int dlm_user_init(void);
 void dlm_user_exit(void);
 int dlm_device_deregister(struct dlm_ls *ls);

[102/111] dlm: send reply before bast

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: David Teigland <teigland@redhat.com>

commit cf6620acc0f6fac57968aafef79ab372bdcf6157 upstream.

When the lock master processes a successful operation (request,
convert, cancel, or unlock), it will process the effects of the
change before sending the reply for the operation.  The "effects"
of the operation are:

- blocking callbacks (basts) for any newly granted locks
- waiting or converting locks that can now be granted

The cast is queued on the local node when the reply from the lock
master is received.  This means that a lock holder can receive a
bast for a lock mode that is doesn't yet know has been granted.

Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/dlm/lock.c |  110 ++++++++++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 84 insertions(+), 26 deletions(-)

--- a/fs/dlm/lock.c
+++ b/fs/dlm/lock.c
@@ -2280,20 +2280,30 @@ static int do_request(struct dlm_rsb *r,
 	if (can_be_queued(lkb)) {
 		error = -EINPROGRESS;
 		add_lkb(r, lkb, DLM_LKSTS_WAITING);
-		send_blocking_asts(r, lkb);
 		add_timeout(lkb);
 		goto out;
 	}
 
 	error = -EAGAIN;
-	if (force_blocking_asts(lkb))
-		send_blocking_asts_all(r, lkb);
 	queue_cast(r, lkb, -EAGAIN);
-
  out:
 	return error;
 }
 
+static void do_request_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
+			       int error)
+{
+	switch (error) {
+	case -EAGAIN:
+		if (force_blocking_asts(lkb))
+			send_blocking_asts_all(r, lkb);
+		break;
+	case -EINPROGRESS:
+		send_blocking_asts(r, lkb);
+		break;
+	}
+}
+
 static int do_convert(struct dlm_rsb *r, struct dlm_lkb *lkb)
 {
 	int error = 0;
@@ -2304,7 +2314,6 @@ static int do_convert(struct dlm_rsb *r,
 	if (can_be_granted(r, lkb, 1, &deadlk)) {
 		grant_lock(r, lkb);
 		queue_cast(r, lkb, 0);
-		grant_pending_locks(r);
 		goto out;
 	}
 
@@ -2334,7 +2343,6 @@ static int do_convert(struct dlm_rsb *r,
 		if (_can_be_granted(r, lkb, 1)) {
 			grant_lock(r, lkb);
 			queue_cast(r, lkb, 0);
-			grant_pending_locks(r);
 			goto out;
 		}
 		/* else fall through and move to convert queue */
@@ -2344,28 +2352,47 @@ static int do_convert(struct dlm_rsb *r,
 		error = -EINPROGRESS;
 		del_lkb(r, lkb);
 		add_lkb(r, lkb, DLM_LKSTS_CONVERT);
-		send_blocking_asts(r, lkb);
 		add_timeout(lkb);
 		goto out;
 	}
 
 	error = -EAGAIN;
-	if (force_blocking_asts(lkb))
-		send_blocking_asts_all(r, lkb);
 	queue_cast(r, lkb, -EAGAIN);
-
  out:
 	return error;
 }
 
+static void do_convert_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
+			       int error)
+{
+	switch (error) {
+	case 0:
+		grant_pending_locks(r);
+		/* grant_pending_locks also sends basts */
+		break;
+	case -EAGAIN:
+		if (force_blocking_asts(lkb))
+			send_blocking_asts_all(r, lkb);
+		break;
+	case -EINPROGRESS:
+		send_blocking_asts(r, lkb);
+		break;
+	}
+}
+
 static int do_unlock(struct dlm_rsb *r, struct dlm_lkb *lkb)
 {
 	remove_lock(r, lkb);
 	queue_cast(r, lkb, -DLM_EUNLOCK);
-	grant_pending_locks(r);
 	return -DLM_EUNLOCK;
 }
 
+static void do_unlock_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
+			      int error)
+{
+	grant_pending_locks(r);
+}
+
 /* returns: 0 did nothing, -DLM_ECANCEL canceled lock */
  
 static int do_cancel(struct dlm_rsb *r, struct dlm_lkb *lkb)
@@ -2375,12 +2402,18 @@ static int do_cancel(struct dlm_rsb *r,
 	error = revert_lock(r, lkb);
 	if (error) {
 		queue_cast(r, lkb, -DLM_ECANCEL);
-		grant_pending_locks(r);
 		return -DLM_ECANCEL;
 	}
 	return 0;
 }
 
+static void do_cancel_effects(struct dlm_rsb *r, struct dlm_lkb *lkb,
+			      int error)
+{
+	if (error)
+		grant_pending_locks(r);
+}
+
 /*
  * Four stage 3 varieties:
  * _request_lock(), _convert_lock(), _unlock_lock(), _cancel_lock()
@@ -2402,11 +2435,15 @@ static int _request_lock(struct dlm_rsb
 		goto out;
 	}
 
-	if (is_remote(r))
+	if (is_remote(r)) {
 		/* receive_request() calls do_request() on remote node */
 		error = send_request(r, lkb);
-	else
+	} else {
 		error = do_request(r, lkb);
+		/* for remote locks the request_reply is sent
+		   between do_request and do_request_effects */
+		do_request_effects(r, lkb, error);
+	}
  out:
 	return error;
 }
@@ -2417,11 +2454,15 @@ static int _convert_lock(struct dlm_rsb
 {
 	int error;
 
