Skip to content

{vis} [intel-2015b] Mesa 11.0.2 (REVIEW)#2038

Merged
boegel merged 8 commits intoeasybuilders:developfrom
hajgato:mesa-bump
Oct 9, 2015
Merged

{vis} [intel-2015b] Mesa 11.0.2 (REVIEW)#2038
boegel merged 8 commits intoeasybuilders:developfrom
hajgato:mesa-bump

Conversation

@hajgato
Copy link
Copy Markdown
Collaborator

@hajgato hajgato commented Oct 7, 2015

Needs PR #2030 #2031

@hpcugentbot
Copy link
Copy Markdown

Easyconfigs unit test suite FAILed.

See https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/4670/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.

@hpcugentbot
Copy link
Copy Markdown

Easyconfigs unit test suite FAILed.

See https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/4691/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
Copy link
Copy Markdown
Collaborator Author

hajgato commented Oct 8, 2015

@boegel I am sorry, its getting to be fat.

@boegel
Copy link
Copy Markdown
Member

boegel commented Oct 8, 2015

@hajgato: well, you did a really nice job on building up towards this with a whole series of other PRs, so let's roll with it

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

@hajgato: M4 here too in Mesa-10.5.5-intel-2015a-Python-2.7.10.eb, so add it here too?

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

@boegel it is a bug or a feature. When I put M4-intel-2015b, then eb said that this conficts with M4-GNU-4.9.3-2.25. And it is a builddep in Autotools. I have checked the path in the eb log, and the path for M4-GNU-4.9.3-2.25 was clearly there. software/M4/1.4.17-GNU-4.9.3-2.25/bin I can put it back, but then it must be the GNU-4.9.3-2.25 version

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

eb produces the conflict, or the tests report a conflict?

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

eb produced the conflict.

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

@boegel anyway, its easy to check, M4-intel-2015b in this PR.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

That means the conflict is caused by one of the other build dependencies, which has M4 as a (runtime) dep?

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

@boegel Ok, I figured out Autotools have libtool dep, and libtool have a M4 runtime dep. So whats now? What M4 builddep should I add to Mesa?
https://gist.github.com/hajgato/750e9cd5dbb23a1b0010

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

OK, so Autotools load libtool, which has M4 as a runtime dep, that's where the conflict comes.

But, that also means there's no need to include it as a direct dep here, since M4 will be loaded as a build dep when building Mesa.

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

@boegel So then is this PR finished?

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

testing it as we speak, should be good to go I think

@hpcugentbot
Copy link
Copy Markdown

Easyconfigs unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/4694/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.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

same here, we already have libXext-1.3.3-intel-2015a-Python-2.7.10.eb, so keep the versionsuffix?

@boegel
Copy link
Copy Markdown
Member

boegel commented Oct 8, 2015

Test report by @boegel
FAILED
Build succeeded for 15 out of 16 (16 easyconfigs in this PR)
Linux centos linux 7.1.1503, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz, Python 2.7.5
See https://gist.github.com/7d11285942dab5f673a5 for a full test report.

@boegel
Copy link
Copy Markdown
Member

boegel commented Oct 8, 2015

Test report by @boegel
SUCCESS
Build succeeded for 1 out of 1 (16 easyconfigs in this PR)
Linux centos linux 7.1.1503, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz, Python 2.7.5
See https://gist.github.com/4f1dd0f2f7792775abcb for a full test report.

@boegel
Copy link
Copy Markdown
Member

boegel commented Oct 8, 2015

@hajgato: it works, just some minor style remarks mostly w.r.t. versionsuffixes; since we already have easyconfigs for those software versions, let's stick to the versionsuffix we're using there?

@hajgato
Copy link
Copy Markdown
Collaborator Author

hajgato commented Oct 9, 2015

@boegel I delibearately deleted the Python versionsuffxes, becase they were only there becasue we had two Python versions for 2015a toolchains. I tried to follow the rule that only those packages get the Python versionsuffix that have a direct Python dependency and not their dependencies have Python dependency. I can put back those versionsuffixes, but at the end every easyconfig will have Python versionsuffix.

@boegel
Copy link
Copy Markdown
Member

boegel commented Oct 9, 2015

@hajgato: well, the real solution for this issue lies elsewhere anyway, I'm just trying to keep things somewhat consistent, but it's a battle we can't win anyway; so, leave it as is

@hajgato
Copy link
Copy Markdown
Collaborator Author

hajgato commented Oct 9, 2015

@boegel I am open to everything, so if you say we should put the versionsuffix because it was there before, it is fine for me.

@boegel
Copy link
Copy Markdown
Member

boegel commented Oct 9, 2015

@hajgato: what we really want as a solution to this is support for version ranges (('Python', '>= 2.7')), since the Python version doesn't really matter in all of this, it's only a build dep (I think)

@boegel
Copy link
Copy Markdown
Member

boegel commented Oct 9, 2015

Test report by @boegel
SUCCESS
Build succeeded for 16 out of 16 (16 easyconfigs in this PR)
Linux centos linux 7.1.1503, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz, Python 2.7.5
See https://gist.github.com/1d0707765cc50c7d2a8f for a full test report.

@boegel
Copy link
Copy Markdown
Member

boegel commented Oct 9, 2015

Test report by @boegel
SUCCESS
Build succeeded for 16 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/159e87eb0e2248c97fc7 for a full test report.

@hpcugentbot
Copy link
Copy Markdown

Easyconfigs unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/4703/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.

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.

3 participants