Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1407797

Re: [PATCH 2/2] sched/fair: Skip detach and attach load avgs for new group task

From Yuyang Du <yuyang.du@intel.com>
Newsgroups linux.kernel
Subject Re: [PATCH 2/2] sched/fair: Skip detach and attach load avgs for new group task
Date 2016-05-27 05:50 +0200
Message-ID <rDgWC-6Eh-11@gated-at.bofh.it> (permalink)
References <rCZCp-4xq-3@gated-at.bofh.it> <rCZCp-4xq-1@gated-at.bofh.it> <rD27f-5YW-13@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


Hi Vincent,

On Thu, May 26, 2016 at 01:50:56PM +0200, Vincent Guittot wrote:
> On 26 May 2016 at 03:14, Yuyang Du <yuyang.du@intel.com> wrote:
> > Vincent reported that the first task to a new task group's cfs_rq will
> > be attached in attach_task_cfs_rq() and once more when it is enqueued
> > (see https://lkml.org/lkml/2016/5/25/388).
> >
> > Actually, it is worse, attach_task_cfs_rq() is called for new task even
> > way before init_entity_runnable_average().
> >
> > Solve this by avoiding attach as well as detach new task's sched avgs
> > in task_move_group_fair(). To do it, we need to know whether the task
> > is forked or not, so we pass this info all the way from sched_move_task()
> > to attach_task_cfs_rq().
> 
> Not sure that this is the right way to solve this problem because you
> continue to attach the task twice without detaching it in the mean
> time:
> - once during the copy of the process in cpu_cgroup_fork (you skip the
> attach of load average but the task is still attached to the local
> cpu)

Sorry, the task's what is still attached, and how? You mean the vruntime
thingy? But the load/util avgs are not.

> In the mean time, sched_entity is initialized and the last_update_time is reset

last_update_time is set to 0 in initialization, and this is the first time
it is touched, no?

> - one more time when the task is enqueued because the last_update_time
> has been reset (this time you don't skip the attache of load_avg

This is expected/wanted. We don't skip this because this will be the first-time
attach.

> Should you better detach the sched_entity with a copy of its parent
> metrics before initializing it and attaching it to the new cpu ?

Thanks,
Yuyang

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH 2/2] sched/fair: Skip detach and attach load avgs for new group task Yuyang Du <yuyang.du@intel.com> - 2016-05-26 11:20 +0200
  Re: [PATCH 2/2] sched/fair: Skip detach and attach load avgs for new  group task Vincent Guittot <vincent.guittot@linaro.org> - 2016-05-26 14:00 +0200
    Re: [PATCH 2/2] sched/fair: Skip detach and attach load avgs for new  group task Yuyang Du <yuyang.du@intel.com> - 2016-05-27 05:50 +0200
      Re: [PATCH 2/2] sched/fair: Skip detach and attach load avgs for new  group task Vincent Guittot <vincent.guittot@linaro.org> - 2016-05-27 15:40 +0200
        Re: [PATCH 2/2] sched/fair: Skip detach and attach load avgs for new  group task Yuyang Du <yuyang.du@intel.com> - 2016-05-30 11:40 +0200

csiph-web