Abstract data types and high-level synthesis

  • Authors:
  • Gregory S. Whitcomb;A. Richard Newton

  • Affiliations:
  • Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA;Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA

  • Venue:
  • DAC '90 Proceedings of the 27th ACM/IEEE Design Automation Conference
  • Year:
  • 1991

Quantified Score

Hi-index 0.00

Visualization

Abstract

Previous data abstraction efforts in software and hardware description have approached abstract data types from a specification and simulation standpoint. This paper addresses the use of abstract data types for high-level synthesis. Data types are extended to support don't care information and new synthesis steps are presented. Results from a number of realistic design examples indicate that the area and delay of logic implementations can be substantially reduced when data abstraction and appropriate synthesis procedures are incorporated into the high-level synthesis process.