bzr push into mysql-trunk-runtime branch (kostja:3037 to 3038) Bug#52289

[Konstantin Osipov]

 3038 Konstantin Osipov	2010-06-03
      A follow up for the previous patch, titled:
      A code review comment for Bug#52289.
      Encapsulate the deadlock detection functionality into
      a visitor class...
      
      Remove a race introduced by omission: 
      initialize iterators under a read lock on the object.
     @ sql/mdl.cc
        Initialize iterators under a read lock on the object.

    modified:
      sql/mdl.cc
 3037 Konstantin Osipov	2010-06-03
      A code review comment for Bug#52289.
      
      Encapsulate the deadlock detection functionality into 
      a visitor class, and separate it from the wait-for graph
      traversal code.
      
      Use "Internal iterator" and "Visitor" patterns to 
      achieve the desired separation of responsibilities.
      
      Add comments.
     @ sql/mdl.cc
        Encapsulate deadlock detection into a class.
     @ sql/mdl.h
        Adjust for a rename of a class.

    modified:
      sql/mdl.cc
      sql/mdl.h
=== modified file 'sql/mdl.cc'
--- a/sql/mdl.cc	2010-06-03 14:08:22 +0000
+++ b/sql/mdl.cc	2010-06-03 14:32:56 +0000
@@ -1854,15 +1854,16 @@ bool MDL_lock::find_deadlock(MDL_ticket 
   MDL_ticket *ticket;
   MDL_context *src_ctx= waiting_ticket->get_ctx();
   bool result= TRUE;
-  Ticket_iterator granted_it(m_granted);
-  Ticket_iterator waiting_it(m_waiting);
-
 
   if (dvisitor->enter_node(src_ctx))
     return TRUE;
 
   mysql_prlock_rdlock(&m_rwlock);
 
+  /* Must be initialized after taking a read lock. */
+  Ticket_iterator granted_it(m_granted);
+  Ticket_iterator waiting_it(m_waiting);
+
   /*
     We do a breadth-first search first -- that is, inspect all
     edges of the current node, and only then follow up to the next


Attachment: [text/bzr-bundle] bzr/[e-mail]