logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

Efficient utilization of SIMD extensions

delete2005-02-01
delete47
delete
OA
AI
F
Franz Franchetti
Š
Štefan Král
J
Juergen Lorenz
C
Christoph W. Ueberhuber
DOI：10.1109/JPROC.2004.840491delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
This paper targets automatic perfomance tuning of numerical kernels in the presence of multilayered memory hierarchies and single-instruction, multiple-data (SIMD) parallelisin. The studied SIMD instruction set extensions include Intel's SSE family, AMD's 3DNow!, Motorola's AltiVec, and IBM's BlueGene/L SIMD instructions. FFTW ATLAS, and SPIRAL demonstrate that near-optimal performance of numerical kernels across a variety of modern computers featuring deep memory hierarchies can be achieved only by means of automatic performance tuning. These software pack-ages generate and optimize ANSI C code and feed it into the target machine's general-purpose C compiler to maintain portability. The scalar C code produced by performance tuning systems poses a severe challenge for vectorizing compilers. The particular code structure hampers automatic vectorization and, thus, inhibits satisfactory performance on processors featuring short vector extensions. this paper describes special-purpose compiler technology that supports automatic performance tuning on machines with vector instructions. The work described includes; 1) symbolic vectorization of digital signal processing transforms; 2) straight-line code vectorization for numerical kernels; and 3) compiler back ends for straight-line code with vector,instructions. Methods from all three areas were combined with FFTW. SPIRAL, and ATLAS to optimize both for memory hierarchy and vector instructions. Experiments show that the presented methods lead to substantial speedups (up to 1.8 for two-way and 3.3 for four-way vector extensions) over the best scalar C codes generated by the original systems as well as roughly matching the performance of hand-tuned vendor libraries.
Keywords:
autontatic vectorization
digital signal processing (DSP)
fast Fourier transform (FFT)
short vector single instruction
multiple data (SIMD)
symbolic vectorization

Journal

Proceedings of the IEEE cover
Proceedings of the IEEE
IF:
25.9
Papers:
9.9K
Citations:
4.5W

Organization

No organization information available