From: Tejun Heo <tj@kernel.org>
To: Andrea Righi <arighi@nvidia.com>
Cc: David Vernet <void@manifault.com>,
Changwoo Min <changwoo@igalia.com>,
Emil Tsalapatis <emil@etsalapatis.com>,
Daniel Hodges <hodgesd@meta.com>,
sched-ext@lists.linux.dev, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/2] sched_ext: Fix ops.dequeue() semantics
Date: Sun, 28 Dec 2025 13:38:01 -1000 [thread overview]
Message-ID: <aVG_WZCq5QlRgrqK@slm.duckdns.org> (raw)
In-Reply-To: <aVG9BBWIQLL8uJrQ@slm.duckdns.org>
Hello again, again.
On Sun, Dec 28, 2025 at 01:28:04PM -1000, Tejun Heo wrote:
...
> So, please ignore that part. That's non-sense. I still wonder whether we can
> create some interlocking between scx_bpf_dsq_insert() and ops.dequeue()
> without making hot path slower. I'll think more about it.
And we can't create an interlocking between scx_bpf_dsq_insert() and
ops.dequeue() without adding extra atomic operations in hot paths. The only
thing shared is task rq lock and dispatch path can't do that synchronously.
So, yeah, it looks like the best we can do is always letting the BPF sched
know and let it figure out locking and whether the task needs to be
dequeued from BPF side.
Thanks.
--
tejun
next prev parent reply other threads:[~2025-12-28 23:38 UTC|newest]
Thread overview: 83+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-19 22:43 [PATCH 0/2] sched_ext: Implement proper " Andrea Righi
2025-12-19 22:43 ` [PATCH 1/2] sched_ext: Fix " Andrea Righi
2025-12-28 3:20 ` Emil Tsalapatis
2025-12-29 16:36 ` Andrea Righi
2025-12-29 18:35 ` Emil Tsalapatis
2025-12-28 17:19 ` Tejun Heo
2025-12-28 23:28 ` Tejun Heo
2025-12-28 23:38 ` Tejun Heo [this message]
2025-12-29 17:07 ` Andrea Righi
2025-12-29 18:55 ` Emil Tsalapatis
2025-12-28 23:42 ` Tejun Heo
2025-12-29 17:17 ` Andrea Righi
2025-12-29 0:06 ` Tejun Heo
2025-12-29 18:56 ` Andrea Righi
2025-12-19 22:43 ` [PATCH 2/2] selftests/sched_ext: Add test to validate ops.dequeue() Andrea Righi
2025-12-28 3:28 ` Emil Tsalapatis
2025-12-29 16:11 ` Andrea Righi
2026-01-21 12:25 [PATCHSET v2 sched_ext/for-6.20] sched_ext: Fix ops.dequeue() semantics Andrea Righi
2026-01-21 12:25 ` [PATCH 1/2] " Andrea Righi
2026-01-21 12:54 ` Christian Loehle
2026-01-21 12:57 ` Andrea Righi
2026-01-22 9:28 ` Kuba Piecuch
2026-01-23 13:32 ` Andrea Righi
2026-01-26 8:41 [PATCHSET v3 sched_ext/for-6.20] " Andrea Righi
2026-01-26 8:41 ` [PATCH 1/2] " Andrea Righi
2026-01-27 16:38 ` Emil Tsalapatis
2026-01-27 16:41 ` Kuba Piecuch
2026-01-30 7:34 ` Andrea Righi
2026-01-30 13:14 ` Kuba Piecuch
2026-01-31 6:54 ` Andrea Righi
2026-01-31 16:45 ` Kuba Piecuch
2026-01-31 17:24 ` Andrea Righi
2026-01-28 21:21 ` Tejun Heo
2026-01-30 11:54 ` Kuba Piecuch
2026-01-31 9:02 ` Andrea Righi
2026-01-31 17:53 ` Kuba Piecuch
2026-01-31 20:26 ` Andrea Righi
2026-02-02 15:19 ` Tejun Heo
2026-02-02 15:30 ` Andrea Righi
2026-02-01 17:43 ` Tejun Heo
2026-02-02 15:52 ` Andrea Righi
2026-02-02 16:23 ` Kuba Piecuch
2026-02-01 9:08 [PATCHSET v4 sched_ext/for-6.20] " Andrea Righi
2026-02-01 9:08 ` [PATCH 1/2] " Andrea Righi
2026-02-01 22:47 ` Christian Loehle
2026-02-02 7:45 ` Andrea Righi
2026-02-02 9:26 ` Andrea Righi
2026-02-02 10:02 ` Christian Loehle
2026-02-02 15:32 ` Andrea Righi
2026-02-02 10:09 ` Christian Loehle
2026-02-02 13:59 ` Kuba Piecuch
2026-02-04 9:36 ` Andrea Righi
2026-02-04 9:51 ` Kuba Piecuch
2026-02-02 11:56 ` Kuba Piecuch
2026-02-04 10:11 ` Andrea Righi
2026-02-04 10:33 ` Kuba Piecuch
2026-02-04 16:05 [PATCHSET v5] " Andrea Righi
2026-02-04 16:05 ` [PATCH 1/2] " Andrea Righi
2026-02-04 22:14 ` Tejun Heo
2026-02-05 9:26 ` Andrea Righi
2026-02-05 15:32 [PATCHSET v6] " Andrea Righi
2026-02-05 15:32 ` [PATCH 1/2] " Andrea Righi
2026-02-05 19:29 ` Kuba Piecuch
2026-02-05 21:32 ` Andrea Righi
2026-02-06 13:54 [PATCHSET v7] " Andrea Righi
2026-02-06 13:54 ` [PATCH 1/2] " Andrea Righi
2026-02-06 20:35 ` Emil Tsalapatis
2026-02-07 9:26 ` Andrea Righi
2026-02-09 17:28 ` Tejun Heo
2026-02-09 19:06 ` Andrea Righi
2026-02-10 21:26 [PATCHSET v8] " Andrea Righi
2026-02-10 21:26 ` [PATCH 1/2] " Andrea Righi
2026-02-10 23:20 ` Tejun Heo
2026-02-11 16:06 ` Andrea Righi
2026-02-11 19:47 ` Tejun Heo
2026-02-11 22:34 ` Andrea Righi
2026-02-11 22:37 ` Tejun Heo
2026-02-11 22:48 ` Andrea Righi
2026-02-12 10:16 ` Andrea Righi
2026-02-12 14:32 ` Christian Loehle
2026-02-12 15:45 ` Andrea Righi
2026-02-12 17:07 ` Tejun Heo
2026-02-12 18:14 ` Andrea Righi
2026-02-12 18:35 ` Tejun Heo
2026-02-12 22:30 ` Andrea Righi
2026-02-14 10:16 ` Andrea Righi
2026-02-14 17:56 ` Tejun Heo
2026-02-14 19:32 ` Andrea Righi
2026-02-10 23:54 ` Tejun Heo
2026-02-11 16:07 ` 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=aVG_WZCq5QlRgrqK@slm.duckdns.org \
--to=tj@kernel.org \
--cc=arighi@nvidia.com \
--cc=changwoo@igalia.com \
--cc=emil@etsalapatis.com \
--cc=hodgesd@meta.com \
--cc=linux-kernel@vger.kernel.org \
--cc=sched-ext@lists.linux.dev \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome