[bugzilla-daemon@xxxxxxxxxxxxx]

http://bugs.wireshark.org/bugzilla/show_bug.cgi?id=1430





------- Comment #35 from anders.blomdell@xxxxxxxxxxxxxx  2007-10-03 21:35 GMT -------
(In reply to comment #34)
> Created an attachment (id=977)
 --> (http://bugs.wireshark.org/bugzilla/attachment.cgi?id=977&action=view) [details]
> SVN diff
> 
> packet-ams.c did not rebuild for me for some obscure reason. Once having done a
> clean and attempting to build again I got the same problem as Anders did (of
> course).
> Corrected now.
OK, now it compiles. My initial comments:

---

The 'E88A4' part in details view only contains 2 bytes, but marks all bytes,
why not move those 2 bytes into the EtherCAT protocol part (or AMS, Mailbox,
etc)? E88A4 is also not very descriptive...

---

Since many EtherCAT frames contains many frames, and there are usually many
ethernet frames en a capture, I find lines with 'Info' fields like:

NPWR#4099[4]@502, BRD[2]@130, APRD#5[2]@130, APRD#4[2]@130, APRD#3[2]@130,
APRD#2[2]@130, APRD#1[2]@130

More useful than lines with 'Info' fields like:

7 Cmds, SumLen 16, 'FPWR'... 

Since with the latter one has to look at details view to find out what is going
on (test with the dump on the Wiki, and try to find find out what goes on in
packets 23 - 32).

---

The summaries for each telegram in the detail view is much better in the new
version.

---

Somebody (who knows them) has to document ams, ecatmb, ioraw, and nv protocols
on the wiki (and put up dumps)...

---

I'm personally not very keen on the 

  typedef struct Txxx
  {
  } xxx, *Pxxx;

style, but since airpcap already uses it, I guess it is OK :-)


---

What is still lacking in both (and a big challenge) is trying to track node
state (address assignment), CANopen conversations, EEPROM reads, etc. Probably
far beyond what wireshark should do since one has to have a full capture from
power on to do a proper analysis, and also keeps state for all nodes (I has
done some python hacks that analyzes  XML dumps from wireshark, but still very
rudimentary). Better leave that...

---

Summary:

I don't have any knowledge of ams, ecatmb, ioraw or nv protocols, and hence
can't test them.

I would like to get more info in the 'Info' field for multi-telegram packets,
but my version is not optimal either (just realized that node addressing is
decimal, and memory addressing is hexadecimal in my Info field, confusing at
best :-( ...)

OK for me to commit it, but if nobody else does, I will probably hack the Info
field the next time I need to do some serious EtherCAT debugging...


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