[exir] Materialize alloc shapes in ToOutVarPass#19806
Merged
oscarandersson8218 merged 3 commits intoMay 28, 2026
Merged
Conversation
Fix a dynamic-shape lowering bug in exir. ConstraintBasedSymShapeEvalPass concretizes TensorSpec metadata, but ToOutVarPass was still building memory.alloc nodes from symbolic FakeTensor/tensor_meta shapes. That let symbolic dims leak into the generated ExecuTorch GraphModule and caused runtime failures when the lowered module was executed in Python. Build memory.alloc specs from concrete upper-bounded integer shapes instead. If an alloc shape is still not concretely bounded, raise a clear error. Add an EXIR regression test that exports a dynamic-shape model, runs ConstraintBasedSymShapeEvalPass + ToOutVarPass, and verifies that memory.alloc shapes are concrete integers. Signed-off-by: Oscar Andersson <oscar.andersson@arm.com> Change-Id: If9a7b4b9aad93c1d594f9f9178d33d7df944c5e6
oscarandersson8218
requested review from
JacobSzwejbka and
larryliu0820
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/19806
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ✅ You can merge normally! (3 Unrelated Failures)As of commit 2309f03 with merge base 7fd21f2 ( BROKEN TRUNK - The following jobs failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
oscarandersson8218
added
help wanted
Remove nonzero test case from xfails in test_torch_functions.py. Signed-off-by: Oscar Andersson <oscar.andersson@arm.com> Change-Id: I5768429c6e289e114c55a1f77822cc03a619b8ab
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
3 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.
Fix a dynamic-shape lowering bug in exir.
ConstraintBasedSymShapeEvalPass concretizes TensorSpec metadata, but ToOutVarPass was still building memory.alloc nodes from symbolic FakeTensor/tensor_meta shapes. That let symbolic dims leak into the generated ExecuTorch GraphModule and caused runtime failures when the lowered module was executed in Python.
Build memory.alloc specs from concrete upper-bounded integer shapes instead. If an alloc shape is still not concretely bounded, raise a clear error.
Add an EXIR regression test that exports a dynamic-shape model, runs ConstraintBasedSymShapeEvalPass + ToOutVarPass, and verifies that memory.alloc shapes are concrete integers.
cc @digantdesai @freddan80 @per @zingo @mansnils @Sebastian-Larsson @robell @rascani