xennet0: rx no buf again

[Stephen Borrill <netbsd%precedence.co.uk@localhost>]

This has been discussed previously, but no resolution or workaround was 
proposed:

https://mail-index.netbsd.org/netbsd-users/2012/11/16/msg011969.html
http://mail-index.netbsd.org/port-xen/2015/03/28/msg008611.html

I'm seeing this with 7.0_RC3 amd64. The VM in question has NMBCLUSTERS set 
to 32768, but has 7 NICs attached. When the problem happens, the machine 
is so busy printing the xennetX messages (for they happen on more than one 
NIC), it's hard to do any debugging (cannot type any commands). The 
machine is effectively locked solid.

--
Stephen