Apple Macbook Air touchpad device name?

[guttermonk]

When I run libinput list-devices I get the following two devices:

Device:           Apple Inc. Apple Internal Keyboard / Trackpad
Kernel:           /dev/input/event1
Group:            7
Seat:             seat0, default
Capabilities:     keyboard
Tap-to-click:     n/a
Tap-and-drag:     n/a
Tap drag lock:    n/a
Left-handed:      n/a
Nat.scrolling:    n/a
Middle emulation: n/a
Calibration:      n/a
Scroll methods:   none
Click methods:    none
Disable-w-typing: n/a
Disable-w-trackpointing: n/a
Accel profiles:   n/a
Rotation:         0.0
  
Device:           bcm5974
Kernel:           /dev/input/event10
Group:            7
Seat:             seat0, default
Size:             105x75mm
Capabilities:     pointer gesture
Tap-to-click:     disabled
Tap-and-drag:     enabled
Tap drag lock:    disabled
Left-handed:      disabled
Nat.scrolling:    enabled
Middle emulation: disabled
Calibration:      n/a
Scroll methods:   *two-finger edge
Click methods:    button-areas *clickfinger
Disable-w-typing: enabled
Disable-w-trackpointing: enabled
Accel profiles:   flat *adaptive custom
Rotation:         n/a

I tried adding device bcm5974 to my config, which worked fine at first but, after hibernating, gestures stopped working. So I tried changing it to device Apple Inc. Apple Internal Keyboard / Trackpad but that killed gestures all together. When I set it to device all it works fine and survives hibernation, but as I understand it, setting it to device all isn't ideal. Any suggestions?

[bulletmark]

Please read the TROUBLE-SHOOTING section, particularly the opening paragraph.

[guttermonk]

Thank you for getting back to me. I read through the trouble-shooting section but it wasn't much help. Looking at the raw data for device bcm5974, gestures are received normally but as soon as I hibernate it stops receiving anything.

I also read through the following issues on the same topic: 68, 160, 177, 194, and 228. Looks like most people are either using a restart script for Gnome and KDE, or they end up giving up and using device all. Since I'm using NixOS, unfortunately neither of those scripts work on my system.

Here's the output of my libinput-gestures -l.

Here's this too:

ls -la /dev/input/by-path/
drwxr-xr-x - root 25 Aug 06:33 .
drwxr-xr-x - root 25 Aug 06:33 ..
lrwxrwxrwx - root 25 Aug 06:17 pci-0000:00:14.0-usb-0:1:1.0-event-kbd -> ../event0
lrwxrwxrwx - root 25 Aug 06:16 pci-0000:00:14.0-usb-0:5:1.1-event-kbd -> ../event1
lrwxrwxrwx - root 25 Aug 06:33 pci-0000:00:14.0-usb-0:5:1.2-event-mouse -> ../event2
lrwxrwxrwx - root 25 Aug 06:33 pci-0000:00:14.0-usb-0:5:1.2-mouse -> ../mouse0
lrwxrwxrwx - root 25 Aug 06:17 pci-0000:00:14.0-usbv2-0:1:1.0-event-kbd -> ../event0
lrwxrwxrwx - root 25 Aug 06:16 pci-0000:00:14.0-usbv2-0:5:1.1-event-kbd -> ../event1
lrwxrwxrwx - root 25 Aug 06:33 pci-0000:00:14.0-usbv2-0:5:1.2-event-mouse -> ../event2
lrwxrwxrwx - root 25 Aug 06:33 pci-0000:00:14.0-usbv2-0:5:1.2-mouse -> ../mouse0

ls -la /dev/input/by-id/
drwxr-xr-x - root 25 Aug 06:33 .
drwxr-xr-x - root 25 Aug 06:33 ..
lrwxrwxrwx - root 25 Aug 06:16 usb-Apple_Inc._Apple_Internal_Keyboard___Trackpad_DQ63363GAVRF94RATBE-if01-event-kbd -> ../event1
lrwxrwxrwx - root 25 Aug 06:33 usb-Apple_Inc._Apple_Internal_Keyboard___Trackpad_DQ63363GAVRF94RATBE-if02-event-mouse -> ../event2
lrwxrwxrwx - root 25 Aug 06:33 usb-Apple_Inc._Apple_Internal_Keyboard___Trackpad_DQ63363GAVRF94RATBE-if02-mouse -> ../mouse0
lrwxrwxrwx - root 25 Aug 06:17 usb-Yubico_YubiKey_OTP+FIDO+CCID-event-kbd -> ../event0

[guttermonk]

Today I tried setting each of the /dev/input/ paths listed above, and got the following results.

