<?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 Re: ifx 2024.2.1 produces wrong code when using pFUnit in Intel® Fortran Compiler</title>
    <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638463#M173994</link>
    <description>&lt;P&gt;Something went wrong in the ctest step&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 98%] Building Fortran object tests/funit-core/CMakeFiles/new_tests.x.dir/Test_TestMethod_before.F90.o&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Building Fortran object tests/funit-core/CMakeFiles/new_tests.x.dir/new_tests.x_driver.F90.o&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Linking Fortran executable new_tests.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Built target new_tests.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Built target build-tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Constructing a list of tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Done constructing a list of tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Updating test list for fixtures&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Added 0 tests to meet fixture requirements&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Checking test dependency graph...&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Checking test dependency graph end&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;No tests were found!!!&lt;/SPAN&gt;&lt;/P&gt;</description>
    <pubDate>Mon, 21 Oct 2024 18:01:10 GMT</pubDate>
    <dc:creator>Ron_Green</dc:creator>
    <dc:date>2024-10-21T18:01:10Z</dc:date>
    <item>
      <title>ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1637192#M173962</link>
      <description>&lt;P&gt;&lt;A href="https://github.com/Goddard-Fortran-Ecosystem/pFUnit" target="_self"&gt;pFUnit&lt;/A&gt; is a &lt;SPAN&gt;unit testing framework for Fortran. Consider the following minimal example using this&amp;nbsp;framework (two files in the same directory):&lt;/SPAN&gt;&lt;/P&gt;&lt;UL&gt;&lt;LI&gt;&lt;SPAN&gt;`m.pf`:&lt;/SPAN&gt;&lt;/LI&gt;&lt;/UL&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;LI-CODE lang="fortran"&gt;module m
   use FUnit
   type tt
      integer,allocatable::a(:),b(:)  ! works without `b`
   contains
      procedure::f
   end type
contains
   subroutine f(o)
      class(tt)o
      if(allocated(o%a))deallocate(o%a)
      allocate(o%a(1))  ! `ifx` produces code that considers this array as allocated here
   end
   &lt;a href="https://community.intel.com/t5/user/viewprofilepage/user-id/46752"&gt;@Test&lt;/a&gt;
   subroutine s()
      class(tt),allocatable::o  ! works with `type` instead of `class`
      o=tt()
      call o%f()  ! works with `f(o)` instead of `o%f()`
   end
end&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;UL&gt;&lt;LI&gt;`CMakeLists.txt`:&lt;/LI&gt;&lt;/UL&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;LI-CODE lang="markup"&gt;cmake_minimum_required(VERSION 3.12)
project(test Fortran)
include(CTest)
include(FetchContent)
set(FETCHCONTENT_BASE_DIR ${CMAKE_BINARY_DIR}/pFUnit)
FetchContent_Declare(
        PFUNIT
        GIT_REPOSITORY "https://github.com/Goddard-Fortran-Ecosystem/pFUnit"
        GIT_TAG v4.8.0
)
FetchContent_MakeAvailable(PFUNIT)
add_pFUnit_ctest(
    funit_test
    TEST_SOURCES m.pf
    LABELS test
)&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;The test can be built and executed with the following commands (`ifx` 2024.2.1):&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;LI-CODE lang="bash"&gt;mkdir build
cd build
FC=ifx cmake ..
cmake --build .
ctest -L test -V&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;The test will fail with an error `forrtl: severe (151): allocatable array is already allocated`,&amp;nbsp; because the executable thinks that `o%a` is allocated at line 12 in `m.pf`, even though `allocated(o%a)` was `.false.` at the previous line. I think, it is a bug of `ifx`, because this error doesn't happen, when using `gfortran` instead, and it can be fixed by changing seemingly unrelated parts of code (like removing `b` from `tt` or changing `class(tt),allocatable` to `type(tt),allocatable`).&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Unfortunately, I could not reduce the example further, since I don't know how the pFUnit code works.&lt;/P&gt;</description>
      <pubDate>Mon, 14 Oct 2024 21:51:33 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1637192#M173962</guid>
      <dc:creator>V-T</dc:creator>
      <dc:date>2024-10-14T21:51:33Z</dc:date>
    </item>
    <item>
      <title>Re: ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638463#M173994</link>
      <description>&lt;P&gt;Something went wrong in the ctest step&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 98%] Building Fortran object tests/funit-core/CMakeFiles/new_tests.x.dir/Test_TestMethod_before.F90.o&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Building Fortran object tests/funit-core/CMakeFiles/new_tests.x.dir/new_tests.x_driver.F90.o&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Linking Fortran executable new_tests.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Built target new_tests.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Built target build-tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Constructing a list of tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Done constructing a list of tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Updating test list for fixtures&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Added 0 tests to meet fixture requirements&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Checking test dependency graph...&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Checking test dependency graph end&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;No tests were found!!!&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Mon, 21 Oct 2024 18:01:10 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638463#M173994</guid>
