LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Xunlei Pang <pang.xunlei@linaro.org>
To: Peter Zijlstra <peterz@infradead.org>,
	Steven Rostedt <rostedt@goodmis.org>
Cc: lkml <linux-kernel@vger.kernel.org>, Juri Lelli <juri.lelli@gmail.com>
Subject: Re: [PATCH 3/5] sched/deadline: Fix wrong cpudl_find() in check_preempt_equal_dl()
Date: Wed, 28 Jan 2015 23:18:34 +0800	[thread overview]
Message-ID: <CADcy93XVn9akNoHaFJfWxRtszRpF_O0dnGa7iZHBFcYsfBWpCg@mail.gmail.com> (raw)
In-Reply-To: <20150127164716.GH21418@twins.programming.kicks-ass.net>

Hi Peter,

On 28 January 2015 at 00:47, Peter Zijlstra <peterz@infradead.org> wrote:
> On Mon, Jan 19, 2015 at 04:49:38AM +0000, Xunlei Pang wrote:
>> In check_preempt_equal_dl(), cpudl_find() is called with a NULL
>> later_mask, thus cpudl_find() here doesn't check cpudl::free_cpus
>> at all.
>>
>> This patch takles this issue by always passing a non-NULL later_mask
>> to cpudl_find(), thereby fixing this issue.
>
> Fix what issue? Afaict this is an optimization not a fix.

Currently, check_preempt_equal_dl() invokes cpudl_find() with a NULL mask,
so cpudl_find() won't check cpudl::free_cpus. For example, 4 core
system, CPU0~CPU2
are all idle(free of deadline tasks), a deadline task is woken on CPU3
which already
has one running deadline task with the same deadline value, then
cpudl_find() will fail
causing CPU3 holding 2 deadline tasks while other cpus are idle,
obviously it should
be placed on one idle cpu.

Thanks,
Xunlei

  reply	other threads:[~2015-01-28 21:27 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-19  4:49 [PATCH 1/5] sched/deadline: Modify cpudl::free_cpus to reflect rd->online Xunlei Pang
2015-01-19  4:49 ` [PATCH 2/5] sched/deadline: Remove cpu_active_mask from cpudl_find() Xunlei Pang
2015-01-27 15:04   ` Peter Zijlstra
2015-02-04 14:36   ` [tip:sched/core] " tip-bot for Xunlei Pang
2015-01-19  4:49 ` [PATCH 3/5] sched/deadline: Fix wrong cpudl_find() in check_preempt_equal_dl() Xunlei Pang
2015-01-27 12:48   ` Peter Zijlstra
2015-01-27 14:15     ` Peter Zijlstra
2015-01-27 16:47   ` Peter Zijlstra
2015-01-28 15:18     ` Xunlei Pang [this message]
2015-01-19  4:49 ` [PATCH 4/5] sched/rt: Consider deadline tasks in cpupri_find() Xunlei Pang
2015-01-27 12:58   ` Peter Zijlstra
2015-01-27 14:18     ` Peter Zijlstra
2015-01-27 23:04     ` Steven Rostedt
2015-01-28 15:21       ` Xunlei Pang
2015-01-19  4:49 ` [PATCH 5/5] sched/rt: Optimize find_lowest_rq() to select a cache hot cpu Xunlei Pang
2015-01-27 14:21   ` Peter Zijlstra
2015-01-27 14:56     ` Steven Rostedt
2015-01-27 16:28       ` Peter Zijlstra
2015-01-29 16:42       ` Xunlei Pang
2015-01-29 17:17         ` Steven Rostedt
2015-01-29 19:23         ` Peter Zijlstra
2015-02-04 13:07           ` Xunlei Pang
2015-01-23 18:09 ` [PATCH 1/5] sched/deadline: Modify cpudl::free_cpus to reflect rd->online Xunlei Pang
2015-02-01 17:53 ` [tip:sched/core] sched/deadline: Modify cpudl:: free_cpus " tip-bot for Xunlei Pang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CADcy93XVn9akNoHaFJfWxRtszRpF_O0dnGa7iZHBFcYsfBWpCg@mail.gmail.com \
    --to=pang.xunlei@linaro.org \
    --cc=juri.lelli@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).