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.

Sign up for GitHub

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

Jump to bottom

0.56.4 lints, try to convice Travis CI to build for ppc64le #109

Draft
wants to merge 14 commits into
base: main
Choose a base branch
from
Draft

0.56.4 lints, try to convice Travis CI to build for ppc64le #109

wants to merge 14 commits into from

Conversation

mbargull
Copy link
Member

Checklist

  • Used a personal fork of the feedstock to propose changes
  • Bumped the build number (if the version is unchanged)
  • Reset the build number to 0 (if the version changed)
  • Re-rendered with the latest conda-smithy (Use the phrase @conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
  • Ensured the license file is being packaged.

closes gh-92
fixes gh-102

@conda-forge-linter
Copy link

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

@mbargull
Remove redundant llvm-openmp test dependency
b1de8fc 
llvm-openmp uses strong run_exports entry as as such is a run dependency
@mbargull
Use pin_compatible for tbb
d821153 
So we can have the version constraint in one place.
@mbargull
Remove platform selectors from run_constrained tbb
98641b7 
1. Recipe was inconsistent since selectors were missing for tbb-devel
   (so we probably wanted the run_constrained everywhere anyway).
2. 32-bit platform selectors are removed in upstream recipe for >=0.57.
3. aarch64/ppc64le TBB available on conda-forge/anaconda for a while now
@mbargull
Use newer build container
06c16cc 
This was introduced to fix an issue that turned out to be actually only
relevant for pkgs/main::numba but not our conda-forge::numba builds.
refs:
 - conda-forge#89 (comment)
@gmarkall gmarkall mentioned this pull request Nov 22, 2022
Closed
5 tasks
@mbargull
Copy link
Member Author

ppc64le seems to work now with verbose test output and redirecting stderr to stdout (the oddest thing...). (The PYTHONUNBUFFERED=1 might or might not be needed.)

But I'm still leery of those Travis/ppc64le runs, so I'll just restart those a couple of times to see how it behaves.

@mbargull
Copy link
Member Author

BTW:
The ppc64le builds took around 30-45 minutes, i.e., the -j --random 0.15 seems to be quite appropriate given the 50 minute limit.
The aarch64 builds seem to finish in under 15 minutes for the last few months, i.e., we could maybe try to up the portion of selected tests to something around 0.3 to 0.5 for that platform.

@mbargull
Copy link
Member Author

@conda-forge-admin, please restart ci

@mbargull
Copy link
Member Author

@conda-forge-admin, please restart ci

@mbargull
Copy link
Member Author

@conda-forge-admin, please restart ci

@jakirkham
Copy link
Member

Restarted one of the jobs that errored without a log

@gmarkall
Copy link
Contributor

The PPC64le builds that error with no output after 10 minutes all seem to have a hostname in common, *.lxd-ppc64le-travis-ci-production-1-worker1-saopaulo. Examples:

@gmarkall
Copy link
Contributor

The PPC64le builds that error with no output after 10 minutes all seem to have a hostname in common

(Just to be clear this is the ones that don't even appear to start cloning, not the ones that were actually running tests and timing out during the tests as in earlier runs, before #109 (comment))

@jakirkham
Copy link
Member

jakirkham commented Nov 23, 2022 •
edited
Loading

FWIW the restarted one passed

Maybe we should follow up with Travis CI on the problematic worker?

Edit: Sent an email to Travis CI with a link to Graham's comment above ( #109 (comment) )

Edit 2: Link to issue ( https://support.travis-ci.com/hc/requests/41747 ). Idk if others can view this (guessing not). In any event they say this issue is fixed and we should retry. So restarting on main ( https://app.travis-ci.com/github/conda-forge/numba-feedstock/builds/258052004 )

Edit 3: Restarting failed with the same timeout issue. Asking Travis CI to take another look.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@henryiii henryiii Awaiting requested review from henryiii henryiii will be requested when the pull request is marked ready for review henryiii is a code owner

@jakirkham jakirkham Awaiting requested review from jakirkham jakirkham will be requested when the pull request is marked ready for review jakirkham is a code owner

@marcelotrevisani marcelotrevisani Awaiting requested review from marcelotrevisani marcelotrevisani will be requested when the pull request is marked ready for review marcelotrevisani is a code owner

@souravsingh souravsingh Awaiting requested review from souravsingh souravsingh will be requested when the pull request is marked ready for review souravsingh is a code owner

@step21 step21 Awaiting requested review from step21 step21 will be requested when the pull request is marked ready for review step21 is a code owner

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Numba 0.56.2 ppc64le builds failing on main numba appears to find cos7 symbols
4 participants
@mbargull @conda-forge-linter @jakirkham @gmarkall