<?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 Re: IntelMPITEST-1.0 compiled with icc in heterogeneous environ in Intel® MPI Library</title>
    <link>https://community.intel.com/t5/Intel-MPI-Library/IntelMPITEST-1-0-compiled-with-icc-in-heterogeneous-environment/m-p/916456#M2339</link>
    <description>&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;Clotho -&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;I'm not familiar with the IntelMPITEST, at least, not by that name. Can you be more specific about what this application is and where you got? I know about the Intel MPI Benchmark suite, but this doesn't have the function that you mentioned by name. Also, are you running this under Windows or Linux and which MPI package are you using?&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;Usually, problems with the Intel compiler are best submitted to the Intel Premier Support site. Ultimately, you'll likely need to go there, but I'd like to try this out on my own if I could to see if I might be able to get some insight into the problem.&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;-- clay&lt;/DIV&gt;</description>
    <pubDate>Fri, 11 Mar 2005 08:41:14 GMT</pubDate>
    <dc:creator>ClayB</dc:creator>
    <dc:date>2005-03-11T08:41:14Z</dc:date>
    <item>
      <title>IntelMPITEST-1.0 compiled with icc in heterogeneous environment</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/IntelMPITEST-1-0-compiled-with-icc-in-heterogeneous-environment/m-p/916455#M2338</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I would like to ask a question about using icc to compile&lt;BR /&gt;IntelMPITEST-1.0&lt;BR /&gt;and run the program in heterogeneous environment.&lt;BR /&gt;&lt;BR /&gt;I have a i386 node and a x86_64 node.&lt;BR /&gt;I have configed and compiled IntelMPITEST-1.0 testsuite at the i386&lt;BR /&gt;node.&lt;BR /&gt;I run the testsuite in the i386 node, and use "mpirun -machinefile" to&lt;BR /&gt;run&lt;BR /&gt;the binary on both nodes.&lt;BR /&gt;&lt;BR /&gt;I have tried the test with gcc and pgi compilers, they work.&lt;BR /&gt;But for icc8, I have encounter error in&lt;BR /&gt;c/blocking/functional/MPI_Ssend_ator&lt;BR /&gt;&lt;BR /&gt;The error message is very long,  but has similar pattern like:&lt;BR /&gt;&lt;BR /&gt;MPITEST error (3): i=0, long double value=  -0.0000000000, expected    &lt;BR /&gt;0.0000000000&lt;BR /&gt;MPITEST error (3): 10 errors in buffer (3,0,13) len 8 commsize 4 &lt;BR /&gt;commtype -10 data_type 13 root 3&lt;BR /&gt;MPITEST error (3): Send/Receive lengths differ - &lt;BR /&gt;Sender(node/length)=0/8,  Receiver(node/length)=3/-32766&lt;BR /&gt;MPITEST error (3): i=0, long double value=  -0.0000000000, expected    &lt;BR /&gt;0.0000000000&lt;BR /&gt;MPITEST error (3): 117 errors in buffer (4,0,13) len 83 commsize 4 &lt;BR /&gt;commtype -10 data_type 13 root 3&lt;BR /&gt;MPITEST error (3): Send/Receive lengths differ - &lt;BR /&gt;Sender(node/length)=0/83,  Receiver(node/length)=3/-32766&lt;BR /&gt;&lt;BR /&gt;All the errors are related to data_type 13 and 14.&lt;BR /&gt;This error does not happen when I run the tests on 2 i386 nodes.&lt;BR /&gt;&lt;BR /&gt;Have you any idea on the problem? Thank you.&lt;BR /&gt;&lt;BR /&gt;PS.&lt;BR /&gt;I find that the error message is produced from "libmpitest.c" line 2361.&lt;BR /&gt;&lt;BR /&gt;And I find that one of the many compilation warning is related to the&lt;BR /&gt;line&lt;BR /&gt;./libmpitest.c(2361): warning #181: argument is incompatible with &lt;BR /&gt;corresponding format string conversion&lt;BR /&gt;                          i, ((derived1 *)buffer)&lt;I&gt;.LongDouble&lt;K&gt;,&lt;BR /&gt;&lt;BR /&gt;May be it's related, I am not sure.&lt;/K&gt;&lt;/I&gt;</description>
      <pubDate>Wed, 02 Feb 2005 09:46:56 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/IntelMPITEST-1-0-compiled-with-icc-in-heterogeneous-environment/m-p/916455#M2338</guid>
      <dc:creator>clotho</dc:creator>
      <dc:date>2005-02-02T09:46:56Z</dc:date>
    </item>
    <item>
      <title>Re: IntelMPITEST-1.0 compiled with icc in heterogeneous environ</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/IntelMPITEST-1-0-compiled-with-icc-in-heterogeneous-environment/m-p/916456#M2339</link>
      <description>&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;Clotho -&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;I'm not familiar with the IntelMPITEST, at least, not by that name. Can you be more specific about what this application is and where you got? I know about the Intel MPI Benchmark suite, but this doesn't have the function that you mentioned by name. Also, are you running this under Windows or Linux and which MPI package are you using?&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;Usually, problems with the Intel compiler are best submitted to the Intel Premier Support site. Ultimately, you'll likely need to go there, but I'd like to try this out on my own if I could to see if I might be able to get some insight into the problem.&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;-- clay&lt;/DIV&gt;</description>
      <pubDate>Fri, 11 Mar 2005 08:41:14 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/IntelMPITEST-1-0-compiled-with-icc-in-heterogeneous-environment/m-p/916456#M2339</guid>
      <dc:creator>ClayB</dc:creator>
      <dc:date>2005-03-11T08:41:14Z</dc:date>
    </item>
  </channel>
</rss>

