On Thu, Jun 06, 2002 at 08:27:34PM +0200, Georg Acher wrote: > On Thu, Jun 06, 2002 at 10:17:38AM -0700, David Brownell wrote: > > This info needs to get to HID and/or usb-uhci-hcd folk, not me ... :) > > > > > > Marcelo de Paula Bezerra wrote: > > >On Thu, 2002-06-06 at 00:55, David Brownell wrote: > > > > > > > > >>Modify the dbg statement -- it's just a regular printf format, > > >>so add a "%d" and pass the return value in the argument list. > > >> > > >>- Dave > > > > > >It returned -6 (-ENXIO) > > > > > >And usb-uhci-hcd also spits those messages: > > >Jun 6 00:56:06 thunderbird kernel: usb-uhci-hcd.c: ENXIO (Control) > > >80000300, flags 0, urb efc87540, burb efc874c0, probably device driver > > >bug... > > >Jun 6 00:56:06 thunderbird kernel: input: USB HID v1.10 Keyboard > > >[045e:002b] on usb-00:04.2-2.1 > > >Jun 6 00:56:06 thunderbird kernel: usb-uhci-hcd.c: ENXIO (Control) > > >80000380, flags 0, urb efc876c0, burb efc874c0, probably device driver > > >bug... > > >Jun 6 00:56:06 thunderbird kernel: hid-core.c usb_hid_configure: > > >reading report descriptor failed. (-6) <6>hub.c: new USB device > > >00:04.2-2.3, assigned address 4 > > >Jun 6 00:56:06 thunderbird kernel: usb-uhci-hcd.c: ENXIO (Control) > > >80000400, flags 0, urb efc879c0, burb efc87940, probably device driver > > >bug > > > Ok, this is the usual ENXIO-problem: > > <bold&blink> > UHCI cannot queue control transfers! > </(bold&blink)>
Well, the HID code tries quite hard (and has a lot of complexity added because of that) to issue Control URBs one by one. But there still might be a bug in it ... I'll take another look. > usb-uhci* has this error message for aeons, uhci* AFAIK is silent in this > case. So it's HID (or something else, like the hotplug manager) which has > two parallel transfers to the same pipe running, and that is a quite new > behavior. The UHCI-drivers can't queue control transfers since their > invention (surely a lack in completeness), but that should be well known. So > please don't kill the messenger ;-) > > > When usb-uhci survives the evaluation, I will add the control queuing, but > for now I don't want to introduce a new complex code path that can seriously > affect USB's most important transfer mode. > > -- > Georg Acher, [EMAIL PROTECTED] > http://www.in.tum.de/~acher/ > "Oh no, not again !" The bowl of petunias > > _______________________________________________________________ > > Don't miss the 2002 Sprint PCS Application Developer's Conference > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > _______________________________________________ > [EMAIL PROTECTED] > To unsubscribe, use the last form field at: > https://lists.sourceforge.net/lists/listinfo/linux-usb-devel -- Vojtech Pavlik SuSE Labs _______________________________________________________________ Multimillion Dollar Computer Inventory Live Webcast Auctions Thru Aug. 2002 - http://www.cowanalexander.com/calendar _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
