sched: Resched proper CPU on yield_to()

yield_to_task_fair() has code to resched the CPU of yielding task when the
intention is to resched the CPU of the task that is being yielded to.

Change here fixes the problem and also makes the resched conditional on
rq != p_rq.

Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1299025701-22168-1-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/kernel/sched.c b/kernel/sched.c
index f3030709d82660f0fbe803adc31881c1ef5646b0..61452e86c73b2b074bf43c0a0c9bdd1d4a0d8acc 100644 (file)
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5522,8 +5522,15 @@ again:
                goto out;
 
        yielded = curr->sched_class->yield_to_task(rq, p, preempt);
-       if (yielded)
+       if (yielded) {
                schedstat_inc(rq, yld_count);
+               /*
+                * Make p's CPU reschedule; pick_next_entity takes care of
+                * fairness.
+                */
+               if (preempt && rq != p_rq)
+                       resched_task(p_rq->curr);
+       }
 
 out:
        double_rq_unlock(rq, p_rq);

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 1438e13cf8bef024ab476a9592e37eef3ba996e8..3f7ec9e27ee1a259f223549211ab9fe6ec75c3fa 100644 (file)
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1987,10 +1987,6 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p, bool preemp
        /* Tell the scheduler that we'd really like pse to run next. */
        set_next_buddy(se);
 
-       /* Make p's CPU reschedule; pick_next_entity takes care of fairness. */
-       if (preempt)
-               resched_task(rq->curr);
-
        yield_task_fair(rq);
 
        return true;