Re: [PATCH 1/3] sched_ext: Add ops.sub_cid_sched_updated() to report the sched running on a cid

[Tejun Heo <tj@kernel.org>]

On Mon, Oct 05, 2026 at 06:10:12PM +0000, sashiko-bot@kernel.org wrote:
> Will this cause the final NONE state notification to be silently dropped for
> the last BPF scheduler unloaded?

Intended. When the count drops to zero, the sched being disabled is the only
one with the op, so it is the only possible receiver, and its tasks are
already gone with ops.exit() next. An ancestor with the op keeps the key on
and gets its notification from the loop. What the loop has to do in the last
case is clear every rq->scx.sched, which it does.

Thanks.

--
tejun