[Testing Update] 2022-12-04 - Kernels, Systemd, Mesa, GNOME

[openminded]

Luckily I have no AMD hardware.
And I have some experience of turning Manjaro into Arch, so in case of emergency …

[A4orce84]

My systemd package did NOT get downgraded to version 251.8 and it currently on 252.2-2 manjaro. Did something fail during the update for me?

Thanks in advance to everyone for your time and help, it is greatly appreciated!

[SDN]

You need to “downgrade” using pacman -Syyuu command, or set enable downgrade in pamac gui settings and then update.

[zomberman]

Or you can update adding the systemd packages after the -Suy, in my case:

pacman -Syu systemd systemd-libs systemd-sysvcompat

[brannon]

What are the specifics on the systemd v252 issues? I’m anxious to get this fix included: https://github.com/systemd/systemd/pull/23348

[Yochanan]

That PR was merged back in June. The shutdown script mentioned does not actually work. We used to include it, but haven’t for several months.

[brannon]

I recognize that the nVidia up/down situation is a mess of failures, especially from the suspend state, and that their shutdown script does nothing to fix the situation. I just don’t want the “umount oldroot” warning messages that show up on every reboot.

[omano]

This was working for me

@DAC324

Create a script with the following contents

#!/bin/sh
# Remove all Nvidia modules on shutdown
# to avoid errors like
# [35509.312925] sd-umoun[56020]: Failed to unmount /oldroot: Device or resource busy
# [35509.317505] sd-umoun[56021]: Failed to unmount /oldroot/sys: Device or resource busy
# [35509.322349] shutdown[1]: Failed to finalize file systems, ignoring.
#
for MODULE in nvidia_drm nvidia_modeset nvidia_uvm nvidia
do
rmmod $MODULE
done

Name this script, for example, nvidia.shutdown, make it executable as usual and place it in /usr/lib/systemd/system-shutdown.

[brannon]

Will that shutdown script get executed on suspend as well as shutdown? I have issues where my laptop commonly won’t wake up after suspend, getting stuck somewhere in the early initialization. I’m sure it’s nvidia-related.

[openminded]

@Yochanan

shutdown script mentioned does not actually work

I’m not sure. It worked for me, then I started seeing these oldroot messages again, now I know why:

@Yochanan

but haven’t for several months.

Adding

#!/bin/sh
# remove nvidia modules
/usr/bin/modprobe -r nvidia_drm nvidia_modeset nvidia_uvm && /usr/bin/modprobe -r nvidia

as /usr/lib/systemd/system-shutdown/nvidia.shutdown and making it executable resolved it again.

[omano]

Indeed, as I said it was working. But I think some people said it wasn’t working with their laptop or whatever and the fix was promptly removed soon after. I re-added it manually back then as it was definitely working on my side…