In particular, coarse-grained arrays generated using our approach are intended to be integrated into customizable processors that use application-specific instruction set extensions to accelerate ...