Abstract
In this paper we present an intermediate program representation, called a program dependence graph or PDG, which summarizes not only the data dependences of each operation but also summarizes the control dependences of the operations. Data dependences represent only the relevant data flow relationships of the program. Analagously, control dependences represent only the relevant control flow relationships of the program, in contrast to the usual control flow graph. The PDG allows transformations such as vectorization, which previously required special treatuent of control dependence, to be performed in a manner which is uniform for both control and data dependences. Program transformations which require interaction of the two can also be easily handled by the representation. As an example, a new incremental approach to modifying data dependences resulting from branch deletion is introduced. Another value of our representation is that many traditional optimizations operate more efficiently on the PDG. Since dependences in the PDG connect computationally relevant parts of the program, a single walk of these dependences is sufficient to perform many optimizations.
Karl Ottenstein was supported in part by the National Science Foundation under Grant No. MCS-8203487.
Preview
Unable to display preview. Download preview PDF.
References
Allen, Frances E. and Cocke, John. A catalogue of optimizing transformations in Design and Optimization of Compilers (Randall Rustin, Ed.) Prentice-Hall (1972) 1–30.
Allen, J.R., Kennedy, Ken, Porterfield, Carrie, and Warren, Joe. Conversion of control dependence to data dependence. 10th Ann. ACM Symp. on Princ. of Prog. Lang. Austin, Texas (January, 1983) 177–189.
Arvind, Gostelow, Kim, and Plouffe, Wil. An asynchronous programming language and computing machine. University of California at Irvine Tech Report TR114A (December, 1978)
Dennis, Jack B. Data flow supercomputers. IEEE Computer 13, 11 (Nov. 1980) 48–56.
Ferrante, Jeanne, and Ottenstein, Karl. A program form based on data dependency in predicate regions. 10th Ann. ACM Symp. on Princ. of Prog. Lang. Austin, Texas (January, 1983) 217–231.
Ferrante, Jeanne, Ottenstein, Karl, and Warren, Joe D. The program dependence graph and its use in optimization. In preparation.
Graham, Susan and Wegman, Mark. A fast and usually linear algorithm for global flow analysis. JACM 231 (January, 1976) 172–202.
Kas'janov, V.N. Distinguishing hammocks in a directed graph. Soviet Math. Doklady 165 (1975) 448–450.
Kuck, D. J; Kuhn, R. H., Padua, D. A.; Leasure, B. and Wolfe, M. Dependence graphs and compiler optimizations. 8th Ann. ACM Symp. on Princ. of Prog. Lang. Williamsburg, VA (Jan. 26–28, 1981) 207–218.
Ottenstein, Karl J. Data-flow graphs as an intermediate program form. Ph.D. Thesis. Computer Sciences Dept., Purdue Univ. (August 1978) 283 pages.
Padua, David A., Kuck, David J., and Lawrie, Duncan. High-Speed multiprocessors and their compilers. IEEE Transactions on Computers 29,9 (September, 1980) 763–776.
Padua Haiek, David Alejan. Multiprocessors: Discussion of some theoretical and practical problems. Ph.D. Thesis. Computer Sciences Dept., University of Illinois (1980).
Ryder, Barbara G. Incremental data flow analysis. 10th Ann. ACM Symp. on Princ. of Prog. Lang. Austin, Texas (January, 1983) 167–176.
Warren, Joe. A hierarchical basis for reordering transformations. 11th Ann. ACM Symp. on Princ. of Prog. Lang. Salt Lake City, Utah (January, 1984), 272–282.
Wegman, Mark. Summarizing graphs by regular expressions. 10th Ann. ACM Symp. on Princ. of Prog. Lang. Austin, Texas (January, 1983) 203–212.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1984 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ferrante, J., Ottenstein, K.J., Warren, J.D. (1984). The program dependence graph and its use in optimization. In: Paul, M., Robinet, B. (eds) International Symposium on Programming. Programming 1984. Lecture Notes in Computer Science, vol 167. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-12925-1_33
Download citation
DOI: https://doi.org/10.1007/3-540-12925-1_33
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-12925-7
Online ISBN: 978-3-540-38809-8
eBook Packages: Springer Book Archive