Controller connects successfully, but generates no input

[nassek]

I installed the driver by following the readme, controller is able to connect and rumbles but it doesn't generate any keypresses. I tried running Steam Big Picture to check it. I also ran evtest; output is empty after the line Testing ... (interrupt to exit).

System info:
Xubuntu 16.04, Linux 4.13.0-43-generic #48~16.04.1-Ubuntu SMP Thu May 17 12:56:46 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

dmesg output: here

[atar-axis]

hi @nassek! I'm sorry to hear that.
The problem seems to be the battery initialization:

[ 2166.217028] xpadneo 0005:045E:02FD.000B: battery initialization failed
[ 2166.217115] xpadneo: probe of 0005:045E:02FD.000B failed with error -26908

I think I may already know where the problem is, will get back as soon as I fixed it (or if I need further debugging information).

[atar-axis]

Okay new version online, could you please give it a try? ;D

Please perform the following steps to do so:

• run ./uninstall.sh (optional, to remove the last version)
• sudo modprobe -r hid_xpadneo to unload the old version from the kernel
• git pull to get the new version
• ./install to install the new version

If the problem persists, it would be helpful if you enable debug output - if it isn't already (see https://github.com/atar-axis/xpadneo#troubleshooting).

Your dmesg was very helpful btw.

Thank you for getting back in advance!

[nassek]

It's working now! Thanks for fixing the issue so fast :D
I still have some button mapping problems, but I think I'll be able to sort them out myself.

[atar-axis]

Ah, wonderful - you are welcome! :)

Regarding the button mapping:

Could you please run evtest /dev/input/event<#> and tell me which buttons are wrong - if any?
E.g. is it BTN_A which is reported if you are pressing Button A?

Furthermore the behaviour of jstest /dev/input/js<#> would be very helpful too.

I am always very interested in wrong mappings because this is one of the main goals of this driver - something we aim to fix :)

If evtest and jstest behave the right way but Steam does not, then the problem may be the one which is desribed in #27 - which is already fixed (somehow, it's more a workaround) but not yet accepted by @kakra (who reported the bug)

[atar-axis]

--deleted comment--

[kakra]

@nassek Please, under all circumstances consider the pitfalls outlined in the README regarding button mapping. There's all sorts of corrections in different layers applied or not applied depending on versions of libraries used. The one and only fix is to first get rid of all "corrections" done at other layers.

[atar-axis]

you can also give this branch a try: https://github.com/atar-axis/xpadneo/tree/input_version (git clone https://github.com/atar-axis/xpadneo.git -b input_version)

it includes the fix/workaround I mentioned above, it at least works on my machine and I would really appreciate if you could confirm that it is working on yours too.

[nassek]

First of all, in my case only Steam's mappings were messed up - evtest and jstest reported correct keypresses.
I tried the input_version branch, and had to change one line to get the module to compile, as in pull request #36

After installing the new module, Steam handles keypresses correctly.

[atar-axis]

perfect ;) thanks for that!

[kakra]

So my proposed version hack idea works around this whole mess successfully... Nice. :-)

[atar-axis]

@kakra Looks like it - wonderful :D I will therefore add the parameter, merge the version back to master and close your issue #27. Thanks to both of you!

[atar-axis]

@nassek May I ask you to test the new master one last time? I merged your pull request into input_version, added some other things and merged everything back to master then.
Thank you very much :)

[nassek]

Updated to current master c2885f4, it works as supposed to. Good job guys ;)

[atar-axis]

Thanks!