Actually... perhaps that wasn't an answer but
it did give me some insights.
- I didn't notice that the alias it set was for
mpiicc... at first glance I thought it was an
alias pointing mpicc to mpiicc (w/intel compilers).
So that wasn't the setting I was seeking. [though
it did finally lead me to understand Fotis's question,
which was another clue that I was going down a
stray path]
- I discovered that EB controls mpicc default compiler
via I_MPI_CC. That was the part of equation I was missing.
I was concerned that all this time when building with
ictce/intel toolchains, since I only saw EB using mpicc,
that EB was using the default gcc wrapper. Now I see
that the variable above is set to icc. So, mpicc DOES use
icc by default when building with EB. Good!
Obviously I should've known (and trusted) the EB team
to do the right thing. I just needed to confirm to alleviate
my concerns.
Thanks...
jack
On 06/03/2015 09:46 AM, Jack Perdue wrote:
Howdy Alan,
Actually, the link did answer my question since it
had an example.
So I can just add something:
versionsuffix = '-mpiicc' # use mpiicc instead of mpicc
set_mpi_wrapper_aliases_intel=True
to achieve the desired effect? At least the
module looks right.
Jack Perdue
Lead Systems Administrator
High Performance Research Computing
TAMU Division of Research
[email protected] http://sc.tamu.edu
SC Helpdesk: [email protected]
On 06/03/2015 08:26 AM, Alan O'Cais wrote:
Ah sorry, that's not what you're asking...
On 3 June 2015 at 15:24, Alan O'Cais <[email protected]
<mailto:[email protected]>> wrote:
Have you checked out
https://github.com/hpcugent/easybuild-easyblocks/pull/493 ?
That supports flags to override the intel mpi defaults in the
module files.
Alan
On 3 June 2015 at 13:11, Fotis Georgatos <[email protected]
<mailto:[email protected]>> wrote:
Hi Jack,
I guess you ask this either as a safety measure or you have a
specific use-case atm.
Esp. if in the later case, can you please clarify the situation?
IMHO, this kind of things should rather be site-specific
policy matters (ie. tunable).
cheers,
F.
--
echo "sysadmin know better bash than english" | sed
s/min/mins/ \
| sed 's/better bash/bash better/' # Yelling in a CERN
forum
________________________________________
From: [email protected]
<mailto:[email protected]>
[[email protected]
<mailto:[email protected]>] on behalf of Jack
Perdue [[email protected] <mailto:[email protected]>]
Sent: Tuesday, June 02, 2015 06:07
To: [email protected] <mailto:[email protected]>
Subject: [easybuild] Intel toolchains and
mpiicc/mpiicpc/mpiifort
Howdy EBers,
Is there a (site-configurable) way to make the Intel toolchains
(ictce/intel/etc) use mpiicc (two 'i's) and friends instead of
just mpicc.
I see mention of mpiicc in easyblocks/i/impi.py, but
if that is usuable, I 'm missing some info If it isn't usable,
has someone spent any time seeing how to make EB support this?
I searched the mailing list but got limited results.
Just curious.
jack
-- Dr. Alan O'Cais
Application Support
Juelich Supercomputing Centre
Forschungszentrum Juelich GmbH
52425 Juelich, Germany
Phone: +49 2461 61 5213 <tel:%2B49%202461%2061%205213>
Fax: +49 2461 61 6656 <tel:%2B49%202461%2061%206656>
E-mail: [email protected] <mailto:[email protected]>
WWW: http://www.fz-juelich.de/ias/jsc/EN
--
Dr. Alan O'Cais
Application Support
Juelich Supercomputing Centre
Forschungszentrum Juelich GmbH
52425 Juelich, Germany
Phone: +49 2461 61 5213
Fax: +49 2461 61 6656
E-mail: [email protected] <mailto:[email protected]>
WWW: http://www.fz-juelich.de/ias/jsc/EN
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzender des Aufsichtsrats: MinDir Dr. Karl Eugen Huthmacher
Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender),
Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt,
Prof. Dr. Sebastian M. Schmidt
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
