Comment 29 for bug 1965219

Same behavior with an Entroware laptop on 5.15.0-52-generic (both touchpad and internal keyboard, function keys work)

Hard to tell who's the culprit among the kernel, libusb, udev and gdm

One weird thing in the context of resuming : The Xorg.log generated by gdm shows:

(II) UnloadModule: "libinput" message
[...]
(II) XINPUT: Adding extended input device "Power Button" (type: KEYBOARD, id 6)
[...]
(II) XINPUT: Adding extended input device "Video Bus" (type: KEYBOARD, id 7)
[...]
(II) XINPUT: Adding extended input device "Video Bus" (type: KEYBOARD, id 8)
(II) XINPUT: Adding extended input device "Power Button" (type: KEYBOARD, id 9)
(II) XINPUT: Adding extended input device "Sleep Button" (type: KEYBOARD, id 10)

Typically:

[ 79.317] (II) config/udev: Adding input device Chicony USB2.0 Camera: Chicony (/dev/input/event9)
[ 79.317] (**) Chicony USB2.0 Camera: Chicony : Applying InputClass "libinput keyboard catchall"
[ 79.317] (II) Using input driver 'libinput' for 'Chicony USB2.0 Camera: Chicony '
[ 79.317] (II) systemd-logind: got fd for /dev/input/event9 13:73 fd 43 paused 0
[ 79.317] (**) Chicony USB2.0 Camera: Chicony : always reports core events
[ 79.317] (**) Option "Device" "/dev/input/event9"
[ 79.319] (II) event9 - Chicony USB2.0 Camera: Chicony : is tagged by udev as: Keyboard
[ 79.319] (II) event9 - Chicony USB2.0 Camera: Chicony : device is a keyboard
[ 79.319] (II) event9 - Chicony USB2.0 Camera: Chicony : device removed
[ 79.319] (**) Option "config_info" "udev:/sys/devices/pci0000:00/0000:00:14.0/usb1/1-2/1-2:1.0/input/input17/event9"
[ 79.319] (II) XINPUT: Adding extended input device "Chicony USB2.0 Camera: Chicony " (type: KEYBOARD, id 12)
[ 79.319] (**) Option "xkb_model" "pc105"
[ 79.319] (**) Option "xkb_layout" "fr"
[ 79.319] (**) Option "xkb_variant" "azerty"
[ 79.320] (II) event9 - Chicony USB2.0 Camera: Chicony : is tagged by udev as: Keyboard
[ 79.320] (II) event9 - Chicony USB2.0 Camera: Chicony : device is a keyboard

(a webcam tagged keyboard?)

I don't know if it's related, but if it is it could be a problem with udev rules during resume?