[MERGE] BTreeIndex intern()

Michael Hudson michael.hudson at canonical.com
Thu Apr 9 01:46:32 BST 2009


John Arbash Meinel wrote:
> Michael Hudson wrote:
>> John Arbash Meinel wrote:
> 
>>> - From what I can tell, I think we have a small memory leak somewhere,
>>> because the current 'WorkingSize' doesn't match the size of objects I
>>> can access using PyMemoryDump. I feel like it is something introduced
>>> recently, but I haven't found a great way to actually debug it yet.
>> valgrind is pretty good at this, I think...
> 
>> Cheers,
>> mwh
> 
> 
> I tried and the output was swamped with:
> 
> ==566== Use of uninitialised value of size 4
> ==566==    at 0x808C047: PyObject_Free (obmalloc.c:929)
> ==566==    by 0x810C25B: code_dealloc (codeobject.c:260)
> ==566==    by 0x8099EA8: tupledealloc (tupleobject.c:170)
> ...
> 
> Now maybe I just need to suppress all PyObject_Free statements...

Ah yeah... there's a suppressions file in the Python source
(Misc/vagrind.supp or something?).

Cheers,
mwh



More information about the bazaar mailing list