X-Git-Url: http://pilppa.org/gitweb/gitweb.cgi?a=blobdiff_plain;f=net%2Fipv4%2Ftcp_input.c;h=18e099c6fa62b627c0ab0aef93c5da22a2eb22b2;hb=056834d9f6f6eaf4cc7268569e53acab957aac27;hp=b59da5308ac0ed35c135162fbfcc54cac32c8bd6;hpb=746aa32d280084dbd520249170852e4616799928;p=linux-2.6-omap-h63xx.git diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index b59da5308ac..18e099c6fa6 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -105,6 +105,7 @@ int sysctl_tcp_abc __read_mostly; #define FLAG_SND_UNA_ADVANCED 0x400 /* Snd_una was changed (!= FLAG_DATA_ACKED) */ #define FLAG_DSACKING_ACK 0x800 /* SACK blocks contained D-SACK info */ #define FLAG_NONHEAD_RETRANS_ACKED 0x1000 /* Non-head rexmitted data was ACKed */ +#define FLAG_SACK_RENEGING 0x2000 /* snd_una advanced to a sacked seq */ #define FLAG_ACKED (FLAG_DATA_ACKED|FLAG_SYN_ACKED) #define FLAG_NOT_DUP (FLAG_DATA|FLAG_WIN_UPDATE|FLAG_ACKED) @@ -120,8 +121,7 @@ int sysctl_tcp_abc __read_mostly; /* Adapt the MSS value used to make delayed ack decision to the * real world. */ -static void tcp_measure_rcv_mss(struct sock *sk, - const struct sk_buff *skb) +static void tcp_measure_rcv_mss(struct sock *sk, const struct sk_buff *skb) { struct inet_connection_sock *icsk = inet_csk(sk); const unsigned int lss = icsk->icsk_ack.last_seg_size; @@ -132,7 +132,7 @@ static void tcp_measure_rcv_mss(struct sock *sk, /* skb->len may jitter because of SACKs, even if peer * sends good full-sized frames. */ - len = skb_shinfo(skb)->gso_size ?: skb->len; + len = skb_shinfo(skb)->gso_size ? : skb->len; if (len >= icsk->icsk_ack.rcv_mss) { icsk->icsk_ack.rcv_mss = len; } else { @@ -172,8 +172,8 @@ static void tcp_incr_quickack(struct sock *sk) struct inet_connection_sock *icsk = inet_csk(sk); unsigned quickacks = tcp_sk(sk)->rcv_wnd / (2 * icsk->icsk_ack.rcv_mss); - if (quickacks==0) - quickacks=2; + if (quickacks == 0) + quickacks = 2; if (quickacks > icsk->icsk_ack.quick) icsk->icsk_ack.quick = min(quickacks, TCP_MAX_QUICKACKS); } @@ -198,7 +198,7 @@ static inline int tcp_in_quickack_mode(const struct sock *sk) static inline void TCP_ECN_queue_cwr(struct tcp_sock *tp) { - if (tp->ecn_flags&TCP_ECN_OK) + if (tp->ecn_flags & TCP_ECN_OK) tp->ecn_flags |= TCP_ECN_QUEUE_CWR; } @@ -215,7 +215,7 @@ static inline void TCP_ECN_withdraw_cwr(struct tcp_sock *tp) static inline void TCP_ECN_check_ce(struct tcp_sock *tp, struct sk_buff *skb) { - if (tp->ecn_flags&TCP_ECN_OK) { + if (tp->ecn_flags & TCP_ECN_OK) { if (INET_ECN_is_ce(TCP_SKB_CB(skb)->flags)) tp->ecn_flags |= TCP_ECN_DEMAND_CWR; /* Funny extension: if ECT is not set on a segment, @@ -228,19 +228,19 @@ static inline void TCP_ECN_check_ce(struct tcp_sock *tp, struct sk_buff *skb) static inline void TCP_ECN_rcv_synack(struct tcp_sock *tp, struct tcphdr *th) { - if ((tp->ecn_flags&TCP_ECN_OK) && (!th->ece || th->cwr)) + if ((tp->ecn_flags & TCP_ECN_OK) && (!th->ece || th->cwr)) tp->ecn_flags &= ~TCP_ECN_OK; } static inline void TCP_ECN_rcv_syn(struct tcp_sock *tp, struct tcphdr *th) { - if ((tp->ecn_flags&TCP_ECN_OK) && (!th->ece || !th->cwr)) + if ((tp->ecn_flags & TCP_ECN_OK) && (!th->ece || !th->cwr)) tp->ecn_flags &= ~TCP_ECN_OK; } static inline int TCP_ECN_rcv_ecn_echo(struct tcp_sock *tp, struct tcphdr *th) { - if (th->ece && !th->syn && (tp->ecn_flags&TCP_ECN_OK)) + if (th->ece && !th->syn && (tp->ecn_flags & TCP_ECN_OK)) return 1; return 0; } @@ -289,8 +289,8 @@ static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); /* Optimize this! */ - int truesize = tcp_win_from_space(skb->truesize)/2; - int window = tcp_win_from_space(sysctl_tcp_rmem[2])/2; + int truesize = tcp_win_from_space(skb->truesize) >> 1; + int window = tcp_win_from_space(sysctl_tcp_rmem[2]) >> 1; while (tp->rcv_ssthresh <= window) { if (truesize <= skb->len) @@ -302,8 +302,7 @@ static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb) return 0; } -static void tcp_grow_window(struct sock *sk, - struct sk_buff *skb) +static void tcp_grow_window(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); @@ -317,12 +316,13 @@ static void tcp_grow_window(struct sock *sk, * will fit to rcvbuf in future. */ if (tcp_win_from_space(skb->truesize) <= skb->len) - incr = 2*tp->advmss; + incr = 2 * tp->advmss; else incr = __tcp_grow_window(sk, skb); if (incr) { - tp->rcv_ssthresh = min(tp->rcv_ssthresh + incr, tp->window_clamp); + tp->rcv_ssthresh = min(tp->rcv_ssthresh + incr, + tp->window_clamp); inet_csk(sk)->icsk_ack.quick |= 1; } } @@ -397,10 +397,9 @@ static void tcp_clamp_window(struct sock *sk) sysctl_tcp_rmem[2]); } if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf) - tp->rcv_ssthresh = min(tp->window_clamp, 2U*tp->advmss); + tp->rcv_ssthresh = min(tp->window_clamp, 2U * tp->advmss); } - /* Initialize RCV_MSS value. * RCV_MSS is an our guess about MSS used by the peer. * We haven't any direct information about the MSS. @@ -413,7 +412,7 @@ void tcp_initialize_rcv_mss(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); unsigned int hint = min_t(unsigned int, tp->advmss, tp->mss_cache); - hint = min(hint, tp->rcv_wnd/2); + hint = min(hint, tp->rcv_wnd / 2); hint = min(hint, TCP_MIN_RCVMSS); hint = max(hint, TCP_MIN_MSS); @@ -470,16 +469,15 @@ static inline void tcp_rcv_rtt_measure(struct tcp_sock *tp) goto new_measure; if (before(tp->rcv_nxt, tp->rcv_rtt_est.seq)) return; - tcp_rcv_rtt_update(tp, - jiffies - tp->rcv_rtt_est.time, - 1); + tcp_rcv_rtt_update(tp, jiffies - tp->rcv_rtt_est.time, 1); new_measure: tp->rcv_rtt_est.seq = tp->rcv_nxt + tp->rcv_wnd; tp->rcv_rtt_est.time = tcp_time_stamp; } -static inline void tcp_rcv_rtt_measure_ts(struct sock *sk, const struct sk_buff *skb) +static inline void tcp_rcv_rtt_measure_ts(struct sock *sk, + const struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); if (tp->rx_opt.rcv_tsecr && @@ -502,8 +500,7 @@ void tcp_rcv_space_adjust(struct sock *sk) goto new_measure; time = tcp_time_stamp - tp->rcvq_space.time; - if (time < (tp->rcv_rtt_est.rtt >> 3) || - tp->rcv_rtt_est.rtt == 0) + if (time < (tp->rcv_rtt_est.rtt >> 3) || tp->rcv_rtt_est.rtt == 0) return; space = 2 * (tp->copied_seq - tp->rcvq_space.seq); @@ -579,7 +576,7 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb) } else { int m = now - icsk->icsk_ack.lrcvtime; - if (m <= TCP_ATO_MIN/2) { + if (m <= TCP_ATO_MIN / 2) { /* The fastest case is the first. */ icsk->icsk_ack.ato = (icsk->icsk_ack.ato >> 1) + TCP_ATO_MIN / 2; } else if (m < icsk->icsk_ack.ato) { @@ -591,7 +588,7 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb) * restart window, so that we send ACKs quickly. */ tcp_incr_quickack(sk); - sk_stream_mem_reclaim(sk); + sk_mem_reclaim(sk); } } icsk->icsk_ack.lrcvtime = now; @@ -608,7 +605,7 @@ 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->metrics[RTAX_RTO_MIN - 1]; return rto_min; } @@ -671,14 +668,14 @@ static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt) } if (after(tp->snd_una, tp->rtt_seq)) { if (tp->mdev_max < tp->rttvar) - tp->rttvar -= (tp->rttvar-tp->mdev_max)>>2; + tp->rttvar -= (tp->rttvar - tp->mdev_max) >> 2; tp->rtt_seq = tp->snd_nxt; tp->mdev_max = tcp_rto_min(sk); } } else { /* no previous measure. */ - tp->srtt = m<<3; /* take the measured time to be rtt */ - tp->mdev = m<<1; /* make sure rto = 3*rtt */ + tp->srtt = m << 3; /* take the measured time to be rtt */ + tp->mdev = m << 1; /* make sure rto = 3*rtt */ tp->mdev_max = tp->rttvar = max(tp->mdev, tcp_rto_min(sk)); tp->rtt_seq = tp->snd_nxt; } @@ -732,7 +729,7 @@ void tcp_update_metrics(struct sock *sk) dst_confirm(dst); - if (dst && (dst->flags&DST_HOST)) { + if (dst && (dst->flags & DST_HOST)) { const struct inet_connection_sock *icsk = inet_csk(sk); int m; @@ -742,7 +739,7 @@ void tcp_update_metrics(struct sock *sk) * Reset our results. */ if (!(dst_metric_locked(dst, RTAX_RTT))) - dst->metrics[RTAX_RTT-1] = 0; + dst->metrics[RTAX_RTT - 1] = 0; return; } @@ -754,9 +751,9 @@ void tcp_update_metrics(struct sock *sk) */ if (!(dst_metric_locked(dst, RTAX_RTT))) { if (m <= 0) - dst->metrics[RTAX_RTT-1] = tp->srtt; + dst->metrics[RTAX_RTT - 1] = tp->srtt; else - dst->metrics[RTAX_RTT-1] -= (m>>3); + dst->metrics[RTAX_RTT - 1] -= (m >> 3); } if (!