Repository navigation
Conversation
meta-cla
Bot
added
the
CLA Signed
Contributor
|
@psiddh has exported this pull request. If you are a Meta employee, you can view the originating Diff in D129036504. |
SS-JIA
approved these changes
Nov 13, 2026
SS-JIA
left a comment
SS-JIA
left a comment
Contributor
There was a problem hiding this comment.
Review automatically exported from Phabricator review in Meta.
This PR needs a
|
Summary: Pull Request resolved: pytorch#25026 The WebGPU partitioner delegated aten.mul.Scalar (from the Vulkan op registry), but the WebGPU runtime had no kernel, so programs with SDPA's math decomposition failed to load. This adds a one-dispatch fp32 mul.Scalar with a resize hook, and the partitioner keeps non-fp32 mul.Scalar on CPU. Use case: unmodified Hugging Face ESM-2 and BERT-style models (SDPA without a fused kernel), including dynamic sequence length. Reviewed By: SS-JIA Differential Revision: D129036504
meta-codesync
Bot
changed the title
psiddh
force-pushed
the
export-D129036504
branch
from
November 13, 2026 06:35
620a852 to
65da4d5
Compare
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary:
The WebGPU partitioner delegated aten.mul.Scalar (from the Vulkan op registry), but the WebGPU runtime had no kernel, so programs with SDPA's math decomposition failed to load.
This adds a one-dispatch fp32 mul.Scalar with a resize hook, and the partitioner keeps non-fp32 mul.Scalar on CPU.
Use case: unmodified Hugging Face ESM-2 and BERT-style models (SDPA without a fused kernel), including dynamic sequence length.
Reviewed By: SS-JIA
Differential Revision: D129036504