On Fri, Mar 10, 2006 at 10:33:45AM -0500, Alan Stern wrote: > On Thu, 9 Mar 2006, Marc Singer wrote: > > > My gadget driver is failing all of the read tests. The trace from > > usbmon looks like this: > > > > ea9379c0 3948991485 S Co:022:00 s 01 0b 0000 0000 0000 0 > > ea9379c0 3948999578 C Co:022:00 0 0 > > e21dcf40 3949000502 S Bi:022:01 -115 512 < > > e21dcf40 3949001572 C Bi:022:01 -121 0 > > > > My cooked version > > > > 0:00.000000 # 1 s> Co 021:00 s 01 0b 0000 0000 0000 ( HtD st in > > SET_INTERFACE ) --- 0 > > 0:00.008419 # 1 <c Co 021:00 --- 0 0 > > 0:00.010549 # 1 s> Bi 021:01 EINPROGRESS --- -115 512 < > > 0:00.011410 # 1 <c Bi 021:01 EREMOTEIO --- -121 0 > > > > The dmesg on the host shows this: > > > > usbtest 2-2:3.0: TEST 2: read 512 bytes 1000 times > > usb 2-2: test2 failed, iterations left 999, status -121 (not 0) > > > > The question I have is where the EREMOTEIO could be coming from. > > EREMOTEIO indicates a short packet response. > > > My interpretation of this sequence is that the host asks for the data > > with the EINPROGRESS message and the UDC is responding with a fault. > > No. The UDC is responding with a packet of length 0 (as you can see from > the usbmon line) whereas the host requested a packet of length 512. > That's why it's called a short response.
OK. Now I get it. I was looking for the code that generate the EREMOTEIO error so that I could inspect the UDC response. > > There is an EREMOTEIO in the zero driver that comes from a buffer > > overrun. printk's show that that it isn't the source. > > > > I can tell that the UDC driver is sending at least one packet. On the > > first interrupt, I load a second packet into the FIFO and tell the UDC > > to go, but it never appears to transmit this packet. Moreover the > > host seems to have received an error so it is likely that the UDC is > > just idle pending a request. > > No, it is actually sending a response. If the UDC were sending NAK (data > not ready), the test would have timed out and given you -ETIMEDOUT, not > -EREMOTEIO. Are these responses generated by the HCD? ------------------------------------------------------- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642 _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
