fix test

Signed-off-by: Yaoyao Ding <[email protected]>

Author: yaoyaoding

tests/lang/test_copy_async_bulk.py

@@ -102,6 +102,7 @@ def __call__(self, bs: int, m_size: int32, n_size: int, x_ptr: ~float16, y_ptr:
         x = self.load_shared(s_x)
         x += self.block_rank_in_cluster + 1
         self.store_shared(s_y, x)
+        self.fence_proxy_copy_async()
         self.sync()
 
         self.copy_async_bulk_shared_to_global(