[mpich-discuss] MPI_Allreduce fail. Please help. (UPDATE) (WINDOWS XP 32bit+gfortran)

[Anthony Chan]

----- Original Message -----

MPI_IN_PLACE in C is defined as a pointer.
MPI_IN_PLACE in Fortran is defined as a Fortran integer
in a common block.  What is important for Fortran
MPI program is actually the MPI_IN_PLACE's memory address.
So the Fortran integer value of MPI_IN_PLACE is irrelvant
in locating the bug.

A.Chan

> 3. Works if not using MPI_IN_PLACE.
> 
> So I guess it could be a MPI_IN_PLACE problem happened in the windows
> version fortran functions.
> 
> Or it could be the problem of MinGW?
> 
> 
> 
> The c code works fine. And in c version MPI_IN_PLACE = 0xffffffff but
> fortran version MPI_IN_PLACE is 0. I can't tell whether they should be
> different since I don't know how does it implemented.
> 
> 
> 
> Please help me out! Thanks.
> 
> Y L
> 
> 
> 
> 
> _______________________________________________
> mpich-discuss mailing list mpich-discuss at mcs.anl.gov
> To manage subscription options or unsubscribe:
> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss