From: Johannes Weiner <hannes@cmpxchg.org>
To: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: Peter Ziljstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Juri Lelli <juri.lelli@redhat.com>,
Vincent Guittot <vincent.guittot@linaro.org>,
Suren Baghdasaryan <surenb@google.com>,
linux-kernel@vger.kernel.org,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Steven Rostedt <rostedt@goodmis.org>,
Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
Valentin Schneider <vschneid@redhat.com>,
Thomas Gleixner <tglx@linutronix.de>,
Klaus Kudielka <klaus.kudielka@gmail.com>,
Chris Bainbridge <chris.bainbridge@gmail.com>,
"Linux regression tracking (Thorsten Leemhuis)"
<regressions@leemhuis.info>,
"Gautham R. Shenoy" <gautham.shenoy@amd.com>,
Youssef Esmat <youssefesmat@google.com>,
Paul Menzel <pmenzel@molgen.mpg.de>,
Bert Karwatzki <spasswolf@web.de>,
regressions@lists.linux.dev
Subject: Re: [PATCH 3/3] sched/core: Indicate a sched_delayed task was migrated before wakeup
Date: Thu, 10 Oct 2024 09:03:16 -0400 [thread overview]
Message-ID: <20241010130316.GA181795@cmpxchg.org> (raw)
In-Reply-To: <20241010082838.2474-4-kprateek.nayak@amd.com>
Hi Prateek,
patches 1 and 2 make obvious sense to me.
On Thu, Oct 10, 2024 at 08:28:38AM +0000, K Prateek Nayak wrote:
> @@ -129,6 +129,13 @@ static inline void psi_enqueue(struct task_struct *p, bool wakeup)
> if (static_branch_likely(&psi_disabled))
> return;
>
> + /*
> + * Delayed task is not ready to run yet!
> + * Wait for a requeue before accounting.
> + */
> + if (p->se.sched_delayed)
> + return;
This one is problematic. It clears sleeping state (memstall, iowait)
during the dequeue of the migration but doesn't restore it until the
wakeup, which could presumably be much later. This leaves a gap in the
accounting.
psi really wants the dequeue and enqueue of the migration, even when a
task is delay-dequeued. We just have to get the context parsing right
to not confuse migration queues with wakeups.
I'll try to come up with a suitable solution as well, please don't
apply this one for now.
next prev parent reply other threads:[~2024-10-10 13:03 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-10 8:28 [PATCH 0/3] sched/core: Fix PSI inconsistent task state splats with DELAY_DEQUEUE K Prateek Nayak
2024-10-10 8:28 ` [PATCH 1/3] sched/core: Dequeue PSI signals for blocked tasks that are delayed K Prateek Nayak
2024-10-10 19:23 ` Johannes Weiner
2024-10-10 8:28 ` [PATCH 2/3] sched/core: Add ENQUEUE_WAKEUP flag alongside ENQUEUE_DELAYED K Prateek Nayak
2024-10-10 8:28 ` [PATCH 3/3] sched/core: Indicate a sched_delayed task was migrated before wakeup K Prateek Nayak
2024-10-10 13:03 ` Johannes Weiner [this message]
2024-10-10 13:06 ` Peter Zijlstra
2024-10-10 19:37 ` Johannes Weiner
2024-10-11 3:31 ` K Prateek Nayak
2024-10-11 8:33 ` Peter Zijlstra
2024-10-11 10:08 ` Johannes Weiner
2024-10-11 10:39 ` Peter Zijlstra
2024-10-14 14:43 ` Johannes Weiner
2024-10-15 3:11 ` K Prateek Nayak
2024-10-28 13:28 ` [tip: sched/core] sched: psi: pass enqueue/dequeue flags to psi callbacks directly tip-bot2 for Johannes Weiner
2024-10-12 14:15 ` [tip: sched/urgent] Since sched_delayed tasks remain queued even after blocking, the load tip-bot2 for Johannes Weiner
2024-10-14 7:28 ` [tip: sched/urgent] sched/psi: Fix mistaken CPU pressure indication after corrupted task state bug tip-bot2 for Johannes Weiner
2024-10-10 15:59 ` [PATCH 3/3] sched/core: Indicate a sched_delayed task was migrated before wakeup K Prateek Nayak
2024-10-10 10:47 ` [PATCH 0/3] sched/core: Fix PSI inconsistent task state splats with DELAY_DEQUEUE Peter Zijlstra
2024-10-10 10:57 ` K Prateek Nayak
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=20241010130316.GA181795@cmpxchg.org \
--to=hannes@cmpxchg.org \
--cc=bsegall@google.com \
--cc=chris.bainbridge@gmail.com \
--cc=dietmar.eggemann@arm.com \
--cc=gautham.shenoy@amd.com \
--cc=juri.lelli@redhat.com \
--cc=klaus.kudielka@gmail.com \
--cc=kprateek.nayak@amd.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=pmenzel@molgen.mpg.de \
--cc=regressions@leemhuis.info \
--cc=regressions@lists.linux.dev \
--cc=rostedt@goodmis.org \
--cc=spasswolf@web.de \
--cc=surenb@google.com \
--cc=tglx@linutronix.de \
--cc=vincent.guittot@linaro.org \
--cc=vschneid@redhat.com \
--cc=youssefesmat@google.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