Re: [RESEND PATCH v4 00/10] Rework REFCOUNT_FULL using atomic_fetch_* operations

[Hanjun Guo]

On 2019/11/21 19:58, Will Deacon wrote:
> Hi everybody,
> 
> This is a resend of version four of the patches I posted here:
> 
>   v4: https://lore.kernel.org/lkml/20191030143035.19440-1-will@xxxxxxxxxx
> 
> Previous versions can be found at:
> 
>   v1: https://lkml.kernel.org/r/20190802101000.12958-1-will@xxxxxxxxxx
>   v2: https://lkml.kernel.org/r/20190827163204.29903-1-will@xxxxxxxxxx
>   v3: https://lkml.kernel.org/r/20191007154703.5574-1-will@xxxxxxxxxx
> 
> I didn't receive any feedback last time around, other than some positive
> noises from Kees, so please consider this for inclusion in mainline.

This time I tested this patch set on both ARM64 and x86 server, against
5.4-rc8.

On ARM64, similar test case and the similar performance improvement.

On X86, I tested on a 24 cores Xeon(R) CPU E5-2620, with kernel compile
(make -j) for test case, and no issue was found.

Tested-by: Hanjun Guo <guohanjun@xxxxxxxxxx>

Thanks
Hanjun