The I Test: An Improved Dependence Test for Automatic Parallelization and Vectorization

Xiangyun Kong, Kleanthis Psarris, David Klappholz

Research output: Contribution to journalArticlepeer-review

82 Scopus citations

Abstract

The GCD and Baneijee tests are the standard subscript dependence tests used to determine whether loops may be parallelized/vectorized. The present work discusses the I Test, a subscript dependence test which extends both the range of applicability and the accuracy of these tests. It promises to be especially useful when, in the event that a positive result must be reported, a definite positive is of more use than a tentative positive, and when insufficient loop iteration limits are known for the Banerjee test to apply.

Original languageEnglish
Pages (from-to)342-349
Number of pages8
JournalIEEE Transactions on Parallel and Distributed Systems
Volume2
Issue number3
DOIs
StatePublished - Jul 1991

Fingerprint

Dive into the research topics of 'The I Test: An Improved Dependence Test for Automatic Parallelization and Vectorization'. Together they form a unique fingerprint.

Cite this