Date: Fri, 19 Mar 1999 18:43:05 -0600 (CST)
From: Tony Skjellum <tony@XXXXXXXXXXXXXXXXXXXXX>
To: Falk Zimmermann <falk@XXXXXXXXXXXXXXXXXXXXXXXXXXX>
cc: mpi-core@XXXXXXXXXXX
Subject: Re: MPI-2
In-Reply-To: <9903190945.ZM1617@XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-mpi-core@XXXXXXXXXXX
Precedence: bulk
X-UIDL: 9b005fa0c3a5c813746d254f4bef9637

I recall that there is an issue with the identify operator being undefined
and would need to be applicable for rank 0...

On Fri, 19 Mar 1999, Falk Zimmermann wrote:

> Date: Fri, 19 Mar 1999 09:45:07 +0000
> From: Falk Zimmermann <falk@XXXXXXXXXXXXXXXXXXXXXXXXXXX>> To: mpi-core@XXXXXXXXXXX
> Subject: MPI-2
> 
> Dear ALL,
> 
> I have a question concerning the definition of the new collective function
> MPI_EXSCAN: Why doesn't the definition allow the "in place" option, while
> the ordinary (inclusive) MPI_SCAN supports it ?
> 
> Best regards
> Falk Zimmermann
> 
> ****************************************************************
> Falk Zimmermann
> E-mail:  falk@ccrl-nece.technopark.gmd.de
> Tel. +49 2241 925235
> Fax. +49 2241 925299
> 				Research Staff Member
> 		                C & C Research Laboratories
>                                 NEC Europe Ltd.
>                                 Rathausallee 10
>                                 D-53757 Sankt Augustin
>                                 Germany
> *****************************************************************
> 
> 

A. Skjellum, PhD, Assoc. Prof. of Computer Science; Mississippi State Univ.
http://www.cs.msstate.edu/~tony; tony@XXXXXXXXXXXXXXX 601-325-8435 (FAX -8997)
"Unpopularity := A failure to poll-test policy before implementation."