GNOME Bugzilla – Bug 740518
Clutter evdev devices should expose capabilities
Last modified: 2021-06-10 11:31:53 UTC
Hello! When I'm trying to launch GNOME on Wayland via gdm, I see black screen with cursor at the top left corner and after a moment it returns back to gdm. Tried to launch with command: gnome-session --session=gnome-wayland And it just does nothing. Tried with this command too: gnome-shell --wayland --display-server I get the following output: (gnome-shell:4836): mutter-WARNING **: failed to unlink stale lock file /tmp/.X1-lock: Operation not permitted glamor: EGL version 1.4 (DRI2): Window manager warning: Failed to set environment variable DISPLAY for gnome-session: GDBus.Error:org.freedesktop.DBus.Error.NameHasNoOwner: Name "org.gnome.SessionManager" does not exist Window manager warning: Failed to set environment variable WAYLAND_DISPLAY for gnome-session: GDBus.Error:org.freedesktop.DBus.Error.NameHasNoOwner: Name "org.gnome.SessionManager" does not exist Gjs-Message: JS LOG: Failed to launch ibus-daemon: Failed to execute child process "ibus-daemon" (No such file or directory) Gjs-Message: JS LOG: No permission to trigger offline updates: Polkit.Error: GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: Action org.freedesktop.packagekit.trigger-offline-update is not registered (gnome-shell:4836): Clutter-CRITICAL **: clutter_input_device_get_pointer_stage: assertion 'CLUTTER_IS_INPUT_DEVICE (device)' failed (EE) Fatal server error: (EE) failed to dispatch Wayland events: Connection reset by peer (EE) Segmentation fault (core dumped) I also see this in dmesg: gnome-shell[5927]: segfault at 0 ip 00007fef0bdf24b6 sp 00007fff7147c740 error 4 in libmutter.so.0.0.0[7fef0bd3f000+103000] OS: Arch Linux (linux 3.17.3, xorg-server 1.16.2, wayland 1.6.0, gnome-shell 3.14.2) Package 'xorg-server-xwayland' is installed. some glxinfo output: OpenGL vendor string: X.Org OpenGL renderer string: Gallium 0.4 on AMD RV770 OpenGL core profile version string: 3.3 (Core Profile) Mesa 10.5.0-devel (git-74f505f) OpenGL core profile shading language version string: 3.30 (This bug happened with Mesa 10.3 and 10.4 too) (GNOME on Wayland worked for me in 3.12 with patched mutter)
Can you try to get a backtrace from the core dump?
Before sending bug report I tried to debug it in VT and the system hanged. However now it worked with gnome-terminal in gnome shell: GNU gdb (GDB) 7.8.1 Copyright (C) 2014 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-unknown-linux-gnu". Type "show configuration" for configuration details. For bug reporting instructions, please see: <http://www.gnu.org/software/gdb/bugs/>. Find the GDB manual and other documentation resources online at: <http://www.gnu.org/software/gdb/documentation/>. For help, type "help". Type "apropos word" to search for commands related to "word"... Reading symbols from gnome-shell...done. (gdb) run Starting program: /usr/bin/gnome-shell --wayland --display-server warning: Could not load shared library symbols for linux-vdso.so.1. Do you need "set solib-search-path" or "set sysroot"? [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/libthread_db.so.1". [New Thread 0x7fffe30e9700 (LWP 24698)] [New Thread 0x7fffe28e8700 (LWP 24699)] [New Thread 0x7fffe0ada700 (LWP 24700)] (gnome-shell:24694): Cogl-WARNING **: ./winsys/cogl-winsys-egl-kms.c:747: Error restoring saved CRTC [New Thread 0x7fffd9641700 (LWP 24701)] [New Thread 0x7fffd8c33700 (LWP 24702)] (EE) Fatal server error: (EE) wl_drm@4: error 0: authenicate failed (EE) (gnome-shell:24694): mutter-ERROR **: Spurious exit of X Wayland server Program received signal SIGTRAP, Trace/breakpoint trap. g_logv (log_domain=0x7ffff69df740 "mutter", log_level=G_LOG_LEVEL_ERROR, format=<optimized out>, args=args@entry=0x7fffffffdfe0) at gmessages.c:1046 1046 g_private_set (&g_log_depth, GUINT_TO_POINTER (depth)); (gdb) bt full
+ Trace 234361
(In reply to comment #0) > (gnome-shell:4836): mutter-WARNING **: failed to unlink stale lock file > /tmp/.X1-lock: Operation not permitted Does it work if you manually clean up this file?
