linux-kernel - Re: [PATCH v6 1/4] sched/fair: Fix attaching task sched avgs twice when switching to fair or changing task group

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: <CAKfTPtDfTKGgQd=ebiKijA0_nBEKzwHL7W_WnMw6AWaVP=TC-g@mail.gmail.com>
Date:	Wed, 15 Jun 2016 16:15:29 +0200
From:	Vincent Guittot <vincent.guittot@...aro.org>
To:	Yuyang Du <yuyang.du@...el.com>
Cc:	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...nel.org>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	Mike Galbraith <umgwanakikbuti@...il.com>,
	Benjamin Segall <bsegall@...gle.com>,
	Paul Turner <pjt@...gle.com>,
	Morten Rasmussen <morten.rasmussen@....com>,
	Dietmar Eggemann <dietmar.eggemann@....com>,
	Matt Fleming <matt@...eblueprint.co.uk>
Subject: Re: [PATCH v6 1/4] sched/fair: Fix attaching task sched avgs twice
 when switching to fair or changing task group

On 15 June 2016 at 02:18, Yuyang Du <yuyang.du@...el.com> wrote:
> On Wed, Jun 15, 2016 at 09:46:53AM +0200, Vincent Guittot wrote:
>> I still have concerned with this change of the behavior that  attaches
>> the task only when it is enqueued. The load avg of the task will not
>> be decayed between the time we move it into its new group until its
>> enqueue. With this change, a task's load can stay high whereas it has
>> slept for the last couple of seconds.
>
> The task will be updated when it is moved in the detach, so if it stays
> high, it is highly expected to be enqueued soon. But if the task is never

you can't do such assumption because if it's not enqueued soon (keep
in mind that a 10ms sleep will decay the value by 20% from max),
metrics become just wrong. Furthermore, the load metrics of the task
becomes different between a task being moved or the same task staying
on same cfs_rq.

> enqueued, we don't bother the new gruop cfs_rq.
>
> So this is not perfect, and nothing is perfect, but a simple solution.