linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Valentin Schneider <valentin.schneider@arm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: mingo@kernel.org, tglx@linutronix.de,
	linux-kernel@vger.kernel.org, jiangshanlai@gmail.com,
	cai@redhat.com, vincent.donnefort@arm.com, decui@microsoft.com,
	paulmck@kernel.org, vincent.guittot@linaro.org,
	rostedt@goodmis.org, tj@kernel.org
Subject: Re: [PATCH -v3 8/9] sched: Fix CPU hotplug / tighten is_per_cpu_kthread()
Date: Thu, 21 Jan 2021 14:36:00 +0000	[thread overview]
Message-ID: <jhj7do6mk6n.mognet@arm.com> (raw)
In-Reply-To: <YAmNIsyaiO6teljH@hirez.programming.kicks-ass.net>

On 21/01/21 15:18, Peter Zijlstra wrote:
> On Thu, Jan 21, 2021 at 02:01:03PM +0000, Valentin Schneider wrote:
>> On 21/01/21 11:17, Peter Zijlstra wrote:
>> > @@ -7504,6 +7525,9 @@ int sched_cpu_deactivate(unsigned int cp
>> >        * preempt-disabled and RCU users of this state to go away such that
>> >        * all new such users will observe it.
>> >        *
>> > +	 * Specifically, we rely on ttwu to no longer target this CPU, see
>> > +	 * ttwu_queue_cond() and is_cpu_allowed().
>> > +	 *
>> 
>> So the last time ttwu_queue_wakelist() can append a task onto a dying
>> CPU's wakelist is before sched_cpu_deactivate()'s synchronize_rcu()
>> returns. 
>> 
>> As discussed on IRC, paranoia would have us issue a
>> 
>>   flush_smp_call_function_from_idle()
>> 
>> upon returning from said sync, but this will require further surgery.
>
> Right, specifically RCU needs a little more help there.
>
>> Do we want something like the below in the meantime? Ideally we'd warn on
>> setting rq->ttwu_pending when !cpu_active(), but as per the above this is
>> allowed before the synchronize_rcu() returns.
>
> I'm not sure I'm brave enough to add that just now :/

I get you; I couldn't come up with a better scheme that would give us a bit
more info than the sched_cpu_dying() splat :/

  reply	other threads:[~2021-01-21 14:37 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-21 10:17 [PATCH -v3 0/9] sched: Fix hot-unplug regression Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 1/9] sched/core: Print out straggler tasks in sched_cpu_dying() Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 2/9] workqueue: Use cpu_possible_mask instead of cpu_active_mask to break affinity Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 3/9] sched: Dont run cpu-online with balance_push() enabled Peter Zijlstra
2021-01-21 14:00   ` Valentin Schneider
2021-01-21 14:14     ` Peter Zijlstra
2021-01-22 17:41   ` [tip: sched/urgent] sched: Don't " tip-bot2 for Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 4/9] kthread: Extract KTHREAD_IS_PER_CPU Peter Zijlstra
2021-01-22 17:41   ` [tip: sched/urgent] " tip-bot2 for Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 5/9] workqueue: Tag bound workers with KTHREAD_IS_PER_CPU Peter Zijlstra
2021-01-21 14:31   ` Valentin Schneider
2021-01-22 17:41   ` [tip: sched/urgent] " tip-bot2 for Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 6/9] workqueue: Restrict affinity change to rescuer Peter Zijlstra
2021-01-22 17:41   ` [tip: sched/urgent] " tip-bot2 for Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 7/9] sched: Prepare to use balance_push in ttwu() Peter Zijlstra
2021-01-22 17:41   ` [tip: sched/urgent] " tip-bot2 for Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 8/9] sched: Fix CPU hotplug / tighten is_per_cpu_kthread() Peter Zijlstra
2021-01-21 14:01   ` Valentin Schneider
2021-01-21 14:18     ` Peter Zijlstra
2021-01-21 14:36       ` Valentin Schneider [this message]
2021-01-22 17:41   ` [tip: sched/urgent] " tip-bot2 for Peter Zijlstra
2021-01-21 10:17 ` [PATCH -v3 9/9] sched: Relax the set_cpus_allowed_ptr() semantics Peter Zijlstra
2021-01-22 17:41   ` [tip: sched/urgent] " tip-bot2 for Peter Zijlstra
2021-01-21 14:36 ` [PATCH -v3 0/9] sched: Fix hot-unplug regression Valentin Schneider
2021-01-21 19:56 ` Paul E. McKenney

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=jhj7do6mk6n.mognet@arm.com \
    --to=valentin.schneider@arm.com \
    --cc=cai@redhat.com \
    --cc=decui@microsoft.com \
    --cc=jiangshanlai@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=paulmck@kernel.org \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=tglx@linutronix.de \
    --cc=tj@kernel.org \
    --cc=vincent.donnefort@arm.com \
    --cc=vincent.guittot@linaro.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;
as well as URLs for NNTP newsgroup(s).