[Bugfix] Fix the test data distribution of cumsum #432
Conversation
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
* Updated block size calculation in Gemm to account for the range of thread bounds, improving accuracy in layout inference. * Simplified layout conflict error messages in ParallelOp for better clarity, enhancing debugging experience. * Removed redundant buffer checks in ParallelOp layout inference logic, streamlining the code.
* Removed unnecessary warning log in Gemm related to WGMMA conditions, streamlining the layout inference process. * Commented out redundant checks in ParallelOp's layout inference, improving code clarity while maintaining functionality. * Enhanced error messages in ParallelOp to provide clearer context for layout conflicts, aiding in debugging efforts.
…ndling * Updated the `cumsum` function to include detailed documentation and error handling for dimension bounds. * Modified the `run_cumsum` test to utilize a random tensor supply type for profiling, enhancing test robustness. * Added annotations to the fused loop in `loop_fusion_utils.h`, ensuring proper metadata is preserved during loop fusion.
LeiWang1999
added a commit
to LeiWang1999/tilelang
that referenced
this pull request
Jul 18, 2025
* [Refactor] Adjust layout inference calculations in Gemm and ParallelOp * Updated block size calculation in Gemm to account for the range of thread bounds, improving accuracy in layout inference. * Simplified layout conflict error messages in ParallelOp for better clarity, enhancing debugging experience. * Removed redundant buffer checks in ParallelOp layout inference logic, streamlining the code. * [Refactor] Clean up layout inference logic in Gemm and ParallelOp * Removed unnecessary warning log in Gemm related to WGMMA conditions, streamlining the layout inference process. * Commented out redundant checks in ParallelOp's layout inference, improving code clarity while maintaining functionality. * Enhanced error messages in ParallelOp to provide clearer context for layout conflicts, aiding in debugging efforts. * lint fix * [Enhancement] Improve cumulative sum functionality and annotations handling * Updated the `cumsum` function to include detailed documentation and error handling for dimension bounds. * Modified the `run_cumsum` test to utilize a random tensor supply type for profiling, enhancing test robustness. * Added annotations to the fused loop in `loop_fusion_utils.h`, ensuring proper metadata is preserved during loop fusion. * lint fix
LeiWang1999
added a commit
to LeiWang1999/tilelang
that referenced
this pull request
Jul 20, 2025
* [Refactor] Adjust layout inference calculations in Gemm and ParallelOp * Updated block size calculation in Gemm to account for the range of thread bounds, improving accuracy in layout inference. * Simplified layout conflict error messages in ParallelOp for better clarity, enhancing debugging experience. * Removed redundant buffer checks in ParallelOp layout inference logic, streamlining the code. * [Refactor] Clean up layout inference logic in Gemm and ParallelOp * Removed unnecessary warning log in Gemm related to WGMMA conditions, streamlining the layout inference process. * Commented out redundant checks in ParallelOp's layout inference, improving code clarity while maintaining functionality. * Enhanced error messages in ParallelOp to provide clearer context for layout conflicts, aiding in debugging efforts. * lint fix * [Enhancement] Improve cumulative sum functionality and annotations handling * Updated the `cumsum` function to include detailed documentation and error handling for dimension bounds. * Modified the `run_cumsum` test to utilize a random tensor supply type for profiling, enhancing test robustness. * Added annotations to the fused loop in `loop_fusion_utils.h`, ensuring proper metadata is preserved during loop fusion. * lint fix
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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 pull request includes updates to improve functionality, maintainability, and documentation across multiple files. Key changes involve enhancing loop fusion behavior, updating test utilities, refining cumulative sum logic, and adding comprehensive documentation for a key function.
Functional Improvements:
annotationsfrom the original loop are now preserved when creating a fused loop inParallelLoopFuser, ensuring that metadata is carried over during fusion. (src/transform/common/loop_fusion_utils.h, src/transform/common/loop_fusion_utils.hL222-R222)test_tilelang_language_cumsum.pyhas been updated to include an additionalcumsumoperation before flipping back, ensuring correctness for reverse operations. (testing/python/language/test_tilelang_language_cumsum.py, testing/python/language/test_tilelang_language_cumsum.pyL67-R68)Testing Enhancements:
tl.TensorSupplyType.Randninstead oftl.TensorSupplyType.One, introducing randomness for more robust test coverage. (testing/python/language/test_tilelang_language_cumsum.py, testing/python/language/test_tilelang_language_cumsum.pyL56-R56)Documentation Improvements:
cumsumfunction inreduce.py, explaining its purpose, arguments, and return value, while also adding validation for thedimparameter to prevent out-of-bounds errors. (tilelang/language/reduce.py, tilelang/language/reduce.pyR129-R146)