Arch has released mkinitcpio33 with some structural changes
There is a pacnew for /etc/mkinitcpio.conf. Make sure you review it. If you make any changes, don’t forget to run sudo mkinitcpio -P afterward.
Here’s a helpful pacnew script by @omano using Meld. Change it to your favorite diff program if you prefer:
pacnew.sh
#!/bin/bash
pacnews=($(/usr/bin/pacdiff --output|grep -v pacsave))
nb="${#pacnews[@]}"
if [[ $nb -gt 0 ]]; then
echo; echo; echo -e "$nb .pacnew found in system"
printf "%s\n" "${pacnews[@]}"
echo; read -p "See the PacDiff? " -n 1 -r
if [[ $REPLY =~ ^[YyOo]$ ]]; then
echo; set -euo pipefail
export PATH=/usr/bin:/usr/sbin
for i in $(/usr/bin/pacdiff --output|grep -v pacsave); do
echo; echo "Processing $i ..."
/usr/bin/meld "admin://$i" "admin://${i/.pacnew/}"
echo; read -p "Delete the file $i? " -n 1 -r
if [[ $REPLY =~ ^[YyOo]$ ]]; then
echo; sudo rm -v "$i"
fi
done
fi
fi
NVIDIA470.161.03 (legacy driver for GKxxx “Kepler” GPUs)
NVIDIA390.157 (legacy driver for GF1xx “Fermi” GPUs)
AUR (Arch User Repository) packages are neither supported by Arch nor Manjaro. Posts about them in Announcement topics are off-topic and will be flagged, moved or removed without warning.
No issue, everything went smoothly
Yes there was an issue. I was able to resolve it myself.(Please post your solution)
Yes i am currently experiencing an issue due to the update. (Please post about it)
I kept my old hooks in the mkinitcpio.conf: HOOKS=(base udev autodetect modconf block filesystems keyboard mdadm_udev) (no fsck here, systemd-fsck is in use).
I notice that keyboard is nowadays placed before block. Am I shooting myself in the foot later down the line by not doing the same? Other hooks, like kms, keymap and consolefont don’t seem relevant to me.
if you want to both “fsck” and mount rootfs in one go, forcing rootfs into RO mode will make it mount twice during boot, first in RO mode to "fsck"and then again in RW mode for regular use.
additionally you should use fsck hook in your mkinitcpio
was only a suggestion, with the sole premise being rootfs only mounted once(pimping boot process :-p) it is the default mechanism afterall. as for, if there is any advantage, don’t know.
who knows what fsck-hook does, maybe it does indeed mount rootfs in RO mode and it’s hidden from the logs (my sole evidence that it is mounted once)
systemd-modules-load[360]: Failed to find module 'amd_pstate'
when using linux61 6.1.0rc7-1
EDIT:
systemd-modules-load[360]: Failed to find module 'amd_pstate'
when using linux61 (6.1.0rc8-1)
…because of Riesen, I know…
EDIT2:
following a hint of stephane
the amd-pstate works in kernel6.1.0-1 by adding additional boot-parameter: amd_pstate=passive
There is a pacnew for 
2022-11-25
) als das AUR.
