Re: [PATCH] memcg: optimize memory.numa_stat like memory.stat

[Johannes Weiner]

On Thu, Apr 23, 2020 at 04:10:09PM -0700, Andrew Morton wrote:
> From: Shakeel Butt <shakeelb@xxxxxxxxxx>
> Subject: mm/memcg: optimize memory.numa_stat like memory.stat
> 
> Currently reading memory.numa_stat traverses the underlying memcg tree
> multiple times to accumulate the stats to present the hierarchical view of
> the memcg tree.  However the kernel already maintains the hierarchical
> view of the stats and use it in memory.stat.  Just use the same mechanism
> in memory.numa_stat as well.
> 
> I ran a simple benchmark which reads root_mem_cgroup's memory.numa_stat
> file in the presense of 10000 memcgs.  The results are:
> 
> Without the patch:
> $ time cat /dev/cgroup/memory/memory.numa_stat > /dev/null
> 
> real    0m0.700s
> user    0m0.001s
> sys     0m0.697s
> 
> With the patch:
> $ time cat /dev/cgroup/memory/memory.numa_stat > /dev/null
> 
> real    0m0.001s
> user    0m0.001s
> sys     0m0.000s
> 
> [akpm@xxxxxxxxxxxxxxxxxxxx: avoid forcing out-of-line code generation]
> Link: http://lkml.kernel.org/r/20200304022058.248270-1-shakeelb@xxxxxxxxxx
> Signed-off-by: Shakeel Butt <shakeelb@xxxxxxxxxx>
> Reviewed-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
> Cc: Roman Gushchin <guro@xxxxxx>
> Cc: Michal Hocko <mhocko@xxxxxxxxxx>
> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>

Ouch, yes. That makes sense.

Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>