<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic MKL library: No vectorization in Software Tuning, Performance Optimization &amp; Platform Monitoring</title>
    <link>https://community.intel.com/t5/Software-Tuning-Performance/MKL-library-No-vectorization/m-p/1108010#M5980</link>
    <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;I am studying different SIMD extensions on convolution network using MKL. I was able to set different SIMD extension using&amp;nbsp;MKL_CBWR. I am wondering is it possible to direct the MKL to use no-vectorization code. I want to quantify the effect of using SIMD extensions against no vectorization.&lt;/P&gt;</description>
    <pubDate>Wed, 27 Apr 2016 15:52:11 GMT</pubDate>
    <dc:creator>Animesh_J_</dc:creator>
    <dc:date>2016-04-27T15:52:11Z</dc:date>
    <item>
      <title>MKL library: No vectorization</title>
      <link>https://community.intel.com/t5/Software-Tuning-Performance/MKL-library-No-vectorization/m-p/1108010#M5980</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;I am studying different SIMD extensions on convolution network using MKL. I was able to set different SIMD extension using&amp;nbsp;MKL_CBWR. I am wondering is it possible to direct the MKL to use no-vectorization code. I want to quantify the effect of using SIMD extensions against no vectorization.&lt;/P&gt;</description>
      <pubDate>Wed, 27 Apr 2016 15:52:11 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Tuning-Performance/MKL-library-No-vectorization/m-p/1108010#M5980</guid>
      <dc:creator>Animesh_J_</dc:creator>
      <dc:date>2016-04-27T15:52:11Z</dc:date>
    </item>
    <item>
      <title>For that purpose (if you want</title>
      <link>https://community.intel.com/t5/Software-Tuning-Performance/MKL-library-No-vectorization/m-p/1108011#M5981</link>
      <description>&lt;P&gt;For that purpose (if you want single thread comparison), you might build public source with no vectorization set for your compiler. You will need to decide about unroll flags. &amp;nbsp;For example, non vector unroll by 4 might be relevant.&lt;/P&gt;</description>
      <pubDate>Wed, 27 Apr 2016 18:34:40 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Tuning-Performance/MKL-library-No-vectorization/m-p/1108011#M5981</guid>
      <dc:creator>TimP</dc:creator>
      <dc:date>2016-04-27T18:34:40Z</dc:date>
    </item>
  </channel>
</rss>

