sched: fix sched_balance_self() smp group balancing

Finding the least idle cpu is more accurate when done with updated shares.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/kernel/sched.c b/kernel/sched.c
index cdd09462fc98e8305e07d7422f64939afe2012cc..39d5495540d25cb74f261c93a17e5b62b90bca99 100644 (file)
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2128,6 +2128,9 @@ static int sched_balance_self(int cpu, int flag)
                        sd = tmp;
        }
 
+       if (sd)
+               update_shares(sd);
+
        while (sd) {
                cpumask_t span, tmpmask;
                struct sched_group *group;