Xorg 1.19 with libinput not recognizing all buttons of the EVGA x5l mouse.
 

Good Day all.

So I am having an issue here with my gaming mouse. I have an EVGA X5L (http://www.evga.com/products/product...901-X1-1051-KR) that for some reason, Xorg using libinput is not recognizing the two thumb buttons on the right hand side of the mouse. I am a lefty and have the mouse configured (via gnome settings) to function as a left handed mouse.

When I was using evdev with Arch and Debian SID in 32 bit mode a year ago, all buttons were recognized. It also seems that in Debian Unstable 64bit with Wayland and libinput, all buttons are recognize. Unfortunately I use that environment for development and have no games installed any more (how would I test this from cli!?).

Anyone have this problem? I would like to use libinput since it is the default and evdev is being depricated.

I can provide information as needed. Unfortunately, the Arch Wiki points to using evdev for configuration and everything else with libinput and Xorg show howto setup touchpads.:(


**Edit**
Need the EVGA Software to remap buttons. This currently requires MS Windows.

Try using the 'xev' utility to determine if the buttons of interest are at least generating events when pressed. It should show the button being pressed. Hopefully, it's just a button-mapping issue...

https://wiki.archlinux.org/index.php...ton_re-mapping

From 'man libinput'

thanks for the quick reply. Have not had a chance to test with xev. I will give it a try today. The Libinput + Xorg stuff is confusing since some of stuff in X is still there while other protions appear to be deprecated.

True, but I/others can probably guide you through the button mapping so long as the button events are being generated/enumerated in the first place.

Update:

It looks like XEV is no longer taking input from the two buttons in question, not even in Debian SID. I will take it to work tomorrow and use the EVGA mouse driver software to see if it can unlock those buttons since it is "Windows" only.

I will post an update to let everyone know how it turns out.

Need the windows tool.
 

Need the windows tool.

So took it to work today and used the EVGA software to apply a firmware upgrade and re-enable the buttons.

Maybe my next project will be to create a *nix tool like the one guy did for Roccat. Seriously though... firmware for a mouse!?

I guess peculiar top-end configurable input devices like this. Anyway, reads like progress?

Yes, that would be a good idea if not too arduous.

Yeah it is progress. The caveat is that you cannot map those buttons to the same function as any other button on the mouse or else it reads as the same button. So Yesterday I set buttons 6 and 7 to match the functions for buttons 8 and 9 (Browser forward/back). XEV showed that button 6 was button 8 and button 7 was button 9. So I had to map the buttons to profile switcher up and profile switcher down. The profile switches happen in software so this has not adverse effects on my GNU/Linux machines since it does not run the EVGA software. They are now reading as buttons 6 and 7 respectively.

So yes, progress.

I am probably going to have to fire up wireshark on my work PC and analyze the packet traffic to see what USB commands the software sends to the mouse and vise versa when I write specific commands.

Well done. I can see a wiki page or how to coming. :)

The adding an item to the mouse configuration wiki is a good idea. Thanks for the suggestion. I'll do it for Debian and Arch.