Arm backend: Document the new Arm runner workflow

- Explain the auto-configured runner build flow and scratch directory
  expectations in examples/arm/README.md.
- Update the Ethos-U notebook to export EXECUTORCH_ROOT before calling
  standalone cmake.

Change-Id: If9f4f456c03b7a36a27ffdd1dfd1873ec286d07b
Signed-off-by: Usamah Zaheer <usamah.zaheer@arm.com>

Author: usamahz

examples/arm/README.md

@@ -1,3 +1,10 @@
+<!--
+Copyright 2023-2026 Arm Limited and/or its affiliates.
+
+This source code is licensed under the BSD-style license found in the
+LICENSE file in the root directory of this source tree.
+-->
+
 ## ExecuTorch for Arm backends Ethos-U, VGF and Cortex-M
 
 This project contains scripts to help you setup and run a PyTorch
@@ -11,12 +18,19 @@ The main scripts are `setup.sh`, `run.sh` and
 `setup.sh` will install the needed tools and with --root-dir <FOLDER> 
 you can change the path to a scratch folder where it will download and generate build
 artifacts. If supplied, you must also supply the same folder to run.sh with
---scratch-dir=<FOLDER> If not supplied both script will use examples/arm/arm-scratch
+--scratch-dir=<FOLDER> If not supplied both scripts will use examples/arm/arm-scratch.
 
 `run.sh` can be used to build, run and test a model in an easy way and it will call cmake for you
 and in cases you want to run a simulator it will start it also. The script will call `aot_arm_compiler.py`
 to convert a model and include it in the build/run.
 
+For bare-metal Ethos-U builds `run.sh` configures the standalone
+`examples/arm/executor_runner/standalone` CMake entry point automatically. If
+`--build-dir` is omitted, the script creates and owns a build tree under
+`arm_test/<target>_<build_type>`. Supplying `--build-dir` reuses an existing tree
+(for example a VGF host build or out-of-tree configuration) and `run.sh`
+verifies it exposes the runner options it needs before compiling.
+
 Build and test artifacts are by default placed under the folder arm_test folder
 this can be changed with --et_build_root=<FOLDER>
 

examples/arm/cortex_m_mv2_example.ipynb

@@ -136,7 +136,7 @@
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": "%%bash \n# Build example executor runner application to examples/arm/cortex_m_mv2_example\n# Note that this is the same runner as used in the Ethos-U example, creating some overlap in the config even though the Ethos-U is not used.\ncmake -DCMAKE_TOOLCHAIN_FILE=$(pwd)/ethos-u-setup/arm-none-eabi-gcc.cmake \\\n      -DCMAKE_BUILD_TYPE=Release \\\n      -DET_PTE_FILE_PATH=cortex_m_mv2_example.bpte \\\n      -DTARGET_CPU=cortex-m55 \\\n      -DETHOSU_TARGET_NPU_CONFIG=ethos-u55-128 \\\n      -DMEMORY_MODE=Shared_Sram \\\n      -DET_BUNDLE_IO=ON \\\n      -DSYSTEM_CONFIG=Ethos_U55_High_End_Embedded \\\n      -Bcortex_m_mv2_example \\\n      executor_runner\ncmake --build cortex_m_mv2_example -j$(nproc) -- arm_executor_runner"
+   "source": "%%bash \n# Build example executor runner application to examples/arm/cortex_m_mv2_example\n# Note that this is the same runner as used in the Ethos-U example, creating some overlap in the config even though the Ethos-U is not used.\ncmake -DCMAKE_TOOLCHAIN_FILE=$(pwd)/ethos-u-setup/arm-none-eabi-gcc.cmake \\\n      -DCMAKE_BUILD_TYPE=Release \\\n      -DET_PTE_FILE_PATH=cortex_m_mv2_example.bpte \\\n      -DTARGET_CPU=cortex-m55 \\\n      -DETHOSU_TARGET_NPU_CONFIG=ethos-u55-128 \\\n      -DMEMORY_MODE=Shared_Sram \\\n      -DET_BUNDLE_IO=ON \\\n      -DSYSTEM_CONFIG=Ethos_U55_High_End_Embedded \\\n      -Bcortex_m_mv2_example \\\n      -S executor_runner/standalone\ncmake --build cortex_m_mv2_example -j$(nproc) -- arm_executor_runner"
   },
   {
    "cell_type": "markdown",
@@ -179,4 +179,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 4
-}
+}

