bpo-46476: Fix memory leak in code objects generated by deepfreeze

[kumaraditya303]

https://bugs.python.org/issue46476

[kumaraditya303]

Valgrind log after this PR:

==7131== HEAP SUMMARY:
==7131==     in use at exit: 0 bytes in 0 blocks
==7131==   total heap usage: 18,468 allocs, 18,468 frees, 2,631,556 bytes allocated
==7131== 
==7131== All heap blocks were freed -- no leaks are possible
==7131== 
==7131== Use --track-origins=yes to see where uninitialised values come from
==7131== For lists of detected and suppressed errors, rerun with: -s
==7131== ERROR SUMMARY: 81 errors from 16 contexts (suppressed: 0 from 0)

[bedevere-bot]

A Python core developer has requested some changes be made to your pull request before we can consider merging it. If you could please address their requests along with any other requests in other reviews from core developers that would be appreciated.

Once you have made the requested changes, please leave a comment on this pull request containing the phrase I have made the requested changes; please review again. I will then notify any core developers who have left a review that you're ready for them to take another look at this pull request.

And if you don't make the requested changes, you will be put in the comfy chair!

[tiran]

The new code is causing a segfault.

[kumaraditya303]

I have made the requested changes; please review again

[bedevere-bot]

Thanks for making the requested changes!

@tiran: please review the changes made to this pull request.

[vstinner]

Before reviewing, I checked that the PR fix my issue: https://bugs.python.org/issue46476#msg411315

Yes, it does :-) Python no longer leaks any memory block at exit!

$ ./python -I -X showrefcount -c pass
[-5 refs, 0 blocks]

[vstinner]

[dealloc_codeobject] can be easily moved when there is more need for such function as currently as I don't think there is any so I think this is good enough for now.

My concern is more that this C code is far from the code responsible to handle PyCodeObject, codeobject.c. I would prefer to make it close to code_dealloc() to ease its maintenance.

The python core currently does not have any immortal objects support so it isn't usable in there too.

Even if it's somehow "hidden", there is already _PyObject_IMMORTAL_INIT() in Include/internal/pycore_object.h.

[kumaraditya303]

My concern is more that this C code is far from the code responsible to handle PyCodeObject, codeobject.c. I would prefer to make it close to code_dealloc() to ease its maintenance.

Ok, I moved it to the codeobject.c added a comment as it is for deepfreeze.

Even if it's somehow "hidden", there is already _PyObject_IMMORTAL_INIT() in Include/internal/pycore_object.h.

That is a partial implementation because if an object is immortal then Py_INCREF and Py_DECREF should be no-op for them but that's another topic.

[bedevere-bot]

A Python core developer has requested some changes be made to your pull request before we can consider merging it. If you could please address their requests along with any other requests in other reviews from core developers that would be appreciated.

Once you have made the requested changes, please leave a comment on this pull request containing the phrase I have made the requested changes; please review again. I will then notify any core developers who have left a review that you're ready for them to take another look at this pull request.

[vstinner]

Thanks @kumaraditya303 for the PR and the updates, the final change is even better! I merged your PR.

Ah right, I hesitated to complain about the braces / PEP 7, @tiran did it ;-)

[gvanrossum]

Sorry I missed the review, but I have a question.