<?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 Please see 3.5 (Listing 1 in Intel® MPI Library</title>
    <link>https://community.intel.com/t5/Intel-MPI-Library/MPI-Test-blocks-on-the-request-from-MPI-Rget/m-p/1072548#M4707</link>
    <description>&lt;P&gt;Please see 3.5 (Listing 1 there is instructive)&amp;nbsp;in&lt;/P&gt;

&lt;P&gt;&lt;A href="http://www.mcs.anl.gov/papers/P4062-0413_1.pdf"&gt;http://www.mcs.anl.gov/papers/P4062-0413_1.pdf&lt;/A&gt;&lt;/P&gt;

&lt;P&gt;quoting: "&amp;nbsp;These operations return an MPI Request object &lt;EM&gt;similar to non-blocking point-to-point communication &lt;/EM&gt;that can be tested or can wait for completion using MPI_Test and MPI_Wait, or the equivalent". So yes, you can initiate these request based RMA calls and then you can wait for their completion (completion means that the remote data delivered to local buffer as described in the reference above).&lt;/P&gt;

&lt;P&gt;BR,&lt;/P&gt;

&lt;P&gt;Mark&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Tue, 01 Mar 2016 02:39:49 GMT</pubDate>
    <dc:creator>Mark_L_Intel</dc:creator>
    <dc:date>2016-03-01T02:39:49Z</dc:date>
    <item>
      <title>MPI_Test blocks on the request from MPI_Rget</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/MPI-Test-blocks-on-the-request-from-MPI-Rget/m-p/1072547#M4706</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;To progress one-sided communication right after MPI_RGet function call, I am explicitly calling 'MPI_Test'.&lt;/P&gt;

&lt;P&gt;The profiled results show that MPI_Test is blocking the thread.&lt;/P&gt;

&lt;P&gt;Is it intended behavior or not?&lt;/P&gt;

&lt;P&gt;(I compiled with -mt_mpi flag and enabled MPICH_ASYNC_PROGRESS and I_MPI_RDMA_SCALABLE_PROGRESS&amp;nbsp;too)&lt;/P&gt;</description>
      <pubDate>Mon, 15 Feb 2016 09:57:04 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/MPI-Test-blocks-on-the-request-from-MPI-Rget/m-p/1072547#M4706</guid>
      <dc:creator>seongyun_k_</dc:creator>
      <dc:date>2016-02-15T09:57:04Z</dc:date>
    </item>
    <item>
      <title>Please see 3.5 (Listing 1</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/MPI-Test-blocks-on-the-request-from-MPI-Rget/m-p/1072548#M4707</link>
      <description>&lt;P&gt;Please see 3.5 (Listing 1 there is instructive)&amp;nbsp;in&lt;/P&gt;

&lt;P&gt;&lt;A href="http://www.mcs.anl.gov/papers/P4062-0413_1.pdf"&gt;http://www.mcs.anl.gov/papers/P4062-0413_1.pdf&lt;/A&gt;&lt;/P&gt;

&lt;P&gt;quoting: "&amp;nbsp;These operations return an MPI Request object &lt;EM&gt;similar to non-blocking point-to-point communication &lt;/EM&gt;that can be tested or can wait for completion using MPI_Test and MPI_Wait, or the equivalent". So yes, you can initiate these request based RMA calls and then you can wait for their completion (completion means that the remote data delivered to local buffer as described in the reference above).&lt;/P&gt;

&lt;P&gt;BR,&lt;/P&gt;

&lt;P&gt;Mark&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Tue, 01 Mar 2016 02:39:49 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/MPI-Test-blocks-on-the-request-from-MPI-Rget/m-p/1072548#M4707</guid>
      <dc:creator>Mark_L_Intel</dc:creator>
      <dc:date>2016-03-01T02:39:49Z</dc:date>
    </item>
  </channel>
</rss>

