-
Notifications
You must be signed in to change notification settings - Fork 682
Build an AAR with QNN #13687
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
Build an AAR with QNN #13687
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/13687
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: ❌ 1 New Failure, 1 Pending, 1 Unrelated FailureAs of commit e33b65f with merge base 785d298 ( NEW FAILURE - The following job has failed:
FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
PYTHON_EXECUTABLE=python bash .ci/scripts/setup-qnn-deps.sh | ||
PYTHON_EXECUTABLE=python bash .ci/scripts/build-qnn-sdk.sh | ||
export EXECUTORCH_BUILD_QNN=ON | ||
export QNN_SDK_ROOT=/tmp/qnn/2.28.0.241029 |
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.
Can we try to reduce the hardcode version and refer to this https://github.com/pytorch/executorch/blob/main/backends/qualcomm/scripts/qnn_config.sh ? If it needs to refactor a bit that's fine
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.
Nice
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.
Oh... This is part of docker image anyways..
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.
Ah yeah I just meant this part /tmp/qnn/2.28.0.241029
, the 2.28.0.241029
can be different when we bump the qnn version
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.
Thanks Hansong. This looks like the right approach. But I was trying to validate something similar and it doesnt quite seem to work yet. One thing is that libQnnModelDlc.so is not distributed in the maven package. I am gonna ask qualcomm to publish one
source backends/qualcomm/scripts/qnn_config.sh | ||
export QNN_SDK_ROOT="/tmp/qnn/${QNN_VERSION}" | ||
export ANDROID_ABIS=arm64-v8a | ||
GRADLE_ARGS+=" -DqnnVersion=2.28.0" |
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.
cc @cccclai make sure to update this when you update the QNN version
@pytorchbot cherry-pick --onto release/0.7 -c release |
(cherry picked from commit e2a6538)
Cherry picking #13687The cherry pick PR is at #13878 Details for Dev Infra teamRaised by workflow job |
No description provided.