Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] making Fortran MPI_Status components public
From: N.M. Maclaren (nmm1_at_[hidden])
Date: 2012-09-27 14:31:37


On Sep 27 2012, Jeff Squyres (jsquyres) wrote:
>
> Fwiw, we have put in many hours of engineering to "that obscene hack"
> *because* compilers all have differing degrees of compatibility suck.
> It's going to be years before compilers fully support f08, for example,
> so we have no choice but to test for various compiler characteristics at
> configure time.

Yes and no, but that wasn't my point anyway. My remarks about the
obscene hack were to the configure mechanism more than its use - I
could describe why I said what I said, but it's not relevant.

> So I'm ok propagating "that obscene hack" when it protects users and
> prevents wasting time on mailing list questions.

That wasn't my point. It may well be the lesser of two evils, given
the world that we are in. It was that there really isn't worth having
a kludge around every little compiler lunacy. Realistically, in the
year 2012, if any Fortran compiler doesn't support Fortran 95 properly,
it should be categorised as defective/obsolescent and only the old
interfaces provided.

In this case, I doubt very much that Oracle's compiler doesn't support
PRIVATE, and I would definitely bet on the problem being caused by
some defect in the configure mechanism. If it isn't, it's almost
certainly some OTHER issue that shows up as a failure with PRIVATE.
Fixing it by putting yet another hack into configure is definitely
not the right solution.

Regards,
Nick Maclaren.