You are not logged in.
Hello,
This is my first post on these forums and while I have read the rules and guidelines feel free to correct me if I happen to apply them incorrectly.
I run a Windows 11/Arch Linux dual-boot on my machine. Two weeks ago, after rebooting from Windows to Arch I was met with an incorrect interface. I did not update my system the last boot I did on Arch, though I have updated my system since.
To go into more details out of my two monitors only one is now detected, and at 1024x768 instead of 2560x1440. This is also the only resolution now available. What's more `xrandr` now shows the display as `None-1` instead of `DP-1` as was previously the case (both my monitors are identical and plugged in via displayport). Full output below:
Screen 0: minimum 320 x 200, current 1024 x 768, maximum 4096 x 4096
None-1 connected 1024x768+0+0 (normal left inverted right x axis y axis) 0mm x 0mm 1024x768 60.00*+I have an Nvidia GTX1080 graphics card, onto which my monitors are plugged. I use the legacy `nvidia-580xx` Nvidia drivers as recommended on the wiki. Although I have tried the newer open drivers, which did not solve my issue and changed nothing I noticed regarding it.
I have tried adding a new mode and enabling for my `None-1` display through xrandr as described in this procedure. It failed.
Tried:
xrandr --newmode "2560x1440_60.00" 312.25 2560 2752 3024 3488 1440 1443 1448 1493 -hsync +vsync
xrandr --addmode None-1 "2560x1440_60.00"
xrandr --output None-1 --mode "2560x1440_60.00"Got:
xrandr: Configure crtc 0 failedWhich just confused me further as I don't know what this is (unless it's cathode tubes but I don't see the link).
I also retrieved the EDID binary for my monitors on my working Windows install with the same hardware and tried to force it as described on the wiki. I failed again, though I only tried setting the EDID after boot. I did not understand how to access the kernel command line yet after all.
I should also mention I'm using the X server and the Qtile WM. My `Xorg.0.log` can be found on this pastebin and the output from `journalctl -b`
If you need anything else feel free to ask ! Good day to you, whenever you may read this ![]()
Last edited by toude (2026-01-13 16:02:17)
Offline
I use the legacy `nvidia-580xx` Nvidia drivers as recommended
Nope.
You're running on the simpledrm software emulation because you're trying to run a pascal chip w/ nvidia-open
janv. 11 23:53:03 s0lst1ce-desktop kernel: NVRM: The NVIDIA GPU 0000:01:00.0 (PCI ID: 10de:1b80)
NVRM: installed in this system is not supported by open
NVRM: nvidia.ko because it does not include the required GPU
NVRM: System Processor (GSP).
NVRM: Please see the 'Open Linux Kernel Modules' and 'GSP
NVRM: Firmware' sections in the driver README, available on
NVRM: the Linux graphics driver download page at
NVRM: www.nvidia.com.https://archlinux.org/news/nvidia-590-d … l-modules/
pacman -Qs 'kernel|nvidia'
dkms statusI run a Windows 11/Arch Linux dual-boot on my machine.
3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.
Offline
Hello Seth, thanks for your quick reply.
Thanks for pointing out my mistake regarding the drivers, I seem to have forgotten to revert to the 58xx. Doing this and disabling fast boot did the trick, thanks !
