Connection details ------------------ * Date: May 30, 2018 * Time: 8-9AM DST, 17:00 CEST: https://www.worldtimebuddy.com/?qm=1&lid=2988507,1816670,5391959,5128581&h=2988507&date=2018-05-30&sln=17-18 * Webex Link: https://cisco.webex.com/ciscosales/j.php?MTID=m356ee15ed34e05dfa9c876815312e1a2 * Meeting number: 202 477 894 * Meeting password: QprtWqpe (77789773 from phones) * Join from a video conferencing system or application * Dial 202477894@cisco.webex.com * Join by phone * +1-866-432-9903 Call-in toll-free number (US/Canada) * +1-408-525-6800 Call-in toll number (US/Canada) * Access code: 202 477 894 * * Global call-in numbers: * https://cisco.webex.com/ciscosales/globalcallin.php?serviceType=MC&ED=410481357&tollFree=1 Agenda ------ * [17:05] Administrivia [10min] o Note-Well, Scribes, Agenda Bashing o Status of drafts * [17:15] SCHC padding - Dominique [20min] * [17:35] SCHC Tickets and Discussed options - Ana + Laurent [20min] * [17:55] AOB [ 5min] Minute takers ------------- * Ana Minaburo * Pascal Thubert * Dominique Barthel * Carles Gomez * Alexander Pelov Attendees --------- * Ana Minaburo * Pascal Thubert * Dominique Barthel * Laurent Toutain * Carles Gomez * Juan-Carlos Zuniga * Alex Pelov * Edgar Ramos * Arun Kandasamy * Flavien Moullec * Diego Dujovne Past Attendees --------- * Alex Pelov * Felipe Díaz-Sánchez * Ana minaburo * Dominique Barthel * Laurent Toutain * Carles Gomez * Ivaylo Petrov * Juan-Carlos Zuniga * Julien Catalano * Orne Brocaar * Pascal Thubert * Vijay Gharge * Orne Brocaar * Paventhan Arumugam * Paul Duffy * Diego Dujovne * Pascal Thubert * Alex Pelov * Laurent Toutain * Ana Minaburo * Vanessa Valderrama * Dominique Barthel * Julien Catalano * Edgar Ramos * Carles Gomez * J Sanchez * Ramon Sanchez * Juan-Carlos Zuniga * Felipe Díaz-Sánchez Action Items ---------------------------- Chairs: find reviewers for drafts Agenda ------ * [17:05] Administrivia [10min] o Note-Well, Scribes, Agenda Bashing o Status of drafts * [17:15] SCHC padding - Dominique [20min] * [17:35] SCHC Tickets and Discussed options - Ana + Laurent [20min] * [17:55] AOB [ 5min] Minutes ------- * [17:05] Administrivia [10min] o Note-Well, Scribes, Agenda Bashing o Status of drafts Alex give the presentation of the meeting Minutes of the last meeting are approved Next IETF Meeting request 2h slot * No more 2.5h slots * We need to make recharter in this time * Agree in topics to be discussed * Only the main points * List of conflicts, please send a mail before this friday to chairs if you have a strong feeling * Actions Items: * Milestones has been updated * We need to find reviewers for the drafts * Adoption of the Technology Specific Draft (TSD) * Authors of TSD should read the CoAP and the IP/UDP drafts * Review need to be done before end of July Pascal: This could help to better unsderstand what do you need for the TSD Alex: Contact the technology authors to ask for review. Ana: ticket 18 has been closed and now it is ticket 27 for the CoAP draft Ticket 18 is closed. * [17:15] SCHC padding - Dominique [20min] Laurent: Welcome Dominique as author of the draft Pascal: Chairs ask Dominique to be author as he has contributed Dominique: This is an idea reach after a discussion with Laurent that can make the text and the mechanisms more simple * current situation: Padding is done at most twice, * Proposal is to do it at most once before transmission, 7 bits max * Figure slide 10: * Byte padding after compression (fill the last byte). If no fragmentation is needed, then OK. If fragmentation is needed, then the last fragment can be with added padding * Figure slide 11 showing the packet format, with the padding P1 at the end of the SCHC Packet and P1 and P2 after the last fragment * Only the last fragment may need additionnal padding. Other fragments don't need added Padding, because of clever treatment of these cases by the existing draft. * at reception, since payload is byte aligned, the paddind can be removed, works for reassembly and decompression * Proposal mechanisms is to pad before transmission, once after SCHC packet because fragmentation is not needed and SCHC Packet needs