On Sat, Aug 14, 2010 at 4:35 AM, Michal Nazarewicz <mina86@xxxxxxxxxx> wrote: > Xiaofan Chen <xiaofanc@xxxxxxxxx> writes: > >> On Fri, Aug 13, 2010 at 5:31 PM, Dries Van Puymbroeck >> <dries.van.puymbroeck@xxxxxxxxx> wrote: >>> When a mutlifunction gadget has more than 1 configuration, Windows does not >>> use its generic driver for multifunctional USB devices. >> >> Sorry but what do you mean by the above? >> As per the previous discussions, it seems that a registry entry hack >> (for Vista and Windows 7) is needed. >> http://msdn.microsoft.com/en-us/library/ff538059%28VS.85%29.aspx >> >>> This commit adds an INF file that will load the required Windows driver >>> as well as a short description in gadget_multi.txt >> >> It seems to me your inf file forces the USB Generic >> Parent Driver (for USB Composite Device) to load. But for which >> configuration? I do not have any device to test myself. But >> I tend to guess it will still be the first configuration (either >> RNDIS or CDC ECM). > > 2-configuration g_multi has configuration with RNDIS as the first one so > that's pretty perfect for us. I guess so. But I am wondering if the inf file really needed or not in this case. Basically I am just wondering if the following statement is true or not. >>> When a mutlifunction gadget has more than 1 configuration, Windows does not >>> use its generic driver for multifunction USB devices. I am thinking the inf file is not necessary. "When the system detects that a USB device is composite, it loads the USB common class generic parent driver (usbccgp.sys)." http://msdn.microsoft.com/en-us/library/ff538059%28VS.85%29.aspx And since 1st configuration RNDIS is what desired, nothing needs to be done. Obviously you have the device to test, I would be interested to know your test results under XP, Vista and Win7. -- Xiaofan -- 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
