]> pilppa.org Git - linux-2.6-omap-h63xx.git/blobdiff - net/ipv4/tcp_input.c
Merge branch 'fixes-davem' of git://git.kernel.org/pub/scm/linux/kernel/git/linville...
[linux-2.6-omap-h63xx.git] / net / ipv4 / tcp_input.c
index 457697a043d9e3710d4ceaed25d87a1572cafb7a..0f0c1c9829a14b41e9654b3a108d2f1aa3b5e6f1 100644 (file)
@@ -1248,6 +1248,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
        int cached_fack_count;
        int i;
        int first_sack_index;
+       int force_one_sack;
 
        if (!tp->sacked_out) {
                if (WARN_ON(tp->fackets_out))
@@ -1268,22 +1269,25 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
        if (before(TCP_SKB_CB(ack_skb)->ack_seq, prior_snd_una - tp->max_window))
                return 0;
 
+       if (!tp->packets_out)
+               goto out;
+
        /* SACK fastpath:
         * if the only SACK change is the increase of the end_seq of
         * the first block then only apply that SACK block
         * and use retrans queue hinting otherwise slowpath */
-       flag = 1;
+       force_one_sack = 1;
        for (i = 0; i < num_sacks; i++) {
                __be32 start_seq = sp[i].start_seq;
                __be32 end_seq = sp[i].end_seq;
 
                if (i == 0) {
                        if (tp->recv_sack_cache[i].start_seq != start_seq)
-                               flag = 0;
+                               force_one_sack = 0;
                } else {
                        if ((tp->recv_sack_cache[i].start_seq != start_seq) ||
                            (tp->recv_sack_cache[i].end_seq != end_seq))
-                               flag = 0;
+                               force_one_sack = 0;
                }
                tp->recv_sack_cache[i].start_seq = start_seq;
                tp->recv_sack_cache[i].end_seq = end_seq;
@@ -1295,7 +1299,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
        }
 
        first_sack_index = 0;
-       if (flag)
+       if (force_one_sack)
                num_sacks = 1;
        else {
                int j;
@@ -1321,9 +1325,6 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                }
        }
 
-       /* clear flag as used for different purpose in following code */
-       flag = 0;
-
        /* Use SACK fastpath hint if valid */
        cached_skb = tp->fastpath_skb_hint;
        cached_fack_count = tp->fastpath_cnt_hint;
@@ -1332,12 +1333,15 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                cached_fack_count = 0;
        }
 
-       for (i=0; i<num_sacks; i++, sp++) {
+       for (i = 0; i < num_sacks; i++) {
                struct sk_buff *skb;
                __u32 start_seq = ntohl(sp->start_seq);
                __u32 end_seq = ntohl(sp->end_seq);
                int fack_count;
                int dup_sack = (found_dup_sack && (i == first_sack_index));
+               int next_dup = (found_dup_sack && (i+1 == first_sack_index));
+
+               sp++;
 
                if (!tcp_is_sackblock_valid(tp, dup_sack, start_seq, end_seq)) {
                        if (dup_sack) {
@@ -1363,7 +1367,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                        flag |= FLAG_DATA_LOST;
 
                tcp_for_write_queue_from(skb, sk) {
-                       int in_sack;
+                       int in_sack = 0;
                        u8 sacked;
 
                        if (skb == tcp_send_head(sk))
@@ -1382,11 +1386,25 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                        if (!before(TCP_SKB_CB(skb)->seq, end_seq))
                                break;
 
-                       in_sack = tcp_match_skb_to_sack(sk, skb, start_seq, end_seq);
-                       if (in_sack < 0)
-                               break;
+                       dup_sack = (found_dup_sack && (i == first_sack_index));
 
