Extending the i test to direction vectors

Kleanthis Psarris, Xiangyun Kong, David Klappholz

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

The GCD and Banerjee-Wolfe tests are the standard data dependence tests used to determine whether a loop may be parallelized/vectorized. In [14] we presented a new data dependence test, the I-test, which extends both the applicability and the accuracy of the GCD and the Banerjee-Wolfe tests. In the original presentation, only the case of general dependence was considered, i.e., the case of dependence with a direction vector of the form (∗, ∗,..., ∗). In the present work we generalize the I-test to check for data dependence subject to an arbitrary direction vector.

Original languageEnglish
Title of host publicationICS 1991 - Proceedings of the 5th International Conference on Supercomputing
EditorsEdward S. Davidson, Friedel Hossfield
Pages330-340
Number of pages11
DOIs
StatePublished - 1 Jun 1991
Event5th International Conference on Supercomputing, ICS 1991 - Cologne, Germany
Duration: 17 Jun 199121 Jun 1991

Publication series

NameProceedings of the International Conference on Supercomputing

Conference

Conference5th International Conference on Supercomputing, ICS 1991
Country/TerritoryGermany
CityCologne
Period17/06/9121/06/91

Cite this