Re: [PATCH RT v2] Fix a lockup in wait_for_completion() and friends
From: Sebastian Andrzej Siewior
Date: Tue Jul 02 2019 - 07:53:33 EST
On 2019-07-02 06:40:28 [-0500], Corey Minyard wrote:
> On Tue, Jul 02, 2019 at 10:35:36AM +0200, Sebastian Andrzej Siewior wrote:
> > On 2019-07-02 09:04:18 [+0200], Kurt Kanzenbach wrote:
> > > > In fact, my system doesn't boot with this commit in 5.0-rt.
> > > >
> > > > If I revert 90e1b18eba2ae4a729 ("swait: Delete the task from after a
> > > > wakeup occured") the machine boots again.
> > > >
> > > > Sebastian, I think that's a bad commit, please revert it.
> > >
> > > I'm having the same problem on a Cyclone V based ARM board. Reverting
> > > this commit solves the boot issue for me as well.
> > Okay. So the original Corey fix as in v5.0.14-rt9 works for everyone.
> > Peter's version as I picked it up for v5.0.21-rt14 is causing problems
> > for two persons now.
> > I'm leaning towards reverting it back to old version for nowâ
> Just to avoid confusion... it wasn't my patch 1921ea799b7dc56
> (sched/completion: Fix a lockup in wait_for_completion()) that caused
> the issue, nor was it Peter's version of it. Instead, it was the patch
> mentioned above, 90e1b18eba2ae4a729 ("swait: Delete the task from after a
> wakeup occured"), which came from someone else. I can verify by visual
> inspection that that patch is broken and it should definitely be removed.
> Just don't want someone to be confused and remove the wrong patch.
The commit 90e1b18eba2ae4a729 is delta of reverting your patch and
adding Peter's patch instead. If you look into the queue
is what Peter suggested in the thread and this
is the removal.