[Lldb-commits] [PATCH] Fixup some issues with Python shutdown / finalization
zturner at google.com
Fri Aug 8 14:59:30 PDT 2014
I just didn't revert the changes to dotest.py. I do agree that it would be
nice to figure out what, if any problem there is with atexit(), but TBH my
python (and even moreso my embedded python) is not the strongest, so I'm
kind of handicapped in that regard, and I also think it just makes for
easier debugging and easier to rationalize about the code-flow without
them. Regarding your original comment about the proposed patch removing
the cleanup for the crashinfo.so, that atexit handler was actually not even
doing anything (empty function).
Let me know if you want me to look into this more though.
On Fri, Aug 8, 2014 at 1:23 PM, Enrico Granata <egranata at apple.com> wrote:
> Do we still get crashes because of the usage of atexit(), or did you
> simply not revert the changes to dotest.py?
> With the exception of those, the rest of your patch LGTM
> On Aug 8, 2014, at 11:19 AM, Zachary Turner <zturner at google.com> wrote:
> Got rid of the Release() method and wrapped the destructor-based decrefs
> inside of Py_IsInitialized() checks.
> lldb-commits mailing list
> lldb-commits at cs.uiuc.edu
> *- Enrico*
> 📩 egranata@.com ☎️ 27683
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the lldb-commits