Skip to content

Commit 54598a9

Browse files
GregoryComerGregoryComer
authored
Merge branch 'main' into reapply-thread-defaults
2 parents 7f0723e + d8a2126 commit 54598a9

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

63 files changed

+1298
-243
lines changed

‎.github/workflows/build-presets.yml‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -109,7 +109,7 @@ jobs:
109109
strategy:
110110
fail-fast: false
111111
matrix:
112-
preset: [pybind, windows, llm]
112+
preset: [pybind, windows]
113113
with:
114114
job-name: build
115115
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}

‎backends/cadence/aot/ops_registrations.py‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -329,7 +329,7 @@
329329
"Tensor bias_scale, float out_scale, int out_zero_point, Tensor out_multiplier, Tensor out_shift, bool channel_last=False) -> (Tensor out)"
330330
)
331331
lib.define(
332-
"avg_pool2d(Tensor input, int[2] kernel_size, int[2] stride=[], int[2] padding=0, bool ceil_mode=False, "
332+
"avg_pool2d(Tensor input, int[2] kernel_size, int[2] stride=[], int[2] padding=[], bool ceil_mode=False, "
333333
"bool count_include_pad=True, int? divisor_override=None, Tensor? in_zero_point=None, bool channel_last=False) -> (Tensor out)"
334334
)
335335
lib.define(
@@ -525,7 +525,7 @@
525525
"Tensor out_multiplier, Tensor out_shift, bool channel_last=False, *, Tensor(a!) out) -> Tensor(a!)"
526526
)
527527
lib.define(
528-
"avg_pool2d.out(Tensor input, int[2] kernel_size, int[2] stride=[], int[2] padding=0, "
528+
"avg_pool2d.out(Tensor input, int[2] kernel_size, int[2] stride=[], int[2] padding=[], "
529529
"bool ceil_mode=False, bool count_include_pad=True, int? divisor_override=None, "
530530
"Tensor? in_zero_point=None, bool channel_last=False, *, Tensor(a!) out) -> Tensor(a!)"
531531
)

‎docs/source/advanced-topics-section.md‎

Lines changed: 112 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,112 @@
1+
(advanced-topics-section)=
2+
3+
# Advanced
4+
5+
Deep dive into ExecuTorch's advanced features for optimization, customization, and integration.
6+
7+
This section covers advanced concepts for developers who need to customize ExecuTorch for specific use cases, optimize performance, or integrate with custom hardware backends.
8+
9+
## Quantization & Optimization
10+
11+
Techniques for model compression and performance optimization.
12+
13+
**→ {doc}`quantization-optimization` — Quantization strategies and performance optimization**
14+
15+
Key topics:
16+
17+
- Quantization strategies and techniques
18+
- Performance profiling and optimization
19+
20+
## Model Export
21+
22+
Learn the core ExecuTorch workflow, exporting PyTorch models to the `.pte` format for edge deployment.
23+
24+
**→ {doc}`using-executorch-export`** - Model Export & Lowering
25+
26+
Key topics:
27+
28+
- Export and Lowering Workflow
29+
- Hardware Backend Selection & Optimization
30+
- Dynamic Shapes & Advanced Model Features
31+
32+
33+
## Kernel Library
34+
35+
Deep dive into ExecuTorch's kernel implementation and customization.
36+
37+
**→ {doc}`kernel-library-advanced` — Kernel library deep dive and customization**
38+
39+
Key topics:
40+
41+
- Kernel library architecture
42+
- Custom kernel implementation
43+
- Selective build and optimization
44+
45+
## Backend & Delegates
46+
47+
**→ {doc}`backend-delegate-advanced` — Backend delegate integration**
48+
49+
Key topics:
50+
51+
- Learn how to integrate Backend Delegate into ExecuTorch and more
52+
- XNNPACK Delegate Internals
53+
- Debugging Delegation
54+
55+
56+
## Runtime & Integration
57+
58+
Advanced runtime features and backend integration.
59+
60+
**→ {doc}`runtime-integration-advanced` — Runtime customization and backend integration**
61+
62+
Key topics:
63+
64+
- Backend delegate implementation
65+
- Platform abstraction layer
66+
- Custom runtime integration
67+
68+
## Compiler & IR
69+
70+
Advanced compiler features and intermediate representation details.
71+
72+
**→ {doc}`compiler-ir-advanced` — Compiler passes and IR specification**
73+
74+
Key topics:
75+
76+
- Custom compiler passes
77+
- Memory planning strategies
78+
- Backend dialect and EXIR
79+
- Ops set definition
80+
81+
82+
## File Formats
83+
84+
ExecuTorch file format specifications and internals.
85+
86+
**→ {doc}`file-formats-advanced` — PTE and PTD file format specifications**
87+
88+
Key topics:
89+
90+
- PTE file format internals
91+
- PTD file format specification
92+
- Custom file format handling
93+
94+
## Next Steps
95+
96+
After exploring advanced topics:
97+
98+
- **{doc}`tools-sdk-section`** - Developer tools for debugging and profiling
99+
- **{doc}`api-section`** - Complete API reference documentation
100+
101+
```{toctree}
102+
:hidden:
103+
:maxdepth: 2
104+
:caption: Advanced Topics
105+
106+
quantization-optimization
107+
using-executorch-export
108+
kernel-library-advanced
109+
backend-delegate-advanced
110+
runtime-integration-advanced
111+
compiler-ir-advanced
112+
file-formats-advanced

‎docs/source/android-arm-vgf.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
```{include} backends-arm-vgf.md

