[tmva][sofie] Reset operator-fusion state for each parsed graph - #23230
Merged
guitargeek merged 1 commit intoSep 3, 2026
Conversation
RModelParser_ONNX::fFusedOperators records operator-fusion decisions (e.g. Gemm+Relu) keyed by node index, but was never cleared. Reusing one parser instance for a second model replayed the stale fusion decisions of the first model against unrelated nodes, leading to wrong models or crashes (e.g. a segfault in ParseFuseGemmRelu when the stale index points to a zero-input Constant node). The map is also per-graph state: ParseONNXGraph is called recursively for the subgraphs of the If operator, whose node indices would collide with those of the enclosing graph. 🤖 Done with the help of AI
Test Results 22 files 22 suites 3d 12h 5m 29s ⏱️ Results for commit 4a6845b. |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
RModelParser_ONNX::fFusedOperators records operator-fusion decisions (e.g. Gemm+Relu) keyed by node index, but was never cleared. Reusing one parser instance for a second model replayed the stale fusion decisions of the first model against unrelated nodes, leading to wrong models or crashes (e.g. a segfault in ParseFuseGemmRelu when the stale index points to a zero-input Constant node).
The map is also per-graph state: ParseONNXGraph is called recursively for the subgraphs of the If operator, whose node indices would collide with those of the enclosing graph.
🤖 Done with the help of AI