func opset(a, b0 obj.As) {
	oprange[a&obj.AMask] = oprange[b0]
}

// Determine if the build configuration requires a TOC pointer.
// It is assumed this always called after buildop.
func NeedTOCpointer(ctxt *obj.Link) bool {
	return !pfxEnabled && ctxt.Flag_shared
}

// Build the opcode table
func buildop(ctxt *obj.Link) {
	// Limit PC-relative prefix instruction usage to supported and tested targets.

	if p == nil || p.Link == nil { // handle external functions and ELF section symbols
		return
	}

	if oprange[AAND&obj.AMask] == nil {
		ctxt.Diag("arm64 ops not initialized, call arm64.buildop first")
	}

	c := ctxt7{ctxt: ctxt, newprog: newprog, cursym: cursym, autosize: int32(p.To.Offset & 0xffffffff), extrasize: int32(p.To.Offset >> 32)}
	p.To.Offset &= 0xffffffff // extrasize is no longer needed

		p1 := a1.Package
		if p1 != nil {
			if a1.built != "" || a1.buildID != "" {
				buildID := a1.buildID
				if buildID == "" {
					buildID = b.buildID(a1.built)
				}
				fmt.Fprintf(h, "packagefile %s=%s\n", p1.ImportPath, contentID(buildID))
			}
			// Because we put package main's full action ID into the binary's build ID,
			// we must also put the full action ID into the binary's action ID hash.

                                            mlir::Builder builder);

  // Returns the inferred output type at index `idx` of the `node` in the
  // context.
  absl::StatusOr<mlir::Type> InferOutputType(const Node& node, int idx,
                                             mlir::Builder builder);

  // Convert deferred TF functions to the MLIR representation.

      RankedTensorType::get(static_cast<int32_t>(1), builder.getI32Type());
  auto out_segids_cst = builder.create<TF::ConstOp>(
      builder.getI32TensorAttr(flattened_out_segids));
  auto contracting_segids_cst = builder.create<TF::ConstOp>(
      builder.getI32TensorAttr(flattened_contracting_segids));
  auto num_segids_tensor =
      builder.create<TF::ConstOp>(builder.getI32IntegerAttr(1));

   */
  LocalCache(
      CacheBuilder<? super K, ? super V> builder, @CheckForNull CacheLoader<? super K, V> loader) {
    concurrencyLevel = Math.min(builder.getConcurrencyLevel(), MAX_SEGMENTS);

    keyStrength = builder.getKeyStrength();
    valueStrength = builder.getValueStrength();

    keyEquivalence = builder.getKeyEquivalence();
    valueEquivalence = builder.getValueEquivalence();

   */
  LocalCache(
      CacheBuilder<? super K, ? super V> builder, @CheckForNull CacheLoader<? super K, V> loader) {
    concurrencyLevel = Math.min(builder.getConcurrencyLevel(), MAX_SEGMENTS);

    keyStrength = builder.getKeyStrength();
    valueStrength = builder.getValueStrength();

    keyEquivalence = builder.getKeyEquivalence();
    valueEquivalence = builder.getValueEquivalence();

                                               Builder *builder) {
  RankedTensorType ty = mlir::RankedTensorType::get(
      {static_cast<int32_t>(values.size())}, builder->getIntegerType(32));
  return DenseIntElementsAttr::get(ty, values);
}

DenseIntElementsAttr GetI64ElementsAttr(ArrayRef<int64_t> values,
                                        Builder *builder) {
  RankedTensorType ty = RankedTensorType::get(

    return non_empty_values[0];
  }

  // Otherwise, build a new concatenation op with non-empty values.
  mlir::OpBuilder builder(getOperation());
  auto new_concat = builder.create<TFL::ConcatenationOp>(
      getLoc(), getType(), non_empty_values,
      builder.getIntegerAttr(builder.getIntegerType(32), getAxis()),
      builder.getStringAttr(getFusedActivationFunction()));
  return new_concat.getResult();
}

      op.getOperation());
}

void NotEqualOp::build(OpBuilder &builder, OperationState &result, Value x,
                       Value y, BoolAttr incompatible_shape_error) {
  auto result_type = DeduceEqualCmpOpType(&builder, result.location, x, y,
                                          incompatible_shape_error);
  return build(builder, result, result_type, x, y, incompatible_shape_error);
}