Hi Dan, I think that I'll disable the UDC soon after jumping into the kernel. Otherwise, the endpoints cannot be reconfigured (as stated in drivers/usb/gadget/pxa27x_udc.h). I would hope that the pxa27x / udc code could then successfully reprogram the endpoints. I don't suppose you could post a few snippets of the code you hacked into your bootloader, could you? Is it something as simple as UDCCR = 0 ? Cheers, C On Mon, May 4, 2009 at 11:14 AM, Daniel Ribeiro <drwyrm@xxxxxxxxx> wrote: > Em Seg, 2009-05-04 às 00:29 +0200, Christopher Friedt escreveu: >> Well, I still haven't gotten the ethernet gadget to show up on my host >> machine, but I did get the framebuffer working :) >> >> It's wierd, the handheld shows that usb0 is configured as an ethernet >> gadget, but on the host machine, it doesn't change (it's still showing >> up as a 'generic ppc flash device'). >> >> I'll have another go at it sometime this week. > > I have a similar issue with my pxa27x device. > > My bootloader initializes the UDC, and gives control to kernel while the > UDC is enabled. > > On previous pxa27x-udc versions this resulted on a kernel panic, with > the current version it does not panic, but the host doesn't detect the > device gadget. If I unplug/plug the USB cable, then the host detects the > gadget driver and enumarates it correctly, but then it simply does not > work (i cant ping the device). > > I hacked the bootloader to disable the UDC before giving control to > kernel and this fixed the issue. But i think that the kernel driver > should be able to work doesn't matter the state at which the UDC is > delivered to it. Just my 2c. > > -- > Daniel Ribeiro > > -- 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
