[Bug c++/49253] New: [4.7 Regression] Debug-mode forward_list broken

[Bug c++/49253] New: [4.7 Regression] Debug-mode forward_list broken

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

           Summary: [4.7 Regression] Debug-mode forward_list broken
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: paolo.carlini@oracle.com


Something changed either in the front-end or the middle end over the next
couple of days which broke the debug-mode forward_list, like this:

...debug/forward_list:650:25: error: request for member
‘__gnu_debug::_Safe_iterator_base::_M_prior’ in ‘__rhs_iterators->’, which is
of non-class type ‘__gnu_debug::_Safe_iterator_base*’

To reproduce just compile, with -std=c++0x -D_GLIBCXX_DEBUG:

////////////
#include <forward_list>
////////////

No recent changes to the library itself can possibly have caused this.

HJ, can you search the offending patch?

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hjl at gcc dot gnu.org

--- Comment #1 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-06-01 11:09:14 UTC ---
Asking HJ help.

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jason at gcc dot gnu.org

--- Comment #2 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-06-01 11:22:57 UTC ---
Tried Rev 174498 and was Ok. Could be 174499.

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

Richard Guenther <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.7.0

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011.06.01 13:46:19
                 CC|hjl at gcc dot gnu.org      |hjl.tools at gmail dot com
     Ever Confirmed|0                           |1

--- Comment #3 from H.J. Lu <hjl.tools at gmail dot com> 2011-06-01 13:46:19 UTC ---
It is caused by revision 174499:

http://gcc.gnu.org/ml/gcc-cvs/2011-05/msg01280.html

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

Jason Merrill <jason at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
         AssignedTo|unassigned at gcc dot       |jason at gcc dot gnu.org
                   |gnu.org                     |

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

--- Comment #4 from Jason Merrill <jason at gcc dot gnu.org> 2011-06-01 17:31:08 UTC ---
It seems unfortunate that make check-libstdc++-v3 doesn't catch obvious
breakage like this; at the least we could make sure that stdc++.h compiles in
debug/parallel modes?

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

--- Comment #5 from Jason Merrill <jason at gcc dot gnu.org> 2011-06-01 18:07:06 UTC ---
Author: jason
Date: Wed Jun  1 18:07:03 2011
New Revision: 174539

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=174539
Log:
    PR c++/49253
    * typeck2.c (build_x_arrow): Don't use build_min_nt.

Modified:
    trunk/gcc/cp/ChangeLog
    trunk/gcc/cp/typeck2.c

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

--- Comment #6 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-06-01 18:39:43 UTC ---
Yes. Running by default the entire testsuite in debug-mode too is not an
option, would take too much time. Honestly, however, in my experience, isn't
*that* common that a front-end issue shows up *only* in debug-mode and, anyway,
the library developers proper run check-debug too rather  often and are willing
to file PRs ;) Thanks for the quick fix!

[Bug c++/49253] [4.7 Regression] Debug-mode forward_list broken

[jason at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49253

Jason Merrill <jason at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

--- Comment #7 from Jason Merrill <jason at gcc dot gnu.org> 2011-06-05 20:22:18 UTC ---
Fixed.