On Fri, 6 Jul 2018, Pradeep Das wrote: > Thanks, Alan Stern for a quick reply. > > Can you please suggest a probable area to look out for the failure of > the endpoint halt test case for the superspeed devices. I cannot. The person you should ask is the maintainer of the xhci-hcd driver, Mathias Nyman <mathias.nyman@xxxxxxxxx>. However, he will be away on vacation for most of July. Alan Stern > On Thu, Jul 5, 2018 at 8:31 PM Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote: > > > > On Thu, 5 Jul 2018, Pradeep Das wrote: > > > > > Hi, > > > > > > Posting this query again as I received some mail delivery failure > > > notification on the previous attempt. > > > > > > I am running usbtest test cases and facing following issues: > > > > > > Setup: > > > I am using two custom board running on Linux. One is configured as > > > host and loaded with usbtest.ko module. Another one is configured as a > > > device with gadget zero drivers. Host and device are interfaced in > > > super speed mode. testusb application is used on host side test > > > features. Both boards uses synopsis USB controller. > > > > > > Observation: > > > With super speed interface, all test cases are passing except test 13 > > > ( set/clear 1 halts). The error is shown in this test case > > > usb 2-1: verify_still_halted failed, iterations left 0, status -71 > > > (not -32) usbtest > > > 2-1:3.0: halts failed, iterations left 999 > > > > > > Analysis: > > > Digging deeper into the issue I have below observation: > > > --- Same test case with the same setup is passing for the high-speed interface > > > --- I tried the same test case only on out endpoint. Seeing the same > > > issue. The test will try to send 1024 byte of data (bulk out) packet. > > > This will lead two trb, one for 1024 byte data followed by zero > > > packets. For first data packet, I am getting EPIPE(-32) error which is > > > expected. Same also expected for zero data packet. But for this trb, > > > we are getting EPROT (-71) which is failing test case. > > > > > > Can we use usbtest for super speed interface? > > > > Yes. It should work. If it doesn't, something is wrong. > > > > > What is expected error code if we will try to send a data packet on a > > > halted pipe? I checked spec but didn't get any specific mention. > > > > The expected error code is -32 (-EPIPE). It is documented in > > Documentation/driver-api/usb/error-codes.rst. > > > > Alan Stern > > > > > Please help regarding the issue. Any input will be highly appreciated. -- 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
