Fix invalid lowerings for ROCm in Pallas #26368
Merged
+8
−2
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
popcount and clz were effectively broken on ROCm, since math_dialect fallbacks were resulting in incorrect lowerings during compilation in XLA. Use the device intrinsics for these functions, as well as for exp and absf, which fixes some accuracy issues in the pallas tests. Docs for OCML/OCKL - https://github.com/ROCm/llvm-project/blob/amd-staging/amd/device-libs/doc/OCML.md - https://github.com/ROCm/llvm-project/blob/amd-staging/amd/device-libs/doc/OCKL.md Co-Authored-By: <[email protected]>
superbobry
approved these changes
Feb 7, 2025
google-ml-butler
bot
added
kokoro:force-run
pull ready
|
Thanks you! |
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Outdated suggestions cannot be applied.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.
popcount and clz were effectively broken on ROCm,
since math_dialect fallbacks were resulting in
incorrect lowerings during compilation in XLA.
Use the device intrinsics for these functions, as
well as for exp and absf, which fixes some accuracy issues in the pallas tests.
Docs for OCML/OCKL