1 files changed, 258 insertions, 314 deletions

diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 0711f92d6a1..8d67900aa00 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -7,8 +7,6 @@
  *	Andi Kleen		<ak@muc.de>
  *	Alexey Kuznetsov	<kuznet@ms2.inr.ac.ru>
  *
- *	$Id: exthdrs.c,v 1.13 2001/06/19 15:58:56 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
@@ -16,7 +14,7 @@
  */
 
 /* Changes:
- *	yoshfuji		: ensure not to overrun while parsing 
+ *	yoshfuji		: ensure not to overrun while parsing
  *				  tlv options.
  *	Mitsuru KANDA @USAGI and: Remove ipv6_parse_exthdrs().
  *	YOSHIFUJI Hideaki @USAGI  Register inbound extension header
@@ -27,12 +25,14 @@
 #include <linux/types.h>
 #include <linux/socket.h>
 #include <linux/sockios.h>
-#include <linux/sched.h>
 #include <linux/net.h>
 #include <linux/netdevice.h>
 #include <linux/in6.h>
 #include <linux/icmpv6.h>
+#include <linux/slab.h>
+#include <linux/export.h>
 
+#include <net/dst.h>
 #include <net/sock.h>
 #include <net/snmp.h>
 
@@ -43,65 +43,23 @@
 #include <net/ndisc.h>
 #include <net/ip6_route.h>
 #include <net/addrconf.h>
-#ifdef CONFIG_IPV6_MIP6
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
 #include <net/xfrm.h>
 #endif
 
 #include <asm/uaccess.h>
 
-int ipv6_find_tlv(struct sk_buff *skb, int offset, int type)
-{
-	int packet_len = skb->tail - skb->nh.raw;
-	struct ipv6_opt_hdr *hdr;
-	int len;
-
-	if (offset + 2 > packet_len)
-		goto bad;
-	hdr = (struct ipv6_opt_hdr*)(skb->nh.raw + offset);
-	len = ((hdr->hdrlen + 1) << 3);
-
-	if (offset + len > packet_len)
-		goto bad;
-
-	offset += 2;
-	len -= 2;
-
-	while (len > 0) {
-		int opttype = skb->nh.raw[offset];
-		int optlen;
-
-		if (opttype == type)
-			return offset;
-
-		switch (opttype) {
-		case IPV6_TLV_PAD0:
-			optlen = 1;
-			break;
-		default:
-			optlen = skb->nh.raw[offset + 1] + 2;
-			if (optlen > len)
-				goto bad;
-			break;
-		}
-		offset += optlen;
-		len -= optlen;
-	}
-	/* not_found */
- bad:
-	return -1;
-}
-
 /*
  *	Parsing tlv encoded headers.
  *
- *	Parsing function "func" returns 1, if parsing succeed
- *	and 0, if it failed.
+ *	Parsing function "func" returns true, if parsing succeed
+ *	and false, if it failed.
  *	It MUST NOT touch skb->h.
  */
 
 struct tlvtype_proc {
 	int	type;
-	int	(*func)(struct sk_buff **skbp, int offset);
+	bool	(*func)(struct sk_buff *skb, int offset);
 };
 
 /*********************
@@ -110,13 +68,11 @@ struct tlvtype_proc {
 
 /* An unknown option is detected, decide what to do */
 
-static int ip6_tlvopt_unknown(struct sk_buff **skbp, int optoff)
+static bool ip6_tlvopt_unknown(struct sk_buff *skb, int optoff)
 {
-	struct sk_buff *skb = *skbp;
-
-	switch ((skb->nh.raw[optoff] & 0xC0) >> 6) {
+	switch ((skb_network_header(skb)[optoff] & 0xC0) >> 6) {
 	case 0: /* ignore */
-		return 1;
+		return true;
 
 	case 1: /* drop packet */
 		break;
@@ -125,83 +81,105 @@ static int ip6_tlvopt_unknown(struct sk_buff **skbp, int optoff)
 		/* Actually, it is redundant check. icmp_send
 		   will recheck in any case.
 		 */
-		if (ipv6_addr_is_multicast(&skb->nh.ipv6h->daddr))
+		if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr))
 			break;
 	case 2: /* send ICMP PARM PROB regardless and drop packet */
 		icmpv6_param_prob(skb, ICMPV6_UNK_OPTION, optoff);
-		return 0;
-	};
+		return false;
+	}
 
 	kfree_skb(skb);
-	return 0;
+	return false;
 }
 
 /* Parse tlv encoded option header (hop-by-hop or destination) */
 
-static int ip6_parse_tlv(struct tlvtype_proc *procs, struct sk_buff **skbp)
+static bool ip6_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb)
 {
-	struct sk_buff *skb = *skbp;
-	struct tlvtype_proc *curr;
-	int off = skb->h.raw - skb->nh.raw;
-	int len = ((skb->h.raw[1]+1)<<3);
+	const struct tlvtype_proc *curr;
+	const unsigned char *nh = skb_network_header(skb);
+	int off = skb_network_header_len(skb);
+	int len = (skb_transport_header(skb)[1] + 1) << 3;
+	int padlen = 0;
 
