[Mpi-forum] MPI_Pcontrol const argument

[Jim Dinan]

I think my confusion is that the level argument is passed by value, so it
should make no difference to the application whether it's const or not.
The const qualifier seems to place a limitation on an implementation of
this routine, but I'm not sure why.

 ~Jim.

On Thu, Apr 6, 2017 at 10:48 AM, William Gropp <wgropp at illinois.edu> wrote:

> There is no rationale, other than that this is an IN argument and hence is
> unchanged.  Of course, declaring it as const achieves nothing new.  I think
> it was just a rote application of that rule, to a routine that is rarely
> used.
>
> Bill
>
> William Gropp
> Acting Director and Chief Scientist, NCSA
> Director, Parallel Computing Institute
> Thomas M. Siebel Chair in Computer Science
> University of Illinois Urbana-Champaign
>
>
>
> On Apr 6, 2017, at 9:42 AM, Jim Dinan <james.dinan at gmail.com> wrote:
>
> Hi All,
>
> I apologize if this is a dumb question -- what's the rationale for the
> const qualifier on the level argument to MPI_Pcontrol?
>
> int MPI_Pcontrol(const int level, ...)
>
> Cheers,
>  ~Jim.
> _______________________________________________
> mpi-forum mailing list
> mpi-forum at lists.mpi-forum.org
> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum
>
>
>
> _______________________________________________
> mpi-forum mailing list
> mpi-forum at lists.mpi-forum.org
> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpi-forum/attachments/20170407/817df8d3/attachment.html>