Message414058
> Good news, the difference on Windows was easy enough to find, bad news total refs are now negative!
Oh wow. How did you find this leak? Did you read all C files and check for code specific to Windows? How did you proceed? Well spotted!
> #define INIT_ALIAS(NAME, TYPE)
I proposed GH-31594 to fix this macro.
> Strange as well, when using dump_refs, the total refs are much more negative (-12 linux, -13 Windows)
Which command do you type? Do you pass -I option to Python?
With my PR, I get exactly 0 on Linux:
$ ./python -I -X showrefcount -c pass
[0 refs, 0 blocks]
> Note that an allocated block is still leaking.
Right, with my PR, I now get 1 leaked memory block on Windows:
> python -I -X showrefcount -c pass
[0 refs, 1 blocks] |
|
Date |
User |
Action |
Args |
2022-02-26 01:50:52 | vstinner | set | recipients:
+ vstinner, jkloth |
2022-02-26 01:50:52 | vstinner | set | messageid: <1645840252.45.0.915416762124.issue46857@roundup.psfhosted.org> |
2022-02-26 01:50:52 | vstinner | link | issue46857 messages |
2022-02-26 01:50:52 | vstinner | create | |
|