On 2017-06-22 19:30:07 [+0200], Mike Galbraith wrote:
> On Thu, 2017-06-22 at 18:34 +0200, Sebastian Andrzej Siewior wrote:
> > On 2017-06-20 09:45:06 [+0200], Mike Galbraith wrote:
> > > See ! and ?
> >
> > See see.
> > What about this:
>
> I'll give it a go, likely during the weekend.
It survi
On Thu, 2017-06-22 at 19:30 +0200, Mike Galbraith wrote:
> On Thu, 2017-06-22 at 18:34 +0200, Sebastian Andrzej Siewior wrote:
> > On 2017-06-20 09:45:06 [+0200], Mike Galbraith wrote:
> > > See ! and ?
> >
> > See see.
> > What about this:
>
> I'll give it a go, likely during the weekend.
What
On Thu, 22 Jun 2017, Mike Galbraith wrote:
> On Thu, 2017-06-22 at 18:34 +0200, Sebastian Andrzej Siewior wrote:
> > On 2017-06-20 09:45:06 [+0200], Mike Galbraith wrote:
> > > See ! and ?
> >
> > See see.
> > What about this:
>
> I'll give it a go, likely during the weekend.
>
> I moved 4.11-rt
On Thu, 2017-06-22 at 18:34 +0200, Sebastian Andrzej Siewior wrote:
> On 2017-06-20 09:45:06 [+0200], Mike Galbraith wrote:
> > See ! and ?
>
> See see.
> What about this:
I'll give it a go, likely during the weekend.
I moved 4.11-rt today (also repros nicely) due to ftrace annoying me.
After y
On 2017-06-20 09:45:06 [+0200], Mike Galbraith wrote:
> See ! and ?
See see.
What about this:
diff --git a/include/linux/sched.h b/include/linux/sched.h
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1014,8 +1014,20 @@ struct wake_q_head {
#define WAKE_Q(name)
On Mon, 2017-06-19 at 18:29 +0200, Mike Galbraith wrote:
> On Mon, 2017-06-19 at 10:41 -0400, Steven Rostedt wrote:
> > On Mon, 19 Jun 2017 16:13:41 +0200
> > Sebastian Andrzej Siewior wrote:
> >
> >
> > > > Hmm, it shouldn't affect futexes, as it's only called by rtmutex when
> > > > waiter->sa
On Mon, 2017-06-19 at 18:27 +0200, Sebastian Andrzej Siewior wrote:
> On 2017-06-19 18:14:52 [+0200], Mike Galbraith wrote:
> >
> > BTW back, I reran virgin 4.9-rt21 on desktop box while off doing the
> > have something resembling a life thing, and it did not stall in 50
> > iterations of performa
On Mon, 2017-06-19 at 10:41 -0400, Steven Rostedt wrote:
> On Mon, 19 Jun 2017 16:13:41 +0200
> Sebastian Andrzej Siewior wrote:
>
>
> > > Hmm, it shouldn't affect futexes, as it's only called by rtmutex when
> > > waiter->savestate is true. And that should always be false for futex.
> >
> >
On 2017-06-19 18:14:52 [+0200], Mike Galbraith wrote:
>
> BTW back, I reran virgin 4.9-rt21 on desktop box while off doing the
> have something resembling a life thing, and it did not stall in 50
> iterations of performance/run.sh (bloody fickle thing). Hohum, take it
> for whatever you think it'
On Mon, 2017-06-19 at 17:03 +0200, Sebastian Andrzej Siewior wrote:
> On 2017-06-19 16:36:22 [+0200], Mike Galbraith wrote:
> > On Mon, 2017-06-19 at 16:06 +0200, Sebastian Andrzej Siewior wrote:
> > >
> > > I am suppressed that your desktop shows any symptoms on rt21. I tried my
> > > smaller AMD
On 2017-06-19 16:36:22 [+0200], Mike Galbraith wrote:
> On Mon, 2017-06-19 at 16:06 +0200, Sebastian Andrzej Siewior wrote:
> >
> > I am suppressed that your desktop shows any symptoms on rt21. I tried my
> > smaller AMD box (A10), an Intel two sockets and a four socket box. Each
> > of them was f
On Mon, 19 Jun 2017 16:13:41 +0200
Sebastian Andrzej Siewior wrote:
> > Hmm, it shouldn't affect futexes, as it's only called by rtmutex when
> > waiter->savestate is true. And that should always be false for futex.
>
> you still have sleeping locks like the hb-lock (which might matter in
> c
On Mon, 2017-06-19 at 16:06 +0200, Sebastian Andrzej Siewior wrote:
>
> I am suppressed that your desktop shows any symptoms on rt21. I tried my
> smaller AMD box (A10), an Intel two sockets and a four socket box. Each
> of them was fine with the run.sh and manual futex_wait invocation.
> Could yo
On 2017-06-19 10:08:38 [-0400], Steven Rostedt wrote:
> > diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> > index 30b24f774198..10e832da70b6 100644
> > --- a/kernel/sched/core.c
> > +++ b/kernel/sched/core.c
> > @@ -2284,7 +2284,7 @@ EXPORT_SYMBOL(wake_up_process);
> > */
> > int wake_u
On Mon, 19 Jun 2017 12:14:51 +0200
Mike Galbraith wrote:
> On Mon, 2017-06-19 at 10:52 +0200, Sebastian Andrzej Siewior wrote:
> > On 2017-06-17 10:14:37 [+0200], Mike Galbraith wrote:
> >
> > > During that rebase, migrate_disable() was changed to no longer map to
> > > preempt_disable() for n
On 2017-06-19 14:55:35 [+0200], Mike Galbraith wrote:
> On Mon, 2017-06-19 at 13:50 +0200, Sebastian Andrzej Siewior wrote:
> >
> > rt20…rt21 is just
> >
> > https://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-devel.git/tree/patches/sched-Prevent-task-state-corruption-by-spurious-lock.p
On Mon, 2017-06-19 at 13:50 +0200, Sebastian Andrzej Siewior wrote:
>
> rt20…rt21 is just
>
> https://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-devel.git/tree/patches/sched-Prevent-task-state-corruption-by-spurious-lock.patch?h=linux-4.9.y-rt-patches
Yup. I got there via git checkou
On 2017-06-19 13:31:28 [+0200], Mike Galbraith wrote:
> > > Scratch that "appears", changing it to TASK_NORMAL just fixed my DL980
> > > running otherwise absolutely pristine 4.9-rt21, after having double
> > > verified that rt20 works fine. Now to go back to 4.11/master/tip-rt,
> > > make sure th
On Mon, 2017-06-19 at 12:44 +0200, Sebastian Andrzej Siewior wrote:
> On 2017-06-19 12:14:51 [+0200], Mike Galbraith wrote:
> > Ok, doesn't matter for RT testing. What does matter, is that...
> >
> > diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> > index 30b24f774198..10e832da70b6 10064
On 2017-06-19 12:14:51 [+0200], Mike Galbraith wrote:
> Ok, doesn't matter for RT testing. What does matter, is that...
>
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 30b24f774198..10e832da70b6 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -2284,7 +2284,7
On Mon, 2017-06-19 at 10:52 +0200, Sebastian Andrzej Siewior wrote:
> On 2017-06-17 10:14:37 [+0200], Mike Galbraith wrote:
>
> > During that rebase, migrate_disable() was changed to no longer map to
> > preempt_disable() for nonrt, but some patches still assume it does. It
> > now depends upon P
On 2017-06-17 10:14:37 [+0200], Mike Galbraith wrote:
> Hi Sebastian,
Hi Mike,
>
> I noticed a couple things wrt migrate_disable() changes...
>
> During that rebase, migrate_disable() was changed to no longer map to
> preempt_disable() for nonrt, but some patches still assume it does. It
> now
On Sat, 2017-06-17 at 10:14 +0200, Mike Galbraith wrote:
>
>... the RT workaround in futex.c induces
> grumbling in nonrt builds with PREEMPT_COUNT enabled.
A trivial way to fix it up is to...
futex: Fix migrate_disable/enable workaround for !PREEMPT_RT_FULL
The imbalance fixed by aed0f50e58eb
On Fri, 2017-06-16 at 12:56 +0200, Sebastian Andrzej Siewior wrote:
> Dear RT folks!
>
> I'm pleased to announce the v4.11.5-rt1 patch set.
> The release has been delayed due to the hotplug rework that was started
> before the final v4.11 release. However the new code has not been
> stabilized ye
Dear RT folks!
I'm pleased to announce the v4.11.5-rt1 patch set.
The release has been delayed due to the hotplug rework that was started
before the final v4.11 release. However the new code has not been
stabilized yet and it was decided to bring back the old patches before
delaying the v4.11-RT
25 matches
Mail list logo