Merge branch 'main' into export-D65911233

Author: zonglinpeng

.github/scripts/check_labels.py

@@ -0,0 +1,62 @@
+#!/usr/bin/env python3
+"""Check whether a PR has required labels."""
+
+import sys
+from typing import Any
+
+from github_utils import gh_delete_comment, gh_post_pr_comment
+from gitutils import get_git_remote_name, get_git_repo_dir, GitRepo
+from label_utils import has_required_labels, is_label_err_comment, LABEL_ERR_MSG
+from trymerge import GitHubPR
+
+
+def delete_all_label_err_comments(pr: "GitHubPR") -> None:
+    for comment in pr.get_comments():
+        if is_label_err_comment(comment):
+            gh_delete_comment(pr.org, pr.project, comment.database_id)
+
+
+def add_label_err_comment(pr: "GitHubPR") -> None:
+    # Only make a comment if one doesn't exist already
+    if not any(is_label_err_comment(comment) for comment in pr.get_comments()):
+        gh_post_pr_comment(pr.org, pr.project, pr.pr_num, LABEL_ERR_MSG)
+
+
+def parse_args() -> Any:
+    from argparse import ArgumentParser
+
+    parser = ArgumentParser("Check PR labels")
+    parser.add_argument("pr_num", type=int)
+    # add a flag to return a non-zero exit code if the PR does not have the required labels
+    parser.add_argument(
+        "--exit-non-zero",
+        action="store_true",
+        help="Return a non-zero exit code if the PR does not have the required labels",
+    )
+
+    return parser.parse_args()
+
+
+def main() -> None:
+    args = parse_args()
+    repo = GitRepo(get_git_repo_dir(), get_git_remote_name())
+    org, project = repo.gh_owner_and_name()
+    pr = GitHubPR(org, project, args.pr_num)
+
+    try:
+        if not has_required_labels(pr):
+            print(LABEL_ERR_MSG)
+            add_label_err_comment(pr)
+            if args.exit_non_zero:
+                sys.exit(1)
+        else:
+            delete_all_label_err_comments(pr)
+    except Exception as e:
+        if args.exit_non_zero:
+            sys.exit(1)
+
+    sys.exit(0)
+
+
+if __name__ == "__main__":
+    main()

.github/workflows/check-labels.yml

@@ -0,0 +1,54 @@
+name: Check Labels
+
+on:
+  # We need pull_request_target to be able to post comments on PRs from forks.
+  # Only allow pull_request_target when merging to main, not some historical branch.
+  #
+  # Make sure to don't introduce explicit checking out and installing/running
+  # untrusted user code into this workflow!
+  pull_request_target:
+    types: [opened, synchronize, reopened, labeled, unlabeled]
+    branches: [main]
+
+  # To check labels on ghstack PRs.
+  # Note: as pull_request doesn't trigger on PRs targeting main,
+  # to test changes to the workflow itself one needs to create
+  # a PR that targets a gh/**/base branch.
+  pull_request:
+    types: [opened, synchronize, reopened, labeled, unlabeled]
+    branches: [gh/**/base]
+
+  workflow_dispatch:
+    inputs:
+      pr_number:
+        description: 'PR number to check labels for'
+        required: true
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.sha }}-${{ github.event_name == 'workflow_dispatch' }}
+  cancel-in-progress: true
+
+jobs:
+  check-labels:
+    permissions:
+      contents: read
+      pull-requests: write
+    name: Check labels
+    if: github.repository_owner == 'pytorch'
+    runs-on: ubuntu-22.04
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+      - uses: actions/setup-python@v4
+        with:
+          python-version: '3.10'
+      # Not the direct dependencies but the script uses trymerge
+      - run: pip install pyyaml==6.0 rockset==1.0.3
+      - name: Check labels
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          PR_NUM: ${{ github.event.number || github.event.inputs.pr_number }}
+        run: |
+          set -ex
+          python3 .github/scripts/check_labels.py --exit-non-zero "${PR_NUM}"

backends/mediatek/scripts/README.md renamed to backends/mediatek/README.md

@@ -1,28 +1,35 @@
-# Build Instructions
+# MediaTek Backend on ExecuTorch
+MediaTek backend empowers ExecuTorch to speed up PyTorch models on edge devices that equips with MediaTek Neuron Processing Unit (NPU). This document offers a step-by-step guide to set up the build environment for the MediaTek ExecuTorch libraries.
 
