All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kuba Piecuch <jpiecuch@google.com>
To: Kuba Piecuch <jpiecuch@google.com>,
	Andrea Righi <arighi@nvidia.com>, Tejun Heo <tj@kernel.org>,
	 David Vernet <void@manifault.com>,
	Changwoo Min <changwoo@igalia.com>
Cc: Emil Tsalapatis <emil@etsalapatis.com>,
	Christian Loehle <christian.loehle@arm.com>,
	 Daniel Hodges <hodgesd@meta.com>, <sched-ext@lists.linux.dev>,
	 <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 sched_ext/for-7.1] sched_ext: Invalidate dispatch decisions on CPU affinity changes
Date: Thu, 19 Mar 2026 13:54:32 +0000	[thread overview]
Message-ID: <DH6T5U5G4AZ1.38Q2K35KN0ET5@google.com> (raw)
In-Reply-To: <DH6OUDJUQNA3.6L4YXJMME4KI@google.com>

On Thu Mar 19, 2026 at 10:31 AM UTC, Kuba Piecuch wrote:
>> @@ -2537,9 +2546,26 @@ static void dispatch_to_local_dsq(struct scx_sched *sch, struct rq *rq,
>>  	}
>>  
>>  	if (src_rq != dst_rq &&
>> -	    unlikely(!task_can_run_on_remote_rq(sch, p, dst_rq, true))) {
>> -		dispatch_enqueue(sch, rq, find_global_dsq(sch, task_cpu(p)), p,
>> -				 enq_flags | SCX_ENQ_CLEAR_OPSS | SCX_ENQ_GDSQ_FALLBACK);
>> +	    unlikely(!task_can_run_on_remote_rq(sch, p, dst_rq, false))) {
>> +		/*
>> +		 * Affinity changed after dispatch decision and the task
>> +		 * can't run anymore on the destination rq.
>> +		 *
>> +		 * Drop the dispatch, the task will be re-enqueued. Set the
>> +		 * task back to QUEUED so dequeue (if waiting) can proceed
>> +		 * using current qseq from the task's rq.
>> +		 */
>> +		if (src_rq != rq) {
>> +			raw_spin_rq_unlock(rq);
>> +			raw_spin_rq_lock(src_rq);
>> +		}
>> +		atomic_long_set_release(&p->scx.ops_state,
>> +			       SCX_OPSS_QUEUED |
>> +			       (src_rq->scx.ops_qseq << SCX_OPSS_QSEQ_SHIFT));
>> +		if (src_rq != rq) {
>> +			raw_spin_rq_unlock(src_rq);
>> +			raw_spin_rq_lock(rq);
>> +		}
>>  		return;
>>  	}
>
> My understanding is that task_can_run_on_remote_rq() can run without src_rq
> locked, so it's possible that @p's cpumask changes after the check, isn't it?
> In that case, I think it's still possible to move the task to the local DSQ
> of a CPU that is outside of its cpumask, triggering a warning in
> move_remote_task_to_local_dsq().

I've looked at the code more carefully and I don't think this is an issue.
It's true that task_can_run_on_remote_rq() can run without src_rq locked, and
it's possible that the cpumask changes after the check, but then the dequeue
preceding the cpumask change must have waited for ops_state to change from
SCX_OPSS_DISPATCHING to SCX_OPSS_NONE and it must have reset holding_cpu to -1,
so thanks to the holding_cpu check later we won't insert the task into the DSQ.

Apologies for the confusion.


  reply	other threads:[~2026-03-19 13:54 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-19  8:35 [PATCH v2 sched_ext/for-7.1] sched_ext: Invalidate dispatch decisions on CPU affinity changes Andrea Righi
2026-03-19 10:31 ` Kuba Piecuch
2026-03-19 13:54   ` Kuba Piecuch [this message]
2026-03-19 21:09   ` Andrea Righi
2026-03-20  9:18     ` Kuba Piecuch
2026-03-23 23:13       ` Tejun Heo
2026-04-22  6:33         ` Cheng-Yang Chou
2026-04-22 11:02           ` Andrea Righi
2026-04-23 13:32           ` Kuba Piecuch
2026-04-26  1:47             ` Cheng-Yang Chou
2026-04-27  9:06               ` Kuba Piecuch
2026-05-01 16:19                 ` Cheng-Yang Chou
2026-05-04  8:00                   ` Kuba Piecuch
2026-05-04 21:24                     ` Tejun Heo
2026-05-04 21:58                       ` Andrea Righi
2026-05-05  8:35                         ` Cheng-Yang Chou
2026-05-05  8:01                       ` Kuba Piecuch
2026-05-05  8:31                         ` Tejun Heo
2026-05-05  9:13                           ` Kuba Piecuch
2026-05-05 15:14                             ` Tejun Heo
2026-05-05 15:58                           ` Cheng-Yang Chou
2026-03-19 15:18 ` Kuba Piecuch
2026-03-19 19:01   ` Andrea Righi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=DH6T5U5G4AZ1.38Q2K35KN0ET5@google.com \
    --to=jpiecuch@google.com \
    --cc=arighi@nvidia.com \
    --cc=changwoo@igalia.com \
    --cc=christian.loehle@arm.com \
    --cc=emil@etsalapatis.com \
    --cc=hodgesd@meta.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sched-ext@lists.linux.dev \
    --cc=tj@kernel.org \
    --cc=void@manifault.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.