Introduction

Software is at the foundation of application specification and development. Whether the end application is targeted towards entertainment, gaming, or medicine, most products available today began as a software model or prototype that needed to be accelerated and executed on a hardware device. From this starting point, the software engineer is tasked with determining the execution device to get a solution to market and to achieve the highest possible degree of acceleration possible.

One traditional approach to accomplish this task has been to rely on processor clock frequency scaling. On its own, this approach has entered a state of diminishing returns, which has in turn led to the development of multi-core and heterogeneous computing devices. These architectures provide the software engineer with the possibility to more effectively trade-off performance and power for different form factors and computational loads. The one challenge in using these new computing architectures is the programming model of each device. At a fundamental level, all multi-core and heterogeneous computing devices require that the programmer rethink the problem to be solved in terms of explicit parallelism.

Recognizing the programming challenge of multi-core and heterogeneous compute devices, the Khronos Group industry consortium has developed the OpenCL™ programming standard. The OpenCL specification for multi-core and heterogeneous compute devices defines a single consistent programming model and system level abstraction for all hardware devices that support the standard. For a software engineer this means a single programming model to learn what can be directly used on devices from multiple vendors.

As specified by the OpenCL standard, any code that complies with the OpenCL specification is functionally portable and will execute on any computing device that supports the standard. Therefore, any code change is for performance optimization. The degree to which an OpenCL program needs to be modified for performance depends on the quality of the starting source code and the execution environment for the application.

Xilinx is an active member of the Khronos Group, collaborating on the OpenCL specification, and supports the compilation of OpenCL programs for Xilinx® FPGAs. The Xilinx SDAccel™ development environment is used for compiling OpenCL programs to execute on a Xilinx FPGA.

There are some differences between compiling a program for execution in an FPGA and a CPU/GPU environment. The following chapters in this guide describe how to use the SDAccel development environment to compile an OpenCL program for a Xilinx FPGA. This book is intended to document the features and usages of the SDAccel development environment. It is assumed that the user already has a working knowledge of OpenCL API. Though it includes some high level OpenCL concepts, it is not intended as an exhaustive technical guide on the OpenCL API. For more information on the OpenCL API, see the OpenCL specification available from the Khronos Group, and the OpenCL API introductory videos available on the Xilinx website.