You are not logged in.
Pages: 1
i'm using gnome/wayland and with linux 6.1.9 wayland is gone and falls back to X11, i can't even select it in the login screen. no clue how to debug this, but downgrading back to 6.1.8 fixes it.
Last edited by espritlibre (2023-02-06 23:16:16)
Offline
Can you start a compositor from the tty? If not, what error messages do you get?
You could also try starting a compositor from any x terminal emulator (for a nested wayland on xorg-backend session).
Last edited by Trilby (2023-02-03 00:41:21)
"UNIX is simple and coherent" - Dennis Ritchie; "GNU's Not Unix" - Richard Stallman
Offline
No issues here. Gnome and Sway both run Wayland mode. Intel graphics. What is your GPU?
Nothing is too wonderful to be true, if it be consistent with the laws of nature -- Michael Faraday
Sometimes it is the people no one can imagine anything of who do the things no one can imagine. -- Alan Turing
---
How to Ask Questions the Smart Way
Offline
i can't even select it in the login screen
Do you mean that the Wayland option is not there? If it is there what happens when you select it? What do you see on the screen?
no clue how to debug this
With the systemd journal, as always.
Sent from sway running kernel 6.1.9 with amdgpu.
Jin, Jîyan, Azadî
Offline
i don't get an error message, wayland is usually the default when starting. on the login screen i usually can select between 4 entries:
- GNOME
- GNOME Classic
both default to wayland
and:
- GNOME Xorg
- GNOME Classic Xorg
with linux 6.1.9 the first two entries default to Xorg the last two options are gone.
My GPUs: NVIDIA T550 Laptop GPU / Mesa Intel® Graphics (ADL GT2)
this is the journal
journalctl -b | curl -F 'file=@-' 0x0.st
Offline
GDM already starts w/ a hybrid setup on X11.
Is that also the case w/ the 6.1.8 kernel? And when you downgraded that, did you also downgrade the nvidia kernel module?
Post a journal from such boot.
Can you start a compositor from the tty?
Sidenote: please fix your hostname, https://wiki.archlinux.org/title/Networ … e_hostname - no, localhost is not a good idea.
Edit: you're not running linux 6.1.9, you're running 6.1.9-hardened1-1-hardened
Did you try the regular kernel?
Last edited by seth (2023-02-05 18:06:52)
Offline
the nvidia module gets rebuild via dkms every time i up- or downgrade the kernel.
this is the journal booted with v6.1.8
Can you start a compositor from the tty?
not sure how to do this right, i tried:
┌──(espritlibre@P14s)-[~]
└─$ sudo XDG_SESSION_TYPE=wayland dbus-run-session gnome-session
Legen Sie Ihren rechten Mittelfinger auf den Fingerabdruckleser
Fehler beim Abgleich des Fingerabdrucks
Legen Sie Ihren rechten Mittelfinger auf den Fingerabdruckleser
dbus-daemon[3823]: [session uid=0 pid=3823] Activating service name='org.gtk.vfs.Daemon' requested by ':1.0' (uid=0 pid=3850 comm="/usr/lib/gnome-session-check-accelerated")
dbus-daemon[3823]: [session uid=0 pid=3823] Successfully activated service 'org.gtk.vfs.Daemon'
dbus-daemon[3823]: [session uid=0 pid=3823] Activating service name='ca.desrt.dconf' requested by ':1.5' (uid=0 pid=3824 comm="/usr/lib/gnome-session-binary")
dbus-daemon[3823]: [session uid=0 pid=3823] Successfully activated service 'ca.desrt.dconf'
A connection to the bus can't be made
┌──(espritlibre@P14s)-[~]
└─$ dbus-run-session -- gnome-shell --display-server --wayland
libmutter-Message: 20:46:42.372: Running GNOME Shell (using mutter 43.2) as a Wayland display server
Failed to setup: Could not take control: GDBus.Error:System.Error.EBUSY: Device or resource busy%
┌──(espritlibre@P14s)-[~]
└─$ sudo dbus-run-session -- gnome-shell --display-server --wayland [1]
libmutter-Message: 20:46:52.216: Running GNOME Shell (using mutter 43.2) as a Wayland display server
Failed to setup: Could not get session ID: User 0 has no sessions%
┌──(espritlibre@P14s)-[~]
└─$ gnome-shell --wayland [1]
libmutter-Message: 20:46:59.045: Running GNOME Shell (using mutter 43.2) as a Wayland display server
Failed to setup: Could not take control: GDBus.Error:System.Error.EBUSY: Device or resource busy%
thanks for the suggestion with the hostname, i'm open for more recommendations
EDIT:
in fact i'm running self compiled linux-hardened with slight changes, but the issue with no wayland is the same with linux 6.1.9 and linux-hardened 6.1.9 from the official repo
Last edited by espritlibre (2023-02-05 20:05:57)
Offline
https://wiki.archlinux.org/title/GNOME#Wayland_sessions
DO NOT USE SUDO HERE!
find ~ -user root
Make sure you didn't plaster it w/ root owned files.
