// Prepare some tensors for shape.
  TF_Tensor* tensor_1X6 = Int32Tensor({1, 6});
  CHECK_EQ(TF_OK, TF_GetCode(status_)) << TF_Message(status_);
  TF_Tensor* tensor_1X1X6 = Int32Tensor({1, 1, 6});
  CHECK_EQ(TF_OK, TF_GetCode(status_)) << TF_Message(status_);

  TFE_Op* reshape_op = TFE_NewOp(tfe_context_, "Reshape", status_);
  CHECK_EQ(TF_OK, TF_GetCode(status_)) << TF_Message(status_);
  TFE_OpSetAttrType(reshape_op, "T", TF_FLOAT);

  # list all .whl files by their modification time (ls -t) and delete anything
  # other than the most recently-modified one (the new one).
  ls -t *.whl | tail -n +2 | xargs rm
fi

# Check if size is too big. TFCI_WHL_SIZE_LIMIT is in find's format, which can be
# 'k' for kilobytes, 'M' for megabytes, or 'G' for gigabytes, and the + to indicate
# "anything greater than" is added by the script.

#include "tensorflow/core/platform/test.h"
#include "tensorflow/core/protobuf/tensorflow_server.pb.h"

using ::tensorflow::string;

string MatMulFunction(const string& matmul_device) {
  tensorflow::FunctionDef def;
  CHECK(tensorflow::protobuf::TextFormat::ParseFromString(
      absl::StrCat("    signature {"
                   "      name: 'MatMulFunction'"
                   "      input_arg {"
                   "        name: 'a'"

# Writes the list of tests that would be run from BAZEL_TEST_COMMAND to OUTPUT.
# Hides all extra output and always exits with success for now.

OUTPUT=$1
shift

    ],
)

tf_cuda_cc_test(
    name = "gradients_test",
    size = "small",
    srcs = [
        "gradients_test.cc",
    ],
    args = ["--heap_check="],
    tags = tf_cuda_tests_tags() + ["nomac"],
    deps = [
        ":abstract_context",
        ":abstract_tensor_handle",
        ":c_api_experimental",
        ":c_api_test_util",

                 absl::Span<AbstractTensorHandle* const> grad_outputs,
                 absl::Span<AbstractTensorHandle*> grad_inputs) override {
    // TODO(b/161805092): Support broadcasting.

    DCHECK(grad_outputs[0]);
    grad_inputs[0] = grad_outputs[0];
    grad_inputs[1] = grad_outputs[0];

    grad_inputs[0]->Ref();
    grad_inputs[1]->Ref();
    return absl::OkStatus();
  }

  }

  tensorflow::Status Shape(
      tensorflow::PartialTensorShape* shape) const override {
    DCHECK(shape != nullptr);
    TF_Status status;
    int num_dims = TF_GraphGetTensorNumDims(graph_, output_, &status);
    DCHECK_GE(num_dims, -1);
    TF_RETURN_IF_ERROR(StatusFromTF_Status(&status));
    if (num_dims == kUnknownRank) {
      return absl::OkStatus();
    }

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

namespace tensorflow {
namespace {

// Specification for expected input/output and its type.
// DataType value of DT_INVALID signifies that we don't want to
// check the data type.
typedef std::pair<string, DataType> IOSpec;

const char* kFeedStackToString = "File \"feed.cc\", line 10, in alpha";
const char* kNegStackToString = "File \"neg.cc\", line 15, in beta";

 protected:
  void SetUp() override {
    TF_StatusPtr status(TF_NewStatus());
    TF_SetTracingImplementation(std::get<0>(GetParam()), status.get());
    Status s = StatusFromTF_Status(status.get());
    CHECK_EQ(errors::OK, s.code()) << s.message();
  }

 public:
  bool UseMlir() const { return strcmp(std::get<0>(GetParam()), "mlir") == 0; }
  bool UseFunction() const { return std::get<2>(GetParam()); }
};

 protected:
  void SetUp() override {
    TF_StatusPtr status(TF_NewStatus());
    TF_SetTracingImplementation(std::get<0>(GetParam()), status.get());

    {
      Status s = StatusFromTF_Status(status.get());
      CHECK_EQ(errors::OK, s.code()) << s.message();
    }

    {
      AbstractContext* ctx_raw = nullptr;
      Status s =
          BuildImmediateExecutionContext(std::get<1>(GetParam()), &ctx_raw);