linux-kernel - Re: [PATCH -v2 1/5] sched: Fix ttwu() race

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] [day] [month] [year] [list]

Message-ID: <20200724175542.GR9247@paulmck-ThinkPad-P72>
Date:   Fri, 24 Jul 2020 10:55:42 -0700
From:   "Paul E. McKenney" <paulmck@...nel.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Chris Wilson <chris@...is-wilson.co.uk>, mingo@...nel.org,
        tglx@...utronix.de, linux-kernel@...r.kernel.org,
        juri.lelli@...hat.com, vincent.guittot@...aro.org,
        dietmar.eggemann@....com, rostedt@...dmis.org, bsegall@...gle.com,
        mgorman@...e.de, frederic@...nel.org,
        torvalds@...ux-foundation.org, hch@....de
Subject: Re: [PATCH -v2 1/5] sched: Fix ttwu() race

On Thu, Jul 23, 2020 at 10:11:28PM +0200, Peter Zijlstra wrote:
> On Thu, Jul 23, 2020 at 08:41:03PM +0100, Chris Wilson wrote:
> 
> > I am very sorry for the wild goose chase.
> 
> *phew*... all good then. I was starting to go a little ga-ga trying to
> make sense of things.
> 
> Arguably we should probably do something like:
> 
> 
> @@ -4555,7 +4572,7 @@ asmlinkage __visible void __sched preempt_schedule_irq(void)
>  int default_wake_function(wait_queue_entry_t *curr, unsigned mode, int wake_flags,
>  			  void *key)
>  {
> -	return try_to_wake_up(curr->private, mode, wake_flags);
> +	return try_to_wake_up(curr->private, mode, wake_flags & WF_SYNC);
>  }
>  EXPORT_SYMBOL(default_wake_function);

If you do:

Tested-by: Paul E. McKenney <paulmck@...nel.org>

This was about nine hours of each of the default rcutorture scenarios.

							Thanx, Paul

> Since I don't think anybody uses anything other than WF_SYNC, ever. And
> the rest of the WF_flags are used internally.
> 
> Thanks Chris!