debian_11_bullseye_et_cuda
Différences
Ci-dessous, les différences entre deux révisions de la page.
| Les deux révisions précédentesRévision précédenteProchaine révision | Révision précédente | ||
| debian_11_bullseye_et_cuda [2021/12/15 18:05] – [Debian 11 Bullseye Mate et CUDA] serge | debian_11_bullseye_et_cuda [2022/06/18 11:58] (Version actuelle) – [Debian 11 Bullseye Mate et CUDA] serge | ||
|---|---|---|---|
| Ligne 1: | Ligne 1: | ||
| ====== Debian 11 Bullseye Mate et CUDA ====== | ====== Debian 11 Bullseye Mate et CUDA ====== | ||
| - | {{ : | + | <WRAP center round box 60% centeralign> |
| + | **How to install CUDA on Debian 11 Bullseye** | ||
| + | </ | ||
| + | |||
| + | <WRAP center round alert 60%> | ||
| + | !!!!!!!!! Ce tuto ne marche pas !!!!!!!!!!! | ||
| + | </ | ||
| ===== Installation Préliminaires ===== | ===== Installation Préliminaires ===== | ||
| * [[debian_installation_configuration|]] | * [[debian_installation_configuration|]] | ||
| Ligne 22: | Ligne 29: | ||
| </ | </ | ||
| <code bash> | <code bash> | ||
| + | sudo dpkg --add-architecture i386 | ||
| sudo apt update | sudo apt update | ||
| sudo apt upgrade | sudo apt upgrade | ||
| sudo apt install mate-menu mate-desktop-environment-extras locales ncdu git gnome-system-tools caja-open-terminal software-properties-common mate-tweak synaptic terminator geany geany-plugins git firmware-iwlwifi firmware-linux-nonfree | sudo apt install mate-menu mate-desktop-environment-extras locales ncdu git gnome-system-tools caja-open-terminal software-properties-common mate-tweak synaptic terminator geany geany-plugins git firmware-iwlwifi firmware-linux-nonfree | ||
| - | sudo dpkg --add-architecture i386 | ||
| - | sudo apt update | ||
| - | sudo apt upgrade | ||
| </ | </ | ||
| =====Cuda et Driver Nvidia===== | =====Cuda et Driver Nvidia===== | ||
| - | L' | + | L' |
| <code bash> | <code bash> | ||
| sudo apt install nvidia-cuda-toolkit nvidia-cuda-dev | sudo apt install nvidia-cuda-toolkit nvidia-cuda-dev | ||
| Ligne 46: | Ligne 51: | ||
| </ | </ | ||
| ====CuDNN==== | ====CuDNN==== | ||
| - | * https:// | ||
| - | |||
| Télécharger CuDNN à [[https:// | Télécharger CuDNN à [[https:// | ||
| Pour pouvoir télécharger, | Pour pouvoir télécharger, | ||
| + | Choisir un deb pour Ubuntu 20.04 ! | ||
| Exemple: | Exemple: | ||
| <code bash> | <code bash> | ||
| - | sudo dpkg -i libcudnn8_8.3.0.98-1+cuda11.5_amd64.deb | + | sudo dpkg -i libcudnn8_8.1.1.33-1+cuda11.2_amd64.deb |
| - | sudo dpkg -i libcudnn8-dev_8.3.0.98-1+cuda11.5_amd64.deb | + | sudo dpkg -i libcudnn8-dev_8.1.1.33-1+cuda11.2_amd64.deb |
| </ | </ | ||
| - | =====Toute la suite ne sert à rien===== | + | =====Compilation de OpenCV |
| - | =====Quel driver possible pour votre carte graphique===== | + | |
| - | <code bash> | + | |
| - | sudo apt -y install nvidia-detect | + | |
| - | nvidia-detect | + | |
| - | </ | + | |
| - | ===GeForce GTX 850M=== | + | |
| - | <code bash> | + | |
| - | Detected NVIDIA GPUs: | + | |
| - | 01:00.0 3D controller [0302]: NVIDIA Corporation GM107M [GeForce GTX 850M] [10de:1391] (rev a2) | + | |
| - | + | ||
| - | Checking card: NVIDIA Corporation GM107M [GeForce GTX 850M] (rev a2) | + | |
| - | Your card is supported by all driver versions. | + | |
| - | Your card is also supported by the Tesla 460 drivers series. | + | |
| - | Your card is also supported by the Tesla 450 drivers series. | + | |
| - | Your card is also supported by the Tesla 418 drivers series. | + | |
| - | It is recommended to install the | + | |
| - | nvidia-driver | + | |
| - | package. | + | |
| - | </ | + | |
| - | + | ||
| - | ===GeForce GTX 1060=== | + | |
| - | <code bash> | + | |
| - | Detected NVIDIA GPUs: | + | |
| - | 01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GP104 [GeForce GTX 1060 6GB] [10de:1b83] (rev a1) | + | |
| - | + | ||
| - | Checking card: NVIDIA Corporation GP104 [GeForce GTX 1060 6GB] (rev a1) | + | |
| - | Your card is only supported up to the 390 legacy drivers series. | + | |
| - | Your card is also supported by the Tesla 440 drivers series. | + | |
