Wireshark-bugs: [Wireshark-bugs] [Bug 5284] new_packet_list: redissection + redraw crashes when
Date: Thu, 11 Oct 2012 03:42:48 -0700 (PDT)
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=5284

--- Comment #33 from Evan Huus <eapache@xxxxxxxxx> 2012-10-11 03:42:47 PDT ---
(In reply to comment #32)
> (In reply to comment #31)
> > (In reply to comment #30)
> > > Created attachment 9334 [details]
> > > Alternative patch for 1.8 branch
> > > 
> > > Using trunk:
> > > I can't reproduce the problem on r45450
> > > (but I agree there's a problem)
> > > 
> > > Using wireshark-1.8 branch:
> > > I can reproduce with reverted edt ref counting.
> > > I can't reproduce the problem with attached patch (patch based on comment #3
> > > and #26)
> > I am seeing exactly the same thing. Based on the original analysis by Eduard,
> > my best guess is that this was fixed accidentally by the custom packet byte
> > field, since it's no longer triggering the same GTK callbacks as the old one.
> > I don't know if it will be possible to find a way to reproduce this
> > consistently (or at all) now.
> 
> Would the best soulution be to backport the custom packet byte widget to not
> let the code divert to much making mantenance harder?

I don't think so. I understand that back-porting the byte list in the first
place would be a non-trivial amount of work, and 1.8 doesn't suffer from the
same bug that makes ref-counting a problem in trunk. Honestly, if we can figure
out for sure why it seems to be fixed in trunk then I'm happy leaving well
enough alone and just closing the bug.

-- 
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
You are watching all bug changes.