sched: minor fast-path overhead reduction

Greetings,

103638d added a bit of avoidable overhead to the fast-path.

Use sysctl_sched_min_granularity instead of sched_slice() to restrict buddy wakeups.

Signed-off-by: Mike Galbraith <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 18fd17172eb66bb567ca4bcc47ca6c0cea923462..67084936b6029037448aba1bdacefaf6b381d99c 100644 (file)
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -747,7 +747,7 @@ pick_next(struct cfs_rq *cfs_rq, struct sched_entity *se)
        struct rq *rq = rq_of(cfs_rq);
        u64 pair_slice = rq->clock - cfs_rq->pair_start;
 
-       if (!cfs_rq->next || pair_slice > sched_slice(cfs_rq, cfs_rq->next)) {
+       if (!cfs_rq->next || pair_slice > sysctl_sched_min_granularity) {
                cfs_rq->pair_start = rq->clock;
                return se;
        }