Abstract

Reconfigurable circuits and systems have evolved from application specific accelerators to a general purpose computing paradigm. But the algorithmic techniques and software tools are also heavily based on the hardware paradigm from which they have evolved. Loop statements in traditional programs consist of regular, repetitive computations which are the most likely candidates for performance enhancement using configurable hardware. This paper develops a formal methodology for mapping loops onto reconfigurable architectures. We develop a parameterized abstract model of reconfigurable architectures which is general enough to capture a wide range of configurable systems. Our abstract model is used to define and solve the problem of mapping loop statements onto reconfigurable architectures. We show a polynomial time algorithm to compute the optimal sequence of configurations for one important variant of the problem. We illustrate our approach by showing the mapping of an example loop statement.

Citations