Fix GroupQueryAttentionFusion crash on ONNX-domain RotaryEmbedding#29707
Open
namgyu-youn wants to merge 1 commit into
Open
Fix GroupQueryAttentionFusion crash on ONNX-domain RotaryEmbedding#29707namgyu-youn wants to merge 1 commit into
namgyu-youn wants to merge 1 commit into
Conversation
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
Azure Pipelines: There may be pipelines that require an authorized user to comment /azp run to run. |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
GroupQueryAttentionFusionmatches GQA input producers byOpType() == "RotaryEmbedding"without a domain check, and readsInputDefs()[2]/[3]assuming the contrib signature(input, position_ids, cos_cache, sin_cache). The ONNX-domain opset-23 op is(X, cos_cache, sin_cache, position_ids?), so a spec-valid 3-input node reads out of bounds — aborting session init on builds with_GLIBCXX_ASSERTIONS(e.g. nightly wheels), silent UB otherwise.Fix: require
Domain() == kMSDomain, so the fusion skips ONNX-domain graphs instead of misreading them. Adds a regression test that aborts without the fix.Motivation and Context
Fixes #28604 (repro and analysis: #28604 (comment)). #29524/#29552 concern a different defect in this function and are not changed here.
Developed with AI assistance (Claude Code); reviewed by the author.