Re: [v2 PATCH 2/2] mm: swap: use smp_mb__after_atomic() to order LRU bit set

[Shakeel Butt]

On Mon, Mar 16, 2020 at 3:24 PM Yang Shi <yang.shi@xxxxxxxxxxxxxxxxx> wrote:
>
> Memory barrier is needed after setting LRU bit, but smp_mb() is too
> strong.  Some architectures, i.e. x86, imply memory barrier with atomic
> operations, so replacing it with smp_mb__after_atomic() sounds better,
> which is nop on strong ordered machines, and full memory barriers on
> others.  With this change the vm-calability cases would perform better
> on x86, I saw total 6% improvement with this patch and previous inline
> fix.
>
> The test data (lru-file-readtwice throughput) against v5.6-rc4:
>         mainline        w/ inline fix   w/ both (adding this)
>         150MB           154MB           159MB
>
> Fixes: 9c4e6b1a7027 ("mm, mlock, vmscan: no more skipping pagevecs")
> Cc: Shakeel Butt <shakeelb@xxxxxxxxxx>
> Acked-by: Vlastimil Babka <vbabka@xxxxxxx>
> Signed-off-by: Yang Shi <yang.shi@xxxxxxxxxxxxxxxxx>

Reviewed-and-Tested-by: Shakeel Butt <shakeelb@xxxxxxxxxx>