This installation is for:
Fedora 33
cuda toolkit 11.4
cuDNN 8
root 24.00
Pre-installation
- Verify the system has a CUDA-capable GPU.
- Verify the system is running a supported version of Linux.
- Verify the system has gcc installed.
- Verify the system has the correct kernel headers and development packages installed.
- Download the NVIDIA CUDA Toolkit.
- Handle conflicting installation methods.
cuda toolkit:
https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html
First:
dnf install gcc-c++ mesa-libGLU-devel libX11-devel libXi-devel libXmu-devel dnf install freeglut freeglut-devel
Get cuda toolkit:wget https://developer.download.nvidia.com/compute/cuda/11.4.0/local_installers/cuda-repo-fedora33-11-4-local-11.4.0_470.42.01-1.x86_64.rpm
sudo rpm -i cuda-repo-fedora33-11-4-local-11.4.0_470.42.01-1.x86_64.rpm
sudo dnf clean all
sudo dnf -y module install nvidia-driver:latest-dkms
sudo dnf -y install cuda
After installation
cat << EOF > /etc/profile.d/cuda.sh
pathmunge /usr/local/cuda-11.1/bin before
if [ -z "${LD_LIBRARY_PATH}" ]; then
LD_LIBRARY_PATH=/usr/local/cuda-11.3/lib64
else
LD_LIBRARY_PATH=/usr/local/cuda-11.3/lib64:$LD_LIBRARY_PATH
fi
export PATH LD_LIBRARY_PATH
EOF
After installation (check all is well):
cd /(insallation location)/NVIDIA_CUDA-11.4_Samples/5_Simulations/nbody
make
./nbody
if you see the simulation all is working with cuda toolkit
Installation of cuDNN libraries:
https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html
https://developer.nvidia.com/cudnn
You can download by registering or get them from our drop box under cuDNN
Download cuDNN v8.2.2 (July 6th, 2021), for CUDA 11.4
These work with Fedora 33 as well:
cuDNN Library for Linux (x86_64)
cuDNN Runtime Library for RedHat/Centos 8 PPC (RPM)
cuDNN Developer Library for RedHat/Centos 8 PPC (RPM)
cuDNN Code Samples and User Guide for RedHat/Centos 8 PPC (RPM)
tar -xzvf cudnn-x.x-linux-x64-v8.x.x.x.tgz
sudo cp cuda/include/cudnn*.h /usr/local/cuda/include
sudo cp -P cuda/lib64/libcudnn* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn*.h /usr/local/cuda/lib64/libcudnn*
Install root:
https://root.cern/install/build_from_source/
git clone --branch v6-24-00 https://github.com/root-project/root.git root_src
mkdir root_build root_install
cd root_build/
cmake -DCMAKE_INSTALL_PREFIX=/home/ptgroup/root_install/ -DCMAKE_CXX_STANDARD=14 -Dminuit2=ON -Dvmc=ON -Dcuda=ON -Dcudnn=ON -Dtmva-gpu=ON -Dtmva-cpu=ON /home/ptgroup/root_src/
cmake --build . --target install
https://docs.nvidia.com/cuda/cuda-quick-start-guide/index.html
https://www.if-not-true-then-false.com/2018/install-nvidia-cuda-toolkit-on-fedora/