lkml.org 
[lkml]   [2015]   [Mar]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [patch-3.18.7-rt1] sched/context-tracking: fix PREEMPT_LAZY explosions
From
Date
On Mon, 2015-03-09 at 15:36 +0100, Mike Galbraith wrote:
> On Mon, 2015-03-09 at 14:45 +0100, Sebastian Andrzej Siewior wrote:
> > * Mike Galbraith | 2015-02-18 12:21:54 [+0100]:
> >
> > >On Mon, 2015-02-16 at 12:18 +0100, Sebastian Andrzej Siewior wrote:
> > >
> > >> Known issues:
> > >>
> > >> - lazy preempt on x86_64 leads to a crash with some load.
> > >
> > >The below still works for me. (it doesn't make nohz_full actually work
> > >in rt, but at least folks who want to tinker with it can do so)
> > So your box still crashes without it?
> >
> > The thing is I tried to reproduce it and it does not occur anymore.
> > Neither in KVM nor on real HW.
>
> Heh, well that's interesting, I just presumed it would still explode
> because you said it would. I'll beat on it any let you know if I'm
> carrying a placebo patch or not :)

Yup, someone made kaboom go away. I was able to easily blow my box out
of the water by running tbench + kbuild with nohz_full active. No more.

nohz_full doesn't work though, due to an otherwise solo task trying to
shut the tick down having just awakened ksoftirqd.. but no more kaboom.

-Mike



\
 
 \ /
  Last update: 2015-03-10 10:41    [W:1.517 / U:0.596 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site