Title :
Performance enhancement by eliminating redundant function execution
Author :
Chen, Peng ; Kavi, Krishna ; Akl, Robert
Author_Institution :
Dept. of Comput. Sci. & Eng., North Texas Univ., Denton, TX, USA
Abstract :
Programs often call the same function with the same arguments, yielding the same results. We call this phenomenon, "function reuse". Previously, we have shown such a behavior for some of the SPEC2000 integer benchmarks using HP ATOM instrumentation tools. However, this required extensive analysis by hand, and assumptions regarding side-effects caused by functions. In this paper, we modified a well-known architecture simulator, SimpleScalar, to analyze multiple benchmarks to investigate the function reuse behavior.
Keywords :
software performance evaluation; software reusability; HP ATOM instrumentation tool; SPEC2000 integer benchmark; SimpleScalar; architecture simulator; basic block reuse; function reuse; instruction level parallelism; instruction reuse; performance enhancement; redundant function execution elimination; speculative execution; value prediction; Analytical models; Computational modeling; Computer architecture; Computer science; Digital arithmetic; Floating-point arithmetic; History; Instruments; Parallel processing; Yarn; Basic Block; Function reuse; Instruction Level Parallelism; Instruction Reuse; Reuse; SimpleScalar; Speculative Execution; Value Prediction;
Conference_Titel :
Simulation Symposium, 2006. 39th Annual
Print_ISBN :
0-7695-2559-8
DOI :
10.1109/ANSS.2006.34
