An efficient data dependence analysis for parallelizing compilers

Journal Article · Mon Jan 01 00:00:00 EST 1990 · IEEE Transactions on Parallel and Distributed Systems; (USA)

DOI:https://doi.org/10.1109/71.80122· OSTI ID:6540901

Li, Z ^[1]; Yew, P C ^[2]; Zhu, C Q ^[3]

---

1. Dept. of Computer Science, York Univ., North York, Ontario M3J 1P3 (CA)
2. Center for Supercomputing Research and Development, Univ. of Illinois at Urbana-Champaign, Urbana, IL (US)
3. Fudan Univ., Shanghai (CN)

An efficient and precise data dependence analysis is the key to the success of a parallelizing compiler because it is required in almost all phases of the parallelism detection and enhancement in such compilers. However, existing test algorithms are quite weak in analyzing multidimensional array references which are usually where the parallelism is in most programs. There are basically two approaches in dependence testing. The first approach is based on numerical methods. This approach tests array references dimension by dimension, thus it may lose accuracy when references have coupled subscripts. The second approach is based on checking the consistency of a set of inequalities. It tests multiple dimensions simultaneously, and hence can really enhance the accuracy. However, this approach is very time consuming. In this paper, a new algorithm is presented for an efficient and accurate data dependence analysis of multidimensional array references.

• OSTI ID: 6540901
• Journal Information: IEEE Transactions on Parallel and Distributed Systems; (USA), Vol. 1:1; ISSN 1045-9219
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
PARALLEL PROCESSING
PERFORMANCE
ACCURACY
ALGORITHMS
ARRAY PROCESSORS
EFFICIENCY
TECHNOLOGY ASSESSMENT
MATHEMATICAL LOGIC
PROGRAMMING
990200* - Mathematics & Computers