Study on Embedded Software Performance Estimation Algorithm
It is necessary to compute the execution time upper bound of embedded hard real-time program under the worst condition in embedded system design, which decides how hardware and software to partition and how to schedule process. Modern microprocessor which use cache memory system and instruction pre-fetching increase the difficulty to compute the upper bound accurately. A new estimation method of embedded software performance based on instruction cache and pre-fetching model is proposed, which use control flow graph and cache conflict graph and combine instruction pre-fetching into cache analysis, it makes the execution time upper bound more accurate.
Guowei Wu Lin Yao
Software School of Dalian University of Technology, Dalian 116023,China
国际会议
首届嵌入式软件与系统国际会议(Proceedings of the First International Conference on Embedded Software and System)
杭州
英文
535-540
2004-12-09(万方平台首次上网日期,不代表论文的发表时间)