(dst_metric_locked(dst, RTAX_RTTVAR))) { @@ -769,7 +766,7 @@ void tcp_update_metrics(struct sock *sk) m = tp->mdev; if (m >= dst_metric(dst, RTAX_RTTVAR)) - dst->metrics[RTAX_RTTVAR-1] = m; + dst->metrics[RTAX_RTTVAR - 1] = m; else dst->metrics[RTAX_RTTVAR-1] -= (dst->metrics[RTAX_RTTVAR-1] - m)>>2; @@ -783,7 +780,7 @@ void tcp_update_metrics(struct sock *sk) dst->metrics[RTAX_SSTHRESH-1] = tp->snd_cwnd >> 1; if (!dst_metric_locked(dst, RTAX_CWND) && tp->snd_cwnd > dst_metric(dst, RTAX_CWND)) - dst->metrics[RTAX_CWND-1] = tp->snd_cwnd; + dst->metrics[RTAX_CWND - 1] = tp->snd_cwnd; } else if (tp->snd_cwnd > tp->snd_ssthresh && icsk->icsk_ca_state == TCP_CA_Open) { /* Cong. avoidance phase, cwnd is reliable. */ @@ -863,6 +860,9 @@ void tcp_enter_cwr(struct sock *sk, const int set_ssthresh) */ static void tcp_disable_fack(struct tcp_sock *tp) { + /* RFC3517 uses different metric in lost marker => reset on change */ + if (tcp_is_fack(tp)) + tp->lost_skb_hint = NULL; tp->rx_opt.sack_ok &= ~2; } @@ -923,7 +923,7 @@ static void tcp_init_metrics(struct sock *sk) } if (dst_metric(dst, RTAX_RTTVAR) > tp->mdev) { tp->mdev = dst_metric(dst, RTAX_RTTVAR); - tp->mdev_max = tp->rttvar = max(tp->mdev, TCP_RTO_MIN); + tp->mdev_max = tp->rttvar = max(tp->mdev, tcp_rto_min(sk)); } tcp_set_rto(sk); tcp_bound_rto(sk); @@ -1112,16 +1112,22 @@ static int tcp_is_sackblock_valid(struct tcp_sock *tp, int is_dsack, * * Search retransmitted skbs from write_queue that were sent when snd_nxt was * less than what is now known to be received by the other end (derived from - * SACK blocks by the caller). Also calculate the lowest snd_nxt among the - * remaining retransmitted skbs to avoid some costly processing per ACKs. + * highest SACK block). Also calculate the lowest snd_nxt among the remaining + * retransmitted skbs to avoid some costly processing per ACKs. */ -static int tcp_mark_lost_retrans(struct sock *sk, u32 received_upto) +static void tcp_mark_lost_retrans(struct sock *sk) { + const struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; - int flag = 0; int cnt = 0; u32 new_low_seq = tp->snd_nxt; + u32 received_upto = tcp_highest_sack_seq(tp); + + if (!tcp_is_fack(tp) || !tp->retrans_out || + !after(received_upto, tp->lost_retrans_low) || + icsk->icsk_ca_state != TCP_CA_Recovery) + return; tcp_for_write_queue(skb, sk) { u32 ack_seq = TCP_SKB_CB(skb)->ack_seq; @@ -1149,9 +1155,8 @@ static int tcp_mark_lost_retrans(struct sock *sk, u32 received_upto) if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) { tp->lost_out += tcp_skb_pcount(skb); TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; - flag |= FLAG_DATA_SACKED; - NET_INC_STATS_BH(LINUX_MIB_TCPLOSTRETRANSMIT); } + NET_INC_STATS_BH(LINUX_MIB_TCPLOSTRETRANSMIT); } else { if (before(ack_seq, new_low_seq)) new_low_seq = ack_seq; @@ -1161,8 +1166,6 @@ static int tcp_mark_lost_retrans(struct sock *sk, u32 received_upto) if (tp->retrans_out) tp->lost_retrans_low = new_low_seq; - - return flag; } static int tcp_check_dsack(struct tcp_sock *tp, struct sk_buff *ack_skb, @@ -1230,34 +1233,205 @@ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb, return in_sack; } +static int tcp_sacktag_one(struct sk_buff *skb, struct sock *sk, + int *reord, int dup_sack, int fack_count) +{ + struct tcp_sock *tp = tcp_sk(sk); + u8 sacked = TCP_SKB_CB(skb)->sacked; + int flag = 0; + + /* Account D-SACK for retransmitted packet. */ + if (dup_sack && (sacked & TCPCB_RETRANS)) { + if (after(TCP_SKB_CB(skb)->end_seq, tp->undo_marker)) + tp->undo_retrans--; + if (sacked & TCPCB_SACKED_ACKED) + *reord = min(fack_count, *reord); + } + + /* Nothing to do; acked frame is about to be dropped (was ACKed). */ + if (!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una)) + return flag; + + if (!(sacked & TCPCB_SACKED_ACKED)) { + if (sacked & TCPCB_SACKED_RETRANS) { + /* If the segment is not tagged as lost, + * we do not clear RETRANS, believing + * that retransmission is still in flight. + */ + if (sacked & TCPCB_LOST) { + TCP_SKB_CB(skb)->sacked &= + ~(TCPCB_LOST|TCPCB_SACKED_RETRANS); + tp->lost_out -= tcp_skb_pcount(skb); + tp->retrans_out -= tcp_skb_pcount(skb); + + /* clear lost hint */ + tp->retransmit_skb_hint = NULL; + } + } else { + if (!(sacked & TCPCB_RETRANS)) { + /* New sack for not retransmitted frame, + * which was in hole. It is reordering. + */ + if (before(TCP_SKB_CB(skb)->seq, + tcp_highest_sack_seq(tp))) + *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; + tp->lost_out -= tcp_skb_pcount(skb); + + /* clear lost hint */ + tp->retransmit_skb_hint = NULL; + } + } + + 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); + + /* Lost marker hint past SACKed? Tweak RFC3517 cnt */ + if (!tcp_is_fack(tp) && (tp->lost_skb_hint != NULL) && + before(TCP_SKB_CB(skb)->seq, + TCP_SKB_CB(tp->lost_skb_hint)->seq)) + tp->lost_cnt_hint += tcp_skb_pcount(skb); + + if (fack_count > tp->fackets_out) + tp->fackets_out = fack_count; + + if (!before(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp))) + tcp_advance_highest_sack(sk, skb); + } + + /* D-SACK. We can detect redundant retransmission in S|R and plain R + * frames and clear it. undo_retrans is decreased above, L|R frames + * are accounted above as well. + */ + if (dup_sack && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS)) { + TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; + tp->retrans_out -= tcp_skb_pcount(skb); + tp->retransmit_skb_hint = NULL; + } + + return flag; +} + +static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk, + struct tcp_sack_block *next_dup, + u32 start_seq, u32 end_seq, + int dup_sack_in, int *fack_count, + int *reord, int *flag) +{ + tcp_for_write_queue_from(skb, sk) { + int in_sack = 0; + int dup_sack = dup_sack_in; + + if (skb == tcp_send_head(sk)) + break; + + /* queue is in-order => we can short-circuit the walk early */ + if (!before(TCP_SKB_CB(skb)->seq, end_seq)) + break; + + if ((next_dup != NULL) && + before(TCP_SKB_CB(skb)->seq, next_dup->end_seq)) { + in_sack = tcp_match_skb_to_sack(sk, skb, + next_dup->start_seq, + next_dup->end_seq); + if (in_sack > 0) + dup_sack = 1; + } + + if (in_sack <= 0) + in_sack = tcp_match_skb_to_sack(sk, skb, start_seq, + end_seq); + if (unlikely(in_sack < 0)) + break; + + if (in_sack) + *flag |= tcp_sacktag_one(skb, sk, reord, dup_sack, + *fack_count); + + *fack_count += tcp_skb_pcount(skb); + } + return skb; +} + +/* Avoid all extra work that is being done by sacktag while walking in + * a normal way + */ +static struct sk_buff *tcp_sacktag_skip(struct sk_buff *skb, struct sock *sk, + u32 skip_to_seq) +{ + tcp_for_write_queue_from(skb, sk) { + if (skb == tcp_send_head(sk)) + break; + + if (!before(TCP_SKB_CB(skb)->end_seq, skip_to_seq)) + break; + } + return skb; +} + +static struct sk_buff *tcp_maybe_skipping_dsack(struct sk_buff *skb, + struct sock *sk, + struct tcp_sack_block *next_dup, + u32 skip_to_seq, + int *fack_count, int *reord, + int *flag) +{ + if (next_dup == NULL) + return skb; + + if (before(next_dup->start_seq, skip_to_seq)) { + skb = tcp_sacktag_skip(skb, sk, next_dup->start_seq); + tcp_sacktag_walk(skb, sk, NULL, + next_dup->start_seq, next_dup->end_seq, + 1, fack_count, reord, flag); + } + + return skb; +} + +static int tcp_sack_cache_ok(struct tcp_sock *tp, struct tcp_sack_block *cache) +{ + return cache < tp->recv_sack_cache + ARRAY_SIZE(tp->recv_sack_cache); +} + static int -tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_una) +tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, + u32 prior_snd_una) { const struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); unsigned char *ptr = (skb_transport_header(ack_skb) + TCP_SKB_CB(ack_skb)->sacked); - struct tcp_sack_block_wire *sp = (struct tcp_sack_block_wire *)(ptr+2); - struct sk_buff *cached_skb; - int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE)>>3; + struct tcp_sack_block_wire *sp_wire = (struct tcp_sack_block_wire *)(ptr+2); + struct tcp_sack_block sp[4]; + struct tcp_sack_block *cache; + struct sk_buff *skb; + int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE) >> 3; + int used_sacks; int reord = tp->packets_out; - int prior_fackets; - u32 highest_sack_end_seq = tp->lost_retrans_low; int flag = 0; int found_dup_sack = 0; - int cached_fack_count; - int i; + int fack_count; + int i, j; int first_sack_index; - int force_one_sack; if (!tp->sacked_out) { if (WARN_ON(tp->fackets_out)) tp->fackets_out = 0; - tp->highest_sack = tp->snd_una; + tcp_highest_sack_reset(sk); } - prior_fackets = tp->fackets_out; - found_dup_sack = tcp_check_dsack(tp, ack_skb, sp, + found_dup_sack = tcp_check_dsack(tp, ack_skb, sp_wire, num_sacks, prior_snd_una); if (found_dup_sack) flag |= FLAG_DSACKING_ACK; @@ -1269,78 +1443,20 @@ 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; - /* 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 */ - 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) - force_one_sack = 0; - } else { - if ((tp->recv_sack_cache[i].start_seq != start_seq) || - (tp->recv_sack_cache[i].end_seq != end_seq)) - force_one_sack = 