From: Con Kolivas <kernel@kolivas.org>
To: Paolo Ornati <ornati@fastwebnet.it>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Ingo Molnar <mingo@elte.hu>,
Nick Piggin <nickpiggin@yahoo.com.au>,
Peter Williams <pwil3058@bigpond.net.au>
Subject: Re: [SCHED] wrong priority calc - SIMPLE test case
Date: Fri, 13 Jan 2006 12:32:01 +1100 [thread overview]
Message-ID: <200601131232.01846.kernel@kolivas.org> (raw)
In-Reply-To: <200601131213.14832.kernel@kolivas.org>
On Friday 13 January 2006 12:13, Con Kolivas wrote:
> On Saturday 31 December 2005 00:52, Paolo Ornati wrote:
> > WAS: [SCHED] Totally WRONG prority calculation with specific test-case
> > (since 2.6.10-bk12)
> > http://lkml.org/lkml/2005/12/27/114/index.html
> >
> > On Wed, 28 Dec 2005 10:26:58 +1100
> >
> > Con Kolivas <kernel@kolivas.org> wrote:
> > > The issue is that the scheduler interactivity estimator is a state
> > > machine and can be fooled to some degree, and a cpu intensive task that
> > > just happens to sleep a little bit gets significantly better priority
> > > than one that is fully cpu bound all the time. Reverting that change is
> > > not a solution because it can still be fooled by the same process
> > > sleeping lots for a few seconds or so at startup and then changing to
> > > the cpu mostly-sleeping slightly behaviour. This "fluctuating"
> > > behaviour is in my opinion worse which is why I removed it.
> >
> > Trying to find a "as simple as possible" test case for this problem
> > (that I consider a BUG in priority calculation) I've come up with this
> > very simple program:
>
> Hi Paolo.
>
> Can you try the following patch on 2.6.15 please? I'm interested in how
> adversely this affects interactive performance as well as whether it helps
> your test case.
I should make it clear. This patch _will_ adversely affect interactivity
because your test case desires that I/O bound tasks get higher priority, and
this patch will do that. This means that I/O bound tasks will be more
noticeable now. The question is how much do we trade off one for the other.
We almost certainly are biased a little too much on the interactive side on
the mainline kernel at the moment.
Cheers,
Con
next prev parent reply other threads:[~2006-01-13 1:32 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-27 18:09 [SCHED] Totally WRONG prority calculation with specific test-case (since 2.6.10-bk12) Paolo Ornati
2005-12-27 21:48 ` Paolo Ornati
2005-12-27 23:26 ` Con Kolivas
2005-12-28 11:01 ` Paolo Ornati
2005-12-28 11:19 ` Con Kolivas
2005-12-28 11:35 ` Paolo Ornati
2005-12-28 17:23 ` Paolo Ornati
2005-12-28 17:39 ` Paolo Ornati
2005-12-30 13:52 ` [SCHED] wrong priority calc - SIMPLE test case Paolo Ornati
2005-12-31 2:06 ` Peter Williams
2005-12-31 10:34 ` Paolo Ornati
2005-12-31 10:52 ` Paolo Ornati
2005-12-31 11:12 ` Con Kolivas
2005-12-31 13:44 ` Peter Williams
2005-12-31 16:31 ` Paolo Ornati
2005-12-31 22:04 ` Peter Williams
2005-12-31 8:13 ` Mike Galbraith
2005-12-31 11:00 ` Paolo Ornati
2005-12-31 15:11 ` Paolo Ornati
2005-12-31 16:37 ` Mike Galbraith
2005-12-31 17:24 ` Paolo Ornati
2005-12-31 17:42 ` Paolo Ornati
2006-01-01 11:39 ` Paolo Ornati
2006-01-02 9:15 ` Mike Galbraith
2006-01-02 9:50 ` Paolo Ornati
2006-01-09 11:11 ` Mike Galbraith
2006-01-09 15:52 ` Mike Galbraith
2006-01-09 16:08 ` Con Kolivas
2006-01-09 18:14 ` Mike Galbraith
2006-01-09 20:00 ` Paolo Ornati
2006-01-09 20:23 ` Paolo Ornati
2006-01-10 7:08 ` Mike Galbraith
2006-01-10 12:07 ` Mike Galbraith
2006-01-10 12:56 ` Paolo Ornati
2006-01-10 13:01 ` Mike Galbraith
2006-01-10 13:53 ` Paolo Ornati
2006-01-10 15:18 ` Mike Galbraith
2006-01-13 1:13 ` Con Kolivas
2006-01-13 1:32 ` Con Kolivas [this message]
2006-01-13 10:46 ` Paolo Ornati
2006-01-13 10:51 ` Con Kolivas
2006-01-13 13:01 ` Mike Galbraith
2006-01-13 14:34 ` Con Kolivas
2006-01-13 16:15 ` Mike Galbraith
2006-01-14 2:05 ` Con Kolivas
2006-01-14 2:56 ` Mike Galbraith
2005-12-27 23:59 ` [SCHED] Totally WRONG prority calculation with specific test-case (since 2.6.10-bk12) Peter Williams
2005-12-28 10:20 ` Paolo Ornati
2005-12-28 13:38 ` Peter Williams
2005-12-28 19:45 ` Paolo Ornati
2005-12-29 3:13 ` Nick Piggin
2005-12-29 3:35 ` Peter Williams
2005-12-29 8:11 ` Nick Piggin
2006-01-27 16:57 [SCHED] wrong priority calc - SIMPLE test case Con Kolivas
2006-01-27 20:06 ` MIke Galbraith
2006-01-27 23:18 ` Con Kolivas
2006-01-28 0:01 ` Peter Williams
2006-01-28 3:43 ` 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=200601131232.01846.kernel@kolivas.org \
--to=kernel@kolivas.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=nickpiggin@yahoo.com.au \
--cc=ornati@fastwebnet.it \
--cc=pwil3058@bigpond.net.au \
/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).