[O-MPI users] [Beowulf] MPI ABI
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Greg Lindahl lindahl at pathscale.comTue Oct 11 22:50:42 PDT 2005
- Previous message: [Beowulf] MorphMPI based on fortran itf (was: MPI ABI)
- Next message: [O-MPI users] [Beowulf] MPI ABI
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Tue, Oct 11, 2005 at 03:44:07PM +0200, Toon Knapen wrote: > The one thing about linking f77 and C that I do not know how to solve > yet is e.g. fortran functions that return a complex. On x86 and AMD64, this is the f2c-abi issue, and is not present in the MPI API. If you experiment on AMD64 systems, you'll find that some compilers do it the way documented in the g77 info files, some follow the platform ABI (PathScale), and some do half and half (probably an accident.) So it's really handy that MPI doesn't have any such calls. Functions returning REAL*4 are the other piece of the f2c-abi issue. -- greg
- Previous message: [Beowulf] MorphMPI based on fortran itf (was: MPI ABI)
- Next message: [O-MPI users] [Beowulf] MPI ABI
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list