[PATCH v2] sched/fair: Fix wakeup_preempt_fair for not waking up task

[PATCH v2] sched/fair: Fix wakeup_preempt_fair for not waking up task

[Vincent Guittot]

The assumption that p is always enqueued and not delayed, is only true for
wakeup. If p was moved while delayed, pick_next_entity will dequeue it
and the cfs might become empty. Test if there are still queued tasks
before trying again to determine if p could be the next one to be picked.

There are at least 2 cases:

When cfs becomes idle, it tries to pull tasks but if those pulled tasks are
delayed, they will be dequeued when attached to cfs.
  attach_tasks() -> attach_task() -> wakeup_preempt(rq, p, 0);

A misfit task running on cfs A triggers a load balance to be pulled on a
better cpu, the load balance on cfs B starts an active load balance to
pulled the running misfit task. If there is a delayed dequeue task on cfs A
, it can be pulled instead of the previously running misfit task.
  attach_one_task() -> attach_task() -> wakeup_preempt(rq, p, 0);

Fixes: ac8e69e69363 ("sched/fair: Fix wakeup_preempt_fair() vs delayed dequeue")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
---

Change since v1:
- Use cfs_rq->nr_queued instead of WF_TTWU flag to cover all cases instead
  of just wakeup.

I'm preparing patches to cover those useless newly idle and active load
balance task migration.

 kernel/sched/fair.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 728965851842..24087b0f25f6 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9145,9 +9145,10 @@ static void wakeup_preempt_fair(struct rq *rq, struct task_struct *p, int wake_f
 
 	/*
 	 * Because p is enqueued, nse being null can only mean that we
-	 * dequeued a delayed task.
+	 * dequeued a delayed task. If there are still entities queued in
+	 * cfs, check if the next one will be p.
 	 */
-	if (!nse)
+	if (!nse && cfs_rq->nr_queued)
 		goto pick;
 
 	if (sched_feat(RUN_TO_PARITY))
-- 
2.43.0

Re: [PATCH v2] sched/fair: Fix wakeup_preempt_fair for not waking up task

[Peter Zijlstra]

On Sun, May 03, 2026 at 12:45:03PM +0200, Vincent Guittot wrote:
> The assumption that p is always enqueued and not delayed, is only true for
> wakeup. If p was moved while delayed, pick_next_entity will dequeue it
> and the cfs might become empty. Test if there are still queued tasks
> before trying again to determine if p could be the next one to be picked.
> 
> There are at least 2 cases:
> 
> When cfs becomes idle, it tries to pull tasks but if those pulled tasks are
> delayed, they will be dequeued when attached to cfs.
>   attach_tasks() -> attach_task() -> wakeup_preempt(rq, p, 0);
> 
> A misfit task running on cfs A triggers a load balance to be pulled on a
> better cpu, the load balance on cfs B starts an active load balance to
> pulled the running misfit task. If there is a delayed dequeue task on cfs A
> , it can be pulled instead of the previously running misfit task.
>   attach_one_task() -> attach_task() -> wakeup_preempt(rq, p, 0);
> 
> Fixes: ac8e69e69363 ("sched/fair: Fix wakeup_preempt_fair() vs delayed dequeue")
> Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
> ---
> 
> Change since v1:
> - Use cfs_rq->nr_queued instead of WF_TTWU flag to cover all cases instead
>   of just wakeup.
> 
> I'm preparing patches to cover those useless newly idle and active load
> balance task migration.
> 
>  kernel/sched/fair.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 728965851842..24087b0f25f6 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -9145,9 +9145,10 @@ static void wakeup_preempt_fair(struct rq *rq, struct task_struct *p, int wake_f
>  
>  	/*
>  	 * Because p is enqueued, nse being null can only mean that we
> -	 * dequeued a delayed task.
> +	 * dequeued a delayed task. If there are still entities queued in
> +	 * cfs, check if the next one will be p.
>  	 */
> -	if (!nse)
> +	if (!nse && cfs_rq->nr_queued)
>  		goto pick;
>  
>  	if (sched_feat(RUN_TO_PARITY))

Right, that works. Thanks!

