Article

Extending LLVM for lightweight SPMD vectorization: using SIMD and vector instructions easily from any language

Authors:

Robin Kruppe,

Julian Oppermann,

Lukas Sommer,

Andreas KochAuthors Info & Claims

CGO 2019: Proceedings of the 2019 IEEE/ACM International Symposium on Code Generation and Optimization

Pages 278 - 279

Published: 16 February 2019 Publication History

Get Access

Abstract

Popular parallel programming environments such as CUDA or OpenMP require considerable compiler support and runtime libraries and are therefore only available for a few programming languages and/or targets. We present an approach to vectorizing kernels written in an existing general-purpose language that requires minimal changes to compiler front-ends. Programmers annotate parallel (SPMD) code regions with a few intrinsic functions, which then guide an ordinary automatic vectorization algorithm. This mechanism allows programming SIMD and vector processors without many of the complexities of more powerful parallel programming environments. Our prototype implementation, based on a custom LLVM pass, is integrated into C, C++ and Rust compilers using only 29--37 lines of frontend-specific code each.

References

[1]

M. Pharr and W. R. Mark, “ispc: A SPMD compiler for high-performance CPU programming,” in Innovative Parallel Computing, 2012.

Crossref

Google Scholar

[2]

R. Karrenberg and S. Hack, “Whole Function Vectorization,” in International Symposium on Code Generation and Optimization, 2011.

Digital Library

Google Scholar

[3]

{Online}. Available: https://github.com/jbush001/NyuziProcessor/

Google Scholar

[4]

{Online}. Available: https://github.com/cdl-saarland/rv

Google Scholar

Recommendations

LLVM Compiler Implementation for Explicit Parallelization and SIMD Vectorization
LLVM-HPC'17: Proceedings of the Fourth Workshop on the LLVM Compiler Infrastructure in HPC

With advances of modern multi-core processors and accelerators, many modern applications are increasingly turning to compiler-assisted parallel and vector programming models such as OpenMP, OpenCL, Halide, Python and TensorFlow. It is crucial to ensure ...
LLVM framework and IR extensions for parallelization, SIMD vectorization and offloading
LLVM-HPC '16: Proceedings of the Third Workshop on LLVM Compiler Infrastructure in HPC

LLVM has become an integral part of the software-development ecosystem for developing advanced compilers, high-performance computing software and tools. This paper presents a small set of LLVM IR extensions for explicitly parallel vector, and offloading ...
SIMD programming using Intel vector extensions
Abstract
Single instruction multiple data (SIMD) extensions are one of the most significant capabilities of recent General Purpose Processors (GPPs) which improves the performance of applications with less hardware modification. Each GPP vendor ...
Highlights
- We provide a review of SIMD technologies in general and Intel’s SIMD in particular.

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

CGO 2019: Proceedings of the 2019 IEEE/ACM International Symposium on Code Generation and Optimization

February 2019

286 pages

ISBN:9781728114361

General Chair:
Mahmut Taylan Kandemir
Penn State University, USA
,
Program Chairs:
Alexandra Jimborean
Uppsala University, USA
,
Tipp Moseley
Google, USA

Publisher

IEEE Press

Publication History

Published: 16 February 2019

Check for updates

Qualifiers

Article

Acceptance Rates

Overall Acceptance Rate 312 of 1,061 submissions, 29%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

0
Total Citations
140
Total Downloads

Downloads (Last 12 months)9
Downloads (Last 6 weeks)0

Reflects downloads up to 12 Dec 2024

Other Metrics

View Author Metrics

Citations

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Recommendations

LLVM Compiler Implementation for Explicit Parallelization and SIMD Vectorization

LLVM framework and IR extensions for parallelization, SIMD vectorization and offloading

SIMD programming using Intel vector extensions