LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Nick Piggin <npiggin@suse.de>
To: "Siddha, Suresh B" <suresh.b.siddha@intel.com>
Cc: mingo@elte.hu, linux-kernel@vger.kernel.org, rostedt@goodmis.org
Subject: Re: [patch 2/2] sched: dynticks idle load balancing - v2
Date: Thu, 22 Feb 2007 04:26:54 +0100	[thread overview]
Message-ID: <20070222032654.GD12137@wotan.suse.de> (raw)
In-Reply-To: <20070216180842.B8744@unix-os.sc.intel.com>

On Fri, Feb 16, 2007 at 06:08:42PM -0800, Suresh B wrote:
> Changes since v1:
> 
>   - Move the idle load balancer selection from schedule()
>     to the first busy scheduler_tick() after restarting the tick.
>     This will avoid the unnecessay ownership changes when 
>     softirq's(which are run in softirqd context in certain -rt
>     configurations) like timer, sched are invoked for every idle tick
>     that happens.
> 
>   - Misc cleanups.
> 
> ---
> Fix the process idle load balancing in the presence of dynticks.
> cpus for which ticks are stopped will sleep till the next event wakes it up.
> Potentially these sleeps can be for large durations and during which today,
> there is no periodic idle load balancing being done.
> 
> This patch nominates an owner among the idle cpus, which does the idle load
> balancing on behalf of the other idle cpus. And once all the cpus are
> completely idle, then we can stop this idle load balancing too. Checks added
> in fast path are minimized.  Whenever there are busy cpus in the system, there
> will be an owner(idle cpu) doing the system wide idle load balancing.

This is really ugly, sorry :(

My suggestion for handling this was to increase the maximum balance
interval for an idle CPU, and just implement a global shutdown when
the entire system goes idle.

The former should take care of the power savings issues for bare metal
hardware, and the latter should solve performance problems for many idle
SMP guests. It should take very little code to implement.

If that approach doesn't cut it, then at least we can have some numbers
to see how much better yours is so we can justify including it.

If you are against my approach, then I can have a try at coding it up
if you like?

  parent reply	other threads:[~2007-02-22  3:26 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-12-11 23:53 [RFC] Patch: dynticks: idle load balancing Siddha, Suresh B
2006-12-13 22:43 ` Ingo Molnar
2006-12-13 23:13   ` Ingo Molnar
2006-12-13 23:03     ` Siddha, Suresh B
2006-12-13 23:31       ` Ingo Molnar
2006-12-13 23:19         ` Siddha, Suresh B
2006-12-14  0:34           ` Ingo Molnar
2006-12-19 20:12           ` Ingo Molnar
2006-12-19 21:12             ` Siddha, Suresh B
2007-01-16 11:35               ` Ingo Molnar
2007-01-30 21:57                 ` Siddha, Suresh B
2007-02-07 22:19                   ` Siddha, Suresh B
2007-02-17  2:03                   ` [patch 1/2] sched: fix idle load balancing in softirqd context Siddha, Suresh B
2007-02-17  2:08                     ` [patch 2/2] sched: dynticks idle load balancing - v2 Siddha, Suresh B
2007-02-21 20:23                       ` Andrew Morton
2007-02-22  3:14                         ` Nick Piggin
2007-02-24  2:01                         ` [patch] sched: dynticks idle load balancing - v3 Siddha, Suresh B
2007-02-22  3:26                       ` Nick Piggin [this message]
2007-02-22 22:33                         ` [patch 2/2] sched: dynticks idle load balancing - v2 Siddha, Suresh B
2007-02-23  3:43                           ` Nick Piggin
2007-02-17 14:42                     ` [patch 1/2] sched: fix idle load balancing in softirqd context Steven Rostedt
2007-02-21  6:25                       ` Siddha, Suresh B
2007-02-21 20:13                     ` Andrew Morton
2006-12-13 23:48     ` [RFC] Patch: dynticks: idle load balancing Ingo Molnar
2006-12-20  0:49 ` Steven Rostedt

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=20070222032654.GD12137@wotan.suse.de \
    --to=npiggin@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=rostedt@goodmis.org \
    --cc=suresh.b.siddha@intel.com \
    --subject='Re: [patch 2/2] sched: dynticks idle load balancing - v2' \
    /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

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).