update xnnpack static docs for alpha (pytorch#2755)

Summary:
Updating static docs for xnnpack

Pull Request resolved: pytorch#2755

Reviewed By: kirklandsign

Differential Revision: D55507499

Pulled By: mcr229

fbshipit-source-id: aaada214b113f224da4ef028f3121585de0c17bc

Author: mcr229

docs/source/native-delegates-executorch-xnnpack-delegate.md

@@ -61,19 +61,19 @@ The XNNPACK delegate uses flatbuffer for serialization. In order to improve runt
 The XNNPACK backend’s runtime interfaces with the ExecuTorch runtime through the custom `init` and `execute` function. Each delegated subgraph is contained in an individually serialized XNNPACK blob. When the model is initialized, ExecuTorch calls `init` on all XNNPACK Blobs to load the subgraph from serialized flatbuffer. After, when the model is executed, each subgraph is executed via the backend through the custom `execute` function. To read more about how delegate runtimes interface with ExecuTorch, refer to this [resource](compiler-delegate-and-partitioner.md).
 
 
-#### XNNPACK Library
-The XNNPACK Library currently used by the delegate is on the following [version](https://github.com/google/XNNPACK/tree/51a987591a6fc9f0fc0707077f53d763ac132cbf). XNNPACK delegate supports CPU's on multiple platforms; more information on the supported hardware architectures can be found on the XNNPACK Library’s [README](https://github.com/google/XNNPACK).
+#### **XNNPACK Library**
+XNNPACK delegate supports CPU's on multiple platforms; more information on the supported hardware architectures can be found on the XNNPACK Library’s [README](https://github.com/google/XNNPACK).
 
-#### Init
+#### **Init**
 When calling XNNPACK delegate’s `init`, we deserialize the preprocessed blobs via flatbuffer. We define the nodes (operators) and edges (intermediate tensors) to build the XNNPACK execution graph using the information we serialized ahead-of-time. As we mentioned earlier, the majority of processing has been done ahead-of-time, so that at runtime we can just call the XNNPACK APIs with the serialized arguments in succession. As we define static data into the execution graph, XNNPACK performs weight packing at runtime to prepare static data like weights and biases for efficient execution. After creating the execution graph, we create the runtime object and pass it on to `execute`.
 
 Since weight packing creates an extra copy of the weights inside XNNPACK, We free the original copy of the weights inside the preprocessed XNNPACK Blob, this allows us to remove some of the memory overhead.
 
 
-#### Execute
+#### **Execute**
 When executing the XNNPACK subgraphs, we prepare the tensor inputs and outputs and feed them to the XNNPACK runtime graph. After executing the runtime graph, the output pointers are filled with the computed tensors.
 
-#### Profiling
+#### **Profiling**
 We have enabled basic profiling for XNNPACK delegate that can be enabled with the following compiler flag `-DENABLE_XNNPACK_PROFILING`. After running the model it will produce basic per-op and total timings. We provide an example of the profiling below. The timings listed are the average across runs, and the units are in microseconds.
 
 ```

docs/source/tutorial-xnnpack-delegate-lowering.md

@@ -12,7 +12,6 @@ In this tutorial, you will learn how to export an XNNPACK lowered Model and run
 :class-card: card-prerequisites
 * [Setting up ExecuTorch](./getting-started-setup.md)
 * [Model Lowering Tutorial](./tutorials/export-to-executorch-tutorial)
-* [Custom Quantization](./quantization-custom-quantization.md)
 * [ExecuTorch XNNPACK Delegate](./native-delegates-executorch-xnnpack-delegate.md)
 :::
 ::::
@@ -23,16 +22,18 @@ In this tutorial, you will learn how to export an XNNPACK lowered Model and run
 import torch
 import torchvision.models as models
 
-from torch.export import export
+from torch.export import export, ExportedProgram
 from torchvision.models.mobilenetv2 import MobileNet_V2_Weights
 from executorch.backends.xnnpack.partition.xnnpack_partitioner import XnnpackPartitioner
-from executorch.exir import to_edge
+from executorch.exir import EdgeProgramManager, ExecutorchProgramManager, to_edge
+from executorch.exir.backend.backend_api import to_backend
 
 
 mobilenet_v2 = models.mobilenetv2.mobilenet_v2(weights=MobileNet_V2_Weights.DEFAULT).eval()
 sample_inputs = (torch.randn(1, 3, 224, 224), )
 
-edge = to_edge(export(mobilenet_v2, sample_inputs))
+exported_program: ExportedProgram = export(mobilenet_v2, sample_inputs)
+edge: EdgeProgramManager = to_edge(exported_program)
 
 edge = edge.to_backend(XnnpackPartitioner())
 ```
@@ -64,7 +65,7 @@ We print the graph after lowering above to show the new nodes that were inserted
 exec_prog = edge.to_executorch()
 
