<?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 The full message should in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-compilation-issue/m-p/1076908#M22611</link>
    <description>&lt;P&gt;The full message should specify where that reference was set.&amp;nbsp; For example, it might have been in the intel_thread library which requires the libiomp5.&amp;nbsp; In that case, you should either specify -liomp5, or the libgomp thread library along with g++ -fopenmp.&amp;nbsp; libiomp5 handles all the libgomp calls, so you may avoid using libgomp.&lt;/P&gt;

&lt;P&gt;I think if you have the current link advisor, and use g++, you must specify whether you will use libgomp or libiomp5, in order to get the corresponding advice.&lt;/P&gt;</description>
    <pubDate>Fri, 29 Jul 2016 15:02:03 GMT</pubDate>
    <dc:creator>TimP</dc:creator>
    <dc:date>2016-07-29T15:02:03Z</dc:date>
    <item>
      <title>Intel MKL compilation issue</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-compilation-issue/m-p/1076907#M22610</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;I'm experiencing link error "undefined reference to '_intel_fast_memcpy'" while trying to compile my program with g++.&lt;/P&gt;

&lt;P&gt;Link and compile option are set using Intel MKL link line advisor for GCC.&lt;/P&gt;

&lt;P&gt;If compiler is set to icc and match the link and compile option, the program builds and runs without any problem.&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;System specs related to the problems are list below.&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;OS : Ubuntu 14.04 64bit OS&lt;/P&gt;

&lt;P&gt;Compiler : GCC 4.8.4&lt;/P&gt;

&lt;P&gt;MKL : Version comes with Intel Parallel Studio 2016&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Suspecting that this issue might be related setup of parallel studio but couldn't narrow it down.&lt;/P&gt;

&lt;P&gt;Anyone experienced similar problem and solved the issue?&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Regards,&lt;/P&gt;

&lt;P&gt;Ilwoo Park&lt;/P&gt;</description>
      <pubDate>Fri, 29 Jul 2016 12:47:45 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-compilation-issue/m-p/1076907#M22610</guid>
      <dc:creator>Ilwoo_P_</dc:creator>
      <dc:date>2016-07-29T12:47:45Z</dc:date>
    </item>
    <item>
      <title>The full message should</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-compilation-issue/m-p/1076908#M22611</link>
      <description>&lt;P&gt;The full message should specify where that reference was set.&amp;nbsp; For example, it might have been in the intel_thread library which requires the libiomp5.&amp;nbsp; In that case, you should either specify -liomp5, or the libgomp thread library along with g++ -fopenmp.&amp;nbsp; libiomp5 handles all the libgomp calls, so you may avoid using libgomp.&lt;/P&gt;

&lt;P&gt;I think if you have the current link advisor, and use g++, you must specify whether you will use libgomp or libiomp5, in order to get the corresponding advice.&lt;/P&gt;</description>
      <pubDate>Fri, 29 Jul 2016 15:02:03 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-compilation-issue/m-p/1076908#M22611</guid>
      <dc:creator>TimP</dc:creator>
      <dc:date>2016-07-29T15:02:03Z</dc:date>
    </item>
    <item>
      <title>Hi </title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-compilation-issue/m-p/1076909#M22612</link>
      <description>&lt;P&gt;Hi&amp;nbsp;&lt;/P&gt;

&lt;P&gt;right, as Tim said, you may use either -fopenmp (libgomp) &amp;nbsp;or intel libiomp5.so.&amp;nbsp;&lt;/P&gt;

&lt;P&gt;for example, using G++ and GNU openmp thread.&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size: 1em; line-height: 1.5;"&gt;-Wl,--start-group ${MKLROOT}/lib/intel64/libmkl_intel_lp64.a ${MKLROOT}/lib/intel64/libmkl_core.a ${MKLROOT}/lib/intel64/libmkl_gnu_thread.a -Wl,--end-group -lpthread -lm -ldl&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 1em; line-height: 1.5;"&gt;&amp;nbsp;-fopenmp -m64 -I${MKLROOT}/include&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;The reference&amp;nbsp;&lt;SPAN style="font-size: 12px; line-height: 18px;"&gt;'_intel_fast_memcpy'" should be in libiomp5.so.&amp;nbsp;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;nm libiomp5* | grep "_intel_fast_memcpy"&lt;BR /&gt;
	0000000000000040 T __kmp_external__intel_fast_memcpy&lt;BR /&gt;
	0000000000000000 T __kmp_external__intel_fast_memcpy.A&lt;BR /&gt;
	0000000000000010 T __kmp_external__intel_fast_memcpy.J&lt;BR /&gt;
	0000000000000020 T __kmp_external__intel_fast_memcpy.M&lt;BR /&gt;
	0000000000000030 T __kmp_external__intel_fast_memcpy.P&lt;BR /&gt;
	&amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;U __kmp_external__intel_fast_memcpy&lt;BR /&gt;
	&amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;U __kmp_external__intel_fast_memcpy&lt;BR /&gt;
	00000000000c5d70 t _intel_fast_memcpy&lt;/P&gt;

&lt;P&gt;Best Regards,&lt;/P&gt;

&lt;P&gt;Ying&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Tue, 02 Aug 2016 07:08:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Intel-MKL-compilation-issue/m-p/1076909#M22612</guid>
      <dc:creator>Ying_H_Intel</dc:creator>
      <dc:date>2016-08-02T07:08:00Z</dc:date>
    </item>
  </channel>
</rss>

