[Mpi-forum] about MPI_Scan

[Jeff Hammond]

This is not the appropriate list for this question.  Please direct
your question to the PETSc list since it is impossible to understand
the context of your usage of MPI_Scan without knowledge of the PETSc
API.

Jeff

On Fri, Sep 2, 2011 at 2:30 PM, Likun Tan <likunt at andrew.cmu.edu> wrote:
> Hello,
>
> I have a rectangular matrix B with size M*N (where N>>M), i use the
> following command to create and compute the matrix by column:
>
> PetscSplitOwnership(PETSC_COMM_WORLD, &n, &N);
> MPI_Scan(&n, &cstart, 1, MPI_INT, MPI_SUM, PETSC_COMM_WORLD);
> cstart -= n;
> MatCreateSeqDense(PETSC_COMM_SELF, M, n, PETSC_NULL, &B);
>
> for(j=0; j<n; j++)
> {
>   for(i=0; i<M; i++)
>   {
>      MatSetValues(B, 1, &i, 1, &j, &value, INSERT_VALUES);
>   }
> }
>
> MatAssemblyBegin(B, MAT_FINAL_ASSEMBLY);
> MatAssemblyEnd(B, MAT_FINAL_ASSEMBLY);
>
> where n is the number of columns in each processor.
> Any problem with this?
>
> Thanks,
> Likun
>
>
>
>
>
> _______________________________________________
> mpi-forum mailing list
> mpi-forum at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
>



-- 
Jeff Hammond
jeff.science at gmail.com