Re: WARN_ON_ONCE() in process_one_work()?

[Tejun Heo]

On Mon, May 01, 2017 at 03:42:37PM -0400, Steven Rostedt wrote:
> On Mon, 1 May 2017 14:42:50 -0400
> Tejun Heo <tj@xxxxxxxxxx> wrote:
> 
> > Steven's involved a rescuer too.  One possibility was cpuset being
> > involved somehow and messing up the affinity of the rescuer kthread
> > unexpectedly.  Is cpuset involved in any way?
> 
> I didn't do anything with cpuset. I did take CPUs offline though.
> 
> I'm currently testing changes, but when its done, I'll remove the
> "WARN_ON() comment out" and see if I can reproduce it again.

Were you also using nohz_full?

Thanks.

-- 
tejun