This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug fortran/16511] Test 19990905-0.f fails with gfortran
- From: "tobi at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 14 Jul 2004 12:03:02 -0000
- Subject: [Bug fortran/16511] Test 19990905-0.f fails with gfortran
- References: <20040713055437.16511.billingd@gcc.gnu.org>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Additional Comments From tobi at gcc dot gnu dot org 2004-07-14 12:02 -------
I disagree, I think this is valid code. We reject it, because we don't resolve
equivalences in any meaningful way.
The expression in the upper or lower bound of an array specification must be a
specification expression. Informally, a specification expression is an
expression that can be evaluated at procedure entry, which is the case here.
In legalese terms:
By R734 a specification expression is a scalar int expression, which is a
restricted expression. A variable is a restricted expression if it is a
"variable that is in a common block or a variable that is a subobject of such a
variable". This is the case here. I couldn't find a quote that this property is
inherited via equivalences, but I would assume that this is the intended meaning.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16511