Created on 2009-11-08 00:00 by pitrou, last changed 2009-11-09 16:54 by pitrou. This issue is now closed.
| Files | ||||
|---|---|---|---|---|
| File name | Uploaded | Description | Edit | |
| rlock_leak.patch | pitrou, 2009-11-08 00:00 | |||
| rlock_leak2.patch | pitrou, 2009-11-09 11:59 | |||
| Messages (3) | |||
|---|---|---|---|
| msg95029 - (view) | Author: Antoine Pitrou (pitrou) *  |
Date: 2009-11-08 00:00 | |
When you use an RLock from a thread which wasn't created through the high-level `threading` API (but, for example, using thread.start_new_thread()), the RLock creates a DummyThread object by calling current_thread(), and that DummyThread is immortal (as well as its attributes etc.). This implementation wart can easily lead to reference leaks for unsuspecting users. One simple solution is not to call current_thread() in RLock, but use get_ident() instead. Patch attached. |
|||
| msg95074 - (view) | Author: Antoine Pitrou (pitrou) * |
Date: 2009-11-09 11:59 | |
An updated patch with test. |
|||
| msg95089 - (view) | Author: Antoine Pitrou (pitrou) * |
Date: 2009-11-09 16:54 | |
Committed in r76172-r76175. |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2009-11-09 16:54:48 | pitrou | set | status: open -> closed resolution: fixed messages: + msg95089 stage: patch review -> committed/rejected |
| 2009-11-09 11:59:17 | pitrou | set | files:
+ rlock_leak2.patch messages: + msg95074 |
| 2009-11-08 00:00:39 | pitrou | create | |