Automatic SIMD vectorization for Haskell

Petersen, Leaf and Orchard, Dominic A. and Glew, Neal (2013) Automatic SIMD vectorization for Haskell. In: Proceedings of the 18th ACM SIGPLAN international conference on Functional programming. Association for Computing Machinery pp. 25-36. ISBN 978-1-4503-2326-0. (doi:https://doi.org/10.1145/2500365.2500605) (Full text available)

PDF - Publisher pdf
Download (246kB) Preview
[img]
Preview
Official URL
http://doi.acm.org/10.1145/2500365.2500605

Abstract

Expressing algorithms using immutable arrays greatly simplifies the challenges of automatic SIMD vectorization, since several important classes of dependency violations cannot occur. The Haskell programming language provides libraries for programming with immutable arrays, and compiler support for optimizing them to eliminate the overhead of intermediate temporary arrays. We describe an implementation of automatic SIMD vectorization in a Haskell compiler which gives substantial vector speedups for arange of programs written in a natural programming style. We compare performance with that of programs compiled by the Glasgow Haskell Compiler

Item Type: Conference or workshop item (Paper)
Uncontrolled keywords: Vectorization, SIMD, Compiler Optimization, Haskell, Functional Languages
Divisions: Faculties > Sciences > School of Computing
Faculties > Sciences > School of Computing > Programming Languages and Systems Group
Depositing User: Dominic Orchard
Date Deposited: 05 May 2017 13:27 UTC
Last Modified: 08 Jun 2017 08:50 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/57494 (The current URI for this page, for reference purposes)
Orchard, Dominic A.: https://orcid.org/0000-0002-7058-7842
  • Depositors only (login required):

Downloads

Downloads per month over past year