On Wed, Feb 06, 2013 at 10:45:13AM -0500, Alan Stern wrote: > On Wed, 6 Feb 2013, Felipe Balbi wrote: > > > This is useful on gadgets that depend on userland > > daemons to function properly. We can delay connection > > to the host until userland is ready. > > > --- a/drivers/usb/gadget/udc-core.c > > +++ b/drivers/usb/gadget/udc-core.c > > @@ -235,7 +235,18 @@ static void usb_gadget_remove_driver(struct usb_udc > > *udc) > > > > kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE); > > > > - usb_gadget_disconnect(udc->gadget); > > + /* > > + * NOTICE: if gadget driver wants to control > > + * pullup, it needs to make sure that when > > + * user tries to rmmod the gadget driver, it > > + * will disconnect the pullups before returning > > + * from its ->unbind() method. > > + * > > + * We are truly trusting the gadget driver here. > > + */ > > + if (!udc->driver->controls_pullups) > > + usb_gadget_disconnect(udc->gadget); > > Even though you trust the gadget driver about when to enable the > pullup, doesn't it make sense to disable the pullup whenever the gadget > driver is removed, regardless?
Maybe, but we don't know what sort of trickery userland has. Also, I wanted to use this as a way to catch 'non-compliant' gadget drivers which say they will control pullup but don't. Maybe a warning if pullup is still connected ?!? -- balbi
signature.asc
Description: Digital signature
