HPE gen10 plus 安装telsa P4驱动
# HPE gen10 plus 安装telsa P4驱动
HPE gen10 plus 安装telsa P4驱动(Ubuntu 20.04.1 LTS x86_64)
# 查看显卡是否安装成功
# 下载驱动
点击下载,然后上传到服务器上。
# 安装驱动
在服务器上执行:
# 禁用Nouveau
sudo vim /etc/modprobe.d/blacklist-nouveau.conf
1
2
2
写入下面内容
blacklist nouveau
options nouveau modeset=0
1
2
2
使之生效:
sudo update-initramfs -u
sudo reboot
1
2
2
安装NVIDIA驱动:
chmod a+x NVIDIA-Linux-x86_64-460.106.00.run
sudo ./NVIDIA-Linux-x86_64-460.106.00.run
1
2
2
# 安装CUDA
最高支持CUDA 11.2
CUDA Toolkit 11.2 Downloads | NVIDIA Developer (opens new window)
wget https://developer.download.nvidia.com/compute/cuda/11.2.0/local_installers/cuda_11.2.0_460.27.04_linux.run
sudo sh cuda_11.2.0_460.27.04_linux.run
1
2
2
配置一下环境变量:
export PATH=$PATH:/usr/local/cuda-11.2/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-11.2/lib64
echo -e 'export PATH=$PATH:/usr/local/cuda-11.2/bin\nexport LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-11.2/lib64' >> /etc/profile
1
2
3
4
2
3
4
# 安装cuDNN
cuDNN Archive | NVIDIA Developer (opens new window)
下载到服务器上,解压后,将include
和lib
中的文件拷贝到cuda
的安装路径下。
tar -xvf cudnn-linux-x86_64-8.9.4.25_cuda11-archive.tar.xz
cd cudnn-linux-x86_64-8.9.4.25_cuda11-archive
sudo cp include/* /usr/local/cuda-11.2/include
sudo cp lib/* /usr/local/cuda-11.2/lib64/
1
2
3
4
2
3
4
# 安装NCCL
NVIDIA Collective Communications Library (NCCL) | NVIDIA Developer (opens new window)
下载到服务器后,解压缩,拷贝到cuda的安装路径下:
tar -xvf nccl_2.18.5-1+cuda11.0_x86_64.txz
cd nccl_2.18.5-1+cuda11.0_x86_64
sudo cp include/* /usr/local/cuda-11.2/include
sudo cp -r lib/* /usr/local/cuda-11.2/lib64/
1
2
3
4
2
3
4
上次更新: 2024/01/16, 22:32:53