Abstract
This paper presents a new static data dependence analysis approach, Dependence Difference Inequality Test, which can deal with coupled subscripts for multi-dimensional array references for software pipelining techniques for nested loops. The Dependence Difference Inequality Test (DDIT) replaces direction vectors with dependence difference inequalities as constraints to variables in a linear system. The method presented in this paper extends the applicable range of the Generalized Lambda Test and seems to be a practical scheme to analyze data dependence. Experimental results show that the number of data independences checked by the DDIT algorithm is slightly smaller than that manually. It is also shown that our method is better than other traditional data dependence analysis methods without increasing time cost: it increases the success rate of the Generalized Lambda Test by approximately 14.19%.
Chapter PDF
Similar content being viewed by others
References
Petersen, P.M., Padua, D.A.: Static and Dynamic Evaluation of Data Dependence Analysis Techniques. IEEE Transactions on Parallel and Distributed Systems 7, 1121–1132 (1996)
Chang, W.L., Chu, C.P., Wu, J.: The Generalized Lambda Test: A Multi-Dimensional Version of Banerjee’s Algorithm. International Journal of Parallel and Distributed Systems and Networks 2, 69–78 (1999)
Qiao, L., Huang, W.T., Tang, Z.Z.: A Static Data Dependence Analysis Approach for Software Pipelining. In: Jin, H., Reed, D., Jiang, W. (eds.) NPC 2005. LNCS, vol. 3779, pp. 213–220. Springer, Heidelberg (2005); Accepted by NPC 2005
Li, Z., Yew, Y.C., Zhu, C.Q.: An Efficient Data Dependence Analysis for Parallelizing Compilers. IEEE Transactions on Parallel and Distributed System 1, 26–34 (1990)
Qiao, L.: On Data Dependencies in Software Pipelining. Doctorial Dissertation, Department of Computer Science. Tsinghua University, Beijing (2001)
Johnson, S.P., Cross, M., Everett, M.G.: Exploitation of Symbolic Information in Interprocedural Dependence Analysis. Parallel Computing 22, 197–226 (1996)
Qiao, L., Huang, W.T., Tang, Z.Z.: A Dynamic Data Dependence Analysis Approach for Software Pipelining. In: Jin, H., Reed, D., Jiang, W. (eds.) NPC 2005. LNCS, vol. 3779, pp. 221–228. Springer, Heidelberg (2005); Accepted by NPC 2005
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 IFIP International Federation for Information Processing
About this paper
Cite this paper
Qiao, L., Huang, W., Tang, Z. (2005). Coping with Data Dependencies of Multi-dimensional Array References. In: Jin, H., Reed, D., Jiang, W. (eds) Network and Parallel Computing. NPC 2005. Lecture Notes in Computer Science, vol 3779. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11577188_40
Download citation
DOI: https://doi.org/10.1007/11577188_40
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-29810-6
Online ISBN: 978-3-540-32246-7
eBook Packages: Computer ScienceComputer Science (R0)