基于国产软硬件的OpenCL计算平台研究  

OpenCL computing platform based on domestic software and hardware

作　　者：安婷玉[1] 郭宝宝 AN Ting-yu;GUO Bao-bao(College of Economics and Management,Taiyuan University of Technology,Taiyuan 030024;Troop 66133,Beijing 100144,China)

机构地区：[1]太原理工大学经济管理学院,太原山西030024 [2]66133部队,北京100144

出　　处：《计算机工程与科学》2019年第11期1919-1923,共5页Computer Engineering & Science

基　　金：太原理工大学校青年基金(2014QN040)

摘　　要：随着智能计算和大数据应用的发展,人们对GPU等加速部件的需求不断增长。基于国产基础软硬件平台运行显控应用做加速计算的需求,研究了OpenCL计算平台的移植和实现途径,就国产软硬件平台进行GPU计算做出了初步探索。研究的计算平台包括Mesa、ROCm、Pocl和Beignet,最后给出了如何将ROCm在国产平台上移植适配的思路和解决方案。With the development of intelligent computing and big data applications,the demand for acceleration components such as GPU is increasing.Based on the requirements for accelerated computing by running display and control applications on domestic fundamental software and hardware platforms,this paper studies the portability and implementation of OpenCL computing platforms,and preliminarily explores the GPU computing on domestic software and hardware platforms.The studied computing platforms include Mesa,ROCm,Pocl and Beignet.Finally,the idea and solution of how to port and adapt ROCm to domestic platforms are given.

关 键 词：国产基础软硬件 OpenCL计算平台 移植 GPU计算 

分 类 号：TP311.54[自动化与计算机技术—计算机软件与理论]