This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug fortran/45777] Alias analysis broken for arrays where LHS or RHS is a component ref
- From: "tkoenig at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Sun, 16 Jan 2011 11:47:12 +0000
- Subject: [Bug fortran/45777] Alias analysis broken for arrays where LHS or RHS is a component ref
- Auto-submitted: auto-generated
- References: <bug-45777-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45777
--- Comment #15 from Thomas Koenig <tkoenig at gcc dot gnu.org> 2011-01-16 11:47:00 UTC ---
Author: tkoenig
Date: Sun Jan 16 11:46:55 2011
New Revision: 168851
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=168851
Log:
2011-01-16 Thomas Koenig <tkoenig@gcc.gnu.org>
Backport from trunk
PR fortran/45777
* symbol.c (gfc_symbols_could_alias): Strip gfc_ prefix,
make static and move in front of its only caller, to ...
* trans-array.c (symbols_could_alias): ... here.
Pass information about pointer and target status as
arguments. Allocatable arrays don't alias anything
unless they have the POINTER attribute.
(gfc_could_be_alias): Keep track of pointer and target
status when following references. Also check if typespecs
of components match those of other components or symbols.
* gfortran.h: Remove prototype for gfc_symbols_could_alias.
2011-01-16 Thomas Koenig <tkoenig@gcc.gnu.org>
Backport from trunk
PR fortran/45777
* gfortran.dg/dependency_39.f90: New test.
Added:
branches/gcc-4_5-branch/gcc/testsuite/gfortran.dg/dependency_39.f90
Modified:
branches/gcc-4_5-branch/gcc/fortran/ChangeLog
branches/gcc-4_5-branch/gcc/fortran/gfortran.h
branches/gcc-4_5-branch/gcc/fortran/symbol.c
branches/gcc-4_5-branch/gcc/fortran/trans-array.c