[REFACTOR][S-TIR] Move remaining data structures to s_tir #18743
Conversation
Summary of ChangesHello @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 undertakes a significant refactoring effort to enhance the organization and clarity of the Schedulable TensorIR (S-TIR) components within the TVM project. By relocating core S-TIR data structures and analysis functions to a dedicated Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Changelog
Activity
Using Gemini Code AssistThe 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
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 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
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request refactors the Schedulable TensorIR (S-TIR) analysis and data layout components by moving them from the tvm.tir namespace to a dedicated tvm.s_tir namespace. Key changes include creating new header files (include/tvm/s_tir/analysis.h, include/tvm/s_tir/sblock_dependence_info.h, include/tvm/s_tir/sblock_scope.h, include/tvm/s_tir/data_layout.h, include/tvm/s_tir/utils.h) and corresponding Python modules (python/tvm/s_tir/analysis/__init__.py, python/tvm/s_tir/analysis/_ffi_api.py, python/tvm/s_tir/sblock_dependence_info.py), and updating existing files to reflect these new paths and namespaces. Specifically, functions like GetSBlockAccessRegion, GetSBlockReadWriteRegion, DetectBufferAccessLCA, and FindAnchorBlock are now part of tvm.s_tir.analysis. Additionally, Layout and BijectiveLayout objects, along with StmtSRef, SBlockScope, and SBlockDependenceInfo, have been moved to the s_tir namespace, involving renaming classes, FFI registration names, and updating include paths across various C++ and Python files. The DataTypeLegalizer test file was also removed as part of this reorganization.
tqchen
force-pushed
the
block-scope
branch
2 times, most recently
from
2969f6f to
f4071a9
Compare
This PR moves remaining related data structures to s_tir. - Moves sblock_dependency_info and sblock_scope. - Moves related analyssis. - Hides the data_type_rewriter to private functions.
tqchen
force-pushed
the
block-scope
branch
3 times, most recently
from
fc01ca2 to
5d9e9ae
Compare
2 participants
This PR moves remaining related data structures to s_tir.