Re: [PATCH tip/locking/core v8 1/5] locking/qspinlock: Use _acquire/_release versions of cmpxchg & xchg

[Boqun Feng]

On Mon, Oct 19, 2015 at 08:46:02PM -0700, Davidlohr Bueso wrote:
> On Tue, 20 Oct 2015, Boqun Feng wrote:
> 
> >>@@ -93,7 +94,7 @@ static __always_inline void queued_spin_unlock(struct qspinlock *lock)
> >> 	/*
> >> 	 * smp_mb__before_atomic() in order to guarantee release semantics
> >> 	 */
> >>-	smp_mb__before_atomic_dec();
> >>+	smp_mb__before_atomic();
> >> 	atomic_sub(_Q_LOCKED_VAL, &lock->val);
> >
> >Just be curious, you don't use atomic_sub_release() here on purpose?
> 
> atomic_sub() does not imply barriers, so there's no relaxed variants; that's
> only for _return() (and such) to the caller.
> 

Ah.. my mistake ;-(

Thank you.

Regards,
Boqun

> Thanks,
> Davidlohr

Attachment: signature.asc
Description: PGP signature