From: Johannes Weiner <hannes@cmpxchg.org>
To: Chengming Zhou <zhouchengming@bytedance.com>
Cc: Tejun Heo <tj@kernel.org>,
surenb@google.com, mingo@redhat.com, peterz@infradead.org,
corbet@lwn.net, akpm@linux-foundation.org, rdunlap@infradead.org,
linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
songmuchun@bytedance.com, cgroups@vger.kernel.org
Subject: Re: [PATCH 8/9] sched/psi: add kernel cmdline parameter psi_inner_cgroup
Date: Thu, 4 Aug 2022 12:56:36 -0400 [thread overview]
Message-ID: <Yuv6RHY0GRGBw+as@cmpxchg.org> (raw)
In-Reply-To: <f8444db4-3235-d108-698a-6772e03a6b67@bytedance.com>
On Thu, Aug 04, 2022 at 09:51:31PM +0800, Chengming Zhou wrote:
> On 2022/8/4 03:22, Johannes Weiner wrote:
> > On Wed, Aug 03, 2022 at 07:58:27AM -1000, Tejun Heo wrote:
> >> Hello,
> >>
> >> On Wed, Aug 03, 2022 at 08:17:22PM +0800, Chengming Zhou wrote:
> >>>> Assuming the above isn't wrong, if we can figure out how we can re-enable
> >>>> it, which is more difficult as the counters need to be resynchronized with
> >>>> the current state, that'd be ideal. Then, we can just allow each cgroup to
> >>>> enable / disable PSI reporting dynamically as they see fit.
> >>>
> >>> This method is more fine-grained but more difficult like you said above.
> >>> I think it may meet most needs to disable PSI stats in intermediate cgroups?
> >>
> >> So, I'm not necessarily against implementing something easier but we at
> >> least wanna get the interface right, so that if we decide to do the full
> >> thing later we can easily expand on the existing interface. ie. let's please
> >> not be too hacky. I don't think it'd be that difficult to implement
> >> per-cgroup disable-only operation that we can later expand to allow
> >> re-enabling, right?
> >
> > It should be relatively straight-forward to disable and re-enable
> > state aggregation, time tracking, averaging on a per-cgroup level, if
> > we can live with losing history from while it was disabled. I.e. the
> > avgs will restart from 0, total= will have gaps - should be okay, IMO.
> >
> > Where it gets trickier is also stopping the tracking of task counts in
> > a cgroup. For re-enabling afterwards, we'd have to freeze scheduler
> > and cgroup state and find all tasks of interest across all CPUs for
> > the given cgroup to recreate the counts. I'm not quite sure whether
> > that's feasible, and if so, whether it's worth the savings.
> >
> > It might be good to benchmark the two disabling steps independently.
> > Maybe stopping aggregation while keeping task counts is good enough,
> > and we can commit to a disable/re-enable interface from the start.
> >
> > Or maybe it's all in the cachelines and iteration, and stopping the
> > aggregation while still writing task counts isn't saving much. In that
> > case we'd have to look closer at reconstructing task counts, to see if
> > later re-enabling is actually a practical option or whether a one-off
> > kill switch is more realistic.
> >
> > Chengming, can you experiment with disabling: record_times(), the
> > test_state() loop and state_mask construction, and the averaging
> > worker - while keeping the groupc->tasks updates?
>
> Hello,
>
> I did this experiment today with disabling record_times(), test_state()
> loop and averaging worker, while only keeping groupc->tasks[] updates,
> the results look promising.
>
> mmtests/config-scheduler-perfpipe on Intel Xeon Platinum with 3 levels of cgroup:
>
> perfpipe
> tip tip patched
> psi=off psi=on only groupc->tasks[]
> Min Time 7.99 ( 0.00%) 8.86 ( -10.95%) 8.31 ( -4.08%)
> 1st-qrtle Time 8.11 ( 0.00%) 8.94 ( -10.22%) 8.39 ( -3.46%)
> 2nd-qrtle Time 8.17 ( 0.00%) 9.02 ( -10.42%) 8.44 ( -3.37%)
> 3rd-qrtle Time 8.20 ( 0.00%) 9.08 ( -10.72%) 8.48 ( -3.43%)
> Max-1 Time 7.99 ( 0.00%) 8.86 ( -10.95%) 8.31 ( -4.08%)
> Max-5 Time 7.99 ( 0.00%) 8.86 ( -10.95%) 8.31 ( -4.08%)
> Max-10 Time 8.09 ( 0.00%) 8.89 ( -9.96%) 8.35 ( -3.22%)
> Max-90 Time 8.31 ( 0.00%) 9.13 ( -9.90%) 8.55 ( -2.95%)
> Max-95 Time 8.32 ( 0.00%) 9.14 ( -9.88%) 8.55 ( -2.81%)
> Max-99 Time 8.39 ( 0.00%) 9.26 ( -10.30%) 8.57 ( -2.09%)
> Max Time 8.56 ( 0.00%) 9.26 ( -8.23%) 8.72 ( -1.90%)
> Amean Time 8.19 ( 0.00%) 9.03 * -10.26%* 8.45 * -3.27%*
Fantastic!
