linux-kernel - Re: [RFC 3/4] sched/fair: replace runnable load average by runnable average

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:   Fri, 14 Feb 2020 08:43:16 +0100
From:   Vincent Guittot <vincent.guittot@...aro.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Ingo Molnar <mingo@...hat.com>, Juri Lelli <juri.lelli@...hat.com>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>,
        linux-kernel <linux-kernel@...r.kernel.org>,
        Phil Auld <pauld@...hat.com>, Parth Shah <parth@...ux.ibm.com>,
        Valentin Schneider <valentin.schneider@....com>
Subject: Re: [RFC 3/4] sched/fair: replace runnable load average by runnable average

On Thu, 13 Feb 2020 at 18:36, Peter Zijlstra <peterz@...radead.org> wrote:
>
> On Tue, Feb 11, 2020 at 06:46:50PM +0100, Vincent Guittot wrote:
> > @@ -367,6 +367,14 @@ struct util_est {
> >   * For cfs_rq, it is the aggregated load_avg of all runnable and
> >   * blocked sched_entities.
> >   *
> > + * [runnable_avg definition]
> > + *
> > + *   runnable_avg = runnable%
>
>                                  * SCHED_CAPACITY_SCALE
>
> right, just like we have for util_avg.
>
> > + *
> > + * where runnable% is the time ratio that a sched_entity is runnable.
> > + * For cfs_rq, it is the aggregated runnable_avg of all runnable and
> > + * blocked sched_entities.
>
> Which is a verbatim repeat of the runnable% definition for load_avg,
> right? Perhaps re-arrange the text such that we only have a single
> definition for each symbol?

yes . will do

>
> > + *
> >   * [util_avg definition]
> >   *
> >   *   util_avg = running% * SCHED_CAPACITY_SCALE
>
> Can you please split this patch in two? First remove everything
> runnable_load_avg, and then introduce runnable_avg.

ok.

>
> I didn't quickly spot anything off, and you're right that runnable vs
> util is an interesting signal.