Merge git://git.infradead.org/mtd-2.6

[linux-2.6] / net / ipv4 / tcp_output.c

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 8197b5e12f1fdc37a1ba9a871012deca60d9f939..07bb5a2b375ec0b02e1be3774a474fbeae24a808 100644 (file)
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -45,6 +45,11 @@
 /* People can turn this off for buggy TCP's found in printers etc. */
 int sysctl_tcp_retrans_collapse = 1;
 
+/* People can turn this on to  work with those rare, broken TCPs that
+ * interpret the window field as a signed quantity.
+ */
+int sysctl_tcp_workaround_signed_windows = 0;
+
 /* This limits the percentage of the congestion window which we
  * will allow a single TSO frame to consume.  Building TSO frames
  * which are too large can cause TCP streams to be bursty.
@@ -54,8 +59,8 @@ int sysctl_tcp_tso_win_divisor = 3;
 int sysctl_tcp_mtu_probing = 0;
 int sysctl_tcp_base_mss = 512;
 
-EXPORT_SYMBOL(sysctl_tcp_mtu_probing);
-EXPORT_SYMBOL(sysctl_tcp_base_mss);
+/* By default, RFC2861 behavior.  */
+int sysctl_tcp_slow_start_after_idle = 1;
 
 static void update_send_head(struct sock *sk, struct tcp_sock *tp,
                             struct sk_buff *skb)
@@ -136,7 +141,8 @@ static void tcp_event_data_sent(struct tcp_sock *tp,
        struct inet_connection_sock *icsk = inet_csk(sk);
        const u32 now = tcp_time_stamp;
 
-       if (!tp->packets_out && (s32)(now - tp->lsndtime) > icsk->icsk_rto)
+       if (sysctl_tcp_slow_start_after_idle &&
+           (!tp->packets_out && (s32)(now - tp->lsndtime) > icsk->icsk_rto))
                tcp_cwnd_restart(sk, __sk_dst_get(sk));
 
        tp->lsndtime = now;
@@ -177,12 +183,18 @@ void tcp_select_initial_window(int __space, __u32 mss,
                space = (space / mss) * mss;
 
        /* NOTE: offering an initial window larger than 32767
-        * will break some buggy TCP stacks. We try to be nice.
-        * If we are not window scaling, then this truncates
-        * our initial window offering to 32k. There should also
-        * be a sysctl option to stop being nice.
+        * will break some buggy TCP stacks. If the admin tells us
+        * it is likely we could be speaking with such a buggy stack
+        * we will truncate our initial window offering to 32K-1
+        * unless the remote has sent us a window scaling option,
+        * which we interpret as a sign the remote TCP is not
+        * misinterpreting the window field as a signed quantity.
         */
-       (*rcv_wnd) = min(space, MAX_TCP_WINDOW);
+       if (sysctl_tcp_workaround_signed_windows)
+               (*rcv_wnd) = min(space, MAX_TCP_WINDOW);
+       else
+               (*rcv_wnd) = space;
+
        (*rcv_wscale) = 0;
        if (wscale_ok) {
                /* Set window scaling on max possible window
@@ -241,7 +253,7 @@ static u16 tcp_select_window(struct sock *sk)
        /* Make sure we do not exceed the maximum possible
         * scaled window.
         */
-       if (!tp->rx_opt.rcv_wscale)
+       if (!tp->rx_opt.rcv_wscale && sysctl_tcp_workaround_signed_windows)
                new_win = min(new_win, MAX_TCP_WINDOW);
        else
                new_win = min(new_win, (65535U << tp->rx_opt.rcv_wscale));
@@ -457,7 +469,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
        TCP_INC_STATS(TCP_MIB_OUTSEGS);
 
        err = icsk->icsk_af_ops->queue_xmit(skb, 0);
-       if (unlikely(err <= 0))
+       if (likely(err <= 0))
                return err;
 
        tcp_enter_cwr(sk);
@@ -525,6 +537,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
        struct tcp_sock *tp = tcp_sk(sk);
        struct sk_buff *buff;
        int nsize, old_factor;
+       int nlen;
        u16 flags;
 
        BUG_ON(len > skb->len);
@@ -543,7 +556,11 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
        buff = sk_stream_alloc_skb(sk, nsize, GFP_ATOMIC);
        if (buff == NULL)
                return -ENOMEM; /* We'll just try again later. */
+
        sk_charge_skb(sk, buff);
+       nlen = skb->len - len - nsize;
+       buff->truesize += nlen;
+       skb->truesize -= nlen;
 
        /* Correct the sequence numbers. */
        TCP_SKB_CB(buff)->seq = TCP_SKB_CB(skb)->seq + len;
@@ -629,7 +646,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss
  * eventually). The difference is that pulled data not copied, but
  * immediately discarded.
  */
-static unsigned char *__pskb_trim_head(struct sk_buff *skb, int len)
+static void __pskb_trim_head(struct sk_buff *skb, int len)
 {
        int i, k, eat;
 
@@ -654,7 +671,6 @@ static unsigned char *__pskb_trim_head(struct sk_buff *skb, int len)
        skb->tail = skb->data;
        skb->data_len -= len;
        skb->len = skb->data_len;
-       return skb->tail;
 }
 
 int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len)
@@ -663,12 +679,11 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len)
            pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
                return -ENOMEM;
 
-       if (len <= skb_headlen(skb)) {
+       /* If len == headlen, we avoid __skb_pull to preserve alignment. */
+       if (unlikely(len < skb_headlen(skb)))
                __skb_pull(skb, len);
-       } else {
-               if (__pskb_trim_head(skb, len-skb_headlen(skb)) == NULL)
-                       return -ENOMEM;
-       }
+       else
+               __pskb_trim_head(skb, len - skb_headlen(skb));
 
        TCP_SKB_CB(skb)->seq += len;
        skb->ip_summed = CHECKSUM_HW;
@@ -1029,7 +1044,8 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len,
        if (unlikely(buff == NULL))
                return -ENOMEM;
 
-       buff->truesize = nlen;
+       sk_charge_skb(sk, buff);
+       buff->truesize += nlen;
        skb->truesize -= nlen;
 
        /* Correct the sequence numbers. */
@@ -1238,8 +1254,8 @@ static int tcp_mtu_probe(struct sock *sk)
                update_send_head(sk, tp, nskb);
 
                icsk->icsk_mtup.probe_size = tcp_mss_to_mtu(sk, nskb->len);
-               icsk->icsk_mtup.probe_seq_start = TCP_SKB_CB(nskb)->seq;
-               icsk->icsk_mtup.probe_seq_end = TCP_SKB_CB(nskb)->end_seq;
+               tp->mtu_probe.probe_seq_start = TCP_SKB_CB(nskb)->seq;
+               tp->mtu_probe.probe_seq_end = TCP_SKB_CB(nskb)->end_seq;
 
                return 1;
        }