From: Andi Kleen <andi@firstfloor.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>,
Andi Kleen <andi@firstfloor.org>,
Thomas Gleixner <tglx@linutronix.de>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
linux-kernel@vger.kernel.org,
Stefani Seibold <stefani@seibold.net>,
Dario Faggioli <raistlin@linux.it>,
Max Krasnyansky <maxk@qualcomm.com>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: [PATCH 6/6] sched: disabled rt-bandwidth by default
Date: Thu, 28 Aug 2008 13:09:23 +0200 [thread overview]
Message-ID: <20080828110923.GL26610@one.firstfloor.org> (raw)
In-Reply-To: <20080828105408.GA4488@elte.hu>
> Even if the system has multiple CPUs, and even if just a single CPU is
> fully utilized by an RT task, without the rt-limit the system will still
> lock up in practice due to various other factors: workqueues and tasks
> being 'stuck' on CPUs that host an RT hog.
The load balancer will not notice that a particular CPU is busy
with real time tasks?
> While there's obviously CPU
> time available on other CPUs, you cannot run 'top', the desktop will
> freeze, work flows of the system can be stuck, etc, etc..
I had such a situation at least once in the past (not due
run away RT but due a kernel bug) and even with 2 out of 4 CPUs blocked
the system was still quite usable. top/kill definitely worked. The system
didn't have a desktop, but I didn't notice many problems in shell use.
Ok it's just one sample.
That said I don't think having such a limit by default is a bad idea actually.
Just handling it in the scheduler anyways is also probably good because
it can happen even due to other issues than just run away RT tasks.
-Andi
next prev parent reply other threads:[~2008-08-28 11:06 UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-08-19 10:33 [PATCH 0/6] sched: rt-bandwidth fixes Peter Zijlstra
2008-08-19 10:33 ` [PATCH 1/6] sched: rt-bandwidth for user grouping interface Peter Zijlstra
2008-08-19 10:33 ` [PATCH 2/6] sched: rt-bandwidth accounting fix Peter Zijlstra
2008-08-19 18:33 ` Max Krasnyansky
2008-08-19 18:38 ` Peter Zijlstra
2008-08-19 10:33 ` [PATCH 3/6] sched: rt-bandwidth group disable fixes Peter Zijlstra
2008-08-19 10:33 ` [PATCH 4/6] sched: extract walk_tg_tree() Peter Zijlstra
2008-08-19 10:33 ` [PATCH 5/6] sched: rt-bandwidth fixes Peter Zijlstra
2008-08-19 10:33 ` [PATCH 6/6] sched: disabled rt-bandwidth by default Peter Zijlstra
2008-08-19 11:05 ` Ingo Molnar
2008-08-19 11:11 ` Ingo Molnar
2008-08-19 11:42 ` [PATCH] sched: extract walk_tg_tree(), fix Ingo Molnar
2008-08-19 11:17 ` [PATCH 6/6] sched: disabled rt-bandwidth by default Nick Piggin
2008-08-19 12:59 ` Ingo Molnar
2008-08-19 18:15 ` Max Krasnyansky
2008-08-20 11:56 ` Nick Piggin
2008-08-26 9:00 ` Nick Piggin
2008-08-26 9:30 ` Ingo Molnar
2008-08-26 9:44 ` Nick Piggin
2008-08-26 10:29 ` Ingo Molnar
2008-08-26 11:03 ` Nick Piggin
2008-08-26 9:54 ` Nick Piggin
2008-08-26 11:09 ` Thomas Gleixner
2008-08-26 11:27 ` Nick Piggin
2008-08-26 12:50 ` Theodore Tso
2008-08-26 13:31 ` Stefani Seibold
2008-08-26 17:55 ` Theodore Tso
2008-08-26 21:37 ` Thomas Gleixner
2008-08-26 22:49 ` Andi Kleen
2008-08-27 10:08 ` Nick Piggin
2008-08-28 10:54 ` Ingo Molnar
2008-08-28 11:09 ` Andi Kleen [this message]
2008-08-28 11:19 ` Peter Zijlstra
2008-08-28 11:28 ` Ingo Molnar
2008-08-28 11:50 ` Andi Kleen
2008-08-28 12:00 ` Peter Zijlstra
2008-08-28 12:14 ` Andi Kleen
2008-08-28 12:18 ` Nick Piggin
2008-08-28 16:19 ` Max Krasnyansky
2008-08-28 16:25 ` Ingo Molnar
2008-08-28 16:33 ` Andi Kleen
2008-08-28 12:03 ` Nick Piggin
2008-08-28 13:07 ` Ingo Molnar
2008-08-28 13:45 ` Nick Piggin
2008-08-28 12:29 ` Nick Piggin
2008-08-27 10:04 ` Nick Piggin
2008-08-26 13:47 ` Mark Hounschell
2008-08-26 23:00 ` Steven Rostedt
2008-08-27 18:55 ` Chris Friesen
2008-08-28 14:15 ` Steven Rostedt
2008-08-28 14:30 ` Ingo Molnar
2008-08-28 14:36 ` Nick Piggin
2008-08-28 15:12 ` Steven Rostedt
2008-08-28 15:34 ` Nick Piggin
2008-08-28 15:50 ` Steven Rostedt
2008-08-28 17:26 ` Linus Torvalds
2008-08-28 18:04 ` Steven Rostedt
2008-08-28 18:10 ` Darren Hart
2008-08-28 18:16 ` Mark Hounschell
2008-08-28 18:42 ` Linus Torvalds
2008-08-28 18:53 ` Steven Rostedt
2008-08-29 7:56 ` Mike Galbraith
2008-08-29 8:06 ` Peter Zijlstra
2008-08-29 8:47 ` Mike Galbraith
2008-08-28 19:39 ` Stefani Seibold
2008-08-28 20:53 ` Alan Cox
2008-08-30 6:33 ` Nick Piggin
2008-08-28 16:33 ` Max Krasnyansky
2008-08-28 17:22 ` John Kacur
2008-08-28 16:05 ` Peter Zijlstra
2008-08-28 16:15 ` Steven Rostedt
2008-08-28 16:29 ` Peter Zijlstra
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=20080828110923.GL26610@one.firstfloor.org \
--to=andi@firstfloor.org \
--cc=a.p.zijlstra@chello.nl \
--cc=linux-kernel@vger.kernel.org \
--cc=maxk@qualcomm.com \
--cc=mingo@elte.hu \
--cc=nickpiggin@yahoo.com.au \
--cc=raistlin@linux.it \
--cc=stefani@seibold.net \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.