[Capture] Logitech G Pro Racing Wheel

[sixtysecondstosmash]

Device Name

Logitech G HUB PRO Racing Wheel for Xbox and PC (USB)

USB Vendor/Product ID

046d:272c

G Hub Version

2026.1.828335

Firmware Version

133.8.113

Captures Included

• G Hub initialization (ghub_init)
• FFB gameplay (ffb_gameplay)
• Rotation range sweep
• FFB strength sweep
• Damping sweep
• TRUEFORCE sweep
• Cold boot sequence
• Other (describe below)

Capture Notes

• ghub_init_gpro_pc : Started G Hub with wheel already connected, waited for the wheel to become active and clicked it to open its settings
• ffb_gameplay_gpro_pc : Probably less than 30s of track time on Brands Hatch Indy in AC Evo. Had to navigate through the menus to start Practice mode. Started in the pit, quick light rev, drove out of pits at partial throttle, swung wheel right to left a few times and induced a little oversteer, hit the kerb on the inside (right) of the first turn, went to the outside of the straight to hit the kerb/grass on the left, floored it and went straight to crash into the back wall. Exit game and stopped capture. This game has a constant TrueForce feedback with the engine sound, so there is possibly a lot of noise in this capture, but hopefully enough for the short playtime
• rotation_range_gpro_pc : Started G Hub with wheel already connected. Opened settings and typed in 0 to set min range, used trackpad to slowly sweep up from min, went past 400 or so before I decided to go back to stop at 360, used the trackpad to sweep up to 900 and stopped, continued on to 1080
• ffb_strength_gpro_pc : Started G Hub with wheel already connected. Opened settings and typed in 0 to set min range, used right arrow to increment each point until it reached max
• damping_gpro_pc : Started G Hub with wheel already connected. Opened settings and typed in 0 to set min range, used right arrow to increment each point until it reached max
• trueforce_gpro_pc : Started G Hub with wheel already connected. Opened settings and typed in 0 to set min range, used right arrow to increment each point until it reached max
• cold_boot_gpro_pc : Plugged in wheel USB and turned it on, heard Windows chime for connected USB and gave a generous 10s count, opened up G Hub, waited for the wheel to become active and clicked it to open its settings

Edit 1: Added another file to capture cold booting the wheel, switching to G923 mode and then opening G Hub. Then clicked the wheel to go to its settings.

I noticed that the G923 mode crashes the wheel in Linux/Bazzite and the system needs to be restarted to do anything, so hoping this can help with something.

Edit 2: Did another set of rotation sweeps at fixed intervals by typing in the max rotation and then turning the wheel left to lock, right to lock, back to center. I don't think turning the wheel captured anything, but I hope it's useful for something. The 360 sweep also includes a cold boot, the other two do not.

• rotation_range_360_gpro_pc
• rotation_range_900_gpro_pc
• rotation_range_1080_gpro_pc

/edits

Capture Files

gpro_captures.zip
cold_boot_gpro_to_g923.zip
rotation_range_sweep_v2.zip

Additional Information

Wheel and game settings used are the recommended on Logitech's official website https://support.logi.com/hc/en-gb/articles/8358055253271-In-Game-Settings-for-Pro-Wheels

Wheel Base

• Pro Mode
• FFB Filter: 5
• Dampener: Off
• TF Audio: 50
• Angle: 1080

AC Evo

• Steering Lock: 1080
• FFB Gain: 100%
• Dynamic Dampening: 50%
• Min Dampener: 0%
• Damper Gain: 30%
• Audio Based FFB Effects: 100%
• Curbs Effects: 30%
• Road Effects: 20%
• Tire Slips Effects: 20%
• ABS Effects: 10%

Side note: the suggested script didn't work for me, so I manually started and stopped the captures in Wireshark. Hopefully I did it right and you have what you need. Captures are zipped because github doesn't accept the files as is.

[mescon]

Thank you! I'll be working on this for a bit next week! Thanks for the detailed report!

[aderumier]

Hi,
I'm also a owner of logitech g pro (pc version) , do you still need capture ?

[mescon]

We've analyzed your captures thoroughly and used them to implement G Pro support, now merged to master.

