<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Feb 18, 2016 at 1:43 PM, Rolf Rabenseifner <span dir="ltr"><<a href="mailto:rabenseifner@hlrs.de" target="_blank">rabenseifner@hlrs.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">> I think you have (1) and (2) backwards in this paragraph.<br>
</span>Yes, I did it with the wrong numbers. I corrected it below.<br>
<span class=""><br>
> I think JeffH and Jed want (1).<br>
> I think I want (1), too.<br>
<br>
</span>I see the point.<br>
I have not a strong opinion and would like to know,<br>
what the other representatives in the Forum would like to see,<br>
because<br>
<br>
(2) MPI_DATATYPE_NULL is what they mainly have today.<br>
(1) "not existing" would be a real change.<br>
<span class=""><br></span></blockquote><div><br></div><div>Real only in the sense that things will break at compile-time instead of run-time. We are not breaking any working programs if we make this change.</div><div><br></div><div>Jeff</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">
Rolf<br>
<br>
----- Original Message -----<br>
> From: "Jeff Squyres" <<a href="mailto:jsquyres@cisco.com">jsquyres@cisco.com</a>><br>
> To: "MPI-WG Fortran working group" <<a href="mailto:mpiwg-fortran@lists.mpi-forum.org">mpiwg-fortran@lists.mpi-forum.org</a>><br>
</span><span class="">> Sent: Thursday, February 18, 2016 10:31:23 PM<br>
> Subject: Re: [MPIWG Fortran] Provide Fortran datatypes if Fortran bindings not provided?<br>
<br>
</span><div><div class="h5">> On Feb 18, 2016, at 3:55 PM, Rolf Rabenseifner <<a href="mailto:rabenseifner@hlrs.de">rabenseifner@hlrs.de</a>> wrote:<br>
>><br>
>> I would prefer a solution where an application or middlware can check at<br>
>> runtime whether the Fortran MPI datatypes exist, i.e., I like mpich's<br>
>> solution. It is more simple than all methods that People must check prior<br>
>> to compile and must remove or Change parts of there software.<br>
><br>
> The reason that JeffH and Jed hate this solution is because it makes it quite<br>
> difficult for an application to know whether the MPI has Fortran support or not<br>
> -- they basically have to compile an MPI application *and then run it* to see<br>
> if the datatypes are actually MPI_DATATYPE_NULL.<br>
><br>
> Doing this inside a GNU Autoconf/configure script is very, very difficult<br>
> because of the huge differences in how you launch MPI applications across<br>
> different HPC environments.<br>
><br>
> Their (strong) preference is that you can have a simple compilation test in the<br>
> GNU Autoconf/configure script that just tries to compile something like the<br>
> following:<br>
><br>
> -----<br>
> #include <mpi.h><br>
> MPI_Datatype foo = MPI_INTEGER;<br>
> -----<br>
><br>
> If that compiles, then your MPI has Fortran support.<br>
> If it doesn't compile, your MPI doesn't have Fortran support.<br>
><br>
>> In the moment,in the case of a missing accompanying Fortran Compiler,<br>
>> an application or middleware that wants to handle<br>
>> MPI Fortran datatypes must handle both cases<br>
>> (1) mpi.h does not provide a specific MPI Fortran datatype<br>
>> (2) mpi.h provides this MPI Fortran datatype as MPI_DATATYPE_NULL<br>
>><br>
>> The additional case, that it is mapped to a real datatype is not<br>
>> an exception, because it defines some sort of accompanying Fortran Compiler.<br>
>><br>
>> I would prefer that the MPI standard defines, which of the two<br>
>> options should be implemented.<br>
><br>
> I think JeffH and Jed want (1).<br>
><br>
> I think I want (1), too.<br>
><br>
</div></div> As said at the beginning, I would prefer (2),<br>
<span class="">>> - as mpich is doing it in general, and<br>
>> - OMPI is doing it partially,<br>
</span> whereas (1) was not done by mpich nor by OMPI.<br>
<span class="im HOEnZb">><br>
><br>
> --<br>
> Jeff Squyres<br>
> <a href="mailto:jsquyres@cisco.com">jsquyres@cisco.com</a><br>
> For corporate legal information go to:<br>
> <a href="http://www.cisco.com/web/about/doing_business/legal/cri/" rel="noreferrer" target="_blank">http://www.cisco.com/web/about/doing_business/legal/cri/</a><br>
><br>
> _______________________________________________<br>
> mpiwg-fortran mailing list<br>
> <a href="mailto:mpiwg-fortran@lists.mpi-forum.org">mpiwg-fortran@lists.mpi-forum.org</a><br>
> <a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-fortran" rel="noreferrer" target="_blank">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-fortran</a><br>
<br>
</span><span class="im HOEnZb">--<br>
Dr. Rolf Rabenseifner . . . . . . . . . .. email <a href="mailto:rabenseifner@hlrs.de">rabenseifner@hlrs.de</a><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 . . . <a href="http://www.hlrs.de/people/rabenseifner" rel="noreferrer" target="_blank">www.hlrs.de/people/rabenseifner</a><br>
Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307)<br>
</span><div class="HOEnZb"><div class="h5">_______________________________________________<br>
mpiwg-fortran mailing list<br>
<a href="mailto:mpiwg-fortran@lists.mpi-forum.org">mpiwg-fortran@lists.mpi-forum.org</a><br>
<a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-fortran" rel="noreferrer" target="_blank">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-fortran</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">Jeff Hammond<br><a href="mailto:jeff.science@gmail.com" target="_blank">jeff.science@gmail.com</a><br><a href="http://jeffhammond.github.io/" target="_blank">http://jeffhammond.github.io/</a></div>
</div></div>