diff options
| author | Will Deacon <will.deacon@arm.com> | 2015-08-06 18:54:41 +0200 |
|---|---|---|
| committer | Ingo Molnar <mingo@kernel.org> | 2015-08-12 11:59:05 +0200 |
| commit | 2b2a85a4d3534b8884fcfa5bb52837f0e1c672bc (patch) | |
| tree | e81e07f44f5018c49ffda73aefd2e3cc9bb19291 /include/asm-generic/qrwlock.h | |
| parent | locking/lockref: Remove homebrew cmpxchg64_relaxed() macro definition (diff) | |
| download | linux-2b2a85a4d3534b8884fcfa5bb52837f0e1c672bc.tar.xz linux-2b2a85a4d3534b8884fcfa5bb52837f0e1c672bc.zip | |
locking/qrwlock: Implement queue_write_unlock() using smp_store_release()
Since the following commit:
536fa402221f ("compiler: Allow 1- and 2-byte smp_load_acquire() and smp_store_release()")
smp_store_release() supports byte accesses, so use that in writer unlock
and remove the conditional macro override.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Waiman Long <Waiman.Long@hp.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: paulmck@linux.vnet.ibm.com
Link: http://lkml.kernel.org/r/1438880084-18856-6-git-send-email-will.deacon@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Diffstat (limited to 'include/asm-generic/qrwlock.h')
| -rw-r--r-- | include/asm-generic/qrwlock.h | 9 |
1 files changed, 1 insertions, 8 deletions
diff --git a/include/asm-generic/qrwlock.h b/include/asm-generic/qrwlock.h index deb9e8b0eb9e..eb673dde8879 100644 --- a/include/asm-generic/qrwlock.h +++ b/include/asm-generic/qrwlock.h @@ -134,21 +134,14 @@ static inline void queued_read_unlock(struct qrwlock *lock) atomic_sub(_QR_BIAS, &lock->cnts); } -#ifndef queued_write_unlock /** * queued_write_unlock - release write lock of a queue rwlock * @lock : Pointer to queue rwlock structure */ static inline void queued_write_unlock(struct qrwlock *lock) { - /* - * If the writer field is atomic, it can be cleared directly. - * Otherwise, an atomic subtraction will be used to clear it. - */ - smp_mb__before_atomic(); - atomic_sub(_QW_LOCKED, &lock->cnts); + smp_store_release((u8 *)&lock->cnts, 0); } -#endif /* * Remapping rwlock architecture specific functions to the corresponding |