Am Donnerstag, 2. Februar 2012, 13:26:56 schrieb Bjørn Mork: > > No, you are misusing the flags here. We need a way to have mutual exclusion. > > The flags may be set after we have passed that point of execution. > > OK, forget the flag and take the global wdm_mutex during reset? Or do > you need another device specific mutex for this purpose to avoid locking > all opens of any cdc-wdm device while one is being reset? > > Or assume that any write blocking will be short, and use the wlock both > for reset and open? No, we cannot assume that. A device may be reset because it is broken. The correct algorithm is a) on the system code path (pre/post ... set a flag wake_up_all mutex_lock b) on the read/write code path mutex_lock wait_event //checking for the flag if (flag) return error mutex_unlock Taking a mutex is not wrong. We just need to make sure everybody holding a mutex makes progress when the system needs to do something. So we need to do wake_up_all() before we take a mutex and set a flag to avoid the race. Regards Oliver -- 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
