Re: Fortran and subdir-objects

[Ralf Wildenhues]

Hi Stepan,

* Stepan Kasal wrote on Mon, Jan 16, 2006 at 04:33:48PM CET:
> On Sat, Jan 14, 2006 at 06:45:51PM +0800, Davyd Madeley wrote:
> > [...] the macros to test for -c/-o compatibility in automake. [...]
> 
> this is an interesting question: is the test for "-c -o" really needed
> for a Fortran or F77 compiler?

The old UNICOS 9 Fortran compiler does not grok `-c -o'.  The manuals for
the recent ones, however, suggest that they still don't implement it.

Likely, non-free Fortran compilers on w32 don't support it (e.g.,
because they spell `-o' differently, as does MSVC for C/C++), but that
is just a guess.

> Currently, _Autoconf_ contains macros to test "-c -o" for C, Fortran,
> and Fortran77 languages.  _Automake_ contains a wrapper for the C macro,
> but not for the latter two ones.
> But the latter two macros have the following comment attached:
> # The usefulness of this macro is questionable, as I can't really see
> # why anyone would use it.  The only reason I include it is for
> # completeness, since a similar test exists for the C compiler.

I would hereby question the usefulness of above comment: I don't see any
reference or even claim to the contrary (that all compilers grok this).
Surely, *without Automake support* the usefulness is limited.

> My impression is that we should remove the artificial macros and write
> new ones if/when there is a real need.

I think there is a need.

> So the questions are:
> Are there any Fortran* compilers which don't understand "-c -o"? 

Yes.

> Are there ones which do understand "-c" yet don't understand "-c -o"?

I have yet to come across one that does not understand `-c'.

> Would it be bad if projects using Automake required GNU Compiler for
> Fortran?

Yes, definitely, as long as you still have to hope that CVS GCC is good
enough to compile some piece of third-party code you happen to need.

Cheers,
Ralf