Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash

Version for 5.0.
It fixes three problems:
1. The cause of the bug was that we did not check the table version for
 the HANDLER ... READ commands. We did not notice when a table was
 replaced by a new one. This can happen during ALTER TABLE, REPAIR
 TABLE, and OPTIMIZE TABLE (there might be more cases). I call the fix
 for this problem "the primary bug fix".
2. mysql_ha_flush() was not always called with a locked LOCK_open.
 Though the function comment clearly said it must.
 I changed the code so that the locking is done when required. I call
 the fix for this problem "the secondary fix".
3. In 5.0 (not in 4.1 or 4.0) DROP TABLE had a possible deadlock flaw in
 concur with FLUSH TABLES WITH READ LOCK. I call the fix for this
 problem "the 5.0 addendum fix".


include/my_pthread.h:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  Added a new macro for the 5.0 addendum fix.
mysql-test/r/handler.result:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  The test result.
mysql-test/t/handler.test:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  The test case.
sql/lock.cc:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  Changed a comment which did confuse me and which is not fully
  correct anymore after the 5.0 addendum fix.
  Added an assertion which would fire without the 5.0 addendum fix.
sql/mysql_priv.h:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  Changed a definition for the secondary fix.
sql/sql_base.cc:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  Changed function calls for the secondary fix.
sql/sql_class.cc:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  Changed a function call for the secondary fix.
sql/sql_handler.cc:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  The first two diffs make the primary bug fix.
  The rest is for the secondary fix.
sql/sql_table.cc:
  Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
  The first diff (four changed places) make the 5.0 addendum fix.
  The other three are changed function calls for the secondary fix.

1 files changed, 75 insertions, 0 deletions

diff --git a/mysql-test/t/handler.test b/mysql-test/t/handler.test
index 1bb9b1d3504..f3e14c3cd2b 100644
--- a/mysql-test/t/handler.test
+++ b/mysql-test/t/handler.test
@@ -347,4 +347,79 @@ drop table t3;
 drop table t4;
 drop table t5;
 
+#
+# Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
+#
+create table t1 (c1 int);
+insert into t1 values (1);
+# client 1
+handler t1 open;
+handler t1 read first;
+# client 2
+connect (con2,localhost,root,,);
+connection con2;
+--exec echo send the below to another connection, do not wait for the result
+send optimize table t1;
+--sleep 1
+# client 1
+--exec echo proceed with the normal connection
+connection default;
+handler t1 read next;
+handler t1 close;
+# client 2
+--exec echo read the result from the other connection
+connection con2;
+reap;
+# client 1
+--exec echo proceed with the normal connection
+connection default;
+drop table t1;
+
 # End of 4.1 tests
+
+#
+# Addendum to Bug#14397 - OPTIMIZE TABLE with an open HANDLER causes a crash
+# Show that DROP TABLE can no longer deadlock against
+# FLUSH TABLES WITH READ LOCK. This is a 5.0 issue.
+#
+create table t1 (c1 int);
+insert into t1 values (14397);
+flush tables with read lock;
+# The thread with the global read lock cannot drop the table itself:
+--error 1223
+drop table t1;
+#
+# client 2
+# We need a second connection to try the drop.
+# The drop waits for the global read lock to go away.
+# Without the addendum fix it locked LOCK_open before entering the wait loop.
+connection con2;
+--exec echo send the below to another connection, do not wait for the result
+send drop table t1;
+--sleep 1
+#
+# client 1
+# Now we need something that wants LOCK_open. A simple table access which
+# opens the table does the trick.
+--exec echo proceed with the normal connection
+connection default;
+# This would hang on LOCK_open without the 5.0 addendum fix.
+select * from t1;
+# Release the read lock. This should make the DROP go through.
+unlock tables;
+#
+# client 2
+# Read the result of the drop command.
+connection con2;
+--exec echo read the result from the other connection
+reap;
+#
+# client 1
+# Now back to normal operation. The table should not exist any more.
+--exec echo proceed with the normal connection
+connection default;
+--error 1146
+select * from t1;
+# Just to be sure and not confuse the next test case writer.
+drop table if exists t1;
+