Merge changes from github.

PiperOrigin-RevId: 184897758

6 files changed, 39 insertions, 65 deletions

diff --git a/tensorflow/contrib/makefile/Makefile b/tensorflow/contrib/makefile/Makefile
index dd5770dc99..81327407d4 100644
--- a/tensorflow/contrib/makefile/Makefile
+++ b/tensorflow/contrib/makefile/Makefile
@@ -377,10 +377,10 @@ $(MARCH_OPTION) \
 
 	ifeq ($(BUILD_FOR_TEGRA),1)
 		NVCC := $(JETPACK)/cuda/bin/nvcc
-		NVCCFLAGS := -x=cu -D__CUDACC__ -DNVCC -DNVIDIA_TEGRA -ccbin $(NDK_ROOT)/toolchains/$(TOOLCHAIN)/prebuilt/$(ANDROID_HOST_OS_ARCH)/bin/$(BIN_PREFIX)-g++ --std c++11 --expt-relaxed-constexpr -m64 -gencode arch=compute_53,\"code=sm_53\" -gencode arch=compute_62,\"code=sm_62\" -DEIGEN_AVOID_STL_ARRAY -DTENSORFLOW_USE_EIGEN_THREADPOOL -DLANG_CXX11 -DEIGEN_HAS_C99_MATH -DGOOGLE_CUDA=1 -DTF_EXTRA_CUDA_CAPABILITIES=5.3
+		NVCCFLAGS := -x=cu -D__CUDACC__ -DNVCC -DANDROID_TEGRA -ccbin $(NDK_ROOT)/toolchains/$(TOOLCHAIN)/prebuilt/$(ANDROID_HOST_OS_ARCH)/bin/$(BIN_PREFIX)-g++ --std c++11 --expt-relaxed-constexpr -m64 -gencode arch=compute_53,\"code=sm_53\" -gencode arch=compute_62,\"code=sm_62\" -DEIGEN_AVOID_STL_ARRAY -DTENSORFLOW_USE_EIGEN_THREADPOOL -DLANG_CXX11 -DEIGEN_HAS_C99_MATH -DGOOGLE_CUDA=1 -DTF_EXTRA_CUDA_CAPABILITIES=5.3
 		CXXFLAGS4NVCC =\
 -DIS_SLIM_BUILD \
--DNVIDIA_TEGRA \
+-DANDROID_TEGRA \
 -fno-exceptions \
 -DNDEBUG $(OPTFLAGS) \
 -march=armv8-a \
@@ -391,7 +391,7 @@ $(MARCH_OPTION) \
 		CXXFLAGS +=\
 -DGOOGLE_CUDA=1 \
 -D__ANDROID_TYPES_FULL__ \
--DNVIDIA_TEGRA \
+-DANDROID_TEGRA \
 -DEIGEN_AVOID_STL_ARRAY \
 -DEIGEN_HAS_C99_MATH \
 -DLANG_CXX11 -DTENSORFLOW_USE_EIGEN_THREADPOOL -DTF_EXTRA_CUDA_CAPABILITIES=5.3
@@ -407,7 +407,7 @@ $(MARCH_OPTION) \
 -I$(JETPACK)/cuda/extras/CUPTI/include
 
 
-		LIBS += \
+		CUDA_LIBS := \
 -ltfcuda \
 -lcudart_static \
 -lcudnn \
@@ -420,10 +420,10 @@ $(MARCH_OPTION) \
 -lculibos \
 -lcurand_static
 
-		OBJDIR := $(OBJDIR)Tegra/
-		LIBDIR := $(LIBDIR)Tegra/
-		BINDIR := $(BINDIR)Tegra/
-		DEPDIR := $(DEPDIR)Tegra/
+		OBJDIR := $(OBJDIR)android_arm64-v8a/
+		LIBDIR := $(LIBDIR)android_arm64-v8a/
+		BINDIR := $(BINDIR)android_arm64-v8a/
+		DEPDIR := $(DEPDIR)android_arm64-v8a/
 
