Subject: Are we ready to vote?
To: mpi-21@XXXXXXXXXXXXX
X-Mailer: Lotus Notes Release 5.0.5  September 22, 2000
From: "Richard Treumann" <treumann@XXXXXXXXXX>
Date: Thu, 31 May 2001 13:25:06 -0400
X-MIMETrack: Serialize by Router on D01ML077/01/M/IBM(Release 5.0.7  SPR #MIAS4UTJ8H &S/390
 SPR #JHEG4V8UT5|April 5, 2001) at 05/31/2001 01:26:06 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: owner-mpi-21@XXXXXXXXXXXXX
Precedence: bulk
Reply-To: mpi-21@XXXXXXXXXXXXX

This email forum has been pretty quiet for 2 weeks now.  Is there any
reason NOT to move toward a first vote on set one of changes?

I would like to get these noncontroversial issues behind us and move on.

In particular, I would like to start the ball rolling on the MPI_ALLTOALLW
issue:

The MPI-2 document shows the two array_of_displacement parameters as
default integers, not as MPI_Aint.  Since these are byte displacements they
fall into the same logical class as the displacements in MPI_Type_struct,
MPI_Type_hindexed and the stride in MPI_Type_hvector.   All of these use
MPI_Aint.  Expressing a byte displacement in a 64 bit MPI and being limited
to using an integer is clearly undesirable.   I do not believe it was a
conscious intent of the Forum to use default integer.  My bet is the text
was taken from MPI_ALLTOALLV and proofreading failed to catch the need for
this little edit.

I urge the Forum to make this correction though I recognize it does change
the documented binding.  That means we must decide between the clearly
better choice for the future of MPI and the "We cannot break a single line
of existing code."  argument.

                Dick

Dick Treumann                RS/6000 SP Development
IBM Poughkeepsie Unix Development Lab
Dept 0lva / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601
Tele (845) 433-7846         Fax (845) 433-8363