Re: hid-pidff bug: fails to find all required reports of saitek gamepad

[Dmitriy Geels <dmitriy.geels@gmail.com>]

Hello!

It took some time to get back to forcefeedback debugging for me.

2009/7/9 Anssi Hannula <anssi.hannula@gmail.com>:
> Dmitriy Geels wrote:
>> 2009/6/12 Dmitriy Geels <dmitriy.geels@gmail.com>:
>>> Hmm... first try failed. After running fftest, then ffmvforce got
>>> system hang in virtual machine. X is completely unresponsive. May be
>>> kernel is still alive, have to find out, how to press ctrl-alt-fx in
>>> virtualbox.
>
> You can run "chvt 1" to switch to a virtual console and test there, so
> you can see any kernel errors. IIRC "echo 9 > /proc/sysrq-trigger" makes
> all kernel messages visible on console.
>
>> Bad... Can't get something out of driver with last patch. Running
>> ffmvforce after fftest/ffcstresstest hangs VM after few seconds (looks
>> like some timeout) with 100% cpu load.
>
> That looks like there is some bug in the code, then.
> The last patch added a 8sec timeout after which the URB is canceled.
>
>> May be it's possible to get logs via serial cable, but will they
>> contain something useful?
>
> Possibly. Alternatively you can do as above with virtualbox.

Remember your last patches with hid_urb_timeout()?
I got kernel panic log after test: http://paste.org.ru/index.pl?fucaex
Looks like something bad happen inside usb_unlink_urb(hid_urb);