linux-kernel - Re: [RFC][PATCH 5/5] sched: Reduce ttwu rq->lock contention

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <AANLkTik86z7NZCqkYU1JY1NUW3qxthaBohY3CNk6awEd@mail.gmail.com>
Date:	Sat, 18 Dec 2010 22:49:19 +0800
From:	Yong Zhang <yong.zhang0@...il.com>
To:	Peter Zijlstra <peterz@...radead.org>
Cc:	Oleg Nesterov <oleg@...hat.com>,
	Chris Mason <chris.mason@...cle.com>,
	Frank Rowand <frank.rowand@...sony.com>,
	Ingo Molnar <mingo@...e.hu>,
	Thomas Gleixner <tglx@...utronix.de>,
	Mike Galbraith <efault@....de>, Paul Turner <pjt@...gle.com>,
	Jens Axboe <axboe@...nel.dk>, linux-kernel@...r.kernel.org
Subject: Re: [RFC][PATCH 5/5] sched: Reduce ttwu rq->lock contention

On Sat, Dec 18, 2010 at 2:15 AM, Peter Zijlstra <peterz@...radead.org> wrote:
> On Fri, 2010-12-17 at 18:43 +0100, Peter Zijlstra wrote:
>>
>> Hrmph, so is it only about serializing concurrent wakeups? If so, we
>> could possibly hold p->pi_lock over the wakeup.
>
> Something like the below.. except it still suffers from the
> __migrate_task() hole you identified in your other email.
>
> By fully serializing all wakeups using ->pi_lock it becomes a lot
> simpler (although I just realized we might have a problem with
> try_to_wake_up_local).
>
> static int
> try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
> {
>        unsigned long flags;
>        int cpu, ret = 0;
>
>        smp_wmb();
>        raw_spin_lock_irqsave(&p->pi_lock, flags);
>
>        if (!(p->state & state))
>                goto unlock;
>
>        ret = 1; /* we qualify as a proper wakeup now */

Could below happen in this __window__?

p is going through wake_event and it first set TASK_UNINTERRUPTIBLE,
then waker see that and above if (!(p->state & state)) passed.
But at this time condition == true for p, and p return to run and
intend to sleep:
          p->state == XXX;
          sleep;

then we could wake up a process which has wrong state, no?

>
>        if (p->se.on_rq && ttwu_force(p, state, wake_flags))
>                goto unlock;
>
>        p->sched_contributes_to_load = !!task_contributes_to_load(p);
>
>        /*
>         * In order to serialize against other tasks wanting to task_rq_lock()
>         * we need to wait until the current task_rq(p)->lock holder goes away,
>         * so that the next might observe TASK_WAKING.
>         */
>        p->state = TASK_WAKING;
>        smp_wmb();
>        raw_spin_unlock_wait(&task_rq(p)->lock);
>
>        /*
>         * Stable, now that TASK_WAKING is visible.
>         */
>        cpu = task_cpu(p);
>
> #ifdef CONFIG_SMP
>        /*
>         * Catch the case where schedule() has done the dequeue but hasn't yet
>         * scheduled to a new task, in that case p is still being referenced
>         * by that cpu so we cannot wake it to any other cpu.
>         *
>         * Here we must either do a full remote enqueue, or simply wait for
>         * the remote cpu to finish the schedule(), the latter was found to
>         * be cheapest.
>         */
>        while (p->oncpu)
>                cpu_relax();
>
>        if (p->sched_class->task_waking)
>                p->sched_class->task_waking(p);
>
>        cpu = select_task_rq(p, SD_BALANCE_WAKE, wake_flags);
> #endif
>        ttwu_queue(p, cpu);
>        ttwu_stat(p, cpu, wake_flags);
> unlock:
>        raw_spin_unlock_irqrestore(&p->pi_lock, flags);
>
>        return ret;
> }
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>



-- 
Only stand for myself.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/