This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c++/19349] [4.0 Regression] Valgrind errors compiling libjava c++ sources
- From: "pinskia at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 18 Jan 2005 14:33:51 -0000
- Subject: [Bug c++/19349] [4.0 Regression] Valgrind errors compiling libjava c++ sources
- References: <20050109203748.19349.mark@gcc.gnu.org>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Additional Comments From pinskia at gcc dot gnu dot org 2005-01-18 14:33 -------
(In reply to comment #4)
> That's strange. I thought ggc_free should poison the memory that is
> freed if the compiler is built with checking enabled. So anything
> that uses it after the ggc_free should segfault or die in a similar
> horrible way.
>
> Did someone remove ENABLE_GC_CHECKING from --enable-checking??
No, we do poison it but since this is only a read we don't get any diagnostic unless you use
valgrind because we poison it even further with valgrind and say the memory is useless.
Mark thought his patch was safe and he even tried to explain why he thought it was safe but this is the
second bug which was effected by it.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19349