Scope binning memory resource fixture to class in Python tests

[bdice]

Description

Follow-up to #2273. The binning memory resource test recreates a BinningMemoryResource (with FixedSizeMemoryResource bins backed by ManagedMemoryResource) for every (dtype, nelem, alloc, upstream_mr) combination. The managed memory slab allocations make each creation take ~0.5–0.9s, totaling ~40s for 147 tests.

Restructure the test into a class with a class-scoped fixture so that each BinningMemoryResource is built once per upstream_mr rather than once per test. A function-scoped autouse fixture sets the current device resource before each test.

Times measured on an RTX 2070 Super (8 GiB) under WSL2, where managed memory operations can be particularly expensive. Binning tests drop from ~39s to ~3s; full suite drops from ~49s to ~11s.

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: d2bda107-a6b6-4bf7-83cc-41f67b288635

📥 Commits

Reviewing files that changed from the base of the PR and between db4f3f6 and 475a745.

📒 Files selected for processing (1)

• python/rmm/rmm/tests/test_binning_memory_resource.py

---

📝 Walkthrough

Summary by CodeRabbit

• Tests
  • Refactored binning memory resource tests to a class-based structure using parameterized fixtures so each upstream memory resource is tested separately.
  • Added helper setup to construct per-upstream test resources and an automatic per-test setup to ensure the correct device resource is active.
  • Consolidated assertions into a class method that consumes the new fixtures.

Walkthrough

Refactors tests for BinningMemoryResource: adds helper _make_binning_mr(upstream_mr), introduces _UPSTREAM_MRS list, replaces top-level parametrized test with TestBinningMemoryResource class, and adds a class-scoped binning_mr fixture plus an autouse device-setup fixture.

Changes

Cohort / File(s)	Summary
Test Refactoring python/rmm/rmm/tests/test_binning_memory_resource.py	Added _make_binning_mr() helper, _UPSTREAM_MRS list, class-scoped @pytest.fixture binning_mr (parameterized over upstream factories), converted top-level parametrized test into class TestBinningMemoryResource with autouse device setup and test_binning_memory_resource method.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Possibly related PRs

• Reduce default pool sizes in Python tests to speed up suite #2273: Similar test refactor pattern converting parametrized tests to class-scoped fixtures and factory helpers.

Suggested reviewers

• TomAugspurger

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 75.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately describes the main change: restructuring a test to use a class-scoped fixture for the binning memory resource, which is the primary focus of the refactoring.
Description check	✅ Passed	The description is directly related to the changeset, explaining the motivation (test performance improvement), the approach (class-scoped fixture), and providing measurable impact data.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[bdice]

/merge