搜索资源列表
turbo
- libjpeg-turbo是对libjpeg的扩展,支持SIMD指令,如X86架构的MMX、SSE、SSE2、3DNOW,ARM架构的NEON,在对jpeg进行编码和解码的过程中能提高速度。-libjpeg-turbo is a JPEG image codec that uses SIMD instructions (MMX, SSE2, NEON) to accelerate baseline JPEG compression and decompression on x86, x86-64
sse2
- SSE-2 implementation of RAID-6 syndrome.
Test_SSE
- SSE测试代码,的确可用,同时加入了openmp加速,性能提升有10x-SSE test code, is indeed available, while adding the openmp acceleration, performance improvements have 10x
viterbi27_sse
- viterbi decoder using sse intrinsics
vp9_subpel_variance_impl_intrin_avx2
- final calculation to sum and sse.load source and another source starting the next following byte. -final calculation to sum and sse.load source and another source starting the next following byte.
fpu_control
- FPU control word bits. x86-64 version.Note that this file sets on x86-64 only the x87 FPU, it does not touch the SSE unit.
traps
- Perform dequantization and inverse DCT on one block of coefficients by SSE.
falloc
- Return a string describing the CPU architexture and extensions that Mesa is using (such as SSE or Altivec).
SkBlurImage_opts_SSE2
- Helper function to spread the components of a 32-bit integer into the lower 8 bits of each 32-bit element of an SSE register.
opts_check_x86
- Function to get the CPU SSE-level in runtime, for different compilers.
t_vertex_sse
- Number of bytes to allocate for generated SSE functions.
kobject
- x86-64 optimizations shamelessy converted x86 sse 3dnow assembly by Mikko Tiihonen. -x86-64 optimizations shamelessy converted x86 sse 3dnow assembly by Mikko Tiihonen.
nfnetlink_compat
- Some functions are not written in SSE-assembly, because the fpu ones are faster.
vpx_psnr
- TODO(dkovalev) change vpx sse to psnr signature: double -> int64 t for Linux v2.13.6.
add
- 利用SIMD指令实现四拍并行计算,主要是SSE,SSE2-using SIMD languanges make the calc more fast,more is SSE,SSE2
51308-1297850-scanner-de-malwares-sse-scanner
- code source-code source.................................
CPUIDInfo
- Quering info about CPU features (MMX, SSE, AVX support etc.) by CPUID
vsyasm-1.3.0-win32
- yasm.exe for compile SSE/MMX *.asm files-yasm.exe for compile SSE/MMX*.asm files
hdcp-0.3
- HDCP加密/解密算法,代码通过了HDC规格提供的测试向量,能在2.33GHz Intel Xeon CPU上以180帧/秒加密640x480视频,但它在单核机器上速度不够快,不能实时解密1080p视频,但在多核机器上可以并行解密。代码未来也能进 一步进行优化,例如使用SSE指令。-HDCP encryption/decryption algorithms, code, test vectors through the HDC specifications provided, can the 2
cxmmand
- MMX,SSE命令程序编制事例研究,运用自如MMX Pentium和PentiumIII被扩张了的命令,-MMX and SSE command programming example research, with ease MMX Pentium and PentiumIII be expanded a command,