为RTX5090安装推荐驱动后报错,Claude辅助快速定位并解决问题

2025年07月12日03:09:57

通过常规的方式安装驱动后执行nvidia-smi报错找不到设备

升级操作系统,并卸载系统自带开源的nouveau驱动

apt update && sudo apt upgrade -y
apt purge *nvidia*

0 upgraded, 0 newly installed, 0 to remove and 1 not upgraded.

自动安装推荐的驱动,最终安装nvidia-driver-570

sudo ubuntu-drivers install

可选驱动列表

root@rtx-5090-beijing-office:~# sudo ubuntu-drivers devices
udevadm hwdb is deprecated. Use systemd-hwdb instead.
udevadm hwdb is deprecated. Use systemd-hwdb instead.
udevadm hwdb is deprecated. Use systemd-hwdb instead.
udevadm hwdb is deprecated. Use systemd-hwdb instead.
ERROR:root:aplay command not found
== /sys/devices/pci0000:00/0000:00:01.0/0000:01:00.0 ==
modalias : pci:v000010DEd00002B85sv000010DEsd00002057bc03sc00i00
vendor   : NVIDIA Corporation
driver   : nvidia-driver-570-server-open - distro non-free
driver   : nvidia-driver-570 - distro non-free recommended
driver   : nvidia-driver-570-server - distro non-free
driver   : nvidia-driver-570-open - distro non-free
driver   : xserver-xorg-video-nouveau - distro free builtin

root@rtx-5090-beijing-office:~#

按照推荐,自动选择安装了如下驱动版本(非开源版本):

driver   : nvidia-driver-570 - distro non-free recommended

通过Claude快速定位到问题

问题原因:安装的驱动版本不匹配

安装的推荐驱动版本,不是开源的版本,需要替换为开源版本。

为什么需要开源驱动?
RTX 5090是较新的显卡,NVIDIA要求某些新架构的GPU必须使用开源内核模块。这是NVIDIA的策略变化,较新的GPU(特别是Ada Lovelace架构及以后)需要使用开源驱动来获得完整支持。

-w600

解决方案:卸载并安装匹配的驱动版本

需要切换到开源版本的NVIDIA驱动:

  1. 卸载当前专有驱动
sudo apt remove nvidia-driver-570
sudo apt autoremove
  1. 安装开源版本驱动
sudo apt install nvidia-driver-570-open
  1. 重启系统
sudo reboot

操作后验证:执行nvidia-smi测试验证OK

root@rtx-5090-beijing-office:~# nvidia-smi
Fri Jul 11 18:39:33 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.133.07             Driver Version: 570.133.07     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 5090        Off |   00000000:01:00.0  On |                  N/A |
| 30%   36C    P8             17W /  575W |      10MiB /  32607MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+
root@rtx-5090-beijing-office:~#

后续:安装CUDA

apt install nvidia-cuda-toolkit

root@rtx-5090-beijing-office:~# nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Fri_Jan__6_16:45:21_PST_2023
Cuda compilation tools, release 12.0, V12.0.140
Build cuda_12.0.r12.0/compiler.32267302_0
root@rtx-5090-beijing-office:~#
posted @ 2025-08-06 01:07  That's_it  阅读(152)  评论(0)    收藏  举报