GPU vectorize pass #970
Merged
GPU vectorize pass #970
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
adam-smnk
force-pushed
the
gpu-vectorize-pass
branch
from
25c9339
to
ef50749
Compare
rengolin
approved these changes
Oct 3, 2024
adam-smnk
force-pushed
the
gpu-vectorize-pass
branch
from
ef50749
to
5bfe7a9
Compare
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Outdated suggestions cannot be applied.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.
Adds GPU kernel vectorization pass.
Extends CUDA lowering pass to support vector operations.
GPU-specific vectorization pass that guides upstream Linalg vectorizer to process operations within a GPU kernel or prepared for outlining. CUDA-specific pass gets extended to allow lowering of vector ops within GPU kernel.
The vectorization is for now disabled within the GPU pipeline due to lack of vector operation unrolling. When vector sizes exceed hardware supported lengths, pipeline gets stuck on GPU binary compilation step. This will be addressed by a separate transformation pass in the future.