ARM: spinlock: avoid exclusive accesses on unlock() path

PATCH New Will Deacon 1 participant 1 message Jan 11, 2013

Subsystems (2)

Arm (32-bit) Locking primitives

lore

Share & tools

lore

Will Deacon

author

2013-01-11 15:07 UTC #0 lore raw

PATCH 1/1 ARM: spinlock: avoid exclusive accesses on unlock() path

When unlocking a spinlock, all we need to do is increment the owner field of the lock. Since only one CPU can be performing an unlock() operation for a given lock, this doesn't need to be exclusive.

This patch simplifies arch_spin_unlock to use non-exclusive accesses when updating the owner field of the lock.

Signed-off-by Will Deacon <[email protected]>

---
 arch/arm/include/asm/spinlock.h | 16 +---------------
 1 file changed, 1 insertion(+), 15 deletions(-)

arch/arm/include/asm/spinlock.h +1 ~~-15~~

index b4ca707..6220e9f 100644
@@ -119,22 +119,8 @@ static inline int arch_spin_trylock(arch_spinlock_t *lock)
 
 static inline void arch_spin_unlock(arch_spinlock_t *lock)
 {
-	unsigned long tmp;
-	u32 slock;
-
 	smp_mb();
-
-	__asm__ __volatile__(
-"	mov	%1, #1\n"
-"1:	ldrex	%0, [%2]\n"
-"	uadd16	%0, %0, %1\n"
-"	strex	%1, %0, [%2]\n"
-"	teq	%1, #0\n"
-"	bne	1b"
-	: "=&r" (slock), "=&r" (tmp)
-	: "r" (&lock->slock)
-	: "cc");
-
+	lock->tickets.owner++;
 	dsb_sev();
 }

Signature · 1 line

1.8.0