memset(&option, 0, sizeof(option));
  option.name = const_cast<char*>(name.c_str());
  TF_Filesystem_Option_Value option_value;
  memset(&option_value, 0, sizeof(option_value));
  option_value.type_tag = TF_Filesystem_Option_Type_Buffer;
  option_value.num_values = values.size();
  std::vector<TF_Filesystem_Option_Value_Union> option_values(values.size());
  for (size_t i = 0; i < values.size(); i++) {

  EXPECT_TF_META("v", -1, TF_ATTR_TENSOR, -1);
  TF_Tensor* value;
  TF_OperationGetAttrTensor(oper, "v", &value, s_);
  ASSERT_EQ(TF_OK, TF_GetCode(s_)) << TF_Message(s_);
  ASSERT_NE(nullptr, value);
  EXPECT_EQ(TF_INT8, TF_TensorType(value));
  EXPECT_EQ(ndims, TF_NumDims(value));
  for (int i = 0; i < TF_NumDims(value); ++i) {
    EXPECT_EQ(dims[i], TF_Dim(value, i)) << i;
  }

  tensorflow::AttrValue padding_value;
  *padding_value.mutable_s() = "VALID";
  tensorflow::SetOpAttrValueScalar(ctx, op, padding_value, "padding", status);

  tensorflow::AttrValue data_format_value;
  *data_format_value.mutable_s() = "NHWC";
  tensorflow::SetOpAttrValueScalar(ctx, op, data_format_value, "data_format",
                                   status);

  BundleEntryProto entry;
  v2_reader_->Seek(kHeaderEntryKey);
  for (v2_reader_->Next(); v2_reader_->Valid(); v2_reader_->Next()) {
    CHECK(entry.ParseFromArray(v2_reader_->value().data(),
                               v2_reader_->value().size()))
        << entry.InitializationErrorString();
    for (int i = 0; i < entry.slices_size(); ++i) {
      const auto& slice_proto = entry.slices(i);
      CHECK(filtered_keys

  *tensor =
      unwrap(TF_CreateAbstractTensorFromEagerTensor(input_eager, status.get()));
  return absl::OkStatus();
}

// Return a scalar tensor handle with given value.
template <class T, TF_DataType datatype>
Status TestScalarTensorHandle(AbstractContext* ctx, const T value,
                              AbstractTensorHandle** tensor) {
  std::unique_ptr<TF_Status, decltype(&TF_DeleteStatus)> status(
      TF_NewStatus(), TF_DeleteStatus);

    // Read the value of tensor handle `handle_1`.
    float value = 0.0f;
    TF_Tensor* t = TFE_TensorHandleResolve(handle_1, status);
    ASSERT_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
    ASSERT_EQ(sizeof(float), TF_TensorByteSize(t));
    memcpy(&value, TF_TensorData(t), sizeof(float));
    TF_DeleteTensor(t);
    EXPECT_EQ(1.2f, value);
    TFE_DeleteTensorHandle(handle_1);

  ASSERT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());
  TF_Tensor* result_tensor = TFE_TensorHandleResolve(result_t, status.get());
  float* result_value = static_cast<float*>(TF_TensorData(result_tensor));
  EXPECT_EQ(*result_value, 4.0);

  TF_DeleteTensor(result_tensor);
  TF_DeleteAbstractTensor(result);
  TF_DeleteOutputList(o);
  TF_DeleteExecutionContext(ctx);
}

// MatMul Test

  std::unordered_map<tensorflow::string, tensorflow::Node*> name_map
      TF_GUARDED_BY(mu);

  // The keys of this map are all the active sessions using this graph. Each
  // value records whether the graph has been mutated since the corresponding
  // session has been run (this is detected in RecordMutation function). If the
  // string is empty, no mutation has occurred. Otherwise the string is a

# Use the public-access TF DevInfra cache (read only)
build:sigbuild_remote_cache --remote_cache="https://storage.googleapis.com/tensorflow-devinfra-bazel-cache/manylinux2014" --remote_upload_local_results=false
# Change the value of CACHEBUSTER when upgrading the toolchain, or when testing
# different compilation methods. E.g. for a PR to test a new CUDA version, set
# the CACHEBUSTER to the PR number.
build --action_env=CACHEBUSTER=20220325

  } else {
    shape_string = "<error computing shape>";
  }
  std::string value_string;
  if (!SummarizeValue(value_string).ok()) {
    value_string = "<error computing value>";
  }
  if (value_string.length() > 100) {
    // The default NumPy-style output can be distractingly long in error
    // messages.
    value_string = absl::StrCat(value_string.substr(0, 100), " [...]");
  }
  Status s;