Re: [PATCHSET sched_ext/for-7.3-fixes] sched_ext: Fix missing ops.dequeue() on remote local DSQ moves

[Kuba Piecuch <jpiecuch@google.com>]

Hi Tejun,

On Tue, Sep 29, 2026 at 07:13:07AM -1000, Tejun Heo wrote:
> The fix looks good to me. It effectively reverts the enqueue_task_scx()
> half of b75aaea24c9f ("sched_ext: Properly mark SCX-internal migrations
> via sticky_cpu"), which as far as I can see only ever suppressed this
> ops.dequeue(). Andrea, can you confirm?

Thanks for the review. Following Andrea's suggestion, v2 clears
p->scx.sticky_cpu right after it's read, as before b75aaea24c9f, so the
fix is now a straight revert of the enqueue side.

> - dequeue_remote.c isn't built until 3/3, so 1/3 can't be built or run.
>   Can you put the fix first, followed by the test with its Makefile entry?

Sure, will do in v2.

> - 2/3: 7.1.y also needs 18d62044cda7 ("sched_ext: Preserve rq tracking
>   across local DSQ dispatch"). Without it, the nested ops.dequeue() trips
>   lockdep when ops.dispatch() uses scx_bpf_dsq_move() to another CPU's
>   local DSQ. It's tagged for stable too, but maybe note it as a
>   prerequisite?

Thanks, I missed that. v2 lists it as a prerequisite:

> - 2/3: With sub-scheds, scx_resolve_local_dsq() can divert the task to the
>   reject or rescue DSQ, so "inserted into the local DSQ" in the comments
>   isn't always accurate. Maybe "destination DSQ"? The new comment in
>   enqueue_task_scx() could be two lines, and the description could lead
>   with the late SCX_DEQ_CORE_SCHED_EXEC and be a lot shorter.

Agreed on all three, will address these in v2.

> - 1/3: A task can only be picked straight out of custody through
>   sched_core_find(), which only returns tasks with a core cookie. Checking
>   p->core_cookie on SCX_DEQ_CORE_SCHED_EXEC would be exact and would
>   remove core_sched_in_use() and the skip.

That's much better, thanks. v2 reads core_cookie through a CO-RE shadow
struct, so the test still builds and loads without CONFIG_SCHED_CORE,
and core_sched_in_use() and the skip are gone. I also ran the test with
the runner and all its workers sharing a core cookie on an SMT guest.
Legitimate core-sched picks out of custody do happen there, and the test
passes.

> - 1/3: _SC_NPROCESSORS_ONLN ignores affinity. With the runner confined to
>   one CPU, the test fails instead of skipping. sched_getaffinity() and
>   CPU_COUNT()?

Done.

> - 1/3: Nits. If the /proc scan stays, PR_SCHED_CORE_GET writes a u64, so
>   the cookie should be u64. ops.dispatch() pops one entry per call, so a
>   stale one idles the CPU until the next kick. Maybe loop a few times?
>   missed_dequeue_cnt and core_sched_exec_dequeue_cnt aren't printed
>   per-scenario like the other counters.

The /proc scan is gone in v2. ops.dispatch() now pops up to 8 entries until
it finds one that isn't stale. All counters are now reset before eac
scenario, so everything printed is per-scenario.

> - 1/3: The variants, error conditions and core-sched caveat are repeated
>   across the cover, description, file header and comments. Can you say
>   each once? Also, single-line comments are usually lowercase in
>   sched_ext.

Done.

Thanks,
Kuba