Manjaro won't boot properly with NVIDIA drivers after GPU change

seems to be a weird motherboard

 Type: Laptop Mobo: Micro-Star model: B350M PRO-VDH (MS-7A38) v: 2.0
    serial: <superuser required> UEFI: American Megatrends v: A.00
    date: 03/21/2017

it says motherboard laptop in a genuine pc. please check if there are newer bios-updates avaiable.

Did sudo pacman-mirrors -f5 && sudo pacman -Syyu again, there’s nothing to update.

try with a newer kernel, the 6.0 - not the rc one…
if it doesnt work, check some settings in bios related to pcie, and power management;
also if it doesnt work, post logs:
journalctl -b0 --no-pager | grep nvidia
sudo dmesg | grep nvidia

there might be that your ryzen 5 1600 is one of them that is affected with the “C6” failure. you can install the “amd-disable-c6” package from the aur to prevent freezing at startup and retry to use a live-iso and boot into with propietary driver.

Installed 6.0.2.2, uninstalled driver and reinstalled through mwdh -a pci nonfree 0300.

In BIOS I only have “PCI Latency Timer (32 PCI Bus Clocks)” and “Resume By PCI-E Device (disabled)”. In Power Management it’s only “ErP Ready (disabled)”, “Restore after AC Power Loss (Power off)” and “System Power Fault Protection (disabled)”

journalctl:

okt   30 16:52:34 kraut kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-6.0-x86_64 root=UUID=a80075b0-2ce2-4998-be05-e3eb48a5191e ro quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 ibt=off nvidia_drm.modeset=1
okt   30 16:52:34 kraut kernel: Kernel command line: BOOT_IMAGE=/boot/vmlinuz-6.0-x86_64 root=UUID=a80075b0-2ce2-4998-be05-e3eb48a5191e ro quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 ibt=off nvidia_drm.modeset=1
okt   30 16:52:34 kraut kernel: audit: type=1400 audit(1667141554.076:9): apparmor="STATUS" operation="profile_load" profile="unconfined" name="nvidia_modprobe" pid=348 comm="apparmor_parser"
okt   30 16:52:34 kraut audit[348]: AVC apparmor="STATUS" operation="profile_load" profile="unconfined" name="nvidia_modprobe" pid=348 comm="apparmor_parser"
okt   30 16:52:34 kraut audit[348]: AVC apparmor="STATUS" operation="profile_load" profile="unconfined" name="nvidia_modprobe//kmod" pid=348 comm="apparmor_parser"
okt   30 16:52:34 kraut systemd-modules-load[335]: Module 'nvidia' is deny-listed
okt   30 16:52:34 kraut systemd-modules-load[335]: Module 'nvidia_drm' is deny-listed
okt   30 16:52:34 kraut systemd-modules-load[335]: Module 'nvidia_uvm' is deny-listed
okt   30 16:52:53 kraut kernel: nvidia: loading out-of-tree module taints kernel.
okt   30 16:52:53 kraut kernel: nvidia: module license 'NVIDIA' taints kernel.
okt   30 16:52:53 kraut kernel: nvidia: module verification failed: signature and/or required key missing - tainting kernel
okt   30 16:52:54 kraut kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 238
okt   30 16:52:54 kraut kernel: nvidia 0000:22:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=io+mem

dmesg:

[    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-6.0-x86_64 root=UUID=a80075b0-2ce2-4998-be05-e3eb48a5191e ro quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 ibt=off nvidia_drm.modeset=1
[    0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-6.0-x86_64 root=UUID=a80075b0-2ce2-4998-be05-e3eb48a5191e ro quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 ibt=off nvidia_drm.modeset=1
[    6.800098] audit: type=1400 audit(1667141554.076:9): apparmor="STATUS" operation="profile_load" profile="unconfined" name="nvidia_modprobe" pid=348 comm="apparmor_parser"
[   26.538197] nvidia: loading out-of-tree module taints kernel.
[   26.538214] nvidia: module license 'NVIDIA' taints kernel.
[   26.568485] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[   27.089665] nvidia-nvlink: Nvlink Core is being initialized, major device number 238
[   27.090915] nvidia 0000:22:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=io+mem

as you can see its being blacklisted…
but why it doesnt work with the live usb?
:thinking:


post output again from:
ls /etc/modprobe.d
find /etc/X11/ -name "*.conf"
pamac list -qm
pacman -Qs 'optimus|bbswitch|bumblebee|prime|gpu'

don’t worry about this messages at boot, they aren’t the issues but you’re right as long as it doesn’t boot in a live the problem isn’t solved.
did you install the package that solves the potential problem with your amd-ryzen that causes memory-errors. (keep in mind that a mismatch in memory allocation will also cause malfunction with additional hardware). is the mobo at the latest bios-version avaiable ?

