MIME-Version: 1.0
References: <20211202101238.33546-1-elver@google.com>
In-Reply-To: <20211202101238.33546-1-elver@google.com>
From:   Marco Elver <elver@google.com>
Date:   Thu, 2 Dec 2021 12:53:14 +0100
Message-ID: <CANpmjNMvPepakONMjTO=FzzeEtvq_CLjPN6=zF35j10rVrJ9Fg@mail.gmail.com>
Subject: Re: [PATCH] locking/mutex: Mark racy reads of owner->on_cpu
To:     elver@google.com, Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>, Will Deacon <will@kernel.org>,
        Waiman Long <longman@redhat.com>,
        Boqun Feng <boqun.feng@gmail.com>, linux-kernel@vger.kernel.org
Cc:     kasan-dev@googlegroups.com, Thomas Gleixner <tglx@linutronix.de>,
        Mark Rutland <mark.rutland@arm.com>,
        "Paul E. McKenney" <paulmck@kernel.org>,
        Kefeng Wang <wangkefeng.wang@huawei.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, 2 Dec 2021 at 11:13, Marco Elver <elver@google.com> wrote:
> One of the more frequent data races reported by KCSAN is the racy read
> in mutex_spin_on_owner(), which is usually reported as "race of unknown
> origin" without showing the writer. This is due to the racing write
> occurring in kernel/sched. Locally enabling KCSAN in kernel/sched shows:
>
>  | write (marked) to 0xffff97f205079934 of 4 bytes by task 316 on cpu 6:
>  |  finish_task                kernel/sched/core.c:4632 [inline]
>  |  finish_task_switch         kernel/sched/core.c:4848
>  |  context_switch             kernel/sched/core.c:4975 [inline]
>  |  __schedule                 kernel/sched/core.c:6253
>  |  schedule                   kernel/sched/core.c:6326
>  |  schedule_preempt_disabled  kernel/sched/core.c:6385
>  |  __mutex_lock_common        kernel/locking/mutex.c:680
>  |  __mutex_lock               kernel/locking/mutex.c:740 [inline]
>  |  __mutex_lock_slowpath      kernel/locking/mutex.c:1028
>  |  mutex_lock                 kernel/locking/mutex.c:283
>  |  tty_open_by_driver         drivers/tty/tty_io.c:2062 [inline]
>  |  ...
>  |
>  | read to 0xffff97f205079934 of 4 bytes by task 322 on cpu 3:
>  |  mutex_spin_on_owner        kernel/locking/mutex.c:370
>  |  mutex_optimistic_spin      kernel/locking/mutex.c:480
>  |  __mutex_lock_common        kernel/locking/mutex.c:610
>  |  __mutex_lock               kernel/locking/mutex.c:740 [inline]
>  |  __mutex_lock_slowpath      kernel/locking/mutex.c:1028
>  |  mutex_lock                 kernel/locking/mutex.c:283
>  |  tty_open_by_driver         drivers/tty/tty_io.c:2062 [inline]
>  |  ...
>  |
>  | value changed: 0x00000001 -> 0x00000000
>
> This race is clearly intentional, and the potential for miscompilation
> is slim due to surrounding barrier() and cpu_relax(), and the value
> being used as a boolean.
>
> Nevertheless, marking this reader would more clearly denote intent and
> make it obvious that concurrency is expected. Use READ_ONCE() to avoid
> having to reason about compiler optimizations now and in future.
>
> Similarly, mark the read to owner->on_cpu in mutex_can_spin_on_owner(),
> which immediately precedes the loop executing mutex_spin_on_owner().
>
> Signed-off-by: Marco Elver <elver@google.com>
[...]

Kefeng kindly pointed out that there is an alternative, which would
refactor owner_on_cpu() from rwsem that would address both mutex and
rwsem:
https://lore.kernel.org/all/b641f1ea-6def-0fe4-d273-03c35c4aa7d6@huawei.com/

Preferences?

Thanks,
-- Marco