<?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 It looks like this is an in Software Tuning, Performance Optimization &amp; Platform Monitoring</title>
    <link>https://community.intel.com/t5/Software-Tuning-Performance/Inaccuracy-from-Memory-read-write-bytes-when-using-PCM-2-11-on/m-p/1106652#M5963</link>
    <description>&lt;P&gt;It looks like this is an attempt to measure behavior for the initial access to the array.&amp;nbsp;&amp;nbsp; The behavior of the system for page instantiation may not be what you expect.&amp;nbsp;&lt;/P&gt;

&lt;P&gt;I recommend that you add more loops to read/write the data after it is initialized, and that you make the number of times the data is accessed after initialization an easy-to-modify variable.&amp;nbsp; You should then see that you get the expected memory access counts for subsequent accesses.&lt;/P&gt;

&lt;P&gt;&lt;BR /&gt;
	&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Mon, 10 Oct 2016 19:23:17 GMT</pubDate>
    <dc:creator>McCalpinJohn</dc:creator>
    <dc:date>2016-10-10T19:23:17Z</dc:date>
    <item>
      <title>Inaccuracy from Memory read/write bytes when using PCM-2.11 on Core i7 5960X Haswell-E</title>
      <link>https://community.intel.com/t5/Software-Tuning-Performance/Inaccuracy-from-Memory-read-write-bytes-when-using-PCM-2-11-on/m-p/1106651#M5962</link>
      <description>&lt;P style="margin-top:0cm;margin-right:0cm;margin-bottom:18.0pt;margin-left:
0cm"&gt;&lt;SPAN style="font-size: 10pt; font-family: Arial, sans-serif;"&gt;Hi All,&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P style="margin: 0cm 0cm 0.0001pt; box-sizing: border-box; word-wrap: break-word;"&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;I try to use PCM-2.11 to measure the memory read/write bytes on Intel I7 5960X&lt;SPAN class="apple-converted-space"&gt;&amp;nbsp;&lt;/SPAN&gt;Processor Extreme Edition.&amp;nbsp;&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;The example “for (int i = 0; i &amp;lt; 200000000; i++) (int *)p_data&lt;I&gt; = i;” is compiled with ICC 16.0.3. &amp;nbsp;&amp;nbsp;&lt;P&gt;&lt;/P&gt;&lt;/I&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;I have checked the assemble of the store instruction: movdqa. It means each cache line will be read first.&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;The PCM result shows:&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;BytesReadFromMC: 47615872 (wrong number)&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;BytesWrittenToMC: 808465920 (right number).&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P style="margin-top:0cm;margin-right:0cm;margin-bottom:15.0pt;margin-left:
0cm"&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;Do I need to modify the lines in cpucounters.cpp: &amp;nbsp;&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P style="margin: 0cm 0cm 0.0001pt; box-sizing: border-box; word-wrap: break-word;"&gt;&lt;EM style="box-sizing: border-box"&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;
color:#666666"&gt;void ServerPCICFGUncore::program()&lt;/SPAN&gt;&lt;/EM&gt;&lt;BR style="box-sizing: border-box" /&gt;
	&lt;I&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;&lt;EM&gt;{&lt;/EM&gt;&lt;BR style="box-sizing: border-box" /&gt;
	&lt;EM&gt;&amp;nbsp; &amp;nbsp; uint32 MCCntConfig[4] = {&lt;/EM&gt;&lt;BR style="box-sizing: border-box" /&gt;
	&lt;EM&gt;&amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; MC_CH_PCI_PMON_CTL_EVENT(0x04) + MC_CH_PCI_PMON_CTL_UMASK(3), &amp;nbsp;&lt;/EM&gt;&lt;BR style="box-sizing: border-box" /&gt;
	&lt;EM&gt;&amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; MC_CH_PCI_PMON_CTL_EVENT(0x04) + MC_CH_PCI_PMON_CTL_UMASK(12),&amp;nbsp;&amp;nbsp; &amp;nbsp;&lt;/EM&gt;&lt;BR style="box-sizing: border-box" /&gt;
	&lt;EM&gt;&amp;nbsp; &amp;nbsp; };&lt;/EM&gt;&lt;/SPAN&gt;&lt;/I&gt;&lt;SPAN style="font-size:10.5pt;font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P style="margin: 0cm 0cm 15pt; box-sizing: border-box; word-wrap: break-word;"&gt;&lt;SPAN style="font-size:10.5pt;
font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P style="margin: 0cm 0cm 15pt; box-sizing: border-box; word-wrap: break-word;"&gt;&lt;SPAN style="font-size:10.5pt;
font-family:&amp;quot;Arial&amp;quot;,sans-serif;color:#666666"&gt;Thanks.&amp;nbsp;&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Sat, 08 Oct 2016 03:28:13 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Tuning-Performance/Inaccuracy-from-Memory-read-write-bytes-when-using-PCM-2-11-on/m-p/1106651#M5962</guid>
      <dc:creator>ZWang45</dc:creator>
      <dc:date>2016-10-08T03:28:13Z</dc:date>
    </item>
    <item>
      <title>It looks like this is an</title>
      <link>https://community.intel.com/t5/Software-Tuning-Performance/Inaccuracy-from-Memory-read-write-bytes-when-using-PCM-2-11-on/m-p/1106652#M5963</link>
      <description>&lt;P&gt;It looks like this is an attempt to measure behavior for the initial access to the array.&amp;nbsp;&amp;nbsp; The behavior of the system for page instantiation may not be what you expect.&amp;nbsp;&lt;/P&gt;

&lt;P&gt;I recommend that you add more loops to read/write the data after it is initialized, and that you make the number of times the data is accessed after initialization an easy-to-modify variable.&amp;nbsp; You should then see that you get the expected memory access counts for subsequent accesses.&lt;/P&gt;

&lt;P&gt;&lt;BR /&gt;
	&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 10 Oct 2016 19:23:17 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Tuning-Performance/Inaccuracy-from-Memory-read-write-bytes-when-using-PCM-2-11-on/m-p/1106652#M5963</guid>
      <dc:creator>McCalpinJohn</dc:creator>
      <dc:date>2016-10-10T19:23:17Z</dc:date>
    </item>
  </channel>
</rss>

