Newsgroups: comp.parallel.mpi
From: lederman@romano.cs.wisc.edu (Steve Huss-Lederman)
Subject: Re: PMPI
Organization: CS Department, University of Wisconsin
Date: 15 Dec 1995 14:57:48 GMT
Message-ID: <LEDERMAN.95Dec15085748@romano.cs.wisc.edu>

We have used the PMPI_ mechanism and it seems to work fine.  As far as
I know, MPICH does support this.  They even have default wrappers to
get profile info for upshot.  See the wrappergen stuff.  The MPI
standard requires the PMPI_ routines so any compliant implementation
must have it.  Thus, all MPIs should let you do this.  It is being
used by many people and packages.

Steve