XDG_SESSION_TYPE=wayland dbus-run-session gnome-session
XDG_SESSION_TYPE=wayland gnome-session
are acceptable.
GDM starts w/o further comment on wayland as well as it starts w/o further comment on X11 in teh other journal.
So let's see the manual way.
Offline
https://wiki.archlinux.org/title/GNOME#Wayland_sessions
DO NOT USE SUDO HERE!
Noted!
$ find ~ -user root
...
/home/espritlibre/eventlog
/home/espritlibre/.gnuradio/config.conf
/home/espritlibre/lshw.lshw
/home/espritlibre/nano
the output is cropped, there was a whole lot going on because of timeshift
$ XDG_SESSION_TYPE=wayland dbus-run-session gnome-session
dbus-daemon[9268]: [session uid=1000 pid=9268] Activating service name='org.gtk.vfs.Daemon' requested by ':1.0' (uid=1000 pid=9291 comm="/usr/lib/gnome-session-check-accelerated")
dbus-daemon[9268]: [session uid=1000 pid=9268] Successfully activated service 'org.gtk.vfs.Daemon'
fusermount3: failed to access mountpoint /run/user/1000/gvfs: Permission denied
dbus-daemon[9268]: [session uid=1000 pid=9268] Activating service name='ca.desrt.dconf' requested by ':1.4' (uid=1000 pid=9269 comm="/usr/lib/gnome-session-binary")
dbus-daemon[9268]: [session uid=1000 pid=9268] Successfully activated service 'ca.desrt.dconf'
A connection to the bus can't be made
$ XDG_SESSION_TYPE=wayland gnome-session
produces a blinking cursor, when i Ctrl+C i get logged out and thrown back to the login screen. which then starts X11 again
Offline
when i Ctrl+C i get logged out and thrown back to the login screen. which then starts X11 again
This does not work from within an X11 or wayland session.
Boot the multi-user.target, 2nd link below. Log into the console and try there.
Offline
sorry, i'm doing this the first time and thanks for your help!
i booted multi-user.target and
XDG_SESSION_TYPE=wayland dbus-run-session gnome-session
starts the wayland session
XDG_SESSION_TYPE=wayland gnome-session
does nothing
Offline
So it's only GDM …
https://wiki.archlinux.org/title/GDM#Use_Xorg_backend
Obviously, you rather want to use
WaylandEnable=true
Offline
i changed the setting in /etc/gdm/custom.conf from
#WaylandEnable=false
to
WaylandEnable=true
i changed back to graphical.target
sudo systemctl set-default graphical.target
and rebooted, now i have the same behaviour again. wayland is gone and i'm on X again.
Last edited by espritlibre (2023-02-05 21:35:02)
Offline
/usr/lib/udev/rules.d/61-gdm.rules has a bunch of nvidia (and other) specific rules to run gdm_disable_wayland
# The vendor nvidia driver has multiple modules that need to be loaded before GDM can make an
# informed choice on which way to proceed, so force GDM to wait until NVidia's modules are
KERNEL!="nvidia", GOTO="gdm_nvidia_end"
SUBSYSTEM!="module", GOTO="gdm_nvidia_end"
ACTION!="add", GOTO="gdm_nvidia_end"
RUN+="/usr/bin/touch /run/udev/gdm-machine-has-vendor-nvidia-driver"
TEST{0711}!="/usr/bin/nvidia-sleep.sh", GOTO="gdm_disable_wayland"
TEST{0711}!="/usr/lib/systemd/system-sleep/nvidia", GOTO="gdm_disable_wayland"
IMPORT{program}="/bin/sh -c \"sed -e 's/: /=/g' -e 's/\([^[:upper:]]\)\([[:upper:]]\)/\1_\2/g' -e 's/[[:lower:]]/\U&/g' -e 's/^/NVIDIA_/' /proc/driver/nvidia/params\""
ENV{NVIDIA_PRESERVE_VIDEO_MEMORY_ALLOCATIONS}!="1", GOTO="gdm_disable_wayland"
IMPORT{program}="/bin/sh -c 'echo NVIDIA_HIBERNATE=`systemctl is-enabled nvidia-hibernate`'"
ENV{NVIDIA_HIBERNATE}!="enabled", GOTO="gdm_disable_wayland"
IMPORT{program}="/bin/sh -c 'echo NVIDIA_RESUME=`systemctl is-enabled nvidia-resume`'"
ENV{NVIDIA_RESUME}!="enabled", GOTO="gdm_disable_wayland"
IMPORT{program}="/bin/sh -c 'echo NVIDIA_SUSPEND=`systemctl is-enabled nvidia-suspend`'"
ENV{NVIDIA_SUSPEND}!="enabled", GOTO="gdm_disable_wayland"
LABEL="gdm_nvidia_end"
GOTO="gdm_hybrid_nvidia_laptop_check"
