<?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 Feature requests in OpenCL* for CPU</title>
    <link>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817947#M1033</link>
    <description>Could you explain a bit about why you need the memory migration APIs?</description>
    <pubDate>Wed, 23 May 2012 12:41:01 GMT</pubDate>
    <dc:creator>Doron_S_Intel</dc:creator>
    <dc:date>2012-05-23T12:41:01Z</dc:date>
    <item>
      <title>Feature requests</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817946#M1032</link>
      <description>Just wanted to add two feature requests:&lt;BR /&gt;&lt;BR /&gt;1. FP64 support has already been mentioned and would be the most important request for me.&lt;BR /&gt;2. Memory migration APIs from OpenCL 1.2 would be very nice to have&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Tue, 22 May 2012 05:41:33 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817946#M1032</guid>
      <dc:creator>rahul_garg</dc:creator>
      <dc:date>2012-05-22T05:41:33Z</dc:date>
    </item>
    <item>
      <title>Feature requests</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817947#M1033</link>
      <description>Could you explain a bit about why you need the memory migration APIs?</description>
      <pubDate>Wed, 23 May 2012 12:41:01 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817947#M1033</guid>
      <dc:creator>Doron_S_Intel</dc:creator>
      <dc:date>2012-05-23T12:41:01Z</dc:date>
    </item>
    <item>
      <title>Feature requests</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817948#M1034</link>
      <description>Could you explain a bit about why you need the memory migration APIs?</description>
      <pubDate>Wed, 23 May 2012 12:41:04 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817948#M1034</guid>
      <dc:creator>Doron_S_Intel</dc:creator>
      <dc:date>2012-05-23T12:41:04Z</dc:date>
    </item>
    <item>
      <title>Feature requests</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817949#M1035</link>
      <description>Actually, disregard the memory migration API request. After looking at Intel's OpenCL optimization guide, I am now looking at avoiding copies altogether by allocating properly aligned memory :)&lt;BR /&gt;&lt;BR /&gt;To explain why I initially requested memory migration APIs: Consider a GPU device. Without memory migration APIs, some drivers don't provide hard guarantees about where exactly a memory object "lives", and only that it will be made available when you launch a kernel. The driver is otherwise free to swap stuff in/out of GPU memory. This causes my application to do somewhat unpredicatable memory copies between CPU and GPU on other platforms. However, on Ivy Bridge, it looks like I can avoid memory transfers altogether if the alignment conditions are met. &lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 24 May 2012 03:17:48 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/Feature-requests/m-p/817949#M1035</guid>
      <dc:creator>rahul_garg</dc:creator>
      <dc:date>2012-05-24T03:17:48Z</dc:date>
    </item>
  </channel>
</rss>

