pytorch · lucylq · Oct 11, 2023
@@ -7,7 +7,7 @@ on a Android device.
 
 ## Prerequisite
 
-Please finish tutorial [Setting up executorch](../../docs/website/docs/tutorials/00_setting_up_executorch.md).
+Please finish tutorial [Setting up executorch](../../docs/source/getting-started-setup.md).
 
 
 ## Conventions

@@ -1,7 +1,7 @@
 ## ExecuTorch on ARM Cortex-M55 + Ethos-U55
 
 This dir contains scripts to help you prepare setup needed to run a PyTorch
-model on a ARM Corstone-300 platform via ExecuTorch. Corstone-300 platform
+model on an ARM Corstone-300 platform via ExecuTorch. Corstone-300 platform
 contains the Cortex-M55 CPU and Ethos-U55 NPU.
 
 We will start from a PyTorch model in python, export it, convert it to a `.pte`

@@ -16,11 +16,11 @@ examples/portable
 
 ## Using portable mode
 
-We will walk through an example model to generate a `.pte` file in [portable mode](/docs/website/docs/basics/terminology.md) from a python `torch.nn.module`
+We will walk through an example model to generate a `.pte` file in [portable mode](../../docs/source/concepts.md#portable-mode-lean-mode) from a python `torch.nn.module`
 from the [`models/`](../models) directory using scripts in the `portable/scripts` directory. Then we will run on the `.pte` model on the ExecuTorch runtime. For that we will use `executor_runner`.
 
 
-1. Following the setup guide in [Setting up ExecuTorch from GitHub](/docs/website/docs/tutorials/00_setting_up_executorch.md)
+1. Following the setup guide in [Setting up ExecuTorch from GitHub](../../docs/source/getting-started-setup.md)
 you should be able to get the basic development environment for ExecuTorch working.
 
 2. Using the script `portable/scripts/export.py` generate a model binary file by selecting a

@@ -3,12 +3,13 @@ This folder contains examples to register custom operators into PyTorch as well
 
 ## How to run
 
-Prerequisite: finish the [setting up wiki](https://github.com/pytorch/executorch/blob/main/docs/website/docs/tutorials/00_setting_up_executorch.md).
+Prerequisite: finish the [setting up wiki](../../../docs/source/getting-started-setup.md).
 
 Run:
 
 ```bash
-bash test_custom_ops.sh
+cd executorch
+bash examples/portable/custom_ops/test_custom_ops.sh [cmake|buck2]
 ```
 
 ## AOT registration
@@ -27,7 +28,7 @@ By linking them both with `libtorch` and `executorch` library, we can build a sh
 
 ## C++ kernel registration
 
-After the model is exported by EXIR, we need C++ implementations of these custom ops in order to run it. For example, `custom_ops_1_out.cpp` is C++ kernel that can be plugged in to ExecuTorch runtime. Other than that, we also need a way to bind the PyTorch op to this kernel. This binding is specified in `custom_ops.yaml`:
+After the model is exported by EXIR, we need C++ implementations of these custom ops in order to run it. For example, `custom_ops_1_out.cpp` is a C++ kernel that can be plugged into the ExecuTorch runtime. Other than that, we also need a way to bind the PyTorch op to this kernel. This binding is specified in `custom_ops.yaml`:
 ```yaml
 - func: my_ops::mul3.out(Tensor input, *, Tensor(a!) output) -> Tensor(a!)
   kernels:
@@ -57,4 +58,4 @@ et_operator_library(
 
 We then let the custom ops library depend on this target, to only register the ops we want.
 
-For more information about selective build, please refer to [`docs/tutorials/selective_build.md`](https://github.com/pytorch/executorch/blob/main/docs/website/docs/tutorials/selective_build.md).
+For more information about selective build, please refer to [`selective_build.md`](../../../docs/source/kernel-library-selective_build.md).
@@ -21,7 +21,7 @@ test_buck2_custom_op_1() {
   # should save file custom_ops_1.pte
 
   echo 'Running executor_runner'
-  buck2 run //examples/portable/executor_runner:executor_runner \
+  $BUCK2 run //examples/portable/executor_runner:executor_runner \
       --config=executorch.register_custom_op=1 -- --model_path="./${model_name}.pte"
   # should give correct result
 
@@ -37,7 +37,7 @@ test_cmake_custom_op_1() {
   (rm -rf cmake-out \
     && mkdir cmake-out \
     && cd cmake-out \
-    && retry cmake -DBUCK2=buck2 \
+    && retry cmake -DBUCK2=$BUCK2 \
         -DREGISTER_EXAMPLE_CUSTOM_OP=1 \
         -DPYTHON_EXECUTABLE="$PYTHON_EXECUTABLE" ..)
 
@@ -52,13 +52,13 @@ test_buck2_custom_op_2() {
   local model_name='custom_ops_2'
 
   echo 'Building custom ops shared library'
-  SO_LIB=$(buck2 build //examples/portable/custom_ops:custom_ops_aot_lib_2 --show-output | grep "buck-out" | cut -d" " -f2)
+  SO_LIB=$($BUCK2 build //examples/portable/custom_ops:custom_ops_aot_lib_2 --show-output | grep "buck-out" | cut -d" " -f2)
 
   echo "Exporting ${model_name}.pte"
   ${PYTHON_EXECUTABLE} -m "examples.portable.custom_ops.${model_name}" --so_library="$SO_LIB"
   # should save file custom_ops_2.pte
 
-  buck2 run //examples/portable/executor_runner:executor_runner \
+  $BUCK2 run //examples/portable/executor_runner:executor_runner \
       --config=executorch.register_custom_op=2 -- --model_path="./${model_name}.pte"
   # should give correct result
   echo "Removing ${model_name}.pte"
@@ -88,7 +88,7 @@ test_cmake_custom_op_2() {
   (rm -rf cmake-out \
     && mkdir cmake-out \
     && cd cmake-out \
-    && retry cmake -DBUCK2=buck2 \
+    && retry cmake -DBUCK2=$BUCK2 \
       -DREGISTER_EXAMPLE_CUSTOM_OP=2 \
       -DCMAKE_PREFIX_PATH="$CMAKE_PREFIX_PATH" \
       -DPYTHON_EXECUTABLE="$PYTHON_EXECUTABLE" ..)
@@ -109,6 +109,12 @@ if [[ -z $PYTHON_EXECUTABLE ]];
 then
   PYTHON_EXECUTABLE=python3
 fi
+
+if [[ -z $BUCK2 ]];
+then
+  BUCK2=buck2
+fi
+
 if [[ $1 == "cmake" ]];
 then
   test_cmake_custom_op_1

@@ -8,7 +8,7 @@ Here are some general information and limitations.
 
 ## Prerequisite
 
-Please finish tutorial [Setting up executorch](../../docs/website/docs/tutorials/00_setting_up_executorch.md).
+Please finish tutorial [Setting up executorch](../../docs/source/getting-started-setup.md).
 
 Please finish [setup QNN backend](../../backends/qualcomm/setup.md).
 
@@ -57,8 +57,7 @@ but the performance and accuracy number can differ.
 set of SoCs. Please check QNN documents for details.
 
 3. The mobilebert example needs to train the last classifier layer a bit, so it takes
-time to run. 
+time to run.
 
 4. [**Important**] Due to the numerical limits of FP16, other use cases leveraging mobileBert wouldn't
 guarantee to work.
-
@@ -1,14 +1,15 @@
 # Selective Build Examples
-To optimize binary size of ExecuTorch runtime, selective build can be used. This folder contains examples to select only the operators needed for ExecuTorch build. We provide APIs for both CMake build and buck2 build. This example will demonstrate both. You can find more information on how to use buck2 macros in [wiki](https://github.com/pytorch/executorch/blob/main/docs/website/docs/tutorials/selective_build.md).
+To optimize binary size of ExecuTorch runtime, selective build can be used. This folder contains examples to select only the operators needed for ExecuTorch build. We provide APIs for both CMake build and buck2 build. This example will demonstrate both. You can find more information on how to use buck2 macros in [wiki](../../docs/source/kernel-library-selective_build.md).
 
 ## How to run
 
-Prerequisite: finish the [setting up wiki](https://github.com/pytorch/executorch/blob/main/docs/website/docs/tutorials/00_setting_up_executorch.md).
+Prerequisite: finish the [setting up wiki](../../docs/source/getting-started-setup.md).
 
 Run:
 
 ```bash
-bash test_selective_build.sh [cmake|buck2]
+cd executorch
+bash examples/selective_build/test_selective_build.sh [cmake|buck2]
 ```
 
 ## BUCK2 examples