From: Peter Schaefer <schaefer@malaga.math.uni-augsburg.de>
Newsgroups: comp.parallel.mpi
Subject: Re: MPI_Barrier - how about argument checking ?
Date: Tue, 15 Sep 1998 10:10:48 +0200
Organization: Spammers against death penalty (SPADE)
Distribution: world
Message-Id: <35FE2108.41C6@malaga.math.uni-augsburg.de>
References: <35FD350E.41C6@malaga.math.uni-augsburg.de>
    <35FE193D.44BD5EAF@lfbs.rwth-aachen.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit


To be more clear:

I wrote 
MPI_Barrier(nprocs);

and tested with nprocs=4;
MPI is mpich Version 1.0.13, IRIX/ch_p4 (SGI)

correct is
MPI_Barrier(MPI_COMM_WORLD);
MPI_COMM_WORLD is an integer.

I thought this was worthwhile to note, since 
I got no error message, no immediate seg'fault,
and the program continued to run till it overwrote
data used by the loop (with 0x10009da6 values (CC -g ) ).

Thanks
--
Peter Schaefer           mailto:schaefer@malaga.math.uni-augsburg.de     
X-Url: http://wwwhoppe.math.uni-augsburg.de/~schaefer/