Acer 6920: Some scancodes not/wrong registered

Bug #407940 reported by Alexander Hartl
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
udev (Ubuntu)
Fix Released
Medium
Martin Pitt

Bug Description

Binary package hint: udev

I have an Acer Aspire 6920G. This notebook has on its left side a "CineDash" panel (see http://www.acer.com/gemstoneblue/eng/) on which there are some keys with scancodes that udev doesn't know yet:
Acer Arcade: e012
Arcade - Back button: e01e
Hold: e04a

Rewind: e003
Fast-forward: e009
(these two keys are NOT next and previous track)

The $ and € keys above the arrow keys also don't work - pressing them doesn't give me an "Unknown key pressed" entry in syslog and they're displayed neither in showkey -s nor in showkey -k. Scancodes:
$: e033
€: e034

Additionally the bluetooth toggle-off switch (not toggle-on) currently increases the brightness
scancode: e054
keycode: 225

It would be nice if this could be changed in a future release, of course it isn't anything urgent.

Best regards,
Alex

Related branches

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

Martin, I haven't caught up with the stuff that moved in from udev-extras yet - do you know what we do about this kind of thing?

Changed in udev (Ubuntu):
assignee: nobody → Martin Pitt (pitti)
importance: Undecided → Low
status: New → Confirmed
Revision history for this message
Martin Pitt (pitti) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better.

Please exercise the steps at https://wiki.ubuntu.com/Hotkeys/Troubleshooting to determine DMI information and on which device these hotkeys occur, and attach the collected information. Thanks!

Changed in udev (Ubuntu):
importance: Low → Medium
status: Confirmed → Incomplete
Revision history for this message
Alexander Hartl (alx321) wrote :

alex@alex-note:~$ killall gnome-settings-daemon gnome-power-manager
alex@alex-note:~$ xev | sed -n 's/^.*state \([0-9].*\), keycode *\([0-9]\+\) *\(.*\), .*$/keycode \2 = \3, state = \1/p'
keycode 36 = (keysym 0xff0d, Return), state = 0x0
** Pressing all new keys, €, $ and bluetooth-switch, but nothing happens **
^C
alex@alex-note:~$ /lib/udev/findkeyboards
AT keyboard: input/event5
alex@alex-note:~$ sudo /lib/udev/keymap input/event5 > /tmp/orig-map.txt
alex@alex-note:~$ sudo /lib/udev/keymap -i input/event5
Press ESC to finish
** Pressing new keys but nothing happens **
** Pressing €, $ **
scan code: 0xB3 key code: euro
scan code: 0xB4 key code: dollar
** turning bluetooth on and off **
scan code: 0xD4 key code: bluetooth
scan code: 0xD9 key code: brightnessup
scan code: 0x01 key code: esc
alex@alex-note:~$ cat /sys/class/dmi/id/sys_vendor
Acer
alex@alex-note:~$ cat /sys/class/dmi/id/product_name
Aspire 6920
alex@alex-note:~$ udevadm info --export-db > /tmp/udev-db.txt

I just found out that the mapping for the bluetooth issue is already done in the file acer-travelmate_6292.
The file for Aspire 6920 should look something like that:

0xD9 bluetooth # (toggle) on-to-off
0x92 media # this should start the media player or so
0x9E back
0xCA hold # don't really know what to map this to
0x83 rewind
0x89 fastforward

Revision history for this message
Alexander Hartl (alx321) wrote :
Revision history for this message
Martin Pitt (pitti) wrote :

alex@alex-note:~$ sudo /lib/udev/keymap -i input/event5
** Pressing new keys but nothing happens **

OK, that means that these keys are on a different input device. Apparently you have something called a "Logitech USB receiver". Can you please try whether you see these keys on input/event6 (unlikely, but possible) or input/event7 (more likely)?

EUR, $, and the other keys are already correct, except for 0xD9 which should be "bluetooth", not "brightnessup". I can change that with the available data.

> The file for Aspire 6920 should look something like that:
>
> 0xD9 bluetooth # (toggle) on-to-off

Correct. I made the change in udev upstream git now.

> 0x92 media # this should start the media player or so
> 0x9E back
> 0xCA hold # don't really know what to map this to
> 0x83 rewind
> 0x89 fastforward

Where do these come from? You did not mention them in your testing above?

Revision history for this message
Alexander Hartl (alx321) wrote :

> OK, that means that these keys are on a different input device. Apparently you have something called a "Logitech
> USB receiver".

That's my Logitech wireless mouse.

> Can you please try whether you see these keys on input/event6 (unlikely, but possible) or input/event7 (more
> likely)?

No, they appear on none of them.

> EUR, $, and the other keys are already correct, except for 0xD9 which should be "bluetooth", not "brightnessup".

Yes, € and $ are correct but they don't work anyway...

> Where do these come from? You did not mention them in your testing above?

From syslog. For each unknown key pressed it throws messages like (e.g. Acer Arcade-key)

Aug 7 18:00:35 alex-note kernel: [ 305.874799] atkbd.c: Unknown key pressed (translated set 2, code 0x92 on isa0060/serio0).
Aug 7 18:00:35 alex-note kernel: [ 305.874807] atkbd.c: Use 'setkeycodes e012 <keycode>' to make it known.
Aug 7 18:00:35 alex-note kernel: [ 305.885437] atkbd.c: Unknown key released (translated set 2, code 0x92 on isa0060/serio0).
Aug 7 18:00:35 alex-note kernel: [ 305.885445] atkbd.c: Use 'setkeycodes e012 <keycode>' to make it known.

Revision history for this message
Martin Pitt (pitti) wrote : Re: [Bug 407940] Re: Acer 6920: Some scancodes not/wrong registered

Hello Alex,

Alexander Hartl [2009-08-07 16:05 -0000]:
> > OK, that means that these keys are on a different input device.
> > Apparently you have something called a "Logitech USB receiver".
>
> That's my Logitech wireless mouse.

Ah, then the input devices I mentioned are not related to the missing
keys.

> > Can you please try whether you see these keys on input/event6
> > (unlikely, but possible) or input/event7 (more likely)?
>
> No, they appear on none of them.

Can you check "sudo lsinput" and check if there's any other device
which isn't obviously something else (such as the power button), and
you didn't try yet?

> > EUR, $, and the other keys are already correct, except for 0xD9 which
> should be "bluetooth", not "brightnessup".
>
> Yes, € and $ are correct but they don't work anyway...

They can't work in X.org because they have a key code > 255, that's a
known limitation of X (there's a bug report for it, but I'm offline
right now and can't find it). Nothing we can fix in udev and the
keymaps, unfortunately.

> > Where do these come from? You did not mention them in your testing above?
>
> From syslog. For each unknown key pressed it throws messages like (e.g.
> Acer Arcade-key)
>
> Aug 7 18:00:35 alex-note kernel: [ 305.874799] atkbd.c: Unknown key pressed (translated set 2, code 0x92 on isa0060/serio0).
> Aug 7 18:00:35 alex-note kernel: [ 305.874807] atkbd.c: Use 'setkeycodes e012 <keycode>' to make it known.
> Aug 7 18:00:35 alex-note kernel: [ 305.885437] atkbd.c: Unknown key released (translated set 2, code 0x92 on isa0060/serio0).
> Aug 7 18:00:35 alex-note kernel: [ 305.885445] atkbd.c: Use 'setkeycodes e012 <keycode>' to make it known.

Ah, this at least proves that they should appear on _some_ input
device, and that it doesn't need a kernel fix. Did adding these keys
to above map actually make these keys work? If so, this would prove
that they are indeed normal keys on the input/event5 device (as I
would expect).

FYI, I'll be on holidays the next two weeks, so my next response will
be delayed a little.

Thanks!

Revision history for this message
Alexander Hartl (alx321) wrote :

> > Yes, € and $ are correct but they don't work anyway...

> They can't work in X.org because they have a key code > 255, that's a
> known limitation of X (there's a bug report for it, but I'm offline
> right now and can't find it). Nothing we can fix in udev and the
> keymaps, unfortunately.

Ah, ok. Then that's a question of time, don't actually need them anyway.

> Did adding these keys
> to above map actually make these keys work? If so, this would prove
> that they are indeed normal keys on the input/event5 device (as I
> would expect).

Yes, they do (except that fastforward and rewind don't do what they're supposed to in rhythmbox). Then the keys also appear in
sudo /lib/udev/keymap -i input/event5
which actually is no surprise as in 2.6 kernels the scancodes are always mapped to keycodes and mapped back if the scancode is requested, so they can't show up as long as they don't have a keycode. Or am I wrong there?

Revision history for this message
Martin Pitt (pitti) wrote :

Thanks! I added the remaining keys to udev trunk. However, "hold" doesn't exist, I mapped it to "stop" for now. Please let me know if that isn't appropriate.

Changed in udev (Ubuntu):
status: Incomplete → Fix Committed
Revision history for this message
Alexander Hartl (alx321) wrote :

Hold locks the panel, so that no keys can be pressed by mistake. "stop" is something different (there's another key for it).
Actually it isn't necessary to map this key to anything because it works by itself, but as it produces a scancode I think we can also map it to something.
Maybe some universal key like prog2 would be more appropriate.

Revision history for this message
Martin Pitt (pitti) wrote :

Alexander Hartl [2009-08-24 15:48 -0000]:
> Hold locks the panel, so that no keys can be pressed by mistake.

Ah, thanks for pointing this out.

> Actually it isn't necessary to map this key to anything because it works by itself

I agree. It shouldn't be overloaded with something else, it would just
be confusing. I removed it from the map.

Revision history for this message
Alexander Hartl (alx321) wrote :

That is, I think I've mixed that up with "stopcd" now. I don't know what "stop" actually is but it doesn't sound very accurate for me.

Revision history for this message
Alexander Hartl (alx321) wrote :

Ok.
Thanks a lot!

Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package udev - 147~-1

---------------
udev (147~-1) karmic; urgency=low

  FFE LP: #427356.

  * Update to GIT HEAD (pre 147 release):
    - worker signal mask corrected. LP: #407428.
    - database format change to avoid path length issues. LP: #377121.
    - multiple devices may not claim the same /dev names, except with
      symlinks
    - NAME="%k" produces a warning
    - symlinks to udevadm no longer resolve to the original command
    - rules updates. LP: #281335, LP: #407940, #420015, #426647.

  * Build-depend on gawk, since build fails with mawk.

  * Replace init scripts with Upstart jobs.
  * debian/control:
    - Add missing ${misc:Depends}
    - Bump build-dependency on debhelper for Upstart-aware dh_installinit

 -- Scott James Remnant <email address hidden> Tue, 15 Sep 2009 03:22:11 +0100

Changed in udev (Ubuntu):
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.