[LLVMGPU] Support i8 MFMA intrinsics in GPUTileAndFuse pipeline #18104
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
Signed-off-by: Max Dawkins <max.dawkins@gmail.com>
Max191
force-pushed
the
tile-and-fuse-i8-intrinsics
branch
from
d769502
to
d074a42
Compare
qedawkins
approved these changes
Aug 5, 2024
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Outdated suggestions cannot be applied.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 simply adds
MFMA_I8_16x16x32_I32to the list of allowed enums inMMAAttr::populateOperandOffsetsSizesStrides. This enables the logic to pack and lower a contraction to this intrinsic with theGPUTileAndFuse. The layout forMFMA_I8_16x16x32_I32is very similar to the layout forMFMA_F16_16x16x16_F32, which is already supported, so the pipeline is already ready to handle the i8 intrinsic case. Numerical correctness has been verified on an MI300 card for a 256x256x256 matmul.