0; - } - tp->recv_sack_cache[i].start_seq = start_seq; - tp->recv_sack_cache[i].end_seq = end_seq; - } - /* Clear the rest of the cache sack blocks so they won't match mistakenly. */ - for (; i < ARRAY_SIZE(tp->recv_sack_cache); i++) { - tp->recv_sack_cache[i].start_seq = 0; - tp->recv_sack_cache[i].end_seq = 0; - } + if (!tp->packets_out) + goto out; + used_sacks = 0; first_sack_index = 0; - if (force_one_sack) - num_sacks = 1; - else { - int j; - tp->fastpath_skb_hint = NULL; - - /* order SACK blocks to allow in order walk of the retrans queue */ - for (i = num_sacks-1; i > 0; i--) { - for (j = 0; j < i; j++){ - if (after(ntohl(sp[j].start_seq), - ntohl(sp[j+1].start_seq))){ - struct tcp_sack_block_wire tmp; - - tmp = sp[j]; - sp[j] = sp[j+1]; - sp[j+1] = tmp; - - /* Track where the first SACK block goes to */ - if (j == first_sack_index) - first_sack_index = j+1; - } - - } - } - } - - /* Use SACK fastpath hint if valid */ - cached_skb = tp->fastpath_skb_hint; - cached_fack_count = tp->fastpath_cnt_hint; - if (!cached_skb) { - cached_skb = tcp_write_queue_head(sk); - cached_fack_count = 0; - } - 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)); + int dup_sack = !i && found_dup_sack; - sp++; + sp[used_sacks].start_seq = ntohl(get_unaligned(&sp_wire[i].start_seq)); + sp[used_sacks].end_seq = ntohl(get_unaligned(&sp_wire[i].end_seq)); - if (!tcp_is_sackblock_valid(tp, dup_sack, start_seq, end_seq)) { + if (!tcp_is_sackblock_valid(tp, dup_sack, + sp[used_sacks].start_seq, + sp[used_sacks].end_seq)) { if (dup_sack) { if (!tp->undo_marker) NET_INC_STATS_BH(LINUX_MIB_TCPDSACKIGNOREDNOUNDO); @@ -1349,172 +1465,153 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ } else { /* Don't count olds caused by ACK reordering */ if ((TCP_SKB_CB(ack_skb)->ack_seq != tp->snd_una) && - !after(end_seq, tp->snd_una)) + !after(sp[used_sacks].end_seq, tp->snd_una)) continue; NET_INC_STATS_BH(LINUX_MIB_TCPSACKDISCARD); } + if (i == 0) + first_sack_index = -1; continue; } - skb = cached_skb; - fack_count = cached_fack_count; - - /* Event "B" in the comment above. */ - if (after(end_seq, tp->high_seq)) - flag |= FLAG_DATA_LOST; - - tcp_for_write_queue_from(skb, sk) { - int in_sack = 0; - u8 sacked; - - if (skb == tcp_send_head(sk)) - break; - - cached_skb = skb; - cached_fack_count = fack_count; - if (i == first_sack_index) { - tp->fastpath_skb_hint = skb; - tp->fastpath_cnt_hint = fack_count; - } + /* Ignore very old stuff early */ + if (!after(sp[used_sacks].end_seq, prior_snd_una)) + continue; - /* The retransmission queue is always in order, so - * we can short-circuit the walk early. - */ - if (!before(TCP_SKB_CB(skb)->seq, end_seq)) - break; + used_sacks++; + } - dup_sack = (found_dup_sack && (i == first_sack_index)); + /* order SACK blocks to allow in order walk of the retrans queue */ + for (i = used_sacks - 1; i > 0; i--) { + for (j = 0; j < i; j++) { + if (after(sp[j].start_seq, sp[j + 1].start_seq)) { + struct tcp_sack_block tmp; - /* 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); + tmp = sp[j]; + sp[j] = sp[j + 1]; + sp[j + 1] = tmp; - 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; - } + /* Track where the first SACK block goes to */ + if (j == first_sack_index) + first_sack_index = j + 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; + skb = tcp_write_queue_head(sk); + fack_count = 0; + i = 0; - sacked = TCP_SKB_CB(skb)->sacked; + if (!tp->sacked_out) { + /* It's already past, so skip checking against it */ + cache = tp->recv_sack_cache + ARRAY_SIZE(tp->recv_sack_cache); + } else { + cache = tp->recv_sack_cache; + /* Skip empty blocks in at head of the cache */ + while (tcp_sack_cache_ok(tp, cache) && !cache->start_seq && + !cache->end_seq) + cache++; + } - /* Account D-SACK for retransmitted packet. */ - if ((dup_sack && in_sack) && - (sacked & TCPCB_RETRANS) && - after(TCP_SKB_CB(skb)->end_seq, tp->undo_marker)) - tp->undo_retrans--; + while (i < used_sacks) { + u32 start_seq = sp[i].start_seq; + u32 end_seq = sp[i].end_seq; + int dup_sack = (found_dup_sack && (i == first_sack_index)); + struct tcp_sack_block *next_dup = NULL; - /* The frame is ACKed. */ - if (!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una)) { - if (sacked&TCPCB_RETRANS) { - if ((dup_sack && in_sack) && - (sacked&TCPCB_SACKED_ACKED)) - reord = min(fack_count, reord); - } + if (found_dup_sack && ((i + 1) == first_sack_index)) + next_dup = &sp[i + 1]; - /* Nothing to do; acked frame is about to be dropped. */ - fack_count += tcp_skb_pcount(skb); - continue; - } + /* Event "B" in the comment above. */ + if (after(end_seq, tp->high_seq)) + flag |= FLAG_DATA_LOST; - if (!in_sack) { - fack_count += tcp_skb_pcount(skb); - continue; + /* Skip too early cached blocks */ + while (tcp_sack_cache_ok(tp, cache) && + !before(start_seq, cache->end_seq)) + cache++; + + /* Can skip some work by looking recv_sack_cache? */ + if (tcp_sack_cache_ok(tp, cache) && !dup_sack && + after(end_seq, cache->start_seq)) { + + /* Head todo? */ + if (before(start_seq, cache->start_seq)) { + skb = tcp_sacktag_skip(skb, sk, start_seq); + skb = tcp_sacktag_walk(skb, sk, next_dup, + start_seq, + cache->start_seq, + dup_sack, &fack_count, + &reord, &flag); } - if (!(sacked&TCPCB_SACKED_ACKED)) { - if (sacked & TCPCB_SACKED_RETRANS) { - /* If the segment is not tagged as lost, - * we do not clear RETRANS, believing - * that retransmission is still in flight. - */ - if (sacked & TCPCB_LOST) { - TCP_SKB_CB(skb)->sacked &= ~(TCPCB_LOST|TCPCB_SACKED_RETRANS); - tp->lost_out -= tcp_skb_pcount(skb); - tp->retrans_out -= tcp_skb_pcount(skb); + /* Rest of the block already fully processed? */ + if (!after(end_seq, cache->end_seq)) + goto advance_sp; - /* clear lost hint */ - tp->retransmit_skb_hint = NULL; - } - } else { - 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); + skb = tcp_maybe_skipping_dsack(skb, sk, next_dup, + cache->end_seq, + &fack_count, &reord, + &flag); - /* 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; - tp->lost_out -= tcp_skb_pcount(skb); - - /* clear lost hint */ - tp->retransmit_skb_hint = NULL; - } - } - - 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; - - if (after(TCP_SKB_CB(skb)->seq, tp->highest_sack)) { - tp->highest_sack = TCP_SKB_CB(skb)->seq; - highest_sack_end_seq = TCP_SKB_CB(skb)->end_seq; - } - } else { - if (dup_sack && (sacked&TCPCB_RETRANS)) - reord = min(fack_count, reord); - - fack_count += tcp_skb_pcount(skb); + /* ...tail remains todo... */ + if (tcp_highest_sack_seq(tp) == cache->end_seq) { + /* ...but better entrypoint exists! */ + skb = tcp_highest_sack(sk); + if (skb == NULL) + break; + fack_count = tp->fackets_out; + cache++; + goto walk; } - /* D-SACK. We can detect redundant retransmission - * in S|R and plain R frames and clear it. - * undo_retrans is decreased above, L|R frames - * are accounted above as well. - */ - if (dup_sack && - (TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS)) { - TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; - tp->retrans_out -= tcp_skb_pcount(skb); - tp->retransmit_skb_hint = NULL; - } + skb = tcp_sacktag_skip(skb, sk, cache->end_seq); + /* Check overlap against next cached too (past this one already) */ + cache++; + continue; + } + + if (!before(start_seq, tcp_highest_sack_seq(tp))) { + skb = tcp_highest_sack(sk); + if (skb == NULL) + break; + fack_count = tp->fackets_out; } + skb = tcp_sacktag_skip(skb, sk, start_seq); +walk: + skb = tcp_sacktag_walk(skb, sk, next_dup, start_seq, end_seq, + dup_sack, &fack_count, &reord, &flag); + +advance_sp: /* 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; + + i++; + } + + /* Clear the head of the cache sack blocks so we can skip it next time */ + for (i = 0; i < ARRAY_SIZE(tp->recv_sack_cache) - used_sacks; i++) { + tp->recv_sack_cache[i].start_seq = 0; + tp->recv_sack_cache[i].end_seq = 0; } + for (j = 0; j < used_sacks; j++) + tp->recv_sack_cache[i++] = sp[j]; - if (tp->retrans_out && - after(highest_sack_end_seq, tp->lost_retrans_low) && - icsk->icsk_ca_state == TCP_CA_Recovery) - flag |= tcp_mark_lost_retrans(sk, highest_sack_end_seq); + tcp_mark_lost_retrans(sk); tcp_verify_left_out(tp); - if ((reord < tp->fackets_out) && icsk->icsk_ca_state != TCP_CA_Loss && + if ((reord < tp->fackets_out) && + ((icsk->icsk_ca_state != TCP_CA_Loss) || tp->undo_marker) && (!tp->frto_highmark || after(tp->snd_una, tp->frto_highmark))) tcp_update_reordering(sk, tp->fackets_out - reord, 0); +out: + #if FASTRETRANS_DEBUG > 0 BUG_TRAP((int)tp->sacked_out >= 0); BUG_TRAP((int)tp->lost_out >= 0); @@ -1560,10 +1657,10 @@ static void tcp_remove_reno_sacks(struct sock *sk, int acked) if (acked > 0) { /* One ACK acked hole. The rest eat duplicate ACKs. */ - if (acked-1 >= tp->sacked_out) + if (acked - 1 >= tp->sacked_out) tp->sacked_out = 0; else - tp->sacked_out -= acked-1; + tp->sacked_out -= acked - 1; } tcp_check_reno_reordering(sk, acked); tcp_verify_left_out(tp); @@ -1597,10 +1694,10 @@ int tcp_use_frto(struct sock *sk) tcp_for_write_queue_from(skb, sk) { if (skb == tcp_send_head(sk)) break; - if (TCP_SKB_CB(skb)->sacked&TCPCB_RETRANS) + if (TCP_SKB_CB(skb)->sacked & TCPCB_RETRANS) return 0; /* Short-circuit when first non-SACKed skb has been checked */ - if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED)) + if (!