From: Mike Galbraith <efault@gmx.de>
To: Nikhil Rao <ncrao@google.com>
Cc: Ingo Molnar <mingo@elte.hu>,
Peter Zijlstra <peterz@infradead.org>,
Venkatesh Pallipadi <venki@google.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 0/3][RFC] Improve load balancing when tasks have large weight differential
Date: Mon, 04 Oct 2010 05:08:37 +0200 [thread overview]
Message-ID: <1286161717.7410.12.camel@marge.simson.net> (raw)
In-Reply-To: <AANLkTi=TQnYGLPC=Pi0o1As83W6VdWdjvimZNSrGT4qW@mail.gmail.com>
Sorry for the late reply. (fired up your patchlet bright and early so
it didn't rot in my inbox any longer;)
On Wed, 2010-09-29 at 12:32 -0700, Nikhil Rao wrote:
> On Tue, Sep 28, 2010 at 6:45 PM, Mike Galbraith <efault@gmx.de> wrote:
> > On Tue, 2010-09-28 at 14:15 -0700, Nikhil Rao wrote:
> >
> >> Thanks for running this. I've not been able to reproduce what you are
> >> seeing on the few test machines that I have (different combinations of
> >> MC, CPU and NODE domains). Can you please give me more info about
> >> your setup?
> >
> > It's a plain-jane Q6600 box, so has only MC and CPU domains.
> >
> > It doesn't necessarily _instantly_ "stick", can take a couple tries, or
> > a little time.
>
> The closest I have is a quad-core dual-socket machine (MC, CPU
> domains). And I'm having trouble reproducing it on that machine as
> well :-( I ran 5 soaker threads (one of them niced to -15) for a few
> hours and didn't see the problem. Can you please give me some trace
> data & schedstats to work with?
Booting with isolcpus or offlining the excess should help.
> Looking at the patch/code, I suspect active migration on the CPU
> scheduling domain pushes the nice 0 task (running on the same socket
> as the nice -15 task) to the other socket. This leaves you with an
> idle core on the nice -15 socket, and with soaker threads there is no
> way to come back to a 100% utilized state. One possible explanation is
> the group capacity for a sched group in the CPU sched domain is
> rounded to 1 (instead of 2). I have a patch below that throws a hammer
> at the problem and uses group weight instead of group capacity (this
> is experimental, will refine it if it works). Can you please see if
> that solves the problem?
Nope, didn't help. I'll poke at it, but am squabbling elsewhere atm.
-Mike
next prev parent reply other threads:[~2010-10-04 3:08 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-09-28 0:29 Nikhil Rao
2010-09-28 0:29 ` [PATCH 1/3] sched: set group_imb only a task can be pulled from the busiest cpu Nikhil Rao
2010-09-28 0:29 ` [PATCH 2/3] sched: drop group_capacity to 1 only if remote group has no running tasks Nikhil Rao
2010-09-28 23:04 ` Suresh Siddha
2010-10-11 21:20 ` Nikhil Rao
2010-09-28 0:29 ` [PATCH 3/3] sched: do not consider SCHED_IDLE tasks to be cache hot Nikhil Rao
2010-09-28 13:57 ` [PATCH 0/3][RFC] Improve load balancing when tasks have large weight differential Mike Galbraith
2010-09-28 21:15 ` Nikhil Rao
2010-09-29 1:45 ` Mike Galbraith
2010-09-29 19:32 ` Nikhil Rao
2010-10-04 3:08 ` Mike Galbraith [this message]
2010-10-06 8:23 ` Nikhil Rao
2010-10-08 7:22 ` Mike Galbraith
2010-10-08 20:34 ` Nikhil Rao
2010-10-10 10:15 ` Mike Galbraith
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=1286161717.7410.12.camel@marge.simson.net \
--to=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=ncrao@google.com \
--cc=peterz@infradead.org \
--cc=venki@google.com \
/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
Powered by JetHome