--- Begin Message ---
Hi Denis,

There are two length fields in the draft Z-Wave G.9959 TAP Specification. The 
TAP Header contains the length of the TLVs section (all the TLVs) and each TLV 
has it’s own length. It occurred to me that this may be confusing to the reader 
so I have given these different names in the 0.95 draft.

However, I think you are referring to the length of the MPDU itself. The Z-Wave 
G.9959 TAP Specification does not contain a length field for the MPDU because 
there is already a length provided in G.9959. I did not want to duplicate this 
value in the Z-Wave G.9959 TAP Specification as this could be problematic, for 
instance if these two length fields are different how should this be handled? 
To avoid this my preference is to re-use the length already in the Z-Wave MPDU.

To clarify that the MPDU length is contained within the MPDU itself I have 
modified this section to read:
The length of the Z-Wave MPDU Payload is contained within MPDU as defined in 
the Z-Wave MPDU format, see G.9959 [R1] section and figures 8-5 and 
8-6. If needed, the Z-Wave MPDU Payload is padded at the end with zero octets 
to the next whole multiple of 4 octets.

The latest 0.95 draft of Z-Wave G.9959 TAP Specification with these changes is 
available at: https://gitlab.com/exegin/zwave-g9959-tap 

Thank-you very much for your feedback. Does this addresses your concern?

Hopefully this is sufficient to allocate a new DLT.


+ Chris Brandson

> On Feb 6, 2023, at 3:10 PM, Denis Ovsienko via tcpdump-workers 
> <tcpdump-workers@lists.tcpdump.org> wrote:
> From: Denis Ovsienko <de...@ovsienko.info>
> Subject: Re: [tcpdump-workers] Z-Wave G.9959 TAP Specification version 0.94
> Date: February 6, 2023 at 3:10:30 PM PST
> To: tcpdump-workers@lists.tcpdump.org
> On Mon, 12 Dec 2022 17:06:07 -0800
> Chris Brandson via tcpdump-workers <tcpdump-workers@lists.tcpdump.org>
> wrote:
>> Hello Tcp Dump Workers,
>> The latest 0.94 draft Z-Wave G.9959 TAP Specification is available at:
>> https://gitlab.com/exegin/zwave-g9959-tap
>> <https://gitlab.com/exegin/zwave-g9959-tap>
>> This version incorporates feedback from Denis Ovsienko, James Ko, and
>> Sam Freemantle, whose contributions are appreciated.
>> Since there were no signifigant issues identified we requested that a
>> DLT value be assigned to this specification. Once assigned the
>> document will be updated with the DLT value and published as version
>> 1.0.
>> Comments and feedback are still appreciated.
> Hello Chris.
> Thank you for waiting.  I believe the last point I raised about the
> proposed encoding was that the currently specified header design makes
> it non-trivial to access any MPDU data beyond the Length header.
> However, if you find the current design good enough to serve the
> intended purpose, it would be wrong to argue, even more so since the
> header has a version field and can be improved later as and if required.
> Are there any other objections to make this DLT allocation?
> -- 
>    Denis Ovsienko
> _______________________________________________
> tcpdump-workers mailing list
> tcpdump-workers@lists.tcpdump.org
> https://lists.sandelman.ca/mailman/listinfo/tcpdump-workers

--- End Message ---
tcpdump-workers mailing list

Reply via email to