current-users: another rarpd problem

Subject: another rarpd problem
To: None <current-users@sun-lamp.cs.berkeley.edu>
From: Roland McGrath <roland@frob.com>
List: current-users
Date: 12/13/1993 14:03:00

I forgot to mention one other thing I had to "fix" in rarpd.
This one is really cute.

	request.arp_ha.sa_family = AF_UNSPEC;
	/* This is needed #if defined(COMPAT_43) && BYTE_ORDER != BIG_ENDIAN,
	   because AF_UNSPEC is zero and the kernel assumes that a zero
	   sa_family means that the real sa_family value is in sa_len.  */
	request.arp_ha.sa_len = 16; /* XXX */

Luckily SIOCSARP actually ignores this length field (always resets it).

------------------------------------------------------------------------------