It's a regression issue.
The reason of the bug appeared to be an error introduced into 5.1 source code.
A piece of code in Create_file_log_event::do_apply_event() did not have test
coverage which made make test and pb unaware.
      
Fixed with inverting the old value of the return value from 
Create_file_log_event::do_apply_event().
The rpl test suite is extended with `rpl_cross_version' the file to hold 
regression cases similar to the current.

The problem is that the query cache was storing partial results
if the statement failed when sending the results to the client.
This could cause clients to hang when trying to read the results
from the cache as they would, for example, wait indefinitely for
a eof packet that wasn't saved.

The solution is to always discard the caching of a query that
failed to send its results to the associated client.

Extending the existing testcase written for BUG#40949 to verify
repair table operation for compressed tables

clause server fires immediately after creating event and time between create and delete
event sometimes is enough for firing. So adding STARTS clause moves first execution in
future after drop of event
1. Added STARTS clause for CREATE EVENT.
2. Updated result file.

character_set_database !=character_set_server" is fixed.

This does not bring any contents changes, it is purely
metadata which are affected.

Details:

Even within 5.0, most of these changesets did not cause
file contents changes, because they were backports done
for the "service pack" builds of 5.0.66sp1 and 5.0.72sp1.

The "real" changesets are also already present in 5.1,
so this upmerge doesn't change any contents.

The only "real" changeset in 5.0 was a fix of the shell
scripts used to configure bdb (BerkeleyDB).
As we completele removed bdb from the 5.1 sources already,
the affected files are not present in the 5.1 source tree,
so this changeset also does not cause any contents changes.

this should not happen.

This is not the final merge of that release build, but we need early
access to these tool fixes (use of "awk" in the BDB configuration).

with mysql_change_user) to 5.0.

Post-fix test failure: fixed mysqlcheck.test on Windows platforms.