Offline
I had marked this as solved as I believed it to be the case. Unfortunately after a few reboots (3 I believe), always going from and to Arch (not my Windows system) I stumbled upon the previously described issue once more, and it persists. The only change I recall having made to my system was to re-install Steam (`steam` package), removing it again did not help. I was careful not to install any `nvidia-*-open` package as optional dependencies. Here is the output of `paru -Qs nvidia`
❯ paru -Qs nvidia
local/egl-gbm 1.1.2.1-1
The GBM EGL external platform library
local/egl-wayland 4:1.1.21-1
EGLStream-based Wayland external platform
local/egl-wayland2 1.0.0.rc.r53.gf4a2d32-1
EGLStream-based Wayland external platform (2)
local/egl-x11 1.0.4-1
NVIDIA XLib and XCB EGL Platform Library
local/lib32-nvidia-580xx-utils 580.119.02-1
NVIDIA drivers utilities (32-bit) (580xx)
local/libvdpau 1.5-3
Nvidia VDPAU library
local/linux-firmware-nvidia 20260110-1
Firmware files for Linux - Firmware for NVIDIA GPUs
and SoCs
local/nvidia-580xx-dkms 580.119.02-2
NVIDIA kernel modules - module sources (580xx)
local/nvidia-580xx-utils 580.119.02-2
NVIDIA drivers utilities (580xx)One more here are my journals for journalctl and Xorg.
Hopefully I didn't miss anything you had mentioned previously. If so please point it out to me so I may learn from that mistake.
Thanks in advance for your help.
Last edited by toude (2026-01-13 15:28:22)
Offline
janv. 13 16:00:11 s0lst1ce-desktop systemd-modules-load[486]: Failed to find module 'nvidia_uvm'Did the modules actually build when you updated the kernel and headers? Or did they fail for some reason?
Offline
Xorg log is dated (the links are btw. cross-labeled)
pacman -Qs 'kernel|nvidia' dkms status
Offline
Thanks for the quick reply. I don't know why the Xorg log is outdated, I posted `/var/log/Xorg.0.log`. Should it be another one ?
Here is the output of
pacman -Qs 'kernel|nvidia'
dkms statuslocal/dkms 3.3.0-1
Dynamic Kernel Modules System
local/egl-gbm 1.1.2.1-1
The GBM EGL external platform library
local/egl-wayland 4:1.1.21-1
EGLStream-based Wayland external
platform
local/egl-wayland2 1.0.0.rc.r53.gf4a2d32-1
EGLStream-based Wayland external
platform (2)
local/egl-x11 1.0.4-1
NVIDIA XLib and XCB EGL Platform Library
local/embree 4.4.0-1
Collection of high-performance ray
tracing kernels
local/fuse2 2.9.9-5
Interface for userspace programs to
export a filesystem to the Linux kernel
local/fuse3 3.17.4-1
Interface for userspace programs to
export a filesystem to the Linux kernel
local/iptables 1:1.8.11-2
Linux kernel packet control tool (using
legacy interface)
local/kmod 34.2-1
Linux kernel module management tools and
library
local/lib32-libdrm 2.4.131-1
Userspace interface to kernel DRM
services (32-bit)
local/lib32-nvidia-580xx-utils 580.119.02-1
NVIDIA drivers utilities (32-bit)
(580xx)
local/libdrm 2.4.131-1
Userspace interface to kernel DRM
services
local/libnetfilter_conntrack 1.0.9-2
Library providing an API to the
in-kernel connection tracking state
table
local/libnfnetlink 1.0.2-2
Low-level library for netfilter related
kernel/userspace communication
local/libsysprof-capture 49.0-1
Kernel based performance profiler -
capture library
local/libvdpau 1.5-3
Nvidia VDPAU library
local/linux 6.18.5.arch1-1
The Linux kernel and modules
local/linux-api-headers 6.18-1
Kernel headers sanitized for use in
userspace
local/linux-firmware-nvidia 20260110-1
Firmware files for Linux - Firmware for
NVIDIA GPUs and SoCs
local/mtdev 1.1.7-1
A stand-alone library which transforms
all variants of kernel MT events to the
slotted type B protocol
local/nvidia-580xx-dkms 580.119.02-2
NVIDIA kernel modules - module sources
(580xx)
local/nvidia-580xx-utils 580.119.02-2
NVIDIA drivers utilities (580xx)
local/python-comm 0.2.3-2
Python Comm implementation for the
Jupyter kernel protocol
local/python-ipykernel 7.1.0-5
The ipython kernel for JupyterSorry for not providing it earlier, I hadn't understood that was the request.
