<?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 Hello, guys. in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944200#M17978</link>
    <description>&lt;P&gt;Hello, guys.&lt;/P&gt;

&lt;P&gt;I have one more question relates to above theme. :)&amp;nbsp;&lt;/P&gt;

&lt;P&gt;If I use memory aligning, how can I define memory with real data and trash memory (allocated to align 32/64 boundary)? Are there some helper functions to define neccessary and trash mem? Are other functions know about the trash memory? It seems that this done by nStepbytes parameter, isn't it?&lt;/P&gt;

&lt;BLOCKQUOTE&gt;
	&lt;P&gt;const int nHeight = 15, nWidth = 8;&lt;/P&gt;

	&lt;P&gt;IppiSize tsWholePic = {nWidth, nHeight};&lt;/P&gt;

	&lt;P&gt;&amp;nbsp;&lt;/P&gt;

	&lt;P&gt;Ipp32f pSrc[nHeight * nWidth];&lt;/P&gt;

	&lt;P&gt;ippiSet_32f_C1R (2., pSrc, nWidth * sizeof (Ipp32f), tsWholePic);&lt;/P&gt;

	&lt;P&gt;&amp;nbsp;&lt;/P&gt;

	&lt;P&gt;int nDstStepBytes = 0;&lt;/P&gt;

	&lt;P&gt;Ipp32f* pDst = ippiMalloc_32f_C1R (nWidth, nHeight, &amp;amp;nDstStepBytes); /&lt;STRONG&gt;/ In due of mem align pDst has trash memory parts. See pic&lt;/STRONG&gt;&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;ippiCopy_32f_C1R (pSrc, nWidth * sizeof (Ipp32f), pDst, nDstStepBytes, tsWholePic);&lt;STRONG&gt; // Is this right copying? Step bytes are different &amp;nbsp;// for pSrc and pDst&lt;/STRONG&gt;&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;Regards,&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;Mark&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;&amp;nbsp;&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;&amp;nbsp;&lt;/BLOCKQUOTE&gt;</description>
    <pubDate>Tue, 24 Dec 2013 10:00:06 GMT</pubDate>
    <dc:creator>m_a_</dc:creator>
    <dc:date>2013-12-24T10:00:06Z</dc:date>
    <item>
      <title>(ippiMalloc_XXX | malloc) &amp; ippiFilterGauss*</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944196#M17974</link>
      <description>&lt;P&gt;Hello.&lt;/P&gt;

&lt;P&gt;Do I need to allocate memory for arrays &lt;STRONG&gt;only&lt;/STRONG&gt; with ippiMalloc (it allocates 32-bytes aligned memory) or not?&lt;/P&gt;

&lt;P&gt;Does using other memory allocating functions (malloc, ...) affects to performance?&lt;/P&gt;

&lt;P&gt;Code:&lt;/P&gt;

&lt;BLOCKQUOTE&gt;
	&lt;P&gt;IppStatus GaussFilter (Ipp32f* pSrc, const int nWidth, const int nHeight, const Ipp32f fSigma, Ipp32f* pDst)&lt;/P&gt;

	&lt;P&gt;{&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; int nKernelSize = 7;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; IppiSize tWholePic = {nWidth, nHeight};&lt;/P&gt;

	&lt;P&gt;&amp;nbsp;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; int nStepBytes = 0;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; pDst = ippiMalloc_32f_C1 (nWidth, nHeight, &amp;amp;nStepBytes);&lt;/P&gt;

	&lt;P&gt;&amp;nbsp;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; int nBorderBufferSize = 0;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; Ippi8u* pBorderBuffer = ippiFilterGaussGetBufferSetSize_32f_C1R (tWholePic, nKernelSize, &amp;amp;nBorderBufferSize);&lt;/P&gt;

	&lt;P&gt;&amp;nbsp;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; ippiFilterGaussBorder_32f_C1R (pSrc, nWidth * sizeof (Ipp32f)&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;, pDst, nWidth * sizeof (Ipp32f) &amp;nbsp;// &lt;STRONG&gt;Have I use this one or nStepBytes receipt from ippiMalloc?&lt;/STRONG&gt;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;, tWholePic&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;, nKernelSize, fSigma, ippBorderRepl, 0.&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;, pBorderBuffer);&lt;/P&gt;

	&lt;P&gt;}&lt;/P&gt;

	&lt;P&gt;void tmain(...)&lt;/P&gt;

	&lt;P&gt;{&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; int nWidth = 12, nHeight = 15;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; Ipp32f fSigma = 1.;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; Ipp32f pSrc[nWidth * nHeight] ;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; Ipp32f* pDst = NULL;&lt;/P&gt;

	&lt;P&gt;&amp;nbsp;// Initialize pSrc&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; GaussFilter (pSrc, nWidth, nHeight, pDst);&lt;/P&gt;

	&lt;P&gt;// Do something with pDst.&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; if (pDst != NULL)&lt;/P&gt;

	&lt;P&gt;&amp;nbsp; &amp;nbsp; ippFree (pDst);&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;
	&lt;P&gt;}&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;

&lt;P&gt;Regards,&lt;/P&gt;

&lt;P&gt;Mark&lt;/P&gt;

&lt;BLOCKQUOTE&gt;
	&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;</description>
      <pubDate>Mon, 23 Dec 2013 11:15:17 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944196#M17974</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2013-12-23T11:15:17Z</dc:date>
    </item>
    <item>
      <title>Hello,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944197#M17975</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Other malloc functions also work. ippsMalloc is actually calling the system malloc function, and&amp;nbsp;make the memory 32 bit/64bit alignment.&amp;nbsp; From the performance point, it is the better if the input data is address is 32bit or 64 bit alignment( for the machine support AVX instructions).&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thanks,&lt;BR /&gt;
	Chao&lt;/P&gt;</description>
      <pubDate>Tue, 24 Dec 2013 03:31:18 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944197#M17975</guid>
      <dc:creator>Chao_Y_Intel</dc:creator>
      <dc:date>2013-12-24T03:31:18Z</dc:date>
    </item>
    <item>
      <title>Hi,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944198#M17976</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;In the source code, where you ask "Have I use...", you need to use nStepBytes, because the step is&amp;nbsp;not always equal to nWidth*sizeof. Otherwise, there is a risk of missing of memory alignment benefits.&lt;/P&gt;

&lt;P&gt;Regards,&lt;BR /&gt;
	Sergey&lt;/P&gt;</description>
      <pubDate>Tue, 24 Dec 2013 04:14:50 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944198#M17976</guid>
      <dc:creator>Sergey_K_Intel</dc:creator>
      <dc:date>2013-12-24T04:14:50Z</dc:date>
    </item>
    <item>
      <title>Thanks.
 
regards,
Mark.</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944199#M17977</link>
      <description>&lt;P&gt;Thanks.&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;regards,&lt;/P&gt;

&lt;P&gt;Mark.&lt;/P&gt;</description>
      <pubDate>Tue, 24 Dec 2013 05:51:03 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944199#M17977</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2013-12-24T05:51:03Z</dc:date>
    </item>
    <item>
      <title>Hello, guys.</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944200#M17978</link>
      <description>&lt;P&gt;Hello, guys.&lt;/P&gt;

&lt;P&gt;I have one more question relates to above theme. :)&amp;nbsp;&lt;/P&gt;

&lt;P&gt;If I use memory aligning, how can I define memory with real data and trash memory (allocated to align 32/64 boundary)? Are there some helper functions to define neccessary and trash mem? Are other functions know about the trash memory? It seems that this done by nStepbytes parameter, isn't it?&lt;/P&gt;

&lt;BLOCKQUOTE&gt;
	&lt;P&gt;const int nHeight = 15, nWidth = 8;&lt;/P&gt;