(In reply to comment #2) > Before sending bug report I tried to debug it in VT and the system hanged. > However now it worked with gnome-terminal in gnome shell: That just means that it couldn't get permission to take over the screen. Running it from a VT won't work either, since fbdev will get stuck. Try something like this. $ gdb gnome-shell --wayland --display-server & (sleep 10; sudo chvt 2) Quickly type "r" then enter, and wait 10 seconds. Make sure you have chvt installed first.
(In reply to comment #3) > (In reply to comment #0) > > (gnome-shell:4836): mutter-WARNING **: failed to unlink stale lock file > > /tmp/.X1-lock: Operation not permitted > > Does it work if you manually clean up this file? Still crashes but without this warning. (In reply to comment #4) > (In reply to comment #2) > > Before sending bug report I tried to debug it in VT and the system hanged. > > However now it worked with gnome-terminal in gnome shell: > > That just means that it couldn't get permission to take over the screen. > Running it from a VT won't work either, since fbdev will get stuck. Try > something like this. > > $ gdb gnome-shell --wayland --display-server & (sleep 10; sudo chvt 2) > > Quickly type "r" then enter, and wait 10 seconds. Make sure you have chvt > installed first. It didn't work ("r" went to bash instead gdb). I tried this too: gdb -ex=r --args gnome-shell --wayland --display-server & (sleep 10; sudo chvt 8) It switches to VT 8 and nothing happens. I tried to reproduce the bug again using gnome-terminal in gnome-shell, when gnome-shell is started via startx without root. I got different backtrace this time: $ gdb -ex=r --args gnome-shell --replace --wayland --display-server GNU gdb (GDB) 7.8.1 Copyright (C) 2014 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-unknown-linux-gnu". Type "show configuration" for configuration details. For bug reporting instructions, please see: <http://www.gnu.org/software/gdb/bugs/>. Find the GDB manual and other documentation resources online at: <http://www.gnu.org/software/gdb/documentation/>. For help, type "help". Type "apropos word" to search for commands related to "word"... Reading symbols from gnome-shell...done. Starting program: /usr/bin/gnome-shell --replace --wayland --display-server warning: Could not load shared library symbols for linux-vdso.so.1. Do you need "set solib-search-path" or "set sysroot"? [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/libthread_db.so.1". [New Thread 0x7fffe30e9700 (LWP 2593)] [New Thread 0x7fffe28e8700 (LWP 2594)] (gnome-shell:2589): mutter-WARNING **: Could not take control: GDBus.Error:System.Error.EBUSY: Device or resource busy radeon: Failed to get PCI ID, error number -13 [New Thread 0x7fffe0ada700 (LWP 2595)] [New Thread 0x7fffd9eea700 (LWP 2596)] (gnome-shell:2589): Cogl-WARNING **: ./winsys/cogl-winsys-egl-kms.c:747: Error restoring saved CRTC (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event4: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event3: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event6: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event0: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event1: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event7: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event8: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event9: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event10: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event11: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event12: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event13: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event14: Permission denied (gnome-shell:2589): Clutter-WARNING **: Could not open device /dev/input/event5: Permission denied [New Thread 0x7fffd8e40700 (LWP 2597)] radeon: Failed to get PCI ID, error number -13 [New Thread 0x7fffcffff700 (LWP 2598)] [New Thread 0x7fffcf7fe700 (LWP 2599)] [New Thread 0x7fffceffd700 (LWP 2600)] (gnome-shell:2589): mutter-WARNING **: failed to unlink stale lock file /tmp/.X1-lock: Operation not permitted Disabling glamor and dri3, EGL setup failed Failed to initialize glamor, falling back to sw [New Thread 0x7fffce7fc700 (LWP 2606)] [New Thread 0x7fffcdffb700 (LWP 2607)] [New Thread 0x7fffcd7fa700 (LWP 2608)] Gjs-Message: JS LOG: Failed to launch ibus-daemon: Failed to execute child process "ibus-daemon" (No such file or directory) ** (gnome-shell:2589): WARNING **: Unable to register authentication agent: GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: An authentication agent already exists for the given subject Gjs-Message: JS LOG: Failed to register AuthenticationAgent Gjs-Message: JS LOG: No permission to trigger offline updates: Polkit.Error: GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: Action org.freedesktop.packagekit.trigger-offline-update is not registered (gnome-shell:2589): Clutter-CRITICAL **: clutter_input_device_get_pointer_stage: assertion 'CLUTTER_IS_INPUT_DEVICE (device)' failed Program received signal SIGSEGV, Segmentation fault. sync_focus_surface (pointer=0x8abaf8, pointer=0x8abaf8) at wayland/meta-wayland-pointer.c:277 277 wayland/meta-wayland-pointer.c: No such file or directory. (gdb) bt full
+ Trace 234362
(In reply to comment #5) > It switches to VT 8 and nothing happens. I meant to say that you run this from VT1, switch to VT2, and then switch back to VT1 and then look at gdb. fbcon should kick back in. If not, you can always just do `gdb attach -p $(pidof gnome-shell)` from your other VT.
Oh, and yeah, that's just another error message that happens when gnome-shell can't take over control of the display (since Xorg has taken it already). Also, you are on a systemd system with logind enabled, right?
(In reply to comment #6) > I meant to say that you run this from VT1, switch to VT2, and then switch back > to VT1 and then look at gdb. fbcon should kick back in. If not, you can always > just do `gdb attach -p $(pidof gnome-shell)` from your other VT. OK, it worked. I got similar backtrace as in comment #5. Tried as root too and got similar backtrace as in comment #2. (In reply to comment #7) > Also, you are on a systemd system with logind enabled, right? Yes
Did you see any warnings when you ran it without root, or just the same backtrace as comment #5 all alone?
(In reply to comment #9) > Did you see any warnings when you ran it without root, or just the same > backtrace as comment #5 all alone? Different warnings, here is the full output: (gdb) Starting program: /usr/bin/gnome-shell --wayland --display-server warning: Could not load shared library symbols for linux-vdso.so.1. Do you need "set solib-search-path" or "set sysroot"? [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/libthread_db.so.1". [New Thread 0x7fffe30e8700 (LWP 818)] [New Thread 0x7fffe28e7700 (LWP 819)] [New Thread 0x7fffe0ad9700 (LWP 820)] [New Thread 0x7fffd9cdd700 (LWP 822)] glamor: EGL version 1.4 (DRI2): Window manager warning: Failed to set environment variable DISPLAY for gnome-session: GDBus.Error:org.freedesktop.DBus.Error.NameHasNoOwner: Name "org.gnome.SessionManager" does not exist Window manager warning: Failed to set environment variable WAYLAND_DISPLAY for gnome-session: GDBus.Error:org.freedesktop.DBus.Error.NameHasNoOwner: Name "org.gnome.SessionManager" does not exist [New Thread 0x7fffd92d5700 (LWP 846)] [New Thread 0x7fffd8ad4700 (LWP 847)] [New Thread 0x7fffcbffe700 (LWP 848)] [New Thread 0x7fffcadff700 (LWP 849)] Gjs-Message: JS LOG: Failed to launch ibus-daemon: Failed to execute child process "ibus-daemon" (No such file or directory) Gjs-Message: JS LOG: No permission to