Boots – shoes – and leggings
Patent
1994-01-25
1996-09-17
Harrell, Robert B.
Boots, shoes, and leggings
364DIG1, 364DIG2, 364280, 3642804, 3642805, 364973, G06F 945
Patent
active
055577610
ABSTRACT:
A system and method of generating object code from an intermediate representation of source code is described. The intermediate representation includes a plurality of basic blocks each being represented by a plurality dam dependency graphs, wherein each data dependency graph comprises a plurality of nodes each corresponding to an instruction from the target computer instruction set. The present invention operates by selecting a source basic block (that is one of the basic blocks of the intermediate representation) and a target basic block (that is another of the basic blocks of the intermediate representation), and by identifying a maximal set of instructions contained in the source basic block that are movable from the source basic block to the target basic block without violating any data dependency relationships of the data dependency graphs. An overall cost model of aggregately moving instructions of the maximal set from the source basic block to the target basic block is generated. This cost model specifies an executable cost of moving each of the instructions of the maximal set from the source basic block to the target basic block. Then, the present invention aggregately moves one or more instructions of the maximal set from the source basic block to the target basic block according to the cost model to form the object code.
REFERENCES:
Ford, Jr., L. R. and D. R. Fulkerson, "Static Maximal Flow", Flows in Networks, Publisher: Princeton University Press, pp. 1-22, 1962.
Fisher, Joseph A., "Trace Scheduling: A Technical for Global Microcode Compaction", IEEE Transactions on Computers, vol. c-30, No. 7, pp. 478-490, Jul. 1981.
Charlesworth, Alan E., "An Approach to Scientific Array Processing: The Architectural Design of the AP-120B/FPS-164 Family", Computer, pp. 18-27, 1981.
Aho, Alfred V., Ravi Sethi and Jeffery D. Ullman, Compilers: Principles, Techniques, and Tools, Publisher: Addison-Wesley Publishing Company, pp. 10-23, 1986.
Ellis, John R., Bulldog: A Compiler for VLIW Architectures, Publisher: The Massachusettes Institute of Technology, 1986.
Nicolau, Alexandru, "A Fine-Grain Parallelizing Compiler", Scheduling, Computer Science Department, Cornell University, 1986.
Ebcioglu, Kemal and Nicolau Alexandru, "A global resource-constrained parallelization technique", Scheduling, pp. 154-163, 1989.
James C. Dehnert et al., "Compiling for the Cydra 5", The Journal of Supercomputing, 7, pp. 181-227 (1993).
Bernstein, David and Michael Rodeh, "Global Instruction Scheduling for Superscalar Machines", Proceedings of the ACM SIGPLAN '91 Conference on Programming Language Design and Implementation, Toronto, Canada, Jun. 26-28, 1991, pp. 241-255, 1991.
Lowney, P. Geoffrey et al., "The Multiflow Trace Scheduling Compiler", The Journal of Supercomputing, vol. 7, pp. 51-55, 1993.
Chan Sun C.
Dehnert James C.
Lo Raymond W.
Towle Ross A.
Harrell Robert B.
Silicon Graphics Inc.
LandOfFree
System and method of generating object code using aggregate inst does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method of generating object code using aggregate inst, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method of generating object code using aggregate inst will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-421730