Abstract
Fisher's trace scheduling procedure for global compaction has proven to be able to produce significant reduction in execution time of compacted microcode, however extra space may be sometimes required during bookkeeping, and the efficacy of compaction of microprogram loop is lower than that of hand compaction.
This paper introduces an improved trace scheduling compaction algorithm to mitigate the drawbacks mentioned above. The improved algorithm is based on a modified menu of moving microoperations, an improved trace scheduling algorithm, and a special loop compaction algorithm. Preliminary tests indicate that this global compaction algorithm gives shorter execution time and less space requirement in comparison with Fisher's algorithm.
- 1 A. V. Aho and J. D. Ullman, Principles of Compiler Design. Reading, MA: Addison-Wesley, 1977. Google Scholar
Digital Library
- 2 S. Davidson, D. Landskov, B. D. Shriver, and P. W. Mallett, "Some Experiments in Local Microcode Compaction for Horizontal Machine," IEEE Trans. Comput., vol.C-30, pp.460-477, July 1981.Google Scholar
Digital Library
- 3 J. A. Fisher, "The Optimization of Horizontal Microcode Within and Beyond Basic Block: An Application of Processor Scheduling with Resource," Courant Math. Comput. Lab., New York University, U.S. Dep. of Energy Rep.C00-3077-161, Oct. 1979.Google Scholar
- 4 J. A. Fisher, "Trace Scheduling: A Technique for Global Microcode Compaction," IEEE Trans. Comput., vol.C-30, pp.478-490, July 1981.Google Scholar
Digital Library
- 5 J. A. Fisher, "Very Long Instruction Word Architecture and the ELI-512," Research Report YALEU/DCS/RR-253, April 1983.Google Scholar
- 6 J. A. Fisher, D. Landskov, and B.D. Shriver, "Microcode Compaction: Looking Backward and Looking Forward," AFIPS National Conference, pp.95-102, 1981.Google Scholar
- 7 R. Grishman and Su Bogong, "An Preliminary Evaluation of Trace Scheduling for Global Microcode Compaction," Dept. of Computer Science, Courant Inst. of Math. Science, NYU, Report No.042, May 1982.Google Scholar
- 8 R. Grishman and Su Bogong, "An Preliminary Evaluation of Trace Scheduling for Global Microcode Compaction," IEEE Trans. Comput., vol.C-32, pp.1191-1194, Dec. 1983.Google Scholar
Digital Library
- 9 T. R. Gross and J. L. Hennessy, "Optimizing Delayed Branches," In 15th Annual Workshop on Microprogramming, pp.114-120, ACM Special Interest Group on Microprogramming, Oct. 1982. Google Scholar
Digital Library
- 10 J. Lah and E. Atkin, "Tree Compaction of Microprograms," In 16th Annual Workshop on Microprogramming, pp.23-33, ACM Special Interest Group on Microprogramming, Oct. 1983.Google Scholar
- 11 D. Landskov, S. Davidson, B. D. Shriver, and P. W. Mallett, "Local Microcode Compaction Techniques," Computing Survey, pp.261-294, Sept. 1980. Google Scholar
Digital Library
- 12 J. L. Linn, "SRDAG Compaction - A Generalization of Trace Scheduling to Increase the Use of Global Context Information," In 16th Annual Workshop on Microprogramming, pp.11-22, ACM Special Interest Group on Microprogramming, Oct. 1983.Google Scholar
Index Terms
An improvement of trace scheduling for global microcode compaction
Recommendations
An improvement of trace scheduling for global microcode compaction
MICRO 17: Proceedings of the 17th annual workshop on MicroprogrammingFisher's trace scheduling procedure for global compaction has proven to be able to produce significant reduction in execution time of compacted microcode, however extra space may be sometimes required during bookkeeping, and the efficacy of compaction ...
A Preliminary Evaluation of Trace Scheduling for Global Microcode Compaction
Fisher has recently described a new procedure for global microcode compaction which he calls "trace scheduling." We have implemented this procedure and tested it on several microcode sequences. We report in this correspondence on the relative ...
Global microcode compaction under timing constraints
MICRO 21: Proceedings of the 21st annual workshop on Microprogramming and microarchitectureExisting global microcode compaction algorithms have all been based on the assumption that the parallelism exploitation is constrained only by data dependency and resource limitation. However, the timing constraint also has great impact upon microcode ...






Comments