trigger offline updates: Polkit.Error: GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: Action org.freedesktop.packagekit.trigger-offline-update is not registered (gnome-shell:814): Clutter-CRITICAL **: clutter_input_device_get_pointer_stage: assertion 'CLUTTER_IS_INPUT_DEVICE (device)' failed Program received signal SIGSEGV, Segmentation fault. sync_focus_surface (pointer=0x7a64a8, pointer=0x7a64a8) at wayland/meta-wayland-pointer.c:277 277 wayland/meta-wayland-pointer.c: No such file or directory. (gdb) #0 sync_focus_surface (pointer=0x7a64a8, pointer=0x7a64a8) at wayland/meta-wayland-pointer.c:277 display = <optimized out> focus_surface = 0x0 interface = <error reading variable interface (Cannot access memory at address 0x0)>
+ Trace 234363
So I'm a bit stumped. interface should never be 0 there. Can you try these commands? p *pointer p pointer->default_grab p default_pointer_grab_interface
Output: (gdb) $1 = {display = 0x0, resource_list = {prev = 0x0, next = 0x0}, focus_resource_list = {prev = 0x0, next = 0x0}, focus_surface = 0x0, focus_surface_listener = {link = {prev = 0x0, next = 0x0}, notify = 0x0}, focus_serial = 0, click_serial = 0, cursor_tracker = 0x0, cursor_surface = 0x0, cursor_surface_destroy_listener = {link = {prev = 0x0, next = 0x0}, notify = 0x0}, hotspot_x = 0, hotspot_y = 0, grab = 0x0, default_grab = {interface = 0x0, pointer = 0x0}, grab_button = 0, grab_serial = 0, grab_time = 0, grab_x = 0, grab_y = 0, device = 0x0, current = 0x0, button_count = 0} (gdb) $2 = {interface = 0x0, pointer = 0x0} (gdb) $3 = {focus = 0x7ffff69c6130 <default_grab_focus>, motion = 0x7ffff69c5b60 <default_grab_motion>, button = 0x7ffff69c5ff0 <default_grab_button>}
That is extremely bizarre. Can you see if meta_wayland_pointer_init or meta_wayland_pointer_release are being called before the crash, by setting breakpoints?
Both never get called :)
Does your machine have a mouse? Can you put a breakpoint on meta_wayland_seat_set_capabilities and see if that's called? If it is, can you: p flags ?
Yes, it does have a mouse. Output: Breakpoint 1, meta_wayland_seat_set_capabilities (seat=seat@entry=0x7a6380, flags=2) at wayland/meta-wayland-seat.c:143 143 wayland/meta-wayland-seat.c: No such file or directory. $1 = 2
That's very bizarre. Apparently, we don't think your system has a mouse, never initialize it, and then crash. There's not much I can tell you to try without you having to compile your own Clutter/Mutter.
It's fine, really. I have already recompiled glib2, gnome-shell, mutter for this bug report so they have debug symbols. Arch build system makes this really easy.
Can you look here: https://git.gnome.org/browse/clutter/tree/clutter/evdev/clutter-device-manager-evdev.c?h=clutter-1.20#n827 And see if it gets called, and it takes the POINTER_DEVICE path? Can you add print statements, etc. to see what kinds of devices are added? Under a traditional X11 session, can you run "xinput list" and show the results? That's the simplest way to show the set of evdev devices.
(In reply to comment #19) > Can you look here: > > https://git.gnome.org/browse/clutter/tree/clutter/evdev/clutter-device-manager-evdev.c?h=clutter-1.20#n827 > > And see if it gets called, and it takes the POINTER_DEVICE path? Can you add > print statements, etc. to see what kinds of devices are added? > > Under a traditional X11 session, can you run "xinput list" and show the > results? That's the simplest way to show the set of evdev devices. Will post the clutter results soon. Anyway, I unplugged my mouse & keyboard (one wireless USB stick for both) and plugged other wired mouse and keyboard, and the GNOME Wayland session worked! So this is about my specific mouse. It may be related to another issue I have with this mouse & keyboard, they are getting detected as three input devices. In games the third input