On Thu, Apr 28, 2005 at 02:36:39PM -0400, Alan Stern wrote: > Here is a fix for driver_detach(). It's a little ugly because it avoids > using the klist iterator, but there's no way around it -- the iterator > simply can't be made to work here. Not only does it prevent > device_release_driver() from calling klist_remove(), but also it prevents > doing get_device() while holding the klist spinlock.
I was about to cook up something very much like this :) > With this patch in place, the code in usb_driver_release_interface() > doesn't really need to be changed. Well, except for the inversion of the test, as given in my original post. > However I think it should be. As it > is now, it relies on internal knowledge of how the driver-model core > works, Agreed. IMHO if the driver model strives to take care of everything on caller's behalf, it also has to gracefully handle recursive calls of one of its exported interfaces, device_release_driver(). Cheers, Roman. ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20 _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