1 Like

Installed it through AUR and tried to boot into LiveUSB, got stuck at the same place.

may you post the actual

glxinfo | grep server

Still the same on nvidia drivers.

ls /etc/modprobe.d
mhwd-gpu.conf

find /etc/X11/ -name "*.conf"
90-mhwd.conf
30-touchpad.conf
00-keyboard.conf

pamac list -qm

amd-disable-c6
angular-cli
aurman
breath2-wallpaper
celt
ceph-libs
ffcast
gnome-icon-theme
gnome-icon-theme-symbolic
jdk
jdk12-openj9-bin
jre
libax25
libcanberra-gstreamer
libkipi
libkipi-debug
libopenaptx
linux414-headers
manjaro-documentation-en
manjaro-firmware
ocs-url
postman-bin
pycharm-professional
python-sphinx-intl
rar
sphinx
sprunge
teams
telegram-desktop-bin
visual-studio-code-bin
wireguard-go-git
woeusb-ng
xrectsel

pacman -Qs 'optimus|bbswitch|bumblebee|prime|gpu'

local/libplacebo 4.208.0-1
    Reusable library for GPU-accelerated video/image rendering primitives
local/mhwd-amdgpu 19.1.0-1
    MHWD module-ids for amdgpu
local/optimus-manager 1.4+r12-2
    Management utility to handle GPU switching for Optimus laptops (Git version)
local/optimus-manager 1.4+r12-2

you have it installed from ollis advice…
and when you were providing logs, you were providing them when you were on video-nvidia?
were you also booting the live usb with the ibt=off parameter?

it is still disabled

@krautman

meanwhile i believe it’s a problem of your motherboard-hardware. please try this: add the following to your
GRUB_CMDLINE_LINUX_DEFAULT parameter in grub

acpi_osi='Windows 2018'

run again

sudo mkinitcpio -P
sudo update-grub

and reboot to your system on hard disk (not live). i know it’s a hassle but that’s the problem with digging out a problem.

yes, I haven’t removed the ibt=off parameter from the GRUB file but I don’t know if it affects LiveUSB. Uninstalled optimus and remembered I had change the modules in mkinitcpio.conf per the advice of another thread, changed that back to its original state.

Logs now:

