On Sun, 4 Sep 2005, Chris Clayton wrote:
> On Saturday 03 Sep 2005 19:17, Stephen J. Gowdy wrote:
> > Make sure you have USB Legacy or a similar option off in your BIOS.
>
> Thanks for the suggestion Stephen, but it didn't make any difference.
>
> I've been doing some more digging around and I believe the problem was
> introduced
> in 2.6.13-rc1. 2.6.12{,.1,.2,.3,.4,.5,.6} all work fine but I get the
> keyboard lock-up
> with 2.6.13-rc1. I've also come across another symptom. Occasionally, a
> keypress will repeat as if the key is being held down. The only way to stop it
> is to unplug the keyboard. As before.once the keyboard is plugged in again, it
> works without problem.
>
> I've also got some more diagnostics by "enabling" some code that was
> previously
> guarded by an #ifdef 0/#endif pair. An extract from /var/log/debug is
> attached.
It would help if you included _all_ the kernel log messages in that file,
not just the debugging messages.
> The USB error message that I mentioned in my original post is at line 329 of
> this
> log, but is now accompanied by additional diagnostics. Does this mean anything
> to anyone, please? The error is immediately preceded by something about Babble
> and something being Stalled.
Those messages mean the keyboard sent back a reply packet that was longer
than it was supposed to be. More accurately, the keyboard's internal hub
did this. The computer asked it for a 1-byte packet, and the response was
more than 1 byte. This was on the interrupt-in endpoint. Note that
according to the information you posted from lsusb, the maxpacket size for
that endpoint is 1, so the hardware is definitely misbehaving.
> I've also backed out one or two at a time, all the uhci related patches (plus
> any
> dependent patches to host or core) that went into -rc1 but that hasn't made
> any
> difference. Bearing in mind the new diagnostics, does anyone have a suggestion
> about where I should look next, please?
There are lots of other things you can try. For example, what happens
when you plug the keyboard into a different computer running the same
versions of Linux? What happens if you don't plug the keyboard in until
well after the bootup procedure is finished?
Assuming you another means of getting in to your computer (logging in over
a network, for instance), you can use the usbmon facility to trace the
data being sent to/from the keyboard. Read about usbmon in
Documentation/usb/usbmon.txt in the kernel source. This patch will help:
http://marc.theaimsgroup.com/?l=linux-usb-devel&m=112561058119237&w=2
Alan Stern
-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-users
