Source-directed transformations for hardware compilation

摘要

This paper presents the Haydn-C language and its parallel programming model. They have been developed to support modular hardware design, to improve designer productivity, and to enhance design quality and maintainability. The principal innovation of Haydn-C is a framework of optional annotations to enable users to describe design constraints, and to direct source-level transformations such as scheduling and resource allocation. We have automated such transformations so that a single high-level design can be used to produce many implementations with different design trade-offs. The effectiveness of this approach has been evaluated using various case studies, including FIR filters, fractal generators, and morphological operators. For instance, the fastest morphological erosion design is 129 times faster and 3.4 times larger than the smallest design.