<?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 vdSqrt (64-bit Windows) throws illegal instruction on certain C in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798166#M2834</link>
    <description>It looks like the older version selected the wrong instruction set, due to mistake in identifying your CPU. That bug would not be specific to vdSqrt.</description>
    <pubDate>Wed, 16 Jun 2010 19:00:55 GMT</pubDate>
    <dc:creator>TimP</dc:creator>
    <dc:date>2010-06-16T19:00:55Z</dc:date>
    <item>
      <title>vdSqrt (64-bit Windows) throws illegal instruction on certain CPUS (MKL Update 4)</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798164#M2832</link>
      <description>Every call to 64-bit vdSqrt throws an illegal instruction on an Intel U7300 CPU ( a ULV core 2 duo variant ).&lt;BR /&gt;Is this fixed in Update 5?&lt;BR /&gt;The 32-bit version of MKL has no such problem, nor does the problem surface on a number of 64-bit desktop Intel CPU variants we use for QA.&lt;BR /&gt;&lt;PRE&gt;[bash]&lt;BR /&gt;e.g. &lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;	double a[N_SQRT];&lt;BR /&gt;	for (int i=0;i&lt;N_SQRT&gt;=i*i;&lt;BR /&gt;	double b[N_SQRT];&lt;BR /&gt;	vdSqrt(4,a,b);&lt;BR /&gt;[/bash]&lt;/N_SQRT&gt;&lt;/PRE&gt; &lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Tue, 15 Jun 2010 18:12:20 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798164#M2832</guid>
      <dc:creator>AndrewC</dc:creator>
      <dc:date>2010-06-15T18:12:20Z</dc:date>
    </item>
    <item>
      <title>vdSqrt (64-bit Windows) throws illegal instruction on certain C</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798165#M2833</link>
      <description>I downloaded MKL Update 5 and this bug has been fixed. It would be nice if the Update 5 release notes had mentioned this - there are some references to VML problems being fixed, but no specific reference to vdSqrt</description>
      <pubDate>Tue, 15 Jun 2010 21:47:03 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798165#M2833</guid>
      <dc:creator>AndrewC</dc:creator>
      <dc:date>2010-06-15T21:47:03Z</dc:date>
    </item>
    <item>
      <title>vdSqrt (64-bit Windows) throws illegal instruction on certain C</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798166#M2834</link>
      <description>It looks like the older version selected the wrong instruction set, due to mistake in identifying your CPU. That bug would not be specific to vdSqrt.</description>
      <pubDate>Wed, 16 Jun 2010 19:00:55 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798166#M2834</guid>
      <dc:creator>TimP</dc:creator>
      <dc:date>2010-06-16T19:00:55Z</dc:date>
    </item>
    <item>
      <title>vdSqrt (64-bit Windows) throws illegal instruction on certain C</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798167#M2835</link>
      <description>Tim right,&lt;DIV&gt;this is MKL's dispatching ptoblem "&lt;B&gt;Illegal instruction Intel Core2 Duo CPU E8400&lt;/B&gt;". It has been disscussed &lt;A href=" Illegal instruction Intel Core2 Duo CPU E840"&gt;here&lt;/A&gt;. The problem has been fixed in Update5.&lt;/DIV&gt;&lt;DIV&gt;We memtioned about that in MKL 10.2 Fix list as&lt;/DIV&gt;&lt;DIV&gt;&lt;TABLE width="700" border="0" cellpadding="0" cellspacing="0"&gt;&lt;TBODY&gt;&lt;TR&gt;&lt;TD&gt;&lt;A href="http://software.intel.com/en-us/forums/showthread.php?t=71647"&gt;DPD200090101&lt;/A&gt;&lt;/TD&gt;&lt;TD&gt;Illegal instruction Intel Core2 Duo CPU E8400 for VML function&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;/TD&gt;&lt;/TR&gt;&lt;/TBODY&gt;&lt;/TABLE&gt;&lt;/DIV&gt;&lt;DIV&gt;--Gennady&lt;/DIV&gt;</description>
      <pubDate>Thu, 17 Jun 2010 10:40:34 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798167#M2835</guid>
      <dc:creator>Gennady_F_Intel</dc:creator>
      <dc:date>2010-06-17T10:40:34Z</dc:date>
    </item>
    <item>
      <title>vdSqrt (64-bit Windows) throws illegal instruction on certain C</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798168#M2836</link>
      <description>The bug fix description should have said something like , in big red letters!!&lt;BR /&gt;&lt;BR /&gt;"Many (all?) VML functions on (many types of ?) Intel Core2 Duo CPUs may fail with illegal instruction"&lt;BR /&gt;&lt;BR /&gt;You will note my problem machine is not a E8400&lt;BR /&gt;</description>
      <pubDate>Thu, 17 Jun 2010 22:14:26 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/vdSqrt-64-bit-Windows-throws-illegal-instruction-on-certain-CPUS/m-p/798168#M2836</guid>
      <dc:creator>AndrewC</dc:creator>
      <dc:date>2010-06-17T22:14:26Z</dc:date>
    </item>
  </channel>
</rss>

