Last update broke Nvidia drivers, unable to reinstall them

galure · March 23, 2023, 10:47am

I’m on a Kde spin install of Fedora 37.

Last update (6.2.7-200.fc37.x86_64) completely messed up my Nvidia settings and drivers, and I have a hard time installing them again.

I followed this tutorial for a fresh install : https://www.linuxcapable.com/how-to-install-nvidia-drivers-on-fedora-linux/

Here are the steps:

sudo dnf autoremove nvidia* --purge

but the --purge option doesn’t exist… anyway, I continued
I added the Nvidia CUDA Repo:

sudo dnf config-manager --add-repo
https://developer.download.nvidia.com/compute/cuda/repos/fedora37/x86_64/cuda-fedora37.repo

installed the dependencies:

sudo dnf install kernel-headers kernel-devel tar bzip2 
make automake gcc gcc-c++ pciutils elfutils-libelf-devel
libglvnd-opengl libglvnd-glx libglvnd-devel acpid pkgconfig dkms

and finally installed the drivers:

sudo dnf module install nvidia-driver:latest-dkms

I rebooted but the installation didn’t work. nvidia-settingstalked about some “core failure” but I don’t remember the exact output.

So I tried the other solution.
I first uninstalled nvidia:

sudo dnf autoremove nvidia*

I imported the repos:

sudo dnf install https://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-$(rpm -E %fedora).noarch.rpm

and

sudo dnf install https://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-$(rpm -E %fedora).noarch.rpm

Although they were already imported because my previous nvidia setup was via rpm.

Then

sudo dnf install akmod-nvidia

I tried

sudo dnf install xorg-x11-drv-nvidia-cuda

But it fails, reporting this:

All matches were filtered out by modular filtering 
for argument: xorg-x11-drv-nvidia-cuda
Error: Unable to find a match: xorg-x11-drv-nvidia-cuda

I rebooted and it doesn’t work. My system doesn’t recognize my screens and nvidia-settings is not installed. If I install it, it opens but doesn’t recognize anything either.

My graphic card is a GeForce RTX 3070
lspci -vnn | grep VGA returns:

2d:00.0 VGA compatible controller [0300]: NVIDIA Corporation GA104 
[GeForce RTX 3070] [10de:2484] (rev a1) (prog-if 00 [VGA controller])

I should say it worked perfectly fine before the last update and I don’t know why I’m not able to re-install the Nvidia drivers.

Am I doing something wrong? Is there somewhere some guidelines that I can follow to reinstall those drivers? Can I prevent this from happening in the future?

galure · March 23, 2023, 11:44am

following this discussion: Nvidia driver installation in F38 beta

I tried those steps:

sudo dnf remove '*nvidia*'
sudo dnf install nvidia-gpu-firmware
sudo dnf install akmod-nvidia --best --allowerasing

I had the same error return for

sudo dnf install xorg-x11-drv-nvidia-cuda
Last metadata expiration check: 0:50:57 ago on jeu. 23 mars 2023 11:31:30.
All matches were filtered out by modular filtering for argument: xorg-x11-drv-nvidia-cuda
Error: Unable to find a match: xorg-x11-drv-nvidia-cuda

I rebooted, then:

nvidia-smi
bash: nvidia-smi: command not found...
Install package 'nvidia-driver-cuda' to provide command 'nvidia-smi'? [N/y]

The following packages have to be installed:
 nvidia-driver-NVML-3:530.30.02-1.fc37.x86_64   NVIDIA Management Library (NVML)
 nvidia-driver-cuda-3:530.30.02-1.fc37.x86_64   CUDA integration for nvidia-driver
 nvidia-driver-cuda-libs-3:530.30.02-1.fc37.x86_64      Libraries for nvidia-driver-cuda
 nvidia-persistenced-3:530.30.02-1.fc37.x86_64  A daemon to maintain persistent software state in the NVIDIA driver
Proceed with changes? [N/y] y

And it installed nvidia-driver-cuda. Then nvidia-smi returned:

nvidia-smi
Thu Mar 23 12:19:20 2023
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 530.30.02              Driver Version: 530.30.02    CUDA Version: 12.1     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                  Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf            Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA GeForce RTX 3070         Off| 00000000:2D:00.0 Off |                  N/A |
|  0%   43C    P8               21W / 220W|     48MiB /  8192MiB |      0%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
|    0   N/A  N/A      2101      G   /usr/libexec/Xorg                            46MiB |
+---------------------------------------------------------------------------------------+

So it seems to work? Although it says NVIDIA GeForce RTX 3070 Off

Although nvidia-settings returns:

nvidia-settings
bash: nvidia-settings: command not found...
Packages providing this file are:
'nvidia-settings-390xx'
'nvidia-settings-470xx'
'xorg-x11-drv-nvidia-340xx'
'nvidia-settings'

I can install nvidia-settings, then it returns:

nvidia-settings
Segmentation fault (core dumped)

But I suppose I have to reboot at this point.

EDIT: after a reboot nvidia-settings returns the same
The nvidia-settings application doesn’t open
and this:

nvidia-settings --v

nvidia-settings:  version 530.30.02

So it “seems” that the driver is installed, right?

galure · March 23, 2023, 12:22pm

Ok, switching to a X11 session works. Wayland is all messed up.
Was that the solution all the time? Wayland seemed to work previously, but now I’m not sure of my configuration…

ilikelinux · March 23, 2023, 12:38pm

Now while changing the kernel from 6.1 > 6.2 before of releasing F38, it would really make sense, that you feed us with your Hardware info and error messages filter out from your logs.

Participation on debugging is important now.

Nvidia drivers are not opensource. Till they are up to date it takes some time!

straycat · March 23, 2023, 12:49pm

Wayland is still prone to have issues with nvidia.I was curious where you installed the 530.30.02 drivers from as rawhide’s repo on rpmfusion was the only place I saw that version.

galure · March 23, 2023, 12:52pm

lspci

00:00.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse Root Complex
00:00.2 IOMMU: Advanced Micro Devices, Inc. [AMD] Starship/Matisse IOMMU
00:01.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Hos
t Bridge
00:01.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse GPP Bridge
00:01.2 PCI bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse GPP Bridge
00:02.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Hos
t Bridge
00:03.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Hos
t Bridge
00:03.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse GPP Bridge
00:04.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Hos
t Bridge
00:05.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Hos
t Bridge
00:07.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Hos
t Bridge
00:07.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse Internal PCIe G
PP Bridge 0 to bus[E:B]
00:08.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Hos
t Bridge
00:08.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Starship/Matisse Internal PCIe G
PP Bridge 0 to bus[E:B]
00:14.0 SMBus: Advanced Micro Devices, Inc. [AMD] FCH SMBus Controller (rev 61)
00:14.3 ISA bridge: Advanced Micro Devices, Inc. [AMD] FCH LPC Bridge (rev 51)
00:18.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: De
vice 18h; Function 0
00:18.1 Host bridge: Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: De
vice 18h; Function 1
00:18.2 Host bridge: Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: De
vice 18h; Function 2
00:18.3 Host bridge: Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: De
vice 18h; Function 3
00:18.4 Host bridge: Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: De
vice 18h; Function 4
00:18.5 Host bridge: Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: De
vice 18h; Function 5
00:18.6 Host bridge: Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: De
vice 18h; Function 6
00:18.7 Host bridge: Advanced Micro Devices, Inc. [AMD] Matisse/Vermeer Data Fabric: De
vice 18h; Function 7
01:00.0 Non-Volatile memory controller: Samsung Electronics Co Ltd NVMe SSD Controller
PM9A1/PM9A3/980PRO
20:00.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Matisse Switch Upstream
21:04.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Matisse PCIe GPP Bridge
21:06.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Matisse PCIe GPP Bridge
21:08.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Matisse PCIe GPP Bridge
21:09.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Matisse PCIe GPP Bridge
21:0a.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Matisse PCIe GPP Bridge
26:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller
(rev 04)
28:00.0 Network controller: Intel Corporation Wi-Fi 6 AX200 (rev 1a)
2a:00.0 Non-Essential Instrumentation [1300]: Advanced Micro Devices, Inc. [AMD] Starsh
ip/Matisse Reserved SPP
2a:00.1 USB controller: Advanced Micro Devices, Inc. [AMD] Matisse USB 3.0 Host Control
ler
2a:00.3 USB controller: Advanced Micro Devices, Inc. [AMD] Matisse USB 3.0 Host Control
ler
2b:00.0 SATA controller: Advanced Micro Devices, Inc. [AMD] FCH SATA Controller [AHCI m
ode] (rev 51)
2c:00.0 SATA controller: Advanced Micro Devices, Inc. [AMD] FCH SATA Controller [AHCI m
ode] (rev 51)
2d:00.0 VGA compatible controller: NVIDIA Corporation GA104 [GeForce RTX 3070] (rev a1)
2d:00.1 Audio device: NVIDIA Corporation GA104 High Definition Audio Controller (rev a1
)
2e:00.0 Non-Essential Instrumentation [1300]: Advanced Micro Devices, Inc. [AMD] Starsh
ip/Matisse PCIe Dummy Function
2f:00.0 Non-Essential Instrumentation [1300]: Advanced Micro Devices, Inc. [AMD] Starsh
ip/Matisse Reserved SPP
2f:00.1 Encryption controller: Advanced Micro Devices, Inc. [AMD] Starship/Matisse Cryp
tographic Coprocessor PSPCPP
2f:00.3 USB controller: Advanced Micro Devices, Inc. [AMD] Matisse USB 3.0 Host Control
ler
2f:00.4 Audio device: Advanced Micro Devices, Inc. [AMD] Starship/Matisse HD Audio Cont
roller

