Software Pipelining for Coarse-Grained Reconfigurable Instruction Set Processors

ABSTRACT

This paper shows that software pipelining can be an effective technique for code generation for coarse-grained reconfigurable instruction set processors. The paper describes a technique, based on adding an operation assignment phase to software pipelining, that performs reconfigurable instruction generation and instruction scheduling on a combined algorithm. Although typical compilers for reconfigurable processors perform these steps separately, results show that the combination enables a successful usage of the reconfigurable resources. The assignment algorithm is the key for using software pipelining on the reconfigurable processor.The technique presented is also able to exploit spatial computation inside the reconfigurable functional unit by which the output of a processing element is directly connected to the input of another processing element without the need of an intermediate register. Results show that it is possible to reduce the cycle count by using this spatial computation.

REFERENCES

• 1: Margarida F. Jacome , Gustavo de Veciana, Design Challenges for New Application-Specific Processors, IEEE Design & Test, v.17 n.2, p.40-50, April 2000 [doi>10.1109/54.844333 ]
• 2: Francisco Barat , Rudy Lauwereins, Reconfigurable Instruction Set Processors: A Survey, Proceedings of the 11th IEEE International Workshop on Rapid System Prototyping (RSP 2000), p.168, June 21-23, 2000
• 3: André DeHon, The Density Advantage of Configurable Computing, Computer, v.33 n.4, p.41-49, April 2000 [doi>10.1109/2.839320 ]
• 4: Cesare Alippi , William Fornaciari , Laura Pozzi , Mariagiovanna Sami, A DAG-based design approach for reconfigurable VLIW processors, Proceedings of the conference on Design, automation and test in Europe, p.57-es, January 1999, Munich, Germany [doi>10.1145/307418.307504]
• 5: Timothy J. Callahan , John Wawrzynek, Instruction-Level Parallelism for Reconfigurable Computing, Proceedings of the 8th International Workshop on Field-Programmable Logic and Applications, From FPGAs to Computing Paradigm, p.248-257, August 31-September 03, 1998
• 6: Mihai Budiu , Seth Copen Goldstein, Fast compilation for pipelined reconfigurable fabrics, Proceedings of the 1999 ACM/SIGDA seventh international symposium on Field programmable gate arrays, p.195-205, February 21-23, 1999, Monterey, California, United States [doi>10.1145/296399.296459]
• 7: Pieter Op de Beeck , Francisco Barat , Murali Jayapala , Rudy Lauwereins, CRISP: A Template for Reconfigurable Instruction Set Processors, Proceedings of the 11th International Conference on Field-Programmable Logic and Applications, p.296-305, August 27-29, 2001
• 8: Vicki H. Allan , Reese B. Jones , Randall M. Lee , Stephen J. Allan, Software pipelining, ACM Computing Surveys (CSUR), v.27 n.3, p.367-432, Sept. 1995 [doi>10.1145/212094.212131]
• 9: B. Ramakrishna Rau, Iterative modulo scheduling: an algorithm for software pipelining loops, Proceedings of the 27th annual international symposium on Microarchitecture, p.63-74, November 30-December 02, 1994, San Jose, California, United States [doi>10.1145/192724.192731]

INDEX TERMS

Keywords:
reconfigurable processor, code generation, coarse grained logic, software pipelining, vliw, spatial computation