Update on "[ET-VK] Add symint infrastructure to VulkanBackend and ComputeGraph"

Extend the Vulkan backend runtime infrastructure to better support symbolic
integer (symint) arguments. This is a prerequisite for operators that need
to handle dynamic shapes via symint values.

Changes:
- VulkanBackend.cpp: Compute output offset from end of args instead of
  assuming outputs follow inputs directly. Add scalar-to-tensor input
  handling so that Int/Bool EValues can populate tensor inputs. Support
  symint inputs provided as raw Int EValues (not just scalar tensors).
  Add symint output handling to write values back as tensor or Int EValue.
- ComputeGraph.h: Add SymInt case to extract_scalar<T>() so operators
  can transparently read symint values as scalars.
- ComputeGraph.cpp: Add Int fallback in read_symint() so values stored
  as plain Int (rather than SymInt objects) can be read uniformly.

Differential Revision: [D95970167](https://our.internmc.facebook.com/intern/diff/D95970167/)

cc manuelcandales digantdesai cbilgin

[ghstack-poisoned]

Author: ssjia

.ci/scripts/test_wheel_package_qnn.sh

@@ -18,6 +18,9 @@ import argparse
 
 import torch
 from executorch.backends.qualcomm.quantizer.quantizer import QnnQuantizer
+from executorch.backends.qualcomm.serialization.qc_schema import (
+    QnnExecuTorchBackendType,
+)
 from executorch.backends.qualcomm.utils.utils import (
     generate_htp_compiler_spec,
     generate_qnn_executorch_compiler_spec,
@@ -50,7 +53,7 @@ def main() -> None:
     example_inputs = model.get_example_inputs()
 
     if args.quantization:
-        quantizer = QnnQuantizer()
+        quantizer = QnnQuantizer(backend=QnnExecuTorchBackendType.kHtpBackend, soc_model=get_soc_to_chipset_map()[args.soc])
         m = torch.export.export(model.eval(), example_inputs, strict=True).module()
         if args.quantization == "qat":
             m = prepare_qat_pt2e(m, quantizer)

.claude/skills/building/SKILL.md

@@ -1,23 +1,223 @@
 ---
 name: building
-description: Build ExecuTorch runners or C++ libraries. Use when compiling runners for Llama, Whisper, or other models, or building the C++ runtime.
+description: Build ExecuTorch from source — Python package, C++ runtime, runners, cross-compilation, and backend-specific builds. Use when compiling anything in the ExecuTorch repo, diagnosing build failures, or setting up platform-specific builds.
 ---
 
