// Set the shapes and types of the output's handle.
//
// The lengths of the arrays pointed to by `shapes`, `ranks`, and `types` must
// all be equal to `num_shapes_and_types`. If `ranks[i] != -1`, (i.e., if the
// rank is known), then it must be equal to the length of `shapes[i]`; if
// `ranks[i] == 1`, then `shapes[i]` may be nullptr.
//
// TODO(akshayka): Implement a corresponding getter method.

    }                                                             \
  } while (0)

#define TF_CLEANUP_AND_RETURN_IF_ERROR(C_STATUS, BUFFER, __VA_ARGS__) \
  do {                                                                \
    ::tensorflow::Status _s(__VA_ARGS__);                             \
    if (TF_PREDICT_TRUE(!_s.ok())) {                                  \
      TF_DeleteStatus(C_STATUS);                                      \

  //
  // For example some tensor handles may represent distributed values, in which
  // case placement information is lost when resolving the handle.
  //
  // If false, a caller might implement pretty-printing by resolving and
  // iterating over the resulting tensor. This may still be viable if resolving
  // the handle loses information, but `SummarizeValue` would be more precise.
  virtual bool PreferCustomSummarizer() const { return false; }

                                   absl::Span<AbstractTensorHandle*>)>;

// Runs `model` maybe wrapped in a function call op. This can be thought as
// being equivalent to the following python code.
//
// if use_function:
//   outputs = tf.function(model)(inputs)
// else:
//   outputs = model(inputs)
Status RunModel(Model model, AbstractContext* ctx,
                absl::Span<AbstractTensorHandle* const> inputs,

class CGraphOptimizer : public CustomGraphOptimizer {
 public:
  explicit CGraphOptimizer(TP_Optimizer optimizer, const char* device_type)
      : optimizer_(optimizer), device_type_(device_type) {
    if (optimizer.create_func != nullptr) {
      c_optimizer_ = (*optimizer_.create_func)();
    } else {
      c_optimizer_ = nullptr;
    }
  }
  std::string name() const override { return "PluggableGraphOptimizer"; }

{
	int i, max;
	
	max = *(int*)p;
	for(i=0; i<max; i++)
		Add(i);
	return 0;
}

void
doAdd(int max, int nthread)
{
	enum { MaxThread = 20 };
	int i;
	uintptr_t thread_id[MaxThread];
	
	if(nthread > MaxThread)
		nthread = MaxThread;
	for(i=0; i<nthread; i++)
		thread_id[i] = _beginthreadex(0, 0, addThread, &max, 0, 0);
	for(i=0; i<nthread; i++) {
		WaitForSingleObject((HANDLE)thread_id[i], INFINITE);

// is for performance optimization by reusing an exiting unused op rather than
// creating a new op every time. If `raw_device_name` is `NULL` or empty, it
// does not set the device name. If it's not `NULL`, then it attempts to parse
// and set the device name. It's effectively `TFE_OpSetDevice`, but it is faster
// than separately calling it because if the existing op has the same
// `raw_device_name`, it skips parsing and just leave as it is.

    TF_GraphProperties* graph_properties);

// Infer tensor shapes through abstract interpretation.
// If assume_valid_feeds is true, it can help infer shapes in the fanout of fed
// nodes. This may cause incorrectness in graph analyses, but is useful for
// simulation or scheduling.
// If aggressive_shape_inference is true, nodes are executed on the host to

==============================================================================*/

#ifndef TENSORFLOW_C_C_API_MACROS_H_
#define TENSORFLOW_C_C_API_MACROS_H_

#ifdef SWIG
#define TF_CAPI_EXPORT
#else
#if defined(_WIN32)
#ifdef TF_COMPILE_LIBRARY
#define TF_CAPI_EXPORT __declspec(dllexport)
#else
#define TF_CAPI_EXPORT __declspec(dllimport)
#endif  // TF_COMPILE_LIBRARY
#else
#ifdef TF_CAPI_WEAK

                          ImmediateExecutionTensorHandle* input) = 0;

  virtual ImmediateExecutionContext* GetContext() const = 0;

  // Following two methods are used to support custom device.
  // Return true if the inputs contain custom device tensor handle. It means
  // that the argument need to be handled by a custom device.
  virtual bool HasCustomDeviceInput() const = 0;

  virtual const tensorflow::OpDef* OpDef() const = 0;