linux-kernel - Re: [2.6.31-rc9] hotplug SATA vs lockdep: false positive?

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]

Message-ID: <d82e647a0909101859t48f4a494l75040b47a7474760@mail.gmail.com>
Date:	Fri, 11 Sep 2009 09:59:23 +0800
From:	Ming Lei <tom.leiming@...il.com>
To:	Peter Zijlstra <peterz@...radead.org>
Cc:	Daniel J Blueman <daniel.blueman@...il.com>,
	Linux Kernel <linux-kernel@...r.kernel.org>,
	linux-acpi@...r.kernel.org
Subject: Re: [2.6.31-rc9] hotplug SATA vs lockdep: false positive?

2009/9/10 Peter Zijlstra <peterz@...radead.org>:
> On Wed, 2009-09-09 at 22:20 +0100, Daniel J Blueman wrote:
>
>> When hot-plugging my SATA DVD drive into my laptop, I see a lockdep
>> warning [1]. On closer inspection, both flush_workqueue() and
>> worker_thread() do tricks with lockdep maps. False positive?
>
> No, looks like a typical case of a workqueue trying to flush itself,
> something that can easily deadlock for real.

Hi,Peter

IMHO, it seems flushing other workqueues in one workqueue, so it
may be a false positive. Since the three workqueue instances share one
lock class, maybe lockdep_set_class*() or other similar annotations
is needed in acpi_os_initialize1() to avoid the warning.

Thanks.

>
>> =======================================================
>> [ INFO: possible circular locking dependency detected ]
>> 2.6.31-rc9-290cd #1
>> -------------------------------------------------------
>> kacpi_hotplug/198 is trying to acquire lock:
>>  (kacpid){+.+.+.}, at: [<ffffffff81073a70>] flush_workqueue+0x0/0xf0
>> (workqueue.c:292)
>>
>> but task is already holding lock:
>>  (&dpc->work){+.+.+.}, at: [<ffffffff81072a42>]
>> worker_thread+0x1f2/0x3c0 (bitops.h:101)
>> which lock already depends on the new lock.
>>
>> the existing dependency chain (in reverse order) is:
>>
>> -> #1 (&dpc->work){+.+.+.}:
>>        [<ffffffff8108eb19>] __lock_acquire+0xe29/0x1240
>>        [<ffffffff8108f04e>] lock_acquire+0x11e/0x170
>>        [<ffffffff81072a92>] worker_thread+0x242/0x3c0 (workqueue.c:291)
>>        [<ffffffff81077456>] kthread+0xa6/0xc0
>>        [<ffffffff8100d29a>] child_rip+0xa/0x20
>>        [<ffffffffffffffff>] 0xffffffffffffffff
>>
>> -> #0 (kacpid){+.+.+.}:
>>        [<ffffffff8108ebe1>] __lock_acquire+0xef1/0x1240
>>        [<ffffffff8108f04e>] lock_acquire+0x11e/0x170
>>        [<ffffffff81073acc>] flush_workqueue+0x5c/0xf0 (workqueue.c:403)
>>        [<ffffffff812f6eaf>] acpi_os_wait_events_complete+0x10/0x1e
>>        [<ffffffff812f6ee7>] acpi_os_execute_hp_deferred+0x2a/0x3e
>>        [<ffffffff81072a98>] worker_thread+0x248/0x3c0 (workqueue.c:292)
>>        [<ffffffff81077456>] kthread+0xa6/0xc0
>>        [<ffffffff8100d29a>] child_rip+0xa/0x20
>>        [<ffffffffffffffff>] 0xffffffffffffffff
>>
>> other info that might help us debug this:
>>
>> 2 locks held by kacpi_hotplug/198:
>>  #0:  (kacpi_hotplug){+.+...}, at: [<ffffffff81072a42>]
>> worker_thread+0x1f2/0x3c0
>>  #1:  (&dpc->work){+.+.+.}, at: [<ffffffff81072a42>] worker_thread+0x1f2/0x3c0
>>
>> stack backtrace:
>> Pid: 198, comm: kacpi_hotplug Tainted: G         C 2.6.31-rc9-290cd #1
>>
>> Call Trace:
>>  [<ffffffff8108c8a7>] print_circular_bug_tail+0xa7/0x100
>>  [<ffffffff8108ebe1>] __lock_acquire+0xef1/0x1240
>>  [<ffffffff8108a808>] ? add_lock_to_list+0x58/0xf0
>>  [<ffffffff8108f04e>] lock_acquire+0x11e/0x170
>>  [<ffffffff81073a70>] ? flush_workqueue+0x0/0xf0 (workqueue.c:397)
>>  [<ffffffff812f6ebd>] ? acpi_os_execute_hp_deferred+0x0/0x3e
>>  [<ffffffff81073acc>] flush_workqueue+0x5c/0xf0 (workqueue.c:403)
>>  [<ffffffff81073a70>] ? flush_workqueue+0x0/0xf0 (workqueue.c:397)
>>  [<ffffffff812f6eaf>] acpi_os_wait_events_complete+0x10/0x1e drivers/acpi/osl.c
>>  [<ffffffff812f6ee7>] acpi_os_execute_hp_deferred+0x2a/0x3e
>>  [<ffffffff81072a98>] worker_thread+0x248/0x3c0 (workqueue.c:292)
>>  [<ffffffff81072a42>] ? worker_thread+0x1f2/0x3c0
>>  [<ffffffff81077900>] ? autoremove_wake_function+0x0/0x40
>>  [<ffffffff81072850>] ? worker_thread+0x0/0x3c0
>>  [<ffffffff81077456>] kthread+0xa6/0xc0
>>  [<ffffffff8100d29a>] child_rip+0xa/0x20
>>  [<ffffffff8100cbd4>] ? restore_args+0x0/0x30
>>  [<ffffffff810773b0>] ? kthread+0x0/0xc0
>>  [<ffffffff8100d290>] ? child_rip+0x0/0x20
>
> --
> 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/
>



-- 
Lei Ming
--
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/