Protocol Analysis

Your captures confirmed that the G Pro uses HID++ Feature 0x8123 FFB (same as the G920/G923), not dedicated-endpoint FFB like the RS50. Key findings:

• 36 HID++ features discovered on device 0xFF (vs 14 on RS50)
• Settings features share the same page IDs as RS50 (0x8138 Range, 0x8136 Strength, 0x8133 Damping, 0x8139 Trueforce, etc.)
• SET function numbers differ per feature: fn=2 for range/strength, fn=1 for damping, fn=3 for trueforce
• No traffic on interface 2 during gameplay (unlike RS50 which uses it for FFB)
• Multi-device structure: G Hub queries 4 device indices (0xFF, 0x01, 0x02, 0x05)
• G923 compat mode: Different feature index mapping, crashes in Linux (we'll investigate separately)

What's Implemented (on master)

1. Device IDs: Both Xbox/PC (046d:c272) and PS/PC (046d:c268) variants
2. FFB: Using the existing G920 HID++ 0x8123 code path, which provides FF_CONSTANT, FF_PERIODIC, FF_SPRING, FF_DAMPER, and more
3. Sysfs settings: Rotation range, FFB strength, damping, trueforce, brake force, FFB filter, profile/mode
4. Per-feature function mapping: G Pro-specific SET function numbers (fn=1 for damping, fn=3 for trueforce)
5. Sysfs naming: Attributes use wheel_* prefix (e.g., wheel_range) for multi-device compatibility

What's NOT yet implemented

• LIGHTSYNC LED control (needs more investigation on G Pro)
• Trueforce audio streaming (research documented in docs/TRUEFORCE_PROTOCOL.md)
• FFB command queue optimization (the JacKeTUs fork had overflow issues with 2700+ queued commands; may need community testing to reproduce and fix)

Testing needed

We don't have a G Pro to test with. If you could try the latest master and report:

1. Does the driver load and detect your wheel?
2. Does fftest /dev/input/by-id/*Logitech*event* produce force feedback?
3. Do sysfs settings work? (echo 900 | sudo tee /sys/class/hidraw/hidrawN/device/wheel_range)
4. Does FFB work in games?

That would help us validate the implementation.

Thank you again for the incredibly detailed captures. The rotation sweep v2 files and the G923 compat mode boot were especially useful.

[sixtysecondstosmash]

Appreciate the quick wok on this. I'm pretty sure I installed the module and enrolled it with MOK correctly. Maybe the below output can tell you if it's loaded properly or not. The system is Fedora 43, kernel 6.19.11-200.fc43.x86_64

dmesg output:

sudo dmesg | grep -i wheel
[  451.984624] usb 1-1.3: Product: PRO Racing Wheel
[  452.137695] Logitech  PRO Racing Wheel: Invalid code 768 type 1
[  452.137702] Logitech  PRO Racing Wheel: Invalid code 769 type 1
[  452.137704] Logitech  PRO Racing Wheel: Invalid code 770 type 1
[  452.137705] Logitech  PRO Racing Wheel: Invalid code 771 type 1
[  452.137707] Logitech  PRO Racing Wheel: Invalid code 772 type 1
[  452.137708] Logitech  PRO Racing Wheel: Invalid code 773 type 1
[  452.137709] Logitech  PRO Racing Wheel: Invalid code 774 type 1
[  452.137710] Logitech  PRO Racing Wheel: Invalid code 775 type 1
[  452.137712] Logitech  PRO Racing Wheel: Invalid code 776 type 1
[  452.137713] Logitech  PRO Racing Wheel: Invalid code 777 type 1
[  452.137783] input: Logitech  PRO Racing Wheel as /devices/pci0000:00/0000:00:08.1/0000:65:00.3/usb1/1-1/1-1.3/1-1.3:1.0/0003:046D:C272.000A/input/input51
[  452.138179] logitech-hidpp-device 0003:046D:C272.000A: input,hidraw6: USB HID v1.11 Joystick [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input0
[  452.199352] logitech-hidpp-device 0003:046D:C272.000B: hiddev98,hidraw7: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input1
[  452.210097] logitech-hidpp-device 0003:046D:C272.000C: hiddev99,hidraw8: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input2
[  679.307635] usb 1-1.3: Product: PRO Racing Wheel
[  679.466600] Logitech  PRO Racing Wheel: Invalid code 768 type 1
[  679.466604] Logitech  PRO Racing Wheel: Invalid code 769 type 1
[  679.466605] Logitech  PRO Racing Wheel: Invalid code 770 type 1
[  679.466606] Logitech  PRO Racing Wheel: Invalid code 771 type 1
[  679.466607] Logitech  PRO Racing Wheel: Invalid code 772 type 1
[  679.466608] Logitech  PRO Racing Wheel: Invalid code 773 type 1
[  679.466609] Logitech  PRO Racing Wheel: Invalid code 774 type 1
[  679.466610] Logitech  PRO Racing Wheel: Invalid code 775 type 1
[  679.466611] Logitech  PRO Racing Wheel: Invalid code 776 type 1
[  679.466612] Logitech  PRO Racing Wheel: Invalid code 777 type 1
[  679.466679] input: Logitech  PRO Racing Wheel as /devices/pci0000:00/0000:00:08.1/0000:65:00.3/usb1/1-1/1-1.3/1-1.3:1.0/0003:046D:C272.000F/input/input60
[  679.466955] logitech-hidpp-device 0003:046D:C272.000F: input,hidraw6: USB HID v1.11 Joystick [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input0
[  679.529704] logitech-hidpp-device 0003:046D:C272.0010: hiddev98,hidraw7: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input1
[  679.540931] logitech-hidpp-device 0003:046D:C272.0011: hiddev99,hidraw8: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input2

I never tried running the dmesg command for the wheel before installing the driver module so I don't know if the output is any different than without the driver. I just figured the invalid codes are the things you hadn't implemented yet.

fftest didn't seem to work. Here's the output:

sudo fftest /dev/input/by-id/*Racing*event*
Force feedback test program.
HOLD FIRMLY YOUR WHEEL OR JOYSTICK TO PREVENT DAMAGES

Device /dev/input/by-id/usb-Logitech_PRO_Racing_Wheel_**-event-joystick opened
Features:
  * Absolute axes: X, Y, Z, RX, RY, RZ, Throttle, Rudder, Hat 0 X, Hat 0 Y, 
    [FF 00 03 00 00 00 00 00 ]
  * Relative axes: 
    [00 00 ]
  * Force feedback effects types: 
    Force feedback periodic effects: 
    [00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ]
  * Number of simultaneous effects: 0

Uploading effect #0 (Periodic sinusoidal) ... Error:: Function not implemented
Uploading effect #1 (Constant) ... Error: Function not implemented
Uploading effect #2 (Spring) ... Error: Function not implemented
Uploading effect #3 (Damper) ... Error: Function not implemented
Uploading effect #4 (Strong rumble, with heavy motor) ... Error: Function not implemented
Uploading effect #5 (Weak rumble, with light motor) ... Error: Function not implemented
Enter effect number, -1 to exit

None of the selections have any effect if I try them. I didn't bother with tests 3 (sysfs settings) and 4 (in-game) since 2 (fftest) didn't seem to work.

Side note: using "Logitech" as the input ID would pick up the wheel, but interestingly default to testing the mouse so I changed it to "Racing"

[mescon]

@sixtysecondstosmash Found and fixed the issue in 1c75a5b.

The problem was that the G Pro (like the RS50) has 3 USB interfaces, but the G920 FFB init code assumed a single-interface device. The driver was trying to initialize FFB on interface 0 (joystick) which has no HID++ support, so the HID++ handshake never completed and FFB was never set up.

The fix:

• Interface 0 is now properly claimed for input only (fixes the "Invalid code 768" phantom button errors too)
• Interface 1 (HID++) now handles FFB init and sysfs settings
• Interface 2 is delegated to hid-generic

Please pull the latest master and rebuild. You should now see:

• No more "Invalid code 768" messages in dmesg
• "HID++ 4.x device connected" on interface 1
• "Force feedback initialized" or similar
• fftest should show available effects and produce force feedback

[sixtysecondstosmash]

It doesn't look like anything has changed in the output, except for maybe "input47" but I don't know if that's dynamic.

