[CYTHON] Fix stream passing bug

[tqchen]

This PR fixes a bug in stream passing which breaks expected stream passing behavior. Also added a regression case via load_inline_cuda to guard this issue (needs CUDA env to run atm).

[gemini-code-assist]

Summary of Changes

Hello @tqchen, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses and resolves a critical bug in the Cython-based Foreign Function Interface (FFI) that affected the proper handling and passing of CUDA streams. The fix ensures that device contexts are correctly identified, preventing issues with stream management. To safeguard against future regressions, comprehensive test cases have been introduced for CUDA stream passing, validating the behavior with both default and explicitly managed PyTorch streams. Additionally, the package version has been updated to reflect these changes.

Highlights

• Stream Passing Bug Fix: Corrected a bug in the Cython FFI where the device context check for stream passing was incorrectly using device_id instead of device_type, leading to unexpected stream behavior.
• Regression Test for CUDA Stream Passing: Added a new regression test case using load_inline_cuda to validate correct CUDA stream passing. This test now explicitly passes a raw_stream argument and verifies that the stream used within the CUDA kernel matches the provided stream.
• Expanded Test Coverage: The new regression test includes scenarios for both default (raw stream 0) and explicit PyTorch CUDA streams, ensuring robust validation of the stream passing mechanism.
• Version Update: The package version has been incremented from 0.1.0b10 to 0.1.0b11 in pyproject.toml and python/tvm_ffi/__init__.py.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request addresses a bug in stream passing within the Cython FFI layer by switching from checking device_id to the more robust device_type to determine if a device context has been established. This is a good improvement for correctness. The PR also introduces a valuable regression test for CUDA stream handling, which validates behavior for both default and user-created streams. Additionally, the package version is updated. My review includes one suggestion to remove a redundant check in the new test code to improve its clarity.