Hi Dave, It turns out that usb_function_deactivate() and usb_function_activate() doesn't seem to be enough for preventing enumeration of a composite gadget driver. Imagine we have a composite gadget driver with, say, 2 f_obex instances, they would sure prevent enumeration by calling usb_function_deactivate() right away, which, in the end, goes to usb_gadget_pull(g, 0) which disables the pullups on d+ But what would happen if only one obex server opens /dev/ttyGS? That would already allow enumeration and only one interface would be usable, right ? How about leaving the decision on when to turn on the pullups to userland ? Would that be bad ? I guess it would be better since userland can know when all the necessary daemons are ready. Comments ?? ps: hope I didn't get usb_gadget_function_deactive() wrong -- balbi -- 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