      <dc:creator>Ron_Green</dc:creator>
      <dc:date>2024-10-21T18:01:10Z</dc:date>
    </item>
    <item>
      <title>Re: ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638467#M173995</link>
      <description>&lt;P&gt;same with a NAG fortran run&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;$ ctest -L test -V&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;UpdateCTestConfiguration&lt;SPAN class="Apple-converted-space"&gt;&amp;nbsp; &lt;/SPAN&gt;from :/nfs/pdx/proj/icl/devshare/rwgreen/pFUnit/pFUnit/build/DartConfiguration.tcl&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt; Add coverage exclude regular expressions.&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;UpdateCTestConfiguration&lt;SPAN class="Apple-converted-space"&gt;&amp;nbsp; &lt;/SPAN&gt;from :/nfs/pdx/proj/icl/devshare/rwgreen/pFUnit/pFUnit/build/DartConfiguration.tcl&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Test project /nfs/pdx/proj/icl/devshare/rwgreen/pFUnit/pFUnit/build&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Run command: /usr/bin/cmake --build . --config "Release" --target "build-tests"&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[&lt;SPAN class="Apple-converted-space"&gt;&amp;nbsp; &lt;/SPAN&gt;1%] Built target posix_predefined.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[&lt;SPAN class="Apple-converted-space"&gt;&amp;nbsp; &lt;/SPAN&gt;1%] Built target generate_posix_parameters&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 10%] Built target m4_type_includes&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 36%] Built target gftl-shared-v2&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 62%] Built target gftl-shared-v2-as-default&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 65%] Built target fargparse&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 72%] Built target funit-core&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 77%] Built target fhamcrest&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 86%] Built target asserts&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 86%] Built target funit-main&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 86%] Built target funit&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 90%] Built target fhamcrest_tests.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 90%] Built target robust&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 90%] Built target remote.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 91%] Built target robust_tests.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 91%] Built target other_shared&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 93%] Built target funit_tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[ 93%] Built target funit_tests.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Built target new_tests.x&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;[100%] Built target build-tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Constructing a list of tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Done constructing a list of tests&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Updating test list for fixtures&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Added 0 tests to meet fixture requirements&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Checking test dependency graph...&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;Checking test dependency graph end&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="p1"&gt;&lt;SPAN class="s1"&gt;No tests were found!!!&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Mon, 21 Oct 2024 18:07:57 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638467#M173995</guid>
      <dc:creator>Ron_Green</dc:creator>
      <dc:date>2024-10-21T18:07:57Z</dc:date>
    </item>
    <item>
      <title>Re: ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638607#M174000</link>
      <description>&lt;P&gt;Apparently, pFUnit uses internally an environment variable `PFUNIT_DIR`, which should be unset before starting the CMake configuration. I'm also getting `No tests were found!!!`, when&amp;nbsp;`PFUNIT_DIR` is defined.&lt;/P&gt;&lt;P&gt;If it's not the issue, then maybe CMake or build software are causing this problem. I'm using CMake&amp;nbsp;3.20.2 and&amp;nbsp;GNU Make 4.2.1 (on&amp;nbsp;Rocky Linux 8.7).&lt;/P&gt;</description>
      <pubDate>Tue, 22 Oct 2024 08:59:40 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638607#M174000</guid>
      <dc:creator>V-T</dc:creator>
      <dc:date>2024-10-22T08:59:40Z</dc:date>
    </item>
    <item>
      <title>Re: ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638706#M174004</link>
      <description>&lt;P&gt;I'm not exactly sure of the steps.&amp;nbsp; If you can get it to build and run the tests, can you update the step by step build and run instructions?&amp;nbsp; I had a look at the README.md and nothing struck me as an obvious fix we need.&lt;/P&gt;</description>
      <pubDate>Tue, 22 Oct 2024 23:28:23 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638706#M174004</guid>
      <dc:creator>Ron_Green</dc:creator>
      <dc:date>2024-10-22T23:28:23Z</dc:date>
    </item>
    <item>
      <title>Re: ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638800#M174008</link>
      <description>&lt;P&gt;I'm able to build and run tests&amp;nbsp;on my system with the commands in my first post. Alternatively, I can also use the following commands:&lt;/P&gt;&lt;LI-CODE lang="bash"&gt;mkdir build
cd build
unset PFUNIT_DIR
FC=ifx cmake ..