-	if (is_remote(r))
+	if (is_remote(r)) {
 		/* receive_convert() calls do_convert() on remote node */
 		error = send_convert(r, lkb);
-	else
+	} else {
 		error = do_convert(r, lkb);
+		/* for remote locks the convert_reply is sent
+		   between do_convert and do_convert_effects */
+		do_convert_effects(r, lkb, error);
+	}
 
 	return error;
 }
@@ -2432,11 +2473,15 @@ static int _unlock_lock(struct dlm_rsb *
 {
 	int error;
 
-	if (is_remote(r))
+	if (is_remote(r)) {
 		/* receive_unlock() calls do_unlock() on remote node */
 		error = send_unlock(r, lkb);
-	else
+	} else {
 		error = do_unlock(r, lkb);
+		/* for remote locks the unlock_reply is sent
+		   between do_unlock and do_unlock_effects */
+		do_unlock_effects(r, lkb, error);
+	}
 
 	return error;
 }
@@ -2447,11 +2492,15 @@ static int _cancel_lock(struct dlm_rsb *
 {
 	int error;
 
-	if (is_remote(r))
+	if (is_remote(r)) {
 		/* receive_cancel() calls do_cancel() on remote node */
 		error = send_cancel(r, lkb);
-	else
+	} else {
 		error = do_cancel(r, lkb);
+		/* for remote locks the cancel_reply is sent
+		   between do_cancel and do_cancel_effects */
+		do_cancel_effects(r, lkb, error);
+	}
 
 	return error;
 }
@@ -3191,6 +3240,7 @@ static void receive_request(struct dlm_l
 	attach_lkb(r, lkb);
 	error = do_request(r, lkb);
 	send_request_reply(r, lkb, error);
+	do_request_effects(r, lkb, error);
 
 	unlock_rsb(r);
 	put_rsb(r);
@@ -3226,15 +3276,19 @@ static void receive_convert(struct dlm_l
 		goto out;
 
 	receive_flags(lkb, ms);
+
 	error = receive_convert_args(ls, lkb, ms);
-	if (error)
-		goto out_reply;
+	if (error) {
+		send_convert_reply(r, lkb, error);
+		goto out;
+	}
+
 	reply = !down_conversion(lkb);
 
 	error = do_convert(r, lkb);
- out_reply:
 	if (reply)
 		send_convert_reply(r, lkb, error);
+	do_convert_effects(r, lkb, error);
  out:
 	unlock_rsb(r);
 	put_rsb(r);
@@ -3266,13 +3320,16 @@ static void receive_unlock(struct dlm_ls
 		goto out;
 
 	receive_flags(lkb, ms);
+
 	error = receive_unlock_args(ls, lkb, ms);
-	if (error)
-		goto out_reply;
+	if (error) {
+		send_unlock_reply(r, lkb, error);
+		goto out;
+	}
 
 	error = do_unlock(r, lkb);
- out_reply:
 	send_unlock_reply(r, lkb, error);
+	do_unlock_effects(r, lkb, error);
  out:
 	unlock_rsb(r);
 	put_rsb(r);
@@ -3307,6 +3364,7 @@ static void receive_cancel(struct dlm_ls
 
 	error = do_cancel(r, lkb);
 	send_cancel_reply(r, lkb, error);
+	do_cancel_effects(r, lkb, error);
  out:
 	unlock_rsb(r);
 	put_rsb(r);

[103/111] ocfs2: Find proper end cpos for a leaf refcount block.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Tao Ma <tao.ma@oracle.com>

commit 38a04e432768ec0b016f3c687b4de31ac111ae59 upstream.

ocfs2 refcount tree is stored as an extent tree while
the leaf ocfs2_refcount_rec points to a refcount block.

The following step can trip a kernel panic.
mkfs.ocfs2 -b 512 -C 1M --fs-features=refcount $DEVICE
mount -t ocfs2 $DEVICE $MNT_DIR
FILE_NAME=$RANDOM
FILE_NAME_1=$RANDOM
FILE_REF="${FILE_NAME}_ref"
FILE_REF_1="${FILE_NAME}_ref_1"
for((i=0;i<305;i++))
do
# /mnt/1048576 is a file with 1048576 sizes.
cat /mnt/1048576 >> $MNT_DIR/$FILE_NAME
cat /mnt/1048576 >> $MNT_DIR/$FILE_NAME_1
done
for((i=0;i<3;i++))
do
cat /mnt/1048576 >> $MNT_DIR/$FILE_NAME
done

for((i=0;i<2;i++))
do
cat /mnt/1048576 >> $MNT_DIR/$FILE_NAME
cat /mnt/1048576 >> $MNT_DIR/$FILE_NAME_1
done

cat /mnt/1048576 >> $MNT_DIR/$FILE_NAME

for((i=0;i<11;i++))
do
cat /mnt/1048576 >> $MNT_DIR/$FILE_NAME
cat /mnt/1048576 >> $MNT_DIR/$FILE_NAME_1
done
reflink $MNT_DIR/$FILE_NAME $MNT_DIR/$FILE_REF
# write_f is a program which will write some bytes to a file at offset.
# write_f -f file_name -l offset -w write_bytes.
./write_f -f $MNT_DIR/$FILE_REF -l $[310*1048576] -w 4096
./write_f -f $MNT_DIR/$FILE_REF -l $[306*1048576] -w 4096
./write_f -f $MNT_DIR/$FILE_REF -l $[311*1048576] -w 4096
./write_f -f $MNT_DIR/$FILE_NAME -l $[310*1048576] -w 4096
./write_f -f $MNT_DIR/$FILE_NAME -l $[311*1048576] -w 4096
reflink $MNT_DIR/$FILE_NAME $MNT_DIR/$FILE_REF_1
./write_f -f $MNT_DIR/$FILE_NAME -l $[311*1048576] -w 4096
#kernel panic here.

