My view is that the kernel-modules-hook package should be a required dependency of the linux-meta package.
That would preserve the modules of the currently running kernel after a kernel update until the next reboot.
2 Likes
So I found another post experiencing the same issue, and followed their suggestion to try pacman, which reported the conflict in clear text. Removing tlp worked for me also, as they reported. Upgrade completed without any problem. Thanks again for the help.
@purplemunchies and @thwright can you both do the following steps on your PC and try to boot into 6.12 kernel again. Most likely early KMS might be needed with newer kernels as @fonic pointed out. Step one is not needed as we enable that by default now.
More information about it here: NVIDIA - ArchWiki
Normally there shouldnāt be any issue when removing a running kernel and replacing it with a new one. What exactly happened on your end @gamert? Did the laptop crash during the update or it simply didnāt work anymore after a reboot into the new kernel?
The only issue Iāve seen is that immediately after the update, the icon+text icons in all toolbars in all programs using KDE toolbars reverted to icon+blank. There was space there as if white text was on white background, and on hovering, the text appeared as tooltip.
So I went to Plasma settings, looked in āFontsā, looked at āToolbarā, and found that no font was selected. So I selected āDejaVu Sans Condensed 11ptā, saved change, then exited-and-relaunched the apps that were having the problem, and voila, the problem is fixed. So apparently whatever font Iād been using for toolbars has been discontinued in the latest update.
1 Like
KDE Plasma becomes black after sleep and wake up. Opened windows works, but wallpaper, bottom bar are black now. Iām not sure how to provide more detail to locate the cause?
Iām using intel NUC, no Nvidia hardware
plasmashell can bring Plasma back, but this command must keep running. (Iām not familiar with these stuff)
2 Likes
Given that I read somewhere above that kernel 6.12 would probably go LTS, I gave a spin to it. I have a couple of problems with it, one important and the other oneā¦ I think less, but I am not sure.
Here is my system, running flawlessly with 6.6:
Result of inxi -v7 -G
System:
Host: ramoth Kernel: 6.12.1-4-MANJARO arch: x86_64 bits: 64 compiler: gcc
v: 14.2.1 clocksource: tsc
Desktop: GNOME v: 47.2 tk: GTK v: 3.24.43 wm: gnome-shell
tools: gsd-screensaver-proxy dm: GDM v: 47.0 Distro: Manjaro
base: Arch Linux
Machine:
Type: Desktop Mobo: N/A model: HX90 serial:
uuid: UEFI: American Megatrends LLC. v: 5.19
date: 10/11/2021
Battery:
Device-1: hidpp_battery_0 model: Logitech Wireless Keyboard
serial: 00-00-00-00 charge: 55% (should be ignored) rechargeable: yes
status: discharging
Memory:
System RAM: total: 32 GiB available: 30.77 GiB used: 3.92 GiB (12.7%)
Message: For most reliable report, use superuser + dmidecode.
Array-1: capacity: 32 GiB slots: 2 modules: 2 EC: None
max-module-size: 16 GiB note: est.
