TFE_MonitoringNewCounterReader(const char* name);

// Reads the value of a counter that was created with 0 labels.
TF_CAPI_EXPORT extern int64_t TFE_MonitoringReadCounter0(
    TFE_MonitoringCounterReader*);

// Reads the value of specific cell of a counter that was created with 1 label.
TF_CAPI_EXPORT extern int64_t TFE_MonitoringReadCounter1(
    TFE_MonitoringCounterReader*, const char* label_value);

#ifdef __cplusplus
} /* end extern "C" */

  std::vector<DataType> result;
  result.reserve(set.size());
  for (DataType dt : set) {
    result.push_back(dt);
  }
  return result;
}

std::vector<std::vector<int64_t>> InterestingShapes() {
  std::vector<std::vector<int64_t>> interesting_shapes;
  interesting_shapes.push_back({});             // Scalar
  interesting_shapes.push_back({10});           // 1D Vector
  interesting_shapes.push_back({3, 3});         // 2D Matrix

TF_CAPI_EXPORT extern void TF_SetShape(TF_Tensor* tensor, const int64_t* dims,
                                       int num_dims);

// Return the number of dimensions that the tensor has.
TF_CAPI_EXPORT extern int TF_NumDims(const TF_Tensor*);

// Return the length of the tensor in the "dim_index" dimension.
// REQUIRES: 0 <= dim_index < TF_NumDims(tensor)
TF_CAPI_EXPORT extern int64_t TF_Dim(const TF_Tensor* tensor, int dim_index);

  DataType Type() const override;
  int NumDims() const override;
  int64_t Dim(int dim_index) const override;
  int64_t NumElements() const override;
  size_t ByteSize() const override;
  void* Data() const override;
  bool IsAligned() const override;
  bool CanMove() const override;
  std::string SummarizeValue() const override;

  void SetShape(const int64_t* dims, int num_dims);
  Status ToTensor(tensorflow::Tensor* dst) const;

                                     bool cifg) {
  SmallVector<int64_t, 2> input_shape{1, 2};
  SmallVector<int64_t, 2> weight_shape{3, 12};
  SmallVector<int64_t, 1> bias_shape{2};
  SmallVector<int64_t, 2> projection_shape{1, 2};
  SmallVector<int64_t, 1> layer_norm_scale{4};
  SmallVector<int64_t, 2> output_shape{1, 2};
  auto input_type = RankedTensorType::get(input_shape, builder->getF32Type());

    ::testing::Values(
        std::make_tuple(
            2, 1, "", std::vector<int64_t>{0},
            "'device_assignment' must not be set when 'topology' is not set"),
        std::make_tuple(8, 1, "", std::vector<int64_t>(),
                        "'num_replicas' must be equal to 1 or 2, got 8"),
        std::make_tuple(2, 2, "", std::vector<int64_t>(),
                        "'num_cores_per_replica' must be equal to 1, got 2")));

 public:
  // Does not take any ownership. Caller must ensure channel_id is valid during
  // life-cylce of this object.
  CollectiveRewritePattern(MLIRContext* context, int64_t* channel_id)
      : OpRewritePattern<T>(context), channel_id_(*channel_id) {}

 protected:
  int64_t& channel_id_;  // A unique channel_id shared by all rewrite patterns
                         // in this pass. Not thread-safe.
};

Status SetAttrString(AbstractOperation*, const char* attr_name,
                     const char* data, size_t length, ForwardOperation*);
Status SetAttrInt(AbstractOperation*, const char* attr_name, int64_t value,
                  ForwardOperation*);
Status SetAttrFloat(AbstractOperation*, const char* attr_name, float value,
                    ForwardOperation*);

#include "tensorflow/core/platform/mutex.h"

namespace tensorflow {
namespace device_compilation_cache_internal {
template <typename ExecutableType>
int64_t ExecutableSize(const ExecutableType* executable) {
  return 0;
}

template <>
inline int64_t ExecutableSize<xla::LocalExecutable>(
    const xla::LocalExecutable* executable) {
  if (executable != nullptr && executable->executable() != nullptr) {

      %0 = "tf.Const"() {value = dense<1000> : tensor<1xi32>} : () -> tensor<1xi32>
      return %0 : tensor<1xi32>
    }
  })";

  CellReader<int64_t> dynamism_op_counter(kDynamismOpCounterStreamzName);
  CellReader<int64_t> dynamism_function_counter(
      kDynamismFunctionCounterStreamzName);

  CreateModule(kMlirModuleStr);
  auto result = Run();

  EXPECT_TRUE(result.succeeded());