Subject: Re: MPI-2.1 corrections, Batch 2 / AlltoallW To: mpi-21@XXXXXXXXXXXXX X-Mailer: Lotus Notes Release 5.0.5 September 22, 2000 From: "Richard Treumann" Date: Fri, 20 Jul 2001 09:58:00 -0400 X-MIMETrack: Serialize by Router on D01ML064/01/M/IBM(Release 5.0.8 SPR #MIAS4UTJ8H |June 21, 2001) at 07/20/2001 09:59:10 AM MIME-Version: 1.0 Content-type: text/plain; charset=us-ascii Sender: owner-mpi-21@XXXXXXXXXXXXX Precedence: bulk Reply-To: mpi-21@XXXXXXXXXXXXX I like the name Rolf suggests. MPI_Alltoallx could stand for MPI_Alltoall (eXtended). I assume MPI_Alltoall-> MPI_Alltoallv in MPI-1 was to suggest the addition of vectors of displacements. I never knew what the choice of "w" in going from Alltoallv -> Alltoallw was about. Did the "w" stand for something? If we use Alltoallx and someday decide that is deficient we can do Alltoally and Alltoallz (;>). MPI_Alltoallve (Alltoall vector enhanced) or MPI_Alltoallvx (Alltoall vector extended) are a couple more options that would even be mnemonic. 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 ----- Forwarded by Richard Treumann/Poughkeepsie/IBM on 07/20/2001 09:26 AM ----- Rolf Rabenseifner cc: Sent by: Subject: Re: MPI-2.1 corrections, Batch 2 / AlltoallW owner-mpi-21@mpi-forum.or g 07/20/2001 08:48 AM Please respond to mpi-21 I started the discussion in Nov. 2000 and now, after reading all the arguments, I would also prefer to define a new function, e.g., MPI_Alltoallx, with MPI_Aint for the displacement arguments. And MPI_Alltoallw can go to the deprecated function list. Rolf Dr. Rolf Rabenseifner High Performance Computing Parallel Computing Center Stuttgart (HLRS) Rechenzentrum Universitaet Stuttgart (RUS) Phone: ++49 711 6855530 Allmandring 30 FAX: ++49 711 6787626 D-70550 Stuttgart rabenseifner@XXXXXXXXXXXXXXXXXXXX Germany http://www.hlrs.de/people/rabenseifner