Eric,

     I checked through all the uses and it appears that it is actually not a 
collective operation so I have changed the manual page in master to reflect 
this.

     Thanks for pointing out the inconsistency

    Barry

> On Jan 16, 2017, at 11:22 AM, Eric Chamberland 
> <[email protected]> wrote:
> 
> Hi,
> 
> shouldn't there be a MPI_Reduce in MatMissingDiagonal_MPIAIJ so it is really 
> collective as stated in documentation?
> 
> Thanks,
> 
> Eric

Reply via email to