-                       fack_count += tcp_skb_pcount(skb);
+                       /* Due to sorting DSACK may reside within this SACK block! */
+                       if (next_dup) {
+                               u32 dup_start = ntohl(sp->start_seq);
+                               u32 dup_end = ntohl(sp->end_seq);
+
+                               if (before(TCP_SKB_CB(skb)->seq, dup_end)) {
+                                       in_sack = tcp_match_skb_to_sack(sk, skb, dup_start, dup_end);
+                                       if (in_sack > 0)
+                                               dup_sack = 1;
+                               }
+                       }
+
+                       /* DSACK info lost if out-of-mem, try SACK still */
+                       if (in_sack <= 0)
+                               in_sack = tcp_match_skb_to_sack(sk, skb, start_seq, end_seq);
+                       if (unlikely(in_sack < 0))
+                               break;
 
                        sacked = TCP_SKB_CB(skb)->sacked;
 
@@ -1402,19 +1420,17 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                                        if ((dup_sack && in_sack) &&
                                            (sacked&TCPCB_SACKED_ACKED))
                                                reord = min(fack_count, reord);
-                               } else {
-                                       /* If it was in a hole, we detected reordering. */
-                                       if (fack_count < prior_fackets &&
-                                           !(sacked&TCPCB_SACKED_ACKED))
-                                               reord = min(fack_count, reord);
                                }
 
                                /* Nothing to do; acked frame is about to be dropped. */
+                               fack_count += tcp_skb_pcount(skb);
                                continue;
                        }
 
-                       if (!in_sack)
+                       if (!in_sack) {
+                               fack_count += tcp_skb_pcount(skb);
                                continue;
+                       }
 
                        if (!(sacked&TCPCB_SACKED_ACKED)) {
                                if (sacked & TCPCB_SACKED_RETRANS) {
@@ -1431,12 +1447,17 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                                                tp->retransmit_skb_hint = NULL;
                                        }
                                } else {
-                                       /* New sack for not retransmitted frame,
-                                        * which was in hole. It is reordering.
-                                        */
-                                       if (!(sacked & TCPCB_RETRANS) &&
-                                           fack_count < prior_fackets)
-                                               reord = min(fack_count, reord);
+                                       if (!(sacked & TCPCB_RETRANS)) {
+                                               /* New sack for not retransmitted frame,
+                                                * which was in hole. It is reordering.
+                                                */
+                                               if (fack_count < prior_fackets)
+                                                       reord = min(fack_count, reord);
+
+                                               /* SACK enhanced F-RTO (RFC4138; Appendix B) */
+                                               if (!after(TCP_SKB_CB(skb)->end_seq, tp->frto_highmark))
+                                                       flag |= FLAG_ONLY_ORIG_SACKED;
+                                       }
 
                                        if (sacked & TCPCB_LOST) {
                                                TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST;
@@ -1445,24 +1466,13 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                                                /* clear lost hint */
                                                tp->retransmit_skb_hint = NULL;
                                        }
-                                       /* SACK enhanced F-RTO detection.
-                                        * Set flag if and only if non-rexmitted
-                                        * segments below frto_highmark are
-                                        * SACKed (RFC4138; Appendix B).
-                                        * Clearing correct due to in-order walk
-                                        */
-                                       if (after(end_seq, tp->frto_highmark)) {
-                                               flag &= ~FLAG_ONLY_ORIG_SACKED;
-                                       } else {
-                                               if (!(sacked & TCPCB_RETRANS))
-                                                       flag |= FLAG_ONLY_ORIG_SACKED;
-                                       }
                                }
 
                                TCP_SKB_CB(skb)->sacked |= TCPCB_SACKED_ACKED;
                                flag |= FLAG_DATA_SACKED;
                                tp->sacked_out += tcp_skb_pcount(skb);
 
+                               fack_count += tcp_skb_pcount(skb);
                                if (fack_count > tp->fackets_out)
                                        tp->fackets_out = fack_count;
 