lscpu

Architecture:            x86_64
  CPU op-mode(s):        32-bit, 64-bit
  Address sizes:         48 bits physical, 48 bits virtual
  Byte Order:            Little Endian
CPU(s):                  16
  On-line CPU(s) list:   0-15
Vendor ID:               AuthenticAMD
  Model name:            AMD Ryzen 7 5800X 8-Core Processor
    CPU family:          25
    Model:               33
    Thread(s) per core:  2
    Core(s) per socket:  8
    Socket(s):           1
    Stepping:            0
    Frequency boost:     enabled
    CPU(s) scaling MHz:  52%
    CPU max MHz:         4850,1948
    CPU min MHz:         2200,0000
    BogoMIPS:            7599,68
    Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov 
                         pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_
                         opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc c
                         puid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 f
                         ma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand l
                         ahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse
                          3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core pe
                         rfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate 
                         ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 
                         erms invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni 
                         xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_t
                         otal cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd ara
                         t npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid
                          decodeassists pausefilter pfthreshold avic v_vmsave_vmload vg
                         if v_spec_ctrl umip pku ospke vaes vpclmulqdq rdpid overflow_r
                         ecov succor smca fsrm
Virtualization features: 
  Virtualization:        AMD-V
Caches (sum of all):     
  L1d:                   256 KiB (8 instances)
  L1i:                   256 KiB (8 instances)
  L2:                    4 MiB (8 instances)
  L3:                    32 MiB (1 instance)
NUMA:                    
  NUMA node(s):          1
  NUMA node0 CPU(s):     0-15
Vulnerabilities:         
  Itlb multihit:         Not affected
  L1tf:                  Not affected
  Mds:                   Not affected
  Meltdown:              Not affected
  Mmio stale data:       Not affected
  Retbleed:              Not affected
  Spec store bypass:     Mitigation; Speculative Store Bypass disabled via prctl
  Spectre v1:            Mitigation; usercopy/swapgs barriers and __user pointer saniti
                         zation
  Spectre v2:            Mitigation; Retpolines, IBPB conditional, IBRS_FW, STIBP alway
                         s-on, RSB filling, PBRSB-eIBRS Not affected
  Srbds:                 Not affected
  Tsx async abort:       Not affected