Device-1: Channel-A DIMM 0 type: DDR4 detail: synchronous unbuffered
(unregistered) size: 16 GiB speed: 3200 MT/s volts: 1 note: check
width (bits): data: 64 total: 64 manufacturer: Kingston
part-no: CBD32D4S2S8MF-16 serial: FD9C42F5
Device-2: Channel-B DIMM 0 type: DDR4 detail: synchronous unbuffered
(unregistered) size: 16 GiB speed: 3200 MT/s volts: 1 note: check
width (bits): data: 64 total: 64 manufacturer: Kingston
part-no: CBD32D4S2S8MF-16 serial: FF9C48CF
CPU:
Info: 8-core model: AMD Ryzen 9 5900HX with Radeon Graphics bits: 64
type: MT MCP smt: enabled arch: Zen 3 rev: 0 cache: L1: 512 KiB L2: 4 MiB
L3: 16 MiB
Speed (MHz): avg: 1099 min/max: 400/4680 boost: enabled cores: 1: 1099
2: 1099 3: 1099 4: 1099 5: 1099 6: 1099 7: 1099 8: 1099 9: 1099 10: 1099
11: 1099 12: 1099 13: 1099 14: 1099 15: 1099 16: 1099 bogomips: 105441
Flags: 3dnowprefetch abm adx aes aperfmperf apic arat avic avx avx2 bmi1
bmi2 bpext cat_l3 cdp_l3 clflush clflushopt clwb clzero cmov cmp_legacy
constant_tsc cpb cppc cpuid cqm cqm_llc cqm_mbm_local cqm_mbm_total
cqm_occup_llc cr8_legacy cx16 cx8 de debug_swap decodeassists erms
extapic extd_apicid f16c flushbyasid fma fpu fsgsbase fsrm fxsr fxsr_opt
ht hw_pstate ibpb ibrs ibs invpcid irperf lahf_lm lbrv lm mba mca mce
misalignsse mmx mmxext monitor movbe msr mtrr mwaitx nonstop_tsc nopl npt
ā¦but basically a AMD Ryzen 9 with integrated graphics, running Gnome over Wayland.
The big problem is that the system is not able to resume from suspend. With kernel 6.6 there is no problem. No debug info also; the last lines of the journal before going to sleep are:
Dec 04 08:33:32 ramoth systemd[1]: Reached target Sleep.
Dec 04 08:33:32 ramoth wpa_supplicant[1411]: wlp3s0: CTRL-EVENT-DSCP-POLICY clear_all
Dec 04 08:33:32 ramoth wpa_supplicant[1411]: wlp3s0: CTRL-EVENT-DSCP-POLICY clear_all
Dec 04 08:33:32 ramoth wpa_supplicant[1411]: nl80211: deinit ifname=wlp3s0 disabled_11b_rates=0
Dec 04 08:33:32 ramoth geoclue[2508]: Failed to query location: Query location SOUP error: Unknown >
Dec 04 08:33:32 ramoth systemd[1]: Reached target Sleep.
Dec 04 08:33:32 ramoth systemd[1]: Starting System Suspend...
Dec 04 08:33:32 ramoth systemd-sleep[9953]: Successfully froze unit 'user.slice'.
Dec 04 08:33:32 ramoth systemd-sleep[9953]: Performing sleep operation 'suspend'...
Dec 04 08:33:32 ramoth kernel: PM: suspend entry (deep)
Dec 04 08:33:32 ramoth kernel: Filesystems sync: 0.026 seconds
ā¦and then there is no way to have the system back. If I try to resume, the āonā light stops blinking, and the fan starts, but nothing happens. There are no logs on the journal.
The less critical problem is that the WiFi interface is complaining all the time about noise (it does not with 6.6); it seems to work anyway (but it would be nice to be able to silence it). The WiFi is up and working.
Dec 04 08:33:32 ramoth systemd[1]: Reached target Sleep.
Dec 04 08:33:32 ramoth wpa_supplicant[1411]: wlp3s0: CTRL-EVENT-DSCP-POLICY clear_all
Dec 04 08:33:32 ramoth wpa_supplicant[1411]: wlp3s0: CTRL-EVENT-DSCP-POLICY clear_all
Dec 04 08:33:32 ramoth wpa_supplicant[1411]: nl80211: deinit ifname=wlp3s0 disabled_11b_rates=0
Dec 04 08:33:32 ramoth geoclue[2508]: Failed to query location: Query location SOUP error: Unknown >
Dec 04 08:33:32 ramoth systemd[1]: Reached target Sleep.
Dec 04 08:33:32 ramoth systemd[1]: Starting System Suspend...
Dec 04 08:33:32 ramoth systemd-sleep[9953]: Successfully froze unit 'user.slice'.
Dec 04 08:33:32 ramoth systemd-sleep[9953]: Performing sleep operation 'suspend'...
