raleighlittles / tourbox_neo_linux_driver Goto Github PK

Hey it's me again. I was wondering what I could do to resolve this issue. It happens in Arch as well as Linux Mint.

I've been trying to get my elite controller to work and have discovered something unusual:
I can read input data from /dev/ttyACM0 initially when plugged into USB, however, if I disconnect the device and reconnect it, I can no longer read data. Only if I boot into Windows and launch the tourbox driver, then I can boot back into Linux and read from the device. Note that simply rebooting Linux does not seem to fix the issue.

Simple reproduction:

# Connect controller
sudo stty -F /dev/ttyACM0 raw
sudo cat /dev/ttyACM0 # and press some inputs
# Disconnect controller and repeat

The vibration motor also does not engage on input so I expect that this is triggered by writing some data to the device. I've been looking at the data produced in Wireshark, however, I expect this is overkill since this also includes all the USB header information as well, although I'm not sure if there is a simpler sniffing method.

So I read that you can remap the buttons using Xmodmap. I've read up on the Arch wiki on how to use it, but it only seems to relate to the keyboard. I've also run xev just to get an idea as to the input, but some of them read the same (dial clockwise, knob counterclockwise). Anybody know where to point me for mapping this bad boy?

• Refactor uinput_helper.hpp -- stop passing the file descriptor around so much.
• Check return value of read(2), ioctl, write and all other syscalls
• Don't use array of single value
• Make keymap const (then use at instead of operator[])

Whenever I launch this, all the command prompt gives me is:

Scroll wheel clicked
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
00 �
01 �
0
�
Scroll wheel clicked

And it keeps looping.

Upon successfully installing the driver, my programmed Logitech MX Master 3 (LogiOps) key binds revert to their default config despite the logid cfg file still having my bind configurations. Restarting and shutting down the system does not improve the issue.

Checking dmesg, the virtual device seems to stay despite already exiting the program.

Hey whenever I try to use the make command, here's what happens:
Consolidate compiler generated dependencies of target TourBox_Linux_Driver [ 50%] Building CXX object CMakeFiles/TourBox_Linux_Driver.dir/main.cpp.o In file included from /home/wayne/Tourbox_Linux_Driver/cpp/main.cpp:23: /home/wayne/Tourbox_Linux_Driver/cpp/uinput_helper.hpp: In function ‘void registerKeyboardEvents(int)’: /home/wayne/Tourbox_Linux_Driver/cpp/uinput_helper.hpp:87:21: error: loop variable ‘keyType’ creates a copy from type ‘const std::pair<const KeyType, int>’ [-Werror=range-loop-construct] 87 | for (const auto keyType : keyMap) | ^~~~~~~ /home/wayne/Tourbox_Linux_Driver/cpp/uinput_helper.hpp:87:21: note: use reference type to prevent copying 87 | for (const auto keyType : keyMap) | ^~~~~~~ | & cc1plus: all warnings being treated as errors make[2]: *** [CMakeFiles/TourBox_Linux_Driver.dir/build.make:76: CMakeFiles/TourBox_Linux_Driver.dir/main.cpp.o] Error 1 make[1]: *** [CMakeFiles/Makefile2:83: CMakeFiles/TourBox_Linux_Driver.dir/all] Error 2 make: *** [Makefile:91: all] Error 2