On Thu, Oct 7, 2010 at 10:39 AM, Ming Lei <tom.leiming@xxxxxxxxx> wrote: > 2010/10/7 David Brownell <david-b@xxxxxxxxxxx>: >> I'm quite certain I've sent *TWO* messages which >> asked you explicitly *NOT* to take this approach. > > You has replied only one message for V1, see the link below > > http://marc.info/?l=linux-usb&m=128318744211444&w=2 > > I also replied you in the link below: > > http://marc.info/?l=linux-usb&m=128334955300554&w=2. > >> >> I want to see more focussed test cases and software, >> instead of just glomming everything into a big mess. > > You did not describe the more focussed test cases > in your previous message, could you give a explicit > description about that now? > >> >> Write a new gadget driver, specifically for testing >> ISO capabilities, and stuffed with logic which will >> trigger and evaluate ISO-specific failure modes. > > You mean we need to write a new udc controller driver, > which is used to test ISO capabilities only? > > This patch only introduces test cases for ISO transfer, > which is in function driver layer, not in udc controller driver. If I understand correctly, I believe David wants a new function driver written which behaves similar to g_zero or gadgetfs, but uses isochronous endpoints instead of bulk. He wants us to not modify g_zero for this purpose. He does not mean a new udc controller driver. The new function driver should work on existing device controllers, as long as they are capable of setting up isochronous endpoints. - Anand -- 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
