One more specific information : in bare OpenBox X session, none of these keys have a visible effect handled by Ubuntu, except XF86WLAN.
Also, this laptop was sold with Windows, and there all function keys work, including the brightness keys.
## kernel-level showkey test
I've tried https://wiki.ubuntu.com/Hotkeys/Troubleshooting but it seems outdated.
No gconf-editor, used dconf-editor instead. Tree paths do not exist, /usr/share/doc/udev/README.keymap.txt does not exist.
Anyway, following https://wiki.ubuntu.com/Hotkeys/Architecture I tested showkey.
Fn-* combinations produced scancodes, except for the 3 malfunctioning brightness key combinations.
## acpi-level test
Killing acpid would see it respawned immediately, so I moved the executable away for test.
Suspend button yields acpi events, but not brightness buttons.
One more specific information : in bare OpenBox X session, none of these keys have a visible effect handled by Ubuntu, except XF86WLAN.
Also, this laptop was sold with Windows, and there all function keys work, including the brightness keys.
## kernel-level showkey test
I've tried https:/ /wiki.ubuntu. com/Hotkeys/ Troubleshooting but it seems outdated. doc/udev/ README. keymap. txt does not exist.
No gconf-editor, used dconf-editor instead. Tree paths do not exist, /usr/share/
Anyway, following https:/ /wiki.ubuntu. com/Hotkeys/ Architecture I tested showkey.
Fn-* combinations produced scancodes, except for the 3 malfunctioning brightness key combinations.
## acpi-level test
Killing acpid would see it respawned immediately, so I moved the executable away for test.
Suspend button yields acpi events, but not brightness buttons.
$ sudo cat /proc/acpi/event
button/sleep SLPB 00000080 0000000a
battery BAT0 00000081 00000001
ac_adapter AC0 00000080 00000001
button/power PWRB 00000080 00000001
## Now what ?
This suggests that the problem should be fixed at kernel level, but it looks like kernel is never aware about those keypresses.
https:/ /wiki.ubuntu. com/Hotkeys/ Troubleshooting says :
> if there is neither an ACPI event nor an input event, this is probably also a kernel bug, though probably harder to diagnose (WMI, perhaps?)
I'm clueless now. Thank you for any hint.