Clean up some code

torch_xla/core/dynamo_bridge.py

@@ -171,6 +171,15 @@ def _maybe_move_tensors_to_device(tensors: tuple,
   return tuple(moved_tensors)
 
 
+# Given a list of args, returns a tuple of all the args' shapes.
+def _get_arg_input_shapes(args):
+  arg_input_shapes = []
+  for arg in args:
+    if isinstance(arg, torch.Tensor):
+      arg_input_shapes.append(tuple(arg.shape))
+  return tuple(arg_input_shapes)
+
+
 class Deduper:
 
   def __init__(self):
@@ -445,6 +454,19 @@ def extract_internal(xla_model: torch.fx.GraphModule):
   (xla_args_sharding_spec, args_and_out, graph_hash,
    arg_index_to_need_update_index, none_remover, graph_input_matcher,
    dumb_return_handler, xla_args_need_update) = extract_graph_helper(xla_model)
+
+  # If dynamic=True, we cache the result to avoid recompilation.
+  if not torch._dynamo.config.assume_static_by_default:
+    arg_input_shapes = _get_arg_input_shapes(xla_model.xla_args)
+    print(
+        f'[WONJOO] extract_internal before optimized_mod, {arg_input_shapes=}')
+    input_shape_mappings[arg_input_shapes] = (xla_args_sharding_spec,
+                                              args_and_out, graph_hash,
+                                              arg_index_to_need_update_index,
+                                              none_remover, graph_input_matcher,
+                                              dumb_return_handler,
+                                              xla_args_need_update)
+
   skip_checking_input_sharding_threashold = xu.getenv_as(
       'XLA_DYNAMO_INPUT_SHARDING_CHECK_THRESHOLD', int, 5)
 
@@ -463,34 +485,15 @@ def optimized_mod(*args: tuple):
 
     # See [Note: Dynamo real-time input-shape cache look-up] above.
     if not torch._dynamo.config.assume_static_by_default:
-      # TODO figure out why we need this mark_step or sync here.
-      # Without it, the results are off slightly.
-      xm.mark_step()
-      # input_tensors_to_sync = [
-      #     args[i] for i, x in enumerate(
-      #         torch_xla._XLAC._check_tensor_need_materialization(
-      #             [a for a in args if isinstance(a, torch.Tensor)])) if x
-      # ]
-      # if len(input_tensors_to_sync) > 0:
-      #   torch_xla._XLAC._xla_increment_counter('DynamoSyncInputExecuteTime', 1)
-      #   torch_xla._XLAC._xla_sync_multi(
-      #       input_tensors_to_sync, devices=[], wait=True, sync_xla_data=True)
-
-      arg_input_shapes = []
-      # When dynamic=True in torch.compile call, TorchDynamo will directly
-      # call optimized_mod without compiling. Hence, xla_model's xla_args will
-      # be different from optimized_mod's args. So we manually set them here.
       xla_model.xla_args = args
-      for arg in args:
-        if isinstance(arg, torch.Tensor):
-          arg_input_shapes.append(tuple(arg.shape))
-      arg_input_shapes = tuple(arg_input_shapes)
+      arg_input_shapes = _get_arg_input_shapes(xla_model.xla_args)
       if arg_input_shapes in input_shape_mappings:
         (xla_args_sharding_spec, args_and_out, graph_hash,
          arg_index_to_need_update_index, none_remover, graph_input_matcher,
          dumb_return_handler,
          xla_args_need_update) = input_shape_mappings[arg_input_shapes]
       else:
+        xm.mark_step()
         (xla_args_sharding_spec, args_and_out, graph_hash,
          arg_index_to_need_update_index, none_remover, graph_input_matcher,
          dumb_return_handler,