[Mpi-comments] typo in MPI_*_SET_ATTR parameter

[William Gropp]

The definitions of IN and INOUT are given in section 2.3, page 10, and are not identical to the Fortran terms.  INOUT means that the call may update the arguments - see lines 23-27 on page 10.  The MPI standard is correct as is.

Bill

William Gropp
Director, Parallel Computing Institute
Thomas M. Siebel Chair in Computer Science
Chief Scientist, NCSA
University of Illinois Urbana-Champaign





On Apr 27, 2016, at 6:37 PM, Gilles Gouaillardet <gilles at rist.or.jp> wrote:

> Folks,
> 
> i noticed 3 similar and minor typos in the MPI 3.1 standard
> 
> 
> 1) MPI_COMM_SET_ATTR page 270 line 43 reads
> 
> "INOUT comm ..."
> 
> but should be
> 
> "IN comm ..."
> 
> 
> 2) MPI_WIN_SET_ATTR page 274 line 33 reads
> 
> "INOUT win ..."
> 
> but should be
> 
> "IN win ..."
> 
> 
> 3) MPI_TYPE_SET_ATTR page 278 line 3 reads
> 
> "INOUT datatype ..."
> 
> but should be
> 
> "IN datatype ..."
> 
> 
> note Fortran 2008 correctly reads "INTENT(IN)"
> 
> 
> Cheers,
> 
> 
> Gilles
> _______________________________________________
> mpi-comments mailing list
> mpi-comments at mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-comments