==============================================================================*/
#include "tensorflow/c/experimental/gradients/tape/tape_operation.h"

#include "tensorflow/c/eager/abstract_context.h"
#include "tensorflow/c/eager/gradients.h"

namespace tensorflow {
namespace gradients {
TapeOperation::TapeOperation(AbstractOperation* parent_op, Tape* tape,
                             const GradientRegistry& registry)
    : AbstractOperation(kTape),

// watched and not the result of any operation in the tape.
using TensorTape = std::unordered_map<int64_t, int64_t>;

// Map from operation-id to tape entry.
template <typename BackwardFunction, typename TapeTensor>
using OpTape =
    std::unordered_map<int64_t, OpTapeEntry<BackwardFunction, TapeTensor>>;

// Operations the tape needs to perform on tensors to do backpropagation. Named

	{as: AMOVFL, a1: C_REG, a6: C_CREG, type_: 69, size: 4},
	{as: AMOVFL, a1: C_REG, a6: C_32CON, type_: 69, size: 4},

	{as: ASYSCALL, type_: 5, size: 4},
	{as: ASYSCALL, a1: C_REG, type_: 77, size: 12},
	{as: ASYSCALL, a1: C_U15CON, type_: 77, size: 12},
	{as: ABEQ, a6: C_BRA, type_: 16, size: 4},
	{as: ABEQ, a1: C_CREG, a6: C_BRA, type_: 16, size: 4},

  explicit TaggedValue(float f32) : type_(FLOAT32), data_(Float32(f32)) {}
  /// TaggedValue constructor for type INT64.
  explicit TaggedValue(int64_t i64) : type_(INT64), data_(Int64(i64)) {}
  /// TaggedValue constructor for type FLOAT32.
  explicit TaggedValue(Float32 f32) : type_(FLOAT32), data_(f32) {}
  /// TaggedValue constructor for type INT64.
  explicit TaggedValue(Int64 i64) : type_(INT64), data_(i64) {}

    mode: Literal["validation", "serialization"] = "validation",
) -> ModelField:
    """
    Create a new response field. Raises if type_ is invalid.
    """
    class_validators = class_validators or {}
    if PYDANTIC_V2:
        field_info = field_info or FieldInfo(
            annotation=type_, default=default, alias=alias
        )
    else:
        field_info = field_info or FieldInfo()

             absl::Span<AbstractTensorHandle*> outputs) -> Status {
    Tape tape(/*persistent=*/false);
    for (size_t i{}; i < inputs.size(); ++i) {
      tape.Watch(inputs[i]);
    }
    std::vector<AbstractTensorHandle*> temp_outputs(1);
    AbstractContextPtr tape_ctx(new TapeContext(ctx, &tape, grad_registry));
    TF_RETURN_IF_ERROR(

  bool passed() const { return type_ == kSuccess; }

  // Returns true iff the test part failed.
  bool failed() const { return type_ != kSuccess; }

  // Returns true iff the test part non-fatally failed.
  bool nonfatally_failed() const { return type_ == kNonFatalFailure; }

  // Returns true iff the test part fatally failed.
  bool fatally_failed() const { return type_ == kFatalFailure; }

 private:

  bool passed() const { return type_ == kSuccess; }

  // Returns true iff the test part failed.
  bool failed() const { return type_ != kSuccess; }

  // Returns true iff the test part non-fatally failed.
  bool nonfatally_failed() const { return type_ == kNonFatalFailure; }

  // Returns true iff the test part fatally failed.
  bool fatally_failed() const { return type_ == kFatalFailure; }

 private:

  Tape tape(/*persistent=*/false);
  tape.Watch(inputs[0]);  // Watch x.
  AbstractTensorHandle* exp_output;
  TF_RETURN_IF_ERROR(ops::Exp(ctx, inputs[0], &exp_output, "Exp"));
  std::unique_ptr<GradientFunction> gradient_function(
      new PassThroughGradientFunction);
  tape.RecordOperation(inputs, {exp_output}, gradient_function.release());
  TF_RETURN_IF_ERROR(tape.ComputeGradient(ctx,

			return errorf("%s: malformed elf file: %v", pn, err)
		}
		symbols[i] = elfsym.sym
		if elfsym.type_ != elf.STT_FUNC && elfsym.type_ != elf.STT_OBJECT && elfsym.type_ != elf.STT_NOTYPE && elfsym.type_ != elf.STT_COMMON {
			continue
		}
		if elfsym.shndx == elf.SHN_COMMON || elfsym.type_ == elf.STT_COMMON {
			sb := l.MakeSymbolUpdater(elfsym.sym)
			if uint64(sb.Size()) < elfsym.size {