select_idle_sibling() has a special case for tasks woken up by a per-CPU
kthread where the selected CPU is the previous one. For asymmetric CPU
capacity systems, the assumption was that the wakee couldn't have a
bigger utilization during task placement than it used to have during the
last activation. That was not considering uclamp.min which can completely
change between two task activations and as a consequence mandates the
fitness criterion asym_fits_capacity(), even for the exit path described
above.
Fixes: b4c9c9f15649 ("sched/fair: Prefer prev cpu in asymmetric wakeup path")
Signed-off-by: Vincent Donnefort <[email protected]>
Reviewed-by: Valentin Schneider <[email protected]>
Reviewed-by: Dietmar Eggemann <[email protected]>
---
V2 -> V3:
* A more verbose commit message.
V1 -> V2:
* Point to the correct fixed patch.
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 6291876a9d32..b90dc6fd86ca 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6410,7 +6410,8 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
*/
if (is_per_cpu_kthread(current) &&
prev == smp_processor_id() &&
- this_rq()->nr_running <= 1) {
+ this_rq()->nr_running <= 1 &&
+ asym_fits_capacity(task_util, prev)) {
return prev;
}
--
2.25.1
On Mon, Nov 29, 2021 at 05:31:15PM +0000, Vincent Donnefort wrote:
> select_idle_sibling() has a special case for tasks woken up by a per-CPU
> kthread where the selected CPU is the previous one. For asymmetric CPU
> capacity systems, the assumption was that the wakee couldn't have a
> bigger utilization during task placement than it used to have during the
> last activation. That was not considering uclamp.min which can completely
> change between two task activations and as a consequence mandates the
> fitness criterion asym_fits_capacity(), even for the exit path described
> above.
>
> Fixes: b4c9c9f15649 ("sched/fair: Prefer prev cpu in asymmetric wakeup path")
> Signed-off-by: Vincent Donnefort <[email protected]>
> Reviewed-by: Valentin Schneider <[email protected]>
> Reviewed-by: Dietmar Eggemann <[email protected]>
Thanks!
The following commit has been merged into the sched/core branch of tip:
Commit-ID: 014ba44e8184e1acf93e0cbb7089ee847802f8f0
Gitweb: https://git.kernel.org/tip/014ba44e8184e1acf93e0cbb7089ee847802f8f0
Author: Vincent Donnefort <[email protected]>
AuthorDate: Mon, 29 Nov 2021 17:31:15
Committer: Peter Zijlstra <[email protected]>
CommitterDate: Sat, 04 Dec 2021 10:56:21 +01:00
sched/fair: Fix per-CPU kthread and wakee stacking for asym CPU capacity
select_idle_sibling() has a special case for tasks woken up by a per-CPU
kthread where the selected CPU is the previous one. For asymmetric CPU
capacity systems, the assumption was that the wakee couldn't have a
bigger utilization during task placement than it used to have during the
last activation. That was not considering uclamp.min which can completely
change between two task activations and as a consequence mandates the
fitness criterion asym_fits_capacity(), even for the exit path described
above.
Fixes: b4c9c9f15649 ("sched/fair: Prefer prev cpu in asymmetric wakeup path")
Signed-off-by: Vincent Donnefort <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Valentin Schneider <[email protected]>
Reviewed-by: Dietmar Eggemann <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
---
kernel/sched/fair.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 5cd2798..0672218 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6400,7 +6400,8 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
if (is_per_cpu_kthread(current) &&
in_task() &&
prev == smp_processor_id() &&
- this_rq()->nr_running <= 1) {
+ this_rq()->nr_running <= 1 &&
+ asym_fits_capacity(task_util, prev)) {
return prev;
}