sudo dmesg | grep -i wheel
[   81.816728] usb 1-1.3: Product: PRO Racing Wheel
[   81.972099] Logitech  PRO Racing Wheel: Invalid code 768 type 1
[   81.972106] Logitech  PRO Racing Wheel: Invalid code 769 type 1
[   81.972108] Logitech  PRO Racing Wheel: Invalid code 770 type 1
[   81.972109] Logitech  PRO Racing Wheel: Invalid code 771 type 1
[   81.972111] Logitech  PRO Racing Wheel: Invalid code 772 type 1
[   81.972112] Logitech  PRO Racing Wheel: Invalid code 773 type 1
[   81.972113] Logitech  PRO Racing Wheel: Invalid code 774 type 1
[   81.972115] Logitech  PRO Racing Wheel: Invalid code 775 type 1
[   81.972116] Logitech  PRO Racing Wheel: Invalid code 776 type 1
[   81.972117] Logitech  PRO Racing Wheel: Invalid code 777 type 1
[   81.972191] input: Logitech  PRO Racing Wheel as /devices/pci0000:00/0000:00:08.1/0000:65:00.3/usb1/1-1/1-1.3/1-1.3:1.0/0003:046D:C272.0009/input/input47
[   81.972395] hid-generic 0003:046D:C272.0009: input,hidraw1: USB HID v1.11 Joystick [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input0
[   81.983585] hid-generic 0003:046D:C272.000A: hiddev97,hidraw2: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input1
[   81.994431] hid-generic 0003:046D:C272.000B: hiddev98,hidraw6: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input2

sudo fftest /dev/input/by-id/*Racing*event*
Force feedback test program.
HOLD FIRMLY YOUR WHEEL OR JOYSTICK TO PREVENT DAMAGES

Device /dev/input/by-id/usb-Logitech_PRO_Racing_Wheel_**-event-joystick opened
Features:
  * Absolute axes: X, Y, Z, RX, RY, RZ, Throttle, Rudder, Hat 0 X, Hat 0 Y, 
    [FF 00 03 00 00 00 00 00 ]
  * Relative axes: 
    [00 00 ]
  * Force feedback effects types: 
    Force feedback periodic effects: 
    [00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ]
  * Number of simultaneous effects: 0

Uploading effect #0 (Periodic sinusoidal) ... Error:: Function not implemented
Uploading effect #1 (Constant) ... Error: Function not implemented
Uploading effect #2 (Spring) ... Error: Function not implemented
Uploading effect #3 (Damper) ... Error: Function not implemented
Uploading effect #4 (Strong rumble, with heavy motor) ... Error: Function not implemented
Uploading effect #5 (Weak rumble, with light motor) ... Error: Function not implemented
Enter effect number, -1 to exit

To verify I had the latest updates I did git pull a couple of times, with subsequent attempts showing I already had the latest. I initially did a --force on both build and install. When nothing changed in the output I did unbuild, build and install of the module, I rebooted a few times too. Is there any other way to verify the latest build is installed and active?

[mescon]

The dmesg you shared shows all three interfaces claimed by hid-generic, not our logitech-hidpp-device driver. That means our module isn't actually binding to the G Pro at all, which would explain why fftest shows 0 effects.

Could you run these diagnostic commands and share the output:

# 1. Is our module loaded?
lsmod | grep hid_logitech_hidpp

# 2. What version/path is DKMS using?
modinfo hid-logitech-hidpp | head -5

# 3. Check DKMS status
sudo dkms status | grep hid-logitech

# 4. Did the in-kernel module get blacklisted?
cat /etc/modprobe.d/*.conf 2>/dev/null | grep -i hid-logitech

# 5. Full kernel module messages
sudo dmesg | grep -iE "hid_logitech|hid-logitech" | head -20

Also: are you sure the module is loading? After rebooting, run lsmod | grep hid_logitech_hidpp before plugging in the wheel. If the module isn't loaded, hid-generic claims the device by default.

On Fedora with secure boot, MOK enrollment is required for out-of-tree modules to load. The earlier log showed logitech-hidpp-device claiming the device, so MOK was working then. Did something change with secure boot or MOK between attempts?

