This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: [Patch, Fortran] PR 41872 - Fixes for allocatable scalars (part 1)
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Tobias Burnus <burnus at net-b dot de>
- Cc: Daniel Kraft <d at domob dot eu>, gcc patches <gcc-patches at gcc dot gnu dot org>, gfortran <fortran at gcc dot gnu dot org>
- Date: Mon, 2 Nov 2009 07:51:24 -0800
- Subject: Re: [Patch, Fortran] PR 41872 - Fixes for allocatable scalars (part 1)
- References: <4AEDB241.7030401@net-b.de> <4AEDB6E3.2010105@domob.eu> <4AEDCAA3.5090906@net-b.de>
On Sun, Nov 1, 2009 at 9:51 AM, Tobias Burnus <burnus@net-b.de> wrote:
> Daniel Kraft wrote:
>> Tobias Burnus wrote:
>>> allocatable scalars have several issues. This patch fixes the some of
>>> them. [...]
>> I'm just wondering if that "large" new code block for deallocating
>> allocatable scalars passed to INTENT(OUT) can't be somehow be
>> simplified -- actually, there should already be code to do this for
>> non-scalar allocatables; is it possible to merge this somehow? ?But I
>> guess it's not, because stuff is probably too different there.
>
> Committed as Rev. 153795. Thanks for the review. I think it is not easy
> to simplify the code: Large parts are the same both there are enough
> differences that one only trades one kind of ifs with another one.
>
> I think the proper way would be to rewrite some of the allocatable
> handling stuff. One might consider this for 4.6 if GCC gets a new array
> descriptor. Actually, I do not like that the caller and not the callee
> deallocates intent(OUT) variables. I wonder how this should work with
> BIND(C) and the new TR 29113.
>
This may have caused:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41907
--
H.J.