On Tue, 19 Aug 2003, John Levon wrote: > On Tue, Aug 19, 2003 at 01:45:18PM -0700, David Brownell wrote: > > > On the other hand, I can imagine some of Alan's updates might > > turn up bugs with drivers accessing endpoints that don't exist > > This could well be it; adding the hunk below causes a change from > success to failure. ... > + * In addition to unbinding the driver, this re-initializes the interface > + * by selecting altsetting 0, the default alternate setting. ... > + usb_set_interface(interface_to_usbdev(iface), > + iface->altsetting[0].desc.bInterfaceNumber, > + 0);
If a program is trying to use an interface after a driver has released it, this part of the patch could easily mess things up. Perhaps it should be removed. Or perhaps the ppp code should explicitly select the correct altsetting for itself. I can no longer remember exactly why I added that code in there. It was probably in order to make sure that the next driver to bind the interface would see it in a properly initialized (i.e., default) state. And after all, once a driver has unbound it has no more say over what happens to the interface, so it shouldn't care if the altsetting gets changed. I'm open to arguments on this. Alan Stern ------------------------------------------------------- This SF.net email is sponsored by Dice.com. Did you know that Dice has over 25,000 tech jobs available today? From careers in IT to Engineering to Tech Sales, Dice has tech jobs from the best hiring companies. http://www.dice.com/index.epl?rel_code=104 _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
