<?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 Questions on Itanium Branch analysis in Analyzers</title>
    <link>https://community.intel.com/t5/Analyzers/Questions-on-Itanium-Branch-analysis/m-p/964766#M8992</link>
    <description>&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;I've beentrying to get a detailed analysis of the BE_FLUSH_BUBBLE counts on my Itanium application. VTUNE reports the following Samples counts:&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-ALL = 8064&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-BRU = 8064&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-XPN = 0&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;So far so good. This tells me that all the Back-end flush bubbles are due to mispredicted branches. But when I drill down into some branch analysis, I get:&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_BR_MISPREDICTED_DETAIL-ANY = 0 !&lt;BR /&gt;&lt;BR /&gt;Yes, zero! How is this possible? (shouldn't it be 8064?)&lt;BR /&gt;&lt;BR /&gt;Also:&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BR_MISPRED_DETAIL-ALL  ALL_PRED = 5315&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt; BR_MISPRED_DETAIL-ALL  CORRECT_PRED = 4037&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt; BR_MISPRED_DETAIL-ALL  WRONG_PATH = 23&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt; BR_MISPRED_DETAIL-ALL  WRONG_TARGET = 1241&lt;BR /&gt;&lt;BR /&gt;...and:&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BR_MISPRED_DETAIL2-ALL = 0&lt;BR /&gt;&lt;BR /&gt;According to the Itanium 2 Ref Manual for Software Dev &amp;amp; Optimization, the misprediction count should be equal to:&lt;BR /&gt;&lt;BR /&gt;BR_MISPRED_DETAIL.[umask] - BR_MISPRED_DETAIL2.[umask]&lt;BR /&gt;&lt;BR /&gt;...and I cant correlate these counts with the grand total of 8064 Flush Bubbles.&lt;BR /&gt;&lt;BR /&gt;I would appreciate any help for my understanding of all this.&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; m
so-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-ALL = 8064&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-BRU = 8064&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-XPN = 0&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;&lt;/SPAN&gt;&lt;/DIV&gt;</description>
    <pubDate>Fri, 02 Apr 2004 00:05:17 GMT</pubDate>
    <dc:creator>jak-amado</dc:creator>
    <dc:date>2004-04-02T00:05:17Z</dc:date>
    <item>
      <title>Questions on Itanium Branch analysis</title>
      <link>https://community.intel.com/t5/Analyzers/Questions-on-Itanium-Branch-analysis/m-p/964766#M8992</link>
      <description>&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;I've beentrying to get a detailed analysis of the BE_FLUSH_BUBBLE counts on my Itanium application. VTUNE reports the following Samples counts:&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-ALL = 8064&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-BRU = 8064&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-XPN = 0&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;So far so good. This tells me that all the Back-end flush bubbles are due to mispredicted branches. But when I drill down into some branch analysis, I get:&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_BR_MISPREDICTED_DETAIL-ANY = 0 !&lt;BR /&gt;&lt;BR /&gt;Yes, zero! How is this possible? (shouldn't it be 8064?)&lt;BR /&gt;&lt;BR /&gt;Also:&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BR_MISPRED_DETAIL-ALL  ALL_PRED = 5315&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt; BR_MISPRED_DETAIL-ALL  CORRECT_PRED = 4037&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt; BR_MISPRED_DETAIL-ALL  WRONG_PATH = 23&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt; BR_MISPRED_DETAIL-ALL  WRONG_TARGET = 1241&lt;BR /&gt;&lt;BR /&gt;...and:&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BR_MISPRED_DETAIL2-ALL = 0&lt;BR /&gt;&lt;BR /&gt;According to the Itanium 2 Ref Manual for Software Dev &amp;amp; Optimization, the misprediction count should be equal to:&lt;BR /&gt;&lt;BR /&gt;BR_MISPRED_DETAIL.[umask] - BR_MISPRED_DETAIL2.[umask]&lt;BR /&gt;&lt;BR /&gt;...and I cant correlate these counts with the grand total of 8064 Flush Bubbles.&lt;BR /&gt;&lt;BR /&gt;I would appreciate any help for my understanding of all this.&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; m
so-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-ALL = 8064&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-BRU = 8064&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;BE_FLUSH_BUBBLE-XPN = 0&lt;/SPAN&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;SPAN style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA"&gt;&lt;/SPAN&gt;&lt;/DIV&gt;</description>
      <pubDate>Fri, 02 Apr 2004 00:05:17 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/Questions-on-Itanium-Branch-analysis/m-p/964766#M8992</guid>
      <dc:creator>jak-amado</dc:creator>
      <dc:date>2004-04-02T00:05:17Z</dc:date>
    </item>
  </channel>
</rss>

