Re: [PATCH] s390: more efficient smp barriers

From: Martin Schwidefsky
Date: Mon Jan 11 2016 - 03:07:49 EST


On Sun, 10 Jan 2016 13:53:26 +0200
"Michael S. Tsirkin" <mst@xxxxxxxxxx> wrote:

> As per: lkml.kernel.org/r/20150921112252.3c2937e1@mschwide
> atomics imply a barrier on s390, so s390 should change
> smp_mb__before_atomic and smp_mb__after_atomic to barrier() instead of
> smp_mb() and hence should not use the generic versions.
>
> Suggested-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Suggested-by: Martin Schwidefsky <schwidefsky@xxxxxxxxxx>
> Signed-off-by: Michael S. Tsirkin <mst@xxxxxxxxxx>
> ---
>
> This is on top of arch barrier cleanup patchset,
> will be included in v3 of that patchset.
>
> arch/s390/include/asm/barrier.h | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/arch/s390/include/asm/barrier.h b/arch/s390/include/asm/barrier.h
> index 4d26fa4..5c8db3c 100644
> --- a/arch/s390/include/asm/barrier.h
> +++ b/arch/s390/include/asm/barrier.h
> @@ -45,6 +45,9 @@ do { \
> ___p1; \
> })
>
> +#define __smp_mb__before_atomic() barrier()
> +#define __smp_mb__after_atomic() barrier()
> +
> #include <asm-generic/barrier.h>
>
> #endif /* __ASM_BARRIER_H */

Feel free to add:
Acked-by: Martin Schwidefsky <schwidefsky@xxxxxxxxxx>

--
blue skies,
Martin.

"Reality continues to ruin my life." - Calvin.