The reason is that if the ocfs2_extent_rec is the last record
in a leaf extent block, the old solution fails to find the
suitable end cpos. So this patch try to walk through the b-tree,
find the next sub root and get the c_pos the next sub-tree starts
from.

btw, I have runned tristan's test case against the patched kernel
for several days and this type of kernel panic never happens again.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ocfs2/alloc.c        |   10 ++--
 fs/ocfs2/alloc.h        |    5 ++
 fs/ocfs2/refcounttree.c |  117 ++++++++++++++++++++++++++++++++++++++++++++----
 3 files changed, 119 insertions(+), 13 deletions(-)

--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -1765,9 +1765,9 @@ set_and_inc:
  *
  * The array index of the subtree root is passed back.
  */
-static int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
-				   struct ocfs2_path *left,
-				   struct ocfs2_path *right)
+int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
+			    struct ocfs2_path *left,
+			    struct ocfs2_path *right)
 {
 	int i = 0;
 
@@ -2872,8 +2872,8 @@ out:
  * This looks similar, but is subtly different to
  * ocfs2_find_cpos_for_left_leaf().
  */
-static int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
-					  struct ocfs2_path *path, u32 *cpos)
+int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
+				   struct ocfs2_path *path, u32 *cpos)
 {
 	int i, j, ret = 0;
 	u64 blkno;
--- a/fs/ocfs2/alloc.h
+++ b/fs/ocfs2/alloc.h
@@ -317,4 +317,9 @@ int ocfs2_path_bh_journal_access(handle_
 int ocfs2_journal_access_path(struct ocfs2_caching_info *ci,
 			      handle_t *handle,
 			      struct ocfs2_path *path);
+int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
+				   struct ocfs2_path *path, u32 *cpos);
+int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
+			    struct ocfs2_path *left,
+			    struct ocfs2_path *right);
 #endif /* OCFS2_ALLOC_H */
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -969,6 +969,103 @@ out:
 }
 
 /*
+ * Find the end range for a leaf refcount block indicated by
+ * el->l_recs[index].e_blkno.
+ */
+static int ocfs2_get_refcount_cpos_end(struct ocfs2_caching_info *ci,
+				       struct buffer_head *ref_root_bh,
+				       struct ocfs2_extent_block *eb,
+				       struct ocfs2_extent_list *el,
+				       int index,  u32 *cpos_end)
+{
+	int ret, i, subtree_root;
+	u32 cpos;
+	u64 blkno;
+	struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
+	struct ocfs2_path *left_path = NULL, *right_path = NULL;
+	struct ocfs2_extent_tree et;
+	struct ocfs2_extent_list *tmp_el;
+
+	if (index < le16_to_cpu(el->l_next_free_rec) - 1) {
+		/*
+		 * We have a extent rec after index, so just use the e_cpos
+		 * of the next extent rec.
+		 */
+		*cpos_end = le32_to_cpu(el->l_recs[index+1].e_cpos);
+		return 0;
+	}
+
+	if (!eb || (eb && !eb->h_next_leaf_blk)) {
+		/*
+		 * We are the last extent rec, so any high cpos should
+		 * be stored in this leaf refcount block.
+		 */
+		*cpos_end = UINT_MAX;
+		return 0;
+	}
+
+	/*
+	 * If the extent block isn't the last one, we have to find
+	 * the subtree root between this extent block and the next
+	 * leaf extent block and get the corresponding e_cpos from
+	 * the subroot. Otherwise we may corrupt the b-tree.
+	 */
+	ocfs2_init_refcount_extent_tree(&et, ci, ref_root_bh);
+
+	left_path = ocfs2_new_path_from_et(&et);
+	if (!left_path) {
+		ret = -ENOMEM;
+		mlog_errno(ret);
+		goto out;
+	}
+
+	cpos = le32_to_cpu(eb->h_list.l_recs[index].e_cpos);
+	ret = ocfs2_find_path(ci, left_path, cpos);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	right_path = ocfs2_new_path_from_path(left_path);
+	if (!right_path) {
+		ret = -ENOMEM;
+		mlog_errno(ret);
+		goto out;
+	}
+
+	ret = ocfs2_find_cpos_for_right_leaf(sb, left_path, &cpos);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	ret = ocfs2_find_path(ci, right_path, cpos);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	subtree_root = ocfs2_find_subtree_root(&et, left_path,
+					       right_path);
+
+	tmp_el = left_path->p_node[subtree_root].el;
+	blkno = left_path->p_node[subtree_root+1].bh->b_blocknr;
+	for (i = 0; i < le32_to_cpu(tmp_el->l_next_free_rec); i++) {
+		if (le64_to_cpu(tmp_el->l_recs[i].e_blkno) == blkno) {
+			*cpos_end = le32_to_cpu(tmp_el->l_recs[i+1].e_cpos);
+			break;
+		}
+	}
+
+	BUG_ON(i == le32_to_cpu(tmp_el->l_next_free_rec));
+
+out:
+	ocfs2_free_path(left_path);
+	ocfs2_free_path(right_path);
+	return ret;
+}
+
+/*
  * Given a cpos and len, try to find the refcount record which contains cpos.
  * 1. If cpos can be found in one refcount record, return the record.
  * 2. If cpos can't be found, return a fake record which start from cpos
@@ -983,10 +1080,10 @@ static int ocfs2_get_refcount_rec(struct
 				  struct buffer_head **ret_bh)
 {
 	int ret = 0, i, found;
-	u32 low_cpos;
+	u32 low_cpos, uninitialized_var(cpos_end);
 	struct ocfs2_extent_list *el;
-	struct ocfs2_extent_rec *tmp, *rec = NULL;
-	struct ocfs2_extent_block *eb;
+	struct ocfs2_extent_rec *rec = NULL;
+	struct ocfs2_extent_block *eb = NULL;
 	struct buffer_head *eb_bh = NULL, *ref_leaf_bh = NULL;
 	struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
 	struct ocfs2_refcount_block *rb =
@@ -1034,12 +1131,16 @@ static int ocfs2_get_refcount_rec(struct
 		}
 	}
 
-	/* adjust len when we have ocfs2_extent_rec after it. */
-	if (found && i < le16_to_cpu(el->l_next_free_rec) - 1) {
-		tmp = &el->l_recs[i+1];
+	if (found) {
+		ret = ocfs2_get_refcount_cpos_end(ci, ref_root_bh,
+						  eb, el, i, &cpos_end);
+		if (ret) {
+			mlog_errno(ret);
+			goto out;
+		}
 
-		if (le32_to_cpu(tmp->e_cpos) < cpos + len)
-			len = le32_to_cpu(tmp->e_cpos) - cpos;
+		if (cpos_end < low_cpos + len)
+			len = cpos_end - low_cpos;
 	}
 
 	ret = ocfs2_read_refcount_block(ci, le64_to_cpu(rec->e_blkno),

[104/111] ocfs2: Set MS_POSIXACL on remount

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Jan Kara <jack@suse.cz>

commit 57b09bb5e492c37c1e4273fe4e435ffd1d2ddbe0 upstream.

We have to set MS_POSIXACL on remount as well. Otherwise VFS
would not know we started supporting ACLs after remount and
thus ACLs would not work.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ocfs2/super.c |    4 ++++
 1 file changed, 4 insertions(+)

--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -701,6 +701,10 @@ unlock_osb:
 
 		if (!ocfs2_is_hard_readonly(osb))
 			ocfs2_set_journal_params(osb);
+
+		sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
+			((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ?
+							MS_POSIXACL : 0);
 	}
 out:
 	unlock_kernel();

[105/111] [PATCH] Skip check for mandatory locks when unlocking

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Sachin Prabhu <sprabhu@redhat.com>

commit ee860b6a650360c91f5d5f9a94262aad9be90015 upstream.

ocfs2_lock() will skip locks on file which has mode set to 02666. This
is a problem in cases where the mode of the file is changed after a
process has obtained a lock on the file.

ocfs2_lock() should skip the check for mandatory locks when unlocking a
file.

Signed-off-by: Sachin Prabhu <sprabhu@redhat.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 fs/ocfs2/locks.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/fs/ocfs2/locks.c
+++ b/fs/ocfs2/locks.c
@@ -133,7 +133,7 @@ int ocfs2_lock(struct file *file, int cm
 
 	if (!(fl->fl_flags & FL_POSIX))
 		return -ENOLCK;
-	if (__mandatory_lock(inode))
+	if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK)
 		return -ENOLCK;
 
 	return ocfs2_plock(osb->cconn, OCFS2_I(inode)->ip_blkno, file, cmd, fl);

[106/111] loop: Update mtime when writing using aops

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Nikanth Karthikesan <knikanth@suse.de>

commit 02246c41171097ceab3246f6dc251ac89de6004b upstream.

Update mtime when writing to backing filesystem using the address space
operations write_begin and write_end.

Signed-off-by: Nikanth Karthikesan <knikanth@suse.de>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/block/loop.c |    2 ++
 1 file changed, 2 insertions(+)

--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -238,6 +238,8 @@ static int do_lo_send_aops(struct loop_d
 		if (ret)
 			goto fail;
 
+		file_update_time(file);
+
 		transfer_result = lo_do_transfer(lo, WRITE, page, offset,
 				bvec->bv_page, bv_offs, size, IV);
 		copied = size;

[107/111] [SCSI] aic79xx: check for non-NULL scb in ahd_handle_nonpkt_busfree

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Hannes Reinecke <hare@suse.de>

commit 534ef056db8a8fb6b9d50188d88ed5d1fbc66673 upstream.

When removing several devices aic79xx will occasionally Oops
in ahd_handle_nonpkt_busfree during rescan. Looking at the
code I found that we're indeed not checking if the scb in
question is NULL. So check for it before accessing it.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/scsi/aic7xxx/aic79xx_core.c |   53 +++++++++++++++++++++---------------
 1 file changed, 31 insertions(+), 22 deletions(-)

--- a/drivers/scsi/aic7xxx/aic79xx_core.c
+++ b/drivers/scsi/aic7xxx/aic79xx_core.c
@@ -3171,13 +3171,16 @@ ahd_handle_nonpkt_busfree(struct ahd_sof
 				tinfo->curr.transport_version = 2;
 				tinfo->goal.transport_version = 2;
 				tinfo->goal.ppr_options = 0;
-				/*
-				 * Remove any SCBs in the waiting for selection
-				 * queue that may also be for this target so
-				 * that command ordering is preserved.
-				 */
-				ahd_freeze_devq(ahd, scb);
-				ahd_qinfifo_requeue_tail(ahd, scb);
+				if (scb != NULL) {
+					/*
+					 * Remove any SCBs in the waiting
+					 * for selection queue that may
+					 * also be for this target so that
+					 * command ordering is preserved.
+					 */
+					ahd_freeze_devq(ahd, scb);
+					ahd_qinfifo_requeue_tail(ahd, scb);
+				}
 				printerror = 0;
 			}
 		} else if (ahd_sent_msg(ahd, AHDMSG_EXT, MSG_EXT_WDTR, FALSE)
@@ -3194,13 +3197,16 @@ ahd_handle_nonpkt_busfree(struct ahd_sof
 				      MSG_EXT_WDTR_BUS_8_BIT,
 				      AHD_TRANS_CUR|AHD_TRANS_GOAL,
 				      /*paused*/TRUE);
-			/*
-			 * Remove any SCBs in the waiting for selection
-			 * queue that may also be for this target so that
-			 * command ordering is preserved.
-			 */
-			ahd_freeze_devq(ahd, scb);
-			ahd_qinfifo_requeue_tail(ahd, scb);
+			if (scb != NULL) {
+				/*
+				 * Remove any SCBs in the waiting for
+				 * selection queue that may also be for
+				 * this target so that command ordering
+				 * is preserved.
+				 */
+				ahd_freeze_devq(ahd, scb);
+				ahd_qinfifo_requeue_tail(ahd, scb);
+			}
 			printerror = 0;
 		} else if (ahd_sent_msg(ahd, AHDMSG_EXT, MSG_EXT_SDTR, FALSE)
 			&& ppr_busfree == 0) {
@@ -3217,13 +3223,16 @@ ahd_handle_nonpkt_busfree(struct ahd_sof
 					/*ppr_options*/0,
 					AHD_TRANS_CUR|AHD_TRANS_GOAL,
 					/*paused*/TRUE);
-			/*
-			 * Remove any SCBs in the waiting for selection
-			 * queue that may also be for this target so that
-			 * command ordering is preserved.
-			 */
-			ahd_freeze_devq(ahd, scb);
-			ahd_qinfifo_requeue_tail(ahd, scb);
+			if (scb != NULL) {
+				/*
+				 * Remove any SCBs in the waiting for
+				 * selection queue that may also be for
+				 * this target so that command ordering
+				 * is preserved.
+				 */
+				ahd_freeze_devq(ahd, scb);
+				ahd_qinfifo_requeue_tail(ahd, scb);
+			}
 			printerror = 0;
 		} else if ((ahd->msg_flags & MSG_FLAG_EXPECT_IDE_BUSFREE) != 0
 			&& ahd_sent_msg(ahd, AHDMSG_1B,
@@ -3251,7 +3260,7 @@ ahd_handle_nonpkt_busfree(struct ahd_sof
 	 * the message phases.  We check it last in case we
 	 * had to send some other message that caused a busfree.
 	 */
-	if (printerror != 0
+	if (scb != NULL && printerror != 0
 	 && (lastphase == P_MESGIN || lastphase == P_MESGOUT)
 	 && ((ahd->msg_flags & MSG_FLAG_EXPECT_PPR_BUSFREE) != 0)) {
 

[108/111] [SCSI] ibmvfc: Fix command completion handling

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Brian King <brking@linux.vnet.ibm.com>

commit f5832fa2f8dc39adcf3ae348d2d6383163235e79 upstream.

Commands which are completed by the VIOS are placed on a CRQ
in kernel memory for the ibmvfc driver to process. Each CRQ
entry is 16 bytes. The ibmvfc driver reads the first 8 bytes
to check if the entry is valid, then reads the next 8 bytes to get
the handle, which is a pointer the completed command. This fixes
an issue seen on Power 7 where the processor reordered the
loads from memory, resulting in processing command completion
with a stale handle. This could result in command timeouts,
and also early completion of commands.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/scsi/ibmvscsi/ibmvfc.c |    6 ++++++
 1 file changed, 6 insertions(+)

--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -2720,6 +2720,7 @@ static struct ibmvfc_async_crq *ibmvfc_n
 	if (crq->valid & 0x80) {
 		if (++async_crq->cur == async_crq->size)
 			async_crq->cur = 0;
+		rmb();
 	} else
 		crq = NULL;
 
@@ -2742,6 +2743,7 @@ static struct ibmvfc_crq *ibmvfc_next_cr
 	if (crq->valid & 0x80) {
 		if (++queue->cur == queue->size)
 			queue->cur = 0;
+		rmb();
 	} else
 		crq = NULL;
 
@@ -2790,12 +2792,14 @@ static void ibmvfc_tasklet(void *data)
 		while ((async = ibmvfc_next_async_crq(vhost)) != NULL) {
 			ibmvfc_handle_async(async, vhost);
 			async->valid = 0;
+			wmb();
 		}
 
 		/* Pull all the valid messages off the CRQ */
 		while ((crq = ibmvfc_next_crq(vhost)) != NULL) {
 			ibmvfc_handle_crq(crq, vhost);
 			crq->valid = 0;
+			wmb();
 		}
 
 		vio_enable_interrupts(vdev);
@@ -2803,10 +2807,12 @@ static void ibmvfc_tasklet(void *data)
 			vio_disable_interrupts(vdev);
 			ibmvfc_handle_async(async, vhost);
 			async->valid = 0;
+			wmb();
 		} else if ((crq = ibmvfc_next_crq(vhost)) != NULL) {
 			vio_disable_interrupts(vdev);
 			ibmvfc_handle_crq(crq, vhost);
 			crq->valid = 0;
+			wmb();
 		} else
 			done = 1;
 	}

[109/111] [SCSI] ibmvfc: Reduce error recovery timeout

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Brian King <brking@linux.vnet.ibm.com>

commit daa142d1773dd3a986f02a8a4da929608d24daaa upstream.

