<?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 Hello Adiaan van Os, in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1132291#M25809</link>
    <description>&lt;P&gt;Hello Adiaan van Os,&lt;/P&gt;&lt;P&gt;Thanks for your proposal.&lt;/P&gt;&lt;P&gt;For the cache usage, does the link below meet your expectation?&lt;/P&gt;&lt;P&gt;&lt;A href="https://software.intel.com/en-us/ipp-dev-guide-cache-optimizations" target="_blank"&gt;https://software.intel.com/en-us/ipp-dev-guide-cache-optimizations&lt;/A&gt;&lt;/P&gt;&lt;P&gt;And, if possible, could you help create ticket for the feature request or enhancement in our online service center through&amp;nbsp;https://supporttickets.intel.com/?lang=en-US&lt;/P&gt;&lt;P&gt;Best Regards,&lt;/P&gt;&lt;P&gt;Ruqiu&lt;/P&gt;</description>
    <pubDate>Thu, 27 Feb 2020 03:05:56 GMT</pubDate>
    <dc:creator>Ruqiu_C_Intel</dc:creator>
    <dc:date>2020-02-27T03:05:56Z</dc:date>
    <item>
      <title>Composite ippi functions</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1132290#M25808</link>
      <description>&lt;P&gt;When I need a sequence of ippi functions for a composite operation I could, of course, call those ippi functions in sequence. However, for&amp;nbsp;large images&amp;nbsp;that would&amp;nbsp;be a rather bad usage of the CPU&amp;nbsp;Ln caches.&lt;/P&gt;&lt;P&gt;Another solution is to create a composite "ippi" function&amp;nbsp;that calls for each row the equivalent series of ipps functions. An added advantage is that this can be threaded easily, by giving each thread its own set of rows.&lt;/P&gt;&lt;P&gt;If the RowBytes (StepSize) of the image(s) equal(s) the PixelSize multiplied by the ImageWidth, we can work per chunk rather than per row. Let's assume that&amp;nbsp;sysctlbyname with "hw.l1dcachesize" returns the&amp;nbsp;L1 cache per core.&amp;nbsp;The chunk size can be chosen to be that size for&amp;nbsp;in-place&amp;nbsp;operations or half that size&amp;nbsp;for operations from a source to a target.&lt;/P&gt;&lt;P&gt;Is that optimal ? Or what does Intel recommend ?&lt;/P&gt;&lt;P&gt;Regards,&lt;/P&gt;&lt;P&gt;Adriaan van Os&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 26 Feb 2020 10:05:03 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1132290#M25808</guid>
      <dc:creator>Adriaan_van_Os</dc:creator>
      <dc:date>2020-02-26T10:05:03Z</dc:date>
    </item>
    <item>
      <title>Hello Adiaan van Os,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1132291#M25809</link>
      <description>&lt;P&gt;Hello Adiaan van Os,&lt;/P&gt;&lt;P&gt;Thanks for your proposal.&lt;/P&gt;&lt;P&gt;For the cache usage, does the link below meet your expectation?&lt;/P&gt;&lt;P&gt;&lt;A href="https://software.intel.com/en-us/ipp-dev-guide-cache-optimizations" target="_blank"&gt;https://software.intel.com/en-us/ipp-dev-guide-cache-optimizations&lt;/A&gt;&lt;/P&gt;&lt;P&gt;And, if possible, could you help create ticket for the feature request or enhancement in our online service center through&amp;nbsp;https://supporttickets.intel.com/?lang=en-US&lt;/P&gt;&lt;P&gt;Best Regards,&lt;/P&gt;&lt;P&gt;Ruqiu&lt;/P&gt;</description>
      <pubDate>Thu, 27 Feb 2020 03:05:56 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1132291#M25809</guid>
      <dc:creator>Ruqiu_C_Intel</dc:creator>
      <dc:date>2020-02-27T03:05:56Z</dc:date>
    </item>
    <item>
      <title>I am not aware that my</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1132292#M25810</link>
      <description>&lt;P&gt;I am not aware that my question implies a feature request or proposal. And I did read the page you refer to, of course.&lt;/P&gt;&lt;P&gt;Regards,&lt;/P&gt;&lt;P&gt;Adriaan van Os&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 27 Feb 2020 08:43:53 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1132292#M25810</guid>
      <dc:creator>Adriaan_van_Os</dc:creator>
      <dc:date>2020-02-27T08:43:53Z</dc:date>
    </item>
    <item>
      <title>Re: Composite ippi functions</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1280463#M27678</link>
      <description>&lt;P class="sub_section_element_selectors"&gt;&lt;SPAN class="sub_section_element_selectors"&gt;We will put the feature request in our developer plan if receive more similar request.&amp;nbsp;Or users can raise a new feature request through Intel online service center&amp;nbsp;&lt;A class="sub_section_element_selectors" href="https://supporttickets.intel.com/servicecenter?lang=en-US" target="_blank" rel="nofollow noopener noreferrer"&gt;https://supporttickets.intel.com/servicecenter?lang=en-US&lt;/A&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="sub_section_element_selectors"&gt;&lt;SPAN class="sub_section_element_selectors"&gt;For this thread, we will no longer respond anymore.&amp;nbsp;&lt;/SPAN&gt;&lt;SPAN class="sub_section_element_selectors"&gt;Any further interaction in this thread will be considered community only. &lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="sub_section_element_selectors"&gt;&lt;SPAN class="sub_section_element_selectors"&gt;Thanks&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Tue, 25 May 2021 04:57:36 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Composite-ippi-functions/m-p/1280463#M27678</guid>
      <dc:creator>Ruqiu_C_Intel</dc:creator>
      <dc:date>2021-05-25T04:57:36Z</dc:date>
    </item>
  </channel>
</rss>