-	if ((skb->h.raw + len) - skb->data > skb_headlen(skb))
+	if (skb_transport_offset(skb) + len > skb_headlen(skb))
 		goto bad;
 
 	off += 2;
 	len -= 2;
 
 	while (len > 0) {
-		int optlen = skb->nh.raw[off+1]+2;
+		int optlen = nh[off + 1] + 2;
+		int i;
 
-		switch (skb->nh.raw[off]) {
-		case IPV6_TLV_PAD0:
+		switch (nh[off]) {
+		case IPV6_TLV_PAD1:
 			optlen = 1;
+			padlen++;
+			if (padlen > 7)
+				goto bad;
 			break;
 
 		case IPV6_TLV_PADN:
+			/* RFC 2460 states that the purpose of PadN is
+			 * to align the containing header to multiples
+			 * of 8. 7 is therefore the highest valid value.
+			 * See also RFC 4942, Section 2.1.9.5.
+			 */
+			padlen += optlen;
+			if (padlen > 7)
+				goto bad;
+			/* RFC 4942 recommends receiving hosts to
+			 * actively check PadN payload to contain
+			 * only zeroes.
+			 */
+			for (i = 2; i < optlen; i++) {
+				if (nh[off + i] != 0)
+					goto bad;
+			}
 			break;
 
 		default: /* Other TLV code so scan list */
 			if (optlen > len)
 				goto bad;
 			for (curr=procs; curr->type >= 0; curr++) {
-				if (curr->type == skb->nh.raw[off]) {
-					/* type specific length/alignment 
-					   checks will be performed in the 
+				if (curr->type == nh[off]) {
+					/* type specific length/alignment
+					   checks will be performed in the
 					   func(). */
-					if (curr->func(skbp, off) == 0)
-						return 0;
+					if (curr->func(skb, off) == false)
+						return false;
 					break;
 				}
 			}
 			if (curr->type < 0) {
-				if (ip6_tlvopt_unknown(skbp, off) == 0)
-					return 0;
+				if (ip6_tlvopt_unknown(skb, off) == 0)
+					return false;
 			}
+			padlen = 0;
 			break;
 		}
 		off += optlen;
 		len -= optlen;
 	}
+
 	if (len == 0)
-		return 1;
+		return true;
 bad:
 	kfree_skb(skb);
-	return 0;
+	return false;
 }
 
 /*****************************
   Destination options header.
  *****************************/
 
-#ifdef CONFIG_IPV6_MIP6
-static int ipv6_dest_hao(struct sk_buff **skbp, int optoff)
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
+static bool ipv6_dest_hao(struct sk_buff *skb, int optoff)
 {
-	struct sk_buff *skb = *skbp;
 	struct ipv6_destopt_hao *hao;
 	struct inet6_skb_parm *opt = IP6CB(skb);
-	struct ipv6hdr *ipv6h = (struct ipv6hdr *)skb->nh.raw;
+	struct ipv6hdr *ipv6h = ipv6_hdr(skb);
 	struct in6_addr tmp_addr;
 	int ret;
 
@@ -212,7 +190,7 @@ static int ipv6_dest_hao(struct sk_buff **skbp, int optoff)
 	opt->dsthao = opt->dst1;
 	opt->dst1 = 0;
 
-	hao = (struct ipv6_destopt_hao *)(skb->nh.raw + optoff);
+	hao = (struct ipv6_destopt_hao *)(skb_network_header(skb) + optoff);
 
 	if (hao->length != 16) {
 		LIMIT_NETDEBUG(
@@ -222,7 +200,7 @@ static int ipv6_dest_hao(struct sk_buff **skbp, int optoff)
 
 	if (!(ipv6_addr_type(&hao->addr) & IPV6_ADDR_UNICAST)) {
 		LIMIT_NETDEBUG(
-			KERN_DEBUG "hao is not an unicast addr: " NIP6_FMT "\n", NIP6(hao->addr));
+			KERN_DEBUG "hao is not an unicast addr: %pI6\n", &hao->addr);
 		goto discard;
 	}
 
@@ -232,43 +210,35 @@ static int ipv6_dest_hao(struct sk_buff **skbp, int optoff)
 		goto discard;
 
 	if (skb_cloned(skb)) {
-		struct sk_buff *skb2 = skb_copy(skb, GFP_ATOMIC);
-		struct inet6_skb_parm *opt2;
-
-		if (skb2 == NULL)
+		if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
 			goto discard;
 
-		opt2 = IP6CB(skb2);
-		memcpy(opt2, opt, sizeof(*opt2));
-
-		kfree_skb(skb);
-
 		/* update all variable using below by copied skbuff */
-		*skbp = skb = skb2;
-		hao = (struct ipv6_destopt_hao *)(skb2->nh.raw + optoff);
-		ipv6h = (struct ipv6hdr *)skb2->nh.raw;
+		hao = (struct ipv6_destopt_hao *)(skb_network_header(skb) +
+						  optoff);
+		ipv6h = ipv6_hdr(skb);
 	}
 
 	if (skb->ip_summed == CHECKSUM_COMPLETE)
 		skb->ip_summed = CHECKSUM_NONE;
 
-	ipv6_addr_copy(&tmp_addr, &ipv6h->saddr);
-	ipv6_addr_copy(&ipv6h->saddr, &hao->addr);
-	ipv6_addr_copy(&hao->addr, &tmp_addr);
+	tmp_addr = ipv6h->saddr;
+	ipv6h->saddr = hao->addr;
+	hao->addr = tmp_addr;
 
-	if (skb->tstamp.off_sec == 0)
+	if (skb->tstamp.tv64 == 0)
 		__net_timestamp(skb);
 
-	return 1;
+	return true;
 
  discard:
 	kfree_skb(skb);
-	return 0;
+	return false;
 }
 #endif
 
-static struct tlvtype_proc tlvprocdestopt_lst[] = {
-#ifdef CONFIG_IPV6_MIP6
+static const struct tlvtype_proc tlvprocdestopt_lst[] = {
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
 	{
 		.type	= IPV6_TLV_HAO,
 		.func	= ipv6_dest_hao,
@@ -277,36 +247,32 @@ static struct tlvtype_proc tlvprocdestopt_lst[] = {
 	{-1,			NULL}
 };
 
-static int ipv6_destopt_rcv(struct sk_buff **skbp)
+static int ipv6_destopt_rcv(struct sk_buff *skb)
 {
-	struct sk_buff *skb = *skbp;
 	struct inet6_skb_parm *opt = IP6CB(skb);
-#ifdef CONFIG_IPV6_MIP6
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
 	__u16 dstbuf;
 #endif
-	struct dst_entry *dst;
+	struct dst_entry *dst = skb_dst(skb);
 
-	if (!pskb_may_pull(skb, (skb->h.raw-skb->data)+8) ||
-	    !pskb_may_pull(skb, (skb->h.raw-skb->data)+((skb->h.raw[1]+1)<<3))) {
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+	if (!pskb_may_pull(skb, skb_transport_offset(skb) + 8) ||
+	    !pskb_may_pull(skb, (skb_transport_offset(skb) +
+				 ((skb_transport_header(skb)[1] + 1) << 3)))) {
+		IP6_INC_STATS_BH(dev_net(dst->dev), ip6_dst_idev(dst),
 				 IPSTATS_MIB_INHDRERRORS);
 		kfree_skb(skb);
 		return -1;
 	}
 
-	opt->lastopt = skb->h.raw - skb->nh.raw;
-	opt->dst1 = skb->h.raw - skb->nh.raw;
-#ifdef CONFIG_IPV6_MIP6
+	opt->lastopt = opt->dst1 = skb_network_header_len(skb);
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
 	dstbuf = opt->dst1;
 #endif
 
-	dst = dst_clone(skb->dst);
-	if (ip6_parse_tlv(tlvprocdestopt_lst, skbp)) {
-		dst_release(dst);
-		skb = *skbp;
-		skb->h.raw += ((skb->h.raw[1]+1)<<3);
+	if (ip6_parse_tlv(tlvprocdestopt_lst, skb)) {
+		skb->transport_header += (skb_transport_header(skb)[1] + 1) << 3;
 		opt = IP6CB(skb);
-#ifdef CONFIG_IPV6_MIP6
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
 		opt->nhoff = dstbuf;
 #else
 		opt->nhoff = opt->dst1;
@@ -314,73 +280,46 @@ static int ipv6_destopt_rcv(struct sk_buff **skbp)
 		return 1;
 	}
 
-	IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_INHDRERRORS);
-	dst_release(dst);
+	IP6_INC_STATS_BH(dev_net(dst->dev),
+			 ip6_dst_idev(dst), IPSTATS_MIB_INHDRERRORS);
 	return -1;
 }
 
-static struct inet6_protocol destopt_protocol = {
-	.handler	=	ipv6_destopt_rcv,
-	.flags		=	INET6_PROTO_NOPOLICY | INET6_PROTO_GSO_EXTHDR,
-};
-
-void __init ipv6_destopt_init(void)
-{
-	if (inet6_add_protocol(&destopt_protocol, IPPROTO_DSTOPTS) < 0)
-		printk(KERN_ERR "ipv6_destopt_init: Could not register protocol\n");
-}
-
-/********************************
-  NONE header. No data in packet.
- ********************************/
-
-static int ipv6_nodata_rcv(struct sk_buff **skbp)
-{
-	struct sk_buff *skb = *skbp;
-
-	kfree_skb(skb);
-	return 0;
-}
-
-static struct inet6_protocol nodata_protocol = {
-	.handler	=	ipv6_nodata_rcv,
-	.flags		=	INET6_PROTO_NOPOLICY,
-};
-
-void __init ipv6_nodata_init(void)
-{
-	if (inet6_add_protocol(&nodata_protocol, IPPROTO_NONE) < 0)
-		printk(KERN_ERR "ipv6_nodata_init: Could not register protocol\n");
-}
-
 /********************************
   Routing header.
  ********************************/
 
-static int ipv6_rthdr_rcv(struct sk_buff **skbp)
+/* called with rcu_read_lock() */
+static int ipv6_rthdr_rcv(struct sk_buff *skb)
 {
-	struct sk_buff *skb = *skbp;
 	struct inet6_skb_parm *opt = IP6CB(skb);
 	struct in6_addr *addr = NULL;
 	struct in6_addr daddr;
+	struct inet6_dev *idev;
 	int n, i;
-
 	struct ipv6_rt_hdr *hdr;
 	struct rt0_hdr *rthdr;
+	struct net *net = dev_net(skb->dev);
+	int accept_source_route = net->ipv6.devconf_all->accept_source_route;
+
+	idev = __in6_dev_get(skb->dev);
+	if (idev && accept_source_route > idev->cnf.accept_source_route)
+		accept_source_route = idev->cnf.accept_source_route;
 
-	if (!pskb_may_pull(skb, (skb->h.raw-skb->data)+8) ||
-	    !pskb_may_pull(skb, (skb->h.raw-skb->data)+((skb->h.raw[1]+1)<<3))) {
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+	if (!pskb_may_pull(skb, skb_transport_offset(skb) + 8) ||
+	    !pskb_may_pull(skb, (skb_transport_offset(skb) +
+				 ((skb_transport_header(skb)[1] + 1) << 3)))) {
+		IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 				 IPSTATS_MIB_INHDRERRORS);
 		kfree_skb(skb);
 		return -1;
 	}
 
-	hdr = (struct ipv6_rt_hdr *) skb->h.raw;
+	hdr = (struct ipv6_rt_hdr *)skb_transport_header(skb);
 
-	if (ipv6_addr_is_multicast(&skb->nh.ipv6h->daddr) ||
+	if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr) ||
 	    skb->pkt_type != PACKET_HOST) {
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+		IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 				 IPSTATS_MIB_INADDRERRORS);
 		kfree_skb(skb);
 		return -1;
@@ -389,13 +328,13 @@ static int ipv6_rthdr_rcv(struct sk_buff **skbp)
 looped_back:
 	if (hdr->segments_left == 0) {
 		switch (hdr->type) {
-#ifdef CONFIG_IPV6_MIP6
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
 		case IPV6_SRCRT_TYPE_2:
 			/* Silently discard type 2 header unless it was
 			 * processed by own
 			 */
 			if (!addr) {
-				IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+				IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 						 IPSTATS_MIB_INADDRERRORS);
 				kfree_skb(skb);
 				return -1;
@@ -406,29 +345,22 @@ looped_back:
 			break;
 		}
 
-		opt->lastopt = skb->h.raw - skb->nh.raw;
-		opt->srcrt = skb->h.raw - skb->nh.raw;
-		skb->h.raw += (hdr->hdrlen + 1) << 3;
+		opt->lastopt = opt->srcrt = skb_network_header_len(skb);
+		skb->transport_header += (hdr->hdrlen + 1) << 3;
 		opt->dst0 = opt->dst1;
 		opt->dst1 = 0;
-		opt->nhoff = (&hdr->nexthdr) - skb->nh.raw;
+		opt->nhoff = (&hdr->nexthdr) - skb_network_header(skb);
 		return 1;
 	}
 
 	switch (hdr->type) {
-	case IPV6_SRCRT_TYPE_0:
-		if (hdr->hdrlen & 0x01) {
-			IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
-					 IPSTATS_MIB_INHDRERRORS);
-			icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->hdrlen) - skb->nh.raw);
-			return -1;
-		}
-		break;
-#ifdef CONFIG_IPV6_MIP6
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
 	case IPV6_SRCRT_TYPE_2:
+		if (accept_source_route < 0)
+			goto unknown_rh;
 		/* Silently discard invalid RTH type 2 */
 		if (hdr->hdrlen != 2 || hdr->segments_left != 1) {
-			IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+			IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 					 IPSTATS_MIB_INHDRERRORS);
 			kfree_skb(skb);
 			return -1;
@@ -436,10 +368,7 @@ looped_back:
 		break;
 #endif
 	default:
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
-				 IPSTATS_MIB_INHDRERRORS);
-		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->type) - skb->nh.raw);
-		return -1;
+		goto unknown_rh;
 	}
 
 	/*
@@ -450,9 +379,11 @@ looped_back:
 	n = hdr->hdrlen >> 1;
 
 	if (hdr->segments_left > n) {
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+		IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 				 IPSTATS_MIB_INHDRERRORS);
-		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, (&hdr->segments_left) - skb->nh.raw);
+		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD,
+				  ((&hdr->segments_left) -
+				   skb_network_header(skb)));
 		return -1;
 	}
 
@@ -460,18 +391,14 @@ looped_back:
 	   Do not damage packets queued somewhere.
 	 */
 	if (skb_cloned(skb)) {
-		struct sk_buff *skb2 = skb_copy(skb, GFP_ATOMIC);
 		/* the copy is a forwarded packet */
-		if (skb2 == NULL) {
-			IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+		if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) {
+			IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 					 IPSTATS_MIB_OUTDISCARDS);
 			kfree_skb(skb);
 			return -1;
 		}
-		kfree_skb(skb);
-		*skbp = skb = skb2;
-		opt = IP6CB(skb2);
-		hdr = (struct ipv6_rt_hdr *) skb2->h.raw;
+		hdr = (struct ipv6_rt_hdr *)skb_transport_header(skb);
 	}
 
 	if (skb->ip_summed == CHECKSUM_COMPLETE)
@@ -484,18 +411,18 @@ looped_back:
 	addr += i - 1;
 
 	switch (hdr->type) {
-#ifdef CONFIG_IPV6_MIP6
+#if IS_ENABLED(CONFIG_IPV6_MIP6)
 	case IPV6_SRCRT_TYPE_2:
 		if (xfrm6_input_addr(skb, (xfrm_address_t *)addr,
-				     (xfrm_address_t *)&skb->nh.ipv6h->saddr,
+				     (xfrm_address_t *)&ipv6_hdr(skb)->saddr,
 				     IPPROTO_ROUTING) < 0) {
-			IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+			IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 					 IPSTATS_MIB_INADDRERRORS);
 			kfree_skb(skb);
 			return -1;
 		}
-		if (!ipv6_chk_home_addr(addr)) {
-			IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+		if (!ipv6_chk_home_addr(dev_net(skb_dst(skb)->dev), addr)) {
+			IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 					 IPSTATS_MIB_INADDRERRORS);
 			kfree_skb(skb);
 			return -1;
@@ -507,182 +434,176 @@ looped_back:
 	}
 
 	if (ipv6_addr_is_multicast(addr)) {
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+		IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 				 IPSTATS_MIB_INADDRERRORS);
 		kfree_skb(skb);
 		return -1;
 	}
 
-	ipv6_addr_copy(&daddr, addr);
-	ipv6_addr_copy(addr, &skb->nh.ipv6h->daddr);
-	ipv6_addr_copy(&skb->nh.ipv6h->daddr, &daddr);
+	daddr = *addr;
+	*addr = ipv6_hdr(skb)->daddr;
+	ipv6_hdr(skb)->daddr = daddr;
 
-	dst_release(xchg(&skb->dst, NULL));
+	skb_dst_drop(skb);
 	ip6_route_input(skb);
-	if (skb->dst->error) {
-		skb_push(skb, skb->data - skb->nh.raw);
+	if (skb_dst(skb)->error) {
+		skb_push(skb, skb->data - skb_network_header(skb));
 		dst_input(skb);
 		return -1;
 	}
 
-	if (skb->dst->dev->flags&IFF_LOOPBACK) {
-		if (skb->nh.ipv6h->hop_limit <= 1) {
-			IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+	if (skb_dst(skb)->dev->flags&IFF_LOOPBACK) {
+		if (ipv6_hdr(skb)->hop_limit <= 1) {
+			IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)),
 					 IPSTATS_MIB_INHDRERRORS);
 			icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT,
-				    0, skb->dev);
+				    0);
 			kfree_skb(skb);
 			return -1;
 		}
-		skb->nh.ipv6h->hop_limit--;
+		ipv6_hdr(skb)->hop_limit--;
 		goto looped_back;
 	}
 
-	skb_push(skb, skb->data - skb->nh.raw);
+	skb_push(skb, skb->data - skb_network_header(skb));
 	dst_input(skb);
 	return -1;
+
+unknown_rh:
+	IP6_INC_STATS_BH(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_INHDRERRORS);
+	icmpv6_param_prob(skb, ICMPV6_HDR_FIELD,
+			  (&hdr->type) - skb_network_header(skb));
+	return -1;
 }
 
-static struct inet6_protocol rthdr_protocol = {
+static const struct inet6_protocol rthdr_protocol = {
 	.handler	=	ipv6_rthdr_rcv,
-	.flags		=	INET6_PROTO_NOPOLICY | INET6_PROTO_GSO_EXTHDR,
+	.flags		=	INET6_PROTO_NOPOLICY,
 };
 
-void __init ipv6_rthdr_init(void)
-{
-	if (inet6_add_protocol(&rthdr_protocol, IPPROTO_ROUTING) < 0)
-		printk(KERN_ERR "ipv6_rthdr_init: Could not register protocol\n");
+static const struct inet6_protocol destopt_protocol = {
+	.handler	=	ipv6_destopt_rcv,
+	.flags		=	INET6_PROTO_NOPOLICY,
 };
 
-/*
-   This function inverts received rthdr.
-   NOTE: specs allow to make it automatically only if
-   packet authenticated.
-
-   I will not discuss it here (though, I am really pissed off at
-   this stupid requirement making rthdr idea useless)
-
-   Actually, it creates severe problems  for us.
-   Embryonic requests has no associated sockets,
-   so that user have no control over it and
-   cannot not only to set reply options, but
-   even to know, that someone wants to connect
-   without success. :-(
-
-   For now we need to test the engine, so that I created
-   temporary (or permanent) backdoor.
-   If listening socket set IPV6_RTHDR to 2, then we invert header.
-                                                   --ANK (980729)
- */
+static const struct inet6_protocol nodata_protocol = {
+	.handler	=	dst_discard,
+	.flags		=	INET6_PROTO_NOPOLICY,
+};
 
-struct ipv6_txoptions *
-ipv6_invert_rthdr(struct sock *sk, struct ipv6_rt_hdr *hdr)
+int __init ipv6_exthdrs_init(void)
 {
-	/* Received rthdr:
+	int ret;
 
-	   [ H1 -> H2 -> ... H_prev ]  daddr=ME
+	ret = inet6_add_protocol(&rthdr_protocol, IPPROTO_ROUTING);
+	if (ret)
+		goto out;
 
-	   Inverted result:
-	   [ H_prev -> ... -> H1 ] daddr =sender
+	ret = inet6_add_protocol(&destopt_protocol, IPPROTO_DSTOPTS);
+	if (ret)
+		goto out_rthdr;
 
-	   Note, that IP output engine will rewrite this rthdr
-	   by rotating it left by one addr.
-	 */
+	ret = inet6_add_protocol(&nodata_protocol, IPPROTO_NONE);
+	if (ret)
+		goto out_destopt;
 
-	int n, i;
-	struct rt0_hdr *rthdr = (struct rt0_hdr*)hdr;
-	struct rt0_hdr *irthdr;
-	struct ipv6_txoptions *opt;
-	int hdrlen = ipv6_optlen(hdr);
-
-	if (hdr->segments_left ||
-	    hdr->type != IPV6_SRCRT_TYPE_0 ||
-	    hdr->hdrlen & 0x01)
-		return NULL;
+out:
+	return ret;
+out_destopt:
+	inet6_del_protocol(&destopt_protocol, IPPROTO_DSTOPTS);
+out_rthdr:
+	inet6_del_protocol(&rthdr_protocol, IPPROTO_ROUTING);
+	goto out;
+};
 
-	n = hdr->hdrlen >> 1;
-	opt = sock_kmalloc(sk, sizeof(*opt) + hdrlen, GFP_ATOMIC);
-	if (opt == NULL)
-		return NULL;
-	memset(opt, 0, sizeof(*opt));
-	opt->tot_len = sizeof(*opt) + hdrlen;
-	opt->srcrt = (void*)(opt+1);
-	opt->opt_nflen = hdrlen;
-
-	memcpy(opt->srcrt, hdr, sizeof(*hdr));
-	irthdr = (struct rt0_hdr*)opt->srcrt;
-	irthdr->reserved = 0;
-	opt->srcrt->segments_left = n;
-	for (i=0; i<n; i++)
-		memcpy(irthdr->addr+i, rthdr->addr+(n-1-i), 16);
-	return opt;
+void ipv6_exthdrs_exit(void)
+{
+	inet6_del_protocol(&nodata_protocol, IPPROTO_NONE);
+	inet6_del_protocol(&destopt_protocol, IPPROTO_DSTOPTS);
+	inet6_del_protocol(&rthdr_protocol, IPPROTO_ROUTING);
 }
 
-EXPORT_SYMBOL_GPL(ipv6_invert_rthdr);
-
 /**********************************
   Hop-by-hop options.
  **********************************/
 
+/*
+ * Note: we cannot rely on skb_dst(skb) before we assign it in ip6_route_input().
+ */
+static inline struct inet6_dev *ipv6_skb_idev(struct sk_buff *skb)
+{
+	return skb_dst(skb) ? ip6_dst_idev(skb_dst(skb)) : __in6_dev_get(skb->dev);
+}
+
+static inline struct net *ipv6_skb_net(struct sk_buff *skb)
+{
+	return skb_dst(skb) ? dev_net(skb_dst(skb)->dev) : dev_net(skb->dev);
+}
+
 /* Router Alert as of RFC 2711 */
 
-static int ipv6_hop_ra(struct sk_buff **skbp, int optoff)
+static bool ipv6_hop_ra(struct sk_buff *skb, int optoff)
 {
-	struct sk_buff *skb = *skbp;
+	const unsigned char *nh = skb_network_header(skb);
 
-	if (skb->nh.raw[optoff+1] == 2) {
-		IP6CB(skb)->ra = optoff;
-		return 1;
+	if (nh[optoff + 1] == 2) {
+		IP6CB(skb)->flags |= IP6SKB_ROUTERALERT;
+		memcpy(&IP6CB(skb)->ra, nh + optoff + 2, sizeof(IP6CB(skb)->ra));
+		return true;
 	}
 	LIMIT_NETDEBUG(KERN_DEBUG "ipv6_hop_ra: wrong RA length %d\n",
-	               skb->nh.raw[optoff+1]);
+		       nh[optoff + 1]);
 	kfree_skb(skb);
-	return 0;
+	return false;
 }
 
 /* Jumbo payload */
 
-static int ipv6_hop_jumbo(struct sk_buff **skbp, int optoff)
+static bool ipv6_hop_jumbo(struct sk_buff *skb, int optoff)
 {
-	struct sk_buff *skb = *skbp;
+	const unsigned char *nh = skb_network_header(skb);
+	struct net *net = ipv6_skb_net(skb);
 	u32 pkt_len;
 
-	if (skb->nh.raw[optoff+1] != 4 || (optoff&3) != 2) {
+	if (nh[optoff + 1] != 4 || (optoff & 3) != 2) {
 		LIMIT_NETDEBUG(KERN_DEBUG "ipv6_hop_jumbo: wrong jumbo opt length/alignment %d\n",
-		               skb->nh.raw[optoff+1]);
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst),
+			       nh[optoff+1]);
+		IP6_INC_STATS_BH(net, ipv6_skb_idev(skb),
 				 IPSTATS_MIB_INHDRERRORS);
 		goto drop;
 	}
 
-	pkt_len = ntohl(*(__be32*)(skb->nh.raw+optoff+2));
+	pkt_len = ntohl(*(__be32 *)(nh + optoff + 2));
 	if (pkt_len <= IPV6_MAXPLEN) {
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS);
+		IP6_INC_STATS_BH(net, ipv6_skb_idev(skb),
+				 IPSTATS_MIB_INHDRERRORS);
 		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff+2);
-		return 0;
+		return false;
 	}
-	if (skb->nh.ipv6h->payload_len) {
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_INHDRERRORS);
+	if (ipv6_hdr(skb)->payload_len) {
+		IP6_INC_STATS_BH(net, ipv6_skb_idev(skb),
+				 IPSTATS_MIB_INHDRERRORS);
 		icmpv6_param_prob(skb, ICMPV6_HDR_FIELD, optoff);
-		return 0;
+		return false;
 	}
 
 	if (pkt_len > skb->len - sizeof(struct ipv6hdr)) {
-		IP6_INC_STATS_BH(ip6_dst_idev(skb->dst), IPSTATS_MIB_INTRUNCATEDPKTS);
+		IP6_INC_STATS_BH(net, ipv6_skb_idev(skb),
+				 IPSTATS_MIB_INTRUNCATEDPKTS);
 		goto drop;
 	}
 
 	if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr)))
 		goto drop;
 
-	return 1;
+	return true;
 
 drop:
 	kfree_skb(skb);
-	return 0;
+	return false;
 }
 
-static struct tlvtype_proc tlvprochopopt_lst[] = {
+static const struct tlvtype_proc tlvprochopopt_lst[] = {
 	{
 		.type	= IPV6_TLV_ROUTERALERT,
 		.func	= ipv6_hop_ra,
@@ -694,27 +615,26 @@ static struct tlvtype_proc tlvprochopopt_lst[] = {
 	{ -1, }
 };
 
-int ipv6_parse_hopopts(struct sk_buff **skbp)
+int ipv6_parse_hopopts(struct sk_buff *skb)
 {
-	struct sk_buff *skb = *skbp;
 	struct inet6_skb_parm *opt = IP6CB(skb);
 
 	/*
-	 * skb->nh.raw is equal to skb->data, and
-	 * skb->h.raw - skb->nh.raw is always equal to
+	 * skb_network_header(skb) is equal to skb->data, and
+	 * skb_network_header_len(skb) is always equal to
 	 * sizeof(struct ipv6hdr) by definition of
 	 * hop-by-hop options.
 	 */
 	if (!pskb_may_pull(skb, sizeof(struct ipv6hdr) + 8) ||
-	    !pskb_may_pull(skb, sizeof(struct ipv6hdr) + ((skb->h.raw[1] + 1) << 3))) {
+	    !pskb_may_pull(skb, (sizeof(struct ipv6hdr) +
+				 ((skb_transport_header(skb)[1] + 1) << 3)))) {
 		kfree_skb(skb);
 		return -1;
 	}
 
 	opt->hop = sizeof(struct ipv6hdr);
-	if (ip6_parse_tlv(tlvprochopopt_lst, skbp)) {
-		skb = *skbp;
-		skb->h.raw += (skb->h.raw[1]+1)<<3;
+	if (ip6_parse_tlv(tlvprochopopt_lst, skb)) {
+		skb->transport_header += (skb_transport_header(skb)[1] + 1) << 3;
 		opt = IP6CB(skb);
 		opt->nhoff = sizeof(struct ipv6hdr);
 		return 1;
@@ -740,7 +660,7 @@ static void ipv6_push_rthdr(struct sk_buff *skb, u8 *proto,
 	int hops;
 
 	ihdr = (struct rt0_hdr *) opt;
-	
+
 	phdr = (struct rt0_hdr *) skb_push(skb, (ihdr->rt_hdr.hdrlen + 1) << 3);
 	memcpy(phdr, ihdr, sizeof(struct rt0_hdr));
 
@@ -750,7 +670,7 @@ static void ipv6_push_rthdr(struct sk_buff *skb, u8 *proto,
 		memcpy(phdr->addr, ihdr->addr + 1,
 		       (hops - 1) * sizeof(struct in6_addr));
 
-	ipv6_addr_copy(phdr->addr + (hops - 1), *addr_p);
+	phdr->addr[hops - 1] = **addr_p;
 	*addr_p = ihdr->addr;
 
 	phdr->rt_hdr.nexthdr = *proto;
@@ -782,6 +702,7 @@ void ipv6_push_nfrag_opts(struct sk_buff *skb, struct ipv6_txoptions *opt,
 	if (opt->hopopt)
 		ipv6_push_exthdr(skb, proto, NEXTHDR_HOP, opt->hopopt);
 }
+EXPORT_SYMBOL(ipv6_push_nfrag_opts);
 
 void ipv6_push_frag_opts(struct sk_buff *skb, struct ipv6_txoptions *opt, u8 *proto)
 {
@@ -796,20 +717,19 @@ ipv6_dup_options(struct sock *sk, struct ipv6_txoptions *opt)
 
 	opt2 = sock_kmalloc(sk, opt->tot_len, GFP_ATOMIC);
 	if (opt2) {
-		long dif = (char*)opt2 - (char*)opt;
+		long dif = (char *)opt2 - (char *)opt;
 		memcpy(opt2, opt, opt->tot_len);
 		if (opt2->hopopt)
-			*((char**)&opt2->hopopt) += dif;
+			*((char **)&opt2->hopopt) += dif;
 		if (opt2->dst0opt)
-			*((char**)&opt2->dst0opt) += dif;
+			*((char **)&opt2->dst0opt) += dif;
 		if (opt2->dst1opt)
-			*((char**)&opt2->dst1opt) += dif;
+			*((char **)&opt2->dst1opt) += dif;
 		if (opt2->srcrt)
-			*((char**)&opt2->srcrt) += dif;
+			*((char **)&opt2->srcrt) += dif;
 	}
 	return opt2;
 }
-
 EXPORT_SYMBOL_GPL(ipv6_dup_options);
 
 static int ipv6_renew_option(void *ohdr,
@@ -822,14 +742,14 @@ static int ipv6_renew_option(void *ohdr,
 		if (ohdr) {
 			memcpy(*p, ohdr, ipv6_optlen((struct ipv6_opt_hdr *)ohdr));
 			*hdr = (struct ipv6_opt_hdr *)*p;
-			*p += CMSG_ALIGN(ipv6_optlen(*(struct ipv6_opt_hdr **)hdr));
+			*p += CMSG_ALIGN(ipv6_optlen(*hdr));
 		}
 	} else {
 		if (newopt) {
 			if (copy_from_user(*p, newopt, newoptlen))
 				return -EFAULT;
 			*hdr = (struct ipv6_opt_hdr *)*p;
-			if (ipv6_optlen(*(struct ipv6_opt_hdr **)hdr) > newoptlen)
+			if (ipv6_optlen(*hdr) > newoptlen)
 				return -EINVAL;
 			*p += CMSG_ALIGN(newoptlen);
 		}
@@ -927,4 +847,28 @@ struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space,
 
 	return opt;
 }
+EXPORT_SYMBOL_GPL(ipv6_fixup_options);
 
+/**
+ * fl6_update_dst - update flowi destination address with info given
+ *                  by srcrt option, if any.
+ *
+ * @fl6: flowi6 for which daddr is to be updated
+ * @opt: struct ipv6_txoptions in which to look for srcrt opt
+ * @orig: copy of original daddr address if modified
+ *
+ * Returns NULL if no txoptions or no srcrt, otherwise returns orig
+ * and initial value of fl6->daddr set in orig
+ */
+struct in6_addr *fl6_update_dst(struct flowi6 *fl6,
+				const struct ipv6_txoptions *opt,
+				struct in6_addr *orig)
+{
+	if (!opt || !opt->srcrt)
+		return NULL;
+
+	*orig = fl6->daddr;
+	fl6->daddr = *((struct rt0_hdr *)opt->srcrt)->addr;
+	return orig;
+}
+EXPORT_SYMBOL_GPL(fl6_update_dst);