Re: [PATCH v4 sched_ext/for-7.3 27/40] sched_ext: Add the SCX_CAP_ENQ_IMMED cap

On Wed, 08 Jul 2026 21:56:57 +0000, [email protected] wrote:
> - [High] Tasks in a bypassing descendant's runqueue can be permanently
> stranded if their nearest non-bypassing ancestor lacks SCX_CAP_BASE on
> the CPU.

Real limitation and already on the known-limitations list in the cover
letter - there's no fallback yet for tasks a sub-scheduler can't run
anywhere. The plan is a root-reserved fallback slice in a follow-up
series. Not new to this patch.

> - [High] scx_caps_for_enq() unconditionally maps all local DSQ enqueues
> to SCX_CAP_ENQ_IMMED, allowing sub-schedulers to bypass isolation
> guarantees and swamp the CPU with non-IMMED tasks.

Same as the v3 round - at this patch ENQ_IMMED is the sole baseline cap;
the ENQ vs ENQ_IMMED split that gates non-IMMED work lands later in the
series:

  https://lore.kernel.org/r/[email protected]

Thanks.

-- 
tejun