Messages in this thread |  | | Date | Mon, 1 May 2017 12:33:09 -0700 | From | "Paul E. McKenney" <> | Subject | Re: [patch] timer: Fix timers_update_migration(), and call it in tmigr_init() |
| |
On Mon, May 01, 2017 at 09:54:10AM +0200, Thomas Gleixner wrote: > On Sun, 30 Apr 2017, Paul E. McKenney wrote: > > > > And sadly, the splats are still there. Please see the following for > > the relevant console output and .config files: > > Right. We are working on fixing the issues. I'll zap it from tip for now.
In case it helps...
I looked over the rest of the rcutorture scenarios, and it looks like the problems occur for NO_HZ_FULL and more than two CPUs (though larger numbers of CPUs might well just be increasing the probability of this happening). The TEST07 scenario (which had the segfaults) also had CPU hotplug enabled, but !NO_HZ_FULL scenarios with CPU hotplug enabled did just fine (or at least had a much lower probability of occurrence).
Thanx, Paul
|  |