As for the error messages in my logs, what should I look for? If you have a command I should run, I’m happy to do so.

bennyisaiah · March 23, 2023, 1:02pm

I also found x11 works better. Wayland is not ready for nvidia from my experience especially for video conferencing

galure · March 23, 2023, 1:09pm

I’m pretty sure I was using wayland before the last update. If I wasn’t, then the last update switched me to wayland, I don’t know… I remember at some point wayland was sort of working but had issues, then it worked fine and I think I moved to it 6 months ago, butagain, my memory is fuzzy.
I didn’t encounter any real issues, even when gaming.

But with the last update, when I start a wayland session it doesn’t even allow me to change the display configuration. I just have 3 mirror displays at very low resolution and I can’t change anything.

ilikelinux · March 23, 2023, 1:13pm

I think because of your older card and because the support for Wayland is newer than for X11. Till F35/36 by default was used X.11.

bennyisaiah · March 23, 2023, 1:13pm

Oh okay perhaps its been updated a lot since I last used it.

I do a lot of video conferencing on jitsi and skype and the screen sharing had issues last year on wayland with my nvidia system (clients couldnt see my screen and vica versa), I switched to x11 and it didnt happen

Perhaps its now working much better, ill edit the configuration and try wayland again and check it out.

What are the pro’s vs cons of wayland as a desktop user, if not for gaming is it really noticable? colour gamut any improvements?

ilikelinux · March 23, 2023, 1:20pm

X11 is what is was used since long and gets more and more security issues. While Wayland is trying to adapt to new technologies and security standards.

If you not really need the power of Nvidia for Gaming and or extreme Graphic specific apps, I propose you to use hardware who works with opensource drivers. This definitely gives you more peace with your Fedora Linux experience alias Linux in general.

Edit:

computersavvy · March 23, 2023, 3:32pm

There have been several cases where that repo and installing nvidia from it seem to have caused problems with nvidia drivers and the newer GPUs.

This may assist

galure · March 23, 2023, 5:41pm

Ok reading quickly what you suggested this may be it:

It seems to explain this prompt:

sudo dnf install xorg-x11-drv-nvidia-cuda
Last metadata expiration check: 0:50:57 ago on jeu. 23 mars 2023 11:31:30.
All matches were filtered out by modular filtering for argument: xorg-x11-drv-nvidia-cuda
Error: Unable to find a match: xorg-x11-drv-nvidia-cuda

For what I understand I have too much repo sources, and they conflict over cuda.

If I check my repos:

cuda-fedora36-x86_64              cuda-fedora36-x86_64
cuda-fedora37-x86_64              cuda-fedora37-x86_64
fedora                            Fedora 37 - x86_64
rpmfusion-free                    RPM Fusion for Fedora 37 - Free
rpmfusion-free-updates            RPM Fusion for Fedora 37 - Free - Updates
rpmfusion-nonfree                 RPM Fusion for Fedora 37 - Nonfree
rpmfusion-nonfree-updates         RPM Fusion for Fedora 37 - Nonfree - Updates
updates                           Fedora 37 - x86_64 - Updates
updates-modular                   Fedora Modular 37 - x86_64 - Updates

Among others, I do have 2 versions of cuda (36 and 37) and rpmfusion-nonfree. Akashdeep Dhar then says this:

So what exactly is happening here? The device on which the drivers are attempted to be installed has had rpmfusion-nonfree-nvidia-driver module enabled first, and then the cuda-fedora33 module was enabled due to which an attempt to fetch one of the akmod-nvidia dependencies (called xorg-x11-drv-nvidia) could not come to fruition. One might say that the inability to find xorg-x11-drv-nvidia makes no sense as it is present with the rpmfusion-nonfree-nvidia-driver module, which is in fact, enabled here.
But here’s the thing, the cuda-fedora33 module seems to have an overriding effect on the rpmfusion-nonfree-nvidia-driver module due to which dnf states to be unable to find xorg-x11-drv-nvidia even when it is present. This essentially means that both the repositories are incompatible with each other, cannot coexist simultaneously and any attempts made to install or upgrade your drivers would result in a failure

