This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: A patch to fix the most obvious failures with -fwhole-file
- From: dominiq at lps dot ens dot fr (Dominique Dhumieres)
- To: fortran at gcc dot gnu dot org
- Cc: sgk at troutmask dot apl dot washington dot edu
- Date: Mon, 14 Dec 2009 18:04:55 +0100
- Subject: Re: A patch to fix the most obvious failures with -fwhole-file
> This is a clear example of why changing the testsuite when
> someone uses nonstandard testsuite options is questionable.
Let me first restate that I have posted this patch because Richard Guenther
wanted to reduce the clutter he saw in regtesting gfortran with -flto.
Second, it is my understanding that -fwhole-file is intended to be the
default someday. Third, I don't know if the option is nonstandard, but
it just detected codes that are nonstandard conforming. For most of the
tests I looked at, I doubt it was intended (but for the last one) and
I think they should made standard conforming (hence they will pass
the tests with -fwhole-file), unless one wants to check how many
invalid codes go through gfortran (in this case it should be documented).
Dominique