Add support for torch.pow in the Arm backend #9309
Merged
Conversation
This file contains hidden or 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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/9309
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 1 PendingAs of commit 5379416 with merge base feb3fcd ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
@pytorchbot label "topic: not user facing" |
|
@pytorchbot label "partner: arm" |
pytorch-bot
bot
added
the
partner: arm
|
@pytorchbot label ciflow/trunk |
zingo
force-pushed
the
main
branch
2 times, most recently
from
6b8a46f to
446fe7f
Compare
|
@digantdesai I have updated this PR so you might want to check it again. |
|
Rebase? |
|
@digantdesai I'm on it. Had some test failures after rebasing locally but the patch is on it's way. |
Implement support for torch.pow in the MI and BI profile of TOSA. For MI, the operator works as Pytorch's reference implementation except for that the base operand cannot be a scalar but must be a tensor. For BI, the exponent operand must be a scalar and a constant value. The base operand must be a tensor. Split the ReplaceScalarWithTensorArgsPass into two subclasses: one for MI and one for BI. For MI, the pow operator's exponent will converted to a tensor in case it is a scalar. For BI, the scalar will be kept, but instead it will be consumed in the InsertTableOpsPass, meaning that the operator will be converted into a table operation with one input and output. This still enforces the exponent to be constant for the BI profile. Change-Id: I464ab91ff46c0a6ad28d0fb84735a403a74e6323
kirklandsign
pushed a commit
that referenced
this pull request
Apr 11, 2025
Co-authored-by: Martin Lindström <[email protected]>
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.
cc @digantdesai @freddan80 @per @zingo @oscarandersson8218