<?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 OFFCORE_RESPONSE_0 and in Software Tuning, Performance Optimization &amp; Platform Monitoring</title>
    <link>https://community.intel.com/t5/Software-Tuning-Performance/Understanding-OFFCORE-RESPONSE-0-and-1/m-p/1023280#M4108</link>
    <description>&lt;P&gt;OFFCORE_RESPONSE_0 and OFFCORE_RESPONSE_1&amp;nbsp; provide identical functionality.&amp;nbsp; The reason that there are two of them is that these events are associated with a separate MSR that is used to program the types of requests/responses that you want to count (instead of being able to include this information in the Umask field of the PERFEVT_SELx MSR).&amp;nbsp;&amp;nbsp; The performance counter event OFFCORE_RESPONSE_0 (Event 0xB7) is associated with MSR 0x1A6, while the performance counter event OFFCORE_RESPONSE_1 (Event 0xBB) is associated with MSR 0x1A7.&lt;/P&gt;

&lt;P&gt;So having two events (with different associated MSRs) allows you to count two different offcore response events at the same time.&lt;/P&gt;</description>
    <pubDate>Tue, 02 Jun 2015 16:32:26 GMT</pubDate>
    <dc:creator>McCalpinJohn</dc:creator>
    <dc:date>2015-06-02T16:32:26Z</dc:date>
    <item>
      <title>Understanding OFFCORE_RESPONSE_0 and 1</title>
      <link>https://community.intel.com/t5/Software-Tuning-Performance/Understanding-OFFCORE-RESPONSE-0-and-1/m-p/1023279#M4107</link>
      <description>&lt;P&gt;I am new to hardware counter measurements on intel. I want to measure L3 misses to local and remote DRAM on Intel Ivy bridge &amp;nbsp;Model 62(Intel(R) Xeon(R) CPU E7-4860):&lt;/P&gt;

&lt;P&gt;I need to understand the difference between OFFCORE_RESPONSE_0 and&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;OFFCORE_RESPONSE_1 as both have the same sub-fields. Can I use either of these two to measure L3_LOCAL_MISSES and L3_REMOTE_MISSES? I am using PAPI low-level interface to add these two events and measure their counters for a section of my code. I have read/write access to MSR registers. Do I need to set up any lower level bits or anything related to offcore request?&amp;nbsp;&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Tue, 02 Jun 2015 16:00:29 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Tuning-Performance/Understanding-OFFCORE-RESPONSE-0-and-1/m-p/1023279#M4107</guid>
      <dc:creator>Mahwish_A_</dc:creator>
      <dc:date>2015-06-02T16:00:29Z</dc:date>
    </item>
    <item>
      <title>OFFCORE_RESPONSE_0 and</title>
      <link>https://community.intel.com/t5/Software-Tuning-Performance/Understanding-OFFCORE-RESPONSE-0-and-1/m-p/1023280#M4108</link>
      <description>&lt;P&gt;OFFCORE_RESPONSE_0 and OFFCORE_RESPONSE_1&amp;nbsp; provide identical functionality.&amp;nbsp; The reason that there are two of them is that these events are associated with a separate MSR that is used to program the types of requests/responses that you want to count (instead of being able to include this information in the Umask field of the PERFEVT_SELx MSR).&amp;nbsp;&amp;nbsp; The performance counter event OFFCORE_RESPONSE_0 (Event 0xB7) is associated with MSR 0x1A6, while the performance counter event OFFCORE_RESPONSE_1 (Event 0xBB) is associated with MSR 0x1A7.&lt;/P&gt;

&lt;P&gt;So having two events (with different associated MSRs) allows you to count two different offcore response events at the same time.&lt;/P&gt;</description>
      <pubDate>Tue, 02 Jun 2015 16:32:26 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Tuning-Performance/Understanding-OFFCORE-RESPONSE-0-and-1/m-p/1023280#M4108</guid>
      <dc:creator>McCalpinJohn</dc:creator>
      <dc:date>2015-06-02T16:32:26Z</dc:date>
    </item>
    <item>
      <title>Hello Mahwish,</title>
      <link>https://community.intel.com/t5/Software-Tuning-Performance/Understanding-OFFCORE-RESPONSE-0-and-1/m-p/1023281#M4109</link>
      <description>&lt;P&gt;Hello Mahwish,&lt;/P&gt;

&lt;P&gt;I don't know much about PAPI. I can tell you something about the events. Yes, you should be able to count L3_LOCAL_MISSES in one of the offcore_response counters and L3_REMOTE_MISSES in the other offcore_response counter. Exactly what papi programs for these 2 events... I don't know.&lt;/P&gt;

&lt;P&gt;You can look at &lt;A href="https://download.01.org/perfmon/IVB/IvyBridge_matrix_V14.tsv" target="_blank"&gt;https://download.01.org/perfmon/IVB/IvyBridge_matrix_V14.tsv&lt;/A&gt; and see the LLC_MISS.LOCAL_DRAM and LLC_MISS.ANY_DRAM events. After you program the appropriate PERFEVTSEL0/1 register, then the MSR_OFFCORE_RSP_0/1 register needs to be programmed. If you are going to use events from this tsv file, then the value in the MATRIX_VALUE column must be programmed into the 'extra' offcore_response MSR (MSR_OFFCORE_RSP_0/1... these are MSR num 0x1a6 and 0x1a7 respectively).&lt;/P&gt;

&lt;P&gt;Pat&lt;/P&gt;</description>
      <pubDate>Tue, 02 Jun 2015 18:31:02 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Tuning-Performance/Understanding-OFFCORE-RESPONSE-0-and-1/m-p/1023281#M4109</guid>
      <dc:creator>Patrick_F_Intel1</dc:creator>
      <dc:date>2015-06-02T18:31:02Z</dc:date>
    </item>
  </channel>
</rss>

