Re: changing argument types for four functions
- From: "James Van Buskirk" <not_valid@xxxxxxxxxxx>
- Date: Fri, 4 Nov 2005 11:57:56 -0700
"Lynn McGuire" <nospam@xxxxxxxxxx> wrote in message
news:11mn1e5ko7an22d@xxxxxxxxxxxxxxxxxxxxx
> > and build *only* that file (if the compiler can swallow what is
> > in effect one source file of 550,000 lines). That will likely
> > catch ALL mismatches. If you put a MODULE x/END MODULE x around
> > it AND all routines are properly terminated with END SUBROUTINE,
> > you will get even more problem diagnostics.
> I like this approach most of all ! I can just cat all the subroutines
into
> one monster f file. No telling how much stuff will blow up. Except I
> have to put END SUBROUTINE on 3500 subroutines.
If you wanna blow up the compiler with monster code, try using
a code generator such as
http://home.comcast.net/~kmbtib/scaled2i.f90
which seems to wipe out compilers at a relatively small code size.
I guess complexity of code is more important than size.
--
write(*,*) transfer((/17.392111325966148d0,6.5794487871554595D-85, &
6.0134700243160014d-154/),(/'x'/)); end
.
- References:
- changing argument types for four functions
- From: Lynn McGuire
- Re: changing argument types for four functions
- From: Jugoslav Dujic
- Re: changing argument types for four functions
- From: Lynn McGuire
- changing argument types for four functions
- Prev by Date: Re: FortranCompiler and IMSL
- Next by Date: Re: A StepWise Regression Subroutine
- Previous by thread: Re: changing argument types for four functions
- Next by thread: Decoding real variables expressed as binary
- Index(es):
Relevant Pages
|
