A method is proposed for optimizing a Control I Data Flow Graph (CDFG), to generate shorter schedules, given a freed number of functional units.A counter-intuitive approach is taken of adding more operations to the CDFG to reduce the number of cycles in the final schedule.In this paper, this technique is applied to individual straight-line code sections (stateblocks) to reduce their execution times (as opposed to recently published methods for optim"zations that transcend state block boundaries).In many cases this technique has produced schedules significantly better (up to 30$%0) than those previously published.The algorithm rearranges the CDFG, modifjing the syntactical structure, while preserving the semantic meaning.The focus of attack is on the critical path, where most of the transformations are attempted to decrease the length of the critical path and hence the length of the resultant CDFG.The benchmarks were run on a MIPS RC3240 Riscomputer (-18 MIPS) and each schedule was derived at in less than 1 cpu second!

This paper addresses relevant issues in connectivity synthesis particularly with respect to other design concerns in high-level behavioral synthesis.A restricted model is used to run a series of design experimerus to provide insight into the relationship between scheduling, conrwctivity, and placement & routing.Contrary to the common approach of counting multiplexer, multiplexer inputs, and busses, this approach looks at more germanz issues, namely: I) How to decide how much a "bus" is really going to cost, 2) How much does "load balanced" scheduling help, especially in the case of resource constrained scheduling, and 3) Should the number of multiplexer inputs really be the main criterion for judging the connectivity quality of the designs?