<html><body>
<p>I am OK with the proposed changes<br>
<br>
Dick Treumann - MPI Team/TCEM <br>
IBM Systems & Technology Group<br>
Dept 0lva / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601<br>
Tele (845) 433-7846 Fax (845) 433-8363<br>
<br>
<br>
<tt>"Rolf Rabenseifner" <rabenseifner@hlrs.de> wrote on 01/29/2008 04:06:49 AM:<br>
<br>
> This is a proposal for MPI 2.1, Ballot 4.<br>
> <br>
> I'm asking especially <br>
> Karl Feind, Bill Gropp, Andrew Lumsdaine, Dick Treumann,<br>
> the participants of the email-discussion in 2001, to review this proposal.<br>
> <br>
> This is a follow up to:<br>
> MPI_Abort<br>
> in <a href="http://www.cs.uiuc.edu/homes/wgropp/projects/parallel/MPI/mpi-">http://www.cs.uiuc.edu/homes/wgropp/projects/parallel/MPI/mpi-</a><br>
> errata/index.html<br>
> with mail discussion in<br>
> <a href="http://www.cs.uiuc.edu/homes/wgropp/projects/parallel/MPI/mpi-">http://www.cs.uiuc.edu/homes/wgropp/projects/parallel/MPI/mpi-</a><br>
> errata/discuss/abort/<br>
> ___________________________________<br>
> <br>
> Proposal:<br>
> MPI-1.1 Sect. 7.5, MPI_Abort, page 200, lines 23-26 read:<br>
> This routine makes a "best attempt" to abort all tasks in the <br>
> group of comm. This function does not require that the invoking <br>
> environment take any action with the error code. However, a Unix or <br>
> POSIX environment should handle this as a return errorcode from the<br>
> main program or an abort(errorcode).<br>
> but should read (" or an abort(errorcode)" removed):<br>
> This routine makes a "best attempt" to abort all tasks in the <br>
> group of comm. This function does not require that the invoking <br>
> environment take any action with the error code. However, a Unix or <br>
> POSIX environment should handle this as a return errorcode from the<br>
> main program.<br>
> ___________________________________<br>
> Rationale for this clarification:<br>
> POSIX defines void abort(void). The routine void exit(int status)<br>
> may be used to implement "handle this as a return errorcode from the<br>
> main program". abort(errorcode) was not substituted by exit(errorcode)<br>
> because this is technically not enough, if the MPI implementation<br>
> wants to return it also from mpiexec, see next proposal. <br>
> ___________________________________<br>
> <br>
> Proposal:<br>
> Add after MPI-1.1 Sect. 7.5, MPI_Abort, page 200, line 34 (end of rationale):<br>
> Advice to users. Whether the errorcode is returned from the executable<br>
> or from the MPI process startup mechanism (e.g., mpiexec), is an aspect<br>
> of quality of the MPI library but not mandatory. (End of advice tousers.) <br>
> ___________________________________<br>
> Rationale for this clarification:<br>
> The intent of word "should" in "should handle this as a return <br>
> errorcode from the main program" is only a quality of implementation<br>
> aspect and not a must. This is not clear and can be misinterpreted.<br>
> ___________________________________ <br>
> <br>
> <br>
> Best regards<br>
> Rolf<br>
> <br>
> <br>
> Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner@hlrs.de<br>
> High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530<br>
> University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832<br>
> Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner<br>
> Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)<br>
</tt></body></html>