[Bug c++/55226] New: [C++11] ICE regression in regard to anonymous unions and constexpr

[Bug c++/55226] New: [C++11] ICE regression in regard to anonymous unions and constexpr

[daniel.kruegler at googlemail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55226

             Bug #: 55226
           Summary: [C++11] ICE regression in regard to anonymous unions
                    and constexpr
    Classification: Unclassified
           Product: gcc
           Version: 4.8.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: daniel.kruegler@googlemail.com


The following code gives me now a regressive ICE when switching from my
previous gcc 4.8 (last successfully tested: 4.8.0 20121014 (experimental)) to
gcc 4.8.0 20121104 (experimental) using the compiler flags

-Wall -pedantic -std=c++11

//----------------------
template<class T>
struct optional_data
{
  union {
    unsigned char for_value_init_;
    T value_;
  };

  constexpr optional_data() : for_value_init_() {}
};

struct ncnm
{
  ncnm(ncnm&&) = delete;
  ncnm(const ncnm&) = delete;
};

int main()
{
  optional_data<ncnm> o; // #20
}
//----------------------

"In function 'int main()':|
20|internal compiler error: in gimplify_init_ctor_eval, at gimplify.c:3787|
"

A tentative guess is that this could have been a side-effect of the fix applied
to bug 54922.

[Bug c++/55226] [4.8 Regression] ICE regression in regard to anonymous unions and constexpr

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55226

Paolo Carlini <paolo.carlini at oracle dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2012-11-07
             Blocks|                            |54922
            Summary|[C++11] ICE regression in   |[4.8 Regression] ICE
                   |regard to anonymous unions  |regression in regard to
                   |and constexpr               |anonymous unions and
                   |                            |constexpr
     Ever Confirmed|0                           |1

--- Comment #1 from Paolo Carlini <paolo.carlini at oracle dot com> 2012-11-07 10:52:16 UTC ---
I'm going to revert for now the fix for 54922. This one is essentially just a
duplicate of that one, simply trying to use the user defined type:

class nullable_int
{
  bool init_;
  union {
    unsigned char for_value_init;
    int value_;
  };
public:
  constexpr nullable_int() : init_(false), for_value_init() {}
};

int main()
{
  nullable_int n;
}

[Bug c++/55226] [4.8 Regression] ICE regression in regard to anonymous unions and constexpr

[paolo at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55226

--- Comment #2 from paolo at gcc dot gnu.org <paolo at gcc dot gnu.org> 2012-11-07 11:15:48 UTC ---
Author: paolo
Date: Wed Nov  7 11:15:40 2012
New Revision: 193292

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=193292
Log:
/cp
2012-11-07  Paolo Carlini  <paolo.carlini@oracle.com>

    PR c++/55226
    Revert:
    2012-10-23  Paolo Carlini  <paolo.carlini@oracle.com>

    PR c++/54922
    * semantics.c (cx_check_missing_mem_inits): Handle anonymous union
    members.

/testsuite
2012-11-07  Paolo Carlini  <paolo.carlini@oracle.com>

    PR c++/55226
    Revert:
    2012-10-23  Paolo Carlini  <paolo.carlini@oracle.com>

    PR c++/54922
    * g++.dg/cpp0x/constexpr-union4.C: New.

Removed:
    trunk/gcc/testsuite/g++.dg/cpp0x/constexpr-union4.C
Modified:
    trunk/gcc/cp/ChangeLog
    trunk/gcc/cp/semantics.c
    trunk/gcc/testsuite/ChangeLog

[Bug c++/55226] [4.8 Regression] ICE regression in regard to anonymous unions and constexpr

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55226

Paolo Carlini <paolo.carlini at oracle dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |paolo.carlini at oracle dot
                   |                            |com
         Resolution|                            |DUPLICATE

--- Comment #3 from Paolo Carlini <paolo.carlini at oracle dot com> 2012-11-07 11:16:37 UTC ---
See PR54922.

*** This bug has been marked as a duplicate of bug 54922 ***