If a command times out resulting in EH getting invoked, we wait for the
aborted commands to come back after sending the abort. Shorten
the amount of time we wait for these responses, to ensure we don't
get stuck in EH for several minutes.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/scsi/ibmvscsi/ibmvfc.c |    2 +-
 drivers/scsi/ibmvscsi/ibmvfc.h |    1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -1969,7 +1969,7 @@ static int ibmvfc_wait_for_ops(struct ib
 	DECLARE_COMPLETION_ONSTACK(comp);
 	int wait;
 	unsigned long flags;
-	signed long timeout = init_timeout * HZ;
+	signed long timeout = IBMVFC_ABORT_WAIT_TIMEOUT * HZ;
 
 	ENTER;
 	do {
--- a/drivers/scsi/ibmvscsi/ibmvfc.h
+++ b/drivers/scsi/ibmvscsi/ibmvfc.h
@@ -38,6 +38,7 @@
 #define IBMVFC_ADISC_PLUS_CANCEL_TIMEOUT	\
 		(IBMVFC_ADISC_TIMEOUT + IBMVFC_ADISC_CANCEL_TIMEOUT)
 #define IBMVFC_INIT_TIMEOUT		120
+#define IBMVFC_ABORT_WAIT_TIMEOUT	40
 #define IBMVFC_MAX_REQUESTS_DEFAULT	100
 
 #define IBMVFC_DEBUG			0

[110/111] md/raid1: delay reads that could overtake behind-writes.

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: NeilBrown <neilb@suse.de>

commit e555190d82c0f58e825e3cbd9e6ebe2e7ac713bd upstream.

When a raid1 array is configured to support write-behind
on some devices, it normally only reads from other devices.
If all devices are write-behind (because the rest have failed)
it is possible for a read request to be serviced before a
behind-write request, which would appear as data corruption.

So when forced to read from a WriteMostly device, wait for any
write-behind to complete, and don't start any more behind-writes.

Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 drivers/md/bitmap.c |    4 +++-
 drivers/md/bitmap.h |    3 +++
 drivers/md/raid1.c  |   25 ++++++++++++++++++-------
 3 files changed, 24 insertions(+), 8 deletions(-)

--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -1317,7 +1317,8 @@ void bitmap_endwrite(struct bitmap *bitm
 {
 	if (!bitmap) return;
 	if (behind) {
-		atomic_dec(&bitmap->behind_writes);
+		if (atomic_dec_and_test(&bitmap->behind_writes))
+			wake_up(&bitmap->behind_wait);
 		PRINTK(KERN_DEBUG "dec write-behind count %d/%d\n",
 		  atomic_read(&bitmap->behind_writes), bitmap->max_write_behind);
 	}
@@ -1629,6 +1630,7 @@ int bitmap_create(mddev_t *mddev)
 	atomic_set(&bitmap->pending_writes, 0);
 	init_waitqueue_head(&bitmap->write_wait);
 	init_waitqueue_head(&bitmap->overflow_wait);
+	init_waitqueue_head(&bitmap->behind_wait);
 
 	bitmap->mddev = mddev;
 
--- a/drivers/md/bitmap.h
+++ b/drivers/md/bitmap.h
@@ -254,6 +254,9 @@ struct bitmap {
 	wait_queue_head_t write_wait;
 	wait_queue_head_t overflow_wait;
 
+#ifndef __GENKSYMS__
+	wait_queue_head_t behind_wait;
+#endif
 };
 
 /* the bitmap API */
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -845,6 +845,15 @@ static int make_request(struct request_q
 		}
 		mirror = conf->mirrors + rdisk;
 
+		if (test_bit(WriteMostly, &mirror->rdev->flags) &&
+		    bitmap) {
+			/* Reading from a write-mostly device must
+			 * take care not to over-take any writes
+			 * that are 'behind'
+			 */
+			wait_event(bitmap->behind_wait,
+				   atomic_read(&bitmap->behind_writes) == 0);
+		}
 		r1_bio->read_disk = rdisk;
 
 		read_bio = bio_clone(bio, GFP_NOIO);
@@ -922,9 +931,13 @@ static int make_request(struct request_q
 		set_bit(R1BIO_Degraded, &r1_bio->state);
 	}
 
-	/* do behind I/O ? */
+	/* do behind I/O ?
+	 * Not if there are too many, or cannot allocate memory,
+	 * or a reader on WriteMostly is waiting for behind writes
+	 * to flush */
 	if (bitmap &&
 	    atomic_read(&bitmap->behind_writes) < bitmap->max_write_behind &&
+	    !waitqueue_active(&bitmap->behind_wait) &&
 	    (behind_pages = alloc_behind_pages(bio)) != NULL)
 		set_bit(R1BIO_BehindIO, &r1_bio->state);
 
@@ -2105,15 +2118,13 @@ static int stop(mddev_t *mddev)
 {
 	conf_t *conf = mddev->private;
 	struct bitmap *bitmap = mddev->bitmap;
-	int behind_wait = 0;
 
 	/* wait for behind writes to complete */
-	while (bitmap && atomic_read(&bitmap->behind_writes) > 0) {
-		behind_wait++;
-		printk(KERN_INFO "raid1: behind writes in progress on device %s, waiting to stop (%d)\n", mdname(mddev), behind_wait);
-		set_current_state(TASK_UNINTERRUPTIBLE);
-		schedule_timeout(HZ); /* wait a second */
+	if (bitmap && atomic_read(&bitmap->behind_writes) > 0) {
+		printk(KERN_INFO "raid1: behind writes in progress on device %s, waiting to stop.\n", mdname(mddev));
 		/* need to kick something here to make sure I/O goes? */
+		wait_event(bitmap->behind_wait,
+			   atomic_read(&bitmap->behind_writes) == 0);
 	}
 
 	raise_barrier(conf);

[111/111] mm: fix corruption of hibernation caused by reusing swap during image saving

[Greg KH]

2.6.32-stable review patch.  If anyone has any objections, please let us know.

