Re: [ipwave] Mirja Kühlewind's Discuss on draft-ietf-ipwave-ipv6-over-80211ocb-49: (with DISCUSS and COMMENT)

[Alexandre Petrescu <[e-mail]>]

Le 09/07/2019 à 23:29, Nabil Benamar a écrit :
> Hi Mirja,
> 
> Thank you for your review and comments.
> 
> You raised a very important point that was discussed extensively on the 
> ML and then we asked the IEEE 802.11 members (thanks to Dorothy Stanly) 
> to provide us with a review to help us clarify this point.
> 
> Here is what we got from them:
> 
> . *Suggest to simply state that the data is transmitted with “User 
> Priority” of Background (numerically 1 or 2)*, and leave the internal 
> details of how this is accomplished to the 802.11 specification.
> 
> User Priority is typically described as a simple integer (not a binary 
> value), and the mapping of this User Priority to TID header value is 
> another 802.11 detail, best left to the 802.11 specification. For 
> example: in the 802.11 specification the TID field is specified to be *4 
> bits* in the header.  The use of these 4 bits to carry the User Priority 
> information is an internal specification of 802.11 and potentially 
> subject to change.
> 
> Suggest using terminology from the MAC SAP in IEEE Std 802.11-2016 
> Clause 5.2.  This clause intentionally abstracts the exact details of 
> 802.11’s internal operation, while describing specifically the behavior 
> required by the user. For example, the following text:
> 
> “In the 802.11 header, the value of the Subtype sub-field in the Frame 
> Control field MUST be set to 8 (i.e. 'QoS Data'); the value of the 
> Traffic Identifier (TID) sub-field of the QoS Control field of the 
> 802.11 header MUST be set to binary 001 (i.e.  User Priority 
> 'Background', QoS Access Category 'AC_BK').”

On my side I do not agree with it.

Only the expensive non open source implementations implement 802.11 QoS 
Data headers, including the QoS fields you suggest (background user 
priority).

The cheap open source implementations, including the ones during 
HAckathon, do not exhibit QoS headers in wireshark packet dumps in 
monitor mode.

I think I will have to write another draft called IPv6-over-OCB-without 
QoS : that is what is implemented in open source and cheap implementations.

There is however interoperability: an RA sent by an IP-RSU with QoS 
header is understood well by an IP-OBU who does not understand QoS 
headers; vice-versa as well.

The fact that QoS headers are there in the specificatioin of IP over OCB 
comes from implementers that do not share the code publicly.  It is for 
implementations that are much more expensive and that try to suggest a 
higher quality is present.  However, I strongly doubt the quality 
improvements.

Quality here comes from many people doing something and the best to be 
elected naturally; it does not come from commercial interest.

Alex
> 
> _could be replaced by:_
> 
> “*The mapping to the 802.11 data service MUST use a ‘priority’ value of 
> 1, which specifies the use of QoS with a “Background” user priority*.”
> 
> 
> Thanks again.
> 
> 
> On Tue, Jul 9, 2019 at 5:43 PM Mirja Kühlewind via Datatracker 
> <[e-mail] <mailto:[e-mail]>> wrote:
> 
>     Mirja Kühlewind has entered the following ballot position for
>     draft-ietf-ipwave-ipv6-over-80211ocb-49: Discuss
> 
>     When responding, please keep the subject line intact and reply to all
>     email addresses included in the To and CC lines. (Feel free to cut this
>     introductory paragraph, however.)
> 
> 
>     Please refer to
>     https://www.ietf.org/iesg/statement/discuss-criteria.html
>     for more information about IESG DISCUSS and COMMENT positions.
> 
> 
>     The document, along with other ballot positions, can be found here:
>     https://datatracker.ietf.org/doc/draft-ietf-ipwave-ipv6-over-80211ocb/
> 
> 
> 
>     ----------------------------------------------------------------------
>     DISCUSS:
>     ----------------------------------------------------------------------
> 
>     One point on this sentence, which I believe was also commented in
>     the TSV-ART
>     review (Thanks Jörg!):
> 
>     sec 4.2: "The mapping to the 802.11 data service MUST use a
>         'priority' value of 1, which specifies the use of QoS with a
>         'Background' user priority."
>     I don't think this should be a MUST requirement. I assume the
>     assumption here
>     is that IP traffic is always some "random" data that is less
>     important than
>     other V2V communication. However, this is a generic mapping document
>     and should
>     therefore probably not make such an assumption (or at least it would
>     need to be
>     spelled out).
> 
> 
>     ----------------------------------------------------------------------
>     COMMENT:
>     ----------------------------------------------------------------------
> 
>     One editorial high level comment: I seams like all text that was
>     somehow deemed
>     as out fo scope for the main body of this document got stuffed into the
>     appendix. Please consider removing what is really not needed in this
>     document
>     as these pages also take review and RFC Editor time, especially as
>     they seem to
>     have received less review and therefore have more nits.
> 
>     nit: sec 4.5.2 s/in OCB mode.A  A future improvement/in OCB mode. A
>     future
>     improvement/
> 
> 
> 
> 
> -- 
> 
> Best Regards
> 
> Nabil Benamar
> Associate Professor
> Department of Computer Sciences
> School of Technology
> Moulay Ismail University
> Meknes. Morocco
> 
> 
> 
> _______________________________________________
> its mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/its
>