(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) break; } return 1; @@ -1704,11 +1801,13 @@ 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)... */ - if ((tp->frto_counter == 1) && !(flag&FLAG_DATA_ACKED)) { + if ((tp->frto_counter == 1) && !(flag & FLAG_DATA_ACKED)) { /* For some reason this R-bit might get cleared? */ if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) tp->retrans_out += tcp_skb_pcount(skb); @@ -1717,11 +1816,11 @@ 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 */ - if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED) && + if (!(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) && !after(TCP_SKB_CB(skb)->end_seq, tp->frto_highmark)) { TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; tp->lost_out += tcp_skb_pcount(skb); @@ -1736,7 +1835,7 @@ static void tcp_enter_frto_loss(struct sock *sk, int allowed_segments, int flag) tp->bytes_acked = 0; tp->reordering = min_t(unsigned int, tp->reordering, - sysctl_tcp_reordering); + sysctl_tcp_reordering); tcp_set_ca_state(sk, TCP_CA_Loss); tp->high_seq = tp->frto_highmark; TCP_ECN_queue_cwr(tp); @@ -1803,7 +1902,7 @@ void tcp_enter_loss(struct sock *sk, int how) if (skb == tcp_send_head(sk)) break; - if (TCP_SKB_CB(skb)->sacked&TCPCB_RETRANS) + if (TCP_SKB_CB(skb)->sacked & TCPCB_RETRANS) tp->undo_marker = 0; TCP_SKB_CB(skb)->sacked &= (~TCPCB_TAGBITS)|TCPCB_SACKED_ACKED; if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED) || how) { @@ -1815,7 +1914,7 @@ void tcp_enter_loss(struct sock *sk, int how) tcp_verify_left_out(tp); tp->reordering = min_t(unsigned int, tp->reordering, - sysctl_tcp_reordering); + sysctl_tcp_reordering); tcp_set_ca_state(sk, TCP_CA_Loss); tp->high_seq = tp->snd_nxt; TCP_ECN_queue_cwr(tp); @@ -1823,18 +1922,15 @@ void tcp_enter_loss(struct sock *sk, int how) tp->frto_counter = 0; } -static int tcp_check_sack_reneging(struct sock *sk) +/* If ACK arrived pointing to a remembered SACK, it means that our + * remembered SACKs do not reflect real state of receiver i.e. + * receiver _host_ is heavily congested (or buggy). + * + * Do processing similar to RTO timeout. + */ +static int tcp_check_sack_reneging(struct sock *sk, int flag) { - struct sk_buff *skb; - - /* If ACK arrived pointing to a remembered SACK, - * it means that our remembered SACKs do not reflect - * real state of receiver i.e. - * receiver _host_ is heavily congested (or buggy). - * Do processing similar to RTO timeout. - */ - if ((skb = tcp_write_queue_head(sk)) != NULL && - (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) { + if (flag & FLAG_SACK_RENEGING) { struct inet_connection_sock *icsk = inet_csk(sk); NET_INC_STATS_BH(LINUX_MIB_TCPSACKRENEGING); @@ -1850,7 +1946,27 @@ static int tcp_check_sack_reneging(struct sock *sk) static inline int tcp_fackets_out(struct tcp_sock *tp) { - return tcp_is_reno(tp) ? tp->sacked_out+1 : tp->fackets_out; + return tcp_is_reno(tp) ? tp->sacked_out + 1 : tp->fackets_out; +} + +/* Heurestics to calculate number of duplicate ACKs. There's no dupACKs + * counter when SACK is enabled (without SACK, sacked_out is used for + * that purpose). + * + * Instead, with FACK TCP uses fackets_out that includes both SACKed + * segments up to the highest received SACK block so far and holes in + * between them. + * + * With reordering, holes may still be in flight, so RFC3517 recovery + * uses pure sacked_out (total number of SACKed segments) even though + * it violates the RFC that uses duplicate ACKs, often these are equal + * but when e.g. out-of-window ACKs or packet duplication occurs, + * they differ. Since neither occurs due to loss, TCP should really + * ignore them. + */ +static inline int tcp_dupack_heurestics(struct tcp_sock *tp) +{ + return tcp_is_fack(tp) ? tp->fackets_out : tp->sacked_out + 1; } static inline int tcp_skb_timedout(struct sock *sk, struct sk_buff *skb) @@ -1973,13 +2089,13 @@ static int tcp_time_to_recover(struct sock *sk) return 1; /* Not-A-Trick#2 : Classic rule... */ - if (tcp_fackets_out(tp) > tp->reordering) + if (tcp_dupack_heurestics(tp) > tp->reordering) return 1; /* Trick#3 : when we use RFC2988 timer restart, fast * retransmit can be triggered by timeout of queue head. */ - if (tcp_head_timedout(sk)) + if (tcp_is_fack(tp) && tcp_head_timedout(sk)) return 1; /* Trick#4: It is still not OK... But will it be useful to delay @@ -2003,17 +2119,18 @@ static int tcp_time_to_recover(struct sock *sk) * retransmitted past LOST markings in the first place? I'm not fully sure * about undo and end of connection cases, which can cause R without L? */ -static void tcp_verify_retransmit_hint(struct tcp_sock *tp, - struct sk_buff *skb) +static void tcp_verify_retransmit_hint(struct tcp_sock *tp, struct sk_buff *skb) { if ((tp->retransmit_skb_hint != NULL) && before(TCP_SKB_CB(skb)->seq, - TCP_SKB_CB(tp->retransmit_skb_hint)->seq)) + TCP_SKB_CB(tp->retransmit_skb_hint)->seq)) tp->retransmit_skb_hint = NULL; } -/* Mark head of queue up as lost. */ -static void tcp_mark_head_lost(struct sock *sk, int packets) +/* Mark head of queue up as lost. With RFC3517 SACK, the packets is + * is against sacked "cnt", otherwise it's against facked "cnt" + */ +static void tcp_mark_head_lost(struct sock *sk, int packets, int fast_rexmit) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; @@ -2035,8 +2152,13 @@ static void tcp_mark_head_lost(struct sock *sk, int packets) /* this is not the most efficient way to do this... */ tp->lost_skb_hint = skb; tp->lost_cnt_hint = cnt; - cnt += tcp_skb_pcount(skb); - if (cnt > packets || after(TCP_SKB_CB(skb)->end_seq, tp->high_seq)) + + if (tcp_is_fack(tp) || + (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) + cnt += tcp_skb_pcount(skb); + + if (((!fast_rexmit || (tp->lost_out > 0)) && (cnt > packets)) || + after(TCP_SKB_CB(skb)->end_seq, tp->high_seq)) break; if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_SACKED_ACKED|TCPCB_LOST))) { TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; @@ -2049,17 +2171,22 @@ static void tcp_mark_head_lost(struct sock *sk, int packets) /* Account newly detected lost packet(s) */ -static void tcp_update_scoreboard(struct sock *sk) +static void tcp_update_scoreboard(struct sock *sk, int fast_rexmit) { struct tcp_sock *tp = tcp_sk(sk); - if (tcp_is_fack(tp)) { + if (tcp_is_reno(tp)) { + tcp_mark_head_lost(sk, 1, fast_rexmit); + } else if (tcp_is_fack(tp)) { int lost = tp->fackets_out - tp->reordering; if (lost <= 0) lost = 1; - tcp_mark_head_lost(sk, lost); + tcp_mark_head_lost(sk, lost, fast_rexmit); } else { - tcp_mark_head_lost(sk, 1); + int sacked_upto = tp->sacked_out - tp->reordering; + if (sacked_upto < 0) + sacked_upto = 0; + tcp_mark_head_lost(sk, sacked_upto, fast_rexmit); } /* New heuristics: it is possible only after we switched @@ -2067,7 +2194,7 @@ static void tcp_update_scoreboard(struct sock *sk) * Hence, we can detect timed out packets during fast * retransmit without falling to slow start. */ - if (!tcp_is_reno(tp) && tcp_head_timedout(sk)) { + if (tcp_is_fack(tp) && tcp_head_timedout(sk)) { struct sk_buff *skb; skb = tp->scoreboard_skb_hint ? tp->scoreboard_skb_hint @@ -2098,7 +2225,7 @@ static void tcp_update_scoreboard(struct sock *sk) static inline void tcp_moderate_cwnd(struct tcp_sock *tp) { tp->snd_cwnd = min(tp->snd_cwnd, - tcp_packets_in_flight(tp)+tcp_max_burst(tp)); + tcp_packets_in_flight(tp) + tcp_max_burst(tp)); tp->snd_cwnd_stamp = tcp_time_stamp; } @@ -2118,15 +2245,15 @@ static void tcp_cwnd_down(struct sock *sk, int flag) struct tcp_sock *tp = tcp_sk(sk); int decr = tp->snd_cwnd_cnt + 1; - if ((flag&(FLAG_ANY_PROGRESS|FLAG_DSACKING_ACK)) || - (tcp_is_reno(tp) && !(flag&FLAG_NOT_DUP))) { - tp->snd_cwnd_cnt = decr&1; + if ((flag & (FLAG_ANY_PROGRESS | FLAG_DSACKING_ACK)) || + (tcp_is_reno(tp) && !