public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	"Rafael J. Wysocki" <rjw@sisk.pl>,
	Thomas Gleixner <tglx@linutronix.de>,
	Mike Galbraith <efault@gmx.de>, Ingo Molnar <mingo@elte.hu>,
	LKML <linux-kernel@vger.kernel.org>,
	pm list <linux-pm@lists.linux-foundation.org>,
	Greg KH <gregkh@suse.de>, Jesse Barnes <jbarnes@virtuousgeek.org>,
	Andi Kleen <ak@linux.intel.com>
Subject: Re: GPF in run_workqueue()/list_del_init(cwq->worklist.next) on resume
Date: Thu, 12 Nov 2009 19:35:08 +0100	[thread overview]
Message-ID: <20091112183508.GA14661@redhat.com> (raw)
In-Reply-To: <4AFB9595.1030002@kernel.org>

On 11/12, Tejun Heo wrote:
>
> Oleg Nesterov wrote:
> >
> > This is even documented, the comment above queue_work() says:
> >
> > 	* We queue the work to the CPU on which it was submitted, but if the CPU dies
> > 	* it can be processed by another CPU.
> >
> > We can improve things, see http://marc.info/?l=linux-kernel&m=125562105103769
> >
> > But then we should also change workqueue_cpu_callback(CPU_POST_DEAD).
> > Instead of flushing, we should carefully move the pending works to
> > another CPU, otherwise the self-requeueing work can block cpu_down().
>
> That looks like an excellent idea and I don't think it will add
> noticeable overhead even done by default and it will magically make
> the "how to implement single-threaded wq semantics in conccurrency
> managed wq" problem go away.  I'll work on it.

I am still not sure all work_structs should single-threaded by default.

To clarify, I am not arguing. Just I don't know. I mean, this change can
break the existing code, and it is not easy to notice the problem.

> If you look at the workqueue code itself very closely, all subtleties
> are properly defined and described.  The problem is that it's not very
> clear and way too subtle when seen from outside and workqueue is
> something used very widely.

Yes, agreed.

> making flush_work() behave as
> flush_work_sync() by default should be doable without too much
> overhead.  I'll give it a shot.

Well, I disagree. Imho it is better to have both flush_work() and
flush_work_sync(). flush_work() is "special" and should be used with
care. But this is minor, and if the work_stuct is single-threaded then
flush_work() == flush_work_sync().

(Perhaps this is what you meant)

> > Not sure this patch will help, but I bet that the actual reason for
> > this bug is much simpler than the subtle races above ;)
>
> And yes it was but still I'm fairly sure unexpected races described
> above are happening.

Yes, sure, I never argued.

My only point was, it is not that workqueues are buggy, they were
designed (and even documented) to work this way. I can't judge if it
was right or not, but personally I think everything is "logical".

That said, I agree that we have too many buggy users, perhaps we
should simplify the rules.

I just noticed that schedule_on_each_cpu() was recently changed by

	HWPOISON: Allow schedule_on_each_cpu() from keventd
	commit: 65a64464349883891e21e74af16c05d6e1eeb4e9

Surprisingly, even this simple change is not exactly right.

	/*
	 * when running in keventd don't schedule a work item on itself.
	 * Can just call directly because the work queue is already bound.
	 * This also is faster.
	 * Make this a generic parameter for other workqueues?
	 */
	if (current_is_keventd()) {
		orig = raw_smp_processor_id();
		INIT_WORK(per_cpu_ptr(works, orig), func);
		func(per_cpu_ptr(works, orig));
	}

OK, but this code should be moved down, under get_online_cpus().

schedule_on_each_cpu() should guarantee that func() can't race with
CPU hotplug, can safely use per-cpu data, etc. That is why flush_work()
is called before put_online_cpus().

Another reason to move this code down is that current_is_keventd()
itself is racy, the "preempt-safe: keventd is per-cpu" comment is not
right. I think do_boot_cpu() case is fine though.

(off-topic, but we can also simplify the code a little bit, the second
 "if (cpu != orig)" is not necessary).


Perhaps you and Linus are right, and we should simplify the rules
unconditionally. But note that the problem above has nothing to do with
single-threaded behaviour, and I do not think it is possible to guarantee
work->func() can't be moved to another CPU.