make
ctest -L test -V&lt;/LI-CODE&gt;&lt;P&gt;If you are unable to reproduce this behavior (and no tests are generated), then either there are some other environment variables that interfere with the process or the CMake configuration in pFUnit is system-dependent, which complicates the situation.&lt;/P&gt;&lt;P&gt;I should then look into the pFUnit code and reduce the example further to a single source code file. But I would need a lot more time for that...&lt;/P&gt;</description>
      <pubDate>Wed, 23 Oct 2024 08:13:07 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1638800#M174008</guid>
      <dc:creator>V-T</dc:creator>
      <dc:date>2024-10-23T08:13:07Z</dc:date>
    </item>
    <item>
      <title>Re: ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1639041#M174023</link>
      <description>&lt;P&gt;I reduced the example further. After you built all pFUnit files (with `&lt;SPAN&gt;cmake --build &lt;/SPAN&gt;&lt;SPAN class=""&gt;.&lt;/SPAN&gt;`), you can create&amp;nbsp;in the build directory a `a.f90` file with contents:&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;LI-CODE lang="fortran"&gt;module m
   type t
      integer,allocatable::a(:),b(:)  ! works without `b`
   contains
      procedure::f
   end type
contains
   subroutine f(o)
      class(t)o
      if(allocated(o%a))deallocate(o%a)
      allocate(o%a(1))  ! `ifx` produces code that considers this array as allocated here
   end 
   subroutine s()
      class(t),allocatable::o  ! works with `type` instead of `class`
      o=t()
      call o%f()  ! works with `f(o)` instead of `o%f()`
   end 
end    
       
program main
   use FUnit,only:stub,TestSuite,TestMethod
   use m,only:s
   procedure(),pointer::i=&amp;gt;stub,f=&amp;gt;stub
   call funit_main(r,i,f)
contains
   function r()
      type(TestSuite)::r
      r=TestSuite('r')
      call r%addTest(TestMethod('s',s))                                                                                                                                                                     
   end 
end&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;It is a simplified version of a code that pFUnit generates using Python. It can then be compiled and executed (also in the build directory) with the following command (`/opt/intel/oneapi/compiler/2024.2/lib/libiomp5.so` should be replaced with the right path to this library):&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;LI-CODE lang="bash"&gt;ifx -I./pFUnit/pfunit-build/src/funit/mod -fiopenmp -c a.f90 -o a.o &amp;amp;&amp;amp; ifx a.o pFUnit/pfunit-build/src/funit/libfunit.a /share/apps/intel/oneapi/compiler/2024.2/lib/libiomp5.so pFUnit/pfunit-build/extern/fArgParse/src/libfargparse.a pFUnit/pfunit-build/extern/fArgParse/extern/gFTL-shared/src/v2/libgftl-shared-v2.a &amp;amp;&amp;amp; ./a.out&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;It's also a simplified version of a building command from CMake.&lt;/P&gt;&lt;P&gt;This executable (`a.out`) produces the error that I described in the first post.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 24 Oct 2024 14:25:21 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1639041#M174023</guid>
      <dc:creator>V-T</dc:creator>
      <dc:date>2024-10-24T14:25:21Z</dc:date>
    </item>
    <item>
      <title>Re: ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1639072#M174025</link>
      <description>&lt;P&gt;I believe, I reduced the code to the absolute minimum (now without pFUnit). `ifx` produces an executable with the same error as in the first post, when the following code is compiled with the `-fiopenmp` option:&lt;/P&gt;&lt;LI-CODE lang="fortran"&gt;module m
   type t
      integer,allocatable::a(:)
   end type
contains
   subroutine f(o)
      class(t)o
      if(allocated(o%a))deallocate(o%a)
      allocate(o%a(1))  ! `ifx` produces code that considers this array as allocated here
   end
   subroutine s()
      type(t),allocatable::o
      o=t()
      call f(o)
   end
end

module n
   type u
   end type
   type v
      class(u),allocatable::e
      procedure(i),nopass,pointer::m
   end type
   interface
      subroutine i()
      end
   end interface
contains
   function c(m)result(o)
      procedure(i)m
      type(v)o
      o%m=&amp;gt;m
   end
end

program p
   use m
   use n
   type(v)o
   o=c(s)
   call o%m()
end&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Thu, 24 Oct 2024 17:14:22 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1639072#M174025</guid>
      <dc:creator>V-T</dc:creator>
      <dc:date>2024-10-24T17:14:22Z</dc:date>
    </item>
    <item>
      <title>Re: ifx 2024.2.1 produces wrong code when using pFUnit</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1643487#M174250</link>
      <description>&lt;P&gt;This issue seems to have be fixed&amp;nbsp;only partially in the newest version of ifx (2025.0.0): my last minimal example doesn't cause this error, but the version with pFUnit still doesn't work.&lt;/P&gt;</description>
      <pubDate>Fri, 15 Nov 2024 14:06:17 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/ifx-2024-2-1-produces-wrong-code-when-using-pFUnit/m-p/1643487#M174250</guid>
      <dc:creator>V-T</dc:creator>
      <dc:date>2024-11-15T14:06:17Z</dc:date>
    </item>
  </channel>
</rss>

