Hi, I am really wondering why this fujitsu-tablet module seems to replicate some of the behavior of the fujitsu-laptop module in the mainline kernel? At least for me I have now two modules caring for the brightness control. But only this fujitsu-tablet module allows me to use my additional buttons on the T2020. But I need two modules for supporting all required things.
In the end this is a bit strange, I think it would be more useful if this one could extend the module inside the kernel. Is there any work going on in this part? I could do it, however I never contributed to mainline and not sure if there is any progress made in this direction? Two conflicting drivers is not really a good idea…
The kernel module ./drivers/platform/x86/fujitsu-laptop.c uses the following ACPI devices:
"FUJ02bf", "FUJ02B1", "FUJ02E3"
and fujitsu-tablet uses "FUJ02BD and "FUJ02BF". Not sure if there is a case-sensitivity between FUJ02bf and FUJ02BF though. My T2020 does have FUJ02bf and FUJ02e5 if I call lspnp, but the kernel log also shows FUJ02B1 handled by fujitsu-laptop as an input device (not sure what buttons those are really at the moment)
Thanks for giving me information about the current progress, if there is any?
Otherwise I will probably start to take efforts merging this two drivers and commiting to mainline, because for me it's not really a good situation (if fujitsu-tablet is loaded additionally, i have some strange issues sometimes, maybe because there is some overlapping between the drivers?).
I've looked at it several times. I think brings no benefits. Both modules (and the supported hardware) are quiet different and share no code. The fujitsu-laptop module defines FUJ02BF in the PNP struct (for auto-loading I think), but don't has code to support this device. I'll continue monitoring it, but it is not being considered a priority (for me). But feel free to work on this, I'll help as best I can.
> if fujitsu-tablet is loaded additionally, i have some strange issues sometimes
Please check if apanel is loaded into the kernel. If so, unload and blacklist this module.
Ah sorry, I didn't look so close. Anyway, I think if this driver works good it should be put into mainline.
The apanel driver tries to load on boot and fails, but I blacklisted it to be sure.
I have found that problems also occur without this driver (I had a strangely stuttering mouse cursor sometimes, sometimes either resume or suspend freezes the system, the pen does sometimes need a double console switch until it works - not any easily reproduceable issues :-/ )
There are also some issues with the ACPI tables - sadly Fujitsu didn't entirely follow the specifications… probably nothing this driver is going to solve though.