bool test_mode_;
  Option<OpSet> op_set_{
      *this, "target-opset", llvm::cl::init(OpSet::TF),
      llvm::cl::desc("Choose target opset."),
      llvm::cl::values(
          clEnumValN(OpSet::TF, "TF",
                     "Uses TF ops that mimic quantization behavior"),
          clEnumValN(OpSet::XLA, "XLA", "Uses TF XLA ops"),
          clEnumValN(OpSet::UNIFORM_QUANTIZED, "UNIFORM_QUANTIZED",

    const QuantizationSpecs& quant_specs,
    tensorflow::quantization::OpSet op_set);

// Creates an instance of the PreprocessOp pass, which will perform op
// preprocessing to allow multi-axis quantization, prior to quantization.
std::unique_ptr<OperationPass<ModuleOp>> CreatePreprocessOpPass(
    tensorflow::quantization::OpSet op_set,
    tensorflow::quantization::QuantizationMethod::PresetMethod
        quantization_method,

      (
          quantization_options.op_set == quant_opts_pb2.OpSet.UNIFORM_QUANTIZED
          or quantization_options.quantization_method.preset_method
          == _PresetMethod.METHOD_STATIC_RANGE_WEIGHT_ONLY_INT8
      )
      or (
          quantization_options.op_set
          in (quant_opts_pb2.OpSet.XLA, quant_opts_pb2.OpSet.STABLEHLO)
          and quantization_options.quantization_method.preset_method

    const absl::flat_hash_map<std::string, std::string> &function_aliases,
    absl::string_view calibration_data_dir) {
  const bool is_stablehlo = quantization_options.op_set() == OpSet::STABLEHLO;
  // Use StableHLO Quantizer option if opset is specified.
  if (is_stablehlo) {
    const QuantizationConfig quantization_config =
        GetQuantizationConfigForStaticRangePtq(quantization_options);

        ),
        tags=tags,
        signature_keys=['serving_default'],
        op_set=target_opset,
    )

    if target_opset != quant_opts_pb2.XLA:
      # Uniform quantized opset is not supported for weight-only
      with self.assertRaisesRegex(
          ValueError, 'TF/Uniform quantized opset does not support weight-only.'
      ):
        converted_model = quantize_model.quantize(

QuantizedType CalculateUniformQuantParams(
    PatternRewriter& rewriter, TF::ConstOp op,
    tensorflow::quantization::QuantizationComponentSpec& weight_spec) {
  // TODO - b/278949920: Enable Per-Channel Quantization for XLA Opset
  // Currently, support symmetric, per-tensor, signed int8
  const bool kIsNarrowRange = true;
  const bool kIsSigned = true;
  const int kBitWidth = 8;

  DenseFPElementsAttr attr;

		spcs := &spcSpace[i]
		if spcs.lo <= spc && spc < spcs.hi {
			return spcs.SPCconv(spc)
		}
	}
	return fmt.Sprintf("SPC???%d", spc)
}

type opSet struct {
	lo    As
	names []string
}

// Not even worth sorting
var aSpace []opSet

// RegisterOpcode binds a list of instruction names
// to a given instruction number range.
func RegisterOpcode(lo As, Anames []string) {

// RUN: tf-quant-opt %s -split-input-file -quant-lift-quantizable-spots-as-functions -quant-quantize='target-opset=XLA' -verify-each=false | FileCheck %s

func.func private @conv(%input: tensor<1x3x4x3xf32> {tf._user_specified_name = "input_tensor"}) -> tensor<*xf32> attributes {tf._construction_context = "kEagerRuntime", tf._input_shapes = [#tf_type.shape<1x3x4x3>]} {
  %weight = arith.constant dense_resource<__elided__> : tensor<2x3x3x2xf32>

  ShapedType input_shape =
      mlir::dyn_cast<ShapedType>(op->getOperand(0).getType());
  if (!input_shape) {
    return op->emitError(
        "Only input with known shape is supported for Uniform Quantized "
        "opset.");
  }

  if (op->getParentOfType<func::FuncOp>().getName().contains("depthwise_")) {
    feature_group_cnt = input_shape.getDimSize(3);
  }

  attrs.push_back(rewriter.getNamedAttr(

// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

package ipset

import (
	"errors"
	"fmt"
	"net/netip"
)

type IPSet struct {
	V4Name string
	V6Name string
	Prefix string
	Deps   NetlinkIpsetDeps
}

const (
	V4Name = "%s-v4"
	V6Name = "%s-v6"
)