About my modules:

sudo dnf module list --enabled
cuda-fedora36-x86_64
Name          Stream             Profiles             Summary                          
nvidia-driver latest-dkms [d][e] default [d] [i], fm, Nvidia driver for latest-dkms bra
                                  ks                  nch                              
cuda-fedora37-x86_64
Name          Stream             Profiles             Summary                          
nvidia-driver latest-dkms [d][e] default [d] [i], fm, Nvidia driver for latest-dkms bra
                                  ks                  nch                              
Hint: [d]efault, [e]nabled, [x]disabled, [i]nstalled

So if i understand (I don’t, I’m just guessing here), I should back off from the cuda-fedora36-x86_64 and cuda-fedora37-x86_64 repos, uninstall everything using sudo dnf remove '*nvidia*' and run sudo dnf install akmod-nvidia xorg-x11-drv-nvidia-cuda.

Does that seem like a reasonable way to install the drivers?

computersavvy · March 23, 2023, 6:07pm

I would do that in 4 steps.

disable the cuda repos.
dnf repolist should allow you to see which cuda repo you have enabled, then
sudo dnf config-manager --disable <repoid> will usually disable that repo.
Reset the modules so they are not shown in your list. This is a reference I found.
QA:Testcase Modularity enable-disable module - Fedora Project Wiki
Now remove the still installed nvidia packages. sudo dnf remove '*nvidia*' --exclude nvidia-gpu-firmware may do it, but I cannot be sure since I have never worked directly with the cuda-fedora repos.
Finally install the packages from rpmfusion and verify the firmware package is still installed with sudo dnf install akmod-nvidia xorg-x11-drv-nvidia-cuda nvidia-gpu-firmware

If there are remaining errors about modular filtering on the cuda drivers that would need resolved as suggested in the link I provided above or the link in step 2.

galure · March 23, 2023, 6:53pm

So I followed your steps

1. Disable cuda repos

sudo dnf config-manager --disable cuda-fedora37-x86_64
sudo dnf config-manager --disable cuda-fedora36-x86_64

then

sudo dnf repolist

to verify they where not active anymore.

2. Reset the modules

sudo dnf module list --enabled
@modulefailsafe
Name           Stream           Profiles             Summary
nvidia-driver  latest-dkms [e]  default [i], fm, ks  Nvidia driver for latest-dkms branch

the command return was different from previously which was, at that time:

cuda-fedora36-x86_64
Name          Stream             Profiles             Summary
nvidia-driver latest-dkms [d][e] default [d] [i], fm, Nvidia driver for latest-dkms bra
                                  ks                  nch

cuda-fedora37-x86_64
Name          Stream             Profiles             Summary
nvidia-driver latest-dkms [d][e] default [d] [i], fm, Nvidia driver for latest-dkms bra
                                  ks                  nch

But since the nvidia-driver module was active I did this (following the QA page);

sudo dnf module reset nvidia-driver
Dependencies resolved.
==========================================================================================
 Package              Architecture        Version              Repository            Size
==========================================================================================
Disabling module profiles:
 nvidia-driver/default

Resetting modules:
 nvidia-driver

Transaction Summary
==========================================================================================

Is this ok [y/N]: y
Complete!

And I verified that the module wasn’t listed in either --installed, enabled or disabled.

3. Remove Nvidia packages

sudo dnf remove '*nvidia*' --exclude nvidia-gpu-firmware

4. Install packages from rpmfusion

sudo dnf install akmod-nvidia xorg-x11-drv-nvidia-cuda nvidia-gpu-firmware

That returned no error stating:

Package nvidia-gpu-firmware-20230310-148.fc37.noarch is already installed.
Dependencies resolved.

