在可编程片上系统(System on P rogramm ab le Ch ip,SOPC)中,特定应用程序中用到的指令是软核CPU指令集的子集,如果在FPGA中实现软核CPU时仅保留应用程序用到的指令子集,将可以提高硬件资源利用率.文中分析了对应用程序进行指令统计的...在可编程片上系统(System on P rogramm ab le Ch ip,SOPC)中,特定应用程序中用到的指令是软核CPU指令集的子集,如果在FPGA中实现软核CPU时仅保留应用程序用到的指令子集,将可以提高硬件资源利用率.文中分析了对应用程序进行指令统计的方法,重点介绍了通过修改HDL文件对8051软核CPU指令集进行删减和扩充的技术.采用这一技术设计的SOPC芯片硬件资源利用率得到了显著的提高,降低了系统成本,适用于可编程逻辑资源受到限制和对成本敏感的嵌入式应用中.展开更多
This paper illustrates the importance of the configuration of function units and the change of an application’s critical path when using instruction set extension （ISE） with multi-issue architectures. This paper al...This paper illustrates the importance of the configuration of function units and the change of an application’s critical path when using instruction set extension （ISE） with multi-issue architectures. This paper also presents an automatic identification approach for customized instruction without input/output number constraints for multi-issue architectures. The approach identifies customized instructions using multiple attribute decision-making based on the analysis of several attributes for each candidate node. Tests indicate that the approach achieves higher speedup ratios than previous approaches, as well as less area cost. In addition, this approach provides designers with multiple candidate designs.展开更多
本文针对常见启发式算法中忽略指令与指令实例区别的问题,改进了一个已有启发式算法Greedy-Heur:根据指令实例的启发式函数值得出相应指令的权值,并根据指令的优先级关系以贪心策略进行指令实例选择.针对启发式算法无法找到最优解的问题...本文针对常见启发式算法中忽略指令与指令实例区别的问题,改进了一个已有启发式算法Greedy-Heur:根据指令实例的启发式函数值得出相应指令的权值,并根据指令的优先级关系以贪心策略进行指令实例选择.针对启发式算法无法找到最优解的问题,本文引入基于群体搜索的差分进化算法,并结合贪心策略,提出了ISDE(Instruction Selection Based on Differential Evolution)算法.ISDE算法通过简单的编码和高效的适应度评价机制,快速地迭代搜索最优指令组合.实验结果表明,GreedyHeur和ISDE算法能快速有效地找到比已有启发式算法更优的候选指令组合.展开更多
文摘在可编程片上系统(System on P rogramm ab le Ch ip,SOPC)中,特定应用程序中用到的指令是软核CPU指令集的子集,如果在FPGA中实现软核CPU时仅保留应用程序用到的指令子集,将可以提高硬件资源利用率.文中分析了对应用程序进行指令统计的方法,重点介绍了通过修改HDL文件对8051软核CPU指令集进行删减和扩充的技术.采用这一技术设计的SOPC芯片硬件资源利用率得到了显著的提高,降低了系统成本,适用于可编程逻辑资源受到限制和对成本敏感的嵌入式应用中.
基金Supported by the Basic Research Fund of Tsinghua University
文摘This paper illustrates the importance of the configuration of function units and the change of an application’s critical path when using instruction set extension （ISE） with multi-issue architectures. This paper also presents an automatic identification approach for customized instruction without input/output number constraints for multi-issue architectures. The approach identifies customized instructions using multiple attribute decision-making based on the analysis of several attributes for each candidate node. Tests indicate that the approach achieves higher speedup ratios than previous approaches, as well as less area cost. In addition, this approach provides designers with multiple candidate designs.
文摘本文针对常见启发式算法中忽略指令与指令实例区别的问题,改进了一个已有启发式算法Greedy-Heur:根据指令实例的启发式函数值得出相应指令的权值,并根据指令的优先级关系以贪心策略进行指令实例选择.针对启发式算法无法找到最优解的问题,本文引入基于群体搜索的差分进化算法,并结合贪心策略,提出了ISDE(Instruction Selection Based on Differential Evolution)算法.ISDE算法通过简单的编码和高效的适应度评价机制,快速地迭代搜索最优指令组合.实验结果表明,GreedyHeur和ISDE算法能快速有效地找到比已有启发式算法更优的候选指令组合.