‎docs/source/android-backends.md‎

Lines changed: 28 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,28 @@
1+
(android-backends)=
2+
# Backends
3+
4+
Available hardware acceleration backends for Android deployment.
5+
6+
## CPU Acceleration
7+
8+
- {doc}`android-xnnpack` — XNNPACK CPU acceleration
9+
10+
## GPU Acceleration
11+
12+
- {doc}`android-vulkan` — Vulkan GPU acceleration
13+
14+
## NPU/Accelerator Backends
15+
16+
- {doc}`android-qualcomm` — Qualcomm AI Engine (NPU)
17+
- {doc}`android-mediatek` — MediaTek NPU acceleration
18+
- {doc}`android-arm-vgf` — ARM VGF Backend
19+
- {doc}`android-samsung-exynos` — Samsung Exynos NPU
20+
21+
```{toctree}
22+
:hidden:
23+
android-xnnpack
24+
android-vulkan
25+
android-qualcomm
26+
android-mediatek
27+
android-arm-vgf
28+
android-samsung-exynos

‎docs/source/android-examples.md‎

Lines changed: 9 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
# Examples & Demos
2+
3+
- [Working with LLMs - Android Examples](https://github.com/meta-pytorch/executorch-examples/tree/main/llm/android)
4+
- [Demo Apps](https://github.com/meta-pytorch/executorch-examples/tree/main/dl3/android/DeepLabV3Demo#executorch-android-demo-app)
5+
- {doc}`tutorial-arm-vgf` — Export a simple PyTorch model for the ExecuTorch VGF backend
6+
7+
```{toctree}
8+
:hidden:
9+
tutorial-arm-vgf

‎docs/source/android-mediatek.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
```{include} backends-mediatek.md

‎docs/source/android-qualcomm.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
```{include} backends-qualcomm.md

‎docs/source/android-samsung-exynos.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
```{include} backends-samsung-exynos.md

‎docs/source/android-section.md‎

Lines changed: 23 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,23 @@
1+
(android-section)=
2+
3+
# Android
4+
5+
Deploy ExecuTorch on Android devices with hardware acceleration support.
6+
7+
## Quick Start & Integration
8+
9+
- {doc}`using-executorch-android` — Complete Android integration guide
10+
11+
## Backends
12+
13+
- {doc}`android-backends` — Available Android backends and acceleration options
14+
15+
## Examples & Demos
16+
17+
- {doc}`android-examples` — Explore Android Examples & Demos
18+
19+
```{toctree}
20+
:hidden:
21+
using-executorch-android
22+
android-backends
23+
android-examples

0 commit comments

Comments
 (0)