 with open("xnnpack_mobilenetv2.pte", "wb") as file:
-    file.write(exec_prog.buffer)
+    exec_prog.write_to_file(file)
 ```
 After lowering to the XNNPACK Program, we can then prepare it for executorch and save the model as a `.pte` file. `.pte` is a binary format that stores the serialized ExecuTorch graph.
 
@@ -117,14 +118,14 @@ edge = edge.to_backend(XnnpackPartitioner())
 exec_prog = edge.to_executorch()
 
 with open("qs8_xnnpack_mobilenetv2.pte", "wb") as file:
-    file.write(exec_prog.buffer)
+    exec_prog.write_to_file(file)
 ```
 
 ## Lowering with `aot_compiler.py` script
 We have also provided a script to quickly lower and export a few example models. You can run the script to generate lowered fp32 and quantized models. This script is used simply for convenience and performs all the same steps as those listed in the previous two sections.
 
 ```
-python3 -m examples.xnnpack.aot_compiler --model_name="mv2" --quantize --delegate
+python -m examples.xnnpack.aot_compiler --model_name="mv2" --quantize --delegate
 ```
 
 Note in the example above,
@@ -134,13 +135,48 @@ Note in the example above,
 
 The generated model file will be named `[model_name]_xnnpack_[qs8/fp32].pte` depending on the arguments supplied.
 
-## Running the XNNPACK Model
-We will use `buck2` to run the `.pte` file with XNNPACK delegate instructions in it on your host platform. You can follow the instructions here to install [buck2](getting-started-setup.md#building-a-runtime). You can now run it with the prebuilt `xnn_executor_runner` provided in the examples. This will run the model on some sample inputs.
+## Running the XNNPACK Model with CMake
+After exporting the XNNPACK Delegated model, we can now try running it with example inputs using CMake. We can build and use the xnn_executor_runner, which is a sample wrapper for the ExecuTorch Runtime and XNNPACK Backend. We first begin by configuring the CMake build like such:
+```bash
+# cd to the root of executorch repo
+cd executorch
+
+# Get a clean cmake-out directory
+rm- -rf cmake-out
+mkdir cmake-out
+
+# Configure cmake
+cmake \
+    -DCMAKE_INSTALL_PREFIX=cmake-out \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DEXECUTORCH_BUILD_EXTENSION_MODULE=ON \
+    -DEXECUTORCH_BUILD_XNNPACK=ON \
+    -DEXECUTORCH_BUILD_EXTENSION_DATA_LOADER=ON \
+    -DEXECUTORCH_ENABLE_LOGGING=1 \
+    -DPYTHON_EXECUTABLE=python \
+    -Bcmake-out .
+```
+Then you can build the runtime componenets with
+
+```bash
+cmake --build cmake-out -j9 --target install --config Release
+```
+
+Now you should be able to find the executable built at `./cmake-out/backends/xnnpack/xnn_executor_runner` you can run the executable with the model you generated as such
+```bash
+./cmake-out/backends/xnnpack/xnn_executor_runner --model_path=./mv2_xnnpack_fp32.pte
+# or to run the quantized variant
+./cmake-out/backends/xnnpack/xnn_executor_runner --model_path=./mv2_xnnpack_q8.pte
+```
+
+
+## Running the XNNPACK Model with Buck
+Alternatively, you can use `buck2` to run the `.pte` file with XNNPACK delegate instructions in it on your host platform. You can follow the instructions here to install [buck2](getting-started-setup.md#building-a-runtime). You can now run it with the prebuilt `xnn_executor_runner` provided in the examples. This will run the model on some sample inputs.
 
 ```bash
 buck2 run examples/xnnpack:xnn_executor_runner -- --model_path ./mv2_xnnpack_fp32.pte
 # or to run the quantized variant
-buck2 run examples/xnnpack:xnn_executor_runner -- --model_path ./mv2_xnnpack_qs8.pte
+buck2 run examples/xnnpack:xnn_executor_runner -- --model_path ./mv2_xnnpack_q8.pte
 ```
 
 ## Building and Linking with the XNNPACK Backend