Fix Conv1d w8a32 operator (pytorch#16607)

Summary:

#### Summary

This diff fixes the Conv1d w8a32 operator by adding a transformation to the `val` attribute of the `other_inputs[0].meta` dictionary. Specifically, the `permute` operation is applied to the `original_val` tensor with the `fake_mode` context, and the resulting `transposed_val` is assigned to `transposed_inputs.meta["val"]`.

Reviewed By: mcremon-meta

Differential Revision: D89863750

Author: Marco Giordano

backends/cadence/aot/quantizer/fusion_pass.py

@@ -438,26 +438,36 @@ def get_args_and_kwargs_mixed_w8a32_conv(
         torch.ops.aten.permute.default,
         (other_inputs[0], [0, 2, 1]),  # NCL -> NLC
     )
-    assert "val" in other_inputs[0].meta, "Missing val metadata on input node"
-    original_val = other_inputs[0].meta["val"]
-    assert original_val.fake_mode is not None, "fake_mode is None on input node"
-    with original_val.fake_mode:
-        transposed_inputs.meta["val"] = torch.ops.aten.permute.default(
-            original_val, [0, 2, 1]
-        )
+    # Propagate val metadata for transposed_inputs
+    if "val" in other_inputs[0].meta:
+        original_val = other_inputs[0].meta["val"]
+        fake_mode = original_val.fake_mode
+        if fake_mode is not None:
+            with fake_mode:
+                transposed_val = torch.ops.aten.permute.default(original_val, [0, 2, 1])
+            transposed_inputs.meta["val"] = transposed_val
+        else:
+            transposed_inputs.meta["val"] = torch.ops.aten.permute.default(
+                original_val, [0, 2, 1]
+            )
     copy_node_metadata(transposed_inputs, other_inputs[0])
 
     transposed_weights = graph_module.graph.call_function(
         torch.ops.aten.permute.default,
         (weights_inputs[0], [2, 0, 1]),  # NCL -> LNC
     )
-    assert "val" in weights_inputs[0].meta, "Missing val metadata on weight node"
-    original_val = weights_inputs[0].meta["val"]
-    assert original_val.fake_mode is not None, "fake_mode is None on weight node"
-    with original_val.fake_mode:
-        transposed_weights.meta["val"] = torch.ops.aten.permute.default(
-            original_val, [2, 0, 1]
-        )
+    # Propagate val metadata for transposed_weights
+    if "val" in weights_inputs[0].meta:
+        original_val = weights_inputs[0].meta["val"]
+        fake_mode = original_val.fake_mode
+        if fake_mode is not None:
+            with fake_mode:
+                transposed_val = torch.ops.aten.permute.default(original_val, [2, 0, 1])
+            transposed_weights.meta["val"] = transposed_val
+        else:
+            transposed_weights.meta["val"] = torch.ops.aten.permute.default(
+                original_val, [2, 0, 1]
+            )
     copy_node_metadata(transposed_weights, weights_inputs[0])
 
     args = (

backends/cadence/aot/quantizer/patterns.py

@@ -744,6 +744,18 @@ def get_anchors(
                     conv_layer,
                 )
 
+            inputs = conv_layer.args[0]
+            if "tensor_meta" in inputs.meta:
+                inputs_shape = inputs.meta["tensor_meta"].shape
+                # Bail if length != kernel size - Not yet supported
+                if inputs_shape[-1] != cnn_weights_shape[2]:
+                    return (
+                        PartitionAnchors(
+                            empty=True,
+                        ),
+                        conv_layer,
+                    )
+
         return (
             PartitionAnchors(
                 inputs=[],