gnrc_ipv6: Forward multicast packets even if they are registered with the receiving netif by DipSwitch · Pull Request #4528 · RIOT-OS/RIOT

DipSwitch · 2015-12-20T14:17:20Z

Currently if you send a broadcast to ff05::1 for example a routing node doesn't forward the packed if the router itself has the address assigned to it's receiving interface.

Thoughts on the subject are welcome :)

miri64 · 2015-12-20T14:24:53Z

sys/net/gnrc/network_layer/ipv6/gnrc_ipv6.c

This ignores the flags of a multicast address. ff12::1 would also be a legal link-local address.

Learned something new. Fixed it

miri64 · 2015-12-20T14:27:03Z

Fixes #4527

miri64 · 2015-12-20T14:42:46Z

Another problem is, that we currently don't implement MLP, so this configuration will most likely flood a network.

DipSwitch · 2015-12-20T14:45:49Z

Would I also need to change the >= to > since forwarding link-local speaks against itself...

miri64 · 2015-12-20T14:46:56Z

Yes, thanks for pointing that out

miri64 · 2015-12-20T14:47:26Z

oh and while we're are at nitpicky stuff: the module you're changing is gnrc_ipv6 not gnrc_sixlowpan

DipSwitch · 2015-12-20T14:56:05Z

What does MLP stand for? Multinode Link Layer Protocol?

And sorry about that, first thought it was 6Low specific.

cgundogan · 2015-12-20T15:07:16Z

What does MLP stand for? Multinode Link Layer Protocol?

I believe @authmillenon was referring to MLD

miri64 · 2015-12-20T15:17:31Z

Yes, sorry for the confusion.

DipSwitch · 2015-12-20T15:30:37Z

Np, I have build an IGMP querier for Linux before. This seems like fun to implement :D

OlegHahm · 2016-02-27T18:56:08Z

@authmillenon, what's the state here?

miri64 · 2016-02-28T05:50:45Z

I have not tested it yet, but code-wise it looks alright. Will test at the Hack'n'ACK :-)

miri64 · 2016-03-01T17:40:00Z

Wait a minute: what if the router has the address assigned to one of its interfaces. Then it won't forward it though it might make sense.

miri64 · 2016-03-06T20:06:50Z

@DipSwitch ping?

DipSwitch · 2016-03-06T22:13:01Z

Whit this PR it would forward it? That was my intention at leased. But maybe I misunderstood what you said.

miri64 · 2016-03-07T10:36:50Z

Ah okay, yes. I forgot about the ||... But shouldn't it also receive it?

DipSwitch · 2016-03-07T12:04:43Z

Yeah, I thought it tries to process it anyway. Will fix tonight. :)
On 7 Mar 2016 11:36, "Martine Lenders" notifications@github.com wrote:

Ah okay, yes. I forgot about the ||... But shouldn't it also receive it?

—
Reply to this email directly or view it on GitHub
#4528 (comment).

DipSwitch · 2016-03-24T20:57:06Z

Rebased, squashed the rest, and fixed to also processing the packet ourselfs part.

miri64 · 2016-03-29T15:09:51Z

Will test tonight.

miri64 · 2016-03-29T17:36:47Z

sys/net/gnrc/network_layer/ipv6/gnrc_ipv6.c


+static inline bool _pkt_none_local_mcast(ipv6_hdr_t *hdr)
+{
+    return (hdr->dst->u8[0] == 0xff) &&


s/dst->u8/dst.u8/

DipSwitch · 2016-03-29T20:34:03Z

Addressed, rebased, squashed, pushed

miri64 · 2016-03-29T21:45:04Z

When I try to ping an added ff0e::1 on another node the receiver hard faulted due to a failed assertion in packet buffer.

