Hi Michael, On 10/13/2011 8:31 PM, Michael Grzeschik wrote: > On Thu, Oct 13, 2011 at 01:51:52PM +0200, Marc Kleine-Budde wrote: >> > On 10/11/2011 06:54 AM, Pavan Kondeti wrote: >>>> > >> while playing around with gadgetfs and the ci13xxx_udc driver on mx28 >>>> > >> I found some problems regarding locking and DMA handling. Another problem >>>> > >> arises when using the ci13xxx_udc driver on mx28, because it has only 8 >>>> > >> compared to 16 endpoints as the driver expects. >> > >>>> > >> With these patches g_ether and gadgetfs work on the mx28, however I was >>>> > >> not able to test these patches on any other hardware. Voluntaries welcome. >> > >>> > > Thanks for the improvements and bug fixes in ci13xxx_udc driver. All the >>> > > patches looks good to me. I am slightly busy with other stuff. Will test >>> > > these patches on MSM platform next week. >> > >> > We've discovered a problem with the driver (on a mx23). >> > g_zero test 2 fails. > We now are able to get all g_zero tests [1] working, but for that we have to > load the g_zero module with buflen=512. > > Has this to do with the limitation in the _hardware_enqueue routine that > it currently can not handle requests which would spawn into several TDs? ci13xxx_udc can handle up to 16k size request. gadget zero default request size is only 4k. I am not sure why you are seeing test#2 (512 bytes IN transfer) failure. I will try to run the test multiple times. Thanks, Pavan -- Sent by a consultant of the Qualcomm Innovation Center, Inc. The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum. -- 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
