Re: [memcg] a8c49af3be: hackbench.throughput -13.7% regression

[Shakeel Butt]

On Tue, Apr 26, 2022 at 9:34 PM Yosry Ahmed <yosryahmed@xxxxxxxxxx> wrote:
>
[...]
>
> The only seemingly expensive operation in the mod_memcg_state() path
> is the call to cgroup_rstat_updated() (through memcg_rstat_updated()).
> One idea off the top of my head is to batch calls to
> cgroup_rstat_updated(), similar to what 11192d9c124d ("memcg: flush
> stats only if updated") did on the flush side. I am interested to see
> what memcg maintainers think about this problem (and the proposed
> solution).
>

I am suspecting this is some code alignment changes which is causing
this difference. Let's first confirm the issue before thinking about
the optimizations.