From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: Yao Kai <yaokai34@huawei.com>
Cc: linux-kernel@vger.kernel.org, tglx@kernel.org, mingo@redhat.com,
peterz@infradead.org, dvhart@infradead.org, dave@stgolabs.net,
andrealmeid@igalia.com, liuyongqiang13@huawei.com
Subject: Re: [PATCH 2/2] futex/requeue: Prevent rcuwait use-after-free during requeue PI
Date: Tue, 21 Jul 2026 09:19:14 +0200 [thread overview]
Message-ID: <20260721071914.48_P-BjT@linutronix.de> (raw)
In-Reply-To: <c8d2128a-cb6d-4817-99e7-01dc380b579c@huawei.com>
On 2026-07-21 10:19:57 [+0800], Yao Kai wrote:
> >
> > You will miss to wake T1 if T2 skips the wake, as suggested. Or do I
> > miss something?
> >
>
> T1 T2
> futex_requeue_pi_wakeup_sync()
> old = Q_REQUEUE_PI_IN_PROGRESS
> new = Q_REQUEUE_PI_WAIT
> cmpxchg()
> requeue_pi_wake_futex()
> task = READ_ONCE(q->task)
> futex_requeue_pi_complete()
> old = Q_REQUEUE_PI_WAIT
> new = Q_REQUEUE_PI_LOCKED
> cmpxchg()
> if (old == Q_REQUEUE_PI_IN_PROGRESS)
> rcuwait_wait_event()
> rcuwait_wake_up(&q->requeue_wait);
> wake_up_state(task, TASK_NORMAL)
> <leave>
>
> Only requeue_pi_wake_futex() can publish Q_REQUEUE_PI_LOCKED. After it calls
> futex_requeue_pi_complete(), it will call wake_up_state(task, TASK_NORMAL)
> to wake up T1 sleeping on rcuwait_wait_event().
That one is intended for the futex_do_wait() but it seems that the outer
wake_up_state() would save the day.
It would deserve a comment and explanation at the least.
> Yao
Sebastian
prev parent reply other threads:[~2026-07-21 7:19 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-17 8:49 [PATCH 0/2] futex/requeue: Fix requeue PI races Yao Kai
2026-07-17 8:49 ` [PATCH 1/2] futex/requeue: Fix rtmutex schedule preparation for requeue PI Yao Kai
2026-07-17 8:55 ` Sebastian Andrzej Siewior
2026-07-20 2:40 ` Yao Kai
2026-07-20 14:58 ` Sebastian Andrzej Siewior
2026-07-21 1:51 ` Yao Kai
2026-07-21 7:23 ` Sebastian Andrzej Siewior
2026-07-21 9:02 ` Yao Kai
2026-07-21 9:45 ` Sebastian Andrzej Siewior
2026-07-17 8:49 ` [PATCH 2/2] futex/requeue: Prevent rcuwait use-after-free during " Yao Kai
2026-07-17 9:38 ` Sebastian Andrzej Siewior
2026-07-20 2:50 ` Yao Kai
2026-07-20 14:55 ` Sebastian Andrzej Siewior
2026-07-21 2:19 ` Yao Kai
2026-07-21 7:19 ` Sebastian Andrzej Siewior [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=20260721071914.48_P-BjT@linutronix.de \
--to=bigeasy@linutronix.de \
--cc=andrealmeid@igalia.com \
--cc=dave@stgolabs.net \
--cc=dvhart@infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=liuyongqiang13@huawei.com \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=tglx@kernel.org \
--cc=yaokai34@huawei.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.