-# Building
+# Building ExecuTorch
 
-## Runners (Makefile)
+## Step 1: Ensure Python environment (detect and fix automatically)
+
+**Path A — conda (preferred):**
+```bash
+# Initialize conda for non-interactive shells (required in Claude Code / CI)
+eval "$(conda shell.bash hook 2>/dev/null)"
+
+# Check if executorch conda env exists; create if not
+conda env list 2>/dev/null | grep executorch || \
+  ls "$(conda info --base 2>/dev/null)/envs/" 2>/dev/null | grep executorch || \
+  conda create -yn executorch python=3.12
+
+# Activate
+conda activate executorch
+```
+
+**Path B — no conda (fall back to venv):**
+```bash
+# Find a compatible Python (3.10–3.13). On macOS with only Homebrew Python 3.14+,
+# install a compatible version first: brew install python@3.12
+python3.12 -m venv .executorch-venv   # or python3.11, python3.10, python3.13
+source .executorch-venv/bin/activate
+pip install --upgrade pip
+```
+
+**Then verify (either path):**
+
+Run `python --version` and `cmake --version`. Fix automatically:
+- **Python not 3.10–3.13**: recreate the env with a correct Python version.
+- **cmake missing or < 3.24**: run `pip install 'cmake>=3.24'` inside the env.
+- **cmake >= 4.0**: works in practice, no action needed.
+
+Parallel jobs: `$(sysctl -n hw.ncpu)` on macOS, `$(nproc)` on Linux.
+
+## Step 2: Build
+
+Route based on what the user asks for:
+- User mentions **Android** → skip to [Cross-compilation: Android](#cross-compilation)
+- User mentions **iOS** or **frameworks** → skip to [Cross-compilation: iOS](#cross-compilation)
+- User mentions a **model name** (llama, whisper, etc.) → skip to [LLM / ASR model runner](#llm--asr-model-runner-simplest-path-for-running-models)
+- User mentions **C++ runtime** or **cmake** → skip to [C++ runtime](#c-runtime-standalone)
+- Otherwise → default to **Python package** below
+
+### Python package (default)
 ```bash
-make help              # list all targets
-make llama-cpu         # Llama
-make whisper-metal     # Whisper on Metal
-make gemma3-cuda       # Gemma3 on CUDA
+conda activate executorch
+./install_executorch.sh --editable    # editable install from source
 ```
+This handles everything: submodules, deps, C++ build, Python install. Takes ~10 min on Apple Silicon.
+
+For subsequent rebuilds (deps already present): `pip install -e . --no-build-isolation`
+
+For minimal install (skip example deps): `./install_executorch.sh --minimal`
+
+Enable additional backends:
+```bash
+CMAKE_ARGS="-DEXECUTORCH_BUILD_COREML=ON -DEXECUTORCH_BUILD_MPS=ON" ./install_executorch.sh --editable
+```
+
+Verify: `python -c "from executorch.exir import to_edge_transform_and_lower; print('OK')"`
+
+### LLM / ASR model runner (simplest path for running models)
+
+```bash
+conda activate executorch
+make <model>-<backend>
+```
+
+Available targets (run `make help` for full list):
+
+| Target | Backend | macOS | Linux |
+|--------|---------|-------|-------|
+| `llama-cpu` | CPU | yes | yes |
+| `llama-cuda` | CUDA | — | yes |
+| `llama-cuda-debug` | CUDA (debug) | — | yes |
+| `llava-cpu` | CPU | yes | yes |
+| `whisper-cpu` | CPU | yes | yes |
+| `whisper-metal` | Metal | yes | — |
+| `whisper-cuda` | CUDA | — | yes |
+| `parakeet-cpu` | CPU | yes | yes |
+| `parakeet-metal` | Metal | yes | — |
+| `parakeet-cuda` | CUDA | — | yes |
+| `voxtral-cpu` | CPU | yes | yes |
+| `voxtral-cuda` | CUDA | — | yes |
+| `voxtral-metal` | Metal | yes | — |
+| `voxtral_realtime-cpu` | CPU | yes | yes |
+| `voxtral_realtime-cuda` | CUDA | — | yes |
+| `voxtral_realtime-metal` | Metal | yes | — |
+| `gemma3-cpu` | CPU | yes | yes |
+| `gemma3-cuda` | CUDA | — | yes |
+| `sortformer-cpu` | CPU | yes | yes |
+| `sortformer-cuda` | CUDA | — | yes |
+| `silero-vad-cpu` | CPU | yes | yes |
+| `clean` | — | yes | yes |
 
