<?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: dss routines - sequence in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727655#M37463</link>
    <description>&lt;P&gt;No, it can not. Internal results are stored in the handle. Using it for different matrices before finishing the cycle leads to undefined behavior.&lt;/P&gt;</description>
    <pubDate>Mon, 24 Nov 2025 21:40:48 GMT</pubDate>
    <dc:creator>Fengrui</dc:creator>
    <dc:date>2025-11-24T21:40:48Z</dc:date>
    <item>
      <title>dss routines - sequence</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1726979#M37451</link>
      <description>&lt;P&gt;Hi everyone,&lt;BR /&gt;I am testing the dss routines. Everything seems fine: I can call dss_create, dss_define_structure, dss_reorder, dss_factor_real and dss_slove_real&amp;nbsp;&lt;SPAN&gt;in that sequence and get the expected result. I can also repeat the sequence.&lt;BR /&gt;&lt;BR /&gt;However, if I again call define_structure after factor_real and before solve_real I get error -13 ( MKL_DSS_TERM_LVL_ERR )&lt;BR /&gt;&lt;BR /&gt;Did I miss something ? Is this not allowed ? Is there a way to undo the results of dss_reorder and dss_factor_real ?&amp;nbsp;&lt;BR /&gt;Any help appreciated&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;Harald&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Wed, 19 Nov 2025 12:06:08 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1726979#M37451</guid>
      <dc:creator>Harald31415</dc:creator>
      <dc:date>2025-11-19T12:06:08Z</dc:date>
    </item>
    <item>
      <title>Re: dss routines - sequence</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727009#M37452</link>
      <description>&lt;P&gt;I solved it pragmatically: I start from scratch&amp;nbsp; before each&amp;nbsp;&lt;SPAN class=""&gt;define_structure. Yet I am interested in the concept.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 19 Nov 2025 18:53:10 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727009#M37452</guid>
      <dc:creator>Harald31415</dc:creator>
      <dc:date>2025-11-19T18:53:10Z</dc:date>
    </item>
    <item>
      <title>Re: dss routines - sequence</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727428#M37456</link>
      <description>&lt;P&gt;Please read the documentation, which lists the order in which the individual DSS routines should be called. Usually, the structure of the matrix does not change during a calculation. Often, even the matrix is constant but the right hand side vector changes. If so, the matrix may be factorized just once, and only the solution plase needs to be repeated for each right hand side. The factorization is the most time-consuming part, so it makes sense to do it as few times as possible -- just once if that is correct to do.&lt;/P&gt;</description>
      <pubDate>Sun, 23 Nov 2025 13:33:39 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727428#M37456</guid>
      <dc:creator>mecej4O</dc:creator>
      <dc:date>2025-11-23T13:33:39Z</dc:date>
    </item>
    <item>
      <title>Re: dss routines - sequence</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727431#M37457</link>
      <description>&lt;P&gt;thank you mecey40. I did read the documentation; my question was, if the same handle can be used for different matrices, even before finishing the cycle.&lt;/P&gt;</description>
      <pubDate>Sun, 23 Nov 2025 14:19:46 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727431#M37457</guid>
      <dc:creator>Harald31415</dc:creator>
      <dc:date>2025-11-23T14:19:46Z</dc:date>
    </item>
    <item>
      <title>Re: dss routines - sequence</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727655#M37463</link>
      <description>&lt;P&gt;No, it can not. Internal results are stored in the handle. Using it for different matrices before finishing the cycle leads to undefined behavior.&lt;/P&gt;</description>
      <pubDate>Mon, 24 Nov 2025 21:40:48 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727655#M37463</guid>
      <dc:creator>Fengrui</dc:creator>
      <dc:date>2025-11-24T21:40:48Z</dc:date>
    </item>
    <item>
      <title>Re: dss routines - sequence</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727657#M37465</link>
      <description>&lt;P&gt;Thank you. That is a clear answer. I thought, there is some clear or free or release command which I missed.&lt;/P&gt;</description>
      <pubDate>Mon, 24 Nov 2025 21:48:58 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/dss-routines-sequence/m-p/1727657#M37465</guid>
      <dc:creator>Harald31415</dc:creator>
      <dc:date>2025-11-24T21:48:58Z</dc:date>
    </item>
  </channel>
</rss>