@@ -1473,6 +1483,8 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                        } else {
                                if (dup_sack && (sacked&TCPCB_RETRANS))
                                        reord = min(fack_count, reord);
+
+                               fack_count += tcp_skb_pcount(skb);
                        }
 
                        /* D-SACK. We can detect redundant retransmission
@@ -1487,6 +1499,12 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
                                tp->retransmit_skb_hint = NULL;
                        }
                }
+
+               /* SACK enhanced FRTO (RFC4138, Appendix B): Clearing correct
+                * due to in-order walk
+                */
+               if (after(end_seq, tp->frto_highmark))
+                       flag &= ~FLAG_ONLY_ORIG_SACKED;
        }
 
        if (tp->retrans_out &&
@@ -1498,7 +1516,9 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 
        if ((reord < tp->fackets_out) && icsk->icsk_ca_state != TCP_CA_Loss &&
            (!tp->frto_highmark || after(tp->snd_una, tp->frto_highmark)))
-               tcp_update_reordering(sk, ((tp->fackets_out + 1) - reord), 0);
+               tcp_update_reordering(sk, tp->fackets_out - reord, 0);
+
+out:
 
 #if FASTRETRANS_DEBUG > 0
        BUG_TRAP((int)tp->sacked_out >= 0);
@@ -1654,6 +1674,9 @@ void tcp_enter_frto(struct sock *sk)
        }
        tcp_verify_left_out(tp);
 
+       /* Too bad if TCP was application limited */
+       tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp) + 1);
+
        /* Earlier loss recovery underway (see RFC4138; Appendix B).
         * The last condition is necessary at least in tp->frto_counter case.
         */
@@ -1686,6 +1709,8 @@ static void tcp_enter_frto_loss(struct sock *sk, int allowed_segments, int flag)
        tcp_for_write_queue(skb, sk) {
                if (skb == tcp_send_head(sk))
                        break;
+
+               TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST;
                /*
                 * Count the retransmission made on RTO correctly (only when
                 * waiting for the first ACK and did not get it)...
@@ -1699,7 +1724,7 @@ static void tcp_enter_frto_loss(struct sock *sk, int allowed_segments, int flag)
                } else {
                        if (TCP_SKB_CB(skb)->sacked & TCPCB_RETRANS)
                                tp->undo_marker = 0;
-                       TCP_SKB_CB(skb)->sacked &= ~(TCPCB_LOST|TCPCB_SACKED_RETRANS);
+                       TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
                }
 
                /* Don't lost mark skbs that were fwd transmitted after RTO */
@@ -2061,7 +2086,7 @@ static void tcp_update_scoreboard(struct sock *sk)
                        if (!tcp_skb_timedout(sk, skb))
                                break;
 
-                       if (!(TCP_SKB_CB(skb)->sacked&TCPCB_TAGBITS)) {
+                       if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_SACKED_ACKED|TCPCB_LOST))) {
                                TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
                                tp->lost_out += tcp_skb_pcount(skb);
                                tcp_verify_retransmit_hint(tp, skb);
@@ -2613,7 +2638,8 @@ static u32 tcp_tso_acked(struct sock *sk, struct sk_buff *skb)
  * is before the ack sequence we can discard it as it's confirmed to have
  * arrived at the other end.
  */
-static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p)
+static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p,
+                              int prior_fackets)
 {
        struct tcp_sock *tp = tcp_sk(sk);
        const struct inet_connection_sock *icsk = inet_csk(sk);
@@ -2622,6 +2648,8 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p)
        int fully_acked = 1;
        int flag = 0;
        int prior_packets = tp->packets_out;
+       u32 cnt = 0;
+       u32 reord = tp->packets_out;
        s32 seq_rtt = -1;
        ktime_t last_ackt = net_invalid_timestamp();
 
