Thursday, June 23, 2016< ^ >
mcr/credil has set the subject to: IETF96 ROLL WG meeting
Room Configuration
Room Occupants

[07:10:56] eintopf joins the room
[08:00:18] <eintopf> mcr: I send you something for adding short address dump functionality
[08:00:22] <eintopf> for linux and iproute2
[09:12:57] eintopf leaves the room: Disconnected: closed
[10:50:25] eintopf joins the room
[13:22:25] eintopf leaves the room: Disconnected: closed
[15:48:55] eintopf joins the room
[17:28:56] <mcr> thanks!
[17:34:31] <eintopf> I don't know if this will be part of real UAPI, I like to have a 6lowpan netsted nl attr in NDA_PRIVATE
[17:35:04] <eintopf> it's all in my UAPI for 6LoWPAN mail which I sent recently, but nobody will reply on that - for me is that "just do it"
[17:35:32] <eintopf> after sending patches somebody will scream if I do stupid things there
[17:36:15] <eintopf> mcr: one question is also, do I filtering the 0x8000 in kernelspace  or shall I filter it in userspace?
[17:36:32] <mcr> hmm.
[17:36:33] <eintopf> because 0x8000 short addr will den not be set in netlink
[17:36:42] <eintopf> I think that's fine but could be useful for debugging
[17:37:01] <mcr> NDA_PRIVATE... should be able to get a type defined for netlink, no?
[17:37:51] <eintopf> NDA_PRIVATE is a nested attribute depends on dev->type e.g. ARPHRD_6LOWPAN
[17:37:57] <mcr> I think, filter in kernelspace. If 0x8000 set, then maybe "do something", yes, like debugging.
[17:38:23] <eintopf> if you like to parse it in userspace you need to figure out the dev->type at first
[17:38:33] <eintopf> and running different netlink parsing stuff
[17:38:42] <eintopf> the same is NDA_DST
[17:38:55] <eintopf> because on AF_INET6 NDA_DST is ipv6 address
[17:39:00] <eintopf> on AF_INET it's ipv4
[17:39:05] <eintopf> it's similar
[17:39:10] <eintopf> but not the same :-)
[17:39:12] <mcr> Ask Jamal's opinion maybe. Because FORCES should have a way to assign you a value.
[17:39:50] <eintopf> what's FORCES?
[17:39:51] <mcr> yeah, but we know it's an address...
[17:40:11] <mcr>
[17:40:15] <mcr> netlink over internet.
[17:40:29] <eintopf> ohhh
[17:40:43] <eintopf> I already thought about if something does exists
[17:40:54] <eintopf> and how to handle different host byte order things
[17:41:12] <mcr> forces solves this already....
[17:41:24] <eintopf> ah, so I don't need to think about that :-)
[17:41:35] <mcr> exactly.
[17:41:59] <eintopf> okay, but back to 0x8000 - then I change it to don't filter it on kernelside
[17:42:05] <mcr> Are you near Berlin for the IETF95 and plugfest beforehand?
[17:42:11] <mcr> next month.
[17:42:18] <eintopf> every people ask me that
[17:42:21] <eintopf> but I don't know
[17:42:31] <eintopf> I need to ask my boss, because I can't pay it myself
[17:43:27] <mcr> so, the plugfest is no cost.
[17:43:28] <eintopf> Jamal told me IETF meetings are boring
[17:43:35] <eintopf> :P
[17:43:45] <mcr> They are boring for Jamal.
[17:44:02] <mcr> He also tends to keep to himself and lot, and tries to be a good muslim and not drink anymore.
[17:44:16] <mcr> (not the case back in 1997...)
[17:44:32] <eintopf> do you know Werner Almesberger?
[17:44:40] <eintopf> linux-atm maintainer?
[17:45:04] <eintopf> he told me that I need to told Jamal that Jamal should take many drinks
[17:45:18] <eintopf> because they drunk a lot (what I heard from Werner)
[17:45:26] <eintopf> but was in the 90th
[17:45:26] <mcr> I'm sure that I've met him some years ago, but we aren't on a first name, see him across a busy bar to recognize him.
[17:46:06] <eintopf>
[17:46:10] <eintopf> he did this hardware
[17:46:29] <eintopf> which is also support in mainline linux as one of the only one usb 802154 solutions
[17:46:34] <mcr> IETF also has a Day pass system, and a student rate.
[17:47:04] <eintopf> yea, I will told that my boss that I get some discount as student
[20:23:40] eintopf leaves the room: Disconnected: closed
Powered by ejabberd - robust, scalable and extensible XMPP server Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!