# If this is a hybrid graphics laptop with vendor nvidia driver, disable wayland
LABEL="gdm_hybrid_nvidia_laptop_check"
TEST!="/run/udev/gdm-machine-is-laptop", GOTO="gdm_hybrid_nvidia_laptop_check_end"
TEST!="/run/udev/gdm-machine-has-hybrid-graphics", GOTO="gdm_hybrid_nvidia_laptop_check_end"
TEST!="/run/udev/gdm-machine-has-vendor-nvidia-driver", GOTO="gdm_hybrid_nvidia_laptop_check_end"
LABEL="gdm_hybrid_nvidia_laptop_check_end"
# Disable wayland when nvidia modeset is disabled or when drivers are a lower
KERNEL!="nvidia_drm", GOTO="gdm_nvidia_drm_end"
SUBSYSTEM!="module", GOTO="gdm_nvidia_drm_end"
ACTION!="add", GOTO="gdm_nvidia_drm_end"
# disable wayland if nvidia-drm modeset is not enabled
# disable wayland for nvidia drivers versions lower than 470
# For nvidia drivers versions Above 510, keep Wayland by default
# For nvidia drivers versions 470-495, prefer Xorg by default
LABEL="gdm_nvidia_drm_end"
nb.
# If this is a hybrid graphics laptop with vendor nvidia driver, disable wayland
So check the status if the tested services and files
stat /usr/bin/nvidia-sleep.sh
systemctl status nvidia-hibernate nvidia-resume nvidia-suspend
And you can edit the file and at the end comment
…
# RUN+="/usr/lib/gdm-runtime-config set daemon PreferredDisplayServer xorg"
GOTO="gdm_end"
LABEL="gdm_disable_wayland"
# RUN+="/usr/lib/gdm-runtime-config set daemon WaylandEnable false"
…
But it very much seems that you're not supposed to run GDM on wayland w/ this config and w/ the older kernel, perhaps because of the module loading timing, GDM is in error.
Offline
thank you seth and all the others for taking the time to help me.
i enabled all the nvidia services and edited "/usr/lib/udev/rules.d/61-gdm.rules" now gnome starts again with wayland on linux 6.1.9. i wonder why wayland gets disabled, since i use this card only with prime-run..?
a weird side effect i have now is that gnome now always starts with bluetooth enabled, even though it should remember its last status on boot.
$ systemctl status bluetooth
● bluetooth.service - Bluetooth service
Loaded: loaded (/usr/lib/systemd/system/bluetooth.service; enabled; preset>
Active: active (running) since Mon 2023-02-06 23:58:51 CET; 12min ago
Docs: man:bluetoothd(8)
Main PID: 1060 (bluetoothd)
Status: "Running"
Tasks: 1 (limit: 38087)
Memory: 4.1M
CPU: 35ms
CGroup: /system.slice/bluetooth.service
└─1060 /usr/lib/bluetooth/bluetoothd
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
Feb 06 23:59:03 P14s bluetoothd[1060]: Endpoint unregistered: sender=:1.64 path>
$ systemctl status systemd-rfkill
○ systemd-rfkill.service - Load/Save RF Kill Switch Status
Loaded: loaded (/usr/lib/systemd/system/systemd-rfkill.service; static)
Active: inactive (dead) since Mon 2023-02-06 23:59:38 CET; 11min ago
Duration: 5.006s
TriggeredBy: ● systemd-rfkill.socket
Docs: man:systemd-rfkill.service(8)
Process: 2994 ExecStart=/usr/lib/systemd/systemd-rfkill (code=exited, statu>
Main PID: 2994 (code=exited, status=0/SUCCESS)
CPU: 4ms
Feb 06 23:59:33 P14s systemd[1]: Starting Load/Save RF Kill Switch Status...
Feb 06 23:59:33 P14s systemd[1]: Started Load/Save RF Kill Switch Status.
Feb 06 23:59:38 P14s systemd[1]: systemd-rfkill.service: Deactivated successful>
i'm marking this thread as solved
Last edited by espritlibre (2023-02-06 23:15:49)
Offline
i wonder why wayland gets disabled, since i use this card only with prime-run..?
If this is a hybrid graphics laptop with vendor nvidia driver, disable wayland
You meet that condition.
I still suspect it previously only worked because of a race condition.
Keep that in mind if you've other weird GDM or graphics related issues.
a weird side effect
That wasn't the case before the previous changes (w/ GDM on Xorg)?