Dec 04 08:33:32 ramoth kernel: PM: suspend entry (deep)
Dec 04 08:33:32 ramoth kernel: Filesystems sync: 0.026 seconds
It always take time to stabilize a kernel. 6.6 is a save bet and you can keep running it a little longer @Rmano
1 Like
Everything worked flawless. I didnāt believe it would when I saw the list of packages pamac wanted to change, but I let pamac do how it wanted and it delivered against my expectations. I would have thought before, that itās better to have a reboot in between and not uninstall a running kernel. Thanks for the clarification, that this is false knowledge.
Hi Philm,
I tried the steps you wanted me to do and it still results with kernel 6.12.x booting to a black screen.
I had the same issue (the pamac GUI disappeared), but while I was looking for solutions, it just kept going in the backgroundā¦so when I finally ran pacman - Syu I just got the output, that everything was up to date already 
I was a bit nervous to restart afterwards, but it worked and so far everything seems fineā¦
Question: Would it have been okay, to use the ākillā command to stop pamac and then start the update again with pacman? Or would it cause just more trouble to kill pamac in the middle of an update?
1 Like
It might be best to revert to kernel 6.6 until 6.12 settles a bit.
A lot of application buttons have been ātransparentā since the update. Like the āauthenticateā and ācancelā buttons in pamac and other application input buttons. Theyāre barely readable because the colors of the fonts and backgrounds are almost identical.
Would begin with checking if /usr/lib/qt6/QtWebEngineProcess is executable. [2787:2787:1203/161719.326285:FATAL:zygote_host_impl_linux.cc(201)] seems like an error from QtWebEngine which should not prevent Plasma from running.
Arey you on X11 or Wayland? Can you check kwin logs?
A post was split to a new topic: SSH connection refused following update
Thatāll cause only troubles. Better is to wait while pamac finish update.
3 Likes
In my opinion, itās better to update from the command line, the pamac GUI often fails and causes problems
2 Likes
I havenāt updated yet but the wiki post mentions GTK4 apps not running due to GNOME switching to the GSK Renderer by default. It also says:
The thing is I donāt know if any of my apps uses GTK4 and how I verify if Vulkan support is configured properly. inxi -G doesnāt list any vulkan API:
inxi -G
Graphics:
Device-1: Advanced Micro Devices [AMD/ATI] Navi 23 [Radeon RX 6600/6600 XT/6600M] driver: amdgpu v: kernel
Display: wayland server: X.org v: 1.21.1.14 with: Xwayland v: 24.1.4
compositor: gnome-shell v: 46.5 driver: X: loaded: amdgpu
unloaded: modesetting,radeon dri: radeonsi gpu: amdgpu
resolution: 1920x1080~144Hz
API: EGL v: 1.5 drivers: kms_swrast,radeonsi,swrast
platforms: gbm,wayland,x11,surfaceless,device
API: OpenGL v: 4.6 compat-v: 4.5 vendor: amd mesa v: 24.2.4-arch1.0.1
renderer: AMD Radeon RX 6600 (radeonsi navi23 LLVM 18.1.8 DRM 3.54
6.6.54-2-MANJARO)
āvulkan-radeonā as well as the āvulkan-icd-loaderā are installed (according to pamac) but Iām not sure if this is considered as āsetup properlyā.
I think itās as simple as making sure the vulkan-tools pakagges is installed:
pamac install vulkan-tools
ā¦and running vkcube:
$ vkcube
Selected GPU 0: NVIDIA GeForce GTX 960, type: DiscreteGpu
And/or see:
2 Likes
Thanks for the tip. Iāve installed vulkan-tools and vkcube is returning the expected result.
$ vkcube
Selected GPU 0: AMD Radeon RX 6600 (RADV NAVI23), type: DiscreteGpu
Same goes for vulkaninfo mentioned in your linked website. I keep my fingers crossed that the update will finish without any issues.
After reading all the posts in this thread, Iām so glad I decided for a Radeon card in the past. phew! 
2 Likes
Have you done this?
Add a config command to
etc/modprobe.d/nouveau.conf
Specifically this command to the nouveau.conf file
nouveau.config=NvGspRm=1
Then install vulkan-nouveau from the pkg mgr.
Reboot. Like a new video card.