[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.
Steffen Persvold sp at scali.comMon Oct 10 02:49:09 PDT 2005
- Previous message: [Beowulf] MPI ABI
- Next message: [O-MPI users] Fwd: Fwd: [Beowulf] MorphMPI based on fortran itf
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Toon,
You seem to be forgetting all the different constants even in the Fortran
interface:
For example :
! return codes
INTEGER MPI_SUCCESS,MPI_ERR_BUFFER,MPI_ERR_COUNT,MPI_ERR_TYPE
INTEGER MPI_ERR_TAG,MPI_ERR_COMM,MPI_ERR_RANK,MPI_ERR_ROOT
INTEGER MPI_ERR_GROUP
INTEGER MPI_ERR_OP,MPI_ERR_TOPOLOGY,MPI_ERR_DIMS,MPI_ERR_ARG
INTEGER MPI_ERR_UNKNOWN,MPI_ERR_TRUNCATE,MPI_ERR_OTHER
INTEGER MPI_ERR_INTERN,MPI_ERR_IN_STATUS,MPI_ERR_PENDING
INTEGER MPI_ERR_REQUEST, MPI_ERR_LASTCODE
PARAMETER (MPI_SUCCESS=0,MPI_ERR_BUFFER=1,MPI_ERR_COUNT=2)
PARAMETER (MPI_ERR_TYPE=3,MPI_ERR_TAG=4,MPI_ERR_COMM=5)
PARAMETER (MPI_ERR_RANK=6,MPI_ERR_ROOT=7,MPI_ERR_GROUP=8)
PARAMETER (MPI_ERR_OP=9,MPI_ERR_TOPOLOGY=10,MPI_ERR_DIMS=11)
PARAMETER (MPI_ERR_ARG=12,MPI_ERR_UNKNOWN=13)
PARAMETER (MPI_ERR_TRUNCATE=14,MPI_ERR_OTHER=15)
PARAMETER (MPI_ERR_INTERN=16,MPI_ERR_IN_STATUS=17)
PARAMETER (MPI_ERR_PENDING=18,MPI_ERR_REQUEST=19)
PARAMETER (MPI_ERR_LASTCODE=1073741823)
These are not the same on all MPI Implementations. I could also say
something on Fortran vs. C in general, but I won't ;)
Cheers,
Steffen Persvold
Technical Director Americas
tel. 508-281-7100 x401
fax. 508-281-7171
http://www.scali.com/
Scaling the Linux datacenter
> -----Original Message-----
> From: beowulf-bounces at beowulf.org [mailto:beowulf-bounces at beowulf.org] On
> Behalf Of Toon Knapen
> Sent: Monday, October 10, 2005 4:58 AM
> To: users at open-mpi.org; 'beowulf at beowulf.org'
> Subject: [Beowulf] MPI ABI
>
> Coming back to the MPI ABI discussion (which dates back from a long time
> ago though), just one additional question (to which MPI implementers
> certainly have an interesting opinion):
>
> Why don't we use the fortran interface instead of the C interface.
> Different C interfaces for MPI are likely incompatible because the type
> behind the MPI_Group etc. might all be different. Same goes for the
> named constants. In the fortran interface however the type's are always
> guaranteed to be int's.
>
> 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?
>
> toon
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit
> http://www.beowulf.org/mailman/listinfo/beowulf
- Previous message: [Beowulf] MPI ABI
- Next message: [O-MPI users] Fwd: Fwd: [Beowulf] MorphMPI based on fortran itf
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
