Comment 5 for bug 703020

Revision history for this message
Herton R. Krzesinski (herton) wrote :

I had a look into the DSDT from the attached acpi tables, and fujitsu-laptop should be handling hotkeys, and indeed from posted dmesg it loads succesfuly and register the input device for hotkeys:

[ 29.212978] input: Fujitsu FUJ02B1 as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/device:43/FUJ02B1:00/input/input4
[ 29.213034] ACPI: Fujitsu FUJ02B1 [FJEX] (on)
[ 29.213848] input: Fujitsu FUJ02E3 as /devices/LNXSYSTM:00/LNXSYBUS:00/FUJ02E3:00/input/input5
[ 29.213888] ACPI: Fujitsu FUJ02E3 [FEXT] (on)
[ 29.214569] fujitsu-laptop: BTNI: [0xf0001]
[ 29.214808] fujitsu-laptop: driver 0.6.0 successfully loaded.

The input device which receives input events is input5, FUJ02E3, which is Device (FEXT) in DSDT.

Probably fujitsu-laptop needs an update to work with this machine, can be the case of newer hardware which needs more handling in the code.

My bet is that on acpi_fujitsu_hotkey_notify function, switch (irb & 0x4ff) doesn't match any option on switch, so it goes to default which sets keycode = -1 and thus no keycode is reported. I think this make sense, probably this laptop reports different/new codes.

To confirm, we need to enable CONFIG_FUJITSU_LAPTOP_DEBUG, then load the module with parameter debug=0xf to catch all debug messages, press hotkey and check dmesg output. Of interest will be the messages with "Unknown GIRB result"..., which then means that indeed keycode was set to -1 and nothing reported.

Can you handle rebuilding the module with CONFIG_FUJITSU_LAPTOP_DEBUG=y? Otherwise I can provide a test kernel, so you can enable debug and report back the dmesg.