作者: Andrew Wolfe , Alex Chanin
关键词: Encoding (memory) 、 Classic RISC pipeline 、 Cache 、 Parallel computing 、 Complex instruction set computing 、 Computer science 、 Memory bandwidth 、 Workstation 、 Reduced instruction set computing 、 Systems architecture
摘要: The difference in code size between RISC and CISC processors appears to be a significant factor limiting the use of architectures embedded systems. Fortunately, programs can effectively compressed. An ideal solution is design system that directly execute compressed programs. A new architecture called Compressed Code Processor presented. This processor depends on code-expanding instruction cache manage compression transparent since all instructions are executed from cache. Experimental simulations show degree achieved fixed encoding scheme. impact performance slight for some memory implementations reduced bandwidth actually increases performance.