Fix optimized binary op broadcast rank mismatch#21052
Conversation
The treat-as-1D path already resizes out to the broadcast target, but the last-dimension and planned broadcast paths still resize it to an operand shape. With leading dimensions of size one, that can change the immutable output rank and fail with InvalidArgument. Use the broadcast target for those paths and cover both operand orders in add and mul. Fixes pytorch#10959
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21052
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: ✅ No FailuresAs of commit c13e4a6 with merge base 5828778 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
Hi @Kilbex! Thank you for your pull request and welcome to our community. Action RequiredIn order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks! |
|
|
|
@pytorchbot label "release notes: ops & kernels" |
|
Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks! |
Summary
The optimized last-dimension and planned broadcast paths resize
outto one input's shape after the operator has already selected the broadcast target. When leading dimensions of size one make the target rank larger, that second resize tries to change the immutable rank and fails at runtime.Resize
outto the broadcast target in both paths and add regression coverage for add and mul in both argument orders.Fixes #10959
Test plan
./cmake-out/kernels/test/optimized_kernels_test --gtest_brief=1lintrunner kernels/optimized/cpu/binary_ops.cpp kernels/optimized/cpu/binary_ops.h kernels/test/op_add_test.cpp kernels/test/op_mul_test.cppcc @larryliu0820 @manuelcandales