sf上的SIMD的程序库

与SIMD相关的大概有这样一些：

Vector3D SSE  Updated 2011-01-09
A C++ header library for fast operations on vectors/matrices (3D/3x3) using Streaming SIMD Extensions (SSE, SSE2, SSE3, SSE4); Tends to be used in 3D graphics applications and game developement.
http://sourceforge.net/projects/v3d/
http://nchc.dl.sourceforge.net/project/v3d/v3d-0.1b_r2.tar.gz

parallel for  Updated 2008-05-29
A data parallel scientific programming model. Compiles efficiently to different platforms like distributed memory (MPI), shared memory multi-processor (pthreads), Cell BE processor, Nvidia Cuda, SIMD vectorization (SSE, Altivec), and sequential C++ code.
http://sourceforge.net/projects/parallel-for/

SSEPlus  Updated 2008-05-21
SSEPlus is a SIMD function library. It provides optimized emulation for newer SSE instructions. It also provides a rich set of high performance routines for common operations such as arithmetic, bitwise logic, and data packing and unpacking.
http://sourceforge.net/projects/sseplus/
http://nchc.dl.sourceforge.net/project/sseplus/SSEPlus/0.3.67/SSEPlus_0.3.67.zip

(这个库居然是AMD开发的，有丰富的文档。看看，Intel，向人家AMD好好学学！)

GENIAL  Updated 2008-04-01
GENIAL is a C++ library for signal and image processing. It uses template-generic techniques, multi threading, cache optimization and SIMD instructions for Pentium (MMX, SSE, SSE2, SSE3) to achieve high performance: FFT,DCT,Convolution,Linear Algebra...
http://sourceforge.net/projects/genial/
http://ncu.dl.sourceforge.net/project/genial/genial/genial-2.1.0/genial-2.1.0.zip

(使用了一下，但是因为没有下载最新的GCC及其头文件中的xmmintrin.h等文件导致无法编译通过。有很详细的文档，简单看了下，提供了各种高度简化的向量计算函数)

SIMDx86  Updated 2006-10-31
This library is meant for high performance calculations for science or 3D games/rasterizers using SIMD instructions of x86 processors to allow an unparalleled level of optimization. This takes advantage of MMX, 3DNow!, 3DNow!+/MMX+, & SSE/SSE2/SSE3/SSSE3
http://sourceforge.net/projects/simdx86/
http://nchc.dl.sourceforge.net/project/simdx86/SIMDx86-src/libSIMDx86-devel-src-0.4.0.zip

NVec  Updated 2005-09-28
NVec is a C++ Library that provides SIMD accelerated functions without the user needing to know that SIMD functionality is available. Currently supports x86 and x86_64 SSE/SSE2/SSE3, for Linux and Windows (mingw) Requires a recent version of gcc. (~3.4)
http://sourceforge.net/projects/nvec/
http://ncu.dl.sourceforge.net/project/nvec/nvec/alpha-20050928/nvec.zip

libSIMD  Updated 2005-04-05
Mathematical library utilising SIMD features of common processors to accelerate many commonly-used algorithms where compilers fear to tread.
http://sourceforge.net/projects/libsimd/
http://nchc.dl.sourceforge.net/project/libsimd/libsimd/20050405/libSIMD-20050405.tar.gz

Cross-platform SIMD C Headers  Updated 2004-10-27
A cross-platform, cross-compiler, cross-CPU C header library for programming with SIMD instruction sets. X86 (MMX/SSE/SSE2) GCC and MSVC, PPC Altivec GCC, WMMX ARM GCC, and software emulated SIMD are supported.
http://sourceforge.net/projects/simd-cph/
http://nchc.dl.sourceforge.net/project/simd-cph/simd-cph/2004.10.26/simd_cph_20041026.tgz

sf上的SIMD程序库