From: Peter Zijlstra <peterz@infradead.org>
To: Juri Lelli <juri.lelli@redhat.com>
Cc: John Stultz <jstultz@google.com>,
soolaugust@gmail.com, mingo@redhat.com,
linux-kernel@vger.kernel.org, zhidao su <suzhidao@xiaomi.com>,
Andrea Righi <arighi@nvidia.com>,
luca.abeni@santannapisa.it
Subject: Re: [PATCH] sched/deadline: Fix stale dl_defer_running in update_dl_entity() if-branch
Date: Tue, 7 Apr 2026 17:00:27 +0200 [thread overview]
Message-ID: <20260407150027.GI3738010@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <adT3IndtrOTxESDF@jlelli-thinkpadt14gen4.remote.csb>
On Tue, Apr 07, 2026 at 02:22:58PM +0200, Juri Lelli wrote:
> > Random brain wave...
> >
> > Since the dl_server is LLF (deferred), it will pretty much always trip
> > the dl_entity_overflow() when interrupted, right? Does it make sense to
> > use the revised wake-up rule for it, when appropriate?
> >
> > ---
> > diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
> > index d08b00429323..674de6a48551 100644
> > --- a/kernel/sched/deadline.c
> > +++ b/kernel/sched/deadline.c
> > @@ -1027,7 +1027,7 @@ static void update_dl_entity(struct sched_dl_entity *dl_se)
> > if (dl_time_before(dl_se->deadline, rq_clock(rq)) ||
> > dl_entity_overflow(dl_se, rq_clock(rq))) {
> >
> > - if (unlikely(!dl_is_implicit(dl_se) &&
> > + if (unlikely((!dl_is_implicit(dl_se) || dl_se->dl_defer) &&
> > !dl_time_before(dl_se->deadline, rq_clock(rq)) &&
> > !is_dl_boosted(dl_se))) {
> > update_dl_revised_wakeup(dl_se, rq);
> >
>
> So to keep boosting, by reducing runtime appropriately, until the end of
> the current dl-server period. Makes sense to me.
OK, like so then.
---
Subject: sched/deadline: Use revised wakeup rule for dl_server
From: Peter Zijlstra <peterz@infradead.org>
Date: Sat, 4 Apr 2026 12:22:44 +0200
John noted that commit 115135422562 ("sched/deadline: Fix 'stuck' dl_server")
unfixed the issue from commit a3a70caf7906 ("sched/deadline: Fix dl_server
behaviour").
The issue in commit 115135422562 was for wakeups of the server after the
deadline; in which case you *have* to start a new period. The case for
a3a70caf7906 is wakeups before the deadline.
Now, because the server is effectively running a least-laxity policy, it means
that any wakeup during the runnable phase means dl_entity_overflow() will be
true. This means we need to adjust the runtime to allow it to still run until
the existing deadline expires.
Use the revised wakeup rule for dl_defer entities.
Fixes: 115135422562 ("sched/deadline: Fix 'stuck' dl_server")
Reported-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Tested-by: John Stultz <jstultz@google.com>
Link: https://patch.msgid.link/20260404102244.GB22575@noisy.programming.kicks-ass.net
---
kernel/sched/deadline.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -1027,7 +1027,7 @@ static void update_dl_entity(struct sche
if (dl_time_before(dl_se->deadline, rq_clock(rq)) ||
dl_entity_overflow(dl_se, rq_clock(rq))) {
- if (unlikely(!dl_is_implicit(dl_se) &&
+ if (unlikely((!dl_is_implicit(dl_se) || dl_se->dl_defer) &&
!dl_time_before(dl_se->deadline, rq_clock(rq)) &&
!is_dl_boosted(dl_se))) {
update_dl_revised_wakeup(dl_se, rq);
prev parent reply other threads:[~2026-04-07 15:00 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-02 13:30 [PATCH] sched/deadline: Fix stale dl_defer_running in dl_server else-branch soolaugust
2026-04-03 0:05 ` John Stultz
2026-04-03 1:30 ` John Stultz
2026-04-03 8:12 ` [PATCH] sched/deadline: Fix stale dl_defer_running in update_dl_entity() if-branch soolaugust
2026-04-03 13:42 ` Peter Zijlstra
2026-04-03 13:58 ` Andrea Righi
2026-04-03 19:31 ` John Stultz
2026-04-03 22:46 ` Peter Zijlstra
2026-04-03 22:51 ` John Stultz
2026-04-03 22:54 ` John Stultz
2026-04-04 10:22 ` Peter Zijlstra
2026-04-05 8:37 ` zhidao su
2026-04-06 20:01 ` John Stultz
2026-04-06 20:03 ` John Stultz
2026-04-07 12:22 ` Juri Lelli
2026-04-07 15:00 ` Peter Zijlstra [this message]
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=20260407150027.GI3738010@noisy.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=arighi@nvidia.com \
--cc=jstultz@google.com \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luca.abeni@santannapisa.it \
--cc=mingo@redhat.com \
--cc=soolaugust@gmail.com \
--cc=suzhidao@xiaomi.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