	&lt;P&gt;IppiSize tsWholePic = {nWidth, nHeight};&lt;/P&gt;

	&lt;P&gt;&amp;nbsp;&lt;/P&gt;

	&lt;P&gt;Ipp32f pSrc[nHeight * nWidth];&lt;/P&gt;

	&lt;P&gt;ippiSet_32f_C1R (2., pSrc, nWidth * sizeof (Ipp32f), tsWholePic);&lt;/P&gt;

	&lt;P&gt;&amp;nbsp;&lt;/P&gt;

	&lt;P&gt;int nDstStepBytes = 0;&lt;/P&gt;

	&lt;P&gt;Ipp32f* pDst = ippiMalloc_32f_C1R (nWidth, nHeight, &amp;amp;nDstStepBytes); /&lt;STRONG&gt;/ In due of mem align pDst has trash memory parts. See pic&lt;/STRONG&gt;&lt;/P&gt;
&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;ippiCopy_32f_C1R (pSrc, nWidth * sizeof (Ipp32f), pDst, nDstStepBytes, tsWholePic);&lt;STRONG&gt; // Is this right copying? Step bytes are different &amp;nbsp;// for pSrc and pDst&lt;/STRONG&gt;&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;Regards,&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;Mark&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;&amp;nbsp;&lt;/BLOCKQUOTE&gt;

&lt;BLOCKQUOTE&gt;&amp;nbsp;&lt;/BLOCKQUOTE&gt;</description>
      <pubDate>Tue, 24 Dec 2013 10:00:06 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944200#M17978</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2013-12-24T10:00:06Z</dc:date>
    </item>
    <item>
      <title>Mark,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944201#M17979</link>
      <description>&lt;P&gt;Mark,&lt;/P&gt;

&lt;P&gt;That's correct. Any "step_bytes" parameter in IPP&amp;nbsp;image processing function defines how many bytes to add to the beginning of previous image row to position to the beginning of next image row. So, "nWidth*sizeof(Ipp32f)" and "nDstStepBytes" both are correct as src and dst steps.&lt;/P&gt;

&lt;P&gt;Regards,&lt;BR /&gt;
	Sergey&lt;/P&gt;</description>
      <pubDate>Tue, 24 Dec 2013 10:22:57 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944201#M17979</guid>
      <dc:creator>Sergey_K_Intel</dc:creator>
      <dc:date>2013-12-24T10:22:57Z</dc:date>
    </item>
    <item>
      <title>I see. Payment for speed and</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944202#M17980</link>
      <description>&lt;P&gt;I see. Payment for speed and comfort. :)&lt;/P&gt;

&lt;P&gt;Merry Christmass,&lt;/P&gt;

&lt;P&gt;Thanks a lot,&lt;/P&gt;

&lt;P&gt;Mark&lt;/P&gt;</description>
      <pubDate>Wed, 25 Dec 2013 03:06:49 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944202#M17980</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2013-12-25T03:06:49Z</dc:date>
    </item>
    <item>
      <title>hello</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944203#M17981</link>
      <description>&lt;P&gt;hello&lt;/P&gt;

&lt;P&gt;I see. Thanks a lot.&lt;/P&gt;

&lt;P&gt;Merry Christmas, guys (yesterday I couldn't add post to the forum, something happened with site.)&lt;/P&gt;

&lt;P&gt;Regards,&lt;/P&gt;

&lt;P&gt;Mark&lt;/P&gt;</description>
      <pubDate>Thu, 26 Dec 2013 02:48:19 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944203#M17981</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2013-12-26T02:48:19Z</dc:date>
    </item>
    <item>
      <title>Ok, thanks a lot.</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944204#M17982</link>
      <description>&lt;P&gt;Ok, thanks a lot.&lt;/P&gt;

&lt;P&gt;regards,&lt;/P&gt;

