CUDA Toolkit 版本选择

• CUDA Toolkit 历史版本

1. NVIDIA-SMI 525.60.11

wget https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda_12.0.0_525.60.13_linux.run
sudo sh cuda_12.0.0_525.60.13_linux.run

静默安装

bash cuda_12.0.0_525.60.13_linux.run --toolkit --silent

# tail .bashrc
export PATH=$PATH:/usr/local/cuda-12.0/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-12.0/lib64

# source .bashrc
# nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Mon_Oct_24_19:12:58_PDT_2022
Cuda compilation tools, release 12.0, V12.0.76
Build cuda_12.0.r12.0/compiler.31968024_0

2. CUDA Toolkit 12.6.0 安装

禁用 nouveau

lsmod | grep nouveau
echo "blacklist nouveau" >> /etc/modprobe.d/blacklist.conf
update-initramfs -u
reboot

依赖安装

sudo apt install gcc git jq pkg-config curl clang build-essential wget -y

gcc - GNU C编译器，用于编译C语言程序
git - 版本控制系统，用于代码管理
jq - 命令行JSON处理工具
pkg-config - 编译应用程序时查找库的编译参数的辅助工具
curl - 命令行文件传输工具
clang - LLVM C/C++/Objective-C编译器
build-essential - 包含编译软件所需的基本包（包括gcc、g++、make等）
wget - 命令行下载工具

下载安装 cuda

wget https://developer.download.nvidia.com/compute/cuda/12.6.0/local_installers/cuda_12.6.0_560.28.03_linux.run
sudo sh cuda_12.6.0_560.28.03_linux.run

显卡驱动安装成功

# bash cuda_12.6.0_560.28.03_linux.run
===========
= Summary =
===========

Driver:   Installed
Toolkit:  Installed in /usr/local/cuda-12.6/

Please make sure that
 -   PATH includes /usr/local/cuda-12.6/bin
 -   LD_LIBRARY_PATH includes /usr/local/cuda-12.6/lib64, or, add /usr/local/cuda-12.6/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the CUDA Toolkit, run cuda-uninstaller in /usr/local/cuda-12.6/bin
To uninstall the NVIDIA Driver, run nvidia-uninstall

设置环境变量

export PATH=$PATH:/usr/local/cuda/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64
export CUDA_PATH=$CUDA_PATH:/usr/local/cuda

# nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Fri_Jun_14_16:34:21_PDT_2024
Cuda compilation tools, release 12.6, V12.6.20
Build cuda_12.6.r12.6/compiler.34431801_0

3. 安装失败

# bash cuda_12.6.0_560.28.03_linux.run
 Installation failed. See log at /var/log/cuda-installer.log for details.

切换到多用户文本模式

systemctl get-default
# 如果显示 graphical.target 说明系统运行在图形模式
# 如果显示 multi-user.target 说明系统运行在文本模式

# 切换到多用户文本模式
sudo systemctl set-default multi-user.target

# 重启系统使更改生效
sudo reboot

参考

1. CUDA Toolkit Archive
2. CUDA