From: Jason Xing <kerneljasonx...@linux.alibaba.com> Only when calling the poll syscall the first time can user receive POLLPRI correctly. After that, user always fails to acquire the event signal.
Reproduce case: 1. Get the monitor code in Documentation/accounting/psi.txt 2. Run it, and wait for the event triggered. 3. Kill and restart the process. The question is why we can end up with poll_scheduled = 1 but the work not running (which would reset it to 0). And the answer is because the scheduling side sees group->poll_kworker under RCU protection and then schedules it, but here we cancel the work and destroy the worker. The cancel needs to pair with resetting the poll_scheduled flag. Signed-off-by: Jason Xing <kerneljasonx...@linux.alibaba.com> Reviewed-by: Caspar Zhang <cas...@linux.alibaba.com> Reviewed-by: Suren Baghdasaryan <sur...@google.com> Acked-by: Johannes Weiner <han...@cmpxchg.org> Signed-off-by: Joseph Qi <joseph...@linux.alibaba.com> --- v3: Change the description as Johannes Weiner suggested. kernel/sched/psi.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index 23fbbcc..6e52b67 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -1131,7 +1131,15 @@ static void psi_trigger_destroy(struct kref *ref) * deadlock while waiting for psi_poll_work to acquire trigger_lock */ if (kworker_to_destroy) { + /* + * After the RCU grace period has expired, the worker + * can no longer be found through group->poll_kworker. + * But it might have been already scheduled before + * that - deschedule it cleanly before destroying it. + */ kthread_cancel_delayed_work_sync(&group->poll_work); + atomic_set(&group->poll_scheduled, 0); + kthread_destroy_worker(kworker_to_destroy); } kfree(t); -- 1.8.3.1