[PATCH 04/20 v3] sched/eevdf: Reset lag when waking up on idle cpu

[Vincent Guittot]

When several tasks wake up simultaneously on an idle CPU but their prev
CPUs were not idle, their final vlag will depend on the enqueue ordering
as the first one will lose its lag but not the next ones.

Reset the lag during the enqueue if no fair task has already been picked
and set as the running task.

As a typical example:
- CPU0 is idle
- TA with vlag 0ms and TB with vlag 5ms wake up on CPU0 simultaneously.

Depending which task grabs the lock 1st, the behavior will be different:
 - If TA is enqueued 1st, TB will be enqueued with a positive lag and will
be picked 1st.
 - If TB is enqueued 1st, it will lose its positive vlag and both TA and
TB will have 0 vlag when eevdf picks next task.

Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
---
 kernel/sched/fair.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 945de0215abb..6ce4162a7703 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -967,6 +967,14 @@ void decay_entity_lag(struct cfs_rq *cfs_rq, struct task_struct *p, int flags)
 	if (vlag <= 0)
 		return;
 
+	rq = rq_of(cfs_rq);
+
+	/* You can't claim any lag when waking on idle CPU */
+	if (rq->curr == rq->idle) {
+		se->vlag = 0;
+		return;
+	}
+
 	/* No fast access to prev rq task clock */
 	if (flags & ENQUEUE_MIGRATED)
 		return;
@@ -977,8 +985,6 @@ void decay_entity_lag(struct cfs_rq *cfs_rq, struct task_struct *p, int flags)
 		return;
 	}
 
-	rq = rq_of(cfs_rq);
-
 	/* Compute sleep time */
 	delta_exec = rq_clock_task(rq) - se->exec_start;
 	if (unlikely(delta_exec <= 0))
-- 
2.53.0