On Wed, 16 Dec 2009, Oliver Neukum wrote: > Am Dienstag, 15. Dezember 2009 16:42:54 schrieb Alan Stern: > > On Tue, 15 Dec 2009, Oliver Neukum wrote: > > > > > Am Montag, 14. Dezember 2009 23:12:06 schrieb Alan Stern: > > > > Yes, I have some code to fall back to usb_reset_configuration(). > > > Should we do so also for storage? > > > > I don't think that's such a good idea. For instance, I doubt > > usb_reset_configuration() will be powerful enough to recover from the > > errors usb-storage encounters. In fact, even a port reset isn't always > > powerful enough. > > True, but we cannot do a port reset for such devices. Are you certain? Does the mode-switch state get reset along with everything else? Or does the device remain in the same mode until a power cycle? > > Also it doesn't notify drivers about the reset. Really it was meant > > for use only with non-composite devices, where there's never more than > > one driver involved. > > Yes, we'd need an extended version that calls pre/post_reset. > I am attaching the stuff I wrote before I looked closely at the > reset logic. This really is too much complication. We shouldn't need multiple ways to reset a device. (And if we do, they should share more code.) > > Even if mode switching were in the kernel (and in a few cases it is), > > there still would be problems. Mode switching often involves changing > > descriptors. The core interprets these changes as meaning that the > > device has disconnected and another device was plugged into the port. > > We'd have to save copies of descriptors in both modes, compare, > initiate a reswitch and wait for it. Extremely ugly. Yes indeed. Much more trouble than it is worth, IMO. > > Face it, mode switching is a bad idea. These devices should have been > > designed to use multiple configurations instead. Since they weren't, > > their functionality is severely compromised: It's very difficult to > > reset them for error recovery. > > I do face it. I am asking whether the unease you feel is due to sucking > hardware or have I overlooked something? It's the hardware. The new quirk is acceptable, but I don't like usb_multiple_reset_configuration(). Alan Stern -- 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
