From mboxrd@z Thu Jan  1 00:00:00 1970
From: Mike Galbraith <efault@gmx.de>
Subject: Re: [PATCH PREEMPT RT] rt-mutex: fix deadlock in device mapper
Date: Tue, 21 Nov 2017 10:18:04 +0100
Message-ID: <1511255884.7672.49.camel@gmx.de>
References: <alpine.LRH.2.02.1711131245580.23280@file01.intranet.prod.int.rdu2.redhat.com>
          <20171117145744.t366d2ztxj2qqnco@linutronix.de>
          <1511030230.12841.42.camel@gmx.de>
          <alpine.LRH.2.02.1711201607230.24006@file01.intranet.prod.int.rdu2.redhat.com>
         <1511234430.7672.26.camel@gmx.de>
         <alpine.DEB.2.20.1711210931460.1782@nanos>
Mime-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-15"
Content-Transfer-Encoding: 8BIT
Cc: Mikulas Patocka <mpatocka@redhat.com>,
        Sebastian Siewior <bigeasy@linutronix.de>,
        linux-kernel@vger.kernel.org, Ingo Molnar <mingo@redhat.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        linux-rt-users@vger.kernel.org
To: Thomas Gleixner <tglx@linutronix.de>
Return-path: <linux-rt-users-owner@vger.kernel.org>
Received: from mout.gmx.net ([212.227.15.15]:54947 "EHLO mout.gmx.net"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1751223AbdKUJSW (ORCPT <rfc822;linux-rt-users@vger.kernel.org>);
        Tue, 21 Nov 2017 04:18:22 -0500
In-Reply-To: <alpine.DEB.2.20.1711210931460.1782@nanos>
Sender: linux-rt-users-owner@vger.kernel.org
List-ID: <linux-rt-users.vger.kernel.org>

On Tue, 2017-11-21 at 09:37 +0100, Thomas Gleixner wrote:
> On Tue, 21 Nov 2017, Mike Galbraith wrote:
> > On Mon, 2017-11-20 at 16:33 -0500, Mikulas Patocka wrote:
> > > 
> > > Is there some specific scenario where you need to call 
> > > blk_schedule_flush_plug from rt_spin_lock_fastlock?
> > 
> > Excellent question.  What's the difference between not getting IO
> > started because you meet a mutex with an rt_mutex under the hood, and
> > not getting IO started because you meet a spinlock with an rt_mutex
> > under the hood?  If just doing the mutex side puts this thing back to
> > sleep, I'm happy.
> 
> Think about it from the mainline POV.
> 
> The spinlock cannot ever go to schedule and therefore cannot create a
> situation which requires an unplug. The RT substitution of the spinlock
> with a rtmutex based sleeping spinlock should not change that at all.
> 
> A regular mutex/rwsem etc. can and will unplug when the lock is contended
> and the caller blocks. The RT conversion of these locks to rtmutex based
> variants creates the problem: Unplug cannot be called when the task has
> pi_blocked_on set because the unplug path might content on yet another
> lock. So unplugging in the slow path before setting pi_blocked_on is the
> right thing to do.

Sure.  What alarms me about IO deadlocks reappearing after all this
time is that at the time I met them, I needed every last bit of that
patchlet I showed to kill them, whether that should have been the case
or not.  'course that tree contained roughly a zillion patches..

Whatever, time will tell if I'm properly alarmed, or merely paranoid :)

	-Mike