Remove error anti-pattern when dealing with dynamic shape output (#12…

…1681)

Summary:
There are cases where capture_dynamic_output_shape_ops=True and we will still see DynamicOutputShapeException. For example, when an op doesn't have a meta kernel implemented to return the correct dynamic shape output. If we blindly give users instructions to set capture_dynamic_output_shape_ops to True, users would try it and see no change. As witnessed in this issue:
pytorch/pytorch#121036 (comment)

X-link: pytorch/pytorch#121681
Approved by: https://github.com/tugsbayasgalan

Reviewed By: osalpekar

Differential Revision: D54919382

Pulled By: gmagogsfm

fbshipit-source-id: 9d022c13a22a4201f26afec844a822467a63d71e

userbenchmark/dynamo/dynamobench/_dynamo/utils.py

@@ -1684,10 +1684,17 @@ def get_fake_value(node, tx, allow_non_graph_fake=False):
         elif isinstance(
             cause, torch._subclasses.fake_tensor.DynamicOutputShapeException
         ):
-            unimplemented(
-                f"dynamic shape operator: {cause.func}; "
-                "to enable, set torch._dynamo.config.capture_dynamic_output_shape_ops = True"
-            )
+            if not torch._dynamo.config.capture_dynamic_output_shape_ops:
+                unimplemented(
+                    f"dynamic shape operator: {cause.func}; "
+                    "to enable, set torch._dynamo.config.capture_dynamic_output_shape_ops = True"
+                )
+            else:
+                unimplemented(
+                    f"dynamic shape operator: {cause.func}; "
+                    "Operator does not have a meta kernel that supports dynamic output shapes, "
+                    "please report an issue to PyTorch"
+                )
         elif isinstance(
             cause, torch._subclasses.fake_tensor.UnsupportedOperatorException
         ):