2011/1/28 MichaÅ Nazarewicz <mina86@xxxxxxxxxx>:
> On Jan 28, 2011 6:19 PM, "Alan Stern" <stern@xxxxxxxxxxxxxxxxxxx> wrote:
>> Right. ÂThat was the point of the original message. ÂThe test failed
>> because MSF didn't respond correctly, because it didn't get called,
>> because the test's setup packet had the wrong wIndex value.
>
> Ah, OK, my understanding was that MSF got called with wIndex having invalid
> value.
Alan, Michal,
Thanks for your comments.
On further investigation I found the below code in f_rndis.c
rndis_bind_config() which disables rndis by default.
ifdef CONFIG_USB_ANDROID_RNDIS
/* start disabled */
rndis->port.func.disabled = 1;
#endif
It looks like on enumeration with a Windows PC, since the RNDIS
function is disabled, Windows does not know about this interface at
all.
It therefore considers UMS as the first interface and issues SETUP
packet with wIndex = 0. But on the device side UMS is at interface 2
and hence the check in fsg_setup() fails.
Since RNDIS is disabled by default the Configuration descriptor also
shows UMS as the first interface.
Do you agree with my understanding here?
Also if I enable RNDIS by default, Windows sees all the interfaces
correctly i.e rndis(interface 0 and 1), UMS as interface 2 and ADB as
interface 3. Although the enumeration is not complete. Seems like some
issues when all functions are enabled together by default.
Any idea why is RNDIS disabled by default for android gadget?
Regards,
Maulik
--
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
