<?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 Quote:Kevin Davis (Intel) in Software Archive</title>
    <link>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088727#M64494</link>
    <description>&lt;P&gt;&lt;/P&gt;&lt;BLOCKQUOTE&gt;Kevin Davis (Intel) wrote:&lt;BR /&gt;&lt;P&gt;&lt;/P&gt;

&lt;P&gt;Please pardon the delay in responding to your post. I will ask an MKL expert to assist here. Please stand-by.&lt;/P&gt;

&lt;P&gt;&lt;/P&gt;&lt;/BLOCKQUOTE&gt;&lt;P&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thanks Kevin. I've found that there is no problem. Nothing goes wrong with the code or compiler.&amp;nbsp;&lt;SPAN style="font-size: 1em; line-height: 1.5;"&gt;There are 240 threads running.&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;However, I don't know how to set this discussion thread as &amp;lt;solved&amp;gt;. Thanks for replying anyway.&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Thu, 03 Mar 2016 10:01:46 GMT</pubDate>
    <dc:creator>lu__shengliang</dc:creator>
    <dc:date>2016-03-03T10:01:46Z</dc:date>
    <item>
      <title>MKL BLAS function only run 1 thread on MIC</title>
      <link>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088725#M64492</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;I am trying to use MKL routines on MIC.&lt;/P&gt;

&lt;P&gt;But I noticed that the performance is slower than my CPU version. And there are only 1 thread running.&lt;/P&gt;

&lt;P&gt;Is that limited by MKL? I did some settings on the environment variables.&lt;/P&gt;

&lt;PRE class="brush:cpp;"&gt;__attribute__(( target (mic) )) void offload_check(void) {
#ifdef __MIC__
        printf("Check Func: Run on MIC!\n");
#else
        printf("Check Func: Run on CPU...\n");
#endif
}


#pragma offload target(mic:0)           \
        in(A:length(A_m*A_n))           \
        in(B:length(B_m*B_n))           \
        inout(C:length(C_m*C_n))
        {
                offload_check();
                cblas_sgemm(CblasRowMajor, CblasNoTrans, CblasNoTrans, A_m, B_n, B_m, \
                            alpha, A, A_n, B, B_n, beta, C, C_n);
        }
}
&lt;/PRE&gt;

&lt;P&gt;What I'm doing here is matrix multiplication. The N, M, K for matrix is 2048.&lt;/P&gt;

&lt;P&gt;Is &lt;SPAN style="font-size: 13.008px; line-height: 15.6096px;"&gt;it because my matrix is too small and data transmission takes a lot of time? I'm running it on a single node belonging to a cluster, where I could not use vtune to profile it...&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size: 13.008px; line-height: 15.6096px;"&gt;Thanks for your help&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Fri, 26 Feb 2016 09:39:02 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088725#M64492</guid>
      <dc:creator>lu__shengliang</dc:creator>
      <dc:date>2016-02-26T09:39:02Z</dc:date>
    </item>
    <item>
      <title>Please pardon the delay in</title>
      <link>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088726#M64493</link>
      <description>&lt;P&gt;Please pardon the delay in responding to your post. I will ask an MKL expert to assist here. Please stand-by.&lt;/P&gt;</description>
      <pubDate>Thu, 03 Mar 2016 09:57:12 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088726#M64493</guid>
      <dc:creator>Kevin_D_Intel</dc:creator>
      <dc:date>2016-03-03T09:57:12Z</dc:date>
    </item>
    <item>
      <title>Quote:Kevin Davis (Intel)</title>
      <link>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088727#M64494</link>
      <description>&lt;P&gt;&lt;/P&gt;&lt;BLOCKQUOTE&gt;Kevin Davis (Intel) wrote:&lt;BR /&gt;&lt;P&gt;&lt;/P&gt;

&lt;P&gt;Please pardon the delay in responding to your post. I will ask an MKL expert to assist here. Please stand-by.&lt;/P&gt;

&lt;P&gt;&lt;/P&gt;&lt;/BLOCKQUOTE&gt;&lt;P&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thanks Kevin. I've found that there is no problem. Nothing goes wrong with the code or compiler.&amp;nbsp;&lt;SPAN style="font-size: 1em; line-height: 1.5;"&gt;There are 240 threads running.&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;However, I don't know how to set this discussion thread as &amp;lt;solved&amp;gt;. Thanks for replying anyway.&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 03 Mar 2016 10:01:46 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088727#M64494</guid>
      <dc:creator>lu__shengliang</dc:creator>
      <dc:date>2016-03-03T10:01:46Z</dc:date>
    </item>
    <item>
      <title>Ok, thank you for the update</title>
      <link>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088728#M64495</link>
      <description>&lt;P&gt;Ok, thank you for the update/reply. We do apologize though for our lack of a reply earlier.&lt;/P&gt;

&lt;P&gt;Not being able to mark a thread resolved/answered/etc. is something lacking with our Forums at present. There are discussions/ideas being considered for that but currently it requires posting a reply :-(&lt;/P&gt;</description>
      <pubDate>Thu, 03 Mar 2016 10:22:55 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088728#M64495</guid>
      <dc:creator>Kevin_D_Intel</dc:creator>
      <dc:date>2016-03-03T10:22:55Z</dc:date>
    </item>
    <item>
      <title>No problem, Kevin!</title>
      <link>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088729#M64496</link>
      <description>&lt;P&gt;No problem, Kevin!&lt;/P&gt;

&lt;P&gt;Indeed, our developers need a lot of help from your side. thanks for your effort.&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;I will probably go for further questions later :)&lt;/P&gt;</description>
      <pubDate>Thu, 03 Mar 2016 11:01:52 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/MKL-BLAS-function-only-run-1-thread-on-MIC/m-p/1088729#M64496</guid>
      <dc:creator>lu__shengliang</dc:creator>
      <dc:date>2016-03-03T11:01:52Z</dc:date>
    </item>
  </channel>
</rss>

