From: Steven Rostedt <rostedt@goodmis.org>
To: Mike Galbraith <efault@gmx.de>
Cc: Chris Mason <chris.mason@fusionio.com>,
"Chris L. Mason" <clmason@fusionio.com>,
"linux-rt-users@vger.kernel.org" <linux-rt-users@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: 3.4.4-rt13: btrfs + xfstests 006 = BOOM.. and a bonus rt_mutex deadlock report for absolutely free!
Date: Tue, 17 Jul 2012 00:27:58 -0400 [thread overview]
Message-ID: <1342499278.24525.13.camel@gandalf.stny.rr.com> (raw)
In-Reply-To: <1342498722.7353.37.camel@marge.simpson.net>
On Tue, 2012-07-17 at 06:18 +0200, Mike Galbraith wrote:
>
> > There's that too. But the issue I was talking about is with all trylock
> > loops. As holding an rt-mutex now disables migration, if a high priority
> > process preempts a task that holds the lock, and then the high prio task
> > starts spinning waiting for that lock to release, the lower priority
> > process will never get to run to release it. The cpu_chill() doesn't
> > help.
>
> Hrm. I better go make a testcase, this one definitely wants pounding
> through thick skull.
Actually, I was mistaken. I forgot that we defined 'cpu_chill()' as
msleep(1) on RT, which would keep a deadlock from happening.
It doesn't explain the performance enhancement you get :-/
-- Steve
next prev parent reply other threads:[~2012-07-17 4:27 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-12 5:47 3.4.4-rt13: btrfs + xfstests 006 = BOOM.. and a bonus rt_mutex deadlock report for absolutely free! Mike Galbraith
2012-07-12 8:44 ` Mike Galbraith
2012-07-12 9:53 ` Mike Galbraith
2012-07-12 11:43 ` Thomas Gleixner
2012-07-12 11:57 ` Mike Galbraith
2012-07-12 13:31 ` Thomas Gleixner
2012-07-12 13:37 ` Mike Galbraith
2012-07-12 13:43 ` Thomas Gleixner
2012-07-12 13:48 ` Mike Galbraith
2012-07-12 13:51 ` Mike Galbraith
2012-07-13 6:31 ` Mike Galbraith
2012-07-13 9:52 ` Thomas Gleixner
2012-07-13 10:14 ` Mike Galbraith
2012-07-13 10:26 ` Thomas Gleixner
2012-07-13 10:47 ` Chris Mason
2012-07-13 12:50 ` Mike Galbraith
2012-07-12 11:07 ` Thomas Gleixner
2012-07-12 17:09 ` Chris Mason
2012-07-13 10:04 ` Thomas Gleixner
2012-07-13 12:50 ` Chris Mason
2012-07-13 14:47 ` Thomas Gleixner
2012-07-14 10:14 ` Mike Galbraith
2012-07-15 17:56 ` Chris Mason
2012-07-16 2:02 ` Mike Galbraith
2012-07-16 16:02 ` Steven Rostedt
2012-07-16 16:26 ` Mike Galbraith
2012-07-16 16:35 ` Chris Mason
2012-07-16 16:36 ` Mike Galbraith
2012-07-16 17:03 ` Steven Rostedt
2012-07-17 4:18 ` Mike Galbraith
2012-07-17 4:27 ` Steven Rostedt [this message]
2012-07-17 4:34 ` Steven Rostedt
2012-07-17 4:46 ` Mike Galbraith
2012-07-17 4:44 ` Mike Galbraith
2012-07-17 12:54 ` Mike Galbraith
2012-07-16 10:55 ` Mike Galbraith
2012-07-16 15:43 ` Chris Mason
2012-07-16 16:16 ` Mike Galbraith
2012-07-14 13:38 ` 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=1342499278.24525.13.camel@gandalf.stny.rr.com \
--to=rostedt@goodmis.org \
--cc=chris.mason@fusionio.com \
--cc=clmason@fusionio.com \
--cc=efault@gmx.de \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rt-users@vger.kernel.org \
--cc=tglx@linutronix.de \
/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).