On the use of data flow analysis in static profiling

Static profiling is a technique that produces estimates of exe-cution likelihoods or frequencies based on source code anal-ysis only. It is frequently used in determining cost/benefit ratios for certain compiler optimizations. In previous work, we introduced a simple algorithm to compute execution like-lihoods, based on a control flow graph and heuristic branch prediction. In this paper we examine the benefits of using more in-volved analysis techniques in such a static profiler. In par-ticular, we explore the use of value range propagation to im-prove the accuracy of the estimates, and we investigate the differences in estimating execution likelihoods and frequen-cies. 1