Thanks Alan for your incredible insight and willingness to teach! I had spent about a month and half reverse engineering the way Test 8 works and what its parameters meant but to no avail. Thanks once again! About toggling, Maxim 3420e automatically clears all data toggles during chip or USB bus reset. The CPU clears individual endpoint data toggle under two conditions: 1. The host issues a Set_Configuration request. 2. The host issues a Clear_Feature (endpoint stall) request. I was clearing the data toggle under condition #2 but not #1. Fixing this and setting -v 64 enabled the driver to pass Test 8!! I'm very grateful for the help you provided, Alan! Now, I'm off to debugging the rest of the failing tests! [root@sunburst testusb]# ./testusb -a -c 50 -v 64 unknown speed /proc/bus/usb/002/029 param->iterations 50 /proc/bus/usb/002/029 test 0, 0.000005 secs param->iterations 50 /proc/bus/usb/002/029 test 1, 4.489105 secs param->iterations 50 /proc/bus/usb/002/029 test 2, 3.404535 secs param->iterations 50 /proc/bus/usb/002/029 test 3, 2.496662 secs param->iterations 50 /proc/bus/usb/002/029 test 4, 1.739709 secs param->iterations 50 /proc/bus/usb/002/029 test 5, 149.778182 secs param->iterations 50 /proc/bus/usb/002/029 test 6, 108.884599 secs param->iterations 50 /proc/bus/usb/002/029 test 7, 73.637804 secs param->iterations 50 /proc/bus/usb/002/029 test 8, 53.577384 secs param->iterations 50 /proc/bus/usb/002/029 test 9 --> 33 (Numerical argument out of domain) param->iterations 50 /proc/bus/usb/002/029 test 10 --> 75 (Value too large for defined data type) param->iterations 50 /proc/bus/usb/002/029 test 11, 0.199343 secs param->iterations 50 /proc/bus/usb/002/029 test 12, 0.199452 secs param->iterations 50 /proc/bus/usb/002/029 test 13 --> 22 (Invalid argument) param->iterations 50 /proc/bus/usb/002/029 test 14 --> 74 (Bad message) Edwin Mercado On Sat, Mar 3, 2012 at 10:54 AM, Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote: > On Sat, 3 Mar 2012, Alan Stern wrote: > >> Subsequent tests probably behaved exactly the same, except that the >> endpoint data toggle values were not synchronized. > > I forgot to mention... The toggles _should_ have been synchronized, > since each test is preceded by a Set-Interface request. The fact that > they aren't indicates a bug in your UDC driver; each time an endpoint > is newly enabled, its data toggle should be set back to 0. > > Alan Stern > -- 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
