New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Added support for Parastation MPICH toolchains with Intel compilers #1072
Conversation
Automatic reply from Jenkins: Can I test this? |
Intel Math Kernel Library (MKL) and Intel FFTW wrappers. | ||
""" | ||
NAME = 'intel-para' | ||
BLACS_LIB = ["mkl_blacs_intelmpi%(lp64)s"] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
maybe add a comment here why you're overriding what is being set by IntelMKL
?
is this a fix specific to Parastation?
also: you're including the 'generic' MPICH
component here, yet naming the toolchain specifically for Parastation.
Maybe here we should be using:
class Impmkl(IntelIccIfort, Mpich, IntelMKL, IntelFFTW):
NAME = 'impmkl'
BLACS_LIB = [...]
and then on your end:
class IntelPara(Impmkl):
NAME = 'intel-para'
I'm OK with including the 'empty' intel-para
toolchain definition in the framework as well if that makes things easier for you
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not sure if the blacs fix is specific to Parastation or not. Perhaps it's best to put it in the empty toolchain
Jenkins: ok to test |
Test PASSed. |
I've updated things so that generic patterns are created first and then wrapped with ParaStation specific extras. |
Test PASSed. |
2 similar comments
Test PASSed. |
Test PASSed. |
@ocaisa: as far as I can tell, there are no easyconfigs yet for |
""" | ||
NAME = 'intel-para' | ||
# Parastation MPI needs to be matched with the IntelMPI blacs library | ||
BLACS_LIB = ["mkl_blacs_intelmpi%(lp64)s"] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
is this still needed?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@ocaisa: as discussed, this should be fixed in toolchains/linalg/intelmkl.py
instead, since the MPICH2
value is now also wrong?
… and MPICH2 and corrected toolchain to match
Note https://software.intel.com/en-us/articles/intel-mkl-link-line-advisor to help figure out BLACS library for mpi implementation |
Test FAILed. |
Simple support for Parastation MPICH with intel compilers