From: Vincent Guittot <vincent.guittot@linaro.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>,
Huangzhaoyang <huangzhaoyang@gmail.com>,
Andrew Morton <akpm@linux-foundation.org>,
Michal Hocko <mhocko@kernel.org>,
Vladimir Davydov <vdavydov.dev@gmail.com>,
Zhaoyang Huang <zhaoyang.huang@unisoc.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [Resend PATCH] psi : calc cfs task memstall time more precisely
Date: Wed, 10 Nov 2021 09:49:44 +0100 [thread overview]
Message-ID: <CAKfTPtAVQ+mtYkGv5xJnbjBO9L9z7jSKOvzhObd0MvVpxakezw@mail.gmail.com> (raw)
In-Reply-To: <YYqMJLXcQ4a+Lh/4@hirez.programming.kicks-ass.net>
On Tue, 9 Nov 2021 at 15:56, Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Tue, Nov 02, 2021 at 03:47:33PM -0400, Johannes Weiner wrote:
> > CC peterz as well for rt and timekeeping magic
> >
> > On Fri, Oct 15, 2021 at 02:16:52PM +0800, Huangzhaoyang wrote:
> > > From: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
> > >
> > > In an EAS enabled system, there are two scenarios discordant to current design,
> > >
> > > 1. workload used to be heavy uneven among cores for sake of scheduler policy.
> > > RT task usually preempts CFS task in little core.
> > > 2. CFS task's memstall time is counted as simple as exit - entry so far, which
> > > ignore the preempted time by RT, DL and Irqs.
>
> It ignores preemption full-stop. I don't see why RT/IRQ should be
> special cased here.
>
> > > With these two constraints, the percpu nonidle time would be mainly consumed by
> > > none CFS tasks and couldn't be averaged. Eliminating them by calc the time growth
> > > via the proportion of cfs_rq's utilization on the whole rq.
>
>
> > > +static unsigned long psi_memtime_fixup(u32 growth)
> > > +{
> > > + struct rq *rq = task_rq(current);
> > > + unsigned long growth_fixed = (unsigned long)growth;
> > > +
> > > + if (!(current->policy == SCHED_NORMAL || current->policy == SCHED_BATCH))
> > > + return growth_fixed;
> > > +
> > > + if (current->in_memstall)
> > > + growth_fixed = div64_ul((1024 - rq->avg_rt.util_avg - rq->avg_dl.util_avg
> > > + - rq->avg_irq.util_avg + 1) * growth, 1024);
> > > +
> > > + return growth_fixed;
> > > +}
> > > +
> > > static void init_triggers(struct psi_group *group, u64 now)
> > > {
> > > struct psi_trigger *t;
> > > @@ -658,6 +675,7 @@ static void record_times(struct psi_group_cpu *groupc, u64 now)
> > > }
> > >
> > > if (groupc->state_mask & (1 << PSI_MEM_SOME)) {
> > > + delta = psi_memtime_fixup(delta);
> >
> > Ok, so we want to deduct IRQ and RT preemption time from the memstall
> > period of an active reclaimer, since it's technically not stalled on
> > memory during this time but on CPU.
> >
> > However, we do NOT want to deduct IRQ and RT time from memstalls that
> > are sleeping on refaults swapins, since they are not affected by what
> > is going on on the CPU.
>
> I think that focus on RT/IRQ is mis-guided here, and the implementation
> is horrendous.
>
> So the fundamental question seems to be; and I think Johannes is the one
> to answer that: What time-base do these metrics want to use?
>
> Do some of these states want to account in task-time instead of
> wall-time perhaps? I can't quite remember, but vague memories are
> telling me most of the PSI accounting was about blocked tasks, not
> running tasks, which makes all this rather more complicated.
I tend to agree with this.
Using rq_clock_task(rq) instead of cpu_clock(cpu) will remove the time
spent under interrupt as an example
and AFAICT, rq->clock_task is updated before calling psi function
>
> Randomly scaling time as proposed seems almost certainly wrong. What
> would that make the stats mean?
next prev parent reply other threads:[~2021-11-10 8:49 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-15 6:16 [Resend PATCH] psi : calc cfs task memstall time more precisely Huangzhaoyang
2021-11-02 19:47 ` Johannes Weiner
2021-11-03 7:07 ` Zhaoyang Huang
2021-11-03 7:08 ` Zhaoyang Huang
2021-11-04 8:58 ` Dietmar Eggemann
2021-11-05 5:58 ` Zhaoyang Huang
2021-11-05 16:42 ` Dietmar Eggemann
2021-11-08 8:49 ` Xuewen Yan
2021-11-08 9:20 ` Zhaoyang Huang
2021-11-09 12:29 ` Dietmar Eggemann
2021-11-10 5:38 ` Xuewen Yan
2021-11-09 9:43 ` Dietmar Eggemann
2021-11-10 5:36 ` Xuewen Yan
2021-11-12 14:16 ` Dietmar Eggemann
2021-11-09 14:56 ` Peter Zijlstra
2021-11-10 1:37 ` Zhaoyang Huang
2021-11-10 8:36 ` Peter Zijlstra
2021-11-10 8:47 ` Zhaoyang Huang
2021-11-10 8:49 ` Vincent Guittot [this message]
2021-11-10 9:04 ` Zhaoyang Huang
2021-11-12 16:36 ` Johannes Weiner
2021-11-12 19:23 ` Peter Zijlstra
2021-11-15 2:24 ` Zhaoyang Huang
-- strict thread matches above, loose matches on Subject: below --
2021-09-26 3:27 Huangzhaoyang
2021-09-18 5:25 Huangzhaoyang
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=CAKfTPtAVQ+mtYkGv5xJnbjBO9L9z7jSKOvzhObd0MvVpxakezw@mail.gmail.com \
--to=vincent.guittot@linaro.org \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=huangzhaoyang@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=peterz@infradead.org \
--cc=vdavydov.dev@gmail.com \
--cc=zhaoyang.huang@unisoc.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).