<?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 OpenCL GPU compiler fails on valid code in OpenCL* for CPU</title>
    <link>https://community.intel.com/t5/OpenCL-for-CPU/OpenCL-GPU-compiler-fails-on-valid-code/m-p/1051068#M4033</link>
    <description>&lt;P&gt;This code fails to compile when using Intel SDK for OpenCL Applications 2014 with an Intel HD 4000 GPU:&lt;/P&gt;

&lt;PRE style="font-family:Consolas;font-size:13;color:black;background:white;"&gt;&lt;SPAN style="color:blue;"&gt;kernel&lt;/SPAN&gt;&amp;nbsp;&lt;SPAN style="color:blue;"&gt;void&lt;/SPAN&gt;&amp;nbsp;f1(&lt;SPAN style="color:blue;"&gt;global&lt;/SPAN&gt;&amp;nbsp;&lt;SPAN style="color:blue;"&gt;double&lt;/SPAN&gt;*&amp;nbsp;W)
{
&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&lt;SPAN style="color:blue;"&gt;const&lt;/SPAN&gt;&amp;nbsp;&lt;SPAN style="color:blue;"&gt;double&lt;/SPAN&gt;&amp;nbsp;t&amp;nbsp;=&amp;nbsp;1.;
&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;min(W[0],&amp;nbsp;t);
}&lt;/PRE&gt;

&lt;P&gt;It complains about the call to min being ambiguous. Changing it to fmin fixes the compilation. The same code compiles fine when targeting the CPU device (Core i7-3820QM), or an NVIDIA GPU.&lt;/P&gt;</description>
    <pubDate>Tue, 16 Sep 2014 13:56:19 GMT</pubDate>
    <dc:creator>Lubomir_A_</dc:creator>
    <dc:date>2014-09-16T13:56:19Z</dc:date>
    <item>
      <title>OpenCL GPU compiler fails on valid code</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/OpenCL-GPU-compiler-fails-on-valid-code/m-p/1051068#M4033</link>
      <description>&lt;P&gt;This code fails to compile when using Intel SDK for OpenCL Applications 2014 with an Intel HD 4000 GPU:&lt;/P&gt;

&lt;PRE style="font-family:Consolas;font-size:13;color:black;background:white;"&gt;&lt;SPAN style="color:blue;"&gt;kernel&lt;/SPAN&gt;&amp;nbsp;&lt;SPAN style="color:blue;"&gt;void&lt;/SPAN&gt;&amp;nbsp;f1(&lt;SPAN style="color:blue;"&gt;global&lt;/SPAN&gt;&amp;nbsp;&lt;SPAN style="color:blue;"&gt;double&lt;/SPAN&gt;*&amp;nbsp;W)
{
&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&lt;SPAN style="color:blue;"&gt;const&lt;/SPAN&gt;&amp;nbsp;&lt;SPAN style="color:blue;"&gt;double&lt;/SPAN&gt;&amp;nbsp;t&amp;nbsp;=&amp;nbsp;1.;
&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;min(W[0],&amp;nbsp;t);
}&lt;/PRE&gt;

&lt;P&gt;It complains about the call to min being ambiguous. Changing it to fmin fixes the compilation. The same code compiles fine when targeting the CPU device (Core i7-3820QM), or an NVIDIA GPU.&lt;/P&gt;</description>
      <pubDate>Tue, 16 Sep 2014 13:56:19 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/OpenCL-GPU-compiler-fails-on-valid-code/m-p/1051068#M4033</guid>
      <dc:creator>Lubomir_A_</dc:creator>
      <dc:date>2014-09-16T13:56:19Z</dc:date>
    </item>
    <item>
      <title>GPU device does not support</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/OpenCL-GPU-compiler-fails-on-valid-code/m-p/1051069#M4034</link>
      <description>&lt;P&gt;GPU device does not support cl_khr_fp64 currently, so switching to fmin will only help you so much. You should probably switch to floats instead of doubles on the GPU device.&lt;/P&gt;</description>
      <pubDate>Thu, 09 Oct 2014 01:58:03 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/OpenCL-GPU-compiler-fails-on-valid-code/m-p/1051069#M4034</guid>
      <dc:creator>Robert_I_Intel</dc:creator>
      <dc:date>2014-10-09T01:58:03Z</dc:date>
    </item>
  </channel>
</rss>

