A new heuristic for SOA problem based on effective tie break function

Abstract

Producing efficient and compact code for embedded DSP processors is very important for nowadays faster and smaller size devices. Because such processors have highly irregular data-path, conventional code generation techniques typically result in inefficient code. Embedded software compilers are expected to make use of the Address Generation Unit (AGU); a feature commonly found in modern embedded DSP processors. This helps in generating optimized offset assignments to program variables in memory, and consequently minimize the overhead instructions dedicated for addresses computations. This paper addresses one of the problems of code optimizations; namely Simple Offset Assignment (SOA) problem. In this paper, we study the tie break function introduced by Leupers and Marwedel [1] and show that this function does not represent the actual tie break that could happen in the graph. Then we introduce the notion of Effective Tie Break Function (ETBF) and use it in proposing a new algorithm for solving the SOA problem. We apply the algorithm to randomly generated graphs. Our results show improvement in offset assignment cost of up to 7% over well known offset assignment algorithms [1,2,3].