When p is allowed on none of the CPUs in the sched_domain, we
currently return NULL from find_idlest_group, and pointlessly
continue the search on lower sched_domain levels (where p is also not
allowed) before returning prev_cpu regardless (as we have not updated
new_cpu).
Add an explicit check for this case, and a comment to
find_idlest_group. Now when find_idlest_group returns NULL, it always
means that the local group is allowed and idlest.
Signed-off-by: Brendan Jackman <brendan.jack...@arm.com>
Cc: Dietmar Eggemann <dietmar.eggem...@arm.com>
Cc: Vincent Guittot <vincent.guit...@linaro.org>
Cc: Josef Bacik <jo...@toxicpanda.com>
Cc: Ingo Molnar <mi...@redhat.com>
Cc: Morten Rasmussen <morten.rasmus...@arm.com>
Cc: Peter Zijlstra <pet...@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guit...@linaro.org>
Reviewed-by: Josef Bacik <jba...@fb.com>
---
kernel/sched/fair.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 933f5c9d95d1..4a14ebca4d79 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5768,6 +5768,8 @@ static unsigned long capacity_spare_wake(int cpu, struct
task_struct *p)
/*
* find_idlest_group finds and returns the least busy CPU group within the
* domain.
+ *
+ * Assumes p is allowed on at least one CPU in sd.
*/
static struct sched_group *
find_idlest_group(struct sched_domain *sd, struct task_struct *p,
@@ -5955,6 +5957,9 @@ static inline int find_idlest_cpu(struct sched_domain
*sd, struct task_struct *p
{
int new_cpu = prev_cpu;
+ if (!cpumask_intersects(sched_domain_span(sd), &p->cpus_allowed))
+ return prev_cpu;
+
while (sd) {
struct sched_group *group;
struct sched_domain *tmp;
--
2.14.1
