Repository navigation
Java: release provider option values with the correct strings - #31828
Scott McKay (skottmckay) merged 6 commits into
Conversation
|
Azure Pipelines: There may be pipelines that require an authorized user to comment /azp run to run. |
|
Azure Pipelines: There may be pipelines that require an authorized user to comment /azp run to run. |
|
@microsoft-github-policy-service agree |
|
Yep, that's a bug, this fix is fine. |
822111e
|
Can we keep CI changes out of this simple PR, especially ones that add new components? |
|
I removed the unrelated Chromium-install workflow change and pushed commit e17acd8, leaving the provider-option string ownership fix focused on the two native binding files. The existing maintainer approval was later dismissed after the CI changes, so this new head still needs a fresh maintainer review. Please run the authorized native Java/provider checks and confirm the CUDA/TensorRT value release loops against the jValueArr objects. |
|
There are still CI changes here that shouldn't be. |
|
Follow-up: I also removed the remaining unrelated CUDA CI thread-count change and pushed commit dd9c386. The PR now changes only these two files: java/src/main/native/ai_onnxruntime_providers_OrtCUDAProviderOptions.c and java/src/main/native/ai_onnxruntime_providers_OrtTensorRTProviderOptions.c. The live diff is now focused solely on fetching provider option values from jValueArr before releasing their UTF-8 buffers. It still needs fresh maintainer review and authorized native Java/provider CI. |
|
/azp run |
|
Commenter does not have sufficient privileges for PR 31828 in repo microsoft/onnxruntime |
|
Validation update: the live diff is clean against upstream/main (2 files, 2 insertions, 2 deletions; git diff --check passes). I triggered /azp run, but Azure replied that this contributor account lacks sufficient privileges for PR 31828. The GitHub workflows for this head are also paused as action_required. A maintainer needs to approve/rerun the protected CI and refresh the review on dd9c386; no further code changes are needed from my side. |
|
Scott McKay (@skottmckay) Thanks for updating the branch with current main. All checks on the current head are now green, and the PR diff remains limited to the two intended JNI provider-option string ownership corrections. When convenient, could you please provide the final review/approval? |
Summary
jstringobjects they came fromFixes #31706
Testing
git diff --checkAI assistance disclosure
I identified and reproduced the issue, then used OpenAI Codex to assist with the implementation and regression-test work. I personally reviewed the resulting diff and ran the validation commands listed above to verify the fix. Any full-suite or local-environment limitations are documented in the validation section.