------------------

From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>

commit 966cca029f739716fbcc8068b8c6dfe381f86fc3 upstream.

Since 2.6.31, swap_map[]'s refcounting was changed to show that a used
swap entry is just for swap-cache, can be reused.  Then, while scanning
free entry in swap_map[], a swap entry may be able to be reclaimed and
reused.  It was caused by commit c9e444103b5e7a5 ("mm: reuse unused swap
entry if necessary").

But this caused deta corruption at resume. The scenario is

- Assume a clean-swap cache, but mapped.

- at hibernation_snapshot[], clean-swap-cache is saved as
  clean-swap-cache and swap_map[] is marked as SWAP_HAS_CACHE.

- then, save_image() is called.  And reuse SWAP_HAS_CACHE entry to save
  image, and break the contents.

After resume:

- the memory reclaim runs and finds clean-not-referenced-swap-cache and
  discards it because it's marked as clean.  But here, the contents on
  disk and swap-cache is inconsistent.

Hance memory is corrupted.

This patch avoids the bug by not reclaiming swap-entry during hibernation.
This is a quick fix for backporting.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Rafael J. Wysocki <rjw@sisk.pl>
Reported-by: Ondreg Zary <linux@rainbow-software.org>
Tested-by: Ondreg Zary <linux@rainbow-software.org>
Tested-by: Andrea Gelmini <andrea.gelmini@gmail.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 mm/swapfile.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -330,8 +330,10 @@ checks:
 	if (offset > si->highest_bit)
 		scan_base = offset = si->lowest_bit;
 
-	/* reuse swap entry of cache-only swap if not busy. */
-	if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) {
+	/* reuse swap entry of cache-only swap if not hibernation. */
+	if (vm_swap_full()
+		&& cache == SWAP_CACHE
+		&& si->swap_map[offset] == SWAP_HAS_CACHE) {
 		int swap_was_freed;
 		spin_unlock(&swap_lock);
 		swap_was_freed = __try_to_reclaim_swap(si, offset);

[000/111] 2.6.32.19 -stable review

[Greg KH]

This is the start of the stable review cycle for the 2.6.32.19 release.
There are 111 patches in this series, all will be posted as a response
to this one.  If anyone has any issues with these being applied, please
let us know.  If anyone is a maintainer of the proper subsystem, and
wants to add a Signed-off-by: line to the patch, please respond with it.

Responses should be made by Friday, August 13, 2010, 20:00:00 UTC.
Anything received after that time might be too late.

The whole patch series can be found in one patch at:
	kernel.org/pub/linux/kernel/v2.6/stable-review/patch-2.6.32.19-rc1.gz
and the diffstat can be found below.

