<div dir="ltr">Helvi<div><br></div><div>The preferred way would be to create a Simfactory entry. The "unset MPI" command would also go there.</div><div><br></div><div>If you are unfamiliar with Simfactory, then you would open an issue ticket on the ET web site and attach the option list.</div><div><br></div><div>-erik</div><div><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Oct 19, 2017 at 8:26 AM, hwitek <span dir="ltr"><<a href="mailto:hwitek@ffn.ub.es" target="_blank">hwitek@ffn.ub.es</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi everyone,<br>
<br>
we could compile the toolkit on MareNostrum4 without any further<br>
problems ("unset MPI" did the trick), and I want to run some further<br>
tests before providing the configuration script.<br>
<br>
What is the standard way to include new configuration files into the<br>
toolkit? Should I send it to someone in particular?<br>
<br>
cheers,<br>
Helvi<br>
<div class="HOEnZb"><div class="h5"><br>
On 2017-10-18 19:16, hwitek wrote:<br>
> Hi Miguel, everyone,<br>
><br>
> regarding your first point: I had a similar problem on its<br>
> predecessor,<br>
> MareNostrum 3.<br>
> The trick was to unset the global MPI variable, so in my bashrc I had<br>
> something like<br>
><br>
> module load intel<br>
> ...<br>
> unset MPI<br>
><br>
> I hope this helps! I'll also have a go at the compilation myself<br>
> tonight.<br>
><br>
> cheers,<br>
> Helvi<br>
><br>
><br>
> On 2017-10-18 18:52, Miguel Zilhão wrote:<br>
>> hi all,<br>
>><br>
>> i've been compiling the latest ET on marenostrum 4, and there were a<br>
>> couple of issues i'd like to ask:<br>
>><br>
>> 1. on marenostrum, they bundle a lot of mpi-related tools on modules.<br>
>> in particular, one of the<br>
>> default modules that is loaded is called 'impi'. this module sets the<br>
>> global variable $MPI to 'impi', ie<br>
>><br>
>> $ echo $MPI<br>
>> impi<br>
>><br>
>> this seems to severely interfere with the Cactus building process. at<br>
>> the configuration stage i get<br>
>> the following:<br>
>><br>
>> Configuring with flesh MPI<br>
>> Warning: use of flesh MPI via MPI option is deprecated and should<br>
>> be replaced with the thorn<br>
>> ExternalLibraries/MPI and its MPI_DIR option<br>
>> MPI selected, but no known MPI method - what is "impi" ?<br>
>><br>
>> to workaround this problem, i've unloaded the module (module unload<br>
>> impi), which undefines the $MPI<br>
>> variable. after this, the configuration stage works fine.<br>
>> this seems to me like a bug, though. after all, a cluster-wide global<br>
>> variable called $MPI seems<br>
>> like a natural thing to exist in a lot of these machines... should<br>
>> the building of Cactus rely on<br>
>> the non-existence of such a variable?<br>
>><br>
>> the other inconvenient thing is that, by unloading the impi module,<br>
>> one also removes from $PATH the<br>
>> mpirun command. so one has to unload the module for compiling the<br>
>> code, and then load it back to be<br>
>> able to run it.<br>
>><br>
>> 2. i was not able to compile with any of the provided intel compilers<br>
>> (with gcc it worked fine). i<br>
>> know there are known issues with some versions of the intel<br>
>> compilers; is there some sort of list of<br>
>> intel compilers that are known to work? i could maybe try to ask the<br>
>> technical support if they could<br>
>> make those specific versions available...<br>
>><br>
>> thanks,<br>
>> Miguel<br>
>> ______________________________<wbr>_________________<br>
>> Users mailing list<br>
>> <a href="mailto:Users@einsteintoolkit.org">Users@einsteintoolkit.org</a><br>
>> <a href="http://lists.einsteintoolkit.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.einsteintoolkit.<wbr>org/mailman/listinfo/users</a><br>
<br>
--<br>
===========================<br>
Dr. Helvi Witek<br>
Marie-Curie Research Fellow<br>
Dep. Fisica Quantica i Astrofisica<br>
Universitat de Barcelona<br>
===========================<br>
______________________________<wbr>_________________<br>
Users mailing list<br>
<a href="mailto:Users@einsteintoolkit.org">Users@einsteintoolkit.org</a><br>
<a href="http://lists.einsteintoolkit.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.einsteintoolkit.<wbr>org/mailman/listinfo/users</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div>Erik Schnetter <<a href="mailto:schnetter@cct.lsu.edu" target="_blank">schnetter@cct.lsu.edu</a>><br><a href="http://www.perimeterinstitute.ca/personal/eschnetter/" target="_blank">http://www.perimeterinstitute.ca/personal/eschnetter/</a></div><div><br></div></div></div>
</div>