[TLERaw] Remove Ruff Comments#338
Merged
sgjzfzzf merged 1 commit intotriton_v3.5.xfrom Feb 11, 2026
Merged
Conversation
89293c7 to
43d08db
Compare
Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn>
ccb0f82 to
eba3e68
Compare
zhzhcookie
pushed a commit
that referenced
this pull request
Feb 27, 2026
Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn>
sgjzfzzf
added a commit
that referenced
this pull request
Mar 6, 2026
Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn>
sgjzfzzf
added a commit
that referenced
this pull request
Mar 9, 2026
* [TLERaw] revert llvm::CallOp related codes Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * [TLERaw] enable topk tle in ci Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * remove flake8 comments (#338) Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn>
sunnycase
pushed a commit
that referenced
this pull request
Mar 9, 2026
* [TLERaw] revert llvm::CallOp related codes Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * [TLERaw] enable topk tle in ci Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * remove flake8 comments (#338) Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn>
zhzhcookie
pushed a commit
that referenced
this pull request
Mar 11, 2026
* Add initial tle dist impl * Add tle docs * Implement submesh distributed barrier support in TLE - Added support for submesh distributed barriers in the TLE dialect, allowing for more granular synchronization across submeshes. - Introduced new attributes for the `DistributedBarrierOp` to handle group metadata, including `group_kind`, `group_rank`, `group_shape`, `group_axes`, and `group_mask`. - Enhanced the `remote` function to validate and handle buffered tensors with appropriate metadata. - Updated the `device_mesh` class to include launch shape and dimension names for better submesh handling. - Implemented verification logic for the new attributes in `DistributedBarrierOp`. - Added tests to ensure correct behavior of the new submesh barrier functionality and validation of remote buffered tensors. * Add AxisInfoExt and RemotePointerUtils for Triton dialect enhancements - Introduced AxisInfoExt.h and AxisInfoExt.cpp to extend axis information analysis with new visitors for LocalPointersOp and RemotePointersOp. - Implemented RemotePointerUtils.h and RemotePointerUtils.cpp to handle remote pointer information extraction and inference of vector sizes/layouts. - Added utility functions for managing remote pointer metadata and determining vectorization hints based on tensor properties. - Enhanced axis analysis capabilities to support new Triton operations, improving optimization opportunities in the Triton dialect. * Fix TLE local pointer encoding and retune fused launch params * tle: add grid distributed barrier + optimize moe atomic fused * remove tle docs from repo and ignore them * ignore tle docs paths * remove unintended triton_shared gitlink * revert unnecessary cluster wait fence insertion * refine markers and relocate tle alignment test * Apply code-format changes * restore core alignment test and split tle coverage * update tle moe tuning and backend/lowering adjustments * Apply code-format changes * Fix * [TLERaw] Revert `LLVM::CallOp`-Related Modification (#387) * [TLERaw] revert llvm::CallOp related codes Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * [TLERaw] enable topk tle in ci Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * remove flake8 comments (#338) Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * Apply code-format changes * Update .gitignore to remove specific entries Remove unnecessary files from .gitignore * Clean up .gitignore by removing unused entries Removed entries for Test and Agent from .gitignore * Enhance TLE Support in TritonNVIDIAGPUToLLVM - Added conditional compilation for TLE support in LoadStoreOpToLLVM.cpp, TargetInfo.cpp, TargetInfo.h, and TritonGPUToLLVM.cpp. - Introduced new functions and modified existing ones to handle remote pointer information and shared memory access for TLE. - Updated the TLE documentation to include examples for scalar shared-memory lookups and clarified the usage of local pointers. - Ensured compatibility with both TLE and non-TLE builds by using preprocessor directives. * Apply code-format changes * Fix * Refactor TLE-related code in TritonNVIDIAGPUToLLVM - Removed unnecessary comments marking the beginning and end of TLE sections. - Consolidated TLE-specific code under preprocessor directives to improve readability. - Streamlined the inclusion of TLE headers and related logic in LoadStoreOpToLLVM, TargetInfo, and TritonGPUToLLVM files. - Ensured consistent handling of shared and cluster shared pointers across various operations. - Enhanced the clarity of cache policy handling in load and store operations. * Apply code-format changes --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> Co-authored-by: flagtree-bot <flagtree_ai@163.com> Co-authored-by: Jinjie Liu <jjliu@baai.ac.cn>
