[Bug c++/12316] New: [3.4 regression] trouble with garbage collection

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

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12316

           Summary: [3.4 regression] trouble with garbage collection
           Product: gcc
           Version: 3.4
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: critical
          Priority: P1
         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

Consider the following preprocessed code snippet:

---------------------------------------------------------
inline void FOO() {}

template<typename> struct A
{
    A() {}
    ~A() throw() {}
};

template<typename> struct B
{
    static void foo();
    static void bar() { foo(); }
};

struct C {};

template<typename> struct D : C
{
    D() {}
    ~D() { B<void>::bar(); }
};

template<typename> struct E : D<void>
{
    static void baz() {}
    E(A<void>) { baz(); }
};

void BAR()
{
    new E<void>(A<void>());
}
---------------------------------------------------------

Compiling it on i686-pc-linux-gnu with mainline using the command

  g++ --param ggc-min-expand=0 --param ggc-min-heapsize=0 -c gc.ii

I get the message

  virtual memory exhausted: Invalid argument

which looks like a garbage collection problem.
(Although we don't get an "Internal comiler error" message, I'd
call that an internal error, hence the "ice-on-valid-code" keyword.)


It's a regression that appeared on 20030910 with Richard's patch

  http://gcc.gnu.org/ml/gcc-cvs/2003-09/msg00275.html

I don't know whether it's the cause for the problems or just
triggers the bug.