%1 = "tf.Const"() {device = "/device:CPU:0", value = dense<[1.0]> : tensor<1xbf16>} : () -> tensor<4xbf16>
  // CHECK: corert.executeop({{.*}}) "tf.Const"() {dtype = ui64, value = dense<[1, 2, 3, 4]> : tensor<4xui64>} : 1
  %2 = "tf.Const"() {device = "/device:GPU:0", value = dense<[1, 2, 3, 4]> : tensor<4xui64>} : () -> tensor<4xui64>
  func.return %0 : tensor<4xui64>
}

  HLO_NO_METADATA,
  HLO_SERIALIZED,
  OPTIMIZED_HLO,
  OPTIMIZED_HLO_SERIALIZED,
  OPTIMIZED_HLO_PROTO_SERIALIZED,
  OPTIMIZED_HLO_DOT
};

struct ArgShapeAndDType {
  TensorShape shape;
  DataType dtype;
};

enum class CompilerArgSource {
  TENSOR_SPEC,
  CONCRETE_INPUT,
};

// Returns the IR format of the selected stage for a given function `func_name`

    %outputs_0, %control_0 = tf_executor.island wraps "tf.Placeholder"() {device = "", dtype = "tfdtype$DT_FLOAT", name = "y", shape = "tfshape$dim { }"} : () -> tensor<0xf32>

    // Set up an arg per arg_shape with the specified type.
    for (int i = 0; i < arg_shapes.size(); ++i) {
      auto metadata_arg = metadata_proto.add_args();
      metadata_arg->set_kind(
          tensorflow::tpu::TPUCompileMetadataProto::Arg::PARAMETER);
      metadata_arg->set_dtype(dtype);
    }
    // Set up one dummy retval.
    metadata_proto.add_retvals();

    auto& var_op = std::get<0>(var_and_tensor);
    auto& resource_tensor = std::get<1>(var_and_tensor);
    bool is_variable_initialized = false;
    if (resource_tensor.dtype() != tensorflow::DT_RESOURCE) {
      is_variable_initialized = true;
    } else {
      auto handle = resource_tensor.scalar<tensorflow::ResourceHandle>()();
      is_variable_initialized =

	_NOTE_TRIGGER = 0x1000000
)

type rtprio struct {
	_type uint16
	prio  uint16
}

type lwpparams struct {
	start_func uintptr
	arg        unsafe.Pointer
	stack      uintptr
	tid1       unsafe.Pointer // *int32
	tid2       unsafe.Pointer // *int32
}

type sigset struct {
	__bits [4]uint32
}

type stackt struct {
	ss_sp     uintptr
	ss_size   uintptr

  void Watch(const AbstractTensorHandle*);
  // Records an operation with given inputs and outputs
  // on the tape and marks all its outputs as watched if at
  // least one input of the op is watched and has a trainable dtype.
  // op_name is optional and is used for debugging only.
  void RecordOperation(absl::Span<AbstractTensorHandle* const> inputs,
                       absl::Span<AbstractTensorHandle* const> outputs,

  tensorflow::AsyncValueTensor* av_tensor =
      tensorflow::AsyncValueTensor::FromTensor(tensor);
  if (av_tensor == nullptr) {
    TF_ASSIGN_OR_RETURN(
        *tensor, MakeTensorFromPjRtBuffer(tensor->dtype(), tensor->shape(),
                                          std::move(buffer)));
  } else {
    av_tensor->SetBuffer(std::move(buffer));
  }
  return absl::OkStatus();
}

}

Status VerifySupportedSignature(TaggedValue signature) {
  if (signature.type() == TaggedValue::Type::TENSOR_SPEC) {
    return ::tensorflow::OkStatus();
  }
  if (signature.type() == TaggedValue::Type::TUPLE) {
    for (const auto& t : signature.tuple()) {
      if (t.type() != TaggedValue::Type::TENSOR_SPEC) {
        break;
      }
    }
    return ::tensorflow::OkStatus();
  }

// Copyright 2016 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

package syntax

type Token uint

type token = Token

//go:generate stringer -type token -linecomment tokens.go

const (
	_    token = iota
	_EOF       // EOF

	// names and literals
	_Name    // name
	_Literal // literal

	// operators and operations