zhzhcookie
pushed a commit
that referenced
this pull request
Mar 13, 2026
Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn>
zhzhcookie
pushed a commit
that referenced
this pull request
Mar 13, 2026
* [TLERaw] revert llvm::CallOp related codes Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * [TLERaw] enable topk tle in ci Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * remove flake8 comments (#338) Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn>
zhzhcookie
pushed a commit
that referenced
this pull request
Mar 13, 2026
* Add initial tle dist impl * Add tle docs * Implement submesh distributed barrier support in TLE - Added support for submesh distributed barriers in the TLE dialect, allowing for more granular synchronization across submeshes. - Introduced new attributes for the `DistributedBarrierOp` to handle group metadata, including `group_kind`, `group_rank`, `group_shape`, `group_axes`, and `group_mask`. - Enhanced the `remote` function to validate and handle buffered tensors with appropriate metadata. - Updated the `device_mesh` class to include launch shape and dimension names for better submesh handling. - Implemented verification logic for the new attributes in `DistributedBarrierOp`. - Added tests to ensure correct behavior of the new submesh barrier functionality and validation of remote buffered tensors. * Add AxisInfoExt and RemotePointerUtils for Triton dialect enhancements - Introduced AxisInfoExt.h and AxisInfoExt.cpp to extend axis information analysis with new visitors for LocalPointersOp and RemotePointersOp. - Implemented RemotePointerUtils.h and RemotePointerUtils.cpp to handle remote pointer information extraction and inference of vector sizes/layouts. - Added utility functions for managing remote pointer metadata and determining vectorization hints based on tensor properties. - Enhanced axis analysis capabilities to support new Triton operations, improving optimization opportunities in the Triton dialect. * Fix TLE local pointer encoding and retune fused launch params * tle: add grid distributed barrier + optimize moe atomic fused * remove tle docs from repo and ignore them * ignore tle docs paths * remove unintended triton_shared gitlink * revert unnecessary cluster wait fence insertion * refine markers and relocate tle alignment test * Apply code-format changes * restore core alignment test and split tle coverage * update tle moe tuning and backend/lowering adjustments * Apply code-format changes * Fix * [TLERaw] Revert `LLVM::CallOp`-Related Modification (#387) * [TLERaw] revert llvm::CallOp related codes Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * [TLERaw] enable topk tle in ci Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * remove flake8 comments (#338) Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * Apply code-format changes * Update .gitignore to remove specific entries Remove unnecessary files from .gitignore * Clean up .gitignore by removing unused entries Removed entries for Test and Agent from .gitignore * Enhance TLE Support in TritonNVIDIAGPUToLLVM - Added conditional compilation for TLE support in LoadStoreOpToLLVM.cpp, TargetInfo.cpp, TargetInfo.h, and TritonGPUToLLVM.cpp. - Introduced new functions and modified existing ones to handle remote pointer information and shared memory access for TLE. - Updated the TLE documentation to include examples for scalar shared-memory lookups and clarified the usage of local pointers. - Ensured compatibility with both TLE and non-TLE builds by using preprocessor directives. * Apply code-format changes * Fix * Refactor TLE-related code in TritonNVIDIAGPUToLLVM - Removed unnecessary comments marking the beginning and end of TLE sections. - Consolidated TLE-specific code under preprocessor directives to improve readability. - Streamlined the inclusion of TLE headers and related logic in LoadStoreOpToLLVM, TargetInfo, and TritonGPUToLLVM files. - Ensured consistent handling of shared and cluster shared pointers across various operations. - Enhanced the clarity of cache policy handling in load and store operations. * Apply code-format changes --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> Co-authored-by: flagtree-bot <flagtree_ai@163.com> Co-authored-by: Jinjie Liu <jjliu@baai.ac.cn>
