8 files changed, 50 insertions, 41 deletions

diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 68b72a7a1806..418862f1bf22 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -570,7 +570,7 @@ struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip,
          *      Allocate a buffer
          */
 
-        skb = alloc_skb(arp_hdr_len(dev) + LL_RESERVED_SPACE(dev), GFP_ATOMIC);
+        skb = alloc_skb(arp_hdr_len(dev) + LL_ALLOCATED_SPACE(dev), GFP_ATOMIC);
         if (skb == NULL)
                 return NULL;
 
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index 05afb576d935..2c0e4572cc90 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -338,7 +338,7 @@ static int cipso_v4_cache_check(const unsigned char *key,
                 return -ENOENT;
 
         hash = cipso_v4_map_cache_hash(key, key_len);
-        bkt = hash & (CIPSO_V4_CACHE_BUCKETBITS - 1);
+        bkt = hash & (CIPSO_V4_CACHE_BUCKETS - 1);
         spin_lock_bh(&cipso_v4_cache[bkt].lock);
         list_for_each_entry(entry, &cipso_v4_cache[bkt].list, list) {
                 if (entry->hash == hash &&
@@ -417,7 +417,7 @@ int cipso_v4_cache_add(const struct sk_buff *skb,
         atomic_inc(&secattr->cache->refcount);
         entry->lsm_data = secattr->cache;
 
-        bkt = entry->hash & (CIPSO_V4_CACHE_BUCKETBITS - 1);
+        bkt = entry->hash & (CIPSO_V4_CACHE_BUCKETS - 1);
         spin_lock_bh(&cipso_v4_cache[bkt].lock);
         if (cipso_v4_cache[bkt].size < cipso_v4_cache_bucketsize) {
                 list_add(&entry->list, &cipso_v4_cache[bkt].list);
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 6250f4239b61..2769dc4a4c84 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -292,7 +292,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size)
         struct iphdr *pip;
         struct igmpv3_report *pig;
 
-        skb = alloc_skb(size + LL_RESERVED_SPACE(dev), GFP_ATOMIC);
+        skb = alloc_skb(size + LL_ALLOCATED_SPACE(dev), GFP_ATOMIC);
         if (skb == NULL)
                 return NULL;
 
@@ -653,7 +653,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc,
                 return -1;
         }
 
-        skb=alloc_skb(IGMP_SIZE+LL_RESERVED_SPACE(dev), GFP_ATOMIC);
+        skb=alloc_skb(IGMP_SIZE+LL_ALLOCATED_SPACE(dev), GFP_ATOMIC);
         if (skb == NULL) {
                 ip_rt_put(rt);
                 return -1;
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 7b4bad6d572f..ff77a4a7f9ec 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -397,7 +397,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
         iph = ip_hdr(skb);
 
         /*
-         *      RFC1122: 3.1.2.2 MUST silently discard any IP frame that fails the checksum.
+         *      RFC1122: 3.2.1.2 MUST silently discard any IP frame that fails the checksum.
          *
          *      Is the datagram acceptable?
          *
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 89dee4346f60..ed45037ce9be 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -710,14 +710,14 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d
         struct net_device *dev = d->dev;
         struct sk_buff *skb;
         struct bootp_pkt *b;
-        int hh_len = LL_RESERVED_SPACE(dev);
         struct iphdr *h;
 
         /* Allocate packet */
-        skb = alloc_skb(sizeof(struct bootp_pkt) + hh_len + 15, GFP_KERNEL);
+        skb = alloc_skb(sizeof(struct bootp_pkt) + LL_ALLOCATED_SPACE(dev) + 15,
+                        GFP_KERNEL);
         if (!skb)
                 return;
-        skb_reserve(skb, hh_len);
+        skb_reserve(skb, LL_RESERVED_SPACE(dev));
         b = (struct bootp_pkt *) skb_put(skb, sizeof(struct bootp_pkt));
         memset(b, 0, sizeof(struct bootp_pkt));
 
