Re: SCX_ENQ_IMMED potentially leaving dispatched tasks lingering on local DSQs

[Tejun Heo]

Hello,

On Thu, Apr 23, 2026 at 08:03:51PM +0000, Kuba Piecuch wrote:
...
> I don't like the idea of maintaining the separate checks, precisely because
> of how subtle and finnicky they are. I worry that it will be difficult to keep
> all cases covered as the codebase evolves.
> 
> As an optimization, we could skip wakeup_preempt() in local_dsq_post_enq() if
> rq->next_class == &ext_sched_class. That should make the performance impact
> negligible in the common case, WDYT?

Yeah, let's do that. Please send a patch and can you also cc PeterZ?

Thanks.

-- 
tejun