[Bug fortran/32980] Vendor extension: Intrinsic functions (D)GAMMA, LGAMMA (ALGAMA/DLGAMA)

["burnus at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>]

------- Comment #7 from burnus at gcc dot gnu dot org  2007-08-21 19:04 -------
> * Will we want a -std=f2008 separate from f2003, or do we consider f2003 an
> alias for f2008 (F2008 being, after all, a minor improvement to F2003)?

Who said that it will be a minor change compared to Fortran 2003 [except that
it was supposed to be one]? With Co-Arrays and submodules and a couple of other
changes I wouldn't claim so. I'm definitely in favour of having -std=f2008,
however, I'm not sure it is already time to introduce such features. (I need
the vendor intrinsic GAMMA which I want to be compatible to Fortran 2008;
currently I don't care of Fortran 2008 nor do I want to promote it, yet.)

The current time line of Fortran 2008 is:
2007-12 First working draft available
2008-02 WG5 review of working draft
2008-04 WG5 approval of draft CD [...]
2009-07 Final CD submitted for approval

I think we should wait at least for the first working draft until we start with
-std=f2008. Only shortly ago (2007-08-13), the BITS type and the macros were
(fortunately) dropped which shows that things are still in a flux -- not that I
expect LOG_GAMMA and GAMMA to change/disappear esp. as they are compatible with
C. The other mentioned intrinsic changes are also unlikely to change [better
ask Toon before adding].

We already have a F2008 feature: We allow "CONTAINS" followed by no contained
procedure. When we introduce STD_F2008, one should change the error message
there.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32980