diff --git a/tensorflow/python/ops/linalg_ops.py b/tensorflow/python/ops/linalg_ops.py
index 03b7b98119d..08e734e9091 100644
--- a/tensorflow/python/ops/linalg_ops.py
+++ b/tensorflow/python/ops/linalg_ops.py
@@ -24,6 +24,7 @@ from tensorflow.python.framework import dtypes
 from tensorflow.python.framework import ops
 from tensorflow.python.ops import array_ops
 from tensorflow.python.ops import control_flow_ops
+from tensorflow.python.ops import gen_array_ops
 from tensorflow.python.ops import gen_linalg_ops
 from tensorflow.python.ops import linalg_ops_impl
 from tensorflow.python.ops import map_fn
@@ -732,7 +733,8 @@ def norm(tensor,
             ops.convert_to_tensor(axis))
         axes = math_ops.range(rank)
         perm_before = array_ops.concat(
-            [array_ops.setdiff1d(axes, positive_axis)[0], positive_axis],
+            [gen_array_ops.list_diff(axes, positive_axis, dtypes.int32)[0],
+             positive_axis],
             axis=0)
         perm_after = map_fn.map_fn(
             lambda i: math_ops.cast(
diff --git a/tensorflow/python/ops/math_grad.py b/tensorflow/python/ops/math_grad.py
index 463ad8337c7..389f6f8dce9 100644
--- a/tensorflow/python/ops/math_grad.py
+++ b/tensorflow/python/ops/math_grad.py
@@ -297,7 +297,7 @@ def _ProdGrad(op, grad):
     reduction_indices = (reduction_indices + rank) % rank
     reduced = math_ops.cast(reduction_indices, dtypes.int32)
     idx = math_ops.range(0, rank)
-    other, _ = array_ops.setdiff1d(idx, reduced)
+    other, _ = gen_array_ops.list_diff(idx, reduced, dtypes.int32)
     perm = array_ops.concat([reduced, other], 0)
     reduced_num = math_ops.reduce_prod(array_ops.gather(input_shape, reduced))
     other_num = math_ops.reduce_prod(array_ops.gather(input_shape, other))
diff --git a/tensorflow/python/ops/math_ops.py b/tensorflow/python/ops/math_ops.py
index c29fd8f1477..2d14c6f8b65 100644
--- a/tensorflow/python/ops/math_ops.py
+++ b/tensorflow/python/ops/math_ops.py
@@ -4514,7 +4514,7 @@ def tensordot(a, b, axes, name=None):
         rank_a = array_ops.rank(a)
         axes = ops.convert_to_tensor(axes, dtype=dtypes.int32, name="axes")
         axes = array_ops.where(axes >= 0, axes, axes + rank_a)
-        free, _ = array_ops.setdiff1d(range(rank_a), axes)
+        free, _ = gen_array_ops.list_diff(range(rank_a), axes, dtypes.int32)
       free_dims = array_ops.gather(shape_a, free)
       axes_dims = array_ops.gather(shape_a, axes)
       prod_free_dims = reduce_prod(free_dims)