From: Suren Baghdasaryan <surenb@google.com>
To: Hillf Danton <hdanton@sina.com>
Cc: Pavan Kondeti <quic_pkondeti@quicinc.com>,
Johannes Weiner <hannes@cmpxchg.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
quic_charante@quicinc.com
Subject: Re: PSI idle-shutoff
Date: Mon, 10 Oct 2022 14:16:26 -0700 [thread overview]
Message-ID: <CAJuCfpGN4TooWJuvoh9RZM96hW4BmuKjwnuT2pM4rGPLcGEinQ@mail.gmail.com> (raw)
In-Reply-To: <20221010105710.171-1-hdanton@sina.com>
On Mon, Oct 10, 2022 at 3:57 AM Hillf Danton <hdanton@sina.com> wrote:
>
> On 13 Sep 2022 19:38:17 +0530 Pavan Kondeti <quic_pkondeti@quicinc.com>
> > Hi
> >
> > The fact that psi_avgs_work()->collect_percpu_times()->get_recent_times()
> > run from a kworker thread, PSI_NONIDLE condition would be observed as
> > there is a RUNNING task. So we would always end up re-arming the work.
> >
> > If the work is re-armed from the psi_avgs_work() it self, the backing off
> > logic in psi_task_change() (will be moved to psi_task_switch soon) can't
> > help. The work is already scheduled. so we don't do anything there.
> >
> > Probably I am missing some thing here. Can you please clarify how we
> > shut off re-arming the psi avg work?
>
> Instead of open coding schedule_delayed_work() in bid to check if timer
> hits the idle task (see delayed_work_timer_fn()), the idle task is tracked
> in psi_task_switch() and checked by kworker to see if it preempted the idle
> task.
>
> Only for thoughts now.
>
> Hillf
>
> +++ b/kernel/sched/psi.c
> @@ -412,6 +412,8 @@ static u64 update_averages(struct psi_gr
> return avg_next_update;
> }
>
> +static DEFINE_PER_CPU(int, prev_task_is_idle);
> +
> static void psi_avgs_work(struct work_struct *work)
> {
> struct delayed_work *dwork;
> @@ -439,7 +441,7 @@ static void psi_avgs_work(struct work_st
> if (now >= group->avg_next_update)
> group->avg_next_update = update_averages(group, now);
>
> - if (nonidle) {
> + if (nonidle && 0 == per_cpu(prev_task_is_idle, raw_smp_processor_id())) {
This condition would be incorrect if nonidle was set by a cpu other
than raw_smp_processor_id() and
prev_task_is_idle[raw_smp_processor_id()] == 0. IOW, if some activity
happens on a non-current cpu, we would fail to reschedule
psi_avgs_work for it. This can be fixed in collect_percpu_times() by
considering prev_task_is_idle for all other CPUs as well. However
Chengming's approach seems simpler to me TBH and does not require an
additional per-cpu variable.
> schedule_delayed_work(dwork, nsecs_to_jiffies(
> group->avg_next_update - now) + 1);
> }
> @@ -859,6 +861,7 @@ void psi_task_switch(struct task_struct
> if (prev->pid) {
> int clear = TSK_ONCPU, set = 0;
>
> + per_cpu(prev_task_is_idle, cpu) = 0;
> /*
> * When we're going to sleep, psi_dequeue() lets us
> * handle TSK_RUNNING, TSK_MEMSTALL_RUNNING and
> @@ -888,7 +891,8 @@ void psi_task_switch(struct task_struct
> for (; group; group = iterate_groups(prev, &iter))
> psi_group_change(group, cpu, clear, set, now, true);
> }
> - }
> + } else
> + per_cpu(prev_task_is_idle, cpu) = 1;
> }
>
> /**
>
next prev parent reply other threads:[~2022-10-10 21:16 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-13 14:08 PSI idle-shutoff Pavan Kondeti
2022-09-15 6:20 ` Pavan Kondeti
2022-09-17 5:45 ` Suren Baghdasaryan
2022-10-03 6:11 ` Suren Baghdasaryan
2022-10-05 16:32 ` Suren Baghdasaryan
2022-10-09 12:41 ` Chengming Zhou
2022-10-09 13:17 ` Chengming Zhou
2022-10-10 6:18 ` Pavan Kondeti
2022-10-10 6:43 ` Pavan Kondeti
2022-10-10 6:57 ` [External] " Chengming Zhou
2022-10-10 8:30 ` Chengming Zhou
2022-10-10 9:09 ` Pavan Kondeti
2022-10-10 9:22 ` Chengming Zhou
2022-10-10 20:59 ` Suren Baghdasaryan
2022-10-10 20:33 ` Suren Baghdasaryan
2022-10-10 5:57 ` Pavan Kondeti
2022-10-10 9:01 ` Pavan Kondeti
2022-10-10 6:25 ` Pavan Kondeti
2022-10-10 10:42 ` [PATCH] sched/psi: Fix avgs_work re-arm in psi_avgs_work() Chengming Zhou
2022-10-10 21:21 ` Suren Baghdasaryan
2022-10-11 0:07 ` Chengming Zhou
2022-10-11 17:00 ` Suren Baghdasaryan
2022-10-12 2:10 ` Chengming Zhou
2022-10-12 18:24 ` Suren Baghdasaryan
2022-10-13 2:23 ` Chengming Zhou
2022-10-13 11:06 ` Chengming Zhou
2022-10-13 15:52 ` Johannes Weiner
2022-10-13 16:10 ` Suren Baghdasaryan
2022-10-14 2:03 ` Chengming Zhou
2022-10-14 2:02 ` Chengming Zhou
2022-10-28 6:42 ` [tip: sched/core] " tip-bot2 for Chengming Zhou
2022-10-28 6:50 ` [External] " Chengming Zhou
2022-10-28 15:58 ` Suren Baghdasaryan
2022-10-28 16:05 ` Chengming Zhou
2022-10-28 19:53 ` [External] " Peter Zijlstra
2022-10-29 11:55 ` Peter Zijlstra
2022-10-29 12:40 ` Chengming Zhou
2022-10-29 18:46 ` Suren Baghdasaryan
[not found] ` <20221010105710.171-1-hdanton@sina.com>
2022-10-10 21:16 ` Suren Baghdasaryan [this message]
[not found] ` <20221011113818.340-1-hdanton@sina.com>
2022-10-11 17:11 ` PSI idle-shutoff Suren Baghdasaryan
[not found] ` <20221012062034.486-1-hdanton@sina.com>
2022-10-12 15:40 ` Suren Baghdasaryan
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=CAJuCfpGN4TooWJuvoh9RZM96hW4BmuKjwnuT2pM4rGPLcGEinQ@mail.gmail.com \
--to=surenb@google.com \
--cc=hannes@cmpxchg.org \
--cc=hdanton@sina.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=quic_charante@quicinc.com \
--cc=quic_pkondeti@quicinc.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).