Oh okay that explains the behaviour of MPI_SHORT, guess sizeof is going to give me the same value no matter what MPI_Datatype I use? Thanks for the quick response!
On Mon, Nov 23, 2009 at 2:33 PM, Edmund Sumbar <esum...@ualberta.ca> wrote: > On Mon, 23 Nov 2009, Natarajan CS wrote: > > [...] > > 1) When I use MPI_SHORT instead of short to count the number of data >> transferred I get an array that is half the size. (ie >> MPI_Isend(&data2[0][0][0], (int) (sizeof(data2)/sizeof(MPI_SHORT)), >> MPI_SHORT, rank, 123, >> cartcomm,&request[rank-1]); >> > > MPI_SHORT and the other data types are actually macros that resolve to > MPI_Datatype which is a pointer to a struct. > > [...] > > -- > Edmund Sumbar > AICT Research Support Group > esum...@ualberta.ca > 780.492.9360 > _______________________________________________ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users >
