Techniques for Dynamically Mapping Computations to Coprocessors

In embedded reconfigurable computing systems, general purpose processors (GPPs) are typically extended with coprocessors to meet specific goals, such as higher performance and/or energy savings. Coprocessors can range from specialized modules which execute a specific task to reconfigurable arrays of ALUs. This paper presents our ongoing work on techniques to dynamically offload computations being executed by a GPP to a coprocessor. We present our method for identifying repetitive instruction traces, named as Mega blocks, as well as transformations which can be applied over those Mega blocks. We also present a proof-of-concept implementation of a system which transparently moves computations from a GPP to a Specialized Reconfigurable Array (SRA). Finally, we present our current and planned work.