<?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 IppiFilterBorderSpec in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181132#M27192</link>
    <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;I'm working on&amp;nbsp;a parallel working programm, which initializes the ippiFilterBorderSpec struct&amp;nbsp;using ippiFilterBorderInit_32f function in a other thread, than it is later used. When the initialization of all ippiFilterBorderSpecs is done (I use a arbitrary number), they were copyed at a syncronized place, so that the thread using ippiFilterBorder_32f_C1R function doesn't have to do the initialzation due to performance. After any parameter change the ippiFilterBorderSpecs&amp;nbsp;were new initialized and I see, that the content of the ippiFilterBorderSpec struct differs on two places, although the same input parameters were used. The differences appear at&amp;nbsp;byte &amp;nbsp;80 and byte 96 in the structure, counting from 0. I don't understand why they should differ? Can somebody please tell me more about the composition of the ippiFilterBorderSpec?&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thank &amp;amp; regards,&lt;/P&gt;

&lt;P&gt;Hans&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Tue, 09 Jan 2018 11:28:25 GMT</pubDate>
    <dc:creator>Bock__Thomas</dc:creator>
    <dc:date>2018-01-09T11:28:25Z</dc:date>
    <item>
      <title>IppiFilterBorderSpec</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181132#M27192</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;I'm working on&amp;nbsp;a parallel working programm, which initializes the ippiFilterBorderSpec struct&amp;nbsp;using ippiFilterBorderInit_32f function in a other thread, than it is later used. When the initialization of all ippiFilterBorderSpecs is done (I use a arbitrary number), they were copyed at a syncronized place, so that the thread using ippiFilterBorder_32f_C1R function doesn't have to do the initialzation due to performance. After any parameter change the ippiFilterBorderSpecs&amp;nbsp;were new initialized and I see, that the content of the ippiFilterBorderSpec struct differs on two places, although the same input parameters were used. The differences appear at&amp;nbsp;byte &amp;nbsp;80 and byte 96 in the structure, counting from 0. I don't understand why they should differ? Can somebody please tell me more about the composition of the ippiFilterBorderSpec?&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thank &amp;amp; regards,&lt;/P&gt;

&lt;P&gt;Hans&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Tue, 09 Jan 2018 11:28:25 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181132#M27192</guid>
      <dc:creator>Bock__Thomas</dc:creator>
      <dc:date>2018-01-09T11:28:25Z</dc:date>
    </item>
    <item>
      <title>Hi Hans.</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181133#M27193</link>
      <description>&lt;P&gt;Hi Hans.&lt;/P&gt;

&lt;P&gt;Are your pointers provided to ippiFilterBorder_32f_C1R aligned on 64 byte?&amp;nbsp;It is not necessary because ippiFilterBorderGetSize requests memory size with reserve for such alignment. But in this case contents of two specs may differ.&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thanks for your question.&lt;/P&gt;</description>
      <pubDate>Mon, 22 Jan 2018 13:43:53 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181133#M27193</guid>
      <dc:creator>Andrey_B_Intel</dc:creator>
      <dc:date>2018-01-22T13:43:53Z</dc:date>
    </item>
    <item>
      <title>Maybe IppiFilterBorderSpec​</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181134#M27194</link>
      <description>&lt;P&gt;Maybe&amp;nbsp;IppiFilterBorderSpec​ has internal pointers, like IppiResizeSpec. Don't memcpy the struct, just share the pointer.&lt;/P&gt;</description>
      <pubDate>Mon, 22 Jan 2018 13:49:25 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181134#M27194</guid>
      <dc:creator>BMart1</dc:creator>
      <dc:date>2018-01-22T13:49:25Z</dc:date>
    </item>
    <item>
      <title> A "spec" structure can</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181135#M27195</link>
      <description>&lt;P&gt;&amp;nbsp;A "spec" structure can shared by the different calls if they are having the same parameters.&amp;nbsp; so it does not need to create another copy.&lt;BR /&gt;
	&lt;BR /&gt;
	Thanks,&lt;BR /&gt;
	Chao&lt;/P&gt;</description>
      <pubDate>Tue, 23 Jan 2018 02:03:26 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IppiFilterBorderSpec/m-p/1181135#M27195</guid>
      <dc:creator>Chao_Y_Intel</dc:creator>
      <dc:date>2018-01-23T02:03:26Z</dc:date>
    </item>
  </channel>
</rss>

