#include "tensorflow/c/experimental/grappler/grappler_internal.h"
#include "tensorflow/c/tf_buffer.h"
#include "tensorflow/c/tf_buffer_internal.h"
#include "tensorflow/c/tf_status.h"
#include "tensorflow/c/tf_status_helper.h"
#include "tensorflow/core/framework/function.h"
#include "tensorflow/core/framework/op.h"
#include "tensorflow/core/framework/op_def.pb.h"
#include "tensorflow/core/grappler/costs/graph_properties.h"

      vspace, target_tensor_ids, source_tensor_ids, sources_that_are_targets,
      output_gradients, result, /*build_default_zeros_grads*/ false));
  return absl::OkStatus();
}

// Helper functions which delegate to `AbstractOperation`, update
// the state of the ForwardOperation and call the tape as appropriate.
// These APIs are mainly to facilitate testing and are subject to change.
namespace internal {

  float* values = new float[1];
  values[0] = v;
  return TF_NewTensor(TF_FLOAT, nullptr, 0, values, num_bytes,
                      &FloatDeallocator, nullptr);
}

// All the *Helper methods are used as a workaround for the restrictions that
// one cannot call ASSERT_* methods in non-void-returning functions (when
// exceptions are disabled during compilation)

#include "tensorflow/c/eager/unified_api_testutil.h"
#include "tensorflow/c/experimental/gradients/grad_test_helper.h"
#include "tensorflow/c/experimental/gradients/tape/tape_context.h"
#include "tensorflow/c/experimental/ops/math_ops.h"
#include "tensorflow/c/tf_status_helper.h"
#include "tensorflow/core/platform/tensor_float_32_utils.h"
#include "tensorflow/core/platform/test.h"

namespace tensorflow {

#include <sys/mman.h>
#include <sys/stat.h>
#include <unistd.h>

#include "tensorflow/c/experimental/filesystem/filesystem_interface.h"
#include "tensorflow/c/experimental/filesystem/plugins/posix/posix_filesystem_helper.h"
#include "tensorflow/c/tf_file_statistics.h"
#include "tensorflow/c/tf_status.h"

// Implementation of a filesystem for POSIX environments.
// This filesystem will support `file://` and empty (local) URI schemes.

  // scheduled on that same threadpool, causing a deadlock in cases where the
  // caller of event_mgr->ThenExecute() blocks on the completion of the callback
  // (as in the case of ConstOp kernel creation on GPU, which involves copying a
  // CPU tensor to GPU).
  // Setting a larger thread pool does not help with the Swift caller, as we use
  // a different TFE context for each thread of execution (for running graph

                   &FloatDeallocator, nullptr),
      TF_DeleteTensor);
  return TensorHandlePtr(TFE_NewTensorHandle(tensor.get(), status));
}

// Helper to un-pack `num_replicas` TFE_TensorHandles from one parallel handle.
template <std::size_t num_replicas>
void ExtractPerDeviceValues(
    TFE_Context* context, TFE_TensorHandle* input,

TF_Buffer* ProcessGetKeyValueResult(absl::StatusOr<std::string> value,
                                    TF_Status* status) {
  status->status = value.status();
  if (!value.ok()) {
    return nullptr;
  }
  // Caller is responsible to call `TF_DeleteBuffer` to release the buffer.
  TF_Buffer* result = TF_NewBuffer();
  const std::string& value_str = *value;
  void* data = malloc(value_str.length());

#include "tensorflow/c/eager/parallel_device/parallel_device_lib.h"
#include "tensorflow/c/eager/tfe_tensorhandle_internal.h"
#include "tensorflow/c/tf_status.h"
#include "tensorflow/c/tf_status_helper.h"

namespace tensorflow {
namespace parallel_device {
namespace {

class OpDeleter {
 public:
  void operator()(TFE_Op* to_delete) const { TFE_DeleteOp(to_delete); }
};

  // finished, if any.
  //
  // `cancellation_manager` must live until after `Join` finishes and pending
  // `is_async` operations finish. In addition to allowing the caller to cancel
  // the operation, its `StartCancel` method will be called if op execution
  // fails on any device in order to cancel the others.
  void StartExecute(TFE_Context* context, const char* operation_name,