Skip to content

Disable Mooncake on prerelease; bump min Julia to 1.10.2; regroup CI tests #2569

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

Merged
merged 8 commits into from
May 30, 2025

Conversation

penelopeysm
Copy link
Member

@penelopeysm penelopeysm commented May 29, 2025

Closes #2570

Mooncake doesn't precompile on 1.12, so this is a very similar fix to TuringLang/DynamicPPL.jl#921. I was getting tired of CI not passing, and also the whole env not precompiling meant that potential real issues would be masked.

In the process, we discovered that sample() was taking a very long time to compile on v1.10.0. This PR therefore also bumps the minimum version to 1.10.2, which is the first patch where this performance bug is fixed. See #2570 for more discussion (and #2571 where I was bisecting Julia patch versions).

I also edited CI to balance out the test times. The new groups are:

  • mcmc/gibbs: ~ 45 minutes
  • mcmc/Inference: ~10 minutes
  • ad: ~30 minutes
  • everything else: ~ 30 minutes

Copy link
Contributor

Turing.jl documentation for PR #2569 is available at:
https://TuringLang.github.io/Turing.jl/previews/PR2569/

@penelopeysm penelopeysm force-pushed the py/no-mooncake-pre branch from 9339cbb to 22569ee Compare May 29, 2025 16:30
Copy link

codecov bot commented May 29, 2025

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 85.86%. Comparing base (1a70627) to head (30706a8).
Report is 1 commits behind head on main.

Additional details and impacted files
@@           Coverage Diff           @@
##             main    #2569   +/-   ##
=======================================
  Coverage   85.86%   85.86%           
=======================================
  Files          21       21           
  Lines        1429     1429           
=======================================
  Hits         1227     1227           
  Misses        202      202           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@coveralls
Copy link

coveralls commented May 29, 2025

Pull Request Test Coverage Report for Build 15346521153

Details

  • 0 of 0 changed or added relevant lines in 0 files are covered.
  • No unchanged relevant lines lost coverage.
  • Overall coverage remained the same at 85.924%

Totals Coverage Status
Change from base Build 15328534629: 0.0%
Covered Lines: 1227
Relevant Lines: 1428

💛 - Coveralls

@penelopeysm penelopeysm changed the title Disable Mooncake on prerelease Disable Mooncake on prerelease; regroup CI tests May 29, 2025
@penelopeysm penelopeysm requested a review from mhauru May 29, 2025 21:06
Copy link
Member

@mhauru mhauru left a comment

Choose a reason for hiding this comment

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

Looks good. I have a slight preference for keeping more than 3 groups for quicker feedback.

@penelopeysm
Copy link
Member Author

Regarding test grouping, stdlib/distributions has dropped to only 5 minutes, so I lumped it into "everything else". The rest is the same.

Between #2564 and this PR, we've gone from 6 test groups and 2.5 hours to 4 test groups and 45 minutes 😄

@penelopeysm penelopeysm changed the title Disable Mooncake on prerelease; regroup CI tests Disable Mooncake on prerelease; bump min Julia to 1.10.2; regroup CI tests May 30, 2025
@penelopeysm penelopeysm merged commit 43a614b into main May 30, 2025
33 checks passed
@penelopeysm penelopeysm deleted the py/no-mooncake-pre branch May 30, 2025 13:07
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.

Specifically, calling sample() seems inordinately slow on 1.10
3 participants