<?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 Finding L1 bound Using Vtune in Analyzers</title>
    <link>https://community.intel.com/t5/Analyzers/Finding-L1-bound-Using-Vtune/m-p/1706423#M25867</link>
    <description>&lt;P&gt;Hi, I look at Vtune logs and documents. L1 bound in top-down metrics is defined and calculated as follows&amp;nbsp;&lt;/P&gt;&lt;P&gt;L1_Bound = 100 * max(0, (EXE_ACTIVITY.BOUND_ON_LOADS - MEMORY_ACTIVITY.STALLS_L1D_MISS) / CPU_CLK_UNHALTED.THREAD)&lt;BR /&gt;&lt;BR /&gt;Are EXE_ACTIVITY.BOUND_ON_LOADS, and&amp;nbsp;MEMORY_ACTIVITY.STALLS_L1D_MISS counted in slots not cycles? If yes, shouldn't we divide it by the total number of slots?&lt;BR /&gt;&lt;BR /&gt;The problem is that when I get these event counters, I can get these&amp;nbsp;&lt;BR /&gt;&lt;BR /&gt;EXE_ACTIVITY.BOUND_ON_LOADS 69010262370.0&lt;/P&gt;&lt;P&gt;MEMORY_ACTIVITY.STALLS_L1D_MISS 64751552228.0&lt;/P&gt;&lt;P&gt;CPU_CLK_UNHALTED.THREAD 200677632402.0&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;So L1_Bound = 2.2% but Vtune reports 21%?&lt;BR /&gt;Is there a factor of 10 missing?!!&lt;/P&gt;</description>
    <pubDate>Wed, 30 Jul 2025 04:11:47 GMT</pubDate>
    <dc:creator>meta1360</dc:creator>
    <dc:date>2025-07-30T04:11:47Z</dc:date>
    <item>
      <title>Finding L1 bound Using Vtune</title>
      <link>https://community.intel.com/t5/Analyzers/Finding-L1-bound-Using-Vtune/m-p/1706423#M25867</link>
      <description>&lt;P&gt;Hi, I look at Vtune logs and documents. L1 bound in top-down metrics is defined and calculated as follows&amp;nbsp;&lt;/P&gt;&lt;P&gt;L1_Bound = 100 * max(0, (EXE_ACTIVITY.BOUND_ON_LOADS - MEMORY_ACTIVITY.STALLS_L1D_MISS) / CPU_CLK_UNHALTED.THREAD)&lt;BR /&gt;&lt;BR /&gt;Are EXE_ACTIVITY.BOUND_ON_LOADS, and&amp;nbsp;MEMORY_ACTIVITY.STALLS_L1D_MISS counted in slots not cycles? If yes, shouldn't we divide it by the total number of slots?&lt;BR /&gt;&lt;BR /&gt;The problem is that when I get these event counters, I can get these&amp;nbsp;&lt;BR /&gt;&lt;BR /&gt;EXE_ACTIVITY.BOUND_ON_LOADS 69010262370.0&lt;/P&gt;&lt;P&gt;MEMORY_ACTIVITY.STALLS_L1D_MISS 64751552228.0&lt;/P&gt;&lt;P&gt;CPU_CLK_UNHALTED.THREAD 200677632402.0&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;So L1_Bound = 2.2% but Vtune reports 21%?&lt;BR /&gt;Is there a factor of 10 missing?!!&lt;/P&gt;</description>
      <pubDate>Wed, 30 Jul 2025 04:11:47 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/Finding-L1-bound-Using-Vtune/m-p/1706423#M25867</guid>
      <dc:creator>meta1360</dc:creator>
      <dc:date>2025-07-30T04:11:47Z</dc:date>
    </item>
  </channel>
</rss>

