libinput and waynergy

[ackstorm23]

Does waynegy's method of communicating with wayland bypass libinput?

I've noticed if I use libinput debug-events, keyboard and mouse clicks are not shown.

I am trying to debug a frequent issue with swaywm where mouse clicks no longer register.

[ackstorm23]

This is the output from initial startup

# sudo libinput debug-events --set-click-method=clickfinger --verbose
libinput version: 1.21.0
event2  - Video Bus: is tagged by udev as: Keyboard
event2  - Video Bus: device is a keyboard
event0  - Lid Switch: is tagged by udev as: Switch
event0  - Lid Switch: device is a switch device
event0  - Lid Switch: device is a switch device
event1  - Intel Touch Host Controller: is tagged by udev as: Touchscreen
event1  - Intel Touch Host Controller: device is a touch device
event266 - Generic TX 384kb Hifi Type_C Audio: is tagged by udev as: Keyboard
event266 - Generic TX 384kb Hifi Type_C Audio: device is a keyboard
event31 - Generic USB Audio Consumer Control: is tagged by udev as: Keyboard
event31 - Generic USB Audio Consumer Control: device is a keyboard
event256 - Generic USB Audio: is tagged by udev as: Keyboard
event256 - Generic USB Audio: device is a keyboard
event13 - HDA Intel PCH Mic: is tagged by udev as: Switch
event13 - not using input device '/dev/input/event13'
event14 - HDA Intel PCH Headphone: is tagged by udev as: Switch
event14 - not using input device '/dev/input/event14'
event15 - HDA Intel PCH HDMI/DP,pcm=3: is tagged by udev as: Switch
event15 - not using input device '/dev/input/event15'
event16 - HDA Intel PCH HDMI/DP,pcm=7: is tagged by udev as: Switch
event16 - not using input device '/dev/input/event16'
event17 - HDA Intel PCH HDMI/DP,pcm=8: is tagged by udev as: Switch
event17 - not using input device '/dev/input/event17'
event18 - HDA Intel PCH HDMI/DP,pcm=9: is tagged by udev as: Switch
event18 - not using input device '/dev/input/event18'
event19 - HDA Intel PCH HDMI/DP,pcm=10: is tagged by udev as: Switch
event19 - not using input device '/dev/input/event19'
event20 - HDA Intel PCH HDMI/DP,pcm=11: is tagged by udev as: Switch
event20 - not using input device '/dev/input/event20'
event21 - HDA Intel PCH HDMI/DP,pcm=12: is tagged by udev as: Switch
event21 - not using input device '/dev/input/event21'
event22 - HDA Intel PCH HDMI/DP,pcm=13: is tagged by udev as: Switch
event22 - not using input device '/dev/input/event22'
event23 - HDA Intel PCH HDMI/DP,pcm=14: is tagged by udev as: Switch
event23 - not using input device '/dev/input/event23'
event24 - HDA Intel PCH HDMI/DP,pcm=15: is tagged by udev as: Switch
event24 - not using input device '/dev/input/event24'
event25 - HDA Intel PCH HDMI/DP,pcm=16: is tagged by udev as: Switch
event25 - not using input device '/dev/input/event25'
event26 - HDA Intel PCH HDMI/DP,pcm=17: is tagged by udev as: Switch
event26 - not using input device '/dev/input/event26'
event4  - gpio-keys: is tagged by udev as: Keyboard
event4  - gpio-keys: device is a keyboard
event5  - gpio-keys: is tagged by udev as: Keyboard
event5  - gpio-keys: device is a keyboard
event9  - Microsoft Surface 045E:09AE Keyboard: is tagged by udev as: Keyboard
event9  - Microsoft Surface 045E:09AE Keyboard: device is a keyboard
  ... event0  - lid: keyboard paired with Lid Switch<->Microsoft Surface 045E:09AE Keyboard
event10 - Microsoft Surface 045E:09B0 Keyboard: is tagged by udev as: Keyboard
event10 - Microsoft Surface 045E:09B0 Keyboard: device is a keyboard
event11 - Microsoft Surface 045E:09B0: not tagged as supported input device
event11 - not using input device '/dev/input/event11'
event12 - Microsoft Surface 045E:09B0: not tagged as supported input device
event12 - not using input device '/dev/input/event12'
event27 - Microsoft Surface 045E:09AF Mouse: is tagged by udev as: Mouse
event27 - Microsoft Surface 045E:09AF Mouse: device is a pointer
event28 - Microsoft Surface 045E:09AF Touchpad: is tagged by udev as: Touchpad
  ... event28 - thumb: enabled thumb detection (area)
event28 - Microsoft Surface 045E:09AF Touchpad: device is a touchpad
  ... event28 - lid: activated for Microsoft Surface 045E:09AF Touchpad<->Lid Switch
