This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
RE: PR26038 and the pre-gomp patch
- From: "THOMAS Paul Richard 169137" <paul dot richard dot thomas at cea dot fr>
- To: "Jakub Jelinek" <jakub at redhat dot com>
- Cc: <fortran at gcc dot gnu dot org>, <paulthomas2 at wanadoo dot fr>
- Date: Tue, 31 Jan 2006 15:58:09 +0100
- Subject: RE: PR26038 and the pre-gomp patch
Jakub,
> it a few days ago I found it causes a regression in a recently added
> testcase (committed just a day before I wanted to commit my patch):
>
> FAIL: gfortran.dg/assumed_charlen_function_3.f90 -O (test
> for excess errors)
Oh, so it does....
>
> I haven't found time to debug that yet unfortunately, am now
> busy with the
> long double stuff that has higher priority. But I hope I'll
> have time for
> that at the end of the week or so.
>
If it helps, the problem lies with
function more_OK (ch, fcn) ! { dg-warning "is obsolescent in fortran 95" }
character(*) more_OK, ch
character (*), external :: fcn ! Fixed charlen => no regression.
more_OK = fcn (ch)
end function more_OK
I just rechecked the standard, DF6.0 and Lahey, just to make sure that this is OK. I am convinced that the fortran is fine.
Paul