Huge thanks to our Platinum Members Endace and LiveAction,
and our Silver Member Veeam, for supporting the Wireshark Foundation and project.

Wireshark-bugs: [Wireshark-bugs] [Bug 6921] GMR-1: Add dissections for various GMR 04.08 element

Date: Thu, 8 Mar 2012 03:02:01 -0800 (PST)
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6921

--- Comment #6 from Anders Broman <anders.broman@xxxxxxxxxxxx> 2012-03-08 03:02:00 PST ---
(In reply to comment #5)
> (In reply to comment #3)
> > Could you please attach a sample capture for testing purposes?
> 
> Well this adds the dissection code, but it's not called from anywhere just yet.
> I have two upcoming patches that will use it:
>  - CCCH message dissection
>  - DTAP handler that will call it for RR messages and delegate to GSM dissector
> for most CC/MM mesages.
> 
> 
> (In reply to comment #4)
> > Wouldn't it be much better to use ELEM_MAND_TLV and friends from
> > packet-gsm_a_common.h? At a glance some IEs looks to be shared with
> > packet-gsm_a_rr.c
> > and could be reused from there rather than duplicate value_strings etc.
> 
> In RR a lot of the IEs have the same name but are different things. Some are
> really common tough, like "spare nibble" or "TMSI", but a lot others are subtly
> different (like LAC which is also 16 bits, but had a different meaning, or PD
> which can have different values as well, ...)
> 
> I'm looking into using the gsm tlv code now. I have a working version but I
> need to check everything works as expected.
> 
> I'm not a big fan of it because of some of the legacy hacks/quirks but fixing
> those would be a huge amount of work due to the number of callers ... (things
> like the 'signed' cast, the fact some stuff that should really be const are
> not, the mixed types gint/guint/guint32 used for the same value thorough the
> processing path, the need to put all the tables in the global namespace instead
> of just one 'get_params' function, ...)

Well the advantage of having the same format is that duplication of dissection
can be avoided, if things should be fixed in the legacy code we should probably
try to do that. Could a workable path be to rename the legaly stuff to "old"
introduce new fixed functions and convert the legacy stuff piece by piece?

Any way I think it's a waste to introduce new IE dissection functions for GMR
and not beeing able to share common functionality, but I don't realy know how
much is actually shared one could also forse that more stuff will be shared in
the future.

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