mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Mike Galbraith <efault@gmx.de>
To: Con Kolivas <con@kolivas.org>
Cc: "Artem S. Tashkinov" <t.artem@lycos.com>, linux-kernel@vger.kernel.org
Subject: Re: HT (Hyper Threading) aware process scheduling doesn't work as it should
Date: Mon, 31 Oct 2011 12:42:28 +0100	[thread overview]
Message-ID: <1320061348.6389.18.camel@marge.simson.net> (raw)
In-Reply-To: <1479015.f2tEjKr9vR@quad>

On Mon, 2011-10-31 at 21:06 +1100, Con Kolivas wrote:
> On Sun, 30 Oct 2011 07:57:12 PM Artem S. Tashkinov wrote:
> > I've found out that even on Linux 3.0.8 the process scheduler doesn't
> > correctly distributes the load amongst virtual CPUs. E.g. on a 4-core
> > system (8 total virtual CPUs) the process scheduler often run some
> > instances of four different tasks on the same physical CPU.
> 
> > Any thoughts? comments? I think this is quite a serious problem.
> 
> Intense cache locality logic, power saving concepts, cpu frequency governor 
> behaviour and separate runqueues per CPU within the current CPU process 
> scheduler in the current mainline linux kernel will ocasionally do this. Some 
> workloads will be better, while others will be worse. Feel free to try my BFS 
> cpu scheduler if you wish a CPU process scheduler that spreads work more 
> evenly across CPUs. 
> 
> Alas the last version I synced up with will not apply cleanly past about 3.0.6 
> I believe:
> 
> http://ck.kolivas.org/patches/bfs/3.0.0/3.0-sched-bfs-413.patch

Yeah, it handles independent tasks well, but cache misses can be
excruciatingly painful for the others.

Q6600 box, configs as identical as possible, tbench 8

3.0.6-bfs413       728.6 MB/sec
3.0.8             1146.7 MB/sec

	-Mike


  reply	other threads:[~2011-10-31 11:42 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-10-30 19:57 Artem S. Tashkinov
2011-10-30 21:26 ` Henrique de Moraes Holschuh
2011-10-30 21:51   ` Artem S. Tashkinov
2011-10-31  9:16     ` Henrique de Moraes Holschuh
2011-10-31  9:40       ` Artem S. Tashkinov
2011-10-31 11:58         ` Henrique de Moraes Holschuh
2011-11-01  4:14           ` Zhu Yanhai
2011-11-01  5:15         ` ffab ffa
2011-10-31 18:59   ` Chris Friesen
2011-11-01  6:01     ` Mike Galbraith
2011-10-30 22:12 ` Arjan van de Ven
2011-10-30 22:29   ` Artem S. Tashkinov
2011-10-31  3:19     ` Yong Zhang
2011-10-31  8:18       ` Artem S. Tashkinov
2011-10-31 10:06 ` Con Kolivas
2011-10-31 11:42   ` Mike Galbraith [this message]
2011-11-01  0:41     ` Con Kolivas
2011-11-01  0:58       ` Gene Heskett
2011-11-01  5:08       ` Mike Galbraith
2011-11-03  8:18 ` Ingo Molnar
2011-11-03  9:44   ` Artem S. Tashkinov
2011-11-03 10:29     ` Ingo Molnar
2011-11-03 12:42     ` Henrique de Moraes Holschuh
2011-11-03 13:06       ` Artem S. Tashkinov
2011-11-03 13:00   ` 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=1320061348.6389.18.camel@marge.simson.net \
    --to=efault@gmx.de \
    --cc=con@kolivas.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=t.artem@lycos.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