-
-
Save fghaas/3406be59095de212182f1803a503a64b to your computer and use it in GitHub Desktop.
# /etc/udev/rules.d/75-input.rules | |
# local udev rules for input devices | |
ACTION!="add|change", GOTO="input_end" | |
# ALPS DualPoint Stick: Ignore as input device | |
ENV{ID_BUS}=="i8042", ENV{NAME}=="AlpsPS/2 ALPS DualPoint Stick", ENV{ID_INPUT}="", ENV{ID_INPUT_MOUSE}="", ENV{ID_INPUT_POINTINGSTICK}="" | |
LABEL="input_end" |
Note that there's a difference between ATTR
and ATTRS
- the former cannot work because it'll never match the event node. But that's the device we care about. ATTRS
is supposed to search upwards from the current device, so it should have worked. Not sure what's going on there.
The device you need to run udevadm
against is /sys/class/input/eventX
- whichever is the event node (event5 in the above output), not the input node.
@whot OK, I have no idea what I did differently this time, but I changed my udev rule back to matching on ATTRS{name}
again, and now things appear to be working. The stick pointer seems to have been permanently disabled and is now off, including in a Wayland session. Thanks for the help!
Hello, thanks for the script! This technique also works with hwdb like so:
# /etc/udev/hwdb.d/98-latitude-trackpoint-disable.hwdb
# Dell Latitude E7470
# AlpsPS/2 ALPS DualPoint Stick
# udevadm info -a -q all /sys/devices/platform/i8042/serio1/input/input3
evdev:input:b0011v0002p0008e0800-e0,1,2,3,k110,111,112,r0,1,a18,mlsfw
ID_INPUT=0
ID_INPUT_MOUSE=0
ID_INPUT_POINTINGSTICK=0
Afterwards run below to update the hwdb database and reload.
systemd-hwdb update
udevadm trigger
The following udev rule worked for me. You can use wildcards in the device name. The rule is case sensitive, i.e., ATTRS{name} works but ATTRS{NAME} does not.
# ALPS DualPoint Stick: Ignore as input device
ATTRS{name}=="*DualPoint Stick", ENV{ID_INPUT}="", ENV{ID_INPUT_MOUSE}="", ENV{ID_INPUT_POINTINGSTICK}=""
@whot Matching on just
ATTRS{name}
doesn't apply the udev rule at all. I changed it toATTR{name}
, because that is whatudevadm info -a
suggests, and that does unset theID_*
variables, but still doesn't appear to hide the device from libinput.udev rules:
udevadm info
after reboot:libinput-list-devices
:And of course, the stick does remain enabled in a Wayland session.