Commit 9619def9 authored by Tim Peters's avatar Tim Peters

Collector 1309: ZODB.DB.DB.cacheExtremeDetail reports wrong reference count

The refcount reported for a ghost was one too small, because the code for
ghosts and non-ghosts was the same, and presumably "it's a feature" that
the code for non-ghosts deliberately lies about the true Python refcount,
in order (guessing here) to report what the count would have been if the
cPickleCache didn't exist.  But while the cPickleCache holds on to a real
reference to non-ghost objects, it does not hold a real reference to ghost
objects, so subtracting "an extra" count for all objects made it appear
that non-referenced ghosts exist in the cache (which doesn't actually
happen).

What a tangled web we weave ...
parent c35a1295
...@@ -9,6 +9,8 @@ fsrecover.py couldn't work, because it referenced attributes that no ...@@ -9,6 +9,8 @@ fsrecover.py couldn't work, because it referenced attributes that no
longer existed after the MVCC changes. Repaired that, and added new longer existed after the MVCC changes. Repaired that, and added new
tests to ensure it continues working. tests to ensure it continues working.
Collector #1309: The reference counts reported by DB.cacheExtremeDetails()
for ghosts were one too small.
What's new in ZODB3 3.3 alpha 3 What's new in ZODB3 3.3 alpha 3
......
...@@ -13,7 +13,7 @@ ...@@ -13,7 +13,7 @@
############################################################################## ##############################################################################
"""Database objects """Database objects
$Id: DB.py,v 1.78 2004/04/23 17:26:37 gintautasm Exp $""" $Id: DB.py,v 1.79 2004/05/07 19:10:59 tim_one Exp $"""
import cPickle, cStringIO, sys import cPickle, cStringIO, sys
from thread import allocate_lock from thread import allocate_lock
...@@ -243,29 +243,39 @@ class DB(object): ...@@ -243,29 +243,39 @@ class DB(object):
return detail return detail
def cacheExtremeDetail(self): def cacheExtremeDetail(self):
detail=[] detail = []
conn_no = [0] # A mutable reference to a counter conn_no = [0] # A mutable reference to a counter
def f(con, detail=detail, rc=sys.getrefcount, conn_no=conn_no): def f(con, detail=detail, rc=sys.getrefcount, conn_no=conn_no):
conn_no[0] = conn_no[0] + 1 conn_no[0] += 1
cn = conn_no[0] cn = conn_no[0]
for oid, ob in con._cache_items(): for oid, ob in con._cache_items():
id='' id = ''
if hasattr(ob,'__dict__'): if hasattr(ob, '__dict__'):
d=ob.__dict__ d = ob.__dict__
if d.has_key('id'): if d.has_key('id'):
id=d['id'] id = d['id']
elif d.has_key('__name__'): elif d.has_key('__name__'):
id=d['__name__'] id = d['__name__']
module = getattr(ob.__class__, '__module__', '') module = getattr(ob.__class__, '__module__', '')
module = module and '%s.' % module or '' module = module and ('%s.' % module) or ''
# What refcount ('rc') should we return? The intent is
# that we return the true Python refcount, but as if the
# cache didn't exist. This routine adds 3 to the true
# refcount: 1 for binding to name 'ob', another because
# ob lives in the con._cache_items() list we're iterating
# over, and calling sys.getrefcount(ob) boosts ob's
# count by 1 too. So the true refcount is 3 less than
# sys.getrefcount(ob) returns. But, in addition to that,
# the cache holds an extra reference on non-ghost objects,
# and we also want to pretend that doesn't exist.
detail.append({ detail.append({
'conn_no': cn, 'conn_no': cn,
'oid': oid, 'oid': oid,
'id': id, 'id': id,
'klass': "%s%s" % (module, ob.__class__.__name__), 'klass': "%s%s" % (module, ob.__class__.__name__),
'rc': rc(ob)-4, 'rc': rc(ob) - 3 - (ob._p_changed is not None),
'state': ob._p_changed, 'state': ob._p_changed,
#'references': con.references(oid), #'references': con.references(oid),
}) })
......
...@@ -205,7 +205,7 @@ class LRUCacheTests(CacheTestBase): ...@@ -205,7 +205,7 @@ class LRUCacheTests(CacheTestBase):
# XXX The above gc.collect call is necessary to make this test # XXX The above gc.collect call is necessary to make this test
# pass. # pass.
# #
# This test then only works because the other of computations # This test then only works because the order of computations
# and object accesses in the "noodle" calls is such that the # and object accesses in the "noodle" calls is such that the
# persistent mapping containing the MinPO objects is # persistent mapping containing the MinPO objects is
# deactivated before the MinPO objects. # deactivated before the MinPO objects.
...@@ -230,12 +230,15 @@ class LRUCacheTests(CacheTestBase): ...@@ -230,12 +230,15 @@ class LRUCacheTests(CacheTestBase):
self.assertEqual(count, CONNS) self.assertEqual(count, CONNS)
for details in self.db.cacheExtremeDetail(): for details in self.db.cacheExtremeDetail():
# one dict per object. keys: # one 'details' dict per object
if details['klass'].endswith('PersistentMapping'): if details['klass'].endswith('PersistentMapping'):
self.assertEqual(details['state'], None) self.assertEqual(details['state'], None)
else: else:
self.assert_(details['klass'].endswith('MinPO')) self.assert_(details['klass'].endswith('MinPO'))
self.assertEqual(details['state'], 0) self.assertEqual(details['state'], 0)
# The cache should never hold an unreferenced ghost.
if details['state'] is None: # i.e., it's a ghost
self.assert_(details['rc'] > 0)
class StubDataManager: class StubDataManager:
def setklassstate(self, object): def setklassstate(self, object):
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment