// it's double-declared (it would cost extra in the common case to verify
// this). But the MethodSet computation will not find it due to the name
// collision caused by the double-declaration, leading to an internal
// inconsistency while we are verifying one computation against the other.
// var _ = T1c{}.Pointer

// T2's method declared before the type.
func (*T2) f /* ERROR "field and method" */ () {}

// Test simple serialized computation consisting of a function named `main`
// with a tf.Identity op forwarding the function single argument to the function
// single result.

// CHECK-LABEL: module

      op = op->getParentOp();
    }
  });
}

int OutsideCompilationOrdering(Operation* predecessor, Operation* op) {
  // Actual compilations go first.
  if (op->hasAttr("_is_compilation")) return 2;
  // Followed by nested ops that contain compilations.
  if (op->hasAttr("_wraps_compilation")) return 1;
  // Followed by everything else.
  return 0;
}

void MoveTpuCompileToFrontPass::runOnOperation() {

import os
import math
from flask import Flask, request
app = Flask(__name__)


@app.route('/hello')
def hello():
    version = os.environ.get('SERVICE_VERSION')

    # do some cpu intensive computation
    x = 0.0001
    for i in range(0, 1000000):
        x = x + math.sqrt(x)

    return 'Hello version: %s, instance: %s\n' % (version, os.environ.get('HOSTNAME'))


@app.route('/health')
def health():

should be used with caution when working with untrusted models.

### Saved graphs and checkpoints

When loading untrusted serialized computation graphs (in form of a `GraphDef`,
`SavedModel`, or equivalent on-disk format), the set of computation primitives
available to TensorFlow is powerful enough that you should assume that the
TensorFlow process effectively executes arbitrary code.

// CHECK-NEXT:  for node {{[{][{]node Add[}][}]}}

func.func @main() {
  tf_executor.graph {
    %0 = tf_executor.island wraps "tf._TPUReplicate"() {computation = @foo, Tinputs = [], Tbroadcast_inputs = [], NumVariables = 0, Tguaranteed_constants = [], output_types = []} : () -> () loc("_TPUReplicate")
    tf_executor.fetch
  }
  func.return
}

func.func @foo() {

  %cst_3 = "tf.Const"() {value = dense<4> : tensor<1xi32>} : () -> tensor<1xi32>
  %0 = "tf.XlaReduceWindow"(%arg0, %arg1, %cst_0, %cst_1, %cst_2, %cst_3, %cst) {computation = @sum_reducer} : (tensor<7xf32>, tensor<f32>, tensor<1xi32>, tensor<1xi32>, tensor<1xi32>, tensor<1xi32>, tensor<1x2xi32>) -> tensor<10xf32>
  func.return %0 : tensor<10xf32>
}

#include "tensorflow/cc/framework/scope.h"
#include "tensorflow/core/graph/while_context.h"

// Utility functions for constructing while loop gradients

namespace tensorflow {

// Adds the gradient computation for the while loop associated with
// `while_ctx`. `grad_inputs` are the partial derivatives w.r.t. the loop
// outputs, i.e. the exit nodes.  The partial derivatives w.r.t. the loop

namespace tensorflow {
namespace tfcompile {

using QuantizeXlaFn = std::function<Status(const tf2xla::Config& config,
                                           xla::XlaComputation* computation)>;

// Set the static quantization function to the `fn` if it hasn't been set.
// Return false if the static function has been set.
bool RegisterQuantizeFn(const QuantizeXlaFn& fn);

}  // namespace tfcompile

 */

package com.google.common.collect;

import com.google.common.annotations.GwtCompatible;
import javax.annotation.CheckForNull;

/**
 * Wraps an exception that occurred during a computation.
 *
 * @author Bob Lee
 * @since 2.0
 * @deprecated This exception is no longer thrown by {@code com.google.common}. Previously, it was