gnrc_ipv6_nib: port to gnrc_netif2

[miri64]

This ports the NIB to the new network interface API (since I don't want to make the unnecessary effort to port the old neighbor discovery to gnrc_netif2 I did it this way around).

Depends on #7014 (merged) and #7424 (merged).

This PR is part of the network layer remodelling effort:

[miri64]

All later PRs after #7014 in the NIB series I will port directly to gnrc_netif2, so all NDP-related parts will be based on this PR.

[miri64]

Rebased and adapted to current master and current dependencies

[miri64]

Rebased to current master and dependencies.

[miri64]

Rebased to current master and current dependencies (again...)

[miri64]

Oh boy rebasing this will be .... fun...

[miri64]

Rebased to current master and dependencies.

[miri64]

Addressed comments

[bergzand]

Thanks!

Should this already result in a functional gnrc_networking example? From what I've tested so far, it refuses to do ndp.

[miri64]

Should this already result in a functional gnrc_networking example? From what I've tested so far, it refuses to do ndp.

Link-local should work with this PR, global and everything involved with routers (except routing protocols) with #7479.

[bergzand]

Link-local should work with this PR, global and everything involved with routers (except routing protocols) with #7479.

Okay, then everything behaves as expected :)

You've got my ACK, code looks good to me overall, but I'm not going to claim that I have enough insight in the subtleties of the NDP code to spot the more difficult functional errors.

[miri64]

Wait! I think I need to fix something I messed up in the rebase of #7424.

[miri64]

Let me squash my work so far and work from there.

[miri64]

I somehow managed to forget about the embargo on Monday and merged #7722. That's what is causing the conflict in #7925. #7960 should take care of it so let's see what other issues Murdock reveals here ;-)

[miri64]

I added two commits to fix some issues in applications I overlooked in #7424. I hope it is alright to add them here instead of adding an extra PR.

@bergzand, please re-ACK

[bergzand]

As long as it is not yet squashed, I don't mind the extra fixes here. Makes my work easier if I can trust the CI output on these PRs again :)

I have time to recheck this PR this evening.

[miri64]

As long as it is not yet squashed, I don't mind the extra fixes here. Makes my work easier if I can trust the CI output on these PRs again :)

That won't be the case. #7414, #7417, and #7462 are also required to build properly. And there is some nasty bug in tests/gnrc_ipv6_ext (i.e. it does not work... not sure this is the case because RPL isn't ported to gnrc_netif2 yet)... can't fix that now however. I will give a follow-up for that

[miri64]

I decided to take out the application fixes after all (see #7966 for that). As I said: they won't fix the Murdock state and need some work anyway.

[bergzand]

ACK

[bergzand]

I decided to take out the application fixes after all (see #7966 for that). As I said: they won't fix the Murdock state and need some work anyway.

Okay, still looks good and the static-test is now also (almost) happy with the PR.

[miri64]

Squashed

[miri64]

Ok. All errors seem to be unrelated. Let's merge this and fix the rest later.