<?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 Total time vs. file duration  in Analyzers</title>
    <link>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872173#M3898</link>
    <description>&lt;BLOCKQUOTE&gt;
&lt;P&gt;I think that Call graph will spend more time than application itself, since Call graph will call fun ctions in itslibraries.&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;
&lt;P&gt;I agree with you, Peter and I'm confused with the result observed from the profiling.&lt;/P&gt;
&lt;BLOCKQUOTE&gt;
&lt;P&gt;You said that total time was from main function only, I suppose that is not enough! You  have to count spending timefrom other threads.&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;
&lt;P&gt;I had a correct result when running in multi-thread. However, the result with total time shorter than the duration time was the one in mono-thread mode. Therefore, I think that observing the main total time would be enough since it corresponds to the application total time. Am I wrong?&lt;/P&gt;
&lt;BLOCKQUOTE&gt;
&lt;P&gt;See example for application vtunedemo on Linux*:&lt;/P&gt;
&lt;P&gt;...&lt;/P&gt;
&lt;P&gt;Can you please verify data again on your app, such as above example?&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;
&lt;P&gt;I'll do as you suggest in multi-thread mode :-)&lt;/P&gt;
&lt;P&gt;Thanks, Peter.&lt;/P&gt;
&lt;P&gt;Regards,&lt;/P&gt;
&lt;P&gt;Lina&lt;/P&gt;</description>
    <pubDate>Fri, 12 Mar 2010 13:25:47 GMT</pubDate>
    <dc:creator>linadaud</dc:creator>
    <dc:date>2010-03-12T13:25:47Z</dc:date>
    <item>
      <title>Total time vs. file duration</title>
      <link>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872171#M3896</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;
&lt;P&gt;I ran a profiling of a .mp4 file with the information below:&lt;/P&gt;
&lt;P&gt;Initial Object Descriptor ID 1&lt;BR /&gt;Duration &lt;B&gt;00:00:15&lt;/B&gt;&lt;BR /&gt;Service Handler: GPAC IsoMedia Reader&lt;BR /&gt;Service URL: C:\\Documents and Settings\\daud\\Bureau\\gpac\\test_sequences\\kangaroo.mp4&lt;BR /&gt;Scene Description: Width 400 - Height 300&lt;BR /&gt;Scene Codec GPAC BIFS Decoder&lt;/P&gt;
&lt;P&gt;However, in the call graph, I observed a total time of &lt;B&gt;9 677 035&lt;/B&gt; &lt;B&gt;s &lt;/B&gt;(in Windows via MS Visual Studio) and &lt;B&gt;10 069 015 s &lt;/B&gt;(in Ubuntu Linux) the main function!&lt;/P&gt;
&lt;P&gt;Could someone explain how can this be possible? How can I observe a total time that is faster than the file duration itself?&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Lina&lt;B&gt; &lt;/B&gt;&lt;/P&gt;</description>
      <pubDate>Thu, 11 Mar 2010 10:22:06 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872171#M3896</guid>
      <dc:creator>linadaud</dc:creator>
      <dc:date>2010-03-11T10:22:06Z</dc:date>
    </item>
    <item>
      <title>Total time vs. file duration</title>
      <link>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872172#M3897</link>
      <description>&lt;P&gt;Hi Lina,&lt;/P&gt;
&lt;P&gt;I think that Call graph will spend more time than application itself, since Call graph will call functions in itslibraries.&lt;/P&gt;
&lt;P&gt;You said that total time was from main function only, I suppose that is not enough! You have to count spending timefrom other threads.&lt;/P&gt;
&lt;P&gt;See example for application vtunedemo on Linux*:&lt;/P&gt;
&lt;P&gt;1. Run "time ./vtunedemo"&lt;/P&gt;
&lt;P&gt;real 0m0.287s (real execution time)&lt;/P&gt;
&lt;P&gt;&lt;STRONG&gt;user 0m0.812s&lt;/STRONG&gt; (user's time consumed in eight cores of my system,was partialllyparallelized)&lt;/P&gt;
&lt;P&gt;sys 0m0.002s&lt;/P&gt;
&lt;P&gt;2. Run Call graph for vtunedemo, total time -&lt;/P&gt;
&lt;P&gt;__libc_start_main (thread_0): 4,937,232 s&lt;/P&gt;
&lt;P&gt;main_thread (thread_1) : 1,226,465 s&lt;/P&gt;
&lt;P&gt;main_thread (thread_2) : 1,027,404 s&lt;/P&gt;
&lt;P&gt;main_thread (thread_3) : 1,037,240 s&lt;/P&gt;
&lt;P&gt;Therefore, total time = 8,228,341 s = &lt;STRONG&gt;8.22s&lt;/STRONG&gt;. It seems there are big overheads - but I have to deduct &lt;STRONG&gt;total wait time&lt;/STRONG&gt; from four threads. They are&lt;/P&gt;
&lt;P&gt;__libc_start_main (thread_0): 4,936,834 s&lt;/P&gt;
&lt;P&gt;main_thread (thread_1) : 749,068 s&lt;/P&gt;
&lt;P&gt;main_thread (thread_2) :746,968 s&lt;/P&gt;
&lt;P&gt;main_thread (thread_3) : 761,892 s&lt;/P&gt;
&lt;P&gt;Total wait time = 7,194,762 s.&lt;/P&gt;
&lt;P&gt;&lt;STRONG&gt;Total no-wait time&lt;/STRONG&gt; = 8,228,341 us - 7,194,762 s = 1,033,579 s = 1.03s&lt;/P&gt;
&lt;P&gt;So real overheads = 1.03s - 0.81s = 0.22s&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;Can you please verify data again on your app, such as above example?&lt;/P&gt;
