Yielding repeatedly without resetting timeslice and deadline can lead to tasks bouncing to another CPU immediately instead of yielding leading to infinitely bouncing between CPUs. Improve behaviour by positively calling timeslice_expired.

1 files changed, 5 insertions, 4 deletions

diff --git a/kernel/sched/MuQSS.c b/kernel/sched/MuQSS.c
index 7d48f697e564..525505f9212b 100644
--- a/kernel/sched/MuQSS.c
+++ b/kernel/sched/MuQSS.c
@@ -5074,6 +5074,7 @@ SYSCALL_DEFINE0(sched_yield)
 
 	p = current;
 	rq = this_rq_lock();
+	time_slice_expired(p, rq);
 	schedstat_inc(task_rq(p), yld_count);
 
 	/*
@@ -5208,19 +5209,19 @@ again:
 	}
 
 	double_rq_lock(rq, p_rq);
-	if (task_rq(p) != p_rq) {
+	if (unlikely(task_rq(p) != p_rq)) {
 		double_rq_unlock(rq, p_rq);
 		goto again;
 	}
 
 	yielded = 1;
-	if (p->deadline > rq->rq_deadline)
-		p->deadline = rq->rq_deadline;
 	rq_p = rq->curr;
+	if (p->deadline > rq_p->deadline)
+		p->deadline = rq_p->deadline;
 	p->time_slice += rq_p->time_slice;
-	rq_p->time_slice = 0;
 	if (p->time_slice > timeslice())
 		p->time_slice = timeslice();
+	time_slice_expired(rq_p, rq);
 	if (preempt && rq != p_rq)
 		resched_task(p_rq->curr);
 	double_rq_unlock(rq, p_rq);