Wireshark-bugs: [Wireshark-bugs] [Bug 9933] Questionable calling of ethernet dissector by encaps
Date Index · Thread Index · Other Months · All Mailing Lists
Date Prev · Date Next · Thread Prev · Thread Next
From: bugzilla-daemon@xxxxxxxxxxxxx
Date: Mon, 18 Jan 2016 01:34:49 +0000

changed bug 9933


What Removed Added
Status RESOLVED CONFIRMED
Resolution FIXED ---
Ever confirmed   1

Comment # 31 on bug 9933 from 
At this point, the only dissectors still using the "eth" dissector are:

    the ATM dissector, for VC-encapsulated Ethernet, where we really *don't*
know whether there's an FCS or not;

    the Cisco FabricPath MiM dissector, where some more work is necessary -
this dissector is an override for dissection of Ethernet frames with the
FabricPath MiM Ethernet type, to dissect the addresses specially, and ideally
the FCS should be handled by the regular Ethernet dissector, so we really need
a better hook into the Ethernet dissector;

    the EPON dissector, which uses its own LINKTYPE_ value, so I need to ask
the person who asked for the LINKTYPE_ value and submitted the dissector
whether packets using that linotype should always have an FCS, never have an
FCS, or do as LINKTYPE_ETHERNET does and have packets of both types.

The first of those should use a "eth_maybefcs" dissector that does *NOT* look
at the pseudo-header.  The second of those needs work that will probably
involve the Ethernet dissector making the "FCS or not" decision.  The third of
those needs clarification and perhaps its own pseudo-header in case any capture
file formats *other* than pcap and pcapng offer EPON (or maybe the "FCS length"
needs to be provided via a different path, given that pcapng can provide the
FCS length).

You are receiving this mail because:
  • You are watching all bug changes.