MySQL Bugs: #42558: maria: race condition in deadlock detector

Bug #42558	maria: race condition in deadlock detector
Submitted:	2 Feb 2009 20:58	Modified:	26 Feb 2009 17:33
Reporter:	Sergei Golubchik	Email Updates:
Status:	Won't fix	Impact on me:	None
Category:	MySQL Server: Maria storage engine	Severity:	S3 (Non-critical)
Version:		OS:	Any
Assigned to:	Sergei Golubchik	CPU Architecture:	Any

View
Add Comment
Files
Developer
Edit Submission
View Progress Log
Contributions

Description:
there is a possible race condition in deadlock detector when there's a custom killing strategy (not LATEST, that is). deadlock() can signal a victim when the latter isn't waiting on pthread_cond_timedwait() and wakeup signal will be lost.

How to repeat:
.

No, it's actually safe. The caller has mutexes that prevent blockers to disappear unexpectedly, they are held over the wt_* function calls and will prevent signals from being lost (exactly like a mutex over pthread_cond_wait does).