From 3b5064d67769ae79f0edebe5dad89a0221c086a0 Mon Sep 17 00:00:00 2001
From: caozhou <48191911+Caozhou1995@users.noreply.github.com>
Date: Tue, 11 Oct 2022 16:42:15 +0800
Subject: [PATCH] update instantiate for auto parallel (#46883)

---
 python/paddle/distributed/auto_parallel/process_group.py | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/python/paddle/distributed/auto_parallel/process_group.py b/python/paddle/distributed/auto_parallel/process_group.py
index 10ff0d36fce..ba63a716434 100644
--- a/python/paddle/distributed/auto_parallel/process_group.py
+++ b/python/paddle/distributed/auto_parallel/process_group.py
@@ -21,7 +21,7 @@ from ..collective import _get_global_env
 from ..collective import _new_ring_id
 from ...fluid.framework import _non_static_mode
 from ...fluid.layers.tensor import fill_constant
-from paddle.fluid.framework import _enable_legacy_dygraph
+from paddle import _legacy_C_ops
 
 
 def get_all_process_groups():
@@ -145,14 +145,15 @@ class ProcessGroup:
             # TODO(shenliang03): This is a temporary solution to solve the problem of
             # hang caused by cross-creation of new_group
             paddle.disable_static()
-            _enable_legacy_dygraph()
             paddle.set_device('gpu:%d' %
                               paddle.distributed.ParallelEnv().dev_id)
             tmp = paddle.to_tensor(
                 [1], dtype="int32") if _non_static_mode() else fill_constant(
                     [0], dtype="int32", value="1")
-            paddle.distributed.all_reduce(tmp, sync_op=True, group=self)
-            paddle.distributed.wait(tmp, group=self)
+            # use legacy ops
+            _legacy_C_ops.c_allreduce_sum_(tmp, 'use_calc_stream', True,
+                                           'ring_id', self.id)
+            _legacy_C_ops.c_sync_calc_stream(tmp, tmp)
             paddle.enable_static()
 
         self._is_instantiate = True
-- 
GitLab