1 目前机子显卡驱动和cuda 版本目前看来有些旧了，需要升级，

cat /etc/os-release

PRETTY_NAME="Ubuntu 22.04.5 LTS"

NAME="Ubuntu"

VERSION_ID="22.04"

VERSION="22.04.5 LTS (Jammy Jellyfish)"

VERSION_CODENAME=jammy

ID=ubuntu

ID_LIKE=debian

HOME_URL="https://www.ubuntu.com/"

SUPPORT_URL="https://help.ubuntu.com/"

BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"

PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy"

UBUNTU_CODENAME=jammy

uname -a

Linux ubuntu-workstation 6.8.0-54-generic #56~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Sat Feb 8 11:41:24 UTC 2 x86_64 x86_64 x86_64 GNU/Linux

#nvidia-smi

hu Dec 11 18:02:27 2025

+---------------------------------------------------------------------------------------+

| NVIDIA-SMI 535.171.04 Driver Version: 535.171.04 CUDA Version: 12.2 |

|-----------------------------------------+----------------------+----------------------+

| 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 RTX 4090 Off | 00000000:04:00.0 On | Off |

| 0% 35C P8 14W / 450W | 1529MiB / 24564MiB | 3% Default |

| | | N/A |

+-----------------------------------------+----------------------+----------------------+

nvcc -v

nvcc: NVIDIA (R) Cuda compiler driver

Built on Wed_Sep_21_10:33:58_PDT_2022

Cuda compilation tools, release 11.8, V11.8.89

Build cuda_11.8.r11.8/compiler.31833905_0

可以看出显卡驱动535.171.04,nvcc 版本 11.8

准备升级显卡驱动，cuda 版本升级至12.6

1 下载并安装驱动

我的是消费级显卡，4090,

目前按照链接https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/

给的参考资料，cuda 12.6的驱动版本要大于等于560.28.03

在以下链接中https://www.nvidia.cn/geforce/drivers/

搜索驱动程序，并最终选择570.195.03这个版本驱动程序，一是这个版本驱动程序，可以兼容cuda 12.8,同时驱动程序版本不是太高，此处没有选择最新的590的驱动程序

下一步是安装驱动程序，安装驱动可能导致机子重启，重要的服务请做好处理

需要关闭X server,ctrl+alt+F1 切换至命令行,然后sudo init 3 关闭X server,然后正常执行驱动安装，这个提示就不会有了，升级前，使用显卡的驱动程序要提前关掉。

There appears to already be a driver installed on you system(version:535.171.04).As part of installing this driver (version:535.1271.04),the existing driver will be uninstalled ,Are you sure you want to continue ? yes (默认)

An alternate method of installing the NVIDIA driver was detected.(This is usually a package provided by your distributor.)A driver installed via that method may intergrate better with your systemthan a driver installed by nvidia-istaller.Please review the message provided by the maintainer of this alternate installation method and decide how to process: Continue Installation (默认) The NVIDIA driver by Ubuntu can be installed by Lanching the "Software &Update" application ,and by selecting the NVIDIA driver from "Additional Drivers" tab

Install NVIDIA's 32-bit compatiibility libraries ? yes(默认)

Would you like to register the kernel module sources with DKMS? This will allow DKMS to automatically build a new module,if you kernel changes later Yes(默认)

Would you like to run the nvidia-xconfig utility to automatically update you X configuration file so that the NVIDIA X driverw will be userd when you restart X7 Any pre-existing X configuration file will be backed up No(默认)

Installation of NVIDIA Accelerated Graphics Driver for linux-X86_64(version 570.195.03 is now complete).Please update your xorg.conf file as appropriate;see the file /use/share/doc/NVIFDIA_GLX-1.0/README.txt for deatils

安装成功， sudo init 5 返回图形界面，查看驱动

Thu Dec 11 18:48:17 2025

+-----------------------------------------------------------------------------------------+

| NVIDIA-SMI 570.195.03 Driver Version: 570.195.03 CUDA Version: 12.8 |

|-----------------------------------------+------------------------+----------------------+

| 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 RTX 4090 Off | 00000000:04:00.0 On | Off |

| 0% 32C P2 69W / 450W | 597MiB / 24564MiB | 0% Default |

| | | N/A |

+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+

| Processes: |

| GPU GI CI PID Type Process name GPU Memory |

| ID ID Usage |

|=========================================================================================|

2 更新cuda

先删除之前安装的cuda

sudo apt purge nvidia-cuda-toolkit cuda*

sudo apt autoremove

在链接中https://developer.nvidia.com/cuda-12-6-3-download-archive?target_os=Linux&target_arch=x86_64&Distribution=Ubuntu&target_version=22.04&target_type=runfile_local

下载安装文件，此处选择是runfile(local)按照下面的教程

复制代码

wget https://developer.download.nvidia.com/compute/cuda/12.6.3/local_installers/cuda_12.6.3_560.35.05_linux.runsudo 
sh cuda_12.6.3_560.35.05_linux.run

安装过程中：

驱动取消勾选，然后选择Install,回车，

Existing installation of CUDA Toolkit 12.6 found: │

│ Upgrade all │

│ Choose components to upgrade │

│ No, abort installation

选择第一个即可，安装后出现如下提示

===========

= Summary =

===========

Driver: Not Selected

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

***WARNING: Incomplete installation! This installation did not install the CUDA Driver. A driver of version at least 560.00 is required for CUDA 12.6 functionality to work.

To install the driver using this installer, run the following command, replacing <CudaInstaller> with the name of this run file:

sudo <CudaInstaller>.run --silent --driver

Logfile is /var/log/cuda-installer.log

代表已经安装成功，只是没有安装驱动

更新环境变量 cuda ->/usr/local/cuda-12.6,source ~/.bashrc

查看nvcc --version

nvcc: NVIDIA (R) Cuda compiler driver

Built on Tue_Oct_29_23:50:19_PDT_2024

Cuda compilation tools, release 12.6, V12.6.85

Build cuda_12.6.r12.6/compiler.35059454_0

然后进行测试，一般目录是

/usr/local/cuda/extras/demo_suite/deviceQuery，执行这个可执行文件，得到如下输出