&lt;P&gt;If you still have any concern, please let me know. Or you can submit a ticket with your example program to &lt;A href="https://premier.intel.com"&gt;https://premier.intel.com&lt;/A&gt;. Thank you.&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;Regards, Peter&lt;/P&gt;</description>
      <pubDate>Fri, 12 Mar 2010 06:56:06 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872172#M3897</guid>
      <dc:creator>Peter_W_Intel</dc:creator>
      <dc:date>2010-03-12T06:56:06Z</dc:date>
    </item>
    <item>
      <title>Total time vs. file duration</title>
      <link>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872173#M3898</link>
      <description>&lt;BLOCKQUOTE&gt;
&lt;P&gt;I think that Call graph will spend more time than application itself, since Call graph will call fun ctions in itslibraries.&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;
&lt;P&gt;I agree with you, Peter and I'm confused with the result observed from the profiling.&lt;/P&gt;
&lt;BLOCKQUOTE&gt;
&lt;P&gt;You said that total time was from main function only, I suppose that is not enough! You  have to count spending timefrom other threads.&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;
&lt;P&gt;I had a correct result when running in multi-thread. However, the result with total time shorter than the duration time was the one in mono-thread mode. Therefore, I think that observing the main total time would be enough since it corresponds to the application total time. Am I wrong?&lt;/P&gt;
&lt;BLOCKQUOTE&gt;
&lt;P&gt;See example for application vtunedemo on Linux*:&lt;/P&gt;
&lt;P&gt;...&lt;/P&gt;
&lt;P&gt;Can you please verify data again on your app, such as above example?&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;
&lt;P&gt;I'll do as you suggest in multi-thread mode :-)&lt;/P&gt;
&lt;P&gt;Thanks, Peter.&lt;/P&gt;
&lt;P&gt;Regards,&lt;/P&gt;
&lt;P&gt;Lina&lt;/P&gt;</description>
      <pubDate>Fri, 12 Mar 2010 13:25:47 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872173#M3898</guid>
      <dc:creator>linadaud</dc:creator>
      <dc:date>2010-03-12T13:25:47Z</dc:date>
    </item>
    <item>
      <title>Total time vs. file duration</title>
      <link>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872174#M3899</link>
      <description>&lt;P&gt;Peter,&lt;/P&gt;
&lt;P&gt;I ran the application as instructed. Below are the results:&lt;/P&gt;
&lt;P&gt;1. run time ./application&lt;/P&gt;
&lt;P&gt;real 15.95&lt;B&gt;&lt;BR /&gt;&lt;/B&gt;user 0.20&lt;BR /&gt;sys 0.65&lt;/P&gt;
&lt;P&gt;2. Run Call graph for the application (still in mono-threading)&lt;/P&gt;
&lt;P&gt;__libc_start_main = 10, 012, 504 s&lt;/P&gt;
&lt;P&gt;main = 10,012,458 s&lt;/P&gt;
I observed real execution time longer than the profiled time! How could this be?&lt;BR /&gt;</description>
      <pubDate>Fri, 12 Mar 2010 16:15:26 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872174#M3899</guid>
      <dc:creator>linadaud</dc:creator>
      <dc:date>2010-03-12T16:15:26Z</dc:date>
    </item>
    <item>
      <title>Total time vs. file duration</title>
      <link>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872175#M3900</link>
      <description>&lt;DIV id="tiny_quote"&gt;
&lt;DIV&gt;Quoting &lt;A class="basic" href="https://community.intel.com/en-us/profile/463998/" rel="/en-us/services/profile/quick_profile.php?is_paid=&amp;amp;user_id=463998"&gt;linadaud&lt;/A&gt;&lt;/DIV&gt;
&lt;DIV&gt;
&lt;P&gt;Peter,&lt;/P&gt;
&lt;P&gt;I ran the application as instructed. Below are the results:&lt;/P&gt;
&lt;P&gt;1. run time ./application&lt;/P&gt;
&lt;P&gt;real 15.95&lt;B&gt;&lt;BR /&gt;&lt;/B&gt;user 0.20&lt;BR /&gt;sys 0.65&lt;/P&gt;
&lt;P&gt;2. Run Call graph for the application (still in mono-threading)&lt;/P&gt;
&lt;P&gt;__libc_start_main = 10, 012, 504 s&lt;/P&gt;
&lt;P&gt;main = 10,012,458 s&lt;/P&gt;
I observed real execution time longer than the profiled time! How could this be?&lt;/DIV&gt;
&lt;/DIV&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;Ifyour application is single application, total wait time is almost "zero", main function is execuation time.&lt;/P&gt;
&lt;P&gt;Please submit a ticket to &lt;A href="https://premier.intel.com"&gt;https://premier.intel.com&lt;/A&gt; with you application (test case) to us for investigating ifthis is a Call graph bug.&lt;/P&gt;
&lt;P&gt;Remindto chekcif you are using VTune Analyzer v9.1 Update 7, which is latest product.&lt;/P&gt;
&lt;P&gt;Regards, Peter&lt;/P&gt;</description>
      <pubDate>Sat, 13 Mar 2010 01:41:54 GMT</pubDate>
      <guid>https://community.intel.com/t5/Analyzers/Total-time-vs-file-duration/m-p/872175#M3900</guid>
      <dc:creator>Peter_W_Intel</dc:creator>
      <dc:date>2010-03-13T01:41:54Z</dc:date>
    </item>
  </channel>
</rss>

