<?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 Thanks all. I am interested in Software Archive</title>
    <link>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015724#M36005</link>
    <description>&lt;P&gt;Thanks all. I am interested in Blender (with OpenCL enabled) and LuxRender. I didn't want to specify too early in this conversation since sometimes it's hard to get broad information when specifics are provided. I already know that nobody has tried this yet, and am trying to get enough information here to ask the right questions.&lt;BR /&gt;
	&lt;BR /&gt;
	I know that Lux will "see" the Xeon processors, but as for if lux can see openCL nodes that are "accelerators" I don't know.&lt;/P&gt;

&lt;P&gt;Either way, it seems like a relatively easy thing to do. I wonder if anyone at the Blender Foundation has a promo card yet.&lt;/P&gt;</description>
    <pubDate>Tue, 09 Dec 2014 01:48:25 GMT</pubDate>
    <dc:creator>Shawn_K_</dc:creator>
    <dc:date>2014-12-09T01:48:25Z</dc:date>
    <item>
      <title>Xeon Phi under OpenCL</title>
      <link>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015721#M36002</link>
      <description>&lt;P&gt;I am curious if the Xeon Phi coprocessors, specifically if they can be used "out of the box" in existing OpenCL applications that support the Xeon OpenCL Framework?&lt;/P&gt;

&lt;P&gt;I use software that uses OpenCL, and this software recognizes&amp;nbsp;the&amp;nbsp;Xeons in my workstation. I was wondering if the same would go for the Phi without recoding.&lt;/P&gt;</description>
      <pubDate>Mon, 08 Dec 2014 19:06:05 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015721#M36002</guid>
      <dc:creator>Shawn_K_</dc:creator>
      <dc:date>2014-12-08T19:06:05Z</dc:date>
    </item>
    <item>
      <title>Hello,</title>
      <link>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015722#M36003</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;First, make sure that your system supports the Intel (R) Xeon Phi(TM) coprocessor.&lt;/P&gt;

&lt;P&gt;If you use the Intel(R) SDK for OpenCL Applications (&lt;A href="https://software.intel.com/en-us/intel-opencl"&gt;https://software.intel.com/en-us/intel-opencl&lt;/A&gt;), and the MPSS version 3.3 (&lt;A href="https://software.intel.com/en-us/articles/intel-manycore-platform-software-stack-mpss-archive"&gt;https://software.intel.com/en-us/articles/intel-manycore-platform-software-stack-mpss-archive&lt;/A&gt;), then your application should recognize Intel Xeon Phi coprocessor as an OpenCL node.&lt;/P&gt;

&lt;P&gt;.&lt;/P&gt;</description>
      <pubDate>Mon, 08 Dec 2014 19:28:23 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015722#M36003</guid>
      <dc:creator>Loc_N_Intel</dc:creator>
      <dc:date>2014-12-08T19:28:23Z</dc:date>
    </item>
    <item>
      <title>I'd like to add one extra</title>
      <link>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015723#M36004</link>
      <description>&lt;P&gt;I'd like to add one extra comment:&lt;/P&gt;

&lt;P&gt;Most OpenCL code that I've seen thus far has trouble selecting the right cpu or card; some older code is NVidia biased, some code is AMD biased.&lt;/P&gt;

&lt;P&gt;Also, the Xeon Phi is not registered as a GPU but as an ACCELERATOR, this confuses some simple OpenCL samples that I have seen as well. Luckily, only a very minor change is usually needed to get the code to recognize and use the Xeon Phi. And, wiht *proper* coding the Phi will work out of the box - but without knowing what type of application you're running it is impossible to tell if it's properly coded ;)&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 08 Dec 2014 20:12:27 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015723#M36004</guid>
      <dc:creator>JJK</dc:creator>
      <dc:date>2014-12-08T20:12:27Z</dc:date>
    </item>
    <item>
      <title>Thanks all. I am interested</title>
      <link>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015724#M36005</link>
      <description>&lt;P&gt;Thanks all. I am interested in Blender (with OpenCL enabled) and LuxRender. I didn't want to specify too early in this conversation since sometimes it's hard to get broad information when specifics are provided. I already know that nobody has tried this yet, and am trying to get enough information here to ask the right questions.&lt;BR /&gt;
	&lt;BR /&gt;
	I know that Lux will "see" the Xeon processors, but as for if lux can see openCL nodes that are "accelerators" I don't know.&lt;/P&gt;

&lt;P&gt;Either way, it seems like a relatively easy thing to do. I wonder if anyone at the Blender Foundation has a promo card yet.&lt;/P&gt;</description>
      <pubDate>Tue, 09 Dec 2014 01:48:25 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015724#M36005</guid>
      <dc:creator>Shawn_K_</dc:creator>
      <dc:date>2014-12-09T01:48:25Z</dc:date>
    </item>
    <item>
      <title>I have installed my Xeon Phi</title>
      <link>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015725#M36006</link>
      <description>&lt;P&gt;I have installed my Xeon Phi card on CentOS 6.6 successfully. I can log into the device using "ssh mic0" and micinfo returns correct values.&lt;/P&gt;

&lt;P&gt;However, the OpenCL runtime crashes upon building my kernel program for this device. When selecting the cpu device it builds the program successfuly. MPSS is running and the board is updated with the latest firmware. Can you please advise?&lt;/P&gt;

&lt;PRE class="brush:bash;"&gt;sudo service mpss status
mpss is running
&lt;/PRE&gt;</description>
      <pubDate>Wed, 05 Aug 2015 23:30:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015725#M36006</guid>
      <dc:creator>PouryaShirazian</dc:creator>
      <dc:date>2015-08-05T23:30:00Z</dc:date>
    </item>
    <item>
      <title>your question is slightly off</title>
      <link>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015726#M36007</link>
      <description>&lt;P&gt;your question is slightly off-topic , compared to the original question in this thread.&lt;/P&gt;

&lt;P&gt;I have seen the same thing you're seeing , however, and never found a solution for it. It has little to do with the Xeon Phi itself, but more with the OpenCL runtime for the Xeon Phi.&lt;/P&gt;

&lt;P&gt;A larger problem is that Intel's support for OpenCL on the Xeon Phi seems to have ended: the last version of the MPSS stack for which OpenCL was supported was mpss 3.3 , and that version is no longer supported by Intel. It would be nice to hear from Intel what their current standpoint is on OpenCL for the Phi :)&lt;/P&gt;</description>
      <pubDate>Thu, 06 Aug 2015 21:10:30 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/Xeon-Phi-under-OpenCL/m-p/1015726#M36007</guid>
      <dc:creator>JJK</dc:creator>
      <dc:date>2015-08-06T21:10:30Z</dc:date>
    </item>
  </channel>
</rss>

