resolve bug (#6781)

pytorch_lightning/plugins/training_type/tpu_spawn.py

@@ -118,11 +118,13 @@ def new_process(self, process_idx: int, trainer, mp_queue) -> None:
         self.__save_end_of_training_weights(self.lightning_module)
         self.transfer_distrib_spawn_state_on_fit_end(results)
 
+        # https://github.com/pytorch/xla/issues/1801#issuecomment-602799542
+        self.barrier("end-process")
+
+        # https://github.com/pytorch/xla/issues/2190#issuecomment-641665358
         if self.global_rank == 0:
             time.sleep(2)
 
-        self.barrier("end-process")
-
     def __save_end_of_training_weights(self, model: LightningModule) -> None:
         # when training ends on these platforms dump weights to get out of the main process
         if on_colab_kaggle():
@@ -158,16 +160,7 @@ def transfer_distrib_spawn_state_on_fit_end(self, results):
                 self.mp_queue.put(results)
 
     def save(self, state_dict: Dict, path: str) -> None:
-        """
-        Saving with ``xm.save`` can be unstable and miss the rendez-vous after ``torch.save``.
-        The rendez-vous doesn't affect directly saving.
-        We can ignore the ``RuntimeError`` to reduce friction with TPUs.
-        """
-        try:
-            xm.save(state_dict, path)
-        except RuntimeError as e:
-            if "Failed to meet rendezvous" not in str(e):
-                raise e
+        xm.save(state_dict, path)
 
     def broadcast(self, obj: object, src: int = 0) -> object:
         buffer = io.BytesIO()