1 系统

# lsb_release -a
No LSB modules are available.
Distributor ID:	Ubuntu
Description:	Ubuntu 18.04.6 LTS
Release:	18.04
Codename:	bionic

# uname -a
Linux k016 4.15.0-202-generic #213-Ubuntu SMP Thu Jan 5 19:19:12 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux

2 查看显卡

# lspci | grep -i nvidia
41:00.0 VGA compatible controller: NVIDIA Corporation Device 2204 (rev a1)
41:00.1 Audio device: NVIDIA Corporation Device 1aef (rev a1)

2.1 更新 pci.ids 文件

识别pci设备厂商名称、设备型号

# update-pciids
Downloaded daily snapshot dated 2023-01-26 03:15:02

# lspci | grep -i nvidia
41:00.0 VGA compatible controller: NVIDIA Corporation GA102 [GeForce RTX 3090] (rev a1)
41:00.1 Audio device: NVIDIA Corporation GA102 High Definition Audio Controller (rev a1)

3 安装显卡驱动 510.54

wget https://us.download.nvidia.com/XFree86/Linux-x86_64/510.54/NVIDIA-Linux-x86_64-510.54.run
sudo sh NVIDIA-Linux-x86_64-510.54.run

# nvidia-smi
Mon Jan 30 12:57:50 2023
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 510.54       Driver Version: 510.54       CUDA Version: 11.6     |
|-------------------------------+----------------------+----------------------+
| 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 ...  Off  | 00000000:42:00.0 Off |                  N/A |
| 37%   33C    P0    N/A / 350W |      0MiB / 24576MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

3.1 安装 nvtop

# nvtop

Command 'nvtop' not found, but can be installed with:

snap install nvtop

# snap install nvtop

在这里插入图片描述

4 禁用 nouveau

# lsmod | grep nouveau
nouveau              1708032  0
mxm_wmi                16384  1 nouveau
wmi                    24576  2 mxm_wmi,nouveau
video                  49152  1 nouveau
ttm                   106496  2 ast,nouveau
drm_kms_helper        172032  3 ast,nvidia_drm,nouveau
i2c_algo_bit           16384  3 igb,ast,nouveau
drm                   401408  7 drm_kms_helper,ast,nvidia,nvidia_drm,ttm,nouveau

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

lsmod | grep nouveau

5 安装 cuda 11.6.1

wget https://developer.download.nvidia.com/compute/cuda/11.6.1/local_installers/cuda_11.6.1_510.47.03_linux.run
sudo sh cuda_11.6.1_510.47.03_linux.run

显卡驱动已经安装，只需要选择cuda-toolkit 安装

5.1 设置环境变量

vim .bashrc

# CUDA
export PATH=$PATH:/usr/local/cuda-11.6/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-11.6/lib64

# source .bashrc
# nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Thu_Feb_10_18:23:41_PST_2022
Cuda compilation tools, release 11.6, V11.6.112
Build cuda_11.6.r11.6/compiler.30978841_0