[sixtysecondstosmash]

So I noticed something interesting. The module isn't automatically loaded after reboot. I ran sudo dmesg | grep -i wheel after rebuilding the module but before rebooting and saw 'logitech-hidpp-device'. Then below is what happened after the reboot (I did run the steps before rebooting and the results for steps 1–5 were the same) :

lsmod | grep hid_logitech_hidpp doesn't return anything at first.

After running sudo modprobe hid-logitech-hidpp it showed a version number per below...

1. Is our module loaded?

lsmod | grep hid_logitech_hidpp
hid_logitech_hidpp    200704  0

2. What version/path is DKMS using?

modinfo hid-logitech-hidpp | head -5
filename:       /lib/modules/6.19.11-200.fc43.x86_64/extra/hid-logitech-hidpp.ko.xz
author:         Bastien N <email>
author:         Nestor C <email>
author:         Benjamin T <email>
license:        GPL

Truncated the names and emails just in case, even though I know it's publicly available

3. Check DKMS status

sudo dkms status | grep hid-logitech
hid-logitech-hidpp/1.0, 6.19.11-200.fc43.x86_64, x86_64: installed (Original modules exist)

4. Did the in-kernel module get blacklisted?

cat /etc/modprobe.d/*.conf 2>/dev/null | grep -i hid-logitech
blacklist hid-logitech-hidpp
blacklist hid-logitech

5. Full kernel module messages

(doesn't return anything)

sudo dmesg | grep -iE "hid_logitech|hid-logitech" | head -20

6. Made sure the module was signed

modinfo hid-logitech-hidpp | grep signer
signer:         DKMS module signing key

This output shows before and after sudo modprobe hid-logitech-hidpp was run (both outputs show up the second time I ran the command, and the second set shows logitech-hidpp-device. Also, I may have unplugged and restarted the wheel base in between)

sudo dmesg | grep -i wheel
[  326.266633] usb 1-1.3: Product: PRO Racing Wheel
[  326.433779] Logitech  PRO Racing Wheel: Invalid code 768 type 1
[  326.433784] Logitech  PRO Racing Wheel: Invalid code 769 type 1
[  326.433786] Logitech  PRO Racing Wheel: Invalid code 770 type 1
[  326.433787] Logitech  PRO Racing Wheel: Invalid code 771 type 1
[  326.433788] Logitech  PRO Racing Wheel: Invalid code 772 type 1
[  326.433790] Logitech  PRO Racing Wheel: Invalid code 773 type 1
[  326.433791] Logitech  PRO Racing Wheel: Invalid code 774 type 1
[  326.433792] Logitech  PRO Racing Wheel: Invalid code 775 type 1
[  326.433794] Logitech  PRO Racing Wheel: Invalid code 776 type 1
[  326.433795] Logitech  PRO Racing Wheel: Invalid code 777 type 1
[  326.433856] input: Logitech  PRO Racing Wheel as /devices/pci0000:00/0000:00:08.1/0000:65:00.3/usb1/1-1/1-1.3/1-1.3:1.0/0003:046D:C272.0009/input/input47
[  326.434154] hid-generic 0003:046D:C272.0009: input,hidraw6: USB HID v1.11 Joystick [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input0
[  326.444879] hid-generic 0003:046D:C272.000A: hiddev98,hidraw7: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input1
[  326.455837] hid-generic 0003:046D:C272.000B: hiddev99,hidraw8: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input2
[  443.694640] Logitech  PRO Racing Wheel: Invalid code 768 type 1
[  443.694643] Logitech  PRO Racing Wheel: Invalid code 769 type 1
[  443.694644] Logitech  PRO Racing Wheel: Invalid code 770 type 1
[  443.694645] Logitech  PRO Racing Wheel: Invalid code 771 type 1
[  443.694646] Logitech  PRO Racing Wheel: Invalid code 772 type 1
[  443.694647] Logitech  PRO Racing Wheel: Invalid code 773 type 1
[  443.694648] Logitech  PRO Racing Wheel: Invalid code 774 type 1
[  443.694649] Logitech  PRO Racing Wheel: Invalid code 775 type 1
[  443.694650] Logitech  PRO Racing Wheel: Invalid code 776 type 1
[  443.694652] Logitech  PRO Racing Wheel: Invalid code 777 type 1
[  443.694698] input: Logitech  PRO Racing Wheel as /devices/pci0000:00/0000:00:08.1/0000:65:00.3/usb1/1-1/1-1.3/1-1.3:1.0/0003:046D:C272.0009/input/input49
[  443.694841] logitech-hidpp-device 0003:046D:C272.0009: input,hidraw6: USB HID v1.11 Joystick [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input0
[  443.745618] logitech-hidpp-device 0003:046D:C272.000A: hiddev98,hidraw7: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input1
[  443.746268] logitech-hidpp-device 0003:046D:C272.000B: hiddev99,hidraw8: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input2