Check the journal, make sure you didn't invalidate GDMs udev plugin and also see whether you can get away w/o the edits comment the gdm-runtime-config calls (ie. just by enabling the services)
Offline
i wonder why wayland gets disabled, since i use this card only with prime-run..?
GDM wrote:If this is a hybrid graphics laptop with vendor nvidia driver, disable wayland
You meet that condition.
I still suspect it previously only worked because of a race condition.
Keep that in mind if you've other weird GDM or graphics related issues.
alright thanks, i'll keep it in mind
a weird side effect
That wasn't the case before the previous changes (w/ GDM on Xorg)?
no, i keep bluetooth disabled and only turn it on when i need it. until kernel 6.1.8 this worked properly like it should and it remebered its last state on boot. i downgraded back to 6.1.8 and bluetooth stays disabled on boot again.
Check the journal, make sure you didn't invalidate GDMs udev plugin and also see whether you can get away w/o the edits comment the gdm-runtime-config calls (ie. just by enabling the services)
this is what i did first when you suggested me to do so, enabling the nvidia services didn't fix the no wayland issue, only after editing usr/lib/udev/rules.d/61-gdm.rules wayland was back. i don't see any udev errors related to gdm.
i compared the journals with Linux 6.1.8-hardened1-1-hardened and Linux 6.1.9-hardened1-2-hardened but the bluetooth error messages are the same
this is the journal
Offline
until kernel 6.1.8 this worked properly like it should and it remebered its last state on boot. i downgraded back to 6.1.8 and bluetooth stays disabled on boot again
The question was whether this was a consequence of your udev edits to force GDM to use wayland (or is a problem w/ the new kernel, even if you run GDM on X11)
Feb 07 19:56:05 P14s systemd[1]: Listening on Load/Save RF Kill Switch Status /dev/rfkill Watch.
Feb 07 19:56:05 P14s kernel: thinkpad_acpi: rfkill switch tpacpi_bluetooth_sw: radio is unblocked
Feb 07 19:56:06 P14s NetworkManager[987]: <info> [1675796166.2137] rfkill2: found Wi-Fi radio killswitch (at /sys/devices/pci0000:00/0000:00:14.3/ieee80211/phy0/rfkill2) (driver iwlwifi)
Feb 07 19:56:06 P14s NetworkManager[987]: <info> [1675796166.2143] manager[0xf49300c6000]: rfkill: Wi-Fi hardware radio set disabled
Feb 07 19:56:06 P14s NetworkManager[987]: <info> [1675796166.2143] manager[0xf49300c6000]: rfkill: WWAN hardware radio set disabled
Feb 07 19:56:06 P14s NetworkManager[987]: <info> [1675796166.2313] manager: rfkill: Wi-Fi enabled by radio killswitch; disabled by state file
Feb 07 19:56:06 P14s NetworkManager[987]: <info> [1675796166.2313] manager: rfkill: WWAN enabled by radio killswitch; disabled by state file
Feb 07 19:56:06 P14s NetworkManager[987]: <info> [1675796166.4411] manager: rfkill: Wi-Fi now disabled by radio killswitch
Feb 07 19:56:08 P14s kernel: rfkill: input handler disabled
Feb 07 19:56:11 P14s systemd[1]: systemd-rfkill.service: Deactivated successfully.
Feb 07 19:56:13 P14s systemd-logind[972]: New session 3 of user espritlibre.
Feb 07 19:56:15 P14s kernel: rfkill: input handler enabled
Feb 07 19:56:17 P14s systemd[1595]: Starting GNOME RFKill support service...
Feb 07 19:56:17 P14s systemd[1595]: Started GNOME RFKill support service.
Feb 07 19:56:17 P14s systemd[1595]: Reached target GNOME RFKill support target.
Feb 07 19:56:17 P14s kernel: rfkill: input handler disabled
Feb 07 19:56:26 P14s systemd[1]: systemd-rfkill.service: Deactivated successfully.
=====> Feb 07 19:56:31 P14s gsd-rfkill[2102]: Failed to set RFKill: Datenstrom hat noch einen ausstehenden Vorgang
Feb 07 19:56:36 P14s systemd[1]: systemd-rfkill.service: Deactivated successfully.
Feb 07 19:56:38 P14s NetworkManager[987]: <info> [1675796198.7504] manager: rfkill: Wi-Fi hardware radio set enabled
Feb 07 19:56:39 P14s NetworkManager[987]: <info> [1675796199.0344] manager: rfkill: Wi-Fi now enabled by radio killswitch
Feb 07 19:56:43 P14s systemd[1]: systemd-rfkill.service: Deactivated successfully.
The kernel starts the device unblocked, NM initially blocks it, then you login, and then there's somehow a collision on the BT device.
Compare the journal from a 6.1.8 boot.
If this is *not* related to the 61-gdm.rules edits, please open a new thread if you want to continue the topic.
Offline
Pages: 1