linux-kernel - Re: [PATCH v4 17/19] sched: Add migrate

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]

Date:   Thu, 29 Oct 2020 18:59:23 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     Valentin Schneider <valentin.schneider@....com>
Cc:     tglx@...utronix.de, mingo@...nel.org, linux-kernel@...r.kernel.org,
        bigeasy@...utronix.de, qais.yousef@....com, swood@...hat.com,
        juri.lelli@...hat.com, vincent.guittot@...aro.org,
        dietmar.eggemann@....com, rostedt@...dmis.org, bsegall@...gle.com,
        mgorman@...e.de, bristot@...hat.com, vincent.donnefort@....com,
        tj@...nel.org, ouwen210@...mail.com
Subject: Re: [PATCH v4 17/19] sched: Add migrate_disable() tracepoints

On Thu, Oct 29, 2020 at 05:56:12PM +0000, Valentin Schneider wrote:
> 
> On 29/10/20 17:43, Peter Zijlstra wrote:
> > On Thu, Oct 29, 2020 at 04:27:26PM +0000, Valentin Schneider wrote:
> >> Don't you want those directly after the ->migration_disabled write?
> >> esp. for migrate_enable(), if that preempt_enable() leads to a context
> >> switch then the disable->enable deltas won't reflect the kernel view.
> >> 
> >> That delta may indeed include the time it took to run the stopper and
> >> fix the task's affinity on migrate_enable(), but it could include all
> >> sorts of other higher-priority tasks.
> >
> > I can put them in the preempt_disable() section I suppose, but these
> > tracers should be looking at task_sched_runtime(), not walltime, and
> > then the preemption doesn't matter.
> >
> 
> True. I was thinking of how to process it downstream, and the first thing
> that came to mind was that rd->overutilized flag which we do monitor
> fairly closely; however that is system-wide while migrate_disable() is
> task-specific.
> 
> > Also, a distinct lack of actual users atm.. :/
> 
> If you'd rather ditch this one altogether until someone asks for it, that
> also works for me.

Yeah, I can pull this patch until we get someone that actually needs it.