Re: [PATCH v3 2/2] cgroup/cpuset: Skip security check for hotplug induced v1 task migration

[Michal Koutný]

On Mon, Apr 20, 2026 at 01:44:39PM -0400, Waiman Long <longman@xxxxxxxxxx> wrote:
> I believe the scheduler has a fallback mechanism in that particular case,
> but it can be any CPU. So I don't think we should rely on that.

I'd say that if the patch resolved a potential lockup it'd be worth it
but if there's a fallback to _run somewhere_, what's the issue?
(v1 behavior with hotplug is "path dependent" anyway.)

AFAICS the cpuset_hotplug_test.sh is v1-only and not touched in the
last 5 years. BTW I checked some old runs with v5.3-based SLES and there
doesn't seem to be the described issue.

Do you have more details why this should be changed?

Thanks,
Michal

Attachment: signature.asc
Description: PGP signature