Abstract

Compile-time code transformations which expose instruction-level parallelism (ILP) typically take into account the constraints imposed byallexecution scenarios in the program. However, there areaddi tional opportunities to increase ILP along some execution sequences if the constraints from alternative execution sequences can be ignored. Traditionally, pro le information has been used to identify important execution sequences for aggressive compiler optimization and scheduling. This paper presents a set of static program analysis heuristics used in the IMPACT compiler to identify execution sequences for aggressive optimization. We show that the static program analysis heuristics identify execution sequences without hazardous conditions that tend to prohibit compiler optimizations. As a result, the static program analysis approach often achieves optimization results comparable to pro le information in spite of its inferior branch prediction accuracies. This observation makes a strong case for using static program analysis with or without pro le information to facilitate aggressive compiler optimization and scheduling.

Citations