MDEV-15152 Optimistic parallel slave doesnt cope well with START SLAVE UNTIL

The immediate bug was caused by a failure to recognize a correct
position to stop the slave applier run in optimistic parallel mode.
There were the following set of issues that the analysis unveil.
1 incorrect estimate for the event binlog position passed to
  is_until_satisfied
2 wait for workers to complete by the driver thread did not account non-group events
  that could be left unprocessed and thus to mix up the last executed
  binlog group's file and position:
  the file remained old and the position related to the new rotated file
3 incorrect 'slave reached file:pos' by the parallel slave report in the error log
4 relay log UNTIL missed out the parallel slave branch in
  is_until_satisfied.

The patch addresses all of them to simplify logics of log change
notification in either the master and relay-log until case.
P.1 is addressed with passing the event into is_until_satisfied()
for proper analisis by the function.
P.2 is fixed by changes in handle_queued_pos_update().
P.4 required removing relay-log change notification by workers.
Instead the driver thread updates the notion of the current relay-log
fully itself with aid of introduced
bool Relay_log_info::until_relay_log_names_defer.

An extra print out of the requested until file:pos is arranged
with --log-warning=3.

1 files changed, 18 insertions, 3 deletions

diff --git a/sql/rpl_rli.h b/sql/rpl_rli.h
index 4ec4821b67d..2bc0a80268a 100644
--- a/sql/rpl_rli.h
+++ b/sql/rpl_rli.h
@@ -219,7 +219,7 @@ public:
   */
   char future_event_master_log_name[FN_REFLEN];
 
-  /* 
+  /*
      Original log name and position of the group we're currently executing
      (whose coordinates are group_relay_log_name/pos in the relay log)
      in the master's binlog. These concern the *group*, because in the master's
@@ -419,7 +419,7 @@ public:
   void close_temporary_tables();
 
   /* Check if UNTIL condition is satisfied. See slave.cc for more. */
-  bool is_until_satisfied(my_off_t);
+  bool is_until_satisfied(Log_event *ev);
   inline ulonglong until_pos()
   {
     DBUG_ASSERT(until_condition == UNTIL_MASTER_POS ||
@@ -427,7 +427,13 @@ public:
     return ((until_condition == UNTIL_MASTER_POS) ? group_master_log_pos :
 	    group_relay_log_pos);
   }
-
+  inline char *until_name()
+  {
+    DBUG_ASSERT(until_condition == UNTIL_MASTER_POS ||
+                until_condition == UNTIL_RELAY_POS);
+    return ((until_condition == UNTIL_MASTER_POS) ? group_master_log_name :
+	    group_relay_log_name);
+  }
   /**
     Helper function to do after statement completion.
 
@@ -564,6 +570,15 @@ private:
     relay_log.info had 4 lines. Now it has 5 lines.
   */
   static const int LINES_IN_RELAY_LOG_INFO_WITH_DELAY= 5;
+  /*
+    Hint for when to stop event distribution by sql driver thread.
+    The flag is set ON by a non-group event when this event is in the middle
+    of a group (e.g a transaction group) so it's too early
+    to refresh the current-relay-log vs until-log cached comparison result.
+    And it is checked and to decide whether it's a right time to do so
+    when the being processed group has been fully scheduled.
+  */
+  bool until_relay_log_names_defer;
 
   /*
     Holds the state of the data in the relay log.