&lt;P&gt;Mark&lt;/P&gt;</description>
      <pubDate>Thu, 26 Dec 2013 02:54:37 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944204#M17982</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2013-12-26T02:54:37Z</dc:date>
    </item>
    <item>
      <title>Hello,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944205#M17983</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;As I understand, in &lt;STRONG&gt;ippiMallocated&lt;/STRONG&gt; structures any row in image is aligned to 32/64 border, so bytes are added to the end of the previous row. And the situation seems follow: for&lt;STRONG&gt; static and dynamic (allocated with malloc)&lt;/STRONG&gt; I can use&lt;STRONG&gt; nWidth * sizeof (Type_Of_Array)&lt;/STRONG&gt;. For &lt;STRONG&gt;dynamic arrays, allocated with ippiMalloc&lt;/STRONG&gt; I have to use &lt;STRONG&gt;nStepbytes&lt;/STRONG&gt;.&lt;/P&gt;

&lt;P&gt;schema in ippiMallocated array&lt;/P&gt;

&lt;P&gt;&lt;S&gt;xxxx&lt;/S&gt;1............a............b............c............&lt;S&gt;xxxxxxxxxx&lt;/S&gt;2............a............b............c............i&lt;S&gt;xxxxxxxxxx&lt;/S&gt;&lt;/P&gt;

&lt;P&gt;1, 2 - address in memory aligned to 32/64&lt;/P&gt;

&lt;P&gt;&lt;S&gt;xxxxx&lt;/S&gt; - trash memory, added to align&lt;/P&gt;

&lt;P&gt;2 - 1 = &lt;STRONG&gt;nStepBytes&lt;/STRONG&gt;,&lt;/P&gt;

&lt;P&gt;&amp;amp;c - 1 =&lt;STRONG&gt; nWidth * sizeof(Type_Of_Array)&lt;/STRONG&gt;&lt;/P&gt;

&lt;P&gt;regards,&lt;/P&gt;

&lt;P&gt;Mark&lt;/P&gt;</description>
      <pubDate>Thu, 26 Dec 2013 02:59:32 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944205#M17983</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2013-12-26T02:59:32Z</dc:date>
    </item>
    <item>
      <title>Hello,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944206#M17984</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;

&lt;P&gt;Thanks a lot.&lt;/P&gt;

&lt;P&gt;Regards,&lt;/P&gt;

&lt;P&gt;Mark&lt;/P&gt;</description>
      <pubDate>Fri, 27 Dec 2013 03:23:38 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944206#M17984</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2013-12-27T03:23:38Z</dc:date>
    </item>
    <item>
      <title>Mark,</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944207#M17985</link>
      <description>&lt;P&gt;Mark,&lt;/P&gt;

&lt;P&gt;There are aligned mallocs in various OSes (_align_malloc, posix_memalign and others), but they provide only alignment of the very first byte of allocated memory, whereas in image processing the beginnings of each image line should be aligned for better performance.&lt;/P&gt;

&lt;P&gt;Regards,&lt;BR /&gt;
	Sergey&lt;/P&gt;</description>
      <pubDate>Fri, 17 Jan 2014 10:51:04 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944207#M17985</guid>
      <dc:creator>Sergey_K_Intel</dc:creator>
      <dc:date>2014-01-17T10:51:04Z</dc:date>
    </item>
    <item>
      <title>I see. I have made a lot of</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944208#M17986</link>
      <description>&lt;P&gt;I see. I have made a lot of tests and discovered this feature of ippiMalloc. :)&amp;nbsp;&lt;BR /&gt;
	&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Thank a lot.&lt;/P&gt;

&lt;P&gt;best regards,&lt;/P&gt;

&lt;P&gt;Mark.&lt;BR /&gt;
	&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 20 Jan 2014 10:46:24 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/ippiMalloc-XXX-malloc-ippiFilterGauss/m-p/944208#M17986</guid>
      <dc:creator>m_a_</dc:creator>
      <dc:date>2014-01-20T10:46:24Z</dc:date>
    </item>
  </channel>
</rss>

