On Mon, Jun 22, 2020 at 10:11:22AM +0100, Mel Gorman wrote:
> On Mon, Jun 15, 2020 at 02:56:55PM +0200, Peter Zijlstra wrote:
> > Where the condition:
> >
> > !cpus_share_cache(smp_processor_id(), cpu)
> >
> > already implies 'cpu != smp_processor_id()', because a CPU always
> > shares cache with itself, the secondary condition added in commit:
> >
> > 2ebb17717550 ("sched/core: Offload wakee task activation if it the wakee
> > is descheduling")
> >
> > voids that implication, resulting in attempting to do local wake-ups
> > through the queue mechanism.
> >
> > Fixes: 2ebb17717550 ("sched/core: Offload wakee task activation if it the
> > wakee is descheduling")
> > Reported-by: Paul E. McKenney <[email protected]>
> > Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
> > Tested-by: Paul E. McKenney <[email protected]>
>
> Yep, I mistakenly though this would be covered by the self-wakeup check
> early in try_to_wake_up() but it is not
It is, one should not be able to observe 'p->on_cpu && task_cpu(cpu) ==
smp_processor_id()); I've since found the actual problem, find here:
https://lkml.kernel.org/r/20200620184622.GA19696@paulmck-ThinkPad-P72
I'm currently polishing the changelog a little, and will shortly post a
new version of that.