Oleg.


  reply	other threads:[~2009-11-12 18:41 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-11-09 11:50 Help needed: Resume problems in 2.6.32-rc, perhaps related to preempt_count leakage in keventd Rafael J. Wysocki
2009-11-09 12:02 ` Ingo Molnar
2009-11-09 12:24   ` Rafael J. Wysocki
2009-11-09 12:49     ` Ingo Molnar
2009-11-09 14:02       ` Thomas Gleixner
2009-11-09 14:16         ` Mike Galbraith
2009-11-09 14:27           ` Rafael J. Wysocki
2009-11-09 14:30             ` Mike Galbraith
2009-11-09 15:47               ` Rafael J. Wysocki
2009-11-09 16:19                 ` Mike Galbraith
2009-11-09 17:36                   ` Rafael J. Wysocki
2009-11-09 18:50                     ` Thomas Gleixner
2009-11-09 20:00                       ` Rafael J. Wysocki
2009-11-09 20:31                         ` [linux-pm] " Alan Stern
2009-11-09 20:48                           ` Rafael J. Wysocki
2009-11-09 21:24                             ` Alan Stern
2009-11-09 20:45                         ` GPF in run_workqueue()/list_del_init(cwq->worklist.next) on resume (was: Re: Help needed: Resume problems in 2.6.32-rc, perhaps related to preempt_count leakage in keventd) Rafael J. Wysocki
2009-11-09 21:42                           ` Linus Torvalds
2009-11-10  0:19                             ` Rafael J. Wysocki
2009-11-10 22:02                               ` Linus Torvalds
2009-11-11  8:08                                 ` GPF in run_workqueue()/list_del_init(cwq->worklist.next) on resume Tejun Heo
2009-11-11 18:13                                   ` Oleg Nesterov
2009-11-12  4:56                                     ` Tejun Heo
2009-11-12 18:35                                       ` Oleg Nesterov [this message]
2009-11-12 19:14                                         ` Tejun Heo
2009-11-16 11:01                                           ` Tejun Heo
2009-11-11 11:52                                 ` GPF in run_workqueue()/list_del_init(cwq->worklist.next) on resume (was: Re: Help needed: Resume problems in 2.6.32-rc, perhaps related to preempt_count leakage in keventd) Rafael J. Wysocki
2009-11-11 19:52                                   ` Linus Torvalds
2009-11-11 20:18                                     ` Marcel Holtmann
2009-11-11 20:25                                       ` Linus Torvalds
2009-11-11 21:18                                         ` Rafael J. Wysocki
2009-11-11 21:13                                       ` Oliver Neukum
2009-11-11 21:38                                         ` Linus Torvalds
2009-11-11 21:44                                           ` Oliver Neukum
2009-11-11 16:13                                 ` Oleg Nesterov
2009-11-11 20:00                                   ` Rafael J. Wysocki
2009-11-11 20:11                                     ` Linus Torvalds
2009-11-11 20:20                                       ` Marcel Holtmann
2009-11-11 20:24                                     ` Oleg Nesterov
2009-11-11 21:15                                       ` Oliver Neukum
2009-11-11 17:17                                 ` Oleg Nesterov
2009-11-12 17:33                                   ` Thomas Gleixner
2009-11-12 19:17                                     ` GPF in run_workqueue()/list_del_init(cwq->worklist.next) on resume Tejun Heo
2009-11-12 20:53                                       ` Thomas Gleixner
2009-11-12 20:53                                     ` GPF in run_workqueue()/list_del_init(cwq->worklist.next) on resume (was: Re: Help needed: Resume problems in 2.6.32-rc, perhaps related to preempt_count leakage in keventd) Rafael J. Wysocki
2009-11-12 20:55                                       ` Thomas Gleixner
2009-11-12 22:55                                         ` Rafael J. Wysocki
2009-11-12 23:08                                           ` Thomas Gleixner
2009-11-15 23:37                                     ` Frederic Weisbecker
2009-11-15 23:40                                       ` Frederic Weisbecker
2009-11-09 19:13                     ` Help needed: Resume problems in 2.6.32-rc, perhaps related to preempt_count leakage in keventd Thomas Gleixner
2009-11-09 20:03                       ` Rafael J. Wysocki
2009-11-09 14:26         ` Rafael J. Wysocki
2009-11-09 14:44           ` Mike Galbraith
2009-11-09 15:47             ` Rafael J. Wysocki
2009-11-09 15:57         ` Linus Torvalds

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=20091112183508.GA14661@redhat.com \
    --to=oleg@redhat.com \
    --cc=ak@linux.intel.com \
    --cc=efault@gmx.de \
    --cc=gregkh@suse.de \
    --cc=jbarnes@virtuousgeek.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@lists.linux-foundation.org \
    --cc=mingo@elte.hu \
    --cc=rjw@sisk.pl \
    --cc=tglx@linutronix.de \
    --cc=tj@kernel.org \
    --cc=torvalds@linux-foundation.org \
    /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