ChatGLM2-6B GPU 驱动
由于 GPU 驱动存在每次重新启动，都需要重新安装的问题，有空予以解决。
Quick Start
# 确认显卡状态
$ nvidia-smi
# 查看驱动版本
$ ls -l /usr/src/
total 16
drwxr-xr-x 2 root root 4096 Jul 4 15:03 annobin
drwxr-xr-x. 2 root root 4096 Feb 9 2022 debug
drwxr-xr-x. 4 root root 4096 Jul 4 15:01 kernels
drwxr-xr-x 8 root root 4096 Jul 4 10:52 nvidia-525.105.17
# 使用 dkms 管理驱动
$ yum install dkms
# 安装指定版本驱动（ /usr/src/ 下的驱动版本）
$ sudo dkms install -m nvidia -v 525.105.17
# 确认显卡状态
$ nvidia-smi
History
[root@iZ8vbgjtg291j3aek7yk6bZ ~]# nvidia-smi
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
[root@iZ8vbgjtg291j3aek7yk6bZ ~]# uname -r
5.10.134-14.1.al8.x86_64
[root@iZ8vbgjtg291j3aek7yk6bZ ~]# ls -l /usr/src/
total 16
drwxr-xr-x 2 root root 4096 Jul 4 15:03 annobin
drwxr-xr-x. 2 root root 4096 Feb 9 2022 debug
drwxr-xr-x. 4 root root 4096 Jul 4 15:01 kernels
drwxr-xr-x 8 root root 4096 Jul 4 10:52 nvidia-525.105.17
[root@iZ8vbgjtg291j3aek7yk6bZ ~]# yum install dkms
Last metadata expiration check: 0:02:28 ago on Tue 11 Jul 2023 09:06:09 PM CST.
Dependencies resolved.
==========================================================================================================================================================================================================================================================
Package Architecture Version Repository Size
==========================================================================================================================================================================================================================================================
Installing:
dkms noarch 3.0.11-1.el8 epel 90 k
Transaction Summary
==========================================================================================================================================================================================================================================================
Install 1 Package
Total download size: 90 k
Installed size: 192 k
Is this ok [y/N]: y
Downloading Packages:
dkms-3.0.11-1.el8.noarch.rpm 386 kB/s | 90 kB 00:00
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Total 384 kB/s | 90 kB 00:00
Extra Packages for Enterprise Linux 8 - x86_64 1.6 MB/s | 1.6 kB 00:00
Importing GPG key 0x2F86D6A1:
Userid : "Fedora EPEL (8) <epel@fedoraproject.org>"
Fingerprint: 94E2 79EB 8D8F 25B2 1810 ADF1 21EA 45AB 2F86 D6A1
From : /etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-8
Is this ok [y/N]: y
Key imported successfully
Running transaction check
Transaction check succeeded.
Running transaction test
Transaction test succeeded.
Running transaction
Preparing : 1/1
Installing : dkms-3.0.11-1.el8.noarch 1/1
Running scriptlet: dkms-3.0.11-1.el8.noarch 1/1
Verifying : dkms-3.0.11-1.el8.noarch 1/1
Installed:
dkms-3.0.11-1.el8.noarch
Complete!
[root@iZ8vbgjtg291j3aek7yk6bZ ~]# sudo dkms install -m nvidia -v 525.105.17
Sign command: /lib/modules/5.10.134-14.1.al8.x86_64/build/scripts/sign-file
Signing key: /var/lib/dkms/mok.key
Public certificate (MOK): /var/lib/dkms/mok.pub
Certificate or key are missing, generating self signed certificate for MOK...
Creating symlink /var/lib/dkms/nvidia/525.105.17/source -> /usr/src/nvidia-525.105.17
Building module:
Cleaning build area...
'make' -j8 NV_EXCLUDE_BUILD_MODULES='' KERNEL_UNAME=5.10.134-14.1.al8.x86_64 modules................
Signing module /var/lib/dkms/nvidia/525.105.17/build/nvidia.ko
Signing module /var/lib/dkms/nvidia/525.105.17/build/nvidia-uvm.ko
Signing module /var/lib/dkms/nvidia/525.105.17/build/nvidia-modeset.ko
Signing module /var/lib/dkms/nvidia/525.105.17/build/nvidia-drm.ko
Signing module /var/lib/dkms/nvidia/525.105.17/build/nvidia-peermem.ko
Cleaning build area...
nvidia.ko:
Running module version sanity check.
- Original module
- No original module exists within this kernel
- Installation
- Installing to /lib/modules/5.10.134-14.1.al8.x86_64/extra/
nvidia-uvm.ko:
Running module version sanity check.
- Original module
- No original module exists within this kernel
- Installation
- Installing to /lib/modules/5.10.134-14.1.al8.x86_64/extra/
nvidia-modeset.ko:
Running module version sanity check.
- Original module
- No original module exists within this kernel
- Installation
- Installing to /lib/modules/5.10.134-14.1.al8.x86_64/extra/
nvidia-drm.ko:
Running module version sanity check.
- Original module
- No original module exists within this kernel
- Installation
- Installing to /lib/modules/5.10.134-14.1.al8.x86_64/extra/
nvidia-peermem.ko:
Running module version sanity check.
- Original module
- No original module exists within this kernel
- Installation
- Installing to /lib/modules/5.10.134-14.1.al8.x86_64/extra/
Adding any weak-modules
depmod...
[root@iZ8vbgjtg291j3aek7yk6bZ ~]# nvidia-smi
Tue Jul 11 21:10:47 2023
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 525.105.17 Driver Version: 525.105.17 CUDA Version: 12.0 |
|-------------------------------+----------------------+----------------------+
| 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 Tesla V100-SXM2... Off | 00000000:00:07.0 Off | 0 |
| N/A 32C P0 39W / 300W | 0MiB / 16384MiB | 2% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
+-----------------------------------------------------------------------------+