 		TEGRA_LIBS := \
 -L$(JETPACK)/cuda/targets/aarch64-linux-androideabi/lib \
@@ -606,7 +606,8 @@ $(wildcard tensorflow/core/util/*/*.cc) \
 tensorflow/core/util/version_info.cc
 # Remove duplicates (for version_info.cc)
 CORE_CC_ALL_SRCS := $(sort $(CORE_CC_ALL_SRCS))
-CORE_CC_EXCLUDE_SRCS := \
+
+CORE_CC_EXCLUDE_SRCS_NON_GPU := \
 $(wildcard tensorflow/core/*/*test.cc) \
 $(wildcard tensorflow/core/*/*testutil*) \
 $(wildcard tensorflow/core/*/*testlib*) \
@@ -626,49 +627,31 @@ $(wildcard tensorflow/core/lib/jpeg/*) \
 $(wildcard tensorflow/core/lib/png/*) \
 $(wildcard tensorflow/core/util/events_writer.*) \
 $(wildcard tensorflow/core/util/reporter.*) \
-$(wildcard tensorflow/core/platform/default/cuda_libdevice_path.*) \
-$(wildcard tensorflow/core/platform/default/stream_executor.*) \
 $(wildcard tensorflow/core/platform/default/test_benchmark.*) \
-$(wildcard tensorflow/core/platform/cuda.h) \
-$(wildcard tensorflow/core/platform/cuda_libdevice_path.*) \
 $(wildcard tensorflow/core/platform/cloud/*) \
 $(wildcard tensorflow/core/platform/google/*) \
 $(wildcard tensorflow/core/platform/google/*/*) \
 $(wildcard tensorflow/core/platform/jpeg.*) \
 $(wildcard tensorflow/core/platform/png.*) \
 $(wildcard tensorflow/core/platform/s3/*) \
-$(wildcard tensorflow/core/platform/stream_executor.*) \
 $(wildcard tensorflow/core/platform/windows/*) \
-$(wildcard tensorflow/core/user_ops/*.cu.cc) \
-$(wildcard tensorflow/core/common_runtime/gpu/*) \
-$(wildcard tensorflow/core/common_runtime/gpu_device_factory.*) \
 $(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.*)
+$(wildcard tensorflow/core/grappler/clusters/single_machine.*) \
+tensorflow/core/util/cuda_kernel_helper_test.cu.cc
+
+CORE_CC_EXCLUDE_SRCS := \
+$(CORE_CC_EXCLUDE_SRCS_NON_GPU) \
+$(wildcard tensorflow/core/platform/stream_executor.*) \
+$(wildcard tensorflow/core/platform/default/cuda_libdevice_path.*) \
+$(wildcard tensorflow/core/platform/cuda.h) \
+$(wildcard tensorflow/core/platform/cuda_libdevice_path.*) \
+$(wildcard tensorflow/core/user_ops/*.cu.cc) \
+$(wildcard tensorflow/core/common_runtime/gpu/*) \
+$(wildcard tensorflow/core/common_runtime/gpu_device_factory.*)
 
 ifeq ($(BUILD_FOR_TEGRA),1)
-CORE_CC_ALL_SRCS := \
-$(wildcard tensorflow/core/*.cc) \
-$(wildcard tensorflow/core/common_runtime/*.cc) \
-$(wildcard tensorflow/core/common_runtime/gpu/*.cc) \
-$(wildcard tensorflow/core/framework/*.cc) \
-$(wildcard tensorflow/core/graph/*.cc) \
-$(wildcard tensorflow/core/platform/*.cc) \
-$(wildcard tensorflow/core/platform/*/*.cc) \
-$(wildcard tensorflow/core/platform/*/*/*.cc) \
-$(wildcard tensorflow/core/util/*.cc) \
-$(wildcard tensorflow/core/util/*/*.cc) \
-$(wildcard tensorflow/cc/training/*.cc) \
-$(wildcard tensorflow/stream_executor/*.cc) \
-$(wildcard tensorflow/stream_executor/*/*.cc) \
-$(wildcard tensorflow/core/grappler/optimizers/*.cc) \
-$(wildcard tensorflow/core/grappler/*.cc) \
-$(wildcard tensorflow/core/grappler/costs/*.cc) \
-$(wildcard tensorflow/core/grappler/clusters/*.cc) \
-$(wildcard tensorflow/core/grappler/utils/*.cc) \
-$(wildcard tensorflow/core/lib/core/*.cc) \
-$(wildcard tensorflow/core/lib/*/*.cc) \
-tensorflow/core/grappler/inputs/utils.cc \
+CORE_CC_ALL_SRCS := $(CORE_CC_ALL_SRCS) \
 tensorflow/core/kernels/concat_lib_gpu.cc \
 tensorflow/core/kernels/cuda_solvers.cc \
 tensorflow/core/kernels/cudnn_pooling_gpu.cc \
@@ -677,28 +660,14 @@ tensorflow/core/kernels/fractional_avg_pool_op.cc \
 tensorflow/core/kernels/fractional_max_pool_op.cc \
 tensorflow/core/kernels/fractional_pool_common.cc \
 tensorflow/core/kernels/pooling_ops_3d.cc \
-tensorflow/core/kernels/sparse_fill_empty_rows_op.cc
+tensorflow/core/kernels/sparse_fill_empty_rows_op.cc \
+tensorflow/core/kernels/list_kernels.cc \
+$(wildcard tensorflow/core/common_runtime/gpu/*.cc) \
+$(wildcard tensorflow/stream_executor/*.cc) \
+$(wildcard tensorflow/stream_executor/*/*.cc)
 
