[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CACT4Y+bTYUVbO56LpKHjGD_aUnSCqE-gft4idj-LOzr8vzF48A@mail.gmail.com>
Date: Sat, 26 May 2018 12:19:11 +0200
From: Dmitry Vyukov <dvyukov@...gle.com>
To: Shawn Lin <shawn.lin@...k-chips.com>
Cc: netdev <netdev@...r.kernel.org>,
"David S. Miller" <davem@...emloft.net>,
syzkaller-bugs <syzkaller-bugs@...glegroups.com>
Subject: Re: System hung for reg_check_changs_work()-> rtnl_lock()->mutex_lock()
On Mon, May 21, 2018 at 5:47 AM, Shawn Lin <shawn.lin@...k-chips.com> wrote:
> Hi,
>
> I found this hung for several times these days, and seems syzbot already
> reported a similar problem. Is there any patch(es) for that?
>
> Successfully initialized wpa_supplicant
> [ 240.091941] INFO: task kworker/u8:1:39 blocked for more than 120 seconds.
> [ 240.092004] Not tainted 4.4.126 #1
> [ 240.092026] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables
> this message.
> [ 240.092047] kworker/u8:1 D ffffff8008084dfc 0 39 2
> 0x00000000
> [ 240.092116] Workqueue: events_power_efficient reg_check_chans_work
> [ 240.092153] Call trace:
> [ 240.092191] [<ffffff8008084dfc>] __switch_to+0x84/0xa0
> [ 240.092228] [<ffffff80086299f0>] __schedule+0x428/0x45c
> [ 240.092260] [<ffffff8008629a98>] schedule+0x74/0x94
> [ 240.092295] [<ffffff8008629e44>] schedule_preempt_disabled+0x20/0x38
> [ 240.092332] [<ffffff800862b13c>] __mutex_lock_slowpath+0xc0/0x138
> [ 240.092364] [<ffffff800862b1e0>] mutex_lock+0x2c/0x40
> [ 240.092399] [<ffffff80084eb820>] rtnl_lock+0x14/0x1c
> [ 240.092428] [<ffffff80085ce2a0>] reg_check_chans_work+0x2c/0x1f0
> [ 240.092463] [<ffffff80080abbac>] process_one_work+0x1b0/0x294
> [ 240.092494] [<ffffff80080ac904>] worker_thread+0x2d8/0x398
> [ 240.092524] [<ffffff80080b0f04>] kthread+0xc8/0xd8
> [ 240.092567] [<ffffff8008082e80>] ret_from_fork+0x10/0x50
> [ 240.092594] Kernel panic - not syncing: hung_task: blocked tasks
> [ 240.101163] CPU: 0 PID: 30 Comm: khungtaskd Not tainted 4.4.126 #1
> [ 240.101729] Hardware name: Rockchip RK3308 evb analog mic board (DT)
> [ 240.102302] Call trace:
> [ 240.102546] [<ffffff800808731c>] dump_backtrace+0x0/0x1c4
> [ 240.103044] [<ffffff80080874f4>] show_stack+0x14/0x1c
> [ 240.103521] [<ffffff800823f4b4>] dump_stack+0x94/0xbc
> [ 240.104000] [<ffffff800810a80c>] panic+0xd8/0x228
> [ 240.104446] [<ffffff80080fb228>] proc_dohung_task_timeout_secs+0x0/0x40
> [ 240.105050] [<ffffff80080b0f04>] kthread+0xc8/0xd8
> [ 240.105500] [<ffffff8008082e80>] ret_from_fork+0x10/0x50
> [ 240.106065] CPU1: stopping
> [ 240.106348] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.4.126 #1
Syzbot has reported whole bunch of hangs on rtnl lock, but there is no
resolution:
https://syzkaller.appspot.com/bug?id=2503c576cabb08d41812e732b390141f01a59545
I suspect this can be related to hangs in unregister_netdevice:
https://syzkaller.appspot.com/bug?id=1a97a5bd119fd97995f752819fd87840ab9479a9
They happen all the time too, there is no resolution for this either.
Also see this thread:
https://groups.google.com/d/msg/syzkaller/-06_laheMF0/xqezy58kAwAJ
Powered by blists - more mailing lists