<?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 libguide problems with serial MKL library in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/libguide-problems-with-serial-MKL-library/m-p/911420#M12185</link>
    <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I'm having the following problems with the libguide library. I'm using the SERIAL version of the MKL blas and lapack libraries together with the threaded version of PARDISO v3.2 available from the developers. &lt;BR /&gt;&lt;BR /&gt;I link all of them with the following commands (part of my Makefile) on a Clovertown machine:&lt;BR /&gt;&lt;BR /&gt;FC  := ifort -O3 -r8 -i8 -xT -traceback&lt;BR /&gt;LIBS := -L ../lib/x86_64 -lpardiso_INTEL_91_EM64T_P -lskit -lmkl_lapack -lmkl_em64t&lt;BR /&gt;&lt;BR /&gt;HOMER: $(HOMER_Objects)&lt;BR /&gt; @echo Compiling HOMER for the $(shell arch) architecture ...&lt;BR /&gt; @$(FC) -module ../obj/ $^ $(LIBS) -openmp -o ../bin/HOMER&lt;BR /&gt;&lt;BR /&gt;Then I run the program and everything works fine. But then, after about 4 hours computing using four threads the program stops and I get the following message:&lt;BR /&gt;&lt;BR /&gt;OMP abort: Initializing libguide.so, but found libguide.so already initialized.&lt;BR /&gt;This may cause performance degradation and correctness issues.&lt;BR /&gt;Set environment variable KMP_DUPLICATE_LIB_OK=TRUE to ignore&lt;BR /&gt;this problem and force the program to continue anyway.&lt;BR /&gt;Please note that the use of KMP_DUPLICATE_LIB_OK is unsupported&lt;BR /&gt;and using it may cause undefined behavior.&lt;BR /&gt;For more information, please contact Intel Premier Support.&lt;BR /&gt;&lt;BR /&gt;forrtl: severe (174): SIGSEGV, segmentation fault occurred&lt;BR /&gt;Image PC Routine Line Source&lt;BR /&gt;libpthread.so.0 0000003C2E80C430 Unknown Unknown Unknown&lt;BR /&gt;libpthread.so.0 0000003C2E80C2CE Unknown Unknown Unknown&lt;BR /&gt;libguide.so 0000002A9588820E Unknown Unknown Unknown&lt;BR /&gt;&lt;BR /&gt;I switched to the serial version of the MKL libraries to avoid this problem (after reading the MKL user guide), but now I do not know what to do, as I do not understand from where does the duplicate libguide.so come from ...&lt;BR /&gt;&lt;BR /&gt;Somebody knows a solution?&lt;BR /&gt;&lt;BR /&gt;Thanks in advance and greetings from Spain,&lt;BR /&gt;&lt;BR /&gt;Miguel&lt;BR /&gt;</description>
    <pubDate>Thu, 30 Aug 2007 13:39:14 GMT</pubDate>
    <dc:creator>mhermanns</dc:creator>
    <dc:date>2007-08-30T13:39:14Z</dc:date>
    <item>
      <title>libguide problems with serial MKL library</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/libguide-problems-with-serial-MKL-library/m-p/911420#M12185</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I'm having the following problems with the libguide library. I'm using the SERIAL version of the MKL blas and lapack libraries together with the threaded version of PARDISO v3.2 available from the developers. &lt;BR /&gt;&lt;BR /&gt;I link all of them with the following commands (part of my Makefile) on a Clovertown machine:&lt;BR /&gt;&lt;BR /&gt;FC  := ifort -O3 -r8 -i8 -xT -traceback&lt;BR /&gt;LIBS := -L ../lib/x86_64 -lpardiso_INTEL_91_EM64T_P -lskit -lmkl_lapack -lmkl_em64t&lt;BR /&gt;&lt;BR /&gt;HOMER: $(HOMER_Objects)&lt;BR /&gt; @echo Compiling HOMER for the $(shell arch) architecture ...&lt;BR /&gt; @$(FC) -module ../obj/ $^ $(LIBS) -openmp -o ../bin/HOMER&lt;BR /&gt;&lt;BR /&gt;Then I run the program and everything works fine. But then, after about 4 hours computing using four threads the program stops and I get the following message:&lt;BR /&gt;&lt;BR /&gt;OMP abort: Initializing libguide.so, but found libguide.so already initialized.&lt;BR /&gt;This may cause performance degradation and correctness issues.&lt;BR /&gt;Set environment variable KMP_DUPLICATE_LIB_OK=TRUE to ignore&lt;BR /&gt;this problem and force the program to continue anyway.&lt;BR /&gt;Please note that the use of KMP_DUPLICATE_LIB_OK is unsupported&lt;BR /&gt;and using it may cause undefined behavior.&lt;BR /&gt;For more information, please contact Intel Premier Support.&lt;BR /&gt;&lt;BR /&gt;forrtl: severe (174): SIGSEGV, segmentation fault occurred&lt;BR /&gt;Image PC Routine Line Source&lt;BR /&gt;libpthread.so.0 0000003C2E80C430 Unknown Unknown Unknown&lt;BR /&gt;libpthread.so.0 0000003C2E80C2CE Unknown Unknown Unknown&lt;BR /&gt;libguide.so 0000002A9588820E Unknown Unknown Unknown&lt;BR /&gt;&lt;BR /&gt;I switched to the serial version of the MKL libraries to avoid this problem (after reading the MKL user guide), but now I do not know what to do, as I do not understand from where does the duplicate libguide.so come from ...&lt;BR /&gt;&lt;BR /&gt;Somebody knows a solution?&lt;BR /&gt;&lt;BR /&gt;Thanks in advance and greetings from Spain,&lt;BR /&gt;&lt;BR /&gt;Miguel&lt;BR /&gt;</description>
      <pubDate>Thu, 30 Aug 2007 13:39:14 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/libguide-problems-with-serial-MKL-library/m-p/911420#M12185</guid>
      <dc:creator>mhermanns</dc:creator>
      <dc:date>2007-08-30T13:39:14Z</dc:date>
    </item>
  </channel>
</rss>

