diff options
author | 2023-02-06 14:06:12 +0000 | |
---|---|---|
committer | 2023-02-11 11:18:10 +0100 | |
commit | 7ea98dfa44917a201e76d4fe96bf61d76e60f524 (patch) | |
tree | 90a38374272d9e4b55482a703ec47f76f65002ef /kernel | |
parent | sched/fair: sanitize vruntime of entity being placed (diff) | |
download | linux-7ea98dfa44917a201e76d4fe96bf61d76e60f524.tar.gz linux-7ea98dfa44917a201e76d4fe96bf61d76e60f524.tar.bz2 linux-7ea98dfa44917a201e76d4fe96bf61d76e60f524.zip |
sched/deadline: Add more reschedule cases to prio_changed_dl()
I've been tracking down an issue on a ~5.17ish kernel where:
CPUx CPUy
<DL task p0 owns an rtmutex M>
<p0 depletes its runtime, gets throttled>
<rq switches to the idle task>
<DL task p1 blocks on M, boost/replenish p0>
<No call to resched_curr() happens here>
[idle task keeps running here until *something*
accidentally sets TIF_NEED_RESCHED]
On that kernel, it is quite easy to trigger using rt-tests's deadline_test
[1] with the test running on isolated CPUs (this reduces the chance of
something unrelated setting TIF_NEED_RESCHED on the idle tasks, making the
issue even more obvious as the hung task detector chimes in).
I haven't been able to reproduce this using a mainline kernel, even if I
revert
2972e3050e35 ("tracing: Make trace_marker{,_raw} stream-like")
which gets rid of the lock involved in the above test, *but* I cannot
convince myself the issue isn't there from looking at the code.
Make prio_changed_dl() issue a reschedule if the current task isn't a
deadline one. While at it, ensure a reschedule is emitted when a
queued-but-not-current task gets boosted with an earlier deadline that
current's.
[1]: https://git.kernel.org/pub/scm/utils/rt-tests/rt-tests.git
Signed-off-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://lore.kernel.org/r/20230206140612.701871-1-vschneid@redhat.com
Diffstat (limited to 'kernel')
-rw-r--r-- | kernel/sched/deadline.c | 42 |
1 files changed, 27 insertions, 15 deletions
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 0d97d54276cc..71b24371a6f7 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -2663,17 +2663,20 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p) static void prio_changed_dl(struct rq *rq, struct task_struct *p, int oldprio) { - if (task_on_rq_queued(p) || task_current(rq, p)) { + if (!task_on_rq_queued(p)) + return; + #ifdef CONFIG_SMP - /* - * This might be too much, but unfortunately - * we don't have the old deadline value, and - * we can't argue if the task is increasing - * or lowering its prio, so... - */ - if (!rq->dl.overloaded) - deadline_queue_pull_task(rq); + /* + * This might be too much, but unfortunately + * we don't have the old deadline value, and + * we can't argue if the task is increasing + * or lowering its prio, so... + */ + if (!rq->dl.overloaded) + deadline_queue_pull_task(rq); + if (task_current(rq, p)) { /* * If we now have a earlier deadline task than p, * then reschedule, provided p is still on this @@ -2681,15 +2684,24 @@ static void prio_changed_dl(struct rq *rq, struct task_struct *p, */ if (dl_time_before(rq->dl.earliest_dl.curr, p->dl.deadline)) resched_curr(rq); -#else + } else { /* - * Again, we don't know if p has a earlier - * or later deadline, so let's blindly set a - * (maybe not needed) rescheduling point. + * Current may not be deadline in case p was throttled but we + * have just replenished it (e.g. rt_mutex_setprio()). + * + * Otherwise, if p was given an earlier deadline, reschedule. */ - resched_curr(rq); -#endif /* CONFIG_SMP */ + if (!dl_task(rq->curr) || + dl_time_before(p->dl.deadline, rq->curr->dl.deadline)) + resched_curr(rq); } +#else + /* + * We don't know if p has a earlier or later deadline, so let's blindly + * set a (maybe not needed) rescheduling point. + */ + resched_curr(rq); +#endif } DEFINE_SCHED_CLASS(dl) = { |