[Bug c++/15264] New: [3.3 regression] ICE in maybe_push_cleanup_level

["reichelt at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>]

Hi Gaby,

your patch

http://gcc.gnu.org/ml/gcc-cvs/2004-05/msg00020.html

causes ICEs on invalid code like the following (see PR14810):

===========================================
template <class T> void make_pair(T x);

void foo(){
  struct fps_chan_ID fps; // { dg-error "incomplete" "" }
  make_pair(fps); // { dg-error "no matching function" "" }
}
===========================================

But fear not, there's hope, i.e. Mark's patch, which should be trivial
enough for the 3.3 branch:

http://gcc.gnu.org/ml/gcc-cvs/2004-04/msg00122.html

I'll a attach a backported version shortly.
(It does cure the ICE, but I didn't do a full bootstrap or run the testsuite).

-- 
           Summary: [3.3 regression] ICE in maybe_push_cleanup_level
           Product: gcc
           Version: 3.3.4
            Status: UNCONFIRMED
          Keywords: ice-on-invalid-code, error-recovery, patch
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: reichelt at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org,gdr at integrable-
                    solutions dot net


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15264