When ran from the sched-out path (preempt_notifier or perf_event),
p->state is irrelevant to determine preemption. You can get preempted
with !task_is_running() just fine.
The right indicator for preemption is if the task is still on the
runqueue in the sched-out path.
Signed-off-by: Peter
On Wed, Jun 02, 2021 at 09:59:07AM -0400, Mathieu Desnoyers wrote:
> - On Jun 2, 2021, at 9:12 AM, Peter Zijlstra pet...@infradead.org wrote:
>
> > When ran from the sched-out path (preempt_notifier or perf_event),
> > p->state is irrelevant to determine preemption. You can get preempted
> >
On Wed, Jun 02, 2021 at 04:10:29PM +0200, Peter Zijlstra wrote:
> On Wed, Jun 02, 2021 at 09:59:07AM -0400, Mathieu Desnoyers wrote:
> > - On Jun 2, 2021, at 9:12 AM, Peter Zijlstra pet...@infradead.org wrote:
> >
> > > When ran from the sched-out path (preempt_notifier or perf_event),
> > >
- On Jun 2, 2021, at 9:12 AM, Peter Zijlstra pet...@infradead.org wrote:
[...]
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -8568,13 +8568,12 @@ static void perf_event_switch(struct tas
> },
> };
>
> - if (!sched_in && task->state == TASK_RUNNING)
> +
- On Jun 2, 2021, at 9:12 AM, Peter Zijlstra pet...@infradead.org wrote:
> When ran from the sched-out path (preempt_notifier or perf_event),
> p->state is irrelevant to determine preemption. You can get preempted
> with !task_is_running() just fine.
>
> The right indicator for preemption is
On Wed, Jun 02, 2021 at 10:15:16AM -0400, Mathieu Desnoyers wrote:
> - On Jun 2, 2021, at 9:12 AM, Peter Zijlstra pet...@infradead.org wrote:
> [...]
> > --- a/kernel/events/core.c
> > +++ b/kernel/events/core.c
> > @@ -8568,13 +8568,12 @@ static void perf_event_switch(struct tas
> >
