
{      The status of path zero determines the capability to send or
{ receive control messages.  The state and clarifier for this path
{ should always be normal (2,1).  The event processor examines the
{ I/O threshold flags to determine if any action is required.  The
{ following algorithm is used:
{
{      IF  output is below threshold
{          AND   control messages are waiting to be sent   THEN
{        queue a request to send the next control message.
{      IFEND
{      IF  input is available  THEN
{        queue a request to receive a control message.
{      IFEND
{
{      The response handler for the receive control message request validates
{ the control message and queues the control message for the corresponding
{ application to retrieve (except for backs).  In this case a 'BACK' counter
{ is incremented in the corresponding connection entry.  If either a control
{ message for an illegal connection or an unsupported control message
{ is received the event processor discards the control message after
{ logging an appropriate diagnostic.  If a control message is
{ successfully received the response handler posts another request
{ to receive an incoming control message.  This continues until
{ a fatal error is encountered or all incoming control messages have
{ been retrieved.
{
{      The response handler for the send control message request
{ verifies that the control message was successfully sent.  If the
{ control message was successfully sent, the response handler posts
{ a request to send another control message.  This continues until;
{ all queued control messages are sent, a fatal error is encountered,
{ or the NAD resources are exhausted.
{
{    nad_index: (input) This parameter specifies the index of the
{      local NAD to check for control message threshold changes.
