Conversation
Summary:
Remove the stale @unittest.skip decorator on TestQNNFloatingPointOperator.test_qnn_backend_mha. The skip dates to Sep 2024 ('failed to lower in QNN 2.26'); the quantized twin has no skip, MHA lowering support has since landed, and the rework framework covers MHA. This un-skips mha across test_qnn_delegate_x86, the fb simulator, and the recipe-session simulator.
Test Plan:
ast.parse passes on the edited file. Needs CI run of test_qnn_delegate_x86 and fb:test_qnn_delegate_simulator mha tests to confirm pass.
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/22872
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit 39d67ba with merge base 13ec99a ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
psiddh
added
the
module: qnn
This PR needs a
|
meta-cla
Bot
added
the
CLA Signed
There was a problem hiding this comment.
🟢 Approval recommended
The change is focused and has no unresolved review issues.
Pull request overview
Re-enables the floating-point MHA QNN delegate test by removing its obsolete skip.
Changes:
- Removes the stale
unittest.skipdecorator. - Restores existing FP MHA test coverage.
File summaries
| File | Description |
|---|---|
backends/qualcomm/tests/test_qnn_delegate.py |
Re-enables the floating-point MHA test. |
Review details
- Files reviewed: 1/1 changed files
- Comments generated: 0
- Review effort level: Lite
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
Remove the stale @unittest.skip decorator on TestQNNFloatingPointOperator.test_qnn_backend_mha (backends/qualcomm/tests/test_qnn_delegate.py). The skip dates to Sep 2024 ('failed to lower in QNN 2.26'); the quantized twin has no skip, MHA lowering support has since landed, and the rework framework covers MHA.
This un-skips mha across test_qnn_delegate_x86, the fb simulator, and the recipe-session simulator.
Fixes SKIPPING bucket rows for ai_infra_mobile_platform (executorch/backends/qualcomm): test_qnn_backend_mha x4.
cc @cbilgin