Qualcomm AI Engine Direct - Verify Direct Build in External CI

[winskuo-quic]

Summary

QNN Backend supports direct build, which is to build library with hexagon tool chain. Since it is using its own tool chain, some of the C/C++ files or functions are not accessible. For example, in this PR, extension/data_loader/mman.h uses some MACRO that is not no Hexagon toolchain. Due to this reason, mainline often breaks direct-build when someone included functions that's not supported by Hexagon tool chain. To prevent this to happen, this PR added:

1. direct-build to CI test to ensure changes doesn't break direct build
2. Ensure the direct build artifact size is smaller than 200kb

Test plan

Passing test-qnn-direct-build-linux under pull.yml

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/19763

• 📄 Preview Python docs built from this PR

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

❌ 1 Cancelled Job, 4 Unrelated Failures, 2 Unclassified Failures

As of commit c978d61 with merge base 2b9e9bf ():

UNCLASSIFIED FAILURES - DrCI could not classify the following jobs because the workflow did not run on the merge base. The failures may be pre-existing on trunk or introduced by this PR:

• Build Windows Wheels / pytorch/executorch / build-wheel-py3_10-cpu (gh) (this job did not run on the merge base, so DrCI cannot tell whether the failure is pre-existing)
  Process completed with exit code 1.
• Build Windows Wheels / pytorch/executorch / upload / upload-wheel-py3_10-cpu (gh) (this job did not run on the merge base, so DrCI cannot tell whether the failure is pre-existing)
  Unable to download artifact(s): Artifact not found for name: pytorch_executorch__3.10_cpu_x64

CANCELLED JOB - The following job was cancelled. Please retry:

• trunk / unittest-release / macos / macos-job (gh)
  ##[error]The operation was canceled.

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

• pull / test-sqnr-static-llm-qnn-linux (smollm2_135m) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
• pull / unittest-editable / windows / windows-job (gh) (matched win rule in flaky-rules.json)
  ##[error]The operation was canceled.
• Test XNNPACK Backend / test-xnnpack / test-backend-linux (xnnpack, models) / linux-job (gh) (matched linux rule in flaky-rules.json)
  The process '/usr/bin/git' failed with exit code 128

BROKEN TRUNK - The following job failed but were present on the merge base:

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

• pull / test-openvino-linux / linux-job (gh) (trunk failure)
  curl: (22) The requested URL returned error:

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

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: winskuo-quic / name: Winston Kuo (55ef49d, 9c838bb)

[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.

[psiddh]

Should we add --skip_x86_64 --skip_linux_android to this? iiuc, build.sh still builds the x86 and full Android stacks (those default on), so a job meant to verify just the direct build does a lot of extra work. Not sure..

[winskuo-quic]

Hi @psiddh,
I think both are fine. I was doing a full build just to double check that direct_runner for build-android can also be properly build. Since the goal here is to verify .so is built successfully, I have skipped x86 and android in newest change.
Thanks

[psiddh]

nits: Should we add a SHA-256 check on the downloaded Hexagon SDK zip? as a follow up PR maybe ?

[winskuo-quic]

I have added this check in newest commit. Thanks for the suggestion on safety check.

[psiddh]

I think we should we gate this job on QNN-related paths (e.g. backends/qualcomm/**, CMakeLists.txt, the build scripts) instead of running it on every PR?

[winskuo-quic]

That is the intial idea. But the general layer can also introduce changes that does not work with Hexagon toolchain. For this reason, I have set it to run on every PR.