okt   30 17:44:23 kraut kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-6.0-x86_64 root=UUID=a80075b0-2ce2-4998-be05-e3eb48a5191e ro quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 ibt=off nvidia_drm.modeset=1
okt   30 17:44:23 kraut kernel: Kernel command line: BOOT_IMAGE=/boot/vmlinuz-6.0-x86_64 root=UUID=a80075b0-2ce2-4998-be05-e3eb48a5191e ro quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 ibt=off nvidia_drm.modeset=1
okt   30 17:44:24 kraut audit[343]: AVC apparmor="STATUS" operation="profile_load" profile="unconfined" name="nvidia_modprobe" pid=343 comm="apparmor_parser"
okt   30 17:44:24 kraut audit[343]: AVC apparmor="STATUS" operation="profile_load" profile="unconfined" name="nvidia_modprobe//kmod" pid=343 comm="apparmor_parser"
okt   30 17:44:25 kraut kernel: nvidia: loading out-of-tree module taints kernel.
okt   30 17:44:25 kraut kernel: nvidia: module license 'NVIDIA' taints kernel.
okt   30 17:44:25 kraut kernel: nvidia: module verification failed: signature and/or required key missing - tainting kernel
okt   30 17:44:25 kraut kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 238
okt   30 17:44:25 kraut kernel: nvidia 0000:22:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=io+mem
okt   30 17:44:26 kraut systemd-modules-load[330]: Inserted module 'nvidia'
okt   30 17:44:26 kraut kernel: nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms  520.56.06  Thu Oct  6 21:22:53 UTC 2022
okt   30 17:44:26 kraut kernel: [drm] [nvidia-drm] [GPU ID 0x00002200] Loading driver
okt   30 17:44:27 kraut systemd-modules-load[330]: Inserted module 'nvidia_drm'
okt   30 17:44:27 kraut kernel: [drm:nv_drm_load [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00002200] Failed to allocate NvKmsKapiDevice
okt   30 17:44:27 kraut kernel: [drm:nv_drm_probe_devices [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00002200] Failed to register device
okt   30 17:44:28 kraut kernel: nvidia_uvm: module uses symbols nvUvmInterfaceDisableAccessCntr from proprietary module nvidia, inheriting taint.
okt   30 17:44:28 kraut systemd-modules-load[330]: Inserted module 'nvidia_uvm'
okt   30 17:44:28 kraut kernel: nvidia-uvm: Loaded the UVM driver, major device number 236.
[    0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-6.0-x86_64 root=UUID=a80075b0-2ce2-4998-be05-e3eb48a5191e ro quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 ibt=off nvidia_drm.modeset=1
[    0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-6.0-x86_64 root=UUID=a80075b0-2ce2-4998-be05-e3eb48a5191e ro quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 ibt=off nvidia_drm.modeset=1
[    7.732930] nvidia: loading out-of-tree module taints kernel.
[    7.732943] nvidia: module license 'NVIDIA' taints kernel.
[    7.754049] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[    8.186455] nvidia-nvlink: Nvlink Core is being initialized, major device number 238
[    8.187486] nvidia 0000:22:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=io+mem
[    8.507367] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms  520.56.06  Thu Oct  6 21:22:53 UTC 2022
[    8.576267] [drm] [nvidia-drm] [GPU ID 0x00002200] Loading driver
[    9.653976] [drm:nv_drm_load [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00002200] Failed to allocate NvKmsKapiDevice
[    9.654107] [drm:nv_drm_probe_devices [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00002200] Failed to register device
[   10.267875] nvidia_uvm: module uses symbols nvUvmInterfaceDisableAccessCntr from proprietary module nvidia, inheriting taint.
[   10.403073] nvidia-uvm: Loaded the UVM driver, major device number 236.

Didn’t change anything.

yes, I haven’t removed the ibt=off parameter

delete it and run

sudo mkinitcpio -P
sudo update-grub

also please post the output of your actual grub

cat /etc/default/grub | grep GRUB_CMDLINE_LINUX_DEFAULT

1 Like
[drm:nv_drm_load [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00002200] Failed to allocate NvKmsKapiDevice
[drm:nv_drm_probe_devices [nvidia_drm]] *ERROR* [nvidia-drm] [GPU ID 0x00002200] Failed to register device

you have these … looks like it could be fixed by a bios update as olli suggests…


provide also logs from these:
sudo dmesg | grep -E 'NVRM|nvrm'
journalctl -b0 | grep -E 'NVRM|nvrm'

GRUB_CMDLINE_LINUX_DEFAULT=“quiet apparmor=1 security=apparmor resume=UUID=fc7ae34c-e872-448b-91e6-f76ef484cc3a udev.log_priority=3 nvidia_drm.modeset=1 acpi_osi=‘Windows 2018’”

Oh gawd, I hope I wont brick my PC with the BIOS update :smiley:
Weirdly enough, I tried to LiveUSB into Garuda with prop drivers and it basically had the same results so at least it’s not manjaro specific.

[    8.339148] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  520.56.06  Thu Oct  6 21:38:55 UTC 2022
[    9.758663] NVRM: GPU 0000:22:00.0: RmInitAdapter failed! (0x26:0x56:1423)
[    9.758719] NVRM: GPU 0000:22:00.0: rm_init_adapter failed, device minor number 0
okt   30 18:11:14 kraut kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  520.56.06  Thu Oct  6 21:38:55 UTC 2022
okt   30 18:11:15 kraut kernel: NVRM: GPU 0000:22:00.0: RmInitAdapter failed! (0x26:0x56:1423)
okt   30 18:11:15 kraut kernel: NVRM: GPU 0000:22:00.0: rm_init_adapter failed, device minor number 0