@@ -2662,10 +2690,14 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p)
                                if ((flag & FLAG_DATA_ACKED) ||
                                    (packets_acked > 1))
                                        flag |= FLAG_NONHEAD_RETRANS_ACKED;
-                       } else if (seq_rtt < 0) {
-                               seq_rtt = now - scb->when;
-                               if (fully_acked)
-                                       last_ackt = skb->tstamp;
+                       } else {
+                               if (seq_rtt < 0) {
+                                       seq_rtt = now - scb->when;
+                                       if (fully_acked)
+                                               last_ackt = skb->tstamp;
+                               }
+                               if (!(sacked & TCPCB_SACKED_ACKED))
+                                       reord = min(cnt, reord);
                        }
 
                        if (sacked & TCPCB_SACKED_ACKED)
@@ -2676,12 +2708,16 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p)
                        if ((sacked & TCPCB_URG) && tp->urg_mode &&
                            !before(end_seq, tp->snd_up))
                                tp->urg_mode = 0;
-               } else if (seq_rtt < 0) {
-                       seq_rtt = now - scb->when;
-                       if (fully_acked)
-                               last_ackt = skb->tstamp;
+               } else {
+                       if (seq_rtt < 0) {
+                               seq_rtt = now - scb->when;
+                               if (fully_acked)
+                                       last_ackt = skb->tstamp;
+                       }
+                       reord = min(cnt, reord);
                }
                tp->packets_out -= packets_acked;
+               cnt += packets_acked;
 
                /* Initial outgoing SYN's get put onto the write_queue
                 * just like anything else we transmit.  It is not
@@ -2713,13 +2749,18 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p)
                tcp_ack_update_rtt(sk, flag, seq_rtt);
                tcp_rearm_rto(sk);
 
+               if (tcp_is_reno(tp)) {
+                       tcp_remove_reno_sacks(sk, pkts_acked);
+               } else {
+                       /* Non-retransmitted hole got filled? That's reordering */
+                       if (reord < prior_fackets)
+                               tcp_update_reordering(sk, tp->fackets_out - reord, 0);
+               }
+
                tp->fackets_out -= min(pkts_acked, tp->fackets_out);
                /* hint's skb might be NULL but we don't need to care */
                tp->fastpath_cnt_hint -= min_t(u32, pkts_acked,
                                               tp->fastpath_cnt_hint);
-               if (tcp_is_reno(tp))
-                       tcp_remove_reno_sacks(sk, pkts_acked);
-
                if (ca_ops->pkts_acked) {
                        s32 rtt_us = -1;
 
@@ -3002,6 +3043,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
        u32 ack_seq = TCP_SKB_CB(skb)->seq;
        u32 ack = TCP_SKB_CB(skb)->ack_seq;
        u32 prior_in_flight;
+       u32 prior_fackets;
        s32 seq_rtt;
        int prior_packets;
        int frto_cwnd = 0;
@@ -3026,6 +3068,8 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
                        tp->bytes_acked += min(ack - prior_snd_una, tp->mss_cache);
        }
 
+       prior_fackets = tp->fackets_out;
+
        if (!(flag&FLAG_SLOWPATH) && after(ack, prior_snd_una)) {
                /* Window is constant, pure forward advance.
                 * No more checks are required.
@@ -3067,13 +3111,13 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
        prior_in_flight = tcp_packets_in_flight(tp);
 
        /* See if we can take anything off of the retransmit queue. */
-       flag |= tcp_clean_rtx_queue(sk, &seq_rtt);
+       flag |= tcp_clean_rtx_queue(sk, &seq_rtt, prior_fackets);
 
+       if (tp->frto_counter)
+               frto_cwnd = tcp_process_frto(sk, flag);
        /* Guarantee sacktag reordering detection against wrap-arounds */
        if (before(tp->frto_highmark, tp->snd_una))
                tp->frto_highmark = 0;
-       if (tp->frto_counter)
-               frto_cwnd = tcp_process_frto(sk, flag);
 
        if (tcp_ack_is_dubious(sk, flag)) {
                /* Advance CWND, if state allows this. */