<?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 Hi Elad, in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125147#M25192</link>
    <description>&lt;P&gt;Hi Elad,&lt;/P&gt;&lt;P&gt;You can call mkl_jit_create_cgemm once and use it in all calling threads.&lt;/P&gt;&lt;P&gt;Since generating the jit kernel using this function will&amp;nbsp;involve runtime overhead, you will see performance improve if this kernel gets called several hundred&amp;nbsp;of times.&lt;/P&gt;&lt;P&gt;Best regards,&lt;/P&gt;&lt;P&gt;Khang&lt;/P&gt;</description>
    <pubDate>Wed, 10 Jul 2019 17:43:02 GMT</pubDate>
    <dc:creator>Khang_N_Intel</dc:creator>
    <dc:date>2019-07-10T17:43:02Z</dc:date>
    <item>
      <title>mkl_jit_create_cgemm on multi cores</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125146#M25191</link>
      <description>&lt;P&gt;Hi all,&lt;/P&gt;&lt;P&gt;I want to use&amp;nbsp;mkl_jit_create_cgemm on my setup where each thread is pinned to single core.&lt;/P&gt;&lt;P&gt;In each thread I'll do the&amp;nbsp;cgemm with the created Jitter.&lt;/P&gt;&lt;P&gt;Do I need to create jit kernel specific for each thread? or create just 1 and use it on all calling threads?&lt;/P&gt;&lt;P&gt;Elad&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 10 Jul 2019 13:41:18 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125146#M25191</guid>
      <dc:creator>Yosef__Elad</dc:creator>
      <dc:date>2019-07-10T13:41:18Z</dc:date>
    </item>
    <item>
      <title>Hi Elad,</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125147#M25192</link>
      <description>&lt;P&gt;Hi Elad,&lt;/P&gt;&lt;P&gt;You can call mkl_jit_create_cgemm once and use it in all calling threads.&lt;/P&gt;&lt;P&gt;Since generating the jit kernel using this function will&amp;nbsp;involve runtime overhead, you will see performance improve if this kernel gets called several hundred&amp;nbsp;of times.&lt;/P&gt;&lt;P&gt;Best regards,&lt;/P&gt;&lt;P&gt;Khang&lt;/P&gt;</description>
      <pubDate>Wed, 10 Jul 2019 17:43:02 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125147#M25192</guid>
      <dc:creator>Khang_N_Intel</dc:creator>
      <dc:date>2019-07-10T17:43:02Z</dc:date>
    </item>
    <item>
      <title>What about the func pointer</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125148#M25193</link>
      <description>&lt;P&gt;What about the func pointer that returned by&amp;nbsp;mkl_jit_get_cgemm_ptr ?&lt;/P&gt;&lt;P&gt;It is thread safe?&amp;nbsp;&lt;/P&gt;&lt;P&gt;Can I init single func pointer and use it on all threads?&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 11 Jul 2019 07:31:49 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125148#M25193</guid>
      <dc:creator>Yosef__Elad</dc:creator>
      <dc:date>2019-07-11T07:31:49Z</dc:date>
    </item>
    <item>
      <title>Hi Elad,</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125149#M25194</link>
      <description>&lt;P&gt;Hi Elad,&lt;/P&gt;&lt;P&gt;Yes, it is thread safe and you can use it on all threads.&lt;/P&gt;&lt;P&gt;Khang&lt;/P&gt;</description>
      <pubDate>Thu, 11 Jul 2019 17:57:23 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/mkl-jit-create-cgemm-on-multi-cores/m-p/1125149#M25194</guid>
      <dc:creator>Khang_N_Intel</dc:creator>
      <dc:date>2019-07-11T17:57:23Z</dc:date>
    </item>
  </channel>
</rss>