[tip: sched/urgent] sched/fair: Fix wakeup_preempt_fair() for not waking up task

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/urgent branch of tip:

Commit-ID:     3ed412adc88d914acf45c5a304366be0b30cfd0a
Gitweb:        https://git.kernel.org/tip/3ed412adc88d914acf45c5a304366be0b30cfd0a
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Sun, 03 May 2026 12:45:03 +02:00
Committer:     Peter Zijlstra <peterz@infradead.org>
CommitterDate: Tue, 05 May 2026 12:44:24 +02:00

sched/fair: Fix wakeup_preempt_fair() for not waking up task

Make sure to only call pick_next_entity() on an non-empty cfs_rq.

The assumption that p is always enqueued and not delayed, is only true for
wakeup. If p was moved while delayed, pick_next_entity() will dequeue it and
the cfs might become empty. Test if there are still queued tasks before trying
again to determine if p could be the next one to be picked.

There are at least 2 cases:

When cfs becomes idle, it tries to pull tasks but if those pulled tasks are
delayed, they will be dequeued when attached to cfs. attach_tasks() ->
attach_task() -> wakeup_preempt(rq, p, 0);

A misfit task running on cfs A triggers a load balance to be pulled on a better
cpu, the load balance on cfs B starts an active load balance to pulled the
running misfit task. If there is a delayed dequeue task on cfs A, it can be
pulled instead of the previously running misfit task. attach_one_task() ->
attach_task() -> wakeup_preempt(rq, p, 0);

Fixes: ac8e69e69363 ("sched/fair: Fix wakeup_preempt_fair() vs delayed dequeue")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260503104503.1732682-1-vincent.guittot@linaro.org
---
 kernel/sched/fair.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index b91c8b2..3ebec18 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9174,9 +9174,10 @@ pick:
 
 	/*
 	 * Because p is enqueued, nse being null can only mean that we
-	 * dequeued a delayed task.
+	 * dequeued a delayed task. If there are still entities queued in
+	 * cfs, check if the next one will be p.
 	 */
-	if (!nse)
+	if (!nse && cfs_rq->nr_queued)
 		goto pick;
 
 	if (sched_feat(RUN_TO_PARITY))

[tip: sched/urgent] sched/fair: Fix wakeup_preempt_fair() for not waking up task

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/urgent branch of tip:

Commit-ID:     0c0cf14be490975a7431e9d49fcd7b190fff337f
Gitweb:        https://git.kernel.org/tip/0c0cf14be490975a7431e9d49fcd7b190fff337f
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Sun, 03 May 2026 12:45:03 +02:00
Committer:     Peter Zijlstra <peterz@infradead.org>
CommitterDate: Tue, 05 May 2026 16:03:12 +02:00

sched/fair: Fix wakeup_preempt_fair() for not waking up task

Make sure to only call pick_next_entity() on an non-empty cfs_rq.

The assumption that p is always enqueued and not delayed, is only true for
wakeup. If p was moved while delayed, pick_next_entity() will dequeue it and
the cfs might become empty. Test if there are still queued tasks before trying
again to determine if p could be the next one to be picked.

There are at least 2 cases:

When cfs becomes idle, it tries to pull tasks but if those pulled tasks are
delayed, they will be dequeued when attached to cfs. attach_tasks() ->
attach_task() -> wakeup_preempt(rq, p, 0);

A misfit task running on cfs A triggers a load balance to be pulled on a better
cpu, the load balance on cfs B starts an active load balance to pulled the
running misfit task. If there is a delayed dequeue task on cfs A, it can be
pulled instead of the previously running misfit task. attach_one_task() ->
attach_task() -> wakeup_preempt(rq, p, 0);

Fixes: ac8e69e69363 ("sched/fair: Fix wakeup_preempt_fair() vs delayed dequeue")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260503104503.1732682-1-vincent.guittot@linaro.org
---
 kernel/sched/fair.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index b91c8b2..3ebec18 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9174,9 +9174,10 @@ pick:
 
 	/*
 	 * Because p is enqueued, nse being null can only mean that we
-	 * dequeued a delayed task.
+	 * dequeued a delayed task. If there are still entities queued in
+	 * cfs, check if the next one will be p.
 	 */
-	if (!nse)
+	if (!nse && cfs_rq->nr_queued)
 		goto pick;
 
 	if (sched_feat(RUN_TO_PARITY))