Hi, Am Wed, 16 Sep 2009 13:14:05 +0100 schrieb David Vrabel <david.vrabel@xxxxxxx>: > Andreas Kemnade wrote: > > Hi, > > > > Am Tue, 15 Sep 2009 18:09:31 +0100 > > schrieb David Vrabel <david.vrabel@xxxxxxx>: > > > >> Andreas Kemnade wrote: > >>> Here is some debug output. > >>> [ 4086.784579] uwb_rc uwb0: SET-DRP-IE failed: time out (wa) or > >>> invalid ie data (whci) (7) > >> Every SET-DRP-IE radio controller command is failing. This means > >> the format is wrong. Add your HWA to hwarc_id_table at the end of > >> drivers/uwb/hwa-rc.c and make sure .driver_info = > >> WUSB_QUIRK_WHCI_CMD_EVT. > >> > > Now I get further with that quirk. After writing -1 and then 15 to > > beacon and then 15 2 to scan the devices seem to talk to each other, > > but still with errors, see the log file, the diff for the quirk is > > also appended. > > The "hwa-hc 2-9:1.0: HWA: unknown notification 0x94, 3 bytes; > discarding" messages are harmless and can be ignored. It's an > notification saying the HWA has adjusted it's BPST. > > I can't tell why it's why it's not connecting correctly or why the > first control transfer fails without an air trace. > I have looked again into the logs and wonder why you think the first control transfer fails. It gives EINPROGRESS as error which does not make sense at all IMHO in blocking io. I would understand things like ETIME, EPIPE, ENOENT but not EINPROGRESS. So perhaps the control transfer was successful, but usb_control_msg did for some reason not wait for the result. I have never seen EINPROGRESS on usb_control_msg() on wired USB, either. So I think there are still issues to look at without rf sniffing equipment. But where does EINPROGRESS come from? Greetings Andreas Kemnade -- Andreas Kemnade (akemnade@xxxxxx) Universität Bremen Technologiezentrum Informatik - [WearLab] Am Fallturm 1 28359 Bremen http://www.tzi.de/ http://www.wearlab.de/ Voice: +49 (421) 218-8919 -- 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