Still no dice with fftest

sudo fftest /dev/input/by-id/*Racing*event*
[sudo] password for chandler: 
Force feedback test program.
HOLD FIRMLY YOUR WHEEL OR JOYSTICK TO PREVENT DAMAGES

Device /dev/input/by-id/usb-Logitech_PRO_Racing_Wheel_**-event-joystick opened
Features:
  * Absolute axes: X, Y, Z, RX, RY, RZ, Throttle, Rudder, Hat 0 X, Hat 0 Y, 
    [FF 00 03 00 00 00 00 00 ]
  * Relative axes: 
    [00 00 ]
  * Force feedback effects types: 
    Force feedback periodic effects: 
    [00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ]
  * Number of simultaneous effects: 0

Uploading effect #0 (Periodic sinusoidal) ... Error:: Function not implemented
Uploading effect #1 (Constant) ... Error: Function not implemented
Uploading effect #2 (Spring) ... Error: Function not implemented
Uploading effect #3 (Damper) ... Error: Function not implemented
Uploading effect #4 (Strong rumble, with heavy motor) ... Error: Function not implemented
Uploading effect #5 (Weak rumble, with light motor) ... Error: Function not implemented
Enter effect number, -1 to exit

Short update: I was curious about the MOK signing and verificatoin so I asked Gemini to verify the correct steps for my system. I asked if signing keys were module specific or system-wide. Since I have Nvidia drivers installed that were already registered with MOK, it was recommended to use that to keep things clean. It also said the steps on github aren't specific to Fedora with secure boot.

In short, I'm not sure if the original signing key mentioned above actually did anything. I've now verified that this module and the Nvidia driver are now registered to the system with the same signer. I'm not a CLI pro so I'm not sure if this makes any difference from DKMS as the signer, but I just wanted to shoot a short update where I'm at.

I went through your diag commands again and none of the output besides the signer has changed though.

I should also add that I'd pulled the latest updates in case you'd done some regressions in your latest edits. I saw some red minuses in the log when I did the pull.
/update

[mescon]

Thanks a lot for this set of G Pro captures, they're exactly the kind of contribution that moves things along.

Quick update on where we are: native Linux Trueforce haptics are now running on the bench and general FFB support has come a long way. To close the remaining gaps we've opened two new help-wanted issues with small, step-by-step batch scripts that contributors can run:

• Help wanted: Trueforce USB captures from BeamNG.drive and Assetto Corsa #14 for the game-required Trueforce captures (BeamNG / Assetto Corsa).
• Help wanted: RS50 wheel / G Hub USB captures (no game required) #15 for the no-game wheel and G Hub captures (calibrate, FFB filter, D-pad, settings commit, persistence, profile button, LIGHTSYNC).

Both scripts just drive tshark + USBPcap, so they should work equally well on G Pro as on RS50 (the new versions should also be more robust than the earlier one you hit trouble with). If you have time to run either set on your G Pro and attach the pcapng files to the matching issue, that would directly help us get G Pro parity too.

[mescon]

@sixtysecondstosmash Good news: between your last report and now, two commits on master land directly on what you were seeing.

• d1a1bd4 slims the interface 0 probe path for RS50-family and G Pro wheels, which is the kind of setup noise that was producing the "Invalid code 768-777" phantom button lines in your dmesg.
• 8106b3a makes the G920-class FFB init walk sibling interfaces to find the input device instead of hardcoding interface 0. That's a plausible cause for fftest showing "Number of simultaneous effects: 0" and "Function not implemented" on every effect: the FFB callbacks were being attached to the wrong evdev node.

