<?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 Bruno, in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/stride-misalignment-on-purpose/m-p/1161039#M26573</link>
    <description>&lt;P&gt;Hi Bruno,&lt;/P&gt;

&lt;P&gt;​Do you mean the cache bank conflict, generally the architecture and software were discussed in IA&amp;nbsp;software developer manual:&lt;/P&gt;

&lt;P&gt;&lt;A href="https://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-optimization-manual.pdf" target="_blank"&gt;https://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-optimization-manual.pdf&lt;/A&gt;&lt;BR /&gt;
	&lt;BR /&gt;
	section of 3.6.8.&lt;/P&gt;

&lt;P&gt;&lt;A href="https://software.intel.com/en-us/forums/software-tuning-performance-optimization-platform-monitoring/topic/280663" target="_blank"&gt;https://software.intel.com/en-us/forums/software-tuning-performance-optimization-platform-monitoring/topic/280663&lt;/A&gt;&lt;BR /&gt;
	&lt;BR /&gt;
	Best Regards,&lt;BR /&gt;
	&lt;BR /&gt;
	Ying&lt;/P&gt;</description>
    <pubDate>Thu, 23 Nov 2017 05:42:43 GMT</pubDate>
    <dc:creator>Ying_H_Intel</dc:creator>
    <dc:date>2017-11-23T05:42:43Z</dc:date>
    <item>
      <title>stride misalignment on purpose</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/stride-misalignment-on-purpose/m-p/1161038#M26572</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;I remember reading a whitepaper I can't now find that argued that you shouldn't over-align, specially for Pentium 4. Image stride should be a multiple of 64 but not 128, to prevent columns fighting for the same cache lines. Otherwise most of the cache is useless. For example, if a 32KB cache is 4-way associative and each line 64 bytes long, only 4x64 = 256 bytes are used.&lt;/P&gt;

&lt;P&gt;iw doesn't take this precaution. Maybe it should? Any pointers to read more on this and why Pentium 4 had it worse than other CPUs?&lt;/P&gt;

&lt;P&gt;Bruno&lt;/P&gt;</description>
      <pubDate>Tue, 21 Nov 2017 15:35:45 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/stride-misalignment-on-purpose/m-p/1161038#M26572</guid>
      <dc:creator>BMart1</dc:creator>
      <dc:date>2017-11-21T15:35:45Z</dc:date>
    </item>
    <item>
      <title>Hi Bruno,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/stride-misalignment-on-purpose/m-p/1161039#M26573</link>
      <description>&lt;P&gt;Hi Bruno,&lt;/P&gt;

&lt;P&gt;​Do you mean the cache bank conflict, generally the architecture and software were discussed in IA&amp;nbsp;software developer manual:&lt;/P&gt;

&lt;P&gt;&lt;A href="https://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-optimization-manual.pdf" target="_blank"&gt;https://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-optimization-manual.pdf&lt;/A&gt;&lt;BR /&gt;
	&lt;BR /&gt;
	section of 3.6.8.&lt;/P&gt;

&lt;P&gt;&lt;A href="https://software.intel.com/en-us/forums/software-tuning-performance-optimization-platform-monitoring/topic/280663" target="_blank"&gt;https://software.intel.com/en-us/forums/software-tuning-performance-optimization-platform-monitoring/topic/280663&lt;/A&gt;&lt;BR /&gt;
	&lt;BR /&gt;
	Best Regards,&lt;BR /&gt;
	&lt;BR /&gt;
	Ying&lt;/P&gt;</description>
      <pubDate>Thu, 23 Nov 2017 05:42:43 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/stride-misalignment-on-purpose/m-p/1161039#M26573</guid>
      <dc:creator>Ying_H_Intel</dc:creator>
      <dc:date>2017-11-23T05:42:43Z</dc:date>
    </item>
  </channel>
</rss>

