linux-kernel - Re: [SchedulerWakeupLatency] Per-task vruntime wakeup bonus

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, 17 Jul 2020 14:20:36 +0200
From:   Vincent Guittot <vincent.guittot@...aro.org>
To:     Dietmar Eggemann <dietmar.eggemann@....com>
Cc:     Patrick Bellasi <patrick.bellasi@...bug.net>,
        LKML <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...hat.com>,
        "Peter Zijlstra (Intel)" <peterz@...radead.org>,
        Juri Lelli <juri.lelli@...hat.com>,
        Paul Turner <pjt@...gle.com>, Ben Segall <bsegall@...gle.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Jonathan Corbet <corbet@....net>,
        Dhaval Giani <dhaval.giani@...cle.com>,
        Josef Bacik <jbacik@...com>,
        Chris Hyser <chris.hyser@...cle.com>,
        Parth Shah <parth@...ux.ibm.com>
Subject: Re: [SchedulerWakeupLatency] Per-task vruntime wakeup bonus

On Thu, 16 Jul 2020 at 18:48, Dietmar Eggemann <dietmar.eggemann@....com> wrote:
>
> On 13/07/2020 14:59, Vincent Guittot wrote:
> > On Fri, 10 Jul 2020 at 21:59, Patrick Bellasi
> > <patrick.bellasi@...bug.net> wrote:
> >>
> >>
> >> On Fri, Jul 10, 2020 at 15:21:48 +0200, Vincent Guittot <vincent.guittot@...aro.org> wrote...
>
> [...]
>
> >>> Instead, it should weight the decision in wakeup_preempt_entity() and
> >>> wakeup_gran()
> >>
> >> In those functions we already take the task prio into consideration
> >> (ref details at the end of this message).
> >>
> >> Lower nice value tasks have more chances to preempt current since they
> >> will have a smaller wakeup_gran, indeed:
> >
> > yes, and this is there to ensure a fair distribution of running time
> > and prevent a task to increase significantly its vruntime compare to
> > others
> >
> > -1 min that se already got more runtime than current
> > 0 that se's vruntime will go above current's vruntime after a runtime
> > duration of  sched_min_granularity
> > and 1 means that se got less runtime than current and its vruntime
> > will still be lower than current ones even after a runtime duration of
> > sched_min_granularity
> >
> > IMHO, latency_nice should impact the decision only for case 0 but not
> > the case -1 and 1.
> > That being said, we can extend the case 0 a bit to include the
> > situation where current's vruntime will become greater than se's
> > vruntimes after a runtime duration of  sched_min_granularity like
> > below:
> >
> >                                                            curr->vruntime
> >                                           |<-- wakeup_gran(se) -->|<--
> > wakeupgran(curr) -->|
> > current range: se->vruntime          +1   |        0              |    -1
> > new range:     se->vruntime          +1   |                       0
> >                     |   -1
> >
>
> I assume this got messed up by line break somehow:

yes

>
>                                                    curr->vruntime
>                                  |<-- wakeup_gran(se) -->|<-- wakeup_gran(curr) -->|
> current range: se->vruntime  +1  |        0              |    -1
> new range:     se->vruntime  +1  |                       0                         | -1
>
> IMHO, with the current use of wakeup_preempt_entity() I don't see what
> will change with that.
> We check 'wakeup_preempt_entity() == 1' in check_preempt_wakeup() and
> 'wakeup_preempt_entity() < 1' in pick_next_entity().
>
> How should the mapping between se's latency_nice value to the consideration of
> wakeup_gran(curr) look like?

IMO, the latency_nice can be used to move the +1|0 boundary on the
right  and the 0|-1 on the left with a formula that need to be defined
And we also need to review where and how wakeup_preempt_entity is used

>
> [...]