sched: Optimize cgroup vs wakeup a bit
authorPeter Zijlstra <a.p.zijlstra@chello.nl>
Wed, 16 Sep 2009 11:44:33 +0000 (13:44 +0200)
committerIngo Molnar <mingo@elte.hu>
Wed, 16 Sep 2009 14:44:32 +0000 (16:44 +0200)
commit3b6408942206f940dd538e980e9904e48f4b64f8
tree7e0a0dc37d07e0308ef2294a5de1cd1ea42a388c
parent7c423e98856df9b941223a7e7845b2502ad84b00
sched: Optimize cgroup vs wakeup a bit

We don't need to call update_shares() for each domain we iterate,
just got the largets one.

However, we should call it before wake_affine() as well, so that
that can use up-to-date values too.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/sched.c
kernel/sched_fair.c
kernel/sched_features.h