Re: Questions about the gfortran test suite

[Tobias Schlüter <tobias dot schlueter at physik dot uni-muenchen dot de>]

Dominique Dhumieres wrote:
> Note that the Intel extension was OK for me if detected by the
> standard conformance flags.

Every extension must be tested, documented and -- in the future -- maintained.
 This extension has no benefits, except allowing for a sloppy coding standard,
for a function without arguments, though possible, is a somewhat
counterintuitive concept and therefore likely appears quite seldomly.  Unless
there are actually users of this extension, I don't think it's worth it.  And
even then I'd be tempted to urge them to fix their their sources, as the
required change is completely mechanical.

- Tobi