[PATCH] sched_ext: Fix inconsistent NUMA node lookup in scx_select_cpu_dfl()

[PATCH] sched_ext: Fix inconsistent NUMA node lookup in scx_select_cpu_dfl()

[Cheng-Yang Chou]

In the WAKE_SYNC path of scx_select_cpu_dfl(), waker_node was computed
with cpu_to_node(), while node (for prev_cpu) was computed with
scx_cpu_node_if_enabled(). When scx_builtin_idle_per_node is disabled,
node is NUMA_NO_NODE but waker_node would be the actual NUMA node,
causing two issues:

1. The (waker_node == node) check always fails when SCX_PICK_IDLE_IN_NODE
   is set, preventing the waker CPU optimization from ever triggering.
2. idle_cpumask(waker_node) is called with a real node ID even though
   per-node idle tracking is disabled, resulting in undefined behavior.

Fix by using scx_cpu_node_if_enabled() for waker_node as well, ensuring
both variables are computed consistently.

Fixes: 48849271e6611 ("sched_ext: idle: Per-node idle cpumasks")
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
---
 kernel/sched/ext_idle.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c
index c7e405262697..8436c7df0a56 100644
--- a/kernel/sched/ext_idle.c
+++ b/kernel/sched/ext_idle.c
@@ -543,7 +543,7 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags,
 		 * piled up on it even if there is an idle core elsewhere on
 		 * the system.
 		 */
-		waker_node = cpu_to_node(cpu);
+		waker_node = scx_cpu_node_if_enabled(cpu);
 		if (!(current->flags & PF_EXITING) &&
 		    cpu_rq(cpu)->scx.local_dsq.nr == 0 &&
 		    (!(flags & SCX_PICK_IDLE_IN_NODE) || (waker_node == node)) &&
-- 
2.48.1

Re: [PATCH] sched_ext: Fix inconsistent NUMA node lookup in scx_select_cpu_dfl()

[Andrea Righi]

Hi Cheng-Yang,

On Sat, Mar 21, 2026 at 06:54:58PM +0800, Cheng-Yang Chou wrote:
> In the WAKE_SYNC path of scx_select_cpu_dfl(), waker_node was computed
> with cpu_to_node(), while node (for prev_cpu) was computed with
> scx_cpu_node_if_enabled(). When scx_builtin_idle_per_node is disabled,
> node is NUMA_NO_NODE but waker_node would be the actual NUMA node,
> causing two issues:
> 
> 1. The (waker_node == node) check always fails when SCX_PICK_IDLE_IN_NODE
>    is set, preventing the waker CPU optimization from ever triggering.

When scx_builtin_idle_per_node is disabled, SCX_PICK_IDLE_IN_NODE won't be
set, which means !(flags & SCX_PICK_IDLE_IN_NODE) should be always true,
short-circuiting the ||, and the waker_node == node comparison is never
evaluated.

However, ...

> 2. idle_cpumask(waker_node) is called with a real node ID even though
>    per-node idle tracking is disabled, resulting in undefined behavior.

...this looks like a legit bug. I'm wondering how this fix impacts
performance, will do some testing. Nice catch!

> 
> Fix by using scx_cpu_node_if_enabled() for waker_node as well, ensuring
> both variables are computed consistently.
> 
> Fixes: 48849271e6611 ("sched_ext: idle: Per-node idle cpumasks")
> Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>

Reviewed-by: Andrea Righi <arighi@nvidia.com>

We should also add:
Cc: stable@vger.kernel.org # v6.15+

Thanks!
-Andrea

> ---
>  kernel/sched/ext_idle.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c
> index c7e405262697..8436c7df0a56 100644
> --- a/kernel/sched/ext_idle.c
> +++ b/kernel/sched/ext_idle.c
> @@ -543,7 +543,7 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags,
>  		 * piled up on it even if there is an idle core elsewhere on
>  		 * the system.
>  		 */
> -		waker_node = cpu_to_node(cpu);
> +		waker_node = scx_cpu_node_if_enabled(cpu);
>  		if (!(current->flags & PF_EXITING) &&
>  		    cpu_rq(cpu)->scx.local_dsq.nr == 0 &&
>  		    (!(flags & SCX_PICK_IDLE_IN_NODE) || (waker_node == node)) &&
> -- 
> 2.48.1
> 