| - | Your card is also supported by the Tesla 418 drivers series. | + | |
| - | It is recommended to install the | + | |
| - | nvidia-legacy-390xx-driver | + | |
| - | package. | + | |
| - | + | ||
| - | </ | + | |
| - | =====Installation numéro 11===== | + | |
| - | * https:// | + | |
| - | * **[[https:// | + | |
| - | + | ||
| - | ====Installation du driver NVIDIA | + | |
| - | <code bash> | + | |
| - | sudo apt install nvidia-driver | + | |
| - | sudo systemctl reboot | + | |
| - | </ | + | |
| - | + | ||
| - | ====Installation du driver NVIDIA avec les paquets officiels nvidia.com==== | + | |
| - | Récupération du driver dans les archives nvidia.com\\ | + | |
| - | * https:// | + | |
| - | Je choisis de télécharger le driver Latest Legacy GPU version (460.xx series) | + | |
| - | + | ||
| - | <code bash> | + | |
| - | sudo apt -y install linux-headers-$(uname -r) build-essential libglvnd-dev pkg-config | + | |
| - | + | ||
| - | sudo nano / | + | |
| - | # Ajouter | + | |
| - | blacklist nouveau | + | |
| - | options nouveau modeset=0 | + | |
| - | # Save Quit | + | |
| - | + | ||
| - | sudo update-initramfs -u | + | |
| - | sudo systemctl set-default multi-user.target | + | |
| - | sudo systemctl reboot | + | |
| - | # Se logguer dans le terminal | + | |
| - | sh NVIDIA-Linux-x86_64-460.91.03.run | + | |
| - | systemctl set-default graphical.target | + | |
| - | systemctl reboot | + | |
| - | </ | + | |
| - | + | ||
| - | ====Nvidia Settings==== | + | |
| - | Dans les 2 cas nvidia-settings n'a pas les infos habituelles. | + | |
| - | + | ||
| - | + | ||
| - | + | ||
| - | + | ||
| - | + | ||
| - | =====Ressources===== | + | |
| - | * **[[https:// | + | |
| - | + | ||
| - | =====Recherches===== | + | |
| - | L' | + | |
| - | A confirmer sur une prochaine install propre, et virer ce paragraphe si c'est confimé. | + | |
| - | + | ||
| - | <code bash> | + | |
| - | # Téléchargement dans le home de l' | + | |
| - | wget https:// | + | |
| - | + | ||
| - | sudo apt install linux-headers-$(uname -r) build-essential libglvnd-dev pkg-config -y | + | |
| - | sudo nano / | + | |
| - | + | ||
| - | # Ajouter ça: | + | |
| - | blacklist nouveau | + | |
| - | options nouveau modeset=0 | + | |
| - | # Enregister quitter | + | |
| - | + | ||
| - | sudo update-initramfs -u | + | |
| - | # Pour être en terminal au redémarrage: | + | |
| - | sudo systemctl set-default multi-user.target | + | |
| - | sudo reboot now | + | |
| - | + | ||
| - | # Dans la console, se logger | + | |
| - | sudo bash NVIDIA-Linux-x86_64-495.29.05.run | + | |
| - | + | ||
| - | # Quand c'est fini | + | |
| - | startx | + | |
| - | + | ||
| - | # Pour ne plus être en terminal au redémarrage | + | |
| - | sudo systemctl set-default graphical.target | + | |
| - | sudo reboot now | + | |
| - | </ | + | |
| - | + | ||
| - | + | ||
| - | ===Sélection du driver=== | + | |
| - | sudo apt install primus | + | |
| - | prime-select | + | |
| - | + | ||
| - | ===CUDA=== | + | |
| - | + | ||
| - | <code bash> | + | |
| - | wget https:// | + | |
| - | sudo dpkg -i cuda-repo-debian11-11-5-local_11.5.1-495.29.05-1_amd64.deb | + | |
| - | sudo apt-key add / | + | |
| - | sudo apt update | + | |
| - | sudo apt install cuda | + | |
| - | </ | + | |
| - | + | ||
| - | + | ||
| - | ===Tensorflow with GPU=== | + | |
| - | + | ||
| - | source mon_env/ | + | |
| - | python3 -m pip install tensorflow-gpu | + | |
| - | + | ||
| - | ===CuDNN=== | + | |
| - | Télécharger CuDNN à [[https:// | + | |
| - | Pour pouvoir télécharger, | + | |
| - | + | ||
| - | Exemple: | + | |
| - | <code bash> | + | |
| - | sudo dpkg -i libcudnn8_8.3.0.98-1+cuda11.5_amd64.deb | + | |
| - | sudo dpkg -i libcudnn8-dev_8.3.0.98-1+cuda11.5_amd64.deb | + | |
| - | </ | + | |
| - | ===OpenCV=== | + | |
| - | + | ||
| - | + | ||
| - | wget -O opencv.zip https:// | + | |
| - | wget -O opencv_contrib.zip https:// | + | |
| {{tag> cuda ia linux opencv sb }} | {{tag> cuda ia linux opencv sb }} | ||
debian_11_bullseye_et_cuda.1639591531.txt.gz · Dernière modification : de serge
