Merge branch 'main' into export-D85064213

Author: zonglinpeng

.gitignore

@@ -62,7 +62,6 @@ xcuserdata/
 /include/
 /share/
 /version.py
-*.csv
 *_etdump
 
 # Android

CONTRIBUTING.md

@@ -34,7 +34,7 @@ executorch
 │   ├── <a href="backends/qualcomm">qualcomm</a> - Qualcomm-specific backends. See <a href="docs/source/backends-qualcomm.md">doc</a>.
 │   ├── <a href="backends/transforms">transforms</a> - Transformations for backend optimization.
 │   ├── <a href="backends/vulkan">vulkan</a> - Vulkan backend for cross-platform GPU support. See <a href="docs/source/backends-vulkan.md">doc</a>.
-│   └── <a href="backends/xnnpack">xnnpack</a> - XNNPACK backend for optimized neural network operations. See <a href="docs/source/backends-xnnpack.md">doc</a>.
+│   └── <a href="backends/xnnpack">xnnpack</a> - XNNPACK backend for optimized neural network operations. See <a href="docs/source/backends/xnnpack/xnnpack-overview.md">doc</a>.
 ├── <a href="codegen">codegen</a> - Tooling to autogenerate bindings between kernels and the runtime.
 ├── <a href="configurations">configurations</a> - Configuration files.
 ├── <a href="devtools">devtools</a> - Model profiling, debugging, and inspection. Please refer to the <a href="docs/source/devtools-overview.md">tools documentation</a> for more information.

README-wheel.md

@@ -11,7 +11,7 @@ The `executorch` pip package is in beta.
 The prebuilt `executorch.runtime` module included in this package provides a way
 to run ExecuTorch `.pte` files, with some restrictions:
 * Only [core ATen operators](docs/source/ir-ops-set-definition.md) are linked into the prebuilt module
-* Only the [XNNPACK backend delegate](docs/source/backends-xnnpack.md) is linked into the prebuilt module.
+* Only the [XNNPACK backend delegate](docs/source/backends/xnnpack/xnnpack-overview.md) is linked into the prebuilt module.
 * \[macOS only] [Core ML](docs/source/backends/coreml/coreml-overview.md) and [MPS](docs/source/backends/mps/mps-overview.md) backend
   are also linked into the prebuilt module.
 

backends/cadence/aot/ops_registrations.py

@@ -53,7 +53,6 @@ def _validate_ref_impl_exists() -> None:
     # 1. be removed
     # 2. have a reference implementation added to ref_implementations.py
     _WARN_ONLY = {
-        "cadence::quantized_w8a32_linear",
         "cadence::quantized_add",  # We should only support per_tensor variant, should remove
         "cadence::_softmax_f32_f32",
         "cadence::requantize",  # We should only support per_tensor variant, should remove
@@ -2706,6 +2705,9 @@ def quantized_w8a32_linear_meta(
     # output comes in empty with shape [leading_dims, out_dim]
     src_shape = list(src.shape)
     weight_shape = weight.shape
+    assert (src_shape[-1] % 4) == 0
+    if len(src_shape) >= 2:
+        assert src_shape[-2] == 1
     assert len(weight_shape) == 2
     assert src_shape[-1] == weight_shape[-1]
     src_shape[-1] = weight_shape[0]
@@ -2720,12 +2722,12 @@ def quantized_w8a32_conv_meta(
     bias: torch.Tensor,
     b_scale: float,
 ) -> torch.Tensor:
-    # src comes in shape [batch, in_channel, in_length]
-    # weight comes in shape [out_ch, in_ch, kernel_dim]
+    # src comes in shape [batch, in_length, in_channels]
+    # weight comes in shape [kernel_dim, out_ch, in_ch]
     # output comes in empty with shape [batch, out_ch, in_length - kernel_dim + 1]
     assert len(src.shape) == 3
 
-    out_channels, in_channels, kernel_size = weight.shape
+    kernel_size, out_channels, in_channels = weight.shape
     assert kernel_size == 3
     assert (out_channels % 4) == 0
     assert (in_channels % 4) == 0

backends/cadence/aot/quantizer/fusion_pass.py

@@ -397,7 +397,7 @@ def get_args_and_kwargs_mixed_w8a32_conv(
     )
     transposed_weights = graph_module.graph.call_function(
         torch.ops.aten.permute.default,
-        (weights_inputs[0], [2, 0, 1]),  # NCL -> NLC
+        (weights_inputs[0], [2, 0, 1]),  # NCL -> LNC
     )
 
     args = (

backends/cadence/aot/ref_implementations.py

@@ -854,18 +854,23 @@ def quantized_w8a32_conv(
     if len(weight.shape) != 3:
         raise ValueError("Weight tensor must be 3D")
 
-    out_channels, in_channels, kernel_size = weight.shape
+    kernel_size, out_channels, in_channels = weight.shape
     if kernel_size != 3:
         raise ValueError("Kernel size must be 3")
     if (out_channels % 4) != 0:
         raise ValueError("Out channels must be a multiple of 4")
     if (in_channels % 4) != 0:
         raise ValueError("In channels must be a multiple of 4")
 
-    # src comes in shape [batch, in_channel, in_length]
-    # weight comes in shape [out_ch, in_ch, kernel_dim]
-    # output comes in empty with shape [batch, out_ch, in_length - kernel_dim + 1]
-    # Dequantize weight using scale
+    assert weight.dtype == torch.int8
+    assert bias.dtype == torch.int8
+
+    # To make compliant with torch (LCN -> NCL format)
+    weight = weight.permute(1, 2, 0).contiguous()
+
+    # channels last to channels first
+    src = src.permute(0, 2, 1).contiguous()
+
     dequant_weight = weight.float() * w_scale
 
     # Dequantize bias using scale
@@ -884,6 +889,36 @@ def quantized_w8a32_conv(
     return output
 
 
+@impl_tracked(m, "quantized_w8a32_linear")
+def quantized_w8a32_linear(
+    src: torch.Tensor,
+    weight: torch.Tensor,
+    w_scale: float,
+    bias: torch.Tensor,
+    b_scale: float,
+) -> torch.Tensor:
+    # src comes in shape [leading_dims, in_dim]
+    # weight comes in shape [in_dim, out_dim]
+    # output comes in empty with shape [leading_dims, out_dim]
+    assert weight.dtype == torch.int8
+    assert bias.dtype == torch.int8
+    if len(src.shape) >= 2:
+        assert src.shape[-2] == 1, "Only supporting vector-matrix multiplication"
+
+    # need to transpose to make compliant with torch linear (in, out -> out, in)
+    weight = weight.transpose(1, 0).contiguous()
+    dequant_weight = weight.float() * w_scale
+    dequant_bias = bias.float() * b_scale
+
+    output = torch.nn.functional.linear(
+        src.float(),
+        dequant_weight,
+        dequant_bias,
+    )
+
+    return output
+
+
 @impl_tracked(m, "quantized_conv2d_nhwc.per_tensor")
 def quantized_conv2d_nhwc_per_tensor(
     input_tensor: torch.Tensor,

backends/cadence/aot/tests/test_ref_implementations.py

@@ -1188,7 +1188,7 @@ def test_quantized_conv_per_tensor(
                     dtype=torch.int8,
                 ),  # weight: 4x4x3
                 0.5,  # w_scale
-                torch.tensor([2, 2, 2, 2], dtype=torch.float32),  # bias: 4
+                torch.tensor([2, 2, 2, 2], dtype=torch.int8),  # bias: 4
                 1.0,  # b_scale
                 torch.tensor(
                     [
@@ -1214,6 +1214,12 @@ def test_quantized_w8a32_conv(
         b_scale: float,
         expected_output: torch.Tensor,
     ) -> None:
+
+        # This op takes in channels last src
+        src = src.permute(0, 2, 1)
+
+        # This op takes in LNC format for weights
+        weight = weight.permute(2, 0, 1)
         output = torch.ops.cadence.quantized_w8a32_conv(
             src, weight, w_scale, bias, b_scale
         )
@@ -1236,6 +1242,95 @@ def test_quantized_w8a32_conv(
             f"Output values don't match expected in {name}. Got {output}, expected {expected_output}",
         )
 
+    @expand(
+        [
+            (
+                "multi_input_features",
+                torch.tensor([[1.0, 2.0, 3.0]], dtype=torch.float32),  # src: 1x3
+                torch.tensor([[2, 1], [1, 2], [1, 1]], dtype=torch.int8),  # weight: 3x2
+                0.5,  # w_scale
+                torch.tensor([0, 1], dtype=torch.int8),  # bias: 2
+                1.0,  # b_scale
+                torch.tensor([[3.5, 5.0]], dtype=torch.float32),  # expected
+            ),
+            (
+                "batch_size_2",
+                torch.tensor(
+                    [[[1.0, 2.0]], [[3.0, 4.0]]], dtype=torch.float32
+                ),  # src: 2x2
+                torch.tensor([[1, 2], [1, -1]], dtype=torch.int8),  # weight: 2x2
+                1.0,  # w_scale
+                torch.tensor([0, 0], dtype=torch.int8),  # bias: 2
+                1.0,  # b_scale
+                torch.tensor(
+                    [[[3.0, 0.0]], [[7.0, 2.0]]], dtype=torch.float32
+                ),  # expected
+            ),
+            (
+                "shape_assertion_error",
+                torch.tensor(
+                    [[[1.0, 2.0], [3.0, 4.0]]], dtype=torch.float32
+                ),  # src: 1x2x2
+                torch.tensor([[1, 2], [1, -1]], dtype=torch.int8),  # weight: 2x2
+                1.0,  # w_scale
+                torch.tensor([0, 1], dtype=torch.int8),  # bias: 2
+                1.0,  # b_scale
+                torch.tensor(
+                    [[[3.0, 1.0], [7.0, 3.0]]], dtype=torch.float32
+                ),  # expected
+            ),
+            (
+                "negative_weights",
+                torch.tensor([[2.0, 4.0]], dtype=torch.float32),  # src: 1x2
+                torch.tensor([[-2, -3], [-1, -2]], dtype=torch.int8),  # weight: 2x2
+                0.5,  # w_scale
+                torch.tensor([2, 1], dtype=torch.int8),  # bias: 2
+                1.0,  # b_scale
+                torch.tensor([[-2.0, -6.0]], dtype=torch.float32),  # expected
+            ),
+        ]
+    )
+    def test_quantized_w8a32_linear(
+        self,
+        name: str,
+        src: torch.Tensor,
+        weight: torch.Tensor,
+        w_scale: float,
+        bias: torch.Tensor,
+        b_scale: float,
+        expected_output: torch.Tensor,
+    ) -> None:
+        if name == "shape_assertion_error":
+            with self.assertRaisesRegex(
+                AssertionError, "Only supporting vector-matrix multiplication"
+            ):
+                torch.ops.cadence.quantized_w8a32_linear(
+                    src, weight, w_scale, bias, b_scale
+                )
+            return
+
+        output = torch.ops.cadence.quantized_w8a32_linear(
+            src, weight, w_scale, bias, b_scale
+        )
+
+        # Verify output properties
+        self.assertEqual(
+            output.dtype,
+            torch.float32,
+            f"Output dtype should be float32 in {name}",
+        )
+        self.assertEqual(
+            output.shape,
+            expected_output.shape,
+            f"Output shape should match expected shape in {name}",
+        )
+
+        # Verify output matches expected values
+        self.assertTrue(
+            torch.allclose(output, expected_output, rtol=1e-4, atol=1e-4),
+            f"Output values don't match expected in {name}. Got {output}, expected {expected_output}",
+        )
+
     @expand(
         [
             # Test case 1: Basic int8 case with negative scale

backends/xnnpack/README.md

@@ -134,4 +134,4 @@ create an issue on [github](https://www.github.com/pytorch/executorch/issues).
 ## See Also
 For more information about the XNNPACK Backend, please check out the following resources:
 - [XNNPACK Backend](https://pytorch.org/executorch/main/backends-xnnpack)
-- [XNNPACK Backend Internals](https://pytorch.org/executorch/main/backend-delegates-xnnpack-reference)
+- [XNNPACK Backend Internals](https://pytorch.org/executorch/main/backends/xnnpack/backend-delegates-xnnpack-reference)

docs/source/android-backends.md

@@ -16,7 +16,7 @@ Available hardware acceleration backends for Android deployment.
 - {doc}`android-qualcomm` — Qualcomm AI Engine (NPU)
 - {doc}`android-mediatek` — MediaTek NPU acceleration
 - {doc}`android-arm-vgf` — ARM VGF Backend
-- {doc}`android-samsung-exynos` — Samsung Exynos NPU
+- {doc}`backends/samsung/samsung-overview` — Samsung Exynos NPU
 
 ```{toctree}
 :hidden:
@@ -25,4 +25,4 @@ android-vulkan
 android-qualcomm
 android-mediatek
 android-arm-vgf
-android-samsung-exynos
+backends/samsung/samsung-overview

docs/source/android-xnnpack.md

@@ -1 +1 @@
-```{include} backends-xnnpack.md
+```{include} backends/xnnpack/xnnpack-overview.md