// tensor, for setting depth_multiplier attribute, etc.).
    auto filter = tf_op.getFilter();
    auto filter_type = mlir::dyn_cast<RankedTensorType>(filter.getType());
    if (!filter_type || filter_type.getRank() != 4 ||
        !filter_type.hasStaticShape())
      return failure();

    Value input = tf_op.getInput();
    RankedTensorType input_type =
        mlir::dyn_cast<RankedTensorType>(input.getType());

    CPred<"$0.getType().cast<ShapedType>().hasRank() && "
          "$0.getType().cast<ShapedType>().getRank() <= " # n>>;

// Checks if the value has rank 'n'.
class HasRank<int n> : Constraint<
    CPred<"$0.getType().cast<ShapedType>().hasRank() && "
          "$0.getType().cast<ShapedType>().getRank() == " # n>>;

class FloatValueEquals<string val> : Constraint<CPred<
  "FloatValueEquals($0, " # val # ")">>;

    // Input should be rank 4.
    if (!input_ty.hasRank() || input_ty.getRank() != 4) {
      return failure();
    }

    // Check that out_size is rank-1, length-2. Otherwise the size is not legal.
    if (!out_size_ty.hasRank() || out_size_ty.getRank() != 1 ||
        out_size_ty.getShape()[0] != 2) {
      return failure();
    }

                        communication_key_index++)
              .str();
      auto& cond = host_while.getCond();
      cond.push_back(new Block);
      auto condition =
          while_op.getCond().front().getTerminator()->getOperand(0);
      builder.setInsertionPoint(while_op.getCond().front().getTerminator());
      builder.create<mlir::TF::XlaSendToHostOp>(while_op.getLoc(), condition,

    @Override
    public AntBuilder ant(Closure configureClosure) {
        return ConfigureUtil.configure(configureClosure, getAnt());
    }

    @Override
    public AntBuilder ant(Action<? super AntBuilder> configureAction) {
        AntBuilder ant = getAnt();
        configureAction.execute(ant);
        return ant;
    }

    @Override

    if (matchPattern(op_state.operands[1], m_Constant(&shape_attr))) {
      auto shape_ty =
          op_state.operands[1].getType().dyn_cast<RankedTensorType>();
      if (shape_ty != nullptr && shape_ty.hasRank() && shape_ty.getRank() > 1) {
        llvm::SmallVector<mlir::Attribute, 4> shape;
        int32_t dim_size = 0;
        for (const auto& dim :
             llvm::enumerate(shape_attr.getValues<llvm::APInt>())) {

		Name:      name,
		Namespace: namespace,
	}
	if fun, ok := f.DefinitionMatchFuncs[key]; ok {
		return fun(definition, a), a.GetResource(), a.GetKind(), nil
	}

	// Default is match everything
	return f.DefaultMatch, a.GetResource(), a.GetKind(), nil
}

// Matches says whether this policy definition matches the provided admission
// resource request

            do {
                curReq.setResponse(curResp);

                ServerMessageBlock nextReq = curReq.getAndx();
                if ( nextReq == null ) {
                    break;
                }
                ServerMessageBlock nextResp = curResp.getAndx();
                nextReq.setResponse(nextReq);

    // the first `TensorListSetItemOp`.
    if (auto shaped_type = element_shape.getType().dyn_cast<ShapedType>()) {
      if (shaped_type.hasRank() && shaped_type.getRank() == 0) {
        bool element_shape_acquired = false;
        auto uses = op.getResult().getUses();
        for (auto &use : llvm::make_early_inc_range(uses)) {
          if (TF::TensorListSetItemOp set_op =

     */
    AntBuilder getAnt();

    /**
     * <p>Creates an additional <code>AntBuilder</code> for this project. You can use this in your build file to execute
     * ant tasks.</p>
     *
     * @return Creates an <code>AntBuilder</code> for this project. Never returns null.
     * @see #getAnt()
     */
    AntBuilder createAntBuilder();

    /**