}

void ReplaceTaskInServerDef(tensorflow::ServerDef* server_def, int task_index,
                            const string& host, int port) {
  tensorflow::JobDef* job_def = server_def->mutable_cluster()->mutable_job(0);
  job_def->mutable_tasks()->at(task_index) =
      tensorflow::strings::StrCat(host, ":", port);
}

std::vector<std::string> ListDeviceNames(TFE_Context* ctx) {
  TF_Status* status = TF_NewStatus();

  (*sess_options.config.mutable_device_count())["CPU"] = num_cpus;
  status->status =
      tensorflow::DeviceFactory::AddCpuDevices(sess_options, prefix, &devices);

  // Remove the device that has the host device name since host device is alreay
  // in an initialized context.
  for (auto d = devices.begin(); d != devices.end();) {
    if (absl::StrContains(d->get()->name(), "CPU:0")) {
      d = devices.erase(d);
    } else {

*   DeviceFactory's AddDevices and CreateDevices functions now return a Status
    instead of void.
*   Int32 elements of list(type) arguments are no longer placed in host memory
    by default. If necessary, a list(type) argument to a kernel can be placed in
    host memory using a HostMemory annotation.
*   `uniform_unit_scaling_initializer()` no longer takes a `full_shape` arg,

    visibility = ["//visibility:public"],
)

# Sometimes Bazel reports darwin_x86_64 as "darwin" and sometimes as
# "darwin_x86_64". The former shows up when building on a Mac x86_64 host for a Mac x86_64 target.
# The latter shows up when cross-compiling for Mac x86_64 from a Mac ARM machine and in internal
# Google builds.
config_setting(
    name = "macos_x86_64_default",

# Set execution platform to Linux x86
# Note: Lot of the "host_" flags such as "host_cpu" and "host_crosstool_top"
# flags seem to be actually used to specify the execution platform details. It
# seems it is this way because these flags are old and predate the distinction
# between host and execution platform.
build:cross_compile_base --host_cpu=k8

      environ_cp,
      var_name='GCC_HOST_COMPILER_PATH',
      var_default=default_gcc_host_compiler_path,
      ask_for_var='Please specify which gcc should be used by nvcc as the host '
      'compiler.',
      check_success=os.path.exists,
      resolve_symlinks=True,
      error_msg='Invalid gcc path. %s cannot be found.',
  )

  write_action_env_to_bazelrc('GCC_HOST_COMPILER_PATH', gcc_host_compiler_path)

+      __attribute_nonstring__;	/* Devicename.  */
+    char ut_user[UT_NAMESIZE]
+      __attribute_nonstring__;	/* Username.  */
 #define ut_name ut_user
-    char ut_host[UT_HOSTSIZE];
+    char ut_host[UT_HOSTSIZE]
+      __attribute_nonstring__;	/* Hostname for remote login.  */
     long int ut_time;
   };
 
diff --git a/misc/sys/cdefs.h b/misc/sys/cdefs.h
index 30482a1..551d5fd 100644
--- a/misc/sys/cdefs.h
+++ b/misc/sys/cdefs.h

        // broadcast, i.e. set as the input to each parallel operation. This
        // allows code like "tf.constant(1.)" or "tf.reduce_sum(..., axis=1)"
        // (where the value starts on the host), without allowing other implicit
        // copies/broadcasts. Other implicit copies may be supported eventually,
        // but need special handling for gradients (gradient of copy-on is not

tensorflow::ServerDef ReplaceTaskInServerDef(
    const tensorflow::ServerDef& server_def, int task_index);

void ReplaceTaskInServerDef(tensorflow::ServerDef* server_def, int task_index,
                            const std::string& host, int port);

std::vector<std::string> ListDeviceNames(TFE_Context* ctx);

// Set the target in TF_SessionOptions.options.
// target can be empty, a single entry, or a comma separated list of entries.
// Each entry is in one of the following formats :
// "local"
// ip:port
// host:port
TF_CAPI_EXPORT extern void TF_SetTarget(TF_SessionOptions* options,
                                        const char* target);

// Set the config in TF_SessionOptions.options.