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 11d7f753a820..fead049daf43 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -322,7 +322,6 @@ static int raw_send_hdrinc(struct sock *sk, void *from, size_t length,
                         unsigned int flags)
 {
         struct inet_sock *inet = inet_sk(sk);
-        int hh_len;
         struct iphdr *iph;
         struct sk_buff *skb;
         unsigned int iphlen;
@@ -336,13 +335,12 @@ static int raw_send_hdrinc(struct sock *sk, void *from, size_t length,
         if (flags&MSG_PROBE)
                 goto out;
 
-        hh_len = LL_RESERVED_SPACE(rt->u.dst.dev);
-
-        skb = sock_alloc_send_skb(sk, length+hh_len+15,
-                                  flags&MSG_DONTWAIT, &err);
+        skb = sock_alloc_send_skb(sk,
+                                  length + LL_ALLOCATED_SPACE(rt->u.dst.dev) + 15,
+                                  flags & MSG_DONTWAIT, &err);
         if (skb == NULL)
                 goto error;
-        skb_reserve(skb, hh_len);
+        skb_reserve(skb, LL_RESERVED_SPACE(rt->u.dst.dev));
 
         skb->priority = sk->sk_priority;
         skb->mark = sk->sk_mark;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 5e3685c5c407..92f90ae46f4a 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1468,14 +1468,14 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph,
 
                                         /* BSD 4.2 compatibility hack :-( */
                                         if (mtu == 0 &&
-                                            old_mtu >= rth->u.dst.metrics[RTAX_MTU-1] &&
+                                            old_mtu >= dst_metric(&rth->u.dst, RTAX_MTU) &&
                                             old_mtu >= 68 + (iph->ihl << 2))
                                                 old_mtu -= iph->ihl << 2;
 
                                         mtu = guess_mtu(old_mtu);
                                 }
-                                if (mtu <= rth->u.dst.metrics[RTAX_MTU-1]) {
-                                        if (mtu < rth->u.dst.metrics[RTAX_MTU-1]) {
+                                if (mtu <= dst_metric(&rth->u.dst, RTAX_MTU)) {
+                                        if (mtu < dst_metric(&rth->u.dst, RTAX_MTU)) {
                                                 dst_confirm(&rth->u.dst);
                                                 if (mtu < ip_rt_min_pmtu) {
                                                         mtu = ip_rt_min_pmtu;
@@ -1497,7 +1497,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph,
 
 static void ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu)
 {
-        if (dst->metrics[RTAX_MTU-1] > mtu && mtu >= 68 &&
+        if (dst_metric(dst, RTAX_MTU) > mtu && mtu >= 68 &&
             !(dst_metric_locked(dst, RTAX_MTU))) {
                 if (mtu < ip_rt_min_pmtu) {
                         mtu = ip_rt_min_pmtu;
@@ -1613,7 +1613,7 @@ static void rt_set_nexthop(struct rtable *rt, struct fib_result *res, u32 itag)
                        sizeof(rt->u.dst.metrics));
                 if (fi->fib_mtu == 0) {
                         rt->u.dst.metrics[RTAX_MTU-1] = rt->u.dst.dev->mtu;
-                        if (rt->u.dst.metrics[RTAX_LOCK-1] & (1 << RTAX_MTU) &&
+                        if (dst_metric_locked(&rt->u.dst, RTAX_MTU) &&
                             rt->rt_gateway != rt->rt_dst &&
                             rt->u.dst.dev->mtu > 576)
                                 rt->u.dst.metrics[RTAX_MTU-1] = 576;
@@ -1624,14 +1624,14 @@ static void rt_set_nexthop(struct rtable *rt, struct fib_result *res, u32 itag)
         } else
                 rt->u.dst.metrics[RTAX_MTU-1]= rt->u.dst.dev->mtu;
 
