<?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: dgels in 10.0.1.015 -&amp;gt; 10.0.2.019 in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913840#M12386</link>
    <description>lda and ldb are the "leading dimensions" of matrices a and b (the length of the columns as actually stored, in case they are stored in a matrix which is dimensioned larger than necessary). The minimum values are OK:&lt;BR /&gt;&lt;A href="http://www.netlib.org/lapack/double/dgels.f"&gt;http://www.netlib.org/lapack/double/dgels.f&lt;/A&gt;&lt;BR /&gt;Also, check that your lwork value is sufficient, preferably at least "optimal."&lt;BR /&gt;&lt;BR /&gt;</description>
    <pubDate>Mon, 10 Mar 2008 13:38:30 GMT</pubDate>
    <dc:creator>TimP</dc:creator>
    <dc:date>2008-03-10T13:38:30Z</dc:date>
    <item>
      <title>dgels in 10.0.1.015 -&gt; 10.0.2.019</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913839#M12385</link>
      <description>&lt;P&gt;Dear Board!&lt;/P&gt;
&lt;P&gt;I just installed the 10 Update 2 Version of MKL, and the dgels function seems not to be working anymore. When I do a debug-break in Visual Studio 2005, I see that the Call stack is something like (highest being most current)&lt;/P&gt;
&lt;P&gt;kmp_wait_sleep&lt;/P&gt;
&lt;P&gt;kmp_join_barrier&lt;/P&gt;
&lt;P&gt;kmp_internal_join&lt;/P&gt;
&lt;P&gt;kmp_join_call&lt;/P&gt;
&lt;P&gt;kmpc_fork_call&lt;/P&gt;
&lt;P&gt;_mkl_lapack_dgeqrf&lt;/P&gt;
&lt;P&gt;?&lt;/P&gt;
&lt;P&gt;mkl_lapack_dgels&lt;/P&gt;
&lt;P&gt;degls&lt;/P&gt;
&lt;P&gt;Dont know if I call the function correctly, since the documentation on this function is not clear for me (didnt understand lda and ldb), but it certainly works under 10 Update 1.&lt;/P&gt;
&lt;P&gt;Other functions like the fft ones still work.&lt;/P&gt;
&lt;P&gt;Thanks for any help&lt;/P&gt;</description>
      <pubDate>Mon, 10 Mar 2008 10:10:45 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913839#M12385</guid>
      <dc:creator>andipandi</dc:creator>
      <dc:date>2008-03-10T10:10:45Z</dc:date>
    </item>
    <item>
      <title>Re: dgels in 10.0.1.015 -&gt; 10.0.2.019</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913840#M12386</link>
      <description>lda and ldb are the "leading dimensions" of matrices a and b (the length of the columns as actually stored, in case they are stored in a matrix which is dimensioned larger than necessary). The minimum values are OK:&lt;BR /&gt;&lt;A href="http://www.netlib.org/lapack/double/dgels.f"&gt;http://www.netlib.org/lapack/double/dgels.f&lt;/A&gt;&lt;BR /&gt;Also, check that your lwork value is sufficient, preferably at least "optimal."&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Mon, 10 Mar 2008 13:38:30 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913840#M12386</guid>
      <dc:creator>TimP</dc:creator>
      <dc:date>2008-03-10T13:38:30Z</dc:date>
    </item>
    <item>
      <title>Re: dgels in 10.0.1.015 -&gt; 10.0.2.019</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913841#M12387</link>
      <description>&lt;P&gt;Thx for your reply!&lt;/P&gt;
&lt;P&gt;Lwork is big enough, after all, from the documentation I understood that the function generates an error in the info variable but returns immediately?!&lt;/P&gt;
&lt;P&gt;But in my case, the dgels function just hangs (with the call stack as written in my last post).&lt;/P&gt;
&lt;P&gt;For lda I simply take m, as for ldb as well (since it should be max(m,n)?) since I always have an overdetermined matrix with m &amp;gt; n.&lt;/P&gt;
&lt;P&gt;But, like I said, it hangs, without an error message. And I tried with 2 different installs of the MKL 10 Update 2, where the same code works with MKL 10 Update 1.&lt;/P&gt;
&lt;P&gt;If this adds any information, I have an overdetermined system that I try to approximate, with just a 1-dimensional result vector b. trans = 'N'.&lt;/P&gt;</description>
      <pubDate>Mon, 10 Mar 2008 17:44:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913841#M12387</guid>
      <dc:creator>andipandi</dc:creator>
      <dc:date>2008-03-10T17:44:00Z</dc:date>
    </item>
    <item>
      <title>Re: dgels in 10.0.1.015 -&gt; 10.0.2.019</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913842#M12388</link>
      <description>&lt;P&gt;Anyone at Iintel willing to answer my question?&lt;/P&gt;
&lt;P&gt;Or maybe ask some question where I need to provide further details?&lt;/P&gt;
&lt;P&gt;By the way, when debugging/testing, I call this code during an "attach dll". Might this cause any problems (no threads can be created at this point).&lt;/P&gt;</description>
      <pubDate>Tue, 25 Mar 2008 20:58:57 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913842#M12388</guid>
      <dc:creator>andipandi</dc:creator>
      <dc:date>2008-03-25T20:58:57Z</dc:date>
    </item>
    <item>
      <title>Re: dgels in 10.0.1.015 -&gt; 10.0.2.019</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913843#M12389</link>
      <description>It looks like you should file a problem report on premier.intel.com, providing the test case which shows the different behavior of these MKL versions.&lt;BR /&gt;</description>
      <pubDate>Wed, 26 Mar 2008 06:38:12 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913843#M12389</guid>
      <dc:creator>TimP</dc:creator>
      <dc:date>2008-03-26T06:38:12Z</dc:date>
    </item>
    <item>
      <title>Re: dgels in 10.0.1.015 -&gt; 10.0.2.019</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913844#M12390</link>
      <description>&lt;P&gt;So.. like I mentioned, the problem occured during startup, so I suppose that in the new version you have some kind of clever threading implemented which - in the startup (dll attachment) does not work (creation of threads is delayed till after attachment).&lt;/P&gt;
&lt;P&gt;When calling the code at a later point, there is no problem.&lt;/P&gt;
&lt;P&gt;I also changed to custom building an MKL dll due to other problems..&lt;/P&gt;
&lt;P&gt;Problem solved, and many thanks!&lt;/P&gt;</description>
      <pubDate>Thu, 27 Mar 2008 11:01:55 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913844#M12390</guid>
      <dc:creator>andipandi</dc:creator>
      <dc:date>2008-03-27T11:01:55Z</dc:date>
    </item>
    <item>
      <title>Re: dgels in 10.0.1.015 -&gt; 10.0.2.019</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913845#M12391</link>
      <description>&lt;P&gt;I also encountered the same problem. So how do you fix it?&lt;/P&gt;</description>
      <pubDate>Fri, 24 Oct 2008 09:41:28 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dgels-in-10-0-1-015-gt-10-0-2-019/m-p/913845#M12391</guid>
      <dc:creator>rahewin</dc:creator>
      <dc:date>2008-10-24T09:41:28Z</dc:date>
    </item>
  </channel>
</rss>