Could you:

1. Pull master, rebuild, and reinstall via DKMS.
2. Also remove these lines from /etc/modprobe.d/*, they're what's preventing the module from autoloading on boot:

   blacklist hid-logitech-hidpp
   blacklist hid-logitech
   

   A quick sudo grep -rl 'blacklist hid-logitech' /etc/modprobe.d/ will find the file. Those were probably left over from G Hub or Oversteer.
3. Reboot, verify lsmod | grep hid_logitech_hidpp shows the module without needing modprobe.
4. Check that sudo dmesg | grep -i wheel no longer shows the "Invalid code 768-777" lines.
5. Rerun sudo fftest /dev/input/by-id/*Racing*event*. With the interface-iteration fix you should now see a non-zero effect count and effects should upload.

If any of those fail, paste the new dmesg + fftest output and we'll take it from there.

---

@aderumier Yes, G Pro captures are still very welcome, thanks for checking in. We've just opened two help-wanted issues with step-by-step batch scripts that drive tshark + USBPcap, so contributing is just "install Wireshark with USBPcap, run the script, attach the pcapng files":

• Help wanted: Trueforce USB captures from BeamNG.drive and Assetto Corsa #14 for game-required Trueforce captures (BeamNG / Assetto Corsa).
• Help wanted: RS50 wheel / G Hub USB captures (no game required) #15 for no-game wheel and G Hub captures (calibrate, FFB filter, D-pad, settings commit, persistence, profile button, LIGHTSYNC).

Both scripts work equally well on G Pro as on RS50. Any subset you can run helps.

[sixtysecondstosmash]

Thanks. As expected no more "Invalid code" lines. But fftest looks the same as before

Edit blacklist

Removed the blacklisting (I deleted the file /etc/modprobe.d/blacklist-hid-logitech-hidpp.conf since it was added it per the install instructions on the main page and only contained the 2 lines)

Check blacklist

cat /etc/modprobe.d/*.conf 2>/dev/null | grep -i hid-logitech

Now returns blank

Is module loaded

lsmod | grep hid_logitech_hidpp
hid_logitech_hidpp    196608  0

It loads without running modprobe now. Result is fresh off a reboot. New number shown

Checked DKMS status

sudo dkms status | grep hid-logitech
hid-logitech-hidpp/1.0, 6.19.11-200.fc43.x86_64, x86_64: installed (Original modules exist)
hid-logitech-hidpp/1.0, 6.19.12-200.fc43.x86_64, x86_64: installed (Original modules exist)

There was a kernel update

Full kernel module messages

sudo dmesg | grep -iE "hid_logitech|hid-logitech" | head -20
[   11.772632] hid_logitech_hidpp: loading out-of-tree module taints kernel.

No longer blank result

Wheel detection

sudo dmesg | grep -i wheel
[  348.777362] usb 1-1.3: Product: PRO Racing Wheel
[  348.940565] input: Logitech  PRO Racing Wheel as /devices/pci0000:00/0000:00:08.1/0000:65:00.3/usb1/1-1/1-1.3/1-1.3:1.0/0003:046D:C272.0008/input/input49
[  348.941421] logitech-hidpp-device 0003:046D:C272.0008: input,hidraw5: USB HID v1.11 Joystick [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input0
[  349.003717] logitech-hidpp-device 0003:046D:C272.0009: hiddev98,hidraw6: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input1
[  349.016012] logitech-hidpp-device 0003:046D:C272.000A: hiddev99,hidraw7: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input2

Invalid code lines are gone

Run fftest

sudo fftest /dev/input/by-id/*Racing*event*
Force feedback test program.
HOLD FIRMLY YOUR WHEEL OR JOYSTICK TO PREVENT DAMAGES

Device /dev/input/by-id/usb-Logitech_PRO_Racing_Wheel_**-event-joystick opened
Features:
  * Absolute axes: X, Y, Z, RX, RY, RZ, Throttle, Rudder, Hat 0 X, Hat 0 Y, 
    [FF 00 03 00 00 00 00 00 ]
  * Relative axes: 
    [00 00 ]
  * Force feedback effects types: 
    Force feedback periodic effects: 
    [00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ]
  * Number of simultaneous effects: 0