-        if (rt->u.dst.metrics[RTAX_HOPLIMIT-1] == 0)
+        if (dst_metric(&rt->u.dst, RTAX_HOPLIMIT) == 0)
                 rt->u.dst.metrics[RTAX_HOPLIMIT-1] = sysctl_ip_default_ttl;
-        if (rt->u.dst.metrics[RTAX_MTU-1] > IP_MAX_MTU)
+        if (dst_metric(&rt->u.dst, RTAX_MTU) > IP_MAX_MTU)
                 rt->u.dst.metrics[RTAX_MTU-1] = IP_MAX_MTU;
-        if (rt->u.dst.metrics[RTAX_ADVMSS-1] == 0)
+        if (dst_metric(&rt->u.dst, RTAX_ADVMSS) == 0)
                 rt->u.dst.metrics[RTAX_ADVMSS-1] = max_t(unsigned int, rt->u.dst.dev->mtu - 40,
                                        ip_rt_min_advmss);
-        if (rt->u.dst.metrics[RTAX_ADVMSS-1] > 65535 - 40)
+        if (dst_metric(&rt->u.dst, RTAX_ADVMSS) > 65535 - 40)
                 rt->u.dst.metrics[RTAX_ADVMSS-1] = 65535 - 40;
 
 #ifdef CONFIG_NET_CLS_ROUTE
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index eda4f4a233f3..b54d9d37b636 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -66,6 +66,7 @@
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/sysctl.h>
+#include <net/dst.h>
 #include <net/tcp.h>
 #include <net/inet_common.h>
 #include <linux/ipsec.h>
@@ -113,8 +114,6 @@ int sysctl_tcp_abc __read_mostly;
 #define FLAG_FORWARD_PROGRESS   (FLAG_ACKED|FLAG_DATA_SACKED)
 #define FLAG_ANY_PROGRESS       (FLAG_FORWARD_PROGRESS|FLAG_SND_UNA_ADVANCED)
 
-#define IsSackFrto() (sysctl_tcp_frto == 0x2)
-
 #define TCP_REMNANT (TCP_FLAG_FIN|TCP_FLAG_URG|TCP_FLAG_SYN|TCP_FLAG_PSH)
 #define TCP_HP_BITS (~(TCP_RESERVED_BITS|TCP_FLAG_PSH))
 
@@ -605,7 +604,7 @@ static u32 tcp_rto_min(struct sock *sk)
         u32 rto_min = TCP_RTO_MIN;
 
         if (dst && dst_metric_locked(dst, RTAX_RTO_MIN))
-                rto_min = dst->metrics[RTAX_RTO_MIN - 1];
+                rto_min = dst_metric(dst, RTAX_RTO_MIN);
         return rto_min;
 }
 
@@ -769,7 +768,7 @@ void tcp_update_metrics(struct sock *sk)
                                 dst->metrics[RTAX_RTTVAR - 1] = m;
                         else
                                 dst->metrics[RTAX_RTTVAR-1] -=
-                                        (dst->metrics[RTAX_RTTVAR-1] - m)>>2;
+                                        (dst_metric(dst, RTAX_RTTVAR) - m)>>2;
                 }
 
                 if (tp->snd_ssthresh >= 0xFFFF) {
@@ -788,21 +787,21 @@ void tcp_update_metrics(struct sock *sk)
                                 dst->metrics[RTAX_SSTHRESH-1] =
                                         max(tp->snd_cwnd >> 1, tp->snd_ssthresh);
                         if (!dst_metric_locked(dst, RTAX_CWND))
-                                dst->metrics[RTAX_CWND-1] = (dst->metrics[RTAX_CWND-1] + tp->snd_cwnd) >> 1;
+                                dst->metrics[RTAX_CWND-1] = (dst_metric(dst, RTAX_CWND) + tp->snd_cwnd) >> 1;
                 } else {
                         /* Else slow start did not finish, cwnd is non-sense,
                            ssthresh may be also invalid.
                          */
                         if (!dst_metric_locked(dst, RTAX_CWND))
-                                dst->metrics[RTAX_CWND-1] = (dst->metrics[RTAX_CWND-1] + tp->snd_ssthresh) >> 1;
-                        if (dst->metrics[RTAX_SSTHRESH-1] &&
+                                dst->metrics[RTAX_CWND-1] = (dst_metric(dst, RTAX_CWND) + tp->snd_ssthresh) >> 1;
+                        if (dst_metric(dst, RTAX_SSTHRESH) &&
                             !dst_metric_locked(dst, RTAX_SSTHRESH) &&
-                            tp->snd_ssthresh > dst->metrics[RTAX_SSTHRESH-1])
+                            tp->snd_ssthresh > dst_metric(dst, RTAX_SSTHRESH))
                                 dst->metrics[RTAX_SSTHRESH-1] = tp->snd_ssthresh;
                 }
 
                 if (!dst_metric_locked(dst, RTAX_REORDERING)) {
-                        if (dst->metrics[RTAX_REORDERING-1] < tp->reordering &&
+                        if (dst_metric(dst, RTAX_REORDERING) < tp->reordering &&
                             tp->reordering != sysctl_tcp_reordering)
                                 dst->metrics[RTAX_REORDERING-1] = tp->reordering;
                 }
@@ -1685,6 +1684,11 @@ static inline void tcp_reset_reno_sack(struct tcp_sock *tp)
         tp->sacked_out = 0;
 }
 
+static int tcp_is_sackfrto(const struct tcp_sock *tp)
+{
+        return (sysctl_tcp_frto == 0x2) && !tcp_is_reno(tp);
+}
+
 /* F-RTO can only be used if TCP has never retransmitted anything other than
  * head (SACK enhanced variant from Appendix B of RFC4138 is more robust here)
  */
@@ -1701,7 +1705,7 @@ int tcp_use_frto(struct sock *sk)
         if (icsk->icsk_mtup.probe_size)
                 return 0;
 
-        if (IsSackFrto())
+        if (tcp_is_sackfrto(tp))
                 return 1;
 
         /* Avoid expensive walking of rexmit queue if possible */
@@ -1791,7 +1795,7 @@ void tcp_enter_frto(struct sock *sk)
         /* Earlier loss recovery underway (see RFC4138; Appendix B).
          * The last condition is necessary at least in tp->frto_counter case.
          */
-        if (IsSackFrto() && (tp->frto_counter ||
+        if (tcp_is_sackfrto(tp) && (tp->frto_counter ||
             ((1 << icsk->icsk_ca_state) & (TCPF_CA_Recovery|TCPF_CA_Loss))) &&
             after(tp->high_seq, tp->snd_una)) {
                 tp->frto_highmark = tp->high_seq;
@@ -1838,9 +1842,16 @@ static void tcp_enter_frto_loss(struct sock *sk, int allowed_segments, int flag)
                         TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
                 }
 
-                /* Don't lost mark skbs that were fwd transmitted after RTO */
-                if (!(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) &&
-                    !after(TCP_SKB_CB(skb)->end_seq, tp->frto_highmark)) {
+                /* Marking forward transmissions that were made after RTO lost
+                 * can cause unnecessary retransmissions in some scenarios,
+                 * SACK blocks will mitigate that in some but not in all cases.
+                 * We used to not mark them but it was causing break-ups with
+                 * receivers that do only in-order receival.
+                 *
+                 * TODO: we could detect presence of such receiver and select
+                 * different behavior per flow.
+                 */
+                if (!(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) {
                         TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
                         tp->lost_out += tcp_skb_pcount(skb);
                 }
@@ -1856,7 +1867,7 @@ static void tcp_enter_frto_loss(struct sock *sk, int allowed_segments, int flag)
         tp->reordering = min_t(unsigned int, tp->reordering,
                                sysctl_tcp_reordering);
         tcp_set_ca_state(sk, TCP_CA_Loss);
-        tp->high_seq = tp->frto_highmark;
+        tp->high_seq = tp->snd_nxt;
         TCP_ECN_queue_cwr(tp);
 
         tcp_clear_retrans_hints_partial(tp);
@@ -2478,7 +2489,7 @@ static void tcp_try_to_open(struct sock *sk, int flag)
 
         tcp_verify_left_out(tp);
 
-        if (tp->retrans_out == 0)
+        if (!tp->frto_counter && tp->retrans_out == 0)
                 tp->retrans_stamp = 0;
 
         if (flag & FLAG_ECE)
@@ -3123,7 +3134,7 @@ static int tcp_process_frto(struct sock *sk, int flag)
                 return 1;
         }
 
-        if (!IsSackFrto() || tcp_is_reno(tp)) {
+        if (!tcp_is_sackfrto(tp)) {
                 /* RFC4138 shortcoming in step 2; should also have case c):
                  * ACK isn't duplicate nor advances window, e.g., opposite dir
                  * data, winupdate