fix: Improve logging and kwarg passing in Dynamo (#2052)

Author: gs-olive

py/torch_tensorrt/dynamo/backend/__init__.py

@@ -8,7 +8,6 @@
 from torch_tensorrt import EngineCapability, Device
 from torch_tensorrt.fx.utils import LowerPrecision
 
-from torch_tensorrt.dynamo.backend._settings import CompilationSettings
 from torch_tensorrt.dynamo.backend.utils import prepare_inputs, prepare_device
 from torch_tensorrt.dynamo.backend.backends import torch_tensorrt_backend
 from torch_tensorrt.dynamo.backend._defaults import (
@@ -62,6 +61,10 @@ def compile(
 
     inputs = prepare_inputs(inputs, prepare_device(device))
 
+    if not isinstance(enabled_precisions, collections.abc.Collection):
+        enabled_precisions = [enabled_precisions]
+
+    # Parse user-specified enabled precisions
     if (
         torch.float16 in enabled_precisions
         or torch_tensorrt.dtype.half in enabled_precisions
@@ -123,19 +126,12 @@ def create_backend(
     Returns:
         Backend for torch.compile
     """
-    if debug:
-        logger.setLevel(logging.DEBUG)
-
-    settings = CompilationSettings(
+    return partial(
+        torch_tensorrt_backend,
         debug=debug,
         precision=precision,
         workspace_size=workspace_size,
         min_block_size=min_block_size,
         torch_executed_ops=torch_executed_ops,
         pass_through_build_failures=pass_through_build_failures,
     )
-
-    return partial(
-        torch_tensorrt_backend,
-        settings=settings,
-    )

py/torch_tensorrt/dynamo/backend/backends.py

@@ -77,13 +77,12 @@ def _pretraced_backend(
         )
         return trt_compiled
     except:
-        logger.error(
-            "FX2TRT conversion failed on the subgraph. See trace above. "
-            + "Returning GraphModule forward instead.",
-            exc_info=True,
-        )
-
         if not settings.pass_through_build_failures:
+            logger.warning(
+                "TRT conversion failed on the subgraph. See trace above. "
+                + "Returning GraphModule forward instead.",
+                exc_info=True,
+            )
             return gm.forward
         else:
             raise AssertionError(