Uploading effect #0 (Periodic sinusoidal) ... Error:: Function not implemented
Uploading effect #1 (Constant) ... Error: Function not implemented
Uploading effect #2 (Spring) ... Error: Function not implemented
Uploading effect #3 (Damper) ... Error: Function not implemented
Uploading effect #4 (Strong rumble, with heavy motor) ... Error: Function not implemented
Uploading effect #5 (Weak rumble, with light motor) ... Error: Function not implemented
Enter effect number, -1 to exit

Errors persist. No effect is sent to the wheel

[mescon]

@sixtysecondstosmash friendly bump on this: when you get a chance, could you git pull master, rebuild/reinstall the DKMS module, and try sudo fftest /dev/input/by-id/*Racing*event* again? Since we last spoke the driver has two fixes that directly target "fftest shows 0 effects" on the G Pro (commits d1a1bd4 and 8106b3a), plus a bunch more G Pro-specific corrections from capture analysis. If fftest now shows available effects and force kicks in, that closes a major open question for us. If it still returns "Function not implemented" for every effect, the new dmesg will tell us where to look next.

[sixtysecondstosmash]

Still no luck with fftest

fftest

sudo fftest /dev/input/by-id/*Racing*event*
Force feedback test program.
HOLD FIRMLY YOUR WHEEL OR JOYSTICK TO PREVENT DAMAGES

Device /dev/input/by-id/usb-Logitech_PRO_Racing_Wheel_**-event-joystick opened
Features:
  * Absolute axes: X, Y, Z, RX, RY, RZ, Throttle, Rudder, Hat 0 X, Hat 0 Y, 
    [FF 00 03 00 00 00 00 00 ]
  * Relative axes: 
    [00 00 ]
  * Force feedback effects types: 
    Force feedback periodic effects: 
    [00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ]
  * Number of simultaneous effects: 0

Uploading effect #0 (Periodic sinusoidal) ... Error:: Function not implemented
Uploading effect #1 (Constant) ... Error: Function not implemented
Uploading effect #2 (Spring) ... Error: Function not implemented
Uploading effect #3 (Damper) ... Error: Function not implemented
Uploading effect #4 (Strong rumble, with heavy motor) ... Error: Function not implemented
Uploading effect #5 (Weak rumble, with light motor) ... Error: Function not implemented
Enter effect number, -1 to exit

Wheel detection

sudo dmesg | grep -i wheel
[  177.263350] usb 1-1.3: Product: PRO Racing Wheel
[  177.431552] input: Logitech  PRO Racing Wheel as /devices/pci0000:00/0000:00:08.1/0000:65:00.3/usb1/1-1/1-1.3/1-1.3:1.0/0003:046D:C272.0009/input/input50
[  177.431827] logitech-hidpp-device 0003:046D:C272.0009: input,hidraw6: USB HID v1.11 Joystick [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input0
[  177.493449] logitech-hidpp-device 0003:046D:C272.000A: hiddev98,hidraw7: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input1
[  177.503677] logitech-hidpp-device 0003:046D:C272.000B: hiddev99,hidraw8: USB HID v1.11 Device [Logitech  PRO Racing Wheel] on usb-0000:65:00.3-1.3/input2

This looks ok to me, but maybe you'll see something I don't

Module info

lsmod | grep hid_logitech_hidpp
hid_logitech_hidpp    188416  0

DKMS status

sudo dkms status | grep hid-logitech
hid-logitech-hidpp/1.0, 6.19.12-200.fc43.x86_64, x86_64: installed (Original modules exist)

Module messages

sudo dmesg | grep -iE "hid_logitech|hid-logitech" | head -20
[   15.235798] hid_logitech_hidpp: loading out-of-tree module taints kernel.

Other
I still haven't figured out the optimal way to update the module (least amount of steps).

For the most surefire way to get it updated I end up doing...

• git pull
• dkms remove the module
• rm -rf the folder in /usr/src
• mkdir p the module folder in /usr/src again
• cp -r the mainline/* directory into the newly created folder
• dkms build it
• dkms install it
• reboot (modprobe [module] isn't enough)

Is this the correct way or is there a more efficient set of steps? I'll keep playing with it, but without removing the old module first... at the dkms install step I was getting bad exit status 3, so I wasn't sure if it was installing properly.