[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220815132330.GA24237@blackbody.suse.cz>
Date: Mon, 15 Aug 2022 15:23:30 +0200
From: Michal Koutný <mkoutny@...e.com>
To: Chengming Zhou <zhouchengming@...edance.com>
Cc: hannes@...xchg.org, tj@...nel.org, corbet@....net,
surenb@...gle.com, mingo@...hat.com, peterz@...radead.org,
vincent.guittot@...aro.org, dietmar.eggemann@....com,
rostedt@...dmis.org, bsegall@...gle.com, cgroups@...r.kernel.org,
linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
songmuchun@...edance.com
Subject: Re: [PATCH v2 09/10] sched/psi: per-cgroup PSI stats
disable/re-enable interface
On Fri, Aug 12, 2022 at 08:36:17PM +0800, Chengming Zhou <zhouchengming@...edance.com> wrote:
> Could you help take a look if there is anything to improve?
Thanks, just a little nit.
> + The reason this control attribute exists is that PSI accounts stalls for
> + each cgroup separately and aggregates it at each level of the hierarchy.
> + This may cause non-negligible overhead for some workloads when under
> + deep level of the hierarchy, in which case this control attribute can
> + be used to disable PSI accounting in the cgroups.
s/in the cgroups/in the non-leaf cgroups/
or
s/in the cgroups/in the uninteresting cgroups/
(I'm concerned that it may result in lots of disabling if you want the
performance. I'll expand on it in 2nd subthread.)
Michal
Powered by blists - more mailing lists