diff options
| author | Dmitry Lenev <dlenev@mysql.com> | 2010-02-10 18:46:03 +0300 |
|---|---|---|
| committer | Dmitry Lenev <dlenev@mysql.com> | 2010-02-10 18:46:03 +0300 |
| commit | f229ac073ab52741b3660b468f2ed4aba4e27adf (patch) | |
| tree | 6c8f7502b8b58b23eb754e7a3508781afa2d9436 /sql/mdl.cc | |
| parent | 73ef80f47cd4eda0b93b0f532846a8e16e021699 (diff) | |
| download | mariadb-git-f229ac073ab52741b3660b468f2ed4aba4e27adf.tar.gz | |
Fix for bug #50998 "Deadlock in MDL code during test
rqg_mdl_stability".
When start of statement's waiting on a metadata lock
created more than one loop in waiters graph server might
have entered deadlock condition.
The problem was that in the case described above MDL deadlock
detector had to perform several searches for deadlock but
forgot to reset Deadlock_detection_context before performing
new search.
Failure to do so has broken assumption in code resposible for
choosing victim that if Deadlock_detection_context::victim
is set we also have read lock on m_waiting_for_lock for this
context. As result this lock could have been unlocked more
times than it was acquired which corrupted rwlock's state
which led to server deadlock.
This fix ensures that such reset is done before each attempt
to find a deadlock.
Diffstat (limited to 'sql/mdl.cc')
| -rw-r--r-- | sql/mdl.cc | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/sql/mdl.cc b/sql/mdl.cc index 54931f89e40..e48ae8a3f59 100644 --- a/sql/mdl.cc +++ b/sql/mdl.cc @@ -1755,10 +1755,15 @@ bool MDL_context::find_deadlock(Deadlock_detection_context *deadlock_ctx) bool MDL_context::find_deadlock() { - Deadlock_detection_context deadlock_ctx(this); - while (1) { + /* + The fact that we use fresh instance of deadlock_ctx for each + search performed by find_deadlock() below is important, code + responsible for victim selection relies on this. + */ + Deadlock_detection_context deadlock_ctx(this); + if (! find_deadlock(&deadlock_ctx)) { /* No deadlocks are found! */ |