<?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: DCOPY failing on Linux in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865150#M7834</link>
    <description>&lt;P&gt;Well, I just tried out all copies from 1 to 10 and they all worked. But the environment was not identical. What is the size of n? What are the values of resid(1):resid(9)? Same for workd.&lt;/P&gt;
&lt;P&gt;Bruce&lt;/P&gt;</description>
    <pubDate>Wed, 23 May 2007 23:17:20 GMT</pubDate>
    <dc:creator>Intel_C_Intel</dc:creator>
    <dc:date>2007-05-23T23:17:20Z</dc:date>
    <item>
      <title>DCOPY failing on Linux</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865149#M7833</link>
      <description>I have some well tested FORTRAN code (ARPACK) that I use with MKL 9.0 on Windows and OS X. I have been having problems on RHEL 4 on em64T -  the code is running in 32-bit mode.
&lt;BR /&gt;
&lt;BR /&gt;
&lt;BR /&gt;call dcopy (n, resid, 1, workd(irj), 1)
&lt;BR /&gt;It seems somewhat bizarre, but this call to dcopy fails to copy element resid(9) into the  workd(irj) array correctly leaving a NAN in the workd(irj+8)
&lt;BR /&gt;
&lt;BR /&gt;From the Intel debugger , here are the values after the call. Note the NAN value of workd(23), but other values were copied correctly.
&lt;BR /&gt;
&lt;BR /&gt;(idb) print n
&lt;BR /&gt;14
&lt;BR /&gt;(idb) print resid(9)
&lt;BR /&gt;0.00060894080186124633
&lt;BR /&gt;(idb) print workd(irj+8)
&lt;BR /&gt;nan 
&lt;BR /&gt;(idb) print irj
&lt;BR /&gt;15
&lt;BR /&gt;
&lt;BR /&gt;Some other values...
&lt;BR /&gt;
&lt;BR /&gt;(idb) print resid(1)
&lt;BR /&gt;0.00036273709783252173
&lt;BR /&gt;(idb) print workd(irj)
&lt;BR /&gt;0.00036273709783252173
&lt;BR /&gt;(idb) print workd(irj+13)
&lt;BR /&gt;4.3187914434694719e-05
&lt;BR /&gt;(idb) print resid(14)
&lt;BR /&gt;4.3187914434694719e-05
&lt;BR /&gt;
&lt;BR /&gt;Setting a memory watch point on the variable shows MKL copying what looks like garbage data into the variable.
&lt;BR /&gt;
&lt;BR /&gt;(idb) print &amp;amp;workd(23)
&lt;BR /&gt;0x80afa38
&lt;BR /&gt;[2] Address 0x080afa38 was accessed at: 
&lt;BR /&gt; function mkl_blas_dcopy(...)
&lt;BR /&gt;[0xf2cb9761] mkl_blas_dcopy+0x1b1: fstpl (%ebp)
&lt;BR /&gt;0x080afa38: Old value = 0xbff568517ad130c3
&lt;BR /&gt;0x00000000: New value = 0xfff8000000000000
&lt;BR /&gt;
&lt;BR /&gt;Anyone got a clue? This has been a debugging nightmare.
&lt;BR /&gt;
&lt;BR /&gt;</description>
      <pubDate>Mon, 30 Apr 2007 23:58:53 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865149#M7833</guid>
      <dc:creator>AndrewC</dc:creator>
      <dc:date>2007-04-30T23:58:53Z</dc:date>
    </item>
    <item>
      <title>Re: DCOPY failing on Linux</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865150#M7834</link>
      <description>&lt;P&gt;Well, I just tried out all copies from 1 to 10 and they all worked. But the environment was not identical. What is the size of n? What are the values of resid(1):resid(9)? Same for workd.&lt;/P&gt;
&lt;P&gt;Bruce&lt;/P&gt;</description>
      <pubDate>Wed, 23 May 2007 23:17:20 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865150#M7834</guid>
      <dc:creator>Intel_C_Intel</dc:creator>
      <dc:date>2007-05-23T23:17:20Z</dc:date>
    </item>
    <item>
      <title>Re: DCOPY failing on Linux</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865151#M7835</link>
      <description>Hi,
&lt;BR /&gt;This has turned out to be a very strange problem.I am building a 32 bit Linux tool on 64 bit linux .When I rebuilt everything as a 64-bit , the problem disappeared.
&lt;BR /&gt;( using ifort 9.1, icc 9.1 and MKL 9.0)
&lt;BR /&gt;I suspect compiler problems
&lt;BR /&gt;
&lt;BR /&gt;To answer your question, all the values in the array were copied correctly , except the one entry.
&lt;BR /&gt;</description>
      <pubDate>Fri, 01 Jun 2007 16:35:29 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865151#M7835</guid>
      <dc:creator>AndrewC</dc:creator>
      <dc:date>2007-06-01T16:35:29Z</dc:date>
    </item>
    <item>
      <title>Re: DCOPY failing on Linux</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865152#M7836</link>
      <description>&lt;P&gt;So is everything OK as far as MKL is concerned?&lt;/P&gt;
&lt;P&gt;Bruce&lt;/P&gt;</description>
      <pubDate>Tue, 05 Jun 2007 16:59:23 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/DCOPY-failing-on-Linux/m-p/865152#M7836</guid>
      <dc:creator>Intel_C_Intel</dc:creator>
      <dc:date>2007-06-05T16:59:23Z</dc:date>
    </item>
  </channel>
</rss>

