[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZampZPLsXsdybtY9@P9FQF9L96D.corp.robot.car>
Date: Thu, 18 Jan 2024 14:42:44 -0800
From: Roman Gushchin <roman.gushchin@...ux.dev>
To: Shakeel Butt <shakeelb@...gle.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>, Jens Axboe <axboe@...nel.dk>,
Johannes Weiner <hannes@...xchg.org>, Tejun Heo <tj@...nel.org>,
Jan Kara <jack@...e.cz>, Michal Hocko <mhocko@...nel.org>,
Muchun Song <muchun.song@...ux.dev>, cgroups@...r.kernel.org,
linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mm: writeback: ratelimit stat flush from
mem_cgroup_wb_stats
On Thu, Jan 18, 2024 at 06:42:35PM +0000, Shakeel Butt wrote:
> One of our workloads (Postgres 14) has regressed when migrated from 5.10
> to 6.1 upstream kernel. The regression can be reproduced by sysbench's
> oltp_write_only benchmark. It seems like the always on rstat flush in
> mem_cgroup_wb_stats() is causing the regression. So, rate limit that
> specific rstat flush. One potential consequence would be the dirty
> throttling might be decided on stale memcg stats. However from our
> benchmarks and production traffic we have not observed any change in the
> dirty throttling behavior of the application.
>
> Signed-off-by: Shakeel Butt <shakeelb@...gle.com>
Acked-by: Roman Gushchin <roman.gushchin@...ux.dev>
Thanks!
Powered by blists - more mailing lists