[RFC PATCH] ARM: cmpxchg: define __HAVE_ARCH_CMPXCHG for armv6 and later

From: Yong Zhang <redacted>

Both pi_stress and sigwaittest in rt-test show performance gain with
__HAVE_ARCH_CMPXCHG. Testing result on coretile_express_a9x4:

pi_stress -p 99 --duration=300 (on linux-3.4-rc5; bigger is better)
  vanilla:     Total inversion performed: 5493381
  patched:     Total inversion performed: 5621746

sigwaittest -p 99 -l 100000 (on linux-3.4-rc5-rt6; less is better)
  3.4-rc5-rt6: Min   24, Cur   27, Avg   30, Max   98
  patched:     Min   19, Cur   21, Avg   23, Max   96

Signed-off-by: Yong Zhang <redacted>
Cc: Russell King <redacted>
Cc: Nicolas Pitre <redacted>
Cc: Will Deacon <redacted>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Thomas Gleixner <redacted>
Cc: linux-arm-kernel at lists.infradead.org
---
 arch/arm/include/asm/cmpxchg.h |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/arch/arm/include/asm/cmpxchg.h b/arch/arm/include/asm/cmpxchg.h
index 7eb18c1..a91b44e 100644
--- a/arch/arm/include/asm/cmpxchg.h
+++ b/arch/arm/include/asm/cmpxchg.h

@@ -127,6 +127,8 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size
 
 #else	/* min ARCH >= ARMv6 */
 
+#define __HAVE_ARCH_CMPXCHG 1
+
 extern void __bad_cmpxchg(volatile void *ptr, int size);
 
 /*

-- 
1.7.5.4

Hello,

On Wed, May 30, 2012 at 08:51:26AM +0100, Yong Zhang wrote:

As long as this only requires 4-byte cmpxchg support then I think this is
fine. It looks like that's currently the case, so:

Acked-by: Will Deacon <redacted>

Will

On Thu, May 31, 2012 at 02:24:50AM +0100, Will Deacon wrote:

Hello,

On Wed, May 30, 2012 at 08:51:26AM +0100, Yong Zhang wrote:

quoted

diff --git a/arch/arm/include/asm/cmpxchg.h b/arch/arm/include/asm/cmpxchg.h
index 7eb18c1..a91b44e 100644
--- a/arch/arm/include/asm/cmpxchg.h
+++ b/arch/arm/include/asm/cmpxchg.h

@@ -127,6 +127,8 @@ static inline unsigned long __xchg(unsigned long x, volatile void *ptr, int size
 
 #else	/* min ARCH >= ARMv6 */
 
+#define __HAVE_ARCH_CMPXCHG 1
+
 extern void __bad_cmpxchg(volatile void *ptr, int size);

As long as this only requires 4-byte cmpxchg support then I think this is
fine. It looks like that's currently the case, so:

Yes and thanks for review.

Yong

Acked-by: Will Deacon <redacted>

Will