Efficient pipelining of nested loops : unroll-and-squash

Abstract

The size and complexity of current custom VLSI have forced the use of high-level programming languages to describe hardware, and compiler and synthesis technology to map abstract designs into silicon. Many applications operating on large streaming data usually require a custom VLSI because of high performance or low power restrictions. Since the data processing is typically described by loop constructs in a high-level language, loops are the most critical portions of the hardware description and special techniques are developed to optimally synthesize them. In this thesis, we introduce a new method for mapping nested loops into hardware and pipelining them efficiently. The technique achieves fine-grain parallelism even on strong intra- and inter-iteration data-dependent inner loops and, by economically sharing resources, improves performance at the expense of a small amount of additional area. We implemented the transformation within the Nimble Compiler environment and evaluated its performance on several signal-processing benchmarks. The method achieves up to 2x increase in the area efficiency compared to the best known optimization techniques.