From: Andrea Righi <arighi@nvidia.com>
To: Tejun Heo <tj@kernel.org>
Cc: David Vernet <void@manifault.com>,
Changwoo Min <changwoo@igalia.com>,
sched-ext@lists.linux.dev, Emil Tsalapatis <emil@etsalapatis.com>,
Peter Zijlstra <peterz@infradead.org>,
ElXreno <elxreno@gmail.com>,
linux-kernel@vger.kernel.org, stable@vger.kernel.org
Subject: Re: [PATCH 3/6] sched_ext: Replace SCX_RQ_BAL_KEEP with a dispatch verdict return
Date: Tue, 11 Aug 2026 09:43:05 +0200 [thread overview]
Message-ID: <anrSiZkg167gQvgR@gpd4> (raw)
In-Reply-To: <20260807210221.232543-4-tj@kernel.org>
Hi Tejun,
On Fri, Aug 07, 2026 at 11:02:18AM -1000, Tejun Heo wrote:
> SCX_RQ_BAL_KEEP tells the pick to keep running the previous task, a leftover
> from when balancing and picking were separate operations. An rq-level flag
> only works while dispatches and picks pair up one to one, which core
> scheduling breaks: selections interleave through dispatch's lock drops and a
> pick can consume a stale flag, keeping a task that has since been dequeued.
> Fixing core scheduling support requires the decision to travel with the
> dispatch that made it. Make scx_dispatch_sched() and balance_one() return an
> explicit verdict instead.
>
> Also factor the pick-side invocation, its follow-up queueing and the
> post-dispatch checks out of do_pick_task_scx() into dispatch_pick(). No
> functional changes intended.
>
> Fixes: 4c95380701f5 ("sched/ext: Fold balance_scx() into pick_task_scx()")
> Cc: stable@vger.kernel.org # v6.19+
> Signed-off-by: Tejun Heo <tj@kernel.org>
> ---
...
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index 450b6a04669f..14c2df6fd9af 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -784,7 +784,6 @@ enum scx_rq_flags {
> */
> SCX_RQ_ONLINE = 1 << 0,
> SCX_RQ_CAN_STOP_TICK = 1 << 1,
> - SCX_RQ_BAL_KEEP = 1 << 3, /* balance decided to keep current */
> SCX_RQ_CLK_VALID = 1 << 5, /* RQ clock is fresh and valid */
> SCX_RQ_BAL_CB_PENDING = 1 << 6, /* must queue a cb after dispatching */
Should we update also tools/sched_ext/include/scx/*.h to remove SCX_RQ_BAL_KEEP?
Thanks,
-Andrea
next prev parent reply other threads:[~2026-08-11 7:43 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-07 21:02 [PATCHSET sched_ext/for-7.2-fixes] sched_ext: Fix core scheduling Tejun Heo
2026-08-07 21:02 ` [PATCH 1/6] sched/core: Handle pick_task() releasing the rq lock Tejun Heo
2026-08-07 21:26 ` sashiko-bot
2026-08-10 11:00 ` Peter Zijlstra
2026-08-07 21:02 ` [PATCH 2/6] sched/core: Make core-sched flips wait for in-flight selections Tejun Heo
2026-08-10 11:15 ` Peter Zijlstra
2026-08-10 22:10 ` Tejun Heo
2026-08-11 16:05 ` Peter Zijlstra
2026-08-07 21:02 ` [PATCH 3/6] sched_ext: Replace SCX_RQ_BAL_KEEP with a dispatch verdict return Tejun Heo
2026-08-11 7:43 ` Andrea Righi [this message]
2026-08-07 21:02 ` [PATCH 4/6] sched_ext: Fix this_rq() assumptions in dispatch kfuncs Tejun Heo
2026-08-07 21:02 ` [PATCH 5/6] sched_ext: Count rq lock releases in rq->scx.lock_drop_seq Tejun Heo
2026-08-07 21:02 ` [PATCH 6/6] sched_ext: Fix rq->core_pick corruption under core scheduling Tejun Heo
2026-08-07 21:22 ` sashiko-bot
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=anrSiZkg167gQvgR@gpd4 \
--to=arighi@nvidia.com \
--cc=changwoo@igalia.com \
--cc=elxreno@gmail.com \
--cc=emil@etsalapatis.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=sched-ext@lists.linux.dev \
--cc=stable@vger.kernel.org \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox