This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: gfortran feature request: internal procedures as actual arguments


Paul Thomas wrote:

Bill,

I could be wrong, but this is prohibited by at least the Fortran
2003 standard.

Yup, I know that. It is (imho) the biggest blunder in the design. I can only imagine that it was proposed by the Intel folks (since they went ahead and implemented it anyway) and rejected by others who didn't immediately see how to make it work.

I have to be careful, because I'm working from memory here - I think the restriction was due to the fact that we had a hard time coming up with a complete proof that nothing the internal procedure could access would go out of scope.

I think that I can see pretty quickly how to make it work if I forget about optimization - once function inlining starts, I do not know how to keep the procedure itself in play. In spite of my opposition (or, perhaps, because of it!), I'll take a look.

I think that in the 2008 Standard, we've warmed our collective cold feet about the above mentioned problem. See for instance Note 12.20 in:
http://j3-fortran.org/ -> WG5 -> N1678.pdf


Cheers,

--
Toon Moene - e-mail: toon@moene.indiv.nluug.nl - phone: +31 346 214290
Saturnushof 14, 3738 XG Maartensdijk, The Netherlands
At home: http://moene.indiv.nluug.nl/~toon/
Who's working on GNU Fortran: http://gcc.gnu.org/ml/gcc/2007-01/msg00059.html



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]