-
Notifications
You must be signed in to change notification settings - Fork 3.7k
[onnxruntime/build] Add CI testing for ORT build with generic interface #23530
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You can commit the suggested changes from lintrunner.
b6e951e to
64c3a26
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You can commit the suggested changes from lintrunner.
fbc3229 to
d53cf44
Compare
|
Hi @karim-vad, I'm not fully understanding what this PR is testing. Is it essentially running CI/tests with the combinations:
|
@adrianlizarraga this PR is forcing all the onnxruntime.dll builds with enable_generic_interface flags when the existing CI tests uses one of the primary EPs ( OpenVino, TensorRT, QNN) . The successful execution of the existing CI test validates the Generic interface for ORT is build correctly. In addition to that, cmake system is incorrectly settings certain pre-processor macros when generic interface is enabled. so this PR fixes those as well. The next steps is to figure out how we want to enable CI testing with generic interface flag of ORT ?
|
I see. If these cmake changes should be merged into main, they should probably be extracted and put into a separate PR. (since we are not on merging this experimental PR). This would allow us to get reviews from others who maintain build configs.
From this comment, @snnn indicated that testing all EP combos would not be acceptable due to doubling of resources. Perhaps we can limit this to QNN, TensorRT, OpenVINO. We can start with a PR that only enables this with QNN (as an example). We can use the yaml "strategy matrix" to run tests with onnxruntime/tools/ci_build/github/azure-pipelines/win-qnn-arm64-ci-pipeline.yml Lines 49 to 54 in 989c96c
strategy:
matrix:
SHARED_LIB:
QnnLibKind: 'shared_lib'
qnn_build_args: '--use_qnn'
STATIC_LIB:
QnnLibKind: 'static_lib'
qnn_build_args: '--use_qnn'
SHARED_LIB_GENERIC_INTERFACE:
QnnLibKind: 'shared_lib'
qnn_build_args: '--use_qnn --use_generic_interface'Maybe something like the above to get us going. |
Summary: - Remove unused cmake variables - Add target specific logic when generic interface is used. - Add QNN EP test case that use ORT generic interface build
d53cf44 to
ee5cf8b
Compare
|
/azp run Linux GPU CI Pipeline |
|
Azure Pipelines successfully started running 1 pipeline(s). |
…ce (#23530) [onnxruntime/build] Add CI testing for ORT build with generic interface Summary: - Remove unused cmake variables - Add target specific logic when generic interface is used. - Add QNN EP test case that use ORT generic interface build
…ce (#23530) [onnxruntime/build] Add CI testing for ORT build with generic interface Summary: - Remove unused cmake variables - Add target specific logic when generic interface is used. - Add QNN EP test case that use ORT generic interface build
[onnxruntime/build] Add CI testing for ORT build with generic interface