ANNOUNCEMENT: Live Wireshark University & Allegro Packets online APAC Wireshark Training Session
April 17th, 2024 | 14:30-16:00 SGT (UTC+8) | Online

Wireshark-bugs: [Wireshark-bugs] [Bug 2453] segmentation fault with wslua script

Date: Wed, 30 Apr 2008 00:10:33 -0700 (PDT)
http://bugs.wireshark.org/bugzilla/show_bug.cgi?id=2453


Márton Németh <nm127@xxxxxxxxxxx> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Attachment #1738|review_for_checkin?         |
               Flag|                            |
Attachment #1738 is|0                           |1
           obsolete|                            |
   Attachment #1758|                            |review_for_checkin?
               Flag|                            |




--- Comment #18 from Márton Németh <nm127@xxxxxxxxxxx>  2008-04-30 00:10:28 GMT ---
Created an attachment (id=1758)
 --> (http://bugs.wireshark.org/bugzilla/attachment.cgi?id=1758)
remove the "outstanding_stuff" handling completely

I would remove the "outstanding_stuff" completely. I tend to think that
implementing the __gc method can clear all the data. In pushXxx(L, xxx)
functions (see epan/wslua/wslua.h) a new userdata is created with the function
lua_newuserdata(). Then the pointer is actually not checked if it is NULL or
not, but this is an other story. So we have a pointer and use this pointer to
fill the content of the buffer.

In the __gc function this area could be filled with zeros, if it is needed at
all.

What is not clear for me that what should be the pair of the
luaL_getmetatable(), lua_setmetatable(), and the "push_code" parameter in __gc.


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