device is treated as joystick which cause the games to misbehave. Maybe this assertion: (gnome-shell:814): Clutter-CRITICAL **: clutter_input_device_get_pointer_stage: assertion 'CLUTTER_IS_INPUT_DEVICE (device)' failed fails on this third input device? xinput output: (under X11 GNOME Shell) ⎡ Virtual core pointer id=2 [master pointer (3)] ⎜ ↳ Virtual core XTEST pointer id=4 [slave pointer (2)] ⎜ ↳ Microsoft Microsoft® 2.4GHz Transceiver v6.0 id=9 [slave pointer (2)] ⎜ ↳ Microsoft Microsoft® 2.4GHz Transceiver v6.0 id=10 [slave pointer (2)] ⎣ Virtual core keyboard id=3 [master keyboard (2)] ↳ Virtual core XTEST keyboard id=5 [slave keyboard (3)] ↳ Power Button id=6 [slave keyboard (3)] ↳ Power Button id=7 [slave keyboard (3)] ↳ Microsoft Microsoft® 2.4GHz Transceiver v6.0 id=8 [slave keyboard (3)] also, relevant dmesg output: [ 1.448453] input: Microsoft Microsoft® 2.4GHz Transceiver v6.0 as /devices/pci0000:00/0000:00:1a.0/usb1/1-1/1-1:1.0/0003:045E:0745.0001/input/input2 [ 1.448639] hid-generic 0003:045E:0745.0001: input,hidraw0: USB HID v1.11 Keyboard [Microsoft Microsoft® 2.4GHz Transceiver v6.0] on usb-0000:00:1a.0-1/input0 [ 1.449206] input: Microsoft Microsoft® 2.4GHz Transceiver v6.0 as /devices/pci0000:00/0000:00:1a.0/usb1/1-1/1-1:1.1/0003:045E:0745.0002/input/input3 [ 1.449243] hid-generic 0003:045E:0745.0002: input,hidraw1: USB HID v1.11 Mouse [Microsoft Microsoft® 2.4GHz Transceiver v6.0] on usb-0000:00:1a.0-1/input1 [ 1.466570] input: Microsoft Microsoft® 2.4GHz Transceiver v6.0 as /devices/pci0000:00/0000:00:1a.0/usb1/1-1/1-1:1.2/0003:045E:0745.0003/input/input4 [ 1.466649] hid-generic 0003:045E:0745.0003: input,hiddev0,hidraw2: USB HID v1.11 Device [Microsoft Microsoft® 2.4GHz Transceiver v6.0] on usb-0000:00:1a.0-1/input2
clutter results: Keyboard device added Keyboard device added Keyboard device added Keyboard device added
You can use G_DEBUG=fatal-warnings to track down where the Clutter warning is coming from, and see if that's relevant. I'd try to figure out why we're not getting a pointer device added here.
gdb output with G_DEBUG=fatal-warnings: warning: Could not load shared library symbols for linux-vdso.so.1. Do you need "set solib-search-path" or "set sysroot"? [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/libthread_db.so.1". [New Thread 0x7fffe30e3700 (LWP 28474)] [New Thread 0x7fffe28e2700 (LWP 28475)] [New Thread 0x7fffe0ad4700 (LWP 28476)] [New Thread 0x7fffd9cdd700 (LWP 28478)] glamor: EGL version 1.4 (DRI2): Window manager warning: Failed to set environment variable DISPLAY for gnome-session: GDBus.Error:org.freedesktop.DBus.Error.NameHasNoOwner: Name "org.gnome.SessionManager" does not exist Window manager warning: Failed to set environment variable WAYLAND_DISPLAY for gnome-session: GDBus.Error:org.freedesktop.DBus.Error.NameHasNoOwner: Name "org.gnome.SessionManager" does not exist [New Thread 0x7fffd92d5700 (LWP 28514)] [New Thread 0x7fffd8ad4700 (LWP 28515)] [New Thread 0x7fffcbffe700 (LWP 28516)] [New Thread 0x7fffcadff700 (LWP 28517)] Gjs-Message: JS LOG: Failed to launch ibus-daemon: Failed to execute child process "ibus-daemon" (No such file or directory) Gjs-Message: JS LOG: No permission to trigger offline updates: Polkit.Error: GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: Action org.freedesktop.packagekit.trigger-offline-update is not registered Keyboard device added Keyboard device added Keyboard device added Keyboard device added (gnome-shell:28470): Clutter-CRITICAL **: clutter_input_device_get_pointer_stage: assertion 'CLUTTER_IS_INPUT_DEVICE (device)' failed Program received signal SIGTRAP, Trace/breakpoint trap. g_logv (log_domain=0x7ffff5df7e65 "Clutter", log_level=G_LOG_LEVEL_CRITICAL, format=<optimized out>, args=args@entry=0x7fffffffdf10) at gmessages.c:1046 1046 gmessages.c: No such file or directory.
