Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-12 Thread Neil Storer
Toon Knapen wrote: Fortran LOGICAL could you elaborate? Toon, There is no universally-agreed Fortran convention for how .TRUE. and .FALSE. boolean values are represented and how a value is checked for .TRUE. and .FALSE. Some Fortran implementations use 0 for .FALSE. and 1 for .TRUE.

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-12 Thread Greg Lindahl
> The government is one of the few forces that could mandate a proper MPI > ABI at this point in time; They certainly aren't the only ones -- vendors of proprietary applications that use MPI plus vendors of interconnect hardware get significant benefits from an ABI. Anyone who wants to distribute

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-11 Thread Toon Knapen
William Gropp wrote: > > Fortran name mangling here means how are Fortran routine names in the > source code mapped to names in the object library. For example, is > > MPI_Init > > in the Fortran source mapped to > > MPI_INIT > mpi_init > mpi_init_ > mpi_init__ > MPI_Init_ > > Each of these h

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-11 Thread Toon Knapen
Greg Lindahl wrote: >>ignoring the politics for a moment, what are the technical sticking points? > > Fortran name-mangling Up to the MPI 1.2, f77 is used so only functions that are new in MPI2 will be available in a mangled version only. So all others can be linked with using C linkage conventio

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-10 Thread Patrick Geoffray
Hi Mark, Mark Hahn wrote: There are only 2 ways to achieve ABI compatibility: 1) you impose/agree on a single one. ignoring the politics for a moment, what are the technical sticking points? for instance, I have the impression that the linux x86_64 ABI is reasonably well-defined, and it seem

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-10 Thread Greg Lindahl
> ignoring the politics for a moment, what are the technical sticking points? MPI types values of constants Fortran name-mangling Fortran LOGICAL program startup (optional) > for instance, I have the impression that the linux x86_64 ABI is reasonably > well-defined, It mostly works. We have ru

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-10 Thread Toon Knapen
Patrick Geoffray wrote: > The Fortran interface is actually worse than the C interface. Instead of > using pointers to opaque structures, the Fortran interface may use > integers as indexes into array of structures, into array of pointers, as > pointers casted to integers, etc. I can imagine that

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-10 Thread Patrick Geoffray
Hi Toon, Toon Knapen wrote: If only the 'named constants' are a problem: It's very easy to create a few functions (like 'int mpi_comm_world()') that will return the correct value for a given implementation. This is not the only problem, but this one can be fixed more or less easily by having

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-10 Thread Toon Knapen
Joachim Worringen wrote: > Wrong - i.e., the value of MPI_COMM_WORLD is not defined in Fortran, > either. This won't work if one MPI implementation sets > MPI_COMM_WORLD to 35 and another expects 626. > > Of course, you are right for opaque datatypes like MPI_Group, but this > is not sufficient fo

Re: [O-MPI users] [Beowulf] MPI ABI

2005-10-10 Thread Joachim Worringen
Toon Knapen wrote: Thus AFAICT if my app uses the fortran interface of MPI I can link my app with a dynamic library (say libmpi.so) which I can make point to either e.g. an MPICH implementation or POE. right? Wrong - i.e., the value of MPI_COMM_WORLD is not defined in Fortran, either. This won