Installing Prerequisite Products

To use GPU Coder™ for CUDA^®code generation, you must install and setup the following products. For setup instructions , seeSetting Up the Prerequisite Products.

MathWorksProducts and Support Packages

MATLAB^®(required).
MATLAB Coder™(required).
Parallel Computing Toolbox™ (required).
金宝app^®(required for generating code from Simulink models).
Computer Vision Toolbox™ (recommended).
Deep Learning Toolbox™ (required for deep learning).
Embedded Coder^®(recommended).
Image Processing Toolbox™ (recommended).
Simulink Coder(required for generating code from Simulink models).
GPU Coder Interface for Deep Learning Librariessupport package (required for deep learning).
MATLAB Coder Support Package for NVIDIA^®Jetson^®and NVIDIA DRIVE^®Platforms(required for deployment to embedded targets such as NVIDIA Jetson and Drive).

For instructions on installing MathWorks^®products, see the MATLAB installation documentation for your platform. If you have installed MATLAB and want to check which other MathWorks products are installed, enterverin the MATLAB Command Window. To install the support packages, use Add-On Explorer in MATLAB.

If MATLAB is installed on a path that contains non 7-bit ASCII characters, such as Japanese characters, GPU Coder does not work because it cannot locate code generation library functions.

Third-Party Hardware

NVIDIA GPU enabled for CUDA with a compatible graphics driver. For more information, seeCUDA GPUs (NVIDIA).

To see the CUDA compute capability requirements for code generation, consult the following table.

Target	Compute Capability
CUDA MEX	SeeGPU Support by Release.
Source code, static or dynamic library, and executables	3.2 or higher.
Deep learning applications in 8-bit integer precision	6.1, 6.3 or higher.
Deep learning applications in half-precision (16-bit floating point)	5.3, 6.0, 6.2 or higher.

ARM^®Mali graphics processor.
For the Mali device, GPU Coder supports code generation for only deep learning networks.

Third-Party Software

Required

C/C++ Compiler:

Linux^®	Windows^®
GCC C/C++ compiler. For supported versions, seeSupported and Compatible Compilers	Microsoft^®Visual Studio^®2013
	Microsoft Visual Studio2015
	Microsoft Visual Studio2017
	Microsoft Visual Studio2019

Optional

For CUDA MEX, the code generator uses the NVIDIA compiler and libraries installed with MATLAB. Standalone code (static library, dynamically linked library, or executable program) generation has additional software requirements.

Software Name	Information
CUDA toolkit	GPU Coder has been tested with CUDA toolkit v9.x-v11.0. To download the CUDA toolkit, seeCUDA Toolkit Archive (NVIDIA).
NVIDIA Nsight™ systems	Generate an execution profiling report for the generated CUDA code. The report provides metrics that help you analyze your application algorithms and identify opportunities to optimize performance. GPU Coder has been tested with Nsight 2021.1.1
NVIDIA CUDA deep neural network library (cuDNN) for NVIDIA GPUs	主机GPU设备,GPU编码器已经测试with cuDNN v8.1.0. To download cuDNN, seecuDNN (NVIDIA).
NVIDIA TensorRT™ high performance inference optimizer and runtime library	主机GPU设备,GPU编码器已经测试with TensorRT v7.2.x. To download TensorRT, seeTensorRT (NVIDIA).
ARM Compute Library for Mali GPUs	GPU Coder has been tested with v19.05. For more information, seeCompute Library (ARM).
Open Source Computer Vision Library (OpenCV)	Required for deep learning examples. For examples targeting NVIDIA GPUs on the host development computer, use OpenCV v3.1.0. For examples targeting ARM GPUs, use OpenCV v2.4.9 on the ARM target hardware. For more information, seeOpenCV.