<?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 Write combining in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/Write-combining/m-p/1067280#M24384</link>
    <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;Does ipp read from the bytes it writes? I'm contemplating resizing directly to graphics memory. It's fine to write to GPU memory, but slow to read from it. Can ipp output to write only memory? Does it overwrite previously written bytes?&lt;/P&gt;

&lt;P&gt;Bruno&lt;/P&gt;</description>
    <pubDate>Mon, 27 Mar 2017 13:33:34 GMT</pubDate>
    <dc:creator>BMart1</dc:creator>
    <dc:date>2017-03-27T13:33:34Z</dc:date>
    <item>
      <title>Write combining</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Write-combining/m-p/1067280#M24384</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;Does ipp read from the bytes it writes? I'm contemplating resizing directly to graphics memory. It's fine to write to GPU memory, but slow to read from it. Can ipp output to write only memory? Does it overwrite previously written bytes?&lt;/P&gt;

&lt;P&gt;Bruno&lt;/P&gt;</description>
      <pubDate>Mon, 27 Mar 2017 13:33:34 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Write-combining/m-p/1067280#M24384</guid>
      <dc:creator>BMart1</dc:creator>
      <dc:date>2017-03-27T13:33:34Z</dc:date>
    </item>
    <item>
      <title>Hi Bruno,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Write-combining/m-p/1067281#M24385</link>
      <description>&lt;P&gt;Hi Bruno,&lt;/P&gt;

&lt;P&gt;This may happen, so I wouldn't rely on this. Why you don't use temporary image store and, then copy temporary image to graphics memory?&lt;/P&gt;

&lt;P&gt;It won't take too much of time.&lt;/P&gt;</description>
      <pubDate>Tue, 28 Mar 2017 05:10:48 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Write-combining/m-p/1067281#M24385</guid>
      <dc:creator>Sergey_K_Intel</dc:creator>
      <dc:date>2017-03-28T05:10:48Z</dc:date>
    </item>
    <item>
      <title>The problem with copying is</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/Write-combining/m-p/1067282#M24386</link>
      <description>&lt;P&gt;The problem with copying is that I need to tile considering cache sizes. If I just do one pass it's always optimum.&lt;/P&gt;

&lt;P&gt;Is CopyManaged bettr for write combining memory?&lt;/P&gt;

&lt;P&gt;Bruno&lt;/P&gt;</description>
      <pubDate>Tue, 28 Mar 2017 13:49:57 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/Write-combining/m-p/1067282#M24386</guid>
      <dc:creator>BMart1</dc:creator>
      <dc:date>2017-03-28T13:49:57Z</dc:date>
    </item>
  </channel>
</rss>

