[ python-Bugs-808756 ] refleaks in _hotshot.c
SourceForge.net
noreply at sourceforge.net
Tue Aug 3 11:03:19 CEST 2004
Bugs item #808756, was opened at 2003-09-18 18:01
Message generated for change (Comment added) made by arigo
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=105470&aid=808756&group_id=5470
Category: Extension Modules
Group: Python 2.3
>Status: Closed
Resolution: Accepted
Priority: 5
Submitted By: Michael Hudson (mwh)
Assigned to: Armin Rigo (arigo)
Summary: refleaks in _hotshot.c
Initial Comment:
My regrtest patches revealed a few leaks in
Modules/_hotshot.c.
Attached patch seems to fix.
----------------------------------------------------------------------
>Comment By: Armin Rigo (arigo)
Date: 2004-08-03 09:03
Message:
Logged In: YES
user_id=4771
Checked in:
trunk _hotshot.c 1.37
release23-maint _hotshot.c 1.34.10.1
----------------------------------------------------------------------
Comment By: Fred L. Drake, Jr. (fdrake)
Date: 2004-08-02 15:32
Message:
Logged In: YES
user_id=3066
I just looked over Armin's version of the patch, and it
looks good to me. Armin, please go ahead and commit, and
close this report.
This should be backported to 2.3.x as well; I think the
patch should apply cleanly to the branch as well as the trunk.
----------------------------------------------------------------------
Comment By: Michael Hudson (mwh)
Date: 2004-08-02 14:22
Message:
Logged In: YES
user_id=6656
This patch is still necessary, still applies and still fixes the leaks.
Any objections to just checking it in?
----------------------------------------------------------------------
Comment By: Tim Peters (tim_one)
Date: 2003-09-22 13:28
Message:
Logged In: YES
user_id=31435
Just attaching Armin's patch (hotshot2.diff).
----------------------------------------------------------------------
Comment By: Armin Rigo (arigo)
Date: 2003-09-22 12:01
Message:
Logged In: YES
user_id=4771
Unless I'm mistaken there is a misplaced DECREF.
I'd love to attach my patch but cannot do so. Here it is:
http://arigo.tunes.org/hotshot2.diff
A more general note is about using 'N' in Py_BuildValue. Its
implementation seems to be buggy because it does not always
consume a reference: it won't do so if an error occurs
before it gets there. I guess I'll submit a separate bug
report/patch for this too.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=105470&aid=808756&group_id=5470
More information about the Python-bugs-list
mailing list