Newsgroups: comp.parallel.mpi
From: James Cownie <jcownie@bbn.com>
Subject: Re: MPI_PROC_NULL and invalid message tags
Organization: BBN
Date: 6 Mar 1996 09:16:05 GMT
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <4hjl4l$e7f@newsgate.dircon.co.uk>

>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.
>

If that's the case, then the standard has taken a decision and said that this
is an erroneous program. Then Raja is right, and whether the
error gets diagnosed becomes a quality of implementation issue. The other 
possiblity is that this is a valid program, in which case implementations
must not raise an error. 

There's a clear difference between whether this is an erroneous program or not.
Whether erroneous programs are diagnosed, or start world war 3 is up to 
the implementation, but the implementation is not allowed to start world war 3
on valid programs, so we do need to be clear about this.

-- Jim
James Cownie 
BBN


