lkml.org 
[lkml]   [2017]   [Nov]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: [PATCH PREEMPT RT] rt-mutex: fix deadlock in device mapper


    On Tue, 21 Nov 2017, Mike Galbraith wrote:

    > 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

    So, drop the spinlock unplugging and leave only mutex unplugging,
    reproduce the deadlock and send the stacktraces.

    Mikulas
    \
     
     \ /
      Last update: 2017-11-21 17:13    [W:3.214 / U:0.848 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site