On Fri, 2013-03-15 at 16:20 +0100, Johan Hovold wrote: > Either way, this series moves the wake-up from the individual driver's > port remove to usb-serial disconnect and before unregistering the device > from the bus. > > However, looking at this again now, I realise that the wake up must be > done in hangup rather than disconnect (the common source for hangup in > usb-serial) in order to fully fix the problem with processes not being > woken up on hangup. (Perhaps, this is what you referred to with > "robust".) I didn't review the other patches yet, so I wasn't specifically referring to that. I just meant that current and future contributors may not have as much comprehension of the tty layer and the subtleties of tty lifetimes (which has obviously been a recurrent problem). As you point out, the problem is fixed by consolidating this construct into the glue layer. Thanks again, Peter Hurley -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
