GithubHelp home page GithubHelp logo

Comments (7)

gridhead avatar gridhead commented on May 30, 2024

Fresh installation of fedora 32.

Fedora Workstation or any one of the spins? (Or as told by you later, specifically KDE Plasma spin?) The question might sound irrelevant but it still can give leads to me as to where I can start reproducing this bug.

SELinux off (same problem than issue : "Generalize and test the tool in KDE Plasma spin #5")

The tool worked on devices I tested (both laptop and desktop PCs having Workstation 32) without actually needing to tweak the SELinux settings. It might be necessary for the KDE Plasma spin as mentioned in #5.

Plus, it does not matter if you fetched the updates first or not as the tool would do that for you, if you haven't. Also, it would check for active internet connection too and there is no need to have the drivers installed first and then other applications and stuff (no chronology stuff needed here - the tool simply fetches proprietary drivers from the RPM Fusion repositories and installs them), so you can do it anytime.

Erreur : Erreur de la transaction de test :
le fichier /usr/lib/udev/rules.d/60-nvidia.rules de l'installation de nvidia-kmod-common-3:440.95.01-1.fc29.x86_64 entre en conflit avec le fichier du paquet xorg-x11-drv-nvidia-3:440.82-1.fc32.x86_64

This thing really dazzles me. If you have a fresh installation of Fedora 32, there is literally no reason why it would fall back into even checking with the Fedora 29 repos. The tool was explicitly designed for Fedora 32 Workstation and is reported to be working in the spins as well. You might want to give a fresh installation of the KDE Plasma spin again (if that is what you'd use) and use steps listed in #5 during the installation.

If you make any other observation as to why this might have happened, please let know here. It will be some time before we get people to actually contribute by testing it on spins. There are open issues for that. 🙂

from nvidia-auto-installer-for-fedora-linux.

sylvoie avatar sylvoie commented on May 30, 2024

Hello,

Fresh installation of fedora 32.

Fedora workstation 32 base on Gnome that I downloaded last friday. Yes I will probably try the "KDE Plasma spin" later.

SELinux off (same problem than issue : "Generalize and test the tool in KDE Plasma spin #5")

I understand, and I tried several times with always the same result when SELinux is enabled. SELinux was "enforcing" and I need to disable to successfully install.

Erreur : Erreur de la transaction de test :
le fichier /usr/lib/udev/rules.d/60-nvidia.rules de l'installation de nvidia-kmod-common-3:440.95.01-1.fc29.x86_64 entre en conflit avec le fichier du paquet xorg-x11-drv-nvidia-3:440.82-1.fc32.x86_64>

Before using "KDE Plasma spin", I will reinstall "Fedora workstation 32" without installing KDE and XFCE to see.

Thank's

Sylvain

from nvidia-auto-installer-for-fedora-linux.

sylvoie avatar sylvoie commented on May 30, 2024

Hello,

I reinstalled from scratch and followed these steps :

  • Install Fedora 32 workstation
  • sudo yum update
  • SELinux = permissive
  • reboot
  • sudo ./NVAutoInstFedora32 --rpmadd
  • sudo ./NVAutoInstFedora32 --driver
  • sudo ./NVAutoInstFedora32 --x86lib
  • sudo ./NVAutoInstFedora32 --nvrepo
  • sudo ./NVAutoInstFedora32 --ffmpeg
  • sudo ./NVAutoInstFedora32 --vidacc
  • reboot
  • sudo ./NVAutoInstFedora32 --plcuda
  • sudo ./NVAutoInstFedora32 --vulkan
  • reboot
  • sudo yum update

Last yum update result :

==================================================================
Paquet Architecture Version Dépôt Taille

Mise à jour:
akmod-nvidia x86_64 3:440.95.01-1.fc29 cuda 13 M
Installation des dépendances:
nvidia-kmod-common x86_64 3:440.95.01-1.fc29 cuda 10 k
Ignorer les paquets en conflit :
(ajouter « --best --allowerasing » à la ligne de commande pour forcer leur mise à niveau):
nvidia-driver-cuda x86_64 3:440.95.01-1.fc29 cuda 313 k
Ignorer les paquets ayant des dépendances cassées :
nvidia-driver x86_64 3:440.95.01-1.fc29 cuda 2.3 M
nvidia-persistenced x86_64 3:440.95.01-1.fc29 cuda 91 k

Résumé de la transaction

Installer 1 Paquet
Mettre à niveau 1 Paquet
Ignorer 3 Paquets

Did I do something wrong?

Why these fc29 come from cuda repository?

Is-it something with the french installed language?

Thank's

Sylvain

from nvidia-auto-installer-for-fedora-linux.

gridhead avatar gridhead commented on May 30, 2024

Why these fc29 come from cuda repository?

That is okay. CUDA is not actively maintained but the thing made in F29 works just as great in F32.

Is it something with the french installed language?

Definitely not. French is one of the leading languages in Weblate with much of the part already translated and accommodated to.

CUDA might just be the core reason behind this. As the F29 CUDA repos require drivers exclusively for F29, it is turning up even for akmods-nvidia package. There is nothing wrong with the package though but I believe F29 is not maintained anymore so using their repositories is not advised.

from nvidia-auto-installer-for-fedora-linux.

gridhead avatar gridhead commented on May 30, 2024

The CUDA problem is definitely an issue with the RPM package available in the official RPM Fusion repositories and not with the installer (as it simply automates the steps listed here and here in a convenient way.) The CUDA package is not up-to-date and pulls outdated packages as its dependencies.

from nvidia-auto-installer-for-fedora-linux.

sylvoie avatar sylvoie commented on May 30, 2024

Hello,

I agree with you, that problem is not with the installer. Thank you for this installer, it saved my personal project to run first in Linux.

I installed and I confirm that this installer works well with Fedora 32 spins Xfce without CUDA (--nvrepo --plcuda). I can install CUDA, and that is working fine, but after that updates are stuck because CUDA package is not up-to-date.

At this time, I have driver 440.82, dnf tries to install 440.95 with CUDA Repo. On NVidia site, the driver are at 440.100. snif!

With my research, I found this that I share with you : https://www.youtube.com/watch?v=IVpOyKCNZYw.

Thank's

Sylvain

from nvidia-auto-installer-for-fedora-linux.

gridhead avatar gridhead commented on May 30, 2024

At this time, I have driver 440.82, dnf tries to install 440.95 with CUDA Repo. On NVidia site, the driver are at 440.100. snif!

The differences between the versions made available on the NVIDIA site and on the RPM Fusion repositories is due to the fact that recompiling the versions acquired from the NVIDIA site into an RPM package takes time, which then creates an offset. This should not be a concern though as you would always have a slightly older but rock stable drivers as a result of Fedora-specific configurations and thorough testing made before release by the folks at RPM Fusion.

With my research, I found this that I share with you : https://www.youtube.com/watch?v=IVpOyKCNZYw.

Thanks for the research. 🤣

In a matter of the next few days, a new version of the tool would roll out packaged in a COPR which would have a newer CUDA repository (the one which would not hold you back into using dependencies from as old as F29). I would be modifying the instructions and releases shortly so stay tuned.

from nvidia-auto-installer-for-fedora-linux.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.