Logo image
Compile?time scheduling of multithread data localities on multiple vector processors
Journal article   Peer reviewed

Compile?time scheduling of multithread data localities on multiple vector processors

Chih‐Yung Chang and Jang‐Ping Sheu
Concurrency: Practice and Experience, Vol.7(5), pp.349-369
1995

Abstract

A large class of loop programs applied in solving differential equations, Fourier transforms, image processing and neural processing can be translated or rewritten into a vector execution form with a ??block dependence graph. In the paper we propose a multithreading strategy to partition such vectorized loops into multithread execution form. Each partitioned thread consists of instances of statements with localities in vector registers. The multithreading scheme gives a novel combination of loop unrolling, statement instances reordering, index shifting, vector register reuse exploiting and multithreading. For some cases of loop program with ??block dependence graph, experimental results show that our scheme assists vector compilers of the Convex C38 series to reduce the number of memory accesses and synchronizations among CPUs. Copyright © 1995 John Wiley & Sons, Ltd

Metrics

1 Record Views

Details

Logo image