Mechanically convert if_hn(4) to DrvAPI

[zlei]

sys/dev/hyperv/netvsc/if_hn.c

944

in if_llmaddr_count() there's check against AF_LINK, is this line change identical ?

u_int   
if_llmaddr_count(if_t ifp)
{       
...
        CK_STAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link)
                 if (ifma->ifma_addr->sa_family == AF_LINK)
                         count++;
...
        return (count);
}

1347

Set flags to wrong if. Should be VF.

1912–1913

"in transparent VF mode.\n" is truncated.

1935–1936

Should use if_name()

1953–1954

if_getinputfn()

2057

2088

SYSCTL_ADD_UINT(ctx, child, OID_AUTO, "tso_max",
	    CTLFLAG_RD, &ifp->if_hw_tsomax, 0, "max TSO size");

tso_max sysctl node is readonly, is if_sethwtsomax() really needed ?

2107

Ditto

2126

Ditto

4091

CC @whu
Originally use atomic_clear_int(), is the change sufficient ?

5786

Atomics @whu

[jhibbits]

sys/dev/hyperv/netvsc/if_hn.c

944

You're right. Maybe I need a new API, if_has_multi(), for this.

2088

Oops, I misread the sysctl declaration.

[jhibbits]

Update diff after (several) rebases. Intended to do this 2 months ago. I hope I've addressed everything.

[zlei]

Generally looks good to me, once D39493 is committed.

sys/dev/hyperv/netvsc/if_hn.c

944

2094

2106