WRT the failing ATF tests (some of them)

[Robert Elz <kre%munnari.OZ.AU@localhost>]

A bunch of the net yet fixed, relatively recent ATF test failures are
caused by:

	rn_init: radix functions require max_keylen be set

T%he code that sets max_keylen in net/radix.c is (from rn_init()) ...

#ifdef _KERNEL
        struct domain *dp;
 
        if (radix_initialized)
                panic("radix already initialized"); 
        radix_initialized = 1;
 
        DOMAIN_FOREACH(dp) {
                if (dp->dom_maxrtkey > max_keylen)
                        max_keylen = dp->dom_maxrtkey;
        }
#endif


Note the #ifdef _KERNEL

That means that that code is absent in rump (while it pretends to be
a kernel it isn't one).  Without it, nothing ever sets max_keylen
(it remains 0) and the immediately following code in rn_init():

        if (max_keylen == 0) {
                log(LOG_ERR,
                    "rn_init: radix functions require max_keylen be set\n");
                return;
        }

unsurprisingly fires.

I will leave it to someone who has a clue what is supposed to be
happening here to fix this one.

kre