Bug 670729

not sure why this is filed against kernel.
sounds more like an X key mapping problem ?

I think it's a kernel problem...
Well, how to check X key mapping?

Thanks for the bug report.  We have reviewed the information you have provided above, and there is some additional information we require that will be helpful in our diagnosis of this issue.

Please attach

* your X server config file (/etc/X11/xorg.conf, if available, and all files in /etc/X11/xorg.conf.d/ directory),
* X server log file (/var/log/Xorg.*.log),
* the output of
  xkbcomp -xkb $DISPLAY -
* output of
  evtest /dev/input/<find out in Xorg.0.log which device corresponds to your keyobard>
, and
* system log (/var/log/messages)

to the bug report as individual uncompressed file attachments using the bugzilla file attachment link above.

We will review this issue again once you've had a chance to attach this information.

Thanks in advance.

Created attachment 512458 [details]
/etc/X11/xorg.conf

Created attachment 512459 [details]
/etc/X11/xorg.conf.d/00-nvidia.conf

Created attachment 512460 [details]
/etc/X11/xorg.conf.d/00-system-setup-keyboard.conf

Created attachment 512461 [details]
the output of   xkbcomp -xkb $DISPLAY -

Created attachment 512462 [details]
/var/log/Xorg.*.log

Created attachment 512463 [details]
/var/log/messages

Created attachment 512466 [details]
evtest /dev/include/event4

(In reply to comment #10)
> Created attachment 512466 [details]
> evtest /dev/include/event4

please reattach the evtest output showing the caps lock key being pressed. this output here only shows alt+tab and ctrl+c to terminate evtest. The caps lock key needs to be in the output for further debugging.

Created attachment 516290 [details]
reattached evtest log

evtest /dev/input/event4 log.

interesting. not a kernel problem, that shows it normally. keymap you attached also has the correct mapping, not sure where that mail key comes from. What keysym does xev show when you press Caps?

KeyPress event, serial 32, synthetic NO, window 0x2e00001,
    root 0x15d, subw 0x0, time 289358, (688,264), root:(689,351),
    state 0x2, keycode 163 (keysym 0x1008ff19, XF86Mail), same_screen YES,
    XLookupString gives 0 bytes: 
    XmbLookupString gives 0 bytes: 
    XFilterEvent returns: False

KeyRelease event, serial 32, synthetic NO, window 0x2e00001,
    root 0x15d, subw 0x0, time 289358, (688,264), root:(689,351),
    state 0x2, keycode 163 (keysym 0x1008ff19, XF86Mail), same_screen YES,
    XLookupString gives 0 bytes: 
    XFilterEvent returns: False


When I press CapsLock, xev prints these to stdout. But the keysym is XF86Mail.

ok, this is just weird now. evtest shows the right keycode (58), but the keycode seems to get changed somewhere to 163. That is the real issue here but I have no idea how this can happen. The keycode is about the only thing we leave alone and just forward on.

Sorry, really don't know how this can happen. Does it happen in a plain X server? 

sudo init 3, log in, run xinit -- to get a plain X server with xterm, then run xev. 
quit xterm, sudo init 5 to get back to normal

and please confirm that if you're running evtest and xev at the same time, they show the different keycodes (X keycode should be evtest keycode + 8)

Created attachment 516465 [details]
xev log in plain X

the xev log in a plain X.

BTW, when i press Capslock, i can type capital letters normally.

how many devices do you see in /proc/bus/input/devices (attach that file and the output of xinput --list), then please attach the output of xinput --test-xi2 when pressing capslock and see if evtest sees events on any other device when hitting the key.

my _guess_ here is that this is broken or just weird hardware. the caps lock comes through normally but is interleaved with that mail key so maybe the hardware triggers that key too whenever you hit caps?

Created attachment 516730 [details]
devices file.

I asked keyboard manufacturer about that but they didn't answer, because they think Linux is not popular.

Created attachment 516731 [details]
xinput --list

Created attachment 516733 [details]
xinput --text-xi2

There is a LED light on the key. Maybe about the light?

yeah, this is a broken device. in the xinput test-xi2 output you can see the capslock key press (keycode 66) from device 11 (the AT Translated Set 2 keyboard) immediately followed by the mail key (keycode 163) from device 13 (Topstar Laptop extra buttons). This is either a broken hardware or broken hardware design.

you can work around this by adding a xorg.conf.d snippet along the lines of
Section "InputClass"
   Identifier "stop mail button"
   MatchProduct "Topstar Laptop extra buttons"
   Option "Ignore" "on"
EndSection

This disables the device completely. Note that you'll likely lose other buttons that this device sends too (volume up/down for example).

The only other (run-time) fix is to run xkbcomp -xkb $DISPLAY some_file.xkb, then edit some_file.xkb, remove the line setting I163 to XF86Mail and loading this file with xkbcomp -xkb some_file.xkb $DISPLAY. This needs to be done after gnome applies the keyboard layout however.

closing as CANTFIX, requires the above workarounds. sorry.

Thank you for your helping. I will report this to the manufacturer.