// version.

// CHECK-LABEL: allows_chlo
func.func @allows_chlo(%arg0: tensor<1x32x10x32xi32>, %arg1: tensor<32xi32>) -> tensor<1x32x10x32xi32> {
  %0 = "chlo.broadcast_add"(%arg0, %arg1) {broadcast_dimensions = array<i64: 3>} : (tensor<1x32x10x32xi32>, tensor<32xi32>) -> tensor<1x32x10x32xi32>
  func.return %0 : tensor<1x32x10x32xi32>

	rv := reflect.ValueOf(v)
	switch rv.Kind() {
	case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64:
		i64 := rv.Int()
		if i64 > (1<<31)-1 || i64 < -(1<<31) {
			return nil, fmt.Errorf("sql/driver: value %d overflows int32", v)
		}
		return i64, nil
	case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64:
		u64 := rv.Uint()
		if u64 > (1<<31)-1 {

// CHECK-LABEL: _tfrt_resource_init
// CHECK: tf.VarHandleOp
// CHECK: tf.ReadVariableOp
// CHECK: tfrt_fallback_async.set_resource
// CHECK-SAME: {device = "/device:CPU:0", index = 0 : i64}


// CHECK-LABEL: func @init
// CHECK-SAME: {tfrt.cost_threshold = 1 : i64}
func.func @func_init() attributes {tf_saved_model.exported_names = ["init"]} {
  func.return
}

// CHECK-LABEL: func @basic
// CHECK-SAME: ([[in_chain:%.*]]: !tfrt.chain

module attributes {tf.versions = {bad_consumers = [], min_consumer = 0 : i32, producer = 121 : i32}} {
  func.func @main(%arg0: tensor<*xf32> {tf.aliasing_output = 0 : i64}, %arg1: tensor<2x4x6x8xi32>) -> (tensor<*xf32>, tensor<2x4x6x8xi32> {tf.device = "/CPU:1"})
  attributes  {tf.entry_function = {inputs = "args_0,args_1", outputs = "rets_0,rets_1"}} {
    %0:2 = tf_executor.graph {

// CHECK-SAME: <{metadata = "", output_shapes = [#tf_type.shape<>], output_types = [!tf_type.string], predicate = @{{.*}}}> {Targuments = [!tf_type.resource, !tf_type.resource, i64, !tf_type.resource, !tf_type.resource, !tf_type.resource, !tf_type.resource, i64]}
}

// CHECK-LABEL: FailureOnInvalidOp
func.func @FailureOnInvalidOp(%arg0: tensor<4xf64>, %arg1: tensor<4xf64>) -> tensor<4xf64> {

# CHECK-DAG: func private @custom_embedding_matmul{{[0-9]*}}(){{.+}}tf._implements = #tf_type.func<@tensorflow.embedding_matmul, {key1 = 2 : i64, key2 = false}>, tf._original_func_name = "custom_embedding_matmul"}...

  // CHECK-SAME: ancestors = []
  // CHECK-SAME: cost_estimate_ns = 1000000 : i64
  // CHECK-SAME: key = ""
  // CHECK-SAME: recv_key = "host_compute_channel_recv"
  // CHECK-SAME: send_key = "host_compute_channel_send"
  // CHECK-SAME: shape_inference_graph = @host_func
  // CHECK-SAME: shapes = [#tf_type.shape<*>, #tf_type.shape<3x?>]
  // CHECK-SAME: tpu_core = 0 : i64
  // CHECK: func @host_func

	}

	switch headerType {
	case "integer":
		switch typed := value.(type) {
		case int64:
			return typed
		case float64:
			return int64(typed)
		case json.Number:
			if i64, err := typed.Int64(); err == nil {
				return i64
			}
		}
	case "number":
		switch typed := value.(type) {
		case int64:
			return float64(typed)
		case float64:
			return typed
		case json.Number:

import (
	"fmt"
	"os"
)

var (
	i8 int8 = 0x12
	i16 int16 = 0x1234
	i32 int32 = 0x12345678
	i64 int64 = 0x123456789abcdef0
	ui8 uint8 = 0x12
	ui16 uint16 = 0x1234
	ui32 uint32 = 0x12345678
	ui64 uint64 = 0x123456789abcdef0

	ni8 = ^i8
	ni16 = ^i16
	ni32 = ^i32
	ni64 = ^i64
	nui8 = ^ui8
	nui16 = ^ui16
	nui32 = ^ui32
	nui64 = ^ui64
)

var nfail = 0

func main() {

    %1:2 = stablehlo.composite "odml.update_kv_cache" %0#0, %0#1, %arg2, %arg3, %arg4 {composite_attributes = {kv_cache_max = 500 : i64}, decomposition = @odml.update_kv_cache.impl_0} : (tensor<1x500x4x4xf32>,...