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Outdated suggestions cannot be applied.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 several changes to the
bitblas
module, focusing on enhancing tensor core utilization, optimizing memory layout, and improving the base class for tensor layout hints. The most important changes include modifications to tensor core handling in matrix multiplication, memory layout optimizations, and extensive documentation and functionality additions to theBaseTLHint
class.Enhancements to tensor core utilization:
bitblas/ops/general_matmul/tilelang/dense/matmul_tensorcore.py
: Added a conditional to setM
to 16 if it is less than 16, ensuring better utilization of tensor cores.Memory layout optimizations:
bitblas/ops/general_matmul/tilelang/dense/matmul_tensorcore_s4.py
: Modified themake_swizzle_layout
function call forB_shared
to include theis_smooth=True
parameter, optimizing memory layout.Improvements to
BaseTLHint
class:bitblas/tl/base_hint.py
: Added detailed docstrings and several new methods (__getitem__
,__getattr__
,__iter__
,keys
) to enhance the interface and functionality of theBaseTLHint
class.