And for the Grand Final:

Installed:
  akmod-nvidia-3:525.89.02-1.fc37.x86_64
  akmods-0.5.7-9.fc37.noarch
  debugedit-5.0-7.fc37.x86_64
  egl-gbm-1.1.0-3.fc37.x86_64
  egl-wayland-1.1.11-1.fc37.x86_64
  fakeroot-1.30.1-1.fc37.x86_64
  fakeroot-libs-1.30.1-1.fc37.x86_64
  http-parser-2.9.4-7.fc37.x86_64
  kmodtool-1.1-5.fc37.noarch
  koji-1.32.0-1.fc37.noarch
  libgit2-1.3.2-1.fc37.x86_64
  libglvnd-gles-1:1.5.0-1.fc37.i686
  libglvnd-opengl-1:1.5.0-1.fc37.i686
  nvidia-persistenced-3:525.89.02-1.fc37.x86_64
  nvidia-settings-3:525.89.02-1.fc37.x86_64
  openssl-1:3.0.8-1.fc37.x86_64
  patch-2.7.6-17.fc37.x86_64
  python3-babel-2.10.3-3.fc37.noarch
  python3-decorator-5.1.1-4.fc37.noarch
  python3-gssapi-1.7.3-3.fc37.x86_64
  python3-koji-1.32.0-1.fc37.noarch
  python3-progressbar2-3.53.2-6.fc37.noarch
  python3-pygit2-1.7.1-4.fc37.x86_64
  python3-pytz-2022.7.1-1.fc37.noarch
  python3-requests-gssapi-1.2.3-6.fc37.noarch
  python3-rpmautospec-0.3.5-1.fc37.noarch
  python3-utils-3.1.0-3.fc37.noarch
  rpm-build-4.18.0-1.fc37.x86_64
  rpmdevtools-9.6-2.fc37.noarch
  xorg-x11-drv-nvidia-3:525.89.02-1.fc37.x86_64
  xorg-x11-drv-nvidia-cuda-3:525.89.02-1.fc37.x86_64
  xorg-x11-drv-nvidia-cuda-libs-3:525.89.02-1.fc37.i686
  xorg-x11-drv-nvidia-cuda-libs-3:525.89.02-1.fc37.x86_64
  xorg-x11-drv-nvidia-kmodsrc-3:525.89.02-1.fc37.x86_64
  xorg-x11-drv-nvidia-libs-3:525.89.02-1.fc37.i686
  xorg-x11-drv-nvidia-libs-3:525.89.02-1.fc37.x86_64
  xorg-x11-drv-nvidia-power-3:525.89.02-1.fc37.x86_64

Complete!

5. Reboot and see
I’ll give you the final verdict in a minute or so.

galure · March 23, 2023, 7:53pm

After rebooting, everything seems to work very well.

. I tried Wayland and although some application like Steam were not very happy, the session did identify my screens. I made a script for my keyboards so one would be in azerty and one in qwerty and the script didn’t run, but that was to be expected, it’s a wayland thing.

. In x11 session all seems good.
[ I ran a heavy-graphics game that blocked for 2 minutes on a black screen using 80% of my CPU. That should be the shaders but 1) previously the game wasn’t loading the shaders on a black screen, and 2) just after the black screen the game was showing the shaders loading and using 20% of the CPU… that should be a new problem for the Proton team.]

So as I unterstand it, my nvidia drivers should update themselves now, every time I update to a new kernel. I mean, that was the case for the last 2 years, something happened in the last update that I can’t figure out.

Maybe it has to do with the way I update my system?
I run sudo dnf update about every time I see new updates available. Sometimes, even when dnf update returns nothing to do! there is more updates available in Discovery (the software center for KDE). I don’t use snaps, just flatpaks, but those updates in Discovery are most of the time for Gnome or Kde desktop, sometimes for other system-wide applications. And on rare occasions for nvidia drivers.
I never understood why sudo dnf update couldn’t update those packages, it is a mystery to me.
So maybe I updated the kernel without doing a very needed update to the nvidia drivers, causing chaos and forcing me to follow wrong installation guides of nvidia drivers through the cuda-fedora repos? I don’t know.

Anyway, thank you computersavvy for saving the day, again and again, you’re a beast.

gnwiii · March 23, 2023, 10:06pm

The internet almost never deletes outdated installation guides, so it is important to look for kernel and device versions. Guides that don’t include such details are not to be trusted.

There can be a significant lag while Nvidia figures out the changes they need to make to support a new kernel, followed by another delay before linux packages are generated. Since nouveau is included in the kernel packages, it has the advantage that the changes are generally made before a kernel update is released (even then, testers can miss a change that impacts old display hardware they don’t have, which is where users can help).

You can either remove the non-free packages and use nouveau or delay upgrading while you watch to see if people have problems with your hardware and the non-free drivers.

computersavvy · March 24, 2023, 12:46am

You noted issues with updates using discover and said it sometimes shows updates that dnf missed.

It seems to me that both gnome software and discover may update flatpaks while dnf does not.
If that is the issue then the extra updates are likely related to whatever flatpak software you may have installed.

With the workstation you would update flatpaks with flatpak upgrade

ilikelinux · March 24, 2023, 2:28am

I observed the same in gnome. When I open software there is always some updates like extensions or apps. Also applications where dnf not has the necessary information’s from gnome.

galure · March 27, 2023, 7:32am

Ok, I waited for an update to show the issue.

Here’s the command return for dnf update this morning

====================================================================================================
 Package                     Architecture      Version                     Repository          Size
====================================================================================================
Upgrading:
 cmake                       x86_64            3.26.1-1.fc37               updates            6.8 M
 cmake-data                  noarch            3.26.1-1.fc37               updates            2.2 M
 cmake-filesystem            x86_64            3.26.1-1.fc37               updates             17 k
 cmake-rpm-macros            noarch            3.26.1-1.fc37               updates             16 k
 curl                        x86_64            7.85.0-8.fc37               updates            316 k
 libadwaita                  x86_64            1.2.4-1.fc37                updates            431 k
 libcurl                     i686              7.85.0-8.fc37               updates            331 k
 libcurl                     x86_64            7.85.0-8.fc37               updates            304 k
 libqalculate                x86_64            4.6.0-1.fc37                updates            2.2 M

Transaction Summary
====================================================================================================

//I'm cutting the uninteresting parts//

Upgraded:
  cmake-3.26.1-1.fc37.x86_64                       cmake-data-3.26.1-1.fc37.noarch                 
  cmake-filesystem-3.26.1-1.fc37.x86_64            cmake-rpm-macros-3.26.1-1.fc37.noarch           
  curl-7.85.0-8.fc37.x86_64                        libadwaita-1.2.4-1.fc37.x86_64                  
  libcurl-7.85.0-8.fc37.i686                       libcurl-7.85.0-8.fc37.x86_64                    
  libqalculate-4.6.0-1.fc37.x86_64    

Complete!

Here’s a screenshot of discover just after I ran the dnf update (I clicked on the icon tray notifying updates after running the command):

updatefedora2

I can see the 8 “System upgrade” I just did.
After a refresh of the updates in Discover:

updatefedora3

So Gnome was not updated by dnf.

Now, the return command for flatpak upgrade

Looking for updates…
        ID                               Branch        Op       Remote        Download
 1.     org.gnome.Platform.Locale        43            u        flathub       < 340,5 MB (partial)
 2.     org.gnome.Platform               43            u        flathub       < 329,0 MB

Proceed with these changes to the system installation? [Y/n]:

So yes, dnf doesn’t update flatpaks, which is a pain.

So I created an alias in .bashrc

alias update="sudo dnf update && flatpak upgrade"

And now it works. thank you again computersavvy.