[Dev] Fix GEMV Dynamic Scheduling with Splitk #52
Conversation
This file contains 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Outdated suggestions cannot be applied.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 pull request primarily involves changes to the
bitblasPython package, specifically to thegemv.py,gemv_dequantize.py,general.py, and__init__.pyfiles. The changes include version updates, modifications to function conditions to handle additional cases, and adjustments to test parameters.Version updates:
VERSIONandpython/bitblas/__init__.py: The version number has been updated from0.0.1.dev8to0.0.1.dev9. [1] [2]Function condition modifications:
python/bitblas/gpu/gemv.py: The condition in theapply_configfunction has been modified to accept a length of 4 forblock_info.iters, along with an explanatory comment.python/bitblas/gpu/gemv_dequantize.py: Theget_vectorize_factorfunction has been modified to handle cases where the length ofsch.get_loops(block_b)is 4, including binding operations and an assertion for theSplitKcase. [1] [2]Function call modifications:
python/bitblas/wrapper/general.py: Thelegalize_cfunction has been modified to include a conditional return statement before the kernel launch string. [1] [2]Test parameter modifications:
testing/python/operators/test_general_matmul_splitk_ops.py: Thetest_matmul_torch_forward_fp8e4m3function has been modified to use a list[1, 16]as the value ofMinmatmul_config.