No subject
Sun Dec 9 17:34:17 CET 2012
>
>
>> @@ -165,7 +166,7 @@ get_descriptor (usbdev_t *dev, unsigned char bmReque=
stType, int descType,
>> + =A0 =A0 =A0 =A0 control (ep, IN, sizeof (dr), &dr, size, result)) {
>
> Same one..
Same as above.
>
>
>> @@ -183,7 +184,7 @@ set_configuration (usbdev_t *dev)
>> + =A0 =A0 dev->controller->control (&dev->endpoints[0], OUT, sizeof (dr)=
, &dr, 0, 0);
>
> Is index 0 in endpoints guaranteed to always be the default endpoint?
Yes, in fact it should be used in every call commented previously.
>
>
>> @@ -201,7 +202,7 @@ clear_stall (endpoint_t *ep)
>> + =A0 =A0 dev->controller->control (ep, OUT, sizeof (dr), &dr, 0, 0);
>
> Good.
>
>
>> @@ -246,7 +247,7 @@ set_address (hci_t *controller, int lowspeed)
>> + =A0 =A0 if (dev->controller->control (&dev->endpoints[0], OUT, sizeof =
(dr), &dr, 0, 0)) {
>
> Again with index 0. And it happens a few more times.
Ok, it should always be 0.
>
>
> //Peter
>
> --
> coreboot mailing list: coreboot at coreboot.org
> http://www.coreboot.org/mailman/listinfo/coreboot
>
Thanks a lot for reviewing. Maybe the whole API should be changed to
match devices with more than one control endpoint, but this is a first
change. Please tell me if I`m missing something about the uhci_reg_*
functions.
Regards...
--=20
(=B0=3D Leandro Dorileo
//\ ldorileo at gmail.com - http://www.dorilex.net
V_/ Software is a matter of freedom.
More information about the coreboot
mailing list