Arm backend: Add Zephyr Cortex-M STM Nucleo n657x0_q example

[zingo]

Summary

Add information in the zephyr readme on how to build and run
a model on STM Nucleo n657x0_q board.

Moves the Ethos-U driver config to Corstone FVPs board files.
This enable smoother building different targets that will enable
or not enable the Ethos-U drivers/backend depending on board.

This makes it possible to build the example both for Ethos-U
and Cortex-M only targets without change.

cc @freddan80 @per @oscarandersson8218 @digantdesai

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/17371

• 📄 Preview Python docs built from this PR

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

❌ 9 New Failures, 2 Pending, 14 Unrelated Failures

As of commit 494784d with merge base 05da1af ():

NEW FAILURES - The following jobs have failed:

• pull / test-samsung-models-linux / linux-job (gh)
  RuntimeError: Command docker exec -t 270f5d70c56677d0b6445ee571a822f38955361a9589337ded3ac98ccb9284fd /exec failed with exit code 1
• pull / test-samsung-quantmodels-linux / linux-job (gh)
  RuntimeError: Command docker exec -t b514c5adc2e39c24d8c57c16ba8b50a54f60802cee4bbcfe23355b1c45b86c35 /exec failed with exit code 1
• trunk / test-qnn-model (fp32, mb) / linux-job (gh)
  RuntimeError: Command docker exec -t cacbdbfc701870bc9ebe52230831cd70c60432b3e1c0f57a715edf0dad45c25b /exec failed with exit code 137
• trunk / test-qnn-model (fp32, w2l) / linux-job (gh)
  RuntimeError: Command docker exec -t 23620f8a7c9d3d57cd1706d7cfc7f8a3c5d61606334416dd8d5c3693bf167153 /exec failed with exit code 137
• trunk / test-qnn-optimum-model (fp32, distilbert) / linux-job (gh)
  RuntimeError: Command docker exec -t b898a2dc4b55ee5e16d2330628fdd47353b73886b77a48ce6866625ee44c398d /exec failed with exit code 137
• trunk / test-qnn-optimum-model (fp32, dit) / linux-job (gh)
  RuntimeError: Command docker exec -t e80a1cacc73d270b5969e879335356c61a14df20550fe2e8b398348e629154b7 /exec failed with exit code 137
• trunk / test-qnn-optimum-model (fp32, efficientnet) / linux-job (gh)
  RuntimeError: Command docker exec -t 09d0528b00943eb3dad4404a086a2c3188e9a71558d3554279aa813bb1f91baa /exec failed with exit code 137
• trunk / test-qnn-optimum-model (fp32, pvt) / linux-job (gh)
  RuntimeError: Command docker exec -t 782c707ea8b2a5d0b6f591a2bf5e55e89e3e802b1533e2711b92d6781de313a8 /exec failed with exit code 137
• trunk / test-qnn-optimum-model (fp32, roberta) / linux-job (gh)
  RuntimeError: Command docker exec -t ccde52aebb816e9fdda80d8276223203cd33c5caf008adebbe8819a9c477067d /exec failed with exit code 137

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

• trunk / test-qnn-optimum-model (fp32, albert) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
• trunk / test-qnn-optimum-model (fp32, swin) / 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.

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

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

• pull / test-llama-runner-qnn-linux (fp32, qnn_16a16w, qnn) / linux-job (gh) (trunk failure)
• trunk / test-llama-runner-qnn-linux (fp32, qnn_16a16w, qnn) / linux-job (gh) (trunk failure)
• trunk / test-qnn-model (fp32, conv_former) / linux-job (gh) (trunk failure)
• trunk / test-qnn-model (fp32, ic3) / linux-job (gh) (trunk failure)
• trunk / test-qnn-model (fp32, mv2) / linux-job (gh) (trunk failure)
• trunk / test-qnn-model (fp32, mv3) / linux-job (gh) (trunk failure)
• trunk / test-qnn-model (fp32, vit) / linux-job (gh) (trunk failure)
• trunk / test-qnn-optimum-model (fp32, bert) / linux-job (gh) (trunk failure)
• trunk / test-qnn-optimum-model (fp32, cvt) / linux-job (gh) (trunk failure)
• trunk / test-qnn-optimum-model (fp32, focalnet) / linux-job (gh) (trunk failure)
• trunk / test-qnn-optimum-model (fp32, mobilevit_v1) / linux-job (gh) (trunk failure)
• trunk / test-qnn-optimum-model (fp32, mobilevit_v2) / linux-job (gh) (trunk failure)

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

[Copilot]

Pull request overview

This PR adds support for building and running ExecuTorch models on the STM Nucleo n657x0_q board, making the example compatible with both Ethos-U and Cortex-M only targets. The changes refactor the Ethos-U driver configuration from the main project configuration into board-specific files, allowing seamless building across different target platforms.

Changes:

• Added documentation for building and running models on STM Nucleo n657x0_q board
• Moved Ethos-U driver configuration to board-specific files for Corstone FVPs
• Fixed type casting issues for size_t and Error types to ensure proper printf formatting

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
zephyr/README.md	Added comprehensive instructions for STM Nucleo n657x0_q board setup and usage
zephyr/CMakeLists.txt	Added CMSIS-NN local path detection and disabled AOT quantized kernels to fix RTTI build issues
examples/arm/zephyr/src/arm_executor_runner.cpp	Added explicit type casts for size_t and Error types to fix printf format warnings
examples/arm/zephyr/prj.conf	Moved Ethos-U configuration to board-specific files and reduced default memory pool sizes
examples/arm/zephyr/boards/mps4_corstone320_fvp.conf	Created board-specific configuration enabling Ethos-U for Corstone 320
examples/arm/zephyr/boards/mps3_corstone300_fvp.conf	Created board-specific configuration enabling Ethos-U for Corstone 300
examples/arm/zephyr/CMakeLists.txt	Removed explicit CONF_FILE setting to allow board-specific configurations

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Corrected spelling of 'consol' to 'console'.

Suggested change

	This will run the simple add model on your hardware one and print the output on the serial consol.
	This will run the simple add model on your hardware one and print the output on the serial console.

[Copilot]

The memory pool sizes have been dramatically reduced from 262144 and 1572864 bytes to 2000 bytes each. The comment states these should 'scale with the compiled model/network', but 2000 bytes seems insufficient for most models. Consider documenting that these are minimal defaults that should be adjusted per model, or provide guidance on sizing these values.

Suggested change

	# Setup memory requirements to scale with the compiled model/network.
	# Setup memory requirements to scale with the compiled model/network.
	# NOTE: The values below are minimal defaults chosen for this example only.
	# Most non-trivial models will require significantly larger pools. You should
	# increase these to match the memory requirements of your compiled model/network
	# (for example, based on profiling or tool-reported memory usage), and leave
	# additional headroom to avoid allocation failures at runtime.