On Tue, Jul 28, 2009 at 06:35:51AM +0200, Balbi Felipe (Nokia-D/Helsinki) wrote: > Hi, > > On Mon, Jul 27, 2009 at 10:50:53PM +0200, ext David Brownell wrote: > > > fail: > > > diff --git a/drivers/usb/gadget/u_serial.c b/drivers/usb/gadget/u_serial.c > > > index fc6e709..ab01440 100644 > > > --- a/drivers/usb/gadget/u_serial.c > > > +++ b/drivers/usb/gadget/u_serial.c > > > @@ -1201,6 +1201,42 @@ void gserial_cleanup(void) > > > } > > > > > > /** > > > + * gserial_start - starts gserial > > > + * @gser: the function to be started > > > + * @port_num: which port to start > > > + * Context: any (usually called from ->bind()) > > > + * > > > + * This should be called in early stage (usualy bind()) > > > + * in order to let the gaget driver connect to the bus > > > + * and let the gser->connect() be called. > > > > I don't follow. What's the problem with just > > having the function call gser_connect() when it's > > activated? Why should there be any bind-time hook > > beyond a single call to gserial_setup()? > > so we load g_serial with f_obex, then obexd opens /dev/ttyGS0 then we > connect the cable and gser->connect() is not called since port->port_usb > only becomes a valid pointer after gserial_connect() is called. And > gserial_connect() is only called when host succesfully sets alternate > setting 1 (on obex case). > > You see the problem ? so, just to clarify: gserial_connect() makes port->port_usb valid and that makes gser->connect() (in this case obex connect) "reachable", but gserial_connect() will never get called unless host sets alternate setting 1 of obex and host will never be able to do that if usb_function_activate() inside obex_connect() isn't called. So there's a little chicken-and-egg problem there, no? -- 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
