Migrate stream APIs from rmm::cuda_stream_view to cuda::stream_ref

[bdice]

Summary

Track the coordinated migration of stream APIs and call sites from rmm::cuda_stream_view to CCCL's cuda::stream_ref. This propagates cuda::stream_ref through RMM containers and memory resources, RAFT resource and handle APIs, downstream C++ interfaces, Python/Cython bindings, benchmarks, tests, and documentation.

This updates cuDF C++ and Cython stream accessors, building on the default-stream work merged in #23770. It continues work from #23636.

Depends on rapidsai/rmm#2372.

Tracked in rapidsai/build-planning#318.

Migrations

• Pass cuda::stream_ref through stream pools, resource accessors, conditionals, and downstream APIs without converting to rmm::cuda_stream_view
• Use cuda::stream_ref constructions for default/legacy/per-thread streams
  • rmm::cuda_stream_default ➡️ cuda::stream_ref{cudaStream_t{cudaStreamDefault}}
  • rmm::cuda_stream_legacy ➡️ cuda::stream_ref{cudaStreamLegacy}
  • rmm::cuda_stream_per_thread ➡️ cuda::stream_ref{cudaStreamPerThread}
• Use .get() when calling an API that requires a raw cudaStream_t, including CUDA runtime, library, CUB, and legacy API boundaries (previously rmm::cuda_stream_view used value())
• Use .sync() when synchronizing a cuda::stream_ref (previously rmm::cuda_stream_view used synchronize())
• Update Cython declarations and call sites to pass stream references directly where supported

Checklist

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

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[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: Enterprise

Run ID: ee72e89d-96d7-4155-b6ae-37da23161b3f

📥 Commits

Reviewing files that changed from the base of the PR and between bb9fc22 and a61eee2.

📒 Files selected for processing (1)

• python/pylibcudf/pylibcudf/nvtext/unicode_normalize.pyx

Included review availability: Your plan provides up to 12 included reviews per hour; 11 remain after this review.

---

📝 Summary

Summary by CodeRabbit

• Bug Fixes
  • Updated CUDA stream handling across GPU-accelerated operations, including joins, sorting, aggregation, text processing, string operations, and file I/O.
  • Improved compatibility with current CUDA stream interfaces across C++, Python, and Java integrations.
  • Preserved existing public APIs, validation, processing behavior, and results while improving stream interoperability.
  • Maintained consistent stream handling for data conversion, table operations, partitioning, and Parquet workflows.

Walkthrough

Changes

The pull request updates CUDA stream integration across C++, Java native code, and Python Cython bindings. It replaces deprecated headers and .value() access with current stream headers, direct views, and .get() access. Operation signatures and behavior remain unchanged.

CUDA stream migration

Layer / File(s)	Summary
Native stream contracts cpp/include/cudf/..., cpp/src/..., cpp/tests/..., java/src/main/native/...	CUDA stream includes now use <cuda/stream>. Native code uses direct stream references where applicable.
Python stream access python/cudf_streaming/..., python/pylibcudf/...	Cython bindings replace .value() with .get() and pass stream views directly in streaming helpers.
Runtime and test call sites cpp/libcudf_streaming/..., cpp/tests/..., python/pylibcudf/...	Callbacks, storage operations, Parquet utilities, and text normalization paths use current CUDA stream accessors.

Estimated code review effort: 3 (Moderate) | ~25 minutes

Merge Risk: ⚪ Minimal · up to 1e03a

This updates CUDA stream-handle access for Unicode normalization while preserving behavior and public interfaces. No current merge-blocking risk is identified.

Possibly related PRs

• NVIDIA/cudf#23727: Migrates related CUDA stream headers, types, and accessors.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 3 functions across 15 files. (1 skipped: …	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title clearly and concisely describes the primary change: migrating stream APIs from rmm::cuda_stream_view to cuda::stream_ref.
Description check	✅ Passed	The description accurately explains the stream API migration, including the use of cuda::stream_ref, .get(), and .sync(), and it relates directly to the changeset.

Full details: Docstring Coverage

Explanation

Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 3 functions across 15 files. (1 skipped: 1 unsupported.)

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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