Abstract
Software pipelining has become a very extended loop execution model since microprocessors include more functional units. Scheduling techniques developed for software pipelining have to look for heuristics to obtain optimal schedulings. In this paper we present a new software pipelining approach based on a methodology that divides the scheduling problem in two phases. The first phase is the subject of this paper and holds the NP-completeness of the problem. It consists on including new edges on the dependence graph which model the scheduling. The second phase takes such edges and generates the parallel code. Results are given for dependence graphs of real innermost loops which show that more than 98% of the times an optimal solution is found.
Chapter PDF
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Rau B.R. and Fisher J.A.“ Instruction Level Parallel Processing: History, Overview and Perspective” Journal of Superccomputing 7. pp.9–50, May. 1993.
Barrado C., Labarta J, Ayguadé E. and Valero M. “Automatic Generation of Loop Scheduling for VLIW”.PACT'95, Limassol (Cyprus), pp.306–309, Jun-95. Extended in 5th Workshop on Compilers for Parallel Computing, Málaga, pp. 485–502, Jun-1995.
Näher S. “The LEDA User Manual. Version 3.1”. Tech. Report Fachbereich Informatik, Universität des Saarlandes, Saarbrücken, Germany, 1995.
Govindarajan R. Altman E.R. and Gao G.R. “Minimizing Register Requirements under Resource-Constrained Rate-Optional Software Pipelining”. Proc. of the 27th Ann. Symp. on Microarchitecture. pp.85–94, San Jose, California, Dec. 1994.
Lam M. “Software Pipelining: An effective scheduling technique for VLIW machines” Proc. of the ACM SIGPLAN 88 Conf. on Programming Languages Dessign and Implementation. pp.318–327. 1988.
Llosa J., Valero M., Ayguadé E and González A.. “Hypernode Reduction Modulo Scheduling”. Proc. of the 28th Ann. Symp. on Microarchitecture. pp.350–360. Dec. 1995.
Huff R.A. “Lifetime-Sensitive Modulo Scheduling”. ACM SIGPLAN 6th Conf. on Programming Languages, Design and Implementation, 6'PLDI. Albuquerque, pp.258–267, 1993.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Barrado, C., Labarta, J. (1997). Hamiltonian recurrence for ILP. In: Lengauer, C., Griebl, M., Gorlatch, S. (eds) Euro-Par'97 Parallel Processing. Euro-Par 1997. Lecture Notes in Computer Science, vol 1300. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0002756
Download citation
DOI: https://doi.org/10.1007/BFb0002756
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63440-9
Online ISBN: 978-3-540-69549-3
eBook Packages: Springer Book Archive