Wireshark · Wireshark-dev: Re: [Wireshark-dev] Patch: Prepare for some cleanups of 802.11ad

[Richard Sharpe <realrichardsharpe@xxxxxxxxx>]

On Sat, Aug 22, 2015 at 12:53 PM, Guy Harris <guy@xxxxxxxxxxxx> wrote:
>
> On Aug 22, 2015, at 12:17 PM, Richard Sharpe <realrichardsharpe@xxxxxxxxx> wrote:
>
>> Hmmm, looking at this again, I wonder if it should not contain an MCS
>> field, because the radiotap dissector thinks that mean
>> PHDR_802_11_PHY_11N.
>
> Well, *if* there's useful radio information for 802.11ad that isn't provided by the MCS or VHT fields, perhaps there should be a DMG field, containing the radio information for 802.11ad, and 11ad captures should have that field rather than MCS or VHT.
>
> Note that MCS and VHT both include certain fields, so presumably the idea is that an 11ac packet doesn't need both MCS *and* VHT fields.  If there's an DMG field that includes all radio information useful for 11ad, an 11ad capture would only need a DMG field, not an MCS or VHT field, and, for radiotap captures, a packet containing a DMG field could be treated as being an 11ad packet.  (Wireshark should probably warn if a packet contains more than one of MCS, VHT, or DMG.)
>
> Note that the A-MPDU information has a separate field, which could be used in 11n, 11ac, or 11ad frames.

Yes, I see that now. My captures do not seem to contain any aggregate MPDUs.

Also, I now see that sections 9.7.5a.4 and 9.7.5a.5 imply that MCS
headers are appropriate for an 802.11ad capture, so the radiotap
dissector will need to change to use the frequency set the PHY type.

-- 
Regards,
Richard Sharpe
(何以解憂？唯有杜康。--曹操)