On Tue, 24 Jul 2012, Lan Tianyu wrote: > 于 2012/7/24 22:59, Alan Stern 写道: > > Yes, good idea. > > > > When you do this, you should also add another little paragraph to > > Documentation/usb/persist.txt. At the end of the "What is the > > solution?" section, explain why the persist attribute cannot be set if > > the avoid_reset_quirk attribute is on. > > > > Hmmm, what about hubs? They don't have a persist attribute because > > persist is_always_ supposed to be enabled for a hub. Maybe this means > > they shouldn't have an avoid_reset_quirk attribute either. > Another proposal is to add USB_CLASS_HUB check in the avoid_reset_quirk > attribute callback. If it was hub, return directly. Or else take the &dev_attr_avoid_reset_quirk.attr entry out of dev_attrs[], and instead make usb_create_sysfs_dev_files() and usb_remove_sysfs_dev_files() explicitly create and remove the avoid_reset_quirk attribute if the device isn't a hub. Adding the check to the avoid_reset_quirk callback is easier, though. 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