Worked at first but didn't survive hibernation:

device bcm5974  
device /dev/input/by-path/pci-0000:00:14.0-usb-0:5:1.2-event-mouse
device /dev/input/by-path/pci-0000:00:14.0-usbv2-0:5:1.2-event-mouse  
device /dev/input/by-path/pci-0000:00:14.0-usbv2-0:5:1.2-mouse  
device /dev/input/by-id/usb-Apple_Inc._Apple_Internal_Keyboard___Trackpad_DQ63363GAVRF94RATBE-if02-event-mouse  

Did not work at all:

device Apple Inc. Apple Internal Keyboard / Trackpad  
device /dev/input/by-path/pci-0000:00:14.0-usb-0:5:1.2-mouse  
device /dev/input/by-id/usb-Apple_Inc._Apple_Internal_Keyboard___Trackpad_DQ63363GAVRF94RATBE-if02-mouse

[bulletmark]

From your output of libinput-gestures -l, I don't recognise the version of libinput-gestures you are running. How did you install it? Can you please install and try the latest version?

[guttermonk]

I've installed it from the Nix package repo. They currently have version 2.76, which is the same version that's in the AUR and it is the same version that I'm currently running.

[bulletmark]

I didn't recognize that hash because the code is different. Nix patches the code to change some things related to their environment although it seems the changes are benign.

[bulletmark]

I've seen others report problems with suspend where, e.g. the device hangs or the device name changes, and for that I created another dbus-action utility which is mentioned in the README here. It works by restarting libinput-gestures (i.e. libinput-gestures-setup restart) after system resume so that it rediscovers where the touchpad is. However, I don't think you can use that on a Mac which I don't think offers D-Bus. You can probably find some other tool that can do something like that though.

[guttermonk]

I'll have to look into the possibility of using D-Bus in NixOS on a Macbook. For some reason, the libinput-gestures-setup restart command isn't working at the moment on my NixOS (it worked fine when I used Arch), but it may be due to some configuration missing from my configuration.nix file. I'll investigate more with the nix community.

In the meantime, do you happen to know if a bug has been submitted for the libinput package? Based on your post on 194, do you still think this is an issue with libinput debug-events not automatically resuming after a suspend for some users/touchpads? Peter Huterrer was suggesting in this post that a work around would be for libinput-gestures to use the udev backend, if I'm not mistaken. I don't know a lot about these systems, so I'm not sure if that's possible. Just curious what your thoughts are.

[guttermonk]

On NixOS running Hyprland, I found that I can issue the following commands using either hypridle or swayidle, before hibernation:
pkill -f gestures

And after hibernation:
libinput-gestures

This solves the issue without having to use the device all method. I hope this helps anyone that stumbles on this in the future.

[bulletmark]

A pkill is crude. If you want to stop libinput-gestures then do a libinput-gestures-setup stop. If you want to start or restart it do a libinput-gestures-setup start or libinput-gestures-setup restart .

[guttermonk]

I started a thread on the Nix forum to get to the bottom of that.

[guttermonk]

I was able to get some help to get libinput-gestures' systemd commands working on NixOS. I added systemctl --user start libinput-gestures & to my Hyprland startup script, and the following to my hyprland.conf:

exec-once = swayidle -w after-resume 'systemctl --user start libinput-gestures' before-sleep 'systemctl --user stop libinput-gestures'

Everything appears to be working correctly. Gestures are coming back after resuming from hibernation. Any down-sides to doing it like this?

[bulletmark]

Assume you are asking me but sorry I am not familiar at all about NixOS, Hyprland, or swayidle. It looks sensible though.