[Developers] Re: [CactusMaint] fortran PROVIDES FUNCTION example?

Mark Miller mamiller at cyga.jpl.nasa.gov
Fri Jan 16 12:00:23 CST 2004


On Fri, 16 Jan 2004, Ian Hawke wrote:

> I do not know of a cactus-proper Fortran provided function, as nearly all
> Cactus* arrangements are in C. Those that exist, such as DriftCorrect and
> Dist are mostly restricted access.
> However, the syntax at the .ccl level is identical independent of the
> language. The only subtleties occur on the calling side where the method
> was changed recently by Erik (largely to ensure that the return type
> definition is automatically included somehow). Some of the recent changes
> haven't propagated up to the users guide yet, largely because there is
> still talk of introducing a "functions.ccl" when the syntax could change
> again. Don't know what the movement on this is.

Tom, is this change of Erik's likely to change in the next few weeks?
If not, I'll put in some elbow grease into getting my things working
with the development version.

> Cactus proper thorns that provide functions include CactusBase/Boundary
> and CactusPUGH/PUGHSlab. The AEIThorns/ADM_BSSN and BSSN_MoL thorns have
> aliased functions in. Some are used from Fortran. However, the provided
> function is C. AEIThorns/BSSN_MoL/Boundaries.F is probably the canonical
> example of using aliased functions from Fortran. Note that you need to
> #include "cctk_Functions.h", that the routine needs the
> DECLARE_CCTK_FUNCTIONS macro, and that from Fortran the checking function
> CCTK_IsFunctionAliased is unfortunately a subroutine.
> Hope this helps,

Yes, I now know where things sort of stand.  I'll try playing around with
providing fortran functions and getting my stuff working.  I'll yell if
I get in over my head.

Thanks for the info,


Mark Miller
Jet Propulsion Laboratory 238-332
4800 Oak Grove Drive
Pasadena, CA  91109
E-mail: Mark.A.Miller at jpl.nasa.gov
office phone: (818) 393-2657, fax: (818) 393-6890

More information about the Developers mailing list