Message237134
| Author | vstinner |
|---|---|
| Recipients | Lothsahn, Mike.Meyer, alex, benjamin.peterson, cvrebert, jcea, maizy, meador.inge, neologix, pitrou, rbcollins, vstinner |
| Date | 2015-03-03.13:25:45 |
| SpamBayes Score | -1.0 |
| Marked as misclassified | Yes |
| Message-id | <1425389146.08.0.652423699016.issue13697@psf.upfronthosting.co.za> |
| In-reply-to |
| Content | |
|---|---|
> The attached test script demonstrates the issue on Python 2.6 and 3.2, and code inspection suggests this is still valid for 2.7 and 3.4. I disagree that Python 3.4 is affected: RLock has been reimplemented in C in Python 3.2. Only the Python implementation of RLock has the bug, but it's not used by fault, you have to explicitly use a private attribute of the threading module to get it. Python 2.6 only accept security fixes, so in fact only Python 2.7 may get a fix. The question is if it worth to backport 300 lines of C code from Python 3 to Python 2 in the _thread module. @Benjamin, release manager of Pythn 2.7: What do you think? I may help to backport the C implementation of RLock. |
|
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2015-03-03 13:25:46 | vstinner | set | recipients: + vstinner, jcea, pitrou, rbcollins, benjamin.peterson, alex, cvrebert, meador.inge, neologix, Mike.Meyer, Lothsahn, maizy |
| 2015-03-03 13:25:46 | vstinner | set | messageid: <1425389146.08.0.652423699016.issue13697@psf.upfronthosting.co.za> |
| 2015-03-03 13:25:46 | vstinner | link | issue13697 messages |
| 2015-03-03 13:25:45 | vstinner | create | |