Refactor CMake build for examples/models/llama

[GregoryComer]

Summary

Update the CMake build for examples/models/llama to be in-tree in the top-level build but excluded from all. This means that it can be built in a single step, but will only be built when explicitly requested with --target llama_main.

This simplifies the build quite a bit (as evidenced by how many lines we remove in this PR) and prevents issues with iterative rebuild or mismatched build options between llama_main and the core runtime.

Note that the top-level make llama-cpu command is unmodified, just simplified under the hood. For direct CMake builds, here's a before and after:

Before:

cmake --preset llm -DCMAKE_INSTALL_PREFIX=cmake-out
cmake --build cmake-out --target install
cd examples/models/llama
cmake -B ../../../cmake-out/examples/models/llama -DCMAKE_FIND_ROOT_PATH=../../../cmake-out
cmake --build ../../../cmake-out/examples/models/llama --target llama_main

After:

cmake -B cmake-out --preset llm
cmake --build cmake-out --target llama_main

Note to reviewers: the MLX failures are unrelated. They are due to running extended CI on a forked PR.

cc @larryliu0820

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/18825

• 📄 Preview Python docs built from this PR

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

❌ 15 New Failures, 2 Cancelled Jobs, 3 Unrelated Failures

As of commit 5bb2f3e with merge base 38fdfae ():

NEW FAILURES - The following jobs have failed:

• MLX / test-mlx-llm (unsloth/gemma-3-1b-it, gemma3-1b, false, 4w) / test-mlx-llm-gemma3-1b-4w (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/gemma-3-1b-it, gemma3-1b, false, nvfp4) / test-mlx-llm-gemma3-1b-nvfp4 (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/gemma-3-1b-it, gemma3-1b, true, 4w) / test-mlx-llm-gemma3-1b-custom-4w (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/gemma-3-1b-it, gemma3-1b, true, nvfp4) / test-mlx-llm-gemma3-1b-custom-nvfp4 (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, false, 4w) / test-mlx-llm-llama-1b-4w (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, false, nvfp4) / test-mlx-llm-llama-1b-nvfp4 (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, true, 4w) / test-mlx-llm-llama-1b-custom-4w (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, true, nvfp4) / test-mlx-llm-llama-1b-custom-nvfp4 (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/Qwen3-0.6B, qwen3-0.6b, false, 4w) / test-mlx-llm-qwen3-0.6b-4w (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/Qwen3-0.6B, qwen3-0.6b, false, nvfp4) / test-mlx-llm-qwen3-0.6b-nvfp4 (gh)
  Library not loaded: @rpath/liblz4.1.dylib
• MLX / test-mlx-llm (unsloth/Qwen3-0.6B, qwen3-0.6b, true, 4w) / test-mlx-llm-qwen3-0.6b-custom-4w (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-llm (unsloth/Qwen3-0.6B, qwen3-0.6b, true, nvfp4) / test-mlx-llm-qwen3-0.6b-custom-nvfp4 (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• MLX / test-mlx-whisper / test-mlx-whisper (gh)
  RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
• pull / android / run-emulator (gh)
  The process '/usr/local/lib/android/sdk/platform-tools/adb' failed with exit code 224
• trunk / unittest-release / macos / macos-job (gh)
  export/tests/test_target_recipes.py::TestTargetRecipes::test_mv3_model

CANCELLED JOBS - The following jobs were cancelled. Please retry:

• Check Labels / Check labels (gh)
• periodic / gather-models (gh)

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

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

• pull / unittest / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.
• pull / unittest-editable / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.
• trunk / unittest-release / windows / windows-job (gh) (trunk failure)
  ##[error]The operation was canceled.

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

[GregoryComer]

CI should be passing now. The MLX failures are failing due to being on a forked PR. The android emulator and unittest failures are flakes.

[larryliu0820]

are these not needed anymore? Do we rely on whole archive flag added in executorch-config.cmake?

[larryliu0820]

Change itself looks ok. I don't think this approach is scalable though, like we should not make all the example cmake target visible to the top level cmake. I tend to make executorch easy to use such that, when a real application needs to depend on it, they just do add_subdirectory(executorch) and can link to all the executorch libraries and when they do that we should not expose example cmake targets to them. I guess since you exclude llama_main from being built by default it's an ok change.