SIMDGiraffe: Visualizing SIMD Functions

P. Ntang, D. Lemire

Abstract

Many common processors offer advanced parallel-processing features to accelerate computations. In particular, most commodity processors support Single Instruction on Multiple Data (SIMD) instructions. Algorithms designed to benefit from these instructions can be several times faster than conventional algorithms. However, they can be difficult to understand, and therefore to review. We build SIMDGiraffe, a tool that can help visualize SIMD code written using the popular Intel intrinsics in C.

Download


Paper Citation


in Harvard Style

Ntang P. and Lemire D. (2021). SIMDGiraffe: Visualizing SIMD Functions.In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: IVAPP, ISBN 978-989-758-488-6, pages 147-154. DOI: 10.5220/0010195201470154


in Bibtex Style

@conference{ivapp21,
author={P. Ntang and D. Lemire},
title={SIMDGiraffe: Visualizing SIMD Functions},
booktitle={Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: IVAPP,},
year={2021},
pages={147-154},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010195201470154},
isbn={978-989-758-488-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: IVAPP,
TI - SIMDGiraffe: Visualizing SIMD Functions
SN - 978-989-758-488-6
AU - Ntang P.
AU - Lemire D.
PY - 2021
SP - 147
EP - 154
DO - 10.5220/0010195201470154