[Patch, Fortran, OOP] PR 50625: [4.6/4.7 Regression] ALLOCATABLE attribute lost for module CLASS variables

[Janus Weil]

>> Regtested successfully on x86_64-unknown-linux-gnu (except for failure
>> on entry_4 and select_type_12, which are known middle-end
>> regressions). Ok for trunk and 4.6?
>
> OK. Thanks for the patch

Thanks for the review. Committed to trunk as r179660.


> - and for telling me that select_type_12 is a
> middle end regression (PR 50640); I already wondered why my being worked-on
> patch showed all of a sudden that failure. (I had self-caused select type
> failures before; thus, I didn't think of a ME problem.)

Yeah, at first I also thought that my local modifications were to
blame. But then I noticed that it only fails with optimization (which
is usually a very reliable sign for a middle-end problem) and that
Dominique had already filed a PR.

Cheers,
Janus



>> 2011-10-07  Janus Weil<janus@gcc.gnu.org>
>>
>>        PR fortran/50625
>>        * class.c (gfc_build_class_symbol): Fix whitespace.
>>        * module.c (mio_symbol): Set 'class_ok' attribute.
>>        * trans-decl.c (gfc_get_symbol_decl): Make sure the backend_decl
>> has
>>        been built for class symbols.
>>
>>
>> 2011-10-07  Janus Weil<janus@gcc.gnu.org>
>>
>>        PR fortran/50625
>>        * gfortran.dg/class_46.f03: New.
>
>