thanks,

greg k-h


 Makefile                                    |    2 +-
 arch/arm/Kconfig                            |   12 +
 arch/arm/include/asm/tlbflush.h             |    8 +
 arch/arm/plat-mxc/include/mach/gpio.h       |    1 +
 arch/ia64/hp/common/sba_iommu.c             |   38 ++-
 arch/ia64/kernel/time.c                     |    4 +-
 arch/powerpc/Makefile                       |   16 +-
 arch/powerpc/include/asm/ppc-pci.h          |    5 +
 arch/powerpc/kernel/time.c                  |    5 +-
 arch/powerpc/platforms/pseries/eeh.c        |    4 +-
 arch/powerpc/platforms/pseries/eeh_driver.c |    4 +-
 arch/powerpc/platforms/pseries/eeh_event.c  |    2 +-
 arch/s390/include/asm/cputime.h             |    8 +
 arch/s390/kernel/s390_ext.c                 |    2 +
 arch/s390/kernel/time.c                     |    3 +-
 arch/s390/kernel/vtime.c                    |    2 +
 arch/x86/include/asm/cmpxchg_32.h           |   86 ++--
 arch/x86/include/asm/cmpxchg_64.h           |   88 ++--
 arch/x86/kernel/apic/io_apic.c              |   28 +-
 arch/x86/kernel/cpu/vmware.c                |   10 +-
 arch/x86/kernel/vsyscall_64.c               |    5 +-
 arch/x86/mm/init_64.c                       |   19 +
 crypto/testmgr.c                            |   84 +++
 drivers/acpi/processor_perflib.c            |    6 +-
 drivers/ata/ata_piix.c                      |    8 +
 drivers/atm/solos-pci.c                     |    7 +-
 drivers/block/loop.c                        |    2 +
 drivers/bluetooth/btusb.c                   |    3 +
 drivers/char/nvram.c                        |   14 +-
 drivers/gpu/drm/drm_edid.c                  |    4 +-
 drivers/gpu/drm/i915/dvo_tfp410.c           |    2 +-
 drivers/ide/ide-cd.c                        |   14 +-
 drivers/md/bitmap.c                         |    4 +-
 drivers/md/bitmap.h                         |    3 +
 drivers/md/raid1.c                          |   25 +-
 drivers/md/raid10.c                         |   18 +
 drivers/net/smsc911x.c                      |   92 ++--
 drivers/net/xen-netfront.c                  |    1 +
 drivers/pci/quirks.c                        |    4 +
 drivers/s390/cio/cio.c                      |    1 +
 drivers/scsi/aic7xxx/aic79xx_core.c         |   53 ++-
 drivers/scsi/ibmvscsi/ibmvfc.c              |    8 +-
 drivers/scsi/ibmvscsi/ibmvfc.h              |    1 +
 drivers/staging/line6/Kconfig               |    1 +
 drivers/staging/panel/panel.c               |    2 +
 drivers/usb/core/hub.c                      |    6 +-
 drivers/usb/core/quirks.c                   |    3 +
 drivers/usb/core/urb.c                      |   50 +--
 drivers/usb/misc/usbtest.c                  |    6 +-
 drivers/usb/mon/mon_bin.c                   |    2 +-
 drivers/usb/serial/cp210x.c                 |    4 +
 drivers/usb/serial/ftdi_sio.c               |    4 +
 drivers/usb/serial/ftdi_sio_ids.h           |   10 +
 drivers/usb/serial/option.c                 |   13 +
 drivers/video/w100fb.c                      |    4 +-
 drivers/xen/events.c                        |    1 +
 fs/block_dev.c                              |   10 +-
 fs/btrfs/acl.c                              |   36 +-
 fs/btrfs/btrfs_inode.h                      |    5 +-
 fs/btrfs/ctree.c                            |  229 ++++++---
 fs/btrfs/ctree.h                            |   40 +-
 fs/btrfs/dir-item.c                         |   19 +-
 fs/btrfs/disk-io.c                          |   40 ++-
 fs/btrfs/extent-tree.c                      |  112 +++--
 fs/btrfs/extent_io.c                        |    3 +-
 fs/btrfs/file.c                             |  729 ++++++++++++---------------
 fs/btrfs/inode.c                            |  624 ++++++++++++++----------
 fs/btrfs/ioctl.c                            |   34 +-
 fs/btrfs/ordered-data.c                     |  117 ++++-
 fs/btrfs/ordered-data.h                     |    5 +-
 fs/btrfs/relocation.c                       |   43 +-
 fs/btrfs/super.c                            |   20 +-
 fs/btrfs/transaction.c                      |   44 +-
 fs/btrfs/transaction.h                      |    6 +-
 fs/btrfs/tree-log.c                         |   86 ++--
 fs/btrfs/volumes.c                          |   19 +-
 fs/btrfs/xattr.c                            |   80 ++-
 fs/btrfs/xattr.h                            |    9 +-
 fs/dlm/ast.c                                |   74 +++-
 fs/dlm/ast.h                                |    4 +-
 fs/dlm/config.c                             |   24 +-
 fs/dlm/debug_fs.c                           |    2 +-
 fs/dlm/dir.c                                |    7 +-
 fs/dlm/dlm_internal.h                       |   11 +-
 fs/dlm/lock.c                               |  120 ++++--
 fs/dlm/lockspace.c                          |   15 +-
 fs/dlm/lowcomms.c                           |    6 +-
 fs/dlm/member.c                             |    8 +-
 fs/dlm/memory.c                             |    6 +-
 fs/dlm/netlink.c                            |    2 +-
 fs/dlm/plock.c                              |    6 +-
 fs/dlm/rcom.c                               |    2 +-
 fs/dlm/requestqueue.c                       |    2 +-
 fs/dlm/user.c                               |   22 +-
 fs/dlm/user.h                               |    4 +-
 fs/ecryptfs/file.c                          |   60 ++-
 fs/ecryptfs/inode.c                         |    5 +-
 fs/ext4/move_extent.c                       |    3 +
 fs/ext4/super.c                             |   27 +-
 fs/jfs/xattr.c                              |   87 ++--
 fs/ocfs2/alloc.c                            |   10 +-
 fs/ocfs2/alloc.h                            |    5 +
 fs/ocfs2/locks.c                            |    2 +-
 fs/ocfs2/refcounttree.c                     |  117 ++++-
 fs/ocfs2/super.c                            |    4 +
 fs/proc/array.c                             |    5 +-
 fs/reiserfs/journal.c                       |   15 +-
 fs/reiserfs/xattr_security.c                |    2 +-
 fs/signalfd.c                               |    2 +
 fs/splice.c                                 |   12 +-
 include/linux/clocksource.h                 |    6 +-
 include/linux/fs.h                          |    6 +-
 include/linux/interrupt.h                   |    7 +-
 include/linux/netdevice.h                   |    2 +
 include/linux/notifier.h                    |    1 +
 include/linux/reiserfs_xattr.h              |    5 +
 include/linux/sched.h                       |    4 +
 include/linux/tick.h                        |    3 +
 include/linux/usb/quirks.h                  |    4 +
 kernel/exit.c                               |   20 +-
 kernel/fork.c                               |    3 +
 kernel/irq/manage.c                         |    2 +-
 kernel/sched.c                              |   71 +++-
 kernel/sys.c                                |   18 +-
 kernel/time/tick-sched.c                    |   75 ++--
 kernel/time/timekeeping.c                   |    6 +-
 mm/memory-failure.c                         |   21 +-
 mm/swapfile.c                               |    6 +-
 net/ipv4/devinet.c                          |    1 +
 net/sched/sch_generic.c                     |   18 +
 130 files changed, 2521 insertions(+), 1551 deletions(-)