From: Peter Zijlstra Date: Thu, 17 Sep 2009 07:01:20 +0000 (+0200) Subject: sched: Stop buddies from hogging the system X-Git-Tag: v2.6.32-rc1~652^2~1 X-Git-Url: https://openfabrics.org/gitweb/?a=commitdiff_plain;h=de69a80b;p=~shefty%2Frdma-dev.git sched: Stop buddies from hogging the system Clear buddies more agressively. The (theoretical, haven't actually observed any of this) problem is that when we do not select either buddy in pick_next_entity() because they are too far ahead of the left-most task, we do not clear the buddies. This means that as soon as we service the left-most task, these same buddies will be tried again on the next schedule. Now if the left-most task was a pure hog, it wouldn't have done any wakeups and it wouldn't have set buddies of its own. That leads to the old buddies dominating, which would lead to bad latencies. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: Signed-off-by: Ingo Molnar --- diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 3e6f78c6687..ffee827fa22 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -764,10 +764,10 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup) static void __clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) { - if (cfs_rq->last == se) + if (!se || cfs_rq->last == se) cfs_rq->last = NULL; - if (cfs_rq->next == se) + if (!se || cfs_rq->next == se) cfs_rq->next = NULL; } @@ -1646,8 +1646,13 @@ static struct task_struct *pick_next_task_fair(struct rq *rq) /* * If se was a buddy, clear it so that it will have to earn * the favour again. + * + * If se was not a buddy, clear the buddies because neither + * was elegible to run, let them earn it again. + * + * IOW. unconditionally clear buddies. */ - __clear_buddies(cfs_rq, se); + __clear_buddies(cfs_rq, NULL); set_next_entity(cfs_rq, se); cfs_rq = group_cfs_rq(se); } while (cfs_rq);