All of lore.kernel.org
 help / color / mirror / Atom feed
From: liuwenfang <liuwenfang@honor.com>
To: 'Tejun Heo' <tj@kernel.org>
Cc: 'David Vernet' <void@manifault.com>,
	'Andrea Righi' <arighi@nvidia.com>,
	'Changwoo Min' <changwoo@igalia.com>,
	'Ingo Molnar' <mingo@redhat.com>,
	'Peter Zijlstra' <peterz@infradead.org>,
	'Juri Lelli' <juri.lelli@redhat.com>,
	'Vincent Guittot' <vincent.guittot@linaro.org>,
	'Dietmar Eggemann' <dietmar.eggemann@arm.com>,
	'Steven Rostedt' <rostedt@goodmis.org>,
	'Ben Segall' <bsegall@google.com>, 'Mel Gorman' <mgorman@suse.de>,
	'Valentin Schneider' <vschneid@redhat.com>,
	"'linux-kernel@vger.kernel.org'" <linux-kernel@vger.kernel.org>
Subject: [PATCH] sched_ext: Fix cpu_released while RT task and SCX task are scheduled concurrently
Date: Sat, 21 Jun 2025 04:09:55 +0000	[thread overview]
Message-ID: <fca528bb34394de3a7e87a873fadd9df@honor.com> (raw)

Supposed RT task(rt1) is running on one CPU with its rq->scx.cpu_released
set to true, if the rt1 becomes sleeping, then the scheduler will balance
the remote SCX task(scx1) because there is no other RT task on its rq,
and rq->scx.cpu_released is false. While one RT task(rt2) is placed on
this rq(maybe rt2 wakeup or migration occurs) before the scx1 is enqueued,
then the scheduler will pick rt2. At last, rt2 will be running on this cpu
with rq->scx.cpu_released being false!
The main reason is that consume_remote_task() will unlock rq lock.

So, expose the switch_class() and check sched class again to fix the value
of rq->scx.cpu_released.

Signed-off-by: liuwenfang liuwenfang@honor.com

---
 kernel/sched/ext.c   | 2 +-
 kernel/sched/sched.h | 6 ++++++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index f5133249f..6bbea0ea1 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -3187,7 +3187,7 @@ preempt_reason_from_class(const struct sched_class *class)
 	return SCX_CPU_PREEMPT_UNKNOWN;
 }
 
-static void switch_class(struct rq *rq, struct task_struct *next)
+void switch_class(struct rq *rq, struct task_struct *next)
 {
 	const struct sched_class *next_class = next->sched_class;
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 47972f34e..d377075d0 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1738,6 +1738,7 @@ static inline void scx_rq_clock_invalidate(struct rq *rq)
 	WRITE_ONCE(rq->scx.flags, rq->scx.flags & ~SCX_RQ_CLK_VALID);
 }
 
+void switch_class(struct rq *rq, struct task_struct *next);
 #else /* !CONFIG_SCHED_CLASS_EXT */
 #define scx_enabled()		false
 #define scx_switched_all()	false
@@ -2470,6 +2471,11 @@ static inline void put_prev_set_next_task(struct rq *rq,
 
 	prev->sched_class->put_prev_task(rq, prev, next);
 	next->sched_class->set_next_task(rq, next, true);
+
+#ifdef CONFIG_SCHED_CLASS_EXT
+	if (scx_enabled())
+		switch_class(rq, next);
+#endif
 }
 
 /*
-- 
2.17.1

             reply	other threads:[~2025-06-21  4:10 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-21  4:09 liuwenfang [this message]
2025-06-23 19:50 ` [PATCH] sched_ext: Fix cpu_released while RT task and SCX task are scheduled concurrently 'Tejun Heo'
2025-06-28  6:50   ` [PATCH v2 1/2] " liuwenfang
2025-07-17 21:38     ` 'Tejun Heo'
2025-07-20  9:20       ` liuwenfang
2025-07-20  9:38         ` [PATCH v3 2/3] " liuwenfang
2025-08-12  1:26           ` 'Tejun Heo'
2025-07-20  9:41         ` [PATCH v3 3/3] sched_ext: Fix cpu_released while changing sched policy of the running task liuwenfang
2025-08-12  1:31           ` 'Tejun Heo'
2025-08-19  6:52           ` [PATCH v4 1/3] sched_ext: Fix pnt_seq calculation when picking the next task liuwenfang
2025-08-19  6:55             ` [PATCH v4 2/3] sched_ext: Fix cpu_released while RT task and SCX task are scheduled concurrently liuwenfang
2025-08-19  7:07               ` [PATCH v4 3/3] sched_ext: Fix cpu_released while changing sched policy of the running task liuwenfang
2025-08-19  7:47               ` [PATCH v4 2/3] sched_ext: Fix cpu_released while RT task and SCX task are scheduled concurrently Peter Zijlstra
2025-08-19  8:47                 ` 回复: " liuwenfang
2025-08-19 10:08                   ` Peter Zijlstra
2025-08-20  0:28                 ` 'Tejun Heo'
2025-08-20  9:18                   ` Peter Zijlstra
2025-08-20 16:52                     ` 'Tejun Heo'
2025-06-28  7:20   ` [PATCH v2 2/2] sched_ext: Fix cpu_released while changing sched policy of the running task liuwenfang
2025-07-17 21:48     ` 'Tejun Heo'
2025-07-18  9:06       ` liuwenfang
2025-07-20  9:36         ` [PATCH v3 1/3] sched_ext: Fix pnt_seq calculation liuwenfang
2025-08-12  0:03           ` 'Tejun Heo'
2025-08-12  0:30             ` 'Tejun Heo'
2025-08-18 10:45               ` liuwenfang
2025-08-18 17:43                 ` 'Tejun Heo'
2025-08-19  7:41                   ` liuwenfang
2025-08-18 17:47           ` Peter Zijlstra
2025-08-19  7:36             ` liuwenfang

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=fca528bb34394de3a7e87a873fadd9df@honor.com \
    --to=liuwenfang@honor.com \
    --cc=arighi@nvidia.com \
    --cc=bsegall@google.com \
    --cc=changwoo@igalia.com \
    --cc=dietmar.eggemann@arm.com \
    --cc=juri.lelli@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=tj@kernel.org \
    --cc=vincent.guittot@linaro.org \
    --cc=void@manifault.com \
    --cc=vschneid@redhat.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.