KILL QUERY not behaving consistently and will hang in some cases

[Laurynas Biveinis]

On 5.6 trunk:

main.show_check w4 [ fail ]
        Test ended at 2016-08-26 19:30:43

CURRENT_TEST: main.show_check
mysqltest: At line 1223: query 'SHOW CREATE TABLE non_existent' failed with wrong errno 1146: 'Table 'test.non_existent' doesn't exist', instead of 1317...

This could be caused by the server still cleaning up the previous executed statement by the time KILL QUERY for that thread arrives, resetting kill flag before SHOW CREATE TABLE starts executing.

Backporting the fix for https://bugs.mysql.com/bug.php?id=45679 would address this and also improve KILL QUERY behavior in general.

[Laurynas Biveinis (laurynas-biveinis)]

3f5ccb3 BUG#16457803 - MAIN.SHOW_CHECK UNSTABLE ON PB2
33eba69 BUG#16457803 - MAIN.SHOW_CHECK UNSTABLE ON PB2
1ff551e BUG#11790057 KILL QUERY WILL AFFECT THE NEXT COMMAND INCORRECTLY Bug#11754124:KILL QUERY NOT BEHAVING CONSISTENTLY AND WILL HANG IN SOME CASES

[Laurynas Biveinis (laurynas-biveinis)]

     https://github.com/percona/percona-server/pull/986
     https://github.com/percona/percona-server/pull/987

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-3551