Re: [PATCH] sched_ext: Fix inconsistent NUMA node lookup in scx_select_cpu_dfl()

[Tejun Heo]

Hello,

Applied to sched_ext/for-7.0-fixes with the following additions:

  Cc: stable@vger.kernel.org # v6.15+
  Reviewed-by: Andrea Righi <arighi@nvidia.com>

Thanks.

--
tejun

Re: [PATCH] sched_ext: Fix inconsistent NUMA node lookup in scx_select_cpu_dfl()

[Cheng-Yang Chou]

Hi Tejun, Andrea,

On Sat, Mar 21, 2026 at 08:42:13AM -1000, Tejun Heo wrote:
> Hello,
> 
> Applied to sched_ext/for-7.0-fixes with the following additions:
> 
>   Cc: stable@vger.kernel.org # v6.15+
>   Reviewed-by: Andrea Righi <arighi@nvidia.com>
> 
> Thanks.
> 
> --
> tejun

Thanks for reviewing and applying the patch!

Regarding Andrea's feedback on the commit message, thanks for pointinig
out the short-circuiting behavior. Since the patch is already applied to
sched_ext/for-7.0-fixes, how would you prefer to handle the commit message
update?

I have sent a v2 patch in reply to the v1 thread with the corrected
description. Please feel free to drop the current one and apply v2.

Alternatively, if it is easier, the commit message can be amended
directly in your tree.

Sorry for the mix-up.

-- 
Thanks,
Cheng-Yang

[PATCH v2] sched_ext: Fix inconsistent NUMA node lookup in scx_select_cpu_dfl()

[Cheng-Yang Chou]

In the WAKE_SYNC path of scx_select_cpu_dfl(), waker_node was computed
with cpu_to_node(), while node (for prev_cpu) was computed with
scx_cpu_node_if_enabled(). When scx_builtin_idle_per_node is disabled,
idle_cpumask(waker_node) is called with a real node ID even though
per-node idle tracking is disabled, resulting in undefined behavior.

Fix by using scx_cpu_node_if_enabled() for waker_node as well, ensuring
both variables are computed consistently.

Fixes: 48849271e6611 ("sched_ext: idle: Per-node idle cpumasks")
Cc: stable@vger.kernel.org # v6.15+
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
---
Changes in v2:
- Update commit message to drop the incorrect short-circuiting claim
  (Andrea Righi)
- Link to v1:
  https://lore.kernel.org/all/20260321105503.869337-1-yphbchou0911@gmail.com/

 kernel/sched/ext_idle.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c
index c7e405262697..8436c7df0a56 100644
--- a/kernel/sched/ext_idle.c
+++ b/kernel/sched/ext_idle.c
@@ -543,7 +543,7 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags,
 		 * piled up on it even if there is an idle core elsewhere on
 		 * the system.
 		 */
-		waker_node = cpu_to_node(cpu);
+		waker_node = scx_cpu_node_if_enabled(cpu);
 		if (!(current->flags & PF_EXITING) &&
 		    cpu_rq(cpu)->scx.local_dsq.nr == 0 &&
 		    (!(flags & SCX_PICK_IDLE_IN_NODE) || (waker_node == node)) &&
-- 
2.48.1

Re: [PATCH v2] sched_ext: Fix inconsistent NUMA node lookup in scx_select_cpu_dfl()

[Tejun Heo]

Hello,

The commit message on sched_ext/for-7.0-fixes has been amended to match
v2.

Thanks.

--
tejun