-
Notifications
You must be signed in to change notification settings - Fork 3.6k
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
[Python][CI] test_dlpack is failing on some nightly jobs allocating unexpected memory #44728
Comments
This started to fail on the main CI as well (eg #44720). I don't directly see any difference in the conda env between the failing build and the last build on main (which is still green) .. |
I can't reproduce locally with:
The test is successful:
The only skipped test is the cuda requiring one, the same skipped in CI:
|
raulcd
added a commit
to raulcd/arrow
that referenced
this issue
Nov 20, 2024
…as / numpy nightly tests
raulcd
added a commit
that referenced
this issue
Nov 21, 2024
…allocated bytes for dlpack tests (#44793) ### Rationale for this change test_dlpack started failing for several jobs on CI due to our tests deallocating memory: ``` allocated_bytes = pa.total_allocated_bytes() try: return f(*args, **kwargs) finally: > assert pa.total_allocated_bytes() == allocated_bytes E assert 3728 == 7440 ``` ### What changes are included in this PR? Trigger a `gc.collect` before checking initial allocated bytes. ### Are these changes tested? Yes via CI ### Are there any user-facing changes? No * GitHub Issue: #44728 Authored-by: Raúl Cumplido <[email protected]> Signed-off-by: Raúl Cumplido <[email protected]>
Issue resolved by pull request 44793 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the bug, including details regarding any error messages, version, and platform.
Several jobs:
are failing for the
test_dlpack
test with some variants of:Component(s)
Continuous Integration, Python
The text was updated successfully, but these errors were encountered: