svn commit: r357786 - head/sys/netgraph

[Eugene Grosbein]

Author: eugen
Date: Wed Feb 12 00:31:00 2020
New Revision: 357786
URL: https://svnweb.freebsd.org/changeset/base/357786

Log:
  ng_nat: avoid panic if attached directly to ng_ether and got short packet
  
  From the beginning, ng_nat safely assumed cleansed traffic
  because of limited ways it could be attached to NETGRAPH:
  ng_ipfw or ng_ppp only.
  
  Now as it may be attached with ng_ether too, the assumption proven wrong.
  Add needed check to the ng_nat. Thanks for markj for debugging this.
  
  PR:		243096
  Submitted by:	Lutz Donnerhacke <lutz at donnerhacke.de>
  Reported by:	Robert James Hernandez <rob at sarcasticadmin.com>
  Reviewed by:	markj and others
  MFC after:	1 week
  Differential Revision:	https://reviews.freebsd.org/D23091

Modified:
  head/sys/netgraph/ng_nat.c

Modified: head/sys/netgraph/ng_nat.c
==============================================================================
--- head/sys/netgraph/ng_nat.c	Wed Feb 12 00:16:56 2020	(r357785)
+++ head/sys/netgraph/ng_nat.c	Wed Feb 12 00:31:00 2020	(r357786)
@@ -806,11 +806,16 @@ ng_nat_rcvdata(hook_p hook, item_p item )
 		panic("Corrupted priv->dlt: %u", priv->dlt);
 	}
 
+	if (m->m_pkthdr.len < ipofs + sizeof(struct ip))
+		goto send;		/* packet too short to hold IP */
+
 	c = (char *)mtodo(m, ipofs);
 	ip = (struct ip *)mtodo(m, ipofs);
 
-	KASSERT(m->m_pkthdr.len == ipofs + ntohs(ip->ip_len),
-	    ("ng_nat: ip_len != m_pkthdr.len"));
+	if (ip->ip_v != IPVERSION)
+		goto send;		/* other IP version, let it pass */
+	if (m->m_pkthdr.len < ipofs + ntohs(ip->ip_len))
+		goto send;		/* packet too short (i.e. fragmented or broken) */
 
 	/*
 	 * We drop packet when: