return absl::OkStatus();
}

Status MlirAbstractOp::Create(ArrayRef<Value> operands,
                              OperationState** state) {
  state_->operands = llvm::to_vector<4>(operands);
  Builder builder(context_);

  if (current_ods_input_ != op_def_->input_arg_size())
    return InvalidArgument(absl::StrCat("Mismatch in operands number: got ",
                                        current_ods_input_, " expected ",

// commutativity of the AddV2: we want the FusedBatchNormV3 on any side.
// Also we need some native calls to handle the "hasOneUse" aspects and the
// optional extra operands for the AddV2 case.
struct ReluToFusedBatchNorm : public OpRewritePattern<ReluOp> {
  using OpRewritePattern<ReluOp>::OpRewritePattern;

  LogicalResult matchAndRewrite(ReluOp relu_op,

# Checks that source debug information is used in the output error message.
# CHECK: error: 'tf.Add' op operands don't have broadcast-compatible shapes
# CHECK: math_ops.add(x, y, name='x_y_sum')
# CHECK: build_graph(out_dir)
node: {
  name: "x"
  op: "Placeholder"
  attr: {
    key: "shape"
    value: {
      shape: {

    parallel_execute = mlir::TF::BuildParallelExecuteOp(cluster_func, &builder);

  // Rewrite results before rewriting operands as `tf.TPUPartitionedInputV2`
  // resource handle results is an indicator for a partitioned resource
  // variable. These `tf.TPUPartitionedInputV2` will be removed when rewriting
  // the operands.
  for (Value result : parallel_execute.getExecuteOutputs()) {
    if (!result.hasOneUse()) continue;

to an aliased device name. Otherwise the device already set in tf_device.launch
is used instead.

Operands are replicated inputs and packed inputs.

replicated_inputs: each group of `n` inputs corresponds to an input for a single
individual replica and is mapped to a single region argument. Inside one group
the operands are matching in order the `devices` attribute. Each replicated
input must have compatible shapes and types.

#include "tensorflow/compiler/mlir/tensorflow/ir/tf_ops.h"

namespace mlir {
namespace quant {

// Copied from tensorflow/compiler/mlir/lite/transforms/prepare_tf.cc.
// By removing identity ops, constant operands with dynamic shapes have static
// shape information which is necessary for correct pattern matching in this
// pass.
struct RemoveIdentity : public OpRewritePattern<TF::IdentityOp> {

    return rewriter.notifyMatchFailure(
        concat, "Concatenate op should have at least two operands");

  auto first = concat.getVal()[0].getDefiningOp<mhlo::SliceOp>();
  auto second = concat.getVal()[1].getDefiningOp<mhlo::SliceOp>();
  if (!first || !second)
    return rewriter.notifyMatchFailure(concat, "operands are not slice ops");
  if (first.getOperand() != second.getOperand())

//
// If the format specifier includes a %w verb with an error operand,
// the returned error will implement an Unwrap method returning the operand.
// If there is more than one %w verb, the returned error will implement an
// Unwrap method returning a []error containing all the %w operands in the
// order they appear in the arguments.
// It is invalid to supply the %w verb with an operand that does not implement

  metadata.ParseFromString(std::string(metadata_str.getValue()));
  int64_t num_replicas = replicate.getN();
  // Find the formattable operands of `execute`, which must be mirrored
  // variables (arguments of `replicate`), and must be pass-throughs from while
  // operands.
  for (const auto& mirrored_index : mirrored_variable_indices_attr) {
    int64_t replicate_arg = mlir::cast<IntegerAttr>(mirrored_index).getInt();

    // NOLINTNEXTLINE
    llvm::SmallVectorImpl<mlir::NamedAttribute> &attributes);

// While the last several tensors could be optional tensors for an tfl op, the
// number of input operands could vary. This function gets the min/max number of
// operands from tflite op name.
llvm::MinMax OperandNumbersMinMax(llvm::StringRef op_name);

// Populates the `custom_code` and `custom_options` to attributes.