padding or * If you do fragment and get this weird size fragment, then you pad * The receiver reassembly the fragments of the buffer and we don't care about the P1 bits, the Decompressor will drop them when decompressing the packet * The problem is that now MIC needs to be computed with the padding in order that the MIC check is correct * Notion of byte with a L2 Word to define byte * There are places that we do define what a "byte" is. So, this notion appears from time to time. The definition of L2 Word can mean a byte, or a bit, or something else. * L2 Word may be bit or byte, the SCHC will works good over both without confussing the description * The MIC will work better (because no need to add padding) * * SCHC must know the L2 word for each underlying technology * Dominique: MIC and padding bits should be defined somewhere. Suggested to be in the technology specific document * Laurent: With CRC, the position doesn't matter (?) * Laurent: We are transmitting the padding anyways, so there is no need to specify them explicitly * Dominique: some lines may prefer a specific value for padding bits * Laurent: better to force them to be zeros or to be ones * Dominique: question is who decides the values for such padding bits? * Dominique: what to do for MIC over padding? Should the padding be covered by the MIC or not? (if yes - simplified treating. if yes - there must be loopback between decompressor and MIC) * Laurent MIC detects that we lost a fragment and not an error because L2 CRC is protecting against the error * Pascal but we need that MIC replace the UDP checksum that will be compressed, we need a proof that this checks * Laurent: MIC and padding are in the All-1 frag, so they can be generated at the "same time" * Pascal: is there a way to make the padding self-describe its length? * Alex: like this idea, go with the simplest possible solution, including tha padding is not difficutl and the second point is how much time the current implementations need to change this * Dominique: the edit is already done (branch on Github). See diff at https://github.com/lp-wan/ip-compression/pull/20/files *Flavien we need bit stream instead of array: this is not a major change * Dominique (slide 19): discussion on layering: "kind of" layer violation, but padding bits are not really processed * Alex: just checking that changes in the document and in implementations are not majors * Ana: fixed the pull request in the github, so we can go forward. * Alex: I don't see big issues on the single-padding proposal, send a mail on the mailing list to announce the changes. * [17:35] SCHC Tickets and Discussed options - Ana + Laurent [20min] * Alex: no much time... Can you focus no tickets that may require more discussion? * Ana: 12 open tickets. Most of them will be closed * Ticket 10: it was possible to have different fragments in different bearers (?). It is possible with DTAG, but is out of scope for SCHC/UDP/IP. This could be added in technology specific documents. * Ticket 12: Padding (lots of discussions) - Dominique presentation. It will be finished to discuss after mailing list * Ticket 14: Legacy devices - this needs to be studied / defined in other document (not ot SCHC UDP/IP). Possible item for rechartering * Ticket 20: Definition of 'Byte boundary' and it's representation is to use "Next Byte Boundary" (instead of Byte Boundary) * Laurent: do we use L2 Word instead? * * Alex: if it is more clear (L2 word), go ahead. Send an email to the WG list, and if there are no objections, go ahead. * Ticket 21: It was necessary to update ticket 15 (the last fragment is 1 bit shorter). * Dominique: are you sure that's what we've decided? * Ana: yes,.. * Dominique: not sure it's what he remembers * Dominique: are the bitmaps changing size? * Ana: no, the header changes size * Can the bitmap be reduced? No? But how does the receiver know that the C-bit was not there? * Dominique, Ana and Laurent take this discussion offline. * Alex: great progress. How many tickets are yet to be closed? Big discussions left? * Laurent: sent an email on the Time Scale option on the CoRE WG list * [17:55] AOB [ 5min]