I've also corrected the swapped links in my previous post.
As for the lnk you posted @Seth, I'm not sure what you're recommending I do ?
Last edited by toude (2026-01-13 15:27:59)
Offline
janv. 13 16:00:11 s0lst1ce-desktop systemd-modules-load[486]: Failed to find module 'nvidia_uvm'Did the modules actually build when you updated the kernel and headers? Or did they fail for some reason?
I got no errors when installing the legacy drivers through pacman, which I think triggers the rebuild you're mentioning ?
Offline
Read it.
Install linux-headers
Rebuild the dkms modules if the header installation doesn't trigger it automatically.
Pay attention to any errors.
Offline
Alright so installing & rebuilding the dkms module for nvidia seems to have done the trick. There were no errors. Thanks for the help ! Sorry if I missed some obvious things, I'm navigating mostly in the dark here,
I do have a few questions though, for which I have not found, or understood, the answer to on the page you linked. Why do I need a dkms module ? From my understanding this is for things outside the kernel. So I suppose the legacy drivers now are ? While the `-open` ones are part of the "main" kernel and thus do not require dkms ? But if so why are the linux-headers not a dependency of the AUR package for the drivers ? Are there some cases when they're necessary and others not ? Why did I have a working system for one or two boots and then back to square one ?
Offline
Why do I need a dkms module ?
Cause the 580xx driver isn't maintained in the repo, you need to build the module yourself, dkms is a very convenient way invented for that exact purpose and installing the driver via the nvidia installer would also bypass any package control.
While the `-open` ones are part of the "main" kernel
No, not yet. It's simply that a package maintainer builds them for the users (using dkms, btw)
But if so why are the linux-headers not a dependency of the AUR package for the drivers ?
Because you need the headers for any kernel you wish to build dkms modules for, not a specific package.
Why did I have a working system for one or two boots and then back to square one ?
Impossible to tell w/o the jorunal but most likely you were running on the simpledrm device.
Did your xorg log btw. update?
You're probably running a wayland session anyway.
Please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.
Thanks.
Offline
No, not yet. It's simply that a package maintainer builds them for the users (using dkms, btw)
So a pre-built module is shipped through the package manager ? Like a pre-built binary for a program would ?
Because you need the headers for any kernel you wish to build dkms modules for, not a specific package.
I'm not sure I understand this. There's a header for each kernel: ok, that makes sense to me. But anyone installing the legacy nvidia drivers would need dkms & linux-headers to make use of it, right ? In which case why not make it a dep of the driver ?
Impossible to tell w/o the jorunal but most likely you were running on the simpledrm device.
Too bad I didn't save it when it happened. Well I didn't expect the issue to come back. Still a mystery to me why it only worked once.
You're probably running a wayland session anyway.
I think not but I do see lots of wayland-related packages installed so I'm not so sure. I don't remember configuring it anyways and since I'm starting my WM with `qtile start` and not `qtile start -b wayland` I would wager I'm on X11.
Then again the Xorg logs are still outdated...
The post should be marked as solved by now. Thanks a lot for your help and answers !
Offline
So a pre-built module is shipped through the package manager
Yes.
anyone installing the legacy nvidia drivers would need dkms & linux-headers
If you're using the LTS kernel you need linux-lts-headers. If you're using the zen kernel you need linux-zen-header. If you're using the hardened kernel you…
I didn't save it when it happened.
sudo journalctl -b -3 # 3 boots agoI'm starting my WM with `qtile start` and not `qtile start -b wayland` I would wager I'm on X11.
Then again the Xorg logs are still outdated...
https://wiki.archlinux.org/title/Xorg#General - you might look at the wrong log then.
Offline
Always always always read what pacman tells you! It told you about the optional deps on the header packages when you installed the dkms package. It also threw errors when it tried to build the modules without the headers. If you ignore pacman, things go wrong.
Offline