Skip to content
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

OpenFOAM with extranal cgal, metis, and paraview, with deps using intel2... #1166

Closed
wants to merge 13 commits into from

Conversation

hajgato
Copy link
Collaborator

@hajgato hajgato commented Oct 28, 2014

@hpcugentbot
Copy link
Contributor

Automatic reply from Jenkins: Can I test this?

@boegel
Copy link
Member

boegel commented Nov 13, 2014

Jenkins: ok to test

@hpcugentbot
Copy link
Contributor

Test FAILed.

@hpcugentbot
Copy link
Contributor

Test FAILed.

1 similar comment
@hpcugentbot
Copy link
Contributor

Test FAILed.

@boegel
Copy link
Member

boegel commented May 31, 2015

Jenkins: test this please

@hpcugentbot
Copy link
Contributor

Refer to this link for build results (access rights to CI server needed):
https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3176/
Easyconfigs unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3176/console for more details).

This pull request is now ready for review/testing.

Please try and find someone who can tackle this; contact @boegel if you're not sure what to do.

@@ -0,0 +1,23 @@
name = 'METIS'
version = '5.1.0'
versionsuffix = '-32bitIDX'
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can you clarify this versionsuffix?

The only 'special' thing I see in here is that the METIS shared libs get built?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

boegel: Well, the other special thing is the absence of the METIS_IDXTYPEWIDTH.patch
For some reason OpenFOAM 2.3.0 needs 32 bit index, probably beacuse it uses fortran code. Instead of patching OpenFOAM, I made the 32bit index version of METIS.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, that explains it.

Maybe add a comment right above the versionsuffix to clarify that? Something like "default 32-bit IDTYPEWIDTH, no patch used" ?

@boegel
Copy link
Member

boegel commented Jun 3, 2015

Test report by @boegel
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in this PR)
Linux SL 6.6, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.6.6
See https://gist.github.com/314ea5ad5201754540f7 for a full test report.

@hajgato
Copy link
Collaborator Author

hajgato commented Jun 3, 2015

@boegel: Comment forom CGAL removed

@hpcugentbot
Copy link
Contributor

Refer to this link for build results (access rights to CI server needed):
https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3255/
Easyconfigs unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3255/console for more details).

This pull request is now ready for review/testing.

Please try and find someone who can tackle this; contact @boegel if you're not sure what to do.

@hpcugentbot
Copy link
Contributor

Refer to this link for build results (access rights to CI server needed):
https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3260/
Easyconfigs unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3260/console for more details).

This pull request is now ready for review/testing.

Please try and find someone who can tackle this; contact @boegel if you're not sure what to do.

@boegel
Copy link
Member

boegel commented Sep 22, 2015

Jenkins: test this please

@boegel boegel modified the milestones: merge sprint 2.0, v2.4.0 Sep 22, 2015
@hpcugentbot
Copy link
Contributor

Easyconfigs unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/4409/console for more details).

This pull request is now ready for review/testing.

Please try and find someone who can tackle this; contact @boegel if you're not sure what to do.

description = "Paraview is a scientific parallel visualizer"

toolchain = {'name': 'intel', 'version': '2014b'}
toolchainopts = {'usempi': False, 'pic': True}
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@hajgato: how does this compare with the Paraview easyconfig in #761 ?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This does not have a mesa dependency, but it should have a mesa osdependency.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

see #1618

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

added old style mesa (10.4.5) with osdependencies = [
'libX11-devel', # Xlibs.h
'xorg-x11-proto-devel', # X.h, glproto, xproto
'libXdamage-devel',
'libXext-devel',
'libXfixes-devel',
]

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

a la. PR #1439

@boegel
Copy link
Member

boegel commented Sep 22, 2015

@hajgato: please sync this with develop (to make sure Jenkins is running the latest version of the tests)

@wpoely86: can you give this a thorough review/test?

@hpcugentbot
Copy link
Contributor

Easyconfigs unit test suite FAILed.

See https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/4428/console for more details.

Please fix the reported issues by pushing additional commits to the branch corresponding with this pull request; contact @boegel if you're not sure what to do.

]

builddeps = [
('CMake', '3.0.0'),
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

upgrade to 3.1.0

@wpoely86
Copy link
Member

General remark: create a Autotools easyconfig version with what you need and use that everywhere instead of the individual tools.

@wpoely86
Copy link
Member

Test report by @wpoely86
FAILED
Build succeeded for 0 out of 16 (16 easyconfigs in this PR)
Linux SL 6.7, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.6.6
See https://gist.github.com/a79681ade8a15b37eb91 for a full test report.

@hajgato
Copy link
Collaborator Author

hajgato commented Sep 23, 2015

We have started to use autotools from 2015a toolchains. Should I change PR to autotools? And if Yes, which version?

@wpoely86
Copy link
Member

I would do that yes. Pick the latest version and change the toolchain version accordingly.

@hajgato
Copy link
Collaborator Author

hajgato commented Sep 23, 2015

configopts = '--enable-nouveau-experimental-api' removed from libdrm (all versions)


dependencies = [
('Qt','4.8.6'),
('Mesa', '10.4.5', '-Python-2.7.8'),
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

wouldn't it be better go with 10.5.5 from #1618, and drop the Mesa-related stuff in this PR?

I'm not very keen on having lots of Mesa easyconfigs around, since they have a huge number of deps...

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@boegel: well, this is 2014b and PR #1618 is 2015a. If I go with the 10.5.5, then I have to make triple the deps than in the case of 10.4.5 (10.4.5 with osdeps, 10.5.5 without osdeps, so much more deps.)

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

or we can promote this to 2015a?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@boegel: yeah, I already started to make this with 2015a, but I have never finished. Either the PR is too big, or it has lots of deps. I would like to finish this as it is, or abandon this PR and finish 2015a vesrion (which should also be bumped to 2015b, and OpenFOAM bumped to 2.4.X), and honestly this OpenFOAM feels for me a never ending story.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm OK with abandoning this in favor of a 2015a or 2015b effort (the latter implies also redoing Mesa with 2015b...)

It's important to keep PRs somewhat minimal; this PR already throws a whole bunch of things together (see the libdrm and libxml2 changes).

Paraview should be handled in a separate PR for example, not involving OpenFOAM

if remarks are made to an easyconfig that should be fixed in all existing easyconfigs for that software (cfr. libdrm) handle that in a separate PR too (to avoid blocking this PR for it)

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OK, I will flesh out libdrm and libxml2 from this pr. and close this one. This means that OpenFOAM external will not be in this sprint.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the sprint is more intended for 'low-hanging fruit', I wouldn't call a PR with 50+ touched files that ;)

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok, originally it was much less. And I meat OpenFOAM with 2015a

@hpcugentbot
Copy link
Contributor

Easyconfigs unit test suite FAILed.

See https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/4456/console for more details.

Please fix the reported issues by pushing additional commits to the branch corresponding with this pull request; contact @boegel if you're not sure what to do.

@hajgato hajgato closed this Sep 25, 2015
@hajgato hajgato deleted the openfoam-intel2014b branch October 9, 2015 15:39
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants