//
//go:systemstack
func checkRanks(gp *g, prevRank, rank lockRank) {
	rankOK := false
	if rank < prevRank {
		// If rank < prevRank, then we definitely have a rank error
		rankOK = false
	} else if rank == lockRankLeafRank {
		// If new lock is a leaf lock, then the preceding lock can
		// be anything except another leaf lock.
		rankOK = prevRank < lockRankLeafRank
	} else {
		// We've now verified the total lock ranking, but we

	}
	return lockNames[rank]
}
`)

	// Create partial order structure.
	fmt.Fprintf(w, `
// lockPartialOrder is the transitive closure of the lock rank graph.
// An entry for rank X lists all of the ranks that can already be held
// when rank X is acquired.
//
// Lock ranks that allow self-cycles list themselves.
var lockPartialOrder [][]lockRank = [][]lockRank{
`)
	for _, rank := range topo {

}

func (rank lockRank) String() string {
	if rank == 0 {
		return "UNKNOWN"
	}
	if rank == lockRankLeafRank {
		return "LEAF"
	}
	if rank < 0 || int(rank) >= len(lockNames) {
		return "BAD RANK"
	}
	return lockNames[rank]
}

// lockPartialOrder is the transitive closure of the lock rank graph.
// An entry for rank X lists all of the ranks that can already be held

namespace tensorflow {

// 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.

//
// $iota = "mhlo.iota"(){iota_dimension = $dimensions[0]},
//
// where $dimensions must have size 1 and iota can have rank>=1.
// It usually used for matching rank 1 iota since the iotaOp will be folded to
// IotaOp + BroadCastInDimOp except for the case when result shape is rank 1.
bool MatchSingleIota(DenseIntElementsAttr dimensions, Value iota) {
  auto iota_op = dyn_cast_or_null<mhlo::IotaOp>(iota.getDefiningOp());

    if (auto result_constraint = results.GetConstraint(op->getResult(0))) {
      // `test.OpA` converts shape constraint to rank constraint.
      if (is_op_a && *result_constraint == ValueConstraint::kShape)
        operands.Insert(op->getOperand(0), ValueConstraint::kRank);

      // `test.OpB` converts value constraint to shape constraint.
      if (*result_constraint == ValueConstraint::kValue)

  auto shape = ranked_ty.getShape();
  int rank = shape.size();

  SmallVector<APInt, 4> dimensions;
  dimensions.reserve(rank);
  for (int i = 0; i < rank; ++i)
    dimensions.push_back(APInt(out_width, shape[i]));

  auto result_type = tensorflow::GetTypeFromTFTensorShape(
      {rank}, IntegerType::get(input_ty.getContext(), out_width));

      current_indices.push_back(i);
      TransposeRecursively(original_values, target_values, current_indices);
      current_indices.pop_back();
    }
  }

  int rank_;                             // Rank of the input values.
  SmallVector<int64_t> original_shape_;  // Shape of the original tensor.
  SmallVector<int64_t> target_shape_;    // Shape of the target tensor.
  SmallVector<int64_t> permutation_;
};

  RankedTensorType tensorType = mlir::cast<RankedTensorType>(value.getType());
  Type element_type = tensorType.getElementType();

  int rank = tensorType.getShape().size();
  int num_rows = tensorType.getShape()[rank - 2];
  int num_cols = tensorType.getShape()[rank - 1];

  std::vector<Value> sliced;

  if (batch_size == 1) {
    // Batch size is 1, no splitting is required

  }

  const int64_t rank = input_type.getRank();
  if (rank <= 0) {
    return emitOptionalError(loc, "input should be of rank larger than 0");
  }

  int64_t axis_value = op.getAxisAttr().getInt();
  if (axis_value < 0) {
    axis_value += rank;
  }
  if (axis_value < 0 || axis_value >= rank) {
    return emitOptionalError(