<?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 Sorry, I finally found what's in OpenCL* for CPU</title>
    <link>https://community.intel.com/t5/OpenCL-for-CPU/Memory-issue-with-HD4000-GPU/m-p/947436#M1920</link>
    <description>&lt;P&gt;Sorry, I finally found what's went wrong with my program:&lt;/P&gt;
&lt;P&gt;I'm using events to profile the exucution time of the kernel (in a loop) and I release the event object just one time=&amp;gt;a memory leakage occurs, to avoid this issue I must release the event object each time I call&amp;nbsp;&amp;nbsp;clEnqueueNDRangeKernel(...........,ev);&lt;/P&gt;
&lt;P&gt;thx,&lt;/P&gt;
&lt;P&gt;Mohamed&lt;/P&gt;</description>
    <pubDate>Mon, 18 Feb 2013 17:01:38 GMT</pubDate>
    <dc:creator>Mohamed_Amine_BERGAC</dc:creator>
    <dc:date>2013-02-18T17:01:38Z</dc:date>
    <item>
      <title>Memory issue with HD4000 GPU</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/Memory-issue-with-HD4000-GPU/m-p/947435#M1919</link>
      <description>&lt;P&gt;I have a memory issue with the HD4000 GPU: I use a simple kernel that performs 2 simple operations, I run it over hundred of times (in a host &amp;nbsp;loop), before each call I release all device and host buffers like this:&lt;/P&gt;
&lt;P&gt;ret = clReleaseMemObject(bufin1_mem_obj);&lt;BR /&gt; ret = clReleaseMemObject(bufin2_mem_obj);&lt;BR /&gt; ret = clReleaseMemObject(bufout1_mem_obj);&lt;BR /&gt; ret = clReleaseMemObject(bufout2_mem_obj);&lt;BR /&gt; ret = clReleaseMemObject(debugbuf_mem_obj);&lt;BR /&gt; ret = clReleaseKernel(kernel);&lt;BR /&gt; ret = clReleaseProgram(program);&lt;BR /&gt; ret = clReleaseEvent(ev);&lt;BR /&gt; ret = clFlush(command_queue);&lt;BR /&gt; ret = clFinish(command_queue);&lt;BR /&gt; ret = clReleaseCommandQueue(command_queue);&lt;BR /&gt; ret = clReleaseContext(context);&lt;/P&gt;
&lt;P&gt;free(bufin1);&lt;BR /&gt; free(bufin2);&lt;BR /&gt; free(buf_out_1);&lt;BR /&gt; free(buf_out_2);&lt;BR /&gt; free(debug_buf);&lt;BR /&gt; free(perfs);&lt;BR /&gt; free(times);&lt;BR /&gt; free(logTitle);&lt;BR /&gt; free(logline);&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;the problem is, when I use the same code with an AMD GPU, I have no memory problem, but when I run it on the HD4000, I have a memory leakage after 22 kernel calls, it seems that the GPU doesn't freed memory.&lt;/P&gt;
&lt;P&gt;is it a known issue ?&lt;/P&gt;
&lt;P&gt;THX,&lt;/P&gt;
&lt;P&gt;Mohamed&lt;/P&gt;</description>
      <pubDate>Mon, 18 Feb 2013 14:55:28 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/Memory-issue-with-HD4000-GPU/m-p/947435#M1919</guid>
      <dc:creator>Mohamed_Amine_BERGAC</dc:creator>
      <dc:date>2013-02-18T14:55:28Z</dc:date>
    </item>
    <item>
      <title>Sorry, I finally found what's</title>
      <link>https://community.intel.com/t5/OpenCL-for-CPU/Memory-issue-with-HD4000-GPU/m-p/947436#M1920</link>
      <description>&lt;P&gt;Sorry, I finally found what's went wrong with my program:&lt;/P&gt;
&lt;P&gt;I'm using events to profile the exucution time of the kernel (in a loop) and I release the event object just one time=&amp;gt;a memory leakage occurs, to avoid this issue I must release the event object each time I call&amp;nbsp;&amp;nbsp;clEnqueueNDRangeKernel(...........,ev);&lt;/P&gt;
&lt;P&gt;thx,&lt;/P&gt;
&lt;P&gt;Mohamed&lt;/P&gt;</description>
      <pubDate>Mon, 18 Feb 2013 17:01:38 GMT</pubDate>
      <guid>https://community.intel.com/t5/OpenCL-for-CPU/Memory-issue-with-HD4000-GPU/m-p/947436#M1920</guid>
      <dc:creator>Mohamed_Amine_BERGAC</dc:creator>
      <dc:date>2013-02-18T17:01:38Z</dc:date>
    </item>
  </channel>
</rss>