 CORE_CC_EXCLUDE_SRCS := \
-$(wildcard tensorflow/core/*/*test.cc) \
-$(wildcard tensorflow/core/*/*testutil*) \
-$(wildcard tensorflow/core/*/*testlib*) \
-$(wildcard tensorflow/core/*/*/*test.cc) \
-$(wildcard tensorflow/core/*/*/*testutil*) \
-$(wildcard tensorflow/core/framework/op_gen_lib.cc) \
-$(wildcard tensorflow/core/lib/gif/*) \
-$(wildcard tensorflow/core/lib/jpeg/*) \
-$(wildcard tensorflow/core/lib/png/*) \
-$(wildcard tensorflow/core/lib/db/*) \
-$(wildcard tensorflow/core/platform/jpeg.*) \
-$(wildcard tensorflow/core/platform/png.*) \
-$(wildcard tensorflow/core/platform/cloud/*) \
-$(wildcard tensorflow/core/platform/s3/*) \
-$(wildcard tensorflow/core/platform/windows/*) \
-$(wildcard tensorflow/core/*/*/*testlib*) \
-$(wildcard tensorflow/cc/training/*test.cc) \
-tensorflow/core/lib/io/record_reader.cc \
-tensorflow/core/util/cuda_kernel_helper_test.cu.cc
+$(CORE_CC_EXCLUDE_SRCS_NON_GPU)
 
 CUDA_CC_SRCS := $(wildcard tensorflow/core/kernels/*.cu.cc)
 CUDA_CC_OBJS := $(addprefix $(OBJDIR), $(CUDA_CC_SRCS:.cc=.o))
@@ -760,7 +729,7 @@ $(BENCHMARK_NAME): $(BENCHMARK_OBJS) $(LIB_PATH) $(CUDA_LIB_DEPS)
 	@mkdir -p $(dir $@)
 	$(CXX) $(CXXFLAGS) $(INCLUDES) \
 	-o $(BENCHMARK_NAME) $(BENCHMARK_OBJS) \
-	$(LIBFLAGS) $(TEGRA_LIBS) $(LIB_PATH) $(LDFLAGS) $(LIBS)
+	$(LIBFLAGS) $(TEGRA_LIBS) $(LIB_PATH) $(LDFLAGS) $(LIBS) $(CUDA_LIBS)
 
 # NVCC compilation rules for Tegra
 ifeq ($(BUILD_FOR_TEGRA),1)
diff --git a/tensorflow/contrib/makefile/build_all_android.sh b/tensorflow/contrib/makefile/build_all_android.sh
index 980a44a595..f67c516186 100755
--- a/tensorflow/contrib/makefile/build_all_android.sh
+++ b/tensorflow/contrib/makefile/build_all_android.sh
@@ -18,7 +18,7 @@
 set -e
 
 usage() {
-  echo "Usage: NDK_ROOT=<path to ndk root> $(basename "$0") [-Es:t:Tx:a:X]"
+  echo "Usage: NDK_ROOT=<path to ndk root> $(basename "$0") [-Es:t:Tx:a]"
   echo "-E enable experimental hexnn ops"
   echo "-s [sub_makefiles] sub makefiles separated by white space"
   echo "-t [build_target] build target for Android makefile [default=all]"
diff --git a/tensorflow/contrib/makefile/build_all_ios.sh b/tensorflow/contrib/makefile/build_all_ios.sh
index a18df256f9..2d99791839 100755
--- a/tensorflow/contrib/makefile/build_all_ios.sh
+++ b/tensorflow/contrib/makefile/build_all_ios.sh
@@ -96,7 +96,7 @@ if [[ "${ONLY_MAKE_TENSORFLOW}" != "true" ]]; then
 
     if [[ -z "${BUILD_ARCH}" ]]; then
         # Compile protobuf for the target iOS device architectures.
-        tensorflow/contrib/makefile/compile_ios_protobuf.sh -a ${DEFAULT_ARCH}
+        tensorflow/contrib/makefile/compile_ios_protobuf.sh
     else
         # Compile protobuf for the target iOS device architectures.
         tensorflow/contrib/makefile/compile_ios_protobuf.sh -a ${BUILD_ARCH}
diff --git a/tensorflow/contrib/makefile/samples/build_and_run_inception_hexagon.sh b/tensorflow/contrib/makefile/samples/build_and_run_inception_hexagon.sh
index 861bb885c7..203ff4f890 100755
--- a/tensorflow/contrib/makefile/samples/build_and_run_inception_hexagon.sh
+++ b/tensorflow/contrib/makefile/samples/build_and_run_inception_hexagon.sh
@@ -76,6 +76,8 @@ GEN_LIBS_DIR="${GEN_DIR}/libs"
 GEN_DOWNLOAD_DIR="${GEN_DIR}/downloads"
 URL_BASE="https://storage.googleapis.com/download.tensorflow.org"
 
+ARCH="armeabi-v7a"
+
 source "${SCRIPT_DIR}/../build_helper.subr"
 
 rm -rf "${GEN_DIR}"
@@ -219,7 +221,7 @@ if [[ "${BUILD_ONLY}" != "true" ]]; then
     adb push "${GEN_LIBS_DIR}/libhexagon_nn_skel.so" "/vendor/lib/rfsa/adsp"
 
     adb push -p \
-        "${TF_ROOT_DIR}/tensorflow/contrib/makefile/gen/bin/hexagon_graph_execution" \
+        "${TF_ROOT_DIR}/tensorflow/contrib/makefile/gen/bin/android_${ARCH}/hexagon_graph_execution" \
         "/data/local/tmp/"
     adb wait-for-device
     adb shell chmod "${ANDROID_EXEC_FILE_MODE}" \
diff --git a/tensorflow/contrib/makefile/sub_makefiles/android/Makefile.in b/tensorflow/contrib/makefile/sub_makefiles/android/Makefile.in
index d9277ed60c..3081084ee7 100644
--- a/tensorflow/contrib/makefile/sub_makefiles/android/Makefile.in
+++ b/tensorflow/contrib/makefile/sub_makefiles/android/Makefile.in
@@ -54,7 +54,7 @@ $(INFERENCE_SO_PATH): $(LIB_OBJS) $(INFERENCE_OBJS) $(CUDA_LIB_DEPS)
 	-o $@ $(INFERENCE_OBJS) $(LIB_OBJS) $(TEGRA_LIBS) \
 	$(LIBFLAGS) $(LDFLAGS) \
 	-shared -Wl,-soname,$(INFERENCE_SO_NAME) \
-	$(LIBS)
+	$(LIBS) $(CUDA_LIBS)
 
 $(INFERENCE_SO_NAME): $(INFERENCE_SO_PATH)
 
diff --git a/tensorflow/contrib/makefile/tf_op_files.txt b/tensorflow/contrib/makefile/tf_op_files.txt
index 5f27566398..5a812af4e9 100644
--- a/tensorflow/contrib/makefile/tf_op_files.txt
+++ b/tensorflow/contrib/makefile/tf_op_files.txt
@@ -91,6 +91,7 @@ tensorflow/core/kernels/reduction_ops_max.cc
 tensorflow/core/kernels/reduction_ops_common.cc
 tensorflow/core/kernels/reduction_ops_any.cc
 tensorflow/core/kernels/reduction_ops_all.cc
+tensorflow/core/kernels/roll_op.cc
 tensorflow/core/kernels/queue_ops.cc
 tensorflow/core/kernels/queue_base.cc
 tensorflow/core/kernels/pooling_ops_common.cc
@@ -270,6 +271,7 @@ tensorflow/core/ops/parsing_ops.cc
 tensorflow/core/ops/no_op.cc
 tensorflow/core/ops/nn_ops.cc
 tensorflow/core/ops/nn_grad.cc
+tensorflow/core/ops/manip_ops.cc
 tensorflow/core/ops/math_ops.cc
 tensorflow/core/ops/math_grad.cc
 tensorflow/core/ops/logging_ops.cc
@@ -291,3 +293,4 @@ tensorflow/core/kernels/batchtospace_op.cc
 tensorflow/core/kernels/warn_about_ints.cc
 tensorflow/core/kernels/segment_reduction_ops.cc
 tensorflow/core/kernels/batch_util.cc
+tensorflow/core/ops/audio_ops.cc