linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Mike Galbraith <umgwanakikbuti@gmail.com>
To: Nicholas Mc Guire <der.herr@hofr.at>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	linux-rt-users <linux-rt-users@vger.kernel.org>,
	LKML <linux-kernel@vger.kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	rostedt@goodmis.org, John Kacur <jkacur@redhat.com>
Subject: Re: [ANNOUNCE] 3.14-rt1
Date: Mon, 28 Apr 2014 11:09:46 +0200	[thread overview]
Message-ID: <1398676186.30930.49.camel@marge.simpson.net> (raw)
In-Reply-To: <1398661784.30930.33.camel@marge.simpson.net>

On Mon, 2014-04-28 at 07:09 +0200, Mike Galbraith wrote: 
> Hi Nicholas,
> 
> On Sat, 2014-04-26 at 15:58 +0200, Mike Galbraith wrote: 
> > On Sat, 2014-04-26 at 10:38 +0200, Mike Galbraith wrote: 
> > > On Fri, 2014-04-25 at 09:40 +0200, Mike Galbraith wrote:
> > > 
> > > > Hotplug can still deadlock in rt trees too, and will if you beat it
> > > > hard.
> > > 
> > > Box actually deadlocks like so.
> > 
> > ...
> > 
> > 3.12-rt looks a bit busted migrate_disable/enable() wise.
> > 
> > /me eyeballs 3.10-rt (looks better), confirms 3.10-rt hotplug works,
> > swipes working code, confirms 3.12-rt now works.  Yup, that was it.
> 
> My boxen, including 64 core DL980 that ran hotplug stress for 3 hours
> yesterday with pre-pushdown rwlocks, say the migrate_disable/enable
> pushdown patches are very definitely busted.

migrate_disable-pushd-down-in-atomic_dec_and_spin_lo.patch

bug: migrate_disable() after blocking is too late.

@@ -1028,12 +1028,12 @@ int atomic_dec_and_spin_lock(atomic_t *a
        /* Subtract 1 from counter unless that drops it to 0 (ie. it was 1) */
        if (atomic_add_unless(atomic, -1, 1))
                return 0;
-       migrate_disable();
        rt_spin_lock(lock);
-       if (atomic_dec_and_test(atomic))
+       if (atomic_dec_and_test(atomic)){
+               migrate_disable();
                return 1;
+       }
        rt_spin_unlock(lock);
-       migrate_enable();
        return 0;
 }
 EXPORT_SYMBOL(atomic_dec_and_spin_lock);

read_lock-migrate_disable-pushdown-to-rt_read_lock.patch

bug: ditto.

@@ -244,8 +246,10 @@ void __lockfunc rt_read_lock(rwlock_t *r
        /*
         * recursive read locks succeed when current owns the lock
         */
-       if (rt_mutex_owner(lock) != current)
+       if (rt_mutex_owner(lock) != current) {
                __rt_spin_lock(lock);
+               migrate_disable();
+       }
        rwlock->read_depth++;
 }

Moving that migrate_disable() up will likely fix my hotplug troubles.
I'll find out when I get back from physical torture (therapy) session.

-Mike


  reply	other threads:[~2014-04-28  9:09 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-04-11 18:57 [ANNOUNCE] 3.14-rt1 Sebastian Andrzej Siewior
2014-04-11 19:34 ` Pavel Vasilyev
2014-04-19 14:46 ` Mike Galbraith
2014-04-21  3:31   ` Mike Galbraith
2014-05-02 10:16     ` Sebastian Andrzej Siewior
2014-04-25  7:40   ` Mike Galbraith
2014-04-26  8:38     ` Mike Galbraith
2014-04-26 13:58       ` Mike Galbraith
2014-04-28  5:09         ` Mike Galbraith
2014-04-28  9:09           ` Mike Galbraith [this message]
2014-04-28 14:18             ` Steven Rostedt
2014-04-28 14:37               ` Mike Galbraith
2014-04-28 14:43                 ` Mike Galbraith
2014-04-29  5:21                 ` Mike Galbraith
2014-04-30  0:13                   ` Steven Rostedt
2014-04-30  7:43                     ` Mike Galbraith
2014-04-30 13:06                       ` Mike Galbraith
2014-04-30 13:15                         ` Steven Rostedt
2014-04-30 14:00                           ` Mike Galbraith
2014-04-30 14:19                             ` Steven Rostedt
2014-04-30 14:33                               ` Steven Rostedt
2014-04-30 14:54                                 ` Mike Galbraith
2014-04-30 15:11                                   ` Steven Rostedt
2014-04-30 15:15                                     ` Mike Galbraith
2014-04-30 15:48                                       ` Steven Rostedt
2014-04-30 15:56                                         ` Mike Galbraith
2014-05-01 17:36                                         ` Mike Galbraith
2014-05-01 18:42                                           ` Steven Rostedt
2014-05-02  1:45                                             ` Mike Galbraith
2014-04-30 15:21                                     ` Mike Galbraith
2014-04-30 14:45                               ` Mike Galbraith
2014-05-02 10:09   ` Sebastian Andrzej Siewior
2014-05-02 11:16     ` Mike Galbraith
2014-04-23 10:37 ` Mike Galbraith
2014-04-23 12:19   ` Steven Rostedt
2014-04-24  4:06 ` Mike Galbraith
2014-04-24  7:12   ` Sebastian Andrzej Siewior
2014-04-24  7:26     ` Mike Galbraith
2014-04-26 18:29 ` Fernando Lopez-Lezcano
2014-05-02 11:37   ` Sebastian Andrzej Siewior
2014-05-15 17:51     ` Fernando Lopez-Lezcano
2014-05-15 20:09       ` Fernando Lopez-Lezcano

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=1398676186.30930.49.camel@marge.simpson.net \
    --to=umgwanakikbuti@gmail.com \
    --cc=bigeasy@linutronix.de \
    --cc=der.herr@hofr.at \
    --cc=jkacur@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rt-users@vger.kernel.org \
    --cc=rostedt@goodmis.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).