<?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 Problem in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775436#M974</link>
    <description>Well, to get correct Fortran code to work with MKL ILP64 library it needs to use compiler options:&lt;BR /&gt; -i8 for Intel Fortran&lt;BR /&gt; -fdefault-integer-8 for GNU Fortran</description>
    <pubDate>Thu, 22 Jul 2010 09:43:20 GMT</pubDate>
    <dc:creator>barragan_villanueva_</dc:creator>
    <dc:date>2010-07-22T09:43:20Z</dc:date>
    <item>
      <title>Problem</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775433#M971</link>
      <description>Hi everyone, I am trying to run a simple fft from dfti routine of the latest MKL modules... I used:&lt;BR /&gt;...&lt;BR /&gt;USE mkl_dfti&lt;BR /&gt;...&lt;BR /&gt;TYPE(DFTI_DESCRIPTOR), POINTER :: My_Desc_Handle&lt;BR /&gt;...&lt;BR /&gt;status = DftiCreateDescriptor(My_Desc_Handle, DFTI_DOUBLE, &amp;amp;&lt;BR /&gt; DFTI_REAL, 1, alpha+2)&lt;BR /&gt; WRITE(*,*) 'Status 1',status,alpha&lt;BR /&gt; status = DftiSetValue(My_Desc_Handle, DFTI_PLACEMENT, DFTI_NOT_INPLACE)&lt;BR /&gt; WRITE(*,*) 'Status 2',status&lt;BR /&gt; status = DftiCommitDescriptor(My_Desc_Handle)&lt;BR /&gt; WRITE(*,*) 'Status 3',status&lt;BR /&gt; status = DftiComputeForward(My_Desc_Handle, Vlm, DFTV)&lt;BR /&gt; WRITE(*,*) 'Status 4',status&lt;BR /&gt; status = DftiFreeDescriptor(My_Desc_Handle)&lt;BR /&gt; WRITE(*,*) 'Status 5',status&lt;BR /&gt;&lt;BR /&gt;... alpha= 20&lt;BR /&gt;I get as status &lt;BR /&gt; 2&lt;BR /&gt; 5&lt;BR /&gt; 5&lt;BR /&gt; 5&lt;BR /&gt; 5&lt;BR /&gt;The DFTV array coming from the routine is zero... &lt;BR /&gt;I compiled with the command on a linux machine:&lt;BR /&gt;&lt;BR /&gt;$(f90) /DGFI16/home/fuchs/intel/Compiler/11.1/072/mkl/include/mkl_dfti.f90 Constants.f90 Flmk.f90 /DGFI16/home/fuchs/intel/Compiler/11.1/072/mkl/include/lapack.f90 SubsTWFD.f90 Legendre.f90 vecmat.f90&lt;BR /&gt;ar -rv $(lib) mkl_dfti.o Constants.o Flmk.o lapack.o SubsTWFD.o Legendre.o vecmat.o&lt;BR /&gt;&lt;BR /&gt;can someone tell me what went wrong ???&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 22 Jul 2010 08:42:35 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775433#M971</guid>
      <dc:creator>Martin_Fuchs</dc:creator>
      <dc:date>2010-07-22T08:42:35Z</dc:date>
    </item>
    <item>
      <title>Problem</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775434#M972</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;It's possibleto get DFTI-messages on given statuses instead of numbers.&lt;BR /&gt;See $MKLROOT/examples/dftf/sources for details.&lt;BR /&gt;&lt;BR /&gt;E.g.&lt;BR /&gt; Status = DftiCreateDescriptor(hand, DFTI_DOUBLE, DFTI_REAL, 1, n)&lt;BR /&gt; if (.NOT. DftiErrorClass(Status, DFTI_NO_ERROR)) then&lt;BR /&gt; &lt;STRONG&gt;print *,'Status 1:', DftiErrorMessage(Status)&lt;BR /&gt;&lt;/STRONG&gt; end if&lt;BR /&gt;&lt;BR /&gt;As to your problem: could you please send small standalone test-case and compilation line with MKL-libraries to reproduce that?</description>
      <pubDate>Thu, 22 Jul 2010 09:23:36 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775434#M972</guid>
      <dc:creator>barragan_villanueva_</dc:creator>
      <dc:date>2010-07-22T09:23:36Z</dc:date>
    </item>
    <item>
      <title>Problem</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775435#M973</link>
      <description>Thanks Victor I think I managed the problem, it might be the ilp64 library; instead I used the lp64 library... Now I good some memory problem... But at least the dfti return is not zero</description>
      <pubDate>Thu, 22 Jul 2010 09:30:53 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775435#M973</guid>
      <dc:creator>Martin_Fuchs</dc:creator>
      <dc:date>2010-07-22T09:30:53Z</dc:date>
    </item>
    <item>
      <title>Problem</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775436#M974</link>
      <description>Well, to get correct Fortran code to work with MKL ILP64 library it needs to use compiler options:&lt;BR /&gt; -i8 for Intel Fortran&lt;BR /&gt; -fdefault-integer-8 for GNU Fortran</description>
      <pubDate>Thu, 22 Jul 2010 09:43:20 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Problem/m-p/775436#M974</guid>
      <dc:creator>barragan_villanueva_</dc:creator>
      <dc:date>2010-07-22T09:43:20Z</dc:date>
    </item>
  </channel>
</rss>

