基于表调度的Matrix DSP指令调度算法的实现  

Implementation of instruction scheduling algorithm based on list scheduling for Matrix DSP

作　　者：罗杰[1] 陈跃跃[1] 孙海燕[1] 阳柳[1] 淡孝强[1] 辛乃军[1] 王霁[1] 

机构地区：[1]国防科学技术大学计算机学院,湖南长沙410073

出　　处：《计算机工程与科学》2013年第8期25-30,共6页Computer Engineering & Science

基　　金：核高基重大专项资助项目(2009ZX01034-001-006)

摘　　要：指令调度是gcc实现指令并行、提高性能的一种优化策略,gcc目前支持的调度算法主要有表调度算法与模调度算法。主要根据Matrix芯片的体系结构特点,对现有的表调度算法进行了改进,实现了Matrix指令调度算法。实验结果表明,改进后的表调度算法能够编译出正确的指令,充分挖掘指令间的并行性,显式标注指令间的并行关系,指令字间的延迟关系符合硬件要求。Instruction scheduling is an optimization strategy for gcc to achieve instruction level paral- lelism and to improve performance. At present, the scheduling algorithm that gcc supports are list scheduling algorithm and swing model scheduling. According to the architecture feature of Matrix DSP, an improved instruction scheduling algorithm based on the list scheduling algorithm is implemented. The experiment results show that the improved algorithm can compile the c code into correct assemble codes, improve the performance of instruction parallelism, explicitly mark the parallelism between instructions, and the delay between instructions meets the hardware demands.

关 键 词：GCC 超长指令字 MATRIX 表调度算法 

分 类 号：TP314[自动化与计算机技术—计算机软件与理论]