ALTIVEC VECTOR UNIT CUSTOMIZATION FOR EMBEDDED SYSTEMS
Tarik Saidani, Joël Falcou, Lionel Lacassagne, Samir Bouaziz · International journal of computer science and applications · 2008
Vector extensions for general purpose processors are an efficient feature to address the growing performance demand of multimedia and computer vision applications. Embed ded processors are the most widespread architectures for such applications. While provid ing sufficient computing power for these applications, they must take into account power, area and real- time constraints. In this paper, we propose two hardware optimization techniques to address those constraints: RISCization and instruction set customization. Experimental results show that those techniques both reduce time and power consump tion by up to 50% when compared to the original ISA.