examples/arm/ethos_u_minimal_example.ipynb

@@ -171,26 +171,8 @@
    "source": [
     "## Build executor runtime\n",
     "\n",
-    "After the AOT compilation flow is done, the runtime can be cross compiled and linked to the produced .pte-file using the Arm cross-compilation toolchain. This is done in two steps:\n",
-    "1. Build and install the executorch libraries and EthosUDelegate.\n",
-    "2. Build and link the `arm_executor_runner` and generate kernel bindings for any non delegated ops."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "%%bash\n",
-    "# Ensure the arm-none-eabi-gcc toolchain and FVP:s are available on $PATH\n",
-    "source arm-scratch/setup_path.sh\n",
-    "\n",
-    "# Build executorch libraries cross-compiled for arm baremetal to executorch/cmake-out-arm\n",
-    "cmake --preset arm-baremetal \\\n",
-    "-DCMAKE_BUILD_TYPE=Release \\\n",
-    "-B../../cmake-out-arm ../..\n",
-    "cmake --build ../../cmake-out-arm --target install -j$(nproc) "
+    "After the AOT compilation flow finishes, cross-compile and link the runtime by configuring the standalone `examples/arm/executor_runner/standalone` CMake project with the Arm toolchain.\n",
+    "It automatically pulls the ExecuTorch checkout in as a dependency so the delegate, kernels, and runner util are rebuilt alongside the application, and it generates kernel bindings for any non-delegated ops found in the `.pte`.\n"
    ]
   },
   {
@@ -201,6 +183,8 @@
    "source": [
     "%%bash \n",
     "source arm-scratch/setup_path.sh\n",
+    "# Ensure CMake resolves the ExecuTorch checkout root regardless of caller env\n",
+    "export EXECUTORCH_ROOT=$(cd ../.. && pwd)\n",
     "\n",
     "# Build example executor runner application to examples/arm/ethos_u_minimal_example\n",
     "cmake -DCMAKE_TOOLCHAIN_FILE=$(pwd)/ethos-u-setup/arm-none-eabi-gcc.cmake \\\n",
@@ -211,7 +195,7 @@
     "      -DMEMORY_MODE=Shared_Sram \\\n",
     "      -DSYSTEM_CONFIG=Ethos_U55_High_End_Embedded \\\n",
     "      -Bethos_u_minimal_example \\\n",
-    "      executor_runner\n",
+    "      -S executor_runner/standalone\n",
     "cmake --build ethos_u_minimal_example -j$(nproc) -- arm_executor_runner"
    ]
   },
@@ -232,6 +216,8 @@
    "source": [
     "%%bash \n",
     "source arm-scratch/setup_path.sh\n",
+    "# Ensure CMake resolves the ExecuTorch checkout root regardless of caller env\n",
+    "export EXECUTORCH_ROOT=$(cd ../.. && pwd)\n",
     "\n",
     "# Run the example\n",
     "../../backends/arm/scripts/run_fvp.sh --elf=ethos_u_minimal_example/arm_executor_runner --target=ethos-u55-128"

examples/arm/pruning_minimal_example.ipynb

@@ -453,7 +453,7 @@
     "      -DMEMORY_MODE=Shared_Sram \\\n",
     "      -DSYSTEM_CONFIG=Ethos_U85_SYS_DRAM_Mid \\\n",
     "      -Bethos_u_original_model \\\n",
-    "      executor_runner\n",
+    "      -S executor_runner/standalone\n",
     "cmake --build ethos_u_original_model -j$(nproc) -- arm_executor_runner"
    ]
   },
@@ -499,7 +499,7 @@
     "      -DMEMORY_MODE=Shared_Sram \\\n",
     "      -DSYSTEM_CONFIG=Ethos_U85_SYS_DRAM_Mid \\\n",
     "      -Bethos_u_pruned_model \\\n",
-    "      executor_runner\n",
+    "      -S executor_runner/standalone\n",
     "cmake --build ethos_u_pruned_model -j$(nproc) -- arm_executor_runner"
    ]
   },