Add some tests for raw sockets.

Update raw socket option handlers as well.

Handle multicast sockets as well.

In D48659#1110770, @emaste wrote:

does sysctl -o still pass all the data to userspace but then only print the first bit?

• Use -o instead of -x: with the latter sysctl(8) will print a huge amount of output and is a couple orders of magnitude slower. We get the same kernel coverage with -o, so just use that.
• Avoid using atf_check: it causes kyua to buffer all of the output in memory and then store it in the test results DB, which is somewhat unwieldy. Check the exit status manually.

In D48666#1109935, @glebius wrote:

The change definitely improves, but some questions remain :) Why do we need to store fib both in socket and inpcb? I believe it belongs only to the inpcb, cause many sockets (unix, netlink, ...) doesn't have a notion of forwarding. The rtsock(4) doesn't have notion of forwarding, but we use it configure other protocols forwarding, hence it needs fib. But it has its own pcb that of course can store fib.

Use RT_ALL_FIBS instead of -1 as a sentinel value.

Add -x to the sysctl flags

This was committed in 44f8e1e8530e1d2e95e84bbbe3d22ac9cb2557fe

This was committed in 629a72376d51aad812d6f1279403bc81c38c35d2

Improve test descriptions.

Go a bit further: add a test which does the same thing in a jail.

It seems more intuitive to specify the requested size first.

I'm not very familiar with netlink nor RPC internals, so it's hard to make useful comments.

Fix MAX_JIFFY_OFFSET, remove an uneeded cast, add an XXX comment for jiffies_64.

This change addresses the failing test, but there's a new failure in sys/netinet/fibs_test:same_ip_multiple_ifaces_fib0. The test creates two interfaces, then assigns the same IP address to each, with different masks. With this change, the second address assignment fails with EEXIST.

Panic after each search for biggest ranges if no range was found at all.