(flag & FLAG_NOT_DUP))) { + tp->snd_cwnd_cnt = decr & 1; decr >>= 1; if (decr && tp->snd_cwnd > tcp_cwnd_min(sk)) tp->snd_cwnd -= decr; - tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp)+1); + tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp) + 1); tp->snd_cwnd_stamp = tcp_time_stamp; } } @@ -2170,7 +2297,7 @@ static void tcp_undo_cwr(struct sock *sk, const int undo) if (icsk->icsk_ca_ops->undo_cwnd) tp->snd_cwnd = icsk->icsk_ca_ops->undo_cwnd(sk); else - tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh<<1); + tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh << 1); if (undo && tp->prior_ssthresh > tp->snd_ssthresh) { tp->snd_ssthresh = tp->prior_ssthresh; @@ -2189,8 +2316,7 @@ static void tcp_undo_cwr(struct sock *sk, const int undo) static inline int tcp_may_undo(struct tcp_sock *tp) { - return tp->undo_marker && - (!tp->undo_retrans || tcp_packet_delayed(tp)); + return tp->undo_marker && (!tp->undo_retrans || tcp_packet_delayed(tp)); } /* People celebrate: "We love our President!" */ @@ -2240,7 +2366,7 @@ static int tcp_try_undo_partial(struct sock *sk, int acked) { struct tcp_sock *tp = tcp_sk(sk); /* Partial ACK arrived. Force Hoe's retransmit. */ - int failed = tcp_is_reno(tp) || tp->fackets_out>tp->reordering; + int failed = tcp_is_reno(tp) || (tcp_fackets_out(tp) > tp->reordering); if (tcp_may_undo(tp)) { /* Plain luck! Hole if filled with delayed @@ -2309,7 +2435,7 @@ static void tcp_try_to_open(struct sock *sk, int flag) if (tp->retrans_out == 0) tp->retrans_stamp = 0; - if (flag&FLAG_ECE) + if (flag & FLAG_ECE) tcp_enter_cwr(sk, 1); if (inet_csk(sk)->icsk_ca_state != TCP_CA_CWR) { @@ -2355,7 +2481,6 @@ static void tcp_mtup_probe_success(struct sock *sk, struct sk_buff *skb) tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); } - /* Process an event, which can update packets-in-flight not trivially. * Main goal of this function is to calculate new estimate for left_out, * taking into account both packets sitting in receiver's buffer and @@ -2367,38 +2492,35 @@ static void tcp_mtup_probe_success(struct sock *sk, struct sk_buff *skb) * It does _not_ decide what to send, it is made in function * tcp_xmit_retransmit_queue(). */ -static void -tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) +static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) { struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); - int is_dupack = !(flag&(FLAG_SND_UNA_ADVANCED|FLAG_NOT_DUP)); - int do_lost = is_dupack || ((flag&FLAG_DATA_SACKED) && - (tp->fackets_out > tp->reordering)); + int is_dupack = !(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP)); + int do_lost = is_dupack || ((flag & FLAG_DATA_SACKED) && + (tcp_fackets_out(tp) > tp->reordering)); + int fast_rexmit = 0; - /* Some technical things: - * 1. Reno does not count dupacks (sacked_out) automatically. */ - if (!tp->packets_out) + if (WARN_ON(!tp->packets_out && tp->sacked_out)) tp->sacked_out = 0; - if (WARN_ON(!tp->sacked_out && tp->fackets_out)) tp->fackets_out = 0; /* Now state machine starts. * A. ECE, hence prohibit cwnd undoing, the reduction is required. */ - if (flag&FLAG_ECE) + if (flag & FLAG_ECE) tp->prior_ssthresh = 0; /* B. In all the states check for reneging SACKs. */ - if (tp->sacked_out && tcp_check_sack_reneging(sk)) + if (tcp_check_sack_reneging(sk, flag)) return; /* C. Process data loss notification, provided it is valid. */ - if ((flag&FLAG_DATA_LOST) && + if (tcp_is_fack(tp) && (flag & FLAG_DATA_LOST) && before(tp->snd_una, tp->high_seq) && icsk->icsk_ca_state != TCP_CA_Open && tp->fackets_out > tp->reordering) { - tcp_mark_head_lost(sk, tp->fackets_out - tp->reordering); + tcp_mark_head_lost(sk, tp->fackets_out - tp->reordering, 0); NET_INC_STATS_BH(LINUX_MIB_TCPLOSS); } @@ -2458,7 +2580,7 @@ tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) do_lost = tcp_try_undo_partial(sk, pkts_acked); break; case TCP_CA_Loss: - if (flag&FLAG_DATA_ACKED) + if (flag & FLAG_DATA_ACKED) icsk->icsk_retransmits = 0; if (!tcp_try_undo_loss(sk)) { tcp_moderate_cwnd(tp); @@ -2508,7 +2630,7 @@ tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) tp->undo_retrans = tp->retrans_out; if (icsk->icsk_ca_state < TCP_CA_CWR) { - if (!(flag&FLAG_ECE)) + if (!(flag & FLAG_ECE)) tp->prior_ssthresh = tcp_current_ssthresh(sk); tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk); TCP_ECN_queue_cwr(tp); @@ -2517,10 +2639,11 @@ tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) tp->bytes_acked = 0; tp->snd_cwnd_cnt = 0; tcp_set_ca_state(sk, TCP_CA_Recovery); + fast_rexmit = 1; } - if (do_lost || tcp_head_timedout(sk)) - tcp_update_scoreboard(sk); + if (do_lost || (tcp_is_fack(tp) && tcp_head_timedout(sk))) + tcp_update_scoreboard(sk, fast_rexmit); tcp_cwnd_down(sk, flag); tcp_xmit_retransmit_queue(sk); } @@ -2584,11 +2707,10 @@ static inline void tcp_ack_update_rtt(struct sock *sk, const int flag, tcp_ack_no_tstamp(sk, seq_rtt, flag); } -static void tcp_cong_avoid(struct sock *sk, u32 ack, - u32 in_flight, int good) +static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight) { const struct inet_connection_sock *icsk = inet_csk(sk); - icsk->icsk_ca_ops->cong_avoid(sk, ack, in_flight, good); + icsk->icsk_ca_ops->cong_avoid(sk, ack, in_flight); tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp; } @@ -2602,7 +2724,8 @@ static void tcp_rearm_rto(struct sock *sk) if (!tp->packets_out) { inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS); } else { - inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, inet_csk(sk)->icsk_rto, TCP_RTO_MAX); + inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, + inet_csk(sk)->icsk_rto, TCP_RTO_MAX); } } @@ -2631,8 +2754,7 @@ 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, - int prior_fackets) +static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets) { struct tcp_sock *tp = tcp_sk(sk); const struct inet_connection_sock *icsk = inet_csk(sk); @@ -2640,31 +2762,32 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p, u32 now = tcp_time_stamp; int fully_acked = 1; int flag = 0; - int prior_packets = tp->packets_out; - u32 cnt = 0; + u32 pkts_acked = 0; u32 reord = tp->packets_out; s32 seq_rtt = -1; + s32 ca_seq_rtt = -1; ktime_t last_ackt = net_invalid_timestamp(); while ((skb = tcp_write_queue_head(sk)) && skb != tcp_send_head(sk)) { struct tcp_skb_cb *scb = TCP_SKB_CB(skb); u32 end_seq; - u32 packets_acked; + u32 acked_pcount; u8 sacked = scb->sacked; + /* Determine how many packets and what bytes were acked, tso and else */ if (after(scb->end_seq, tp->snd_una)) { if (tcp_skb_pcount(skb) == 1 || !after(tp->snd_una, scb->seq)) break; - packets_acked = tcp_tso_acked(sk, skb); - if (!packets_acked) + acked_pcount = tcp_tso_acked(sk, skb); + if (!acked_pcount) break; fully_acked = 0; end_seq = tp->snd_una; } else { - packets_acked = tcp_skb_pcount(skb); + acked_pcount = tcp_skb_pcount(skb); end_seq = scb->end_seq; } @@ -2674,43 +2797,34 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p, tcp_mtup_probe_success(sk, skb); } - if (sacked) { - if (sacked & TCPCB_RETRANS) { - if (sacked & TCPCB_SACKED_RETRANS) - tp->retrans_out -= packets_acked; - flag |= FLAG_RETRANS_DATA_ACKED; - seq_rtt = -1; - 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; - } - if (!(sacked & TCPCB_SACKED_ACKED)) - reord = min(cnt, reord); - } - - if (sacked & TCPCB_SACKED_ACKED) - tp->sacked_out -= packets_acked; - if (sacked & TCPCB_LOST) - tp->lost_out -= packets_acked; - - if ((sacked & TCPCB_URG) && tp->urg_mode && - !before(end_seq, tp->snd_up)) - tp->urg_mode = 0; + if (sacked & TCPCB_RETRANS) { + if (sacked & TCPCB_SACKED_RETRANS) + tp->retrans_out -= acked_pcount; + flag |= FLAG_RETRANS_DATA_ACKED; + ca_seq_rtt = -1; + seq_rtt = -1; + if ((flag & FLAG_DATA_ACKED) || (acked_pcount > 1)) + flag |= FLAG_NONHEAD_RETRANS_ACKED; } else { + ca_seq_rtt = now - scb->when; + last_ackt = skb->tstamp; if (seq_rtt < 0) { - seq_rtt = now - scb->when; - if (fully_acked) - last_ackt = skb->tstamp; + seq_rtt = ca_seq_rtt; } - reord = min(cnt, reord); + if (!(sacked & TCPCB_SACKED_ACKED)) + reord = min(pkts_acked, reord); } - tp->packets_out -= packets_acked; - cnt += packets_acked; + + if (sacked & TCPCB_SACKED_ACKED) + tp->sacked_out -= acked_pcount; + if (sacked & TCPCB_LOST) + tp->lost_out -= acked_pcount; + + if (unlikely(tp->urg_mode && !before(end_seq, tp->snd_up))) + tp->urg_mode = 0; + + tp->packets_out -= acked_pcount; + pkts_acked += acked_pcount; /* Initial outgoing SYN's get put onto the write_queue * just like anything else we transmit. It is not @@ -2730,12 +2844,14 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p, break; tcp_unlink_write_queue(skb, sk); - sk_stream_free_skb(sk, skb); + sk_wmem_free_skb(sk, skb); tcp_clear_all_retrans_hints(tp); } + if (skb && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) + flag |= FLAG_SACK_RENEGING; + if (flag & FLAG_ACKED) { - u32 pkts_acked = prior_packets - tp->packets_out; const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops; @@ -2751,9 +2867,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p, } 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 (ca_ops->pkts_acked) { s32 rtt_us = -1; @@ -2765,8 +2879,8 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p, net_invalid_timestamp())) rtt_us = ktime_us_delta(ktime_get_real(), last_ackt); - else if (seq_rtt > 0) - rtt_us = jiffies_to_usecs(seq_rtt); + else if (ca_seq_rtt > 0) + rtt_us = jiffies_to_usecs(ca_seq_rtt); } ca_ops->pkts_acked(sk, pkts_acked, rtt_us); @@ -2796,7 +2910,6 @@ static int tcp_clean_rtx_queue(struct sock *sk, s32 *seq_rtt_p, } } #endif - *seq_rtt_p = seq_rtt; return flag; } @@ -2807,8 +2920,7 @@ static void tcp_ack_probe(struct sock *sk) /* Was it a usable window open? */ - if (!after(TCP_SKB_CB(tcp_send_head(sk))->end_seq, - tp->snd_una + tp->snd_wnd)) { + if (!after(TCP_SKB_CB(tcp_send_head(sk))->end_seq, tcp_wnd_end(tp))) { icsk->icsk_backoff = 0; inet_csk_clear_xmit_timer(sk, ICSK_TIME_PROBE0); /* Socket must be waked up by subsequent tcp_data_snd_check(). @@ -2837,8 +2949,9 @@ static inline int tcp_may_raise_cwnd(const struct sock *sk, const int flag) /* Check that window update is acceptable. * The function assumes that snd_una<=ack<=snd_next. */ -static inline int tcp_may_update_window(const struct tcp_sock *tp, const u32 ack, - const u32 ack_seq, const u32 nwin) +static inline int tcp_may_update_window(const struct tcp_sock *tp, + const u32 ack, const u32 ack_seq, + const u32 nwin) { return (after(ack, tp->snd_una) || after(ack_seq, tp->snd_wl1) || @@ -2907,7 +3020,7 @@ static void tcp_ratehalving_spur_to_response(struct sock *sk) static void tcp_undo_spur_to_response(struct sock *sk, int flag) { - if (flag&FLAG_ECE) + if (flag & FLAG_ECE) tcp_ratehalving_spur_to_response(sk); else tcp_undo_cwr(sk, 1); @@ -2950,7 +3063,7 @@ static int tcp_process_frto(struct sock *sk, int flag) tcp_verify_left_out(tp); /* Duplicate the behavior from Loss state (fastretrans_alert) */ - if (flag&FLAG_DATA_ACKED) + if (flag & FLAG_DATA_ACKED) inet_csk(sk)->icsk_retransmits = 0; if ((flag & FLAG_NONHEAD_RETRANS_ACKED) || @@ -2967,16 +3080,16 @@ static int tcp_process_frto(struct sock *sk, int flag) * ACK isn't duplicate nor advances window, e.g., opposite dir * data, winupdate */ - if (!(flag&FLAG_ANY_PROGRESS) && (flag&FLAG_NOT_DUP)) + if (!(flag & FLAG_ANY_PROGRESS) && (flag & FLAG_NOT_DUP)) return 1; - if (!(flag&FLAG_DATA_ACKED)) { + if (!(flag & FLAG_DATA_ACKED)) { tcp_enter_frto_loss(sk, (tp->frto_counter == 1 ? 0 : 3), flag); return 1; } } else { - if (!(flag&FLAG_DATA_ACKED) && (tp->frto_counter == 1)) { + if (!(flag & FLAG_DATA_ACKED) && (tp->frto_counter == 1)) { /* Prevent sending of new data. */ tp->snd_cwnd = min(tp->snd_cwnd, tcp_packets_in_flight(tp)); @@ -2984,10 +3097,12 @@ static int tcp_process_frto(struct sock *sk, int flag) } if ((tp->frto_counter >= 2) && - (!(flag&FLAG_FORWARD_PROGRESS) || - ((flag&FLAG_DATA_SACKED) && !(flag&FLAG_ONLY_ORIG_SACKED)))) { + (!(flag & FLAG_FORWARD_PROGRESS) || + ((flag & FLAG_DATA_SACKED) && + !(flag & FLAG_ONLY_ORIG_SACKED)))) { /* RFC4138 shortcoming (see comment above) */ - if (!(flag&FLAG_FORWARD_PROGRESS) && (flag&FLAG_NOT_DUP)) + if (!(flag & FLAG_FORWARD_PROGRESS) && + (flag & FLAG_NOT_DUP)) return 1; tcp_enter_frto_loss(sk, 3, flag); @@ -2996,17 +3111,13 @@ static int tcp_process_frto(struct sock *sk, int flag) } if (tp->frto_counter == 1) { - /* Sending of the next skb must be allowed or no F-RTO */ - if (!tcp_send_head(sk) || - after(TCP_SKB_CB(tcp_send_head(sk))->end_seq, - tp->snd_una + tp->snd_wnd)) { - tcp_enter_frto_loss(sk, (tp->frto_counter == 1 ? 2 : 3), - flag); - return 1; - } - + /* tcp_may_send_now needs to see updated state */ tp->snd_cwnd = tcp_packets_in_flight(tp) + 2; tp->frto_counter = 2; + + if (!tcp_may_send_now(sk)) + tcp_enter_frto_loss(sk, 2, flag); + return 1; } else { switch (sysctl_tcp_frto_response) { @@ -3037,7 +3148,6 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) 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; @@ -3058,12 +3168,14 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) tp->bytes_acked += ack - prior_snd_una; else if (icsk->icsk_ca_state == TCP_CA_Loss) /* we assume just one segment left network */ - tp->bytes_acked += min(ack - prior_snd_una, tp->mss_cache); + tp->bytes_acked += min(ack - prior_snd_una, + tp->mss_cache); } prior_fackets = tp->fackets_out; + prior_in_flight = tcp_packets_in_flight(tp); - if (!(flag&FLAG_SLOWPATH) && after(ack, prior_snd_una)) { + if (!(flag & FLAG_SLOWPATH) && after(ack, prior_snd_una)) { /* Window is constant, pure forward advance. * No more checks are required. * Note, we use the fact that SND.UNA>=SND.WL2. @@ -3101,29 +3213,28 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) if (!prior_packets) goto no_queue; - 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, prior_fackets); + flag |= tcp_clean_rtx_queue(sk, 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. */ if ((flag & FLAG_DATA_ACKED) && !frto_cwnd && tcp_may_raise_cwnd(sk, flag)) - tcp_cong_avoid(sk, ack, prior_in_flight, 0); - tcp_fastretrans_alert(sk, prior_packets - tp->packets_out, flag); + tcp_cong_avoid(sk, ack, prior_in_flight); + tcp_fastretrans_alert(sk, prior_packets - tp->packets_out, + flag); } else { if ((flag & FLAG_DATA_ACKED) && !frto_cwnd) - tcp_cong_avoid(sk, ack, prior_in_flight, 1); + tcp_cong_avoid(sk, ack, prior_in_flight); } - if ((flag & FLAG_FORWARD_PROGRESS) || !(flag&FLAG_NOT_DUP)) + if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) dst_confirm(sk->sk_dst_cache); return 1; @@ -3148,22 +3259,22 @@ uninteresting_ack: return 0; } - /* Look for tcp options. Normally only called on SYN and SYNACK packets. * But, this can also be called on packets in the established flow when * the fast version below fails. */ -void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, int estab) +void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, + int estab) { unsigned char *ptr; struct tcphdr *th = tcp_hdr(skb); - int length=(th->doff*4)-sizeof(struct tcphdr); + int length = (th->doff * 4) - sizeof(struct tcphdr); ptr = (unsigned char *)(th + 1); opt_rx->saw_tstamp = 0; while (length > 0) { - int opcode=*ptr++; + int opcode = *ptr++; int opsize; switch (opcode) { @@ -3252,7 +3363,7 @@ void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, static int tcp_fast_parse_options(struct sk_buff *skb, struct tcphdr *th, struct tcp_sock *tp) { - if (th->doff == sizeof(struct tcphdr)>>2) { + if (th->doff == sizeof(struct tcphdr) >> 2) { tp->rx_opt.saw_tstamp = 0; return 0; } else if (tp->rx_opt.tstamp_ok && @@ -3337,7 +3448,8 @@ static int tcp_disordered_ack(const struct sock *sk, const struct sk_buff *skb) (s32)(tp->rx_opt.ts_recent - tp->rx_opt.rcv_tsval) <= (inet_csk(sk)->icsk_rto * 1024) / HZ); } -static inline int tcp_paws_discard(const struct sock *sk, const struct sk_buff *skb) +static inline int tcp_paws_discard(const struct sock *sk, + const struct sk_buff *skb) { const struct tcp_sock *tp = tcp_sk(sk); return ((s32)(tp->rx_opt.ts_recent - tp->rx_opt.rcv_tsval) > TCP_PAWS_WINDOW && @@ -3369,16 +3481,16 @@ static void tcp_reset(struct sock *sk) { /* We want the right error as BSD sees it (and indeed as we do). */ switch (sk->sk_state) { - case TCP_SYN_SENT: - sk->sk_err = ECONNREFUSED; - break; - case TCP_CLOSE_WAIT: - sk->sk_err = EPIPE; - break; - case TCP_CLOSE: - return; - default: - sk->sk_err = ECONNRESET; + case TCP_SYN_SENT: + sk->sk_err = ECONNREFUSED; + break; + case TCP_CLOSE_WAIT: + sk->sk_err = EPIPE; + break; + case TCP_CLOSE: + return; + default: + sk->sk_err = ECONNRESET; } if (!sock_flag(sk, SOCK_DEAD)) @@ -3411,43 +3523,43 @@ static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th) sock_set_flag(sk, SOCK_DONE); switch (sk->sk_state) { - case TCP_SYN_RECV: - case TCP_ESTABLISHED: - /* Move to CLOSE_WAIT */ - tcp_set_state(sk, TCP_CLOSE_WAIT); - inet_csk(sk)->icsk_ack.pingpong = 1; - break; + case TCP_SYN_RECV: + case TCP_ESTABLISHED: + /* Move to CLOSE_WAIT */ + tcp_set_state(sk, TCP_CLOSE_WAIT); + inet_csk(sk)->icsk_ack.pingpong = 1; + break; - case TCP_CLOSE_WAIT: - case TCP_CLOSING: - /* Received a retransmission of the FIN, do - * nothing. - */ - break; - case TCP_LAST_ACK: - /* RFC793: Remain in the LAST-ACK state. */ - break; + case TCP_CLOSE_WAIT: + case TCP_CLOSING: + /* Received a retransmission