Move QNN SDK into backends/qualcomm/CMakeLists.txt

[larryliu0820]

Summary

[PLEASE REMOVE] See CONTRIBUTING.md's Pull Requests for ExecuTorch PR guidelines.

[PLEASE REMOVE] If this PR closes an issue, please add a Fixes #<issue-id> line.

[PLEASE REMOVE] If this PR introduces a fix or feature that should be the upcoming release notes, please add a "Release notes: " label. For a list of available release notes labels, check out CONTRIBUTING.md's Pull Requests.

Test plan

[PLEASE REMOVE] How did you test this PR? Please write down any manual commands you used and note down tests that you have written if applicable.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15546

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 8 New Failures, 3 Unrelated Failures

As of commit 1acb151 with merge base 3826f44 ():

NEW FAILURES - The following jobs have failed:

• pull / unittest / linux / linux-job (gh)
  exir/backend/test/test_lowered_backend_module.py::TestBackendAPI::test_emit_nested_lowered_backend_module
• pull / unittest-editable / linux / linux-job (gh)
  exir/backend/test/test_lowered_backend_module.py::TestBackendAPI::test_emit_nested_lowered_backend_module
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-large-v3-turbo, non-quantized) / linux-job (gh)
  RuntimeError: Command docker exec -t 99ed4fc7a2bbc096752a7972921d2a87419d5e3b034e5f8e3a7bc8b85aa64cc0 /exec failed with exit code 1
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-large-v3-turbo, quantized-int4-tile-packed) / linux-job (gh)
  RuntimeError: Command docker exec -t 656453a3c4bf74180979fa94d4a6ee34b88a394af20add9c04f5edd0b4a05388 /exec failed with exit code 1
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-large-v3-turbo, quantized-int4-weight-only) / linux-job (gh)
  RuntimeError: Command docker exec -t 4fdfecfcf3ebb57fa7bb7dceca1927ab892a3e37af586fe42ba58837b4cfb8f0 /exec failed with exit code 1
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-small, non-quantized) / linux-job (gh)
  RuntimeError: Command docker exec -t 375dcfc936159cb7e2e610fd1446546a7805cb196c3f434d67f4393c6826395f /exec failed with exit code 1
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-small, quantized-int4-tile-packed) / linux-job (gh)
  RuntimeError: Command docker exec -t 0b395b9795d2549be440f3f9a3a7d228cc25b03691749ab56f9beac0b2c31063 /exec failed with exit code 1
• Test CUDA Builds / test-model-cuda-e2e (openai, whisper-small, quantized-int4-weight-only) / linux-job (gh)
  RuntimeError: Command docker exec -t 9ec00eedb26e8add8d68a6ead018d543167347657a7994a658396ceedf64aad2 /exec failed with exit code 1

FLAKY - The following job failed but was likely due to flakiness present on trunk:

• pull / test-models-linux (resnet18, portable, linux.2xlarge) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)

BROKEN TRUNK - The following jobs failed but was present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• Test Metal Backend / test-model-metal-e2e (openai, whisper-large-v3-turbo, non-quantized) / macos-job (gh) (trunk failure)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 1
• Test Metal Backend / test-model-metal-e2e (openai, whisper-small, non-quantized) / macos-job (gh) (trunk failure)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 1

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[github-actions]

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[cccclai]

Thanks! I added the binary ci job to make sure the wheel package publishing path is still working

[cccclai]

Hi @haowhsu-quic, can you review the changes inside backends/qualcomm?

[shewu-quic]

I’m not sure I understand the reason for this condition, as both Android and x86 builds require the QNN SDK.

[larryliu0820]

This is for pip wheels only. Like Chen said, maybe I should condition on that

[larryliu0820]

@shewu-quic can you give some advice on how to move forward? The purpose of this PR is quite simple: to move the qnn build logic from setup.py to backends/qualcomm/CMakeLists.txt. Notice that the logic in setup.py is unique to wheel building and I'm trying to wrap my head on why building the runtime doesn't require downloading the sdk.

[shewu-quic]

Based on my understanding, you want to include libqnn_executorch_backend.so and the pybind files in the wheel build so users can install via pip without needing to set QNN_SDK_ROOT or download qnn sdk.

We could add a step to download the QNN SDK if QNN_SDK_ROOT is not set, similar to your approach, and use "if is_linux_x86():" in setup.py to avoid installation on unsupported systems for now. Does this approach work for you?

[larryliu0820]

Thank you for the reply, I think that should work. Let me sync with @cccclai to have more understanding.

[larryliu0820]

I think it makes sense to gate this change by EXECUTORCH_BUILD_PYBIND. After talking to @cccclai I think we agree that this code path should not be used by cases other than building ET wheel.

[cccclai]

I think the logic below is for wheel only, so probably better to have a separate env variable to control

[shewu-quic]

LGTM. Thanks!

[cccclai]

Thank you for making the changes