An evaluation of functional unit lengths for single-chip processors

ABSTRACT

When designing a pipelined single-chip processor (SCP) with pipelined functional units of varying length, the processor issue logic must deal with scheduling of the result bus. In order to prevent serious performance degradation due to result bus conflicts, some pipeline scheduling techniques developed in the 1970's may need to be incorporated into the issue logic. Since this is a non-trivial complication of the issue logic, a set of simulations were performed in order to evaluate the effectiveness of the combination of multiple length functional units and scheduling techniques. Analysis of the simulation results indicates that providing relatively short multiple length functional units is not worthwhile. Multiple length functional unit configurations employing result bus scheduling do perform slightly better than uniform length configurations, but the difference is often less than 1%. Thus, the SCP designer should not waste valuable time improving the performance of each functional unit, but rather should produce a good design for the most complicated unit and design all other units to match it.

REFERENCES

• Farr89: Matthew Karl Farrens , Andrew R. Pleszkun, The design and analysis of a high-performance single-chip processor, 1989
• FaPl89: M. K. Farrens , a. R. Pleszkun, Improving performance of small on-chip instruction caches, Proceedings of the 16th annual international symposium on Computer architecture, p.234-241, April 1989, Jerusalem, Israel
• GHLP85: J. R. Goodman , Jian-tu Hsieh , Koujuch Liou , Andrew R. Pleszkun , P. B. Schechter , Honesty C. Young, PIPE: a VLSI decoupled architecture, Proceedings of the 12th annual international symposium on Computer architecture, p.20-27, June 17-19, 1985, Boston, Massachusetts, United States
• HsPG84: J. T. Hsieh, A. R. Pleszkun and J. R. Goodman, "Performance Evaluation of the PIPE Computer Architecture", Computer Science Department Technical Report #566, University of Wisconsin-Madison , Madison, Wisconsin (November 1984).
• McMa84: F. H. McMahon, LLNL FORTRAN KERNELS: MFLOPS, Lawrence Livermore Laboratories, Livermore, California, (March 1984).
• PaDa76: Janak H. Patel , Edward S. Davidson, Improving the throughput of a pipeline by insertion of delays, ACM SIGARCH Computer Architecture News, v.4 n.4, p.159-164, January 1976
• Patt85: David A. Patterson, Reduced instruction set computers, Communications of the ACM, v.28 n.1, p.8-21, Jan. 1985 [doi>10.1145/2465.214917]
• Russ78: Richard M. Russell, The CRAY-1 computer system, Communications of the ACM, v.21 n.1, p.63-72, Jan. 1978 [doi>10.1145/359327.359336]
• Smit88: J. E. Smith, Characterizing computer performance with a single number, Communications of the ACM, v.31 n.10, p.1202-1206, Oct. 1988 [doi>10.1145/63039.63043]
• Thor70: J. E. Thornton, Design of a Computer—The Control Data 6600, Scott Foresman & Co, 1970
• Toma67: R. M. Tomasulo, "An Efficient Algorithm for Exploiting Multiple Arithmetic Units", IBM Journal, vol. 11 (January 1967), pp. 25-33.
• YoGo84: H. C. Young and J. R. Goodman, "A Simulation Study of Architectural Data Queues and Prepare-to-Branch Instruction", Proceedings of the IEEE INT Conference on Computer Design: VLSI in Computers, Port Chester, New York (October 1984), pp. 544-549.