of the FIN, do + * nothing. + */ + break; + case TCP_LAST_ACK: + /* RFC793: Remain in the LAST-ACK state. */ + break; - case TCP_FIN_WAIT1: - /* This case occurs when a simultaneous close - * happens, we must ack the received FIN and - * enter the CLOSING state. - */ - tcp_send_ack(sk); - tcp_set_state(sk, TCP_CLOSING); - break; - case TCP_FIN_WAIT2: - /* Received a FIN -- send ACK and enter TIME_WAIT. */ - tcp_send_ack(sk); - tcp_time_wait(sk, TCP_TIME_WAIT, 0); - break; - default: - /* Only TCP_LISTEN and TCP_CLOSE are left, in these - * cases we should never reach this piece of code. - */ - printk(KERN_ERR "%s: Impossible, sk->sk_state=%d\n", - __FUNCTION__, sk->sk_state); - break; + case TCP_FIN_WAIT1: + /* This case occurs when a simultaneous close + * happens, we must ack the received FIN and + * enter the CLOSING state. + */ + tcp_send_ack(sk); + tcp_set_state(sk, TCP_CLOSING); + break; + case TCP_FIN_WAIT2: + /* Received a FIN -- send ACK and enter TIME_WAIT. */ + tcp_send_ack(sk); + tcp_time_wait(sk, TCP_TIME_WAIT, 0); + break; + default: + /* Only TCP_LISTEN and TCP_CLOSE are left, in these + * cases we should never reach this piece of code. + */ + printk(KERN_ERR "%s: Impossible, sk->sk_state=%d\n", + __FUNCTION__, sk->sk_state); + break; } /* It _is_ possible, that we have something out-of-order _after_ FIN. @@ -3456,7 +3568,7 @@ static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th) __skb_queue_purge(&tp->out_of_order_queue); if (tcp_is_sack(tp)) tcp_sack_reset(&tp->rx_opt); - sk_stream_mem_reclaim(sk); + sk_mem_reclaim(sk); if (!sock_flag(sk, SOCK_DEAD)) { sk->sk_state_change(sk); @@ -3464,13 +3576,14 @@ static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th) /* Do not send POLL_HUP for half duplex close. */ if (sk->sk_shutdown == SHUTDOWN_MASK || sk->sk_state == TCP_CLOSE) - sk_wake_async(sk, 1, POLL_HUP); + sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP); else - sk_wake_async(sk, 1, POLL_IN); + sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); } } -static inline int tcp_sack_extend(struct tcp_sack_block *sp, u32 seq, u32 end_seq) +static inline int tcp_sack_extend(struct tcp_sack_block *sp, u32 seq, + u32 end_seq) { if (!after(seq, sp->end_seq) && !after(sp->start_seq, end_seq)) { if (before(seq, sp->start_seq)) @@ -3493,7 +3606,8 @@ static void tcp_dsack_set(struct tcp_sock *tp, u32 seq, u32 end_seq) tp->rx_opt.dsack = 1; tp->duplicate_sack[0].start_seq = seq; tp->duplicate_sack[0].end_seq = end_seq; - tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + 1, 4 - tp->rx_opt.tstamp_ok); + tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + 1, + 4 - tp->rx_opt.tstamp_ok); } } @@ -3533,12 +3647,12 @@ static void tcp_sack_maybe_coalesce(struct tcp_sock *tp) { int this_sack; struct tcp_sack_block *sp = &tp->selective_acks[0]; - struct tcp_sack_block *swalk = sp+1; + struct tcp_sack_block *swalk = sp + 1; /* See if the recent change to the first SACK eats into * or hits the sequence space of other SACK blocks, if so coalesce. */ - for (this_sack = 1; this_sack < tp->rx_opt.num_sacks; ) { + for (this_sack = 1; this_sack < tp->rx_opt.num_sacks;) { if (tcp_sack_extend(sp, swalk->start_seq, swalk->end_seq)) { int i; @@ -3546,16 +3660,19 @@ static void tcp_sack_maybe_coalesce(struct tcp_sock *tp) * Decrease num_sacks. */ tp->rx_opt.num_sacks--; - tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack, 4 - tp->rx_opt.tstamp_ok); - for (i=this_sack; i < tp->rx_opt.num_sacks; i++) - sp[i] = sp[i+1]; + tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + + tp->rx_opt.dsack, + 4 - tp->rx_opt.tstamp_ok); + for (i = this_sack; i < tp->rx_opt.num_sacks; i++) + sp[i] = sp[i + 1]; continue; } this_sack++, swalk++; } } -static inline void tcp_sack_swap(struct tcp_sack_block *sack1, struct tcp_sack_block *sack2) +static inline void tcp_sack_swap(struct tcp_sack_block *sack1, + struct tcp_sack_block *sack2) { __u32 tmp; @@ -3578,11 +3695,11 @@ static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq) if (!cur_sacks) goto new_sack; - for (this_sack=0; this_sack0; this_sack--, sp--) - tcp_sack_swap(sp, sp-1); + for (; this_sack > 0; this_sack--, sp--) + tcp_sack_swap(sp, sp - 1); if (cur_sacks > 1) tcp_sack_maybe_coalesce(tp); return; @@ -3601,14 +3718,15 @@ static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq) sp--; } for (; this_sack > 0; this_sack--, sp--) - *sp = *(sp-1); + *sp = *(sp - 1); new_sack: /* Build the new head SACK, and we're done. */ sp->start_seq = seq; sp->end_seq = end_seq; tp->rx_opt.num_sacks++; - tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack, 4 - tp->rx_opt.tstamp_ok); + tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack, + 4 - tp->rx_opt.tstamp_ok); } /* RCV.NXT advances, some SACKs should be eaten. */ @@ -3626,7 +3744,7 @@ static void tcp_sack_remove(struct tcp_sock *tp) return; } - for (this_sack = 0; this_sack < num_sacks; ) { + for (this_sack = 0; this_sack < num_sacks;) { /* Check if the start of the sack is covered by RCV.NXT. */ if (!before(tp->rcv_nxt, sp->start_seq)) { int i; @@ -3645,7 +3763,9 @@ static void tcp_sack_remove(struct tcp_sock *tp) } if (num_sacks != tp->rx_opt.num_sacks) { tp->rx_opt.num_sacks = num_sacks; - tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack, 4 - tp->rx_opt.tstamp_ok); + tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + + tp->rx_opt.dsack, + 4 - tp->rx_opt.tstamp_ok); } } @@ -3698,14 +3818,14 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) goto drop; - __skb_pull(skb, th->doff*4); + __skb_pull(skb, th->doff * 4); TCP_ECN_accept_cwr(tp, skb); if (tp->rx_opt.dsack) { tp->rx_opt.dsack = 0; tp->rx_opt.eff_sacks = min_t(unsigned int, tp->rx_opt.num_sacks, - 4 - tp->rx_opt.tstamp_ok); + 4 - tp->rx_opt.tstamp_ok); } /* Queue data for delivery to the user. @@ -3721,7 +3841,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) tp->copied_seq == tp->rcv_nxt && tp->ucopy.len && sock_owned_by_user(sk) && !tp->urg_data) { int chunk = min_t(unsigned int, skb->len, - tp->ucopy.len); + tp->ucopy.len); __set_current_state(TASK_RUNNING); @@ -3739,12 +3859,12 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) queue_and_out: if (eaten < 0 && (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf || - !sk_stream_rmem_schedule(sk, skb))) { + !sk_rmem_schedule(sk, skb->truesize))) { if (tcp_prune_queue(sk) < 0 || - !sk_stream_rmem_schedule(sk, skb)) + !sk_rmem_schedule(sk, skb->truesize)) goto drop; } - sk_stream_set_owner_r(skb, sk); + skb_set_owner_r(skb, sk); __skb_queue_tail(&sk->sk_receive_queue, skb); } tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; @@ -3813,9 +3933,9 @@ drop: TCP_ECN_check_ce(tp, skb); if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf || - !sk_stream_rmem_schedule(sk, skb)) { + !sk_rmem_schedule(sk, skb->truesize)) { if (tcp_prune_queue(sk) < 0 || - !sk_stream_rmem_schedule(sk, skb)) + !sk_rmem_schedule(sk, skb->truesize)) goto drop; } @@ -3826,7 +3946,7 @@ drop: SOCK_DEBUG(sk, "out of order segment: rcv_next %X seq %X - %X\n", tp->rcv_nxt, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq); - sk_stream_set_owner_r(skb, sk); + skb_set_owner_r(skb, sk); if (!skb_peek(&tp->out_of_order_queue)) { /* Initial out of order segment, build 1 SACK. */ @@ -3838,7 +3958,7 @@ drop: tp->selective_acks[0].end_seq = TCP_SKB_CB(skb)->end_seq; } - __skb_queue_head(&tp->out_of_order_queue,skb); + __skb_queue_head(&tp->out_of_order_queue, skb); } else { struct sk_buff *skb1 = tp->out_of_order_queue.prev; u32 seq = TCP_SKB_CB(skb)->seq; @@ -3861,10 +3981,10 @@ drop: if (!after(TCP_SKB_CB(skb1)->seq, seq)) break; } while ((skb1 = skb1->prev) != - (struct sk_buff*)&tp->out_of_order_queue); + (struct sk_buff *)&tp->out_of_order_queue); /* Do skb overlap to previous one? */ - if (skb1 != (struct sk_buff*)&tp->out_of_order_queue && + if (skb1 != (struct sk_buff *)&tp->out_of_order_queue && before(seq, TCP_SKB_CB(skb1)->end_seq)) { if (!after(end_seq, TCP_SKB_CB(skb1)->end_seq)) { /* All the bits are present. Drop. */ @@ -3874,7 +3994,8 @@ drop: } if (after(seq, TCP_SKB_CB(skb1)->seq)) { /* Partial overlap. */ - tcp_dsack_set(tp, seq, TCP_SKB_CB(skb1)->end_seq); + tcp_dsack_set(tp, seq, + TCP_SKB_CB(skb1)->end_seq); } else { skb1 = skb1->prev; } @@ -3883,15 +4004,17 @@ drop: /* And clean segments covered by new one as whole. */ while ((skb1 = skb->next) != - (struct sk_buff*)&tp->out_of_order_queue && + (struct sk_buff *)&tp->out_of_order_queue && after(end_seq, TCP_SKB_CB(skb1)->seq)) { - if (before(end_seq, TCP_SKB_CB(skb1)->end_seq)) { - tcp_dsack_extend(tp, TCP_SKB_CB(skb1)->seq, end_seq); - break; - } - __skb_unlink(skb1, &tp->out_of_order_queue); - tcp_dsack_extend(tp, TCP_SKB_CB(skb1)->seq, TCP_SKB_CB(skb1)->end_seq); - __kfree_skb(skb1); + if (before(end_seq, TCP_SKB_CB(skb1)->end_seq)) { + tcp_dsack_extend(tp, TCP_SKB_CB(skb1)->seq, + end_seq); + break; + } + __skb_unlink(skb1, &tp->out_of_order_queue); + tcp_dsack_extend(tp, TCP_SKB_CB(skb1)->seq, + TCP_SKB_CB(skb1)->end_seq); + __kfree_skb(skb1); } add_sack: @@ -3914,7 +4037,7 @@ tcp_collapse(struct sock *sk, struct sk_buff_head *list, /* First, check that queue is collapsible and find * the point where collapsing can be useful. */ - for (skb = head; skb != tail; ) { + for (skb = head; skb != tail;) { /* No new bits? It is possible on ofo queue. */ if (!before(start, TCP_SKB_CB(skb)->end_seq)) { struct sk_buff *next = skb->next; @@ -3952,9 +4075,9 @@ tcp_collapse(struct sock *sk, struct sk_buff_head *list, /* Too big header? This can happen with IPv6. */ if (copy < 0) return; - if (end-start < copy) - copy = end-start; - nskb = alloc_skb(copy+header, GFP_ATOMIC); + if (end - start < copy) + copy = end - start; + nskb = alloc_skb(copy + header, GFP_ATOMIC); if (!nskb) return; @@ -3968,7 +4091,7 @@ tcp_collapse(struct sock *sk, struct sk_buff_head *list, memcpy(nskb->cb, skb->cb, sizeof(skb->cb)); TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(nskb)->end_seq = start; __skb_insert(nskb, skb->prev, skb, list); - sk_stream_set_owner_r(nskb, sk); + skb_set_owner_r(nskb, sk); /* Copy data, releasing collapsed skbs. */ while (copy > 0) { @@ -4064,9 +4187,9 @@ static int tcp_prune_queue(struct sock *sk) tcp_collapse_ofo_queue(sk); tcp_collapse(sk, &sk->sk_receive_queue, sk->sk_receive_queue.next, - (struct sk_buff*)&sk->sk_receive_queue, + (struct sk_buff *)&sk->sk_receive_queue, tp->copied_seq, tp->rcv_nxt); - sk_stream_mem_reclaim(sk); + sk_mem_reclaim(sk); if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf) return 0; @@ -4086,7 +4209,7 @@ static int tcp_prune_queue(struct sock *sk) */ if (tcp_is_sack(tp)) tcp_sack_reset(&tp->rx_opt); - sk_stream_mem_reclaim(sk); + sk_mem_reclaim(sk); } if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf) @@ -4103,7 +4226,6 @@ static int tcp_prune_queue(struct sock *sk) return -1; } - /* RFC2861, slow part. Adjust cwnd, after it was not full during one rto. * As additional protections, we do not touch cwnd in retransmission phases, * and if application hit its sndbuf limit recently. @@ -4165,8 +4287,8 @@ static void tcp_new_space(struct sock *sk) int sndmem = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) + MAX_TCP_HEADER + 16 + sizeof(struct sk_buff), demanded = max_t(unsigned int, tp->snd_cwnd, - tp->reordering + 1); - sndmem *= 2*demanded; + tp->reordering + 1); + sndmem *= 2 * demanded; if (sndmem > sk->sk_sndbuf) sk->sk_sndbuf = min(sndmem, sysctl_tcp_wmem[2]); tp->snd_cwnd_stamp = tcp_time_stamp; @@ -4207,8 +4329,7 @@ static void __tcp_ack_snd_check(struct sock *sk, int ofo_possible) /* We ACK each frame or... */ tcp_in_quickack_mode(sk) || /* We have out of order data. */ - (ofo_possible && - skb_peek(&tp->out_of_order_queue))) { + (ofo_possible && skb_peek(&tp->out_of_order_queue))) { /* Then ack it now */ tcp_send_ack(sk); } else { @@ -4236,7 +4357,7 @@ static inline void tcp_ack_snd_check(struct sock *sk) * either form (or just set the sysctl tcp_stdurg). */ -static void tcp_check_urg(struct sock * sk, struct tcphdr * th) +static void tcp_check_urg(struct sock *sk, struct tcphdr *th) { struct tcp_sock *tp = tcp_sk(sk); u32 ptr = ntohs(th->urg_ptr); @@ -4285,8 +4406,7 @@ static void tcp_check_urg(struct sock * sk, struct tcphdr * th) * buggy users. */ if (tp->urg_seq == tp->copied_seq && tp->urg_data && - !sock_flag(sk, SOCK_URGINLINE) && - tp->copied_seq != tp->rcv_nxt) { + !sock_flag(sk, SOCK_URGINLINE) && tp->copied_seq != tp->rcv_nxt) { struct sk_buff *skb = skb_peek(&sk->sk_receive_queue); tp->copied_seq++; if (skb && !before(tp->copied_seq, TCP_SKB_CB(skb)->end_seq)) { @@ -4295,8 +4415,8 @@ static void tcp_check_urg(struct sock * sk, struct tcphdr * th) } } - tp->urg_data = TCP_URG_NOTYET; - tp->urg_seq = ptr; + tp->urg_data = TCP_URG_NOTYET; + tp->urg_seq = ptr; /* Disable header prediction. */ tp->pred_flags = 0; @@ -4309,7 +4429,7 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, struct tcphdr *th) /* Check if we get a new urgent pointer - normally not. */ if (th->urg) - tcp_check_urg(sk,th); + tcp_check_urg(sk, th); /* Do we wait for any urgent data? - normally not... */ if (tp->urg_data == TCP_URG_NOTYET) { @@ -4351,7 +4471,8 @@ static int tcp_copy_to_iovec(struct sock *sk, struct sk_buff *skb, int hlen) return err; } -static __sum16 __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb) +static __sum16 __tcp_checksum_complete_user(struct sock *sk, + struct sk_buff *skb) { __sum16 result; @@ -4365,14 +4486,16 @@ static __sum16 __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb return result; } -static inline int tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb) +static inline int tcp_checksum_complete_user(struct sock *sk, + struct sk_buff *skb) { return !skb_csum_unnecessary(skb) && - __tcp_checksum_complete_user(sk, skb); + __tcp_checksum_complete_user(sk, skb); } #ifdef CONFIG_NET_DMA -static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, int hlen) +static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, + int hlen) { struct tcp_sock *tp = tcp_sk(sk); int chunk = skb->len - hlen; @@ -4388,7 +4511,9 @@ static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, int hlen if (tp->ucopy.dma_chan && skb_csum_unnecessary(skb)) { dma_cookie = dma_skb_copy_datagram_iovec(tp->ucopy.dma_chan, - skb, hlen, tp->ucopy.iov, chunk, tp->ucopy.pinned_list); + skb, hlen, + tp->ucopy.iov, chunk, + tp->ucopy.pinned_list); if (dma_cookie < 0) goto out; @@ -4470,7 +4595,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, */ if ((tcp_flag_word(th) & TCP_HP_BITS) == tp->pred_flags && - TCP_SKB_CB(skb)->seq == tp->rcv_nxt) { + TCP_SKB_CB(skb)->seq == tp->rcv_nxt) { int tcp_header_len = tp->tcp_header_len; /* Timestamp header prediction: tcp_header_len @@ -4539,7 +4664,8 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, eaten = 1; } #endif - if (tp->ucopy.task == current && sock_owned_by_user(sk) && !copied_early) { + if (tp->ucopy.task == current && + sock_owned_by_user(sk) && !copied_early) { __set_current_state(TASK_RUNNING); if (!tcp_copy_to_iovec(sk, skb, tcp_header_len)) @@ -4586,9 +4712,9 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, NET_INC_STATS_BH(LINUX_MIB_TCPHPHITS); /* Bulk data transfer: receiver */ - __skb_pull(skb,tcp_header_len); + __skb_pull(skb, tcp_header_len); __skb_queue_tail(&sk->sk_receive_queue, skb); - sk_stream_set_owner_r(skb, sk); + skb_set_owner_r(skb, sk); tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; } @@ -4618,7 +4744,7 @@ no_ack: } slow_path: - if (len < (th->doff<<2) || tcp_checksum_complete_user(sk, skb)) + if (len < (th->doff << 2) || tcp_checksum_complete_user(sk, skb)) goto csum_error; /* @@ -4825,7 +4951,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, if (!sock_flag(sk, SOCK_DEAD)) { sk->sk_state_change(sk); - sk_wake_async(sk, 0, POLL_OUT); + sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT); } if (sk->sk_write_pending || @@ -4868,7 +4994,8 @@ discard: } /* PAWS check. */ - if (tp->rx_opt.ts_recent_stamp && tp->rx_opt.saw_tstamp && tcp_paws_check(&tp->rx_opt, 0)) + if (tp->rx_opt.ts_recent_stamp && tp->rx_opt.saw_tstamp && + tcp_paws_check(&tp->rx_opt, 0)) goto discard_and_undo; if (th->syn) { @@ -4903,7 +5030,6 @@ discard: tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); tcp_initialize_rcv_mss(sk); - tcp_send_synack(sk); #if 0 /* Note, we could accept data and URG from this segment. @@ -4935,7 +5061,6 @@ reset_and_undo: return 1; } - /* * This function implements the receiving procedure of RFC 793 for * all states except ESTABLISHED and TIME_WAIT. @@ -5055,9 +5180,9 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, * are not waked up, because sk->sk_sleep == * NULL and sk->sk_socket == NULL. */ - if (sk->sk_socket) { - sk_wake_async(sk,0,POLL_OUT); - } + if (sk->sk_socket) + sk_wake_async(sk, + SOCK_WAKE_IO, POLL_OUT); tp->snd_una = TCP_SKB_CB(skb)->ack_seq; tp->snd_wnd = ntohs(th->window) << @@ -5069,8 +5194,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, * and does not calculate rtt. * Fix it at least with timestamps. */ - if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr && - !tp->srtt) + if (tp->rx_opt.saw_tstamp && + tp->rx_opt.rcv_tsecr && !tp->srtt) tcp_ack_saw_tstamp(sk, 0); if (tp->rx_opt.tstamp_ok)