[alirezamonfared]

Summary:

Summary

This change improves the Samsung ENN backend integration by:

• adding E9945 to SamsungChipset and testing compile-spec serialization for every supported chipset;
• adding host-side AOT and Android runtime build targets, including static backend registration;
• importing CompileSpec from the lightweight schema module;
• hardening EnnExecutor teardown after partial or failed initialization;
• fixing portable size_t formatting and Android variadic logging.

EnnExecutor now avoids model and buffer cleanup before successful initialization or model creation, releases buffers only when an allocation exists, and still closes a valid model when no buffers were allocated. Android logging copies the va_list and uses __android_log_vprint before writing separately to stderr.

Existing E9955 and E9965 behavior is unchanged. The backend continues to load ENN dynamically from the device and does not bundle vendor binaries. This is backend infrastructure and contains no application, model, asset, device-eligibility, or rollout policy.

Authorship

Authored with Codex.

Test Plan:

Test plan

• Unit tests passed for compile-spec serialization of E9945, E9955, and E9965, including case-insensitive E9945 lookup (2 tests).
• Built the Samsung host-side AOT and Android ENN runtime/static-registration targets.
• On E9945 hardware, loaded an ExecuTorch program with one ENN delegate and observed successful model open, commit, and execution.
• Verified output matched the reference within the expected tolerance and placement logs confirmed NPU participation.
• With the ENN runtime library unavailable, verified the backend returned a clean initialization error and the process remained alive.
• Lint and format checks passed.

Reviewed By: sivapatibandla

Differential Revision: D124906831

Pulled By: alirezamonfared

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/24165

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• Planned maintenance affecting GB300 CI jobs

✅ No Failures

As of commit d117543 with merge base 79feb10 ():
💚 Looks good so far! There are no failures yet. 💚

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

[meta-codesync]

@alirezamonfared has exported this pull request. If you are a Meta employee, you can view the originating Diff in D124906831.

[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.