Adding more Threadblock Tiles for Mixed-input TensorOp (BF16 * S8) in cutlass_library #1132
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
manishucsd
changed the title
hwu36
reviewed
Oct 10, 2023
hwu36
reviewed
Oct 10, 2023
hwu36
reviewed
Oct 10, 2023
added 2 commits
October 11, 2023 05:46
hwu36
approved these changes
Oct 13, 2023
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Outdated suggestions cannot be applied.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 adds more ThreadblockShapes (CTA) tile shapes in
cutlass_library. More ThreadblockShapes are need to autotune and choose the most performant tile shapes for mixed-input GEMMs.cutlass_library
For the
cutlass_library, it adds more ThreadblockShapes into twoGenerateSM80*function:GenerateSM80_TensorOp_16816_mixed_input_upcast_a: For upcast on operand A (e.g. S8* BF16)GenerateSM80_TensorOp_16816_mixed_input_upcast_b: For upcast on operand B (e.g. BF16 * S8)Unit tests
More tile shapes needs different warp-configurations and this PR adds more device- and warp-level tests ensuring that those configurations are functionally solid (and remains solid).