On Fri, 23 Nov 2012, Oliver Neukum wrote:
> Hi,
>
> in usb_unbind_interface() we call usb_cancel_queued_reset() before
> restoring altsetting 0. This seems wrong to me. If a driver found it
> necessary to reset a device we cannot simply ignore that. I'd say
> that we should wait for the work to f
On Fri, Nov 23, 2012 at 10:39 PM, Oliver Neukum wrote:
> Hi,
>
> in usb_unbind_interface() we call usb_cancel_queued_reset() before
> restoring altsetting 0. This seems wrong to me. If a driver found it
> necessary to reset a device we cannot simply ignore that. I'd say
> that we should wait for t
Hi,
in usb_unbind_interface() we call usb_cancel_queued_reset() before
restoring altsetting 0. This seems wrong to me. If a driver found it
necessary to reset a device we cannot simply ignore that. I'd say
that we should wait for the work to finish, not cancel it.
What do you say?
Regards