Advertisement

An Efficient Microcode-Compiler for Custom DSP-Processors

  • Gert Goossens
  • Jan Rabaey
  • Joos Vandewalle
  • Hugo De Man

Abstract

In this paper, a microcode compiler for custom DSP-processors is presented. This tool is part of the CATHEDRAL II silicon compiler. Two optimization problems in the microcode compilation process are highlighted: microprogram scheduling and memory allocation. Algorithms to solve them, partly based on heuristics, are presented. Our compiler successfully handles repetitive programs, and is able to decide on hardware binding. In most practical examples, optimal solutions are found. Whenever possible, indications of the complexity are given.

Keywords

Schedule Algorithm Data Path Memory Allocation Incremental Search Architecture Synthesis 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    A.V. Aho, J.D. Ullman, “Principles of compiler design”, Addison-Wesley, Reading, 1977.Google Scholar
  2. [2]
    E.G. Coffman Jr, “Computer and job scheduling theory”, J. Wiley and Sons, New York, 1976.MATHGoogle Scholar
  3. [3]
    H. De Man et al., “CATHEDRAL II: a synthesis and module generation system for multiprocessor systems on a chip”, Proc. NATO-ASIon log. synth. and silicon compilation for VLSI-design, L’Aquila, July 1986.Google Scholar
  4. [4] H. De Man et al., “CATHEDRAL II: a silicon compiler for digital signal processing”, IEEE Design and Test, pp. 13–25, Dec. 1986.Google Scholar
  5. [5]
    A. Gibbons, “Algorithmic graph theory”, pp. 198–201, Cambridge Univ. Press, Cambridge, 1985.MATHGoogle Scholar
  6. [6]
    G. Goossens et al., “An efficient microcode-compiler for custom multiprocessor DSP-systems”, Technical report ESPRIT97/IMEC/3.87/D/d(4)/1, available from IMEC vzw, Leuven, March 1987.Google Scholar
  7. [7]
    G. Goossens, “Techniques for automated hardware assignment”, Technical report ES-PRIT97/IMEC/9.87/D/d(4)/1, available from IMEC vzw, Leuven, Sept. 1987.Google Scholar
  8. [8] C.J. Tseng, D.P. Siewiorek, “Automated synthesis of data paths in digital systems”, IEEE Trans. CAD, pp. 379–395, July 1986.Google Scholar
  9. [9]
    J. Vanhoof et al., “A knowledge-based CAD system for synthesis of multi-processor digital signal processing chips”, Proc. VLSI’87, Vancouver, Aug. 1987.Google Scholar
  10. [10]
    J. Zeman, G.S. Moschytz, “Systematic design and programming of signal processors, using project management techniques”, IEEE Trans. ASSP, pp. 1536–1549, Dec. 1983.Google Scholar

Copyright information

© Springer Science+Business Media New York 2003

Authors and Affiliations

  • Gert Goossens
    • 1
    • 4
  • Jan Rabaey
    • 2
    • 4
  • Joos Vandewalle
    • 3
    • 4
  • Hugo De Man
    • 4
  1. 1.Target Compiler TechnologiesBelgium
  2. 2.University of California at BerkeleyUSA
  3. 3.Katholieke Universiteit LeuvenBelgium
  4. 4.IMEC LaboratoryLeuvenBelgium

Personalised recommendations