 Output: `cmake-out/examples/models/<model>/<runner>`
 
-## C++ Libraries (CMake)
+### C++ runtime (standalone)
+
+**With presets (recommended):**
+
+| Platform | Command |
+|----------|---------|
+| macOS | `cmake -B cmake-out --preset macos` (uses Xcode generator — requires Xcode) |
+| Linux | `cmake -B cmake-out --preset linux -DCMAKE_BUILD_TYPE=Release` |
+| Windows | `cmake -B cmake-out --preset windows -T ClangCL` |
+
+Then: `cmake --build cmake-out --config Release -j$(sysctl -n hw.ncpu)` (macOS) or `cmake --build cmake-out -j$(nproc)` (Linux)
+
+**LLM libraries via workflow presets** (configure + build + install in one command):
+```bash
+cmake --workflow --preset llm-release        # CPU
+cmake --workflow --preset llm-release-metal  # Metal (macOS)
+cmake --workflow --preset llm-release-cuda   # CUDA (Linux/Windows)
+```
+
+**Manual CMake (custom flags):**
+```bash
+cmake -B cmake-out \
+  -DCMAKE_BUILD_TYPE=Release \
+  -DEXECUTORCH_BUILD_XNNPACK=ON \
+  -DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
+  -DEXECUTORCH_BUILD_EXTENSION_MODULE=ON \
+  -DEXECUTORCH_BUILD_EXTENSION_FLAT_TENSOR=ON \
+  -DEXECUTORCH_BUILD_EXTENSION_NAMED_DATA_MAP=ON \
+  -DEXECUTORCH_BUILD_EXTENSION_DATA_LOADER=ON \
+  -DEXECUTORCH_BUILD_EXTENSION_TENSOR=ON
+cmake --build cmake-out --parallel "$(nproc 2>/dev/null || sysctl -n hw.ncpu)"
+```
+
+Run `cmake --list-presets` to see all available presets.
+
+### Cross-compilation
+
+**iOS/macOS frameworks:**
+```bash
+./scripts/build_apple_frameworks.sh --coreml --mps --xnnpack
+```
+Link in Xcode with `-all_load` linker flag.
+
+**Android:**
+
+Requires `ANDROID_NDK` on PATH (typically set by Android Studio or standalone NDK install).
 ```bash
-cmake --list-presets                    # list presets
-cmake --workflow --preset llm-release   # LLM CPU
-cmake --workflow --preset llm-release-metal  # LLM Metal
+# Verify NDK is available
+echo $ANDROID_NDK           # must point to NDK root, e.g. ~/Library/Android/sdk/ndk/<version>
+export ANDROID_ABIS=arm64-v8a BUILD_AAR_DIR=aar-out
+mkdir -p $BUILD_AAR_DIR && sh scripts/build_android_library.sh
 ```
+
+## Key build options
+
+Most commonly needed flags (full list: `CMakeLists.txt`):
+
+| Flag | What it enables |
+|------|-----------------|
+| `EXECUTORCH_BUILD_XNNPACK` | XNNPACK CPU backend |
+| `EXECUTORCH_BUILD_COREML` | Core ML (macOS/iOS) |
+| `EXECUTORCH_BUILD_MPS` | MPS GPU (macOS/iOS) |
+| `EXECUTORCH_BUILD_METAL` | Metal compute (macOS, requires EXTENSION_TENSOR) |
+| `EXECUTORCH_BUILD_CUDA` | CUDA GPU (Linux/Windows, requires EXTENSION_TENSOR) |
+| `EXECUTORCH_BUILD_KERNELS_OPTIMIZED` | Optimized kernels |
+| `EXECUTORCH_BUILD_KERNELS_QUANTIZED` | Quantized kernels |
+| `EXECUTORCH_BUILD_EXTENSION_MODULE` | Module extension (requires DATA_LOADER + FLAT_TENSOR + NAMED_DATA_MAP) |
+| `EXECUTORCH_BUILD_EXTENSION_LLM` | LLM extension |
+| `EXECUTORCH_BUILD_TESTS` | Unit tests (`ctest --test-dir cmake-out --output-on-failure`) |
+| `EXECUTORCH_BUILD_DEVTOOLS` | DevTools (Inspector, ETDump) |
+| `EXECUTORCH_OPTIMIZE_SIZE` | Size-optimized build (`-Os`, no exceptions/RTTI) |
+| `CMAKE_BUILD_TYPE` | `Release` or `Debug` (5-10x slower). Some presets (e.g. `llm-release`) set this; others require it explicitly. |
+
+## Troubleshooting
+
+| Symptom | Fix |
+|---------|-----|
+| Missing headers / `CMakeLists.txt not found` in third-party | `git submodule sync --recursive && git submodule update --init --recursive` |
+| Mysterious failures after `git pull` or branch switch | `rm -rf cmake-out/ pip-out/ && git submodule sync && git submodule update --init --recursive` |
+| `conda env list` PermissionError | Use `CONDA_NO_PLUGINS=true conda env list` or check env dir directly |
+| CMake >= 4.0 | Works in practice despite `< 4.0` in docs; only fix if build actually fails |
+| `externally-managed-environment` / PEP 668 error | You're using system Python, not conda. Activate conda env first. |
+| pip conflicts with torch versions | Fresh conda env; or `./install_executorch.sh --use-pt-pinned-commit` |
+| Missing `Python.h` (Linux) | `sudo apt install python3.X-dev` |
+| Missing operator registrations at runtime | Link kernel libs with `-Wl,-force_load,<lib>` (macOS) or `-Wl,--whole-archive <lib> -Wl,--no-whole-archive` (Linux) |
+| `install_executorch.sh` fails on Intel Mac | No prebuilt PyTorch wheels; use `--use-pt-pinned-commit --minimal` |
+| XNNPACK build errors about cpuinfo/pthreadpool | Ensure `EXECUTORCH_BUILD_CPUINFO=ON` and `EXECUTORCH_BUILD_PTHREADPOOL=ON` (both ON by default) |
+| Duplicate kernel registration abort | Only link one `gen_operators_lib` per target |
+
+## Build output
+
+**From `./install_executorch.sh` (Python package):**
+
+| Artifact | Location |
+|----------|----------|
+| Python package | `site-packages/executorch` |
+
+**From CMake builds** (`cmake --install` with `CMAKE_INSTALL_PREFIX=cmake-out`):
+
+| Artifact | Location |
+|----------|----------|
+| Core runtime | `cmake-out/lib/libexecutorch.a` |
+| XNNPACK backend | `cmake-out/lib/libxnnpack_backend.a` |
+| executor_runner | `cmake-out/executor_runner` (Ninja/Make) or `cmake-out/Release/executor_runner` (Xcode) |
+| Model runners | `cmake-out/examples/models/<model>/<runner>` |
+
+**From cross-compilation:**
+
+| Artifact | Location |
+|----------|----------|
+| iOS frameworks | `cmake-out/*.xcframework` |
+| Android AAR | `aar-out/` |
+
+## Tips
+- Always use `Release` for benchmarking; `Debug` is 5–10x slower
+- `ccache` is auto-detected if installed (`brew install ccache`)
+- `Ninja` is faster than Make (`-G Ninja`) — but `--preset macos` uses Xcode generator
+- For LLM workflows, `make <model>-<backend>` is the simplest path
+- After `git pull`, clean and re-init submodules before rebuilding

.github/workflows/pull.yml

@@ -1057,7 +1057,8 @@ jobs:
 
   test-samsung-quantmodels-linux:
     name: test-samsung-quantmodels-linux
-    # if: github.event.pull_request.head.repo.full_name == github.repository || github.event_name != 'pull_request'
+    # Skip this job if the pull request is from a fork (secrets are not available)
+    if: github.event.pull_request.head.repo.full_name == github.repository || github.event_name != 'pull_request'
     uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@main
     permissions:
       id-token: write
@@ -1094,7 +1095,8 @@ jobs:
 
   test-samsung-models-linux:
     name: test-samsung-models-linux
-    # if: github.event.pull_request.head.repo.full_name == github.repository || github.event_name != 'pull_request'
+    # Skip this job if the pull request is from a fork (secrets are not available)
+    if: github.event.pull_request.head.repo.full_name == github.repository || github.event_name != 'pull_request'
     uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@main
     permissions:
       id-token: write

backends/arm/CMakeLists.txt

@@ -29,7 +29,7 @@ set(ETHOSU_LINUX_DRIVER_GIT_REPO
     CACHE STRING "Git repository that hosts the Ethos-U Linux driver stack"
 )
 set(ETHOSU_LINUX_DRIVER_GIT_TAG
-    "25.11"
+    "26.02"
     CACHE STRING
           "Git tag/branch/commit used to fetch the Ethos-U Linux driver stack"
 )

backends/arm/MODELS.md

@@ -1,4 +1,5 @@
-# The following file contains all models that have been confirmed to be functional and tested for the Arm backend :
+<!-- Copyright 2025-2026 Arm Limited and/or its affiliates. -->
+# The following file contains all models that have been confirmed to be functional and tested for the Arm backend:
 - Conformer
 - Deit Tiny
 - DeepLab v3 (DL3)
@@ -12,6 +13,7 @@
 - Some popular torch.nn.modules models (NN modules)
 - Some popular torch ops (Torch Functions)
 - Neural Super Sampler (NSS)
+- Phi-3
 - ResNet 18
 - Wav2Letter (W2L)
 - Stable Diffusion:

backends/arm/_passes/__init__.py

@@ -135,6 +135,8 @@
 from .rewrite_index_put_pass import RewriteIndexPutPass  # noqa
 from .rewrite_le_lt_to_ge_gt_pass import RewriteLeLtToGeGtPass  # noqa
 from .rewrite_matmul import RewriteMatmulPass  # noqa
+from .rewrite_pad import RewritePadPass  # noqa
+from .rewrite_slice import RewriteSlicePass  # noqa
 from .rewrite_upsample import RewriteUpsamplePass  # noqa
 from .scalars_to_attribute_pass import ScalarsToAttributePass  # noqa
 from .size_adjust_input_pass import SizeAdjustInputPass  # noqa

backends/arm/_passes/arm_pass.py

@@ -107,7 +107,7 @@ def call_submodule(
         return result
 
     def call_shape_operator(
-        self, op, args: tuple, kwargs: dict, meta: NodeMetadata, update: bool
+        self, op, args: tuple, kwargs: dict, meta: NodeMetadata, updated: bool = True
     ) -> ProxyValue:
         """Call operator for shape-producing operators.
 
@@ -123,4 +123,4 @@ def call_shape_operator(
         shape_meta.data = dict(meta.data)
         shape_meta.data[TosaSpecialDtype.meta_key()] = TosaSpecialDtype.SHAPE
         # Call the super (ArmPass) call operator with updated meta
-        return self.call_operator(op, args, kwargs, shape_meta, update)
+        return self.call_operator(op, args, kwargs, shape_meta, updated)

backends/arm/_passes/arm_pass_manager.py

@@ -120,6 +120,8 @@
     RewriteIndexPutPass,
     RewriteLeLtToGeGtPass,
     RewriteMatmulPass,
+    RewritePadPass,
+    RewriteSlicePass,
     RewriteUpsamplePass,
     ScalarsToAttributePass,
     SizeAdjustInputPass,
@@ -372,6 +374,8 @@ def _tosa_pipeline(
                 RewriteUpsamplePass(),
                 RewriteConvPass(exported_program),
                 RewriteMatmulPass(),
+                RewritePadPass(),
+                RewriteSlicePass(),
             ]
         )