sunnycase
pushed a commit
that referenced
this pull request
Mar 17, 2026
* [TLERaw] revert llvm::CallOp related codes Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * [TLERaw] enable topk tle in ci Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * remove flake8 comments (#338) Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn>
sunnycase
added a commit
that referenced
this pull request
Mar 17, 2026
* Add initial tle dist impl * Add tle docs * Implement submesh distributed barrier support in TLE - Added support for submesh distributed barriers in the TLE dialect, allowing for more granular synchronization across submeshes. - Introduced new attributes for the `DistributedBarrierOp` to handle group metadata, including `group_kind`, `group_rank`, `group_shape`, `group_axes`, and `group_mask`. - Enhanced the `remote` function to validate and handle buffered tensors with appropriate metadata. - Updated the `device_mesh` class to include launch shape and dimension names for better submesh handling. - Implemented verification logic for the new attributes in `DistributedBarrierOp`. - Added tests to ensure correct behavior of the new submesh barrier functionality and validation of remote buffered tensors. * Add AxisInfoExt and RemotePointerUtils for Triton dialect enhancements - Introduced AxisInfoExt.h and AxisInfoExt.cpp to extend axis information analysis with new visitors for LocalPointersOp and RemotePointersOp. - Implemented RemotePointerUtils.h and RemotePointerUtils.cpp to handle remote pointer information extraction and inference of vector sizes/layouts. - Added utility functions for managing remote pointer metadata and determining vectorization hints based on tensor properties. - Enhanced axis analysis capabilities to support new Triton operations, improving optimization opportunities in the Triton dialect. * Fix TLE local pointer encoding and retune fused launch params * tle: add grid distributed barrier + optimize moe atomic fused * remove tle docs from repo and ignore them * ignore tle docs paths * remove unintended triton_shared gitlink * revert unnecessary cluster wait fence insertion * refine markers and relocate tle alignment test * Apply code-format changes * restore core alignment test and split tle coverage * update tle moe tuning and backend/lowering adjustments * Apply code-format changes * Fix * [TLERaw] Revert `LLVM::CallOp`-Related Modification (#387) * [TLERaw] revert llvm::CallOp related codes Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * [TLERaw] enable topk tle in ci Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * remove flake8 comments (#338) Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> * Apply code-format changes * Update .gitignore to remove specific entries Remove unnecessary files from .gitignore * Clean up .gitignore by removing unused entries Removed entries for Test and Agent from .gitignore * Enhance TLE Support in TritonNVIDIAGPUToLLVM - Added conditional compilation for TLE support in LoadStoreOpToLLVM.cpp, TargetInfo.cpp, TargetInfo.h, and TritonGPUToLLVM.cpp. - Introduced new functions and modified existing ones to handle remote pointer information and shared memory access for TLE. - Updated the TLE documentation to include examples for scalar shared-memory lookups and clarified the usage of local pointers. - Ensured compatibility with both TLE and non-TLE builds by using preprocessor directives. * Apply code-format changes * Fix * Refactor TLE-related code in TritonNVIDIAGPUToLLVM - Removed unnecessary comments marking the beginning and end of TLE sections. - Consolidated TLE-specific code under preprocessor directives to improve readability. - Streamlined the inclusion of TLE headers and related logic in LoadStoreOpToLLVM, TargetInfo, and TritonGPUToLLVM files. - Ensured consistent handling of shared and cluster shared pointers across various operations. - Enhanced the clarity of cache policy handling in load and store operations. * Apply code-format changes --------- Signed-off-by: Jinjie Liu <jjliu@baai.ac.cn> Co-authored-by: flagtree-bot <flagtree_ai@163.com> Co-authored-by: Jinjie Liu <jjliu@baai.ac.cn>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR is used to remove ‘noqa: F722’ in TLE-Raw development. Close #328.