> ifconfig 6 add ff0e::1
ifconfig 6 add ff0e::1
success: added ff0e::1/64 to interface 6
> 0x49dd7a1
*** RIOT kernel panic:
FAILED ASSERTION.

    pid | name                 | state    Q | pri | stack ( used) | location   
      1 | idle                 | pending  Q |  15 |  8192 ( 6240) | 0x807c2c0 
      2 | main                 | pending  Q |   7 | 12288 ( 9312) | 0x80792c0 
      3 | pktdump              | bl rx    _ |   6 | 12288 ( 9312) | 0x808aac0 
      4 | ipv6                 | running  Q |   4 |  8192 ( 6240) | 0x8086d60 
      5 | udp                  | bl rx    _ |   5 |  8192 ( 6240) | 0x808dae0 
      6 | gnrc_netdev2_tap     | bl rx    _ |   4 |  8192 ( 6240) | 0x8084d40 
        | SUM                  |            |     | 57344 (43584)

*** halted.


Program received signal SIGSEGV, Segmentation fault.
0x00000081 in ?? ()
(gdb) where
#0  0x00000081 in ?? ()
#1  0x00000001 in ?? ()
#2  0x0806d374 in ?? ()
#3  0x08058ff0 in _release_error_locked (pkt=0x50, err=0) at /home/martine/Repositories/RIOT-OS/RIOT/sys/net/gnrc/pktbuf_static/gnrc_pktbuf_static.c:224
#4  0x080590a3 in gnrc_pktbuf_release_error (pkt=0x80892a0 <_pktbuf>, err=0)
    at /home/martine/Repositories/RIOT-OS/RIOT/sys/net/gnrc/pktbuf_static/gnrc_pktbuf_static.c:243
#5  0x0804fece in gnrc_pktbuf_release (pkt=0x80892a0 <_pktbuf>) at /home/martine/Repositories/RIOT-OS/RIOT/sys/include/net/gnrc/pktbuf.h:163
#6  0x08051002 in _receive (pkt=0x80892a0 <_pktbuf>) at /home/martine/Repositories/RIOT-OS/RIOT/sys/net/gnrc/network_layer/ipv6/gnrc_ipv6.c:889
#7  0x08050393 in _event_loop (args=0x0) at /home/martine/Repositories/RIOT-OS/RIOT/sys/net/gnrc/network_layer/ipv6/gnrc_ipv6.c:237
#8  0xf7ddef0b in makecontext () from /usr/lib32/libc.so.6
#9  0x00000000 in ?? ()
(gdb)

DipSwitch · 2016-03-29T22:11:26Z

Last time I tried this was 20 December probably, and I don't have tranceivers at home yet. Moving to next release. :)

OlegHahm · 2016-04-01T22:37:20Z

Hm, since this is a bugfix, we can still merge - and I would like to see this problem solved.

jnohlgard · 2016-04-02T06:08:08Z

sys/net/gnrc/network_layer/ipv6/gnrc_ipv6.c

    }
 }

+static inline bool _pkt_none_local_mcast(ipv6_hdr_t *hdr)


"none" -> "non" or "not"

…with the receiving netif

miri64 · 2016-05-31T19:42:37Z

Please rebase.

miri64 · 2016-06-06T10:54:01Z

Please rebase, we need this fix!

kYc0o · 2016-07-12T08:57:09Z

Can this fix come into the release? Maybe is it related to #5596 ?

OlegHahm · 2016-07-12T13:56:22Z

If @DipSwitch is unavailable right now, we could adopt it.

kYc0o · 2016-07-18T12:48:49Z

Can you @OlegHahm ?

OlegHahm · 2016-07-18T12:55:36Z

Not this week.

kYc0o · 2016-07-18T13:02:47Z

OK, so I need to postpone it.

miri64 · 2016-08-05T12:35:02Z

I go ahead and adapt it myself. Can you review @kYc0o?

kYc0o · 2016-08-05T12:42:55Z

I can try to test, I'm supposed to be off today but I'll have some time.

miri64 · 2016-08-05T12:47:07Z

kk

miri64 · 2016-08-05T12:47:17Z

Need not be today ;-)

kYc0o · 2016-08-05T12:48:34Z

Let's do it then ;-)

miri64 · 2016-08-05T12:54:41Z

See #5729.

DipSwitch added Type: bug The issue reports a bug / The PR fixes a bug (including spelling errors) Area: network Area: Networking Discussion: RFC The issue/PR is used as a discussion starting point about the item of the issue/PR labels Dec 20, 2015

DipSwitch assigned miri64 Dec 20, 2015

DipSwitch mentioned this pull request Dec 20, 2015

gnrc_ipv6: Multicast is not forwarded if routing node listens to the address #4527

Open

miri64 reviewed Dec 20, 2015
View reviewed changes

miri64 changed the title ~~gnrc_sixlowpan: Forward multicast packets even if they are registered with the receiving netif~~ gnrc_ipv6: Forward multicast packets even if they are registered with the receiving netif Dec 20, 2015

OlegHahm added the Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties label Feb 27, 2016

DipSwitch force-pushed the pr/fix_6lo_multicast_route branch from 8f4a78b to be18502 Compare March 24, 2016 20:49

OlegHahm added this to the Release 2016.04 milestone Mar 25, 2016

miri64 reviewed Mar 29, 2016
View reviewed changes

DipSwitch force-pushed the pr/fix_6lo_multicast_route branch from 345e9ec to 203b147 Compare March 29, 2016 20:33

DipSwitch modified the milestones: Release 2016.07, Release 2016.04 Mar 29, 2016

miri64 mentioned this pull request Apr 1, 2016

gnrc ipv6: multicast packets are not dispatched to the upper layers #5230

Open

jnohlgard reviewed Apr 2, 2016
View reviewed changes

gnrc_sixlowpan: Forward multicast packets even if they are registerd …

88532c0

…with the receiving netif

DipSwitch force-pushed the pr/fix_6lo_multicast_route branch from 203b147 to 88532c0 Compare April 17, 2016 15:06

kYc0o removed the Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties label Jul 12, 2016

kYc0o modified the milestones: Release 2016.10, Release 2016.07 Jul 18, 2016

miri64 mentioned this pull request Aug 5, 2016

gnrc_ipv6: Forward multicast packets addressed to self #5729

Closed

miri64 closed this Aug 5, 2016

Conversation

DipSwitch commented Dec 20, 2015

Uh oh!

miri64 Dec 20, 2015

Choose a reason for hiding this comment

Uh oh!

DipSwitch Dec 20, 2015

Choose a reason for hiding this comment

Uh oh!

miri64 commented Dec 20, 2015

Uh oh!

miri64 commented Dec 20, 2015

Uh oh!

DipSwitch commented Dec 20, 2015

Uh oh!

miri64 commented Dec 20, 2015

Uh oh!

miri64 commented Dec 20, 2015

Uh oh!

DipSwitch commented Dec 20, 2015

Uh oh!

cgundogan commented Dec 20, 2015

Uh oh!

miri64 commented Dec 20, 2015

Uh oh!

DipSwitch commented Dec 20, 2015 via email

Uh oh!

OlegHahm commented Feb 27, 2016

Uh oh!

miri64 commented Feb 28, 2016

Uh oh!

miri64 commented Mar 1, 2016

Uh oh!

miri64 commented Mar 6, 2016

Uh oh!

DipSwitch commented Mar 6, 2016

Uh oh!

miri64 commented Mar 7, 2016

Uh oh!

DipSwitch commented Mar 7, 2016

Uh oh!

DipSwitch commented Mar 24, 2016

Uh oh!

miri64 commented Mar 29, 2016

Uh oh!

miri64 Mar 29, 2016

Choose a reason for hiding this comment

Uh oh!

DipSwitch commented Mar 29, 2016

Uh oh!

miri64 commented Mar 29, 2016

Uh oh!

DipSwitch commented Mar 29, 2016

Uh oh!

OlegHahm commented Apr 1, 2016

Uh oh!

jnohlgard Apr 2, 2016

Choose a reason for hiding this comment

Uh oh!

DipSwitch Apr 17, 2016

Choose a reason for hiding this comment

Uh oh!

miri64 commented May 31, 2016

Uh oh!

miri64 commented Jun 6, 2016

Uh oh!

kYc0o commented Jul 12, 2016

Uh oh!

OlegHahm commented Jul 12, 2016

Uh oh!

kYc0o commented Jul 18, 2016

Uh oh!

OlegHahm commented Jul 18, 2016

Uh oh!

kYc0o commented Jul 18, 2016

Uh oh!

miri64 commented Aug 5, 2016

Uh oh!

kYc0o commented Aug 5, 2016

Uh oh!