ip: generate unique IP identificator if local fragmentation is allowed

[ Upstream commit 703133de331a7a7df47f31fb9de51dc6f68a9de8 ]

If local fragmentation is allowed, then ip_select_ident() and
ip_select_ident_more() need to generate unique IDs to ensure
correct defragmentation on the peer.

For example, if IPsec (tunnel mode) has to encrypt large skbs
that have local_df bit set, then all IP fragments that belonged
to different ESP datagrams would have used the same identificator.
If one of these IP fragments would get lost or reordered, then
peer could possibly stitch together wrong IP fragments that did
not belong to the same datagram. This would lead to a packet loss
or data corruption.

Signed-off-by: Ansis Atteka <aatteka@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>

diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c
index ad6a9d91dfdd943df5b841d7454912b9fbb0e9f4..2b349d3355f474d30e73022b362509844d9eed20 100644 (file)
--- a/drivers/net/ppp/pptp.c
+++ b/drivers/net/ppp/pptp.c
@@ -281,7 +281,7 @@ static int pptp_xmit(struct ppp_channel *chan, struct sk_buff *skb)
        nf_reset(skb);
 
        skb->ip_summed = CHECKSUM_NONE;
-       ip_select_ident(iph, &rt->dst, NULL);
+       ip_select_ident(skb, &rt->dst, NULL);
        ip_send_check(iph);
 
        ip_local_out(skb);

diff --git a/include/net/ip.h b/include/net/ip.h
index eca0ef7a495e9d1605b7705e73ee3377bd8aaae5..06aed72023f7f1479dcbb0dbda4e9929b4ad75b2 100644 (file)
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -266,9 +266,11 @@ int ip_dont_fragment(struct sock *sk, struct dst_entry *dst)
 
 extern void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more);
 
-static inline void ip_select_ident(struct iphdr *iph, struct dst_entry *dst, struct sock *sk)
+static inline void ip_select_ident(struct sk_buff *skb, struct dst_entry *dst, struct sock *sk)
 {
-       if (iph->frag_off & htons(IP_DF)) {
+       struct iphdr *iph = ip_hdr(skb);
+
+       if ((iph->frag_off & htons(IP_DF)) && !skb->local_df) {
                /* This is only to work around buggy Windows95/2000
                 * VJ compression implementations.  If the ID field
                 * does not change, they drop every other packet in
@@ -280,9 +282,11 @@ static inline void ip_select_ident(struct iphdr *iph, struct dst_entry *dst, str
                __ip_select_ident(iph, dst, 0);
 }
 
-static inline void ip_select_ident_more(struct iphdr *iph, struct dst_entry *dst, struct sock *sk, int more)
+static inline void ip_select_ident_more(struct sk_buff *skb, struct dst_entry *dst, struct sock *sk, int more)
 {
-       if (iph->frag_off & htons(IP_DF)) {
+       struct iphdr *iph = ip_hdr(skb);
+
+       if ((iph->frag_off & htons(IP_DF)) && !skb->local_df) {
                if (sk && inet_sk(sk)->inet_daddr) {
                        iph->id = htons(inet_sk(sk)->inet_id);
                        inet_sk(sk)->inet_id += 1 + more;

diff --git a/include/net/ipip.h b/include/net/ipip.h
index a32654d52730d751a4c89dd8f8fea98b57add370..4dccfe3bf731f9eb30846ebbf88785c8157ca361 100644 (file)
--- a/include/net/ipip.h
+++ b/include/net/ipip.h
@@ -50,7 +50,7 @@ struct ip_tunnel_prl_entry {
        int pkt_len = skb->len - skb_transport_offset(skb);             \
                                                                        \
        skb->ip_summed = CHECKSUM_NONE;                                 \
-       ip_select_ident(iph, &rt->dst, NULL);                           \
+       ip_select_ident(skb, &rt->dst, NULL);                           \
                                                                        \
        err = ip_local_out(skb);                                        \
        if (likely(net_xmit_eval(err) == 0)) {                          \

diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index c8989a71bbd9a227febcc568de2d1323947731bb..24cf7da4d284112a387d8f8afd9c8135b0389553 100644 (file)
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -342,7 +342,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size)
        pip->saddr    = fl4.saddr;
        pip->protocol = IPPROTO_IGMP;
        pip->tot_len  = 0;      /* filled in later */
-       ip_select_ident(pip, &rt->dst, NULL);
+       ip_select_ident(skb, &rt->dst, NULL);
        ((u8*)&pip[1])[0] = IPOPT_RA;
        ((u8*)&pip[1])[1] = 4;
        ((u8*)&pip[1])[2] = 0;
@@ -683,7 +683,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc,
        iph->daddr    = dst;
        iph->saddr    = fl4.saddr;
        iph->protocol = IPPROTO_IGMP;
-       ip_select_ident(iph, &rt->dst, NULL);
+       ip_select_ident(skb, &rt->dst, NULL);
        ((u8*)&iph[1])[0] = IPOPT_RA;
        ((u8*)&iph[1])[1] = 4;
        ((u8*)&iph[1])[2] = 0;

diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c
index 552e4f8d0922bda08bad3c223b7d37a74117fda2..58c4e696a6a7a211372cf258372b37a76c6e19ee 100644 (file)
--- a/net/ipv4/inetpeer.c
+++ b/net/ipv4/inetpeer.c
@@ -32,8 +32,8 @@
  *  At the moment of writing this notes identifier of IP packets is generated
  *  to be unpredictable using this code only for packets subjected
  *  (actually or potentially) to defragmentation.  I.e. DF packets less than
- *  PMTU in size uses a constant ID and do not use this code (see
- *  ip_select_ident() in include/net/ip.h).
+ *  PMTU in size when local fragmentation is disabled use a constant ID and do
+ *  not use this code (see ip_select_ident() in include/net/ip.h).
  *
  *  Route cache entries hold references to our nodes.
  *  New cache entries get references via lookup by destination IP address in

diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 0bc95f3977d2aecff69f30c871f1b459588a34a7..daf408e8633ce7ca9a5c5f5741b67ef0f3d4610b 100644 (file)
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -162,7 +162,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk,
        iph->daddr    = (opt && opt->opt.srr ? opt->opt.faddr : daddr);
        iph->saddr    = saddr;
        iph->protocol = sk->sk_protocol;
-       ip_select_ident(iph, &rt->dst, sk);
+       ip_select_ident(skb, &rt->dst, sk);
 
        if (opt && opt->opt.optlen) {
                iph->ihl += opt->opt.optlen>>2;
@@ -390,7 +390,7 @@ packet_routed:
                ip_options_build(skb, &inet_opt->opt, inet->inet_daddr, rt, 0);
        }
 
-       ip_select_ident_more(iph, &rt->dst, sk,
+       ip_select_ident_more(skb, &rt->dst, sk,
                             (skb_shinfo(skb)->gso_segs ?: 1) - 1);
 
        skb->priority = sk->sk_priority;
@@ -1334,7 +1334,7 @@ struct sk_buff *__ip_make_skb(struct sock *sk,
        iph->ihl = 5;
        iph->tos = inet->tos;
        iph->frag_off = df;
-       ip_select_ident(iph, &rt->dst, sk);
+       ip_select_ident(skb, &rt->dst, sk);
        iph->ttl = ttl;
        iph->protocol = sk->sk_protocol;
        iph->saddr = fl4->saddr;

diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 006439403213d81359cfa92ef0a53753f5c3b86e..b5e64e4c06462942c9d38c1d96c15b400a8573dd 100644 (file)
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1576,7 +1576,7 @@ static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr)
        iph->protocol   =       IPPROTO_IPIP;
        iph->ihl        =       5;
        iph->tot_len    =       htons(skb->len);
-       ip_select_ident(iph, skb_dst(skb), NULL);
+       ip_select_ident(skb, skb_dst(skb), NULL);
        ip_send_check(iph);
 
        memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));

diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index e1d4f30bb718ab2015d14ecb1e25d4dd66e81b41..28150145ce02aa76a666017f7717cfc7e5f24488 100644 (file)
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -380,7 +380,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
                iph->check   = 0;
                iph->tot_len = htons(length);
                if (!iph->id)
-                       ip_select_ident(iph, &rt->dst, NULL);
+                       ip_select_ident(skb, &rt->dst, NULL);
 
                iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
        }

diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c
index ed4bf11ef9f4709b9418ebeff06789c9b84b77ab..938553efbf8ea647dc586233279dbfc883ed64fe 100644 (file)
--- a/net/ipv4/xfrm4_mode_tunnel.c
+++ b/net/ipv4/xfrm4_mode_tunnel.c
@@ -54,7 +54,7 @@ static int xfrm4_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb)
 
        top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ?
                0 : (XFRM_MODE_SKB_CB(skb)->frag_off & htons(IP_DF));
-       ip_select_ident(top_iph, dst->child, NULL);
+       ip_select_ident(skb, dst->child, NULL);
 
        top_iph->ttl = ip4_dst_hoplimit(dst->child);
 

diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c
index aa2d7206ee8a064a5fbc57e88f655fb42e7fe21f..38c08136baa201e603707cb853d7feb9f881f2fd 100644 (file)
--- a/net/netfilter/ipvs/ip_vs_xmit.c
+++ b/net/netfilter/ipvs/ip_vs_xmit.c
@@ -853,7 +853,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
        iph->daddr              =       cp->daddr.ip;
        iph->saddr              =       saddr;
        iph->ttl                =       old_iph->ttl;
-       ip_select_ident(iph, &rt->dst, NULL);
+       ip_select_ident(skb, &rt->dst, NULL);
 
        /* Another hack: avoid icmp_send in ip_fragment */
        skb->local_df = 1;