linux-kernel - Re: [PATCHv2] workqueue: Catch more locking problems with flush

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Sat, 21 Apr 2012 08:34:01 +0800
From:	Yong Zhang <yong.zhang0@...il.com>
To:	Stephen Boyd <sboyd@...eaurora.org>
Cc:	Tejun Heo <tj@...nel.org>, linux-kernel@...r.kernel.org,
	netdev@...r.kernel.org, Ben Dooks <ben-linux@...ff.org>
Subject: Re: [PATCHv2] workqueue: Catch more locking problems with flush_work()

On Sat, Apr 21, 2012 at 8:28 AM, Stephen Boyd <sboyd@...eaurora.org> wrote:
> If a workqueue is flushed with flush_work() lockdep checking can
> be circumvented. For example:
>
>  static DEFINE_MUTEX(mutex);
>
>  static void my_work(struct work_struct *w)
>  {
>         mutex_lock(&mutex);
>         mutex_unlock(&mutex);
>  }
>
>  static DECLARE_WORK(work, my_work);
>
>  static int __init start_test_module(void)
>  {
>         schedule_work(&work);
>         return 0;
>  }
>  module_init(start_test_module);
>
>  static void __exit stop_test_module(void)
>  {
>         mutex_lock(&mutex);
>         flush_work(&work);
>         mutex_unlock(&mutex);
>  }
>  module_exit(stop_test_module);
>
> would not always print a warning when flush_work() was called.
> In this trivial example nothing could go wrong since we are
> guaranteed module_init() and module_exit() don't run concurrently,
> but if the work item is schedule asynchronously we could have a
> scenario where the work item is running just at the time flush_work()
> is called resulting in a classic ABBA locking problem.
>
> Add a lockdep hint by acquiring and releasing the work item
> lockdep_map in flush_work() so that we always catch this
> potential deadlock scenario.
>
> Signed-off-by: Stephen Boyd <sboyd@...eaurora.org>

Reviewed-by: Yong Zhang <yong.zhang0@...il.com>

> ---
>  kernel/workqueue.c |    3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/kernel/workqueue.c b/kernel/workqueue.c
> index 5abf42f..038cf64 100644
> --- a/kernel/workqueue.c
> +++ b/kernel/workqueue.c
> @@ -2506,6 +2506,9 @@ bool flush_work(struct work_struct *work)
>  {
>        struct wq_barrier barr;
>
> +       lock_map_acquire(&work->lockdep_map);
> +       lock_map_release(&work->lockdep_map);
> +
>        if (start_flush_work(work, &barr, true)) {
>                wait_for_completion(&barr.done);
>                destroy_work_on_stack(&barr.work);
> --
> Sent by an employee of the Qualcomm Innovation Center, Inc.
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
>



-- 
Only stand for myself
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/