Newsgroups: comp.parallel.mpi
From: raja@convex.com (Raja Daoud)
Subject: Re: MPI_PROC_NULL and invalid message tags
Organization: Hewlett-Packard Co.; Convex Technology Center
Date: 5 Mar 1996 18:58:55 -0600
Message-ID: <4hio0f$i8t@tbag.convex.com>

Steve Huss-Lederman <lederman@reggiano.cs.wisc.edu> wrote:
> > What about an invalid pointer for a zero-length message?
>My recollection of the discussion during the MPI Forum is that we
>never addressed this issue directly.  My reading of the standard would
>be that you can choose to generate an error or not.

It depends on what the implementation uses for MPI_BOTTOM in C.  If it
sets it to (void *) 0 (a valid approach), then it cannot distinguish
between NULL and MPI_BOTTOM, so it can't test for this "error".
The main issue is: what is an invalid buffer (choice) argument?  If
an implementation can cheaply test for some cases, go for it.  If it
can't, then it's ok too.  This is a quality of implementation issue,
and I don't think it is something to standardize.

--Raja

-=-
Raja Daoud				Hewlett-Packard Co.
raja@convex.com				Convex Technology Center

