How to build NVIDIA GPU Plugin

Contents

• Introduction
• Building for Linux

Introduction

OpenVINO™ NVIDIA GPU plugin is not included into Intel® Distribution of OpenVINO™. To use the plugin, it should be built from source code.

Building for Linux

How to build

In order to build the plugin, you must prebuild OpenVINO™ package from source using this guideline.

Afterwards plugin build procedure is as following:

1. Install one of the following compilers with support of C++17:

• Install gcc-7 compiler

sudo apt-get update
sudo apt-get install gcc-7 g++7

• Install clang-8 compiler

sudo apt-get update
sudo apt-get install clang-8 clang++8

2. Install NVIDIA 460 version of driver from NVIDIA download drivers

3. Install CUDA 11.2 from How to install CUDA

   Do not forget to add <path_to_cuda>/bin/ in PATH variable for example export PATH="<path_to_cuda>/bin:$PATH"

4. Install cuDNN 8.1.0 from How to install cuDNN

5. Install cuTENSOR 1.3.0 from How to install cuTENSOR

6. Clone openvino_contrib repository:

git clone --recurse-submodules --single-branch --branch=master https://github.com/openvinotoolkit/openvino_contrib.git 

7. Go to plugin directory:

cd openvino_contrib/modules/nvidia_plugin
git checkout develop

8. Prepare a build folder:

mkdir build && cd build

9. Build plugin

   First of all, switch OpenVINO™ to tag 2021.4 and then build it according the instruction How to build

   Then build NVIDIA GPU Plugin with one of 2 options:

• Using build.sh

  Setup the following environment variables:

  export OPENVINO_HOME=<OpenVINO source directory>
  export OPENVINO_CONTRIB=<OpenVINO Contrib packages source directory>
  export OPENVINO_BUILD_PATH=<OpenVINO build directory>

  Then run one of the following commands:

  # Run cmake configuration (if necessary) and then build
  ../build.sh --build
  
  # Run cmake configuration
  ../build.sh --setup
  
  # For old build delete old configuration, generate new one and then build
  ../build.sh --rebuild

• Using OpenVINODeveloperPackage

  Run the following command:

  cmake -DOpenVINODeveloperPackage_DIR=<path to OpenVINO package build folder> -DCMAKE_BUILD_TYPE=Release ..
  cmake --build . --target openvino_nvidia_gpu_plugin -j `nproc`

Docker support

Build docker container

First build docker container:

1. Install docker:

./docker.sh install
su $USER # Relogin for current user

2. Download all *.deb packages for CUDA and put them in one folder
3. Build docker container:

CUDA_PACKAGES_PATH=<path to CUDA packages> ./docker.sh build

Build NVIDIA GPU Plugin in docker container

In order to build NVIDIA GPU Plugin in docker, follow the steps:

1. Enter the docker container:

docker run --gpus all -it openvino/cudaplugin bin/bash

2. Build the OpenVINO and NVIDIA GPU Plugin according the steps described in ## How to build, except 3), 4), 5) steps (this packages already installed in image)
3. Commit all your changes in container:

docker commit openvino/cudaplugin <name of new image>