Alexander Mohr added the comment:
so I just discovered that the object that has the zero refcount has the same
tp_dealloc:
(gdb) print *FROM_GC(gc)->ob_type
$8 = {ob_base = {ob_base = {_ob_next = 0x7f80c8aafc88, _ob_prev =
0x7f80c8aafd00, ob_refcnt = 7, ob_type = 0x7f80cd8c86e0 <PyType_Type>}, ob_size
= 0}, tp_name = 0x7f80c8aa5c38 "_GatheringFuture", tp_basicsize = 104,
tp_itemsize = 0,
tp_dealloc = 0x7f80cd41bee7 <subtype_dealloc>, tp_print = 0x0, tp_getattr =
0x0, tp_setattr = 0x0, tp_as_async = 0x556ba4342d58, tp_repr = 0x7f80c8870a9b
<FutureObj_repr>, tp_as_number = 0x556ba4342d70, tp_as_sequence =
0x556ba4342ea8,
tp_as_mapping = 0x556ba4342e90, tp_hash = 0x7f80cd54c995 <_Py_HashPointer>,
tp_call = 0x0, tp_str = 0x7f80cd423d05 <object_str>, tp_getattro =
0x7f80cd3fec8b <PyObject_GenericGetAttr>, tp_setattro = 0x7f80cd3ff0b5
<PyObject_GenericSetAttr>,
tp_as_buffer = 0x556ba4342ef8, tp_flags = 808449, tp_doc = 0x7f80c8cd7380
"Helper for gather().\n\n This overrides cancel() to cancel all the children
and act more\n like Task.cancel(), which doesn't immediately mark itself
as\n cancelled.\n ",
tp_traverse = 0x7f80cd41baae <subtype_traverse>, tp_clear = 0x7f80cd41bd5c
<subtype_clear>, tp_richcompare = 0x7f80cd423d42 <object_richcompare>,
tp_weaklistoffset = 96, tp_iter = 0x7f80c8871d4f <future_new_iter>,
tp_iternext = 0x7f80cd3fe6d6 <_PyObject_NextNotImplemented>, tp_methods =
0x0, tp_members = 0x556ba4342f28, tp_getset = 0x0, tp_base = 0x7f80c8a7b3c0
<FutureType>, tp_dict = 0x7f80c8aafc88, tp_descr_get = 0x0, tp_descr_set = 0x0,
tp_dictoffset = 88,
tp_init = 0x7f80cd431000 <slot_tp_init>, tp_alloc = 0x7f80cd41b7f8
<PyType_GenericAlloc>, tp_new = 0x7f80cd41b9c6 <PyType_GenericNew>, tp_free =
0x7f80cd583d8b <PyObject_GC_Del>, tp_is_gc = 0x0, tp_bases = 0x7f80c8ab20c0,
tp_mro = 0x7f80c8aafd00,
tp_cache = 0x0, tp_subclasses = 0x0, tp_weaklist = 0x7f80c8aae5d8, tp_del =
0x0, tp_version_tag = 791, tp_finalize = 0x7f80c8870ddb <FutureObj_finalize>}
This is for a GatheringFuture, something tells me perhaps there is more to this
function that needs to be resolved?
----------
_______________________________________
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue26617>
_______________________________________
_______________________________________________
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