-This document provides a step-by-step guide to set up the build environment for the MediaTek ExercuTorch libraries.
+## Supported Chips
 
-## Prerequisites
+The examples provided in this repository are tested and supported on the following MediaTek chip:
+
+- MediaTek Dimensity 9300 (D9300)
+
+## Build Instructions
+
+### Prerequisites
 
 Before you begin, ensure you have the following prerequisites installed and configured:
 
-### 1. Buck2 Build Tool
+#### 1. Buck2 Build Tool
 
 - **Download Buck2**: Obtain Buck2 from the official [releases page](https://github.com/facebook/buck2/releases/tag/2024-02-01).
 - **Add to PATH**: Extract the downloaded file and add the directory to your system's `$PATH` environment variable.
    ```bash
    export PATH=<path_to_buck>:$PATH
    ```
 
-### 2. Android NDK
+#### 2. Android NDK
 
 - **Download Android NDK**: Acquire the Android NDK version 26.3.11579264 from the [Android developer site](https://developer.android.com/ndk/downloads).
 - **Set NDK Path**: Ensure that the `$ANDROID_NDK` environment variable is set to the path where the NDK is located.
    ```bash
    export ANDROID_NDK=<path_to_android_ndk>
    ```
 
-### 3. MediaTek ExercuTorch Libraries
+#### 3. MediaTek ExecuTorch Libraries
 
 Download [NeuroPilot Express SDK](https://neuropilot.mediatek.com/resources/public/npexpress/en/docs/npexpress) from MediaTek's NeuroPilot portal:
 
@@ -31,11 +38,11 @@ Download [NeuroPilot Express SDK](https://neuropilot.mediatek.com/resources/publ
 - `mtk_converter-8.8.0.dev20240723+public.d1467db9-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl`: This library preprocess the model into a MediaTek representation.
 - `mtk_neuron-8.2.2-py3-none-linux_x86_64.whl`: This library converts the model to binaries.
 
-## Setup
+### Setup
 
 Follow the steps below to setup your build environment:
 
-1. **Setup ExercuTorch Environment**: Refer to the [Setting up ExercuTorch](https://pytorch.org/executorch/stable/getting-started-setup) guide for detailed instructions on setting up the ExercuTorch environment.
+1. **Setup ExecuTorch Environment**: Refer to the [Setting up ExecuTorch](https://pytorch.org/executorch/stable/getting-started-setup) guide for detailed instructions on setting up the ExecuTorch environment.
 
 2. **Setup MediaTek Backend Environment**
 - Install the dependent libs. Ensure that you are inside backends/mediatek/ directory
@@ -52,18 +59,21 @@ Follow the steps below to setup your build environment:
    export NEURON_BUFFER_ALLOCATOR_LIB=<path_to_buffer_allocator>
    ```
 
-## Build
+### Build
+1. Navigate to `scripts/` directory.
 
-1. **Build MediaTek Backend**: Once the prerequisites are in place, run the `mtk_build.sh` script to start the build process, MediaTek backend will be built under `cmake-android-out/backends/` as `libneuron_backend.so`
+2. **Build MediaTek Backend**: Once the prerequisites are in place, run the `mtk_build.sh` script to start the build process, MediaTek backend will be built under `cmake-android-out/backends/` as `libneuron_backend.so`
 
    ```bash
    ./mtk_build.sh
    ```
 
-## Run
+### Run
 
-1. **Push MediaTek universal SDK and MediaTek backend to the device**: push `libneuronusdk_adapter.mtk.so` and `libneuron_backend.so` to the phone and export it to the `$LD_LIBRARY_PATH` environment variable before executing ExercuTorch with MediaTek backend.
+1. **Push MediaTek universal SDK and MediaTek backend to the device**: push `libneuronusdk_adapter.mtk.so` and `libneuron_backend.so` to the phone and export it to the `$LD_LIBRARY_PATH` environment variable before executing ExecuTorch with MediaTek backend.
 
    ```bash
    export LD_LIBRARY_PATH=<path_to_usdk>:<path_to_neuron_backend>:$LD_LIBRARY_PATH
    ```
+
+Please refer to `executorch/examples/mediatek/` for export and execution examples of various of models.

backends/xnnpack/test/ops/conv2d.py

@@ -394,3 +394,26 @@ def get_inputs(self):
             quant_config=get_symmetric_quantization_config(),
             conv_count=2,
         )
+
+    def test_qs8_conv2d_relu_multi_users(self):
+        class Conv2dReluMultiUsers(torch.nn.Module):
+            def __init__(self):
+                super().__init__()
+                self.conv1 = torch.nn.Conv2d(1, 1, 1)
+                self.conv2 = torch.nn.Conv2d(1, 64, 1)
+                self.relu = torch.nn.ReLU()
+
+            def forward(self, x):
+                conv_default = self.conv1(x)
+                y = self.relu(conv_default)
+                conv_default_2 = self.conv2(y)
+                return conv_default + conv_default_2
+
+            def get_inputs(self):
+                return (torch.randn(1, 1, 1, 1),)
+
+        self._test(
+            Conv2dReluMultiUsers(),
+            quant_config=get_symmetric_quantization_config(),
+            conv_count=2,
+        )

docs/source/build-run-mediatek-backend.md

@@ -0,0 +1,94 @@
+# Building and Running ExecuTorch with MediaTek Backend
+
+MediaTek backend empowers ExecuTorch to speed up PyTorch models on edge devices that equips with MediaTek Neuron Processing Unit (NPU). This document offers a step-by-step guide to set up the build environment for the MediaTek ExecuTorch libraries.
+
+::::{grid} 2
+:::{grid-item-card}  What you will learn in this tutorial:
+:class-card: card-prerequisites
+* How to export and lower a PyTorch model ahead of time with ExecuTorch for MediaTek devices.
+* How to build MediaTek backend and examples.
+* How to deploy the exported models on device with ExecuTorch runtime.
+:::
+:::{grid-item-card}  Tutorials we recommend you complete before this:
+:class-card: card-prerequisites
+* [Introduction to ExecuTorch](intro-how-it-works.md)
+* [Setting up ExecuTorch](getting-started-setup.md)
+* [Building ExecuTorch with CMake](runtime-build-and-cross-compilation.md)
+:::
+::::
+
+
+## Prerequisites (Hardware and Software)
+
+### Host OS
+- Linux operating system
+
+### Supported Chips:
+- MediaTek Dimensity 9300 (D9300)
+
+### Software:
+
+- [NeuroPilot Express SDK](https://neuropilot.mediatek.com/resources/public/npexpress/en/docs/npexpress) is a lightweight SDK for deploying AI applications on MediaTek SOC devices.
+
+## Setting up your developer environment
+
+Follow the steps below to setup your build environment:
+
+1. **Setup ExecuTorch Environment**: Refer to the [Setting up ExecuTorch](https://pytorch.org/executorch/stable/getting-started-setup) guide for detailed instructions on setting up the ExecuTorch environment.
+
+2. **Setup MediaTek Backend Environment**
+- Install the dependent libs. Ensure that you are inside `backends/mediatek/` directory
+   ```bash
+   pip3 install -r requirements.txt
+   ```
+- Install the two .whl downloaded from NeuroPilot Portal
+   ```bash
+   pip3 install mtk_neuron-8.2.13-py3-none-linux_x86_64.whl
+   pip3 install mtk_converter-8.9.1+public-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
+   ```
+- Set evironment variables for building backend
+   ```bash
+   export NEURON_BUFFER_ALLOCATOR_LIB=<path_to_buffer_allocator.so>
+   ```
+
+## Build
+
+### Ahead of time:
+
+**Exporting a PyTorch Model for MediaTek Backend**:
+1. Lower and export the `.pte` file for on-device execution. The export script samples are povided under `example/mediatek/`. For example, the following commnad exports the `.pte` using the scripts provided.
+```bash
+cd executorch
+
+./examples/mediatek/shell_scripts/export_oss.sh mobilenetv3
+```
+
+2. Find the `.pte` files under the directory named as same as the model.
+
+### Runtime:
+
+**Build MediaTek Backend for ExecuTorch Runtime**
+1. Navigate to `backends/mediatek/scripts/` directory.
+
+2. **Build MediaTek Backend**: Once the prerequisites are in place, run the `mtk_build.sh` script to start the build process:
+   ```bash
+   ./mtk_build.sh
+   ```
+
+3. MediaTek backend will be built under `cmake-android-out/backends/` as `libneuron_backend.so`.
+
+**Build a runner to execute the model on the device**:
+1. Build the runners and the backend by exedcuting the script:
+```bash
+./mtk_build_examples.sh
+```
+
+2. The runners will be built under `cmake-android-out/examples/`
+
+## Deploying and running on a device
+
+1. **Push MediaTek universal SDK and MediaTek backend to the device**: push `libneuronusdk_adapter.mtk.so` and `libneuron_backend.so` to the phone and export it to the `$LD_LIBRARY_PATH` environment variable before executing ExecuTorch with MediaTek backend.
+
+   ```bash
+   export LD_LIBRARY_PATH=<path_to_usdk>:<path_to_neuron_backend>:$LD_LIBRARY_PATH
+   ```

docs/source/index.rst

@@ -111,6 +111,7 @@ Topics in this section will help you get started with ExecuTorch.
       customcarditem entries below.
    executorch-arm-delegate-tutorial
    build-run-coreml
+   build-run-mediatek-backend
    build-run-mps
    build-run-qualcomm-ai-engine-direct-backend
    build-run-xtensa
@@ -331,6 +332,13 @@ ExecuTorch tutorials.
    :link: build-run-coreml.html
    :tags: Export,Backend,Delegation,CoreML
 
+.. customcarditem::
+   :header: Building and Running ExecuTorch with MediaTek Backend
+   :card_description: A tutorial that walks you through the process of building ExecuTorch with MediaTek Backend
+   :image: _static/img/generic-pytorch-logo.png
+   :link: build-run-mediatek-backend.html
+   :tags: Export,Backend,Delegation,MediaTek
+
 .. customcarditem::
    :header: Building and Running ExecuTorch with MPS Backend
    :card_description: A tutorial that walks you through the process of building ExecuTorch with MPSGraph Backend

examples/mediatek/README.md

@@ -72,12 +72,6 @@ bash shell_scripts/export_oss.sh <model_name>
     - `model_name`: deeplabv3/edsr/inceptionv3/inceptionv4/mobilenetv2/mobilenetv3/resnet18/resnet50
 
 # Runtime
-## Supported Chips
-
-The examples provided in this repository are tested and supported on the following MediaTek chip:
-
-- MediaTek Dimensity 9300 (D9300)
-
 ## Environment Setup
 
 To set up the build environment for the `mtk_executor_runner`:

runtime/executor/method.cpp

@@ -963,8 +963,8 @@ Method::set_output_data_ptr(void* buffer, size_t size, size_t output_idx) {
   if (tensor_meta->is_memory_planned()) {
     ET_LOG(
         Error,
-        "Output %zu is memory planned, or is a constant. Cannot override \
-        the existing data pointer.",
+        "Output %zu is memory planned, or is a constant. Cannot override "
+        "the existing data pointer.",
         output_idx);
     return Error::InvalidState;
   }