This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug fortran/34693] [4.3 Regression] gfortran.dg/common_6.f90 -O
- From: "burnus at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 6 Jan 2008 22:23:23 -0000
- Subject: [Bug fortran/34693] [4.3 Regression] gfortran.dg/common_6.f90 -O
- References: <bug-34693-682@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #4 from burnus at gcc dot gnu dot org 2008-01-06 22:23 -------
I don't know. When I investigated PR 33375, I found that that the symbol
existed but was uninitialized. I do not understand why it started to fail with
x86-64 but valgrind showed it used also before uninitialized memory.
I assume that now that one accesses the symbol much more often, the chance of
crashes increases. At least it now also crashes here for -m32. But as the
symptoms are the same (crash for invalid COMMON) and as valgrind shows the same
error (invalid read), I still believe this is a duplicate of PR 33375.
You can try valgrind with an older version.
*** This bug has been marked as a duplicate of 33375 ***
--
burnus at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution| |DUPLICATE
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34693