  ... event28 - palm: dwt activated with Microsoft Surface 045E:09AF Touchpad<->Microsoft Surface 045E:09AE Keyboard
event29 - Microsoft Surface 045E:09AF UNKNOWN: not tagged as supported input device
event29 - not using input device '/dev/input/event29'
event30 - Microsoft Surface 045E:09AF UNKNOWN: not tagged as supported input device
event30 - not using input device '/dev/input/event30'
event8  - Microsoft Surface KIP Tablet Mode Switch: is tagged by udev as: Switch
event8  - Microsoft Surface KIP Tablet Mode Switch: device is a switch device
  ... event9  - tablet-mode: paired Microsoft Surface 045E:09AE Keyboard<->Microsoft Surface KIP Tablet Mode Switch
  ... event28 - tablet-mode: activated for Microsoft Surface 045E:09AF Touchpad<->Microsoft Surface KIP Tablet Mode Switch
event3  - PC Speaker: not tagged as supported input device
event3  - not using input device '/dev/input/event3'
event6  - IPTS Touch: is tagged by udev as: Touchscreen
event6  - IPTS Touch: device is a touch device
event7  - IPTS Stylus: is tagged by udev as: Tablet
event7  - IPTS Stylus: tablet 'IPTS Stylus' unknown to libwacom
event7  - IPTS Stylus: device is a tablet
-event2   DEVICE_ADDED            Video Bus                         seat0 default group1  cap:k
-event0   DEVICE_ADDED            Lid Switch                        seat0 default group2  cap:S
 event0   SWITCH_TOGGLE           +0.000s       switch lid state 1
-event1   DEVICE_ADDED            Intel Touch Host Controller       seat0 default group3  cap:t ntouches 1 calib
-event266  DEVICE_ADDED            Generic TX 384kb Hifi Type_C Audio seat0 default group4  cap:k
-event31  DEVICE_ADDED            Generic USB Audio Consumer Control seat0 default group5  cap:kp scroll-nat
-event256  DEVICE_ADDED            Generic USB Audio                 seat0 default group5  cap:k
-event4   DEVICE_ADDED            gpio-keys                         seat0 default group6  cap:k
-event5   DEVICE_ADDED            gpio-keys                         seat0 default group6  cap:k
-event9   DEVICE_ADDED            Microsoft Surface 045E:09AE Keyboard seat0 default group7  cap:kp scroll-nat
-event10  DEVICE_ADDED            Microsoft Surface 045E:09B0 Keyboard seat0 default group8  cap:k
-event27  DEVICE_ADDED            Microsoft Surface 045E:09AF Mouse seat0 default group9  cap:p left scroll-nat scroll-button
-event28  DEVICE_ADDED            Microsoft Surface 045E:09AF Touchpad seat0 default group9  cap:pg  size 96x52mm tap(dl off) left scroll-nat scroll-2fg-edge click-buttonareas-clickfinger dwt-on dwtp-on
-event8   DEVICE_ADDED            Microsoft Surface KIP Tablet Mode Switch seat0 default group10 cap:S
-event6   DEVICE_ADDED            IPTS Touch                        seat0 default group11 cap:t  size 274x185mm ntouches 11 calib
-event7   DEVICE_ADDED            IPTS Stylus                       seat0 default group12 cap:T  size 274x185mm calib

however nothing can be seen after this output from that point forward.

[ackstorm23]

running latest wlroots-git, sway-git, and waynergy-git from arch aur

[r-c-f]

It will indeed bypass libinput, which is just one of several backends on the wlroots side of things

[ackstorm23]

Thanks!

[r-c-f]

If it's particularly useful, for debugging purposes you can use the uinput waynergy backend with sway as well. This will go through libinput, like any hardware device.

[ackstorm23]

Technically speaking, how does Waynergy control keyboard/mouse input with wlroots compositors like sway?

This issue has existed for ages with no resolution, but everyone reporting it has not found any information from debugging libinput.

As I experience this same problem using Waynergy (which is not using libinput), it may reveal that the bug is in fact in wlroots (or sway)

[r-c-f]

For the mouse input that seems to be the issue here, it uses this Wayland protocol to do so. For debugging purposes you can export WAYLAND_DEBUG=1 to see what it's doing on either end, compositor or client.

In any case it is almost certainly a sway/wlroots bug; to narrow it down one might consider trying to reproduce using other wlroots compositors.