<?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 Thank you for reporting this in Intel® Fortran Compiler</title>
    <link>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970671#M97077</link>
    <description>&lt;P&gt;Thank you for reporting this issue and for the convenient reproducer. I reported it to Development (see internal tracking id below).&lt;/P&gt;

&lt;P&gt;I found the previous 13.1 compiler produces the same result.&amp;nbsp;If needed, as a work around, one can NULLIFY(p) explicitly after the deallocation.&lt;/P&gt;

&lt;P&gt;(Internal tracking id: DPD200249493)&lt;BR /&gt;
	&lt;STRONG&gt;(Resolution Update on 09/16/2014):&lt;/STRONG&gt; This defect is fixed in the Intel® Parallel Studio XE 2015 Initial Release (2015.0.090 - Linux)&lt;/P&gt;</description>
    <pubDate>Thu, 31 Oct 2013 08:15:00 GMT</pubDate>
    <dc:creator>Kevin_D_Intel</dc:creator>
    <dc:date>2013-10-31T08:15:00Z</dc:date>
    <item>
      <title>14.0 Compiler generates incorrect results</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970669#M97075</link>
      <description>&lt;P&gt;Under certain circumstances, a pointer will remain associated, according to ASSOCIATED(), after it is deallocated. &amp;nbsp;I've attached a 17-line example that exhibits this incorrect behavior with the new 14.0 compiler.&lt;/P&gt;

&lt;P&gt;[Are we not allowed to attach files anymore?]&lt;/P&gt;

&lt;P&gt;[fortran]&lt;/P&gt;

&lt;P&gt;!!&lt;/P&gt;

&lt;P&gt;!! COMPILER GENERATES INCORRECT CODE&lt;/P&gt;

&lt;P&gt;!!&lt;BR /&gt;
	!! In the following example, a derived-type scalar pointer P is allocated and&lt;BR /&gt;
	!! then deallocated. At that point, ASSOCIATED(P) should return FALSE but it&lt;BR /&gt;
	!! returns TRUE. Key to the error seems to be that the pointer is a dummy&lt;BR /&gt;
	!! argument and that the derived type contains an allocatable CLASS(*) component.&lt;BR /&gt;
	!!&lt;BR /&gt;
	!! $ ifort --version&lt;BR /&gt;
	!! ifort (IFORT) 14.0.0 20130728&lt;BR /&gt;
	!!&lt;BR /&gt;
	!! $ ifort intel-bug-20131030.f90&lt;BR /&gt;
	!! $ ./a.out&lt;BR /&gt;
	!! associated(p) = T (should be FALSE)&lt;BR /&gt;
	!!&lt;/P&gt;

&lt;P&gt;module foo_mod&lt;BR /&gt;
	type, public :: foo&lt;BR /&gt;
	class(*), allocatable :: value&lt;BR /&gt;
	end type&lt;BR /&gt;
	contains&lt;BR /&gt;
	subroutine sub (p)&lt;BR /&gt;
	type(foo), pointer :: p&lt;BR /&gt;
	allocate(p)&lt;BR /&gt;
	deallocate(p)&lt;BR /&gt;
	print *, 'associated(p) =', associated(p), '(should be FALSE)'&lt;BR /&gt;
	end subroutine&lt;BR /&gt;
	end module&lt;/P&gt;

&lt;P&gt;program main&lt;BR /&gt;
	use foo_mod&lt;BR /&gt;
	type(foo), pointer :: p&lt;BR /&gt;
	call sub (p)&lt;BR /&gt;
	end program[/fortran]&lt;/P&gt;</description>
      <pubDate>Thu, 31 Oct 2013 05:29:46 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970669#M97075</guid>
      <dc:creator>Neil_Carlson</dc:creator>
      <dc:date>2013-10-31T05:29:46Z</dc:date>
    </item>
    <item>
      <title>Here's the example</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970670#M97076</link>
      <description>&lt;P&gt;Here's the example&lt;/P&gt;</description>
      <pubDate>Thu, 31 Oct 2013 05:33:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970670#M97076</guid>
      <dc:creator>Neil_Carlson</dc:creator>
      <dc:date>2013-10-31T05:33:00Z</dc:date>
    </item>
    <item>
      <title>Thank you for reporting this</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970671#M97077</link>
      <description>&lt;P&gt;Thank you for reporting this issue and for the convenient reproducer. I reported it to Development (see internal tracking id below).&lt;/P&gt;

&lt;P&gt;I found the previous 13.1 compiler produces the same result.&amp;nbsp;If needed, as a work around, one can NULLIFY(p) explicitly after the deallocation.&lt;/P&gt;

&lt;P&gt;(Internal tracking id: DPD200249493)&lt;BR /&gt;
	&lt;STRONG&gt;(Resolution Update on 09/16/2014):&lt;/STRONG&gt; This defect is fixed in the Intel® Parallel Studio XE 2015 Initial Release (2015.0.090 - Linux)&lt;/P&gt;</description>
      <pubDate>Thu, 31 Oct 2013 08:15:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970671#M97077</guid>
      <dc:creator>Kevin_D_Intel</dc:creator>
      <dc:date>2013-10-31T08:15:00Z</dc:date>
    </item>
    <item>
      <title>I confirmed this defect is</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970672#M97078</link>
      <description>&lt;P&gt;I confirmed this defect is fixed in our next major release (15.0 compiler) due out later this year.&lt;/P&gt;</description>
      <pubDate>Wed, 18 Jun 2014 10:59:04 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/14-0-Compiler-generates-incorrect-results/m-p/970672#M97078</guid>
      <dc:creator>Kevin_D_Intel</dc:creator>
      <dc:date>2014-06-18T10:59:04Z</dc:date>
    </item>
  </channel>
</rss>

