[Python-bugs-list] [ python-Bugs-485152 ] memory leak in test_scope
noreply@sourceforge.net
noreply@sourceforge.net
Thu, 06 Dec 2001 07:35:12 -0800
Bugs item #485152, was opened at 2001-11-24 11:45
You can respond by visiting:
http://sourceforge.net/tracker/?func=detail&atid=105470&aid=485152&group_id=5470
Category: Python Interpreter Core
Group: Python 2.2
Status: Open
Resolution: None
Priority: 5
Submitted By: Neal Norwitz (nnorwitz)
Assigned to: Jeremy Hylton (jhylton)
Summary: memory leak in test_scope
Initial Comment:
test_scope leaks memory
see attached file for details
----------------------------------------------------------------------
>Comment By: Jeremy Hylton (jhylton)
Date: 2001-12-06 07:35
Message:
Logged In: YES
user_id=31392
And that's it!
The leak is in the code that maps between the frame and the
locals dict when calling the trace function.
----------------------------------------------------------------------
Comment By: Jeremy Hylton (jhylton)
Date: 2001-12-06 07:14
Message:
Logged In: YES
user_id=31392
This one is strange!
I've attached a new file (smaller.py) that reproduces the
leak more quickly than test_scope.py. I've been unable to
simplify further. The leak seems to occur in the presence
of three things:
- an installed trace function
- a string format operation
- a nested lambda that causes the string to be a cell
----------------------------------------------------------------------
Comment By: Neal Norwitz (nnorwitz)
Date: 2001-12-06 07:08
Message:
Logged In: YES
user_id=33168
Attached a new file that was built with
current CVS and w/o optimization so
the line #s should be good
----------------------------------------------------------------------
Comment By: Jeremy Hylton (jhylton)
Date: 2001-12-06 06:48
Message:
Logged In: YES
user_id=31392
Weird. The leak is reported in PyString_FormatString(),
which is called because of a string interpolation call in
test_scope: "_%s__%s" % (klass.__name__, name). That's the
only use of % on a string in the module.
Unfortunately, the line numbers in the Purify output don't
match the current CVS version of stringobject.c. Neal--
Would it be hard to do a new run with the current CVS?
----------------------------------------------------------------------
Comment By: Tim Peters (tim_one)
Date: 2001-11-27 12:04
Message:
Logged In: YES
user_id=31435
Assigned to Barry.
----------------------------------------------------------------------
You can respond by visiting:
http://sourceforge.net/tracker/?func=detail&atid=105470&aid=485152&group_id=5470