void TF_ShapeAndTypeListSetDtype(TF_ShapeAndTypeList* shape_list, int index,
                                 TF_DataType dtype) {
  DCHECK(index >= 0 && index < shape_list->num_items);
  TF_ShapeAndType& shape_and_type = shape_list->items[index];
  shape_and_type.dtype = dtype;
}

void TF_DeleteShapeAndTypeList(TF_ShapeAndTypeList* shape_list) {
  if (shape_list == nullptr) return;
  for (size_t i = 0; i < shape_list->num_items; ++i) {

distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
==============================================================================*/
#include "tensorflow/c/experimental/filesystem/plugins/posix/posix_filesystem.h"

#include <dirent.h>

  // and not for correctness. The only downside of keeping this 1 seems to be
  // that the gradient accumulation is unbounded and we will never
  // aggressively aggregate accumulated gradients to recover memory.
  // Revisit and fix.
  return 1;
}

// Consumes references to the tensors in the gradient_tensors list and returns
// a tensor with the result.

  // 1) nullptr indicates tensor is compact and row-majored.
  // 2) fill in the strides array as the real case for compact row-major data.
  // Here we choose option 2, since some frameworks didn't handle the strides
  // argument properly.
  dlm_tensor->dl_tensor.strides = stride_arr->data();

  dlm_tensor->dl_tensor.byte_offset =
      0;  // TF doesn't handle the strides and byte_offsets here

                               "' has a reference type ", DataTypeString(dt))
             : absl::OkStatus();
}

// Converts `ninputs` and `inputs` into `inputs_tensors` and `input_nodes` and
// does various checks while doing so. `input_nodes` will contain the same
// information as input_tensors just in a different structure to make

  if (!status->status.ok()) {
    return nullptr;
  }

  TF_DataType dtype = TFE_TensorHandleDataType(components[0].get());
  // Verify that the TensorHandle's shape and dtype match all of the component
  // shapes and dtypes.
  for (TensorHandlePtr& component : components) {
    TFE_TensorHandleGetStatus(component.get(), status);
    if (!status->status.ok()) {
      return nullptr;
    }

// This test creates a cluster with two workers, create a variable on the
// second worker, and run a distributed function (VariableAddFunction) whose ops
// span the local and remote workers. If the graph optimization pass is executed
// on both the main function side and the component function side, an error will
// be thrown in the registered graph optimization pass.
TEST(CAPI, DistributedFunctionGraphPassOnlyOnce) {

distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
==============================================================================*/

#include "tensorflow/c/experimental/next_pluggable_device/c_api.h"

#include <cstdint>
#include <cstdlib>

distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
==============================================================================*/
#include "tensorflow/c/experimental/gradients/math_grad.h"

#include "tensorflow/c/eager/abstract_tensor_handle.h"

  TF_Graph* g_;
  std::unique_ptr<TF_OperationDescription> op_;
  // Hold `op_type` and `op_name` till both are available since we need both
  // to build a graph operation.
  string op_type_;
  const char* op_name_ = nullptr;
  // TODO(srbs): Use this.
  string device_name_;
};

// GraphContext wraps a TF_Graph modeling a single function and manages the
// "execution" of operation, i.e. adding them to the function.