diff --git a/tensorflow/contrib/image/kernels/adjust_hsv_in_yiq_op_gpu.cu.cc b/tensorflow/contrib/image/kernels/adjust_hsv_in_yiq_op_gpu.cu.cc
index f97e790b56c..a0542b399fd 100644
--- a/tensorflow/contrib/image/kernels/adjust_hsv_in_yiq_op_gpu.cu.cc
+++ b/tensorflow/contrib/image/kernels/adjust_hsv_in_yiq_op_gpu.cu.cc
@@ -18,7 +18,7 @@ limitations under the License.
 #include "tensorflow/contrib/image/kernels/adjust_hsv_in_yiq_op.h"
 #include "tensorflow/core/kernels/gpu_utils.h"
 #include "tensorflow/core/platform/stream_executor.h"
-#include "tensorflow/core/util/cuda_kernel_helper.h"
+#include "tensorflow/core/util/gpu_kernel_helper.h"
 
 namespace tensorflow {
 
diff --git a/tensorflow/contrib/makefile/Makefile b/tensorflow/contrib/makefile/Makefile
index dc1b106ed91..199863867f4 100644
--- a/tensorflow/contrib/makefile/Makefile
+++ b/tensorflow/contrib/makefile/Makefile
@@ -685,7 +685,7 @@ $(wildcard tensorflow/core/platform/windows/*) \
 $(wildcard tensorflow/core/grappler/inputs/trivial_test_graph_input_yielder.*) \
 $(wildcard tensorflow/core/grappler/inputs/file_input_yielder.*) \
 $(wildcard tensorflow/core/grappler/clusters/single_machine.*) \
-tensorflow/core/util/cuda_kernel_helper_test.cu.cc
+tensorflow/core/util/gpu_kernel_helper_test.cu.cc
 
 CORE_CC_EXCLUDE_SRCS := \
 $(CORE_CC_EXCLUDE_SRCS_NON_GPU) \
diff --git a/tensorflow/contrib/mpi_collectives/kernels/ring.cu.cc b/tensorflow/contrib/mpi_collectives/kernels/ring.cu.cc
index 976317afe26..c8e3e81c8ba 100644
--- a/tensorflow/contrib/mpi_collectives/kernels/ring.cu.cc
+++ b/tensorflow/contrib/mpi_collectives/kernels/ring.cu.cc
@@ -19,9 +19,8 @@ limitations under the License.
 
 #define EIGEN_USE_GPU
 
-#include "tensorflow/core/util/cuda_launch_config.h"
-
 #include "tensorflow/contrib/mpi_collectives/kernels/ring.h"
+#include "tensorflow/core/util/gpu_launch_config.h"
 
 namespace tensorflow {
 namespace contrib {
diff --git a/tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc b/tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc
index 13fbd974e9c..ea4026008ed 100644
--- a/tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc
+++ b/tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc
@@ -21,7 +21,7 @@ limitations under the License.
 #include "tensorflow/core/framework/op.h"
 #include "tensorflow/core/framework/op_kernel.h"
 #include "tensorflow/core/framework/register_types.h"
-#include "tensorflow/core/util/cuda_kernel_helper.h"
+#include "tensorflow/core/util/gpu_kernel_helper.h"
 
 namespace tensorflow {
 
diff --git a/tensorflow/contrib/resampler/kernels/resampler_ops_gpu.cu.cc b/tensorflow/contrib/resampler/kernels/resampler_ops_gpu.cu.cc
index 3b2ee098b3e..ecb6c187a07 100644
--- a/tensorflow/contrib/resampler/kernels/resampler_ops_gpu.cu.cc
+++ b/tensorflow/contrib/resampler/kernels/resampler_ops_gpu.cu.cc
@@ -17,13 +17,13 @@
 
 #define EIGEN_USE_GPU
 
-#include "tensorflow/contrib/resampler/kernels/resampler_ops.h"
-
 #include <stdio.h>
+
 #include <cmath>
 
+#include "tensorflow/contrib/resampler/kernels/resampler_ops.h"
 #include "tensorflow/core/framework/register_types.h"
-#include "tensorflow/core/util/cuda_kernel_helper.h"
+#include "tensorflow/core/util/gpu_kernel_helper.h"
 
 namespace tensorflow {
 
diff --git a/tensorflow/contrib/rnn/kernels/lstm_ops_gpu.cu.cc b/tensorflow/contrib/rnn/kernels/lstm_ops_gpu.cu.cc
index 81beb2942c1..472058b9a9e 100644
--- a/tensorflow/contrib/rnn/kernels/lstm_ops_gpu.cu.cc
+++ b/tensorflow/contrib/rnn/kernels/lstm_ops_gpu.cu.cc
@@ -17,12 +17,11 @@ limitations under the License.
 
 #define EIGEN_USE_GPU
 
-#include "tensorflow/contrib/rnn/kernels/lstm_ops.h"
-
 #include "third_party/eigen3/unsupported/Eigen/CXX11/Tensor"
+#include "tensorflow/contrib/rnn/kernels/lstm_ops.h"
 #include "tensorflow/core/kernels/eigen_activations.h"
 #include "tensorflow/core/platform/logging.h"
-#include "tensorflow/core/util/cuda_kernel_helper.h"
+#include "tensorflow/core/util/gpu_kernel_helper.h"
 
 namespace tensorflow {
 namespace functor {
diff --git a/tensorflow/contrib/seq2seq/kernels/beam_search_ops_gpu.cu.cc b/tensorflow/contrib/seq2seq/kernels/beam_search_ops_gpu.cu.cc
index be2aa4782c3..3af6b1cb766 100644
--- a/tensorflow/contrib/seq2seq/kernels/beam_search_ops_gpu.cu.cc
+++ b/tensorflow/contrib/seq2seq/kernels/beam_search_ops_gpu.cu.cc
@@ -18,7 +18,7 @@ limitations under the License.
 #define EIGEN_USE_GPU
 
 #include "tensorflow/contrib/seq2seq/kernels/beam_search_ops.h"
-#include "tensorflow/core/util/cuda_kernel_helper.h"
+#include "tensorflow/core/util/gpu_kernel_helper.h"
 
 namespace tensorflow {
 namespace functor {
diff --git a/tensorflow/contrib/tensorrt/custom_plugin_examples/inc_op_kernel.cu.cc b/tensorflow/contrib/tensorrt/custom_plugin_examples/inc_op_kernel.cu.cc
index b683c14c0d7..65f99a2eea7 100644
--- a/tensorflow/contrib/tensorrt/custom_plugin_examples/inc_op_kernel.cu.cc
+++ b/tensorflow/contrib/tensorrt/custom_plugin_examples/inc_op_kernel.cu.cc
@@ -24,7 +24,7 @@ limitations under the License.
 #include "cuda/include/cuda_runtime_api.h"
 #include "tensorflow/core/framework/op_kernel.h"
 #include "tensorflow/core/platform/stream_executor.h"
-#include "tensorflow/core/util/cuda_launch_config.h"
+#include "tensorflow/core/util/gpu_launch_config.h"
 
 namespace tensorflow {
 namespace tensorrt {