[PATCH v4 07/26] sched/deadline: Do not lose a throttled task switched out and back

[Mathieu Desnoyers <mathieu.desnoyers@efficios.com>]

A SCHED_DEADLINE task that is throttled, switched to another policy and
switched back before its 0-lag time can be left throttled with nothing
to replenish it: on no dl_rq and with no timer pending, it never runs
again, not even for a fatal signal, until another switch out of
SCHED_DEADLINE requeues it:

 1. the task exhausts its runtime: dl_throttled is set, and the
    replenishment timer armed;
 2. a switch to SCHED_OTHER arms the inactive timer for the 0-lag time,
    task_non_contending(), the parameters staying until then;
 3. the replenishment timer fires, finds a task out of the class and
    returns, dl_throttled still set;
 4. a switch back before the 0-lag time cancels the inactive timer, and
    enqueue_dl_entity() leaves a throttled task to its replenishment
    timer, which fired already.

The window is the overrun scaled by period/runtime: about 100us for an
overrun of 20us at 20% bandwidth. The replenishment timer reset a task
it found out of the class until commit 209a0cbda7a0 ("sched/deadline:
Improve the tracking of active utilization") moved that reset to the
inactive timer.

Give the task the replenishment the timer owes it, as the timer does for
a throttled task that is not queued: the overrun comes off the periods
that follow, the deadline moves past it, and dl_throttled and dl_yielded
are cleared. Clearing dl_throttled alone, as the first version of this
patch did, left the overrun, the deadline and a yield to throttle the
task again, as the Sashiko AI review of the first posting of this series
pointed out.

The deadline_switch_test selftest added later in this series fails its
throttled_round_trip case 8 times out of 8 on v7.3 and v7.4-rc1, and
passes it 8 of 8 with this change. A cgroup selftest under development
found it, hanging about once in 1350 runs.

Fixes: 209a0cbda7a0 ("sched/deadline: Improve the tracking of active utilization")
Cc: stable@vger.kernel.org
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
---
 kernel/sched/deadline.c | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 00983487bff3..6cf33045d93a 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -1232,10 +1232,23 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
 
 	/*
 	 * The task might have changed its scheduling policy to something
-	 * different than SCHED_DEADLINE (through switched_from_dl()).
+	 * different than SCHED_DEADLINE (through switched_from_dl()). It
+	 * keeps its parameters until its 0-lag time and may come back to
+	 * SCHED_DEADLINE before then, switched_to_dl() cancelling the
+	 * inactive timer: give it the replenishment this timer owes it, as
+	 * the !queued case below does, and leave the enqueue to its return.
+	 * A throttle left behind would never be lifted; an overrun left in
+	 * its runtime would be forgiven by the new period its enqueue
+	 * starts; and a yield left set would throttle it again at once.
 	 */
-	if (!dl_task(p))
+	if (!dl_task(p)) {
+		if (dl_se->dl_throttled && !is_dl_boosted(dl_se)) {
+			sched_clock_tick();
+			update_rq_clock(rq);
+			replenish_dl_entity(dl_se);
+		}
 		goto unlock;
+	}
 
 	/*
 	 * The task might have been boosted by someone else and might be in the
-- 
2.47.3