Alma9でminimumインストールからCUDAインストール

[root@gpu2 ~]# grubby –update-kernel=ALL –args=”selinux=0″

[root@gpu2 ~]#

[root@gpu2 ~]# grubby –info=ALL

index=0

kernel=”/boot/vmlinuz-5.14.0-162.6.1.el9_1.x86_64″

args=”ro crashkernel=1G-4G:192M,4G-64G:256M,64G-:512M resume=UUID=35522ad0-d2cf-4a3d-b23c-9161372ba7ac selinux=0

root=”UUID=efe6299e-70fd-4a8f-ae3e-10134cdc17d0″

initrd=”/boot/initramfs-5.14.0-162.6.1.el9_1.x86_64.img”

title=”AlmaLinux (5.14.0-162.6.1.el9_1.x86_64) 9.1 (Lime Lynx)”

id=”03d008a262a34de6ab8845431b0d881f-5.14.0-162.6.1.el9_1.x86_64″

index=1

kernel=”/boot/vmlinuz-0-rescue-03d008a262a34de6ab8845431b0d881f”

args=”ro crashkernel=1G-4G:192M,4G-64G:256M,64G-:512M resume=UUID=35522ad0-d2cf-4a3d-b23c-9161372ba7ac selinux=0

root=”UUID=efe6299e-70fd-4a8f-ae3e-10134cdc17d0″

initrd=”/boot/initramfs-0-rescue-03d008a262a34de6ab8845431b0d881f.img”

title=”AlmaLinux (0-rescue-03d008a262a34de6ab8845431b0d881f) 9.1 (Lime Lynx)”

id=”03d008a262a34de6ab8845431b0d881f-0-rescue”

[root@gpu2 ~]#

 

[root@gpu2 ~]# vi /etc/modprobe.d/blacklist-nouveau.conf

[root@gpu2 ~]# cat /etc/modprobe.d/blacklist-nouveau.conf

blacklist nouveau

options nouveau modeset=0

[root@gpu2 ~]#

[root@gpu2 ~]#

[root@gpu2 ~]# dracut -force

[root@gpu2 ~]# reboot

 

[root@gpu2 ~]# sestatus

SELinux status:                 disabled

[root@gpu2 ~]#

[root@gpu2 ~]# lsmod | grep nouveau

[root@gpu2 ~]#

 

[root@gpu2 ~]# dnf install gcc make gcc-gfortran tar kernel-devel

メタデータの期限切れの最終確認: 2:26:36 時間前の 2023年04月26日 21時54分20秒 に実施しました。

依存関係が解決しました。

======================================================================================================================================================

パッケージ                                 アーキテクチャー       バージョン                            リポジトリー                           サイズ

======================================================================================================================================================

インストール:

gcc                                        x86_64                 11.3.1-2.1.el9.alma                   InstallMedia-AppStream                  32 M

gcc-gfortran                               x86_64                 11.3.1-2.1.el9.alma                   InstallMedia-AppStream                  13 M

make                                       x86_64                 1:4.3-7.el9                           InstallMedia-BaseOS                    531 k

インストール後のサイズ: 188 M

これでよろしいですか? [y/N]: y

パッケージのダウンロード:

完了しました!

[root@gpu2 ~]#

 

[root@gpu2 cuda]# bash cuda_12.1.1_530.30.02_linux.run

lqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqk

x  End User License Agreement                                                  x

x  ————————–                                                  x

x                                                                              x

x  NVIDIA Software License Agreement and CUDA Supplement to                    x

x  Software License Agreement. Last updated: October 8, 2021                   x

x                                                                              x

x  The CUDA Toolkit End User License Agreement applies to the                  x

x  NVIDIA CUDA Toolkit, the NVIDIA CUDA Samples, the NVIDIA                    x

x  Display Driver, NVIDIA Nsight tools (Visual Studio Edition),                x

x  and the associated documentation on CUDA APIs, programming                  x

x  model and development tools. If you do not agree with the                   x

x  terms and conditions of the license agreement, then do not                  x

x  download or use the software.                                               x

x                                                                              x

x  Last updated: October 8, 2021.                                              x

x                                                                              x

x                                                                              x

x  Preface                                                                     x

x  ——-                                                                     x

x                                                                              x

xqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqx

x Do you accept the above EULA? (accept/decline/quit):                         x

x accept                                                                       x

mqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqj

 

lqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqk

x CUDA Installer                                                               x

x – [X] Driver                                                                 x

x      [X] 530.30.02                                                           x

x + [X] CUDA Toolkit 12.1                                                      x

x   [X] CUDA Demo Suite 12.1                                                   x

x   [X] CUDA Documentation 12.1                                                x

x – [ ] Kernel Objects                                                         x

x      [ ] nvidia-fs                                                           x

x   Options                                                                    x

x   Install                                                                    x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x                                                                              x

x Up/Down: Move | Left/Right: Expand | ‘Enter’: Select | ‘A’: Advanced options x

mqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqj

 

 

===========

= Summary =

===========

 

Driver:   Installed

Toolkit:  Installed in /usr/local/cuda-12.1/

 

Please make sure that

–   PATH includes /usr/local/cuda-12.1/bin

–   LD_LIBRARY_PATH includes /usr/local/cuda-12.1/lib64, or, add /usr/local/cuda-12.1/lib64 to /etc/ld.so.conf and run ldconfig as root

 

To uninstall the CUDA Toolkit, run cuda-uninstaller in /usr/local/cuda-12.1/bin

To uninstall the NVIDIA Driver, run nvidia-uninstall

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

[root@gpu2 cuda]#

 

[root@gpu2 cuda]# ldconfig

[root@gpu2 cuda]#

[root@gpu2 cuda]# nvidia-smi

Thu Apr 27 00:29:43 2023

+—————————————————————————————+

| NVIDIA-SMI 530.30.02              Driver Version: 530.30.02    CUDA Version: 12.1     |

|—————————————–+———————-+———————-+

| 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 A2                       Off| 00000000:05:00.0 Off |                    0 |

|  0%   37C    P0               20W /  60W|      0MiB / 15356MiB |      0%      Default |

|                                         |                      |                  N/A |

+—————————————–+———————-+———————-+

 

+—————————————————————————————+

| Processes:                                                                            |

|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |

|        ID   ID                                                             Usage      |

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

|  No running processes found                                                           |

+—————————————————————————————+

[root@gpu2 cuda]#