搜索

搜索结果:

搜索结果: 1 010

  1. ISA 扩展、 英特尔AVX | 英特尔® 开发人员专区

    https://software.intel.com/zh-cn/isa-extensions/intel-avx

    英特尔® AVX 是一个针对英特尔® SIMD 流指令扩展的全新256 位扩展指令集,专为 浮点密集型应用而设计。 它于2011 年年初作为第二代英特尔® 酷睿™ 处理器 ...

  2. AVX horizontal sum and store

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/346695

    The next thing is how to store only, one value of an AVX register. In SSE I just used _mm_store_sd. I could do a cast from __m256d to __m128d and use SSE ...

  3. AVX Optimizations and Performance: VisualStudio vs GCC

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/475462

    Greetings, I have recently written some code using AVX function calls to perform a convolution in my software. I have compiled and run this code on two ...

  4. AVX - Vector shifts

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/343530

    I'm comparing two programmes, one is written using SSE and the other one AVX. My aim is to show that the avx version is running 2 times faster but I'm loosing ...

  5. AVX in Sandy Bridge

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/294209

    A few new public information about the 1st AVX incarnation in Sandy Bridge in the SF09_ARCS002_FIN.pdf document available here ...

  6. AVX

    https://software.intel.com/zh-cn/forums/intel-performance-bottleneck-analyzer/topic/326650

    TITLE: AVX ISSUE_NAME: AVX Finder. ... With AVX, the width of the SIMD register file is increased from 128 bits to 256 bits and renamed from XMM0- XMM15 to ...

  7. Detecting AVX

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/282242

    The program says that this processor DOESN'T supports AVX. When I check the result of cat /proc/cpuinfo in linux, avx is missing from the flags. What is the ...

  8. Do the Intel 11.1 compilers support AVX?

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/295887

    Title says it all. Just wondering if the 11.1 version of the intel C and Fortran compilers generate AVX code? Can someone verify this? Thanks for any assistance.

  9. AVX: _mm_alignr_epi8 equivalent for YMM registers

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/283576

    Hello,I am looking for an equivalent to the most useful intrinsic_mm_alignr_epi8 with AVX registers (I guess its equivalent to PALIGNR or VPALIGNR for the one ...

  10. Welcome to the Intel(R) AVX Forum!

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/302778

    Please take a moment to read the papers and download the guide from the Intel AVX web site. If you have any questions about Intel AVX, AES, or SSE4.2 please  ...

  11. Performance Issues on AVX Instructions

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/280799

    In my understanding vzeroupper is only for mixing SSE and AVX. The biggest ... I' m running a billion AVX instructions and every one of these is delayed! The test ...

  12. PDEP/PEXT operations for AVX

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/370419

    Hi, I was wondering if Intel engineers would consider adding certain BMI instructions to AVX. In particular, I think vector variants of PDEP and PEXT would be of ...

  13. Disabling AVX

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/508413

    Hi all, Is there a way (under Windows 7) to disable the support of AVX. I wan't to make sure that on a pre-SB machine I don't get "Illegal Instruction exception".

  14. AVX sometimes slower than SSE

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/283753

    Has anyone experienced a slow down by a factor of around 2 for certain functions that are converted from SSE to AVX-128?My setup:Intel Compiler icc ...

  15. 如何检测Knights Landing AVX-512 支持(英特尔至强融核处理器 ...

    https://software.intel.com/zh-cn/articles/how-to-detect-knl-instruction-support

    2016年2月22日 ... 英特尔至强融核处理器(代号“Knights Landing”)是第二代英特尔至强融核产品的 一部分。Knights Landing 支持AVX-512 指令,特别是AVX-512F ...

  16. AVX thread safety problem

    https://software.intel.com/zh-cn/forums/intel-math-kernel-library/topic/284912

    Seems like there's some sort of thread safety problem in the AVX code inside MKL. ... The AVX kernels in MKL require 32-byte alignment to obtain consistent ...

  17. AVX Performance Reduction Vs 128 SIMD

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/282731

    Dear All, Legacy 128 Bit SIMD code replaced with AVX code, expected a performance gain or atleast status quo. But the performance has fallen to 10% of the ...

  18. Processing of data in SSE/AVX/AVX2

    https://software.intel.com/zh-cn/forums/intel-isa-extensions/topic/534686

    I expect AVX2 (with FMA) to be faster with 32 Byte datas than AVX, but Im not sure if AVX and AVX2 will be faster with 16 Byte datas vs SSE. @bronxzv - Im not  ...

  19. 在不编写AVX 代码的情况下使用AVX | 英特尔® 开发人员专区

    https://software.intel.com/zh-cn/articles/using-avx-without-writing-avx-code

    2012年3月7日 ... Using AVX Without Writing AVX Code (PDF 260KB) 摘要英特尔® 高级矢量扩展 指令集(AVX)是一套针对英特尔® SIMD 流指令扩展(英特尔® ...

  20. Converging AVX and LRBni

    https://software.intel.com/zh-cn/comment/1492124

    Obviously power consumption is a big issue, but since AVX is specified to process up to 1024-bit registers, it could execute such wide operations using SNB's ...

有关编译器优化的更完整信息,请参阅优化通知