Hypernode reduction modulo scheduling
- 1 November 1995
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
Software pipelining is a loop scheduling technique that extracts parallelism from loops by overlapping the execution of several consecutive iterations. Most prior scheduling research has focused on achieving minimum execution time, without regarding register requirements. Most strategies tend to stretch operand lifetimes because they schedule some operations too early or too late. The paper presents a novel strategy that simultaneously schedules some operations late and other operations early, minimizing all the stretchable dependencies and therefore reducing the registers required by the loop. The key of this strategy is a pre-ordering that selects the order in which the operations will be scheduled. The results show that the method described in this paper performs better than other heuristic methods and almost as well as a linear programming method but requiring much less time to produce the schedules.Keywords
This publication has 15 references indexed in Scilit:
- Minimizing register requirements under resource-constrained rate-optimal software pipeliningPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Non-consistent dual register files to reduce register pressurePublished by Institute of Electrical and Electronics Engineers (IEEE) ,1995
- Optimum modulo schedules for minimum register requirementsPublished by Association for Computing Machinery (ACM) ,1995
- Lifetime-sensitive modulo schedulingPublished by Association for Computing Machinery (ACM) ,1993
- Compiling for the CydraThe Journal of Supercomputing, 1993
- Circular schedulingPublished by Association for Computing Machinery (ACM) ,1991
- Overlapped loop support in the Cydra 5Published by Association for Computing Machinery (ACM) ,1989
- Software pipelining: an effective scheduling technique for VLIW machinesPublished by Association for Computing Machinery (ACM) ,1988
- Conversion of control dependence to data dependencePublished by Association for Computing Machinery (ACM) ,1983
- Some scheduling techniques and an easily schedulable horizontal architecture for high performance scientific computingACM SIGMICRO Newsletter, 1981