<?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 Hi Marc, in OpenCL* for CPU</title>
    <link>https://community.intel.com/t5/OpenCL-for-CPU/How-does-Code-Builder-measures-time/m-p/1000341#M2853</link>
    <description>&lt;P&gt;Hi Marc,&lt;/P&gt;

&lt;P&gt;In Code Builder we give 4 execution times:&lt;/P&gt;

&lt;P&gt;1. Total - measured by calculating the host time (machine time) before the&amp;nbsp;clEnqueueNDRangeKernel and after the clFinish.&lt;/P&gt;

&lt;P&gt;2. Queue - measured by querying the clEvent object we get from&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;clEnqueueNDRangeKernel&amp;nbsp;&lt;/SPAN&gt;for&amp;nbsp;CL_PROFILING_COMMAND_QUEUED,&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;CL_PROFILING_COMMAND_SUBMIT&amp;nbsp;and calculating the difference&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;3. Submit -&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;measured by querying the clEvent object we get from&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;clEnqueueNDRangeKernel&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;for&amp;nbsp;&lt;/SPAN&gt;CL_PROFILING_COMMAND_SUBMIT,&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;CL_PROFILING_COMMAND_START&amp;nbsp;and calculating the difference&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;4. Run -&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;measured by querying the clEvent object we get from&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;clEnqueueNDRangeKernel&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;for&amp;nbsp;&lt;/SPAN&gt;CL_PROFILING_COMMAND_START,&amp;nbsp;CL_PROFILING_COMMAND_END&amp;nbsp;and calculating the difference&lt;/P&gt;

&lt;P&gt;we do that for every iteration and for every configuration we run and then calculate the average, median, minimum, maximum and standard deviation.&lt;/P&gt;

&lt;P&gt;I don't think the time measuring method is documented anywhere, do you think we should do it?&lt;/P&gt;</description>
    <pubDate>Mon, 20 Apr 2015 08:38:26 GMT</pubDate>
    <dc:creator>Arik_Z_Intel1</dc:creator>
    <dc:date>2015-04-20T08:38:26Z</dc:date>
    <item>
      <title>How does Code Builder measures time ?</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/How-does-Code-Builder-measures-time/m-p/1000340#M2852</link>
      <description>&lt;P&gt;Hi everyone,&lt;/P&gt;

&lt;P&gt;I'm willing to know how exactly the Intel Code Builder tool measures time through execution, is there any documentation on this ? I can't find any.&lt;/P&gt;

&lt;P&gt;Thanks !&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 30 Mar 2015 13:58:36 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/How-does-Code-Builder-measures-time/m-p/1000340#M2852</guid>
      <dc:creator>Marc_M_</dc:creator>
      <dc:date>2015-03-30T13:58:36Z</dc:date>
    </item>
    <item>
      <title>Hi Marc,</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/How-does-Code-Builder-measures-time/m-p/1000341#M2853</link>
      <description>&lt;P&gt;Hi Marc,&lt;/P&gt;

&lt;P&gt;In Code Builder we give 4 execution times:&lt;/P&gt;

&lt;P&gt;1. Total - measured by calculating the host time (machine time) before the&amp;nbsp;clEnqueueNDRangeKernel and after the clFinish.&lt;/P&gt;

&lt;P&gt;2. Queue - measured by querying the clEvent object we get from&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;clEnqueueNDRangeKernel&amp;nbsp;&lt;/SPAN&gt;for&amp;nbsp;CL_PROFILING_COMMAND_QUEUED,&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;CL_PROFILING_COMMAND_SUBMIT&amp;nbsp;and calculating the difference&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;3. Submit -&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;measured by querying the clEvent object we get from&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;clEnqueueNDRangeKernel&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;for&amp;nbsp;&lt;/SPAN&gt;CL_PROFILING_COMMAND_SUBMIT,&amp;nbsp;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;CL_PROFILING_COMMAND_START&amp;nbsp;and calculating the difference&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;4. Run -&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;measured by querying the clEvent object we get from&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;clEnqueueNDRangeKernel&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN style="font-size: 13.0080003738403px; line-height: 19.5120010375977px;"&gt;for&amp;nbsp;&lt;/SPAN&gt;CL_PROFILING_COMMAND_START,&amp;nbsp;CL_PROFILING_COMMAND_END&amp;nbsp;and calculating the difference&lt;/P&gt;

&lt;P&gt;we do that for every iteration and for every configuration we run and then calculate the average, median, minimum, maximum and standard deviation.&lt;/P&gt;

&lt;P&gt;I don't think the time measuring method is documented anywhere, do you think we should do it?&lt;/P&gt;</description>
      <pubDate>Mon, 20 Apr 2015 08:38:26 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/How-does-Code-Builder-measures-time/m-p/1000341#M2853</guid>
      <dc:creator>Arik_Z_Intel1</dc:creator>
      <dc:date>2015-04-20T08:38:26Z</dc:date>
    </item>
    <item>
      <title>Hi ARIK Z.,</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/How-does-Code-Builder-measures-time/m-p/1000342#M2854</link>
      <description>&lt;P&gt;Hi &lt;A href="https://software.intel.com/en-us/user/950204"&gt;ARIK Z.,&lt;/A&gt;&lt;/P&gt;

&lt;P&gt;Thank you for your answer, in fact what I wanted to know was an idea about the hardware/software mechanism by which the time is measured and retrieved.&lt;/P&gt;

&lt;P&gt;I've read the whole documentation about Code Builder and I think it could be a good idea to explain briefly how the measurements are done. Because yes we have time measurements but we don't know how they are done by Intel's implementation.&lt;/P&gt;</description>
      <pubDate>Mon, 20 Apr 2015 11:36:35 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/How-does-Code-Builder-measures-time/m-p/1000342#M2854</guid>
      <dc:creator>Marc_M_</dc:creator>
      <dc:date>2015-04-20T11:36:35Z</dc:date>
    </item>
  </channel>
</rss>

