diff options
| author | Srivatsa Vaddagiri <vatsa@codeaurora.org> | 2014-11-28 16:07:19 +0530 |
|---|---|---|
| committer | David Keitel <dkeitel@codeaurora.org> | 2016-03-23 20:01:04 -0700 |
| commit | dc66ef50f5572f811e068a78ef99c92782fbfc0d (patch) | |
| tree | f90f0d11b6f59066ea2a35df38f1e646093ec4b1 | |
| parent | 8e3aa6790ca4ff4744e30720e97c458375a35237 (diff) | |
sched: Limit LBF_PWR_ACTIVE_BALANCE to within cluster
When higher power (performance) cluster has only one online cpu, we
currently let an idle cpu in lower power cluster pull a running task
from performance cluster via active balance. Active balance for
power-aware reasons is supposed to be restricted to balance within
cluster, the check for which is not correctly implemented.
Change-Id: I5fba7f01ad80c082a9b27e89b7f6b17a6d9cde14
Signed-off-by: Srivatsa Vaddagiri <vatsa@codeaurora.org>
| -rw-r--r-- | kernel/sched/fair.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 1238a6825e7f..3614bb03cab5 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7660,8 +7660,8 @@ group_type group_classify(struct sched_group *group, /* Mark a less power-efficient CPU as busy only if we haven't * seen a busy group yet. We want to prioritize spreading * work over power optimization. */ - if (group->group_weight == 1 && sgs->sum_nr_running && - (env->idle != CPU_NOT_IDLE) && + if ((capacity(env->dst_rq) == group_rq_capacity(group)) && + sgs->sum_nr_running && (env->idle != CPU_NOT_IDLE) && power_cost_at_freq(env->dst_cpu, 0) < power_cost_at_freq(cpumask_first(sched_group_cpus(group)), 0)) { env->flags |= LBF_PWR_ACTIVE_BALANCE; |