[colug-432] Keybord input oddity... thoughts?

[Jeff Frontz]

That sounds like a "key UP" event is getting lost.  It's been a few
years since I've delved into the world of USB HID, but vaguely--
everything that happens on the keyboard is sent to the host as an
event.  Key is pressed -- key down event is sent.  Key is released --
key up event is sent.  It's then up to the host to implement things
like shift, shift-lock, and.... repeating keys.  So, if the host gets
a key down but never sees a corresponding key up, the host thinks
you're leaning on a key.

I can't remember how the X server works, but I suspect that it is
keeping a count for each key state-- key-down makes the count
increment; key-up makes the count decrement.  So, no amount of
key-presses would clear it.  You truly would need to pull your
keyboard (or otherwise cause it to disconnect from the USB) and
reconnect it.

You can dork around with which interface the X server uses (kbd vs.
event) -- that might help a bit.  I suspect, though, that your VM is
sending events to the wrong place every so often.

Jeff



On Sat, Mar 31, 2012 at 1:14 PM, William Yang <wyang at gcfn.net> wrote:
> All-
>
> I'm running a dual boot Dell E6510 "desktop replacement" laptop, Ubuntu
> 10.04.4 and Win7 professional (both 64 bit).  This component of my portable
> office is a near constant companion, giving me a big, high density screen
> and plenty of computing power to do whatever tasks I get asked to do.  Of
> late, I'm spending a lot of time (>5 hours/day, every day) on this device.
>  My split of Win vs Ubuntu differs based on the kind of work I'm doing, but
> for the past 2 months or so, it's been about 70% Windows to facilitate
> collaboration on documents with my clients.
>
> My problem...  Occasionally on the Ubuntu side, the system *acts* like a
> keyboard key gets stuck, but no amount of tapping or adjustment of the
> keyboard is sufficient to resolve it until I restart the X server
> (unplugging the keyboard on my laptop requires a screwdriver to get at the
> plug; I've restarted the X server via SSH a couple of times, and that's
> been sufficient to stop the problem, but if I don't have another device on
> the same network that's handy, I have to reboot the box because no keyboard
> input works).  It's a mild annoyance, but given how much time I spend on
> this thing, even mild annoyances are worth a moment's attention if it's
> straightforward to make them go away.
>
> Just from an anomoly identification standpoint, it's a little odd that I
> have *never* had a similar problem on the Win7 side.  That makes me think
> it could be software, rather than hardware.  A cursory search for bugs
> hasn't shown anything common... but I'm starting to wonder if there's an X
> input bug somewhere that I'm running into.  Anyone else ever run into
> anything similar?  Any suggestions as to approaches to resolve it?
>
> Thanks,
>
>        -Bill
>
> _______________________________________________
> colug-432 mailing list
> colug-432 at colug.net
> http://lists.colug.net/mailman/listinfo/colug-432