From: Ingo Molnar <mingo@elte.hu>
To: Mike Galbraith <efault@gmx.de>
Cc: Brian Rogers <brian@xyzw.org>,
linux-kernel@vger.kernel.org,
Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: [BUG] How to get real-time priority using idle priority
Date: Wed, 14 Jan 2009 06:31:06 +0100 [thread overview]
Message-ID: <20090114053106.GA10410@elte.hu> (raw)
In-Reply-To: <1231910014.610.14.camel@marge.simson.net>
* Mike Galbraith <efault@gmx.de> wrote:
> On Tue, 2009-01-13 at 03:58 +0100, Mike Galbraith wrote:
> > On Mon, 2009-01-12 at 17:05 -0800, Brian Rogers wrote:
>
> > > I'll try Mike's "more complete" patch on top of 2.6.29-rc1 and see what
> > > that does.
> >
> > Don't bother. I just tried a SCHED_IDLE make -j8 and had character
> > repeats while typing. Must be another spot.
>
> Hrmph, what an annoying problem. The below works pretty well, but
> _still_ has latency problems in some circumstances.
>
> The more I look at this, the more I think these guys _really_ want to be
> a separate class. The problem is the incredible rate of min_vruntime
> advancement creating absurdly huge spreads.
>
> Hm, maybe I could advance min_vruntime at nice 0 when these guys are
> running, only advance their vruntime at warp 512, but that seems awfully
> hackish. If they were a separate class, they could use the full nice
> spectrum instead of being merely mega-nice.
A separate class has its own set of problems: starvation, etc.
What about increasing the weight of SCHED_IDLE tasks from 1 to 2 or 3?
That still makes them mega-nice (which is more than enough), but should
make the math a lot less borderline.
Ingo
next prev parent reply other threads:[~2009-01-14 5:31 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-01-11 10:58 [BUG] How to get real-time priority using idle priority Brian Rogers
2009-01-12 5:09 ` Mike Galbraith
2009-01-12 13:03 ` Mike Galbraith
2009-01-12 13:14 ` Ingo Molnar
2009-01-12 15:23 ` Mike Galbraith
2009-01-12 15:24 ` Ingo Molnar
2009-01-13 1:05 ` Brian Rogers
2009-01-13 2:58 ` Mike Galbraith
2009-01-14 5:13 ` Mike Galbraith
2009-01-14 5:31 ` Ingo Molnar [this message]
2009-01-14 6:02 ` Mike Galbraith
2009-01-14 7:35 ` Ingo Molnar
2009-01-15 9:28 ` Mike Galbraith
2009-01-15 10:14 ` Peter Zijlstra
2009-01-15 10:30 ` Mike Galbraith
2009-01-15 11:37 ` Mike Galbraith
2009-01-15 11:41 ` Peter Zijlstra
2009-01-15 12:54 ` Ingo Molnar
2009-01-15 13:05 ` Peter Zijlstra
2009-01-15 13:15 ` Ingo Molnar
2009-01-15 13:16 ` Peter Zijlstra
2009-01-15 12:07 ` Brian Rogers
2009-01-12 20:46 ` [patch take 2] " Mike Galbraith
2009-01-12 20:50 ` 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=20090114053106.GA10410@elte.hu \
--to=mingo@elte.hu \
--cc=a.p.zijlstra@chello.nl \
--cc=brian@xyzw.org \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.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).