Re: [syzbot] [usb?] INFO: task hung in usb_port_suspend

[Alan Stern <stern@rowland.harvard.edu>]

On Mon, Oct 21, 2024 at 10:04:52AM +0200, Oliver Neukum wrote:
> On 20.10.24 18:38, syzbot wrote:
> > INFO: task kworker/0:0:8 blocked for more than 143 seconds.
> >        Not tainted 6.12.0-rc3-syzkaller-00051-g07b887f8236e #0
> > "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> > task:kworker/0:0     state:D stack:24544 pid:8     tgid:8     ppid:2      flags:0x00004000
> > Workqueue: pm pm_runtime_work
> > Call Trace:
> >   <TASK>
> >   context_switch kernel/sched/core.c:5322 [inline]
> >   __schedule+0x105f/0x34b0 kernel/sched/core.c:6682
> >   __schedule_loop kernel/sched/core.c:6759 [inline]
> >   schedule+0xe7/0x350 kernel/sched/core.c:6774
> 
> And this sleeps forever. This must not happen.
> >   usb_kill_urb.part.0+0x1ca/0x250 drivers/usb/core/urb.c:713
> >   usb_kill_urb+0x83/0xa0 drivers/usb/core/urb.c:702
> 
> We are changing our mind, presumably due to a timeout
> >   usb_start_wait_urb+0x255/0x4c0 drivers/usb/core/message.c:65
> 
> We are sending a control message, presumably to enable
> remote wakeup
> >   usb_internal_control_msg drivers/usb/core/message.c:103 [inline]
> >   usb_control_msg+0x327/0x4b0 drivers/usb/core/message.c:154
> >   usb_enable_remote_wakeup drivers/usb/core/hub.c:3365 [inline]
> >   usb_port_suspend+0x339/0xf10 drivers/usb/core/hub.c:3472
> 
> Suspending ...
> >   usb_generic_driver_suspend+0xeb/0x1d0 drivers/usb/core/generic.c:302
> >   usb_suspend_device drivers/usb/core/driver.c:1272 [inline]
> >   usb_suspend_both+0x66d/0x9c0 drivers/usb/core/driver.c:1443
> >   usb_runtime_suspend+0x49/0x180 drivers/usb/core/driver.c:1968
> 
> This very much looks like the HC driver used to run these tests
> can hand in unlink. If that happens there is nothing usbcore
> or a driver can do.
> As this is now reproducible I would suggest a bisection. Brute force,
> but I see no good alternative.
> 
> Syzbot is an important tool and if the HC driver it uses is unreliable,
> the whole thing becomes unreliable and that is most undesirable.

This issue should be fixed by commit 5189df7b8088 ("USB: gadget: 
dummy-hcd: Fix "task hung" problem").

Alan Stern