> Tejun suggested using a bitmap in task to remember whether the task is accounted
> at a given level or not, which I think also is a very good idea, but I haven't
> clearly figure out how to do it.
>
> The above performance test result looks good to me, so I think we can implement this
> per-cgroup "cgroup.psi" interface to disable/re-enable PSI stats from the start,
> and we can change to a better implementation if needed later?
Yes, that sounds good to me.
next prev parent reply other threads:[~2022-08-04 16:56 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-21 4:04 [PATCH 0/9] sched/psi: some optimization and extension Chengming Zhou
2022-07-21 4:04 ` [PATCH 1/9] sched/psi: fix periodic aggregation shut off Chengming Zhou
2022-07-25 15:34 ` Johannes Weiner
2022-07-25 15:39 ` Johannes Weiner
2022-07-26 13:28 ` Chengming Zhou
2022-07-21 4:04 ` [PATCH 2/9] sched/psi: optimize task switch inside shared cgroups again Chengming Zhou
2022-07-21 4:04 ` [PATCH 3/9] sched/psi: move private helpers to sched/stats.h Chengming Zhou
2022-07-25 16:39 ` Johannes Weiner
2022-07-21 4:04 ` [PATCH 4/9] sched/psi: don't change task psi_flags when migrate CPU/group Chengming Zhou
2022-07-21 4:04 ` [PATCH 5/9] sched/psi: don't create cgroup PSI files when psi_disabled Chengming Zhou
2022-07-25 16:41 ` Johannes Weiner
2022-07-21 4:04 ` [PATCH 6/9] sched/psi: save percpu memory when !psi_cgroups_enabled Chengming Zhou
2022-07-25 16:47 ` Johannes Weiner
2022-07-21 4:04 ` [PATCH 7/9] sched/psi: cache parent psi_group to speed up groups iterate Chengming Zhou
2022-07-21 4:04 ` [PATCH 8/9] sched/psi: add kernel cmdline parameter psi_inner_cgroup Chengming Zhou
2022-07-25 16:52 ` Johannes Weiner
2022-07-26 13:38 ` [External] " Chengming Zhou
2022-07-26 17:54 ` Tejun Heo
2022-08-03 12:17 ` Chengming Zhou
2022-08-03 17:58 ` Tejun Heo
2022-08-03 19:22 ` Johannes Weiner
2022-08-03 19:48 ` Tejun Heo
2022-08-04 13:51 ` Chengming Zhou
2022-08-04 16:56 ` Johannes Weiner [this message]
2022-08-04 2:02 ` Chengming Zhou
2022-07-21 4:04 ` [PATCH 9/9] sched/psi: add PSI_IRQ to track IRQ/SOFTIRQ pressure Chengming Zhou
2022-07-21 10:00 ` kernel test robot
2022-07-21 22:10 ` kernel test robot
2022-07-22 3:30 ` Abel Wu
2022-07-22 6:13 ` Chengming Zhou
2022-07-22 7:14 ` Abel Wu
2022-07-22 7:33 ` Chengming Zhou
2022-07-25 18:26 ` Johannes Weiner
2022-07-26 13:55 ` [External] " Chengming Zhou
2022-07-27 11:28 ` Chengming Zhou
2022-07-27 13:00 ` Johannes Weiner
2022-07-27 15:09 ` Chengming Zhou
2022-07-27 16:07 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Yuv6RHY0GRGBw+as@cmpxchg.org \
--to=hannes@cmpxchg.org \
--cc=akpm@linux-foundation.org \
--cc=cgroups@vger.kernel.org \
--cc=corbet@lwn.net \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=rdunlap@infradead.org \
--cc=songmuchun@bytedance.com \
--cc=surenb@google.com \
--cc=tj@kernel.org \
--cc=zhouchengming@bytedance.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).