Loop Quantization: Unwinding for Fine-Grain Parallelism Exploitation

Abstract

Loop unwinding is a well known technique for reducing loop overhead, exposing parallelism and increasing the efficiency of pipelining. Traditional loop unwinding is limited to the innermost loop in a group of nested loops and the amount of unwinding is either fixed or has to be specified by the user, on a case by case basis. In this paper we present a general technique for automatically unwinding multply nested loops, explain its advantages over other transformation techniques and illustrates its practical effectiveness.