+ Trace 234376
Yeah, that's just a side effect of Wayland not thinking you have a pointer. Try this: $ sudo evemu-describe Poke around for the three mouse/keyboard devices. Do they all list EV_KEY / EV_REL as supported events?
Created attachment 291500 [details] event2 description event0 has EV_KEY. event1 has both. event2 has both, and also long, weird output (attached).
OK, I think that event2 is not the cause here, I tried "sudo rm /dev/input/event2" and gnome-shell still crashed.
Could you dump out descriptions for event0 and event1 as well?
Created attachment 291503 [details] event0 output
Created attachment 291504 [details] event1 output
I also did another test: connected another wired mouse without unplugging my current mouse and keyboard, and the wayland session didn't crash. So it's probably the mouse (event1) not being detected as a mouse.
After some talking with libinput dev Hans de Goede, it's apparently common for mice in mouse/keyboard combos to claim they have the multimedia keys (as seen in the output of comment 29), this enables in the libinput_device both the keyboard and pointer capabilities, and makes clutter mistakenly prefer a keyboard device for this. I'm moving to Clutter for further discussion, the patch I'm going to attach just inverts the ClutterInputDeviceType guessing out of capabilities, which is slightly more correct, but I have not much idea about how do key events from that device behave like along the stack. Next week I'll get hold of one of these devices, so will be able to try then. As for the event2 device, it really looks like some subproduct, if I'm right it's probably not the source of any events. that might be checked with xinput test-xi2 on x11.
Created attachment 291651 [details] [review] evdev: Prefer pointer/touch devices over keyboard devices In keyboard/mouse wireless combos, it is rather common for the mouse to claim it contains the multimedia keys, this makes libinput enable both the pointer and keyboard capabilities on this device, and Clutter thus to create a keyboard ClutterInputDevice for it. Ideally clutter devices should be able to reflect their full capabilities, or maybe account for the fact that certain events can be sent from seemingly unexpected device types. But this will bring a somewhat better behavior on such devices.
Review of attachment 291651 [details] [review]: Sure. Do we know if this better matches xf86-input-evdev?
mostly. evdev sets anything with keys as XI_KEYBOARD type, but the server gives preference to pointers. so pointer devices with keys will show up as pointer devices in XI2. problem is that you'll also get keyboards with axes which will give you the inverse problem, they're just rarer than mice with keys.
Comment on attachment 291651 [details] [review] evdev: Prefer pointer/touch devices over keyboard devices Ok, I pushed the patch at the moment, still would be nice to reflect capabilities better though, not sure if through evdev specific or generic APIs... Attachment 291651 [details] pushed as ddacf33 - evdev: Prefer pointer/touch devices over keyboard devices
Retitling for clarity
Thanks so much for helping us track this down, yashax! This was a tricky one.
Glad to help :) Thanks for the great support! I can use GNOME on Wayland now :) Should I open another bug report for the crash that occurs when there is no pointer device available? (GNOME on X doesn't crash in that case)
(In reply to comment #38) > Should I open another bug report for the crash that occurs when there is no > pointer device available? (GNOME on X doesn't crash in that case) Please do, crashing is always unintended, and this bug already drifted onto covering the other aspect :)
Bug report for the crash: https://bugzilla.gnome.org/show_bug.cgi?id=741494
GNOME is going to shut down bugzilla.gnome.org in favor of gitlab.gnome.org. As part of that, we are mass-closing older open tickets in bugzilla.gnome.org which have not seen updates for a longer time (resources are unfortunately quite limited so not every ticket can get handled). If you can still reproduce the situation described in this ticket in a recent and supported software version of clutter, then please follow https://wiki.gnome.org/GettingInTouch/BugReportingGuidelines and create a ticket at https://gitlab.gnome.org/GNOME/clutter/-/issues/ Thank you for your understanding and your help.