<?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 The EU is like a mini-cpu in OpenCL* for CPU</title>
    <link>https://community.intel.com/t5/OpenCL-for-CPU/EU-and-Sampler/m-p/955301#M2102</link>
    <description>&lt;P&gt;The EU is like a mini-cpu capable of performing a given operation on a given set of data. EUs can have multiple threads that help to keep the EU busy during a high latency operation. Some times, depending on what the program (shader/kernel) is doing, EUs require a texture read, for example. This is done by the sampling engine, which fetches the data (surface) from memory and filter based on filtering mode. While the EU is waiting for the sampler to return the data (called a stall), the EU can switch processing to another thread. When the data arrives, it can switch back to the original thread.&lt;/P&gt;

&lt;P&gt;Hope this high level explanation is clear.&lt;/P&gt;

&lt;P&gt;Thanks,&lt;BR /&gt;
	Raghu&lt;/P&gt;</description>
    <pubDate>Mon, 24 Mar 2014 20:58:09 GMT</pubDate>
    <dc:creator>Raghupathi_M_Intel</dc:creator>
    <dc:date>2014-03-24T20:58:09Z</dc:date>
    <item>
      <title>EU and Sampler</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/EU-and-Sampler/m-p/955300#M2101</link>
      <description>&lt;P&gt;Hi,&amp;nbsp;&lt;/P&gt;

&lt;P&gt;I am doing GPGPU computing for video codec applications.&amp;nbsp;&lt;/P&gt;

&lt;P&gt;The algorithm adopted may use bi-linear filter which can take advantage of image sampler in OpenCL&lt;/P&gt;

&lt;P&gt;However, I am not clear &amp;nbsp;however it works using underlying hardware,&lt;/P&gt;

&lt;P&gt;say if I use sampler, will it be possible that the sampler and EU execute at the same time?&lt;/P&gt;

&lt;P&gt;Can anyone elaborate EU and sampler's difference/relationship?&lt;/P&gt;

&lt;P&gt;&lt;SPAN style="font-size: 1em; line-height: 1.5;"&gt;Thanks&amp;nbsp;&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Fri, 21 Mar 2014 09:54:08 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/EU-and-Sampler/m-p/955300#M2101</guid>
      <dc:creator>Biao_W_</dc:creator>
      <dc:date>2014-03-21T09:54:08Z</dc:date>
    </item>
    <item>
      <title>The EU is like a mini-cpu</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/EU-and-Sampler/m-p/955301#M2102</link>
      <description>&lt;P&gt;The EU is like a mini-cpu capable of performing a given operation on a given set of data. EUs can have multiple threads that help to keep the EU busy during a high latency operation. Some times, depending on what the program (shader/kernel) is doing, EUs require a texture read, for example. This is done by the sampling engine, which fetches the data (surface) from memory and filter based on filtering mode. While the EU is waiting for the sampler to return the data (called a stall), the EU can switch processing to another thread. When the data arrives, it can switch back to the original thread.&lt;/P&gt;

&lt;P&gt;Hope this high level explanation is clear.&lt;/P&gt;

&lt;P&gt;Thanks,&lt;BR /&gt;
	Raghu&lt;/P&gt;</description>
      <pubDate>Mon, 24 Mar 2014 20:58:09 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/EU-and-Sampler/m-p/955301#M2102</guid>
      <dc:creator>Raghupathi_M_Intel</dc:creator>
      <dc:date>2014-03-24T20:58:09Z</dc:date>
    </item>
    <item>
      <title>Quote:Raghu Muthyalampalli</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/EU-and-Sampler/m-p/955302#M2103</link>
      <description>&lt;P&gt;&lt;/P&gt;&lt;BLOCKQUOTE&gt;Raghu Muthyalampalli (Intel) wrote:&lt;BR /&gt;&lt;P&gt;&lt;/P&gt;

&lt;P&gt;The EU is like a mini-cpu capable of performing a given operation on a given set of data. EUs can have multiple threads that help to keep the EU busy during a high latency operation. Some times, depending on what the program (shader/kernel) is doing, EUs require a texture read, for example. This is done by the sampling engine, which fetches the data (surface) from memory and filter based on filtering mode. While the EU is waiting for the sampler to return the data (called a stall), the EU can switch processing to another thread. When the data arrives, it can switch back to the original thread.&lt;/P&gt;

&lt;P&gt;Hope this high level explanation is clear.&lt;/P&gt;

&lt;P&gt;Thanks,&lt;BR /&gt;
	Raghu&lt;/P&gt;

&lt;P&gt;&lt;/P&gt;&lt;/BLOCKQUOTE&gt;&lt;P&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thanks for your explanation. I understand from the high level but would like to have a more detailed read.&lt;/P&gt;

&lt;P&gt;Do you have something whitepaper introduce the sampling engine, Intel GPU in particular?&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 26 Mar 2014 21:26:58 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/EU-and-Sampler/m-p/955302#M2103</guid>
      <dc:creator>Biao_W_</dc:creator>
      <dc:date>2014-03-26T21:26:58Z</dc:date>
    </item>
  </channel>
</rss>

