[Mpi-forum] Handling of MPI_BOTTOM in Reductions

[Richard Treumann]

In my opinion, it is simply erroneous to use MPI_BOTTOM as both buffer 
addresses just as it is to use &array_1. 

If by "imply" you mean that the user should know enough to pass MPI_BOTTOM 
as the real buffer and MPI_IN_PLACE as the other the I would say "yes". If 
you are hoping the MPI implementation will recognize MPI_BOTTOM as both 
buffers  as intending MPI_IN_PLACE, I would say "no".



Dick Treumann  -  MPI Team 
IBM Systems & Technology Group
Dept X2ZA / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601
Tele (845) 433-7846         Fax (845) 433-8363




From:
Marc-André Hermanns <m.a.hermanns at grs-sim.de>
To:
mpi-forum at lists.mpi-forum.org
Date:
10/07/2010 04:36 AM
Subject:
[Mpi-forum] Handling of MPI_BOTTOM in Reductions
Sent by:
mpi-forum-bounces at lists.mpi-forum.org



Hi all,

I have a question on the use of MPI_BOTTOM for the following function 
calls: 

MPI_Reduce
MPI_Reduce_scatter
MPI_Scan
MPI_Exscan
MPI_Reduce_local
MPI_Reduce_scatter_block

They only have a single datatype, but two buffers. With MPI_BOTTOM as 
the  address of the buffer is encoded in the datatype. As I only have 
a single datatype, does MPI_BOTTOM imply MPI_IN_PLACE as the other
argument? 

Cheers,
Marc-Andre
--
Marc-Andre Hermanns
German Research School for
Simulation Sciences GmbH
c/o Laboratory for Parallel Programming
52056 Aachen | Germany

Tel +49 241 80 99753
Fax +49 241 80 6 99753
Web www.grs-sim.de

Members: Forschungszentrum Jülich GmbH | RWTH Aachen University
Registered in the commercial register of the local court of
Düren (Amtsgericht Düren) under registration number HRB 5268
Registered office: Jülich
Executive board: Prof. Marek Behr Ph.D. | Dr. Norbert Drewes

[attachment "smime.p7s" deleted by Richard Treumann/Poughkeepsie/IBM] 
_______________________________________________
mpi-forum mailing list
mpi-forum at lists.mpi-forum.org
http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpi-forum/attachments/20101007/de4ee3a2/attachment-0001.html>